.'t
                                                                                                                                                       I
                                                                                                                                                                  r

-------
National Conference on
MANAGEMENT
OF UNCONTROLLED
October 28-30,1981
Washington, D.C.
Co-sponsors
U.S. Environmental Protection Agency
Hazardous Materials Control Research institute


In Affiliation with
American Society of Civil Engineers
Chemical Manufacturers Association
National wildlife Federation
U.S. Coast Guard

-------
               ACKNOWLEDGEMENT
  The National Conference and Exhibition on Management of Uncontrolled Hazardous
Waste Sites was made possible through the involvement of many individuals and organiza-
tions. We would like to express our thanks and appreciation to all who assisted.

                  Co-Sponsors:
                  U.S. Environmental Protection Agency
                  Hazardous Materials Control Research Institute
                  Affiliates:
                  American Society of Civil Engineers
                  Chemical Manufacturers Association
                  National Wildlife Federation
                  U.S. Coast Guard


  The Program Committee is comprised of knowledgeable individuals cooperating to
produce an effective and informative program. These individuals are:
Harold J. Snyder, Jr. (Chairman)           Don Sanning
U.S. Environmental Protection Agency      U.S. Environmental Protection Agency

Harold Bernard                           Lucy Sibold
Hazardous Materials Control Research       {/. 5. Environmental Protection Agency
Institute
                                        Jerry  Steinberg, Ph.D.
frank freestone                          Water & Air Research, Inc.
U.S. Environmental Protection Agency
Wayne Harrington, Ph.D.                 Michael Tobbe
Chemical Manufacturers Association        u-s- Coast Guard
Kenneth Kamlet                          Wayne Tusa
National Wildlife Federation               A merican Society of Civil Engineers


  The concentrated effort necessary to publish a Proceedings of this size and scope in the
time allotted is certainly 'above and beyond'. Our special thanks to Dr. Gary Bennett,
Professor of Biochemical Engineering, The University of Toledo, and Hal Bernard, HMCRI
whose editing allowed for a more uniform Proceedings; to the typesetters and graphics team
who managed to meet the impossible deadlines set; and to the staff of HMCRI, in
particular, Beverly Walcoff, Project Manager and Paula Geary, for coordinating the many
aspects and activities of this Conference.

-------
                                               CONTENTS
                     OVERVIEW
CMA Position on Degree of Hazard  	
  Geraldine V. Cox, Ph.D.
The ASCE Environmental Engineering
Division's Role in Uncontrolled Hazardous
Waste Site Management
  Dr. John P. Collins & Wayne K. Tusa
A Dynamic National Response Mechanism
  Captain C.R. Corbett
        IMPACT AND IMPLEMENTATION
                  OF SUPERFUND

The EPA Research Program in Support of
Superfund  	;	
   William Rosenkranz & Michael Heeb, Ph.D.
Private Sector Attitudes and Concerns
with Superfund  	
  Frank B. Ryan, John C. Gorton, Jr. &
  Martin S. Tanzer, Ph.D.
Ranking System for Releases of
Hazardous Substances 	
  Steve Caldwell, Kris W. Barret & S. Steven Chang
.10
.14
                 STATE PROGRAMS
State-Federal Superfund Coordination:
Who Will Run the Program?   	
  Jacqueline M. Rams, Barbara Simcoe &
  David Duncan
Records Management Systems as Utilized by
the Texas Department of Water Resources and the
United States Environmental Protection Agency   ...
  William H. Hupp & Jo Johnson-Bollard
The California Superfund 	
  Peter H. Weiner
The Pennsylvania Approach to Uncontrolled Sites  ..
  MichaelD. LaGrega, Ph.D., Richard H. Demmy&
  John W. Thorsen

               SITE INVESTIGATION
                   & ASSESSMENT

Hydrogeological Investigation of an
Uncontrolled Hazardous Waste Site  	
  Christopher Hugger & Paul F. Clay
Location and Prioritizing of Abandoned Dump
Sites for Future Investigations 	
  Ann B. Nelson & Richard A. Young, Ph.D.
Selection of Monitoring Well Locations in East
and North Woburn, Massachusetts  	
  David K. Cook
Characterization, Impact Assessment, and
Closure Requirements for a Copper Residue
Disposal Site 	
  JeffreyL. Pintenich, JohnM. Mines, Michael R.
  Corn & Frederick G. Ziegler, Ph.D.
.21




.30

.37

.42
.45
.52
.63
.70
        New Assessment Methods to Aid Site Restoration   	79
          G. W. Dawson & S.M. Brown
        Minimizing Cost and Risk in Hazardous Waste
        Site Investigations Using Geophysics 	84
          Richard Benson, Robert Glaccum & Paul Beam
        Installing Groundwater Monitoring Wells at a
        Hazardous Waste Site   	89
          Richard Spear, Ph.D. & Peter Franconeri
      SCREENING OF UNKNOWN WASTES
Management of Analytical Laboratory
Support at Uncontrolled Hazardous Waste Sites   	96
  Michael Gruenfeld, Uwe Frank, David P.
  Remeta & Richard Losche
Use of a Regulated Access Laboratory for
Screening and Preparation of Hazardous
Waste Site Samples 	103
  Barry E. North, Ph.D., Kathleen H. Driscoll
  & Laurence W. Strattan, Ph.D.
The Use of Laboratory Screening Procedures in
the Chemical Evaluation of Uncontrolled Hazardous
Waste Sites  	'.	107
  Robert K. Wyeth
Compatibility Field Testing Procedures for
Unidentified Hazardous Wastes  	110
  Rodney D. Turpin, Joseph P. LaFornara, Harry
  L. Allen & Uwe Frank
Novel Approaches to Screening of Aqueous Effluents
From Hazardous Waste Sites  	114
  IhorLysyj

                AIR MONITORING
Ambient Monitoring for Specific Volatile
Organics Using a Sensitive Portable PID GC    	122
  T.M. Spittler & Alan W. Oi
Measurement of Fugitive Hydrocarbon Emissions
from a Chemical Waste Disposal Site   	123
  James A. Peters, Keith M. Tackett & Edward C. Eimutis
A Study of the Emission Rate of Volatile
Compounds from Lagoons  	129
  Andrew T. McCord
Air Monitoring of a Hazardous Waste Site 	136
  David A. Sullivan & Jerome B. Strauss

                     SAMPLING
Influence of Sampling Techniques on Organic
Water Quality Analyses 	
  Arthur M. Sean or & Larry K. Brannaka
Special Sampling Techniques Used for
Investigating Uncontrolled Hazardous Waste
Sites in California  	
  Howard K. Hatayama
The Stratified Sample Thief—A Device for
Sampling Unknown Fluids  	,
  Michael G. Johnson
                                                           .143
                                                           .149
                                                           .154

-------
                 REMOTE SENSING
 The Complementary Nature of Geophysical
 Techniques for Mapping Chemical Waste Disposal
 Sites: Impulse Radar and Resistivity  	
   Keith A. Horton, RexfordM. Morey, Louis
   Isaacson & RichardH. Beers. Ph.D.
 Utilization and Assessment of a Pulsed RF
 System to Monitor Subsurface Liquids  	
   Robert M. Koerner, Ph.D., Arthur E. Lord,  Jr.,
   Ph.D. & John J. Bonders
 Integration of Remote Sensing Techniques with
 Direct Environmental Sampling for Investigating
 Abandoned Hazardous Waste Sites  	
   Robert W. Pease, Jr. & Stephen C. James
 Survey and Analysis of Present/Potential
 Environmental Impact Sites in Woburn,
 Massachusetts  	
   Susan E. Titus
 Plants as Bioindicators of Environmental Pollution
   C.A". Sharma, Ph.D. & Christy Cooper
               REMEDIAL RESPONSE

 Rationale for Determining Priorities and Extent of
 Cleanup of Uncontrolled Hazardous Waste Sites  ...
   Walter Unterberg, Ph.D., Wayne L. Stone &
   AnthonyN. Tafuri
 Hazardous Substance Response Management Model
   J. Bill Hanson, Richard Stanford, Robert W.
   Pease, Jr. &PaulJ. Stoller
 Surface Sealing to Minimize Leachate Generation
 at Uncontrolled Hazardous Waste Sites  	
   Donald E. Sanning
 Applying Techniques for Solidification and
 Transportation of Radioactive Wastes to
 Hazardous Wastes  	
   J. W. Phillips
 Renovation of a Wood Treating Facility  	
   W. Lawrence Ramsey, Richard R. Steimle & James
   T. Chaconas
 The Fort Miller Site: Remedial Program for
 Securement of an Inactive Disposal Site
 Containing PCB's   	
   Warren V. Blasland, Jr., William H. Bouck,
  Edward R. Lynch & Robert K. Goldman
 Organic Leachate Effects on the Permeability
 of Clay Liners  	
D.C. Anderson, K. W. Brown,  Ph.D. & J. Green
 Multiattribute Decision Making for Remedial
 Action at Hazardous Waste Sites  	
   Terry H.Ess& Chia S. Shih, Ph.D.
 Risk Assessment Near Uncontrolled Hazardous
 Waste Sites: Role of Monitoring Data  	
  Glenn E. Schweitzer
 Costs of Remedial Actions at Uncontrolled Sites 	
  Howard L. Rishel, Sheila M. Kennedy, James
 ' J. Walsh, P.E., Dennis P  GillespieA
  Oscar W. Atbrecht
 Forced Cleanup: A Police Action or a
 Money Judgment  	
  James C. Scott & Robert B. Pearce
         Technical and Financial Aspects of Closure
         and Post Closure Care  	
           John W. Thorsen
                                                     .259
.158



.165




.171



.177

.185
.188


.198



.201



.206

.212




.215



.223


.230


.238

.248




.255
         SITE AND PERSONNEL SAFETY
Practical Considerations for the Protection of
Personnel During the Gathering, Transportation,
Storage and Analysis of Samples from
Hazardous Waste Sites  	263
  Gordon A. Allcott, Robert Vandervort&
  John V. Messick
Safety Procedures for Hazardous Materials Cleanup  	269
  Robert W. Melvold, Steven C. Gibson &
  Michael D. Royer
Initial Site Personnel Protection Levels Based on
Total Vapor Readings  	277
  Rodney D. Turpin
Hazardous Substance Site Ambient Air
Characterization to Evaluate Entry Team Safety   	280
  Martin S. Mathamel

          TREATMENT AND DISPOSAL

Shakedown and Performance Testing of the
EPA-ORD Mobile Incineration System   	285
  John E. Brugger, Ph.D., James J. Yezzi, Jr. &
  Frank J. Freestone
History and Bench Scale Studies for the  Treatment
of Contaminated Groundwater at the Ott/Story
Chemical Site, Muskegon, Michigan  	288
  Stephen C. James, Alan J. Shuckrow, Ph.D.
  & Andrew P. Pajak
Design of a Treatment System for Hazardous Run-Off  	294
Peter B. Lederman, Ph.D., John W. Hammond
  & Joseph P. LaFornara, Ph.D.
      GEOHYDROLOGY MATH MODELING

Computer-Enhanced Geophysical Survey
Techniques for Exploration of Hazardous
Waste Sites  	300
  G.L. McKown, Ph.D. & G.A. Sandness, Ph.D.
The Use of Mathematical Models to Assess and
Design Remedial Action for Chemical Waste Sites  	306
  C.R. Cole & G.L. McKown, Ph.D.
Parametric Analysis of Geological. Hazardous
Waste Disposal  	313
  EduardoA. Figueroa & Frank L. Parker, Ph.D.
Locating of Groundwater Pollution Sources
from Limited Field Data  	317
  Jack C.  Hwang, Ph.D. & Robert M. Koerner, Ph.D.
                  CASE HISTORIES
Dioxin Investigations in Southwest Missouri	
  Daniel J. Harris
The Denney Farm Site Remedial Project: A Model
for the Safe Excavation, Storage and
Elimination of Dioxin  	
  Ray Forrester
.322
.326

-------
A Hazardous Waste Disposal Problem Vs. A
Systematic Approach for Imposing Order onto Chaos  	329
  Thomas O. Dahl
Implications of the Chemical Control Corp. Incident   	341
  AdamM. Finkel & Richard S. Golob
A Fast Track Approach to Management and
Impact Assessment  	348
  Gregory A. Vanderlaan
Hazardous Waste Site Investigation Sylvester
Site, Nashua, New Hampshire   	359
  John J. Gushue, John E. Ayres &
  Alvin J. Snyder
Case Histories in Handling Unknown Hazardous
Materials at Dump Site Locations  	371
  Thomas F. Dalton
Contamination Control at Rocky Mountain
Arsenal, Denver, Colorado  	374
  John Wardell, Ph.D., MargotNielson &
  Judith  Wong
The Effective Use of Resource Recovery in the
Cleanup of Uncontrolled Hazardous Waste Sites—
Based on the California Experience 	380
  William Quan
               LIABILITY, LEGAL &
                  PUBLIC ISSUES

The Generator's Liability for Past Hazardous
Waste Disposal Practices 	
  Joseph M. Manko & Marc E. Gold
Recovering Damages to Natural Resources
Under CERCLA 	
  Edward Yang, Ph.D., Amy Home & Oscar Albrecht
Site Contamination and Liability Audits in the
Era of Superfund  	
  John J. Housman, Jr., David I. Brandwein &
  Dennis F. Unites
Institutional Learning in a Bureaucracy: The
Superfund Community Relations Program   	
  Steven A. Cohen, Thomas G. Ingersoll&
  James R. Janis
The Community Hazardous Waste Coordinator
Program 	.
  Betsy Goggin & Ann Rappaport
Citizen/Government Interaction at Toxic Waste
Sites: Lessons from Love Canal  	,
  L. Gardner Shaw & Lester W. Milbrath
.387
.393
.398
.405
.411
.415

-------
                CMA POSITION ON DEGREE OF HAZARD
                                    GERALDINE V. COX, Ph.D.
                                 Chemical Manufacturers Association
                                          Washington, D.C.
  It  is timely to clarify CMA's position  on so-called
"degree of hazard" in the context of EPA's hazardous
waste management regulations under RCRA. CMA sup-
ports continued recognition of degree of hazard  consid-
erations in these regulations. At the same time it is neces-
sary  to point out that this term has taken on somewhat
different  meaning as  EPA's RCRA  regulations  have
evolved.
  Regulations proposed by EPA in December 1978 con-
tained a complex and broad-reaching scheme for identify-
ing hazardous wastes. It was so sweeping it covered innoc-
uous  materials such  as wood ash and soft drinks. All
wastes identified by this scheme  were to be managed in
facilities subject to stringent, inflexible design standards.
  In  response to these sweeping regulatory  proposals,
CMA recommended  that  the Agency substitute  a "de-
gree  of hazard" waste classification  scheme and couple
it with flexible facility standards appropriate to the vary-
ing nature and degree of hazard posed by different wastes.
It is to this waste classification scheme that the term "de-
gree of hazard" originally referred.
  In its May  1980 interim final regulations, EPA substan-
tially altered  its  scheme for identification of hazardous
wastes so  as to embrace a significantly more limited uni-
verse of materials. In addition it relinquished an approach
based almost exclusively upon "characteristic" protocols
in  favor of one emphasizing identification and listing of
particular waste streams. This change of approach by EPA
had reduced  CMA's concerns about degree of hazard in
identification of  materials  as hazardous waste. Real con-
cern  remains, however, because  of the Agency's failure
to  identify concentration thresholds  with regard to toxic
constituents deemed to make particular wastes hazardous.
In the absence of such  thresholds it is difficult or impos-
sible  to obtain delisting of  particular wastes or waste
residues at particular sites, even though they may, because
of very  low concentrations of toxic constituents, present
a negligible degree of hazard.
  Degree of hazard concerns remain alive in a second,
important respect relating not to classification of wastes
by relative intrinsic hazardousness, but rather to appro-
priate management for differing wastes. EPA shares these
concerns and is examining a scheme for categorization of
wastes in light of appropriate management options for
these wastes. CMA supports this effort, so long as it does
not lead to inflexible results such as outright prohibitions
on land  disposal of particular toxic constituents. Examples
of steps that EPA is actively considering to tailor its facil-
ity  standards  to  various waste categories include  the
following:
•Not subjecting wastes  deemed hazardous solely because
 of ignitability, corrosivity or reactivity to the full array
 of incineration regulations.
•Giving "permits by rule" to impoundments used solely to
 neutralize solely reactive wastes.
•Giving "permits by rule" to impoundments used solely
 to neutralize corrosive wastes.
•Extending permits by rule to tank-type facilities used to
 neutralize solely reactive wastes.
•Amending and clarifying its  "mixture  rule"  to avoid
 unnecessary regulation of wastewater streams involving
 de minimus incidental quantities of hazardous materials.
  CMA supports these and other efforts EPA may under-
take to  frame  facility  requirements  appropriate to the
nature and class of wastes involved and to clarify the reach
of its  RCRA regulations.  Further, it strongly urges EPA
to establish thresholds or de minimis levels below which
residues or mixtures  involving listed hazardous wastes,
no longer need to be managed as hazardous wastes.

-------
   THE ASCE ENVIRONMENTAL ENGINEERING DIVISION'S
          ROLE IN UNCONTROLLED HAZARDOUS WASTE
                                  SITE MANAGEMENT*
                                        DR. JOHN P. COLLINS**
                                          WAYNE K. TUSA
 INTRODUCTION

  The phrases Love Canal, Valley of the Drums, Kin-Buc,
 Burnt Fly Bog, etc. have practically become household
 words. With the grace of hindsight, the reasons for that
 occurrence are abundantly clear.  For example, each year
 300 to 500 new chemicals are produced commercially in
 addition  to the 70,000 chemicals currently produced at
 over 115,000 manufacturing plants throughout the United
 States. It has been estimated that over 60,000,000 tons of
 hazardous wastes  were produced in  1980 alone.  While
 these chemicals have contributed to the high standard of
 living enjoyed by this country, the after effects of their dis-
 posal are only now beginning to have  a perceptable and
 negative impact on the quality of life.
  The Chemical Manufacturer's Association has estimated
 that over the next 10 years the chemical industry will in-
 vest $5.4 billion under Superfund legislation and another
 $3.5 billion for compliance under RCRA. The Environ-
 mental Protection Agency has indicated that up to 30,000
 disposal sites may contain hazardous wastes which  could
 require remedial expenditures. The Superfund program in-
 cludes investigation of up to 15,000 sites, of which approx-
 imately 1,500 sites will require detailed inspection and of
 these approximately 400 will be selected for priority clean-
 up.
  The magnitude of the abandoned site problem offers a
 retrospective look at how engineers and society as a whole
 have made decisions and technical judgements important
 to the continued welfare of that society. In addition, the
 abandoned site problem offers a number of opportunities
 for ASCE to serve as a spokesman for the engineering pro-
 fession, as well as for individual practicing engineers and
companies to contribute to the successful resolution of that
problem.
 •This paper has been prepared by the Hazardous Waste Management
 Committee and  as such represents the views of that  committee
 acting as a spokesman for the Environmental Engineering Division
 of Ihc American Society of Civil Engineers.
••Dr. Collins is the upcoming Vice Chairman of the Solid Waste Man-
 agement Committee and is a past chairman of the  Hazardous Waste
 Committee. Mr. Tusa is currently Chairman of the  Hazardous Waste
 Management Committee.
ASCE ROLE

  ASCE, largely through its Environmental Engineering
Division and its Geotechnical Engineering  Division, has
played  an active  role in hazardous waste  management.
The Hazardous Waste Management Committee, for ex-
ample,  was established as a standing committee of the Sol-
id Waste Management Committee in 1979. Since then, the
Committee has completed the following hazardous waste
efforts:
  1. Developed a policy statement on Superfund for pres-
entation before the House Subcommittee on Commerce
and Transportation
  2. Sponsored approximately ten sessions  on hazardous
waste  management at  four conferences throughout the
United States
  3. Published "The Hazardous Waste Dilemma: Issues
and Solutions," a compendium of papers from the July
1980 ASCE Environmental Engineering Division Confer-
ence
  4. Provided peer review on several draft technical re-
source  documents  for the USEPA Office of Solid Waste
relating to RCRA Subtitle C regulations
  5. Sponsored several local seminars/presentations  on
hazardous waste management
  6. Completed a policy paper on hazardous  facility siting
  Society member companies and  individuals have also
successfully contributed by actively becoming involved in
the development of federal and  state regulations, the de-
velopment and implementation  of site investigation pro-
cedures, and the engineering of  proposed remedial meas-
ures. To a limited extent to date,  Society engineers have
also assisted in remedial construction  and construction
management activities.

PROBLEM IDENTIFICATION

  The  experience  engendered by  the  Hazardous Waste
Management Committee, as well as by practicing society
professionals, has  resulted in the identification of a num-
ber of  problems which must be confronted by  the engi-
neering professional in  applying applicable skills to the
abandoned sites problem.
  Perhaps the most evident  problem relates  to  the fact
that abandoned site management is still an  art and not a

-------
                                                                                               OVERVIEW
science. Developing a set of construction plans and specif-
ications for an abandoned site  remedial  project, while
relying in large part on established skills and procedures,
is much different from  designing a conventional retain-
ing structure or a twelve story office building. The engi-
neer must make technical decisions based upon varying
amounts of data from a number of engineering and scien-
tific disciplines. Final decisions  are  in part judgements
that attempt  to weigh  the  risks  of potential environ-
mental  and public health  impacts against the costs  and
relative effectiveness of alternative  engineering approach-
es. To a large degree, this dilemma is compounded by in-
adequate data in those scientific disciplines which come in-
to play in making those technical decisions. Furthermore,
the potential for leachate migration through a  confining
clay layer, the potential for  health impacts due to long-
term low-level hazardous air emissions and the potential
long-term chemical interaction of waste materials and the
proposed  materials of construction are all issues impacting
the selection of a final engineering alternative.
  In part, the problem of  inadequate data  and prac-
tical experience has been complicated by rapidly develop-
ing and often changing regulatory guidance at the federal
and state levels. This challenge  must also often  be  met
within the constraints of  innumerable regulatory guide-
lines which are often more procedurally than technically
complete. For example,  one of the areas most lacking re-
lated to the necessity of criteria or guidance to determine
acceptable risk levels. For many waste constituents there
are no standards or criteria for acceptable concentrations
in groundwater, surface water or ambient air. As a con-
sequence, these issues are often reconsidered at each aban-
doned site and overall goals or acceptable risks levels may
vary not only from state to state but also from site to site.
  The complexity of technical issues related to abandoned
sites often challenges the "interdisciplinary" approach be-
tween the engineering and scientific  disciplines. While a
large number of training mechanisms have been developed
in the last two to three years (i.e., seminars and short train-
ing  courses),  most  experienced  professionals  appear to
have received the bulk of their training via the well estab-
lished trial and error method. Unfortunately, at the univer-
sity level,  very  few  programs are in  place to specifically
train young professionals in the waste management area.
Information Transfer

   Another  problem is the lack of information transfer
throughout the engineering and scientific professions with
regard to abandoned site management. While this confer-
ence certainly is helpful, not all readers will be familiar
with the turkey baster method of air velocity measurement,
the construction details of a shallow trench-well for moni-
toring relatively impermeable surface soils and the opera-
tion of a ground penetrating radar device? While in part
this lack of knowledge is understandable due to the rap-
idly changing state-of-the-art, for the same reason it is par-
ticularly unacceptable due to the pressing need for new and
better techniques.
  One of the most pressing information transfer needs re-
lates to adequate safety practices for engineers complet-
ing inspections as well as engineering  and construction
teams  actively implementing  proposed  remedial plans.
While  teams currently active under EPA contracts have
addressed these issues in depth, in the near future a much
larger number of engineering firms and employees will be
involved in site-specific efforts. A real need exists for the
dissemination of this type of information both to a wider
range of recipients and to what is rssentially a newly devel-
oping cadre of hazardous waste personnel, i.e., those who
will be directly involved in digging the wells, moving the
earth and treating the waste residuals.

Liability and Insurance

  Another issue of significance is liability and insurance
requirements for  the engineering firm involved in hazar-
dous waste management. As of this date, not all the lia-
bility issues  have been resolved for those firms which will
be involved in implementing EPA's Superfund program.
  Who will be liable if an interim solution  installed  at
one of the 100 EPA interim priority sites does not elim-
inate  all forthcoming environmental and public health
impacts? Which parties will be required to participate  in
any eventual  court settlement: the engineer, the construc-
tion contractor, the land owner, the waste generator?
  Perhaps equally important is the  issue of insurance.
While  a number  of the  larger insurance companies are
offering or  are currently developing programs to offer
non-sudden RCRA insurance, there appears to be  in some
cases  little  interest  in offering  engineering  companies
"errors and omissions insurance" specifically relating  to
hazardous waste work.
  An interesting sidelight of the liability issue and an addi-
tional  problem area relates to the development of inno-
vative  techniques  for abandoned site management.  Un-
fortunately, while there may be a great need  for better
and less costly approaches to abandoned site management,
the development and utilization of new techniques, partic-
ularly  in the private  sector, is thwarted  by  the  liability
issues involved if those techniques cannot be guaranteed
to be 100 percent effective.
CONCLUSIONS

  What needs to be done to resolve some of the problems
identified above? Clearly more waste management train-
ing mechanisms are needed at both the university and pro-
fessional levels. It would be particularly useful to work un-
der a consistent,  equivalent and definitive set of regula-
tions  nationwide.  While obviously that  is not currently
possible, it suggests that  as  much emphasis as possible
must  be  placed on the development of those regulations
in the shortest time span possible.
  In addition, engineers believe that substantial additional
efforts should be particularly directed toward information
transfer activities. The liability and insurance issues also
deserve much closer scrutiny. After all,  should an engi-

-------
     OVERVIEW
neering firm accept the risks associated with a hazardous
waste contract, if it cannot enjoy the privilege of the ade-
quate financial protection it enjoys in other business areas
practiced by the firm?
  In keeping with its policy to provide, where appropriate,
technical guidance to society  as a whole, ASCE, and
specifically the Hazardous Waste Management Commit-
tee, would like to offer its services in resolving  some of
those issues identified above. The committee, for example:
  1. Intends to continue to provide peer review services on
proposed technical resource documents
  2. Is considering sponsoring an ASCE Environmental
Engineering Division Conference  solely on  hazardous
waste management
  3. Is expanding its role at the local chapter level relating
to the development of seminars/presentations on haz-
ardous waste management
  4. Is continuing to monitor  advances in  siting pro-
cedures
  5. Is developing  a  recommended curriculum in waste
management for university level students
  6. Is establishing a  sub-committee to investigate liabil-
ity and insurance issues and to develop appropriate recom-
mendations

  The  Committee is  also  open  to suggestions  from the
scientific community at large and would welcome specific
recommendations.

-------
         A DYNAMIC NATIONAL RESPONSE MECHANISM
                                       CAPTAIN C.R. CORBETT
                                Marine Environmental Response Division
                                        United States Coast Guard
                                            Washington, D.C.
INTRODUCTION

  In preparation for writing this paper, I reflected on the
Management of Uncontrolled Hazardous Waste Sites Con-
ference held here in Washington during October of 1980.
At that time I expressed the hope that Superfund would
soon be operative, and that late 1980 would see the devel-
opment  of an intelligent revision to the National Con-
tingency  Plan which would adequately address  uncon-
trolled waste  sites. Superfund did  not quite make it by
October of that year, but it was not far behind.
  Since  the enactment of the Comprehensive Environ-
mental Response,  Compensation,  and Liability Act  of
1980, or Superfund, the U.S. Coast  Guard and U.S.
Environmental Protection Agency have encountered and
overcome a number of difficult problems which have
proven that the ties between them are strong and endur-
ing.

PROBLEMS SOLVED

311 Fund Transferral

  The first obstacle which had to be overcome was the
transfer of one-half of the Clean Water Act 311(k) fund
balance to the Response Trust Fund established by Super-
fund; initially, this did not appear to present a major prob-
lem. However, when it was learned that the Department
of Treasury believed that those transferred monies could
not be spent  until an appropriation to Superfund was
made, the Coast Guard and the EPA became alarmed that
the Clean Water Act Section 31 l(k) revolving fund balance
of some seven million dollars would become depleted be-
fore its request for supplemental appropriation, expected
in July 1981, was authorized.
  In our view, the Congress  did not intend that those
monies transferred lie dormant. At the very least, they
were intended to remain available for hazardous waste site
response  which met 311(k) criteria, that is,  water-related
and discharging  or threatening to discharge one of the
hazardous polluting substances named pursuant to  the
Clean Water Act. With the help of the National Response
Team and  Congressman Florio of New Jersey, and  a
dangerously low  level of the 311(k) fund, Treasury was
convinced to release the money and we were able to use it
for the intended purpose.
Contracting Procedure

  But the problems were not over. EPA had not yet de-
vised  a contracting mechanism  capable  of addressing
Superfund type problems. Even with the  funds released
by Treasury, there were still only limited monies avail-
able to continue  oil spill, chemical  spill and hazardous
waste site responses.
  What followed was a complex series of events but it
was the "handshake" agreements, Coast Guard contrac-
ting help and a general  spirit of cooperation which per-
mitted the combined funds to continue to address oil and
chemical spills as well as limited numbers of hazardous
waste  sites. This  was a difficult period which required
EPA to take a tough stand on what sites would or would
not be addressed  and the Coast Guard to continue in a
business it really never wanted.
  Eventually, with the help of Congressman Florio and
the Office of Management and Budget, a supplemental
appropriation was authorized nearly two months earlier
than expected.  No responses to acute discharges failed to
occur.
  All  of this activity took place during a period when we
were working  concurrently on developing an  Executive
Order to implement Superfund and the required revision
to the National Contingency Plan.  The same  spirit of
cooperation is continuing as we make the necessary hand-
off s of waste site response cases from the  311(k) fund to
Superfund and as  we work to establish Coast Guard access
to the  Superfund for emergency response.
SUPERFUND IMPLEMENTATION

  The primary interests of the Coast Guard, in Super-
fund implementation, are the revision to the National Con-
tingency  Plan, the establishment of necessary response
authority and the assurance of adequate resources to re-
spond to a broader spectrum of acute releases than pre-
viously afforded under Section 311 of the Clean Water
Act. As you know, the Coast Guard has a long history of
responding, in coastal areas,  to  oil discharges  and re-
leases of the Section 311 designated hazardous polluting
substances into, or threatening, the wasters of the United
States.

-------
     OVERVIEW
  The Coast Guard has executed that responsibility as a
member of the National Response Team established in the
highly successful and acclaimed National  Oil  and  Haz-
ardous Substance  Pollution Contingency  Plan.  In our
view, the  strength  of the National  Response Team has
been based on the equal involvement and standing of the
twelve  Federal  member agencies. This  cooperation has
enabled open and candid discussion of what it has  done
well, what it could have  done better and  to  make ad-
justments when necessary.

NATIONAL CONTINGENCY PLAN REVISION

  The  Coast  Guard is now engaged, as are all of the
members of the National Response Team, in assisting the
Environmental  Protection  Agency  in  redeveloping the
Plan to deal with a broader spectrum of releases into the
environment—not necessarily associated with water or its
adjoining  shorelines. This is a difficult task and one on
which all of our membership has been working very hard
to accomplish.  A few of the features I believe must be
included in the new Plan are:
(1) To the maximum extent possible, the new Plan  must
   draw on the present Plan for its  organizational struc-
   ture. The  terms, acronyms,  organizational  relation-
   ships, etc., do not need re-invention. Likewise, Fed-
   eral, state, or local response authorities do not need the
   confusion of a Plan distinctly different from the one we
   are now using;
(2) The concept of using the coastal and inland areas for
   assigning Federal response officials must be preserved;
   the Coast Guard should respond in the coastal areas,
   and the EPA in the inland areas except  that for waste
   sites it is appropriate for EPA to have the authority in
   both the coastal and inland areas
(3) Response to oil spills must occupy a distinct place in
   the Plan so that those using the Plan in Clean Water
   Act related matters will not be encumbered by terms
   having meaning only when associated with Superfund
(4) The Plan must not become overly complex, for it will
   affect a broader spectrum of interests than just Fed-
   eral or other response oriented agencies
(5) State and  local governments should continue (in fact
   expand their role) as our partners in pollution response
   activities,  rather than  as  bystanders  confused by an
   unnecessary array of Federal officials
(6) The field On-Scene Coordinators must  continue to
   have sufficient latitude and authority to  aggressively
   and adequately protect the public health and welfare.
   We in the Coast Guard also believe that the Plan should
receive intense public review prior to finalization and that
such a review will help us to ultimately publish the very
best Plan possible. The Coast Guard will continue to assist
the Environmental Protection Agency and other members
of our National Response Team in Developing a rational
and effective Plan which  the public can understand. By
the way, we in the Coast Guard are extremely comfortable
with the EPA as the steward of the National Plan.

CONCLUSION

   It is extremely important  that the National Response
Mechanism, which is presently in  place, be preserved. I
have repeatedly said that the National Plan is one of the
most enlightened documents the government has devel-
oped. Its success is a tribute to  all of the Agency mem-
bers who have worked together to provide the public the
excellent service it deserves. All of us must resist any at-
tempt to diminish the importance of this Plan, or to mod-
ify it for purposes other than real service to the public.

-------
                      THE EPA RESEARCH PROGRAM IN
                             SUPPORT OF SUPERFUND

                                      WILLIAM ROSENKRANZ
                                       MICHAEL HEEB, Ph.D.
                         Office of Environmental Engineering and Technology
                                U.S. Environmental Protection Agency
                                           Washington, D.C.
INTRODUCTION

  Hazardous waste is one of the nation's most signifi-
cant environmental problems.  Chemicals once  thought
benign have been shown to  cause cancer, birth defects
and  a  variety of other  harmful human  health effects.
Toxic wastes lie in abandoned and often long-buried sites,
some of which are slowly poisoning underground water
supplies. Practices which have  been and continue to be
widespread include uncontrolled incineration, poorly de-
signed  and sited landfills, industrial waste lagoons  with
leaking bottoms and  weak dikes and the  indiscriminate,
often illicit dumping of dangerous waste materials. These
past, environmentally unsound disposal practices are a
threat to soil, air and water and hence, to human popu-
lations and ecosystems.
  Hazardous materials are also released to  the environ-
ment as a result of accidental releases (spills) of oil and
hazardous  substances.  Such  incidents  originate  from
transportation accidents as well as discharges  from storage
tanks,  pipelines,  holding lagoons and industrial plants
used in the manufacturing and processing of chemicals.
  In December  1980, Congress provided authority for
Federal response to the release of hazardous materials to
the environment from uncontrolled waste  sites as well as
spills and in all media when it passed PL96-510, the Com-
prehensive Environmental Response, Compensation and
Liability Act (CERCLA). The Environmental Protection
Agency (EPA) has been  delegated a lead role in the im-
plementation of this Act. EPA established the Office of
Emergency and Remedial Response (OERR) to carry out
its responsibilities. The Office of Research and Develop-
ment (ORD) in turn established a Hazardous Emergency
Response Research Committee, co-chaired by  ORD and
OERR, to plan the ORD technology programs in support
of response activities implementing CERCLA.


PROGRAM OBJECTIVES

  The  program offices are the primary "users" of the
ORD program  outputs and as such,  they establish the
focus and the  requirements  for technical support and
field evaluations. The EPA CERCLA program is aimed
at providing  effective emergency  cleanup  response in
spill  situations and effective remedy of the  problems at
uncontrolled hazardous waste sites. The  application of
technologies for performing these tasks is relatively new
and their long-term reliability and effectiveness have not
been fully evaluated. If the legislative mandate for rapid
and effective response is to be met, EPA needs to assess
the technologies and methods proposed for those actions.
In addition, the risks  to human health and the environ-
ment from spills of hazardous materials and from uncon-
trolled sites need to be evaluated as accurately as possible.
The ORD program focuses  on these  major  short-term
needs of the Agency.

THE OFFICE OF RESEARCH AND
DEVELOPMENT PROGRAM

   The overall  goal of the ORD support program is to
provide a core  of scientific and technical information and
expertise in response to the implementation require-
ments of CERCLA and the enforcement actions designed
to perform necessary  cleanup and containment actions
under this Act. In general, the program addresses the im-
mediate problems of providing the data required to sup-
port a hazard  assessment program, expert witnesses for
enforcement cost  recovery  actions,  initial assessment,
analysis and evaluation of alternatives as well as con-
tainment and  removal and  ultimate  disposal  and site
restoration. The development of  techniques for evalua-
tion of potential impacts on the  public and the environ-
ment is also addressed.
   Specifically,  this program concentrates  on the evalua-
tion of equipment and techniques  for discovering,  assess-
ing,  preventing, controlling, removing  and  ultimately
disposing of hazardous substances released to the envir-
onment from spills or from uncontrolled waste sites. It
also provides methods for assessment of the  ecological
and  human  population impacts from  releases and the
equipment and techniques suitable for restoring the en-
vironment to a  healthy state.
   The program covers two  major areas, (1) emergency
response (removal  activities)  and  (2)  remedial action re-
sponse. Emergency response generally applies to spills of
hazardous materials but also  applies to waste sites which
are discovered  to be posing an immediate threat to human
health and/or  the environment. Remedial response de-
notes a  planned  effort  for  permanent remedy  of the
problem site and, therefore, more time is afforded before
this type of activity is initiated.

-------
8    IMPACT
   The objectives of the emergency response (removal)
 program are to evaluate and demonstrate new or improved
 equipment,  devices,  systems  and  data  analysis tech-
 niques for the  prevention,  identification, containment,
 control and removal of hazardous substances released to
 the environment. This  includes the cleanup and recovery
 of hazardous  substances from spills as well as from un-
 controlled hazardous waste sites.
   The outputs  must meet the regulation development,
 enforcement and technical support needs of the Office of
 Emergency and Remedial Response (OERR), the Office
 of Waste Programs Enforcement and the Regional Of-
 fices.  Regulation  development  must  be implemented
 through  technical background development, such  as, up-
 dating the list  of  designated  hazardous  materials  and
 the  reportable quantity regulations. Regulation  support
 is provided by evaluating new techniques that will be in-
 corporated in  the regulations by reference.
   In this area, ORD provides user manuals for response
 equipment,  sampling  and  analysis, technology  evalua-
 tion, monitoring techniques, and guidance for on-scene
 response personnel. Enforcement support  is provided by
 ensuring the use of proper chain-of-custody techniques for
 sample collection,  analysis and reporting, providing ex-
 pert  witnesses  for  enforcement  cost  recovery  actions
 and  providing  technical support for  hazardous waste
 enforcement investigations.
   The objectives of the remedial response research pro-
 gram are to assess and validate new or improved remedial
 action technologies or schemes to minimize pollutant re-
 lease from uncontrolled hazardous  waste  sites. This ac-
 tivity  includes  quality assurance programs and pilot,
 field, cost-effectiveness and model studies to validate con-
 trol  technologies as related  to  surface  water control,
 groundwater control,  plume management, chemical im-
 mobilization and excavation and reburial.
   Technical resource documents will be provided  for the
 user communities to describe  the best engineering judg-
 ments for the various technologies.  Enforcement support
 is provided to assist in devising the remedial measures and
 to  testify on  the  need for  and  evaluation of those
 remedies for negotiations and litigation.
   The program is performed at several ORD laboratories.
 Environmental engineering  solutions to  problems  are be-
 ing evaluated  by the Municipal Environmental Research
 Laboratory, Cincinnati and its branch in Edison,  New
 Jersey. Monitoring and quality assurance programs are
 performed at the Environmental Monitoring and Support
 Laboratories in Las Vegas, Nevada,  Warrenton, Virginia,
 Cincinnati,  Ohio and  Research  Triangle Park, North
 Carolina.
   Techniques  for environmental  measurement are being
 evaluated by  the Environmental Research Laboratories
 in Research Triangle Park,  North Carolina;  Ada, Okla-
 homa; Athens,  Georgia;  Corvallis,  Oregon;  Duluth,
 Minnesota; Gulf Breeze, Florida and Narragansett, Rhode
 Island. Finally, hazard assessment  support  is  provided
 by the Environmental  Criteria and Assessment  Office in
 Cincinnati, Ohio.
HAZARD ASSESSMENT

  CERCLA specified a Federal response to releases or
threatened releases  of hazardous materials to the envir-
onment from  spill  sites and waste sites.  Before a "re-
sponse"  can be executed  it is  necessary to characterize
the nature and magnitude of the health  and ecological
hazard. Therefore,  evaluation of adverse  health impacts
and  the  hazard potential  of substances  released from
spills and defective abandoned waste sites is a necessity if
response funds are to be used to their maximum benefit.
  ORD will provide information on the acute and chronic
toxicities of identified compounds in order to select the
best  safety measures for protection of cleanup crews as
well  as to determine acceptable residual levels (extent of
cleanup). Currently, ORD-maintained data on chemicals
for which  adequate acute,  subchronic  or chronic data
exist will be used  to prepare tabular summaries of ad-
verse health effects  for  previously investigated com-
pounds, health profiles and guidelines, and methodologies
for rapid (24 hours) hazard assessment of chemical mix-
tures. Toxicity indexes based on no-effect levels, type of
effect, and chemical class  are also being developed. The
methodologies  are  being  developed for incorporating
route-specific toxicology into a  site ranking system to be
used for identifying the  nation's highest priority sites
for remedial action.

TECHNOLOGY EVALUATION AND ASSESSMENT

  Engineering guidance and control  technology evalua-
tion  will be provided to Federal, State and local  agencies
for immediate response to hazardous emergency  removal
and  remedial action at  spill sites  and abandoned waste
sites. ORD will address  the  construction and  field evalu-
ation of equipment and  technologies  for onsite treat-
ment/destruction of hazardous wastes, leachates and con-
taminated groundwater. It will also evaluate and assess
the best  existing remedial  or abatement action and con-
trol  technologies to minimize pollutant releases from un-
controlled hazardous waste sites and maximize the appli-
cability and  transfer of  this  information  to  the user
community.
  CERCLA specified that response to releases  may  be
removal  or remedial. But  this raises important questions
such as "when is the emergency over, i.e., stabilized, so
that  remedial  action can  begin?" and, "when remedial
action begins, to what extent should the cleanup be exe-
cuted?". Work is currently underway to prepare guidance
manuals for use by  response personnel, such  as On-
Scene Coordinators, to help them  make these critical de-
cisions.

MONITORING AND QUALITY ASSURANCE

  The program will meet the monitoring requirements of
removal  and remedial actions  by  providing  the latest
techniques  in measurement, monitoring and  quality as-
surance.  The tools  and methods for field support and
sampling, laboratory  analysis,  and  remote  monitoring

-------
                                                                                                 IMPACT
will be applied to site discovery and characterization, as
well  as mitigation  and cleanup. The  program  includes
quality  assurance through  laboratory audit and inter-
comparison and chain-of-custory for enforcement actions.
  A comprehensive quality  assurance program has been
initiated to evaluate contractors involved in the  site pro-
gram.  The program includes evaluation and standardi-
zation of sampling and  analysis methods and chain-of-
custody. In an effort to meet the demand for an Agency
capability, a  contractor operated laboratory has been es-
tablished  to  analyze field samples and complex samples
taken by Regional field teams.  The program  includes the
standardization of  mobile emergency  labs  for  removal
operations.  A comprehensive  program  to  monitor  air
quality in and around  sites has been  started,  including
the  development of  a personal exposure  monitor for
volatile organics.
  An extensive nationwide  program has been instituted
to acquire aerial photographs  of sites, characterize the
problems and prioritize the need for cleanup.  The unique,
in-house capabilities of ORD scientists have been made
available  through  this program to accomplish special
monitoring studies. The capability provides the Agency
with special  studies of  air, surface and groundwater
and site characterization.
ENVIRONMENTAL MEASUREMENT

  Assessment and evaluative procedures for determining
the potential ecological effects  of hazardous substances
spills  and releases from abandoned waste sites will be
provided by ORD.  A primary goal will be  to  provide
user-oriented material on situation assessment along with
evaluating and developing ecologically sound isolation,
removal, restoration and disposal methods.
  In support of field  response personnel requirements,
a uniform set  of operational  procedures  for  environ-
mental characterization and assessment will be prepared.
This will provide  uniform  "extent of contamination"
assessments  and cleanup operations planning. A manual
on proper utilization and choice of dispersants for clean-
up, from an environmental point of view, is in prepara-
tion.  This is a critical need because a poor choice of
dispersant could compound and extend the damage.

-------
                         PRIVATE SECTOR ATTITUDES
                    AND CONCERNS WITH SUPERFUND
                                          FRANK B. RYAN
                                   URS/Coverdale & Colpitts, Inc.
                                         New York, New York

                                       JOHN C. GORTON, JR.
                                         URS Company, Inc.
                                          Buffalo, New York

                                     MARTIN S. TANZER, Ph.D.
                                         URS Company, Inc.
                                         New York, New York
BACKGROUND AND PURPOSE
OF RESEARCH

  As  Superfund was proceeding through the legislative
process  to final enactment in  December,  1980  the En-
vironmental Protection Agency commissioned a group of
studies to facilitate its  implementation. One such study,
the subject of this report, was planned to indicate the im-
plementation capabilities of the private sector, obtain a
qualitative appraisal of the likely level of private sector
response to the program, and  identify constraints and
"bottlenecks" with respect to its timely execution. Briefly
then,  the questions to  be answered  were: (1) "Are the
necessary private sector resources available?" (2) "Will
the private sector  respond effectively?" and (3) "What
possible problems  of private sector  firms must be  con-
sidered in program procedures?"
  As the Workplan for the study was being carried out,
it became apparent that before a study of the magnitude
and scope originally contemplated could be completed,
actual  implementation  of Superfund would have  been
started.  The initial results of  this implementation  pro-
cess could logically be  expected to answer the questions
originally posed in the  study endeavor. Under this set of
circumstances, it was considered appropriate to limit this
investigation of Private Sector Attitudes and Concerns
toward Superfund to the work  already accomplished or
immediately to be  accomplished and  prepare this Insight
Report for current use. It has been  the authors' exper-
ience that a timely report often is of  greater value than a
deferred submission of greater depth or scope.
ORGANIZATION AND WORKPLAN

  The original  workplan in its final version was  trans-
mitted to EPA in Washington on December 24, 1980. The
principal elements of this Workplan are briefly described
below.
Review of Capabilities Required by EPA
from Private Sector Firms

  In addition to talks with EPA in Washington, discus-
sions were conducted with EPA officials in Region II and
with staff of  the (New Jersey) Department of Environ-
mental Protection. These discussions provided necessary
information as to the type of services required of the pri-
vate sector by the Superfund program.

Pilot Study Area

  Originally interviews were planned in New York, New
Jersey and California. By the cut-off date for active field
work, interviews were conducted both in New York and
New Jersey with some additional emphasis on the latter
State.

Rosters of Firms Possessing the Required Capabilities

  The  bibliography provides a listing of the rosters and
directories assembled for use in structuring the study. In
addition to the formal sources* the study utilized personal
referrals as appropriate.

Preparation of Preliminary Draft of Interview Form

  Preliminary drafts of interview questions were prepared
and submitted to EPA-Washington for comments. As a re-
sult of these comments and the field experience described
below, a revised interview form was prepared.

Testing of Interview Guidelines and Approach

  The original guidelines were tested with a General Con-
tractor and subsequently modified to the format attached.
Further refinements were  planned, if necessary, as  the
larger scale interviewing was undertaken. A single inter-
view format was designed to be applicable to all the firms
surveyed.
                                                   10

-------
                                                                                                  IMPACT
                                                                                       11
Administration of Interviews

  Prior to the decision to limit the scope of this  study,
URS  offices  in  Philadelphia  and  San  Francisco were
alerted to participate in the interview effort and provided
with copies of the Workplan  and Interview Guidelines.
Contractor associations were interviewed to obtain sugges-
tions as to the firms to be selected for personal contact.
As already noted, the larger effort was discontinued.

Interviews

  The types of firms to be interviewed included, but were
not intended to be limited to:
•General Contractors
•Testing Laboratories
•A/E Firms (Architectural/Engineering)
•Hazardous Waste Management Firms
•Other Professional/Technical Organizations
  The following tabulation summarizes the number of in-
terviews obtained and includes interviews with both Fed-
eral and state (New  Jersey) officials. (Since URS Com-
pany  has  extensive architectural/engineering capabilities,
it was not considered  necessary to interview additional
firms as part of initial effort).
Type of
Firm or Agency

EPA-Region II
N.J. DEP
General Contractors
Testing Laboratories
Hazardous Waste Mgmt
Other Professional
Totals
Number of Persons Interviewed
New York    New Jersey    Total
3(B)
7
 3
 2
 3(A)
 1
 1

10
 5
 2
 5
 1
 1
 3
17
Notes: (A) Includes Associations
      (B) Includes Coast Guard.
  Although the number of interviews was not as great as
would eventually have been obtained,  most interviewees
proved completely responsive and  it is problemmatical
whether a more extensive effort  would alter the report
findings significantly.  The results  of these interviews and
the problem areas uncovered are described in the following
pages.

FINDINGS

  The interview process  generated  information  on atti-
tudes toward Superfund and permitted the  identification
of specific problem areas: (1) General Attitudes, (2) De-
mand for Professional Skills, (3) Some Equipment Limita-
tions and (4) Specific Problem Areas.

General Attitudes

  The impact of Superfund had not, as yet, been felt at
the implementation level. There appeared to be a  "wait
and see" attitude in the private sector; it is probably fair to
                                    say that the same attitude, to a less pronounced extent,
                                    also prevails in the public sector.
                                      As far as contractors are concerned  there was  a per-
                                    vasive attitude that they "can do  anything."  Certainly
                                    there are reservations in regard to some specific problems,
                                    as will be  discussed, but the general attitude was one of
                                    confidence both  in  innovative ability and in equipment
                                    and manpower capabilities and resources.
                                      With regard both to contractors and other professionals,
                                    there was a dichotomy between the public and private sec-
                                    tors.  Government officials apparently believed the private
                                    sector will respond, but without the requisite skills and
                                    commitment. The private sector, on the other hand, ap-
                                    pears to wonder how real  "Superfund" is and what its
                                    specific impact would be on existing operations.
                                      Perhaps these  general attitudes can be  summarized as
                                    follows—both the public and private sectors  were  await-
                                    ing further developments with the public sector somewhat
                                    dubious as to the availability and actual commitment of
                                    required skills and the private sector equally dubious as
                                    to the reality, timing and  scope  of the  program, and its
                                    possible benefits and impacts  with respect to individual
                                    firms of various types.
Demand for Professional Skills

  From discussions  with  representatives  of the private
sector  and  with  government officials  who will be in-
volved with Superfund, it  became apparent that two pro-
fessional categories—chemists and hydrogeologists—may
be limiting  factors during the  program's initial imple-
mentation. These shortages of trained, experienced pro-
fessionals are especially significant because litigation may
be required to establish financial responsibility for some of
the clean-up activities. In such instances,  the  expertise
and professional  qualifications of the witness will  come
under close  scrutiny, and it is the authors' understanding
that many tests and interpretive analyses cannot be del-
egated to technicians subject  only to professional review.
Basically, particularly in the  case of mass spectrographic
techniques, the chemist performing the analyses is respon-
sible for their credibility.
  In the case of hydrogeologists, there are  probably less
than 150 graduated each year and many of these will en-
ter private employment. This shortage can be alleviated to
some extent by approximately two years of training of an
individual with a background in geology or other  scien-
tific discipline. In this regard, because of the current ener-
gy shortage, petroleum companies  have continuing  re-
quirements for graduate geologists so there is competition
for these skills.  Because  of the possible litigations  in-
volved, high levels of demonstrable expertise will  be  re-
quired.
  The  identification of these two professional categories
as possible  problem areas is not intended to imply that
shortages in other skills do not exist. However, the  short-
age of both chemists  and  hydrogeologists are believed to
be the most serious and  least  amenable  to "quick fix"
solutions.

-------
12     IMPACT
Some Equipment Limitations

   Mass spectrographic equipment appeared to be another
problem area. These units with ancillary and related equip-
ment cost approximately $250,000 each and thus repre-
sent a significant investment for a private  laboratory with
no assurance of amortization through use. While it can
be safely assumed that this equipment can be manu-
factured in sufficient quantity to meet  demand, main-
tenance is said to  be a continuing problem, and trained,
experienced chemists, as noted in the previous section,
may be in short supply.
   In addition, some  construction contractors expressed
concern  over the  availability of suitable tank trucks  to
handle possible combinations of corrosive, toxic and flam-
mable liquid wastes.

Specific Problem Areas

   The study disclosed certain specific problem areas which
may hamper  the  initial implementation of Superfund.
These areas of concern, not necessarily arranged in order
of importance, are:
 •Conflict of interest     •Availability of disposal sites
 •Expert testimony      'Financial arrangements
 •Safety training        "Some other concerns
 •Insurance             "Bureaucracy
 •Permits and licenses

CONFLICT OF INTEREST

   Many laboratories and firms specializing in hydrogeol-
ogy have developed longstanding relationships with the
private sector,  often with those chemical companies who
may be the focus  of aspects  of the Superfund program.
Conflict situations may  arise, and even if a conflict does
not  exist, a firm identified with the private sector may not
wish to participate in government prosecuting efforts.


EXPERT TESTIMONY

   Closely related to "conflict of interest," the  require-
ment of expert testimony suggests that it may be  difficult
in some instances to obtain the services of recognized ex-
perts in support of the efforts to assign financial respon-
sibility for cleanup activities. Probably many of the recog-
nized experts are either employed by or closely associated
with the private sector  and  may be  unavailable. How
serious this situation will be is impossible to assess at this
time.

SAFETY TRAINING

  Several private sector  contractors have mentioned the
need for some predetermined procedures to  assure the
safety of personnel both in cleanup and transport and
handling  of hazardous wastes. Presumably, if such  pro-
cedures  do not already  exist, an established check list
could be developed for different categories of hazardous
wastes. It was further suggested by one  respondent that
EPA should provide some type of seminar or training pro-
gram  for contractor supervisory personnel; incidentally,
the response to such program would be an indication of
active interest in the Superfund program.

INSURANCE

  The problem of proper insurance for protection against
personal injury and damages and professional liability was
cited as a possible deterrent to participation in Superfund.
(Other respondents tended to minimize this problem on the
assumption  that the  necessary insurance is or would be-
come  available at a cost which would be included in the
bid price.) It was also suggested that the contractor might
insert a "hold harmless" clause in any contract.
  If insurance develops into  a costly or significantly de-
terring factor, EPA may wish to self-insure, at least above
some  maximum amount  covered by the Contractor's
"normal" insurance. This procedure would have the effect
of eliminating small  "nuisance" claims from Agency re-
sponsibility and significantly reduce insurance costs to the
individual contractors.  (The  Washington  Metropolitan
Area  Transportation Authority  uses  a "self-insurance"
system for its contractors.)

PERMITS AND LICENSES

  Under this general category may be included the "paper-
work" and "documents" necessary to begin or carry out
specific tasks which the contractor is otherwise prepared to
execute. These delays, particularly after mobilization, are
viewed  as arbitrary, costly and detrimental  to efficient
work  performance. The identification of and responsibil-
ity for obtaining necessary documentation should be clear-
ly defined.
  While it   was  suggested  that  contractors  provide
necessary documentation, another more practical  sugges-
tion was that EPA or a designated State agency provide
a comprehensive checklist of federal, state and  local per-
mits and licenses to the contractor for his compliance.
Such a government agency would be more knowledgeable
in such matters than the Contractor and could outline pro-
cedures to be followed to minimize contractor effort and
uncertainty in reflecting such measures in his bid.

AVAILABILITY OF DISPOSAL SITES

  The problem of adequate disposal sites is recognized at
all levels of  government and industry. From the contrac-
tor's point of view, the problem is one of having  a  place
and/or  approved  technique which  is  suitable for his
immediate needs. This problem is in a sense related to
"Permits  and  Licenses",   and reflects the contractor's
concern with delays and "red tape."

FINANCIAL ARRANGEMENTS

  Some concern was expressed about the problem  of
"getting paid". (Some government bodies—federal,  state
and local—have reputations as "slow pays".) Given cur-

-------
                                                                                                 IMPACT     13
rent high interest rates, it has become increasingly diffi-
cult to finance work as well as  carry it out. Assurances
and conformance  to  a definite predetermined payment
basis and schedule should mitigate these fears. However,
almost any payment terms and schedule can be fully re-
flected in a bid and cause a contractor no concern at all.
Compliance with such terms is the crucial aspect that will
determine whether the private sector will continue to re-
spond to Super fund needs.


SOME OTHER CONCERNS

  It was stated that contractors are most effective in their
local areas, and should not be expected to move their men
and equipment to distant sites.  There were suggestions
that some government agencies were unfamiliar with con-
tractors' work and general business problems; thus it was
suggested that working through an overall General Con-
tractor or Trade Association might help communication.
(Some cleanup jobs are not expected to be large in dollar
volume and it is important to minimize procedural difficul-
ties.) Having regional contractors advisory councils pro-
viding "up  front" advice  to EPA  is another method of
addressing this concern.


BUREAUCRACY

  One respondent, who performed work for all levels of
government on an ad hoc basis, would prefer to work for
private clients because of the "bureaucracy" involved.
Additional personnel, it was feared, would have to be
hired to cope with paperwork  and the  administration
attendant on government contracts. There  is probably
nothing to be done about this problem but it suggests that
there is a reservoir of talent available which regards gov-
ernment work as a last resort.
DISCUSSION

  The foregoing problem  areas were identified during the
Study but it is improbable that other concerns exist which
were  not uncovered because of  sample-size limitations.
Despite this caveat, the authors believe the results indicate
that while some significant  (and many less important)
problem areas do, in fact, exist,  most appear capable of
resolution as the Superfund program is fully implemented.
CONCLUSIONS

  Basically, except for certain skills  in the fields  of
hydrogeology and  chemistry, no evidence of insufficient
technical and management capabilities in the private sector
to implement Superfund has been found.
  Problem areas include: a lack of suitable disposal sites,
costly or inadequate insurance coverage, inordinate effort
to identify and obtain required permits and  licenses,  po-
tential conflicts of interest  with respect to needed firms
and experts, the provision of safety training and the need
to develop suitable contract terms which reflect contrac-
tor  concerns and  special considerations  inherent in  the
Superfund work. None of these problems, in and of it-
self, is insuperable, but the aggregation seems likely to in-
hibit  participation to some extent  unless each  is fully
addressed.
  A majority of the firms contacted would be willing to
participate in Superfund. The level of participation  will
depend on general business activity, alternative  uses of
equipment and personnel and successful resolution or mit-
igation of the problem areas previously described.
  In general, based on discussions in New York and New
Jersey, there is sufficient contractor capacity to meet the
likely level of demand  generated by Superfund activities
and no specific areas of expansion appeared necessary.
A likely shortage of mass spectrographic equipment  and
professionals to operate such equipment has been noted.

REFERENCES

1. "The Blue Book  Contractors  Register",  New York,
  New Jersey, Connecticut Edition,  1979-1980; Washing-
  ton, Baltimore,  Philadelphia Edition, 1979-1980.
2. "Directory—The  American  Council of  Independent
  Laboratories, Inc.," 1980.
3. "Directory—Utility  Contractors  Association  of New
  Jersey," 1980.
4. "AGC's Complete  Directory of General Contractor
  and National Associate Members" (Association of Gen-
  eral Contractors), 1980.
5. "Approved  Pollution Contractors,"  Basic Ordering
  Agreement, Third Coast Guard District.
6. "Official Register—American  Society  of Civil Engi-
  neers," 1980.
7. "Directory—American Consulting Engineers Council,"
  1980.

-------
                   RANKING SYSTEM FOR RELEASES OF
                            HAZARDOUS SUBSTANCES

                                          STEVE CALDWELL
                                 U.S. Environmental Protection Agency
                              Office of Emergency and Remedial Response
                                           Washington, D.C.
                                          KRIS W. BARRETT
                                          S. STEVEN CHANG
                                        The MITRE Corporation
                                            McLean, Virginia
INTRODUCTION

   In recent years, hundreds of incidents involving hazard-
ous substances  have occurred in the United  States,  in-
cluding rail and barge spills, hazardous waste explosions,
direct contact poisonings, food crop contamination, toxic
air pollution, as well as ground and surface  water con-
tamination.  The   Environmental   Protection   Agency
(EPA) has  identified over  10,000  inactive  hazardous
waste sites, many of which continue to threaten  the pub-
lic. In addition,  thousands of hazardous substance spills
occur each year. A number of major rivers have become
contaminated with persistent toxic pollutants, often due to
poor waste handling practices long since discontinued.
   The large number of problems and the high  costs of
investigation and cleanup activities have forced those pub-
lic agencies responsible  for  hazardous substances pro-
grams to set priorities for response.  In general, this has
been done at the State level, largely on the basis of pro-
fessional  judgment. In this  paper, the authors  describe
the status of a  system currently under development for
setting priorities for remedial actions to address hazardous
substances releases.
   In  passing  the  Comprehensive  Environmental  Re-
sponse,  Compensation,   and  Liability  Act  of  1980
(CERCLA), Congress recognized the need for a systemat-
ic  approach to setting priorities. CERCLA  Section 105(8)
requires  the  President  to include criteria for setting
priorities  among releases and potential releases of hazard-
ous substances as a part of the National Contingency Plan.
The criteria are to take into account population at risk,
the nature of the hazardous substances, the potential for
contaminating drinking water supplies, the potential for
direct human contact, potential for destruction  of sensi-
tive ecosystems, State preparedness and other appropriate
factors.
   In addition, the Act requires  the States  to apply these
criteria to establish  priorities for remedial actions at  fa-
cilities and submit them to the  President.  The President
must then establish a National Priority List of at  least 400
facilities based on the criteria and taking into considera-
tion the States' priorities. The National Priority List is to
be used  in selecting the  most serious hazardous sub-
stance problems for remedial action.
  In response to the program needs and  legal  require-
ments for a system for setting priorities, EPA along with
the MITRE Corporation undertook development  of a
method for ranking facilities according to risks to health
and the environment. The objectives of the project were:
•To develop a system for ranking facilities according to
 risks
•To develop a system that would give consistent results
 when applied by various user organizations
•To develop a system that could be applied by the States,
 with the results then used by EPA to form a national
 priorities list
  Several other considerations were important in shaping
the development  of the system.  Since approximately 400
out of thousands of facilities are to be listed, the system
should discriminate most accurately among the very worst
problems. In the course of developing  a  list of at least
400,  as  many as  several  thousand  facilities might be
evaluated using the criteria; thus, costs to collect data
and apply the criteria are  a major concern. In  practice
that  means that  accuracy  in results has  been balanced
against costs of data collection. Finally, from  the outset
the EPA established the general policy that public health
considerations would outweigh environmental effects.

GENERAL APPROACH

  Ideally, in order to evaluate the harm to public health
and  environment  for  the  priorities list, investigators
would  simply measure  all  effects and convert them to
some common denominator  (e.g., dollars). In addition,
where effects have not already  happened, the potential
for harm would be expressed as a quantitative risk assess-
ment; for example, one might predict the  number of in-
creased  cancers in a given  population. Unfortunately,
demonstration of health effects due to exposure to harm-
ful substances is  extremely costly, if not generally in-
feasible.  Likewise,  EPA studies indicate that  precise
quantitative  risk  assessments of  hazardous waste  facili-
ties are not generally feasible due to uncertainties sur-
rounding the health effects of mixtures of chemicals and
the difficulties  of  accurately  predicting   migration of
groundwater contaminants.'" Finally, even assuming the
technical ability  to perform such  analyses,  the  costs
                                                    14

-------
                                                                                                 IMPACT
                                                    15
would be entirely too high to warrant their use in imple-
menting CERCLA. A simpler system is required.
  Three  existing general approaches to setting priorities
were then identified: classification systems; judgment by
"experts"; and scoring systems. Classification  systems,
used by  several State  and local  agencies,  have the dis-
tinct disadvantage that in order to  be useful in setting
priorities, every member of one  class must be higher in
the distinguishing characteristics (risk) than every member
of the next lower class. The variety and combinations of
hazardous substance problem types is not amenable to re-
duction to such a classification system.
  Judgment by "experts" has been the most widely used
approach to setting priorities, and within a single organi-
zation may be the best approach. However, it is doubtful
that such an approach would be consistently applied from
State to State. Perhaps even more importantly, while such
a system might accurately rank  facilities within a  State
or region, it offers no mechanism  for merging results from
all  States to create a National  Priorities List. A scoring
system of some sort was  deemed the most desirable ap-
proach.


REVIEW OF EXISTING APPROACHES

  Several models have been developed for rating the rela-
tive hazard to  public health and the environment posed
by  various hazardous  substance  facilities. These include
the LeGrand and Surface Impoundment Assessment (SIA)
models,  the EPA Solid and Hazardous  Waste Research
Division (SHWRD) Predictive Method (SPM),  and the
Rating Methodology Model. The first two of these are
concerned only with groundwater contamination, whereas
the SPM might be  applicable to  a variety of routes and
the Rating Methodology  Model  addresses both ground-
and surface water contamination. The characteristics of
these models are outlined in the following sections.
The LeGrand Model
  The LeGrand Model'2' describes the potential  for con-
tamination of water wells by waste disposal sources. The
final  rating reflects the potential hazard of the wastes,
the likelihood  of the wastes reaching the groundwater,
and the  vulnerability of the groundwater to contamina-
tion. However, the  model does not address air or surface
water pollution problems.
The Surface Impoundment Assessment
(SIA) Model
  The SIA Model(3) expands on the  scope of the Le-
Grand Model to evaluate  the potential threat of contam-
ination  to  groundwater itself,  rather than the potential
threat of contamination of wells. However, the model does
not address surface water pollution, air pollution or  other
types of risks caused by hazardous substances.
  The SIA Model has been adapted in a modified form by
the State of Michigan  Department of Natural Resources
to rank  their hazardous waste sites. The toxicity factor
has been revised, and  factors have been added  for con-
sidering the number and proximity of drinking water wells
and of people using these wells in the immediate proximity
of the site.


SHWRD Predictive Method
  The SHWRD Predictive Method(4) involves the applica-
tion of multivariate analysis to perform two functions:
•Evaluate the relative importance of various rating factors
 (e.g., distance to groundwater)
•Classify sites into 2 or more categories as to their poten-
 tial for an impact (e.g., 3  categories such as  "good",
 "minimal", and  "bad", according to compliance with
 accepted environmental standards)
  Classifying facilities into  a number of classes would
probably not provide adequate discrimination among them
to effectively implement  the priority list requirements of
CERCLA.   In  addition,  constructing  a representative
mathematical  function  resembling a  regression  equa-
tion  to deal with the multiple routes and  an infinite
classification (i.e., no   pre-established  cut-off  points)
would require more information than would generally be
available at each facility. The approach of the method
might be very useful in refining rating factors and weights
once a number of sites have  been scored and data on the
sites is available for multivariate analysis.

The Rating Methodology Model
  The  Rating Methodology  Model(5) was  developed  pri-
marily to  assess landfills,  surface impoundments  and
other types of land-based storage  and disposal facilities
for the purpose  of setting priorities for technical investi-
gations. The hazard is rated in four general areas:
•Receptors
•Pathways
•Waste characteristics
•Waste management practices
  "Receptors"  are humans  and other living  beings  and
their environment  that may  be affected adversely by the
hazardous  wastes.  "Pathways" are the routes  or media
(e.g., groundwater, air) that  the waste is likely to traverse
in reaching the receptors. "Waste characteristics" are the
hazardous  properties  of the  waste including  mobility,
toxicity and ignitability.  "Waste management practices"
refers to the designs and procedures that have been used
in managing and containing the waste.
  Thirty-one factors were identified as a means of rating,
on a scale of 0 to 3, the potential contribution of the four
areas of the overall hazard.  If factors do not apply to a
specific site or cannot be evaluated due to lack of data,
they may be omitted.  Alternatively, extra factors may be
added  to account  for special considerations that are not
addressed by the 31.  A  fixed multiplier associated with
each factor serves  to weight that factor's relative impor-
tance. The  final score is obtained by adding the products
of all the factors and  their corresponding multipliers and
by normalizing the result on a scale of 0 to 100.
  The  State of New Jersey Department of Environmental
Protection  has  adopted the Rating Methodology Model
and  converted it into a classification system. All of the
sites in the State are  grouped into one of three priority

-------
16
IMPACT
classes and there is no distinction drawn within each class
when assigning the remedial action priority to a site.

Hazard Ranking System
  The Hazard Ranking  System is  a scoring system  de-
signed to address the full  range of problems  resulting
from releases of hazardous substances.  Unlike  the four
approaches  previously discussed, it  is  designed to  ad-
dress surface water,  air, fire and  explosion, and direct
contact, in addition to groundwater contamination.
  The  system applies a structured  value  analysis  ap-
proach, similar to the Rating Methodology Model. Three
migration routes of exposure, groundwater, surface water
and air are evaluated and their  scores are  combined to
derive a score representing the relative risk  posed by the
facility. Two additional  routes of exposure, (1) fire and
explosion and (2) direct contact, are measures of the need
for emergency response.
  The  system differs  from  the  Rating  Methodology
Model  primarily in  that the routes are  scored  inde-
pendently and the  dependent nature of the variables is
reflected by multiplying  where appropriate rather than
simply  adding. Also, unlike  the Ranking Methodology
Model, the Hazard Ranking  System  was  designed  spe-
cifically for the purpose  of ranking facilities for remedial
action.  The system therefore requires a greater amount of
more detailed information.
Description
  Application of the Hazard Ranking System results in
three scores for a  hazardous waste facility. One score,
SM,  reflects the potential for harm to  humans or the en-
vironment as a result of migration of a hazardous  sub-
stance away from the facility by routes involving ground-
water, surface water or air. SM is a  composite of separate
scores for each of the three routes. Another score,  Spg,
reflects the potential for harm from  materials  that can
explode or  cause fires. The third,  SDC,  reflects  the  po-
tential for harm as a result of direct human contact with
hazardous materials at the facility (i.e., no migration need
be involved).
  The score for each hazard  mode (migration,  fire and
explosion, and direct contact) or route is obtained by con-
sidering a set of factors  that characterize the hazardous
potential  of  the facility. A  comprehensive listing  of
factors for all of the hazard modes is given in Table I.
  Each  factor is assigned a numerical value  (generally on
a scale of 0 to 3, 5 or 8) according to prescribed guide-
lines. This value is then multiplied  by  a  weighting factor
to yield the  factor score.  The factor scores are then com-
bined by following established guidelines: scores within a
factor category are additive but the  factor category scores
are multiplicative.
  In computing Spg  or S^, or an individual migration
route score, the product of its factor  category scores is
divided by the maximum value the product can have and
the resulting ratio is  multiplied by 100, thus  normalizing
scores to a 100-point scale.
  Computation of SM is slightly more complex since SM is
a composite of the  scores for the three  possible routes:
                                                    groundwater (S^J, surface water (Ssw) and air (Sa). SM is
                                                    obtained from the equation:
                                                          _ 1    fc  2  i  e2  i C 2
                                                          -
(1)
                                                    The factor 1/1.73 arises from the vector addition of the
                                                    three route scores after the individual scores are normal-
                                                    ized to a common denominator.  This means of combin-
                                                    ing them gives added weight to routes with higher scores.
                                                      The Hazard Ranking System does not result in quantita-
                                                    tive estimates of the probability of harm from a facility
                                                    or the magnitude  of the harm that could  result. Rather, it
                                                    is a device for rank-ordering  facilities in  terms  of the po-
                                                    tential hazard they present. Risks  are generally considered
                                                    to be a function of the probability of an event occurring
                                                    and the magnitude or severity should the  event occur. Ap-
                                                    plying this approach to hazardous substance  facilities,
                                                    the probability  and magnitude  of a release are generally
                                                    functions of the following areas:
                                                    •The manner in which the hazardous material is contained
                                                    •The route by which its release would occur
                                                    •The characteristics of the harmful substance
                                                    •The amount of hazardous substance
                                                    •The likely targets
                                                    These areas have been  examined and representative fac-
                                                    tors were chosen to address each.
                                                      The scoring guidelines for each factor and the weight
                                                    accorded  to each  factor  were developed   based  on
                                                    judgment initially. The weights and guidelines  were then
                                                    adjusted based on tests using data from 43 facilities.

                                                    Using the Hazard Ranking System
                                                      Use of the Hazard Ranking  System requires consider-
                                                    able information about the  facility,  its hazardous sub-
                                                    stances content, its surroundings for distances up to three
                                                    miles and the geological character of the  area down to
                                                    the depth of aquifers that may  be at risk. Complete data
                                                    may not be available and the individual assigning scores
                                                    may have to use some judgment in applying the guidance
                                                    provided.
                                                      Geology, hydrology,  chemistry  and the  ecological  sci-
                                                    ences are the most relevant disciplines and  scoring is best
                                                    accomplished using a team of individuals knowledgeable
                                                    in these disciplines.  A  format  for recording general in-
                                                    formation regarding the facility being evaluated is shown
                                                    in Figure 1. This figure can also serve as  a cover sheet for
                                                    the work  sheets  used  in the evaluation.  Sample  work
                                                    sheets are shown in Figures 2 through 7.
                                                      Where data for a factor are unavailable, it  should be
                                                    assigned a value of zero. However, if the factor for which
                                                    the data are missing is  the only factor in a  category (e.g.,
                                                    containment or waste quantity), then the factor is given a
                                                    score of unity. If data are  lacking for more than  one
                                                    factor in connection the evaluation of S™,  S™, Sa, Sen or
                                                                  .          *  !•          gw*   sw»  a»  re
                                                    Spc,tnat score is automatically set at zero.
                                                      Detailed instructions  and guidance are provided in the
                                                    guidance manual  for using the Hazard  Ranking System.
                                                    As mentioned previously, if sufficient data are available,
                                                    three scores are computed, SM, S^ and  S^; SM being a

-------
HAZARD MODE
Migration





















Fire and
Explosion














Direct
Contact





FACTOR CATEGORY

Containment
Route
Characteristics





Waste
Characteristics

Hazardous Waste
Quantity

Targets







Containment

Waste
Characteristics



Hazardous Waste
Quantity
Targets






Accessibility
Containment
Waste
Characteristics
Targets


Ground Water Route
• Containment
FACTORS
Surface Water Route
• Containment
Air Route

• Depth to aquifer of concern • Site slope and Interviewing

• Net precipitation
terrain
• 1 Year 24 hour rainfall


• Permeability of unsaturated • Distance to nearest surface
zone


• Physical State
• Persistence
• Toxicity
• Total waste quantity


• Ground water use

water
• Flood potential

• Physical State
• Persistence
• Toxicity
• Total waste quantity


• Surface water use

• Distance to nearest well • Distance to a sensitive
down gradient
environment


• Volatility/Physical State
• Reactivity
• Incompatibility
• Toxicity
• Total waste quantity


• Distance to nearest human
population
• Population within 1 mile
radius
• Population served by ground • Population served by surface • Distance to sensitive
water drawn with 3 mile radius water drawn within 3 mile


• Containment





radius environment
• Land use



• Direct evidence of Ignitibllity or exploslvlty
• Ignltibillty
• Reactivity
• Incompatibility

• Total waste quantity













• Distance to nearest human population
• Distance to nearest building
• Distance to sensitive environment
• Land use


• Population within 2 mile radius
• Number of buildings within 2 mile radius

• Accessibility
• Containment
• Toxicity











• Population within a 1 mile radius
• Distance to a critical
habitat
















O
o
•I
3
EP
n
HI
~
n
r+
o

i

5
JO

so
n
o
3















-------
18    IMPACT
composite of separate scores for three  release routes:
groundwater, surface water and air.  It is this composite
migration score that is used to rank facilities for remedial
action priorities.

SYSTEM TESTING

   The equations  and factors described above were settled
on after a comprehensive review and selective testing was
performed on previous drafts. The review included EPA
Regional personnel, contractors  and  industry representa-
tives.  Major anomalies were  pointed out, particularly in
the case of a city of 50,000 population with a threatened
drinking water supply  and a composite  migration score
of 28. The  results of  preliminary testing are  shown in
Figure 8.
   The ideal distribution for maximum discrimination is a
straight line  with a slope of 1.0. The  preliminary  data
was skewed toward the higher values because the 43  test
facilities were not chosen randomly  but were  generally
those sites considered "bad" by the EPA regional offices.
This  trend would be expected  even  under  ideal condi-
tions  since it is  unlikely that individuals would spend
time and money to score facilities that are apparently a
very low priority.
   A sensitivity analysis was  done to  analyze the sensitiv-
ity of  the  final  ranking score to changes in  individual
factor  scores. This analysis  showed that the hazardous
waste  quantity score had the greatest effect  on the final
ranking score. A change of  one point in a  factor value
 Facility Name:

 Location:

 EPA Region:
 Person(s) in Charge of the Facility:
 Name of Reviewer:
 General Description of the Facility

 (For example:  landfill, surface impoundment, pile, container; types
  of wastes;  location of the site; contamination route of major con-
  cern; types of information needed for rating; agency action, etc.)
  Scores:  S  •
GROUNDHATER ROUTE WORKSHEET
Rating Factor
Assigned Value Multi-
(Circle One) plier
=" Observed Release 0 45 1
Score

Max.
Score
45
If observed release Is given a score of 45, proceed to line I4|.
If observed release is given a score of 0, proceed to line [ 2|.
- Route Characteristics
Depth to aquifer 0123 2 6
of concern
Net Precipitation 0123 1 3
Permeability of 0123 2 6
unsaturated zone

3!
Containment
Total Route Characteristics Score
0123 1


15
3
J Waste Characteristics
Physical State 0123 1 3
Persistence 0123 2 6
Toxicity 0123 2 6

- Hazardous Waste
Quantity
Total Waste Characteristics Score
012345678 1


15
8
=1 Targets
Groundwater use 0123 3 9
Distance to nearest 0123 3 9
well doungradlent
Population served 012345 6 30




^ IE line [T| is 45, multiply [T] x [T] x (T) x [T]
If line [T] is 0, multiply [I]x[T]x[4]xl5]x(f[] j
" Divide Q] by
48
59,201
259,200 and multiply by 100 Sg« -
                        Figure 1.
                    Model Cover Sheet
                        Figure 2.

could, under worst case, effect the final score by 5.7%.
The sensitivity analysis  assumed  that individuals assign-
ing values  according to the  guidelines would  not err
more than one point on any factor.
  As a result of the  review  comments received, the pre-
liminary test results  and the sensitivity analysis, major
changes were made in the original Hazard Ranking Sys-
tem. Among these were:
•Elimination of fire and explosion and direct contact as
 routes to be considered for facility ranking
•A reduction in the number of factors
•Revision of the population factors to increase the Sys-
 tem sensitivity to human populations
•Revision of the hazardous waste quantity factor.
FUTURE DEVELOPMENT

  Development and testing of the system is continuing,
and some revision is expected prior to implementing the
National  Priorities  List of 400 facilities. Areas of de-
velopment include:

-------
                                                                                                         IMPACT
                                                                                                                     19
a
2JRoute Characteristics

  Site Slope and         0123
   Intervening Terrain
  1-yr. 24-hr. Rainfall   0123
  Distance to Nearest     0123
    Surface Water
  Flood Potential        0123
U
                     SURFACE WATER ROUTE WORK SHEET
 Rating Factor
                       Assigned Value
                        (Circle One)
                                      Muli-
                                     /plier
  Observed Release
                            45
                                                Score
 Max
Score
                                                      45
  If observed release is given a value of 45, proceed to linef
  If observed release is given a value of 0, proceed to line|2
                Total Route Characteristics Score
   Containment
                        0123
   Waste Characteristics

    Physical State        0123

    Persistence          0123

    Toxicity             0123
                Total Waste Characteristics Score
  Hazardous Waste
  Quantity	
                     012345678
 IjTargets

    Surface Water Use

    Distance to a Sen-
     sitive Environment

    Population Served
                     0123

                     0123
                        012345
                      Total Targets Score
 -Zllf line [T| Is 45, multiply [D x IT1 x ITI x ITT

   If line [T| is 0, multiply [TlxlTlxrnxlTlx [Tl
[-^
Divide (T] by 243,000 and multiply by 100
                                                      15
                                                      15
                                                     45
                                                 243,000



AIR ROUTE WORK SHEET

Assigned Value ^
Rating Factor (Circle One) p
— 'Observed
If line
If line
Release 0 45
T"| is 0, then Sa 0. Enter on line
1 is 45, then proceed to line \2\
— 'Volatility/physical 0123
Staff

Reactivity 0123
Incompatibility 0123
Toxicity

3 JHazardoUv

— 'Distance
0123
Total Waste Characteristics
WaSte 012345678

to Nearest 0123
lulti- Max.
Her Score Score
11 45
a

1 i 3

11 3
11 3
22 6
Score 15
1 8

2 6
Human Population
Population Within 012345
1-Mile
Distance
Radius
to Sensitive 0123
5 25

2 6
Env ironmen t
Land Use

— ^Multiply
0123
Total Targets Score
0 . m -• m . H
1 3
40
216,000
— 'Divide [JJ by 216,000 and multiply by 100 Sa
                          Figure 3.
                                                                                      Figure 4.
•Cost of securing data
•Anomalies
•Testing
•Quality assurance
  The  cost of securing data  is  of concern because of
the quantity of detailed sampling and analysis that is re-
quired  to score a facility. A method is needed to screen
sites before spending funds to collect extensive data for a
site that will not rank above a given level.
  Anomalies,  facilities  that  have extreme characteristics
outside  the scope of the system will occur regardless of
the complexity of detail required  by the  Hazard Ranking
System. Where  a class of anomalies can be identified,
the system  can be modified  to more accurately rank fa-
cilities according to risk.
  Field testing is required to measure the ability of the
system to  discriminate  and  rank sites with independent
judgment.  A method has been proposed to use an inde-
pendent panel of experts to  rank  a selected group of
sites  and then compare  these results with  the  Hazard
Ranking System  scores.  The  two  ranking  lists would
then be analyzed for a statistical correlation.
                                                               Groundwater Route Score
                                                                                    ™
                                                              Surface Water Route Score (Ssw)
                                                               Air Route Score  (Sa)

                                                                                                              V
                                                                                      Figure 5.
                                                                             Worksheet for Computing SM
                                                                Quality assurance is necessary to ensure  that there is
                                                             consistency  in the  application of  the Hazard  Ranking
                                                             System among the individuals assigning scores.

-------
 20    IMPACT
                        Figure 6.
FIRE AND EXPLOSION WORKSHEET
Rating Factor

Containment
Assigned Value Multi-
(Circle One) plier S

1 3 1

Max.
core Score

! 3
-) Route Characteristics
Direct Evidence 03 1
Ignltability 0123 1
Reactivity
0123 1
Incompatibility 0123 1

Total Waste Characteristics Score
3) Hazardous Waste 012345678 1
Quantity

-" Targets
Distance to


Nearest 012345 1
Population
Building
Distance to

Sensi- 0123 1
3
3
3
3
12
: 8


5


3
tive Environment
Land Use
0123 1
Population Within 012345 1
3
5
2 Mile Radius
Buildings Within 012345 1
5
2 Mile Radius

Total Targets Score
^ Multiply |TJ x Q] x [T] x [Z]
^ Divide (T] by
6912 and multiply by 100
24
6912
Sfe-

Rating Factor
i> Observed Incident
- If line (T
If line 1
is 45
is 0,
DIRECT CONTACT WORKSHEET
Assigned Value Multi-
(Circle One) plier
0 45 1
Score

Max.
Score
AS
proceed to line HO
proceed to line |2|
2) Route Characteristics
Accessibility 0123 1
3J
Containment

0 15 1
*1 Waste Characteristics
Toxicity 0123 5
^ Targets
Population with:
1-mile radius
Distance to a
critical habit
§1 If line T
If line T

0123 4
at
Total Targets Score
is 45, multiply JT] x J3 x J3 _,
is o. muitioiy mx m x fTTx fFI



3
IS
IS
20
12


32
21,600
^ Divide line 0 by 21,600 and multiply by 100 Sd(, •
                                                                                 Figure?.
                                                                               Facility Ranking Score
REFERENCES

1. "Identification,  Assessment,  and   Evaluation   of
   Hazardous  Waste Facility Siting Risks," Teknekron
 Research, Inc., McLean, Virginia, July 1980.

2. LeGrand, H.E., "A Standardized System for Evaluat-
   ing  Waste-Disposal  Sites,"  National   Water Well
   Association, 1980.

3. Silka,  L.R. and Swearingen, T.L.,  "A Manual  for
   Evaluating  Contamination Potential of Surface  Im-
   poundments,"  U.S. Environmental Protection Agency
   (EPA 570/9-78-003), June 1978.
                       Figure 8.
             Frequency Distribution of Facility
   Ranking Scores for 43 Preliminary Test Facilities (April 1981)


4.  Klee,  A.J.  and  Flanders, M.U.,  "Classification of
   Hazardous Wastes," J.  oftheEnvir. Eng. Div., ASCE
   106, No. I.February 1980.
5.  Kufs,  C., Twedell, D., Paige, S., Wetzel, R., Spooncr,
   P.,  Colonna,  R. and  Kilpatrick,  M.,  "Rating the
   Hazard  Potential of Waste  Disposal Facilities," US
   EPA  Conference on  Management of Uncontrolled
   Hazardous Waste Sites, October 15-17, 1980, Wash-
   ington, D.C., HMCRI, Silver Spring, Md., 1980,30.

-------
           STATE-FEDERAL SUPERFUND COORDINATION:
                        WHO WILL RUN THE PROGRAM?
                                        JACQUELINE M. RAMS
                                          BARBARA SIMCOE
                                            DAVID DUNCAN
                 Association of State and Territorial Solid Waste Management Officials
                                            Washington, D.C.
INTRODUCTION

  The Comprehensive Environmental Response, Compen-
sation and Liability Act of 1980 (CERCLA), better known
as "Superfund," provides the institutional framework for
Federal and State  agencies to respond to emergency re-
sponse needs at abandoned hazardous  waste sites.  It
focuses primarily on waste sites presenting an imminent
danger to  the public,  removal of wastes and/or remedial
actions taken at those sites and coordination between
Federal and State agencies engaged in those activities.
  At present, an implementation strategy for Superfund is
being fine-tuned to specify the Federal  and State roles
for  immediate  or planned removal  efforts  and subse-
quent remedial  actions.  In addition,  the National Con-
tingency Plan is also  being revised to include a national
hazardous substance  response  plan which  will establish
basic  criteria  for emergency action response as well as
those for site evaluation and ranking.
  Since the mid-1970's, several States have taken the in-
itiative in establishing their own implementation  plans
and  funds J;p  deal with past negligent  waste manage-
ment procedures and  emergency response. In this paper,
the  authors examine the  intended role of Superfund and
State funding mechanisms and analyze potential problems
which may be encountered by States as  they attempt to
coordinate their own program efforts with federal Super-
fund activities.

SUMMARY OF SUPERFUND

  The Resource Conservation and Recovery Act (RCRA)
was enacted by Congress in 1976 to upgrade existing state
and local waste management programs to a uniform Fed-
eral level  to  assure protection of human health  and the
environment and conserve valuable materials and energy
resources. Subtitle C  of RCRA specifically sets forth a
regulatory protocol to track hazardous  materials from
"cradle  to grave" (i.e., generation/storage/transporta-
tion/disposal) and any real or potential release to the en-
vironment that might result in these operations.
  The Comprehensive Environmental  Response, Compen-
sation and Liability  Act of 1980 (CEDRCLA),  better
known as superfund, was enacted in 1980  to specifically
address  existing waste sites presenting imminent  health
and environmental dangers as a result of past waste man-
agement  practices. The  law provides  the  institutional
framework for Federal agencies and the States to respond
to hazardous waste incidents and emergency spills not cov-
ered by the rigorous RCRA protocol. It establishes a liabil-
ity  system for specific pollutant  releases, provides for
long-term care for removal of waste and/or  remedial
actions at sites presenting imminent dangers (as well as
short-term cleanup), calls for a coordinated national strat-
egy for quick response in the event of an emergency spill
or accident, and establishes a fund to finance such immed-
iate response.
  The two laws complement each other  and, hopefully,
together with the assistance of the regulated community,
will demonstrate government's concern and intent to the
now highly sensitized and volatile public.
  In the past, the ultimate responsibility for cleaning up
abandoned waste sites or the dangers imposed by hazar-
dous substances and dealing  with the political, legal, and
financial  expenses involved, was borne almost exclusively
by  state  and local governments.  Since  the mid-1970s,
moreover, states have taken the initiative in establishing
their own implementation plans and funds to deal with
both negligent management procedures  and emergency
response.
  It, therefore, seemed practical that when the Superfund
legislation was making its way through the Congressional
process and finally enacted, state and local governments,
because of their prior experiences, should have an intimate
role in shaping the legislation. The public's concern with
the  abandoned site issue has  become  so volatile that state
and local elected officials alone have the ekxperience to
promote the necessary public trust and curb the political
sensitivity that now exists.
  Superfund recognizes the need for a state/federal part-
nedrship and, through Congressional intent,  provides the
flexibility for extensive state participation should EPA be
willing to adopt such an approach.  However, the states
are  finding it difficult to implement the programs man-
dated by  Superfund because of the lack of direction and
clarification of responsibilities provided thus far by EPA.
  To better determine where those difficulties lie, examine
the  Congressional intent  of  Superfund and  the directive
it mandates: The goals of the law include:

(1) An inventory of inactive hazardous waste sites in a sys-
  tematic manner
                                                    21

-------
22     STATE PROGRAMS
(2) Establishment of priorities among the sites based on
   relative danger
(3) Accelerating the elimination of unsafe hazardous waste
   sites
(4) Systematic program of funding to identify, evaluate
   and take responsive actions at inactive hazardous waste
   sites to assure protection of public health and the en-
   vironment in a cost-effective manner.
  Location of sites housing hazardous substances known
or abandoned and spills of hazardous  substances in re-
leasable quantities of one pound or greater as specified
by Section 311 (b) (4) of the Federal Water Pollution Con-
trol Act must be reported to the National Response Center
or the Administrator of EPA or face civil action or fine of
$10,000. (Interim or final RCRA sites are excluded).
  If it is determined that  a hazardous substance or re-
lease of any pollutant or contaminant impinges on the en-
vironment or presents a  substantial threat and endanger-
ment to the public health  or  welfare, the  President is
authorized to act, consistent with the  National  Contin-
gency Plan. In the event that the responsible party can-
not  be found or takes  inadequate or  no  action,  he is
authorized to remove, provide for removal of, or take re-
medial action relating to such  hazardous substances, in-
cluding those specified by RCRA, Section 311 of the Clean
Water Act, Section 112 of Clean Air Act, Section 307 of
Clean Water Act, Section 7 of  TSCA and any others de-
signated by Superfund. Direct response  actions, however,
are limited to a six month or $1,000,000 ceiling, unless an
emergency or immediate risk to public health exists.
  Moreover, remedial action may only  be taken after
consultation  with the affected  state(s),  in the form  of a
cooperative agreement to assure that:
•States will assume all future maintenance of such actions
•There is available an acceptable disposal facility
•The state will assume 10% of the costs, or 50% in the
 case of state-owned facilities
Credit is granted those states having expended funds for
releases since January 1, 1978.  Such costs must be direct
and documented.

National Contingency Plan

  Section 105 of the law mandates promulgation of a Na-
tional Contingency Plan within 180 days after the law's en-
actment to establish  procedures and  standards  for re-
sponding to releases of hazardous substances, pollutants
and contaminants. The plan must include:
•methods for discovery and investigation
•methods for evaluation and remedy of releases
•methods and criteria concerning the appropriate extent of
 response
•appropriate roles of all government or non-governmental
 entities
•provision for equipment and supplies
•provision for reporting of federally owned facilities
•a method to assure that actions are cost-effective
•priority criteria  based on a variety of health  and environ-
 mental risk  factors,  and state preparedness to  assume
 state costs and responsibility
Site Priority List

  Subsequently a national priority list,  composed' of at
least 400 of the highest priority facilities will be established
based on these criteria. The top 100 of these sites must in-
clude at least one site per state, as specified by individual
state priority lists. Abatement action may then be sought
by the Attorney General in the event of danger and threat
or by Presidential orders forcing violators to comply with
such orders or be fined up to $5,000 per day. Promulga-
tion  of guidelines using  this response authority, more-
over, must be published by EPA in consultation with the
Attorney General with 180 days enactment of the bill.
  Section 107  establishes  a mechanism for quick recovery
of monies expended from the Hazardous Waste Response
Fund from the liable parties. Owners and  operators who
arranged for disposal or transportation of hazardous sub-
stances at the time of disposal or transportation of haz-
ardous substances at  the  time  of disposal  or who ac-
cepted such substances for disposal at facilities are deemed
liable in accordance with Section 311 of  the Clean Water
Act for costs incurred consistent with  the  National Con-
tingency Plan  and for damage to natural resources. Mon-
etary limits regarding  liability have  been established for
vessels, motor vehicles, aircraft and pipelines. The remain-
ing limits are to be established by regulations.
  Financial responsibility for vessels must be established
and  maintained by owners and operators, and enforced
by the Department of transportation. For facilities, the
level of financial responsibility will be based on the estab-
lishment  of  facility  classification (within three  years)
and consistent with the degree and duration of risk asso-
ciated  with the production,  transportation, treatment,
storage, or disposal of hazardous substances. Regulations
imposing  such responsibilities  must be  promulgated no
later than six years after the date of Superfund enactment.

Financing and Spending

  Monies  for allowable expenditures are drawn from the
Hazardous Waste Response Fund in the Treasury of the
United States.  Sources are as follows:
•Fees collected on crude oil (received at  a U.S. refinery),
 petrochemical feedstocks and certain  inorganic elements
 and compounds
•Monies recovered under liability provisions
•Reimbursements from actions taken
•Penalties collected under the Act
•Appropriated funds
  Approximately 86% of the  fund will be received from
taxes and  14% from  appropriations.  For the proposed
five-year period, approximately $1.6 billion dollars will be
collected or transferred into the fund. EPA estimates that
this will accomplish at least 85 % of the cleanup actions,
provided claims total  no more than  15% of the total.
Uses are limited to the following:
•Response costs
•Claims for response costs by any persons consistent with
 the National Contingency Plans
•Outstanding Section 311 claims

-------
                                                                                      STATE PROGRAMS     23
•Assessment, restoration and rehabilitation of natural re-
 sources, including
  enforcement and abatement
  epidemiological studies, registry and diagnostic testing
  equipment and supplies for Superfund and Section 311
   (subject to appropriations)
  worker health and safety program
  authorized appropriate advances
  Claims against the fund will be negotiated between the
claimant and responsible party,  should that party be lo-
cated and deemed  responsible.  In  the absence of a re-
sponsible party, the fund will negotiate a settlement with
the claimant. Each will be paid on a first come first serve
basis. The statute of limitations is three years  from the
date of enactment or discovery  of  the loss, whichever is
later. Should a state have its own fund, Superfund exer-
cises the right of preemption, i.e., no contributionf can be
required to any  State fund for claims for  response or
damages compensated  under Superfund.  Furthermore,
double compensation is not allowed.

STATE FUNDING MECHANISMS
  As of August,  1981, twenty-four states had authorized
the creation of thirty-two funding mechanisms to meet this
financial challenge. In the accompanying matrix (Chart I)
the major elements addressed in these state funding laws
are shown, including the purposes for which the  funds
were created, sources  from which  fund monies are col-
lected, functions  for which the funds may be utilized and
other key points  including liability clauses, fund ceilings
and actual fund balances to date.
  In order to illustrate the wide range of responsibilities
and actions which these state funds encompass, the follow-
ing descriptive narrative is included to clarify and detail
major provisions  contained in each state funding law.

Purposes

  Michigan first  enacted legislation establishing a "water
cleanup emergency fund" in 1974,  followed  five years
later by the creation of two other funds to address emer-
gency response and perpetual care activities. Louisiana,
which established a Hazardous Waste Protection Fund
in 1979, amended its law in 1980 to create two additional
funds which  address  general  emergency  response and
abandoned site cleanup. Kansas, Tennessee, Texas, and
Wisconsin  have  each created two  distinct  funds within
their respective laws to meet  specific hazardous  waste
management concerns.
  Most state funds are multi-purpose in nature, and au-
thorize  payment  for a  host of  problems resulting from
hazardous  waste  incidents. Eighteen funds address vary-
ing activities  associated with emergency response, nine
funds can be utilized for the maintenance  and perpetual
care of TSD facilities, and seven are authorized to clean
up spills in navigable waters or on land. Thirteen other
funds may be used to cover such expenses as program ad-
ministration and  management, initial facility closure, site
reclamation,  or state costs associated with the implemen-
tation of the federal Superfund.
Sources of Funds

  State laws authorizing the establishment of funds often
explicitly spell out the sources from which the fund is to
collect its money. Eleven states (AS, CA, FL, GA, IL, KS,
MD,  MI, OH, TN, WI)  assess a fee or tax on a TSD
facility owner/operator. Kansas bases its fee on the vol-
ume of hazardous wastes received at the facility, not to
exceed 25


-------
 24    STATE PROGRAMS
FUND PURPCSE
ADMINISTRATIVE/PROGRAM MANAGEMENT
EMERGENCY RESPONSE
FACILITY CLOSURE
PERPETUAL CARE
SPILL CLEANUP
SUPPLEMENT TO SUPEHFUNO
OTHER

SOURCES OF HOVY
ASSESSMENT ON DISPOSAL FACILITY
ASSESSMENT ON GENERATOR
LEGISLATIVE APPROPRIATION
REIMBURSEMENT FROM LIABLE PARTY
PENALTIES/FINES
BONDS
GRANTS/GIFTS
FEDERAL CONTRIBUTIONS
OTHER

UTlLIZAIIOn OF FUNDS
ADMINISTRATIVE COSTS
EMERGENCY RESPONSE
PERPETUAL CARE
RECLAMATION o SITES
EQUIPMENT/PER ONNEL TRAINING
VICTIM COMPEN AT ION
STATE SHARE o SUPERFUND

LIABILITY ESTABLISHED FO?
RESPONSIBLE PHRTY
CFIIING SET ON FUND RAUWCf
DEPOSITS CREDITED TO FUND

•










^-0-






•













9


0


0










•










•

0





0
•
•



0
0


•
•







•












•






















0


































0
0
•
-J

0






0

0









0





0



0






•
*

















6























0
9




0

•'








•
«
0










0
0





0









•












0







~»-
w
0
0
0




•



0










0




0

r0-









0





•


A







0











0












0






•
0
0



















0
0




0

•








•










pt-








0

0

0
0



+















0




0
•







•


0











0







0



0


•.



0










0

0


































•



















•




•


0









A








•




0





0



0





















0
•








0






0

0



•
•



0





0

•








•
0
0





•
•
                                                    Chart 1.
                                   Major Provisions of State Hazardous Waste Funds
  Missouri will begin levying  a  fee on generators effec-
tive January 1, 1982 of up to $l/ton of hazardous waste
registered with the  Department  of  Natural Resources.
Fees shall not  exceed $10,000 per generator per year, and
no fee will be required of generators who register less than
10 tons of hazardous waste annually. New Hampshire re-
quires generators who  produce 300 kg or more of unre-
cycled hazardous waste within a three-month period to pay
up to IC/kg on a quarterly basis to  its Bureau of Solid
Waste Management.
  Eleven states depend upon legislative appropriations to
support their funds, while twelve states attempt to collect
funds from the parties responsible for the spill or contam-
ination. Seven states deposit all  fines and penalties col-
lected from violators  into  their funds,  while  Georgia,
Louisiana, Massachusetts and Tennessee depend upon
bonds to finance their funding  mechanisms. Louisiana re-
lies partially on grants and state/federal allocations  to
support its  three  funds.  Fifty  percent  of California's
Hazardous Waste Control Account is funded by  RCRA
grants, and the state also relies on monies received from
liens imposed on site property.  Florida, Louisiana,  and
Missouri accept grants and gifts from public and private
agencies  to  deposit into their  funds;  New Jersey taxes
the transfer of petroleum products  and hazardous sub-
stances at a rate of l
-------
                                                                                     STATE PROGRAMS
                                                   25
of its hazardous waste management program. California
also established a temporary Site Closure and Mainten-
ance Revolving Account which was appropriated $370,000
from the General Fund to clean up the Stringfellow Quarry
in 1978. Louisiana can use its Emergency Response Fund
to control/contain/abate pollution sources  and  ensure
adequate scientific/technical support of litigation  arising
from a pollution incident, in addition to defraying emer-
gency response costs.
   Colorado's Emergency Response Cash Fund is used for
emergency response efforts where the fire department or
other public agencies provide assistance outside its own
jurisdiction. New York has established a General Gov-
ernmental Emergency Fund which is used for all govern-
ment emergencies,  including  environmental  incidents.
Monies are expended only when the legislature is not in
session to appropriate special funds itself, and the Com-
missioner of Health must declare a state of emergency in
order for the funds to be used. Texas uses one of its funds
to train personnel and obtain equipment and supplies re-
quired for cleanup activities involving discharges or spills
in coastal  areas  and offshore,  including  restoration  of
beaches and marine resources. One of Wisconsin's funds
can  also be used to train personnel and to procure and
maintain  equipment and supplies  needed in emergency
response cleanup efforts.
   Several states can activate their funds  to  provide for
the cleanup and monitoring of inactive or abandoned haz-
ardous waste facility sites. Connecticut's Emergency Spill
Response Fund will pay for the containment and removal
of oil spills, petroleum, chemical liquids or solids, liquid
or gaseous products or  hazardous wastes. Florida's  fund
can  be used to pay for restoration of areas damaged by
releases from  abandoned sites,  while  Kentucky's Haz-
ardous  Waste  Management Fund will respond to  haz-
ardous substance spills and provide perpetual  monitoring
and maintenance of inactive sites.  Georgia and Tennessee
provide for the detoxification and removal of hazardous
wastes,  site reclamation, cleanup, perpetual care and sur-
veillance of abandoned or inactive sites.
   Louisiana's Abandoned Hazardous Waste Site Fund can
be used to defray containment/control/cleanup costs, ac-
quire rights-of-way easements or title  in order to secure
and quarantine a site and provide money or services as
the state share of matching funds for federal grants. Illi-
nois' Hazardous Waste Fund  can be used  for preven-
tive/corrective  actions necessary and  appropriate when
hazardous  waste  disposal sites can cause danger to  pub-
lic health or the  environment,  and will  pay  for aban-
doned site  cleanup if the Governor decides that an emer-
gency exists. The Kansas Perpetual Care Trust Fund will
pay  for extraordinary costs of monitoring inactive sites,
repair sites and  environmental  damage due  to a post-
closure occurrence, and  authorizes up to 20% of the fund
balance to  be used on an emergency basis for  the investi-
gation, engineering, equipment,  and construction needed
to treat or dispose of hazardous wastes. Missouri's fund
will pay for cleanup of hazardous waste emergencies and
abandoned  or  illegal sites, and will  compensate other
state agencies for services rendered.
  Arizona limits its trust fund to paying for the opera-
tion, maintenance and perpetual care of state-owned haz-
ardous waste sites  and facilities, or for mitigating immi-
nent and substantial threats to public health or the en-
vironment created by the site or facility. Michigan's dual
fund system will pay for both prospective hazardous waste
emergencies as defined by agency rules and for the costs of
long-term care of  inactive disposal sites;  Massachusetts'
Division of Waste Pollution  Control was appropriated
$300,000 in FY 80 to use to clean up oil  and hazardous
waste spills.
  New Mexico's emergency fund can clean up and dis-
pose of hazardous substances  and make repairs to or re-
place state property damaged by a hazardous substance
incident. Louisiana can obligate its protection fund to de-
fray emergency response costs, maintain and care for dis-
posal sites, and assure financial responsibility in case of
damages from an accident or negligence. Wisconsin's
Waste Management Fund is used for the  long term care
of inactive sites after owner liability has terminated, or for
site and environmental reclamation needed for damages
caused by  an unanticipated incident.  Ohio's fund con-
centrates on  facility  closures and  air/water pollution
abatement but also authorizes up to  $500,000 to be ex-
pended over a three year period in the form of grants to
municipalities to encourage the siting  of hazardous waste
facilities in their jurisdiction.
  Four states address spills into waters within their juris-
dictions. Maryland allows its  fund  to be  used to defray
program administrative costs, identify and restore natural
resources, and remove  or mitigate the  effects  of haz-
ardous substances  contained in their waters. Kansas can
use its Pollutant Discharge Cleanup Fund for emergency
responses to spills in waterways but Florida prohibits the
use of its trust fund to clean up spills being removed from
navigable waters by a federal  agency, or  spills being re-
moved from coastal waters by the state. Michigan's "water
cleanup emergency fund" can reimburse contractors called
to the scene of a spill for cleanup expenses, and collects
money from liable parties.
  Florida and New Jersey are the  only states so far to in-
stitute a victim compensation  scheme into their  funding
authority. Florida will pay for provable property damage
resulting from a hazardous waste  release, while New Jer-
sey will  reimburse  the costs of restoring, repairing or re-
placing real or personal property damaged by a discharge.
This includes income derived from the damaged property,
the costs of restoring natural  resources damaged by the
discharge, and the loss of tax  revenue by state and local
governments due to property damage.
  Three states have created funding mechanisms in their
1981  legislative sessions  in direct response to coordina-
tion efforts  with the federal Super fund, and at this writ-
ing, both houses of the  California Legislature are con-
sidering bills which would enable that state to meet its re-
sponsibilities in order to qualify for credit under Super-
fund.
  Maine's Hazardous Waste Fund may be disbursed to
pay for, among other  things,  the purchase of necessary
hazardous waste response  equipment and supplies,  per-

-------
26    STATE PROGRAMS
sonnel, and training for such personnel, but the law stip-
ulates that no money may be  disbursed for response or
damage claims which may be compensated under the fed-
eral Superfund.
  The New Hampshire  Hazardous Waste  Cleanup Fund
shall be used to provide for adequate  and safe contain-
ment and cleanup of sites within that  state, after certif-
ication has been received from  the Governor that a threat
to the environment and public health exists. The Bureau
of Solid Waste  Management may enter  into contracts and
use the fund for  hiring consultants and personnel, pur-
chasing/leasing/renting  necessary equipment and paying
for other necessary expenses  and administrative  costs.
The fund shall not duplicate Superfund  functions, nor pay
compensation for claims which may be compensated under
the federal fund.
   Texas established a Disposal Facility Response Fund to
 be used only to provide  the state's required share of funds
 under Section  104 of Superfund, and to pay for removal
 and remedial  actions as required by the  federal law.
 Funds may not be used  for the normal  administrative and
 operating expenses of the Department of Water Pollution
 Control.

 Liability

   Eight states set liability standards for owners/operators
 of hazardous waste facilities in conjunction with the  es-
 tablishment of a funding mechanism to deal with emergen-
 cy response activities. Connecticut holds the party respon-
sible for the hazardous waste or oilspill liability for up to
 1 Vi times the cost and expenses incurred by the state  for
the spill cleanup.
   Kentucky's 1980 law stipulated that applicants for haz-
ardous waste disposal permits must submit estimates of the
annual cost of  post-closure monitoring  and routine  main-
tenance, which will determine the annual  cash payments
to be made  by the disposer into a post-closure monitor-
ing and maintenance fund for  up to 20 years. The  appli-
cant must also provide evidence of financial responsibility
through liability  insurance, self-insurance or  other  evi-
dence for the purpose of satisfying claims arising out of
injury to persons or property resulting from the release or
escape of hazardous wastes into the environment.
  Massachusetts expects owners/operators  of a faulty ves-
sel, vehicle, railroad car, container or facility to promptly
notify the Division of Water Pollution Control of any spill
or face  a  fine of  up to $5,000. Persons who caused the
spill, and  all owners/operators of vehicles or containers
which released hazardous wastes are liable for the costs of
investigating and cleaning up the spill. Owner/operators of
hazardous waste disposal facilities in Michigan  must file
a surety bond or  other  suitable instrument to cover the
costs  of closing,  monitoring and maintaining a disposal
facility after  its capacity is  reached, based upon a rea-
sonable estimate of the costs over a 15-year period or less.
  New Jersey's Spill Compensation and Control  Act,
amended in  1979, stipulates that  dischargers or persons
responsible for  hazardous substances which result in spills
will be strictly,  jointly and severally liable and all com-
mon law defenses such as contributory negligence by the
government or a  third party liability claim will no longer
be reasonable excuses. Texas will attempt to collect reim-
bursement from the owner/operator and/or other parties
responsible for a spill and Wisconsin owners  of solid or
hazardous waste sites established after May 21, 1978, are
required to care for the sites for 30 years after operations
cease or 20 years if they pay higher fees into the waste
management fund.

Fund Balances

  Fifteen  states specifically establish maximum limits on
the amount of money to be collected for their funds, rang-
ing from $50,000 in the Kansas Pollutant Discharge Clean-
up Fund to $50 million in New Jersey's fund. The remain-
ing funds  are open-ended and do not specify a ceiling on
fund contributions.
  However, the fact that a state has legislatively author-
ized creation of a funding mechanism  to deal with haz-
ardous waste management programs does not necessarily
imply that money is  available in the  fund to carry out
its objectives. Although over half the states have deposited
monies into their funds—CA, CO,  CT, FL, GA,  IL, KS,
KY, LA, MD, MA, MI, MO, NH, NJ, NM, NY, TX, WI
—most of the contributions are meager in comparison to
anticipated revenues, and  would be  quickly depleted if
tapped for a major emergency response effort. Some states
(AZ, OH, TN) are waiting for authority to issue permits
for new (or existing) facilities to trigger their fund collec-
tion process.  Others  depend upon legislative actions to
appropriate needed funding (CT, FL, KS, MI, NH, NM,
NY, TX).  Only a handful of states have fee  systems in
place to generate fund revenues, or have taken  legal ac-
tions which have resulted in  fines being credited to their
fund accounts.
  The matrix in Chart  2 shows specific state funding
mechanisms, maximum limits on fund contributions and
actual fund balances as of July,  1981.
  Clearly, these states have begun to confront their haz-
ardous waste problems by establishing funding options to
meet the  financial  burdens associated with a hazardous
waste incident. Some funding programs are still in the in-
itial stages of development/implementation due to time-
consuming state regulatory processes or delays in obtain-
ing permitting authority from  EPA.  Other states have
forged ahead despite institutional and political barriers to
maintain sizable fund balances.  The states are now looking
to EPA for guidance and/or reassurance as to how their
funding  mechanisms  can  coordinate with  Superfund
efforts when and if emergency  response actions are to be
taken which  prove too enormous for the state alone to
handle.
POTENTIAL PROBLEMS IN STATE/
FEDERAL SUPERFUND COORDINATION

  The ultimate responsibilities for Superfund in adminis-
tration and implementation  should be afforded to the
states, provided technical capabilities and interest on their

-------
                                                                                                         STATE PROGRAMS      27
STATE

ALABAMA

ARIZONA

CALIFORNIA




COLORADO

CONNECTICUT

FLORIDA

GEORGIA

ILLINOIS

KANSAS


KENTUCKY

LOUISIANA



MAINE

MARYLAND

MASSACHUSETTS

MICHIGAN




MISSOURI

NEW HAMPSHIRE

NEW JERSEY
NEW MEXICO

NEW YORK

OHIO


TENNESSEE


TEXAS


WISCONSIN
NAME OF FUND

HAZARDOUS WASTE MANAGEMENT FUND

HAZARDOUS WASTE TRUST FUND

HAZARDOUS WASTE CONTROL ACCOUNT

SITE CLOSURE & MAINTENANCE REVOLVING
   ACCOUNT

EMERGENCY RESPONSE CASH FUND

EMERGENCY SPILL RESPONSE FUND

HAZARDOUS WASTE MANAGEMENT TRUST FUND

HAZARDOUS WASTE TRUST FUND

HAZARDOUS WASTE FUND

PERPETUAL CARE TRUST FUND
POLLUTANT DISCHARGE CLEANUP FUND

HAZARDOUS WASTE MANAGEMENT FUND

HAZARDOUS WASTE PROTECTION FUND
ABANDONED HAZARDOUS WASTE SITE FUND
EME?GENCV RESPONSE FUND

HAZARDOUS WASTE FUND

HAZARDOUS SUBSTANCE CONTROL FUND

SPILL FUND

WATER CLEANUP EMERGENCY FUND
DISPOSAL FACILITY TRUST FUND

HAZARDOUS WASTE SERVICE FUND

HAZARDOUS WASTE FUND

HAZARDOUS WASTE CLEANUP FUND

SPILL COMPENSATION FUND
HAZARDOUS WASTE EMERGENCY FUND

GOVERNMENTAL EMERGENCY FUND

HAZARDOUS WASTE FACILITY MANAGEMENT
   SPECIAL ACCOUNT

PERPETUAL CARE TRUST FUND
HAZARDOUS HASTE TRUST FUND

TEXAS COASTAL PROTECTION FUND
DISPOSAL FACILITY RESPONSE FUND

WASTE MANAGEMENT FUND
HAZARDOUS SUBSTANCES SPILL FUND
FUND CEILING (BY LAW)


NOT SPECIFIED


NOT SPECIFIED


NOT TO EXCEED AMOUNT NEEDED TO
ADMINISTER PROGRAM
NOT TO EXCEED $500,000



NOT SPECIFIED


$200,000 BY APPROPRIATION


UP TO $10 MILLION


NOT SPECIFIED


NOT TO EXCEED $25 MILLION


NOT SPECIFIED
NOT TO EXCEED $50,000


NOT TO EXCEED $6 MILLION


NOT SPECIFIED
NOT SPECIFIED
$1 MILLION


NOT SPECIFIED


NOT SPECIFIED


$300,000 BY APPROPRIATION


$ 25,000 BY APPROPRIATION
LIMITED TO $2 MILLION/YEAR,
   UP TO $30 MILLION
$1 MILLION


NOT SPECIFIED


NOT SPECIFIED


$50 MILLION, TO BE REDUCED TO
$26 MILLION AFTER 3 YEARS FOR
PORTION OF FUND USED FOR INCIDENTS
OCCURRING AFTER 4/1/77;  $6 MILLION
IN FY81 FOR ABANDONED SITE CLEANUP
UP TO $1,5 MILLION/SITE


NOT SPECIFIED


NOT SPECIFIED


$20 MILLION
NOT SPECIFIED
NOT SPECIFIED


NOT TO EXCEED $5 MILLION
NOT SPECIFIED


NOT TO EXCEED $15 MILLION
NOT SPECIFIED
FUND BALANCE

$0 (COLLECTIONS BEGIN JUNE 1981)

$0 (TIED TO SITING NEW FACILITY)

$628,500 (AVAILABLE FOR
             APPROPRIATION)
PHASED OUT BY 1981 LAW


$ 20,000 BY APPROPRIATION

$150,000

$600,000

$ 22,000

$400,000

$ 25,000
$ 50,000 IN FY81

$121,855

$0
$400,000
$1 MILLION   BY  APPROPRIATION

$0 (COLLECTIONS BEGIN  SEPT. 1981)

$348,000

$300,000 (+ $1  MILLION FROM BOND)

$ 25,000
$140,000

$500,000

$ ]2,000

$200,000 BY APPROPRIATION

$12 MILLION
$  5,000 BY APPROPRIATION


$9 MILLION


$0 (TIED TO FACILITY PERMITTING)
$0
$0
(TIED  TO  FACILITY  PERMITTING)
$1 MILLION
$5.6 MILLION BY APPROPRIATION
$180,000
$125,000
                                                                                                                          FIGURES)
                                                                 Chart 2.
                           State Funding Mechanisms: Ceiling Amounts and Fund Balances (as of July, 1981)

-------
28    STATE PROGRAMS
part is demonstrated. This state/federal partnership, how-
ever, precipitates the existence of major differences which
must be resolved in order  to achieve Superfund's objec-
tives. Issues to be resolved include:
•Preemption
•State match of funds
•Selection of contractors
•Definition of the state role
   Interpretation of Section 114 (C) of CERCLA, the pre-
emption clause, may impede successful state/federal fund
coordination efforts  and hinder promotion of a viable
state/federal partnership. To prevent affected industries in
a state from paying a double tax into both a state fund and
Superfund  for the same emergency response activity,  a
state may not require an industry to contribute to a fund
that would be used for the same purposes as CERCLA.
Section  114 (C)  fixes the  prohibition as  any fund "the
 purpose of which is to pay compensation for claims for
 any costs of response or damages or claims which may be
 compensated under this title." State general revenue funds
 may be used  and an industry tax  may  be imposed to
 purchase cleanup equipment, train personnel, or otherwise
 prepare for response to an accident. This will have far-
 reaching ramifications for states with a number of problem
 sites only a fraction of which will qualify under the act.
 Some officials at EPA candidly admit that CERCLA may
 only mitigate  half the number of sites in the  National
 Priority List of 400. Indeed, most of these 400 are con-
 centrated east of the Mississippi in states that are suffering
 from financial crises.
   At the same time, the responsibility for waste manage-
 ment programs is being shifted back to the states without a
corresponding shift in funding support. The capacity of a
state's general revenues to pay for  these environmental
programs, in addition to the site cleanup demanded by
residents, will  likely be severely strained.  There  is good
cause  for concern that a state with a number of sites elig-
ible to receive Superfund monies will not be able to finance
their portion of the cleanup that is needed and called for
by residents because of the prohibition on raising sufficient
funds  through an industry  tax. A narrower interpretation
either  by the courts or by Congressional action might avoid
preempting these states from recovering costs at those non-
Superfund sites utilizing a state-imposed industry tax.
   As  previously  mentioned,  some  states  already have
funds  financed by industry taxes to address hazardous
waste  damages. After CERCLA was enacted, other states
established  funds which have  tried to avoid  the con-
straints imposed by this preemption  clause by including
provisions in their laws that  monies cannot  be used to
compensate claims which will be covered by Superfund.
However, questions still exist as to how states will be able
to  finance their portion  of the cleanup costs, and which
costs may be preempted by Superfund.
   At present, eligibility requirements for Superfund have
not yet been clearly delineated.  States are looking to the
courts for direction. New  Jersey,  for example,  has filed
suit in Federal Court for  a declaration  judgment since
their state fund is financed  by transfer taxes on petroleum
and hazardous waste products and a $100  million bond
issue has been placed on the November ballot. In other
states, Indiana for example, the questionable issue of com-
pensable state costs is of prime importance, since their top
priority site is publicly owned and therefore requires a 50%
match  and long-term maintenance care. Draft strategies
developed by EPA require  the match (10% or 50%) to
be satisfied by a cash commitment. However, when costs
are totalled—previous  expenditures  for cleanup, state
match,  funds for sites  not  covered by Superfund—state
expenditures will probably exceed federal outlays, and that
10 or 50% match will, in actuality, be far greater.  States
just do not have those monies. They overwhelmingly agree
that state costs should be judged similarly to EPA eligible
expenditures. Rather than a  cash commitment, this would
include as eligible:
•state in-kind contributions
•long-term operation and maintenance costs
•past expenditures  for  partial or complete  remediation
 incurred after January  1, 1978
In addition, states feel that use of their own personnel will
prove more  expeditious and  cost-effective than use  of
contractors. This, in turn, will  not only reflect the true
spirit of the  law as Congress intended but also provides
incentive for  states to expand the scope of their own pro-
grams.
  Selection of prime contractors with their own employees
in supervisory capacities is  gaining strong support from
the states. Due to their long-standing experience in the util-
ization of contractors states feel that they can effectively
select and manage contractors cognizant of state specific
problems.
  EPA, however, has recommended to OMB that the Na-
tional Contingency Plan designate  the Army Corps of En-
gineers as the lead agency for design and construction at
Superfund sites, rather than regionalized prime contrac-
tors  managing a number of subcontractors.  Since  the
Corps  is a public agency and would  be accountable to
EPA, overhead costs  (of prime  contractors)  would  be
eliminated. However,  their lack of statutory authority,
personnel,  or past experience would create serious imped-
iments to rapid program implementation.
  Current program estimates by EPA anticipate the  use of
$120 million  over the next five years by states and the re-
maining $700 million to be used  by the Corps for their
clean-up activities. Such a projection is causing  alarm
within states. Concern  has  been expressed that state in-
volvement  in deciding  and implementing state clean-up
activities will be minimal, and that the extent and design
action implemented by  the Corps  will overlook solutions
tailored to specific state needs. All-in-all, the states fore-
see themselves as being  excluded from involvement in  lieu
of a federally dominated program.
  It is critical that optional procedures available to the
states for  participation  in Superfund be delineated  in the
forthcoming  National Contingency Plan. Provision for a
flexible approach with a range  of  options should  be
adopted and reflected in contracts or cooperative  agree-
ments between federal and state governments. Maximum
state participation is desired and provided for under the
law.

-------
                                                                                    STATE PROGRAMS     29
  There are certain issues yet to be resolved which are
currently barriers to expeditious Superfund implementa-
tion. It behooves EPA, therefore to address and consult
with the states on the following:
•Definition of roles available to the states
•Definition  of  fiscal  policies  and requirements:  allow-
 able costs,  credit/reimbursement policies, scope of con-
 tractual agreements
•Definition of response action objectives; criteria for cost-
 effective approaches
•Guidelines  in the event of disagreement concerning ex-
 tent of response actions
•State involvement  as far as technical requirements are
 involved, i.e., safety requirements and procedures, and
 technical assistance to states for training of local officials
 to assure long-term maintenance of Superfund sites
CONCLUSION

  As of August, 1981, the implementation strategy for
the Federal Superfund program has not yet been clearly de-
fined and those problems affecting the overall coordina-
tion  of Superfund with existing State funding programs
remain unresolved. The question posed in the title of this
paper—who will run the program—is clouded by these de-
lays, and therefore remains unanswered.
  However, daily communication with State Solid Waste
Directors and State emergency response actions under-
taken thus far indicate that States are not sitting by idly
and waiting for complete clarification of their role. States
are concerned with the health and welfare of their citizens
and  are expending  monies  when possible to  forestall
further degradation  and deterioration of selected aban-
doned hazardous waste sites. Funds are being  stretched
to their limits, however, and the States will welcome the
opportunity to cooperate with federal agencies and coor-
dinate on-scene  federal cleanup efforts when the Super-
fund program is finally operational.
  States are fully aware that a State/Federal  partnership
must be developed  to provide the adequate funding re-
lief necessary for Superfund activities  and had, there-
fore,  initiated lines  of communication with the EPA/
Superfund office soon after the  change in Administra-
tion. The Association of State and Territorial Solid Waste
Management Officials, likewise, has established a Super-
fund Task Force to work directly with EPA  in this role
definition.
  However, even with these  combined efforts,  the Con-
gressional mandate  that provides  a State  role in Super-
fund and the new Administration's policy  efforts to re-
turn responsibility to the States, a real ambiguity exists
as to whether the States will actually have a substantive
role under CERCLA and what that role will demand.

REFERENCES

 1. Chemical Manufacturers Association,  "Recommen-
    dations of the Chemical Manufacturers Association,"
    Washington, D.C., June 30, 1981.
 2. DeVille,  W.B., "Action Plan: Superfund Implemen-
    tation, The Role of the States," Baton Rouge, March
    18,1981.
 3. Personal Communications with State Solid Waste Di-
    rectors.
 4. Simcoe,  B.T., "Hazardous Waste Management: A
    Survey of State Laws 1976-1980 (Update)", National
    Conference of State Legislatures, Washington, D.C.
    1981.
 5. U.S. Code, Congressional and Administrative News,
    96th Congress, Volume 2, Legislative History, 94 Stat.
    2767-2811, "Comprehensive Environmental Response,
    Compensation and Liability Act of 1980," West Pub-
    lishing Co., St. Paul, 1981.
 6. U.S. Congress, Public Law 96-510, "Comprehensive
    Environmental Response, Compensation and Liabil-
    ity Act of 1980," Washington, D.C. 1980.
 7. U.S. Congress, Public Law 94-580, "Resource Con-
    servation and Recovery Act of 1976," Washington,
    D.C.1976.
 8. U.S. District Court for the District of Columbia, State
    olfNew Jersey et al., vs. The United States of America,
    Civil Action No. 81-0945, "Memorandum of Points of
    Law and Authorities in Support of the United States'
    Motion to Dismiss," July, 1981.
 9. U.S.  Environmental Protection  Agency,  "Interim
    Superfund  Removal Guidance," Washington, D.C.,
    1981.
10. U.S.  Environmental Protection  Agency, "Recom-
    mended  Superfund  Strategy," Washington, D.C.,
    April 10, 1981.
11. U.S. Environmental  Protection Agency, "Resource
    Analysis, Alternative  Responsibility Assignments for
    Accomplishment of Remedial Action Planning, Rem-
    edy   Selection,  Design  and  Construction  Under
    CERCLA," Washington, D.C., 1981.
12. U.S. Environmental Protection Agency,  "State  and
    Local  Representatives'  Superfund   Implementation
    Strategy Meeting Minutes," Washington, D.C., March
    6,1981.

-------
  RECORDS MANAGEMENT SYSTEMS AS UTILIZED BY THE
   TEXAS DEPARTMENT OF WATER RESOURCES AND THE
 UNITED  STATES ENVIRONMENTAL  PROTECTION AGENCY
                                        WILLIAM H. HUPP
                                Texas Department of Water Resources
                                            Austin, Texas

                                      JO JOHNSON-BALLARD
                           U.S. Environmental Protection Agency, Region 6
                                            Dallas, Texas
INTRODUCTION

  In an  effort  to  effectively deal  with  the  problems
posed by uncontrolled hazardous waste sites prior to the
implementation of the National Hazardous Waste Man-
agement program, the United States  Environmental Pro-
tection Agency (EPA)  developed and implemented the
Hazardous Waste Site  Enforcement and Response Sys-
tem. Although not specifically mandated by federal legis-
lation, development was based upon regulations and grant
provisions issued in conformance with the: (1) Resource
Conservation and Recovery  Act (RCRA)  of  1976, (2)
Safe Drinking  Water Act of 1974,  (3) Federal Water
Pollution Control Act of 1972, and  (4) the Comprehen-
sive Environmental Response, Compensation, and Liabil-
ity Act (CERCLA) of 1980, or "Superfund".
  The system was designed for use by regional EPA offices
and the states for records management purposes in the
identification and screening of potentially hazardous waste
sites and in  the  management of resulting enforcement
and remedial actions.'" To verify the need for such a sys-
tem, various  studies performed for the EPA's  Office of
Solid Waste indicated that there were between 32,000 and
50,000 hazardous waste dump sites throughout the United
States. Few of  these sites had been  inventoried and the
risks posed by them were unknown.<2)
  In comparison, the State of Texas  also has its share of
potentially hazardous waste disposal sites. According to
Paul Sweeney in an article in The Texas Observer, of~rhe
twenty largest chemical corporations in the United States,
nineteen have manufacturing operations in Texas. The
largest 53 chemical companies have  159 plants in Texas
and over half of the nation's manufacturing capacity for
the petrochemical industry is located along the upper gulf
coast, from the Louisiana border south to Freeport. Addi-
tionally, the state's petroleum refineries, utility power gen-
erating stations, rubber and tire manufacturing plants and
copper and  lead smelters  also  contribute  enormous
amounts of waste.
  During 1978, the Texas chemical industry produced 3.9
million tons of industrial solid waste, which was approx-
imately ten percent of the nation's total. This large quan-
tity places Texas sixth among the states in yearly output of
chemical waste.'3' Also in 1978, Texas  businesses generated
approximately 60 million tons of Class I industrial solid
waste;'4' "Class I" is the ranking for the most dangerous
of the industrial wastes.(5) Therefore, because of the exten-
sive manufacturing performed in Texas and the resultant
massive amounts of waste, Texas has great potential for
both reported and unreported hazardous waste problem
sites.
  In support of  the  EPA Hazardous Waste  Site En-
forcement and Response  System  and related records
management system, the State of Texas is currently con-
ducting a statewide Imminent Hazard Survey. The survey
of uncontrolled hazardous waste sites  encompasses pre-
scribed EPA  procedures and decision-making points.
Procedural steps are supported by records management
forms designed to assure a common vocabulary and pro-
cedural standardization  relative to the acquisition, re-
cording and reporting of information.
  The purpose of the state survey is to assist the EPA in
identifying, evaluating and initiating corrective actions at
potentially hazardous  waste disposal  sites which present
an imminent hazard to  public health  and the environ-
ment.  In this paper,  the  authors describe the current
records management system utilized by the Texas De-
partment  of Water Resources and discuss the overall EPA
records keeping  system  as  it relates to uncontrolled
hazardous waste sites within the state. General background
information, system differences, similarities and prob-
lems will also be presented.

TEXAS SOLID WASTE MANAGEMENT

  Prior to the enactment of the Texas Solid Waste Dis-
posal Act in 1969,  there were no  effective state regula-
tions governing the disposal of solid wastes. In many cases,
waste was dumped indiscriminately without records or en-
vironmental safeguards. With the passage of the Act, the
61st  Texas Legislature assigned jurisdiction of industrial
solid waste management to the Texas Water Quality Board
which is now the Texas  Department of Water Resources
(TDWR). The Texas Department of Health (TDK) was
given responsibility  for managing municipal  solid waste
and  any  industrial solid  waste mixed with  municipal
waste.(6)
  The  dual jurisdiction over the management  of solid
waste results in both departments (TDWR and TDH) hav-
ing authority to carry out the state's Imminent Hazard
                                                  30

-------
                                                                                     STATE PROGRAMS    31
Survey of industrial and municipal uncontrolled hazardous
waste sites.

IMMINENT HAZARD SURVEY

  The state Imminent Hazard Survey which deals with se-
lected uncontrolled industrial hazardous waste sit* is con-
ducted by the Enforcement and Field Operations Division
of the TDWR. The division maintains thirteen district of-
fices throughout the state (Figure 1) with a combined staff
of approximately 22-25 specially  trained field personnel.
Designated field personnel with backgrounds in biology,
chemistry and geology are solely responsible for  the ac-
quisition of information on the hazard potential posed by
each industrial hazardous waste site evaluated. The evalua-
tion of each site is based upon information obtained from
existing files within each district office and by one or more
on-site inspections.
                        Figure 1.
                TDWR District Field Offices

  An evaluation, as mentioned throughout this presenta-
tion, will refer to the following prescribed sequence of
steps or events which must be followed by field personnel
according to EPA procedural requirements:
•Identification information is obtained and submitted for
 all sites (assigned or unassigned)
•A preliminary assessment to determine potential hazard
 is conducted
•An on-site inspection is performed—depending upon pre-
 liminary assessment results
•Disposition in terms of remedial or enforcement measures
 are determined (optional)
•A final strategy is determined and initiated (district and
 central office effort)
  All acquired information, which includes  identifica-
tion data, is recorded on EPA management forms and sub-
mitted to the division's central office. Each form is ex-
amined for completeness, reproduced for central office use
and originals transmitted to the PEA Region VI office for
review and computer input. All  forms are completed by
district personnel with the exception of the Final Strategy
form(s).  These  are completed by the  central  office en-
forcement staff after  consulting  with district offices and
examining all relevant information.
  Since commencement of the TDWR Imminent Hazard
Survey in December 1979, the number of listed TDWR
sites increased from 148 to approximately 290. The in-
crease from 148, initially identified and selected from the
Eckhardt list,(7) is  attributed to input from the following
sources:
•EPA and TDWR Citizen Complaint Program
•Readjustments of assigned sites (TDK, EPA and TDWR)
•The Wapora and Moffett surveys
•Incidental sites (provided by district personnel)
•Aerial reconnaissance activities
  The discovery and listing  of  additional uncontrolled
hazardous waste sites is also made possible by centralized
coordination of inspections and review of RCRA Open
Dump Inventory results. Should inspection results indi-
cate the  presence of hazardous waste,  the site or facility
is subsequently listed for evaluation under the Imminent
Hazard Survey.
TDWR RECORDS MANAGEMENT SYSTEM

  The unanticipated additional records keeping require-
ments placed on the initial records management system,
by  the rapidly increasing  number of  assignments, re-
sulted in  duplication of effort and operational ineffic-
iency. To resolve these problems, changes were designed
and initiated to enlarge and improve records keeping and
site tracking capabilities and to reduce duplication result-
ing from  assignment confusion. This was  accomplished
by:
•Replacing the existing index card format used for records
keeping purposes  with a more flexible data sheet type
format (Facility Status Sheet)
•Verifying the list of assigned sites
•By coordinating and directing field activities
  Although identical to the EPA  Records Management
System in terms of content and purpose, the TDWR sys-
tem differs in data handling and operational procedures.
In addition, while EPA records keeping relies on an auto-
mated system, the TDWR  system  is not automated and
is based  on the use of a facility status sheet  (Figure 2)
which requires  manual input of information and  main-
tenance.
  Prior to operational use  of the facility status  sheet,
all sites listed for evaluation by the TDWR were examined
for correct assignment. This  effort, with assistance pro-
vided by the EPA  regional office and a contractor (Ecol-
ogy and Environment, Inc.), resulted in a much  more
comprehensive  and error free listing. The  resulting list,
distributed to survey participants  for  comparative pur-
poses, provided a data base for the facility status sheet.

-------
  32     STATE PROGRAMS
Haz.










Facility





















Log





















Inspection










Disposition










Strategy 1










2










Enforcement
Documentation










Misc.
Inf.










                                                       Figure 2.
                                                  Facility Status Sheet
   The facility  status sheet alphabetically lists all sites
 assigned to the TDWR for evaluation, provides  a site
 tracking capability and  also helps to prevent duplicate
 evaluations of the same site by TDWR and other survey
 participants (TDK and  EPA).  Additionally, the  status
 sheet  aids  in assigning field evaluations relative to the
 survey,  and provides a ready index to site information of
 benefit to other division programs and services.
   The facility status sheet offers a more up-to-date activ-
 ity status check capability for site tracking than the EPA
 computer printout. Recorded information is rapidly up-
 dated  and new  sites can be immediately entered into the
 TDWR/EPA systems following a check with the EPA reg-
 ional  office to  avoid duplicate  entries.  The amount  of
 time required by  the EPA system  to  generate  a  com-
 puter printout reflecting updated or new site information
 is usually three  to four weeks. Upon request, the facility
 status  sheet can be updated and submitted within 2 to 3
 days. Completed management forms submitted to the reg-
 ional office are  accompanied by an updated facility status
 sheet.
   Coordinating  and directing field activities  to eliminate
 confusion in the acquisition and  recording of appropriate
 information, and thereby improving efficiency, is accom-
 plished by  the  use of an assignment worksheet.  Each
 worksheet contains a list of sites to be evaluated within
 each district and  the required management  form  to  be
completed and submitted for each entity (Figure 3). Work-
sheets are distributed to each district on a quarterly basis
and the provided guidance assures the proper completion
and submittal of information as prescribed by EPA guide-
lines. Worksheet  assignments are determined  from the
facility status sheet.
  As an  index to available survey information on file, the
 facility  status sheet often  benefits other division  busi-
ness matters.  The Citizens  Complaint  Program (indus-
trial), superfund candidate site selection and enforcement
activities draw upon information contained in the site in-
spection forms  for background information,  updating,
and for general information.

UNCONTROLLED HAZARDOUS
WASTE SITE PROGRAM

  The EPA  uncontrolled  hazardous waste site program
(officially known as the Hazardous Waste Site Enforce-
ment and Response System),  implemented in  1979, has
operated under a variety of names throughout the Region
VI five state area (Figure 4): (1) the problem sites program,
(2) the uncontrolled hazardous waste  sites program, (3)
the potential hazardous waste sites program and (4) the
imminent hazard survey.
  The records management system, an integral part of the
Hazardous Waste  Site Enforcement  and  Response sys-
tem, was developed with assistance from the Hazardous

                   DISTRICT	
    FACILITY/SITF NAMF

    A-l Lee oil - La Porte


    Associated Properties - Sorinq


    Baytown Tunnel Site  Houston


    9ryan Mound Tar Pits - Freeport


    Church Road site - Houston





    Gulf Coast Disposal - Baytown


    Hall street Site - Diclcinson
    Nine Bar Ranch - El Paso
REQUIRED EPA FOBH

Insoection


Assessment


Assessment


Inspection


Inspection


Inspection


Assessment


Assessment


Inspection


Assessment
                       Figure 3.
              TDWR Assignment Worksheet

-------
                                                                                     STATE PROGRAMS    33
     	%-a-	-a—...-	
                       Figure 4.
                     EPA Region VI

Waste Enforcement Task Force (now known as the Office
of Waste Programs Enforcement), EPA Oil and Special
Materials Control Division, National Enforcement Inves-
tigations Center, and the EPA regional offices.*8'
  The system, as currently used in Region VI, involves the
continuous compilation and computer input of informa-
tion received from the states on uncontrolled hazardous
waste sites. Information submitted by the various states to
the regional office is conveyed via Hazardous Waste Site
Enforcement and Response System management forms.
These forms consist of:
  (1) the potential hazardous waste site identification form
  (2) tHe preliminary assessment
  (3) the site inspection report
  (4) the tentative disposition
  (5) the final strategy determination form(9)
  Each potential site  identified and tracked by the com-
puterized records keeping  system requires the completion
and input of every management form. This requirement,
though not always timely,  allows for the complete evalua-
tion of every listed site.
  Of particular concern to Region VI records keeping ac-
tivities is the problem stemming from the dual jurisdiction
in the State of Texas over waste management. As previous-
ly mentioned, the TDWR is authorized the management of
industrial solid waste  and the TDH, the management of
municipal wastes.  Although  Region VI  deals with  both
Texas  state waste management authorities,  the greater
amount of communication is with the TDWR. This is due
to the fact that the TDWR retains authority over most of
the uncontrolled hazardous waste sites in Texas, and has
consequently been assigned  a greater percentage  of sites
for evaluation.
  The uncontrolled hazardous waste sites records manage-
ment system in Region VI began with approximately 467
listed potential sites of which 259 were in Texas alone.
The ratio of Texas sites  to  the number of  sites in the
region has remained somewhat constant since initial imple-
mentation of the records management system (Figure 5).
There are currently 1,600 potential uncontrolled hazardous
waste sites listed in the Region VI  five state area with
approximately 700 assigned to the State of Texas.
  A major portion of the initial group of listed sites in
Region VI were derived from the Eckhardt Subcommittee
listing. Originally listed were 542 sites  in the region; the
number was reduced  to 425 after duplicate listings and
other discrepancies were eliminated.
  The Eckhardt Subcommittee listing represented the re-
sults of a survey dealing  with the process waste disposal
practices (since  1950)  of  the 53  largest chemical com-
panies in the United States.00' Survey  results were made
available to the EPA, and the sites identified in Region VI
were incorporated into the records management system.
Additional sources of information from which sites were
added to the list included the Wapora listing, citizens com-
plaints, and EPA, state or "other" entity identification.
  Information on newly found  uncontrolled sites is sub-
mitted on the potential hazardous waste site forms. Each
potential waste site is assigned its own identifying haz-
ardous site number (HAZSIT) and the information is then
entered on computer coding sheets. This procedure allows
for periodic updates of the computer  listing  of uncon-
2000 ,
1800 •
1600 ••
1400 •
1200 - •
1000 ••
 800 ••
 600 ..
 400 .
 200-.
         EPA TDWR
          1979
EPA  TDWR
  1980
EPA  TDWR
  1981
                       FigureS.
             Ratio of Texas Sites to EPA Sites

-------
34     STATE PROGRAMS
trolled hazardous waste sites. This practice was initially for
the benefit of the Enforcement Division's Hazardous Ma-
terials Enforcement Section and as the number of potential
uncontrolled waste sites in Texas  increased, the amount of
incoming information did also. The increase in site infor-
mation led to a need for greater Agency  accountability
for its own, as well as the state's purposes.  As a result,
the Region's  Hazardous Materials Enforcement Section
decided that the incoming state site-related data should be
shared with the states, and a computer program designed
to generate current  information  for each site was  devel-
oped.
   Computer output  includes the identifying hazardous site
number,  site name  and location, actual dates of com-
pletion  for preliminary  assessments,  site  inspections,
tentative dispositions, and final  strategy determinations,
as well as other pertinent site-specific information. This
listing could reveal,  upon quick examination, the  status
of the site investigation, the degree of hazard at the site,
as  well  as how the  site had been originally identified
(Figure 6).

Site Assignment

   In  order that all sites identified could be properly eval-
uated,  a certain number of sites  were assigned to each
                                                 responsible agency: The EPA, the TDWR, and the TDH.
                                                 Assignments were based upon differing jurisdictions each
                                                 respective  state agency had over various waste types (as
                                                 previously noted).  The  EPA was assigned federal facil-
                                                 ities, sites which the Agency had already begun work on,
                                                 and cases that the state felt could prove to be too political-
                                                 ly volatile to pursue. From this point, sites were  added to
                                                 or taken from each specific agency's designated list of sites
                                                 until all the  uncontrolled  hazardous waste sites had been
                                                 assigned and all agencies involved were  satisfied with the
                                                 site designations.
                                                   Of particular concern to listing procedures was the prob-
                                                 lem dealing  with the reporting (citizen  complaint) of the
                                                 same site to  both the state and EPA. For example, a cit-
                                                 izen complaint could be called in to the Dallas EPA of-
                                                 fice, the information  recorded on a potential hazardous
                                                 waste site form, and the site information entered into the
                                                 records management system.  If the site had no readily
                                                 available information, the contract team for EPA sites in-
                                                 vestigation  (Ecology  and  Environment,  Inc.)  would
                                                 schedule a reconnaissance (fact-finding) inspection at the
                                                 site. A great deal of confusion ensued  in the event that
                                                 the state had received  the  same complaint, TDWR would
                                                 likewise schedule a site inspection.  In  some  cases, visits
                                                 to the  same  site by the  two agencies would be  separated
                                                 by only one week.
                                      UNITED STATES ENVIRONMENTAL PROTECTION  AGENCY
                                                     REGION 6
                                               HAZSIT SYSTEM -- 24
                                        HAZSIT SITES SORTED BY   STATE/SITE NAME
  HAZSJJ__

  TX02526



  TX02534



  TX04545



  TX04BI?



  TX06491




  H04B04


  TX0254J



  TX05428
           SITE NAME
              LOCATION
A-l Garbage
Ois;i. Trash SV
San Antonio

A-l Lee Oil
SVC,  Inc.
La Porte

AAA Wire
Prod. , Inc.
Point

ABC Externi-
nat ing Co.
Beaumont

ACF Indus. , Inc.
Shippers Car
Line Division
lonqvtew

Acock Lab. , LTD
Austin

Addlks rairbk
R S Co.
Houston

Aerial Survey
By Toler
Mineral  Wells
                 ENF
                 ENG
RWB



RWB



LOW



RWB




RWB


RWB



RWB



RWB
         SOURCE
         OF SITE
         IDENT.
STA
CIT
STA
          STRAT
          CODE
SEHIOl'S
COUE
PRELIM      RECON
ASSESS      INSPEC
DATE        DATE
SAMPL      TENT      FINAL
INSPEC     DISPO     DISPO
DATE       DATE      DATE
                   TDH
          TDWR       M
          EPA       M
EPA        EPA
                                    CIT
          TDWR       M


          EPA



          TDH



          EPA       L
                               3/4/81
                               3/21/80
                               1/21/81
           9/29/80
                                3/27/80
                               9/5/80     9/5/80
                                                                   10/16/80
                                                                       12/3/80   12/3/80
                                                      Figure 6.
                                             EPA Computer Output Listing

-------
                                                                                    STATE PROGRAMS    35
  In order to prevent this unnecessary duplication of agen-
cy efforts,  a system to screen all citizen complaints re-
ferred to the agencies was established. Whenever a com-
plaint reaches the EPA office, the TDWR contact per-
son is called and if the complaint is initially made with
the TDWR, the EPA contact is  called. In this way, any
available site information is shared between the agencies.
If the state office is in the process of evaluating the site,
the "lead" on the site is assigned to TDWR for the com-
pletion  of  site investigation activities.  If, however, the
EPA office had previously begun work on the complain-
ant's site, the site would be placed on the Agency's sites
listing for additional  investigation. The development  of
this sites assignment system kept agency effort duplication
from recurring. Communication  with the TDWR is en-
chanced, and much valuable knowledge concerning uncon-
trolled hazardous waste sites complaints  in Texas  is ex-
changed.

Submittal and Review of Management Forms

  Uncontrolled hazardous waste sites tracking informa-
tion submitted by the State of  Texas  was at first sent
through the Region's Air and Hazardous Materials Divis-
ion (Solid Waste Branch) before being sent to the Enforce-
ment Division for review. The  purpose of this was to as-
sure that the  TDWR could account for  the  number  of
sites the agency was committed  to evaluate and inspect
in its RCRA grant provisions.  The forms were mailed to
the Texas RCRA project officer, then were forwarded to
the  Hazardous  Materials Enforcement Section of the
Enforcement Division for review and input into the sys-
tem. The preliminary assessments, tentative dispositions
and  final strategy determinations completed  by the  state
and  by the EPA contract team were evaluated  for com-
pleteness and coherency by the Section's engineers.
  Those records management forms completed by Ecology
and Environment Inc. that were lacking information were
returned to  the  Region's  Surveillance  and  Analysis
Division for referral and additional input from the Ecol-
ogy and Environment personnel. Those forms completed
by the state that did not contain sufficient information to
make a valid site determination were returned to the state
through the Solid Waste Branch  (again, for record-keep-
ing purposes). This method of site forms  transmittal was
inefficient, and at times, very slow.
  The Texas RCRA project officer was contacted by the
TDWR  and  by the Hazardous   Materials  Enforcement
Section's contact persons to discuss a new technique for
forms submittal.  They agreed  that the incoming uncon-
trolled hazardous waste sites forms would be mailed di-
rectly to the section responsible  for their review, with a
copy of the accompanying letter stating  the  amount  of
site  work performed simultaneously forwarded to  the
Texas RCRA project officer.
  All completed inspection reports are reviewed by the
Region's Surveillance and Analysis Division and are di-
rected to the Enforcement  Division for any necessary
action. It is at this point when  the decision  is made  to
either pursue enforcement action, remedial action  or  to
take no action at the site in question. This judgment is
based upon results of samples taken at the site, as well as
past regulatory actions and the actual situation of the site
(i.e., geologic and hydrologic considerations,  proximity
to population, etc.).
  If enforcement action is to be taken at  the site, the ac-
tion is initiated after the  completion  of  a  tentative dis-
position (either by the  state or EPA)  indicating that de-
cision. The file information included in the  records man-
agement system is  considered during the  construction of
a Case Development Plan (CDP).
  The  CDP is a guide used by the Agency in pursuit of en-
forcement action to assure the collection  of information
and interpretative analysis for  use during anticipated en-
forcement proceedings.(11> In order for a valid site determ-
ination to be made, all of the system's management forms
have to be fully and correctly completed. The records man-
agement system's reason for existence is  to provide the
basis from which these site determinations can be derived
and then pursued.

ADDITIONAL USES FOR
HAZARDOUS SITES INFORMATION

  The  need for accuracy in the records management track-
ing system can be  substantiated by the utilization of the
system by the RCRA Permits program,  as well as  the
"Superfund" program. The uncontrolled hazardous waste
sites program has provided information to  the RCRA pro-
gram before that system was able to generate its own data
base.
  It is thought that  those sites that are not included in
the RCRA Permits program would fall under the purview
of the  Superfund.  Although the  CERCLA  will be  more
selective in its use  of information made available by the
records management system, the information involving the
site's consideration for remedial aid must be accurate and
of sufficient value for use in the immediate  future.

CONCLUSION

  The  future for the uncontrolled hazardous waste  sites
program, at best,  appears uncertain.  The program was
ranked fourth (and lowest) priority  for the state RCRA
Hazardous Waste Programs by EPA guidance for  fiscal
year 1982 RCRA grants. Regardless of the  fact that the
number of uncontrolled hazardous waste sites increased
nationwide over a  one  year period (from December  31,
1979 through December  31, 1980) from 4,098 to 8,677, it
appears as though  Superfund allocations alone will be
forced  to provide resources for continuing site investiga-
tion and evaluation. The information base established by
the uncontrolled hazardous waste sites program records
management system has been utilized by at  least two on-
going EPA programs and  may also be used to determine
EPA resources necessary to carry out the future cleanup
and enforcement activities at these sites.
  Despite additional funding provided by  the Superfund,
the amount of support available  for the investigation of
increasing numbers of uncontrolled hazardous waste sites

-------
36    STATE PROGRAMS
is  unknown.  The CERCLA called for the revision of
the National Contingency Plan (pursuant to section 311 of
the FWPCA), a segment of which includes a National
Hazardous  Substances Response Plan. The plan would
encompass, among other things: methods for  discovery
and  investigation of facilities,  methods  for evaluation
and remedy of releases from facilities, methods and cri-
teria for determining the  appropriate  extent of response
measures and criteria for prioritization  among releases for
remedial actions at sites."2' Not until this plan is released,
however, will the impact of Superfund  resources upon the
uncontrolled  hazardous  waste  sites  program be  de-
termined.
REFERENCES

 1. Blum, B.,  "Hazardous Waste Site Enforcement and
    Response  System",  Internal EPA  Memorandum,
    November 9, 1979.
 2. U.S. Environmental Protection Agency, Office of Re-
    search and  Development, "Research Summary—Con-
    trolling Hazardous Wastes", EPA-600/8-80-017, May
    1980.
 3. Sweeney, P., "One Man's Waste...," quoted in Texas
    Legislative  Committee on Environmental Affairs Re-
    port,  Subcommittee on  Hazardous Waste, October
    1980.
 4. TDWR, "The Texas Department of Water Resources
    Industrial Solid Waste Regulatory Program," quoted
    in  Texas Legislative Committee  on Environmental
    Affairs, Subcommittee on Hazardous Waste, October
    1980.
 5. Texas,  "Solid Waste Disposal Act," Texas Revised
   Civil Statues Annotated,  (Vernon), art. 4477-7, sec.
   2(15).
 6. TDWR, "Draft-Solid Waste Management Plan for
   Texas Volume 2  Industrial Solid Waste," November
   1980.
 7. U.S., Congress,  House, Subcommittee on Oversight
   and Investigations, "Waste Disposal Site Survey," Oc-
   tober 1979.
 8. Blum, B., op. cit., November 9,1979.
 9. Miller, J.G., "Management of the Hazardous Waste
   Site Enforcement and Response System and a Defi-
   nition of Terms Used," Internal EPA Memorandum,
   August 7, 1979.
10. U.S., Congress, op. cit., October 1979.
11. Blum, B.,  "Identification and  Assessment of  Haz-
   ardous  Waste Sites," Internal  EPA  Memorandum,
   October 3,1980.
12. U.S.  Comptroller  General's Office,  "Hazardous
   Waste sites Pose Investigation, Evaluation, Scientific,
   and Legal Problems Report," April 24,1981.

-------
                          THE CALIFORNIA SUPERFUND
                                              PETER H. WEINER
                                        Special Assistant to the Governor
                                          for Toxic Substances Control
                                             Sacramento, California
INTRODUCTION

  California has just passed the most comprehensive pro-
gressive superfund legislation in the country.  The Cal-
ifornia legislation  provides  for  victims' compensation,
for reimbursing local agencies for their costs and for clean-
ing up California problems well beyond what the federal
superfund might pay for.  Ironically, after a great strug-
gle, the legislation that was passed was that sponsored by
industry, not by government or  environmentalists. It is
therefore all the more noteworthy that the $100 million the
fund will receive over the  next ten years is entirely from
industry-paid fees on hazardous waste disposal.
  The spectre of abandoned  hazardous waste dumps and
toxic spills now haunt the people of this country. The pos-
sibility of discovering hidden buried wastes  or of  being
assaulted by the contents  of an overturned truck a mile
away  is real,  based on a feeling of powerlessness and of
being violated by unknown  sources. In  many  ways, the
outrage expressed by many citizens is akin to that accom-
panying the fear of burglary  and  other crime. Despite the
fact that occupational exposures  remain the  number one
toxic substances problem throughout the world, including
California, spills and abandoned  hazardous waste dumps
remain  the top political priority because they affect a
broader socioeconomic range of citizens and appear to be
a  less expected hazard. Because the California  exper-
ience  with this problem  appears  similar to that of  many
other  states, it  may be  useful to review the history and
details of California's new superfund legislation.
BACKGROUND OF THE CALIFORNIA SUPERFUND

  California is one of the largest users of hazardous ma-
terials  and producers of hazardous wastes in  the coun-
try. Since the  advent of the Brown Administration,  Cal-
ifornia has acquired a reputation for having the best haz-
ardous waste  control program  in  the country. The Re-
source Conservation and Recovery Act is patterned prin-
cipally on California legislation, including the cradle-to-
grave concept  of tracking wastes. Because California in-
dustrialized later than the northeast and has long used rela-
tively stringent regulatory controls, the state has not been
subject to the same historical horror stories as some east-
ern states.
  However, as Congress studied abandoned waste sites as
a prelude to enacting federal superfund, California took
similar action to search out abandoned and uncontrolled
waste sites. After surveying a small number of counties
with heavy industry, constituting perhaps one-third of the
state's hazardous waste problem, state researchers  sur-
prisingly found 67 sites which needed attention because of
their potential for adverse affects on public health or the
environment.  Some of these sites required costly remed-
ial measures.  For  example, the General Electric site in
Oakland, with a deep pool of PCB's under the ground,
will cost as much as $30 to $60 million to clean up.
  Altogether, it was estimated that the clean up of the 67
sites found so far  would cost up to $330 million. More
significantly,  the state  found that  at  19 of these sites,
whose total estimated cleanup cost was $57 million,  it
was doubtful that those liable for the problem would have
sufficient resources to clean up.
  The sheer number of these sites,  combined with  sub-
stantial publicity about a few problem sites in Southern
California,  generated substantial concern  regarding the
adequacy of the federal superfund to clean up these sites.
The location and distribution of these problems also play-
ed a role in generating widespread concern.
  The Stringfellow Quarry, located in a quasi-rural  area
near Los Angeles, had long been thought to be California's
one major problem. However, during the fall of 1980, it
became apparent that upper middle class residents in the
conservative Republican suburb of  Fullerton were  pro-
foundly upset  by the odors emanating  from the McColl
dump site, which was literally in their backyards. A mas-
sive  state investigation  showed that  residents  suffered
nausea, headaches  and other similar acute symptons, but
that  the amount of hazardous constituents  polluting the
air from the dump site  were so minor as to render proof
of chronic  harm very difficult. Despite the inability of
scientists to prove  the risks involved, the articulate  resi-
dents of the  community continued to  press their  case
for removal or other elimination of the hazard.
  At the same time, residents near a  dump site in the
congressional district of Barry Goldwater, Jr. were  sim-
ilarly upset, eventually causing the closure of the dump site
near them due to earthquake concern. And, in the inner
city near East Los Angeles, one irresponsible waste dis-
poser was  shut down in a flurry  of  litigation and an-
other abandoned site of 18,000 drums was discovered.
                                                      37

-------
38    STATE PROGRAMS
The abandoned site,  in the small industrial city of Santa
Fe Springs, eventually exploded, and the resultant efflu-
ent in the flood channel killed over 250,000 fish in the
Pacific Ocean near Los Angeles.
  While these well publicized stories occupied the media,
local governments were increasingly concerned about the
high cost of responding to emergencies involving highway
spills, industrial fires and other toxic chemical emergen-
cies. Data from one small department  of environmental
health in a relatively less populated county (San Bernar-
dino) showed expenditures of $33,000  in a nine month
period for  personnel and laboratory costs. The costs of
fire fighting and law enforcement agencies were substan-
tially higher.


 POLITICAL HISTORY

   Responding to the problems at  Stringfellow, McColl,
and the many emergency spill sites in the state, the Brown
Administration in February of 1981 prepared draft legisla-
tion which was later introduced by state Senator Presley
as SB 788.
   As originally drafted,  SB 788 would have provided a
fund of up to $35 million through a tax structure iden-
tical to that of the federal superfund. Among other  fea-
tures, the bill declared that the generation, transportation
and disposal of hazardous wastes is an ultra hazardous ac-
tivity, thereby creating a standard  of strict and absolute
liability for all releases.
   Industry was  alarmed  by the draft  of  SB 788,  and
immediately introduced its own bill, SB 618. SB 618 called
for 100% general fund contributions by the state to match
its federal superfund obligations. The bill provided for no
other authority or expenditures.
   After both bills passed out of their first policy commit-
tee, aborted attempts were made to work out a compro-
mise. Intent on pushing its own bill, the petrochemical in-
dustry then began a long process  of amending SB 618.
Interaction between industry and the Governor's Office,
with occasional  input from environmentalists and local
government, eventually produced SB 618 as an acceptable
compromise for cleaning up California problems.
   The eventual compromise reflects most of the provisions
originally contained in the Administration's SB 788, with
one notable exception. Industry was successful in insist-
ing that no new standards of liability be enacted and that
reliance instead be placed on existing standards of liability,
such as those contained  in Section 107 of the  Compre-
hensive Environmental Response, Compensation, and Lia-
bility Act  (CERCLA) of  1980, (42 USC Sec.  9601), or
various provisions of state law.

THE CALIFORNIA HAZARDOUS
SUBSTANCE ACCOUNT ACT

Tax Structure and Funding

  The Act, which is effective  immediately, appropriates
$2 million to a new hazardous substance account, which
is  to be administered by the Director  of the  State  De-
partment of Health Services (Health and Safety Code Sec.
25330(1)), subject to overall site response coordination by
the Governor (Sec. 25355).
  The total tax imposed by the Act is equal to $10 million
per year, minus the existing unobligated balance in the ac-
count as of a specified date. The Board of Equalization
sets the tax rate, according to the following formula, to
raise that amount of money. The tax rate  is as follows:
Base tax rate =
                    $10,000,000-M
                     W2 + 2W3 + 0.01W4)
where:
(1)  "M" is the existing unobligated balance in the fund on
    December 31 of the immediately preceding year.
(2)  Wj is the total amount in tons of waste and material
    disposed of  in California,  the federal regulation of
    which has  been suspended  under the  Resource  Con-
    servation and Recovery Act (RCRA), plus the total
    amount of waste and material on a total  net weight
    basis which has been transferred to a surface impound-
    ment for purposes of reducing the water content by
    evaporation.
(3)  W2 is the total amount of hazardous waste disposed of
    in the state which is also regulated by RCRA.
(4)  W3 is the total amount of extremely hazardous waste
    disposed of  in the state, which  is also  regulated by
    RCRA and is declared extremely hazardous by Depart-
    ment of Health Services regulation.
(5)  W4 is the total amount of hazardous or extremely haz-
    ardous mining wastes.
  The Board of Equalization is  provided this information
by reports filed by all generators of such waste by March
1 of each year (Sec. 25342). Failure  to file the report re-
sults in liability of $500 per day, and $25,000 per day for a
knowing violation.
  This rather  complex  tax  structure was  worked out
among different industry groups. Alternatives included a
graduated tax by waste volume, different rates for on and
off site disposal and equal treatment for mining wastes and
surface impoundments. However, the Administration took
the position that its principal interest was in raising the
necessary funds, not in determining  the allocation of the
tax among different industries.
  Similarly, although the Administration  originally pro-
posed taxing feedstocks, it saw no impediment to taxing
the wastes themselves.  Industry asserted that a waste tax
would provide an incentive for recycling  and  opponents
asserted that a waste tax would provide incentive for mid-
night dumping.  After  some analysis, California decided
that a waste tax would not create a large marginal increase
in the cost of disposal, so that incentives for either activity
were likely to be small.

Coverage

  The California Act generally  covers releases as defined
in federal superfund.  Releases  of hazardous  waste are

-------
                                                                                      STATE PROGRAMS    39
more broadly defined because of generally broader state
regulation of such wastes. For example, geothermal ma-
terials and drilling muds are both regulated by the state,
but not by RCRA.
  On the other hand, the California Act excludes certain
storm water drained  into sewers, ash produced by a re-
source recovery facility utilizing a municipal  solid waste
stream, and the  "normal application of fertilizer, plant
growth regulant, and pesticides." (Sees. 25317  and 25321).
The storm water exception is taken from the exceptions to
reporting requirements for the State Water Resources Con-
trol Board, and should not prove to be any problem.
  It is not clear whether the "normal" application of pes-
ticides refers to application in accordance with statutes or
custom. What is  clear is that the state Act, in sharp con-
trast to the federal Act (Sec. 107(i)),  excludes such pesti-
cide applications  from any response authority, instead of
simply excluding such applications from liability.

Expenditures

  Sections 25351 and 25352  of the Act provide for  the
following types of uses of the state account:
(1)  Costs of administering the Act
(2)  The state's share as determined in any  cooperative
    agreement under the federal fund (42 USC 9604(c)(3))
(3)  Purchase by a state or local agency of hazardous sub-
    stance response equipment or other costs of preparing
    for responses  to releases
(4)  Reimbursement of all costs of remedial and  removal
    action incurred by the state or (with approval of the
    Director) any local agency, in response to  a release or
    threatened release
(5)  Costs of health effects studies or costs of cooperation
    with the Toxic Substances Disease Registry established
    pursuant to  Sec.  104(i)  of  the  federal Act,  up to
    $500,000 per year
(6)  Costs of natural  resources restoration,  rehabilitation
    and replacement,  except where the injury was already
    specified in an environmental impact statement, as set
    forth similarly in the federal Act

  The Act contains relatively few limitations  on this  ex-
penditure  authority.  Expenditure  for natural resources
restoration is the most encumbered, requiring considera-
tion of processes of natural rehabilitation, restoration and
replacement, limiting  the state to act on behalf of the pub-
lic  as  trustee of  such natural resources  and  precluding
funds for natural resources purposes where the damages
occurred prior to the Act (Sec. 25353).
  The Act also prohibits the usage of funds  for remed-
ial  action with respect to  facilities  owned or operated
by government entity at the  time of disposal  but only to
the extent that the government entity would otherwise
be liable for such costs (Sec.  25353). It would  appear that
such liability would be predicated on proprietary rather
than regulatory action or inaction, given broad grants of
sovereign  immunity from liability otherwise contained in
California Government Code Sees. 815 et seq.
  The one provision  that  is  not contained in the  Cali-
fornia Act that was  originally proposed by the  Admin-
istration is authorization to use some portion of the ac-
count to set up a center at the University of California
to perform research regarding alternatives to landfill dis-
posal of wastes, the development of safer substitutes for
toxic materials, and expanded toxicological and epidemio-
logical studies regarding hazardous substances. The author
of the bill and its sponsors indicated their willingness to
set up such a center but their unwillingness to do so with
fees  from industrial disposal of wastes.  It is hoped that
this subject can be addressed in greater detail next year.
  Procedural limitations are placed on the expenditure of
funds from the account. Action may not be initiated un-
til the Director determines that action will not  be  taken
properly or in a timely fashion by  any responsible party
(Sec. 25355). This provision clearly balances the need for
prompt and effective action against the similarly clear need
to assure responsible cleanup by responsible parties  wher-
ever possible.
  Industry spokespersons  also  evinced a strong  desire to
have state priorities coordinated with those of the fed-
eral  government through the National Contingency Plan
(see  Sec.  105(8))  of the federal Act,  (42 USC Sec.  9605
(8)). However, it was clear that the federal Plan would not
reach the same priorities for California as  California  might
reach itself and that the federal Plan would certainly not
call for remedial or removal action for the same number of
sites as called for in California.
  Early drafts of the Plan  also raised grave concern over
the ranking system used for such prioritization. The rank-
ing system appeared to emphasize groundwater and other
concerns and to give too little  emphasis  to potential fire
and  explosion hazards and to the potential for direct pub-
lic contact and health effects stemming from contact with
the site.
  The state's experience at the General  Disposal  site at
Santa Fe Springs bore  out  this assessment, where the ex-
plosion of 15,000 drums caused the immediate destruc-
tion of 250,000 fish and the inundation  of a broad area
with toxic debris. As  a result, the site ranking system
called for by the Act requires  only that the  Department
take into account the criteria established in the  National
Contingency Plan as well as pertinent factors related to
substantial protection of public health and environment
(Sec. 25356).
  Where expenditure of funds  will be made  pursuant to
a cooperative  agreement,  or otherwise  pursuant to the
federal Act, costs are reimbursed only insofar as they are
consistent with the regulations and guidelines contained in
the National Contingency Plan. However, for response ac-
tions taken pursuant to the state Act only,  costs will  be
allowed  to the extent  that the state's program is con-
sistent "to the maximum extent  possible" with the Na-
tional Plan (Sec. 25350). This language, when interpreted
in harmony with  Sec. 25356, will allow California to de-
termine its own needs while assuring maximum consis-
tency with federal edfforts.
  Finally, the Act provides for an appropriation  of  $1
million  per year  to take immediate corrective action to
respond  to substantial dangers to public health and the
environment. To facilitate emergency response, the De-

-------
40     STATE PROGRAMS
partment is empowered to enter into oral and written con-
tract without previous approval from the Department of
General Services, subject to prompt reporting to the De-
partment of Finance  and the Legislature.  It is antici-
pated  that substantial  funds from the  immediate  $2
million loan to the account will be used for this purpose.


Liability and Apportionment of Damages

   As a result of negotiated compromise, the Act no where
provides any standard  of liability  for  releases  of haz-
ardous substances. In a  letter to one legislator, the Pres-
ident of the California Manufacturers Association stated
that no such  statement  was needed because strict liabil-
ity is already the rule in California.
   Students of California law  believe that the standard of
liability may  vary depending on the  circumstance. Sta-
tutory  liability is predicated more on  non-reporting and
negligence theories, except that it would appear that dis-
posers  are strictly liable  for discharges  to surface or
groundwater that is  not permitted. As a  result  of this
crazyquilt  of liability standards at the  state level, it is ex-
 pected  that the standards contained in CERCLA  (Sec.
 107) will be utilized fully by the state in enforcing its own
 Act. There is ample precedent for enforcing federal stand-
 ards of behavior in state court.
   The  Act itself requires recovery from any "liable per-
son or  persons"  (Sec.  25360).  Liable  parties are also
 liable to the Department  for administrative costs in an
amount equal to 10% of  the damages.
   Because it is the Department that recovers monies and
does so only when costs are payable  from  the state ac-
count,  the state account is a party in any action for re-
covery under the Act (Sec. 25361).
   Because the funding mechanism creates what is tanta-
mount  to a pooled insurance fund on the part of indus-
try, industry sought and obtained agreement to set up an
apportionment system for liable parties. Any  party that
can prove by a preponderance of the evidence that only a
portion  of the damages  are attributable  to his or her ac-
tions is  entitled to apportionment. If  the evidence is in-
sufficient to establish such apportionment  by a  prepon-
derance of the evidence,  the Act  directs  the court to
"apportion such costs or expenditures, to the extent prac-
ticable, according to equitable principles..." (Sec. 26363).
California  law already provides that where an injury is in-
divisible in practice, even if divisible in theory, that ap-
portionment does not take place.
   The state expressed substantial concern that the pro-
visions of the Act permitting joinder of interested parties
in apportionment of damages would bar the state from go-
ing after other parties who were later discovered  to  have
been responsible  for  damages, and for  whom the  state
fund paid because of  apportionment with regard to the
other parties.  As a result, Sec. 25365 provides  that no
judgment shall bar the state  from later prosecuting the
responsible parties. The Act also provides that the appor-
tionment provisions in the bill do not affect any other com-
mon law or statutory remedy. The apportionment  pro-
vision affects only actions by the fund  for recovery of
payments made from  this industry-created account. Ef-
forts by some members of the oil industry to broaden the
apportionment provisions to other areas of tort law were
defeated.
  Finally, the  Act exempts  from liability  federally-per-
mitted releases, congruent with the federal Act and also
exempts state-permitted releases from liability. State-per-
mitted releases are defined more broadly than federally-
permitted releases, exempting releases that are allowed by
any  specific permit,  license,  or similar  authorization,
including  recognition of a standard industry practice by
the permitting  agency. Industry sought  an  even broader
exemption, arguing that an "implied" recognition of an
industry practice should lead to exemption.  This industry
effort was unsuccessful.

Compensation of Victims

  One of the principal failings of the federal superfund is
that it cleans up dirt but not people. California informed
industry spokespersons early on that it would demand a
provision  for compensation of victims as part of a Cali-
fornia superfund.  Industry feared that the compensation
provisions could use the entire amount of funding, with-
out ensuring cleanup.  The result was a compromise. The
Act allows up to $2 million each year of the $10 million
to be used for compensation purposes (Sec. 25381).
  A party who has sustained out-of-pocket medical ex-
penses or lost income or wages as a result of exposure to
a release of a toxic substance may obtain compensation
for 100% of those medical expenses  for three years and
80%  of  wages  or  business  income,  up to $15,000 per
year, for three years.
  The injured party must file a claim with the California
Board of Control, which administers a  similar  program
for victims of violent crime. The victim must prove by the
weight of the evidence that the proximate cause  of his or
her injury was the release (Sec. 25375).
The conditions  of compensation exist  when the person
cannot identify the party liable for the injury, the source
cannot be determined with reasonable diligence, the judg-
ment against a liable party cannot be satisfied, or there is
no liable party. The injured parties must describe the re-
lease and their physical injury or illness or loss. They must
also furnish information regarding their  last five years of
residence and medical  history, and figures regarding then-
last three years of income if such expenses are claimed. A
written decision is to be rendered within 90 days.
  Claims  are compensable if they are filed within three
years from the date of discovery. This administrative rem-
edy need not be pursued before filing any  court action.
Double recovery in court and through the Board of Con-
trol is barred. Any determination made before the Board
of Control is inadmissible in a judicial action.
  A major omission from the compensation provision is
the right of a party of obtain out-of-pocket medical ex-
penses or lost income from an allegedly liable party, pend-
ing resolution  of legal action. The  Administration sug-
gested that persons be allowed to make such claims after
submitting a demand to an allegedly liable party and not

-------
                                                                                     STATE PROGRAMS    41
having that demand satisfied within 60 to 90 days. The Ad-
ministration reasons:
(1)  indigent and disabled people need that compensation,
(2)  persons who do not know the identity of the source of
    the release should not be in a better position to make a
    claim than those who do know.
  Rewarding the  ignorant can deprive the fund itself of
valuable witnesses in a later subrogation action. (The fund
is  subrogated to  all claims and rights that  the injured
party has, with regard to claims paid under the fund (Sec.
25380).) In short, the current limitation on the ability to
obtain compensation may be both short-sighted as well as
hardhearted. It is hoped that cleanup legislation next year
can address this problem.

SUNSET

  The California superfund will disappear in 1991, five
years after a similar demise for  the federal Act. Industry
strongly advocated a five year sunset provision. However,
the Administration was able to argue successfully in com-
mittee that a ten year sunset clause was the least that would
be needed to clean up California's subsequent problems.
  California's superfund legislation is an important step
forward in the control and eradication of hazards  result-
ing from irresponsible disposal of hazardous waste. The
State looks forward to cleaning up past mistakes. How-
ever, the State is ever mindful that the most important ele-
ment of any government program  in this area is preven-
tion. To that end the State must provide strong enforce-
ment of hazardous waste regulations,  responsible siting
decisions, and the  facilitation  of  alternative waste  dis-
posal and waste reduction strategies.

REFERENCES

1. Unless otherwise specified,  all  section references are
   to the Health and Safety Code.

-------
                   THE PENNSYLVANIA APPROACH TO
                                UNCONTROLLED SITES
                                   MICHAEL D. LaGREGA, Ph.D.
                                           Bucknell University
                                        Lewisburg, Pennsylvania

                                         RICHARD H. DEMMY
                                          JOHN W. THORSEN
                                           RoyF. Weston, Inc.
                                       West Chester, Pennsylvania
INTRODUCTION
  Pennsylvania is estimated to be the fourth largest gen-
erator of hazardous wastes in the United  States. It cur-
rently has no sites that are legally able to accept hazardous
wastes, although a number of licensed solid waste facili-
ties accept a small amount  of specific wastes. As a result
of past  disposal practices,  the number of uncontrolled
sites in Pennsylvania is estimated to be as high as 400, al-
though some of these may not prove to be hazardous when
evaluated.
  Current state policy is to clean up orphaned sites where
they present imminent  hazard and  to  establish a state
hazardous waste management plan  that will provide a
framework for the orderly handling of these wastes. How-
ever, uncontrolled sites will continue to be a serious prob-
lem until new disposal sites  are established.  Therefore,  the
issue of siting is integral to solving the problem  of  or-
phaned sites.
LEGAL BASIS

  Legislation for the state to assume primacy in regula-
tion of hazardous waste management was passed in the
Solid  Waste Management Act of 1980  (Act 97). This
legislation provides  for the planning  and regulation of
solid waste storage, collection, transportation, processing,
treatment and disposal,  provides for the regulation of
hazardous wastes,  establishes  permit requirements for
operating hazardous waste management facilities, as well
as requiring  licenses for the transportation of hazardous
wastes.
  A unique portion of this legislation is the provision for
override of local zoning  by the issuance of  a  "certifi-
cate of public necessity"  by the Pennsylvania Environ-
mental Quality Board to a treatment or disposal facility
which has:
  (1)  Obtained all necessary permits
  (2)  Is in conformance with the Pennsylvania Hazardous
      Waste Facilities Plan
  (3)  Has satisfied  the Environmental Quality Board re-
      garding:
      a. Identification and mitigation of the impact upon
        adjacent  populated  areas  and  areas  through
        which waste is transported
     b. Identification and mitigation of the impact on the
        municipality in which it is to be located in terms
        of health, safety, cost and  consistency of local
        planning
     c. Participation in a community relations program
        in which the citizens have had a meaningful op-
        portunity to participate in all phases of the site
        selection process
  Includes  in the legislation is the requirement for the
Pennsylvania Department of Environmental  Resources
(DER) through, and with the assistance of, the Hazardous
Waste  Facilities Planning Advisory Committee,  to de-
velop preliminary environmental, social  and economic
criteria for the  siting  of  hazardous waste facilities. The
law  required these criteria to be published within six
months of the passage of  the Act. It also required the de-
velopment of a  hazardous waste management plan within
two years.
  The criteria were developed upon a three tier concept.
Level 1 criteria are exclusionary  for  the location of a
hazardous  waste  treatment or disposal facility. In this
area are  such exclusions  as coastal  flood  hazard areas,
coastal wetlands, water  supply  watersheds, wilderness
scenic areas and historic sites. In Levels 2 and 3, criteria
were developed  to indicate factors that would be positive,
cautionary or negative (exclusionary) and require in depth
evaluation  addressing the  appropriate  environmental,
social and economic concerns.
  In June and  July of 1981 a series of public meetings
were held by the Advisory Committee to provide for pub-
lic input. Final  criteria are  scheduled to be published in
October.


INTERACTION WITH SUPERFUND

  The  Remedial  Action  Program instituted under the
Federal Comprehensive  Emergency  Response, Contin-
gency and Liability Act (CERCLA or Superfund) has been
implemented. Seventeen sites  in the United States have
been identified  for the first remedial action engineering
studies to identify a  feasible  alternative  and design the
clean-up program for these sites.
  The sites were selected by  a screening process developed
by EPA.  The states worked together to identify the high-
est priority sites and to characterize  the sites according
to their potential for environmental harm. One of the 17
                                                    42

-------
                                                                                    STATE PROGRAMS    43
sites is located in the Commonwealth of Pennsylvania at
Bruin Borough, 60 miles north of Pittsburgh.
  The Pennsylvania  site is known as Bruin Lagoon. It is
an abandoned lagoon site that has  been the recipient of
various  wastes from  the 1930s through  1980. The wastes
that were  deposited there included  petroleum refining
wastes, crankcase oil reclamation waste (acid clay sludges)
and coal washing residues.
  The toxic nature of the lagoon contents was shown in
1968 when sludge and supernatent were spilled into Bear
Creek, a  tributary of  the  Allegheny River.  The results
of this 3,000 gal spill were a fish kill of 4,000,000 fish
and visible foaming of Bear Creek  and the Allegheny
River for 200 miles downstream.
  The site is perched above  Bear Creek and  is retained
by  manmade dikes.  The remedial actions to  be  studied
will focus on feasible and implementable alternatives for
both on-site and off-site management of these wastes.
  The state, Federal  and contractor  interrelationships
that must  be established in  this program  are critical to
its successful implementation. For the Bruin Lagoon pro-
ject,  a  management structure has  been  built between
EPA, DER and the engineering services contractor to as-
sure accurate, continuous communication; to define roles
and responsibilities  of  the parties; and to enhance the
ability of the selected alternative to be implemented in a
timely fashion.
  For Superfund, there  is a project officer located in EPA
Headquarters  in Washington. This person manages the
technical aspects of the  contract from EPA's prospective.
A Project Monitor has  been assigned  within EPA Region
III, Division of  Surveillance and Analysis, Environmental
Emergency Branch to act as a local contact with DER and
the  contractor.  The Project Monitor's responsibilities
include resolving logistical problems, providing input to
the work plan and evaluating and selecting the remedial
action alternative in coordination with DER.
  DER has identified a Project Manager who is the re-
gional solid waste manager  in the Pittsburgh Regional
Office to perform the same functions  as the EPA Project
Monitor, only  at the state level.  The  Project Manager
was selected based on his knowledge of the specific prob-
lem at hand.
   The Contractor is responsible for preparing and imple-
menting the work plan so that the most  cost effective,
reliable and feasible  alternatives are evaluated. After eval-
uation, the feasibility study is presented to EPA and DER
for the selection of the remedial cleanup action.  After
selection of the  remedial action alternative, the contractor
then  finalizes the engineering services tasks by develop-
ing an implementation  plan,  plans and specifications and
bidding documents for  the implementation of the chosen
remedial cleanup alternative.
  The Superfund Remedial  Action projects  are action-
oriented. The engineering services,  from  project initia-
tion, data analysis and data collection through prepara-
tion of a bid package,  is scheduled to take 27 weeks  or
less. This time  frame will  limit the number and  type  of
cleanup alternatives  that will be analyzed. This restriction
may  tend to skew  the alternatives  selected to "pro-
curement "-oriented programs where technologies or bid-
ders may be pre-selected to enhance the feasibility or im-
plementability of any chosen alternative.  This  type  of
approach puts emphasis on proper characterization of the
quantity  and  quality  of the material in  the  lagoon.
Further, adequate characterization must be conducted so
that various manners or techniques for management can
be  accurately and  cost-effectively  bid by  prospective
contractors.
  The first remedial action cleanup in Pennsylvania  is
very important to  the state. There are many abandoned
or orphaned sites that must be cleaned up. This project
will  serve as a learning program and model  for future
activities.  Superfund cannot pay for all of these; how-
ever, the most serious imminent hazards can be addressed
and resolved all under this program.


TRAINING

  In order to achieve the goals of  the state  hazardous
waste management program, it is necessary to develop a
staff of individuals trained and sensitive to the problems
in managing hazardous wastes. A number of recent in-
cidents have indicated that the specialized training and
equipment required  to inspect and  evaluate sites is pre-
sently lacking.
  DER has outfitted  a response  trailer  containing all
necessary   protective  equipment  from  self-contained
breathing  apparatus to completely  enclosed  suits. The
trailer will also have analytical monitoring equipment re-
quired to assess sites.
  However, trained  personnel are needed to utilize this
equipment. The  type of training envisioned varies from
brief seminars for DER management to intensive courses
for staff members who will actually go to the site. The
management seminars are intended to sensitize senior staff
to the health and safety problems to which their employees
may be exposed and the Department's  liability for such
exposure.  This program should provide the  knowledge
and  skills  required to safely and efficiently handle most
situations   that  are  likely to  be  encountered  in  the
field.
  The people who inspect uncontrolled sites require spe-
cific training including the use of safety and monitoring
equipment,  background instruction  in toxicology, first
aid and legal  procedures  for  maintaining an unbroken
evidence chain for samples. The Department is currently
reviewing methods to provide the required training. How-
ever, the staff time and funds needed to implement this
extensive training program are not currently available.


PUBLIC PARTICIPATION

  In Pennsylvania as in other states the emotional opposi-
tion to hazardous wastes is intense.  As an example,  at
one  public hearing  on a permit,  opponents  of sites
throughout the  state converged upon the hearing loca-
tion in buses and approximately 2,000 people greeted the
state's representatives.

-------
44     STATE PROGRAMS
  In order to develop an acceptable siting policy for all
affected parties, roundtable discussions  with  two repre-
sentatives from each of 20 groups opposing specific sites
in their communities were initiated.  The goal is to de-
velop acceptable methods of siting faculties. The round-
table has met four times, twice in 1980 and twice in 1981.
  As might be expected, initial meetings of  the  round-
table concentrated on individual concerns of their local
municipalities problems.  The group ultimately did discuss
three items of prime concern to the participants, namely:
1. Why locate a facility in "our" community?
2. State-of-the-art of hazardous waste management tech-
   nologies
3. Facility operator responsiveness and integrity.
  In addressing the financial concerns of the communities
impacted by a facility the roundtable has indicated a need
for the facility operator  to participate in the  tax burden
of  neighbors  and to  protect  existing landholders from
financial loss due to property devaluation. One suggested
approach was that the facility operator would pay:
  (1) All real estate taxes within a one mile radius of the
      facility and 0.25 miles on either side of the access
      road to the facility.
  (2) 75 % of the real estate taxes 1.5 miles away.
  (3) 50% two miles away.
  (4) 25% 2.5 miles away or 0.25  miles either side of the
      access road to the facility.
  In addition,  it was  felt that  the facility developer
should purchase any real estate offered within  the zone of
impact if offered within the first six months of announce-
ment of the  facility. This is to permit those  individuals
who do not desire to be neighbors the option to  move.

A CASE STUDY

As  an example of  the  many  cleanup efforts currently
underway, DER's  activities at the  Enterprise  Avenue
Landfill in Philadelphia will be summarized. The City of
Philadelphia  Water Department owns and operates a 57-
acre landfill in southwest Philadelphia. Over a period of
six  years, approximately 40 acres  were filled  with incin-
erator residues,  fly ash  and  construction/demolition
debris. Later, it was found that industrial/chemical wastes
had also been dumped in the landfill.
  To assess the potential migration of contaminants into
ground- and surface-waters, Philadelphia retained Roy F.
Weston, Inc. to study and characterize the environmental
impact of the  landfill (Phase I). The  investigation was
conducted with heavy emphasis on health and safety pro-
tection. After  determining the contamination levels  in
the soils and  groundwater and collecting  the  necessary
hydrogeological information,  a remedial action plan was
developed for the landfill (Phase II).
  Under this plan, drums and other containers of indus-
trial/chemical wastes are being located using  a Ground
Penetrating Radar Unit. Bid  documents (drawings and
specifications) will be prepared which will allow the City
of Philadelphia to obtain  the services of a contractor
who will  remove  these drums, containers  and contami-
nated bulk material  for transport to approved disposal
sites. During drum removal and  subsequent site closure,
continuous field  inspection  will  be  maintained  by the
consultant. As  in the first phase,  heavy emphasis will be
placed on health and safety protection.
  Concurrently, the ground- and surface-waters are being
monitored to assess  the capability of  the bottom clay
layer to retard contaminant migrations. Appropriate ma-
terials for covering the  landfill after removal  of the in-
dustrial and chemical  wastes are being  identified and
located for use.
  After these wastes have been removed, a resurvey of the
site will be conducted to  update the site plans and prepare
post-closure plans. Post closure activities will  involve
continued  groundwater  monitoring  and maintenance of
the cover materials to  assure their integrity.
SUMMARY

  Pennsylvania  is  evaluating  uncontrolled  sites  and
initiating clean-up activities  wherever imminent hazards
to health are identified.  The  Commonwealth  is in the
process of developing a Hazardous Waste Management
Plan, which envisions the establishment of new disposal
facilities as essential to the management of uncontrolled
sites.

-------
            HYDROGEOLOGICAL INVESTIGATION OF AN
           UNCONTROLLED HAZARDOUS WASTE SITE
                                     CHRISTOPHER HAGGER
                                           PAUL F. CLAY
                                   Ecology and Environment, Inc.
                                       Woburn, Massachusetts
INTRODUCTION

  Hydrogeological investigations of  uncontrolled haz-
ardous waste sites can be time consuming, expensive and
dangerous.  Ecology and Environment, Inc. (E & E) has
developed a unified approach to uncontrolled hazardous
waste site investigations which combines existing geophysi-
cal investigatory methods with field analytical techniques.
This approach tries to provide preliminary information on
the  relative hazards associated with  an uncontrolled haz-
ardous waste site  in a short  period  of time at minimum
expense. Field analytical methods adopted by E & E and
described in more detail in another  section of this paper
are  used to closely control the costs of groundwater well
installation and monitoring  program and provide field
personnel with air characterization data for health, safety
and respiratory protection purposes.
  This case history involves a hydrogeological site inves-
tigation whose purpose was to develop and implement a
groundwater monitoring program to determine the extent
of groundwater contamination. The location  will be re-
ferred to as the Site.*
  The Site is located on privately owned land which was
previously used for mining operations.  About ten years
ago, the owner of the property began to receive 55-gallon
drums  containing  spent solvents,  sludges and other mis-
cellaneous chemical waste. The drums were placed on the
ground and stored on a portion of the owner's land in an
area of approximately six acres. Many drums were in poor
condition and  soil staining indicated that material had
been spilled. It was also suspected that some of the drums
had been buried. At the commencement of the hydrogeo-
logical study,  approximately  half of the total  number of
drums had been removed during initial emergency cleanup
actions. The remaining drums were placed in a bermed and
lined temporary storage area.
  Bulk dumping of waste chemicals was alleged to have
occurred  in an area northeast of the temporary  drum
storage area.  Inspection of  five  year  old aerial photo-
graphs revealed a  depression  or pit in this area which has
since been filled in. The existence  of this pit and the bulk
dumping activity was alleged by several eyewitnesses.
The approach presented in this case hr.tory involves the investigation
 of a site which was  used for the storage and disposal  of hazardous
 wastes and is alleged  to have caused groundwater contamination. The
 actual site characteristics and analytical results described herein are
 fictitious. However, the procedures, engineering approach, team organ-
 ization and field analyses described actually took place.
  The Site is located west of a river which flows south-
ward toward a small city. Two of the city's water supply
production wells, which are located on the west side of the
river about 1.5 miles downgradient from the Site, were
found to contain small  quantities  of  organic solvents.
Activities at  the Site were halted  approximately  eight
months  ago  and cleanup operations  began.  Contam-
inants which had been spilled, leaked or dumped on the
ground have either vaporized or percolated into the under-
lying soil and groundwater.
PRELIMINARY SITE ASSESSMENT

  E & E under contract to the U.S. Environmental Pro-
tection Agency (EPA) began its  hydrogeologic investiga-
tion of the Site by first conducting a preliminary site char-
acterization which included a review of past and present
land uses, preliminary waste characterization and analysis
of the Site topography, cultural landmarks, aerial photo-
graphs, surface hydrology and surficial geology.
  The preliminary site assessment involved an analysis of
all available information on the Site. The location, accessi-
bility, general  area topography,  past/present ownership/
uses  of the Site were determined from maps supplied by
the local assessor's office and the United States Geological
Survey (USGS). USGS and the Soil Conservation Service
(SCS)  provided  preliminary information  regarding the
surficial geology of the Site.
  Hazardous wastes that were disposed of and remain on-
site in 55-gallon drums were sampled, analyzed and iden-
tified by the EPA. The analytical data and other back-
ground information indicated that most of the compounds
identified in the 55-gallon drums were volatile solvents.
The  presence of these types of  substances indicated the
need to consider the fuse of respiratory  protection  gear
while working at the Site.
  After consultation with  personnel at the state and local
levels regarding their knowledge of disposal activities  at
the Site, E & E was able to complete the preliminary Site
assessment  without  the considerable expense  of several
Site visits.

Site Description

  Land elevation is approximately 250 feet above mean
sea level to the  north, west and  southwest  of the Site
                                                    45

-------
46    SITE INVESTIGATION & ASSESSMENT
(Figure 1). The land elevation slopes to approximately 230
feet above mean sea level to the east and southeast of the
Site  which is bounded by the river.  Private residences
whose drinking water is supplied by individual wells are
located northwest, west and southwest of the Site along the
county road.
  Preliminary information also indicated that the Site and
its surrounding environs are underlain by  varying thick-
ness of glacially deposited materials. These unconsolidated
materials provide a medium for groundwater storage and
movement  and as such constitute a significant aquifer.
The  glacial sediments lie on top of the metamorphic and
igneous bedrock that is capable of providing small but re-
liable amounts of groundwater. As previously mentioned,
parts of the Site were also used for mining operations.

SITE MAPPING

  As part of the engineering approach to this site investi-
gation and groundwater well  monitoring  program, the
Site  was mapped extensively to show  site-specific  topog-
raphy, cultural landmarks, areas of surface drainage, veg-
etation, locations of waste storage containers, on-site/off-
                                    1 5 MILES TO
                                    MUNICIPAL WELL
                                    HELD
    KEY.   A I* GRADIENT WELL

         © ON-SiTE WELL

         I DOWN GRADIENT WELL
        D
            PRIVATE RESIDENCE

            DIRECTION Of
            GROUWWATER
            MOVEMENT
            NOT TO SOLE
                        Figure 1.
 site existing wells locations, accessibility to the Site, prop-
 erty lines, easements, building locations, surficial geology
 and groundwater/bedrock topography. All of this infor-
 mation was gathered as part of the next phase of E & E's
 site investigation following the preliminary assessment. It
 involved actual Site visits by personnel with different spe-
 cialties and use of various geophysical measurement tech-
 niques to gather all the necessary data.
   The purpose of the second phase of the overall inves-
 tigation was to provide enough information to develop as
 accurately as possible the scope of the subsequent ground-
 water well drilling and installation program. Drilling and
 installation of monitoring wells can be  the  most expen-
 sive phase  of a hydrogeological investigation and exten-
 sive knowledge of the area where monitoring wells are to
 be located can save both money and time in the course of
 the overall site investigation.
   Groundwater/bedrock mapping was developed and
 combined  with the Site topographical  base map to assist
 in  choosing  optimum  monitoring well  locations.  The
 groundwater  and bedrock data  was augmented  by  the
 measurement of private well  water depths at residences
 adjacent to the Site. A ground penetrating radar survey
 was performed on the  Site to determine the presence of
 any buried 55-gallon drums at selected on-site well loca-
 tions.
   The results of the seismic refraction study and measure-
 ment of private well data depths indicated that the bed-
 rock/groundwater contours  generally reflect the surface
 topography, i.e., gradually sloping southeast toward the
 river.  The corresponding groundwater flow pattern is in-
 dicated by the large directional arrows on Figure  1. The
 ground penetrating radar study revealed no buried hazards
 at the selected locations.

 GROUNDWATER WELL MONITORING PROGRAM

 Location of Monitoring Wells

  The purpose of instituting  the monitoring well program
 was to assess  whether groundwater flowing to the Site is
 uncontaminated and that groundwater which flows off-
 site has  been contaminated  due to past disposal  opera-
 tions.  A monitoring well network consisting of 14 wells
 was installed (Figure 1) on and adjacent to the Site.
  These well locations were selected in order to detect on-
 site contamination and off-site migration. Two clusters of
 wells (well  cluster A and well cluster B), each consist-
 ing two  separate wells  in close proximity and which ter-
 minate at  different depths,  were located downgradient
 off-site to detect contaminants in soil  layers of different
 permeability.
  Six  wells were  located downgradient from the Site in
 line with the groundwater flow from the Site. The majority
 of contaminants moving through the Site are transported
by the groundwater into the path of these proposed mon-
itoring wells.
  Four wells were drilled and installed upgradient from
the Site  to act  as  "control"  wells  to assess  whether
groundwater flowing onto the Site was uncontaminated as

-------
                                                                    SITE INVESTIGATION & ASSESSMENT    47
                   Fixed Level Well
  51 length
  Steel Protective Sleeve-*"
  Layer Concrete Grout*-f° o°
   Length Slotted P.V.C. -

            P.V.C. CAP
                 Threaded Cap with Lock
                       (Existing Ground
                       '
 5' steel Protective
      Sleeve
        -
        Fully  Screened Well


                     -Threaded Cap with Lock

            |T— I    *~
            Llol
                                                                             2' 8entonite
                                        Water Table
                                           V
                                 1 1/2" Nominal Diameter P.V.C.
                                - Schedule 80

                                 Threaded Flush Joints
                                 Concrete Grout or
                                - Impermeable Material
                                 for Backfill
                                2' Layer Bentonite
              5' Layer Ottawa Sand


           fc  1' Ottawa Sand	
                                                                                       • 2' Concrete Grout
                                                                                     Screened }' below
                                                                                     Existing Ground
                                                                                                       Ottawa Sand
                                                                                                  J11 Ottawa Sand
      X
X
X
X
K  »^ -I
X
                                                       Figure 2.
                                            Typical Detail of Monitoring Wells
                                                     (Not to Scale)
suspected and to verify the prevailing groundwater flow
pattern to the Site.
  The remaining four wells were drilled and installed at
various locations on  the Site,  free from buried objects,
however,  downgradient from the drum  storage and bulk
disposal area.
Monitoring Well Construction Specifications

   E & E established construction specifications and pro-
cedures for drilling and installation of monitoring wells for
areas potentially contaminated  with hazardous wastes.
These new specifications and procedures involved issues
never addressed before. Well construction specifications
and procedures included safety precautions, equipment de-
contamination, on-site soil analysis, field determination of
well depth placement, collection and disposal of poten-
tially hazardous wastes  generated by well drilling and in-
stallation, maintenance  of field records and management/
staffing during the drilling and installation. Safety precau-
tions, on-site soil analysis and field determination of well
depth placement will be discussed in the subsequent Field
Analytical Approach section.
                                              These specifications, based on engineering construction
                                            specifications for typical soil borings and well installa-
                                            tions, were adapted for use on hazardous wastes  sites
                                            with emphasis on  the minimum amount  of drilling re-
                                            quired to accomplish the objective of the overall investi-
                                            gation as well as protection of the environment and the
                                            health/safety of those involved in the work.

                                            Well Types

                                              The two types of wells (Figure 2)  that  were installed,
                                            fixed-level and fully-screened, are differentiated by length
                                            of their well screens.
                                              All wells installed on and off-site  were fully screened
                                            groundwater wells from bedrock/refusal to a  point ap-
                                            proximately  4-5 ft below  ground  surface with  the ex-
                                            ception of certain off-site  wells  used to detect contam-
                                            inants in soil layers  of different  permeabilities. These
                                            fixed-level wells were screened only for the length of con-
                                            taminated soil layer.
                                              The monitoring well casing used was PVC schedule 80
                                            with a nominal pipe size of one and  1.5 in. The pipe ex-
                                            tended 2.5 ft above ground and had threaded flush joints.
                                            No solvents  were used as  joining compounds. The pipe

-------
48    SITE INVESTIGATION & ASSESSMENT
used allows passage of a 1.3 in. O.D. bailer for sampling
purposes. Slot size for well screens was 0.010 in. and the
casing at the bottom of all wells was capped.
   All monitoring wells were drilled using either the drive
casing method or hollow-stem  auger method. The wells
were installed using an all-terrain vehicle-mounted drilling
rig. Final well depths were determined in the field by E & E
personnel using a portable gas chromatograph to analyze
soil samples for contamination as they were brought to
the surface. The exact well depths were determined as a
function of contaminant concentrations. This procedure is
described in more detail in  a subsequent section of this
paper.
   As drilling proceeded,  a well log containing the follow-
ing information was  recorded:  well location,  well  num-
ber, depth to water  table, soil classification and depths,
driller's  remarks,  sample depth  and type, well  screen
length and depth installed, gravel pack depths, seal depths,
total  well depth,  date and time of the well drilling and
weather  conditions during drilling, E & E personnel and
contractor personnel  present during drilling, number of
blows required for each penetration of split spoon samp-
ler, weight of drop hammer, height of drop, monitoring
well material used and lengths per well.
   Only  clean,  non-recirculated water was used as wash
 water for drilling  in a new well  location. All  drilling
 equipment, casing, weights,  samplers, and augers, were
 cleaned  by the driller prior to drilling the first hole and
 when drilling equipment was moved to any new well loca-
 tion.
   This cleaning process consisted of:
   (1) High pressure  hot water cleaning  of the  drilling
      equipment
   (2) Rinsing of equipment with a methanol spray
   (3) High pressure hot water final rinse
   All waste generated from this entire washing process was
 collected for possible hazardous waste disposal. E & E per-
 sonnel determined in the field whether the waste material
 required hazardous waste disposal based  on field analyti-
 cal techniques. No oil or grease was used during drilling
 in such a manner as  to contaminate the  drilled hole. No
 equipment was removed from a drilling  site at any time
 without  first being decontaminated as  previously des-
cribed.
   Wash  water remaining after  decontamination of drill-
ing equipment and personnel decontamination was con-
sidered a hazardous  waste and was not discharged into
the environment. These wastes were containerized in 55-
gallon drum containers for final disposal.

Well Drilling Management/Staffing

   The E & E team on-site during well drilling operations
included four specialists. One team member was  a con-
struction engineer as project manager who verified well in-
stallation techniques, maintained a well  log and decided
upon final well placement depths/location. The second
specialist was a geologist who visually analyzed the soil
samples  and maintained the geologic field record. The
third team member was  a chemist who analyzed  the soil
samples for contaminants and who monitored the ambient
air in the immediate vicinity of the well drilling (described
in more detail subsequently).  The fourth team member
was a safety officer, stationed outside the immediate well
drilling location who monitored the overall safety of the
workers, and set  up of  decontamination stations. This
team approach to  well drilling and installation proved in-
valuable for the overall success, safety and expediency of
the groundwater monitoring well program.

FIELD ANALYTICAL APPROACH

  As mentioned earlier, the predominant chemical wastes
at the Site were volatile organic compounds. The presence
of these types of compounds  which were alleged to have
been spilled, leaked and/or dumped on the ground at the
Site gave rise to two considerations for the installation
of groundwater monitoring wells at  and around the Site.
The first consideration was the potential for personnel ex-
posure to organic vapors during drilling operations and the
second consideration was the opportunity to evaluate the
distribution of contaminants  in the soil and groundwater
as the  wells were being installed. The following sections
describe the approach to  air  monitoring respiratory pro-
tection and soil/groundwater  analyses carried out at the
Site during the monitoring well installation.

AIR MONITORING/RESPIRATORY PROTECTION

Background Information

  Except in an area of the Site where drums remain  all
other areas on the Site where wells were installed were free
of any visible sources  of organic vapors. All of the on-
site area is open and  flat, precluding conditions which
could lead to an accumulation of potentially dangerous
levels of organic vapors.
The proposed locations of on-site monitoring wells were
at areas of the Site where it  was  highly probable that
large amounts of contaminants had been leaked, spilled
or dumped onto the ground (Figure 1). Therefore, pene-
tration into contaminant saturated soil and groundwater
by drilling  equipment could lead to  the possibility of re-
leasing organic vapors to the ground  surface. Personnel
working in the direct vicinity of the well hole would then
be subject to an organic vapor respiratory hazard.
  In addition, the locations of six of the wells were down
gradient in the Site's groundwater  flow and although it
was not anticipated  that highly contaminated soil would
be encountered during  drilling, consideration had to  be
given to the possibility  of personnel exposure to organic
vapors  during  the processes  of well development and
sampling which involve bringing groundwater to the sur-
face.

Air Monitoring

  A Century Systems Model 128 Portable Organic Vapor
Analyzer (OVA) was used in the total survey  mode to
continually monitor  organic vapor concentrations in the

-------
                                                                 SITE INVESTIGATION & ASSESSMENT    49
air in the immediate vicinity of each well hole on-site and
down gradient.  In the  total survey  mode, air is drawn
into the instrument through a probe and sampling line to
a  flame  ionization detector (FID). The instrument used
is calibrated to methane so that the concentrations read
out  from the FID response are methane-equivalent,  i.e.,
the total organic vapor concentration (expressed in ppm) is
relative to the concentration of methane.
   Prior to drilling at each location, total organic vapor
concentrations  were  within  the  Site's  normal  ambient
range of 3-4 ppm. As drilling commenced, the air in the
immediate  vicinity of the well hole was monitored. Also,
each time drilling was stopped to allow a  new section of
hollow stem auger or casing to be added, the probe of the
OVA was placed directly at the top of the well hole. Al-
though personnel would not be breathing air directly from
Jhat point, the intent was to anticipate  high levels of
vapors which might emanate from the hole  and possibly
be inhaled by personnel.

Results

   All personnel  at on-site locations wore  full-face air
purifying respirators  to start drilling operations. At the
locations of two on-site wells, total organic vapor  con-
centrations  intermittently  approached    1000  ppm,
measured at the  top of the well hole. Although concen-
trations were still in the range of ambient background (3-
4 ppm)  in the breathing range of personnel  and  favor-
able weather conditions  existed (low  temperature  and
strong, steady winds) all personnel utilized Self Contain-
ed Breathing Apparatus (SCBA) to complete well drilling
at these two locations. Additionally, SCBA was used  dur-
ing the withdrawal of the drilling auger sections, as high-
ly contaminated  soil  was brought to the  surface during
this operation.
   At the other on-site locations, concentrations above am-
bient levels were not  detected at the top of the well  hole
or in the breathing range of personnel who wore air  pur-
ifying respirators throughout the installation procedures.
At the locations  downgradient from the site,  concentra-
tions above ambient levels were not detected and  per-
sonnel were able to complete these installations  without
donning respirators.

FIELD ANALYSIS OF SOIL SAMPLES

   Soil samples were collected during drilling using a 2 in.
O.D. split-spoon sampler every 5 ft or at  each change in
geological  stratum. Soil samples taken over  the entire
length of the bore hole were then analyzed for total vol-
atile organics. The concept of the approach was suggested
bySpittler.(1)

Field Analytical Instrument

   The instrument used  for field analysis for volatile or-
ganic compounds was a Century Systems Model 128  Por-
table Organic Vapor  Analyzer (OVA) with  Gas Chrom-
atograph (GC) capability. The instrument may be used in a
total survey mode as described in the previous section on
Air Monitoring Respiratory Protection and may also be
used as a GC by introducing a sample to a GC column and
recording a chromatogram on a portable strip chart re-
corder. The strip chart recorder used in this instance was
a Rustrak Model 288 with a high and low gain setting and
two chart speeds attained by changing gear assemblies in-
side the recorder.
  The Century OVA utilizes a  flame ionization detector
(FID) which consists of a diffusion flame of pure hydro-
gen mixed with air. This type of detector is commonly used
in laboratory GC applications. The same hydrogen supply
is also used as  the carrier gas in the GC column. The sen-
sitivity of the  instrument when  used in the GC mode in-
creases over that possible  when used in the  total survey
mode. By increasing the size of the sample aliquot injected
and setting the strip chart recorder on high gain, it is pos-
sible to detect volatile  organic compounds in concentra-
tions as low as  30 ppb.
  Upon retrieving the split-spoon at each sampling  inter-
val, a clean, stainless steel spatula was used to place a small
quantity of soil from the core into a clean 40 ml vial  fitted
with a Teflon-backed septum. The depth over which the
sample was collected was recorded by the supervising geol-
ogist and entered on the field analytical record.
  The soil samples were allowed to thermally equilibrate in
the interior of the field laboratory for several minutes prior
to analysis. Each sample was then screened for total vol-
atile organic compounds and, if present, the sample was
analyzed for individual volatile constituents.
  The GC column chosen for the initial analyses of soil
samples was a type T-6 (1 %TCEP on Chromosorb W, HP,
60/80 mesh). Although this column is short in length and
may present separation problems at higher ambient tem-
peratures, it was selected for field use because of the  fairly
low ambient temperatures encountered and the need  to re-
duce the time between analyses so that the information ob-
tained could be used to make expedient decisions  about
well installation and placement.
  The procedure for analysis was to withdraw by syringe
an aliquot of soil sample headspace vapor and inject this
into the GC column with the backflush valve up and strip
chart recorder  running. The size of the aliquot injected was
based upon the response shown during the initial total vol-
atile analysis,  but was  typically 250 pi  for off-site  down
gradient wells  and  25-50 ul for  on-site wells.  The chrom-
atogram was allowed to run for 5-10 min., at which time
the backflush  valve was depressed reversing flow in the
column and the presence of any heavy constituents noted.
  Aliquots of  headspace vapor from aqueous mixtures of
known volatile compounds were then chromatogrammed
in order to establish retention times. The resulting chrom-
atograms were compared with those of the samples  to de-
termine whether there  were any retention time matches.
Where matches were apparent, repeat injections of the
samples and knowns  were made. Two other columns were
then used to further establish retention  time matches. A
schematic representation of  a typical soil sample chrom-
atogram and of subsequent injections of headspace  vapor
from known aqueous mixtures is shown in Figure 3.

-------
50     SITE INVESTIGATION & ASSESSMENT
RESULTS

   A summary of the results of the soil sample analyses is
given in Table I. Concentrations are expressed on a rela-
tive basis and no attempt was made at precise quantita-
tion.  The  field analysis  of  the  soil  samples  bene-
fitted the well installation in the following ways:
•It could be  confirmed that well placement based upon
  previously existing geophysical and hydrological data was
  intercepting the groundwater contamination plume.
•The vertical distribution of contaminants at each  well
  location could be integrated with geological data to assist
  personnel  in selecting optimum depths for well screen
  placement during  well installation, thereby  eliminating
  unnecessary  well drilling or well screen placement which
  would save on the overall costs of the monitoring  well
  program.
•The need to install wells at different depths at the same
  location could be confirmed or eliminated.
•The-need for drilling equipment to be decontaminated in
  between wells could be confirmed or eliminated.

        Column: T-6
        Gain. "HI"
        Chart Speed: 0.657mln.
                          Table I.
              Summary of Soil Sample Analyses
                   During Well Installation
                                         7
  The above schematic representation is of chromatograms ob-
tained by injecting headspace vapor from a soil sample (1) and
from dilute aqueous mixtures of known volatile  organic com-
pounds (2) and (3). Chromatograms of the samples and knowns
were also run on different columns to confirm retention time
matches.

  1.  150 jul Injection: (Small spike marks injection)
    a. 1, 1, 1 Trichloroethane
    b. Trichloroethylene
    c. Tetrachloroethylene
    d. Toluene
    e. Ethylbenzene
    f. Air peak due to backflush valve activation
  2.  100 ul Injection: (Small spike marks injection)
    a. 1, 1, 1 Trichloroethane
    b. Trichloroethylene
    c. Tetrachloroethylene
  3.  100 ul Injection: (Small spike marks injection)
    a. Benzene
    b. Toluene
    c. Ethylbenzene
    d. O-xylene

                         Figure3.
             Typical Soil Sample Chromatogram
Well Location
    Up-gradient
    Up-gradient
    Up-gradient
    Up-gradient
 5. On-site



 6. On-site

 1. On-site

 8. On-site
Depths of
Volatile Detection

None Detected
None Detected
None Detected
None Detected
5.5-15 ft
Total Well Depth 20 ft
10-23 ft
Total Well Depth 23 ft
7-22 ft
Total Well Depth 25 ft
1-16 ft
Total Well Depth 17ft
Compounds Identified
 9. Down-gradient  None Detected
    (Well Cluster A)  Total Well Depth 18 ft
 10. Down-gradient  20-25 ft
    (Well Cluster A)  Total Well Depth 25 ft
                                                              11. Down-gradient
                                                                 (Well Cluster B)
                                                              12. Down-gradient
                                                                 (Well Cluster B)
                                                              13. Down-gradient
 14. Down-gradient
None Detected
Total Well Depth 13 ft
15-21 ft
Total Well Depth 22 ft
12-28 ft
Total Well Depth 30 ft

10-29 ft
Total Well Depth 32 ft
1,1,1 -Trichloroethane,
Trichloroethylene,
Tetrachloroethylene,
Toluene, Ethylbenzene
Tetrachloroethylene,
Toluene, Ethylbenzene
Trichloroethylene,
Tetrachloroethylene
1,1,1 -Trichloroethane,
Trichloroethylene,
Toluene, Ethylbenzene
                    1,1,1-Trichloroethane,
                    Trichloroethylene,
                    Tetrachloroethylene,
                    Toluene, Ethylbenzene
Ethylbenzene, Toluene

1,1,1-Trichloroethane
Trichloroethylene,
Toluene, Ethylbenzene
Tetrachloroethylene
Toluene, Ethylbenzene
INITIAL SAMPLING AND
ANALYSIS OF GROUNDWATER

  Immediately following the development of the wells by
pumping,  a groundwater sample was  obtained for pre-
liminary analysis on the  portable GC.  Samples were an-
alyzed using the headspace vapor technique.1'12)  The re-
sults of this initial field analysis are compared, in Table
II,  to subsequent laboratory  analysis  by gas chromato-
graph/mass spectrometer analyses obtained several weeks
later.
  The advantages afforded by the initial field analysis of
the groundwater samples included the following:
  (1) The data generated by the previous analysis of soil
      samples during well installation could be confirmed.
  (2) Those wells yielding samples with  very high con-
      centrations of contaminants could be identified so
      that the necessary special arrangements for handling
      and analyzing those samples could be made.
  (3) The data obtained by analyzing  samples from the
      newly installed wells could be evaluated for public
      health implications.
  (4) The field analyses provided an  additional  quality
      control  element   for  the subsequent  laboratory
      analysis of samples.

-------
                                                                  SITE INVESTIGATION & ASSESSMENT     51
                                                      Table II.
                                      Comparison of Field Analysis of Groundwater
                                           Samples with Laboratory Analysis'1*
    Well Location
                         1,1,1-Trichloroethane
                                             Trichloroethylene
                                                               Tetrachloroethylene
                                                                                      Toluene
                                                                                                    Ethylbenzene
Up-gradient
Dp-gradient
Up-gradient
Up-gradient
On-site
On-site
Qn-site
On-site
Down-gradient
Down-gradient
Down-gradient
Down-gradient
Down-gradient
Down-gradient
ND (ND)
ND (ND)
ND (ND)
ND (ND)
High (2300)
ND (11)
ND (ND)
Low (80)
ND (ND)
Low (80)
ND (ND)
Low (80)
Moderate (390)
Low (100)
ND (ND)
ND (ND)
ND (ND)
ND (ND)
High (8500)
ND (ND)
Low (15)
Low (300)
ND (ND)
Moderate (1700)
ND (ND)
ND (ND)
Low (35)
Low (60)
ND (ND)
ND (ND)
ND (ND)
ND (ND)
Moderate (1300)
Low (20)
Low (60)
High (3600)
ND (ND)
High (3500)
ND (ND)
ND (ND)
ND (ND)
Moderate (1175)
ND (ND)
ND (ND)
ND (ND)
ND (ND)
High (8600)
Moderate (1000)
Low (15)
High (4000)
ND (ND)
High (9500)
ND (ND)
Moderate (1000)
High (2500)
High (18,000)
ND (ND)
ND (ND)
ND (ND)
ND (ND)
High (2200)
Low (200)
ND (ND)
High (1200)
ND (ND)
High (3600)
ND (ND)
Moderate (400)
Moderate (400)
High (1800)
  ND = Not Detected


  1. Field Analysis Performed with portable GC; concentrations given are relative estimated

    Laboratory analysis performed by GC/MSj concentrations in ppb in parentheses.
Groundwater Sampling Results

   All the newly installed  groundwater monitoring wells
were sampled for EPA priority pollutant analysis. Specific
analytical results cannot be presented in this paper, how-
ever the general trends can be discussed.

   The results indicate that there is no groundwater con-
tamination to  the north, west or southwest of the Site.
There is, however, contamination on-site and to the south-
east and east of the Site.

   These  general observations confirm  the conclusions
drawn from the Site hydrogeology as briefly discussed in
this paper. In  summary, groundwater upgradient  of the
Site is uncontaminated and after entering the Site picks up
contaminants and flows southeasterly toward the river.

   Contaminants with different chemical properties will ex-
hibit different characteristics in groundwater. The results
of the priority pollutant analysis indicated that not  all
volatile organics followed similar on-site/off-site contam-
ination distribution patterns. However, all conformed to
the basic hydrogeologic description of the Site.

  Analysis  of  the  contaminant  data  from  the  wells
screened at different depths indicate that a  large fraction
of the  total contamination in the down-gradient area is
confined in ice-contact deposits overlain by less permeable
soil deposits.
CONCLUSION

  The extent and concentration  of  contaminants in the
groundwater monitoring wells follows the pattern devel-
oped from analysis of the Site performed before the mon-
itoring wells were installed and sampled. This close agree-
ment supports the engineering approach and field analyti-
cal  methods  used by E  & E to  develop  the site hydro-
geologic model.
  The actual levels of volatile organics also closely corre-
late with the results of preliminary soil and groundwater
testing performed by E & E personnel using the  Century
Systems 128 OVA on-site during installation of the well.
The agreement between  the E & E  on-site field  analysis
and the actual  contaminant concentration data supports
E & E's use of the OVA during groundwater investiga-
tions.

REFERENCES

 1.  Spittler, T.M., "Use of Portable Organic Vapor De-
    tectors for  Hazardous  Waste Site  Investigations",
    Second Oil and Hazardous Material Spills Conference
    and Exhibition. Philadelphia, Pennsylvania. Decem-
    ber 2-4,1980.
 2.  Dietz, E.A. and  Singley, K.E., "Determination of
    Chlorinated Hydrocarbons in Water  by  Headspace
    Gas  Chromatography", Analytical  Chemistry,  51,
    September, 1979.

-------
         LOCATION AND PRIORITIZING OF ABANDONED
             DUMP SITES FOR FUTURE INVESTIGATIONS

                                           ANN B. NELSON
                         Monroe County Environmental Management Council
                                         Rochester, New York

                                    RICHARD A. YOUNG, Ph.D.
                                   Department of Geological Sciences
                                        State University College
                                          Geneseo, New York
INTRODUCTION

  The methodology described in this paper to locate and
classify abandoned dump sites was developed in Monroe
County, New York in response to a 1978  county legisla-
ture request to locate potential "Love Canal" type prob-
lems and a 1979 New York State law requiring counties
to identify abandoned sites.  The system  is  designed to
evaluate and prioritize a large number of sites for  more
detailed investigation.  While  not all points will be ap-
plicable in every part of the country, the general approach
will allow government agencies to focus limited resources
on those sites that pose the greatest potential impact on
human health.
  This paper is a brief description of a methodology that
will be fully described in a  final report to EPA to be com-
pleted in 1982.

RATIONALE AND APPROACH

  The highest priority for  the Monroe County study was
to locate sites where potentially hazardous substances in
high concentrations might  be present in areas occupied by
large numbers of people. There are many other situations
where hazardous materials might have a  potential  long-
term effect on humans or  the environment that were ex-
cluded from the highest priority category in the design of
the study, i.e.: the  slow, long-term release of toxic sub-
stances into streams  and lakes  through  ground  water
might be of great concern in  other circumstances. How-
ever, the immediate effect  of these low concentrations on
areas where large numbers  of people live or work might be
negligible. Furthermore, detection of low  concentrations
of hazardous substances in water bodies used for drinking
water or recreation is an area of current  concern for
health or water  quality agencies  and should be handled
under existing programs.  Information bearing on such
long-term  problems is not ignored in the present  study
but would be referred to  the appropriate agency  when
encountered.
  On the other  hand,  public  or  private well water sup-
plies in close proximity to existing or abandoned landfills
are placed in the highest priority category because the po-
tential for a more direct contamination pathway is obvious.
  An integrated study  such  as  the one  undertaken in
Monroe County documents a great deal  of historic in-
formation that could eventually be used for  many pur-
poses. It provides comprehensive baseline data that might
be unrecoverable in future years. Planning activities and
zoning requirements are obvious areas of impact. The in-
formation, once collected, could be used by any agency
whose efforts or priorities might overlap but differ from
the specific goals of the present study.
  The key to the successful implementation of this type of
program  is the careful organization and integration  by
qualified  individuals of all the critical resources that might
be overlooked or ineffectively utilized in a less compre-
hensive effort with  more limited time constraints.  In-
volvement of personnel  from all affected or peripheral
agencies is strongly recommended.
  Historic aerial photographic analysis is the best way to
confirm actual or potential site boundaries. This, in turn,
allows personnel and resources to be focused on those
areas where physical documentation  of sites is reason-
ably accurate.
  The geologic analysis  aids in  both the search for and
prioritization of potential sites, but the  date must be in-
tegrated  with the  information developed  by record
searches, interviews and the careful  development of a
comprehensive site activity profile.
  Finally, this effort is not meant to  replace those tech-
nical studies which  have dealt  with  the  analysis and
prioritization of known hazardous waste sites where a
great deal  of specific information is available on  site
history, contents, and hydrogeology.  The present study
should reduce the need for expensive  drilling and testing
programs and provide a means for their rational  imple-
mentation.

SITE IDENTIFICATION

  In conducting a general survey of abandoned dump
sites, there are many sources of information which must be
used in combination to  identify and/or verify site loca-
tions and to characterize site activity. The Monroe County
methodology uses historic aerial photographs in  stereo-
graphic pairs as the primary resource to develop an in-
formation base, but no single source should be depended
upon entirely. Aerial photographs are the most reliable
resource  for identifying site boundaries,  periods of operar
tion, site drainage, and pre-existing uses or features such
as sand and gravel operations, borrow pits, surface pond-
                                                   52

-------
                                                                  SITE INVESTIGATION & ASSESSMENT    53
ing,  and access routes. However, it is not  always pos-
sible to distinguish dumping  from  "clean"  fill  or con-
struction activity on photographs. The information must
be supplemented by other sources. Reports, maps, records,
interviews and information from the general  public all
provide information that  is useful in constructing a site
activity profile. These sources, used without  air  photo
information, may be insufficient to locate  sites adequately
and generally do not provide accurate information on site
boundaries.  It is the combination  of all  available re-
sources that is the key to accurate site characterization.

Data Sources
  The useful sources of date include:
•Aerial photographs in stereographic pairs  for  selected
years
•Records from local and state  health department and con-
 servation agencies
•Environmental atlases
•Government publications on hazardous waste sites
•County soils maps
•Historic  resources (newspaper  clippings,  articles,  city
 directories, fire insurance maps, plat books, industrial
 surveys)
•Interviews with public officials, agency employees
•Public call-in campaign.

General Survey
  The general  survey is conducted for one USGS quad-
rangle sheet at a time. Historic aerial photographs are in-
terpreted for selected years (generally ten year intervals)
to identify  abandoned  dumps. A  record of  interpreta-
tions is made on mylar sheets attached to one photo in
each stereo pair. These overlay sheets are retained as im-
portant permanent documentation for the study.
  Since it is not always possible to distinguish between
construction, filling, and dumping on the photographs,
sites are initially given  one of six site activity designa-
tions (Table 1). Sites in the "Possible Dump" and "Un-
specified Fill" categories will need supplemental informa-
tion to determine whether dumping actually  occurred.
The mixture of types that occurred for one quadrangle
in Monroe County is shown in Figure 1.
  After the  photos are interpreted for each selected year,
the information from all years must be combined for each
site. A written  record of site activity is compiled and re-
corded on the Site Activity Record (Figure 2). This record
is also used to record detailed information subsequently
gathered for each site. A maximum site boundary map is
prepared using large  scale (1":200') maps. Orthophoto
maps are best for this purpose if available. Lastly, a gen-
eral survey map at the scale of 1:24000 is completed for
each quadrangle by  transferring boundaries  from the
larger scale maps (Figure 1).

GEOLOGIC ANALYSIS

  A general geologic analysis  of  the entire region under
study is  essential to the development of a rational ap-
proach to site prioritization. Such a broad-based analysis
also serves  as  the frame  of  reference for site  specific
studies during the later phases of the project.
  The important components of a general geologic analy-
sis  are  those factors which  directly influence the  pro-
duction,  containment, attenuation or migration of leach-
ate. These generally involve the groundwater system, the
soil or rock permeability,  and the structures  within the
                                          TABLE 1: Site Activity Categories
 D—Identifiable:
   Sites where information on dumping activity is known
 from public records, interviews  with  government or in-
 dustry  officials,  the  public  call-in campaign, industrial
 surveys, or where dumping activity is clearly evident on
 aerial photographs.

 P—Possible:
   Sites where filling activity is evident but there has been
 no confirmation as to  whether or not dumping has oc-
 curred. However, based on the location of the  site and
 peripheral land use,  it would appear that dumping could
 have occurred. Sites located adjacent to industrial or  Com-
 mercial activities, maintenance areas,  large construction
 sites and public  facilities such as sewage treatment plants
 and incinerators should be evaluated as  possible dumps.

 U—Unspecified:
   Sites that are apparent either as  recent surface dis-
 turbances or topographic changes that were not present on
 earlier photographs. Sites that are obviously clean fill for
 construction purposes are not included in this category
nor are they annotated. (Such sites may be identified by
the relatively rapid completion of activity followed by the
appearance  of a highway, new building or structure on
more recent photographs.)

L—Lagoons:
  Potential  liquid waste disposal areas that are either sug-
gested by associated activity on the photographs or are
known to have existed. Standing  water in borrow pits or
quarries  is  not generally  placed  in  this category unless
associated with dumping.

J—Auto Junkyards and Salvage Areas:
  Such sites may contain significant surface disposal or
spills of oil,  transmission and hydraulic fluids, or solvents.

S—Suspicious:
  Areas  where unusual or unidentifiable activity has oc-
curred that is not readily recognizable. Sites are placed in
this category pending more complete analysis that will re-
sult in one of the above designations or elimination.

-------
  54    SITE INVESTIGATION & ASSESSMENT
                                   IS*
                              »:
                                  *
                        Figure 1
   General survey map of sites in the Rochester West quadrangle.


 overburden or rock that control either the direction of
 movement, rate of movement or local  concentration of
 fluids. In most cases, landfills or old dumps are located
 in unconsolidated soils or overburden  but occasionally
 the character of the local bedrock is also significant. Any
 geologist working on  a project  such as this  needs to
 evaluate the critical  factors within the  particular region
 under study.
   When dealing with abandoned landfills that were start-
 ed or completed prior to the late 1970's, it should be as-
 sumed that wastes  were generally placed in natural or
 man-made  depressions without utilization of sophisticated
 preparation or containment techniques. Thus, the proper-
 ties and structures of the natural overburden and  rock
 formations should be assumed to control the natural mi-
gration of leachate and groundwater adjacent to the site.
While the Monroe County study emphasizes the problems
encountered in the glacial overburden deposits bordering
Lake Ontario, the basic approach should be applicable,
with minor modifications, to any regional analysis of hy-
drogeologic conditions.
  The first  phase of the general geologic  analysis  con-
sists of the  collection of all readily available published or
unpublished data for the construction of regional  geologic
maps such as those shown in Figures 3, 4 and 5. The col-
 lection of additional unpublished data should  continue
 throughout the period of study because useful or unique
 information can continue  to  be located.  The  types of
 sources commonly used in the development of a geologic
 data base for both general or site-specific geologic models
 are:
 •Geologic maps and reports:  published literature  and
  student theses/dissertations; open-file reports of agencies
  such as United States Geological Survey; government
  documents.
 •County  soils  maps: old and new  versions  of United
  States Department of Agriculture maps describing  soil
  types and locations  of dumps  or "made land"  (good
  for shallow depth of overburden only).
 •Aerial photographs and topographic maps: used for geo-
  morphological analysis of landforms if need  to supple-
  ment or confirm  published data. Compare old and re-
  cent prints or maps, if available.
 •Engineering data from public  or private agencies or
  firms (especially exploratory boring logs)—
  Highway/bridge plans or surveys
  Town, village or county construction/maintenance pro-
  jects
  Utilities, railroads, pipeline companies, canals
  Drilling and soil testing from engineering firms
  Architectural firms
  Oil, gas and water  well drillers
  Sewer and water district agencies
  Mining, quarrying and tunneling ventures
  Public buildings, airfields and large industrial firms
Geologic Overlay Maps

  Using  all available sources,  maps  such as  those in
Figures 3,  4 and 5 are constructed. It is recommended
that a common scale be adopted for all such general sur-
vey map products and  that they be compiled on trans-
parent overlays so that all types of information can be
easily transposed onto or compared with the general survey
site maps. Topographic  maps with a scale of 1:24,000 are
very convenient for this purpose as much geologic data
are routinely published  on this scale by state and federal
agencies.
  These maps and the data from which they are generated
have the following uses:
•Development and improved understanding of a regional
 groundwater model  that includes depths  to the water
 table and flow directions
•Development of a regional perspective concerning areas
 where landfill or leachate problems would either be re-
 duced or aggravated by local geologic conditions
•Development of a basis upon which to prioritize or or-
 ganize the search for sites with potential for significant
 impact
•Allowing an agency to focus  its limited resources in the
 most critical areas
•Extrapolation of subsurface conditions to sites where no
 data can  be located from similar ones  where detailed
 information exists

-------
                                                                          SITE INVESTIGATION & ASSESSMENT     55
              SITE  ACTIVITY RECORD
                                               OFFICE at CM.Y
               OKOV 
-------
 56    SITE INVESTIGATION & ASSESSMENT
 Geologic Ranking Process
   Once all the general maps have been prepared and sub-
 surface engineering  information compiled,  the  geologic
 ranking process can  be applied using the Geologic Rank-
 ing Sheet (Figure 6). This ranking sheet  has  been devised
 for the following reasons:
 •To provide a means of organizing the process of site
  comparison
 •To minimize inconsistencies and  oversights that could
  occur when dealing with large amounts of data
 •To provide a record for other project personnel that can
  be updated  and will allow discussion or review of site
  characteristics.
   A ranking  scheme such as this is basically an informa-
 tion gathering and documentation  device that would be
 especially useful if project personnel changed.
   Strict numerical site rankings should be considered only
 as approximations of site characteristics  and geologic ex-
 pertise should be substituted where appropriate. The sys-
 tem described here requires knowledgeable decisions to be
 made for each of the "presumed effects" and the ranking
process should be done by an individual with hydrogeo-
logic background.
  The ranking sheet was designed to divide potential sites
into high, intermediate or low priorities based on antici-
pated  geologic  conditions governing leachate  (ground-
water) occurrence,  production, migration  or accumula-
tion, but it is only one component of an integrated assess-
ment system. Its  prime consideration is the potential or
inferred effects that leachate contamination would have on
people near the sites.
  High priority sites are those where hazardous materials
might accumulate, move readily at shallow  depths, or re-
appear  at the  surface in concentrated  amounts. Low
priority sites are those adjacent to lakes, ravines, or on
steep slopes and presumed to have been well drained for
an extended period. Any large site with known hazardous
contents should be evaluated individually and referred to
the appropriate agency regardless  of  its  position  in the
geologic ranking system.
  Intermediate  sites are  those  falling between the ex-
tremes described  above. They would generally  be sites


                             Legend

                       — 5oo- Bedrock contours

                       »•*,  Sands, gravel (shoreline!

                        ;i^J» Moraine

                        —•— Iroquois shoreline

                       *-— Damon shoreline

                           Newfane beach

                           Lockport Dolomite

                       j~i i Solution zone (Lockport

                           Subsurface data point
                                                      Figures.
Subsurface bedrock contour map with selected surface geology for Rochester East and Rochester West 7'/i minute topographic quad-
Angles. Contour interval 25 feet. Horizontal distance across map is 12.6 miles. A: Culver-Ridge Shopping Center Site; B: WeilandRond

-------
                                                                  SITE INVESTIGATION & ASSESSMENT     57
where geologic conditions contain and isolate any leach-
ate or partially attenuate its effects.
  Potential water well contamination involves geologic
considerations but  sites  near public  or  private drinking
water supplies are treated separately due to the potential
direct risks involved.
USE OF THE RANKING SHEET

  Each factor  on  the  ranking sheet is  analyzed as to
whether it increases or  decreases the potential risk for a
site. Available  engineering information and potential in-
terrelations among factors need to be considered. When
the sheet  subtotal is computed, the "additional factors"
are evaluated and a total score computed. It is important
that borderline cases be carefully evaluated and a tenta-
tive rank  assigned if necessary, pending further data col-
lection. Special problems should be noted in the space
provided at the bottom of the sheet.
  This system is designed to deal with a large number of
abandoned sites where information is scarce or difficult to
locate. It is not meant to replace the elaborate systems
for ranking known hazardous  waste sites such as those
by LeGrand(1) and Kufs et al.(2) The Monroe County study
is intended to broadly prioritize sites, rapidly identify po-
tentially high risk sites  and allow the most efficient use of
funds available for later  testing or subsurface investi-
gations.


INTEGRATION OF WATER WELL DATA

  Upon completion of the site  identification  phase, there
will be sites  categorized as "Possible Dumps" or "Un-
specified  Fills." Additional information must be  ob-
tained in order  to clarify which  of these sites were actually
used as dumps. Information can be obtained from existing
agency  records,  interviews  with  local  officials,  waste
haulers,  industrial  employees, historic  documents and
street and business directories.
  Once the site categories are refined, the process outlined
in Figure 7 is  followed to determine which dump sites
could impact private or public water wells. Where private
                                                     Figure 4.
Generalized groundwater contour map for Rochester East and West topographic quadrangles. Twenty-five foot contour interval shows
water elevations in overburden and may reflect some seasonal variations in data or perched water. Same scale as Figure 3.

-------
58     SITE INVESTIGATION & ASSESSMENT
wells are found within 1000 ft. of a known dump, or pub-
lic water wells within 0.5 miles, geologic conditions  are
carefully evaluated to  determine if contamination  could
occur. When this  determination is positive, the sites  are
referred  to the appropriate local  or state agency  for
groundwater testing. Sites which do not appear to impact
drinking water supplies are then classified and prioritized
according to the method described below.
SITE CLASSIFICATION

  Site classification takes place throughout the study. Ini-
tially, each site is classified as Major or Minor at the time
of photo interpretation according to the following criteria:

Major:
•Greater than one acre with measurable depth
•With known chemicals regardless of size
•With drums or containers
•With lagoons, bermed pits
Minor:
•Less than one acre
•Random surface activity greater than one acre with little
 accumulation
•Low volume activity on steep slopes

  Sites are classified according to current land use at the
time the boundaries are transferred  from the photos to
the orthophoto  maps.  One of six categories is  applied.
These will eventually need to be verified by field inspec-
tion.
•24-hour occupancy on or within 100 ft. of the site (com-
 mercial establishments, industry)
•24-hour occupancy within 1000 ft. of site
•Part-time occupancy within 1000 ft. of site
•24-hour occupancy within 1000-2500 ft. of site
•Part-time occupancy within 1000-2500 ft. of site

  At the time the geologic analysis is performed,  the sites
are ranked geologically based on the potential for  leachate
impact and migration. The geologic ranking system is de-
                                                    Figure 5.
Overburden thickness map for Rochester East and West topographic quadrangles from same data points as Figure 3 R designates MX*
outcrops. Same scale as Figures 3 and 4.

-------
                                                                                          SITE INVESTIGATION & ASSESSMENT      59
                                          GEOLOGIC RANKING SHEET
                           FOR GENERAL COMPARISON OF ABANDONED LANDFILL/DUMP SITES
                                                                                                            SITE NAME/NO.
t/i
PRESUMED =
EFFECT1 £
. INCREASED
HAZARD
„ INTERMEDIATE
(UNCERTAIN)
r DECREASED
• HAZARD
OVERBURDEN ,,
GEOLOGY






ESTIMATED ,
PERMEABILITY






RELIEF, .
GEOMORPHOLOGY






a:
UJ
oi
0
n: z
1— 3
LU ce
Q C3






GROUNDWATER
GRADIENT6






BEDROCK ?
CHARACTER






SOIL MINERAL-
OGY TEXTURE8






LJ_
O 
-------
  60      SITE INVESTIGATION & ASSESSMENT
                                                                    OBTAIN PRELIMINARY SITE
                                                                 INFORMATION THROUGH  INTERVIEWS
                                                                AND RESEARCH OF EXISTING RECORDS
                                                                TO CONFIRM PHOTO INTERPRETATIONS
                                                               AND REFINE ACTIVITY CLASSIFICATION
                            TRACK I  (PUBLIC HATED HELLS)
                                                   PROCEED TO TRACK 11
                                                   PROCEED TO TRACK II
                        REFER TO APPROPRIATE
                         AGENCY FOR TESTING
                        OF WELL AND DETAILED
                         SITE INVESTIGATION
PROCEED TO TRACK II
                                                                                                $1 TRACK ii (PRIVATE WATER WELLS)
                                                                                              RESEARCH ADDITIONAL
                                                                                                INFORMATION ON
                                                                                                 SITE CONTENTS
                                                                                              REFER TO APPROPRIATE
                                                                                               AGENCY FOR TESTING
                                                                                              OF WELL AND DETAILED
                                                                                               SITE INVESTIGATION
                •PU81IC WTFR MILS FROM V to J HUES
                 ua.'ot PRIMTI turn KEIUS FBJU wa-
                 it : «u SHOULV K nmmaefiv
                 IMLVlCV F0( ial(M GtDUWIC
                 BELFWVT
                                                                        Figure 7.
Flow chart for investigating water wells  near potential or documented abandoned dump sites.  Distance limits  are arbitrary and may K
changed to fit circumstances.

-------
                                                                  SITE INVESTIGATION & ASSESSMENT    61
scribed under the Geologic Analysis section of this report.
  The information on each site is combined into a final
classification system,  incorporating site  identifying  in-
formation.
Sample Site Code No:
D_
A
10
B
RO
 C
6_
D
3_
G
2_
H
A:  Type of site activity (dump, possible dump, unspeci-
    fied fill, lagoon, junkyard)
B.  Quadrangle sheet number (assigned)
C.  Municipality: two letter code (assigned)
D.  Site number (assigned)
E.  Proximity to wells: (private wells within 1000 ft.; pub-
    lic wells within 0.5 miles)
F.  Major or minor
G.  Land use: 24-hour or part-time occupancy on or adja-
    cent to site
H.  Geology: impact of leachate
  Sites  that could impact nearby wells are  immediately
referred to local and state health agencies for testing. The
remaining  sites  are  prioritized in a matrix  that  weighs
geologic and land use impact, size, and type of activity.

CASE HISTORY

  One of the proposed uses of the general geologic analy-
sis  and map  products is the development  of geologic
models from available engineering data that can be extra-
polated to sites  where subsurface information is  inade-
quate. The following example illustrates this approach.
  A cross-section through an abandoned landfill located
on a former proglacial lake shoreline is shown in Figure
8. This abandoned site,  now beneath a paved shopping
area, was the subject of studies related to construction of
a deep sewer project. Descriptions  of former gravel  de-
posits in borrow pits nearby were also located in old geo-
logic publications.
  The  Weiland  Road landfill, partially abandoned and
built over, but still used, in part, by a large  corporation
as a private solid waste disposal site is shown in Figure 9.
The Weiland Road site is located on a similar, but higher,
abandoned glacial lake shoreline a few miles to the west of
the Culver Road site. The  approximate location of both
areas is shown on Figure 3.
  The  Monroe County study documented photographi-
cally the extent  of the Weiland Road site  and  described
the anticipated subsurface  geologic conditions  using  the
Culver Road site for comparison. This information was
used by state and county agencies to evaluate procedures
proposed by the company to upgrade the existing leach-
ate collection system. It was postulated that wave energy
in the  former  glacial lakes  would have left more perme-
able sediments at  depth than had  been encountered or
described in preliminary borings. Borings for  a county
highway relocation project adjacent to the site strengthen-
ed this hypothesis which resulted in the company drilling
additional borings in suspected critical areas.  Sandy sedi-
ments  were located at depths  of ten  feet or  more below
the existing fill. This new subsurface information con-
    450 -,
    430 -
    410 -
    390 -
    370 -
    350 -
    330 -
                                                                                                         N
                                                                            IDGE ROAD
                             Wei 1  numbers
                     ~~j  FINE    SAND  v.^
                       -payers)     <-->

                          <-,(varves_),S
             200 ft.
                       ROCK
                                             i_
                                    (observati on
                                       wel1s)
                                    GENERALLY
                                      COMPACT
                                        GRAY  TO
                                          BROWN
                                           [lacustrine)
                                                        SILT
                                                                        VERY  DENSE GRAY-BROWN TILL
                                                             LANDFILL
                                                             AND FINE
                                                               SAND
                                                                    COARSE  TO  FINE  GRAVEL
                                                                     AND  SAND:  MANY  COBBLES

                                                                          (beach  gravels)
                                                                  '^MEDIUM TO FINE
                                                                    "'•-..RUNNING SAND
                                                                      "	
                                                                                    Artesian
                                                                                     water
                                                                                      pressure
                                                                            ROCK
                                                     Figures.
 Hydrogeologic conditions and interpretive geologic cross section of Culver-Ridge Shopping Center site developed from sewer interceptor
 contract studies. Groundwater monitoring and soil investigation wells indicated by vertical lines. Location A on Figure 3.

-------
 62    SITE INVESTIGATION & ASSESSMENT
                                           BEACH SANDS  AND
                                              GRAVELS
                                     Weiland    ^  SITE  AREA
                                      Road
   SILTY  SANDS
       Vv

              WEATHERED OR
               BOULDERY ZONE  AT TOP OF ROCK
                                                                                        (Not  to scale)
                                                    Figure 9.
 Diagrammatic geologic cross section of the Weiland Road solid waste disposal site illustrating the inferred relations between ground-
 water and surficial deposits prior to site development. Arrows denote subsurface groundwater migration. Location B on Figure 3.
firmed the preliminary geologic analysis and permitted
an evaluation of the existing leachate collection system.
  As a result of the recommendations of a joint commit-
tee composed of participating state and county agencies,
a more comprehensive groundwater sampling  plan was
voluntarily developed and implemented  by the owner.
Further investigations are being considered to  evaluate
the effects of  the landfill on nearby wells in  bedrock
located near geologic structures striking toward the land-
fill. This structural zone was unsuspected prior to recent
completion of subsurface engineering studies unrelated to
the landfill.

SUMMARY AND CONCLUSIONS

  In the  final analysis, no general program for  identify-
ing abandoned  dump sites  can  ever completely determine
the location  and  contents of  all hazardous waste dis-
posal areas. The  Monroe  County approach is  designed
to provide general  information  on a large number of
previously undocumented sites, as well as a method for
local, state and  federal agencies to prioritize sites for more
detailed site investigation and testing. Application of the
procedures will  reduce the expense of costly drilling and
testing programs by focusing resources on the most criti-
cal sites.

ACKNOWLEDGMENTS

  This study  has  been conducted under  the direction of
the Monroe County Landfill  Review  Committee,  com-
prised of representatives of the Monroe County Environ-
mental Management Council, the Monroe County De-
partments of Health and Planning, the  New York State
Departments of Health and Environmental Conservation,
the City of Rochester Department of  Community De-
velopment and the Industrial Management Council.
  The  development of the methodology has been made
possible through the financial support of the New York
State  Department  of  Environmental  Conservation in
1979 and 1980, and currently through the United States
Environmental Protection Agency's Environmental Mon-
itoring Systems Lab in  Las Vegas, Nevada (Contract No.
14043 administered by Lockheed Engineering  and Man-
agement Services Company, Inc.).
REFERENCES

1. LeGrand, H. E., A Standardized System for Evaluat-
   ing Waste Disposal Sites: A manual to accompany de-
   scription and rating charts. National Water Well As-
   sociation, Worthington, Ohio, 1980.
2. Kufs, C., Twedell, D., Paige, S., Wetzel, R., Spooner,
   P., and Colonna, R., "Rating The Hazard Potential of
   Waste Disposal Facilities." Proceedings,   U.S.  EPA
   National Conference on Management of Uncontrolled
   Hazardous Waste  Sites,  October 15-17, 1980, Wash-
   ington, D.C. Hazardous Materials Control Research
   Institute, Silver Spring, Md. 31-41.

-------
      SELECTION OF MONITORING WELL LOCATIONS IN
         EAST AND NORTH WOBURN, MASSACHUSETTS
                                          DAVID K. COOK
                                    Ecology and Environment, Inc.
                                       Woburn, Massachusetts
INTRODUCTION

  Potentially hazardous materials have been generated,
stored, used and disposed of in East and North Woburn,
Massachusetts and  portions of the  surrounding towns
(Figure 1) for over 120 years.
  Of particular interest is the Industri-plex site situated in
North Woburn  which has a  long history of chemical
manufacturing activity. In 1853, Robert B. Eaton's Chem-
ical Works began to produce chemicals for the textile,
leather and paper industries of New  England. In 1863,
the property was purchased by Merrimac Chemical Com-
pany which, by 1929, grew to be one of the largest chemi-
cal manufacturers in the country. Products of this complex
      t      N        \       ^
             \   WILMINGTON     ^   \ \

                                 L \

                                 \\
                                  \ \  READING
                                   \\
                                    \\
                                    \
            s"
 \
 \
 \
  \


" \
 STUDY \
 AREA *"\
                                         l
                                         \ I STONE HAM
 V
      T
                        WINCHESTER

                                    \   MEDFORD
                     Figure 1.
                   Site Location
                  (Scale: 1:48,000)
included lead-arsenic pesticides, acids, TNT and various
heavy metal compounds.
  Between 1934 and  1968, ownership of the property
changed hands several times, and the production of animal
glue and grease became the primary activity.
  Present industries within the defined study area (Figure
2) include three chemical manufacturers, a tannery,  a
tallow manufacturer, three drum reclaiming operations, a
waste oil company, several auto junkyards, a truck wash-
ing operation and  the City of Woburn sanitary landfill.
In addition, several suspected illegal  hazardous  waste
dump sites have been identified within the study area.
  As a result of present and past industrial and  illegal
dumping activities, a very productive aquifer  tapped by
approximately 200  industrial, private and municipal wells
has been  contaminated. Two  of Woburn's  municipal
drinking water wells were closed in late 1979 when signifi-
cant amounts of chlorinated organic solvents were  de-
tected in the water.
  As a result of the detection of groundwater contamina-
tion in Woburn, Ecology and Environment, Inc. (E & E)
was authorized by EPA (Contract  # 68-01-6056)  to in-
vestigate the sites  of fifteen existing industries to  deter-
mine if they were sources of the contamination. No direct
sources of the previously detected contamination were
discovered during the site inspections. Because of the com-
plex industrial history of the area, it is very likely that the
detected contamination was introduced at some time in
the past from a source no longer in existence.
  E & E was then  asked by the EPA to evaluate ground-
water quality in the entire East and North Woburn area.
It was decided that a complete hydrogeologic investiga-
tion of the area would be necessary to identify past or
present sources of  contamination  and to recommend  ap-
propriate remedial  action. The purpose of this paper is to
outline  the approach developed  by E  & E for this in-
vestigation, especially the siting of additional groundwater
quality monitoring  wells.
                            BACKGROUND INFORMATION GATHERING

                              The study area (Figure 2) of approximately ten square
                            miles is a very large area for a complete hydrogeologic in-
                            vestigation. Route 128 divides the area into North Woburn
                            and East Woburn. The Industri-plex area as well as the
                                                  63

-------
64     SITE INVESTIGATION & ASSESSMENT
                         Figure 2.
        Sites includes in the groundwater investigation of
                East and North Woburn, MA.
 majority of industrial sites are located north of Route 128,
 and this area  was once considered to be the largest in-
 dustrial complex in the world. South of Route 128 is pri-
 marily  residential, and it  is within this  area  that an ele-
 vated level of childhood  leukemia has  raised suspicions
 regarding  the  quality  of  the city's drinking water. The
 Aberjona  River passes from north to south  through the
 industrial  complex and the residential area and empties
 into the Mystic Lakes which  are  used extensively  for
 recreational purposes.
   To determine the need for and locations of monitoring
 wells in an  area  as large  as Woburn,  a detailed  back-
 ground investigation was performed. All chemicals which
 might be encountered during preliminary sampling of ex-
                                                                                   Key for Figure 2
                                                                       Sites Included in the Groundwater Investigation
                                                                          East and North Woburn, Massachusetts

                                                              1.  Truck washing operation
                                                              2.  Drum reconditioning plant
                                                              3.  Chemical warehouse
                                                              4.  Drum reconditioning plant
                                                              5.  Septage receiving station
                                                              6.  Paint and lacquer thinner plant
                                                              7.  Chemical manufacturer
                                                              8.  Chemical manufacturer
                                                              9.  Tannery
                                                              10.  Waste oil company
                                                              11.  Drum reconditioning plant
                                                              12.  Used auto parts
                                                              13.  Degreasing plant
                                                              14.  Former City of Woburn sanitary landfill
                                                              15.  Present location of hide piles
                                                              16.  Chromium lagoon
                                                              17.  Former location of hide piles
                                                              18.  Arsenic pits
                                                              19.  Arsenic trioxide drums (now removed)
                                                              20.  Unlabelled barrels and drums along unpaved road
                                                              21.  Food processing plant
                                                              22.  Stoneham—Sweetwater Brook
                                                              23.  Tallow company
                                                              24.  Drum burial site
                                                              25.  Drum storage area
                                                              26.  Present location of Woburn sanitary landfill
isting wells were identified during a process evaluation of
existing and historical industrial activities. A lexicological
evaluation of the identified chemicals was then performed
to ensure the  safety  of all  individuals  involved  in the
sampling of existing wells and  installation of additional
monitoring wells.
  All available  information regarding the  geology and
hydrology of the area was assembled. Included were topo-
graphic maps, aerial photographs from  1938 to 1980, a
bedrock outcrop map, numerous well logs, USGS hydro-
logic data reports, groundwater quality analyses and litera-
ture on  the  bedrock geology. To  fill in  the gaps  in the
literature, a field survey was performed to investigate the
nature  and  competency  of all  bedrock outcrops and
gather additional data regarding the unconsolidated glacial
deposits overlying the bedrock.
  Although  drillers logs for 123 wells within the  study
area were located, most wells were found to be screened
in the glacial materials and  very little depth to bedrock
data was available. Previous experience regarding the in-
vestigation of  organic solvent contamination of ground-
water has shown that the solvents tend to migrate down-
ward through  the aquifer to the bedrock/overburden in-
terface and into the bedrock if  it is highly fractured. Be-
cause of this, an accurate picture of the  bedrock surface
configuration is essential to an evaluation of contaminant
migration.
  A seismic survey was performed to gather the necessary
depth to bedrock information.  The  performance of thii
survey was hampered by thick deposits (up to 50 feet) of
peat along the Aberjona River  and in the area formerly
occupied by  Lake  Mishawum  in North Woburn. The
presence of heavy  traffic and  various  utilities also in-
creased  the  challenge of locating appropriate sites for

-------
                                                                   SITE INVESTIGATION & ASSESSMENT     65
 Locations of Wells Presently In Use
 Locations of Unused But Accessible Wells
 Locations of Inaccessible Wells       7
 Proposed Monitoring Well Locations
 Proposed Surface Water Sampling Locations
- Swampy Areas
                        FigureS.
    A portion of the base map of the Woburn, MA. study area
seismic lines. However, appropriate data were recorded
and later supplemented by existing well and bedrock out-
crop data to produce a bedrock surface elevation contour
map. The seismic survey also provided data regarding the
water table depth within the study area.
  The locations of  all 123 inventoried wells were in-
spected to determine which of the wells could be sampled.
Of these, 32 wells were found to be accessible  and a
sample from each was taken following  approved EPA
procedures. Due to the complex industrial history of the
site, each sample received a complete priority pollutant
analysis  at  an  EPA  approved  laboratory. Preliminary
screening of all samples was performed on  site by E & E
using a portable organic vapor analyzer. This procedure
ensured  the identification  of highly volatile  organics
which might be lost  if a delay of the complete analyses
occurred.
SELECTION OF MONITORING WELL LOCATIONS

  Using the assembled background information, a series
of maps was generated to aid in the selection of appropri-
ate sites for groundwater quality monitoring wells and to
develop the appropriate  specifications for the installation
of these wells.
                                                              Using color stereo  aerial photographs taken on No-
                                                           vember 6, 1980, a  base  map of major roads and rail-
                                                           roads having a scale of 1:9600 was prepared. The loca-
                                                           tions of all inventoried wells were plotted on this map. In
                                                           addition,  the surface drainage of the area was plotted on
                                                           the base map using  the stereo imagery.  A portion of the
                                                           base map is  shown  in Figure 3 to illustrate the types of
                                                           information contained in it.
                                                              A uniform set of well logs was constructed from the
                                                           existing data to facilitate correlations between logs and
                                                           the construction of geologic cross-sections.  Because of
                                                           the large  area of the  site, twelve cross-sections  were re-
                                                           quired to adequately characterize the area. Figure 4 is an
                                                           index map of the prepared cross-sections  and Figure 5
                                                           shows a portion of one of the sections. The sections were
                                                           prepared  with a horizontal scale of  1:9600 so that  they
                                                           could be used in conjunction with the base map and all
                                                           maps derived from it.
                                                              The geologic map presented in Figure 6 was prepared
                                                           from the literature and field survey of all outcrops within
                                                           the study area performed by E & E personnel.
                                                              Using the  1:9600  scale base map, a  ground elevation
                                                           contour map, water table contour map  and bedrock sur-
                                                           face contour map were prepared.  United  States Geological
                                                           Survey topographic sheets were used to prepare the ground
                                                           elevation  contour map.  Water  table elevation measure-
                                                           ments taken by E & E personnel were  supplemented by
                                                      SCALE: r4sooo
                                                      ©Well Locations  A	A' Cross-Sections
                                                                             Figure 4.
                                                                       Cross-section locations
                                                                          (Scale: 1:48,000)

-------
 66    SITE INVESTIGATION & ASSESSMENT
                                                                           NOTE: SYMBOLS MAY BE COMBINED.
                                                     Figure 5.
                                                 Cross-section D-D'
data  from the available literature logs  and the seismic
survey to produce the water table contour map.
  The surficial deposits are controlled to some extent by
the bedrock topography, the most prominent feature of
which is a fault-controlled valley. This valley was  prob-
ably widened and deepened during Pleistocene glaciation.
Locally, the valley has very steep walls,  but throughout
most of the study area, it has gently sloping walls.
  The areas adjacent to the valley are mantled with glacial
till  consisting  of  poorly sorted sand,  silts, clays  and
cobbles. The bedrock valley contains glacial deposits up to
160 feet thick overlain by  modern  alluvial deposits.  The
glacial material in the valley consists of outwash deposits
of fine- to coarse-grained  sands and gravel interbedded
with fine  sand, silt, clay and till. Well log  data indicate
that the sediments generally become coarser with depth.
Upper and central portions of the valley contain finer
sands, silts and clays, while the  coarser sands, gravels  and
till  are located at  the  bottom and  long the walls of the
valley. Much  of the  finer materials  was deposited by
meltwater streams which sorted and reworked the valley
material.
  Modern alluvium associated  with the Aberjona  River
System consists of peat deposits and bogs located  along
the  current course of the river and its tributaries. Topsoil
and artificial fill make up the top five to ten feet of the
surface deposits except in North Woburn where extensive
reworking of the sediments by man has severely disrupted
the surficial deposits.

Bedrock Surface Topography

  The northern one-quarter of the study area is charac-
terized by four shallow bedrock troughs which meet to
form  one deep  north-south  trending trench. The four
troughs slope  toward the deep  trench from the general
                                SCALE |mllM|
                             0123
  KEY
|+ +1 Salem Gabbro-diorite

|J"_.-ii| Dedham Granodiorite

l/Xv/J Precambrlan  Schists  and Gneisses

 v^»  Thrust Fault

s**  Fault
 .^-^  Contacts
                   Figure 6.
         Geologic map of Woburn, MA

-------
                                                                 SITE INVESTIGATION & ASSESSMENT    67
                       Figure 7.
        A portion of the ground elevation contour map
              of the Woburn, MA. study area
areas of the two branches of the Aberjona River to the
northeast, the East Drainage Ditch to the north and the
Woburn  Sanitary Landfill to  the northwest.  The deep
trench is approximately two miles  long and  reaches a
depth of approximately 100 feet  below the ground sur-
face providing a possible trap  for  groundwater  con-
tamination.
  The southern portion of the area is characterized by a
deep trench sloping from  the  Stoneham/Woburn boun-
dary  to  Whittemore  Pond and  into Winchester  with
depth increasing to 150 feet below the ground  surface at
the Woburn/Winchester boundary. A shallower  trench
under the Sweetwater Brook enters the deep  trench ap-
proximately 1200 feet southeast of Whittemore Pond.
  Between  the two  deep bedrock trenches in  the study
area is a bedrock high. All of the bedrock troughs and
trenches in the study area are most likely fault-controlled
as the regional geology indicates  faults trending parallel
to them.  In general, the groundwater and surface water
within the study area flow parallel to the fault shear zones
and are presumably controlled by them.
  At the east and west borders  of the study area, the
bedrock surface rises sharply, and many bedrock outcrops
are present along Route 93 and  west of the  Boston &
Maine Railroad tracks.
                                                         Bedrock Geology
                                                           The Woburn Study Area is situated within the Appal-
                                                         achian  Mountain orogenic  belt  which  extends  from
                                                         Georgia to Newfoundland.  The belt is comprised of  a
                                                         series of plutonic, sedimentary  and volcanic rocks,  Pre-
                                                         cambrian  through Permian in age, most of which have
                                                         undergone multiple deformation and intense metamorph-
                                                         ism. The  Northern Appalachians (New England and the
                                                         Maritime  Provinces) are divided  by Zen(l) into several
                                                         broad linear  structural  features or zones. The Woburn
                                                         Study Area is located in the easternmost zone called the
                                                         Eastern Avalonian Platform.
                                                           The Eastern Avalonian Platform is characterized by
                                                         Late Precambrian metavolcanic  rocks intruded by a series
                                                         of Late Precambrian granitic plutons.  The age of these
                                                         plutons has been radiometrically set at approximately 630
                                                         million years  by Naylor.(2) Unconformably overlying these
                                                         Precambrian  rocks  are  fossil-bearing  lower  Cambrian
                                                         strata composed of shales and carbonates. Superimposed
                                                         on the Platform are several down-faulted blocks or basins;
                                                         the Boston Basin  being nearest to the Woburn area. A
                                                         large thrust belt of Precambrian metavolcanic and meta-
                                                         sedimentary rocks lies at the western edge of the platform.
                                                         The Woburn  Study Area (Figure 1) lies between the thrust
                                                         belt and the Boston Basin.
             Figures.
A portion of the water table contour map
    of the Woburn, MA study area

-------
68    SITE INVESTIGATION & ASSESSMENT
  Relationships between the units described above are dif-
ficult to determine in the field because extensive glacial
material overlies the bedrock resulting in widely spaced
outcrops. In addition, faults have dissected the region into
slices and small blocks, extensively distorting the regional
geology. As a result, the geology of the study area is not
well known, having been  mapped  in reconnaissance fash-
ion only, by Emerson,'3'  LaForge,(4) Bell and Alvord, (5)
andBarosh. etal.(f>t
   Where faulting  occurs, the original character of the
 rock is severely altered to produce a finer-grained, highly-
 foliated rock that  is very, susceptible to weathering. In
 time,  weathering of the  faulted  materials will  produce
 topographic lows or valleys within  the bedrock surface.
 Glacial plucking is  more easily accomplished in the faulted
 material than in the non-faulted  rocks. Within  the  Wo-
 burn area, several  faults exist, some paralleling the re-
 gional trend of the rocks (northeast-southwest) and some
 perpendicular  to the regional trend.  The bedrock lows
 associated with fault zones appear to control a significant
 amount of the  groundwater and surface waters  in the
 Woburn area.
 Groundwater Flow Direction
   Groundwater  flow closely parallels surface water  flow
 throughout the study area. Groundwater from  the west,
             <>GW5
 WILMINGTON
 \
                                     READING
            WOBURN
       Q3Il-10ppb

       E3 w-50
       1=3 50-100

       (S3 100-500

       S3 500-1000

       CJ >1000
                                                                        WINCHESTER
STONEHAM
                        Figure 9.
        A portion of the bedrock elevation contour map
              of the Woburn, MA. study area
                         Figure 10.
             Areal distribution of trichloroethylene
                East and North Woburn, MA


north and east moves  toward the center of the area and
then south into Winchester. Contours  in the Sweetwater
Brook area are close together indicating a strong gradient
from Stoneham into  South Woburn.  Groundwater de-
pressions near the center  of the area and in the southern
portion of the  area represent  the  drawdowns resulting
from the pumping of industrial wells.

SELECTION OF ADDITIONAL
MONITORING WELLS

  All of the data gathered during the procedure described
earlier in this  paper were  used to select sites that will
provide the  maximum amount of  reliable information.
Of primary concern during the  selection of the sites were
the locations of bedrock depressions and  fault zones,
areas with confirmed high levels of  groundwater con-
tamination,  areas  down gradient from suspected sources
of contamination  and areas where  there are no existing
accessible wells. The  locations  chosen  are  shown in
Figure 11. An  accurate assessment  of  groundwater con-

-------
                                                                SITE INVESTIGATION & ASSESSMENT    69
                       0w,o*Qwl
                        * «GW5
              \    WILMINGTON

                               GZ3    \
                                ^.GZ2    \  READING

                              GZ5-X-GZ6    *

                               •M      \
                                  •••
             >**      • Proposed Well Locations   BB
                     * Existing Well Locations      ,
  V


                      Figure 11.
          Locations of proposed monitoring wells

lamination within, entering  and leaving the study area
will be possible using the existing and proposed monitor-
ing wells. From this information,  recommendations  for
appropriate remedial actions may be formulated.
REFERENCES

1.  Zen, E-An, et al., "Studies of Appalachian Geology:
   Northern and Maritime," Interscience Publishers, New
   York, N.Y., 1968.
2.  Naylor,  R.S., "Age Provinces  in the Northern  Ap-
   palachians." Annual Review of Earth and Planetary
   Sciences, 3, 1975, 387-400.
3.  Emerson, B.K., "Geology of Massachusetts and Rhode
   Island," U.S. Geological Survey Bulletin No. 597, 1917,
   289 p.
4.  LaForge, L., "Geology of the  Boston Area, Massa-
   chusetts," U.S.  Geological Survey Bulletin No. 839,
   1932,105 p.
5.  Bell, K.G.,  and  Alvord, D.C.,  "Pre-silurian Strati-
   graphy of Northeastern Massachusetts," hi Page, L.R.
   ed.,  "Contributions to the Stratigraphy of New Eng-
   land,"  Geol. Soc.  Amer.  Memoir No.   148, 1976,
   179-216.
6.  Barosh,  P.J., Fahey,  R.J., and Pease, M.H., "Pre-
   liminary Compilation  of the Bedrock Geology of the
   land area of the Boston 2-Degree Sheet, Massachusetts,
   Connecticut,  Rhode Island, and New Hampshire,"
   U.S.  Geological  Survey Open  File Report,  Boston,
   Mass., 1977, 138 p.
7.  Kaye, C.A.,  "Outline of the Pleistocene Geology of
   the Boston Basin," in Cameron, B., ed., "Geology of
   Southeastern New England," Guidebook,  68th Meet-
   ing,  New  England  Intercollegiate  Geologic  Con-
   ference, 1976.
8.  Chute, N.E., "Glacial Geology of the Mystic Lakes-
   Fresh Pond Area,  Massachusetts,"  U.S.  Geological
   Survey Bulletin No. 1061-F, 1959, 187-216.

-------
       CHARACTERIZATION, IMPACT ASSESSMENT, AND
      CLOSURE REQUIREMENTS FOR A COPPER RESIDUE
                                      DISPOSAL SITE
                                     JEFFREY L. PINTENICH
                                          JOHN M. HINES
                                        MICHAEL R. CORN
                                 FREDERICK G. ZIEGLER, Ph.D.
                                     The AWARE Corporation
                                        Nashville, Tennessee
INTRODUCTION

  An  11 acre disposal site located in the southeastern
United States was utilized for 50 years for the disposal of
copper-laden sludges and  a  number of other materials
from the production of rayon. This  paper summarizes
the results of a, preliminary engineering/hydrogeological
study which included characterization  of the materials in
the waste fill, an assessment of the impact of the site on
surface and groundwater  and development  of closure
plans for the site so that a light industrial park could be
developed on land adjacent to and possibly on the closure
area.
                          "•"•"eatod rnschartie to -:iver
                      Figure 1.
   Process Flow Diagram—Cuprammonium Rayon Production
  The textile mill adjacent  to the site was utilized for
the production of rayon fiber between 1927 and 1977. The
northern portion of the property includes a filled disposal
area, whose size is estimated to be approximately 11 acres.
The disposal area lies just south of a moderate size river.
Adjacent to the site is a  lagoon with a surface area of
approximately 3.8 acres, which was utilized for controlled
release of treated liquid process wastes while the mill was
in production. There is also a much smaller depression on
the site, which is dry and contains a sludge residue.
  Field reconnaissance indicates that during the years of
operation of the textile facility, virtually all types of solid
wastes generated at the mill  were disposed of in this site,
including rayon and  paper waste, construction debris,
laboratory wastes, scrap  equipment  and some process
sludge residues.  These residues likely resulted from the
spent copper ammonium  sulfate spinning solutions used
in the rayon production. Based upon existing topography
at this site, the disposal area ranges from 1 to 15 ft
above original ground elevations, varying with location.
  The state  environmental  regulatory agency  had ex-
pressed concern about the  disposal area as it presently
exists and about the potential for contamination of sur-
face and groundwaters. The owners  were interested in
reclamation of the site for use as a portion of an industrial
park that could be developed on the mill property. The
goal of the evaluations described herein was to conduct a
preliminary engineering evaluation for site  closure that
will be acceptable to the state agency and also is environ-
mentally  sound  and in conformance with the appropri-
ate regulatory requirements.

SITE HISTORY

  Rayon manufacturing activities commenced at the mill
site in 1927 and continued  until December, 1969, when
the plant was sold. The plant reopened in August, 1971,
but in 1977, production  again ceased and the company
declared bankruptcy.  There have been no rayon manu-
facturing activities conducted at the site since 1977.
  Rayon was manufactured at this mill via the cupraffl-
monium process which involved mixing cellulose (wood
pulp) with an ammoniacal copper  hydroxide solution
(Schweitzer's Reagent). The  copper solution was prepared
by adding caustic to a copper sulfate solution, filtering
and dissolving the precipitate in aqua ammonia. The cd-
                                                  70

-------
                                                                    SITE INVESTIGATION & ASSESSMENT     71
lulose solution was then diluted with water,  filtered and
forced through a spinneret into a coagulating bath  of
caustic. Coagulated cellulose was decoppered with sul-
furic acid, and subjected to a series of washing and dry-
ing steps. Both lime precipitation and ion exchange were
employed for copper recovery and wastewater treatment.
A manufacturing process flow  diagram is presented  in
Figure 1. Principal chemical reactions were as follows:
(CfiH,nO,) + Cu(OH)2
                        NH
                              H2O
                                     solution of cellulose
    cellulose     Schweitzer's Reagent

  Cellulose solution  + NaOh (5 %) —> partly coagulated cellulose

   Partly coagulated cellulose + H2SO4(l!/2%)   (C6H,0O5) x

                                  Cuprammonium rayon


PROJECT SCOPE

  The initial activity on  this project was  a preliminary
site reconnaissance conducted with the client and repre-
sentatives of State and Federal agencies.  A detailed plan
of study was then prepared which outlined major work
tasks,  the extent of field activities,  and a project sche-
dule;  this plan of study was submitted to  the client and
agencies for review and  comment. The project's scope
(Table I) of work included:
  (1)  Review  of information  pertaining to the  manu-
      facturing process that generated the copper residu-
      als
  (2)  Review of agency file information (historical  cor-
      respondence, flood plain maps, meteorological data)
  (3)  Topographical survey of the site and examination of
      original topographic maps for the area
  (4)  Acquisition and review of all  available soils, geo-
      logic and hydrologic information
  (5)  Residue and soil sampling and characterization
  (6)  Groundwater monitoring well  installation,  samp-
      ling and analysis of groundwater samples
                         Table I.
                     Work Task Outline
       Preliminary Engineering for Closure of Disposal Site

 TASK       TITLE
 1.0           Project Management
 2.0           Plan of Study
 3.0           Background Information
 4.0           Topographical and Boundary Survey
 5.0           Hydrogeologic Evaluation
     5.1          Available Data Base
     5.2          Field Investigations
     5.3          Sample Analysis and Data Evaluation
 6.0           Field Reconnaissance, Surface Draining
                 Evaluation, Sampling and Analysis
 7.0           Regulatory Agency Liaison
 8.0           Conceptual Plan for Site Closure
 9.0           Engineering Report
  (7) Detailed field reconnaissance and surface drainage
      evaluations
  (8) Assessment of site environmental impacts
  (9) Preparation of a conceptual closure plan and  en-
      gineering report.

SITE CONDITIONS AND EVALUATIONS
Topography
  The disposal site topography (Figure 2) is fairly irregu-
lar, with existing ground surfaces ranging from elevations
1,525 ft to 1,496 ft. Portions of the site are thinly wooded
and  vegetated,  while the bulk of  the  surface area con-
sists of solids and waste materials. In general, ground sur-
faces slope to the north and northwest, toward the river.
Surficial Waste Materials
  A detailed site reconnaissance was conducted to classi-
fy and quantify surficial debris.  There  are five  cate-
gories of waste products, as follows:
                                                            Waste Produce
                                                            Ash and Cinders
                                                            Refuse (trash, lumber, pallets)
                                                            Construction Material
                                                             (bricks, concrete)
                                                            Rayon
                                                             Spools
                                                             Barrels (Fiberboard)
                                                            Copper Sludge
                                                            The total landfill area is about 11 acres
                                                                                            Approximate Exposed
                                                                                                    Surface Area
                                                                                                         (Acres)
                                                                                                            2.2
                                                                                                            1.7

                                                                                                            1.0

                                                                                                            0.2
                                                                                                            0.3
                                                                                                            1.0
                                                                                                                6.4
                                                        Waste Chemistry .
                                                          Copper was utilized in the rayon manufacturing pro-
                                                        cess as copper sulfate and ammoniacal copper hydroxide.
                                                        Wastewaters from the manufacturing process were treated
                                                        for copper recovery by ion exchange and lime  precipita-
                                                        tion.  Waste products (copper  sludges) from these two
                                                        treatment facilities were disposed in the landfill. Contam-
                                                        inants in these sludges include copper, ammonia,  nitrate
                                                        and sulfate. Copper compounds  in the  sludges are un-
                                                        known, but solubilities for the most probably compounds
                                                        areas follows:
                                                        Compound
                                                        copper sulfate

                                                        copper hydroxide

                                                        copper sulfate,
                                                         ammoniated or
                                                         copper
                                                         ammonium
                                                         sulfate
                                                                             Solubility in Water
                                                                             decomposes; soluble in concentrated acid
                                                                             insoluble; soluble in acid
                                                                             18g/100gwaterat21°C

-------
 72    SITE INVESTIGATION & ASSESSMENT
                        Figure 2.
                        Site Plan

 The solubility of the copper sludges would also be influ-
 enced by natural physical  and biological forces in the
 landfill.
   Copper concentrations in two samples collected from
 the sludge pond were  34,000  and 100,000 mg/kg. The
 concentration in one sediment  sample collected from the
 lagoon was  15,000 mg/kg. The concentrations in the
 sludges of the two ponds are similar to the concentration
 expected from the  solubility of copper ammonia sulfate
 in water—which is 47,000 mg/1. Similar copper concen-
 trations were present in soil samples obtained near the
 sludge pond.


 Geology

  The waste disposal site is  underlain by alluvial deposits
 of Quaternary age and the Honaker dolomite  of Cambri-
 an age. The  Quaternary alluvial deposits which  immedi-
ately underlie the waste  disposal site  are comprised of a
heterogeneous mixture of gravel, sand, silt and clay de-
posited by the  river.  Typically these alluvial deposits
consist primarily of silty clay  with varying amounts  of
silt and minor amounts  of  gravel and cobbles. Distinct,
continuous lithologies were  not apparent based upon the
logs of these  wells.  The thickness of alluvial deposits be-
neath the site ranged from approximately 20 ft to great-
er than 35 ft.
   Underlying the surficial alluvial deposits is the Honaker
dolomite of Middle Cambrian age.  The Honaker con-
sists of  light to dark, fine-to-coarse-grained  shaley and
massive  dolomite. Interbedded limestone is also common
within the formation with shale  becoming dominant in
the lower part of the formation. Structurally, the Honaker
dolomite in the general area of the site  has been folded
into a southerly plunging, northeast trending anticline.
The dip of the underlying bedrock is expected to be to
the northwest.  The thickness of the Honaker  dolomite
ranges up to about  1,300 ft in this region; however, the
actual thickness at the site is undetermined.
Soils
   The original soils underlying the waste disposal site were
primarily Congaree and Masada type soils. The Congaree
soils are present in the northern half of  the site and are
classified as fine sandy loam. This soil type is  an alluvial
soil  containing  mica and minor amounts of gravel and
cobbles.  The  Congaree soils  are well drained and mod-
erately permeable. These  soils have  a low shrink-swell
potential and  a low to medium plasticity index. The
Masada  soils  are predominantly in the southern half of
the  disposal area and are classified  as silt  loam and
gravelly  silt loam. These soils are also alluvial in origin
and contain some gravel zones and cobbles. The Masada
soils  are well  drained and moderately  permeable. The
shrink-swell potential as well as the plasticity index is low
for these soils.
   As part of the site investigation, two  shallow borings
were  made  on the small knoll located just east of the
waste disposal site. These borings were conducted to make
a preliminary determination of the type of soil materials
present  in the  immediate  vicinity  of the disposal site
which might be suitable for cover material. Each boring
was approximately three ft  deep. A composite grab sam-
ple was collected from each boring with  a sieve analysis
and  plasticity index determination being made for each
sample.  The results  of this  limited testing indicated a
sandy silty clay to clayey sand being present in the area of
the borings. The clayey sand was nonplastic, whereas the
sandy silty clay had a plasticity index of only 11. Addi-
tional soil samples from shallow (0.5-1.0 ft) excavations
in the eastern section of the plant property have been col-
lected by client personnel. A single composite sample was
made from six separate sampling locations for laboratory
determination of grain size distribution,  plasticity index,
and permeability. The results from this single  composite
sample indicate  that the material is essentially a clay soil
(CH) with a  relatively high plasticity index  of 25. A
falling head permeability test performed on the composite
sample revealed  an  extremely low permeability of 8.43
x 10 "9 cm/sec.

Subsurface Investigation

   In order to evaluate the  distribution of potential con-
taminants in the soil profile  beneatn the waste disposal
site,  six boreholes were drilled at the site.  These borehoto
were also intended to provide information on  the char-
acteristics and depth of waste materials  across the rite.

-------
                                                                  SITE INVESTIGATION & ASSESSMENT
                                                    73
The location of each borehole is depicted in Figure 2. The
six boreholes ranged in depth from 4 to  18 ft with refusal
being encountered in boreholes AM-B-1, AM-B-4, AM-B-
5 and AM-B-6 at depths of 4, 11,  6.5  and 5  ft, respec-
tively.  Waste materials were  encountered in boreholes
AM-B-2, AM-B-3, AM-B-4 and AM-B-5 to depths  of 4,
11, 5 and 2 ft, respectively. The total fill material added
to this area is estimated to be about 140,000 cu yd, based
upon comparisons of original (1927) and  existing (1980)
site topography.  Soil  borings  indicate that most of this
material is ash and cinders.
  Twelve grab samples were collected from six boreholes
for determination of copper concentration and pH.  Each
sample was placed in a plastic jar with Saran Wrap®  over
the  mouth (moisture  barrier) before   capping  and re-
turned to AWARE's Laboratory in Nashville  for chemi-
cal  analysis.  The results of this sampling program are
summarized in Table II.
  High  concentrations of  copper were  found in the
samples  from boreholes AM-B-2, AM-B-3 and AM-B-4.
As expected, the  highest concentrations  are found in the
waste materials with concentrations ranging from 26,000
mg/kg in AM-B-4 to 75,000 mg/kg in AM-B-2. The con-
centration of copper  decreases drastically in  the under-
lying soil in each borehole.  For example, the  concentra-
tion in AM-B-2  decreases from 75,000 mg/kg  to  3,200
mg/kg in only a 5-ft interval; and in AM-B-4 from 26,000
mg/kg to 570 mg/kg in an interval of 7 ft.  This  sharp
decrease  in the concentration of copper with depth is to
be expected since copper is one of the least mobile trace
elements.
  At this site, the copper is apparently  readily absorbed
by the silt and clay  particles in the underlying  soil. There
appears  to be  no direct correlation between  concentra-
tions of  copper and soil pH,  which ranged from 4.5 to
9.2 This  soil data,  in conjunction with  the groundwater
quality data  generated during this  investigation, appear
to substantiate the low mobility of copper through soils.
  The  three  boreholes with the highest copper concen-
trations are all located immediately adjacent  to the pri-
mary disposal area for  the  copper ammonium sulfate
solution. The other three boreholes (AM-B-1,  AM-B-5
and  AM-B-6) located  in the northern  and western sec-
tions of the site had relatively low copper concentrations
with very low values being detected at a depth of 5 ft in
each of these boreholes. Therefore, the  vast majority of
the copper-bearing  wastes are expected  to be  located in
the southern  section of the disposal site within the  con-
fines of boreholes AM-B-2, AM-B-3 and AM-B-4.

Groundwater Conditions
  Groundwater at the waste disposal site  occurs in both
the surficial,  unconsolidated alluvial deposits and the un-
derlying  Honaker dolomite.  Due  to the heterogeneous
and  discontinuous nature of the unconsolidated alluvial
deposits,  these materials do not typically represent signifi-
cant sources  of groundwater.  However, small yields can
be obtained from the  sand and gravel zones which  com-
monly  occur within these deposits.  The Honaker  dolo-
mite represents a significant  source of groundwater in
Table II.
Analytical Results of Soil and Waste Samples from
Boreholes at the Disposal Site
Borehole
Sample
AM-B-1
AM-B-2- 1
AM-B-2-2
AM-B-3- 1
AM-B-3-2
AM-B-3-3
AM-B-4- 1
AM-B-4-2
AM-B-5- 1
AM-B-5-2
AM-B-6- 1
AM-B-6-2
Depth
4-5'
4-5'
9-10'
7-8'
11-12'
14-15'
1-2'
8-9'
2-3'
4-5'
2-3'
4-5'
Soil
Material
silt
Hll material & loam
clay
fill material (cinder)
sandy clay
silty clay
fill material
silty clay
sandy clay
sandy clay
clay
clay
pH
4.8
6.0
4.5
8.4
9.2
8.1
6.6
9.2
7.7
5.8
7.5
7.4
Copper
Concentration
(mg/kg)
190
75,000
3,200
50,000
5,400
2,800
26,000
570
1,100
840
8,300
140
this area with groundwater movement occurring through
joints, fractures  and  bedding planes in  the  bedrock.
Yields of wells completed into the Honaker dolomite in
this area are reported to be in the range of 500 to 2,500
gal/min.
  Several wells are present on and in the immediate vi-
cinity  of the plant  site.  There are three wells at the
plant;  however, only one well is currently being utilized.
The well currently in use is located a few hundred feet
southeast of the waste disposal area. The plant wells are
all completed into  the Honaker  dolomite  to  depths of
305 ft, 600 ft and 700 ft with each well having a reported
yield of 2,500 gal/min.
  At lease nine wells have been installed at another  plant
just west of the rayon mill. Each of these wells is  com-
pleted into the Honaker dolomite to depths of between
70 and 602 ft. Yields of these wells range from about 100
to 1,500 gal/min.  Additional  industrial,  domestic and
municipal wells are not known to be present in the im-
mediate vicinity of the disposal site.
  The depth to groundwater at the disposal site during
September and October 1980 ranged from about 13 ft to
17 ft in the northern section of the site at wells AM-1 and
AM-2 to about 24 ft to 28 ft in the southern section of
the plant site at well AM-3 (Table III). The elevation of
the water table beneath the disposal site at the time of
water  level measurements in  September  and  October
ranged from about  Elevation  +1481.9  to   Elevation
+ 1485.5. These elevations were approximately 12  to 19
ft below the base of any waste materials at  the site.  Fluc-
tuations in groundwater levels would be expected to be in
the range of 5 to 10 ft annually. Therefore, the base of
the waste materials at the site would not be expected to
come in contact with the underlying groundwater.
  Based upon the  general hydrogeologic setting of the
waste  disposal site, the direction of groundwater flow
would be expected  to be toward the river,   a  natural
groundwater sink.  The three groundwater  monitoring
wells  installed as part  of this investigation verified  a hy-
draulic gradient to the north/northwest toward the  river.
Localized variations in the direction of groundwater flow
commonly  occur in carbonate bedrock such as the Ho-

-------
74    SITE INVESTIGATION & ASSESSMENT
naker dolomite since flow is  controlled by the orienta-
tion of  secondary openings such as joints, fractures and
bedding planes. However, these flow variations would be
relatively  local in nature with the regional groundwater
flow pattern being toward natural groundwater sinks such
as the river.

                       Table III.
            Groundwater Level Data for the Site
Monitoring
Well
AM-1
AM-2
AM-3
Date
9/25/80
10/20/80
9/25/80
10/20/80
9/25/80
10/20/80
Depth of
Groundwater from
Top of Casing
(ft)
17.0
14.4
19.0
14.8
30.0
26.6
Elevation
(ft)
1480.2
1483.5
1479.3
1483.5
1481.9
1485.5
  The rate of groundwater movement in the vicinity of
the waste  disposal  site  is  controlled by  the materials
through which  flow  is occurring. The unconsolidated al-
luvial deposits at the site  would be expected to have mod-
erate  to  moderately low permeabilities in the range of
10~3 to 10~5 cm/sec. The rate of groundwater flow through
these  deposits would be  expected to be on the  order of
100 to 200 ft/yr. However, the underlying carbonate bed-
rock would have a considerably greater permeability where
secondary openings are present.  Flow through these secon-
dary openings in the bedrock would be analogous to flow
through pipes with relatively high velocities in the range
of a foot per day to several feet per day being common.
The exact rates of groundwater flow at the disposal site
were not determined; however, the above values would be
expected to be representative.
  In order to make preliminary determinations of the pre-
sent extent and distribution of potential contaminants at
the site, three groundwater monitoring wells were installed
in the vicinity of the waste disposal site. Information was
sought on the groundwater quality and water level data to
be used as a basis for preliminary determination of po-
tential impacts  and the closure  requirements for the dis-
posal  site.  Two monitoring wells were located  hydraul-
ically  downgradient  of the  waste disposal site  and  one
monitoring well was located hydraulically  upgradient of
the site. The two downgradient wells are located  within
200 ft of the northern boundary of the waste materials in
order to provide information on the potential migration of
contaminants from  the  site via the groundwater flow
system. These two downgradient monitoring wells were
deemed a sufficient number of downgradient sampling sta-
tions based upon the relatively  small size of  the disposal
site, the characteristics of  the  waste materials, and the
scope and intent of the project.
  The   upgradient  monitoring  well   is   located  in
the southwest section  of the mill site at the southwest
corner of the plant parking area. This monitoring well was
installed in  order to obtain groundwater quality data rep-
resentative of background conditions. The upgradient well
was located as far as practicable from the waste disposal
site and from other past plant operations that might in.
terfere with the wells' intended use.
  Each of the groundwater monitoring wells was complet-
ed  into  the  upper part  of  the  uppermost  aquifier be-
neath the site. Monitoring  well  AM-1  was installed ap-
proximately seven ft into the water table, whereas wells
AM-2  and AM-3 were installed  about  18 and  12 ft, re-
spectively, into the water table.  Since  the downgradient
monitoring wells are within  200  ft of the waste disposal
site and the disposal site is situated in a groundwater dis-
charge zone, completion of the monitoring wells into this
uppermost zone allowed  the interception of the contam-
inants potentially migrating from the site.
  Each monitoring well installed as part of the ground-
water monitoring program was constructed of 4-in., sche-
dule 40 PVC pipe. PVC pipe was utilized since organic
contaminants were not expected to be associated with the
dump site in any  significant quantities. It was expected
that copper and inorganic parameters such as sulfates and
nitrates would be of most concern at this site. Each moni-
toring well is equipped with  a 10-ft section of perforated
PVC screen. The  annular space around and  above the
well screen was gravel-packed with the remaining annular
space being backfilled with  cement  grout to the ground
surface.  A 3-  to 5-ft section  of 6-in. steel casing  was
utilized as a protective casing above ground surface.  The
protective steel casing is  equipped with a lockable steel
cap.
  Following  the  installation of the three  groundwater
monitoring wells at the  site, samples were collected for
analysis of the following parameters:
    pH                      Sulfate
    Copper                  Chloride
    Ammonia-Nitrogen       Total Organic Carbon
    Nitrate-Nitrogen

Each of the groundwater monitoring wells was bailed to
remove three to five well volumes prior to sample  col-
lection. Samples  were collected  in appropriate contain-
ers, labeled and packed in ice for shipment to the labora-
tory. The pH of each sample was determined in the field.
All samples  were  delivered  to the laboratory within 24
hr of collection.
  In addition  to the  samples collected  from the three
groundwater monitoring wells,  samples were collected
from  the on-site  plant  well that is still being utili^d
and the river, both upstream and downstream of the we.
Each of these samples was handled in  the same manner
as the groundwater samples and was analyzed for thesanw
parameters.  The analytical  results for all  of the water
samples collected at the site are found in Table IV.
  The samples from the  downgradient  monitoring welll,
AM-1  and AM-2,  have  higher  concentrations of rood
chemicals than the  upgradient monitoring well, AM»3>
and the existing plant well. Those parameters found H>
be  significantly higher  in  the two  downgradient weft
included  sulfate, nitrate and  ammonia. Each of thW
chemicals would logically be  present  since one of tbt
primary waste  materials disposed at the disposal site ***

-------
                                                                 SITE INVESTIGATION & ASSESSMENT    75
                                                     Table IV.
                                  Analytical Results of Water Samples from the Plant Site
Sampl ing
Station
AM-1
AM-2
AM-3
Plant Well
River/upstream
River/upstream
PH
7.9
7.3
6.4
7.4
8.0
7.9
Chloride
15
7
7
2
2
2
Nitrate
35
22
0.77
0.35
0.40
0.42
Ammonia
32
2.0
0.17
<0.1
0.17
0.25
TOC
5
9
5
<5
< 5
<5
Sulfate
650
310
87
26
9
6
Copper
0.08
0.05
0.22
0.40
1 <.02
<.02
 Note:  All  values, except pH, are in mg/1,
spent copper ammonium sulfate solution used in the pro-
duction of rayon.  The nitrate and ammonia detected in
the groundwater are the result of oxidation of the am-
monium compounds in the wastes.
  Of particular significance  are  the  lower concentra-
tions of copper in the two downgradient wells as com-
pared to higher levels in the upgradient monitoring well
and  the plant  well.  Concentrations  of only  0.08 and
0.05 mg/1 were detected in wells AM-1 and AM-2, where-
as AM-3  contained  0.22  mg/1 and  the  plant  well con-
tained  0.40 mg/1 of  copper. The reason for this concen-
tration  difference  is  not  clearly understood;  however,
it might be related to the difference in pH in the  up-
gradient and  downgradient  wells since copper is  less
soluble  at higher pH values. Another possibility is that
reducing  conditions   exist beneath  the  waste  disposal
site.  The solubility of copper is generally lower in reduc-
ing environments,  especially is reduced sulfur species are
present.
  The  high copper concentration detected  in  the plant
well  is  most likely attributable  to the pump,  pipe and
fixtures of  the water distribution system.  Based upon
the soils and groundwater quality data, it is not believed
that the waste disposal site is responsible for the high cop-
per concentration detected in this well.  Due to the rela-
tively low concentrations  of all  parameters in the  up-
stream and  downstream samples  from the  river, no dis-
cernible impact attributable to the waste disposal site can
be determined.
  In summary,  the  uppermost  part  of the underlying
groundwater flow  system immediately downgradient  of
the waste  disposal  site has been impacted to some extent.
The  primary contaminants are sulfate,  nitrate and am-
monia;  copper in the waste materials at the site appears
to be relatively immobile.  In addition, no discernible im-
pact can  be seen in the  deeper  groundwater system
tapped  by the  plant well  or in the  river.  Since the di-
rection  of groundwater flow at  the  site is  toward the
river, contaminants leaching from the site would be ex-
pected to discharge  to  the  river and  not  migrate off-
site.
Surface Waters
  The USGS  maintains  a continuous recording gauge
(river stage) approximately 0.8 miles upstream from the
mill  site. The USGS reports the following  information
for the gauge site:
  (1) Flows at the gauge  are partly regulated by a lake
      located 10.8 miles upstream from the gauge
  (2) Low flows are  regulated by a lake located 8.1 miles
      upstream from the gauge
  (3) Average  discharge  (48  years  of record)  at  the
      gauge is 1,092 ftVsec.
  (4) Minimum daily discharge  of record  is 85  ftVsec
      (December 3, 1953) and the instantaneous minimum
      of record is 42 ftVsec (September 20,1932)
  (5)  Maximum discharge of record since closure of an
      upstream dam  (December 1,  1948) is 14,500 ftVsec
      (March  12, 1963). The corresponding  stage at the
      gauge for this flow is Elevation 1496.93 ft (from
      National Geodetic Vertical Datum of  1929)
  (6) Drainage area at the gauge is 692 mi2
  (7) The river flows into a reservoir about 25 miles down-
      stream from the mill  site  and backwaters of the
      reservoir arm begin  about six to eight  miles down-
      stream from the site.
  TVA reports that the river stage for the 100 year flood
at the site is Elevation 1492. The river at this location is
also  at one of its steepest slopes.  The lowest  fill elevation
present in the site is about  1496 ft. Therefore,  the site
would not  be inundated by the  floods of 100 year fre-
quency.
  Water  samples collected from  the ponds had copper
concentrations of 24 mg/1 (sludge pond) and 0.17  mg/1
(lagoon). The copper concentrations  (5,000 to  100,000
mg/kg) in the sludges of the two ponds are similar to the

-------
76     SITE INVESTIGATION & ASSESSMENT
concentration of 47,000 mg/1 expected from the solubility
of copper ammonia sulfate in water. Similar copper con-
centrations were present in soil samples obtained near the
sludge pond. The lower concentrations of copper (24 and
0.17 mg/1) in the surface waters  of the two ponds indi-
cate that the physical attraction of the copper to soils may
be greater than its potential to solubilize in water.
  The site soils, both sands and  clays, appear to have a
high affinity for the copper. This is evidence from the high
concentrations in the sludge and the significant reduction
in concentrations with depth in the soil borings.
  The concentrations of both ammonia and sulfate were
high (140 and 460 mg/1,  respectively) in the waters of the
sludge pond. No significant nitrification of the ammonia
in the surface waters of the two ponds was apparent from
the water quality data.
  No significant difference in water quality between the
upstream  and downstream  samples  was evident (see
Table IV). The major pathway for contaminants to reach
the river is through surface  water runoff. Groundwater
inflow from the disposal  site is an insignificant portion of
stream flow as can  be seen from  the following simplified
example in which the theoretical  downstream  concentra-
tion is calculated.
  Avg. stream flow =  l,092ftVsec
  Approx. drainage area  = 692 mi2
  Assume 1.58 ftVsec/mi2
  groundwater inflow (avg.)
     (l,092ftVsec/692mi2
  Ten acres of landfill =  0.016 mi2
  Assume 650 mg/1 SO4 in groundwaters (max. SO4 cone.)
  Assume 9 mg/1 SO4 in river (upstream cone.)

        Downstream sulfate concentration =
 (1,092 ftVsec (9 mg/1) + (0.016 mi2 (1.58 ftVsec/mi3 (650 mg/1)

                l,092ftVsec + 0.03ftVsec
                    = 9.01 mg/1 SO4

  During storm events, contaminants can enter the river
through runoff. Copper that might enter the stream during
storms would most likely be attached to colloidal soil par-
ticles and would eventually settle to the stream bottom. No
estimates can be made of what percentage or quantities of
material  this  may represent but it  is not believed to  be
significant compared  to  total streamflow and sediment
volumes.


DISPOSAL SITE CLOSURE

  Surface water runoff is  anticipated to be the major mode
of contaminant transport from the disposal area. The dis-
posal site closure plan described below  is basically de-
signed to limit the amount of water running across or per-
colating into the waste material. The following steps are
contemplated in the closure plan:
•Control-burn the combustible material on the landfill
•Eliminate standing water areas  by appropriate  site
 grading
•Minimize  the  distance  incident   stormwater  runoff
 crosses the landfill
•Minimize run-on (overland flow from other areas) to the
 landfill
•Place an  impermeable cap  on the landfill to  minimize
 infiltration into the groundwater
•Establish  surveillance and maintenance plan for closed
 site
  A conceptual drawing of the site  closure plan is given
in Figure 3, while a generalized cross-section is shown as
Figure 4.

Disposition of Surf icial Debris
  Much of the surficial waste consists of combustible ma-
terials that  could be control-burned during one period
(with appropriate county and state approvals). Estimates
of the combustible material present are:
Waste Material

Paper, Pallets, Lumber
Rayon
  Spools
  Barrels (cardboard)
Estimated Volume (yd1)

        9,000

        1,000
        1,500
  Specific permits for the control burn will have to be
obtained from the city/county Fire Marshal and possibly
the state  air  pollution  agency.  Local  foresty officials
should  also  be  contacted.  The  local  fire department
should be requested to oversee the burning.
  Bricks, concrete and other construction material are lo-
cated in the north fill area and cover about one acre. The
bricks should be salvaged if possible.
Site Grading
  Based upon existing site topography, it is recommended
that the site be graded to a 3 to 4 percent grade as shown
in Figure 4. Low spots  such as the 0.8 acre sludge pond
should be filled in with soil fill. The entire site soils and
materials should  be  compacted in order to decrease the
chances of subsequent soil settlement.
Capping and Ground Cover
  Once the final grade is achieved, a compacted two-foot
thick clay cap should be placed on the fill and graded to
conform to the 3 to  4 percent grade. A potential borrow
area is present on-site from an area with  soils classified
as clay (CH). It will be necessary to obtain soil samplw
for permeability,  compaction and porosity determination!
from several borings in  the prospective borrow area prior
to final decision.  An estimated 40,000 cu yds of clay ma-
terial is needed for the clay cap and potential fill material.
  Next, six inches of top soil should be placed on top of
the clay cap  and seeded with a hardy grass cover (e.g.,
fescue). The final site grade should conform to the 3 to 4
percent slope and no low spots or potential ponded areas
should remain. About 10,000 cu yds of topsoil are needed
for the topsoil cover.

-------
                                                                 SITE INVESTIGATION & ASSESSMENT    77
                                                     Figure 3.
                                       Conceptual Plan for Disposal Site Closure
                       Table V.
          Future Groundwater Monitoring Program
Parameter   Fre-
                                 LOCATION

Disch (a)
or water
elev
PH
Cond.
Cu(b)
So4(b)
NH3-N(b)
NO3-N (b)
quency

M
Q
Q
Q
Q
Q
Q
Obser-
vation
Wells (c)
X
X
X
X
X
X
X
Drink-
ing Water
Well
X
X
X
X
X
X
X
River
Upstream
X
X
X
X
X
X
X
River
Down-
Stream
X
X
X
X
X
X
X
a. Discharge available from USOS gauge       c. Observation wells AM-1, AM-2, AM-3
b. Analysis on soluble portion of water sample    d. M = Monthly  Q = Quarterly
Groundwater Monitoring Program

  Groundwater monitoring should be done quarterly for
at least one year following closure of the landfill. Parame-
ters and  sampling frequency for the groundwater moni-
toring program are given in Table V. River water samples
should be obtained upstream aird downstream from the
site along with the groundwater samples. At the end of
the first  year,  the data from the  monitoring  program
should be reviewed and a continue, curtail or cease moni-
toring program decision should be made.
CONCLUSIONS

  The following conclusions were derived from this study:
•The disposal site covers about 11 surface acres and con-
 tains an estimated 140,000 yd3 of material
•There are approximately 12,000 yd3 of combustible ma-
 terial on the surface of the disposal site.
•A total of about one surface acre of copper sludges has
 been placed on the site.
•The groundwater table is 13  ft to 17 ft below the base of
 the fill material.
•The disposal site is above the 100 year flood plain of the
 river.
•Copper was at or below background concentrations in
 groundwaters downgradient  from the disposal site. Am-
 monia,  nitrate and  sulfate  significantly above  back-
 ground concentrations were present in the downgradient
 wells.
•Surface water  runoff is believed to be the  major trans-
 port pathway for copper. Percolation of surface waters
 and rainfall to the underlying groundwaters is resulting
 in degradation of water quality in downgradient wells by
 ammonia, nitrate and sulfate.
•Water contact with fill materials can be minimized by
 proper grading, capping with a  compacted clay  and
 topsoil layer, and seeding with grass.

-------
 78    SITE INVESTIGATION & ASSESSMENT
     1515


     1510



     1505



     I5OO



     1495



     1490



     1465



     1480


     147!
 HARDY GRASS STAND
 WASTE  FILL
..WITH SOME  SAND SEAMS
                         x XXXX . .            _.,
                         X X X X X X X X X X X X X X X X X X X X^^iW,.,
                         XXXXXXXXXXXXXXXXX XXXXXXTS^-
                         \xxxxxx\xxxxxxxxxxxx\\\\ x xS,,.
                       xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
                      txxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
                      "sXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
                    Jr\xxxxxxx\xxx\x\xxxxx\xxxxx\x\xx\xx\x\xx\x\xxxxxxxxxxxxx\xx
                   ^XXXXXX XXXX XXXXXXXX XX XXX XX XXXXX \\X\X\\XX\XXXXXXX\XXXXXXXXX\\
                  M~\ \XXX\XXX\X\\\XX\\X\\XX\XXX\\X\X\\XXXXXX\\\XX\XX\XXXXXXXXXX\
                  f\ XX\XXXXXXX\\X\Xx\XX\XX\\\XXXXX\\\XXXXXXXXXXXX\X\\\XXXXXXXXXX
                 -*x XXXX\XXXXX\XXXXX\XXX\XXX\XXXX\XXX\XXXXXXX\XXXX\\\XXXXXXXXX\XX
_... „	 - —	     ^_ ^ XNXXXXXXXXXXXXXXXXXXXXXXXXXXX X'X XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
^~"™"    *^fC»!^*" \X\X\X\XXXX\XXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
\ •   \,  \v xx-\  \\XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
•^1 AWC-v Gil -T Tn Cll TV  CAhinv  ^1 AV XXXXXXXXXXXXXXXXXX , \XXXXXXXX\\X\\XXXXXXXXXXXXXXX
 CLAYEY SILT TO SILTY,  SANDY  CLAY\xxxxxxxxxxxxxxxxxs\x\x\\\\xxxxxxxxxxxxxx\xxxxx\
                              	NXXXXXXXXXXXXXXXXXXXXXXXXXXXX
                                                    	X X\\XXXXXXXXXX\\XXXXXX
              XXXXXXXXXXXXXXXX                              " "	
              XXXXXXXXXXXXXXXX
                     XXXXXXXX
        .xxxxxxxxxxxx
 SANDY  CLAY
MESTON
*^'-&


                                                                                                          -0.5 FT. TOPSfflL
                                                                                                          -2 FT. CLAY CAP
                                                                               -EXISTING
                                                                                TOPOGRAPHY
                                                                               -ORIGINAL
                                                                                TOPOGRAPHY
                                                       Figure 4.
                                         Conceptual Plan for Disposal Site Closure
                                                 General Cross-Section
•Proper closure of the site should prevent any significant
 future contamination  to  ground or  surface  waters.
 Closure of the disposal site can be accomplished in such a
 way as  to allow  certain commercial and/or  industrial
                                 uses  of the site.  However, additional  evaluations  and
                                 determinations would be required to ensure that any fu-
                                 ture  development is compatible with the closure plans
                                 described herein.

-------
 NEW ASSESSMENT METHODS TO AID SITE RESTORATION
                                             G.W. DAWSON
                                               S.M. BROWN
                                  Battelle-Pacific Northwest Laboratory
                                           Richland, Washington
INTRODUCTION

  With passage of the Comprehensive Environmental Re-
sponse,  Compensation  and Liability  Act (CERCLA) of
1980, financial and regulatory incentives have accelerated
activity in the area of site restoration. The Federal Gov-
ernment  now  has both  clear authority and  resources
which will allow it to systematically evaluate and, where
necessary, apply remedial action to sites burdened with
chemical contaminants. In turn, industry is motivated to
review areas where former activities may have  left re-
siduals and apply appropriate restoration measures to in-
sure against subsequent impacts on  public health   and
the environment.
  Historically, when such measures have been undertaken,
they have involved a number of activities which were loose-
ly termed environmental assessment. In general, these ac-
tivities related to the collection of baseline information and
the prediction of future impacts. They were typically di-
rected to the preparation of products such as an Environ-
mental Report and ultimately an Environmental Impact
Statement.
  In the early years, environmental impact assessment
was  more art than science.  Recent  developments  have
changed both the image and the nature of assessment ac-
tivities. New methods have made the process more quan-
titative and formalized. They have encouraged the expan-
sion of the assessment role from a relatively passive one
to a very active one wherein all phases of problem reso-
lution are designed, monitored and concluded through in-
put from successive levels of evaluation. Assessment meth-
ods for site restoration are undergoing much the same eval-
uation. This paper addresses how new methods can be used
to enhance the hazardous waste site restoration process.

THE SITE RESTORATION PROCESS

  The site restoration process can be conceptualized as
a composite consisting of three discreet elements: (1) site
characterization, (2) site assessment and (3) site remedia-
tion. (Monitoring is  viewed as a form  of characteriza-
tion occurring during and after implementation of remed-
ial action.) It is  misleading,  however, if this  order of
presentation is taken to connote sequencing of the four
elements. Experience suggests  that the elements  are inter-
active and connected by a series of feedback loops which
sponsor optimization as one progresses through the restor-
ation process. The substance of the control mechanisms
which allow optimization is a suite of assessment methods
which function first to determine the  adequacy of the
previous step and then to design the most cost effective
approach to the succeeding step. This process is illustrated
in Figure 1.

Site Characterization

  Work performed during site characterization is directed
to the description of the source term and the surround-
ing environment. The objective is to enumerate and quan-
tify to the extent possible the physical, chemical, and bio-
logical factors which will affect fate and migration of con-
taminants.  For maximum efficiency, characterization is
performed in an integrated manner with assessment such
that only necessary data are collected.
  The first element of characterization, the preliminary
survey, is focused on gathering all existing data on a site
and the potential contaminants of concern. Reports, maps,
monitoring data and related materials  are collected for
conduct of a preliminary assessment. If data are sufficient,
no new characterization efforts are required. If the pre-
liminary assessment reveals data gaps which must be closed
before a final assessment can be made, then final char-
acterization is implemented. Hence, the preliminary assess-
ment is designed to determine the adequacy of existing
data and the optimum design  for any subsequent  data
collection activities. In particular, the initial assessment
should provide details on the location and type of data re-
quired from the site and the nature of  chemical proper-
ties information which should be sought or generated.
  When active site characterization is necessary, it is divid-
ed among a number of  field activities.  Geophysical sur-
vey techniques  can be employed to map  deposits.  Out-
put details the location of  buried  objects  or  the boun-
daries between significant strata such as the phreatic sur-
face and major impermeable layers. These efforts are com-
plimented by  field geohydrologic investigations which de-
termine the nature and extent of geologic and hydrologic
features which will affect the movement of contaminants.
Finally, sampling and analysis is performed to  detail the
disposition of contaminants in the environment and the
physical/chemical properties of  geologic  media which
would attenuate movement.
                                                     79

-------
80     SITE INVESTIGATION & ASSESSMENT
  All characterization activities may not be required at
every site. Further chemical characterization may include
in depth literature searches or in situ and laboratory ex-
perimentation to  generate  empirical data. As noted pre-
viously,  the type  and extent of activities should be iden-
tified during the preliminary assessment.
  Post-closure characterization can actually be thought of
as two distinct activities: (1) surveillance of remedial ac-
tion implementation and (2) post-construction monitoring.
The first of these is directed  toward assuring safe per-
formance of work and conduct of that work in the man-
ner prescribed in the final plan. The second activity looks
          to confirm  the  effectiveness of the remedial action on a
          continous basis and, in so doing, to   provide a warning
          in the event of failure.
            Post-closure characterization plans are dictated by the
          remedial action approach selected and designed through
          conduct of a post-closure assessment. Once again, the as-
          sessment is  directed to determining the optimum location
          and frequency of sampling activities.  This is done by de-
          termining the likely  location and timing of failures and
          then placing sample  locations to intercept resultant con-
          taminant plumes.
                      CHARACTERIZATION
                          ACTIVITIES
                      Collect Existing
                            Data
    ASSESSMENT
    ACTIVITIES
REMEDIATION
 ACTIVITIES
I CC.-.CUCT: Preliminary
I     Assessment
                                 _j  Specify Minimuir.
                                                       /    Are
                                                     / Data Sufficient
                                                     •A. -o Conduct Final,
                                                      \  Assessment?
! i Data Requirements
Collect New
Data
I

jies
Conduct Final
Assessment
                                                                       Identify Feasible
                                                                       Alternatives for
                                                                          Remediation
                                                  Calculate F.isk After
                                                  Implementation of Each
                                                  Rem?ciatior, Action
                                                  Alternative
                                                   Calculate Risk
                                                  Reduction Potential
                                                  for Each Alternative
                                       Design Surveillance
                                         and Monitoring
                                             Plan
                            Calculate Costs
                          Required for Each
                              Alternative
                     Implement Surveillance
                      and Monitoring Plan
                         Rank Alternatives
                         by Cost Effective-
                         ness
                                                                               Select  Recommended
                                                                                   Alternatives
                                                                               Design  and Construct
                                                                                 Remedial Action
                                                         Figure 1.
                                           The Integrated Site Restoration Process

-------
                                                                  SITE INVESTIGATION & ASSESSMENT     81
Site Assessment

  Activities conducted in an assessment  mode are de-
signed to determine the implications of the available data.
In this  context,  they  are interpretative in  nature.  The
first level of assessment,  preliminary assessment, is con-
ducted using available data to determine if sufficient in-
formation is on hand  to characterize a site and  quantify
the risk posed by that site. Two  levels of output are poss-
ible. The first is a determination that current data are in-
sufficient. In this case, the assessment identifies  the min-
imum necessary data that should be collected to allow con-
duct of a final assessment. The  second possible output is
the result of a final assessment itself.
  The final assessment is conducted only  after sufficient
data are in hand. Narrowly defined, the product of a final
assessment is the description of the time frames, durations
and concentrations of contaminants in each segment of the
environment (i.e., surface water, groundwater, atmospher-
ic and soil). In this context,  the assessment product is  a
prediction of what continuous monitors would record if
the situation is allowed to remain in its current state. When
coupled with data on dose-response functions for the con-
taminants of concern, the assessment output is converted
to an estimate of the risk posed  by the site. When it is de-
termined that remediation is required and conceptual ap-
proaches are identified, assessment is again conducted  to
predict the effectiveness of proposed  alternatives and  to
identify the most cost effective  postclosure monitoring
plan.
  Assessment is performed  through the  use  of models
which 'predict the fate and  migration of contaminants.
Models employed may be  as simple as an equation or algo-
rithm describing the partitioning  of a contaminant be-
tween two phases, or as  complicated  as  a  three-dimen-
sional mathematical model.  Battelle has  found that the
best approach to assessment is to have available a suite  of
models covering each pertinent pathway in successive lev-
els of complexity. The appropriate model is selected at the
time of use on the basis of the  amount of data  available
and the level of resolution required.  Cost-effectiveness
considerations dictate use of the simplest  possible model
that can provide the level of output required.  Only those
pathways of major significance are modeled. The key  to
good assessment is a thorough understanding of environ-
mental pathways and the ability to conceptualize the site
under study prior to modeling.

Site Remediation

   Remediation activities are the major end-product of site
restoration work. They are initiated whenever a final as-
sessment concludes that risks posed by a site are  unaccep-
table. At that time, the preliminary work is begun to iden-
tify all feasible alternatives.
  After a cursory screening, the most promising alterna-
tives are described in a conceptual framework and input to
the post-closure  assessment where estimates are  made  of
the effectiveness of each alternative. Output from the~post-
closure assessment is compared to output from  the final
site assessment to determine the level of risk reduction of-
fered by each alternative. Alternatives not capable of re-
ducing risks  to acceptable levels are  discarded. Remain-
ing candidates are ranked according to cost. The ratio of
risk reduction to cost is then employed to select the most
cost-effective approach.
  The selected alternative is then taken into final design
and construction. Post-closure  characterization is con-
ducted to assure that construction is performed in accor-
dance with final designs and to detect any releases  or ac-
tivities which pose a hazard to the public as a result of re-
mediation.
  The need to select the most cost-effective means of re-
mediation dictates knowledge of the latest technology as
well as a strong interaction with regulatory agencies to in-
sure that recommendations can meet permit  application
requirements. Site restoration is an area of rapid technical
change. Considerable work is presently ongoing to develop
new approaches to remediation. At the same  time, there
are no formal standards for defining clean-up thresholds.

EXPOSURE-RESPONSE ANALYSIS
IN SITE RESTORATION

  As suggested  previously, one  means of tying all ele-
ments of site restoration together is to reduce assessments
to  quantitative results where possible.  One of  the most
effective ways of doing that is to couch results in terms of
risk.  Risk is  a common unit between the  elements of the
restoration process. At the samed time,  it lies at the heart
of the motivation for restoration efforts in the first place,
i.e., site restoration is necessitated because extant con-
tamination poses unacceptable risks and similarly, restora-
tion is  deemed complete when  risks are reduced to ac-
ceptable levels. Given  this background, it is now possible
to discuss the nature of risk assessment in site restoration.
     REQUIRED INPUT DATA
                                      ANALYSIS
   METEOROLOGICAL DATA
   PROPERTIES AND APPLICATION
   RATES OF CONTAMINANTS
   WATERSHED CHARACTERISTICS
OVERLAND CONTAMINANT
 TRANSPORT MODELING
    (ARM MODEL)
   CHANNEL CHARACTERISTICS
   SEDIMENT CHARACTERISTICS
   UPSTREAM FLOW, SEDIMENT
   AND CONTAMINANT CONDITIONS
INSTREAM CONTAMINANT
MODELING (SERATRA AND
   EXPLORE MODELS)
TOXICOLOGICAL DATA
• LC50 VALUES
• MATC


STATISTICAL ANALYSIS
OF PREDICTED INSTREAM
CONTAMINANT CONCENTRATION
IN STREAMS (FRANCO MODEL)
                                    RISK ANALYSIS
                        Figure 2.
           Chemical Migration a"nd Risk Assessment
                  (CRMA) Methodology

-------
82    SITE INVESTIGATION & ASSESSMENT
  By the classical definition, risk is considered the product
of the probability of an occurrence and the consequences
of that occurrence.  Hence, if one in every 1,000,000 air-
plane landings results in an accident, and 200 people died
in the last  100 accidents (150 in one, 50  in one and 0 in
all others) the risk of death during airplane landings is 1 in
5,000 or 2 x 10 '\
  When the activity being  evaluated involves contaminant
migration,  the data are not available in such simple for-
mats. Hence, the definition is altered slightly to accommo-
date the nature of the inputs. Exposure-response analysis
is performed through the integration of results from an
exposure assessment and a hazard assessment. For these
purposes, an exposure assessment is a quantitative estima-
tion of the timing, concentration and duration of contam-
ination exposure at a specified location.  This assessment
is typically performed through use of mathematical models
that predict the transport, transformation  and  fate  of
chemicals in the environment.
  A hazard assessment is the estimation of the effects a re-
ceptor population will evidence if exposed to specified con-
centrations of contaminant for given periods of time. This
is usually done by constructing a time-dose relation chart
for the contaminant of interest.
  As the result of a recent EPA funded  development ef-
fort, (1) a methodology has been devised to integrate the re-
sults of the exposure and hazard assessments. The Chem-
ical  Migration  Risk Assessment (CMRA)  Methodology
couples a number of computerized  codes to accomplish
the integration as illustrated in Figure 2. A statistical code,
FRANCO (Frequency  Analysis of  Concentrations),  re-
duces the results obtained from transport models to a time-
concentration format consistent with the time-dose rela-
tion output from a hazard assessment.  Hence, when the
two are overlayed,  one gets a  numerical estimate of the
percent of time concentrations exceed specified hazard  lev-
els. The segmentation of the time-dose universe into haz-
ard ones is illustrated in Figure 3.
   This  method  has  been  applied to  the pesticide Toxa-
phene in the Four Mile Creek watershed in Iowa'" and the
Yazoo River Basin in Mississippi.(2) Example results are
provided in Figure 4. The numbers in each segment of the
time-dose universe  indicate the percent  of the time that
conditions at a downstream site would fall in each hazard
zone. A similar  approach  can  be employed to look at hu-
man health effects employing available data on acute  and
chronic dose levels and drinking water limits.
  There are two key issues to be addressed by the applica-
tion of exposure-response analysis to the site restoration
problem set:
•Which sites require remediation?
•How clean is clean enough?
The first of these concerns questions both the degree of
risk posed and the timing of the risks. Ideally, after con-
duct of the exposure-response analysis, one will be able to
categorize sites into one of three groupings:
•Sites which do not require remediation
•Sites which pose a long-term risk
•Sites which pose a short term risk
The first refers to sites  where risks fall below some max-
    c% POTENTIAL ACUTE
       DAMAGE        I
                          TIME
                        Figure3.
             Time-Dose Universe Employed for
                Exposure Response Analysis
mg/l


  14 x 10-1 -
  12 x 1CH
  10 x
O
z
   8x10
   6x10°
  1.5 x10-5
                               1 2%
                  13.7%
                        Figure 4.
           Toxaphene Exposure-Response Analysis
                 Results for Rainbow Trout

 imum acceptable level, i.e., risks are comparable to those
 accepted by people on a routine basis. The second group
 would include sites where risks may reach  unacceptable
 levels, but the impacts will not occur for a number of
 years.
   Generally  speaking, time and resource considerations
 will dictate addressing the third group only at the present
 time. There is no sense in using today's monies to avoid a
 problem far in the future at the  expense of other sites
 which will have  immediate effects. Caution is necessary
 here, however. For some  sites, waiting to implement t*

-------
                                                                  SITE INVESTIGATION & ASSESSMENT
                                                   83
medial action will result in disproportionately large costs
because of the nature of the contaminant's migration. As
an example, a contaminant plume in the unsaturated zone
could be much less expensive to exhume than that same
plume once  it is allowed to enter the  aquifer. Therefore,
a part of the  separation  into short and  long term risk
should cover the concomitant cost differential.
  The issue of "How clean is clean  enough?"  is at the
heart of the selection of a remedial action alternative. Be-
fore a preferred approach can be designated, a target  level
of contamination must be established such that areas  with
residuals at  higher concentrations will be addressed while
those at lower concentrations  are left untouched. Specif-
ically, cleanup criteria need to be established so one can de-
termine how extensive restoration should be and which al-
ternatives are capable of meeting the goals.
  No soil criteria presently exists for most toxic chemicals.
For nearly all contaminants, there are no threshold limits
defining when hazardous effects will  begin to be  evi-
denced. In  part,  this reflects the  fact that there are no
simple standard tests to which a contaminated soil can be
subjected for designation as hazardous or nonhazardous.
  Fortunately,  there  is still a relatively straightforward
approach available for establishing criteria. For most con-
taminants, soil residuals are of concern because of  their
ultimate ability to contaminate  the atmosphere (through
volatilization  or  resuspension)  and  the  hydrosphere
(through leaching and runoff). Hence, hazardous chemical
levels in soil can  be defined as those which will   sponsor
hazardous levels in air or water.
   Criteria and guidelines for ambient air  and water have
been suggested for a number of chemicals. By working
backwards  with  minimal data on dilution potential and
distribution coefficients, criteria can be established. Given
these, one can  determine  the subset of alternatives which
can meet objectives and the extent of restoration required.
   Viable remedial  action alternatives are evaluated for
effectiveness using the same models employed in the ex-
posure assessment. Predicted levels of removal on barriers
to migration are input to  the models and output reviewed
to  determine the impact on risk levels.  Once these values
are available, alternatives can be compared as indicated in
Figure 5. Eachf alternative is compared on the basis of cost
                                        EFFECTIVENESS
                                          RATING^
                                      RISK REDUCTION/COST
                        Figure 5.
     Approach to Selection of Remedial Action Alternatives
and risk reduction potential. Selection of the optimum ap-
proach proceeds in a two-step fashion:
•Which alternatives meet acceptable risk thresholds?
•Which of the above alternative offers the  most risk re-
duction per unit of cost?
  Having performed the analyses, one can  conclude  site
restoration. Assessment has allowed design of each step of
work in a cost-effective manner identifying only the neces-
sary and sufficient  data  to proceed  to subsequent steps.
Reduction of data to risk values maintains output in a cur-
rency that can be  communicated with decision makers and
employed directly in environmental  impact statements,
permit applications and other  documents associated with
the approval process.

STATUS

  Environmental-response analysis  is currently available
for application to site restoration.  The exposure assess-
ment elements  of the process  are particularly well devel-
oped. Validated  models are  available off-the-shelf  for
predicting migration in the atmosphere, groundwater and
surface waters. The major  area  of uncertainty  rests in
the ability to describe chemical interactions  with soil and
other attenuation mechanisms. Knowledge is particularly
lacking with respect to organic contaminants.
  Hazard assessment is much less developed. This is large-
ly true because of the paucity of data on effects of chronic
exposure to most contaminants. Hence, it is  impossible to
construct accurate time-dose universes for most contami-
nants. At present, a few data  points (largely acute in na-
ture) can be entered. This does not  invalidate the process
or diminish its ability. It merely points to the significant
efforts that must still be undertaken.
  In the interim,  exposure assessments can be employed as
the basis for  decision making. When hazard data exist, a
complete exposure-response analysis can be made. When
hazard data do not  exist, judgment can be employed as to
the  acceptability  of projected exposure levels. In either
event, decision makers are afforded all available data and
a formal process for utilizing that data.
REFERENCES

1. Onishi, Y., Brown, S.M., Olsen, A.R., and Parkhurst,
   M.A.,  "Chemical Migration and Risk  Assessment
   Methodology", Proc. of the ASCE National  Confer-
   ence on Environmental Engineering, July 8-10,  1981,
   Atlanta, Georgia.
2. Onishi, Y., Whelan, G., Parkhurst, M.A., Olsen, A.R.,
   and Gutknecht, P.J., "Preliminary Assessment of Tox-
   aphene Migration and  Risk Assessment in the Yazoo
   River Basin, Mississippi", Report to the Environmental
   Protection Agency by Battelle, Pacific Northwest Lab-
   oratories, Richland, Wash., April, 1980.

-------
MINIMIZING COST AND RISK IN HAZARDOUS WASTE SITE
                  INVESTIGATIONS USING GEOPHYSICS
                                         RICHARD BENSON
                                        ROBERT GLACCUM
                                             Technos, Inc.
                                            Miami, Florida
                                             PAUL BEAM
                                U.S. Environmental Protection Agency
                             Office of Emergency and Remedial Response
                                           Washington, D.C.
INTRODUCTION

  Hazardous waste sites require the very highest levels of
quality and confidence in site evaluations. All site assess-
ment projects require good information at minimal cost.
  Commonly, drilling monitoring wells and a sampling
program  are recommended to define subsurface condi-
tions at these sites.  This approach  often yields an in-
complete and/or erroneous evaluation. Errors arise due to
the small volume of subsurface space sampled by a moni-
tor well.  This small volume often is not representative
of the site conditions. An accurate analysis of site condi-
tions requires that the number  of sample sites be sub-
stantially greater than are normally used. Conversely, in-
creased drilling activity increases risk of fumes, fire, ex-
plosion and release of hazardous materials into ground-
water or the environment.
  Discrete well  sampling programs  should be designed
with  prior knowledge obtained from preliminary  surveys
using remote sensing geophysical methods.  This systems
approach will clearly minimize risk and cost,  while syner-
gistically improving the understanding of site conditions.
In investigating conditions at such hazardous waste sites,
three types of problems are commonly confronted:
•Determining  the  presence and distribution of buried
 hazardous materials.
•Determining the  presence  and  boundaries of contami-
 nants and leachate plumes.
•Characterizing the natural geohydrologic setting.
  Ground-based remote  sensing geophysical  techniques
are very  effective in providing information  for these
purposes and can be rapidly applied to a site. Having ob-
tained this information, a detailed drilling and sampling
program can then be designed, ensuring the placement of
monitor wells directly in the problem area.  This results
in  much greater confidence in the final data interpreta-
tion with fewer wells.
  Where site investigations have previously been initiated,
the geophysics systems approach also provides significant
benefits. The location of existing monitor wells  relative
to  problem areas can be assessed, providing an  evalua-
tion of the validity  of data previously  acquired. If needed,
additional well locations can be precisely identified to fill
gaps in the overall site coverage.
  Depending upon site objectives and conditions, the re-
mote sensing methods utilized will vary (Significantly. A
wide range of methods and variations have already been
successfully applied to many hazardous sites throughout
the United States.
  Successfully applied techniques include:
•photography
•Infrared scanners
•Ground penetrating radar
•Various electromagnetic
 conductivity measurements
•Metal detection surveys
•Magnetic surveys
•Resistivity surveys
•Seismic refraction surveys
  With the exception of resistivity and seismic refraction,
all of these remote sensing methods provide continuous
measurements  along the survey line.  These continuous
spatial measurements are preferred as they enable site
coverage to  approach 100%. Continuous spatial  mea-
surements have significant benefits when applied to sites
which  are highly variable, as  they avoid incomplete in-
formation obtained from station measurements.
  In this paper, the authors outline the benefits of geo-
physical remote sensing assisted programs compared to
traditional drilling and sampling methods.
SPATIAL SAMPLING REQUIREMENTS

  A critical problem at many hazardous waste sites is the
determination  of the location  of  buried  materials or
plumes. In addition, the investigator often needs to know
the amount of material, its  boundaries and depth of
burial.
  The area of a site under investigation is often many
times the area of the buried target (Figure  1). Ratios of
typical site to target areas (As/At), encountered at hazard-
ous waste sites, range from 10 to 1000. As this ratio in-
creases, the search problem can rapidly become analogous
to "looking for a needle in a haystack." The table belot
illustrates some typical dimensions found for various As/
At ratios, assuming a site area (As) of 25 acres.
                                                   84

-------
                                                                   SITE INVESTIGATION & ASSESSMENT    85
                       Areo of Target
 Area of Site - As
                                                             100 T
                        Figure 1.
                 Site to Target Area Ratios
            (Common ranges of As are 10 to 1000)

  As/At      Target Area
  10          330 ft x 330 ft (2.5 acres)
 100          100 ft x 100 ft (0.25 acre)
1000          33 ft x 33 ft (0.025 acre)
  Numerous wells may be required to detect the presence
of a burial target or plume (Figure 1). A traditional grid
search pattern applied to a site-to-target ratio of 10  is
shown in Figure 2.  More than 16 wells  are required to
approach a 100% confidence level, just to detect the tar-
get. If the target size really is smaller than assumed, the
selected drilling program may not detect it. A "miss" will
obviously lead to an  erroneous assessment of the site.
These steps are concerned only with locating a target and
not providing any definition of it. Definition of the tar-
get's  boundaries will require additional wells, and as
boundary resolution requirements increase,  more wells
will be necessary. As an example, using statistical con-
verging methods, a  minimum of at least eight to twelve
wells would be required to approximately locate four sides
of a simply shaped  burial site.  As many as 20 or more
wells  may be required to define  complex-shaped  boun-
daries. A total of 24 to 36 wells would be  necessary to de-
tect, locate and provide approximately spatial boundaries
for a complex site with an As/At of 10 to obtain high con-
fidence levels. For  the As/At ratio of 100 to 1000, the
number of wells and the related costs would be unthink-
able.
  Often the risk  of drilling is significant and the benefit of
remote sensing prior to drilling  is therefore critically im-
portant.  Breaching  the clay seal of a trench, puncturing
a drum, releasing of fumes and gases and possible ex-
plosions and fire are all real potential risks when drilling
sites with hazardous material present. Although a large
number  of  wells is  necessary to  detect  and define the
target  (Figure 2), a concurrent,  significant  increase in
risk is incurred (Figure 3).
  The discussion of location of buried sites is also appli-
cable  to  the location and mapping of plumes and  defin-
ing natural geohydrologic  conditions. Limited monitor
well  installations and  sampled  collections  to  define
                                                           S  50--
                                                                                             Grid Search Pattern
                                                                                        Curve for As /AT = 10
                            10
                         Number of Wells
                                                   20
                        Figure 2.
         Detection Probability Vs. Number of Holes
-I
E
o •»
jf ir
                           50
                  Probability of Detection (Percent)
                                            100
                        Figures.
           General Curve of Risk Vs. Probability of
                   Detection by Drilling

groundwater plumes at  hazardous  waste sites are often
insufficient to  adequately define the conditions. Extra-
polation beyond and interpolation between data points is
necessary when using this approach.
  Results,  therefore, are highly  dependent upon the uni-
formity of subsurface conditions.  Variations in the na-
tural setting as well as interaction  with specific  contam-
inants make such estimates highly prone to error.

EXAMPLES OF SITE INVESTIGATIONS USING
DRILLING AND ELECTROMAGNETIC METHODS

  The  following examples are included to illustrate the
problems and variations encountered in several hazardous
waste site investigations.

-------
86    SITE INVESTIGATION & ASSESSMENT
                   3-D   Representation  of Conductivity  Data  Showing
                   Buried  Hazardous  Materials
                                                                                    =  Well
                                                   Figure 4.
                                           Six Wells Miss Buried Target
Example 1
  Six wells  were drilled  in  a 25-acre site  with  a site/
target area ratio of 10. These wells were placed using the
traditional "best guess" approach—one near each corner,
one  in the center and  another at a suspected  location
(Figure 4). None of the wells located the buried  wastes.
Two wells  showed  low  levels of contaminants in the
groundwater samples. Consequently, investigators knew
that  a problem existed at the site but still had  not lo-
cated the burial area or had information about the magni-
tude of the problem.
  This result is not surprising as Figure 2 shows that for a
site/target area ratio of 10,  approximately  sixteen wells
would be necessary for a  high probability of detection at
this site.
  The site was  then mapped using electromagnetic (EM)
methods which measure  bulk  subsurface conductivities.
Twenty continuous EM profiles were run across  the site
in parallel lines spaced 50 ft apart.  These results clearly
defined an irregular, anomalous zone in  one portion of
the site (Figure 4). The results of this survey provided the
following information:
•Initial location of the  target area was made within the
 first hour of survey
•Approximate boundaries were identified by the end of the
 first day
•Qualitative EM sounding data indicated that most of the
 buried wastes were less than 10 feet from the surface
•Magnetometer profiles  over  the same area indicated
 that no large concentrations of ferrous materials such as
 steel drums  were present, indicating the contaminants
 were not buried in drums
•Computer processing of the EM data revealed that con-
 taminants were leaving the burial zone

  Based  upon  these results,  two selected  wells  were
drilled to verify and characterize the wastes and to identi-
fy the contaminants leaving the site.
  The geophysical surveys not  only  located the burial
zone, but  resolved the exact boundaries  of the burial site
as well as location of a contaminant plume. An estimate
of contaminant volume and depth was also  made froiD
the data.  To provide information at this  level of detail.
drilling costs  would have been prohibitive,  requiring mort
than three dozen  wells and associated chemical analyse!.
Well installation  time and  turn-around  time for analysB
would have also  been an important factor in the overall
project schedule.  Since hazardous materials were present
the cost of special drilling procedures and the risk factff
of drilling this number of holes would have  been large.
  Geophysical  results  provided  essential  informational"
less  than  one  week  and only  two to  three monitorinj

-------
                                                                SITE INVESTIGATION & ASSESSMENT    87
wells were necessary. A high level of confidence at mini-
mal cost and time was achieved.

Example 2

  The  extent  and  direction of  contaminants  seepage
and leachate plumes and the evaluation of groundwater
flow near hazardous waste sites  are major concerns. In
ideal homogeneous'isotropic soil and rock,  textbook ap-
proaches using water levels and potentiometric heads may
yield predictable and calculable results. The real world,
however, is quite  often not such a simple system. Local
groundwater flow cannot always be equated to  regional
groundwater  flow and  permeabilities  determined  in  a
laboratory from a few samples do not always represent
in situ values.  Even though water level measurement can
yield general flow directions, complex geological settings
may create localized variations in this flow. In addition,
contaminants denser than water may not  follow  "flow
lines"  but  may instead be  controlled by  gravity.  Data
from the same area, one resulting from observations
wells,  the other from electromagnetic geophysical  sur-
veys, are shown in Figures 5A and 5B.
   Figure 5A was developed based on data  from 16 open
hole  wells which  were  also used  for water  quality
sampling. From these data,  the generalized groundwater
flow and assumed plume direction were determined. The
actual plume as measured by EM conductivity is shown
in Figure 5B. A much more detailed and accurate defini-
tion of the plume is provided by the EM method.  Suf-
ficient data were acquired over the site so as to provide
nearly total coverage. The plume shown  in 5B  repre-
sents flow of  a  conductive,  conservative contaminant
through both  the saturated and unsaturated zones. The
confidence level of the detailed flow pattern approaches
100% (shown in Figure 5B), while that of the water level
measurements provides a  considerably  lower confidence
level (Figure 5A).
   To achieve  the high resolution results seen in Figure
5B with monitor wells would have required an extensive
number of strategically placed wells. The problem occurs
in placing the wells in the correct position which is typi-
cally accomplished by a "best guess" method. The geo-
physical  systems  approach  provided  not  only  detailed
data on plume flow and concentrations, but also enabled
an assessment to be made of contaminant data from exist-
ing monitor wells.

Example 3
  Another example of variability in vadose and ground-
water flow can be seen  in  Figure 6. A  point source
contaminant has flowed through a thin vadose zone into
the shallow  groundwater. The site was originally con-
sidered to be  quite homogeneous and isotropic.  How-
ever, an EM conductivity survey showed the distribution
of contaminants within this regime as very complex and
changing drastically with time.
  An attempt  to utilize monitor wells for such mapping
could provide  highly misleading  information on the dis-
tribution and degree of contamination. Any conceivable
    Values
    in  Feet
  A)   Potentiometric Contours Showing
       Flow Directions
    Values in Millimhos/Meter
  B)  Conductivity  Contours Showing
       Plume Flow
                       Figures.
       Difference in Groundwater Flow at the Same Site
    Based upon Monitor Wells (A) and Em Measurements (b)

plan to derive similar results using monitor wells would be
at a low confidence level.
  This example shows that it is possible to  measure time
variables at a site using geophysical methods. Conducting
surveys  at  periodic intervals, the EM geophysical ap-
proach provides an excellent means of monitoring plume
development with time as well as providing an excellent
early warning system. In addition, the information ob-
tained  through mapping plume concentrations  and ge-
ometries  can  be used to significantly improve the use of
predictive models.

COSTS

  The relative costs  and benefits of an integrated remote
sensing  geophysical systems survey combined with se-
lected wells compared to a program consisting of monitor
wells only are shown in Figure 7. The curve shows cost
versus various site-to-target  area ratios. The monitor well
costs are based upon a total program cost of $3,000 to
$10,000 per well which includes drilling, construction of a

-------
88  SITE INVESTIGATION & ASSESSMENT
          Point of Spill
        Regional  Ground-Water Flow
                                                   I  Acre
                A)   I   Day
                  B)   26  Days
                                                    Figure 6.
                                         •Change in Shallow Plume with Time
                                         •Plume Measured by Em Conductivity
 quality monitor  well, initial  sampling and  analysis of
 priority  pollutants  and  a  subsequent quarterly  sample
 program over one year, along  with supporting interpreta-
 tion reports  and project management. All wells  are as-
 sumed to be 30 ft deep. The costs used are conservative
 in that they  do not include the increased direct costs of
 drilling into  highly hazardous  areas, nor do they  include
 consideration of intangible risk factors.
   The geophysical systems  costs are based upon a site
 survey coverage  approaching  100% plus three monitor
 wells as described above and all supporting effort of re-
 ports and project management. One well is typically used
 for background measurement  and two placed with high
 confidence levels to  assess  spatial distribution of con-
 taminants or attenuation along a plume. The curve shows
 the geophysical systems approach is more  effective for
 sites of As/At greater than 5.
   The geophysicals-assisted approach becomes increas-
 ingly more cost-effective than monitor wells when more
 wells are required. As site complexity, unknowns  and
 risks  increase,  the advantage of a systems approach be-
 comes even more significant.

 SUMMARY

   In this paper, the authors have discussed some of the
 problems,  variables,  errors and  risks  encountered in
 hazardous waste sites  evaluations. Some of the trade-offs
 that may exist in planning  and implementing such work
 are shown in Figures 1,2 and 3.
  The number and placement  of monitor wells  are often
 inadequate to describe the natural  geohydrologic setting,
 evaluation  of burial sites or  contaminant plumes from
 hazardous waste sites. As a result, significant errors can
 be made in assessing conditions due to a lack of adequate
 spatial information.
  Alternatives now exist to alleviate problems inherent in
 direct sampling methods alone. High density or continu-
 ous spatial mapping via remote  sensing and geophysical
   1000 y
                     10
                        Area of Site /Area of Target
                       Figure 7.
  Cost Comparison of Monitor Wells Vs. Geophysical Approach

methods  provides  significant  benefits.   No  single
method will solve all site  problems and  geophysical or
remote  sensing  technology is  not  a panacea in itself.
Trained, experienced professionals are required to design
and execute surveys and analyze data. Then, an effective
monitor well system may be implemented to provide di-
rect sampling and analysis  of the contaminants. This in-
tegrated systems approach  provides synergistic .results
unattainable by direct sampling methods alone.
  As the site-to-target area increases and the complexity.
details and hazards increase, the cost/benefit of a geo-
physics-assisted systems approach becomes  more signifi-
cant. The technologies and  benefits are available today to
those wanting  to  optimize  hazardous waste site assess-
ments.

-------
     INSTALLING GROUNDWATER MONITORING WELLS
                       AT A HAZARDOUS WASTE SITE

                                    RICHARD SPEAR, Ph.D.
                               U.S. Environmental Protection Agency
                                        Edison, New Jersey
                                      PETER FRANCONERI
                                    Fred C. Hart Associates, Inc.
                                        Newark, New Jersey
INTRODUCTION

  The installation of monitoring wells at a hazardous
waste site is important in a groundwater investigation.
There is, however, very limited experience using cumber-
some safety equipment and following strict safety pro-
cedures on hazardous waste sites during well drilling.
  Since February 1980, Fred C. Hart Associates, Inc. has
provided a Field Investigation Team to assist in the U.S.
Environmental Protection Agency's investigations of haz-
ardous waste sites in Region II. During the past year and
a half, the Region II Field Investigation Team super-
vised the installation of 42 groundwater monitoring wells
at five hazardous waste sites in New York and New Jer-
sey. Three  of these sites are now Superfund sites.  Based
on this valuable experience, guidance is offered for similar
undertakings.
  The material presented in this paper is specifically in-
tended to provide a very general overview and serve as a
primer for managers, attorneys and other non-hydrogeolo-
gists. Non-hydrogeologists working in hazardous  waste
management should understand the  basics of installing
monitoring wells. More detailed information can be ob-
tained from the sources provided in the Bibliography at
the end of the paper.
  The sequence of activities  typically required to  install
groundwater monitoring wells at a hazardous waste site is
shown in Figure 1.  Each step  in the flow chart is dis-
cussed.
COLLECT
EXISTING
WELL LOGS


CONDUCT
SITE
VISIT


DESIGN WELL
INSTALLATION
PROGRAM
                                                        WRITE SPECIFICATIONS
                                                        FOR TEST BORINGS AND
                                                         WELL INSTALLATION
                                               Figure 1.
                               Block Flow Diagram for Installing Monitoring Wells
                                                 89

-------
90     SITE INVESTIGATION & ASSESSMENT
PRELIMINARY STEPS
Collect Existing Well Logs
  The first step in installing groundwater monitoring wells
is to collect existing well logs. A geologic well log provides
a detailed record of the lithology of the bore hole con-
taining the well. In addition  to geologic well logs, geo-
physical well logs and laboratory test data such as perme-
ability and grain size analysis  should be obtained if avail-
able. The purpose of collecting existing well logs  is to
facilitate well design for the drilling program. Very often,
no  well log exists for a well,  or no one knows who put
the well in the ground.
  Existing well logs can usually be obtained from the fol-
lowing sources:
•States-
  Some states, such  as New Jersey, require that all wells be
  permitted and that well logs for recently completed wells
  be filed with the state. In these cases, obtaining existing
  well logs should be easy.
•U.S. Geological Survey (USGS)—
  The USGS  collects  well  logs in some states.  Their re-
  gional office may have copies of well logs even if the state
  collects them.
•Driller—
  In many cases, the driller keeps logs of the wells he drills.
  The USGS or  the state might know who did the drilling.
•Owner—
  The owner  of the well often receives a copy of the well
  log or knows who  drilled the well.
Conduct Site Visit
   The investigator should conduct a site visit to become
familiar with site conditions and identify potential drilling
constraints. The site visit should determine the data need-
ed by the investigation such as any site specific features.
To facilitate monitoring  well  design, existing  sampling
points such as springs, wells, storm sewers and lagoons
should  be  located. Any feature of the proposed drilling
site which  limits  access of the drill equipment should be
noted. Typical limiting features include steep grades,  for-
ested areas or areas with  drums. The investigator should
also note any area containing surface contaminants. These
must be avoided to protect the safety of the drilling party.
   The investigator should  proceed with caution during the
site visit because at this early point the risks are still rela-
tively unknown.  In some cases, a "windshield  survey"
should first be conducted  by driving the perimeter of the
site to verify aerial photo information and topographic
map data.  There should be adequate personnel on hand
during the actual  site visit to provide assistance in case of
an emergency.

DESIGN OF WELL INSTALLATION PROGRAM

  After conducting a  site visit, the investigator can de-
sign the well installation program. The purpose of design-
ing the well installation program is  to establish the meth-
ods of conducting the groundwater investigation.
  First, the investigator must identify the drilling method
most compatible with the site geology. He must then de-
termine the number of wells necessary to provide the best
spatial coverage of the site with a reasonably high level of
confidence. Using information gathered  from the site
visit and records review, the investigator can identify the
aquifer most susceptible to contamination.
  Next, fully penetrating wells can be located. In this step,
care must  be taken to avoid breaching  confining layers
because this could lead to aquifer cross contamination,
The wells should be designed using non-reactive materials
such as stainless steel or teflon-coated screens.
  If two or more aquifers are potentially subject to con-
tamination, clustered, nested wells  are sometimes neces-
sary to define  differences  and interaquifer leakage. Fin-
ally, the investigator should determine if the wells should
be gravel packed  or naturally developed. If falling head
testing is planned on  the monitoring  wells, gravel pack-
ing should be  avoided because the test results could be
skewed by the  higher permeability of the artificial gravel
pack. As an alternative, the falling head test  can be per-
formed in situ before installing the gravel pack, and the
well can be completed after testing with the gravel pack
in place.
Write Specifications for Test Borings
and Well Installation
  Specifications must be written for test borings and well
installation  to  define the required scope  of work to dril-
lers who bid the drilling contract. The long  term needs of
the groundwater investigation must be considered along
with immediate needs when writing specifications.
  The elements of monitoring well installation which must
be  specified include anticipated lengths  of casing, type
and length of screen, sampling requirements, development
procedures, gravel packing, grouting and  final cement-
ing. The method of construction required under each op-
tion should be fully  described. Options should be in-
cluded to let the driller select the best drilling method to
fit the site characteristics. If use of hydraulic rotary is an
option, the driller must be  told not to use contaminated
water for the drilling operation so the aquifer is not con-
taminated  by  drilling. For  quality control,  samples of
the drilling  water should be taken and sent to  a labora-
tory for analysis. The proper type of drilling mud should
be specified. Organic agents or  other items typically used
by  a driller which have the potential to react with con-
taminants in the groundwater are not always suitable.
  Finally,  a plan for properly abandoning unsuccessful
well sites should  be described. Typical methods include
grouting the unsuccessful borehole with  an impermeable
material such  as  cement or  bentonite. This step is in-
portant because an open hole breaching  a  confining unit
increases  the  potential for cross  contamination of to*
aquifer.

DETERMINATION OF SAFETY REQUIREMENTS

  The first step in determining safety requirements it W
identify  hazards and analyze  their potential  effect A

-------
                                                                     SITE INVESTIGATION & ASSESSMENT    91
hazard analysis should be based on the following consid-
erations:
•Type of waste (toxicity, explosivity, volatility)
•Quantity of waste
•Site topography
•Weather conditions
  Based upon the hazard analysis, a safety plan should be
developed and safety equipment requirements should be
determined.
  When drilling takes place in an area where hazardous
materials  could  be  encountered, the  driller must  be
warned. The safety plan should include provisions to re-
duce the risk of exposure to contaminated soils, water or
vapors  that could be released  during the drilling opera-
tion. Safety equipment and safety training to be provided
by the driller must be specified so the bidders can include
the cost of these requirements in their bids. For example,
at one site, monitoring wells had to be installed in a land-
fill  containing polychlorinated  biphenyls (PCB's)   and
other contaminants. At this site, the investigator and dril-
lers  had to wear  self-contained breathing apparatus  and
butyl rubber rainsuits  for safety.
  The safety plan must consider the strenuous effort that
well drilling involves. Consideration must be given to the
stress and discomfort caused by wearing  of safety equip-
ment during extreme hot and cold  weather.  Stress  and
discomfort must be balanced against  the necessity to limit
exposure.  The driller must be made aware of all safety
requirements and agree to them  upon award of the  con-
tract.
BIDDING
Assemble Bid Package for Drillers
  A drilling bid package must be assembled to give the
drilling contractor enough information to reasonably esti-
mate the costs involved in drilling at the site. Local dril-
lers should  be consulted to confirm the feasibility of the
required drilling.
  The drilling bid package should include a map provid-
ing the precise location of the site, a site sketch, geologic
cross-section sketch, a short description of the geology of
the site and the type of conditions the driller  may en-
counter. The bid package should contain a statement ad-
vising the driller  he is  responsible  for  verifying all  site
conditions.  The driller should be told to visit the site if he
needs more  information to prepare his bid.
  Well construction should  be described. Abandoned
procedures  for unsuccessful borings must be specified to
make certain  that abandoned  wells are sealed to prevent
interaquifer contamination.
  Unit prices should be requested.  One option involves
bidding  on  a  footage-plus-time-and-material basis. An-
other option  is a straight time-and-materials basis. The
latter method tends to be simpler, but it can be abused by
drilling contractors who work  slowly. To avoid this  prob-
lem, time and material cost can be tied to a bonus  to be
paid for every foot drilled  over a specified minimum
footage.
  The driller should be instructed to obtain any necessary
permits. In some states he must prove that he is a duly
licensed driller  in that  state.  In these cases,  the  driller
should be asked to submit his license number with his bid.
  Materials to  be used by the driller, such  as  casing,
screens, bentonite, gravel  and cement must be specified.
The driller should be told when he will be expected to col-
lect split spoon samples  and how to store and label them.
He should also be told  which development methods will
be necessary to insure the wells are  adequately open to
the formation. The driller  must also be told when drilling
should begin and end.
Obtain Bids From Drillers
  The investigator should develop a  list of reliable (and
preferably) local bidders. At least three drillers should be
sent a bid package and  asked to bid. For  sites requiring
high  levels  of safety protection, such as  self-contained
breathing apparatus, bidders may have to be drawn from
a wider geographical area.
  Drillers usually require at least two weeks to  review the
drilling specifications and other requirements. After com-
pleting their review,  the drillers can  prepare a cost esti-
mate  and determine their time requirements  to do the
drilling.
Category

Drilling

Mobilization
Experience


Safety
Project
Comprehension

Liability
Insurance
Costs
References
          Table I.
    Driller Selection Factors

               Selection Factor

•Availability of drilling equipment
•Qualifications of personnel
•Mobilization plan
•Time required from notice to proceed to on-site
 drilling
•Drilling experience in landfills
•Drilling experience in hazardous waste sites
•Experience with split-spoon sampler
•Attitude toward safety plan
•Availability of personnel trained in safety
•Willingness to train personnel in safety
•Availability of safety equipment
•Prior projects involving safety procedures
•Understanding of decontamination responsibility
•Experience with decontamination of personnel and
 equipment

•Understands project requirements
•Understands project complexity
•Amount
•Type coverage
•Realistic breakdown of costs
•Cost limitations
•Lump sum costs
•Hourly rates
•Names
•Phone numbers
Evaluate Bids and Select Driller
  The investigator must identify driller selection factors
as a basis for comparing bids. A suggested list of consid-
erations when evaluating drillers' bids is given in Table I.
  Although price is important, a driller should never be
selected because of a low price alone. References should

-------
92    SITE INVESTIGATION & ASSESSMENT
always be contacted to insure  that the driller's previous
work has been done properly.
Award Drilling Contract
  After the driller is selected, he should  be notified at
once. If he accepts the assignment, he should be sent a
contract to  sign. The drillers who bid but were not  se-
lected should be so notified in writing as a courtesy. This
simple step will encourage them to bid on similar work in
the future.

DRILLING
Meet with Driller to Establish Schedule
  After  the contract  is  signed  by the  driller, the  in-
vestigator should meet  with him to confirm the drilling
schedule. This meeting  should  be held at the site, especi-
ally if the driller has not visited the site.
  The  following  factors  will  determine  the drilling
schedule:
  (1) Accessibility of drilling locations
  (2) Drilling methods
  (3) Safety equipment required at each drilling location
  (4) Weather  conditions,  especially  temperature  ex-
      tremes

  The length of time required to drill at a hazardous waste
site depends  heavily on the safety  equipment require-
ments.  Drilling is strenuous work and safety equipment
can make  the job much more difficult. Wearing  safety
equipment  can double  the  required  length  of time to
drill, especially during very hot and humid weather.
Obtain Well Permits
  Some states  such as New Jersey require the registration
of all wells drilled by the state. This requirement provides
a centralized  information  source  for  well locations and
their respective logs.
  Either the  driller or  the  investigator can assume  re-
sponsibility for acquiring the necessary permits. The per-
mit  should be registered in the name of the investigator's
firm to facilitate later inquiries about the well.
  After  drilling  is completed, a  copy of the  well logs
should be submitted to the state agency which issued  the
permit. This will complete the record concerning the mon-
itor well and  make it easier to compile well information
in the future.
Obtain Clearance to Drill
  Drilling to  detect possible contamination  may require
the  wells to be placed  either on the  site or  on privately
owned land.  For wells  which  are to be drilled on pri-
vately owned land, permission must be  obtained from  the
owner.  State-owned  land  requires  obtaining  clearance
from any  public service companies which   might  have
underground  transmission  lines,  electric lines  or  water
mains  which could be  damaged during the  drilling pro-
cess. The driller is often best qualified  to  obtain  these
clearances  since he will most  likely  be  more familiar
with the local  area and  would know which  utilities have
to be contacted. This is a good reason to use local drillers.
                                     GROUND SURFACE
   BENTONITE
   PELLETS
       SAND
                                          WELL SCREEN
                        Figure 2.
       Simplified Schematic of a Typical Well Installation

  Some states have a centralized agency which handles
clearance matters for a number of large utilities. Even in
these cases, additional information  should be obtained
from the local municipality to help  locate underground
lines.
Drill Test Borings
  Test borings  are  drilled  to  provide information con-
cerning site geology. In turn, this information is used to
determine well locations and depths. Test boring informa-
tion on site geology is  obtained by analyzing split-spoon
samples taken at various depths,  usually every 1.5 m or
when lithology changes.
  Monitoring wells  can be installed in the  test bore hole
after completing the boring. This  method saves a consid-
erable amount of time  and money since it  eliminates  the
need of an additional borehole.

WELL INSTALLATION

  The investigator must supervise the installation of wells
to insure the well design is properly followed. Figure 2
is a simplified  (many details have been purposely omit-
ted) schematic  of a typical  monitoring well installation.
The specific well design used in a  given situation depends
heavily on site geology and  the purpose of the well.  For

-------
                                                                 SITE INVESTIGATION & ASSESSMENT     93
example, a well with a long life  requirement  should  in-
clude a lock to prevent tampering.
  In many cases, the borehole is drilled, the screen is set,
the well is developed and finally the well  is completed.
Each of these steps must be carefully monitored by the
investigator.
  A borehole must be drilled to  contain the well  screen
and casing. Materials used in the drilling process must be
carefully monitored and accurately recorded to avoid in-
fluencing water quality and subsequent chemical analyses.
The investigator must  carefully monitor what goes into
the borehole to prevent any contamination from migrat-
ing further due  to drilling techniques. For example, the
confining layer  between  a contaminated  and  "clean"
aquifer should  not be breached without sealing off the
upper part of the aquifer.
   A well screen must be set in the  zone where ground-
 water is to be monitored. The ideal screen setting depth is
 in the zone  of highest contamination and highest perme-
 ability. Usually, both of these conditions are found in
                        FigureS.
         Setting up drill rig and connecting to air supply
                        Figure 4.
          Drillers and investigators at drill rig location
the same  zone because contaminants tend to be  trans-
ported by groundwater.
  When volatile organic contaminants  are not  present,
test boring information must be used to locate the highest
permeability  zone  and establish the well screen setting
which can be temporarily set at various depths to obtain
groundwater  samples. These groundwater samples from
various depths can be scanned with an organic  vapor
analyzer (OVA). This instrument can be used as an on-
site, portable gas chromatograph to analyze water samples
from various depths.  Representative samples of ground-
water at a specific depth are obtained by pumping the well
at that  depth. A representative water sample from each
depth is analyzed with the OVA to identify contaminated
zones of the aquifer.
  The proper depth for  the permanent screen setting is
determined by scanning the water sample with the OVA to
determine the presence of organic contaminants.  Split-
spoon samples can also be scanned with the OVA to de-
termine proper well screen setting depths.
  After the well screen is set in a zone, the well must be
developed. In sand and gravel formations, the well can be
naturally developed by pumping water through the screen.
This will  allow the formation to  collapse around  the
screen as  fine particles are  removed. When clear water
begins to be pumped, the development process is complete.
  After the well is  developed, the well installation must
be completed by filling the annular space between the bore-
hole and the well casing. For example, sand can be poured
into the space part of the  way, then  bentonite pellets
and finally cement can be poured up to  the ground sur-
face to  prevent contamination by surface runoff. Sand is
often mixed with bentonite to form grout and  this-mix-
ture can be used to prevent downward flow of water.
  The investigator must  also  monitor  safety during the
well installation process.  This important function can be-
come very difficult when there  are very hot  outdoor
temperatures and the drilling team would prefer to relax
safety requirements to avoid bodily harm from  heat ex-
haustion.  The investigator should monitor  safety using
the OVA. He should stop drilling operations if OVA read-
ings of ambient air show high concentrations of volatile
organics.
  Figure 3 shows a driller setting up his drill rig and con-
necting to the breathing air supply. In  Figure 4, a closer
view of the drillers and the investigator wearing protec-
tive clothing at the drill rig location is provided. In Figure
5, one  can see a driller's helper checking the breathing
air connections. Figure 6 shows a  close-up view of the
chemical  resistant suit worn by  the  driller. The drill
auger flight  and  150-lb  hammer are visible behind the
driller.

ESTABLISH WELL LOCATION

  Completed  monitoring wells  must be accurately  lo-
cated horizontally and vertically. Accurate well location
is necessary to determine groundwater gradients and to
tie groundwater data  in with surface water data. In this
effort, accurate horizontal well locations are usually less

-------
94     SITE INVESTIGATION & ASSESSMENT
                        Figures.
       Driller's helper checking breathing air connections
critical than are accurate vertical  well locations.  If ac-
curate large scale maps or controlled  aerial photos are
available for the site, the horizontal location may be de-
termined by chaining to four existing topographic features
such as buildings, culverts or utility poles.
   If surveying expertise is available in-house or can be
developed, significant savings of time and money can be
realized. For example,  EPA Region II conducted an in-
house  training program to enable vertical location surveys
to be done without resorting to subcontract.
   Because  there are significant differences  between do-
ing a vertical location survey in-house and doing it by con-
tracting the work out, guidance for either method is pro-
vided below.

Conduct Vertical Location Survey In-House
   Conducting level surveys typically requires less survey-
ing expertise and experience than do most  other ground
survey tasks. Vertical surveys can often be conducted in-
house  using personnel  who have  been given  adequate
equipment and training.
  The following guidelines are recommended when con-
ducting a vertical survey:
  (1) Decide on a fixed elevation reference point on all
      wells to eliminate confusion. For example, always
      take the elevation at the northernmost point of the
      inner casing.
  (2) Establish a temporary benchmark near each well or
     cluster of wells to facilitate rechecking elevations.
  (3)  If possible, resurvey at the end of long term moni-
     toring campaigns.
  For  level surveys,  all  measurements must  be part of a
complete level circuit, beginning and ending at a properly
documented state or federal monumented benchmark.
Conduct Vertical and Horizontal Location
Survey by Contract

  If in-house surveying expertise is not available, it will be
necessary to  establish horizontal and vertical well  loca-
tions using services of a professional land surveyor. Sur-
veyors may not be familiar with groundwater investiga-
tions, particularly at hazardous waste sites. Several items
must therefore be included in the specifications and bid
package to prevent confusion and minimize costs.
  Accuracy requirements must be specified for horizontal
and vertical location. Typically, absolute vertical accuracy
of 0.03 m and horizontal accuracy of 30 m will be ade-
quate  for most  investigations.  Between wells, relative
horizontal accuracy standards may be more rigorous.
  Required safety equipment  and  training for survey
crews must be specified.  Surveyors will generally require
training if self-contained breathing apparatus is required.
Based  on Region II experience, operation of transit or
level is possible while wearing  full-face respiratory pro-
tection. The investigator should serve as Safety Officer
on-site and provide technical supervision.
  Materials, construction and documentation should be
specified to establish temporary benchmarks at each well
or cluster of wells. Proper temporary benchmarking will
facilitate rechecking well elevations.
  Required documentation such  as large scale location
maps,  location sketches and reports should also be sped-
                        Figure 6.
              Driller in chemical-resistant suit

-------
                                                               SITE INVESTIGATION & ASSESSMENT    95
fied. The bidder should also be told if documents on re-
producible media such as sepia prints or film (Mylar) are
required.
  After bids are received, the bid evaluation for selecting
a surveyor should be based on his understanding of the
special requirements for such surveys, ability to comply
with schedule requirements, and  cost.  Licensed profes-
sional land surveyors  will  have demonstrated necessary
technical surveying expertise as part of the state licensing
process.

ACKNOWLEDGEMENT
  The contributions made by Thomas Hughes and James
Shirk in preparing this paper are gratefully acknowledged.

REFERENCES
  More detailed  information can be obtained from the
following sources:
1. Scalf, M.R., McNabb, J.F., DunLap, W.J., Cosby,
   R.L. and  Fryberger,  J.,  "Manual  of Groundwater
   Sampling Procedures," NWWA/EPA  Series,  1981.
   National Water Well Association, Worthington, Ohio.
2. "Procedures Manual for Groundwater Monitoring at
   Solid Waste Disposal Facilities." U.S. EPA. Publica-
   tion No. 2 PA/530/SW-611. Aug. 1977.
3. DunLap,  W.J.,  McNabb, J.F.,  Scalf, M.R., and
   Cosby, R.L. "Sampling  for Organic Chemicals and
   Microorganisms  in the Subsurface." Publication No.
   EPA-600/2-77—176, Aug. 1977.
4. Pettyjohn, W.A., DunLap, W.J.,  Cosby,  R.L. and
   Keeley, J.R., "Sampling Groundwater  for  Organic
   Contaminants." Groundwater. 19, 1981, 180-191.
5. Middelburg,  "Methods for Sampling Small Diameter
   Wells for Chemical Quality Analysis," National Conf.
   on Quality Assurance  of  Environmental  Measure-
   ments, Denver, November 27-29, 1978.

-------
MANAGEMENT OF ANALYTICAL LABORATORY SUPPORT
         AT UNCONTROLLED HAZARDOUS WASTE SITES

                                     MICHAEL GRUENFELD
                                           UWEFRANK
                                        DAVID P. REMETA
                                        RICHARD LOSCHE
                               U.S. Environmental Protection Agency
                          Municipal Environmental Research Laboratory-Ci
                                         Edison, New Jersey
INTRODUCTION

  The recently enacted  "Superfund" law (PL-96-510:
the Comprehensive Environmental Response, Compen-
sation, and  Liability Act of 1980) provides significantly
broadened legislative authority for response to spills and
other discharges of hazardous chemicals. "Superfund" es-
tablishes a trust fund for the mitigation of hazardous sub-
stances released from all sources including uncontrolled
hazardous waste disposal sites. "Superfund" also pro-
vides for the amelioration of pollution in groundwaters,
surface waters, soils, sediments and the  atmosphere. In
support of this mandate, EPA Research and Development
(ORD) has focused attention on generating new  and im-
proved technology for multi-media cleanup, removal and
remedial action at spills and hazardous waste dumpsites.'"
  The imminent threat posed by the continuing discovery
of abandoned hazardous waste sites necessitated the de-
velopment of a comprehensive management scheme to
coordinate analytical services at such environmental in-
cidents. ORD's Municipal Environmental Research Labor-
atory (MERL), through a program at its Oil and  Hazard-
ous Materials Spills  (OHMS) Branch in Edison,  New
Jersey,  has  developed a unique interlaboratory  manage-
ment approach to provide  broad scope emergency re-
sponse analysis capabilities. The Chemisty Group at the
OHMS Branch serves a dual role:  (1) managing  research
and development  projects  and (2) providing emergency
support  in the form  of analytical assistance.  The R&D
and emergency support programs are interrelated within
the overall management scheme of the laboratory, which
integrates the functions  of central,  mobile and service
laboratory entities, as depicted in Figure 1. Each unit with-
in this matrix may operate on specific or joint tasks with
continual input, feedback and integration with other units
to yield rapid responses to requests for analytical services.
  In this paper, the authors describe a specific approach
for providing accurate and timely analytical services dur-
ing environmental emergencies and include a detailed dis-
cussion of the mechanics involved in managing such a
concerted analysis effort. The paper portrays one of the
essential functions of the OHMS Branch, namely the de-
velopment of  a prototype  emergency response  analysis
capability, consisting of several integral units functioning
as a single entity  in coordinating  sample analyses. The
overall objective of integrating the individual  units is to
establish the capability to respond promptly and efficiently
to emergency situations, using the appropriate equipment,
facilities and methodology. The paper presents-an over-
view of this prototype response capability.

CENTRAL LABORATORY

  The OHMS Branch maintains a highly sophisticated
central analytical laboratory that serves as the nucleus or
coordination center of the individual laboratory entities.
The central laboratory conducts research studies designed
to develop,  evaluate and refine methods and technique!
for the detection, identification and quantitative measure-
ment of oil and hazardous materials in diverse environ-
mental matrices. The central laboratory is also responsible
for managing and supervising the practical application of
prototypal analytical methodology during emergency re-
sponse episodes. Interrelation of research  and develop-
ment responsibilities with analytical assistance needs is
therefore a key function performed by this laboratory.
ANALITICAl
ASSISTANCE
REQUESTS


OHMS BRANCH
                     Figure 1.
    Overall OHMSB Laboratory Management Scheme for
      Providing Emergency Response Analytical Service!

-------
                                                                                             SCREENING     97
Analytical Methodology
  The primary focus of method development activities in-
volves generating exceptionally rapid and direct methods
of chemical analysis for use  within the confined  work
space of mobile  laboratories  operating in remote field
locations. The OHMS Branch has developed a substantial
number of  analytical  methods  and techniques for  the
qualitative and quantitative determination of hazardous
chemicals in water, sediments and the contents of aban-
doned  chemical storage tanks and  drums. The method-
ology is in the form of individual  instrumental analysis
techniques,  sample preparation and analysis  procedures,
and  quality assurance/quality control guidelines.  These
protocols are employed by both the central  and mobile
laboratories to  ensure the  acquisition and  subsequent
dissemination of timely verified data.
  The analytical methodology developed by the laboratory
utilizes an assemblage of instrumental approaches includ-
ing gas chromatography and fluorescence and infrared
spectroscopy. The chromatographic methods specifically
address the analysis  of EPA designated priority  pol-
lutants®, a "template" of organic toxicants comprised of
pesticides, polychlorinated biphenyls  (PCBs), base neu-
trals, phenolics and purgeable compounds. Organic pol-
lutants analyzed by fluorescence and infrared spectroscopy
include aromatic  compounds  and petroleum oils.  An
earlier publication describes specific aspects of the analyti-
cal and instrumental approaches, citing examples of their
application  to  emergency response analytical  assistance
situations.0' The methods  exhibit unique capabilities in
providing rapid and accurate information regarding oil
and  hazardous  material concentrations in a variety of en-
vironmental media.
  The majority of the specialized methods incorporate the
use of single stage extraction techniques which avoid tedi-
ous  and  time consuming solvent and soxhlet extraction,
evaporation and concentration steps. These innovative
techniques  are  applicable  to  the  analysis  of  virtually all
priority pollutant materials.  Rapid spinning techniques
are employed to extract priority pollutants from aqueous
media/4' Extraction is accomplished by agitating the mix-
ture (i.e. organic solvent(s) and water sample) using a
magnetic stir bar and stirring mechanism. Soil and sedi-
ment samples are extracted in a single step procedure util-
izing a gyratory  shaker to ensure  complete  contact  be-
tween solvent and sediment. Gas chromatographic analysis
of the extracts permits rapid quantification of the priority
pollutants.
  The development of rapid analytical methods based
upon the synchronous excitation fluorescence technique
considerably reduces  sample preparation and analysis
time. Direct in  situ quantification of hazardous materials
in aqueous media enables detection of fluorescing organics
at the ug/1 level.(5) Rapid quantitative  analysis of petrol-
eum oil in water is achieved through addition of an equiva-
lent amount of co-solvent to a portion of aqueous sample.
The  synchronous excitation fluorescence technique is also
applicable to the rapid measurement of hazardous mater-
ials(6) and petroleum oils(7) extracted from wet sediments.
The  extraction  procedure incorporates the use of a gyro-
tory shaker which affords a five to ten-fold reduction in
analysis time  over conventional  methods  using  soxhlet
and alcoholic-KOH extraction techniques.
  Infrared spectroscopy  is almost  exclusively employed
for the identification and  quantitative measurement of
petroleum oils extracted  from various matrices.(8'9'10)  An
unusually time efficient method for analyzing petroleum
oils in sediments  entails extracting the samples with car-
bon tetrachloride via gyratory agitation. The extracts  are
then diluted for infrared spectroscopic measurement which
provides accurate  quantification of petroleum oil in  the
sediment.
  The availability of this unique technology greatly  fa-
cilitates emergency response analytical laboratory support
by providing proven applicable methodology, or furnish-
ing the groundwork for the rapid development of an  ap-
propriate  method required for a specific response  ac-
tion.  The response capabilities of the laboratory have
evolved over the last decade according to the operational
needs of both hazardous materials spills and uncontrolled
hazardous  waste  dumpsite  management.  The  central
laboratory is presently compiling a manual of analytical
methods for emergency response to  discharges of hazard-
ous chemicals and petroleum oils. This manual is intended
for use in support of decisions associated with responses
to environmental emergencies.
Operational Protocols
  In addition to coordination the development, evalua-
tion and refinement of analytical methods and techniques,
the laboratory manages the implementation of specific op-
erational protocols for assuring emergency response readi-
ness. The central laboratory established a column quality
control system for maintaining and documenting the use of
gas chromatographic columns. Columns  entered into the
system are performance evaluated in accordance with spec-
ified efficiency and resolution criteria. Certification files
are established for the individual columns which are as-
signed inventory control numbers  for management pur-
poses. Extensive records are maintained regarding column
use and performance, to assure the availability of an ade-
quate  supply of validated  columns  in  anticipation  of
emergency response activations. Effective management of
such a laboratory system is of critical importance for pro-
viding accurate and timely analytical services during en-
vironmental emergencies.
  The laboratory also evaluated a large number of column
packing materials for broad scope application to the gas
chromatographic analysis  of priority pollutants. The ob-
jective of this research was to establish a  "universal"
column packing material suitable for the analysis of nearly
all  priority pollutants. Advantages of employing a single
column for general use include simplification of data ac-
quisition and column inventory, and reduction in time and
cost.  Comprehensive evaluation of commercially  avail-
able column packing materials resulted in the selection of
a mixed phase column consisting  of 1.5%  SP-2250 and
1.95% SP-2401 coated on 100/120 mesh  Supelcoport, for
general use.  This column yields excellent results for the

-------
98     SCREENING
analysis of pesticides, PCBs, base neutrals and a limited
quantity of phenolic and purgeable compounds.(4)


Quality Assurance/Quality Control
  An important function  of the central laboratory is as-
suring interlaboratory compliance with applicable Qual-
ity  Assurance/Quality Control  (QA/QC)  requirements.
Quality assurance defines  the limits of measurements and
monitoring data in terms of sensitivity, reproducibility,
detection  limits and accuracy.  QA/QC  protocols  are
necessary to demonstrate the reliability of data generated
by  the individual laboratory entities.  The central labora-
tory implemented  a  comprehensive QA/QC  program to
ensure the acquisition of high quality data  in method de-
velopment and sample analysis activities.
  The QA/QC program incorporates a novel technique
for defining a suitable quantitative measurement or linear
dynamic range for single point analysis."" Single point
analysis affords significant  capabilities in  terms  of  pro-
cessing  large numbers of  samples rapidly,  and providing
analytical results having known accuracy and precision.
The procedure of defining the calibration range for single
point analysis employs  numerous computations  that are
amenable to computer  processing. Computer programs
developed by the laboratory include calibration range de-
terminations  and quantitative analyses for both spectro-
scopic  and chromatographic  instruments.(I2) Increased
utilization  of  automated instrumentation and  micro-
processors facilities  management  of data  acquisition,
manipulation and processing in sample report format.
  The laboratory employs a method validation procedure
in which replicate synthetic samples are prepared  in a ma-
trix and at a concentration resembling that of the actual
samples. The  synthetic samples  are used to evaluate the
precision and  accuracy of the primary analysis methods.
Method evaluation is essential to confirm the veracity of
analytical results,  especially  when  procedures and  data
are used in an interlaboratory mode. The method valida-
tion techniques are employed extensively during emergency
response activations  of  the  mobile laboratory to ensure
data integrity.
MOBILE LABORATORIES

  As part of a continuing effort to provide prompt ana-
lytical  support  to  environmental  incidents  requiring
emergency response, the OHMS Branch developed a mo-
bile laboratory and an ancillary mobile unit for perform-
ing requisite analytical services  in remote field locations.
The primary function of  the mobile laboratory03' is to
provide rapid on-site  chemical analyses  in support  of
multi-media cleanup activities.  The EPA mobile labora-
tory is equipped with multi-parametric analytical instru-
mentation, including:

•a computerized gas chromatograph/mass spectrometer
•two computerized gas chromatographs furnished with
 automatic sample injectors and flame ionization, electron
 capture and nitrogen-phosphorous detectors
•a computerized atomic absorption spectrometer with a
 graphite furnace accessory and automatic sampling sys-
 tem
•infrared and fluorescence spectrophotometers
•an emission spectrometer
•a total organic carbon analyzer. The broad range of
  The broad range of analytical capabilities  aboard the
mobile  laboratory permits accurate chemical  analysis of
virtually all organic and inorganic  substances potentially
encountered at uncontrolled hazardous waste sites.

Emergency Response
  The mobile laboratory is readily dispatched for trans-
port to any part of the United States within a few hours of
an official request for emergency analytical services. The
laboratory is normally stationed within the proximity of
the cleanup  operation  to enable  rapid  on-site  sample
analyses. Junior level personnel (i.e. highly trained tech-
nicians  and cooperative college students) perform the
"hands-on"  analyses  aboard  remotely located  mobile
laboratories, while senior level chemists provide guidance
and direction from the central laboratory.
  Experience attained over several years  of field opera-
tions  demonstrates that, as a  general  rule, junior level
personnel enthusiastically accept travel and long term as-
signment to mobile laboratory activations. This operation-
al management scheme avoids  the  obvious morale prob-
lems associated with stationing senior level personnel for
prolonged periods of duty in remote field locations.
  The staffing of mobile  laboratories  with junior level
personnel prompted the development of a unique manage-
ment  approach  for coordinating  the  on-site analytical
laboratory support. Senior level personnel situated at the
chemical laboratory manage and direct  the sample analy-
ses performed on mobile laboratories, using a multi-facet-
ed communications system. Specific sample analysis pro-
cedures and analytical methodology developed at the cen-
tral  laboratory  are relayed to  the mobile  laboratory
through use of high speed telefacsimile units. Mobile lab-
oratory  transmission of  "raw" data, such  as  sample
chromatograms  and spectra,  supplies  the  necessary in-
formation  for senior level chemists to accurately inter-
pret analytical  results  and  implement corrective pro-
cedures.
  The central laboratory also evaluates the methodology
employed and generates validation data concurrently with
sample  analyses  performed  aboard the  mobile  labora-
tories. The concept of senior level personnel guiding aM
directing sample analyses  on mobile  laboratories via a
rapid communications system  is a unique and  effective
management technique  for integrating  analytical ser-
vices during environmental response efforts.
Ancillary Field Response Units
  The experience derived through  numerous  activation
of EPA's mobile laboratory served as a precursor to the
novel concept of developing and managing a  network of
mobile  laboratories. The concept involves equipping mo-
bile units with specialized instrumentation for  response W
specific analytical assistance requests. The concept is effl-

-------
                                                                                             SCREENING    99
ployed when analytical needs are specified  in  advance,
such as requests for establishing the  presence of PCBs in
abandoned chemical storage tanks and  drums. The ap-
propriate  instruments  and associated support materials
are rapidly installed  on a mobile unit and dispatched to
the site. Communication lines are established for trans-
mitting  data to  and  receiving  operational  instructions
from, the central laboratory. Development of  a broad-
based management system for coordinagint a network of
mobile laboratories  provides the OHMS Branch  with
enhanced emergency response analytical capabilities.
  An extention of this approach involves the concept of
crating and forwarding specific instruments to particular
environmental cleanup operations. Use of this concept is
restricted  to  incidents in  which  analytical  assistance
needs are  highly specialized.  The procedure entails  pack-
aging the  appropriate  analytical instrument in a custom
designed crate for shipment. A technician is dispatched
from the central  laboratory to unpack, install and oper-
ate the instrument at facilities established  on-site.  Com-
munication is maintained with the central laboratory via
telefacsimile to permit  close coordination of sample anal-
ysis activities. Crating of portable field devices such as
shoulder borne gas chromatagraphs and  redox test kits is
often employed to provide additional  analytical support.
The crating concept  is a unique approach for managing
analytical  services, especially during concurrent emergency
response analytical activities.
 Specialized Analytical Methodology
   The inherent space and time restrictions associated with
 mobile laboratory operations in the field necessitate the
 use of exceptionally rapid  and direct methods of sample
 preparation and analysis.(14) Extensive application of the
 specialized analytical methodology developed by the cen-
 tral laboratory enables rapid processing of numerous sam-
 ples during field activations. The established methodology
 is continually refined, and additional procedures and pro-
 tocols are developed to streamline sample analyses.
   Specialized  sample  preparation techniques  are under
 continual development, in order to maximize the avail-
 able  space  and minimize  analysis time  aboard mobile
 laboratories. A considerable expenditure of analysis time
 is required  for performing sequential sample and stan-
 dard  dilutions with conventional volumetric  pipets. Such
 sequential dilutions are time consuming,  tedious, waste-
 ful of expensive solvents and prone to error magnification
 with each successive step.
   The alternative  of using positive  displacement micro-
 pipets, which afford  a  "one-step"  dilution capability,
 was evaluated  in  accordance with established  precision
 and accuracy criteria.  The  high degree of reproducibility
 and reliability associated with their use,  resulted in the
 incorporation  of positive displacement micropipets into
 the methodology.05' Micropipets are used extensively in
 clinical laboratories, where  quantities of body fluids avail-
 able for analytical tests are at a premium. Application of
 micropipets to environmental analyses reduces problems
 associated with waste  solvent generation, minimizes per-
sonnel exposure to toxic chemicals and conserves costly
Primary-Standards and chemical reagents.
  The instrumental analysis techniques employed by the
mobile laboratories  are augmented through the  use  of
computerized  gas chromatography/mass  spectrometry.
This analytical technique enables separation of complex
mixtures into their individual constituents for compound
identification. Interpretation of mass spectra is accom-
plished by internal library  searching  and matching for
priority pollutant identification. Files of spectra for num-
erous compounds are maintained on cassette tapes to facil-
itate  retrieval and  storage. Functional GC/MS  opera-
tion is accomplished by junior level personnel in the field,
with minimum supervision  from the central laboratory.
Computerized interpretation of analytical results is a time
efficient  managerial  approach for acquiring high quality
data during emergency response activities.

SERVICE LABORATORIES

  Use of commercial service laboratories is an integral
part of the interlaboratory analytical emergency response
management  system.  Commercial  service  laboratories
provide emergency response support vis-a-vis sample anal-
yses,  quality assurance services,  custodial responsibility
for hazardous chemicals and reagents and development of
field  kits and test devices.  Primary  considerations in-
volved in selecting the appropriate service laboratory for a
particular  analysis  request include:  (1)  performance
history, (2) analytical methods used, (3) analysis cost and
time, (4) available instrumentation,  (5) internal quality
control system,  (6) staff and management qualifications
and, (7) results of test sample audits.
Sample Analyses
  The management system for  service laboratories tracks
samples from time of receipt until completion of analysis.
The initial stage of sample documentation involves scan-
ning the sample(s) for the  presence of nuclear radiation.
The next stage  entails completion  of an analytical ser-
vices  request form to execute the following tasks: (1) as-
sign sample  log  numbers, (2)  describe  and photograph
the samples, (3) establish a contract with a service labora-
tory,  (4) determine methods of analysis, (5) evaluate safety
procedures, (6)  maintain  chain of  custody records  and,
(7)  prepare and forward quality assurance samples. This
form provides the means for coordinating emergency re-
sponse sample analyses performed by service laboratories.

Quality Assurance Services
  An essential feature of the service laboratory manage-
ment system is its quality  assurance (QA) protocol. QA
services are performed by the central  laboratory or by a
second commercial service laboratory. The protocol neces-
sitates that split samples be retained and analyzed in the
central laboratory,  while  replicate  synthetic samples are
forwarded to the service laboratory with the actual  sam-
ples. The QA protocol consists of three states:
  (1)  a pre-analysis  meeting with service laboratory per-
      sonnel to evaluate the analysis methods and the in-
      ternal quality control (QC) program

-------
100     SCREENING
  (2)  an on-site visit during the course of sample analyses
      to monitor incoming data
  (3)  a comprehensive review of the final sample analysis
      report.
The QA protocol is specifically designed to generate vali-
dation data concurrently with sample analysis activities,
to provide analytical results  of known  quality in "real
time."
Custodial Services
  Commercial service laboratories are also employed to
stock and  maintain custodial responsibility for  specific
hazardous chemicals and reagents required by the  central
and mobile laboratories. This essential service assures the
availability of an adequate supply of certified reference
standards  and chemical  reagents during  emergency re-
sponse activation. The service laboratory documents the
purity and shelf-life of reference standards,  reorders ex-
pended materials and prepares stock solutions of chemical
reagents upon request.
Field Kits and Test Devices
  Contracts are occasionally arranged with service labora-
tories for  the rapid development  of emergency response
field kits and test devices. These  requests are associated
with specific emergency response episodes and  their en-
suing analytical laboratory support requirements.  Such a
contractual  arrangement was recently used for  the de-
velopment  of a portable Oxidation-Reduction (REDOX)
Potential measurement kit to  classify unknown chemicals
at an uncontrolled  hazardous waste site.  This relatively
low cost project was completed  on an emergency type
basis  within a two-week period using the  service labora-
tory analytical support mechanism.


INTERLABORATORY MANAGEMENT

  The unique interlaboratory management approach de-
veloped by the OHMS  Branch  affords broad  scope
emergency  response analysis capabilities.  The  central
laboratory serves as the coordination center of the analysis
effort and functions as a liaison between mobile and ser-
vice laboratories when  the  latter  support  is  required
(Figure 1). Integration of the analytical services provided
by the individual entities is accomplished through  the use
of specific  laboratory management protocols. These pro-
tocols are essential for assuring prompt and efficient re-
sponse to analytical assistance requests.
Sample Analysis Protocol

  The central laboratory employs a  specialized procedure
for coordinating sample analysis activities and preparing
final reports. Processing of large numbers of samples  is
expedited  through  completion  of a  Sample  Analysis
Form (Figure 2). This documentation  form provides spe-
cific guidelines for directing  and  performing sample an-
alyses. The central laboratory is responsible  for complet-
ing the form and managing  the  overall analysis effort.
The  Sample Analysis  Form incorporates  all  essential
sample documentation and analysis  data and is therefore
an effective management device for coordinating into-
laboratory services during emergency response activities.
  The Sample Analysis Form is included in the final re-
port,  which contains a comprehensive summary of the
analytical results. The report consists of six sections, each
of which addresses  specific aspects of  the analysis ac-
tivity:
•pertinent sample documentation
•chain of custody records
•the sample analysis results
•methodology used
•QA/QC and method validation
•analysis  data,   such  as  sample  chromatograms and
 spectra
  Following final review and approval, the sample analy-
sis results are forwarded to  the Requestor and the analy-
sis report is archived for future reference and use.
QA/QC Protocols
  A complementary series of QA/QC protocols are used
by the OHMSB laboratory to monitor the analytical data
generated by its individual entities, and to assure the in-
tegrity of reported sample analysis results. Management of
the interlaboratory QA/QC program is the responsibility
of the central laboratory,  which coordinates  emergency
response sample  analyses.  All sample documentation is
performed by the central laboratory prior to forwarding
the samples to the service laboratory for an analysis.
  A representative portion  of sample  is separated from
the main sample(s) and retained at  the central laboratory
for QC  purposes. Approximately  25%  of the retained
samples are analyzed at the central laboratory or a second
service laboratory.  Replicate  synthetic samples are pre-
pared in  a matrix and  at a concentration resembling that
of the actual samples  to evaluate the accuracy and pre-
cision  of the  analytical methodology employed by the
service laboratory.
  An abbreviated version of the laboratory QA/QC pro-
gram  is  used during emergency  response activations of
the mobile laboratories. This protocol provides the on-
scene coordinator with sample analysis data of known
quality for establishing specific QC guidelines. All sam-
ple analyses are performed in  such a manner as to per-
mit rapid dissemination of analytical results. The central
laboratory coordinates the QA aspects of the protocol to
ensure that  reported results are of known accuracy and
precision.
  The simultaneous reporting of sample analysis and QA
validation data  furnishes the  Requestor  with analytical
results in "real time." The Requestor is then in a position
to determine the need and practicality  of repeated or
supplementary analyses. This type of management system
permits immediate release of sample analysis results to
the on-scene coordinator, who is responsible for imple-
menting  appropriate remedial techniques during environ-
mental cleanup operations.

CONCLUSIONS
  A comprehensive management  system for providing
analytical laboratory support  to environmental incident*

-------
                                                                                           SCREENING
                                                            101
                                      SAMPLE  ANALYSIS   FORM

                                   MUNICIPAL  ENVIRONMENTAL  RESEARCH  LABORATORY
                             OIL AND HAZARDOUS  MATERIALS  SPILLS  BRANCH -  EDISON, NJ
                                                  CHEMISTRY  STAFF
1. REQUE
OHMSB
LOG I
SAMPL
STED

E

BY:



# OF SAMPLES:
DATE OF
ORGANIZATION: REQUEST:
Mo
bay
Vr
SAMPLE ORIGIN
(LOCATION):
BACKGROUND:



           CHECK ACTIVITY: CENTRAL LAB
MOBILE LAB
                                                                     SERVICE LAB
       2.  DESCRIPTION OF SAMPLES  (TYPE  OF  SAMPLES,  WEIGHT  OR  VOLUME,  TYPE OF CONTAINER):
           	/ATTACH PH
       3.  ANALYSES REQUESTED:
       4.  METHOD(S) USED  (GIVE TITLES):
       5.  SAFETY INFORMATION:
                                  IMPORTANT:   DO  NOT  PROCEED  WITH  ANALYSES UNTIL THE FOLLOWING
                                  INFORMATION  HAS BEEN  OBTAINED.
           IDENTIFICATION OF:
              A.  HEALTH HAZARD:
              B.  FLAMMABILITY: "
              C.  REACTIVITY:  ~
              D.  RAOIOACTIVITYT
                                                                                /HANDLING INSTRUCTIONS:
       6.  QA/QC:                                                YES   NO
           A.  SERVICE LABORATORY  QUALITY  ASSURANCE  PERFORMED:   II   FH   DESCRIBE:,
           B.  OHMSB (IN-HOUSE/MOBILE  LABORATORY)  QUALITY  CONTROL  PERFORMED:
               -CALIBRATION RANGES  ESTABLISHED  /  EC  VALUES  OBTAINED:	
               -COEFFICIENT OF VARIATION  (Cv) OF  MIDPOINT  SOLUTIONS  OBTAINED:
               -ACCURACY  (RECOVERY  STUDY)  Re  VALUES  OBTAINED:	"
               -PRECISION  (REPLICATE ANALYSES)  Cv  VALUES  OBTAT
               -CONFIRMATORY  ANALYSES  OF  SPLIT  REPRESENTATIVE  SAMPLE  PORTIONS:
               -CALCULATION OF  RESULTS CHECKED:	"
       7.  CHAIN OF CUSTODY  RECORD:
RELINQUISHED BY (SIGNATURE)
RELINQUISHED BY (SIGNATURE)
RELINQUISHED BY (SIGNATURE)
DATE / TIME
DATE / TIME
DATE / TIME
RECEIVED BY (.SIGNATURE)
RECEIVED BY (SIGNATURE)
RECEIVED BY (SIGNATURE)
REASON FOR CHANGE OF CUSTODY
REASON FOR CHANGE Oh CUSTODY
REASON FOR CHANGE OF CUSTODY
       8.  SUMMARY OF RESULTS  (ATTACHMENTS  AS  REQUIRED):,
       9.  REMARKS:
                                                    Figure 2.
           An abbreviated version of the Sample Analysis Form employed by the OHMSB Laboratory for documentation
                                        of emergency response sample analyses.
requiring emergency response was developed through ef-
fective coordination  of a series of  interlaboratory ser-
vices. The system incorporates specific operational proto-
cols for coordinating and integrating the sample analyses
performed by central, mobile  and service laboratory fa-
         cilities. Application of the appropriate analytical method-
         ology and operational protocols lends invaluable support
         to management decisions involving  the  procedures  em-
         ployed for environmental cleanup at uncontrolled haz-
         ardous waste sites.

-------
102    SCREENING
REFERENCES

 1.  "Controlling  Hazardous  Wastes,"  U.S.  Environ-
    mental Protection  Agency, Report  No.  EPA-600/
    8-80-017, 1980.
 2.  Federal  Register,  1979,  Vol.  44,  No.  233,  69464,
    December3.
 3.  Gruenfeld,  M., Frank, U.  and Remeta, D., "Rapid
    Methods  of Chemical Analysis  Used in Emergency
    Response Mobile  Laboratory Activities,"  Proc.  of
    U.S. EPA National Conference on Management of
    Uncontrolled  Hazardous  Waste Sites, Washington,
    D.C., October  15-17,  1980.  Hazardous  Materials
    Control  Research Institute, Silver Spring, Md., 165-
    172.
 4.  Gruenfeld, M., Losche, R. and Frederick, R. "Rapid
    Methods  of Analysis for  Emergency Response  to
    Chemical Spills and Waste  Chemical Dumpsites,"
    EPA Quality Assurance Newsletter, 4, January 1981.
 5.  Frank, U. and Gruenfeld, M., "Use of Synchronous
    Excitation  Fluorescence  Spectroscopy  for  in-Situ
    Quantification of Hazardous  Materials in Water,"
    Proc. of the 1978 National Conference on Control of
    Hazardous  Material Spills, Miami Beach, FL, April
    11-13, 1978. Information Transfer,  Rockville, Md.
    119-123.
 6.  Frank, U. and Remeta, D., "Rapid Quantification of
    Hazardous  Materials in  Sediments  by Synchronous
    Excitation Fluorescence Spectroscopy," EPA Quality
    Assurance Newsletter, 1, April 1978.
 7.  Frank, U.,  "Rapid Quantification of Petroleum Oils
    in Sediments," EPA Quality Assurance Newsletter, 1,
    July 1978.
 8.  Gruenfeld,  M., "Quantitative Analysis of Petroleum
    Oil   Pollutants  by  Infrared  Spectrophotometry,"
    Water Quality Parameters, American Society for Test-
    ing and Materials, ASTM STP 573, 1975, 290-308.
 9.  Gruenfeld,  M. and Frederick, R., "The Ultrasonic
    Dispersion, Source  Identification,  and Quantitative
    Analysis  of Petroleum  Oils  in  Water,"  Rapports
    Precedes Verbuux  Reunion  Conseil  Internationale
    Exploration De La Mer, No. 171,1977,33-38.

10.  Gruenfeld,  M., "Extraction of Dispersed Oils from
    Water for Quantitative Analysis by Intrared Spectro-
    photometry," Environmental Science and Technolo-
    gy, 7, 1973, 636-639.

11.  Gruenfeld,  M and Remeta,  D.,  "Selection of a
    Measurement Range for  Quantitative Analyses Using
    Single Point Instrument Calibration," EPA Quality
    Assurance Newsletter, 3, April 1980.

12.  Gruenfeld,  M. and DeMaine, B., "Availability of
    Computer Programs," EPA Quality Assurance New--
    letter, 4, January 1981.

13.  Urban, M. and Losche, R., "Development and Use of
    a Mobile Chemical  Laboratory for Hazardous Ma-
    terial Spill  Response Activities," Proc. of the 1978
    National Conference on Control of Hazardous Ma-
    terial Spills, Miami Beach, FL,  April  11-13, 1978,
    Information Transfer, Rockville, Md. 311-314.

14.  Frank, U. Gruenfeld, M., Losche, R. and Lafornara,
    J., "Mobile Laboratory Safety and Analysis Proto-
    cols Used at Abandoned  Chemical Waste Dump Sites
    and Oil and Hazardous Chemical  Spills,' Proc.  of
    the  1980 National Conference on  Control of Hat"
    ardous  Material  Spills, Louisville,  KY, May 13-li,
    1980, Vanderbilt University, Nashville, Tenn., 259-
    263.

15.  Gruenfeld, M. and Remeta, D., "Limited Evaluation
    of the Accuracy  and Precision of Positive Displace-
    ment Micropipets," EPA  Quality Assurance Neva-
    letter, 4, January 1981.

-------
       USE OF A REGULATED ACCESS LABORATORY FOR
                    SCREENING AND PREPARATION OF
                    HAZARDOUS WASTE SITE SAMPLES

                                     BARRY E. NORTH, Ph.D.
                                    KATHLEEN H. DRISCOLL
                                    Fred C. Hart Associates, Inc.
                                          Denver, Colorado
                                LAURENCE W. STRATTAN, Ph.D.
                               U.S. Environmental Protection Agency
                             National Enforcement Investigations Center
                                          Denver, Colorado
INTRODUCTION

  The investigation and cleanup of uncontrolled hazardous
waste sites create a need for special waste sample screen-
ing and analysis capabilities, presenting some unique chal-
lenges to the analytical laboratory. The potentially highly
hazardous nature of"these samples and the typical lack of
available information about any hazardous properties re-
quire that extreme caution be taken in the analytical pro-
cess during sample preparation.
  However, concentrated hazardous waste samples can be
handled  safely  with the proper equipment and  precau-
tions. A regulated access laboratory at the U.S. Environ-
mental Protection Agency's (EPA) National Enforcement
Investigations Center (NEIC) is being used for analytical
support of the agency's hazardous waste program. Since
April 1980, the Regulated  Access Laboratory at NEIC
has been operated by Fred C. Hart Associates as part of
the EPA's Field Investigations  of Uncontrolled Hazar-
dous Waste Sites contract.

PURPOSE OF LABORATORY

  The Regulated Access Laboratory is designed to handle
highly toxic and carcinogenic materials. The facility pro-
vides a safe workplace for the preparation of hazardous
waste  samples  collected  from  uncontrolled  hazardous
waste sites. The laboratory has been equipped to permit
samples to be examined, screened for various parameters
which indicate hazard level and extracted and diluted as
required so that the prepared extracts  can be analyzed
safely in an environmental laboratory.

LABORATORY PROCEDURES

  The Regulated Access Laboratory receives  many types
of samples, including solid wastes, sludges, and  aqueous
and nonaqueous liquids. These samples come from con-
tainers such as  drums and tanks, as  well as from land-
fill leachate, pits, ponds, lagoons, piles of solid chemi-
cal and contaminated soils. Because of the varied and po-
tentially highly  hazardous nature of these samples, the
laboratory's preparation procedures have been designed to
reduce excessive handling of the concentrated sample, to
accommodate a wide range of sample types and to pro-
duce extracts and other preparations analogous to environ-
mental soil, sediment and water sample extracts.
  The sample handling procedures include sample receipt
and documentation,  initial aliquot preparation,  sample
screening and extract preparation. All procedures incor-
porate quality control.

Sample Receipt and Documentation
  Packaging and shipping procedures for transporting
hazardous waste samples to a laboratory in accordance
with Department of Transportation Regulations (49 CFR
173.2) have been  described  previously.(1) Samples  are
packed in glass bottles sealed in plastic bags and in turn,
sealed inside metal cans filled with  an absorbent, non-
combustible filler material.
  The metal can is brought into the laboratory unopened.
The sample bottle is opened only in a laboratory type hood
designed for this purpose, taking precautions to avoid con-
tamination of the  surrounding area. At this  time, all
documentation, such as sample tags and chain-of-custody
records are verified.
Initial Aliquot Preparation
  All required sample aliquots of the sample for the vari-
ous screening tests and preparations are weighed out into
appropriate  containers, thus eliminating the handling of
large amounts of the sample after this  step. Most pro-
cedures require no more than one or two grams, but a 20g
aliquot is required  for preparations used for analysis of
extractable organics.
Sampling Screening
  All samples are screened by spot tests for pH, cyanide,
sulfide and strong oxidants. Conductivity, alkalinity  and
acidity are determined when appropriate. The spot test re-
sults provide hazard information as well as indicating need
for quantitative testing. Screening for organic constituents
is accomplished by a gas chromatographic scan of organic
extracts using a flame ionization detector. By using an in-
ternal standard, the approximate concentrations of com-
ponents  can be  estimated. Although  samplers  are in-
structed  not to send radioactive samples, all samples are
screened for radioactivity prior to other testing.
                                                  103

-------
104     SCREENING
 Extract Preparation
   Samples  are  routinely prepared for analysis  of total
 metals  and acid extractable  metals, strong  acid  anions,
 volatile  organics  and  semivolatile  (acidic  and base/
 neutral) organics. Typical detection limits are 1 to  10 mg/1
 when the extracts are analyzed by conventional trace level
 analytical methods. The preparation methods and recom-
 mended analysis methods for these parameters are given
 in Table I.

                       TABLE I.
       Preparation of Extracts for Inorganic and Organic
            Analysis of Hazardous Waste Samples
 Analytical Parameter   Method of Preparation
 Total Metals
 Acid Extractable
  Metals
                 Dilution in 2% HNO
                 Metal Fusion (s,n)

                 Extraction with 0. IN
                  HCl(s)
 Strong Acid Anions    Dilution (a, s)
Total Mercury


Volatile Organics

Semi Volatile Organics
                                           Method of Recom-
                                            mended Analysis
ICP-AES
ICP-AES

ICP-AES


Ion Chromatography


CV-AAS


GC/MS (Method 6242)

GC/MS (Method 6253,60S4)
                 Persulfate/Perman-
                  ganate Digestion (s,n)

                 Dilution in Water

                 Shakeout (a)


                 Water and Solvent
                   Extraction (s)
                 Dilution (n)
Key: a = aqueous samples; n = non-aqueous samples; s = solid samples;
ICP-AES = inductively coupled plasma atomic emission spectroscopy;
CV-AAS = cold vapor atomic absorption spectroscopy
GC/MS = gas chromatography/mass spectrometry
Development of New Methods for
Screening Hazardous Waste Samples
  Presently,  newly  developed procedures for screening
hazardous waste samples are being tested in the Regulated
Laboratory. The new approach permits a detailed organic
screening of the sample to determine whether a further
analysis is warranted. The organic extracts and sample ali-
quots prepared  by this method  permit  screening by gas
chromatography with flame ionization and Hall electroly-
tic conductivity detectors, using both packed and  capil-
lary columns for pesticides, PCB's, base/neutral and acid
priority pollutants, volatile organic priority pollutants and
related organic chemicals. Similarly, the inorganic sample
preparations permit screening  by x-ray fluourescence for
priority pollutant  metals.  If the screenings indicate  no
detectable pollutants at the lower limits of detection, the
need for additional and more expensive analysis can be
evaluated.<3)
  The new method has several advantages over the previ-
ous method. First, it requires a smaller  sample size; only
2g are needed for the organic extract preparations. Tenta-
tive  identification and quantitation of  organic  and  in-
organic (metal)  priority pollutants can be obtained and,
perhaps most importantly, identification of samples which
contain contaminants below a specified level can eliminate
the need for further expensive analysis. If further analysis
is desired,  the  sample can then be analyzed using pro.
cedures designed for environmental levels of pollutants.

LABORATORY DESIGN
Laboratory Layout
  A diagram of the laboratory, showing its major fea-
tures and  the location  of major equipment is  given in
Figure 1. The laboratory work area is divided into two
rooms (1) a "dry lab" which contains no running water or
drains, and (2) a "wet lab" equipped with running water,
compressed air and vacuum.
  The "dry lab" is equipped with a glove box attached by
means of double pass-through doors to a four ft. labora-
tory type hood. The "dry lab"  also contains a laminar
flow hood  for biohazard work. The "dry lab" is used for
work involving large amounts of a sample such as open-
ing sample containers and weighing out samples.
  The actual sample preparation occurs in the "wet lab."
It is equipped with two, six ft. laboratory type hoods back
to back. These hoods are primarily used for screening and
preparation  work using  small  quantities or dilutions of
the samples. The laboratory is also equipped with a gas
chromatograph, flammable solvent storage  cabinets, a
centrifuge and explosion-proof refrigerators.
  The "wet lab" and "dry lab" are adjacent to a small
vestibule which serves as contamination reduction area.
An emergency phone is located in the vestibule. The vesti-
bule leads to the adjacent men's and women's shower and
change rooms, as well as to a door which provides direct
access to the laboratory and is  used to  pass equipment in
and out. These three access-ways are provided with elec-
trically interlocked double doors to maintain the labora-
tory's reduced  atmospheric pressure balance. An emer-
gency exit is located in the "wet lab."
                                                              HEW H1NDQU
                                                  Figure 1.
                              Floorplan of EPA/NEIC Regulated Access Laboratory

-------
                                                                                            SCREENING     105
Ventilation System

  Laboratory air is exhausted exclusively through the four
ft. hood and the two, six ft. hoods. The hood exhauts are
equipped with high  efficiency particulate air  filters and
activated carbon filters to remove airborne contaminants.
The incoming air is limited to provide a net negative air
pressure within the laboratory. The system is designed so
that the "dry  lab" is at slightly more  negative pressure
relative to the "wet lab," thus creating a flow of air from
the "wet lab" to the "dry lab", i.e. toward the area of
highest potential contamination.

Laboratory Status Indicators

  The laboratory is  equipped with  filter static pressure
gauges, laboratory differential negative pressure gauges
and hood airflow indicators as a means of checking the
proper functioning of the ventilation system.


Safety Considerations

  A number of factors, described here, contribute to the
safe operation of the Regulated Access Laboratory. These
factors are training of personnel in safe handling  of
hazardous  materials, laboratory design, standard operat-
ing procedures and personal protective equipment.

Training

  Laboratory  personnel  undergo extensive training  to
prepare them to function safely in the  Regulated Access
Laboratory and to assure that all laboratory operations
take place  in a safe manner. Personnel attend courses in
basic  first  aid and cardiopulmonary resuscitation as well
as laboratory safety and  safe handling of chemical car-
cinogens. Training in use of self-contained breathing ap-
paratus and other respiratory protective equipment is also
provided.

SAFETY FEATURES

Design

  The Regulated  Access Laboratory contains the follow-
ing safety and contamination control features:
•Restricted and controlled  access. Only authorized per-
  sonnel are allowed to enter the regulated area  of the
  laboratory. Access is controlled by an electric access con-
  trol  system. A  daily roster of personnel entering  the
  laboratory is maintained.
•Maintenance of reduced atmospheric pressure. The regu-
  lated area is maintained at a  lower atmospheric pres-
  sure  than the  surrounding non-regulated  area.  The
  laboratory entrances are equipped with electrically inter-
  locked double doors, which^allow access to the labora-
  tory while maintaining the Tower inside pressure. Main-
  tenance of the reduced internal pressure is checked daily.
•Ventilation system.  The  laboratory ventilation system
  provides for approximately 12 air changes per hour. All
  exhaust air,  which exits through  the laboratory  type
  hoods, is  treated by filtration through  high  efficiency
  particulate air (HEPA) filters and  activated carbon fil-
  ters to remove particulates and trace organics.
•Laboratory type hoods. Open-face hoods with an average
 linear face velocity of 100 feet per minute provide a work-
 ing space for concentrated hazardous waste samples. The
 face velocity is checked daily.
•Glove  box.  A glove box provides an isolation  area,
 primarily for preparation of highly toxic or carcinogenic
 standards. The glove box is kept under negative pressure
 of 0.5 inches water gauge with respect to the surrounding
 space.
•Shower and change area. Shower and change areas are
 used for thorough personal decontamination  prior to
 leaving the regulated area.
Standard Operation Procedures
  Use of standard operating procedures for all laboratory
operations assures that safety practices are followed. These
standard procedures  include procedures  for handling
samples, housekeeping, waste disposal, and spill contain-
ment  and cleanup. The standard operating procedures
for handling hazardous waste samples include the follow-
ing safety practices:
•Samples are only opened in laboratory type hoods
•Only small aliquots of samples (20 g or less) are handled
 at one time
•The initial step in sample preparation,  involving  exam-
 ing the entire sample  and weighing out small amounts for
 the various  procedures, is  always  performed by two
 people. One person (called the processor) carries out the
 steps required  to transfer and weigh the sample aliquots
 while the other assists  by recording  data, preparing
 labels and obtaining supplies and reagents for  the pro-
 cessor. By restricting the arms of the processor, prop-
 erly protected, to the inside of the four-foot hood during
 aliquot preparation, the chance of spreading contamina-
 tion is reduced.
•No one may work alone in the laboratory.

  Because the waste  samples  handled by the Regulated
Access Laboratory may be highly hazardous, the standard
operating procedure for spill containment and cleanup
require that cleanup personnel are  fully protected against
exposure to  unanalyzed  hazardous  waste samples.  If a
sample is spilled outside a hood or glove box, the labora-
tory must be evacuated, and the cleanup crew must re-
enter  wearing  self-contained breathing apparatus and
other  protective equipment. Emergency escape air  masks
stored inside the laboratory can be  used if required for
respiratory protection during evacuation of the laboratory.
  In general, prudent practices for handling hazardous
chemicals in the laboratory are followed.(6)
Personal Protective Equipment
  Protective clothing is worn at all times while hazardous
waste  samples  are being  processed.  Disposal  Tyvec©
hooded suits are worn over  launderable jumpsuits, and
Tyvec© booties are worn over washable sneakers. Goggles
or  safety glasses and nitrile gloves  are worn routinely.
Heavy gloves, made of various materials, are available and
are used during cleanup of spills. Two pairs of gloves are
always worn  during operations involving high probability
of direct contact with undiluted hazardous waste samples.

-------
106    SCREENING
  Personnel remove all clothing and other gear on leaving
the laboratory  before  washing or showering. Although
the standard operating procedures for  sample prepara-
tion reduce the  potential for contamination of laboratory
personnel,  the  use  of personal  protective  equipment
further eliminates  the  likelihood of skin or eye contact
with toxic and potentially  carcinogenic samples. Respira-
tory protective equipment is not used except for emergen-
cies such as spill containment.

PROBLEMS

   Problems have been encountered due to space limita-
tions  of the laboratory. The most limiting factor is the
availability of laboratory hood space,  since many opera-
tions  must be carried out in a hood. The hood space  limi-
tation requires that work be scheduled so that operations
requiring the same hood are performed at different times.
   Use of separate laboratories in different locations for
sample preparation and analysis is sometimes cumbersome
and creates problems with quality control. Since analytical
quality control results are not immediately available, prob-
lems  with preparation  procedures may not be discovered
rapidly.
   Handling potentially hazardous materials  and being
confined in the laboratory create stressful working condi-
tions. While the work of processing samples on a produc-
tion basis becomes very routine, a high degree of alert-
ness is required at all times. Since laboratory personnel do
not see an end result, i.e., a final analysis of the sample,
they receive no  immediate feedback from their work, and
therefore interest level and morale suffer. Involvement
with the analysis of sample extracts with institution of the
new procedures described above is expected  to improve
this situation. Additionally, rotation of laboratory person-
nel between the Regulated Access Laboratory and the
analytical laboratories  is expected to reduce negative ef-
fects of confinement.
ACCOMPLISHMENTS OF THE PROGRAM

  During approximately one year that the Regulated Ac-
cess  Laboratory has been in operation,  more than 550
hazardous waste  samples,  from almost  100  sites, have
been screened and prepared for analysis. Work on most
sampling projects is completed in  15 work days or less
using the routine procedure described. The laboratory has
handled a number of high priority requests which have
been completed in one week or less.
  The laboratory  has prepared samples for analysis of
2, 3, 7, 8-tetrachlorodibenzo-p-dioxin (TCDD), using the
extraction and cleanup procedures described in the Federal
Register.(7) This work required preparation of pure TCDD
standards. Another accomplishment of the laboratory is
the preparation of samples using the EP Toxicity proce-
dure required by the Resource Conservation and Recov-
ery Act Regulations.(8>
  Since its opening,  the laboratory has operated without
a serious accident  or significant exposure. Several minor
accidents involving cuts from broken glassware have oc-
curred, but procedural changes have been made to reduce
the chance of this type of injury.
  The analytical results from  hazardous  waste  samples
prepared in  the  Regulated Access Laboratory are being
used by both Headquarters EPA and Regional EPA of-
fices for the development of enforcement cases involving
uncontrolled  hazardous  waste sites.  The NEIC facility
has provided a central location where high hazard samples
can be handled  safely and prepared for  analysis using
standard and special request procedures.

REFERENCES

1. Blackman,  W.C.,  Benson,  B.E., Fischer,  K.E., "En-
   forcement and  Safety Procedures for  Evaluation of
   Hazardous Waste Disposal Sites", Proc. of U.S. EPA
   Conference on Management of Uncontrolled Hazard-
   ous Waste Sites,  Washington,  D.C., October 15-17,
   1980, Hazardous Materials Control Research Institute,
   Silver Spring,  Md., 91-106.
2. Federal Register,  44, No. 232, Monday, December 3,
   1979, 69532-695.
3. Federal Register,  44, No. 233, Monday, December 3,
   1979, 69540-69552.
4. Federal Register,  44, No. 233, Monday, December 3,
   1979, 69501-69509.
5. National Enforcement Investigations Center,  Method
   for Preparation  of  High  Concentration  Hazardous
   Waste  Samples:  Organic  Chemical  Extraction Pro-
   cedures, July, 1981.
6. National Research Council. Committee on Hazardous
   Substances in the Laboratory.  Prudent Practices for
   Handling Hazardous Chemicals in Laboratories. Na-
   tional Academy Press, Washington, D.C., 1980.
7. Federal Register,  44, No. 233, Monday, December 3,
   1979, 69526-69530.
8. Federal Register, 45, No. 98, Monday,  May 19, 1980,
   33084-33135.

-------
    THE USE OF LABORATORY SCREENING PROCEDURES
   IN THE CHEMICAL EVALUATION OF UNCONTROLLED
                            HAZARDOUS WASTE SITES

                                         ROBERT K. WYETH
                                          Recra Research Inc.
                                          Amherst, New York
INTRODUCTION

  The disposal of hazardous waste materials  has  been
and remains a major problem for the scientific community
as a whole.  Disposal practices  have and unfortunately
still include the deposition of hazardous waste materials
at geologically/hydrogeologically  unsuitable  locations.
Nationwide, it is believed that thousands of uncontrolled
hazardous  waste  landfills remain undiscovered  and/or
uninvestigated. These sites generally are devoid of disposal
records or the information concerning the types of depos-
ited wastes is incomplete.
  The most favorable and cost effective approach to in-
vestigating these uncontrolled waste sites  is through an
initial exploratory phase and a secondary confirmational
phase. The major factor in the  cost effectiveness of the
exploratory  phase of an investigation manifests itself in
the use of screening procedures.
TYPES OF SCREENING PROCEDURES

  Numerous types of screening procedures, whether or
not recognized or identified by that terminology, have
been used by analytical environmental scientists for many
years. These procedures include,  but are not limited to,
the determination of pH, conductivity, total organic car-
bon, methylene blue active substances and phenols. These
particular  tests  have been and  continue  to be viable
screening procedures in most landfill situations.
  Of greater concern today, however, are the  presence
and ultimate effects and/or impacts of organic materials
in landfills. The types of screening  procedures developed
and presented in this paper represent cost effective and
technically viable means for detection of  volatile or non-
volatile chromatographable organic constituents.
  These  screening procedures rely upon total chromato-
graphic response of samples or sample extracts compared
to the response of appropriate  standard materials. These
scans include halogenated organic scan via either electron
capture detection (BCD) and/or modified Coulson's elec-
trolytic conductivity detection (Coulson's), organic nitro-
gen/phosphorous scan,  volatile  halogenated/non-halo-
genated  scan and organic scan via flame ionization  de-
tection.
Halogenated Organic Scan
  The halogenated organic scan (BCD or Coulson's) is a
gas chromatographic analysis  with electron capture or
electrolytic conductivity detection designed to scan sample
extracts for the possible presence of many hazardous halo-
genated organic compounds. Responses are caused by the
presence of compounds such as the majority of the U.S.
E.P.A. designated  Priority  Pollutant  Pesticides, poly-
chlorinated biphenyls, as  well as many other halogenated
organics.'1' 2) Many non-halogenated organics  containing
the elements oxygen, nitrogen, phosphorus, or sulfur may
also cause a response. Various clean-up procedures, or the
use of the  chlorine specific detector, can be incorporated
into the analysis to remove many of these "interferences".
  Quantification is based upon a summation of chromato-
graphic response calculated as chlorine, using the response
factor and chlorine content of Lindane as a standard. The
selection of the ECD or Coulson's detector for this scan
is generally a function of sensitivity and  specificity. Of-
ten, the halogen specific Coulson's detector is used in the
confirmatory phase of the investigation as opposed to the
exploratory phase.
  This scan test, regardless of the detector selected,  is
based upon a solvent  extract chromatographed on  a
OU17/QF1 (or equivalent) pesticide column. The screen-
ing procedure is operationally defined as beginning at the
retention time of dichlorobenzene and concluding at the
retention time of methoxychlor.

Volatile Halogenated Organic Scan
  The volatile halogenated  organic scan  is a  chromato-
graphic  analysis,  with  electrolytic  conductivity  detec-
tion, operated in  the halogen specific mode, designed to
scan  samples for the possible presence of many  volatile
hazardous halogenated organic compounds.
  This screening procedure  can also be used in conjunc-
tion with a flame ionization detector, in  order to assure
the detection  of both halogenated  and non-halogenated
organics, most notably materials such as benzene, toluene
and xylenes. The process uses chromatographable organic
materials on a carbowax 1500/carbopack  C and/or a SP-
1000/carbopack B column. This scan test is operationally
defined as beginning at the retention time of methylene
chloride and concluding  at  the retention time of mono-
chlorobenzene.
                                                   107

-------
108    SCREENING
  The choice of lindane and carbon tetrachloride as the
standards for these procedures is  somewhat arbitrary.
These materials  do provide approximate mean response
factors under the condition of the test on ECD and flame
ionization detectors, however.
  The presence of compounds such as the majority U.S.
E.P.A.  designated  Priority Pollutant  Volatiles  are de-
tected. This detector is specific to halogenated organics,
thus increasing  the sensitivity.  Quantification is  based
upon a summation of chromatographic response calculat-
ed as chlorine, using the response factor and chlorine con-
tent of carbon tetrachloride as a standard.
Organic Nitrogen/Phosphorus Scan
  Organic nitrogen/phosphorus scans  are again  a gas
chromatographic analysis  with  detection via  a nitrogen-
phosphorus specific detector, designed to scan samples for
a variety of organonitrogen and organophosphorus  com-
pounds.  This analysis is generally custom designed to fit
the specific needs of the  particular site under investiga-
tion. Chromatographic conditions are selected to provide
a scan for the possible presence of organonitrogen and
organophosphorus  compounds  with  similar  chemical
structure  and polarity. This analysis presupposes know-
ledge of the probable identities of the types of compounds
which may be present. In  the absence of any information
on  the specific site being investigated,  a solvent extract
is chromatographed on a SP-2250 column and results are
presented based upon the response  factor and nitrogen
content of dimethylaniline.
Organic Screening Procedure

  The organic screening  procedure  (FID) is completed
with the  use of flame ionization detection and is designed
to scan samples for a variety of organic compounds. This
versatile  analysis is generally custom-designed to fit the
specific needs of  the site.  Chromatographic conditions
are selected to provide a scan for the possible presence of
compounds with similar chemical structure and polarity.
This  analysis presupposes  knowledge  of  the probably
identities of the types of compounds which may be present.
In the absence of any information regarding the site, a
solvent extract is chromatographed on a SP-2250 column
and the  results  are presented based upon the response
factor and carbon content of naphthalene.
  Quantification is based upon a summation of chromato-
graphic response calculated as carbon, using the response
factor and carbon content of a carefully selected specific
organic compound  as a standard. As was  the case with
nitrogen/phosphorus scans,  multiple Organic Scan (FID)
analyses  employing different chromatographic columns
and/or conditions may be  necessary in order to screen for
the variety of chemical structures and polarities which are
suspected to be present.
APPLICATION OF SCREENING PROCEDURES

  The initial or exploratory chemical evaluation of un-
controlled hazardous waste sites has been completed using
numerous  types  of screening  tests.  The  use of these
                         Figure 1.
           Halogenated Organic Screening Procedure
              Typical GLC/ECD Chromatogram
                         Figure 2.
             Volatile Organic Screening Procedure
              Typical GLC/FID Chromatograph

screening tests has provided a cost effective means of it-
fining those areas of a site (soil, sediment, surface and/or
groundwater)  and  the  types/classes of  chemical  com-
pounds which require more in-depth examination prior to
or during  a secondary confirmational phase or before
formulation and implementation of remedial action plans
for a site.
  Landfill  investigations performed by Recra Research,
Inc.  have confirmed the usefulness  of these  procedurei.
A typical Halogenated Organic  Scan chromatograph ii
shown in Figure  1. This scan procedure has  detected
numerous halogenated  organic materials, including but
not limited to the following:
•chorobenzenes (di-, tri-, tetra-, penta-, and hexa-)
•chlorotoluenes
•hexachloro -1,3 - butadiene
•hexachloropentadiene
•lindane (and other BHC isomers)
•pesticides (including DOT's, aldrin, etc.)
•polychlorinated biphenyls
•polybrominated biphenyls
•phthalate esters.
  The volatile halogenated and non-halogenated organic
screening  procedure has been proven via GC/MS con-

-------
                                                                                           SCREENING
                                                  109
firmation to detect the following types  of organic ma-
terials:
•trichloroethylene
•methylene chloride
•1,1,1- trichloroethane
•tetrachloroethylene
•dichloroethylenes
•chlorabenzene
•benzene
•toluene
•xylenes
•styrene
An example of a chromatograph resulting from this pro-
cedure is included as Figure 2.
  The organic  screening procedure  and  the  nitrogen-
phosphorus screening procedures have also been used to
detect numerous organic  constituents  including but not
limited to the following:
•polynuclear aromatics
•chlorobenzenes
•chloronaphthalenes
•aniline (including substituted anilines)
•nitrosoanines
   Based upon the broad scope extractions014) used and the
columns employed, the detection of these types of organics
is not surprising. The advantage of these  screening pro-
cedures lies primarily in the time savings afforded  due to
the total chromatographic character of the test, the use of
only one standard material and hence the reduced cost of
the analysis. Generally screening procedures can be com-
pleted for less than half the cost of analyzing a "laundry
list" of suspected organic materials.
   The advantages of these screening procedures also in-
clude the estimate of halogenated organics, for example,
which have illustrated 14 Mg/g (as chlorine) and yet un-
detectable levels of a list  of 16 specific organic constitu-
ents.  The final advantage of screening procedures  is that
the chromatographs may possibly be suitable  for addi-
tional review and prediction of the specific organic  consti-
tuents or class of constituents which comprise the screen-
ing results, assuming that  operational  conditions are
known. A particular example of this application is the
ability to detect the presence of PCB's in the halogenated
organic scan chromatograph.
   The major disadvantage in the use of  screening pro-
cedures at an uncontrolled hazardous waste site investiga-
tion is in the loss of specificity for detection of particular
organic  materials.  Since most sites, however, are being
investigated via  the primary exploratory and secondary
confirmatory design, this disadvantage does not outweigh
the cost savings realized by use  of  the screening pro-
cedures.

RECOMMENDATION

  Based upon Recra Research, Inc.'s work at various un-
controlled hazardous waste sites the following analytical
program is recommended for use in  exploratory efforts
at other sites.
•pH
•Conductivity
•Total Organic Carbon
•Total Recoverable Phenols
•Organic Scan (FID)
•Halogenated Organic Scan (BCD)
•Volatile Organic Scan
•Nitrogen-Phosphorus Organic Scan
  Additionally, it is recommended that samples from these
sites also be analyzed for the toxic heavy metals including
beryllium,  cadmium,  chromium,  copper,  nickel, lead,
zinc,  silver, arsenic, antimony,  selenium,  thallium and
mercury.

REFERENCES

1. U.S.  Environmental  Protection Agency,  Guidelines
   Establishing Test Procedures  for the Analysis of Pol-
   lutants; proposed regulations,  Federal Register Vol. 44,
   No. 233, December 3, 1979.
2. U.S. Environmental Protection Agency,  Sampling and
   Analysis Procedures for Screening of Industrial Ef-
   fluents for Priority Pollutants, Environmental  Moni-
   toring  and  Support  Laboratory,  Cincinnati,  Ohio,
   April 1977.
3. American  Public   Health  Association,   Standard
   Methods for the Examination of Water and Waste-
   water, 14th Edition, APHA, New York, N.Y., 1975.
4. U.S. Environmental Protection Agency,  Manual of
   Analytical  Methods for  Analysis  of Pesticide Resi-
   dues  in  Human  and Environmental Samples,  Re-
   search Triangle Park, 1974, revised 1980.

-------
     COMPATIBILITY FIELD TESTING PROCEDURES FOR
                  UNIDENTIFIED HAZARDOUS WASTES

                                       RODNEY D. TURPIN
                                    JOSEPH P. LAFORNARA
                                        HARRY L. ALLEN
                                   Environmental Response Team
                               U.S. Environmental Protection Agency
                                           UWE FRANK
                             Oil and Hazardous Materials Spills Branch
                        Municipal Environmental Research Laboratory—Ci.
                                         Edison, New Jersey
INTRODUCTION

  US EPA's Environmental Response Team was estab-
lished in October  1978 to provide technical assistance to
On-Scene Coordinators (OSCs), Regional Response Teams
(RRT),  EPA Headquarters and Regional Offices, as well
as other governmental agencies  in  the area of multi-
media, emergency environmental problems.
  The  Office of Research  and  Development  (ORD),
Municipal Environmental Research Laboratory  (MERL)
in Cincinnati, Ohio, has the lead role for research and
development relating to  hazardous waste  environmental
emergencies. Through  a program at its  Oil and Haz-
ardous  Materials Spills Branch (OHMSB) in Edison, New
Jersey,  MERL is developing prototype equipment and
experimental techniques  for  controlling wastes and pro-
viding analytical support to environmental incidents re-
quiring emergency response.
  In this paper, the authors describe  ERT's field testing
procedures for segregating unidentified hazardous waste
with portable identification  equipment. The compatibil-
ity classification scheme is based on pH,  total organic
vapors, oxidation  potential, flammability, water reac-
tivity and radioactivity.

COMPATIBILITY TESTS

  The primary purpose of the Compatibility Field Test is
to have a method for segregating uncharacterized drums
of waste into separate  storage areas.  The  tests (Table I)
can be either performed  at the drum  head (staging area)
or at a  mobile unit located within the site's "hot zone."
This procedure permits rapid analysis and results in an
expedited drum segregation operation. The selection cri-
teria for the different categories is shown in Table  II.

FLAMMABILITY

  Although  a standard  flash point test  would furnish
helpful  information about  the  unidentified  hazardous
waste,  this  method becomes impractical  when dealing
with large numbers of drums on an emergency opera-
tion. For example, if a test could be performed  every 20
minutes,  it would take approximately  3  man-years to
analyze 20,000 drums.
  Since it is possible to move, stage and sample between
100-200 drums per day, the drum segregation operation
would range between 20-40 weeks. Thus, a simple flam-
mability technique is used to  separate flammables from
non-flammables.
  Such a  technique could simply consist of placing 2-5 ml
representative hazardous waste sample  in a disposable
beaker. The beaker is placed in a large  sand box and a
propane  torch is  slowly passed  over the unidentified
waste. Obviously, the ambient temperatures will affect the
results, therefore, it is  important to simulate ambient
summertime  temperature. If a flame is observed, then
the sample is classified as flammable. A non-flammable
classification is assigned to the waste after the torch has
been passed over the waste several times.

RADIATION

  Radiation monitoring should be one of the initial tests
performed on unidentified hazardous waste drums and it
should be performed as soon as possible  to  prevent
worker exposure. Thus, the monitoring should be conduct-
ed as  soon as drums are placed in the staging area and
opened.
  Since normal environmental gamma  radiation back-
ground is approximately 0.01 to  0.02 milliroentgen per
hour (mR/hr)  on a  gamma survey instrument,  routine
employee exposure should not be more than 2-3 times
background levels. At no time should routine employee
exposure  be 10 mR/hr or above without the advice of a
qualified  health physicist.

PH

  The pH of a waste affects not only its corrosivity, but
also its compatibility with other  wastes. If a barrel of
acidic wastes were to leak and come in contact with 8
barrel of another waste containing a sulfide or cyanide.
accelerated corrosion of the second drum would occur
and the resultant co-mingling of the wastes would evolve
poisonous hydrogen  sulfide  or hydrogen cyanide gM-
                                                 110

-------
                                                                                             SCREENING    111
                        Table I.
                 Compatibility Field Tests
Test                             Category    Category

1. pH*                           Caustic (NF)     A
                                Caustic (F)       B
                                Acid (NF)       C
                                Acid (F)         D
2. Water Reactive                                  E
3. Oxidative/Reductive              Oxidizer(F)      F
                                Oxidizer (NF)    G
4. Radioactive                                    H
5. Volatile vapor/gas
6. Flammable
*the pH level depends on when the release of cyanide, sulfide and sulfide gases pose a threat.
(F) Flammable
(NF) Non-Flammable

Similarly, co-mingling of caustic wastes with ammonium
salts or amine salts could  lead to the evaluation of irri-
tant ammonia gas or amine  vapors.
  The guard against  these possibilities, caustic and acid
wastes must be segregated.  It is widely accepted that both
cyanide and sulfide ions must be kept above a pH of 9 in
order to  remain in aqueous solution. Therefore, caustic
wastes (Categories A  and B) were defined as those with a
pH above 9, and acidic wastes (Categories C and D) were
defined as those with a pH below 9.
  pH  can be determined  by  a variety  of colorimetric
and  electrochemical  techniques, each of which  has  its
disadvantages when used on "dirty" samples containing
organic layers, sludges or concentrated solutions. For in-
stance, standard pH  electrodes  are easily fouled  and re-
quire constant  cleaning and recalibrating. Most colori-
metric indicators and  papers are easily obscured by grease,
sludges or deeply colored solutions. Interfering chemicals
may even cause false color changes.
  The pH method selected for this protocol  utilizes  a
multi-band pH paper  strip  which contains a reaction zone
and a series of indicator colors (fixed) for reference. The
entire strip is dipped into  the waste  and the color reac-
tion allowed to occur. The  strip is withdrawn and any ex-
cess  sludge, organic  or water wiped-off  by squeezing  it
between the thumb and forefinger, while wearing rubber
gloves. The color comparison  of the reaction zone and
the indicator color is made assuming that the indicator
color and the reaction zone are both affected in the same
way by the grease, sludge or concentrated solution.

WATER REACTIVITY

  One purpose of hazardous waste segretation at a site
or a spill is to ensure that  antagonistic effects do not oc-
cur due to contact between incompatible wastes. While
many hazardous substances at a scene may be relatively
stable and compatible in a dry state, the high  probabil-
ity that water may contact these materials warrants con-
sideration of the  consequences  of such  an occurrence.
Thus, water reactivity, aside from being among the chart-
acteristics  identifying  a solid waste as a hazardous waste,
is important in assessing the waste segregation strategy.
  The characteristic of reactivity, as defined in the RCRA
regulations (40 CFR 261.23), is exhibited if a representa-
tive sample of the waste has  any of several properties.
Properties 2,  3 and  4  of the material are  of principal
concern to this discussion:
  (2) It reacts violently with water
  (3) It forms potentially explosive mixtures with water
  (4) When mixed with water, it generates toxic  gases,
vapors  or fumes  in  a  quantity sufficient to present a
danger of human health or the environment.
  The  EPA  laboratory manual,  "Physical  Chemical
Methods for Evaluating Solid Waste," does not include a
specific procedure for evaluating reactivity.  The reason
given is the diversified  nature  of the material properties
listed in the regulation. In practice, however,  water  re-
activity is commonly  observed  during the course of other
tests. The authors believe the dangers posed by the three
properties above explain why  water reactivity should be
determined.
  In routine hazardous waste  management, established
laboratory safety procedures and a general knowledge of
waste composition, are helpful in preventing a potential
disaster due to the  wetting of the reactive substance.
Since the  EPA proposed Comprehensive Hazardous Sub-
stance List contains 21 reactive chemical wastes and nine
additional acutely hazardous  chemical wastes, caution
must obviously be observed in the laboratory. But what
about in the field?
  The same analytical safety precautions should be applied
in the field where reactive substances  are known  to  be
present. Such items as  ammonium picrate or fulminate
of mercury  should be segregated and perhaps  not even
sampled if such materials are labeled or strongly presumed
to be present. In questionable cases, however, the authors
recommend a simple water reactivity screening test be ap-
plied to check for the evaluation of heat or other violent
reaction or the generation of toxic vapors.
  While the authors have not tried this test yet, it should
be  relatively safe if  precautions  against  explosion and
toxic vapor hazards are taken.  Unknown organic vapors,
cyanide, hydrogen sulfide,  chlorine, ammonia and hydro-
gen could be generated in small  amounts.  Hence, the
authors recommend the reactivity test be conducted after
other waste characterization testing has  been evaluated
and some idea of the waste composition has been gained.
  Briefly, the tests should be conducted as follows:

  (1) Place 500 ml distilled water  in  a  one liter  metal
      beaker in an explosion-proof hood, if  available, or
      behind a substantial barrier. Analysts  should wear
      appropriate  protective gear.
  (2) Insert a thermometer in  the beaker and record the
      temperature after equilibration.
  (3) Place an explosive  atmosphere  measuring  device
      and an organic vapor measuring device in position
      to pick up any gases generated in the beaker head-
      space using laboratory clamps, etc.
  (4) Introduce about one gram  (one  ml) of waste into
      the beaker.
  (5) Observe the  beaker measuring devices.

-------
 112     SCREENING
  (6)  If no changes are observed in temperature or head-
      space gases,  add an additional 4  grams (4 ml) of
      waste, added 1 gram (1 ml)  at a time and recheck
      instruments.
  (7)  If still no reaction is observed, add  an additional
      5 grams (5 ml).
  Wastes  testing positive to  water reactivity  should be
isolated and protected from rain or other water sources as
well  as  possible. Complete confinement, however,  (i.e.,
enclosure  in a walled building) is  not recommended be-
cause of a possible explosion hazard from  confining ex-
plosive vapors.

REDOX FIELD TEST KIT

  The oxidation reduction (REDOX) potential field test
kit was developed as a screening  procedure  for analyz-
ing and classifying drums as those containing oxidizing or
reducing agents at uncontrolled  hazardous waste  sites.
The  segregation of drums by redox potential is a neces-
sary  first step in a clean-up activity due to  the danger of
explosion  associated with proximate storage or shipment
of waste chemicals which have strong oxidizing or reduc-
ing properties.  This hazard  was  recently  demonstrated
when approximately 40,000 drums containing chemical
wastes exploded at  the  abandoned  "Chemical  Control
Corporation"  hazardous  waste  site  in Elizabeth,  New
Jersey.'"
                                   The REDOX  Test  Kit permits measurement of the
                                 REDOX potential of drum samples through use of a
                                 portable battery-operated instrument, electrode probes
                                 and electrolyte solutions.  The measurement of oxidation-
                                 reduction potential (ORP) or Electromotive Force (EMF)
                                 of a sample solution can be performed with a platinum
                                 sensing electrode and  a standard  reference electrode,
                                 usually calomel  or  silver-silver  chloride.  Although the
                                 instrument and electrode probes are  readily available
                                 the unique feature of the Test Kit rests in its ability to per-
                                 form REDOX measurements not only in aqueous but also
                                 organic matrices such as  are  found  at hazardous waste
                                 sites.  Operation of the kit involves  using  the electrolyte
                                 solutions to generate a known REDOX potential (EMF)
                                 and monitoring changes in the EMF caused by the addi-
                                 tion of sample  to the electroyle. The entire procedure for
                                 obtaining REDOX measurements requires only a few min-
                                 utes and can be performed by inexperienced operators.
                                   The feasibility of ORP measurements is based on the
                                 ability of the sensing electrode to generate  and the meter
                                 to indicate, 10 mv or less change in potential. At very low
                                 concentrations, microgram quantities can cause measur-
                                 able EMF changes. However, such sensitivity would mean
                                 that most tests  would be positive. Therefore, it is prefer-
                                 able to use a standard  test  solution that will generate a
                                 known EMF, and monitor changes in the EMF caused by
                                 the  addition of the sample. For example, a 0.001 N solu-
                                 tion of ferrous  ammonium sulfate will generate a cell po-
                                                     Table II
                               Selection Criteria for the Compatibility Field Testing Methods
   Category
High
pH9
Low
pH9«
HNU or OVA on
 vapor spice
                                                                  Flam.
Re-
dox        R/A
                                                                                                       B/T
   A-Caustic  (NF)      +

   B-Caustic  (F)       +

   C-Acid (NF)

   D-Acid (F)

   E-Water
     Reactive

   F-Oxidizers  (F)

   G-Oxidizers  (NF)   -

   H-Radioactives
   •At pH 9 Hit release of cyanide, sulfide and sulfide gases post a threat.

   IT)  Flammable     R/A Radioactivity

   (NF) NoD-Flmmmable  B/T Beaker lest for »aier reactivity

-------
                                                                                           SCREENING     113
tential  of approximately  380 mv versus a  silver-silver
chloride reference electrode. When an oxidizing agent is
added to the test solution, the Fe++ of the ferrous ammon-
ium sulfate will be oxidized to Fe+ + +, raising the cell EMF.
  The test is very sensitive and a reaction with only a small
portion of an oxidizing agent will give a positive test.
Failure to obtain a positive test would indicate an ab-
sence of any strong oxidizing agents. Potassium chromate
can be used in place of the ferrous ammonium sulfate as
a standard test solution for determining the presence of
reducing agents.
  The test kit consists of a portable pH  meter capable of
EMF measurements  (Fisher  Scientific  Accumet Model
150, for  example), and a platinum sensing combination
electrode with a silver-silver chloride reference electrode.
Two types of combination electrodes are currently used:
Orion Model 96-78 and Fisher Scientific platinum REDOX
electrodes. In addition, the kit contains 0.001  normal fer-
rous ammonium  sulfate  and 0.001  normal  potassium
chromate standard test  solutions, measuring  flasks, dis-
posable 50 ml volume beakers, and disposable containers
for taking sample measurements.
  In the field,'a chemical waste sample  is  added to a
standard test solution and then the  change in potential
(mv) is measured with the pH meter and electrodes.  Each
standard test solution gives a constant reading. Changes in
those readings either in the negative or positive direction
upon addition of sample, indicates the presence of an oxi-
dizing or reducing agent.
  Thus by using the ferrous ammonium sulfate test solu-
tion, the standard reading with  the silver-silver chloride
reference electrode is 380 mv. A threshold change of 50
mv in  the positive direction (over 430 mv) indicates the
presence of oxidizing agents. The potassium chromate test
solution gives a standard reading of 630 mv. A threshold
change of 50 mv in the negative  direction (less than 580
ms) indicates the presence of reducing agents.
   An actual field evaluation was  performed during Janu-
 ary  1980 at the Ottati and Goss site in Kingston,  N.H.
 Two test kits were successfully used to screen drums con-
 taining chemical wastes for the presence of strong oxidiz-
 ing agents. Two minor problems were encountered: elec-
 trode  probe clogging and freezing  temperatures.  The
 clogging problem was easily resolved, however, by proper
 probe cleaning between samples.  The cold weather, how-
 ever, caused the probe electrolyte to  freeze and therefore
 prevented use of the kit at ambient temperatures. Remed-
 ial measures involved using the kit in a small heated shed,
 located at the site.
VOLATILE VAPOR/GASES

  The objective of taking total vapor concentration values
just inside the barrel's bung  hole is to assist in determin-
ing whether or not the headspace has a potentially ex-
plosive atmosphere. In utilizing total vapor concentrations
as a guide for determining the presence  of organic ma-
terials or explosion materials, a number of factors should
be considered:
  (1) The uses, limitations and operating characteristics of
      the monitoring instrument must be recognized and
      understood. Instruments such as photoionizers and
      organic vapor analyzer (OVA)  do  not respond to
      all substances that may be present or may respond
      differently to identical  substances when compared
      to one another.
  (2) Hazardous other than detectable gases/vapors such
      as phosgene,  HCN, chlorine, liquid/solid particu-
      lates and other harmful conditions may exist.
  (3) Potential for  an explosive atmosphere exists when
      the  instrument  reaches  its maximum  reading  of
      2,000 ppm and should be checked with an explosive
      meter.
  The primary purpose of the total vapor testing is to de-
termine explosion potential.  If the  selected instrument
reads 2,000  ppm, one has to make a decision as to  the
classification. The authors practice has been to classify
the waste as flammable if 2,000 ppm readings are obtained
until further confirmation is available. This decision takes
into  consideration instrument limitations, ambient temp-
eratures, weather  conditions, operator experience and
sampling rate.

ACKNOWLEDGEMENT

  The authors express their appreciation  to all members
of EPA's Environmental Response Team, EPA's Oil and
Hazardous Materials  Spill Branch, Municipal  Environ-
mental  Research  Laboratory, Edison, New  Jersey and
Princeton Testing Laboratory, Princeton, New Jersey.

REFERENCES

1. Frank, U., Gruenfeld, M.,  Losche,  R., and Lafornara,
   J.P., "Mobile Laboratory Safety and Analysis Proto-
   cols Used for Response to Abandoned Chemical Waste
   Dump  Sites and  for  Oil and Hazardous  Chemical
   Spills." Proc. 1980 National Conference on Control
   of  Hazardous Material  Spills,  May  13-15,  1980,
   Louisville, Ky., 259.

-------
       NOVEL APPROACHES  TO SCREENING OF AQUEOUS
            EFFLUENTS FROM HAZARDOUS WASTE SITES

                                              IHOR LYSYJ
                                   Rockwell International Corporation
                              Environmental Monitoring & Services Center
                                        Newbury Park, California
INTRODUCTION

  One of the  central  issues  in  management of uncon-
trolled hazardous waste sites is control of contamination
of surface and ground waters due to  leaching,  followed
by the entry of the leachates into water supplies render-
ing them unsuitable for use. The  extent of this problem is
underscored by the attention it received at the 1980 Na-
tional  Conference  on  Management   of  Uncontrolled
Hazardous  Waste Sites."' The majority of  papers  pre-
sented in this conference addressed this problem in some
manner. Topics dealing with  site assessment  and subsur-
face  investigations  dealt heavily with the  intrusion of
leachates into  groundwaters.  Remedial techniques  and
methods for treatment, as well as case histories, dealt to a
large extent, with containment of aqueous effluent from
the hazardous waste sites.
  Silka12', in a discussion of assessment of groundwater
contamination, states  that "because of their proximity to
population  centers, both industrial and municipal im-
pounds have  the  potential  to  directly affect human
health." The preliminary finding  of surface impoundment
assessment, conducted by the EPA,  indicates that almost
one-third of the 25,000 impounds  utilized  for storage,
treatment, and  disposal of liquid  waste have a high poten-
tial to contaminate usable aquifers.
  Timely detection, chemical characterization, and quan-
tification of toxic contaminants in aqueous effluents from
waste sites  is an essential first step in  alerting the public
to the potential health hazard and institution of preventive
and mitigation  steps. Monitoring of the quality of ground
and surface waters in water supplies affected by toxic con-
tamination  is a necessary activity during decontamination
and  effluent containment periods of  response activity.
During post-closure periods the water  quality of affected
aquifers and/or surface waters must also be monitored to
assure that  no toxic materials are leaching into water sup-
plies.
  Chemical characterization of water samples, in terms of
detection and quantification of toxic materials that might
be present  in  the aqueous  effluent from  uncontrolled
hazardous waste sites  is, unfortunately, complex,  time-
consuming, and an expensive proposition, when carried
out routinely over a period of decades at numerous  haz-
ardous waste sites.
  An operationally simple and cost-effective method is
needed to indicate  whatever ground and/or natural wat-
ers are contaminated or uncontaminated. Such a method,
based on differentiation between  gross chemical char-
acter of  natural waters  and  gross  character of contam-
inated water was developed and is described in this presen-
tation.

CONCEPT

  The organic chemicals contained in natural, unpolluted
water come mainly from biomass activity and are com-
posed of carbohydrates,  proteins, lipids, and their degra-
dation products.(3)  As such,  the  chemicals  are distinctly
different from those supplied by leachates from hazardous
waste dumps, which are composed largely  of refractory
materials. Such differences in overall chemical compo-
sitions provide  a   means for differentiating  between
natural and man-made contributions to receiving waters.
  A  method of multicomponent pattern recognition pro-
vides a means for such differentiation. Using this method,
it is possible to define a complex chamical compositional
a separate entity, and to differentiate it quantitatively and
qualitatively from  other chemical compositions such as
might be found in an industrial waste.
  The pyrographic  met hod of analysis is used as a phys-
ical vehicle for application of this concept to pollution de-
tection and monitoring in water.(4) In this method, organic
matter is subjected  to elevated temperatures in the absence
of oxygen, leading to thermal fragmentation of pareilt
organic  compounds. When such  a  molecular fragmenta-
tion  was carried out in  the presence of a large excess of
water,  the  processes  were   linear with concentratiffll
and independent for each organic compound in a rnixtfiftt.
The pyrographic pattern produced by a mixture of organfc
compounds is a simple arithmetic summation of contrib-
uting patterns of each compound present.
  A  recorded pattern  of pyrolytically  produced  frag-
ments for a given water sample reflects the total nature of
its organic composition, and can be interpreted and dif-
ferentiated in a number of ways. Using a priori established
calibration patterns for individual components to be found
in a  mixture, the pattern produced by a mixture can be
analyzed mathematically. The system can  be calibrated
and then differentially analyzed in terms of unique organic
compounds, classes of organic materials, or in terms of
                                                    114

-------
                                                                                         SCREENING     115
any  other  arbitrarily  defined  organic  compositions,
such as are found in industrial waste or in a biomass-
produced organic matter.
  Normally it is possible to set a linear equation for each
peak of a composite pyrogram, and then solve a series of
simultaneous  equations for unknown  concentrations of
organic materials represented  by the pyrogram. A maxi-
mum number  of  components, into which  data  can be
solved, equals  the number of  peaks observed on a pyro-
gram.

METHODOLOGY

  There are three essential components to this methodol-
ogy: instrumental hardware, mathematical logic, and com-
putation means.

Instrumental Hardware

  The instrumental hardware'5' consists of a water sample
injector, flow stabilization controller, pyrolysis chamber,
gas chromatographic column,  and flame ionization detec-
tor.  Associated components include a supply of carrier
gas, oxygen, and hydrogen, as well as electronic and data
reduction equipment.  Data reduction operations can be
connected by telephone to a remote time-shared computer
or performed in the field using simpler electronic calcula-
tors. A photograph of a compact field instrument is shown
in Figure 1.

Mathematical Logic

  When a single chemical entity (such as an organic com-
pound) is being pyrolyzed, a number of derivative molec-
ular fragments are produced. They are revealed on a chart
of the potentiometric  recorder as  a  series of peaks with
varying retention times and peak intensities.
  Many organic compounds will produce similar products
when pyrolyzed. Showing up on a pyrogram as peaks with
identical retention times. However, the intensities of such
peaks will vary. Those variations in intensities produce a
pattern specific to that  of the parent material.
  If more than one organic compound is pyrolyzed simul-
taneously (such as is the case when a sample containing a
mixture of compounds is analyzed), the pyrographic repre-
sentation will be a summation of pyrograms produced by
individual components. In this case, each common peak
for each component will be a summation of contributions
                                                   Figure 1.
                                           Pyrographic Instrumentation

-------
116    SCREENING
from each component. The specific peaks will, of course,
appear independently.
  Whether two or 2000 compounds are present in analyzed
mixtures, the common peaks will be  summations  of con-
tributions from  each contributing compound, and spe-
cific peaks will  appear with identities of their own. In
strictly mathematical terms, a pyrographic representation
of data becomes a sum of  various numbers. Any mathe-
matical sum, of course, can be divided into smaller num-
bers.
  Normally, it is possible to set a linear equation for each
peak of a pyrogram,  and  then  solve a series  of simul-
taneous equations for  the unknown concentrations of
organic materials represented by the pyrogram. A maxi-
mum number of components, into which pyrographic data
can be solved, equals the number of  peaks observed on a
pyrogram.
  The  organic compositions into which pyrographic  in-
formation can be differentiated are defined by  calibra-
tions, which provide  the coefficients for a series of simul-
taneous equations. The system can be calibrated in terms
of molecular entities, classes of organic materials, or any
other arbitrarily  defined combination of organic mole-
cules.
  The  mathematical  technique  for  treatment of pyro-
graphic data in  the  described manner was described.(6)
The technique is  based on  the use of least-squares solu-
tion to a number of linear equations.

Computations

  In past studies, the data  were  transferred by telephone
lines to time-shared computers which performed the neces-
sary calculations. With the advent of  increased sophistica-
tion in compact  calculators, computing operations could
be performed on site, making the system totally field-suf-
ficient.

APPLICATION

  Application of this methodology to screening of aque-
ous effluents from uncontrolled hazardous waste sites is
based on the assumption that natural, background chem-
ical composition of uncontaminated waters  is definable,
either in empirical terms of pyrographic  patterns or in
fundamental  terms of organic class  composition  (carbo-
hydrates, proteins, lipids), and is distinctly different from
the composition of industrial chemical wastes. Because of
these empirical and  fundamental differences,  industrial
chemical waste can be differentiated from natural chem-
ical background of  unpolluted  water  and characterized
in terms of nature and quantity. Support for this prop-
osition can be found in the literature.n-8)

Natural Waters

  Extensive data on organic composition in inland pollut-
ed waters were collected by Birge and Juday"1  over 50
years ago. Their study included a series of Wisconsin lakes
widely  different  in hydrological  and limnological char-
acteristics (from clear, gravel glaciated bed lakes to lakes
undergoing entrophication) and ranging in organic content
from 4 to 34 mg/1.
  The analysis of water samples from such diverse lakes
disclosed that in all  cases  organic  load was comprised
essentially  of three  classes of  organic  matter:  carbo-
hydrates, proteins, and lipids. Such  natural organic com-
position is derived largely from biomass activity and decay
of vegetable matter. A typical range of organic compounds
in inland lakes as reported  by Birge and Juday  is shown
in Table I.

                       Table I.
          Organic Composition of Natural Waters
              According to Birge and Juday(3>
                    Concentration, (mg/I)
Class 1 Lake

Class 3 Lake

Class 4 Lake
 Carbohy-
 drates

 2.6

18.7

28.6
Proteins


1.0

3.8

4.9
Lipids


0.1

0.2

0.2
 Tolal
 Org.
 Loid
 3.7

22.7

33.6
  As  total organic load in fresh water increases (from
clear, gravel bed lakes to swampy lakes) the increase in
concentration  of carbohydrates is proportionally greater
than increases in concentrations of proteins  and lipids.
This is so because proteins and lipids come primarily
from algal sources, while carbohydrates are also supplied
by decaying vegetation of swampy waters.
  Using this model as an example, a study was designed
and executed in Georgia'9' using natural, unpolluted waters
of wide-ranging  hydrologic and limnologic character. The
organic content of the water was characterized pyrograph-
ically,  both  in empirical form (as a series of  peaks on a
pyrogram) and also in fundamental terms of organic class
composition. The organic class compositions were com-
puted from empirical  pyrograms after the system was cal-
ibrated with model compounds for carbohydrates, pro-
teins, and lipids.
  Included in  the study  were streams, rivers, reservoirs,
and swamp  water sources in the southeast  region of the
United States.  This region is well-suited for such an inves-
tigation because it is neither highly industrialized nor pop-
ulated  and still contains many unpolluted water sources.
Those  waters  also  represent a variety of hydrochemical
types:
COWEETA HYDROLOGICAL STATION: The Coweett
Hydrological Station  is maintained by the  U.S.  Depart-
ment  of  Agriculture  in  the Appalachian Mountains OB
the border of North Carolina  and Georgia. The water
source has a low mineral and organic content. An extensive
hydrological record is  available for this watershed.
PIEDMENT REGION STREAMS: Near Athens, Georg-
ia,  are a number of swift-moving streams,  rivers, and
reservoirs. Alcovy  River and Clark Hill Reservoir were
used as water sources in this study.

-------
SCREENING    117

15

14
13

12

11

10
0
s 9
| 6
tr
< ?
0.
6

5

6

3
2

'

_

—
-


@ Mountain Streams I37mg//organ
— © Foothills Streams (5 5 mg;^ organ
(Q. Coastal Rivers (17 Omg/£ otgan
© Swamps ( ~dOmg/i organ



_



—

—

—
fS) @
—


L.

PEAK








CS|
CS)
CS)
CS)












©



II



























NUMBERS























































Figure 2.






















©


I
|
Table 11.
Pyrographic Data: Empirical Patterns for Natural Waters
(Peak area in microvolts/sec)

Coweeta(a) Coweeta(a) Coweetat3)
Peak Watershed Watershed Watershed Average(a)
No. No. 6 Stream No. 6 Pond Main Stream Response

1 1126 1113 980 1073
2 149 168 156 158
3 2043 1996 2213 2084
4 1029 906 1005 980
5 148 175 212 178
6 158 206 201 188
(b) (b)
Shoal Creek Alcovy River

1 1434 1712 1573
2 170 199 185
3 2400 2778 2589
4 1023 1153 1088
5 279 348 314
6 270 268 269

Little River'c'
Tifton
1 5257 . 5257
7 Qfi<; QCC
3 5246 5246
4 2443 2443
5 697 697
6 914 914
Pyrographic Patterns for Natural, Unpolluted Waters
COAST PLAINS RIVERS: The U.S. Department of
Agriculture, Hydrological Model Watershed in Tifton,
Big Floyd'd) Cravens^'
Water Prairie Island Spillway*"'

Georgia, was used in this study. The watershed comprises 1 12-590 15, 960 15,990 16,100 15,410
the Little River and its tributaries. The tributaries flow \ ^2'.lfo isleo? is^es is! 377 u!?28
through cultivated lands, and
cultural pollution. However,

carry some degree of agri- 4 6,030 9,114 9,876 9,357 8,594
water from the tributaries 5 829 1,236 1,137 1,221 1,106


K i/KK Tina T cvft T 9 at, v a/\ 7
filters through swamps before it reaches the Little River. 7 2^376 2^43 i'lain MOO 2J78
COASTAL SWAMPS: Two principal swamps of the
area are the Everglades in Florida, and the Okefenokee  "ou"*?1i" streams
( b ) Foothi 1 1 St resins
in Georgia. Culturally unpolluted waters of swamps con- (c) coastal Plains River
tain large quantities of organic matter. Water samples (
-------
118    SCREENING
                        Table HI.
    Organic Composition in Natural Waters as Determined by
                  the Pyrographic Method
 Mountain Streams
  Coweeta
 Foothill Streams
  Alcovy R.
  Shoal Cr.
 Coastal Plains
 Rivers
  Little R.
  Mills Cr.
 Swamps
 Everglades
 Moving Streams
  Everglades Ponds
                 Total
                 Organic
                 Load (mg/l)
 3.7

 5.2
 5.8
17.5
16.6
24.7
33.4
                 Organic Class
                 Concentration, mg/l
             Carbohy-  Proteins  Lipids
             drales
 2.2

 3.4
 4.7
11.7
11.2
19.7
27.9
1.4

1.6
1.0
5.1
4.8
3.7
3.9
 0.1

0.2
0.1
0.7
0.6
1.3
1.6
were supplied by 20 major industrial  operations in Ala-
bama, Florida, Georgia, Mississippi and  South Carolina
including: (1) Hercules, Inc., Hattiesburg, Mississippi; (2)
Air Products and Chemicals,  Inc., Escambia Plant, Pen-
sacola, Florida; (3) Merck and Co., Inc., Medicinal Chem-
ical  Plant,  Albany,  Georgia; (4) Monsanto  Company
Plant, Pensacola,  Florida;  and (5) Delta Airlines,  Inc.,
Atlanta, Georgia. The industry sources can be classified
as chemical, food,  textile, paper  and  pulp, oil refinery,
brewing, pharmaceutical and aircraft maintenance.
  Samples of industrial waste  discharges were collected at
the points of discharge into receiving waters and were pre-
served by freezing in dry ice.  The collected samples were
analyzed pyrographically under standard conditions.
  The  calibration  constants  for 19 industrial  waste
effluents (one plant reclaimed its waste, and  could not
technically be considered as a waste discharger) were  ex-
amined with  the object of determining  the  differences
between  pyrographic  patterns. The data  are reported in
Table  IV.  A few representative  patterns are presented
graphically in Figure 3.
  Examination of reported data reveals that pyrographic
patterns of industrial waste are indeed different from pyro-
graphic patterns of natural  unpolluted waters  and hence
can be determined and differentiated from natural organic
background.

DETECTION OF WATER CONTAMINATION

  Liquid and solid chemical wastes are usually either dis-
posed  of in landfills, in open pits (impoundments) or
stored  in  55-gallon drums.  Contaminants enter ground-
waters from impoundments as a result  of ground penetra-
tion. They can also enter surface waters as a result of run-
off. Chemical wastes stored in 55-gallon drums may leak
out with the passage of time and enter surface waters.
  Chemicals that enter natural, unpolluted ground and/or
surface waters alter gross their chemical composition. In
addition  to  carbohydrates, proteins and lipids which  are
found in  natural waters, man-made refractory and per-
sistent chemicals appear.  Due to dilution  by receiving
waters, such man-made contaminants or pollutants are us-
ually found in trace quantities and in many instances could
be present in concentrations lower than  that  of organic
compounds present naturally in water.
  The task is then to detect and quantify trace amounts of
contaminants in the  presence  of  natural organic back-
ground of water, under conditions when natural organic
background of water  can be greater than  that of the con-
taminants. Hence, there could  be two distinct and differ-
ent situations:
•The  nature of  hazardous waste contamination is un-
 known.
•The nature of hazardous waste contamination is known.
  Depending on the  situation, two different approaches
to detection and quantification  of water contamination by
the hazardous waste can be implemented.

Detection of Unknown Contaminants

  In many instances the exact nature of chemicals disposed
of at a given hazardous waste site is unknown and exten-
sive trace  analytical work must be undertaken to establish
whether surrounding  water sources are contaminated or
not.
  Using the  pyrographic method of analysis, it is possible,
however, to  develop an empirical pattern, visual examina-
tion of which will indicate whether or not contamination
of ground or surface water supplies took place. As shown
in Table II  and graphically displayed in  Figure 2, pyro-
graphic representations of unpolluted natural waters are
distinct and characteristic. In cases where water is polluted
by man-made chemicals, pyrographic patterns will become
                                                   .1.   .
                                                                  Figure 3.
                                                  Pyrographic Patterns of Industrial Wastewaters

-------
                                                                                           SCREENING    119
                                                    Table IV.
                                  Pyrographic Fingerprints of Industrial Waste Effluents
                                              (Peak Magnitude x KH)
Food and Beveraoe
Average
Peak Retention
No. Time
1 1.9
2 2.2
3 3.3
4 4.0
5 6.5
6 14.0
Various Industrial
Average
Peak Retention
No. Time
1 1.9
2 2.2
3 3.3
4 4.0
5 6.5
6 14.0
Textiles
Average
Peak Retention
No. Time
1 1.9
2 2.2
3 3.3
4 4.0
5 6.5
6 14.0
Chemical Industrial
Average
Peak Retention
No. Time
1 1.9
2 2.2
3 3.3
4 4.0
5 6.5
6 14.0



Food-A
67.9
2.1
13.3
5.9
0.7
3.8

Paper
and
Pulp
96.6
4.5
32.9
25.0
5.1
29.0


Textile
A
607.0
110.9
576.5
569.2
95.4
156.1


Chemical
A
38.2
4.2
84.3
39.4
10.6
6.5



Food-B
22.5
1.4
31.6
12.0
2.8
8.4


Oil



Food-C
342.1
29.5
147.0
144.6
21.2
61.0



Refining Pharmaceutical
34.6
2.2
46.5
20.9
6.2
16.8


Textile Textile
B C
669.6 2,289.5
83.9 555.2
293.4 3,213.5
152.8 2,364.0
279.0 392.8
117.9 1,252.0


Chemical Chemical
B C
85.9 4008.0
7.2 328.0
75.1 330.5
35.4 1444.7
5.8 27.1
38.8 327.0
2320.7

115.7
87.8
19.5
60.5


Textile
D
1,229.0
328.4
2,050.0
1,571.0
210.1
565.4


Chemical
D
49.6
5.0
42.7
26.2
2.2
12.3



Brewing
248.5
39.4
129.9
110.8
15.8
54.4


Aircraft
Maintenance
237.7
17.6
324.6
122.9
7.6
25.4


Tpxt.il e
E
681.5
102.2
363.1
228.2
36.5
171.5


Chemical Chemical
E F
11.7 45.2
2.4 2.3
8.7 32.6
3.6 31.2
1.1 1.6
3.7 14.7
distorted and the ratio  and  magnitude  of peaks  will
change.
  In practice, it should be possible to match visually a py-
rogram of a water sample that is suspected to be polluted
by hazardous waste with a series of pyrograms typical of
known types of  natural  waters.  Pyrographic patterns
shown in Figure 2 are quite representative of the majority
of types of unpolluted natural water found in the coun-
try.
  If no pollution (intrusion of hazardous waste) took place
in ground or surface water, the pyrographic pattern devel-
oped will match one of the  types of natural, unpolluted
waters, as shown in Figure 2. If ground or surface waters
have been contaminated by man-made hazardous wastes,
then the  pyrographic patterns or representations of water
sample will be distorted and the ratios and magnitudes
of the peaks  will be different from those of natural un-
polluted waters.

-------
120    SCREENING
   A visual examination of pyrographic patterns provides
 the simplest form of detection  of the presence of haz-
 ardous waste (or other man-made pollutant) in the natural
 waters. The magnitude of contamination can be estimated
 visually from a degree of pyrogram distortion or computed
 mathematically as described below.
   The pyrographic instrument can be calibrated for detec-
 tion of principal  natural water organics:  carbohydrates,
 proteins and lipids. By solving a series of simultaneous
 linear equations, the concentration of each class of com-
 pound can be determined.  The computation logic is also
 provided with a statistical measure of the  quality of such
 computations:  these are "residual value" and  "percent
 fit."
   The residual value reflects the quantity  of information
 that could not be fitted into the calibrated model and "per-
 cent  fit" reflects a degree of agreement between calibra-
 tion factors and observed, unknown data.  "Low residual
 values" and high "percent fits" indicate that calibration
 factors and analyzed substances match, and no other than
 calibrated compounds are present in the sample. In case of
 natural water,  it means that only carbohydrates,  proteins
 and lipids are found.
   High "residual values" and low "percent fits" indicate
 that materials other than natural organics  (carbohydrates,
 proteins and lipids) are present in the sample. The higher
 is "residual value" and the lower is "percent fit," greater
 is the probability of contamination of water.
   In case of a natural  water the instrumentation system
 is calibrated for carbohydrates,  proteins and  lipids using
 known chemicals: glucose, albumin, and oleic acid. An ex-
 ample pyrographic analysis of two synthetic solutions of
 standard solutions is shown in Table V. Here  one can see
 that  observed  and actual values  for each  compound are
 close,  "percent  fit" is high, and "residuals" are low.
 Thus, no other than natural organics are present in this
 sample. In  practice, when  "percent fit" falls  below 75%
 and "residual value" increases over 1000, then  one may as-
 sume that the water sample is contaminated by other than
 natural organic matter.
                         Table V.
          Pyrographic Analysis of Synthetic Solutions
         Run*
                   Cbemlcml Concentration (mg/l)
                              Olek
               Albumin   Glucose   Acid  Relldtlll
                                           Percent Fit
e


i
r 2
i 3
S Avg Found
" Actual

s
e
r
e
s
Presenl
1
2
3
A\g. Found

9.69
7.12
8.04
8.28

10
19.14
18.69
19.52
19 12

22.47
20.88
19.44
20.93

20
9.3
9.41
9 59
943

3.17
3.4
3.3
3.29

3
.92
99
.97
.96

117.449 98
343.97 95
140.839 99



128.886 %
149.194 96
152.734 96

Detection of Known Contaminants

  The nature of chemical waste present  at a given haz-
ardous waste site might be known and a sample of waste
matter might be available for calibration. In such cases,
an accurate determination of the natural organics present
in the water and a calculation of amounts of hazardous
waste can be made.  The instrumental system is calibrated
in the usual manner for carbohydrates, proteins and lipids,
but is additionally calibrated  for one or  more suspected
hazardous waste compositions. After the analysis is per-
formed, the concentrations of natural organic and contam-
inants are computed as shown in Table VI.

                        Table VI.
   Analysis of Mixed Industrial Wastes in Natural Water Matrix
 Natural Background
 Proteins
 Carbohydrates
 Lipids
 Waste Source   Run:
 Chemical

 Textile Finishing
  Chemical

  Textile Finishing
1.22 mg/l
7.47 mg/l
 .46 mg/l
                 Volume Percent From Each Source
     2    3
0.48   0.42    .44   .46

0.66   0.66    .87   .91


Conc'n. of Organic Carbon
From Each Source (mg/l)
           .5
               Average   Actuil
               0.46     0.5
           .98   0.86
                      1.0
0.91

2.85   2.67
0.86

2.7
0.87

2.82
0.95   0.87

3.84   2.67
   Here the technician has used a sample of natural water
from the Okonee River, Georgia, contaminated with two
types of industrial waste. The natural organic background
of the water was 9.15 mg carbon/1, to which small quan-
tities of chemical and textile waste were added (equal to
approximately 3.5 mg carbon/1). The man-made contam-
inants represented approximately one-fourth of the organ-
ic matter present in the water, a situation likely to be found
in real life. The water sample was analyzed  five times and
results are shown in Table VI. The computer analysis of
the pyrogram revealed its natural organic composition and
identified and quantified industrial waste present. This ex-
ample illustrates that hazardous  organic wasle present in
quantities smaller than  natural  organic background of
water can be detected and accurately measured.
                                                            CONCLUSIONS

                                                              An instrumental method is described for rapid screening
                                                            of water samples suspected of being contaminated by haz-
                                                            ardous wastes.  The method can be used for analysis of
                                                            water  samples contaminated by unknown organic pollu-
                                                            tants or for the accurate quantification of pollutants, tM
                                                            nature of which is known. The technique is designed for
                                                            field use and can be applied for empirical characterization
                                                            of the sample or for accurate quantitative chemical an-
                                                            alysis. The described procedure is fast (15 to 20 minutes)
                                                            and utilizes water samples  as  received without pretreat-
                                                            ment or other preparations.

-------
                                                                                       SCREENING     121
ACKNOWLEDGMENT

  The material in this paper was extracted in part from the
EPA reports references (Nos.  7 and 9) (Contract  14-12-
802). The authors express our gratitude to Dr. H. Page
Nicholson, EPA Athens,  Ga., for  his support and en-
couragement in this work.
REFERENCES

1. Proceedings, U.S. EPA National Conference on Man-
  agement  of Uncontrolled Hazardous  Waste Sites.,
  October  15-17, 1980. Washington,  DC. Hazardous
  Materials  Control  Research  Institute, Silver Spring,
  Md., 1980.
2. Ibid, p. 45.
3. Birge, E.D. and Juday, C. Ecological Monographs 4,
  No. 9, 1930, 63-80.
4.  Lysyj, I., "Pyrographic Analysis  of Waste Waters,"
  Environmental Science and Technology, 8, 1974, 31.
5.  Lysyj, I, "A Pyrographic Instrument for Analysis of
  Water Pollutants. "American Laboratory. July 1971.
6.  Lysyj, I., Nelson, K.H. and Webb, S.R. "Determina-
  ation of Multicomponent  Organic Composition  in
  Aqueous Media," Water Research, 4, 1970; 157-163.
7.  Pyrographic Gross Characterization of Water Contam-
  inants. "U.S. EPA  Report  R-2-72-227.  Office of Re-
  search and Monitoring, U.S. Environmental Proltection
  Agency.  U.S. Government  Printing Office,  Wash-
  ington, DC 20402.
8.  Lysyj, I, "Pyrography - A New Hydrochemical Tool,"
  International Symposium  on Identification and Meas-
  urement of Environmental Pollutants. Ottawa, Canada
  (1971).
9.  "Pyrographic Gross Characterization of Water Con-
  taminants. Interim Report prepared for the EPA under
  Contract 14-12-802 (1970).

-------
         AMBIENT MONITORING FOR SPECIFIC VOLATILE
        ORGANICS USING A SENSITIVE PORTABLE PID GC
                                            T.M. SPITTLER
                                              ALAN W. OI
                                 U.S. Environmental Protection Agency
                                                Region I
                                        Lexington, Massachusetts
ABSTRACT

  With the intensifying concern over organic chemical
contamination at spills, dump-sites and chemical manu-
facturing  facilities, a need has emerged  for rapid assess-
ment of ambient air pollution by volatile chemicals. We
have recently tested  a  portable  photoionization detector
(PID) gas chromatograph, in several field incidents. The
unit  (Photovac  10A10, Thornhill,  Ont.) is  lightweight,
completely portable,  and requires only compressed air for
operation. It has a measured sensitivity in the .1 to 10 ppb
range for a wide variety of commonly occurring volatile
organic contaminants  such as  vinyl  chloride, benzene,
chlorinated methanes, ethanes and ethylenes and even low
volatility  compounds like nitrobenzene and  xylenes. A
characteristic of the PID system  is enhanced sensitivity to
aromatic and unsaturated molecules over the FID detector.
However, the Photovac PID has an uncharacteristic sen-
sitivity to  halogenated methanes and ethanes. This unusual
and theoretically unexpected sensitivity (e.g., CHC13 can
be detected at the 5-10  ppb level and 1, 2-di chloroethane
at  the 1-5  ppb  level  in l-5c.c.air samples) makes the in-
strument even more useful for general air pollution  mon-
itoring. There is no need to have different detectors  (such
as  Electron Capture) in order to determine all the volatile
haloalkenes and other alkenes of  interest.
  The paper describes  several incidents involving specific
volatile organics.
(1) We used the instrument to measure vinyl chloride in
   the vicinity of a plastics manufacturer. Levels of  VCM
   below 0.2 ppb were easily detected and  semi-quanti-
   tated in the field.
(2) Nitrobenzene was monitored in the vicinity of a ten-
   year-old spill site which was being dredged from a river
   bottom. Maximum ambient contaminant levels never
   exceeded 20 ppb of nitrobenzene 100-300 feet down-
   wind  of the dredge operation. Organic quantitation
   and calibration were conducted in the field.
(3) Acrolein was monitored for an OSHA study against a
   background of  15  or more solvent peaks in  another
   manufacturing facility. The OSHA safety level of 0.1
   ppm was never exceeded and sensitivity was found to
   exceed 30 ppb even in the presence of interfering sol-
   vent peaks.
  In the above studies, and several others that will be dis-
cussed, a key feature is the ability to make in-field meas-
urements. Techniques  for  field  calibration will  be dis-
cussed in detail. One promising technique is the use of the
two-column option which has two distinct field use bene-
fits.
  a.  A short column can be used to enable rapid  'screen-
ing' type analysis with very short retention times at am-
bient temperature (e.g.,  4  min. for detection of nitro-
benzene).
  b.  Two-column referencing  of known  retention times.
This technique of using a second column (and field stand-
ards) to identify retention times, provides a high  level of
confidence in the positive identification of unknown am-
bient contaminants  before  one collects  time-integrated
field samples for laboratory analysis using GC/MS as an
identification tool.
  The combination of field screening and measurements
on a portacle  GC, with integrated field samples  and
follow-up GC/MS  analysis, provides a powerful tool to
settle questions of  ambient organic contamination.  Ow
technique for doing such work will be discussed in de-
tail.
                                                   122

-------
MEASUREMENT OF FUGITIVE HYDROCARBON EMISSIONS
              FROM A  CHEMICAL WASTE DISPOSAL SITE
                                          JAMES A. PETERS
                                         KEITH M. TACKETT
                                        EDWARD C. EIMUTIS
                                    Monsanto Research Corporation
                                              Dayton, Ohio
INTRODUCTION

  One of the primary concerns is air quality near chem-
ical waste disposal sites, especially existing chemical land-
fills that were not constructed and operated according to
criteria now required by the Resource Conservation and
Recovery Act (RCRA).  Since hydrocarbon emissions from
landfills and disposal sites are fugitive emissions, they are
difficult to monitor. Quantifying the emission rates of air
contaminants from landfills has not been practical because
of the  complexity and costs. The  transfer behavior of
organic pollutants between the solid-air interface has been
studied in detail  for pesticide volatilization'1'  and theo-
retical estimation techniques have been reported for the air
pollution aspects  of volatilized organic compounds from
waste disposal sites.(2> 3> 4) However, at present  the extent
of air contamination from chemical waste disposal  sites
remains largely unknown.
  In this paper, the authors describe a technique that can
be used to measure air quality close to waste disposal sites
and to calculate the emission rate of a fugitive emission
area source (open source) at minimum cost.

SOURCE DESCRIPTION

  The Hooker Chemicals and Plastics Corporation is cur-
rently building a clay-lined waste containment facility at a
chlorine production complex in Montague,  Michigan, to
contain  chemical residues and contaminated  soil from
various locations at this site. The Michigan Department of
Natural Resources (DNR)  established several air qual-
ity conditions to be met  during and after the transfer
period.  Before beginning restoration activities,  it was
necessary to  determine background ambient concentra-
tions and emission levels  of the perchlorinated organic
compounds of concern.  The DNR  had chosen hexa-
chlorocyclopentadiene  (C-56) to represent  those com-
pounds.
  C-56 is a nonflammable liquid with a very pungent odor;
the odor threshold is as low as 0.33  ppb; some individ-
uals can detect as little  as 0.15ppb.<5) It is a key intermed-
iate in the production  of the c>dodiene group of chlor-
inated pesticides and flame retardants. Unlike some of the
pesticides derived from it, C-56 degrades rapidly by photo-
lysis, giving water-soluble degradation products. Tests on
its stability towards hydrolysis at ambient temperature in-
dicated a half life of about 11 days at pH 3-6, which was
reduced to 6 days at pH 9. The present TLV for industrial
exposure is 0.01 ppm,  or about 7% of the lowest vapor
concentration shown to produce chronic toxic effects in
laboratory animals.(5)
  Two inactive sites contained C-56 wastes that were to be
relocated to the new waste containment facility. The first
site is  a  roughly triangular above-grade dump partially
covered with soil, varying in height from 0 to 15 ft above
a flat ground surface. Trees were located on two opposite
sides of the site. The second site was an abandoned pro-
duction facility for chlorinated hydrocarbons, approx-
imately 200 ft square with a brine lake to the north. The
remaining areas surrounding the site were open.

MEASUREMENTS AND METHODS

  Measurements of pollutant concentration and meteoro-
logical variables were made at the waste disposal sites in
June 1980. Descriptions of the equipment and methodol-
ogy follow.

Open Source Sampling

  The  major difference between open sources and point
sources is that open sources do not have a definable point
of emissions. There is no established protocol for sampling
open sources; the methods employed depend upon the type
of information needed, the physical configuration of the
source, meteorological conditions at the time of sampling,
and interferences due to other point and fugitive sources.
The following methods  have been developed and were con-
sidered for sampling fugitive and open sources: 1) quasi-
stack approach, 2) grid/flux approach and 3) upwind/
downwind approach.(6> The  quasi-stack  approach was
unsuitable because the two sites  were not isolable and the
grid/flux approach was eliminated because  the disposal
site was heterogeneously soil-covered and the abandoned
production facility was not amenable to gridding.
  The  upwind/downwind approach is generally favored
because it preserves the integrity of the source and any
process associated  with it. It also eliminates any construc-
tion  to capture emissions, which may prove expensive.
Continuity of the emissions is generally of secondary im-
portance since the magnitude of the ambient air volume
into which the emissions are dispersed is large enough to
                                                   123

-------
124    AIR MONITORING
provide a degree of smoothing to cyclic or intermittent
emissions. Best of all, the upwind and downwind measured
ambient  concentrations truly reflect the actual pollutant
levels at the sampler locations.
Meteorological Variables and
Atmospheric Stability

  Dispersion  models,  normally used  to  predict  con-
centrations surrounding a point source  of known emis-
sion rate, are used in reverse for open  source  sampling
calculations.  Several downwind concentration  readings
are taken to calculate the source emission rate (after sub-
traction of the upwind or background concentration).
  Because of the relative ease of application, the  Gaussian
plume model was used in conjunction with Turner's(7) de-
termination of the dispersion coefficients ay and az, the
crosswind and  vertical  standard  deviations of plume
spread.
  Comparison of predicted and measured pollutant con-
centrations from point  sources indicate that, with suffic-
ient meteorological  information, the predictions of the
Gaussian plume model can be generally within a  factor of
two of measured values. The accuracy achieved using the
Gaussian plume model depends to  a  great extent on the
accuracy of predicting or measuring ay  and az
  The detailed meteorological measurements required for
the most accurate determinations of  ay  and  az are gen-
erally not available  for inexpensive, rapid  sampling and
analysis projects. The Pasquill-Gifford determinations of
cry and az as  given by  Turner, where ay and  a are
functions of the distance from source  and an atmosphere
stability parameter only, are strictly applicable only to dis-
persion over relatively smooth surfaces, which fortunately
was the case in this study.
  Pasquill-Gifford  atmospheric  stability categories were
determined using the protocol described in  Figure 1. The
upwind-downwind   method  is  strongly influenced  by
meteorological  conditions, requiring a wind consistent in
direction  and velocity throughout the sampling period.
For this reason upwind-downwind sampling is  futile dur-
ing stable air (stability classes E and F) and very unstable
air (stability class A). Typically then, weather with strong
frontal activity  and  conditions  conducive  to  inversions
are avoided, and upwind-downwind sampling is best prac-
ticed between 9 a.m. and 4 p.m. By confining the sampling
to periods of moderately unstable to  neutral conditions,
DlOltTIOH INDD :
•t
                                                     Figure 1.
                                 Flow Chart for Determining Atmospheric Stability Class.

-------
                                                                                   AIR MONITORING    125
the accuracy of the Gaussian  plume model  may be en-
hanced.
  Given the stability class and the downwind distance of
the sampler from the source, continuous functions are then
used to calculate values for ay and CTZ. (8)
  Wind speed and direction were monitored continuously
with a Climatronics model EWS portable weather station
with a 10-ft tower. An  example of its output is shown later
in Figure 4.

Sampling and Analysis of C-56

  Pollutant concentrations of C-56  were measured  by
collecting  air  samples  in a packed porous polymer tube
at a height of about four feet above ground level. Eight
samplers were used simultaneously; one sampling  posi-
tion was upwind from  the source and seven were position-
ed in an array downwind from the source.
  The C-56 sampling and analysis  procedure  followed
NIOSH Method No. P&CAM 308 for hexachlorocyclo-
pentadiene in air.(8) In this procedure a known volume of
air is drawn through  a tube containing Porapak T (an
ethyleneglycodimethylacrylate monomer), followed by in-
field,  ultrasonic-assisted desorption with hexane, subse-
quent gas chromatographic analysis  with a "Ni electron
capture detector and  comparison of  peak  heights with
those obtained from injection of standards.
  The sorbent tube used for collection of C-56 in  air is
shown in Figure 2. The pumps used to pull air through the
sorbent tubes were MSA Model G personnel pumps, which
are    rechargeable-battery-operated,    diaphragm-type
with a sample flow indicator and a sample  rate control
valve. They were operated for 90-113 minutes at a rate of
1.6-2.0  1/min. During  sampling,  the  flowmeter  float
position of each pump was checked periodically to ensure
that the proper flow rate was maintained. The combined
sampling/analysis detection limit is 0.001 ppb.
  During  the C-56 sampling period,  sulfur hexafluoride
(SF6) tracer gas was released at two known rates at an up-
wind  location that approximated  the virtual point  of
emission for the area source  being sampled, based on the
atmospheric stability  class and the  width of  the area
source. Ground level release rates of 100% SF6 were mea-
sured with a soap bubble flowmeter.
  Downwind samples  were collected in Mylar bags in a
sealed metal container. To collect  an air sample, the bag
was opened to the  atmosphere and a pump connected to
the sealed container.  The pump, operating  at about 3
1/min, created a light  vacuum which pulled the ambient
SF6 sample into the  bag. The sampling  duration was
about three minutes per sample.
  SF6 analysis was conducted  in the field with an AID
Model 511 portable gas chromatograph equipped with a
6 ft x 1/8 in. Molecular Sieve 5A  column  operable at
104°F and an electron capture detector. Analysis followed
a procedure supplied by the equipment manufacturer00'
Standards  of SF6 were  prepared in the field by dilution of
a 1.00 ppm SF6 standard furnished by a commercial sup-
plier. The combined sampling/analysis  detection limit was
0.02 ppb.
                             PLASTIC CAP
                             SILYLATED GLASS WOOL
                             75 mg PORAPAK T
                             25 mg PORAPAK T
                             4 mm i.d./6 mm o.d.
                             TEFLON TAPE
                             PLASTIC CAP
                    Figure 2.
Sorbent Tube for Collection of Hexachlorocyclopentadiene
                  (C-56) in Air.

-------
126    AIR MONITORING
DATA ANALYSIS AND RESULTS

  In this section, the discussion focuses on the use of the
Gaussian plume model  for area sources in combination
with the ratio equation technique for estimating the emis-
sion rate of an inactive chemical waste disposal site.

Plume Dispersion Model

  For a continuous point source, the Gaussian plume dis-
persion model is a rather simple form of calculation. Be-
cause the source under study was at ground level with lit-
tle or no plume rise (H = 0), and since ground level con-
centrations were taken (z  = 0), the model was simplified
to:
X  =
      nayazU
  Where X is the mean concentration at the point of mea-
surement; y is the distance from the sampler to the plume
centerline;  Q is  the source emission rate; u is the mean
horizontal  wind speed at the height of emission; and  cry
and crz are  the Gaussian standard deviations in the pollu-
tant concentration distribution in the horizontal and ver-
tical, respectively, at  the point of measurement, and are
both functions of the distance from the source (X).
  For fugitive source sampling the concentrations are mea-
sured and the emission rate Q is of interest. The model
then becomes:
Q  =
            a  u  exp  [~
  When sampling can be done  along  the plume center-
line (y = 0), the model is further simplified:

Q =  xnayazu


  In dealing with dispersion of pollutants from a defined
area rather than  a point of emission,  an  approximation
can be made by treating the area as a source having an in-
itial  horizontal standard deviation of plume spread, a  .
A virtual distance \j. can then be found that will give this
horizontal dispersion coefficient, depending upon stability
class. Then equations for a point source may be used, de-
termining av as a function of x +  x^. How this is done
is shown in Figure 3.
  In this procedure, the area source is  treated  as a cross-
wind  line source  with an initial normal  distribution, a
fairly good approximation  for the distribution across an
area  source. The  initial standard deviation for a roughly
square area source can be approximated by a   =  S/4.3,
where S is the length of a side of the area.17' For a ground-
level area source, the vertical dispersion coefficient a  is
left untreated as having originated from the actual source
                                                                                                    PUMK
                                                             STATION
                                                                                  Figure 3.
                                                           Schematic Diagram of the Gaussian Point Source Plume Model
                                                                          Applied to an Area Source.

                                                          rather than the virtual point.
                                                            The dispersion model then becomes:
                                                          Q =
                                                                                yo
     *>1/2a.
                                                                                             u
                                                            In this  sampling program the area source dispersion
                                                          model concept was used only to find the virtual point of
                                                          emission in order to site the SF6 tracer gas release point.
                                                          By knowing the disposal site side length S and the atmos-
                                                          pheric stability class, a  distance xy upwind from the dis-
                                                          posal site  edge was found at which to release the tracer
                                                          gas and allow the SF6  plume  to  travel across the area
                                                          source to be measured at several downwind locations.
                                                            During C-56 sampling, two grab samples were collected
                                                          at  each of three  downwind plume centerline  locations
                                                          for on-site analysis of SF6. Emission rates of C-56 were
                                                          then calculated by a ratio equation technique:
                                                           ,       = X(C-S6)
                                                           JC-56    x(SF6)
X  ESF6  X 24.45
                                                                                                      _
                                                                                                    P    T
                                                            Where EC.J6 is the emission rate of C-56 in g/hr; x(C-56)
                                                          is the concentration of C-56 at the sample point; x(SF«) "
                                                          the concentration of SF, in 1/hr, MWC.J6 is the molecular
                                                          weight of C-56 (212.11 f; and 24.45 is the molar volume of
                                                          a gas at STP in 1 /mole.
                                                            The measured concentrations of C-56 and SF6 should be
                                                          a mean  concentration over the same time interval as the
                                                          time interval  for  which the a's and /* are representative,
                                                          typically three minutes. An often-forgotten calculation in
                                                          plume dispersion modeling is to correct for time-averaging
                                                          differences due to  increased meander of wind direction
                                                          with time. When  estimating emission rates (or concentra-
                                                          tions) from a single source for the time  intervals greater
                                                          than a few minutes; the best estimate apparently can be ob-
                                                          tained from07':
                                                          xs  =
                                                            Where xs is the desired concentration estimate for the
                                                          sampling time ts; Xk is the concentration estimate for the

-------
                                                                                   AIR MONITORING     127
                                                    Table I.
                               Calculation of C-56 Emission Rate by SF6 Tracer Gas Ratio
                                   Method, Run 1, Abandoned Production Area.
                       Corrected
                          C-56
           Sample    concentration,
           point	ppb	
                SFe  concentration,  ppb
              97.0 cc/min,94.0  cc/min,
              SF» release   SF» release
                  rate          rate
                            C-56  emission  rate,  g/hr
                           97.0 cc/min,94.0 cc/min,
                           SF« release   SFS  release
                               rate           rate
             S*
             s,
             s.
0.053
0.047
0.032
0.71
1.13
0.75
0.46
0.57
0.30
0.35
0.20
0.20
0.31
0.22
0.27
                                   Average  C-56 emission  rate » 0.26 g/hr
                                   Confidence limits  (95%)     • 10.05
shorter sampling time tk; and  p should be between 0.17
and 0.2.  Equation (6) is most appropriately applied to
sampling times less than two hours.

Ground Level Concentrations

  An example of the average ground-level concentrations
of C-56 and SF6 tracer gas taken during one of the emis-
sion test runs is given in Table I. The form of the meteor-
ological data collected during each run is shown in Figure
4. When  the wind direction is shown to be steady (< 45 °
shift) over the sampling period, only the plume centerline
samplers  need be used in estimating the emission rate of
the source. Of course,  the precision of the calculated emis-
sion rate usually will improve with less wind direction
meandering. The ratio of predicted to observed concentra-
tions of SF6 ranged from 0.4 to 0.8 for all runs, which were
conducted under stability class C.
  The estimated emission rate  of C-56 shown  in Table I
applies only to the ambient conditions experienced during
the sampling period.  Any factor that could change the
volatilization rate, such as increased temperature or distur-
bance of  the soil cover, will cause a corresponding change
in emission rates from a chemical waste disposal  site.

CONCLUSIONS

  The results  of  the calculated  emission rates  from
sampling chemical waste disposal  area sources  indicate
that  a tracer gas-ratio equation technique can estimate
emission  rates with good  precision,  provided a few open
source sampling rules-of-thumb are followed. Seasonal or
weather-dependent trends on  the emission rate can  be
monitored with relative ease. The technique as described is
fully portable and can be assembled rapidly. Knowledge of
gas chromatograph operation in the field is the  only diffi-
cult skill  necessary.  Two  people can make two or three
runs per day, weather permitting.
  Regardless of how inaccurate the emission rate estimata-
tion  is,  the ambient  concentrations measured are real
values and can have an accuracy ascribed  to them. The
concentration of C-56  in test atmospheres was determined
by NIOSH in control  experiments by sampling  with bub-
blers containing hexane and subsequently  analyzing the
                                 bubbler solutions by GC/ECD. The determinations with
                                 Porapak T sorbent sampling gave values averaging 100%
                                 of those found by bubbler sampling over the range of the
                                 method. Sorbent sampling for C-56 has a precision of 8%
                                 over  the concentration  range of  13-865 ng/m};  the pre-
                                 cision of the analytical method is 3 %.
                                   While the collection and analysis method described here-
                                 in is specific for C-56, methods are also available  to quan-
                                 tify emission rates of many organic chemicals simultan-
                                 eously using the tracer gas/area  source technique. Low
                                 concentrations of a wide variety of pollutants are  expected
                                 at chemical landfills and sorbent trap sampling is consid-
                                 ered  to be more reliable, accurate,  precise, and  sensitive
                                 than grab sampling.0'112)
                                   Monsanto Research Corporation has developed a com-
                                 bination sorbent system based on Tenax-GC, Porapak R
                                 and Ambersorb XE-340 for use in collecting a broad range
                                 of organic compounds.03' Over 30 different sorbent ma-
                                 terials were evaluated  in this  EPA-sponsored  program
                                 (Contract No.  68-02-2774), whose goal was to choose a
                                 collection of commercially available materials that, when
                                 used in combination, showed  promise for quantitatively
                                 trapping  compounds  varying  widely in  volatilities and
                                 polarities.
                                   360

                                   315

                                   210

                                   225

                                   180

                                   125

                                    90

                                    45

                                    0
                                        1:00  1:30   2:00   2:30   3:00  3:30  4:00  4:30  5:00
                                                        TIME Of DAY (PMI
                                                            5:30 6:00
                                                       Figure 4.
                                  Meteorological Data for Run 1, Abandoned Production Area.

-------
128    AIR MONITORING
  Tenax-GC is the only high-temperature (750 °F) adsor-
bent (2,  6-diphenyl-p-phenylene oxide)  available which
allows the quantitative thermal desorption of low volatility
organic compounds. Porapak R is one of the highest-ca-
pacity polymeric adsorbents (n-vinyl pyrrolidone) with a
reasonable background level and with an overlap in the
range of utility with Tenax-GC.
  Ambersorb XE-340 (a carbonized  styrenedivinyl ben-
zene) gives less  difficulty with the desorption of com-
pounds of intermediate volatility, fewer detrimental effects
by water, and less reactivity with collected samples than
with charcoal adsorbents. Also, its  range of utility leaves
the smallest gap between polymeric  and carbonaceous ad-
sorbents  in the types of compounds collected. These sor-
bent materials are packed into three separate tapered glass
tubes for sampling, with flow being directed either in series
or in parallel. If sampling is done with the sorbent tubes
in series, the  arrangement is Tenax-GC at the air intake,
Porapak  R in the middle and Ambersorb XE-340 last.
ACKNOWLEDGEMENTS

  The authors are grateful to Hooker Chemical and Plas-
tics Corporation for the opportunity to present the results
of this work.
REFERENCES

  1. Spencer, W.F.  and Cliath,  M.M.,  "Transfer  of
    Organic Pollutants Between the Solid-Air Interface,"
    Fate of Pollutants  in the Air and  Water Environ-
    ments, Part I, I.H.  Saffet,  ed.,  John Wiley & Sons,
    Inc., New York, NY, 1977, 107.
  2. Shen,  T.T.  and  Tofflemire,  T.J.,  "Air  Pollution
    Aspects of Land  Disposal of Toxic Waste," In: Proc.
    of the 1979 National Conference on Hazardous Ma-
    terial Risk  Assessment,  Disposal  and Management.
    Information Transfer Inc., Silver Spring, MD, 1979,
    153.
  3. Shen, T.T., "Emission Estimation of Hazardous Or-
    ganic Compounds from  Waste Disposal Sites," Pre-
    sented  at the 73rd Annual Meeting of the Air Pollu-
    tion Control Association, June 22-27, 1980, Montreal,
    Quebec. Paper No. 80-68.8. 24 p.
 4. Thibodeaux, L.J., "Estimating the Air Emissions of
    Chemicals from Hazardous Waste Landfills," /. Haz-
    ard. Mater., 4, 1981,235.
 5. Bell, M.A., Ewing, R.A., and Lutz, G.A., "Reviews
    of the Environmental Effects of Pollutants: XII Hex-
    achlorocyclopentadiene," U.S. EPA, Health Effects
    Research Laboratory, EPA-600/1-78-047, December
    1978. 94 p.
 6. Kolnsberg,  H.J.,  "Technical Manual for Measure-
    ment  of  Fugitive  Emissions:  Upwind/Downwind
    Sampling Method for  Industrial  Emissions,"  U.S.
    EPA, Industrial Environmental Research Laboratory
    EPA-600/2-76-089a, April 1976. 75 p.
 7. Turner,  D.B.,  "Workbook of Atmospheric Disper-
    sion Estimates," U.S. DHEW, National Air Pollution
    Control  Administration, EPA-AP-26, January 1974.
    84 p.
 8. Eimutis, E.G., and M.G.  Konicek, "Deviations of
    Continuous Functions of the Lateral and Vertical At-
    mospheric  Dispersion  Coefficients," Atmos.  En-
    viron, 6, 1972, 859.
 9. "NIOSH Manual of Analytical Methods, Volume 5,"
    U.S.  Department  of Health, Education, and Wel-
    fare, DHEW-NIOSH Publication No.  79-141, August
    1979.
10. "Analysis of Trace SF6 in Air," Analytical Note (AN)
    116B of Application Information Data, Analytical In-
    strument Development, Inc.
11. Esposito, M.P., and S.M.  Bromberg,  "Fugitive Or-
    ganic Emissions from Chemical Waste Dumpsites,"
    Presented at the  74th  Annual  Meeting of the Air
    Pollution Control Association, June 22-26,  1981,
    Philadelphia, PA.  Paper No. 81-41.2. 14 p.
12. Hwang,  S.T., "Hazardous Air Emissions from Land
    Disposal/Treatment  Facilities,"  Presented at the
    74th  Annual Meeting of the  Air  Pollution Control
    Association, June 22-26,  1981,  Philadelphia,  PA.
    Paper No. 81-41.4. 12 p.
13. Brooks, J.J., D.S.  West, D. J. David, and J.D. Mulik,
    "A  Combination  Sorbent  System for Broad Range
    Organic  Sampling in Air," Proc. of the Symposium
    on the Development and Usage of Personal Monitors
    for Exposure and Health Effect Studies, U.S. EPA,
    Environmental Sciences  Research Laboratory, EPA-
    600/9-79-032, June 1971. 383.

-------
                    A STUDY OF THE EMISSION RATE OF
                VOLATILE COMPOUNDS FROM LAGOONS
                                         ANDREW T. McCORD
                                           Director of Research
                                         Recra Research Institute
                                              Amherst, N.Y.
 INTRODUCTION

 Theory

   Nusselt1  has described the  rate of evaporation of water
 from a lagoon as follows:
            ,0.2
                          . (CW-C0).
                                      .(1)
 G  = kilograms of water evaporated per hour

 Am = surface area of the lagoon in square meters

 D  = diffusivity coefficient of water  in square meters per
      hour

 L  = a dimension of the lagoon surface. For a circular la-
      goon, L,  in meters, is the  diameter. For an irregular
      shaped lagoon, such as square, oblong or elliptical,
      etc., it is the  diameter, in  meters of a circle of area
      equal to the lagoon surface. For long narrow lagoons,
      L will be the smaller dimension in meters.
 WQ = wind velocity in meters per second

 Co = humidity content of the air at a distance above the
     lagoon surface, or the humidity of the air blowing
     across the lagoon.
            	g water	
              dry air containing the water

   Nusselt's equation may be applied to the evaporation rate
of a volatile compound which is dissolved in the lagoon water
if  one replaces Cw, the moisture content of the air at the
immediate surface of the lagoon by Cv, the volatile content of
the air at the immediate surface of the  lagoon. Assuming that
there is no volatile  compound in the incoming air, Nusselt's
C0 for the volatile  compound is zero.  Nusselt's equation for
a volatile  compound dissolved in the lagoon water then be-
comes, in pounds per hour
GA = 39A
        m
•(C2
(W0)°-78 • (Cv) • (2.205)	(2)
                                              Cv is derived as follows:

                                                C,  the molar concentration of a volatile compound in the
                                              lagoon water
                                                                     _%
                                                                     Mv
                                                       100-% +%
                                                       Mw      My

                                   where % is the weight percent of the compound in the lagoon
                                   water. Because this value is most frequently less than one, the
                                   above equation can be reduced, with negligible  error to:
                                             C =
                                                 100
                                           , M
                                            M.
                                                •(3)
                                             Mw = molecular weight of water =18

                                             Mv = molecular weight of compound = 28.96

                                                The mole-fraction of a volatile compound P* at the lagoon
                                             surface (in the air) is the product of the molar concentration
                                             and  the vapor  pressure  (V.P.)  of the pure compound, in
                                             atmospheres, at any temperature, i.e.,
                                   P* =  (C)(V.P.)
From equation (3) P* = I—^
                                                                                             (4)
                                                                 (V.P.)
                                                                                             (5)
                                         _ /Pounds of compound in the lagoon
                                                   _ /
                                                     \
                                   also% =1' "p"" 7 "-•—---- "• "" -e—" )100
                                             Pounds of waste in the lagoon  /
                                                   -I^. 100	
                                                   ~  V    8.34 (sp.gr)
                                                                                   .(6)
   L0 =  Pounds of volatile compound in the lagoon
   V  =  total gallons of waste in the lagoon
sp .gr  =  specific gravity of the waste
                                                   129

-------
130     AIR MONITORING
Combining equations (5) and (6):
  , _  (LnXMw)-(V.P.)
       V-(8.34MMv)(sp.grK
                                                    .(7)
Cv -
       wt. of compound in the air at the lagoon surface
            wt. of air containing this compound

    _  (P*)MV
                                                    .(8)
       (1-P*)MA

 MA =  molecular weight of air


   Again, because the concentration of a volatile compound
 in wastewater is usually less than 1%, P* is small compared
 with unity so equation (8) can be reduced to:
 Cv    M
       P*MV
 Combining equations (7) and (8)

     _  (Ln)(Mw)(V.P.)      Jk
  v    (VX8.34)Mv(sp.gr)    MA
    =  (Ln)(Mw)(V.P.)
  v    (VX8.34)MA(sp.gr) '   '
                                                    •(9)
 Thus
       M,,
        IA   8.34

Therefore

       0.07453   /
                  = 0.07453
E0  =  Initial evaporation rate of volatile compound.

APPLICATION

   In  order to evaluate the emissions in real-life operation in
the lagoon, one assumes the following arbitrary conditions:

   •   The  volume of  waste in  the lagoon is maintained
       constant

   •   The feed volume and composition is constant

   •   The discharge volume is equal to the feed volume

   •   Temperature and wind  velocity are constant

   •   Steady state condition.


                        Table I.

Symbols used in material balance

V  =  lagoon volume in U.S. gal
F   =  feed rate in gal/hr
    =  discharge rate in gal/hr
K  =  volatile compound in the feed, Ib/hr

L0  =  amount of volatile compound in  the lagoon at start,
       i.e., time-zero, Ib
E0  =  initial  evaporation rate of a volatile compound in
       Ib/hr as computed from equation (12)
                                                                                        Table II.
                                                                            Material Balance Over a Lagoon for
"v (sp.gr) W/"'' 	 vw One Volatile Component
The analogous Nusselt expression now becomes m ,
in Feed Disch Evap. Am t
Time Lagoon Rate Rate Rate in
r (*-'hr)-39\ /D\°'22 (0-07453) Lfl Hrs Ibs Ibs/hr Ibs/hr Ibs/hr Lagoon
(W0)°.78(2.205) 	 (11) ° Lo
but L0 = (0.0834)(V)(%)(sp.gr)- - - -from equation (6) lst L°
substituting for L and combining the constants ^ 1

+K -K -EO =L]
+K -L,F -E0L! =1-2
                          0.22
Note, the sp.gr. of the waste is eliminated.
                                           iO.78
                                                   •(12)
                                                             3rd
                                                             nth
                                                                    L-
                                                                      n-l
                                                                              +K
                  +K
                                                                                                -O •
                                                                                                  L

-------
                                                                                        AIR MONITORING    131
summarizing,
          F   F
let X = 1 "TT^f0, a constant because F, V, EQ and L0 are
constants.       °

Then Ln = (L^) X + K	(13)
Using this approach, outlined in Tables 1 and  2, one  can
write:
Time from start  Volatile Content of Lagoon

n = 0                       Ln

n=lhr

n = 2 hrs
                        L0X + K
                                 = L,
n = 3 hrs
n = n hrs
   (L0X + K) X + K =
    L0X2 + KX +K

(L0X2 + KX +K)X +K =
L0X3 + KX2 + KX + K

L0Xn + K + KX + KX2 +
  KX3
                                                   = L
                                                   ~ Ln
Because K + KX + KX2 + KX3   .   + KXn-1 is a geometric
                                        V
progression with X less than 1, the sum
                                       1-X
                                                  .(14)
                  K
                 1-K"
Where n is very large,  Xn becomes very small and so  L0Xn
may be neglected.
Equation (14) becomes Ln = —-
                           K
                         1-X
                                                  .(15)
Equation 15, represents the steady state and Ln will be now
designated Ls

  In arriving at a steady state, i.e., when n is many hours, the
following equations are used:

  initial pounds of a volatile compound in a lagoon

      L0 = (0.0834(V)(%)(sp.gr.) 16

  pounds of a volatile compound in lagoon at a steady state
        _  K  pounds
       s ~ 1-K
                                                               pounds of a volatile compound in feed to lagoon

                                                                  K = 0.0834(F)(%)(sp.gr.)lb/hr
                                                                          E   F
                                                                   1 - X =-r-2 + T73 fraction less than unity
                                                               evaporation rate of a volatile compound in 16/hr.
                                                                             Ib/hr
                                                                   E =
                                                                    S    L
                                                               percent of a volatile compound in a lagoon at the steady
                                                               rate

                                                                  %s = (% at start)Ls
   Initial evaporation rate of a volatile compound in a lagoon
is given in equation (12). D and  V.P. are calculated for the
specific operating temperature.

EMISSIONS OF A VOLATILE COMPOUND IN AN
AERATED LAGOON

   Consider the case of a lagoon containing less than 1% of a
volatile compound in water. At the center of the lagoon, a
water pump withdraws lagoon water and directs  the stream
vertically against a deflector plate  to produce a cone of drop-
lets.  The  maximum  trajectory is attained when  the  initial
direction of the droplets is 45° from horizontal.
                                                                                             -R/4-
                                                                     LAOOON SURFACE   S~   U PUMP
                                                                                 'DEFLECTION
                                                                                    CONE
                                                            Vj  = initial velocity of the droplets in ft/sec

                                                            R   = maximum diameter of the spray falling back into the
                                                                  lagoon

                                                              The horizontal and vertical components of the velocity
                                                            Vh and Vv of a droplet are Vj cos 45° and V0 sin 45° respec-
                                                            tively and are both equal to 0.7071Vj. The maximum height

-------
132    AIR MONITORING
attained  by a droplet will be,  theoretically, at a distance
D
—  from the deflector plate.
   Let t seconds be the time for a droplet to reach maximum
height, i.e., to reach the point A.

Then :
 From Newton's Laws of Gravity:
                                                            e   =  time in seconds = 0.17678R^

                                                            TT   =  3.1416

                                                            Fv =  correction factor, which approaches unit for dilute
                                                                   solutions of volatile compounds

                                                            L0 =  amount of volatile compound in the lagoon. Ib.

                                                            VL =  volume of liquid waste in the lagoon, U.S. gal
                                                            Mw =  Mol. wt. of water = 18

                                                   .(16)     Mv =  Mol. wt. of volatile component

                                                            V.P.=  vapor pressure in atmospheres of the volatile com-
                                                                   pound at T°F sp.gr. = specific gravity of the liquid
                                                                   waste
    32
                                                   •0 7)
 Combining equations  (16) and  (17)  and  letting Vh = Vv =
 0.7071V,:
                                                   (18)
and, from equation (16):

t = 0.08839Rl/i	(19)

The total flight time for a droplet = 2t sec:

2t = 0.17678R'/i	(20)

For a spherical water droplet:

surface area _ 4 ir r-* _ 3
 volume    4_77 r^   r
            3~
 H = water delivered by the pump comp.

    or.(H)'n3785)=63.083Hcm3/sec
          60
The surface area of this mass of water, as spherical drops, r
cm radius
= --63.083H
  r
= 189.25-cm2 .
        r
                                                    (21)
   Arnold- has studied the diffusion of a volatile compound
from a liquid surface into air and arrived at this conclusion:
                                                            Substituting these values in equation (22) one obtains:

                                                              _2H /189.25V L^ . MW . (V.P.)
                                                                 r  V  8.347 VL  Mv  (sp.gr;

                                                            The emission in Ib/hr
                                                                = Q_  3600 sec x Mv
                                                              0  e '   HR
                                                                    	LB.
                                                                    22400 x(460+5)  454 gm
                                                                             460+60
                                                             substituting for Q from equation (23), and simplifying:

                                                                           H  T    rvP"»   d        1
                                                                                                               .(24)
                                                                =
                                                                                                          = .  . .(25)
                                                   .(22)
  0   (460+T)  r    VL   (sp.gr)   TT V0.17678R"

When the value of E0 is obtained, Es may be calculated by the
procedure described previously.

                   NOMENCLATURE

Am c      Surface area, m = m2; c = cm2
C          Molar  concentration  of a  compound in  water

Cv         Gms compound as  gas divided by gms dry air
           containing the compound
Cw        Humidity content  of air = gms water/gmw dry air
           containing the water

D         Diffusivity coefficient of a compound in  air in
           m2/hour

d          Diffusivity coefficient of a compound in  air in
           cm2/second
E0 s       Evaporation rate of a compound in pounds/hour;
           o - initial, s = steady state computed from eqi*
           tion (12)
F          Feed rate, gal/hr =  discharge rate (in a lagoon)

-------
                                                                          AIR MONITORING    133
 G,GA     G = Kg/hr;GA = lb/hr

*H        gpm

 K        Amount of a compound in a feed stream, Ib/hr

 L        A lagoon dimension in meters

 Los      Pounds of a compound in a lagoon; o=at state,
          s=at steady state

 MwvA   Molecular weights  of w = water, v = compound,
          A = air

 P*        Mole-fraction of  compounds  in air at  lagoon
          surface

 Q        Gas volume, vol. of compounds emitted to air in
          6 seconds, cm^/sec

 sp.gr.     Specific gravity of a liquid waste

 V        Operating volume of a lagoon in gal

 V.P.      Vapor pressure of pure compounds, atm

 W0       Wind velocity m/sec

              F   F
 X        I-£.-±Q
              V  L

 y*       partial pressure in  atmospheres of a compound in
          air surrounding water droplets =
          LO.MW.V.P.
                                                IT

                                                F,,
           Weight percent  of compounds in water or lagoon
           liquid

           Time in sec

           3.1416

           Arnold's correction  factor, becomes 1 for dilute
           systems.
VL Mv   8.34(sp.gr)
REFERENCES

1. Nusselt, W., Heat Transfer, Diffusion and Evapora-
   tion, Technical Memorandum 1367, National Advisory
   Committee for Aeronautics, 1975.

2. Arnold,  J.H.,  "Studies  in Diffusion.  Ill Unsteady
   State Vaporization and Adsorption," Trans. Am. Inst.
   Chem. Engrs. 40, 1944, 361-379.

3. Schlessinger,  G.G., "Vapor Pressures, Critical Tem-
   peratures, and  Critical  Pressures  of Organic  Com-
   pounds." Handbook of Chemistry and Physics, 54th
   Ed. D-162toD-188.

4. Fuller, Atomic  and Structural Diffusion Volume In-
   crements. Perry 5th Edition (3-233) CRC.
                                        Appendix 1.
                   Vapor Pressure of Selected Compounds Based on Reference (3)
OCPDB
Ident.No.
90
110
240
380
790
810
890
930
1120
1244
1660
1670
1990
2360
2500
2510
2620
2640
2860
2910
3230
3349
3395
3411
3430
3541
Compound
Acetone
Acetonitrile
Amyl Acetate
Benzene
Carbon dlsulphlde
Carbon tetrachloride
Chlorobenzene
Chlorform
Cyclohexane
dichloroethane
Ethanl
Ethyl Acetate
di-Ethyl ether
iso-Propyl Alcohol
Methanol
Methyl Acetate
Methylene Chloride
Methyl ethyl ketone
Perchloroethylene
Phenol
Styrene
Toluene
1,1,1-trichloroe thane
1,1, 2-trichloroethylene
Freon
Xylene(s)
70°F
0.2229
0.0972
0.0056
0.0796
0.3813
0.1095
0.0094
0.1923
0.0924
0.2171
0.0578
0.0902
0.5214
0.0438
0.1233
0.2065
0.4773
0.1062
0.0192
0.0006
0.0083
0.0310
0.1331
0.0795
0.3280
0.0085
90°F
0.3586
0.1617
0.0108
0.1327
0.5817
0.1761
0.0165
0.3066
0.1504
0.3416
0.1056
0.1513
0.8032
0.0819
0.2155
0.3342
0.7646
0.1763
0.0341
0.0013
0.0151
0.0556
0.2192
0.1334
0.5108
0.0158
110°F
0.5580
0.2595
0.0199
0.2126
0.8614
0.2738
0.0278
0.4732
0.2366
0.5208
0.1848
0.2445
1.2006
0.1467
0.3622
0.5228
1.1849
0.2826
0.0582
0.0026
0.0264
0.0955
0.3485
0.2158
0.7790
0.0280
130°F
0.8426
0.4032
0.0348
0.3288
1.1386
0.4118
0.0452
0.7091
0.3609
0.7715
0.3113
0.3826
1.7461
0.2524
0.5879
0.7933
1.7826
0.4385
0.0957
0.0050
0.0444
0.1583
0.5369
0.3379
1.1315
0.0478
150°F
1.2383
0.6087
0.0594
0.4974
1.7487
0.6068
0.0716
1.0349
0.5355
1.1139
0.5069
0.5813

0.4191
0.9243
1.1714
2.6108
0.6612
0.1525
0.0091
0.0722
0.2538
0.8039
0.5138
0.6195
0.0787

-------
134     AIR MONITORING
                      APPENDIX 2

                   Diffusivity Coefficient

 The diffusivity coefficient  "d", is measured  in cm2/sec. In
 Arnold's equation, the diffusivity coefficient d is in cm2sec.

 In  Nusselt's equation, the diffusivity coefficient  "D" is meas-
 ured in m-/hr.
    D = 0.36d

 Fuller14' has derived values for d, using the relationship
   d.0.OOIT,,5(L    i
1   M2'        ->/
1     *•      cm "/sec
 D  values for a selected group  of  hazardous  compounds is
 presented for a range of temperatures in Appendix (2).
                         Table III
           Values of Diffusivity Coefficient D (mVhr)
        Compound
70 °F
90 °F
                        110 °F
Melhanol
Elhanol
Phenol
Acetone
M.E.K.
Methyl Acetate
Ethyl Acetate
Methyl Aminc
di-Melhyl Amine
Ethyl Amine
Hexane
Xylene
Toluene
Benzene
Di-Ethyl Ether
Melhylene Chloride
Chloroform
Carbon Telrachloride
FreonCCjF.CI.)
PCB(lcfilorine)
0.05687
0.04338
0.03000
0.03724
0.03231
0.03475
0.03067
0.05585
0.04282
0.04284
0.02837
0.02587
0.02827
0.03146
0.03159
0.03689
0.03203
0.02878
0.02578
0.02068
0.06069
0.04630
0.03199
0.03974
0.03448
0.03708
0.03273
0.05960
0.04572
0.04572
0.03027
0.02761
0.03017
0.03357
0.03371
0.03936
0.03418
0.03071
0.02751
0.02207
0.06460
0.04928
0.03405
0.04231
0.03671
0.03948
0.03484
0.06345
0.04867
0.04867
0.03222
0.02939
0.03212
0.03573
0.03589
0.04191
0.03639
0.03270
0.02929
0.02350
 Example 1

   To determine  the  steady  state emissions of methanol
 from a receiving lagoon.

 Basis for Model

   The  receiving lagoon has an operating volume of 600,000
     gallons.
   The feed rate is constant at 6667 gph
   The methanol concentration in the feed averages 0.09%.
   The feed weight is 8.54 Ib/gal
   The temperature is constant at 70°F
   The wind velocity is stead) at 4 m/sec.
                                    The  initial evaporation  rate  E0 will  be determined  usi
                                    Nusselt's modified equation;
                                                       % • (V.P.) • Wg.78  ..... (,2)
                                    E0 = 0.53452 Am •
                                    The the steady state evaporation rate Es will be determined
                                    from the relationship;

                                    Es=E0-i


                                    The parameters needed are:

                                    L0  =  the  pounds of methanol  in  the  lagoon at zero time
                              =  600,000 gallons X
                                                                                      gal
                                                                  X
                                                                                                    = 4503 Ib Methanol
                                    F   =  6667 gph

                                    V   =  600,000 gal

                                    L   =  40m
                                           4503 X 6667
                                                              K   =
                                             600,000
                                                       = 50 Ib/hr
Am =  0.7854 L2= 1256.64m2

W0 =  4 m/sec

%   =  0.09

V.P.=  0.12324 Atm(App 1)

D   =  0.05687,n2/hr @ 70°F
                                    Accordingly,

                                    E0  =  (0.53452X1256.64)(0.23638)(0.09(0.1234) (2.94854)
                                          = 5.193 Ib/hr Methanol.

                                    To determine the emissions at the steady state, Es, first deter-
                                    mine  Ls, the methanol  content of the lagoon at  the steady
                                    state

                                                                  K
                                    K  =  50 Ib
                                    !-X=
                                                            _
                                                V  4503   600,000
                                                                              -2
                           Ls  =  1.2264410-2  = 4076.7 Ib Methanol

                                         = 5.193 (4?Jf;?)  = 4.7 Ib/hr Methanol
                                    Es  =  E0

-------
                                                                                        AIR MONITORING    135
Example 2.

   Ammonia emissions from an aerated lagoon at the steady
state. An aeration lagoon is 45 ft in diameter, has an operating
volume (V) of 300,000 gal.

   A feed stream at 10.5 pH, containing 0.1% NH3 enters the
lagoon at a rate (F) of 210 gph.

   Mass of ammonia in the lagoon at zero time :
   Mass of ammonia entering the lagoon each hour:

 •v   -m  , v 8.34 lbv 0.1 IbNH?   ,,,,,,,,
 is K = 210 gal X  _,   X   , ----- •* = 1.7514 Ib.
                gal
                         lOOlb
  An aeration pump transfers 3390 gpm of lagoon water to a
deflector which produces an umbrella-like spray cover which
completely covers the lagoon.

  The flight time of droplet in seconds will be;
              V
                 lagoon diameter in feet
                        5.6568
   In this  case,  the lagoon diameter is 45 ft, therefore flight
 time (G) 2t = 1 . 1 86 seconds

   Assume the  water droplets  (r) average 0.25 'cms radius.
 The  weight of  water suspended as droplets at any time is:
  min
gal
                    Ib
                          60 sec
   Total surface of the droplets, Ac = 253720
 X 4?rr2 (droplet area)
  _4_
  37rr3 (droplet wt.)

                   = 253720 X— cm2
                              r
 when r = 0.25 cm Ac = 3,044,642 cm2

 The vapor pressure (V.P.) of pure ammonia (NH3) at 70 °F
 is 8. 55 13 Atmospheres

 The molar concentration of 0.1% ammonia is 1.06X 10~~3
                                                               y* = (molar concentration)(vapor pressure) = 9.05x10 3

                                                           The diffusivity coefficient (d) of ammonia from water to air
                                                           at 70°F = 0.23785 cm2/sec.

                                                           The volume of ammonia gas entering the atmosphere from the

                                                           aeration lagoon V cm3 = 2Acy*l /   according to Arnold.


                                                           V  =  16513cm3/!.186 sec

                                                           Amount of ammonia emitted, Ib/hr:

                                                                            3    mole

                                                     ,=16513cr

                                                        3600 sec
                                                          hr
                                                                            X
                                                                                 1
                                                                              1.186 sec
                                                                                           moeH
                                                                                      = 83.8 Ib Ammonia/Hr.
                                                            To determine the emission rate of ammonia at the steady state
                                                            proceed as follows:
                                                            L0  = 2502 Ib
                                                            Ls  = Ib ammonia in the lagoon at the steady state
                                                            T   =
                                                             s
                                                        K
                                                           K  =  1.7514 Ib

                                                           1-K=  *&+L

                                                                  83.8 .
                                                                  2502  300,000
                                                                                   0.0342
                                                           -     0.0342= 5L211b

                                                           Emissions at the steady state

                                                           Es  =  EoJ ii =1.715 Ib/hr.

                                                           Steady state emissions of ammonia is 1.75 Ib/hr.

                                                              Thus, the  ammonia concentration  leaving the  lagoon  in
                                                           the liquid, at the steady state is

                                                                         L^=_^	  =2X10-5
                                                                      (8.34)V   (8.34)(300,000)
                                                                              = 20 mg/1

-------
        AIR MONITORING OF A HAZARDOUS WASTE SITE
                                          DAVID A. SULLIVAN
                                          JEROME B. STRAUSS
                                                Versar Inc.
                                           Springfield, Virginia
INTRODUCTION

  Versar Inc., as a contractor to the U.S. Environmental
Protection Agency for air quality  emergency response,
performs ambient monitoring, dispersion modeling and
other support functions as requested by  the Division  of
Stationary Source Enforcement. In this capacity, Versar
and  GEOMET, Inc.  (prime  contractor)  responded  to a
request to perform ambient air sampling at Rollins Envir-
onmental Services' (RES) waste treatment and  disposal
facility near Baton Rouge, Louisiana. This paper focuses
on Versar's  approach to the identification and quantifi-
cation of airborne pollutants from a hazardous  waste
facility.
  Community  residents  had  complained of odors and
respiratory impairment which they attributed to  the RES
waste treatment and disposal facility. To determine  what
action could be supported by air quality impacts, ambient
air concentrations were needed for hazardous pollutants
emitted from the site. Sampling and analysis were compli-
cated by the wide range of compounds which are brought
to RES for treatment and/or disposal. It was anticipated
that emissions  would be variable and that reactions could
occur among components of the waste in liquid or solid
form, or among gaseous pollutants emitted  from the  fa-
cility. Since the airborne pollutants were not well identi-
fied  at the outset of this program, a two-phase approach
was  designed to identify and quantify organic air pol-
lutants. The study was intended to provide a quick  re-
sponse assessment of airborne emissions from the  facil-
ity,  as  opposed to providing a long-term  ambient  air
quality evaluation.
  Using  solid  adsorption columns composed of Tenax
or activated  charcoal for sampling, and using gas chroma-
tography (GC) and gas chromatography/mass spectrome-
try  for analysis (GC/MS), the Phase I program was  in-
tended  to identify pollutants that were present in  the
highest  concentrations. RES management placed restric-
tions on sampling which allowed  Versar to collect only
around  the  perimeter of the  site (i.e., "fenceline" sam-
ples).
  Upwind samples did not contain detectable concentra-
tions of any pollutants. Therefore, downwind concentra-
tions presented  in this  report also  reflect  downwind
minus upwind concentrations.
  Benzene and toluene were  present  in the highest con-
centrations  and  were quantified during Phase  I of  the
project. The maximum concentrations at the fenceline
were 0.5 and 2.4 ppm, for benzene  and toluene, re-
spectively. Compounds that were only tentatively identi-
fied  and therefore  not quantitied,  included ethyl ben-
zene, 1,1,1-trichloroethane,  methylene  chloride, tetra-
chloroethylene, chloroform,  and meta-xylene. Since the
impact to the general public was an important considera-
tion  for this study, fenceline sampling did allow for a
suitable data base to determine impacts from the facility.
  The Phase II program was designed to provide quanti-
tative air quality data and utilized a portable GC and cali-
bration standards for benzene, toluene and the tentatively
identified compounds of Phase I. The use of a portable
GC greatly assisted in the selection of areas of expected
maximum concentrations  and was  useful  in providing
feedback on appropriate sampling times. Maximum con-
centrations were again  found for benzene  and  toluene,
with maximum fenceline values of 0.2 and  0.4 ppm, re-
spectively.   Tentative   identification  included   several
compounds  that were not observed during Phase I, in-
cluding pentenenitrile and methyl pentene.

DESCRIPTION OF RES FACILITY

  The  RES  facility is  located 12 miles  north of Baton
Rouge on 165 acres of land. It is designed to treat and/or
dispose of liquid and solid wastes generated by area in-
dustries, such as from chemical, petrochemical, refining
and  synthetic polymers plants.  Treatment  and  disposal
practices include landfarming, landfilling,  incineration,
biological treatment and physical/chemical treatment.
  Area residents had filed numerous complaints alleging
that  activities at RES were creating an odor problem and
were adversely affecting the health of  the community. It
appears that many of the complaints were associated with
odors from the biotreatment area and the landfarm area.
  Prior to  the  Phase  I program,  however, RES had
changed to  a subsurface injection system for landfarm
application,  which reduced odors and  organic emissions;
also, the emissions from the biotreatment area were sub-
stantially reduced to meet a State-issued order. Modifi-
cations to the facility between Phase  I and II sampling
included closing  a number  of  landfill pits to reduce
odors, as requested by the State.
  The data collected during the sampling programs were
considered to accurately represent ambient  conditions li
the time they were obtained, (Phase I—September 25-26,
                                                     136

-------
                                                                                    AIR MONITORING     137
1980,  and Phase II—November  12-13, 1980). Versar's
experience with modifications of the type used at the fa-
cility would suggest that the ambient concentrations had
been substantially higher before the corrective action had
been taken during the summer  and fall of  1980. It is
therefore possible that complaints about the facility prior
to these modifications were the result of higher concen-
trations than those observed during the sampling program.
  By  the time  the  Phase II  program began, the most
distinctive and  unpleasant odors were found to  be  as-
sociated with  the mixing pit rather than the landfarm or
biotreatment  areas. The  mixing pit combines ash with
liquid wastes  as a treatment method.  Gaseous and par-
ticulate emissions are emitted from this source.

SAMPLING
Air Quality
   Phase I of the two-phase sampling program involved
collecting 46 air and eight liquid samples. Activated char-
coal and Tenax were each used to collect 23 air samples.
The height of  the  intake in  all samples was approxi-
mately four feet, to represent the approximate breathing
level. In Phase  I, the air sample volumes through the ad-
sorption columns were from 1 to 150 liters. During  this
phase, the strongest odors were detected downwind of
the mixing pit and the biotreatment area.
   These areas were evaluated by selecting upwind  and
downwind sampling sites, where concurrent upwind  and
downwind sampling was performed. As previously stated,
detectable concentrations were not  found in  any of the
upwind samples. Phase I was characterized by fairly vari-
able winds. The highest concentrations and  the  largest
variety of compounds were generally observed downwind
of the mixing  pit.  The sampling sites for Phase I  are
shown in Figure 1.
   In  Phase II, the same  basic  sampling  approach  was
used as in Phase I, but feedback was obtained on site se-
lection and optimum sample volumes through the use of
an on-site Hewlett Packard 5880 dual channel GC. Sample
volumes were increased over Phase I since breakthrough of
pollutants into  the back  sections of the adsorption  col-
umns did not occur.  Volumes in the  range  of  150-400
liters were obtained during Phase II. Wind flow was fairly
consistently out of the east through east-southeast during
Phase II, which was the optimum wind flow  to evaluate
fenceline concentrations downwind of  the mixing pit  and
biotreatment area. The location of the sampling sites for
Phase II is shown in Figure 2.
   Sampling was performed from 3:00 to 6:00 a.m. in order
to  coincide  with minimum dispersion conditions  and
therefore  relatively  high ambient  air concentrations.
Afternoon sampling was also performed to coincide with
maximum activity periods. The  early  morning hours of
November 12, 1980, were particularly  suitable for samp-
ling since stable  conditions (relatively poor  dispersion)
occurred with clear skies and light but reasonably steady
winds, 3.5 to 5.0 mi/hr between 1:00 and 6:00 a.m.
   The sampling  protocol was  essentially the same  for
Phases I and  II, with the exception that a GC was not in
the field during Phase I. The basic protocol was as fol-
lows:
  (1)  Based on current wind direction  data, downwind
      and upwind sites were selected.
  (2)  A steel support post was used to provide a sample
      intake height of approximately four feet.
  (3)  The seals on the adsorption tube were broken, and
      the tube was attached to the end of approximately
      one foot of neoprene tubing. The pump and tub-
      ing were downstream of the sample.
  (4)  Sampling times generally were two hours; however,
      some samples were taken for as long as four hours.
  (5)  The sample was stored in the GC trailer at 4°C un-
      til  it was  analyzed.  Ice was contained in sealed
      plastic bags to minimize sample exposure to mois-
      ture.
                       Figure 1.
              Air Sample Locations—Phase I
                        Figure 2.
            RES Phase II Air Sampling Locations
 (All Sampling Sites Were Within Fenceline, Except 15,16 and 23)

-------
138     AIR MONITORING
Blanks were taken for quality control purposes by break-
ing the seals on the glass adsorption tubes, capping and
then handling the same as the samples. No contamination
was found during either Phase I or II.
  The information associated with each sample including
the sampling period,  sample volume, site location and
other  specifications associated with  each sample  is sum-
marized in Tables  I and II. During Phase I, the results
of the analyses indicated that the charcoal and Tenax ad-
sorption tubes  provides  similar results. During Phase  II
most of the samples were collected on charcoal.

Liquid Samples

  Liquid samples were taken at eight holding pits during
Phase I to assist in the evaluation of emissions from this
facility. GC qualitative analysis of the volatiles from these
samples did not indicate any compounds  different from
those  identified or tentatively identified during  the air
sampling program. For  selecting appropriate adsorption

                         Table I.
                Sample Log Summary, Phase I
      Sample
      Number


       5
       6
       7
       9
       10
       II

       12

       17

       IB
       19
       20

       21
       22
       23
       24

       27
       28
       29

       30

       31
       32
      33
      34

      39
      40
      4!

      3?
      36
      37

      38

      47
      48
      49
      50
     $topc>*3 b»tt*r»
 Pu«p
Number
 Charcoal
 Oarcoal
 Tenax
 Tenax

 Charcoal
 Charcoal
 Tenax
 Tenax

 Tenax
 Tenax
 Charcoal
 Charcoal

 Charcoal
 Charcoal
 Ten ax
 Tenax

 Charcoal
 Charcoal
 Tenax
 Tenax

 Charcoal
 Charcoal
 Tenax
 Tenax

 Charcoal
 Charcoal
 Tenax
 Tenax

 Tenax
 Tenax
 Charcoal
 Charcoal

 Tenax
 Charcoal
 Charcoal

 Charcoal
 Charcoal
 Tenax
 Tenax

 Tenax
 Tenox
 Charcoal
Charcoal

Tenax
Tenax
Charcoa I
Charcoal
Flo. Volume
 (liters!


   51.7

   1.88
   98.7
   11.3

   11.6
  143.0

   2.65
  136.0

   14.4
  126.0
   66.0
   2.40

   9.86
  122.0
  116.0
   2.26

   1.29
   52.8
   95.7
   5.61

   1.36
   37.4

   71.4
   8.16

   2.40


  126.0
   10.2

   16.8
  140.0
  147.0
   2.73

   3.28
   13.9
  1 14.8

   2.34

  126.0
  120.0

   14.3

   10.5
  130.0
  68.2
   2.48

  16.3
 136.0

 143.0

   2.65
                            Start
                            Tine
                           3:40p«
                                  4: 27pm
                                              3:l5p»  4:23p«
                         9/25/80
                                              4:42p«  5:40p«
                                 6:00p»   6:33pn
                                 6:04pn   6:38p»
                                   5e>   7:45a>i
                                 7:50ae   9:!2a>
tubes and to aid in the selection of calibration standards
for field  analysis,  liquid samples can be a useful prelim-
inary step in an overall evaluation. By performing GC/MS
and/or GC analysis  on the volatile  gases  of the liquid
samples,  an  indication of probable  airborne pollutants
can be made.

Meteorological Monitoring

  A portable wind system was set up at the same, well-
exposed location for  both phases, as shown in Figures 1
and 2. This station was expected to represent general flow
for the  entire  facility  except for  conditions associated
with light winds (e.g., less than two miles per hour.) The
terrain is quite flat. The wind vane was oriented by com-
pass readings to the nearest degree, and the reduced data
corrected for the 4  degree east magnetic declination.
  The height of the wind  vane and anemometer was ap-
proximately eight  feet above ground  level which is rela-
tively  low  for monitoring  wind  data.  Although wind
speeds were relatively low and wind flow was somewhat
more variable than at the  more standard ten meter level,
sufficient documentation for general flow was obtained.
  Relative humidity values were obtained from wet bulb
and dry bulb temperature readings from a sling physchro-
meter. On  the sample log sheets, cloud cover  data and
weather events were recorded to provide an indication of
atmosphere stability and  possible  interference from pre-
cipitation . These data are documented in Tables HI and IV.

ANALYSIS

  The procedures  followed  in analyzing the samples  in
the field  (Phase II) and in the Versar analytical labora-
tory are discussed  in  the following section.  This descrip-
tion  pertains specifically to Phase II;  however, similar
laboratory procedures were used for Phase I.
GC Analysis
  GC analysis in  the field was performed  using an HP
5880  gas chromatograph equipped with a flame ioniza-
tion  detector. GC analysis  in  the analytical laboratory
was  performed  using an  HP 5700 gas chromatograph
equipped with a flame ionization detector. GC analytical
conditions were as  follows:

Gas Chromatograph
Detector
Column
Carrier Gas
Hydrogen
Hydrogen
Air
Oven Temp., initial
program rate
final
Field
HP 5880
FID
6.!ms.s. 10WSP-1000
30 ml/min N2
30 ml/min
30 ml/min
350 ml/min
100°Cfor8min
4°C/min
200°Cibrl6min
Laboratory
HP 5700
FID
6.1ms.s. 10WSP-1000
30 ml/min Nj
30 ml/min
30 ml/min
250 ml/min
105°Cfor8mto
4°C/min
200"Cforl6min
                                                 Charcoal  samples were analyzed  using NIOSH Method
                                                 P + CAM 127  as outlined in  the NIOSH  Manual of
                                                 Analytical Methods.  The  use of carbon disulfide as the

-------
                                                                                   AIR MONITORING    139
                       Table II.
             Sample Log Summary, Phase II*
                       Table III.
               Meteorological Data, Phase I
                  September 25-26,1981


Sample No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

16

17
IB
19


Pump No.
4806
4815
5934
4806
4815
5934
5949
4816
5195
5942
5156
5942
5949
5195
4816

5156

5934
5156
5949

Start Time
(CST)
0447
0436
0434
0643
0630
0637
1410
1426
1416
1400
1440
1606
1619
1627
1610

1617

0412
0422
0356

Stop Time
(CST)
0641
0629
0625
0841
0831
0830
1618
1609
1625
1604
1616
1753
1802
1806
1757

1805

0547
0550
—


Operator
Sul I Ivan
Sul llvan
Feldman
Sul llvan
Sul 1 Ivan
Felthan
Koch/AI len
Su 1 1 1 van
Koch/AI len
Koch/AI len
Sul 1 Ivan
Al len
Al len
Al len
Koch/
Su 1 1 1 van
Koch/
Sullivan
Koch
Koch
Al len


Date
11/12
11/12
11/12
11/12
11/12
11/12
11/12
11/12
11/12
11/12
11/12
11/12
11/12
11/12
11/12

11/12

11/13
11/13
11/13
Site
Number/
Comments
Site *I3
Site f!4
Site *15
Site *13
Site 114
Site /16
Site f!8
Sits *19
Site (21
Site 116
Site *20
Site »16
Site f18
Site *21
Site 119

Site »20

Site »20
SI1B <22
Site *16,
Samp le"*
Volume
(liters)
199
210
207
206
225
210
271
176
273
202
188
174
219
211
183

211

175
176
—
pump stopped
20
21
22
23
5942
4816
5942
4815
0406
0611
0557
0404
0556
0734
0727
0615
Al len
Al len
Al len
Kbch
1 1/13
11/13
11/13
11/13
Site »23
Site »16
Site «3
Site *17
180
146
150
244
Date
9/25


9/26


Time
(cst)
1600
1700
1800
0700
0900
1000
Wind
Speed
(mph)
4
4
5
6
9
8
Average
Wind
Direction
(degrees)
250 (variable)
200 (variable)
040
050
055
055
Relative
Humidity
(§)
	
66
70
91
88
75
                                                         GC/MS Analysis
                                                           Prior to the qualitative  analysis of the air samples
                                                         collected on  Tenax  and charcoal  tubes,  the  GC/MS
                                                         calibration was verified by a direct injection of para-
                                                         bromo-fluorobenzene and the characteristic ions and rela-
                                                         tive ion abundances were verified. For samples collected
                                                         on charcoal, a 2-microliter injection of the methanol char-
                                                         coal  tube extract was  chromatographed into  the  mass
                                                         spectrometer. For samples collected on Tenax, the Tenax
                                                         tube was thermally desorbed at 180°C for 4 minutes using
                                                         Matheson purity  helium (99.9999%) to purge the de-
                                                         sorbed components  onto the GC  column. All samples
                                                         were chromatographed  and analyzed using the following
30
31
32
33
34t
35t
36t
37
38
39
40
* Except
tubes.
** Correc
t Tenax
4816
5156
5949
5942
4606
5195
4815
5156
4816
5942
5934
where Indicated

ted to standard t
sample ( low f low)
1741
1738
1754
1605
1934
1935
1945
2121
2132
2149
2218
as a Tenax

emperature
.
2132
2120
—
2143
2234
2235
2244
0015
0018
0034
0000
sample, al

and pressu

Sullivan
Sul llvan
Su 1 1 1 van
Su 1 1 1 van
Al len
Al len
Al len
Koch
Koch
Koch
Allen
1 samp les H

re.

11/13
11/13
11/13
11/13
11/13
11/13
11/13
11/13
11/13
11/13
11/13
Site
Site
Site
pump
Site
Site
Site
Site
Site
Site
Site
Site
'ere col lected






120
124
(18,
stopped
*17
124
124
f18
124
120
125
123
427
469
—
371
109
397
331
367
307
261
195
on charcoal






GC/MS conditions:
Column

Oven Temp.


Scan Range



.initial
initial hold
program rate
final


Ion Source Voltags
Scan Rate

1 % SP-1000 on 60/80 Carbopack B 2.44 m x 2
mm ID glass
45 °C
t min
j mm.
8"C/min
220 °C
34-257 AMU

70 eV
2.7 seconds
Electron Multiplier
Voltage

Run Time



2400V

35 minutes
desorption solvent unfortunately precluded its quantifica-
tion as a possible contaminant.
  During Phase II, three compounds were chosen as in-
dicators for ambient onsite analysis. These were benzene,
toluene and meta-xylene. Work in Phase I had identified
these  as compounds  of interest to be quantified onsite.
All Phase II samples were screened by GC to  determine
those  with the highest concentrations.  The three samples
with the highest concentrations were selected for GC/MS
analysis to qualitatively identify a  wide range  of pollut-
ants. After GC/MS analysis, GC analysis was then used to
quantify the identified pollutants.
  Standards  were prepared in carbon disulfide for those
identified components for which reference materials were
available.  These were benzene, toluene,  ethyl benzene,
1,1,1-tetrochloroethane, methylene chloride, tetrachloro-
ethylene, chloroform and meta-xylene.  Ambient air con-
centrations were calculated from the amount measured in
the extract and the volume of air sampled.
  Background-subtracted spectra were obtained for all
peaks and  compared  to in-house reference spectra and
spectra obtained from EPA/NIH/MSDC Library Search
System. Confirmation of identified  compounds was ob-
tained by analysis of standard reference materials under
the same conditions as the samples. The standards and
samples  were matched  for retention times  and  mass
spectra.
  The following criteria are the minimum  requirements
that were used for the interpretation of the mass spectra
to confirm identification of any pollutant:
•The mass spectra contained  all ions present above 10
 percent relative abundance in the mass spectrum of the
 reference with the general agreement of ±20%  of the
 relative abundance in the reference mass spectrum.
•Ions present in the experimental mass spectrum that are
 not present  in the reference spectrum must  not  exceed
 10% of the total ion abundance in the experimental mass
 spectrum.

-------
 140     AIR MONITORING
                                                           Table IV.
                                            RES Onsile Meteorological Data, Phase II
                     November 12,1980
     Sunny skies in the morning, 3/10 cloud cover, high cloud (thin)
             becoming 8/10 high cloud cover by sunset.
                         Average                Relative
                         Wind Dir.-   Range"    Humidity
                         (Degrees)
                            085
                            075
                            065
                            055
                            055
                            060
                            060
                            045
                            055
                            070
                            065
                            060
                            065
                            050
                            055
                            100
                            095
                            050
Time
(CSD
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
Wind Speed
(mph)
   3.5
   4.0
   5.0
   4.5
   4.5
   5.0
   6.5
   7.0
   7.5
  10.0
   9.0
   9.0
   9.0
   7.0
   6.5
   2.5
   2.5
   2.5
  Calm
  Calm
   2.5
   3.0
   2.5
   2.5
                            035
                            055
                            040
                            040
Range"
(Degrees)
   70
   60
   40
   40
   40
   50
   70
   60
   55
   65
   65
   75
   65
   80
   85
   60
   20
   75
   65
   55
   75
   60
67
44
32
26
41
53
58
 ' Wind direction was corrected for magnetic declination of 4° East, associated with compass
  orientation of wind vane. Wind speed and wind direction averages are based on the period
  60 minutes preceding the indicated hour.
 "The second highest maximum and second lowest minimum were determined and  used to
  estimate the range or the wind direction for the period. This provides a rough indication of
  (he horizontal variability or wind direction which is a function of atmospheric stability.
 —Data not available.
                                                                            November 13 and 14,1980
                                                           9/10 high cloud cover in the morning becoming overcast by afternoon;
                                                                 trace of precipitation occurred between 1900-2000 CST
                                                                                   Average
Time
(CST)
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
Wind Speed
(mph)
4.5
5.0
5.0
5.5
5.5
5.0
4.5
5.0
7.0
7.5
8.0
8.0
12.5
12.0
12.0
8.0
9.0
9.5
9.0
8.0
9.0
12.5
13.0
10.0
Wind Dir.*
(Degrees)
070
065
070
085
075
075
090
055
085
090
080
075
105
110
110
110
105
105
095
095
105
110
110
095
Range**
(Degrees)
55
50
50
50
70
55
50
80
70
60
55
75
55
35
70
60
55
40
40
40
50
40
35
40
R.H.
<*)
—
—
59
...
56
_
—
56
--
._
...
...
...
...
—
.„
...
77
...
86
...
82
86
81
                                                                       0100
                                                                          7.0
                                                                              November 14,1980
                                                                              095         50
 •The  retention  time of the  experimental  mass  spectrum
  must be within ± 5 scans or ± 30 seconds of the reference
  compound.
   Detection limits were 10 mg/1 in solution, and all com-
 ponents that did  not meet  these criteria  are considered
 tentative identification.
  Two charcoal tube samples (Nos.  9 and 38) and one
Tenax tube sample  (No.  35) were  analyzed  by  GC/MS
for  identification  of  major organic  compounds.  The
presence  of methylene  chloride, acetone,  carbon disul-
fide,  chloroform,   1,2-dichloroethane,   1,1,1-trichloro-
ethane, benzene, hexane,  toluene and ethyl benzene was
confirmed by mass spectra in the Tenax sample (No. 35).
Other trace components that were tentatively identified by
library search were:
 2-pentenenitrile                 ethane, 1,2-dichloro
 acetone                          isopropyl ether
 isopropyl alcohol                1-pentene,  2-methyl
 tert-butanol                     2-pentenenitrile
 4-penten-2-ol                    hexane
cyclopentanone, o-methyloxine  butanoic acid, 3-3 dimethyl
 hexanedinitrile                  2-butanone
                                                          Also,  tentatively  identified by  in-house  spectra  were
                                                          tetrachloroethylene and trichlorotrifluoroethane.  Due to
                                                          the desorption  procedures used for  GC/MS analysis,
                                                          quantification of these  samples (9,  35 and 38) was not
                                                          feasible by GC or GC/MS.
                                                                                   Table V.
                                                            Concentrations of Benzene and Toluene in Air as Calculated
                                                                    from the Charcoal Tube Extracts, Phase I
Site
2
2
4
4
8
10
Field No.
1
2
9
10
33
36
Laboratory
Sample
Number
6171
6172
6175
6176
6203
6206
Benzene
(ppm)
0.05
0.05
0.03
0.02
0.004
0.007
TLV = 10
TolncM
(ppm)
2.3
2.4
1.4
1.0
0.02
0.05
TLV-MO

-------
                                                    Table VI.
                                Phase II  GC Results Summary (Concentrations in ppm)*
                                                                                     AIR MONITORING    141

1 13 .006 -006
2 14 .056 .021 .084
3 15 (UPWIND SAMPLE )
4 13 .003 .001
5 14
6 16 ( UPWIND SAMPLE )
7 IB
8 19 .004
10 16 ( UPWIND SAMPLE )
11 20 .089 .240
12
13 IB .060 .063 .055 .059
14 21 .012 .034
15 19 .057 .052 .0132
TLV** 10 200 100
16 20 .215 .428
17 20
IB 22 .031 .108
20 23 (UPWIND SAMPLE )
21 16 ( UPWIND SAMPLE)
22 23 (UPWIND SAMPLE)
23 17
31 24
33 17 .016 .030 .022 .026
37 24
39 25
40 23 (UPWIND SAMPLE >

GC Run Onstte
.019 .241 GC Run Onslte
No significant
peaks
GC Run Onslte
No significant
peaks
GC Run Onslte
No significant
peaks
GC Run Onstte
Early peak
GC Run Ons Its
GC Run Onslte
No significant
peaks
.01 1 GC Run Onslte
No significant
peaks
.064
.020 GC Run Onslte
.042 .138
100 100 50
.026 GC Run Onstte
No significant
peaks
GC Run Onslte
GC Run Onslte
No significant
peaks
No significant
peaks
No significant
peaks
No significant
peaks
GC Run Onslte
No significant
peaks
.015 .039
No significant
peaks
GC Run Onslte
Some smal 1 peaks
No significant
peaks
                   ' Blanks Indicate that pollutant concentrations
                   h Threshold limit value.
                                                           100

                                               i below detection limits.
RESULTS AND CONCLUSIONS

  The quantitative data obtained during the Phase I and
II  programs,  are  summarized in  Tables V and VI, re-
spectively. For each compound,  threshold limit values
(TLVs) are provided only to indicate a point of compari-
son. However, since TLVs are associated with an eight-
hour average applicable to the workplace, they are  not
suitable for exposure assessments  of the general popula-
tion.  Although  the  observed concentrations  for  which
TLVs were in place were substantially below these values,
it is beyond the  scope of this paper to make any evalua-
tion regarding health effects associated with exposure of
the general public to these concentrations.
  For both the  Phase  I and Phase II air sampling pro-
grams at  Rollins Environmental Services, many organic
compounds were identified by GC and GC/MS analyses.
Samples upwind of RES did not indicate significant con-

-------
142    AIR MONITORING
centrations of organic pollutants and odors were generally
not noticeable at the upwind sampling sites. Benzene and
toluene were usually found in the highest concentrations
in the downwind samples. During Phase I,  the  highest
values for  benzene and toluene were found  to be 0.05
ppm and 2.4 ppm, respectively. Phase II sampling indi-
cated a maximum of 0.2 ppm for benzene and 0.4 ppm
for  toluene. A  number  of  organic  compounds  were
present, but in concentrations not high enough to allow
for  positive  identification.  For example, pentenenitrile
which could possible be toxic at relatively low levels, was
tentatively identified.
   To perform a more detailed identification  and  quanti-
fication of the  pollutants emitted from this facility, a
longer-term program would be necessary. A sampling pro-
gram during the summer quarter, when the highest emis-
sion of volatile organics could be expected, may result in
detection of substantially greater pollutant concentrations.
Caution must be exercised in drawing conclusions from a
short-term program such  as performed during Phases I
and II of  the  study.  Although in Phase II, conditions
were such  that  relatively  high concentrations would be
expected (i.e., sampling was performed  from  three to six
a.m. with fairly light winds and clear skies), a short-term
sampling  program would not be  expected  to  contain
values near the annual maximums. The variability of waste
being treated and disposed  of is a factor that makes a
long-term program more meaningful.
   The RES incinerator was  not operational during either
Phase I or Phase II of the Versar sampling program. This
could explain some differences between these results and
those of a  previously performed short-term program. Al-
though that program  produced results that were generally
of the same order of magnitude as our study  for benzene
and toluene,01 one sample from the previous program in-
dicated a fenceline concentration of 6.1 ppm for benzene,
which is  substantially higher  than  any concentrations
measured during our program. In addition, the qualitative
identifications of other compounds  were different from
those observed during this study.
   Although short-term sampling programs have limita-
tions, the results provide an indication of ambient concen-
trations to evaluate the need to collect a more long-term
data base or to take enforcement action. To complete the
evaluation,  a health effects  review of the data obtained
from the  sampling programs would  be necessary. This
will   be  performed   by  the  Environmental   Protection
Agency (Region VI). At the time of this writing, the health
effects review had not been completed.
GENERAL RECOMMENDATIONS FOR
RELATED PROGRAMS

  (1) Perform analysis of volatiles from liquid samples, if
      possible.
  (2) Choose adsorption medium on the basis of pollut-
      ants anticipated to be present.|4i5-6>
  (3) Use screening techniques to choose areas of maxi-
      mum expected concentrations, such as portable Gas
      Chromatography equipment with a flame ionization
      detector to measure total hydrocarbons, obtaining
      grab samples in collection bulbs in conjunction with
      an onsite Gas Chromotrograph, etc.
  (4) Sample collection periods should be chosen to en-
      sure that a sufficient quantity of sample is obtained
      to allow for identification  and quantification, but
      not  so  large  that  saturation of the  absorption
      medium occurs and breakthrough problems  result
      or that the collection  efficiency is significantly re-
      duced by high flow rates.(4|5)
  (5)  Calibration of portable pumps  used to draw the
      samples should be performed at least  on a daily
      basis.
  (6) Proper handling,  such as refrigeration of samples,
      chain-of-custody documentation, etc., are essential
      to ensure defensibility of the data collected.
  (7) The use of an onsite wind system is highly recom-
      mended in lieu  of the use of offsite data sources.
      Upwind and downwind  samples  should always be
      taken concurrently with properly  documented wind
      data. Cloud cover and weather events should also be
      documented.
  (8) The field log  should include an  indication of site
      activities associated with the sampling periods (e.g.,
      incinerator  operation, unloading  of waste,  truck
      traffic, landfarm activity, etc.)
  (9) If possible, sampling periods should include  those
      associated  with  maximum anticipated emission
      terms  and poor dispersion conditions, e.g., high
      activity  periods, maximum  ambient temperature,
      poor dispersion conditions, wind flow toward sen-
      sitive receptors.

REFERENCES

1.  "Phase I  Air  Monitoring Program/Rollins Environ-
   mental Services," Versar  Inc., Subcontract No.  6156-
   12, November 4, 1980.
2.  "Phase II Air Monitoring Program/Rollins Environ-
   mental Services," Versar  Inc., Subcontract No.  6156-
   12, January 19,19981.
3.  "Analysis Report, Rollins  Environmental Services,"
   Contract  No. 21660-80-01,  Enviro-Med Laboratories,
   Inc., Baton Rouge, Louisiana, May 27, 1980.
4.  "Characterization of Sorbent Resins for Use in Envir-
   onmental   Sampling,"   EPA-600/7-78-054,  March
   1978, IERL/RTP.
5.  "Selection  and Evaluation  of Sorbent  Resins for the
   Collection of Organic  Compounds," EPA-600/7-77-
   044, April 1977,  IERL/RTP.
6.  Katz, M., ed., "Methods of Air Sampling and Analy-
   sis,"  American  Public  Health  Association,   Inter-
   society Committee, Byrd  Pre-Press, Inc., Springfield,
   Va., 1977.

-------
            INFLUENCE OF SAMPLING TECHNIQUES ON
                 ORGANIC WATER QUALITY ANALYSES
                                        ARTHUR M. SEANOR
                                       LARRY K. BRANNAKA
                                            Dames & Moore
                                        Baldwinsville, New York
INTRODUCTION

  In the past, the role of a groundwater geologist was
generally confined to the development of techniques to
obtain or remove  sufficient quantities of water to satisfy
the objectives of  a particular project. However, the re-
cent inclusion of approximately 113  organic compounds
on the priority pollutant lists has opened a new concern
for groundwater geologists.  This concern directly relates
to the problem of obtaining a sample  of groundwater that
one can be confident is truly representative of the sub-
surface  fluids.  The  need to  supplement the  existing
knowledge currently available pertaining to the fate and
movement of organic constituents in the subsurface has
been documented by several authors.(1'2)
  The presence of many trace organics in a groundwater
supply system even in very low concentrations may cre-
ate near-term or long-term health hazards. The complexity
of these chemicals has  further confused attempts at un-
derstanding the fate of most trace organics once they enter
the subsurface.  Many times compounds not produced or
known in a facility's waste stream have been identified in
samples of groundwater collected from a particular fa-
cility.
  Because of the potential acute hazards posed  by many of
the trace organic compounds (even in the ug/1 range), the
techniques  for  obtaining and  storing of groundwater
samples containing  these compounds may  differ  sub-
stantially from  classical field sampling techniques. Even
with the vast amount of research currently underway with
respect to sampling of  organics, the state-of-the-art for
obtaining samples  still lags far behind  that of the analytical
laboratory techniques.
  Because the  advanced laboratory techniques are so
costly and labor consuming, the necessity  for accurate,
cost effective sampling  techniques becomes more appar-
ent. The money and time involved in analyzing a suite of
samples for trace organics can be completely wasted if the
samples were incorrectly samples or improperly stored.


AREAS OF SPECIFIC
ORGANIC CONTAMINANTS

  During a series  of recent groundwater investigations,
Dames & Moore has attempted to identify potential prob-
lem areas in order to improve the efficiency of obtaining
and storing samples. Five major areas of concern were
identified whereby specific contaminants may be allowed
to either enter the groundwater or contaminate the samples
before analyses can be instituted.  The areas  of concern
addressed include the following:
•The drilling operations
•Piezometer or well construction materials
•Sampling equipment
•Sampling methods
•Sample preservation
Each of these areas of concern are addressed in more de-
tail in the following sections of this paper.

Drilling
  Generally the drilling of a well is required to obtain
groundwater samples for analyses. A well should be in-
stalled wherever possible  to  minimize the potential for
organic contamination. However,  this  one  operation,
which may in fact be the  most  costly part of a ground-
water  study, has the potential to introduce  the greatest
error!  Additionally,  if error is considered, then it can be
very costly from an economical and labor standpoint to
redrill a new well.  During  a drilling operation several op-
erations may lead, directly or indirectly, to the existence
of contaminants within the groundwater.
  Where depths in excess  of about 50 to 75 ft are neces-
ary to drill, rotary  drilling or  percussion drilling  tech-
niques are normally employed.  Rotary drilling is gen-
erally  favored over percussion drilling because it is much
faster. However, circulation  of drilling fluids or muds
during rotary drilling may significantly alter the chemical,
mircobiological or  hydrogeological makeup  of the well
walls or some depth into the formation.
  The drilling equipment  itself  may present  a source of
contamination.  Safeguards, such  as steam  cleaning of
the drill rig and equipment, may not be sufficient when
investigating trace organics and buying all new equipment
for each hole  is ludricous. Although hydrologists  insist
upon  all downhole  equipment  being  steam  cleaned be-
tween each boring when  drilling in a hazardous or po-
tentially hazardous site, the potential exists for some areas
of the drilling equipment to be missed during any reason-
able steam cleaning operation.
  For this reason, investigators generally attempt to drill
in the upgradient zone first and progress toward the areas
which are assumed to contain the greatest potential for
                                                    143

-------
 144   SAMPLING
contamination.  Additionally, it is a very rare occurrence
that a drill rig  utilizing hydraulic systems does not leak
considerable  quantities  of  hydraulic  fluids  onto  the
ground or into the well bore.
   When a drilling program is designed to drill through a
contaminated zone  into a lower,  possibly less contam-
inated zone,  interaquifer  communication may create  an
avenue of contaminant transfer. In this case extreme cau-
tion and careful placement of intermediate casings will be
required. It is for this  reason that for all investigations
on hazardous wastes sites,  the authors  feel, the addi-
tional expense of maintaining an experienced contaminant
hydrogeologist on the rig at all times is warranted.
   Another  common  source of well contamination occurs
as a result  of surface contamination. Wells that are not
completed or sealed  sufficiently may allow surface water
to be  transported down along the casing material to the
sampling point. Even good development techniques which
may be  sufficient to remove all traces of inorganic con-
taminant may not be sufficient to eliminate the existence
of trace organics.
   Many authors have discussed the merits and problems
associated with  the various drilling techniques and as such
will not be examined here.  Although  certain techniques
may be  favored by  field personnel, no technique  can  be
exempt from being a contaminant source.  Consequently,
it is imperative that a detailed log be maintained at  all
times  during the drilling operations.  The  drilling  log
should note all unusual factors that may ultimately lead
to inconsistent results.

Piezometer or Well Construction Materials
   Discussions by  Pettyjohn,  et al.(]> have  compared the
merits of preferred well construction material versus the
cost of construction versus the quality of data. However,
notwithstanding the  arguments regarding preferred ma-
terials, contaminants may also be introduced into wells
during the  installation of the piezometer or casing ma-
terials.
   A common practice during installation of a well is to lay-
out the well materials along the ground and either install
it en masse or in discreet sections. This generally results in
a considerable quantity of surface material being installed
into the borehole along with the piezometer material. The
authors  have observed  several cases in  the  field where
chemically  inert well  materials and  thermally  welded
joints were utilized in an attempt to limit the introduction
of contaminants.  However,  before installing these ma-
terials  they were stored along  the drill rig prior to installa-
tion.  This combined with the use  of sand  filter packs
from  local  quarries  can result  in very confusing  results
being  obtained  from  samples  collected  in  these  "so
called" monitoring wells.
   The  preferred well materials were described by Petty-
john, et al. as being, in order of acceptance:
•Glass                  'Polypropylene
•Teflon                 Other plastics and metal
•Stainless steel          "Rubber
   This list of preferred materials must also be considered
from a cost benefit point of view. Each particular site
must be examined as to what level of risk may be intro-
duced when one utilizes a more reactive material for well
construction. This is particularly significant for a consul-
tant who must justify to his client  the additional costs in-
volved in utilizing a more inert material, such as glass or
teflon.
  It is essential to evaluate each material type against the
possible effect on the trace organics being evaluated. In
any case where a more  reactive material is utilized in well
construction, one well  of  the same construction material
should be placed outside the contaminated zone to provide
background data on the input of trace organics from the
uncontaminated aquifer and the influence of the well ma-
terial. Moreover, the well drilling and well construction
methods for the background well should be comparable to
that used to install the monitoring well.
  During the  production  of steel casing,  considerable
quantities of oils and  solvents are necessary at various
stages.  If trace amounts  of these materials remain ad-
hered to the casing during installation,  contaminants may
be introduced either directly or indirectly. Besides the ob-
vious case of oil from the casing getting in the sample, an
oil-coated pipe may prevent the complete sealing of the
aquifer  from either the  influence of surface contaminants
or intercommunication with another aquifer.
Sampling Equipment
  The  materials used   to construct  organic sampling
equipment also must be considered. Where possible, rela-

                                      ., ADSORBENT TUBE
               CLASS TUBE-
     i" PRESSURE TUBE
     TEFLON PISTON -
  36"
      TEFLON OR GLASS
      PISTON CHAMBER
                               TEFLON CONNECTOR
                             1" RETURN TUBE (TEFLON)

-)" PIEZOMETER TUBE
    (PLASTIC)
                                  • CHECK VALVES
                                    (TELFON)
                             SCREEN (TEFLON)
                        Figure 1.
         Dames & Moore Positive Displacement Pump
                   for Volatile Organics

-------
                                                                                            SAMPLING     145
lively inert materials are recommended to limit the pos-
sibilities of bleeding organics into the sample or adsorbing
organic  solutes.  The preferred materials  are  similar to
those outlined above for piezometer materials. Pump and
sampling devices that allow groundwater to contact metal,
rubber  or  lubricants  should  not  be  considered  when
sampling for organics.
  Some sampling equipment, by virtue of its design, may
create problems  with respect to interwell contamination.
Bailers,  for example, especially those with triggering de-
vices, are extremely difficult to ensure that they are com-
pletely cleaned after each well has been sampled. Further,
for many clients it cannot be considered cost effective to
dedicate separate samplers for each well.
  The uses of adsorbing  tubes and  positive displacement
(in-hole pumps) can greatly improve the efficiency of the
sampling equipment. Pettyjohn, et  a/.(" and others  have
investigated the use of various adsorbing materials for the
extraction of nonpurgeable organics from groundwaters.
Dames & Moore is  currently investigating the use of ad-
sorbing  columns on the return tube of a positive displace-
ment type pump (Figure 1). The authors feel this would
essentially eliminate the  contamination of groundwater
samples  that are subjected  to various sampling devices
and storage jars, before arriving at the laboratory for
analysis. Additionally, the sample would remain under a
positive pressure until being adsorbed.  This  positive dis-
placement pump will be described in more detail elsewhere
in this paper.
Sampling Technique
  Water sampling techniques should be designed to meet
the primary objective of a sampling program: to obtain a
representative sample of the fluid flowing  in the aquifer.
(That is, obtain a sample retaining the physical and chem-
ical properties of the aquifer fluid.) Preservation methods
and in situ tests will aid in measuring these properties but
are meaningless if the sample of well water is  not repre-
sentative of the actual aquifer.
  Water which has stagnated  in  the well bore has  had
ample opportunity to  change  physical  and  chemical
properties such  as  temperature, pH,  dissolved  oxygen,
total dissolved solids, etc.<3) Volatile organic contaminants
and dissolved gasses may have either volatilized or effer-
vesced in the well bore. Stratification of the water ac-
cording  to temperature and  specific gravity may lead to
different results being obtained from samples collected at
different depths, none of which may necessarily be repre-
sentative of the aquifer.
  Consequently, prior to taking a sample, it is necessary to
evacuate the well of the stored water in order to  allow
aquifer  water to flow  into  the well. The EPA recom-
mends evacuating at least two exchanges  of water prior
to sampling.(4)
  The well storage water may be  removed by bailing,
pumping from the surface, air  lift pumping, or by pump-
ing with a submersible  pump.  However, when pumping,
care must be taken  not to overpump the well  and allow
excessive fines to be drawn into the well.  This results in
increased turbidity  and  possible  damage to the gravel
pack.
  Care must also be taken in the selection of the pump-
ing method and materials. This selection may be dependent
upon the  parameters to be measured  and on  the  fluid
composition.
  Bailers  that resemble long, narrow buckets are avail-
able, or they may be  more sophisticated and allow. a
sample to be collected from a  specific depth. For small
wells with low yield, evacuation by bailing is a feasible
method of  obtaining aquifer  water. Bailing,  however,
becomes labor and time intensive for larger, deeper wells
with large volumes of storage.  In such cases, some sort
of pumping arrangement is more  economical  with the
bailer only utilized in obtaining a sample.
  A sample which is obtained by bailer is  exposed to
the atmosphere  during  transfer to  a sampling  jar. This
may destroy the test credibility for  volatile contaminants
and  dissolved gasses. Additionally, the  sample is exposed
to descending pressures as it approaches the surface.
  Operation on  the  more sophisticated bailers  becomes
difficult in  freezing  weather as the petlocks and  check
valves freeze and often  samples are lost. Bailers also in-
troduce a  certain amount of dissolved  oxygen which, in
turn, may  affect parameters sensitive to dissolved gas com-
position such as pH,  alkalinity and redox-dependent trace
metals.
  Pumps  used for well evacuation include suction  lift,
airlift and submersibles.  When using any type of pump, a
foot valve or check valve  is required  to prevent  con-
tamination of the well water by water which has been in
contact with the suction hoses or the pump itself.
  The suction lift pumps are limited as to the depth from
which they  may  pump (approximately  24 ft). When the
water level in the well is drawn  down beyond the limit of
the pump, confidence that a complete exchange of water
has taken place is reduced.
  Submersible pumps, on the other  hand, are not limited
as to the  depth  of the water level from  which  they can
pump, as are the suction lift pumps. However, the use of
submersible pumps involves the  installation and  with-
drawal of  the pump from each well during each  sampling
episode, assuming that  cost efficiency dictates that the
same pump be used for several wells.
  In the case where the  pump is used for multiple  wells,
there may be contamination from surface soils which ad-
hered to the hose during handling. Withdrawal of a  pump
with a check valve installed becomes laborious, especi-
ally  for deep settings and requires the  use of additional
supporting equipment and extra labor. Additionally, sub-
mersible pumps require the use of an outside power source
requiring additional  bulk to be transported during each
sampling period.
  Air lift pumping  is  a  relatively  old  method that is
gaining increasing attention. This method involves forcing
air down a tube  inside a casing with a resulting air-water
mixture rising up in the annulus.<5) Depending  upon the
applications, this method is an ideal way  to evacuate a
well  in a very cost efficient timely manner. Equipment is
readily obtainable and  relatively  inexpensive. However,
testing by  Dames & Moore has indicated that this method
generally  results  in  increased  dissolved oxygen  (D.O.)

-------
146     SAMPLING
values which in many cases have taken several days to re-
turn to prepumping levels. Additionally, gas stripping and
pH changes may also  occur.
  Other  pumping techniques may also alter the chemical
properties  of the  water through the introduction of dis-
solved gasses. Suction lift pumps subject the water to nega-
tive pressures which will  affect  concentrations  of  dis-
solved gasses. The  water  may   also  be contaminated
through contact with  the pump and the hose.
  An experiment  was designed and performed by Dames
& Moore to evaluate  the cross contamination potential by
using the same submersible pump for several  wells.  The
experiment and its findings are discussed in a later section.
  A positive displacement pump of the type discussed later
in this paper  has the  advantage of being suited for samp-
ling both volatile  contaminants and dissolved gasses.  The
sample is maintained under a positive pressure and is kept
from the atmosphere until it reaches the surface. The dis-
charge tube may be connected to a syringe or an  adsorb-
tion column as in Figure 1 to obtain a sample without at-
mospheric exposure.
                 The positive displacement pump has the added advan-
               tage that it may be installed at various levels within the
               same borehole, sampling  from  a small screened section
               below the pump. The other sampling techniques discussed
               above are not three-dimensional, that is, it cannot be de-
               termined from what level the contaminants are entering the
               well bore.  By  installing  several positive displacement
               pumps in a single borehole, the third  dimension of con-
               taminant plume  can be measured. In  Table I, adapted
               from Barker'2' a comparison of several methods currently
               employed for the sampling of organic compounds is made.
               Sample Preservation
                 Once the problem of how to obtain a sample has  been
               solved,  the next dilemma involves what to do with the
               sample. Normally the samples are submitted to the analy-
               tical laboratory in prerinsed amber glass bottles with tef-
               lon lined lids. For many of the organics holding times are
               critical  and unless the  laboratory receives the samples
               within the allotted time frame, the quality of the resultant
               analyses may be open to scrutiny.
                                                       Table I.
                        Characteristics and Evaluation of Various Methods Used to Sample Groundwaters
                                   for Organic Contaminants [Adapted from Barker 1981]
Method
bailer
2
high-lift
pump
special
piston pump
4
Materials Contamination Depth
Problems Limit
(ft.)
T, M Min

T, G Min

M, R, T Min-Sli


Min.
Piez.
Dia. (in. )
1

2

2.5

1 C
Air
Contact
Y

N

N


Degassing-
Loss of
Volatiles
minor

minor

minor


                     air squeeze
                                    M,  R, T
                     air-lift pump   T,  M

                     piston pump     M,  R, T

                     jet pump        M,  R, T
Sli-Max

Min

Sli-Max

Sli-Max
1.5

1.25

2

1.5
minor

major

minor

minor
submersible
pump
6
suction pump
and flask
peristaltic
pump
positive
displacement
pump
M, R, T

T, G
T, P
P, T

Max 4

Min. 25 0.25
Min-Sli 25 0.25
Sli 1.5

Mjterials— T-teflon, M-melal; G-glass; P-plasuc; R-rubber.
1 teflon extruded tubing with teflon rod end; glass marble check valve.
: all glass-icflon. 2-siage pump driven bv high purnv N7; Tomson el al., 1980. Groundwater, IS. pp
N minor

N major
N major
N minor

Materials:
T - teflon
M - metal
G - glass
P • plastic
444-446. R - nlbbtr
                    1 metal with rubber O-rmg seals, posilne displacement pump driven by compressed gas; Signor (1980); similar to Bennett pump

                    4 rubber and metal positive displacement pump driven by compressed gas; Middelburg pump (Tole Devices Co Ltd., P.O. Box 456, New
                      Albam. Ind. )

                    ? polveihvleneor teflon lubing and metal air lift pump; such as Johnson Water Sampler (Johnson Div., P.O. Box 3118, St. Paul. Minn.).
                    6 pump placed behind flask so thai \>aier contacts onh tubing and glass; hand vacuum pump or peristaltic pump can be used.

-------
                                                                                            SAMPLING
                                                                                            147
  Basic holding times and recommended sample volumes
were published  in the Federal Register*® and are repro-
duced here at Table II.
  During the holding period for the samples, losses or al-
terations of organics may result from any of a number of
causes including:
•Adsorption onto glass jar walls or particulate matter
•Precipitation or co-precipitation
•Biological transformations
•Chemical reactions (oxidation)
Generally, filtering of samples prior to analysis is not
recommended. Precipitates that may  have formed during
the holding period must be processed along with the water
during analyses. This limits the possibility of some con-
taminant being adsorbed onto particulate matter and being
lost  during  the filtration.  Consequently, very  careful
piezometer construction  and development prior  to any
sampling is necessary to reduce the concentrations of par-
ticulate matter that enters a well.

EXPERIMENT ON
PUMP CROSS-CONTAMINATION

  An experiment was designed by Dames & Moore person-
nel specifically to identify and solve the problem associated
with cross-contamination resulting from the use of the
same submersible pump  in several wells.  The tests were
made on a site with known PCB contamination. PCB's are
a compound with known affinity for plastics. Two pumps
were used in the experiment, the first a 0.5 hp submersi-
ble pump with plastic impellers, the second was an all
stainless steel 0.75 hp submersible pump  with  stainless

                       Table  II.<6)
       Sample Volumes, Preservation and Holding Times
               for Organic Priority Pollutants
          (from Federal Register, December 18,1979)
                 Sample
                 Volume   Preservative
    Measurement     (ml)     @4°C
    BOD
    COD

    chlorinated
    organic
    compounds
    organic carbon

    extractables
    (except phenols)

    extractables
    (phenols)

    purgeables (halo-
    carbons and
    aromatics)

    purgeables(ac-
    rolim and
    acrylonitrite)
    pesticides
                        cool
        cool, H,SO4 to
        pH<2
        cool, 0.008%
1000
50
25 to 500  cool, h,SO, to
        PH<1
1000     cool, 0.008%


5000
        cool, H-SO. to
        pH<2;0.008%N2S203
                      Max. Holding Time,
                      Before Analyses or
                      Extraction
                      (days)

                           2
                           28
                           28

                           7


                           7
20 to 500  cool, 0.008%

        N?2S2°3

20 to 800  cool, 0.008%
                1000
                      14
                        cool, 0.008%
    a—should only be added in the presence of residual chlorine
steel impellers. Both pumps utilized the same collapsible
plastic hose for discharge.
  The purpose of  the  experiment was to  estimate  the
amount, if  any, of potential  cross-contamination  be-
tween wells  which occurs  by utilizing the same pump in
the sampling procedures. The contribution of the plastic
hose was also to be estimated. Both pumps  were used in
the experiment to determine the advantages of using all
stainless steel parts, as opposed to plastic.
  The pumps were used to evacuate a 120 ft deep well
known, to be contaminated with PCBs. The pumps were
then placed in a 55 gal drum of potable water. Samples of
the potable  water were  obtained to develop background
levels from the drum prior to placement of the pump. The
hose was  connected and the discharge directed  to  recir-
culate water within the drum. The pump was then run for
1/2 hour, recirculating the potable  water. The water from
the drum was subsequently sampled.
  In order to isolate the contamination from the pump,
the above procedure was repeated, but without the hose
being connected to the pump in the  drum. The first slug of
water discharged from the pump was subsequently sam-
pled. The hose was  again connected, and the first slug of
water through the hose was then sampled. Once the ex-
periment was completed for both pumps, the  samples were
transmitted immediately to a laboratory for analysis.
  The results of the  analyses are  shown in Figure 2. A
significant amount of contamination was measured in the
test  recirculating water using the  plastic impeller pump
(approximately 11.4 ug/1). The samples isolated from this
pump showed approximately 3 ug/1 came from the pump,
and  a similar amount from the hose.  The stainless steel
pump produced insignificant levels considering the back-
ground levels of the "potable" water. When the hose was
connected, however, a marked contamination was noted,
approximately 3 pg/1.  From this very limited experi-
ment, the authors  have concluded that the plastic  im-
pellers appear to have an affinity  for PCBs and may be
responsible  for a small amount of cross-contamination.
Additionally, a plastic discharge hose, although very con-
venient to use, does contribute to cross-contamination.
Therefore, where  multiple  wells must be sampled,  the
choice of pump and hose materials can be significant for
low concentration contaminants.
DISPLACEMENT PUMP DEVELOPMENT

  To limit the problems  outlined in the above sections,
the authors believe that a monitoring program allowing an
individual  positive displace  pump to be  installed  into
each well is truly needed when dealing with the problems
of trace  organics. Such a positive displacement pump is
currently being developed by Dames & Moore  and is
shown in Figure 3.
  This pump consists of a 1.75 in diameter tube with an
inside piston. Connected to the piston chamber is a short
section of  well screen.  A check  valve between the well
screen and the piston chamber allows one  way flow into
the chamber. When pressure is applied to  the piston via

-------
148
SAMPLING
 a 0.25 in tube at the top of the chamber, the check valve
 closes and the fluid in the chamber is forced  through a
 second valve out a 0.5 in discharge tube.  The present de-
 sign will enable one liter of fluid to be  pumped  per stroke
 of the piston.
  A third  tube is connected to the 0.5  in tube between
 the screen  and the piston  chamber just below  the check
 valve. This permits monitoring of the hydrostatic pres-
 sure within the screen.
  The pump is suited for installation in various  materials,
 by  attaching different well screen sections.  By placing a
 filter pack around the screen, problems  from silt migra-
 tion may be reduced, if not eliminated.
                                                                                     1" RETURN TUBE
     15
    10

     0  "
           13 6
                               50
                                   BACK9ROUNO POTABLE WATER

                                      1!!!!!!;=*: =	 2.2
                 PUMP IMPELLER  TYPE

                       Figure 2.
    Cross Contamination Test Results for the Plastic Impeller
         Pump and the Stainless Steel Impeller Pump

  The pump may be constructed  of teflon or plastic, de-
pendent upon proposed operations. All pressure connec-
tions are made with swaglok fittings,  available in teflon.


CONCLUSIONS

  This paper was prepared to assist the reader in under-
standing the complexities  involved  in  sampling ground-
water for organic contaminants. Without careful attention
to all phases of a sampling program,  the results of the
sampling program are worthless. This may result in a client
performing  much  additional  investigative  drilling  and
sampling to answer questions that  may have resulted from
erroneous procedures.
  Additionally,  in many  areas low levels  of  trace  or-
ganics  may remain undetected because of poor sampling

                                                         i"  PRESSURE TUBE
                                                               PISTON
                                                     36"
                                                           PISTON CHAMBER
                                                                                         •i" PIEZOMETER TUB£
                                                                                                 • CHECK VALVES
                                                                                            SCREEN
                                                                          Figure3.
                                                           Dames & Moore Positive Displacement Pump
                                                   programs. If allowed  to  remain undetected,  these low
                                                   levels of trace organics  may create a serious health hazard
                                                   in the long term or at least result in a much more costly
                                                   investigative program.
                                                     Although rapid developments are currently taking place
                                                   in the field of organic sampling, these are currently only
                                                   in developmental stages and  as such, modifications and
                                                   improvements are constantly  being  made. Consequently,
                                                   proven standard cost effective techniques for sampling
                                                   organics  still lags  far  behind the  analytical  laboratory
                                                   techniques.

                                                   REFERENCES

                                                   1. Pettyjohn, W.A., Dunlap,  W.J., Cosby, R.,  andKeely,
                                                     J.W.,  Groundwater, 19, 1981, 180-9.
                                                   2. Barker,  J.F., Short Course,  Field Methods in Con-
                                                     taminant Hydrogeology.  Department  of Earth Sci-
                                                     ences, University of Waterloo, Waterloo, Ont., 1981.
                                                   3. Schuller,  R.M.,  Gibb,   J.P.,  and  Griffin,  R.A.,
                                                     Groundwater Monitoring Review 1 #1, 1981,  p. 42.
                                                   4. U.S.  Environmental Protection Agency, "Procedures
                                                     Manual for  Groundwater  Monitoring at Solid Waste
                                                     Disposal Sites," EPA-530/SW-611, 1977.
                                                   5. Morrison, R.D. and Brewer, P.E.,  Groundwater Mon-
                                                     itoring 1 #1, 1981, p. 52.
                                                   6. U.S.   Environmental   Protection   Agency,  Federal
                                                     Register, Dec. 18, 1979.

-------
           SPECIAL SAMPLING TECHNIQUES USED FOR
  INVESTIGATING UNCONTROLLED HAZARDOUS WASTE
                                SITES IN CALIFORNIA
                                    HOWARD K. HATAYAMA
                              California Department of Health Services
                                         Berkeley, California
INTRODUCTION

  The primary purpose of this paper is  to briefly dis-
cuss some special sampling techniques used to  character-
ize the nature and extent of contamination at selected un-
controlled hazardous waste sites in California. These in-
clude techniques  for sampling soil, waste, groundwater
and gases. The methods are discussed as they were applied
to specific sampling problems at four sites: (1) Occidental
Chemical Company, Lathrop, Ca., (2) McColl  Site, Full-
erton, Ca., (3) Mola Development Site, Huntington Beach,
Ca. and (4) General Electric Company, Oakland, Ca.
  The techniques for soil and waste sampling discussed
are: (1) Hollow-stem  augering with drive  sampling,  (2)
Rotary drilling with drive sampling, (3) Bucket augering
and (4) Trenching.  The two techniques for groundwater
sampling discussed  are: (1) Single completion wells with
submersible pump sampling and (2) Multiple cased wells.
Installation of gas collection wells is covered under gas
sampling. Most of the information presented herein is de-
rived from the files  of the  California  Department of
Health Services.
SITE DESCRIPTIONS

Occidental Chemical Company, Lathrop, Ca.

  The facility is located in the Central Valley  of Cal-
ifornia approximately 50 miles south of Sacramento, the
State capitol. It is built on the flood plain of the San Joa-
quin River with useable groundwater from 20 feet down
to 300 ± feet. It is a fertilizer manufacturing and pes-
ticide formulating plant handling a wide variety of organo-
halogen, organophosphate and carbamate type pest con-
trol agents. For a period of 10-15 years, the nemnatocide
DBCP, (1, l-dibromo-2-chloro-propane), were also man-
ufactured there.
  Wastes from fertilizer manufacturing consisted largely
of a gypsum slurry which was ponded until just recently.
Wastes from the pesticide formulating and manufactur-
ing activities consisted  of equipment washes, unrecover-
able bad  batches, run-off, raw material containers and re-
turned obsolete off-specification consumer products. Until
the mid-1970s,  all the  liquid wastes were  discharged to
an unlined ditch and pond.
  In the  1950s and 1960s,  the solid wastes were  either
burned in open pits or buried in trenches. These trenches
were not apparent from observation of the ground surface
at the start of investigations. These practices resulted in
gross contamination of surface and near surface soils, and
of the ground water to depths greater than 200 ft.

McColl Site, Fullerton, CA.

  This site is located in Southern California about 17 miles
southeast of downtown Los Angeles. It is a seven acre site
bounded on three sides by homes and on the fourth by a
golf course. It consists of six sumps covering about four
acres containing an estimated 50,000 yd3 of acid petrol-
eum tars.
  The wastes were generated by the local petroleum re-
fineries during the 1940's from high octane  fuel produc-
tion. The material is a black tarry substance with a firm
crust that softens upon warming by the sun. On very hot
days, the waste flows and up-wells. It contains large quan-
tities of sulfur dioxide, benzene and other hydrocarbons,
tetrahydrothiophenes and arsenic. It is highly  acidic (pH of
the slurry is less than one). When the crust is broken, large
quantities of sulfur dioxide and other extremely odorous
compounds are released.

Mola Development Site, Huntington Breach, Ca.

  This site is located along the Southern California coast
25 miles southeast of downtown Los Angeles on a plateau
overlooking a wildlife preserve. It is bounded on four sides
by homes and apartment houses.
  It  was originally a stream bed used as a  sand quarry.
Subsequently, it was used as a disposal site for acid petrol-
eum tars and  other petroleum wastes during  the  1940s.
Thereafter, it was operated  as a permitted site for drill-
ing muds and building demolition wastes. It was estimated
to contain approximately 100,000 yd3 of petroleum wastes
and 200,000 yd3 of drilling muds and building demolition
wastes.
  The petroleum waste characteristics are very  similar to
those of the McColl Site wastes, but somewhat diluted by
water. Similar types of gases were being generated  by the
wastes along with significant quantities of  methane and
carbon dioxide.  The rate of gas  generation was  slower
although the gases were no less odorous. The absorptive
                                                  149

-------
150    SAMPLING
capacity of loosely compacted building demolition wastes
and  the pooling of water on the layers of drilling mud
caused the tarry petroleum material to up-well in pools on
the surface. The up-welled material became more flowable
on warming.

General Electric Company, Oakland. CA.

  This  facility is  located in Northern California, near
downtown Oakland. It is bounded on one side by homes
and  on another by a food processing firm that draws water
from 150-200 foot depth. The water table aquifer is at
10-20 ft. The facility was a  transformer repair shop that
handled PCB (polychlorinated biphenyl) and  other types
of transformer oils in bulk quantities.
  Spills and  leaks of PCB's and other transformer oils
resulted in significant soil contamination and a  layer of
PCB containing oil  up to 6 inches thick floating on the
water table.


SOIL AND WASTE SAMPLING TECHNIQUES

Hollow-stem Angering With  Drive Sampling

  This technique (Figure  1) employs  a continuous flight
hollow-stem auger to bore to the desired depth in the soil
or waste. The sample is then obtained using various types
of drive sampling devices. The auger is usually plugged
during drilling and  the plug is removed to advance the
sampler ahead of the auger.
  The unlined ditch and ponds at Occidental were sampled
at surface, 2, 5, and 10  ft  depths  using this technique.
The hollow-stem auger not only provided support for the
bore holes but also controlled sloughing of the more high-
ly contaminated soils into depths of  less contamination.
A split-barrel sampling device was selected for obtaining
samples because it allowed the taking of a relatively un-
disturbed core, the physical  inspection of the sample, and
efficient transfer  of the  sample for splitting and con-
tainerization. The top  1-2 in. were generally discarded as
a further step to control the effects of sloughing.
  The sampler was  thoroughly cleaned with  scap, water
and acetone  between each use. Sampler decontamination
procedures generally depends on the type of waste and the
contaminants of concern. In this case, since contamina-
tion of soils  with  organics and heavy metals was  antici-
pated,  both  an aqueous and  organic  decontamination
solutions were used.

Rotary Drilling with  Drive Sampling

  This  technique  (Figure 2) is  different  from  the pre-
vious one only in  the drilling method. The desired depth
is reached by boring with a  rotary drilling rig using some
type of fluid  to circulate the  cuttings and support the bore
holes. The sampler is advanced ahead of the drill bit.
  This system was used  to  obtain discrete core samples
from the McColl Site petroleum waste sumps. A  "dry"
drilling technique (augering or compressed  air rotary)
could not be used because of the large quantities of sul-
  Soll or M««tt
                       Figure 1.
           Hollow Stem Auger with Drive Sampler.

fur dioxide and other  noxious  gases released by
ing the crust and maintaining an open hole. The  ..._
ing winds carried these gases directly into the tdjoiaiaf
homes. Water was found to be an adequate fluid for
trolling these emissions.
  Because of the flowable nature of the  wastes,  •
weight trailer mounted rig with large surface supports
necessary. The  cleanliness of  the  split barrel
could not be maintained because of the use of th
fluid. The drilling fluid may also have penetrated  the'
disturbed waste material around the bore holes.
effects were controlled to a certain extent  by _
ing the top 1-2 in. of the sample. Any  significant _
ture in the sample was noted. The petroleum waste	
terial was found to be relatively  impermeable under the
circumstances of the sampling.

Bucket Auger Drilling and Sampling

  A bucket auger (Figure 3) is a cylindrical (12-18  in.
diameter) bucket with a  swing-open bottom. A set of

-------
                                                                                SAMPLING     151
           Figure 2.
Rotary Drilling with Drive Sampling.
           Figures.
         Bucket Auger.
angled teeth is fixed to the bottom for augering into the
soil. The teeth guide the  sample into the bucket as the
bucket is rotated by  a turntable. Depending on the length
of the bucket, samples can be collected at 1 to 2 ft inter-
vals. Once a sample is obtained, the bucket is withdrawn
from the bore hole, swung out to the side and the sample
is released onto the ground surface by opening the bottom.
The large sample size generally allows taking  several sub-
samples from each depth.
  This technique was used at the Mola Development site
with the primary objective of obtaining samples of the
waste material deposited  below the building demolition
wastes and drilling muds. From  previous drilling exper-
ience at the site, it  was determined that other techniques
were  not adequate  for penetrating the  stubborn over-
burden and obtaining samples  of the waste.  With much
effort, it was possible to remove the concrete rubble over-
burden and  obtain depth  specific samples of the solidi-
fied petroleum waste material.
  In some cases, the waste was in a semi-solid or slurry
form and it was not possible to obtain representative depth
specific samples. A composite sample was taken of these
borings and  the  bottom  of  the  waste was  roughly de-
termined by the sandy stream bottom.
  No attempt was made  at  decontaminating the auger
between samples because of the time involved and the large
sample size. Sloughing of the walls of the bore hole could
not  be  adequately  controlled.  The rate of  gas genera-
tion was significantly lower, here than at the McColl  Site
and the prevailing winds provided higher dilution such that
gas control measures were not necessary during sampling.


Trenching

  Trenching for purposes of locating and sampling buried
hazardous waste materials usually entails the use of an ex-
tendable back-hoe  with a fully enclosed operator com-
partment (Figure 4). The  unknown nature of the buried
material requires strict attention to the health and safety
of the sampling personnel. It is critical that as  much in-
formation as possible on the location and possible nature
of the buried material be gathered before trenching is be-
gun. A series of cross-trenches is usually dug to intersect
the suspected burial area. The trench walls and diggings
are logged and sampled using manual devices.
  Investigation of historical records at Occidental revealed
the existence and general locations of the burning and bur-
ial  trenches.  After marking  the  approximate locations,
trenching began with the intention of locating  the trenches
more precisely and  sampling their contents. The 3-4 ft of
relatively uncontaminated overburden was placed in  sep-
arate piles while  the contaminated soil  and wastes were
placed on polyethylene sheeting.
  A significant fraction of the wastes were found in the
buried and saturated zone. The wastes consisted of par-
tially full pesticide containers, empty drums and  other sol-
id wastes related to pesticide formulation. The volume of
waste uncovered and the  fact that much of the pesti-
cides were  leaking directly into  the  groundwater man-
dated a change of strategy.

-------
152     SAMPLING
  The wastes were loaded into trucks for immediate dis-
posal classified as extremely hazardous  pesticide  wastes
without benefit of analytical results. Containers with re-
sidual pesticides were overpacked  and empty containers
were placed in dumpsters. Contaminated soil was sampled
and analyzed prior to disposal.
  Although circumstances in this  case dictated that  the
exploration and removal process occur concurrently, it is
generally  more appropriate  to separate  these two. This
allows time to develop as much information as possible to
safely and properly remove  the wastes without incurring
undue risks to workers, the surrounding community, and
causing further contamination of  the soils  and ground-
water.

GROUNDWATER SAMPLING

Single Completion Wells with
Submersible Pump and Packer

  As  shown (Figure 5), this technique involves using a
submersible pump  attached  to  an air inflatable rubber
tube to obtain a representative groundwater sample. After
positioning  the pump/packer assembly  at the desired
depth, the packer is inflated to form  a seal between the
screened section and the  rest of the water column. The
system which also includes a Teflon ® tube  for sampling
is then flushed with 3-5 volumes  (volume of the screened
section) before a sample  is taken.  The sampling tube is
not, however, shown in Figure 5.
      Air, Electrical, Control Lines

             Steel Casing
             Cement Grout
             Air-Inflatable Packer
             Gravel Packing
             Submersible Centrifugal
                Pump
        Figures.
Pump and Packer Assembly.
                        Figure 4.
                       Trenching.
                                                                        Oil Layer
                                                                                 I
                                                                     Shallow Aquifer
                                                                     .Impermeable -;
                                                                     * C. Stratum _*_"
                                                                          Aquifer * »
        &
        :•;»
        $
                                                                                           ,PVC Well Casing
              m
-Cement Grout Casing II
                                                                                              Cement Grout Casing 11
                                                                                             -Hell Screen


                                                                                             -Gravel Packing
        Figure 6.
    Multi-Cased Well.

-------
                                                                                             SAMPLING     153
                                  Gas Sampling Valves

                                  ,— Metal Cap

                                    Cement Grout Cap
                                    Gas Collection Port
                        Figure?.
                   Gas Sampling Well.
  The permeability of the soils in the area, analytical data
from surrounding wells, and the  nature of the operations
at Occidental's  facility indicated that  groundwater  was
contaminated with organics and  inorganics to significant
depths. Based on available hydrogeological and pilot  well
data, further sampling locations, each with three single
completion wells at various depths down to approximate-
ly 200  ft were  selected. This type of  well construction
allowed isolation of the  most significant water bearing
strata. The packer further isolated the specific zone of in-
terest while limiting the amount of water flushed and con-
trolling any mixing of stagnant water in the well column
with water which is more representative of the aquifer.
  The submersible centrifugal pump with relatively inert
component parts allowed sampling for purgeable organ-
ics such as DBCP and EDB (1,  2-dibromoethane) along
with other less volatile constituents. The depth of sampling
and these volatile components prohibited the use of other
sampling methods (bailing, vacuum pumps, and air lift
pumps). The entire sampling assembly was flushed with
distilled,  deionized water between  each location  instead
of providing dedicated tubing for each location to provide
further control over the possibility of cross-contamination.

Multiple Cased Wells

  This is more a method of well  construction rather than
of sampling (Figure 6). A large diameter boring is made
through the upper, highly  contaminated aquifer into the
first impermeable zone. This boring can then be cased and
grouted with cement or plugged with bentonite clay. A
smaller boring is then made through the casing or the seal
into the next aquifer where a single completion well is
constructed.
  Multiple cased wells were employed at the General Elec-
tric facility where the PCB contaminated oil  layer on the
water table threatened to cause contamination of the deep-
er aquifer during  drilling. Representative depth specific
soil  samples  were also obtained from zones below the
water table aquifer using this drilling technique. Such an
approach should be considered in all cases where the prob-
ability exists  of  contaminating to a significant extent
previously uncontaminated zones.

AIR SAMPLING

Gas Well Installation and Sampling

  The landfill gas sampling probes  shown in Figure 7 are
encased in a perforated metal casing and lowered into a
bore hole. The annulus is  packed with gravel and sealed
with cement. All parts are constructed of steel or brass to
minimize adsorption and desorption effects.
  Gas collection parts are provided  at two levels for char-
acterizing gases at these levels and to provide for the poss-
ibility  of inundating the  deeper probe  with leachates.
Samples of the collected gases  are  obtained by pumping
through various types of adsorbents  (i.e., activated car-
bon, or Tenax ® , impingers with selective solutions or by
filling Tedlar ® bags.
  This approach was necessary at the Mola Develop-
ment site to evaluate the rate  of generation of the noxious
gases,  characterize them more fully and  evaluate the po-
tential for subsurface migration to  the surrounding com-
munity. The information was also  necessary to develop
appropriate mitigation measures. These gas collection de-
vices were installed in some  of the  borings made  by the
bucket auger and the well was  allowed to equilibrate for
several weeks. Composite gas samples representing the en-
tire column were  taken using Tenax ®  adsorbent for
organics and impingers for inorganic vapors. This method
allowed extensive characterization of  the gases generated
on-site although it was not sufficient to fully  evaluate the
rate of gas generation for purposes of planning excava-
tion.

CONCLUSIONS

  Technology has generally been available for sampling at
the sites discussed and  other sites.  Investigations at un-
controlled sites often require novel applications of these
technologies.
  Maintaining quality control is a key factor in any sampl-
ing method selected, especially when addressing low levels
of  highly toxic contaminants.  However, compromises
must be made in some cases when sampling waste material.
  Care must  be taken in any investigation that the act of
sampling does not cause significant  contamination of pre-
viously uncontaminated soils, aquifers or air space.

-------
            THE STRATIFIED SAMPLE THIEF— A DEVICE
                     FOR SAMPLING UNKNOWN FLUIDS
                                       MICHAEL G. JOHNSON
                                 Mason & Hanger-Silas Mason Co., Inc.
                                         Leonardo, New Jersey
INTRODUCTION
  The primary reason for the development of the strati-
fied sample thief was the problem of sampling the fluids
recovered by the oil spill recovery  devices tested at the
EPA OHMSETT test facility. The method used since the
facility opened and continued until the 1980 test season
was as follows:
•The test fluids were collected in containers on the auxili-
 ary bridge during a test
•The container fluid level  was measured to obtain total
 volume collected
•The free water was decanted from the bottom of the con-
 tainer
•The remaining fluid was remeasured
•The  sample  was mixed (emulsified) using an  electric
 motor and propeller
•A small sample of the fluids  was taken while mixing
 continued
•The sample was taken to the  lab where it was centri-
 fuged to determine the amount of water in the oil.
  Some of the problems associated with  this procedure
were:
•The measuring, draining and remeasuring took time and
 produced more numbers that could be lost of misinter-
 preted
•The mixing caused severe emulsification that made the
 refurbishment of the oil for reuse difficult and expensive
•The plastic containers used for  the recovered fluids were
 subject to damage and degradation by sunlight.
  To eliminate some of these problems, many  methods
for sampling were discussed and/or tried including:
•Discrete sampling using a rotary table with tubing and a
 solenoid valve
•Discrete grab samples (this was similar to the rotary table
 except that all operations were manual).
  The rotary  table discrete sampler was less than satis-
factory because:
•The fluid in the discharge line from a device could strati-
 fy and allow a non-representative sample to enter the
 tubing
•High viscosity fluids would not flow through the sampler
 tubing fast enough to fill the sample bottles in the time
 allowed for each discrete sample
•Debris sometimes plugged the tubing or solenoid valve
•In some test configurations, there  was not enough pres-
 sure to allow sufficient flow in the sampler system
•The electrical system constantly failed due to corrosion
 and mechanical damage.
  The  manual discrete sampling was better from  the
mechanical/electrical standpoint but it had problems with
stratification, low pressure, debris and high viscosity. An-
other disadvantage was the large amount  of manpower
required.
  Some of the sampling methods discussed for possible
use were:
•Vacuum bottles (similar to those used for blood sampling)
•Multiple valved tube
•Grain sampler tube
•Cork plugged tube (COLWASA).
  These methods were all rejected because  of anticipated
problems such as:
•Cost
•Complexity
•Debris problems
•Viscosity effects
•Sealing
•Susceptibility to damage.
STRATIFIED SAMPLER DEVELOPMENT

  A  solution to the  problems inherent with  the other
sampling devices or methods was finally arrived at after
eliminating numerous other ideas. It is called a  strati-
fied sample thief.<'•2'3)
  This sampler  uses discs and wipers to hold  the strati-
fied liquids in position while the tube is slipped past them.
(Figures 1 and 2.) The wipers keep the inside of the tube
from carrying portions of the upper fluid down into other
layers. They seal well enough so that  the samplers can be
carried to and stored in the laboratory horizontally.
  This procedure allows the transfer  of the sample from
the sample thief to another container to be accomplished
in the laboratory. This is  done by clamping the sample
thief vertically in a rack and removing the sheath.

Advantages
•Simplicity
•Inexpensive
•Ease of use
•Speed of operation
•Representative sample obtained (see Table I)
                                                   154

-------
                                          SAMPLING    155
        i
        2
        3
        4
        5
        6
        7
        8
Spacer
Supporting Washers
Wiper
Sheath
Center Rod
Extension
Bottom Stopper
O-ring
         Figure 1.
Stratified Sample Thief Construction

-------
156    SAMPLING
                             1
                                                   B.  The  outer  sheath
                                                       is  withdrawn to
                                                       expose the center
                                                       section.
                                 A.  The  sampler with  the
                                     extension rod is placed
                                     in the  barrel through
                                     the pour spout.

                                  c.
The outer  sheath
is  slid down  the
center section,
trapping the liquid.
                                             D.  The entire
                                                 sampler is
                                                 withdrawn
                                                 from  the  drum
                                                 with a repre-
                                                 sentative
                                                 sample enclosed
                                                                 ffl
                                       Figure 2.
                                 Use of the stratified sampler

-------
                                                                                            SAMPLING    157
•Horizontal storage
•Useable by personnel wearing bulky safety gear
•Can be made in lengths to eight feet for deep containers.
Disadvantages
•Hard to clean
•Sheath material may be  damaged by organic solvents
 (glass sheaths may be better)
•Use in low overhead area restricted by rod extension
•Bottom o-ring swells  after repeated  usage, requiring re-
 placement.

COLWASA SAMPLER

  This device is similar to the COLWASA(4) device, which
is a tube with a cork or plug in the lower end which can be
pulled into the lower end by a line  or rod attached to the
cork. The line or rod  runs up through the inside of the
tube and extends far enough out of the upper end of the
tube so that it can be grasped. A sample is taken by lower-
ing the assembly slowly into the fluid with the cork lowered
so liquid can  flow  into the  tube. When the  assembly
reaches the bottom of the fluid container, the line or rod is
pulled  up to seal the liquid  inside  the tube. On some
models, a locking mechanism has been added to lock the
rod in place.
  The COLWASA sampler works  well for homogeneous
fluids or for stratified liquids of low viscosity. Stratified
fluids with high viscosity can  cause a problem due to the
rather thick coating that can build up inside the tube. This
coating can prevent the taking of representative samples.
The stratified sample thief offers a solution to this prob-
lem.
  The 2.22-cm (inside diameter) sampler was evaluated in
22 different tests in a 500-gal barrel of salt water with
measured quantities of oil varying  from 10 to 90 percent.
The results of tests with the Johnson sampler and (for
comparison) with grab samples taken after homogenizing
barrel contents with an electric mixer are shown in Table I.
ACKNOWLEDGEMENT

  The work upon which this presentation is based  was
performed by  Mason &  Hanger-Silas  Mason Co.,  Inc.
pursuant to Contract No.  68-30-2642  with the U.S.
Environmental Protection Agency. The mention of trade
names or commercial products does not necessarily con-
stitute endorsement or recommendation for use. The opin-
ions or assertions contained herein are those of the au-
thors and  do not necessarily represent the views of the
U.S. Government.
                        Table I.
      Johnson Sampler Test Results on Oil/Water Samples
                 Actual       Johnson
     Test        Cone.        sampler
     number      (% Oil)       (% Oil)

      1          74.2         73.1
      2          64.3         60.0
      3          53.9         59.8
      4          67.2         66.3
      5          45.9         46.1
      6          56.0         54.7
      7          76.5         71.2
      8          86.9         83.1
      9          66.4         64.9
     10          81.8         77.8
     11          40.3         39.7
     12          49.7         48.3
     13          44.3         44.9
     14          37.0         35.2
     15          39.2         38.8
     16          30.8         29.0
     17          23.2         21.1
     18          13.3          8.8
     19          18.8         18.0
     20          10.9         10.3
     21          14.5          9.7
     22          89.5         90.5
Grab
sample
(% Oil)

81.0
73.1
59.8
65.5
44.5
54.7
75.2
85.1
64.9
79.7
39.7
50.5
42.6
36.2
41.4
29.4
23.7
 9.0
16.3
 9.0
11.2
90.0
REFERENCES

1. Borst, M. "Johnson Sampler Accreditation Test" (un-
   published) Report to  U.S. Environmental Protection
   Agency MERL-Ci, Cincinnati, Ohio, 1981.

2. Nash, J.H. "OHMSETT Evaluation of the Clean At-
   lantic Associates; Fast Response Open Sea Skimming
   System." (Unpublished Report to Clean Atlantic As-
   sociates,  c/o  Halliburton  Services,  P.O.  Box  1431,
   Duncan, Oklahoma 73533) 1980.

3. Farlow, J.S.,  Griffiths, R.A., "OHMSETT Research
   Overview, 1979-1980", In:  Proceedings of the 1981
   Oil Spill Conference,  American  Petroleum Institute,
   Washington, D.C. 664-665.

4. deVera,  e.R., Simmons,  B.P.,  Stephens, R.D., and
   Storm, D.L.,  "Samplers and Sampling Procedures for
   Hazardous Waste Streams", U.S. Environmental Pro-
   tection Agency  (MERL-Ci) Report No.  EPA-600/2-
   80-018, Cincinnati, Ohio, 1980.

-------
      THE COMPLEMENTARY NATURE OF GEOPHYSICAL
         TECHNIQUES FOR MAPPING CHEMICAL WASTE
     DISPOSAL SITES: IMPULSE RADAR AND RESISTIVITY
                                        KEITH A. HORTON
                                       REXFORD M. MOREY
                                         LOUIS ISAACSON
                                    RICHARD H. BEERS, Ph.D.
                                          Geo-Centers, Inc.
                                 Newton Upper Falls, Massachusetts
INTRODUCTION

  Geophysical mapping at waste disposal sites has pro-
vided a direct comparison of results obtained from the
application of two different geophysical techniques: (1)
electrical resistivity (ER) and (2) electromagnetic impulse
or ground penetrating radar (GPR). A correlation be-
tween  data acquired  by  these two methods is reported
in this paper. The two types of data can be used in a com-
plementary manner since the radar signatures give geo-
logical meaning to the resistivity and the resistivity can be
used to estimate radar system capabilities and predict the
depth  of penetration of the  radar signals in the  site
geology."1

GROUND PENETRATING RADAR

  The ground  penetrating  radar  system, illustrated in
block form in Figure 1, is composed of a transmitter, a
receiver, an antenna  (or antennas), an  analog magnetic
tape recorder and graphic display unit. The transmitter
generates repetitive extremely short-time duration  pulses
which are radiated into the ground from a broadband an-
tenna placed in close proximity and electromagnetically
coupled  to the ground. The  radar operates as an echo
sounding device.  Electromagnetic  pulses, reflected from
targets or interfaces within the ground, are detected by the
antenna,  converted into voltage as  a  function of time
waveforms in the  receiver, and displayed on a graphic re-
corder.
  As the antenna is scanned  over the ground surface, a
continuous profile of subsurface electromagnetic condi-
tions is printed on the graphic recorder. The location and
depth of subsurface targets are inferred from the graphic
record. Depth of penetration and velocity of propagation
in the ground are a function of the constitutive electro-
magnetic parameters of the soil. Maximum depth of pene-
tration  ranges from tens of centimeters to hundreds of
meters.
  The  transmitter and  high  frequency receiving elec-
tronics are mounted in the antenna. This allows the use of
several different antenna configurations operating at dif-
ferent frequencies with the same radar control unit and
recording equipment. Even though each antenna config-
uration radiates a broad spectrum of frequencies, a par-
ticular antenna unit is designated by the center frequency
 GROUND SURFACE
TRANSMITTED PULSE
       TARGET
                                    REFLECTED PULSE
                      Figure 1.
     Block Diagram of a Ground Penetrating Radar System

of its spectrum. Typical radar center frequencies vary be-
tween 10 and 1000 MHz.
  The recording equipment and control unit are mounted
in a  van or other appropriate vehicle and the antenna is
towed behind the vehicle. Data are recorded on magnetic
tape  and on the greyscale graphic recorder; the latter in-
formation being compressed dur to the high input data
rate. After completion of a survey, the magnetic tape data
are played back to generate  full resolution hard copy for
analysis.
  An example of a typical transmitted and received pulse
train is shown in Figure 2a while the graphic recorder
display of the same is shown in Figure 2b. The dark bands
occur at signal peaks,  (both positive and negative), while
the narrow  white lines  occur at  the zero crossings be-
tween peaks. Areas to be surveyed are defined by driving
stakes into the ground to establish a reference grid sys-
tem. This grid is used as the coordinate system for the
radar data and for any other  geophysical surveys also
undertaken. Positional information is accurate to within
a few feet. Grass and weeds  are usually mowed and brush
cleared to ease actual travel over the surface.
                                                  158

-------
                                                                                 REMOTE SENSING     159
  Several preliminary scans are made at eacri site to cali-
brate the antennas  with  respect to the electromagnetic
characteristics of the specific soils. In this way, each sys-
tem configuration is optimized  to  provide the desired
depth of penetration and/or resolution. In addition, the
velocity of propagation of the EM signal and the dielec-
tric constant of the soil are calculated by measuring the
two-way travel  time of the signal to objects at known
depths or by a series of wide-angle reflection from an ob-
ject or interface at an unknown depth.
          SIGNAL
        AMPLITUDE
                         I RECORDER  PRINT

                         I THRESHOLDS
 HORIZONTAL
 TRAVEL	
                            TRANSMITTED  PULSE'
                                       SURFACE •
                        •1








        ^^^^^^^"•^^^^^^^^•^^^^B^^^^W^^W^WWn^^^^^"^^*^^^^^^
                                                                                   ._•£;«:
                                      INTERFACE
                                         SIGNAL
                                                                                                        H
                      Figure 2a.
        Sketch of a Single GPR Pulse and Reflections
                 as Seen by the Receiver.
                      Figure 2b.
       Example of Profile Information as Displayed by
                 the Graphic Recorder.

-------
160     REMOTE SENSING
RESISTIVITY

  Earth resistivity surveys have been used for many years
in exploration for groundwater and mineral deposits, and
in the study of engineering properties of earth materials.<2)
Equipment to measure apparent resistivity consists of a
controlled  source of electric current and  a device for
measuring the potential differences generated by the  cur-
rent passing through the earth.
  Typically, four electrodes are used in resistivity measure-
ments. The volume of a subsurface material influencing
the resistivity measurement is controlled by the spacing
and geometric arrangement of the electrodes. While any
array of four or more  electrode contacts can be used in
studying earth resistivity, a relatively few electrode con-
figurations have become accepted as standard  arrays in
practice. The most common electrode arrays used in the
resistivity method are shown in Figure 3.
  Many  factors go into the choice of array configura-
tion for a given problem. Susceptibility to geological noise,
ease of array movement and the nature  of the assumed
structure are a few of the factors to be considered.
   (a)  Wenner  Spread
   (b)   Schlumberger Spread
J f r r r f f r rrrrtrrrr—r~r—7—7—*
  (c)  Double-dipole  Spread
                        Figures.
    Common Electrode Configurations for Resistivity Arrays
RADAR/RESISTIVITY CORRELATION

   Examples of radar signatures obtained at a frequency of
120 MHz are shown in  Figure 4, for "broken slag" and
"clay" present at an uncontrolled chemical waste disposal
site; the corresponding resistivity profile superimposed
on the radar profile is also illustrated. Apparent resistivity
is plotted downward in order that conductivity may be
visualized as  increasing upward. The effective depth of
penetration is greater in "slag" with a corresponding lower
conductivity and less in "clay" with a higher conduc-
tivity.  Although  conductivity is frequency dependent and
the two methods are operating at frequencies differing
about  seven orders  of magnitude, the relative effects of
radar signal attenuation and measured d.c. conductivity
(or resistivity) are shown to be similar when moving from
one material to another.
           85 + 30 85 + 00
                       STATIONS
                        84 + 50  84 + 00
                                      83 + 60
                                           -10
                                                                                                       50
                                                                                                      -100
                                                                   14-11
                                                                          "BROKEN-
                                                                            SLAG"
                                                                                                          ft
                                                                                                          GC
                                            500



                                           -1000


                          •	"CLAY"-	-I

                           A RESISTIVITY DATA POINTS
                                                                                  Figure 4.
                                                                          Radar and Resistivity Profiles.
                                                                   Radar Profile Shows Characteristic Signals of
                                                                          "Broken Slag" and "Clay".
  The relationship between the radar depth  of penetra-
tion and the d.c. conductivity (resistivity) is illustrated in
Figure 5. Based on the radar range equation  and a com-
plex index  of refraction soil model, the maximum depth
of penetration  of the  radar system is calculated for fre-
quencies of 10,  100, and 300 MHz.(M'5)
  Also  plotted in this figure are  the radar  penetration
depths achieved at 120 MHz and d.c. conductivity (resis-
tivity) measurements obtained at the chemical waste site
illustrated in Figure 4. Considering the complexity of this

-------
  f
  sr
  £
      1000
II
r* ^^
•< a-
» a


§ 5
O r-t-


II

3- 2
w 3
jo 51
p C
O. B
Q3 O
S a
 ••
  o.

   '
100 -
      L(m)
                                                 COMPLEX  REFRACTIVE  INDEX  MODEL



                                                        (ROUGH PLANE  REFLECTOR)
                                                                  WATER



                                                                CONTENT
                                                                    30%
                                                                             C. (dry) = 4
                   SLAG 8 GRAVEL



                      CLAYS


-------
I—
o
z   §
o   °

S"   2

<   S
y   o
         1000
  O O
  —i 3


  >
   .
  O C
  CD S

  S 3

  tfl g
  2c
  » >
           100 -
  -s
   QJ

   3
                                                                                              70
                                                                                              m



                                                                                              I
                                                                                              m
                                                                                              C/5
                                                                                              tfl

                                                                                              co


                                                                                              o
H
11
              10
10
10
                                                                  A  (dB/m)
10

-------
                                                                                      REMOTE SENSING     163
site and the extent of modeling involved, excellent agree-
ment is obtained between the soil and radar models and
the experimental data.
  In mapping the substructure  of a number of low-level
nuclear waste disposal sites,  Geo-Centers has had the op-
portunity to test the model comparing both the calculated
and the measured depth of penetration and attenuation of
the radar system. Such a correlation is  illustrated in Figure
6  in which  the depth of penetration and attenuation
measured with radar frequencies between 10 and 300 MHz
are compared with calculated values. The range of expect-
ed attenuation for a  number of typical soils and rocks is
also  indicated  on the figure. The data indicate that at-
tenuations range from  2.5dB/m to 25dB/m for the soils
encountered at the sites. The relationship of attenuation to
depth of penetration encountered  in the  field  surveys
agrees well with that calculated using the model.(1)
   In order to simulate a typical uncontrolled disposal site
in New England and obtain radar signatures from an as-
sortment of objects in known orientations and configura-
tions and at known depths, a  site was developed near
Harvard, Massachusetts  (about 25  miles  west of Bos-
ton).(1) The site chosen was in an undisturbed portion of an
active sand mining operation.
   A trench 150 feet long and 20 feet wide was excavated to
a depth of 10 feet, except for one 20 foot section, 20 feet
deep, in a glacial outwash deposit of well-sorted,  cross-
bedded sand. Sixteen 55-gallon metal barrels, two plastic
drums and a one meter cubic box were buried at the site.
                                                                                            GROUND SURFACE
                                                                              Figure?.
                                                        Portion of a Radar Profile Taken at 300 MHz Over a Vertical
                                                        Metal Barrel (A) and a Horizontal Metal Barrel (B) in a Trench
                                                                            10 Feet Deep.
 W
 w
 Q
     hi
     2-
   I
   I
   I
3-
   I
   I
   i
                                              GROUND SURFACE

                                             • • • • JJUUIJIP • • • • •JLJLJULP • • •
       !   <-
                      TRENCHED AREA
UNDISTURBED
                                                     Figure 8.
  Apparent Resistivity Pseudo-section Across Trench. Resistivity Values Range From 190 ohm-m at the Location of a Metal Barrel (White
                       Area Indicated as "B") to 2400 ohm-m (Black). Horizontal Distance is 15 Meters.

-------
164     REMOTE SENSING
The metal barrels  were representative of typical  objects
found at waste disposal sites. The two plastic barrels and
wooden box were representative of voids of different size
and shape within the soil.
  Comprehensive data were taken with three antenna sys-
tems at  10  MHz,  80 MHz and 300 MHz. Radar scans
were made  along  the  length of the trench  at approxi-
mately five  foot intervals. In addition, one scan perpen-
dicular to the  length of the trench and one diagonal scan
were made  over each  of the targets at each frequency.
Electrical resistivity data were obtained to supplement the
radar data.
  An example of  a radar profile taken over two buried
55-gallon metal  drums, one on its  end (A) and  one on
its side (B) is given in Figure 7. All fifteen target  config-
urations, both metallic and non-metallic, were detected
with the radar. Resistivity data at  this site  are presented
in Figure 8  as a computer plotted pseudo-section show-
ing position across the trench and  apparent  depth.  The
resistivity values differ markedly between the disturbed
area within  the  trench  and the undisturbed geology. In
addition, the effect of the presence of a metal barrel on
the resistivity values is also demonstrated.
SUMMARY

  A correlation has been found between the data  ob-
tained  using ground penetrating radar and electrical re-
sistivity. These  two geophysical techniques provide com-
plementary geophysical information.
  Ground  penetrating  radar,  coupled ^with  ground-
truth,  can be used to give greater meaning to  apparent
resistivity values, both with respect to geological  substruc-
ture identification and to target or object location and
identification. Measurements of bulk soil resistivity can be
used to predict the expected depth of penetration of the
ground penetrating radar. Other geophysical techniques,
such as induced electro-magnetism and induced  polariza-
tion are expected to show similar correlations with ground
penetrating radar.

REFERENCES

1. Horton, K.A., Morey, R.M., Beers, R.H., Jordan, V.,
   Sandier, S.S., and Isaacson, L., U.S. Nuclear Regula-
   tory Commission, "An Evaluation of Ground Pene-
   trating Radar for Assessment of  Low-Level Nuclear
   Waste  Disposal Sites," NUREG/CR-2212 (to be pub-
   lished).
2. Mooney,  H.M.,  "Handbook  of  Engineering  Geo-
   physics, Vol. 2: Electrical Resistivity," Bison Instru-
   ments, 1980.
3. Annan, A.P. and J.L. Davis, "Radar Range  Analysis
   for Geological Materials," Report of Activities, PartB,
   Geological Survey Canada, 1977.
4. Ridenour, L.N. (ed.),  "Radar Systems Engineering,"
   McGraw-Hill, New York, N.Y. 1947.
5. Birchak, J.R., Gardner, C.G., Hipp, J.E., and Vistor,
   J.M.,  "High Dielectric Constant  Microwave Probes
   for  Sensing Soil Moisture," Proc.  IEEE, 62, No. 1,
   1974.

-------
UTILIZATION AND ASSESSMENT OF A PULSED RF SYSTEM
                    TO MONITOR SUBSURFACE LIQUIDS
                                    ROBERT M. KOERNER, Ph.D.
                                    ARTHUR E. LORD, JR., Ph.D.
                                          JOHN J. BOWDERS
                              Department of Civil Engineering and Physics
                                            Drexel University
                                       Philadelphia, Pennsylvania
INTRODUCTION

  Hazardous liquids infiltrating the ground from shallow
or deep injections, storage lagoons, leaking  containers,
tank car spills, etc.,  often  reach the groundwater table
where they then flow downstream to open water, streams
or wells. The necessity of locating  and monitoring such
seepage trajectories  is of paramount importance. Obvi-
ously, this can be done by core borings and  monitoring
wells, but the economics of the situation begs for a more
suitable and cost effective technique. It seems that one of a
number of nondestructive testing (NDT) methods might be
adopted for such  monitoring.  Candidate  techniques,
grouped according to their basic operating principle,  are
as follows:
          •Thermally Induced Methods
          infrared
          heat pulse
          •Elastic Wave Methods
          seismic refraction
          seismic reflection
          ultrasonics
          acoustic emission
          sonar (active and passive)
          •Nuclear Methods
          neutron
          •Electromagnetic Methods
          resistivity
          very low frequency
          eddy current (metal detector)
          pulsed radio frequency
          continuous microwave
          magnetometer
          penetrating radiation
  These methods have  been reviewed as far as their
overall  applicability to hazardous materials problems,01
and the electromagnetic methods have  been  categorized
and compared in a unified manner for location of subsur-
face anomalies, mainly buried drums.(2) This paper is an
extension  of one  of  these electromagnetic methods,  the
pulsed radio frequency (RF) method, for the  purpose of
locating the water table  in  different soil types and over
the complete seasonal change of one year. Water surface
measurements via water observation pipes were also made
so as to obtain "ground truth" information and thus pro-
vide an actual comparison.
OVERVIEW OF THE PULSED RF TECHNIQUE

  A considerable amount of subsurface probing at shal-
low depths has been based on the transmitting of pulsed
RF waves in the frequency range from about 1 MHz to
200 MHz. The transmitted pulse travels through the soil
until it infringes upon an object or material with dissimilar
electrical characteristics. Part of it is then reflected back to
the ground surface where it is received and  the time of


                       Table I.
       Details of Pulsed Ratio Frequency (RF) Methods
             Approx.
             Freq.
             Range
Investigator(s)   (MHz)
Cook (4, 5)
Rosetta (6)

Morey (7)
1-100
100-200

100-200
Dolphin,
 etal. (8)       15-50
Unterberger (9)  230
Harrison (10)    35
Rubin
 etal. (11)      100-200
Rubin and
 Fowler (12)    100-200
Benson and
 Glaccum
 (13,14)       100-200
         Max.
         Depth
         (meters)
 225
  15

  15


  40
 500
2000

  10

  15
          10
Sandness
 etal. (15)      100-200     10

Alongi(16)     1,000       3
Moffat and
 Puskar (17)    6,000       3
          Major
          Application
          Area
locating faults, walls, holes
locating faults, caverns,
water, utilities
locating faults, caverns,
water, utilities

locating rock cavities
salt thickness measurement
determining ice thickness

detecting subway tunnels

drilling guidance, subway
tunnel monitoring, coal
thickness
          general subsurface probing,
          locate and follow pollutants
          in ground, detection of
          buried containers of in-
          dustrial wastes

          general subsurface probing
          as described above
          locating mines, pavement
          thickness, shallow voids,
          pipelines

          locating faults, joints,
          cavities, pipelines
                                                   165

-------
166  REMOTE SENSING
travel is measured. The depth "d" to the interface is then
calculated from d = (vt)/2 where "v" is the wave velocity
(which is equal to c/y/T, where "c" is the velocity of
light and  " er" is the relative dielectric constant of the
material in which the wave is propagating) and "t"  is the
pulse travel time. The relative dielectric constant of num-
erous soils at different water contents has been evaluated
by many investigators, including Okrasinski, et a/.(3) As
shown in Table I, this technique has been studied by many
workers.
  Additional systems of a similar type as just described are
also available at a higher frequency range, i.e.,  above 1
GHz. These are also  listed in Table I. They could, how-
ever, better be classified as microwave methods.  The ad-
vantage of the higher frequency range is that a shorter
wave length gives greater definition to the subsurface ob-
ject. However, the attenuation of the waves is higher, re-
sulting in lower penetration depths, see Lord, et a/.(ll)
  The authors have been active in adapting the pulsed RF
method to problems of  interest in hazardous materials
spills and  incidents. Using a commercially available  sys-
tem which transmits  RF pulses in the 100-300 MHz fre-
quency range and presenting the received signal in  a real
time  printout, a visual  subsurface profile is developed.
Strong return signals appear as the black areas, while weak
(or no) return signals appear as white. Gray areas require
appropriate interpretation. The particular system used in
this study transmitted short (  10~" sec. long) pulses of
carrier  frequency about  120 MHz into  the  ground.  Re-
flected patterns were recorded in real time printout as  pre-
viously discussed. Maximum depth of penetration  at the
sites under investigation  was approximately 6 meters.  The
specific goal was to reflect the RF pulses off the  water
table surface, thereby determining its depth under the wid-
est possible set of conditions and circumstances.
          DETAILS OF TEST SETUP
          AND SITE CONDITIONS

            Each of the three sites selected for monitoring were un-
          paved and originally thought to be relatively homogeneous
          within the upper 3 m of interest. To guarantee the near-
          surface existence of the water table, each site began at a
          creek or river at one end of the traverse and proceeded
          inland. One site went from exposed creek to exposed creek.
          Shown in Figure 1 are photographs of the three sites with
          the water bodies visible in the background.

          Wharton Site
            The Wharton site extends from a 2 m deep cedar water
          creek onto the land for a distance of 100 m. The ground
          surface rises 3 m within this length. Nine water observa-
          tion wells were placed in the ground, each deep enough to
          intercept  the water  table. The  water  observation wells
          were 2.5 cm diameter  open ended pipes with removable
          surface caps to prevent damage or vandalism during time
          of non-use. The spacing varied from 3 m to IS m.
            Subsurface sampling from the site showed that  the soil
          beneath the upper IS cm was a poorly graded sand which
          contained a well defined water table, below which the soil
          was saturated and above  which the  soil was dry or parti-
          ally saturated depending on the recent weather conditions.
          Physical properties of the  soil are given in Table II. Geo-
          technical properties calculated from the date of Table II
          show an in situ porosity of 0.38 and a saturation of 9%
          for the soil above the water table.
          Springfield Site
            The Springfield site extends between  two shallow fresh-
          water creeks spaced 90 m apart. Between these two creeks
          the  land surface rises approximately 1.5 m. The  ground
             (a) Wharton Site
(b) Springfield Site
(c) Penrose Site
                                                     Figure 1.
                                   General Surface Conditions of Sites Being Monitored

-------
                                                                                       REMOTE SENSING
                                                    167
                                                     Table II.
                               Physical Properties of Soils Encountered at Various Test Sites
Location
and
Stratum
Wharton Site
Upper
Lower
Springfield Site
Upper
Lower
Penrose Site
Riverside
Lands id e
Unit
Weight
PCF (g/cc)

111 (1.78)
105 (1.68)

107 (1.72)
125 (2.00)

108 (1.73)
Water
Content
%

2
2

13
23

29
Specific
Gravity
g/cc

2.68
2.68

2.70
2.67

2.20
Effective
Size
mm

0.25
0.10

0.065
N/A

0.19
0.15
Coef. of
Uniformity

18
5

11
N/A

130
20
Liquid
Limit

N/A
N/A

N/A
56

N/A
51
Plastic
Limit

N/A
N/A

N/A
36

N/A
51
Plasticity
Index

N/A
N/A

N/A
20

N/A
0
Classi-
fication*

SW-GP
SP

SW
OH

GW/SW
MH
 N/A = not applicable
 •refers to Unified Soils Classification System
surface is an asphalt paved parking lot with a  crushed
stone base.  Total  pavement thickness  is approximately
15 cm. Eleven water observation wells, of the type pre-
viously described,  were placed along the traverse at in-
tervals of 4 to 15m.
  Soil samples indicated  two  different soil types:  The
upper soil (i.e. beneath the asphalt/stone base pavement)
is a well graded silty sand and the lower soil is an organic
clay of high plasticity. Physical properties of the soils are
given in Table II. Calculated values from these properties
show the upper sand to have a porosity of 0.44 and per-
cent saturation of 45 %, while the lower clay has a porosity
of 0.39 and is fully saturated.

Penrose Site
  The Penrose site  extends from the edge of a major river
(brackish water) up a relatively sharp ridge to gradually
upward sloping land. The traverse selected was 50 m long
and  four water observation pipes  were placed along its
length at spacings from 5 to 10 m.  Soil samples indicated
two different soil types: Closest to the river is a random
fill consisting of cobbles, gravel and miscellaneous fines
which was too coarse to obtain an  undisturbed  sample.
Awaf from the river the soil abruptly changed due to past
CQirstiSiction activity.  The  soil  here was identified as a
high plasticity silt whose physical properties are given in
Table II. The resulting in situ properties of the landside
silt were a porosity  of 0.39 and a saturation of 100%, i.e.,
fuUy saturated. It was observed that the riverside random
fill was also  high in water content and obviously close to
complete saturation.

TEST RESULTS

  During the course of this 12 month study, 20 separate
scans were taken of each test site. Temperatures ranged
from summer high of 29 °C to winter lows of -6°C. Sur-
face conditions varied from dry, to wet, to  snow covered
and included varying amounts of frost penetration. Time
of monitoring varied from morning, to noon,  to late af-
ternoon.
  Actual water table elevations were taken by placing a
stiff coaxial wire down the water observation pipe until it
came into  contact  with  the  water within  the  pipe.  The
shielding of the wire was drawn back from the central lead
about 2 mm so that a shorted condition existed until con-
tact was made with the water surface within the pipe. The
cable was connected to a voltmeter at the ground surface
which  clearly registered a completed  circuit,  hence  the
water surface. The length  of wire was then measured to
determine actual  elevations from the top of the water  ob-
servation pipes.
  Pulsed RF traces of the various  sites were made  im-
mediately after water table measurements. The  entire pro-
cess at each site took less than one hour, thus  changes in
water table elevation between the two sets of readings were
not considered significant.
Wharton Site
  A typical pulsed RF trace at the Wharton site is shown
in Figure 2. As with all of the traces it appears as though
the ground surface is level while, in reality, it rises from
left (the creek) to right (landside). Also noted is a heavy
black band immediately beneath the ground surface where
no data can be obtained. This "main bang" effect is due
to the interaction of the  three pulses traveling directly
across  the transceiver. Its thickness depends  upon  the
frequency of the pulses and varies from 20 to 40 cm.
  Seen reasonably clearly on the trace is a separation from
light (above) to dark (below), which indicates the location
of the water table. Superimposed on the figure are the lo-
cations of the water surfaces within the water observation

-------
168    REMOTE SENSING
                WHARTON  SITE
      WATER OBSERVATION PIPES
                       (J)  (8)
                                        DEPTH
                                        —rO»
                                          In
                                          2m
                               'irtW  i.j  J
                     IOO m
                    Figure 2.
   Pulsed RF Survey of Wharton Site made on April 3,1981
       with actual water table elevations indicated by
               horizontal line in circle
pipes. Correlation has generally been good with accuracy
of 14% during dry ground conditions and 27% during
conditions of high surface moisture. This latter situation
occurred usually during winter readings when the ground
was frozen in the upper 20 to 50 cm. Interestingly, the
ground surface having a snow or ice covering did not ser-
iously influence the readings. Readings taken during dif-
ferent times of the day and temperatures above freezing
did not appear to influence results.
 Springfield Site

  A typical pulsed RF trace of the Springfield  site is
 shown in Figure 3. Here a very different situation is seen to
 occur where no correlation exists between the RF trace
 and the actual water table elevation. This is undoubtedly
 due to the capillary zone which draws moisture above the
 water table in fine grained soils. Since the degree of satur-
 ation in this zone is very high, e.g., completely saturated in
 the lower soil, no distinct boundary exists for the  pulsed
 RF system to define a distinct water table elevation. How-
 ever, within the water observation pipe, the capillary zone
 does not exist, thus the actual water table surface can be
 located.
  What is dominant on the pulsed RF trace of Figure 3 is
 an  undulating discontinuity  showing two well defined
 areas. This discontinuity maps out the distinction between
 the upper granular soil and the lower fine grained soil.
 It was present in this exact location every time that a tra-
 verse was made. The only different noted from time to
 time was in contrast, which was probably due to varying
 performance of the printer.
Penrose Site
  A typical pulsed RF trace of the Penrose site is given in
Figure 4. As with the Springfield site, the capillary zone,
with its high moisture content soils, completely masks
out the presence of the actual water table as determined by
the water observation pipes. What is clearly seen is that the
soils on the left side (adjacent to the river) are very dif-
ferent than the soils on the right side (the inland soils).
  Subsequent soil testing showed these two soil types to be
very different from one another and placed in their respec-
tive locations during past construction activity. The boun-
dary was always present, in the same location, with con-
trast being the only difference between various readings.
SUMMARY AND CONCLUSIONS

  Of the many candidate techniques available to nonde-
structively test for subsurface water, as in seepage plume
detection, the use  of pulsed  RF methods are likely
methods. Indeed, the reflection from a well defined water
table can lead to  reasonable accuracy. Results of 12
months of testing show that the greatest accuracy, approx-
imately 14%, is during dry summer weather where there
is no surface moisture. The accuracy decreases to 27%
during  wet winter  months when  moisture is contained
within the soil above the water table. These results, how-
ever, are only valid  for sandy (or gravelly) soils which do
not contain  a  capillary zone. This capillary zone which
                SPRINGFIELD SITE
  WATER OBSERVATION PIPES
 £"££JLs
 ^^^^^W^^^^p^ JM. JH^^^^III^^^I
 Mw'Vh'X
 M;«^j   K.MJ
    '•fcl
lift'Mil
                   SO m
                     FigureS.
   Pulsed RF Survey of Springfield Site made on April 3,1981
        with actual water table elevations indicated by
                horizontal line in circle

-------
                                                                                    REMOTE SENSING    169
                     PENROSE  SITE

                WATER OBSERVATION  PIPES
                      Figure 4.
    Puked RF Survey of Penrose Site made on April 3,1981
        with actual water table elevations indicated by
                 horizontal line in circle

varies in thickness above the water table is saturated, or
nearly so, and has a theoretical height of:
         T
h,. = cosaj^   1

         yu   dv

where
h,.   = height of capillary rise
ft    = wetting angle of water to soil
T0   = surface tension force of water to soil
Tw   = unit weight of water
d,   = average soil void diameter
Thus, height of capillary rise is inversely proportional to
the average soil void diameter. Easily seen is that coarse
grained soils (gravels,  sand and their mixtures), with large
particle sizes hence proportionately large void sizes, have
very  little capillary zones, e.g., the soils at the Wharton
site.
  This is not the case  for the Springfield and Penrose
sites, both of which had predominantly fine grained soils
(silts, clays and their mixtures). At both of these sites the
actual water table was not possible to locate and was com-
pletely dominated by stratigraphic  changes in soil layers.
Seen at the Springfield site was a undulating horizontal
layering, while at the Wharton site the distinction was
between two different soils placed beside one another.
  In  conclusion,  the use of  pulsed RF methods can  be
used in seepage detection in granular soils, but not in fine
grained soils.  This  is not a particularly disastrous limi-
tation because it is the granular soils which have high per-
meability, hence seepage problems are more acute in these
soil types. For the fine grained soils a spill, or other inci-
dent involving  liquids, will be contained relatively close
to the  spill location due to the much lower soil permea-
bility. Thus, in its proper setting, pulsed RF methods can
be used for groundwater monitoring and the related prob-
lem of spill detection.

ACKNOWLEDGEMENTS

  This project has been supported by the U.S. Environ-
mental Protection Agency Cooperative Agreement (No.
CR-804763) administered through the Municipal/Environ-
mental Research Laboratory in Edison, New Jersey. Our
sincere thanks to the Agency, and in particular, to the
Project Officer, John E. Brugger.

REFERENCES

 1. Lord, A.E. Jr., Tyagi, S. and Koerner, R.M., "Non-
    Destructive Testing (NOT) Methods Applied to En-
    vironmental Problems Involving Hazardous Materi-
    als  Spills,"  Proc.  1980 Natl.  Conf. on  Control of
    Hazardous Material Spills, May 13-15, 1980, Louis-
    ville, Kentucky, 174-179.
 2. Lord, A.E.  Jr., Koerner, R.M. and Brugger, J.E.,
    "Use of Electromagnetic  Wave Methods to Locate
    Subsurface Anomalies," Proc.  U.S. EPA Natl. Conf.
    on Management of Uncontrolled Hazardous  Waste
    Sites, October  15-17, Washington, DC, Hazardous
    Materials Control  Research  Institute, Silver Spring,
    Md., 1980, pp. 119-124.
 3. Okrasinski, T.A., Koerner, R.M. and Lord, A.E. Jr.,
    "Dielectric Constant Determination of Soils at L-
    Band Microwave Frequencies," Geotechnical Testing
    Journal, ASTM, 1, 1979,134-140.
 4. Cook, J.C., "Seeing Through Rock With Radar," in
    Proc. of the North American Rapid Excavation and
    Tunneling Conference, Port City Press, Baltimore,
    1972,89-101.
 5. Cook, J.C., "Ground Probing Radar," Proc., Sub-
    surface Exploration for Underground Excavation and
    Heavy Construction, American Society of Civil En-
    gineers, New York, 1974,172-194.
 6. Rosetta,  J.V.,  Jr., "Detection of Subsurface Cavi-
    ties by Ground  Probing Radar," Symp. on Detection
    of Subsurface  Cavities by Surface Remote Sensing
    Techniques, U.S. Army Waterways Experiment Sta-
    tion, Vicksburg, MS, 1977,120-127.
 7. Morey, R.M., "Continuous Subsurface Profiling by
    Impulse  Radar,"  Proc: Subsurface Exploration for
    Underground Excavation  and Heavy  Construction,
    American Society of Civil Engineers, New York, N.Y.
    1974, 213-232.
 8. Dolphin, L.T.,  Bollen, R.L., and Oetzel, G.N., "An
    Underground Electromagnetic Sounder Experiment,"
    Geophysics, 39, 1974,49-55.

-------
 170     REMOTE SENSING
 9. Unterberger, R.R.,  "Overview of  Electromagnetic
   Methods in Geophysics and Application of Radar to
   Detection of Cavities in Salt," Symp. on Detection of
   Subsurface  Cavities  by Surface Remote  Sensing
    Techniques, U.S. Army Waterways Experiment Sta-
   tion, Vicksburg, MS, 1977, 119.
10. Harrison, C.H., "Reconstruction of Subglacial Relief
   from Radio Echo Sounding Records," Geophysics,
   35, 1970,1099-1115.
11. Rubin,  L.A., Griffin, J.N., and Still, W.L.,  "Sub-
   surface Site Investigation by Electromagnetic Radar,
   Phase I Report: Feasibility," Report NSF/RA-76-
   0187, ENSCO, Inc., Springfield, VA., March 1976.
12. Rubin,  L.A.,  and Fowler, J.C.,  "Ground-Probing
   Radar for Delineation of Rock Features," Engineer-
   ing Geology, 12, 1978, 163-170.
13. Benson, R.C. and Glaccum, R.A., "Remote Assess-
   ment of Pollutants in Soil and Groundwater," Proc.
   of the 1979 Conference on Hazardous Materials Risk
   Assessment, Disposal and Management, Information
   Transfer, Inc., April 1979, Miami Beach,  FL,  188-
    194.
14. Benson, R.C. and Glaccum, R.A.,  "Radar Surveys
    for Geotechnical Site Assessment," Proc. ofConf. on
    Geophysical Methods in Geotechnical  Engineering,
    Atlanta, Oct. 23-25,1979, Preprint #3794.
15.  Sandness, G.A., Dawson, G.W., Mathieu, T.J. and
    Rising, J.L., "The Application of Geophysical Sur-
    vey Techniques to  Mapping of Wastes in Abandoned
    Landfills," Proc. of the 1979 Conference on Hazard-
    ous Materials Risk Assessment, Disposal and Manage-
    ment, Information Transfer, Inc., April, 1979, Miami
    Beach, FL.
16.  Alongi, A.V., "Pavement Thickness Measured, Voids
    Detected by Downward-Looking Radar in New York
    Test," Industrial Research News, Fall, 1973,36-39.
17.  Moffat, D.L. and  Puskar, R.J., "A Subsurface Elec-
    tromagnetic Pulse  Radar," Geophysics, 41, 1976, pp.
    506-518.
18.  Lord, A.E., Jr., Koerner, R.M. and Reif, J.S., "De-
    termination of Attenuation and Penetration Depth of
    Microwaves in  Soil," Geotechnical Testing Journal,
    ASTM, 2, 1979, 77-83.

-------
      INTEGRATION OF REMOTE SENSING TECHNIQUES
       WITH DIRECT ENVIRONMENTAL SAMPLING FOR
INVESTIGATING ABANDONED HAZARDOUS WASTE SITES
                                     ROBERT W. PEASE, JR.
                                      The MITRE Corporation
                               Resource Recovery and Energy Systems
                                      Bedford, Massachusetts
                                       STEPHEN C. JAMES
                               U.S. Environmental Protection Agency
                           Solid and Hazardous Waste Research Division
                                          Cincinnati, Ohio
INTRODUCTION

  Abandoned hazardous waste sites present varying de-
grees of difficulty to  investigators. For example, aban-
doned sites which are really extensive, rural (with hinder-
ing vegetation) or in areas  of complex geology and hy-
drology,  represent troublesome  environments  for in-
vestigation. Therefore, it  is important to  develop ap-
proaches for thorough, but rapid and cost-effective as-
sessments  of these difficult situations.  In most cases, a
well designed and executed investigative program will in-
clude remote  sensing techniques  in addition  to direct
measurement. Premature action to drill wells, collect and
analyze  various  air, water and soil samples or perform
excavations without careful planning and proper integra-
tion of  available techniques may result in an unneces-
sary exposure to hazardous conditions and in an inac-
curate or incomplete understanding of the total problem.
  Remote sensing techniques may be used to provide
reasonably accurate assessments of subsurface contamina-
tion, the location and extent of buried drums and other
data needs for  determining  -appropriate  methods of
abatement. However, not all the critical information can
be obtained remotely, since each of the techniques has
limitations, both theoretical and  site-specific and conse-
quently, direct sampling should be undertaken at every
uncontrolled hazardous waste site.
  To accomplish site investigations in the most efficient
manner, a systematic  approach is necessary to take ad-
vantage  of the information that  can be extracted from
remote sensing methods.  In addition,  a systematic ap-
proach allows a reduction in the time and cost and an in-
crease in the effectiveness of direct  sampling.
  In general,  the following two  objectives must be ad-
dressed by all investigations at uncontrolled hazardous
waste sites:
1. Determination of the nature and extent of the problem
  and the resulting effects on public health (both actual
  and potential)
2. Determination of environmentally  sound  and cost-
  effective methods to effectively abate the problem (if
  abatement is deemed necessary).
The  first activity  of  any investigation should be the
identification of the specific data needed to meet each
objective. After this has been accomplished, the various
techniques available for data acquisition, both remote
and direct, can be evaluated with regard  to the type of
information that can be obtained from each in relation
to the specific conditions at the site.
  Although not always the case, it may be reasonably as-
sumed that remote sensing techniques should be used in
advance of the more direct  data acquisition methods of
borings or excavations. This is not intended to imply, how-
ever, that all direct sampling should be held in abeyance.
There have been numerous instances in which emergency
action is dependent upon immediate  results from air,
water and soil sampling, and for such cases remote sensing
techniques should be used secondarily.
  The purpose,  advantages and limitations of  each of
the four remote sensing methods discussed in this paper
are summarized in Table I. This material should be con-
sulted prior to the development of an investigatory pro-
gram. Even though there are disadvantages inherent to
each  technique,  proper  sequencing  and phased studies
can potentially result in an overall optimized approach.
  As the study progresses,  preliminary conclusions will
necessarily be modified and  the nature of direct  sampling
activities will need to be evaluated continuously.  Final
conclusions should not be drawn solely from the  results of
remote  sensing methods. Direct sample  collection should
be undertaken for all studies.
  Because no single procedure would be appropriate for
all site  investigations,  the conditions at an uncontrolled
hazardous waste site in Coventry,  Rhode Island will be
used  for the basis of development of systematic pro-
cedures  with  the expectation that some of the  concepts
can be applied elsewhere, as appropriate.
  The sequence of activities discussed on the  following
pages represents the idealized case and may not neces-
sarily have been followed in actual practice during an in-
vestigation performed  by MITRE at the above  site. This
is due to the fact that one of the purposes of the site in-
vestigation was  to evaluate the capabilities and limita-
tions of the selected remote sensing methods. The exact
                                                 171

-------
 172     REMOTE SENSING
                                                                  Table I.
                                                Comparison of Remote Sensing Techniques
                       Technique
                                                  Purpose
                                                                             Advancages
                                                                                                          Limitations
                 Electrical Resistivity

                  Lateral Profiling
                                        • determine lateral extent of
                                          contaminated ground water

                                        • facilitate placement of mon-
                                          itoring wells and optimize
                                          their number
• procedure less expensive than
  drilling

• procedure more rapid than
  drilling

* equipment light-weight, able
  to be hand carried
• limited ability to detect
  non-conductIv? pollutants

• technique  unsuitable If no
  sharp contrast between con-
  taminated  and natural ground
  water
                                        • monitor changes in plume
                                          position and direction        ^ ^^ ^ ^ conducted tn     . Interpretation difficult If

                                                                      vegetated anas                water table la d«P
                                                                                                 • Interpretation difficult If
                                                                                                   lateral variations In stra-
                                                                                                   tigraphy exist

                                                                                                 • interpretation difficult if
                                                                                                   radical changes in topogra-
                                                                                                   phy are not accounted for it
                                                                                                   choice of A-spaclng

                                                                                                 * technique unsuitable in paved

                                                                                                   ductlve objects

                  Depth Profiling          * indicate change in contamina-           same as  above                  aame as above
                                          tion with depth

                                        • establish vertical control in
                                          areas of complex stratigraphy

                Seismic Refraction (Non-    • determine depth and topogra-    • procedure less  expensive and    • technique unsuitable if no
                  explosive Method)          phy of bedrock                 safer than coring or excavation   aharp velocity contrast be-
                                        • determine depth of trench      • procedure more  rapid than        twe*n units of Interest
                                          containing buried  drums          coring or excavation            <••*•• trench  containing
                                                                                                   burled drums and surrounding
                                                                    • survey may be conducted In       aoil)
                                                                      vegetated areas
                                                                                                 • survey requires access road
                                                                                                   for vehicle

                                                                                                 * depth of penetration varies
                                                                                                   with strength  of energy
                                                                                                   source

                                                                                                 • low velocity unit obscured
                                                                                                   by overlying high velocity
                                                                                                   units

                                                                                                 • interpretation difficult in
                                                                                                   regions  of complex stratigra-
                                                                                                   phy

                Metal Detection            • locate areas of high metal     * procedure less  expensive and    • technique unsuitable for the
                                          content (e.g., buried drums)      safer thnn excavation or radar    detection of non-metallic

                                                                    * procedure more  rapid than ex-     o j c
                                                                      cavation or radar             • technique unsuitable for ob-

                                                                    . equipment light-weight, able      JectB bel°H flve feet
                                                                      to be hand-carried            • technique unsuitable for dt-
                                                                    • survey may be conducted In       termination of number or ar-
                                                                      vegetated areas                ranyement of buried objects

                Ground-Penetrating Radar    e locate burled objects (e.g.,    • procedure less  expensive and    * technique unsuitable for
                                          buried drums)                  safer than excavation            vegetated areas

                                        • provide qualitative infor-     • procedure more  rapid than ex-   • data requires  sophisticated
                                          nation regarding drum density     cavation                       Interpretation

                                        • detect Interfaces  between      • procedure deeper-penetrating    • underlying object* obscured
                                          disturbed and undisturbed        than metal detection            by those above
                                          soil (e.g.,  bottom of tren-           ,      ,,
                                          Ches)                       • procedure yields more infor-    • survey requires access road
                                                                      macion than metal detection       for vehicle
                                        • detect plunes of high chemi-           ,      .
                                          c.l concentration            ' P""-1-" "X *•« ««
-------
                                                                                            REMOTE SENSING     173
is helpful to keep them both in mind  separately when
planning an investigation. The integration of remote and
direct sampling methods is therefore presented in the con-
text of meeting these two separate goals.

DETERMINATION OF NATURE AND
EXTENT OF PROBLEM

  The following items had  to be determined for a com-
prehensive  understanding of the nature and extent of the
problem at the Coventry site:
•Direction,  rate, and extent of subsurface migration of
 contaminants
•Location  of surface  discharge  areas  of contaminated
 groundwater and the subsequent fate of the contaminants
•Identification of contaminants
•Location of areas of contaminated soil and buried drums
 and determination of the potential of this source of pol-
 lutants for long-term release.
  An example  of a  systematic approach  to  achieve the
above objectives is presented in Table II and the phasing of
site activities is shown by Figure 1. In Figure 1  a two-phase
monitoring-well  installation  and  sampling   program  is
shown.  The purpose of a phased investigation is  to ob-
tain a preliminary understanding  of the problem prior to
final planning of all direct sampling in order to more ef-
fectively guide  subsequent  activities.  Additionally,  this
figure shows that,  in general, remote sensing  precedes
direct sampling in order to reduce the time  and cost of
the latter and to help ensure  that  the full extent  of the
situation is identified.
                              For buried drums, metal detection, rather than ground-
                            penetrating radar should be used to locate the burial areas
                            because  of the relatively  lower cost  and  greater  porta-
                            bility of the former technique. However, radar has great-
                            er penetration  and  should  be used  in  all areas  where
                            drums are suspected but not found by metal detection.
                              Limited  excavation  may then be required to  gain in-
                            formation  about the  depth, condition and contents  of
                            the drums so that the concept of total drum excavation
                            and  chemical disposal can  be evaluated with the  other
                            abatement alternatives. Limited excavation was considered
                            feasible for the Coventry site since total drum excavation
                            did not  appear  significantly more expensive  than  other
                            abatement alternatives.
                              Surface  water,  groundwater and  soil  sampling  are
                            necessary  for  any  hazardous waste  site  investigation.
                            Monitoring wells are best  located once the  extent and di-
                            rection of  the plume have been determined by a resistiv-
                            ity survey. Monitoring wells should be placed at the fol-
                            lowing locations:

                            •Upgradient from the source of contamination (to moni-
                             tor background conditions)
                            •Outside of the plume downgradient  from the source of
                             contamination (to verify the  extent of the plume and to
                             monitor its movement)
                            •Within the plume close to the source of contamination
                             (to obtain  samples before extensive  dispersion, dilution
                             or attenuation)
                            •Within the plume at the outer extent of contamination
                             (to observe dispersion, dilution and attenuation).
                                                         Table II.
                       Systematic Approach to Determine Nature and Extent of Problem at Coventry Site
                   Objective
                                       Data Needs
                                                            Investigatory Methods
                                                                                            Comments
              1. Direction, Rate,
                and Extent of
                Subsurface Con-
                taminants
determine areal extent
of contaminated ground
water
                                   determine vertical ex-
                                   tent of contaminated
                                   ground water
v assess hydrogeologic and geo-
  logic settings of the site by
  reconnaissance and study of
  topographic maps, aerial pho-
  tographs,  and all existing
  data

• conduct vertical resistivity
  profiles in various locations
  to determine approximate
  depths of  contamination

• conduct lateral resistivity
  survey choosing A-spacing
  based on results of vertical
  resistivity profile

• Install monitoring wells in-
  side and outalde of contami-
  nated zones as defined by
  resistivity survey for con-
  firmational purposes

• conduct seismic refraction
  survey over contaminated
  area to determine depths to
  bedrock and vertical subsur-
  face profiles

• install cluster wells con-
  sisting of borings screened
  in soil and in bedrock both
  inside and outside of con-
  taminated  zones
• establish permanent field
  grid  for use by all remote
  sensing surveys; grid loca-
  tions can be transferred to
  site  map by aerial photogra-
  phy or land survey

• choose wide enough A-spaclng
  for lateral survey to mini-
  mize  influence of variations
  of anticipated depth to
  ground water from ground
  surface
                                                 • use established field grid
                                                  for seismic survey

                                                 • use results of seismic re-
                                                  fraction survey to select
                                                  most economic locations for
                                                  bedrock borings

                                                 • determine whether vertical
                                                  gradients exist and whether
                                                  contaminants are present in
                                                  bedrock fractures

                                                 • install bedrock wells as a
                                                  second-phase drilling effort
                                                  and install any additionally-
                                                  needed monitoring wells based
                                                  upon first-phase results

-------
174      REMOTE SENSING
                       Objective
                                                 Data Needs
                                                                             Investigatory Methods
                                                                                                                      Commenta
  drtc-rmine direction and
  rote of subsurface ml-
  prat ion
                              •  Install  minimum  number  of
                                monitoring wells to  define
                                ground water  flow; well lo-
                                cations  should be based on
                                results  of resistivity  survey
                                and  chemical  analytical needs
                                (BCC  third objective  In this
                                cable)
                              •  perform  In-sltu  permeability
                                ti-sts In selected monitoring
                                and bedrock wells
                                                                                                           • because the ground surface
                                                                                                             contours at Coventry site
                                                                                                             Indicate a potentially large
                                                                                                             arc of subsurface travel,
                                                                                                             wells at edge of contaminated
                                                                                                             zones were necessary to de-
                                                                                                             tail direction of outer frlnpe
                                                                                                             of pollutants

                                                                                                           • install cluster well upgradi-
                                                                                                             ont of pollutant source to
                                                                                                             dut ermine  background condi-
                                                                                                             tions
                    Location of Sur-
                    face Discharge
                    Areas of Ground
                    Water Contamina-
                    t ion and Deter-
                    mlnathm of Pate
                    ,»l P,,l lutiints
>  locate surface discharge
  areas
                                           > determine fate of pollu-
                                            tants
                                                                        • reconnoiter site;  locate  dls-     •  use  air quality  measuring
                                charge areas by sight and air
                                quality measuring devices
                                             ,  ,    .   .
                              . use results of electrical re-
                                atsttvlty and water table map
                                to identify surface- receptors
                                of subsurface dls charge

                              • conduct downstream sampling,
                                including rate of flow  of
                                surface water
                                    .     .        .  .        .
                              • sample sediment and air  around
                                discharge area
                 3. Chemical Identi-
                    fication of ill
                    Principal Contam-
                    inants
a determine principal con-     a  analyze  composite samples
  tamlnants                     from selected wells  and sur-
                                face waters  for priority
                                pollutants
                              e  analyze  selected wells and
                                surface  waters for selected
                                compounds based upon priority
                                pollutant analysis
                 it. Location of Areas
                    of Burled Drums
                    and Contaminated
                    Soil and Deter-
                    mination of Life-
                    time of Future
                    Ke leant- of Chcmi-
                    cals
                                          9 determine  dispersion of
                                            contaainants
i locate areas of burled
 druma and contaminated
 BOiL
                                         • determine lifetime of
                                           future release of chemi-
                                           cals
                              9  select  indicator analyses
                                based upon priority pollutant
                                screening: include general
                                water quality testa (pH, con-
                                ductivity, iron, chloride,
                                TOC)
                              •  reconnoiter site; search for
                                areas of disturbed soil or
                                vegetation, or areas of dis-
                                colored soil
                              a  Interview persons involved
                                with dumping activities for
                                information concerning loca-
                                tion of trenches and method
                                of operations
                              •  study previously taken aerial
                                photographs to obtain histor-
                                ic information
                              •  use results of resistivity
                                survey to locate source of
                                contaminated ground water

                              a  conduct metal detection sur-
                                vey over all cleared or dis-
                                turbed areas of site

                              a  conduct ground-penetrating
                                radar survey in burial areas
                                located by metal detection
                                survey

                              •  conduct seismic refraction
                                survey over drum burial area
                                to determine depth of druma

                              a  perform limited excavation
                                of burled druma to determine:
                                condition and contents of
                                druma, density of drums, lowar
                                boundary of drums

                             a  sample soil In drum burial
                                areas and bulk chemical dis-
                                charge area; analyu extracted
                                leachate for specific chemical
                                compounds determined by pri-
                                ority pollutant screening
                                                                  devices to determine areas of
                                                                  poor quality. Indicating sur-
                                                                  face discharge points: also
                                                                  indlcatcs areas requiring
                                                                  brcathlrlR protection devices
                                                                          " K
                                                                • mass balances should be cal-
                                                                  culated for principal pollu-
                                                                  tants to determine ultimate
                                                                  dispersion mechanlsm(s)
                                                                     K
                                                                • upstream sampling necessary
                                                                  for background conditions

                                                                • all water wells and surface
                                                                  waters used for notable water
                                                                  supplies within a mile radius
                                                                  should be precautlonarlly
                                                                  sampled

                                                                • composites made up of wells
                                                                  close to source of pollutants
                                                                  and surface discharge areas;
                                                                  should limit each composite
                                                                  to only two adjacent wells
                                                                                J
• indicator analysis provides
  relatively low coat method
  for non-specific monitoring
  of pollutant levels in moni-
  toring well network

• use established field grid
  for metal detection and ground-
  penetrating radar survey
a metal detection used in ad-
  vance of ground-penetrating
  radar because of lower cost
  and ease of use; however,
  radar has greater penetration
  and should be used in all
  areas where drums are sus-
  pected, but not found by
  metal detection
                                                                a  effectiveness  of  seismic  re-
                                                                  fraction  method to  determine
                                                                  lower  boundary of buried
                                                                  druma  remains  subject  to
                                                                  verification
                                                                a  drum excavation limited to
                                                                  data gathering only and should
                                                                  be  terminated  whan  aufflclcnt
                                                                  Information obtained;  refer to
                                                                  Phase  II  project  raport (ref-
                                                                  erenced In Section  1)  for rec-
                                                                  ommendations  regarding excava-
                                                                  tion procedures
                                                                •  drum excavation should bt con-
                                                                  ducted after  all  other eite
                                                                  activities completed in order
                                                                  to  minimise personnel, on  a tie
                       Coventry elte  procedures  used  for  illustrative purposes only;  investigatory procedure*  end  sequence  mav not
                       necessarily  be directly spplicsble to other sites.

-------
                                                                                             REMOTE SENSING     175
    H:
  Phaae I
Monitoring Welle
                                                             Bedrock Boring*
                                                                                Pnaae I
                                                                             Chewlcal Analyala
                                           - preliminary dtuer-
                                           - Jc I •mill
                                            Jtlulfer
                                            Intlta
Data Havlav






Haalatlvlty


SElBBlc Refraction
Survey
                                                                                   Ion of

                                                                                 roprlate Indi-
- alte »eonnalaa«nea

- ravlen of prior ra-
 porta and chealcal


- Interview of par-
 anal aMoclatad
 with duaplng

- atudy of topo-
 iraphlc Mpa end
 oarlal photagrapha
                 - aatabllahMot of
                  panaoent field grid

                 - production of photo-
                    letrlc tup of
                        rroundlng
>cetlon of cant a
lated ground wet I
                  •It* end a
       - dataralnetlon of
        depth to bedrock

       - determination of
        lower boundary of
        dtuoa and prella-
        Inary eeelnate of
        nunber of druni
                                                                                             Well Inatallatton
                                                                                            and Chenlcil Analyala
                                                                                                             - determination of cc
                                                                                                              tenta, depth, and
                                                                                                              - determination of
                                                                                                               final eatlMte of
                                                                                                               number of burled
                                  - location of burled
                                                   - diceralMcton of hor-
                                                    liontal boundarl**
                                                    of burled druM
                                                          Figure 1.
                                     Recommended Sequence of Activities at Coventry Site
  The depth and topography of the bedrock should be
defined, not only to evaluate certain abatement methods,
but also to economically locate the deep borings that may
be needed for bedrock sampling, permeability testing and
water sampling. Seismic refraction is a very effective tool
for providing remote information on the configuration of
subsurface strata. Additionally,  the results can be used to
locate bedrock wells, the  position of which will depend
upon the objectives of the investigation. For example,
some investigators may wish to install bedrock wells only
to determine formation integrity as an acceptable base for
certain abatement alternatives. Other investigators may be
interested in  locating  bedrock  wells  to  determine  the
presence of contaminants in specific regions or channels.

DETERMINATION OF METHODS
TO ABATE THE PROBLEM

  The data  needs for selecting and determining  the cost
of abatement alternatives (Table III) are similar  to those
required for understanding the nature  and extent  of the
problem at  the  Coventry site. However,  the two  objec-
tives are best addressed separately, since the location for
direct sampling  and remote sensing  may differ  between
the two, as may the use of the information obtained.
  It is  possible  to use  remote  sensing  techniques as  a
"negative screening" step in the evaluation  of certain
abatement options. An example of this concept is given in
the case of  evaluating source  encapsulation  as an abate-
ment technique  for the  problem at Coventry. For  encap-
sulation to be feasible it is  necessary to have a low perme-
ability base within a relatively  shallow depth from the
                                                                                         Table III
                                                                     Major Informational Needs for Implementation of
                                                                   Certain Abatement Activities at Coventry, Rhode Island
                                                               Alternative
                                              Informational Needs
                                                               Removal of Buried   Drum Condition
                                                               Drums and Dis-     Drum Number
                                                               posal of Chemicals   Drum Contents
                                                                                 Trench Location and Geometry
                                                               Encapsulation of
                                                               Source
                                                               Collection and
                                                               Treatment of
                                                               Leachate (trenches
                                                               and/or wells)
                                                              No Action
                                                               Alternative
                                              Drum Contents
                                              Imperviousness of Underlying Strata
                                              Level(s) of Contamination (soil and/or ground-
                                               water and/or bedrock)
                                              Trench Location and Geometry

                                              Areal Extent of Contamination
                                              Type of Contamination
                                              Concentration of Contaminants
                                              Imperviousness of Underlying Strata
                                              Aquifer Characteristics

                                              Drum Contents
                                              Drum Condition
                                              Level(s) of Contamination (soil and/or ground-
                                               water and/or bedrock)
                                              Type of Contamination
                                                              ground surface. Seismic refraction has  the potential to
                                                              determine whether the  bedrock underlying  the  source of
                                                              contamination should be  ruled out  (negative screen) or
                                                              consideration as an acceptable base.
                                                                 If, after a seismic refraction survey it is found that the
                                                              bedrock is either too deep or too fractured to function as

-------
176    REMOTE SENSING
an effective base, then rock coring  is unnecessary. The
information obtained  by this method is  not  sufficient,
however, to prove that the bedrock  is sufficiently sound
for encapsulation without actual test borings  and field-
permeability tests.
  There is  one investigational method for estimation of
the number of buried drums which  is listed in Table II
and not included in the discussion of the preceding sec-
tion. The remote sensing technique of seismic refraction
can determine the lower boundary of buried drums. The
results of an exploratory excavation  can be compared to
the seismic profiles obtained at a particular location and
if valid, the seismic profiles for all  other buried  drum
areas may be used for  drum number  estimates. If there is
no correlation found between the seismic profile and the
excavation (of if an exploratory excavation is not desired),
it is recommended that the maximum feasible excavation
depth or depth to bedrock be used for the lower boundary
of the buried drums.
ACKNOWLEDGEMENTS

  The work on which this paper is based was supported
under MITRE  contracts  with  the U.S. Environmental
Protection Agency, Solid and Hazardous Waste Research
Division, and the State of Rhode Island, Department of
Environmental Management. The encouragement and in-
terest of Mr. Donald  Sanning  of  the EPA is especially
noted. The contributions to the site investigation of Mr.
Paul Stoller and Dr. Harold Yaffe of MITRE and Ms.
Nancy Cichowicz,  formerly  of MITRE  and presently
with ERT, Inc., are also acknowledged.
  The actual results of the Coventry investigation are
found in a paper by Yaffe et al.(4)
REFERENCES

1. Pease, R.W., et al., "Hazardous Waste Investigation:
   Picillo  Property,  Coventry,  Rhode  Island," MTR-
   80W00032,  The   MITRE  Corporation,  Bedford,
   Massachusetts, 1980.
2. Cichowicz, N.L.,  Pease,  R.W.,  Stoller,  P.J., and
   Yaffe, H.J., "Evaluation of Abatement Alternatives:
   Picillo  Property,  Coventry,  Rhode  Island," MTR-
   80W00253,  The   MITRE  Corporation,  Bedford,
   Massachusetts, 1980.
3. Cichowicz, N.L.,  Pease,  R.W.,  Stoller,  P.J., and
   Yaffe, H.J., "Use of Remote Sensing Techniques in
   a Systematic  Investigation of an Uncontrolled Haz-
   ardous  Waste Site",  MTR-80W00244, The MITRE
   Corporation, Bedford, Massachusetts, 1980.
4. Yaffe,  H.J.,  Cichowicz,  N.L.,  and Stoller,  P.J.,
   "Remote Sensing for Investigating Buried Drums and
   Subsurface Contamination at Coventry, Rhode Island,
   Proc.  U.S. EPA  National Conference  on Manage'
   ment of Uncontrolled Waste Sites, Oct.  15-17,  1980,
   Washington, D.C., Hazardous Materials Control Re-
   search Institute, Silver Spring, Md., 239-249.

-------
       SURVEY AND ANALYSIS OF PRESENT/POTENTIAL
                      ENVIRONMENTAL IMPACT SITES
                        IN WOBURN, MASSACHUSETTS
                                          SUSAN E. TITUS
                                      The Bionetics Corporation
                                         Warrenton, Virginia
INTRODUCTION

  The U.S. Environmental Protection Agency, Region I,
initiated a request through EPA Headquarters to the En-
vironmental Photographic Interpretation Center (EPIC)
for a survey and analysis of a heavily industrialized area
(approximately 16 square miles) surrounding North Wo-
burn, Massachusetts. The purpose of the study was to es-
tablish the ambient environmental condition of this area,
where various industries have produced a high output  of
industrial and chemical products for more than a century.
Many of these industries are alleged to have dumped  or
buried spent chemicals and by-products, which are a sus-
pected source of endemic health effects in the vicinity.
  Analysis of historical cartographic  and  photographic
sources and current aerial imagery can supply continu-
ous information to  search  efforts in the field, thus reduc-
ing the time anc cost of locating test sites,  abandoned  or
unpermitted dumps, and areas of environmental damage.
Through the utilization of imagery, sites can be located
and analyzed in a  matter or hours or days by a single
analyst, as opposed to time-consuming ground searches
by inspection crews  themselves.
  Currently, field  tests are  being scheduled to coincide
with the receipt of incremental site assessments forwarded
from EPIC upon completion.
  Located north of Boston, Woburn is a town with a long
history of industrial activity. Since the 1850's, industries
have manufactured products which include: chemicals for
textile, paper and leather processing; explosives, hide and
bone glue and arsenic-based pesticides. The wastes from
these products were frequently disposed of on grounds
surrounding the factories. A tract of approximately 120
acres containing abandoned dump sites, building ruins,
waste pits and a new industrial park is the main focus of an
area-wide  survey to determine the extent of  present or po-
tential environmental hazards and ambient environmental
conditions.
  Presently owned  and controlled by prominent area de-
velopers, this tract  is a valuable piece of commercial real
estate. Its location in close proximity to 1-93  and Route 128
resulted in a surge of expansive redevelopment creating
the industrial facility known as Industri-Plex  128. Pur-
chased in  1968, the tract has yielded vast amounts of  in-
dustrial by-products during excavation for future com-
mercial sites. Public attention was drawn to this  area in
the mid-1970's when hydrogen sulfide odors were emit-
ted by rotting hides and animal wastes unearthed during
construction.  Complaints  were made to the  State  of
Massachusetts by residents of neighboring communities af-
fected by the fumes.  In a separate but coinciding inci-
dent, an EPA employee reported a possible case of illegal
wetlands filling. Unpermitted, the filling was halted; sub-
sequent testing by the EPA and the U.S. Army Corps of
Engineers revealed concentrations  of arsenic, hexavalent
chromium and lead. Research into prior land use along this
strip indicated past utilization for chemical waste disposal
in ponds and lagoons,  as well as for a dump  site  by
various tanners and Tenderers.
   Subsequent sampling of water throughout the Woburn
 area revealed two municipal  wells in North Woburn to
 be contaminated with the solvent trichlorethylene (TCE),
 a suspected carcinogen.  The source of the contamination
 by this compound is unknown.
   Woburn has experienced a  higher than normal rate of
 cancer,  approximately 13% higher than Department of
 Health Statistics predicted, especially in a neighborhood
 known as Walnut Hill located less than a mile south of
 the industrialized  corridor.  Eight of fourteen reported
 cases of leukemia have occurred within a half-mile radius
 of one another in this small residential development/0 As
 yet, no cause-and-effect correlations can be drawn for this
 occurrence.

 AREAS OF CONCERN

   Seven sites and surrounding areas, including  the indus-
 trial park, were chosen for in-depth  analysis of present
 and historical land use. Known industrial and commercial
 uses of these tracts include: the sites for tanneries and glue
 manufacture,  chemical  processing,  and extraction and
 landfilling operations.
   As of July 1981, sets of prints,  overlay series and texts
 concerning the Woburn study area have been  forwarded
 incrementally upon completion to Region I for use by EPA
 Massachusetts  Department  of  Environmental  Quality
 Engineering (DEQE) and the field inspection team (FIT).
   The focus  of critical concern from impact  by  uncon-
 tained wastes is the quality of surface and subsurface water
 in the study area. A review of historical aerial photography
 shows that major  changes have occurred in surface water
 paths, wetlands and terrain throughout the Woburn area.
                                                  177

-------
178    REMOTE SENSING
Although alteration of these features through excavation,
filling, dredging  or construction  may  re-route  surface
drainage or re-shape impoundments, subsurface water will
often resume its flow in the original stream bed, along the
path of least resistance. Fluid or viscous waste and leachate
from buried contaminants  will flow under  gravity to
groundwater levels  and form plumes along  the  stream.
These plumes may extend under many acres of ground if
the source is unchecked.(2) Locating the point source of
subsurface water  contamination can be  facilitated by ex-
amining historical land use in the drainage basin.
  Slightly less critical, because of its more stationary na-
ture, is soil contamination. Sites with possibly contaminat-
ed fill material can also be identified through historical
imagery analysis.
  In the realm of air quality, past and present offenders
may be noted  by the  presence of vents, stacks, fans and
similar exhaust structures.

METHODOLOGY
Use of Historical Cartographic Records
  Since aerial  imagery of the United States is scarce or
non-existent prior to the late 1920's, land use before this
period may be largely unrecorded. Human memory is sub-
ject  to error;  interviews with area residents on  historic
usage may yield conflicting data or place sites far  from
where they actually existed. In regions such as Woburn, it
is apparent that many major alterations probably occurred
between the  inception of industrial land  use in the 1800's
and the first available photography in the 1930's.  In order
to assess past environmental influences,  research was un-
dertaken at the U.S. Library of Congress for plots, docu-
ments  or cartographic records of  properties  which were
recorded in the late 1800's or early 1900's.
  In the Geography and Maps Division,  the Sanborn Fire
Insurance Maps(3) were the focus of attention. Developed
to record safety precautions and hazards associated with
structures in the  towns, they provide valuable informa-
tion on conditions in an area beginning in the 1800's.
  Coverage  of Woburn begins in 1888 and ends in a series
of 1926  maps corrected by paste-ons in the 1930's and
again in  1947. An example of the  type  and extent of in-
formation found  on the maps is the Merrimac Chemical
Company sheet.  In  1888, Merrimac Chemical occupied a
site in North Woburn adjacent to the railroad (Figure 1).
Each structure is plotted,  and the building  contents or
process described. At this time, the major  compounds
present were sulphur, lead, muriatic  acid and alum. By
1904 (Figure 2), the number of buildings had increased
and  substances such  as  chloride  of  aluminum,  bauxite
and  pyrite  had  been added to the  inventory.   Moving
chronologically through  the maps,  it is possible to see the
transition of Merrimac  Chemical (1888) to include New
England  Manufacturing   Company  Chemical  Works
(1918) and  its evolution into Consolidated Chemical In-
dustries Co. Glue Plant (by  the 1940's), along  with de-
tails  in process changes  and on-site materials. The most
recent  transitions were  revealed by aerial  imagery. The
site remains  a  functioning chemical process plant in 1963
(Figure 3), and is finally seen as a razed, abandoned site
in 1980 (Figure 4).
  An example of the value of these cartographic records
in extreme land-use changes is depicted by the reproduc-
tion of a block in the town of Woburn in 1894 (Figure 5).
This lot had been the site of a leather company since 1888.
In  1904, it is recorded as American  Hide and Leather
Co. Factory H  (closed); in 1918, the lot contains scat-
tered residences and open ground. By 1926, the block con-
tains many residential units.  This block, as it appeared
in 1963, is centered in Figure 6.
  In a rough preview count, 14 of 48 selected sites traced
through the maps  became  residential,  commercial  or va-
cant areas by the 1940's.  Each of the 48 sites will be
checked against  the aerial imagery for 1938 and 1963 to
determine  the continued  development  of these historic
industrial grounds.
  The  utilization of Sanborn Maps in hazardous waste
studies is also being  pioneered  in a current project for
Stroudsburg, Pa. Concerned with contamination from no
apparent sources,  Sanborn Maps checked for the years
1905, 1912, 1923, 1930 and 1950 disclosed the location of
a coaltar injection well and two storage tanks which no
longer  exist. No previous  knowledge of the location of
these features was apparent in the region.

Use of Historical Aerial Photography
  In order to  accomplish a comprehensive, detailed data
base on locations of present and former excavation and
fill sites,  streams, bogs, wetlands and impoundments, as
well as industrial locations, a survey  utilizing historical
and current aerial photography was  undertaken. His-
torical black-and-white imagery was  obtained for De-
cember 1938,  April 1963 and  April  1978, from the U.S.
Geological Survey and the National Archives. Current
true-color imagery was obtained in  November 1980 un-
der contract by EPIC.
  The  photo frames were chosen at  the scale of 1:24,000
for the historical, and obtained  at 1:10,000 for the color.
Each of the seven primary areas of concern was plotted on
overlays  to USGS quadrangle maps (scale 1:24,000) and
analyzed from the imagery for both minute and massive al-
terations which may have occurred over the 42-year span.
With the location illustrated on the map overlays, the
specific details of site variation were plotted on clear mylar
overlays  to enlarged photos of each year of coverage per
site. The scale of the enlargements varied from  site to
site. An example of drastic change, illustrating the type of
details found, is evident in Figures 7 and  8. In  1938,
(Figure 7) large lagoons and trenches for liquid waste run
along  the western boundary of the  plant property. This
same section in 1963 (Figure 8) is the site of additional
plant structures, an access  road  and a  filled lot with only
a  single  narrow trench  visible.  The surrounding locale
has experienced development and the addition of major
transportation routes.
   In order to obtain and evaluate  information such as
this from aerial imagery, film transparencies are  backlit
on a standard Richards light table and examined in stereo
through  optics  capable  of magnification  up to  30x. A

-------
                                          REMOTE SENSING    179
                   ffC     CHEMICflL    CO.
                              flc/os.
            Figure 1.
Merrimac Chemical, Sanborn Maps, 1888

-------
180    REMOTE SENSING
                                                                     [11   EH  EH    G
                                               Figure 2.
                                   Merrimac Chemical, Sanborn Maps, 1904

-------
                                                                                       REMOTE SENSING
                                                     181
                       Figure3.
             Old Merrimac Chemical Site, 1963

trained analyst examines each site  for details concerning
terrain, structures,  ground disturbance, impacted drain-
age, discarded material, land use of the locale and "sig-
natures"* of other elements of environmental influence.
In general,  examination of terrain features, structures,
drainage and  land  use is fairly straightforward;  the  de-
tection of drums,  rubble, offal or  evidence of under-
ground structures requires close scrutiny and reliance on
surrounding information for detection. All pertinent in-
formation regarding the site is recorded on the  overlays
for location;  a corresponding text contains  details and
conditions of all possible past and present impact points.
  In  cases of confirmed or suspected  waste disposal or
other pertinent environmental factors, comparisons will
be made by engineers in the region against local hydro-
logic,  geological  and air,  studies  to determine  possible
cause-and-effect correlations,  the extent  of any  effects,
and other related factors.  From this information,  fields
checks will be performed on  soil and  water quality. At
EPIC, when the photographic  analyses have been per-
formed, a computer system is utilized to locate and chart
the exact location of sites to guide  field sampling efforts.
EPIC's Imagery Analysis  System (IAS),  designed  by
Cab"8, is capable  of rectifying photo-to-map scales in

'The term "signature" is used to describe a particular pattern, shape,
tone or color which consistently indicates  the presence of an object,
material, etc. on aerial imagery, even though the object itself may be in-
distinguishable.
                        Figure 4.
              Old Merrimac Chemical Site, 1980

order to plot  points directly from  imagery to a standard
map. The system  is also  able  to determine  the  exact
geocoordinates for an individual site or set of points and
to compute the area of a chosen feature.  Using this in-
formation, the field inspection team can quickly and ef-
fectively locate  precise points  from  which  to extract
samples.
  In addition to the site-by-site analysis, a survey of the
entire Woburn study area  was performed to locate any
dumps,  impoundments,  fill  mounds, extractions,  in-
dustries or junkyards in each year of  imagery coverage.
The sites were located numerically on a  series of clear
overlays to USGS 1:24,000 quadrangle maps, and a brief
description, by site, for each year  was  included in an in-
ventory list (Figure 9). The transformation of a site from
a hillside farm in 1938 to an extraction site and back to
a filled meadow by 1980  can  thus be quickly surveyed.
A survey is  also being performed  at 10-year intervals
from the Sanborn Maps of the study area between 1888
and 1947.
  To accompany the site survey, two additional series of
overlays were developed to illustrate historical  to present
drainage transformation and land-use  patterns. Included
on  the  drainage overlays  were streams, bogs, wetlands,
natural lakes  and ponds, and man-made features includ-
ing dams, industrial lagoons  and impoundments.  This
series can provide much needed data on  original  water
courses in sections of disturbed drainage.  When used in

-------
182    REMOTE SENSING
                                                      Figures.
                                           Block of Chestnut and Scott Streets
                                                 SanbornMaps, 1894
                                                       Figure 6.
                                          Block of Chestnut and Scott Street. 1963

-------
                                                                                     REMOTE SENSING    183
                      Figure 7.
           Waste Lagoons; Industrial Site, 1938

conjunction with the site overlays, possible point sources
of subsurface water impaction  may be  located,  and
abatement procedures formulated.
  The land-use series  employed a system  of mapping
land usage categorized down to Level HI, modified from
the USGS Land Use Classification System/4' For example,
Category 2 in this system covers agricultural land use; at
Level III, sub-categories are determined and broken out
for crops (211), pasture (212), fallow  (213) and orchards
(22). The drainage series and the land-use series together
can show historical patterns of agricultural run-off, pos-


           Sitel—
             1938: Agricultural use; barn, residence
             1963: Extraction site
             1978: Extraction site
             1980: Filled, revegatated meadow
           Site 2—
             1938: Agricultural field
             1963: Auto junkyard
             1978: Construction site
             1980: Residential development

                      Figure 9.
         Sample inventory list for site comparison
                        Figure 8.
              Waste Trench; Industrial Site, 1963

sible paths of industrial waste or leachate flows, and lo-
cations of liquid dump sites. Findings such as these can
assist in  evaluating deposits of contaminants not associ-
ated with present land usage and drainage tendencies.

VERIFICATION OF FINDINGS

  Upon  the completion of the analyses of the seven spe-
cific sites and at mid-point in the area-wide site inventory,
a  ground-truthing  meeting was  held  by  the imagery
analyst and representatives of the Massachusetts DEQE
and EPA Region I office. The comparison of data ob-
tained from the four sets of imagery at EPIC and informa-
tion researched by the DEQE (including different sets  of
aerial imagery and historical information on zoning uses of
some  properties) confirmed a high level of accuracy. Field
inspection of  several sites also  yielded confirmation  of
suspected historic  use not found on the present  color
imagery. An example of this occurred during an inspec-
tion of a tannery site. Historical imagery revealed plumes
of sludge assumed to be chromium wastes along a slope at
the rear of the property. The 1980 color imagery showed
only  natural-toned  earth in  this  section.  From  the
ground at close range, however, a very light tone of pale
yellow was observed emerging in several places. Pale and

-------
184    REMOTE SENSING
bright yellow tones are sometimes indicative of chromate
leachates.
  In  another case,  historic  analysis  showed  "soft"
mounds of waste, which appeared to be hides, on the lot of
a plant. Although the dumping of hides was not reported
or suspected at this site, a walk through the abandoned
site turned up several scraps of old hide in a historically
filled area. Some information given to the state representa-
tive by local residents who recalled past dump sites and
industrial uses also  indicates a correlation with informa-
tion gained from the  imagery.  The results of samples
gathered by the  field inspection  team will determine the
accuracy of selecting sites by imagery analysis.

ASSESSMENT OF MODE OF ANALYSIS

  The procedure of site and area analysis via remote sens-
ing is an effective,  time-saving and  economical method
for determining environmental conditions in both the past
and the present. Although it has been employed in Woburn
as a locating device, imagery analysis is extremely useful in
monitoring control and clean-up operations and in  emer-
gency response for assessing areas of immediate danger.
  The utilization of historical imagery for a current na-
tional concern emphasizes the importance of maintaining
imagery obtained in the past.  It is invaluable for research
in this mode; it does not describe what may have been but
shows what actually existed at a given time and place.
   The use of historic cartographic records such as the San-
born maps, unprecedented in hazardous waste assessment,
also illustrates  the necessity for preserving  historical
records.  The existence of such  a detailed  record of ma-
terials and industrial site locations long before the hazards
of chemical wastes were  realized has contributed  much
information on potential impact sites  where no immediate
signs of danger may otherwise have been noted.

CONCLUSIONS

   To date, the impact of uncontrolled hazardous  wastes
upon Woburn has not been as spectacular nor as immedi-
ately devastating as  at Love Canal although the essence of
the situation is the same. It is the intent of those involved
in the aerial analysis of the area to  provide data  which
may prevent damage from dormant wastes, and aid in the
control and clean-up of sites known to be a problem.
  In  combination,  historical  imagery and  cartographic
references can supply information for all  areas in the
United States where past land usage may pose a potential
threat to the quality of the environment,  as has already
occurred in  the vicinity of the town of Woburn, Massa-
chusetts, in Memphis, Tennessee, and at the Love Canal,
in New York.
  As  proven by studies already completed, this methodol-
ogy is fast, effective, low in cost and it has a high rate of
accuracy when tested in the field. It is hoped that utiliza-
tion of  this  technology will become more widespread as
cities  throughout the United States examine  the potential
dangers of unsound, past environmental practices.

ACKNOWLEDGEMENTS

  The author wishes to acknowledge the assistance of the
following persons in obtaining information on the Woburn
area and developing the methodologies through which the
study is being accomplished: Richard T. Leighton, U.S.
Environmental Protection Agency, Region I; Thomas R.
Osberg, U.S. Environmental Protection Agency, Environ-
mental Photographic  Interpretation Center; and Robert
Cleary,  Massachusetts  Department of Environmental
Quality Engineering.


REFERENCES

1. Tangner, P., "Hazardous Wastes: Ghosts of a Prodigal
   Past" Technology Review, 8.2 (8), Aug/Sept, 1980.

2. Geraghty, J. J., "Evaluation of Hydrogeologic  Con-
   ditions"  U.S.  EPA Conference on Management of
   Uncontrolled Hazardous Waste Sites, Oct. 15-17,1980,
   Washington, D.C., Hazardous Materials  Control Re-
   search Institute, Silver Spring, Md., 49-52.

3. Sanborn  Map Co., Inc., The Sanborn  Fire Insurance
   Map Series for Woburn, Ma., Feb 1888, June  1884,
   Aug  1899, May 1904, June 1910, May 1918, May  1926,
   May 1926-1947.

4. Anderson,  J.  R.,  Hardy, E.E.,  Roach,  J.T. and
   Witmer, R.E., A Land Use and Land Cover Classifica-
   tion System for Use With Remote Sensor Data,  Geo-
   logical Survey Professional Paper 964,1976.

-------
                           PLANTS AS BIOINDICATORS
                      OF ENVIRONMENTAL POLLUTION
                                         G.K. SHARMA, PH.D.
                                    Department of Biological Sciences
                                    University of Tennessee at Martin
                                            Martin, Tennessee

                                          CHRISTY COOPER
                                         University of Tennessee
                                      Center for the Health Sciences
                                           Memphis, Tennessee
INTRODUCTION

  Plant morphological and cuticular features have been
used  for over a century in the interpretation of  taxo-
nomic,  paleobotanical,  and phylogenetic relationships.
However, their use in the understanding of ecological
data is fairly recent. In addition, plant cuticular features
have recently been found to be extremely useful in estab-
lishing relationships between environmental pollution and
various taxa.
  In view of the presence of acid rain and other pollutants
throughout the world, the use of plants as bioindicators
of environmental pollution in  various habitats is an ex-
tremely significant contribution to the understanding of
our ecosystems. With detailed studies on a wide variety
of plant taxa, it is possible to suggest  a  specific relation-
ship between environmental pollution and a  plant taxon
affected by it.
  Numerous studies have revealed the detrimental effects
of environmental pollution on plants under  natural and
controlled conditions. Scheffer and  Hedgcock's  study(6)
of the forests of the Northwestern United States revealed
the characteristic effects of sulfur dioxide injury on leaves.
Solberg and  Adams(9) reported that fluoride and sulfur
dioxide destroyed the spongy mesophyll and the lower leaf
epidermis of plants. This was followed by chloroplast dis-
tortion, and palisade and upper leaf epidermis damage.
Chamberlain0' observed that dirt, smoke, and the gases of
a large city were fatal to conifers, especially Pinus bank-
siana. The   plants illustrated  chlorosis and necrosis.
Pyatt(5) investigated lichens as possible indicators of air
pollution in a steel-producing town in Wales and discov-
ered that generally the lichen flora decreased in number of
species present with increasing proximity of the source of
pollution.
  Although several studies*2-4' indicate the effects  of en-
vironmental pollution on the gross morphological features
of plants, relatively little work has been done to determine
the relationship between environmental pollution and the
cuticular features of plant leaves. Preliminary studies on
the subject'7-8' indicate the usefulness of leaf cuticular fea-
tures as  indicators of  environmental pollution  in  some
plant taxa. The purpose of the present study was to deter-
mine  the gross morphological and  leaf cuticular varia-
tions in American sycamore and their possible relation-
ship to environmental pollution. The present study is,
therefore, a continuation of a comprehensive project in-
volving plants as indicators of environmental pollution.

METHODS

  American sycamore (Platanus occidentalis) is one of the
largest trees in the eastern deciduous forest of the United
States. It may grow to fifty meters in height  and trunks
may be as large as four meters in diameter. The sycamore
family (Platanaceae)  is  composed of a single  genus
(Platanus) of about seven species occurring in the North-
ern Hemisphere. The tree is characterized by broadly ovate
leaves which are 3-5 lobed. The bark of sycamore tree is
deciduous and is in the form of broad, thin brittle plates.
Sycamore grows well in rich soil such as bottom-lands. The
wood of sycamore tree is utilized for many specialized pro-
ducts such as boxes,  crates, and furniture.  Due to the
fungal disease caused by Gnomia veneta that often invades
sycamore trees, planting of the tree is currently not ad-
vised(3).
  Four populations (A, B, C,  D) of sycamore tree were
collected from different sites exhibiting varying degrees of
environmental pollution (Table I). Populations A and B
                       Tablet.
 Distribution and Habitat Features of Sycamore Populations in
Northwest Tennessee, Middle Tennessee, and Northern Missouri,
                       U.S.A.
           Loca-
Pop'n      tion

A          Reelfoot Lake,
           Tenn.
B          Overton Park,
           Memphis, Tenn.
C          Nashville, Tenn.
D          St. Louis, Mo.
                                                        •+ + + +, highest level; +, lowest level.
Rel.
Pol'n
Level
Source

agricultural
operations
automobiles

automobiles
industry
automobiles,
heavy
industry
                                                    185

-------
186    REMOTE SENSING
were  collected respectively  in  the  semi-protected, semi-
wooded vicinity of Reelfoot Lake  and Overton Park in
Tennessee.
  Reelfoot Lake represents an extremely small, rural com-
munity in northwest Tennessee with a minimum level of
pollution. Although the Overton Park area is in the city
of  Memphis, the tree samples  were collected from  a
wooded, semi-protected microhabitat unlike the polluted
macrohabitat of Memphis.
  Populations C and  D  were  collected from Nashville,
Tennessee  and St. Louis,  Missouri  respectively,  where
automobile traffic was an obvious source of environmental
pollution. The tree samples comprising populations C and
D were taken from the industrialized downtown areas of
Nashville and St. Louis, respectively. In both  of these
large metropolitan areas,  environmental pollution has re-
sulted from industrialization and urbanization.
  Each tree population  sample consisted  of 20  leaves
collected at random from sycamore trees.  These leaf sam-
ples were collected in mid-autumn to ensure their matur-
ity  at the time of collection. Gross morphological features
were recorded and  analyzed for  the four populations
(Table II).

                        Table II.
            Gross Morphological Characteristics*
                 of Platanus Occidentals
            Population
Pattern      A          B
D
Leaflength    22.5±3.48    22.0±2.12   14.7±2.46    8.5±1.68
(cm)
Leaf width    24.3 ±3.87    22.6±1.84   I5.2±2.01    9.0±1.68
(cm)
Petiole       7.6±1.04    7.5±0.87    6.5±1.23     1.2±0.63
length (cm)
Internodal    6.9±1.48    6.0±1.01    5.1 ±0.92     2.9±0.78
distance (cm)
Length:      0.9         1.0         1.0         0.9
width
ratio

The values represent means of 20 measurements ± standard deviation.


  Cuticular impressions of the upper and lower leaf sur-
faces were prepared by applying  Duco®  cement to the
washed and air dried leaves'10'. A  small portion from the
central area of the leaf imprints was used to make cuticular
slides for the adaxial and abaxial leaf surfaces of each pop-
ulation. Cuticular features of these four populations were
analyzed by selecting at random  20 fields (n = 20) from
each microscope  slide. The microscopic analysis was per-
formed using a 40x objective and lOx oculars.

RESULTS

  No stomata occurred on the upper epidermis, as is gen-
erally the case in woody dicots. The number of undula-
tions in the epidermal cells were counted only on the lower
leaf surface since the  undulations  were not easily discern-
ible on the upper epidermis. The statistical analysis of the
data is given in Table III.
  Populations A and B growing in relatively less polluted
habitats had longer leaves (22.5 and 22.0 cm) compared to
the leaf lengths (14.7 and 8.5 cm) in populations C and D
respectively. The latter populations exhibited a high de-
gree of environmental pollution in their habitats.
  Leaf width measurements for populations A and B were
also higher (24.3  and 22.6  cm) than for populations C
and D  (15.2  and 9.0 cm),  respectively. Petioled length
and internodal distance were also  greater in plant popula-
tions  of less  polluted habitats.  The  mean  length-width
ratio  in all the populations studied  showed a very in-
significant variation (0.9-1.0) and can  be regarded as con-
stant  for the taxon under investigation.  It is  quite ap-
parent from the measurements of the gross morphological
features of  the four sycamore populations, that polluted
environmental conditions  had an adverse effect  on the
growth of the plant.
  Stomata were absent on the upper  epidermis of syca-
more leaves, as  is the  case in many woody dicotyledons.
The trichome frequency was slightly higher on the upper
leaf surface as compared  to the lower leaf surface. The
highest stomatal frequencies of 30.2 and 29.2 per unit area
(0.152 mm2) were found in  populations A and B. These
two populations illustrated a minimum  level  of environ-
mental  pollution.  The leaves of  sycamore tree seem to
have low stomatal frequency (26.1 and 23.3) in polluted
areas,  an adaptation which may be  of ecotypic signif-
icance in  regulating the  limited and  controlled entry of
harmful gaseous pollutants into the plant tissues, espec-
ially when the tree grows in a polluted area.
  Trichome frequency values of 3.3 and 3.5 for the upper
leaf surfaces of populations C and D respectively were
the highest. These two populations were exposed to the
highest levels  of environmental pollution. Populations A
and B, however, had fewer trichomes. All of the trichomes
were unicellular and the majority of  the trichomes were
found along the veins rather than the lamina.
  The  trichome  shape  was  variable. Most  of the tri-
chomes were star-shaped with 5-6 rays radiating from the
case of the trichome. These star-shaped trichomes were
prominent on the  midvein of sycamore leaves. Also, sev-
eral simple  pointed trichomes were noted in each popula-
tion. Populations C and D possessed the longest trichomes
(106.7 and HO.Oum), respectively. It seems quite apparent
that longer and more numerous trichomes were asso-
ciated with a high degree  of  environmental pollution.
These modifications or adaptations in sycamore may be of
ecotypic or even evolutionary significance. A pubescent
leaf surface may act as an insulator in a  polluted environ-
ment. Leaf hairs are also known to be of value in shading
the living cells and  thus reducing the temperature of the
leaf tissues.
  The size of the largest and smallest stomata on the low-
er  leaf surface  in all the populations  sampled showed
little  variation.  The smallest stomatal size ranged from
7.6um to 8.3 urn while the  largest size ranged from 12.7p
to  14.3pm. Because of this relatively insignificant varia-
tion in the size of the two  extremes  within populations,

-------
                                                                                       REMOTE SENSING
                                                                                                    187
                      Table III.
       Leaf Cuticular Characteristics* of Populations
                 of Platanus occidentalis
Pattern

Stomatal
frequencyt
Stomatal
length (um)
 Largest
 stoma
 Smallest
 stoma
Trichome
density + +

Trichome
length (um)
 Longest
 trichome

 Shortest
 trichome

Epidermal
cell
undulations**
(number)
Subsidiary
cell complex
(number)
               Population
               A        B
                                 D
U
30.2±1.12  29.2±2.91  26.1 ±2.46  23.3 ±1.11




12.9±0.89  12.8±0.67  12.7±0.89  14.3±1.11

7.6±0.89   7.6±0.45   7.9±0.89   8.3 ±0.45


1.3 ±0.45   2.1 ±0.45   3.3 ±0.62   3.5 ±0.89

0.8±0.78   1.7±0.82   2.7±0.89   2.9±0.76
U  90.3±4.47  91.1 ±5.59  106.7±3.21 110.0±4.47
L
U
89.4 ±1.41  90.2 ±1.22  100.9 ±2.41 104.2 + 2.42
40.9±2.12 -56.3±4.47  62.7±4.47  80.0±1.01

39.2±2.11  53.2 + 2.23  60.2±2.82  76.4±1.84

3.5±0.89   3.6±0.78   3.1 ±0.89   3.2±0.68
   3-4
          3-4
                       3-4
3-4
 'The values represent means of 20 measurements ± standard deviation.
"Mean epidermal cell undulations = undulations of the lower surface of leaves.
 tMean Stomatal frequency = stomata of the lower surface of leaves observed through a 40x
 objective and lOx oculars (field area = 0.152 mm1). U= upper surface of leaf; L = lower
 surface of leaf.
+ +per unit area (0.152 mm!).

it may be suggested that  the  stomatal size range remains
fairly uniform in sycamore.
  A count of the number of undulations in the epidermal
cells reveals that environmental pollution had no signifi-
cant effect on this cuticular feature. In all four sycamore
populations, the mean  value for the number of undula-
tions ranged from 3.1 to 3.6. The subsidiary cell complex
remained constant,  with  3-4  cells adjacent to  the guard
cells. The constancy of the subsidiary cell complex in all
the populations sampled may be of taxonomic significance
for the taxon.

CONCLUSIONS

  The study points out leaf cuticular and gross morpho-
logical variations in American sycamore as evidenced by
modification in leaf length, leaf width, petiole length, in-
ternodal distance, stomatal  frequency, and trichome den-
               sity in the four populations studied.  Stomatal frequency
               was  low in the highly polluted habitats.  The data  sug-
               gest that high trichome frequency may be correlated with
               assumed high environmental pollution.
                 The number of  undulations in  the epidermal cells as
               well  as the stomatal size  ranges were  not affected by en-
               vironmental  pollution to  any significant  degree.  Leaf
               length-width ratio  and  the subsidiary cell  complex  re-
               mained the same in all the populations. The latter is no
               doubt a species trait and hence of taxonomic significance.
               Similar results have been found in other plant taxa studied
               by the author. With detailed  investigations on additional
               herbaceous and woody plants growing in polluted habi-
               tats, it may be possible to suggest a more precise correla-
               tion   between plants  and  environmental  pollution  and
               hence their use as indicators of environmental pollution.
REFERENCES

 1. Chamberlain, C.J., "Gymnosperms: Structure and
   Evolution," Dover, New York, 1934.
 2. Feder, W.A., "Plant Response to Chronic Exposure
   to  Low  Levels  of  Oxidant  Type Air  Pollution,"
   Environmental Pollution, 1, 1970.
 3. Gray,  A., "Gray's  Manual of Botany,"  American
   Book Co., New York, 1950.
 4. Mathis,  P.M. and  Tomlinson, G., "Lichens:  Bio-
   assay for Air  Pollution in a Metropolitan Area (Nash-
   ville, Tenn)," Journal Tenn.  Academy  of Science,
   47, 1972.
 5. Pyatt,  B.F.,  "Lichens as Indicators of Air Pollution
   in a Steel  Producing  Town in South Wales," En-
   vironmental Pollution, 1,  1970.
 6. Scheffer, T.C. and Hedgcock, G.C., "Injury to North-
   western Forest Trees by  Sulfur Dioxide from Smelt-
   ers," U.S. Dept. Agr. Tech. Bull., 1117, 1955.
 7. Sharma,  O.K., "Cuticular Features as Indicators  of
   Environmental Pollution," Water, Air, and Soil Pollu-
   tion, 8, 1977.
 8. Sharma,  O.K. and Butler, J.,  "Leaf Cuticular  Var-
   iation in Trifolium repens L.  as  Indicators of En-
   vironmental Pollution,"  Environmental Pollution  5,
   1973.
 9. Solberg,  R.A. and Adams, D.F.,  "Histological Re-
   sponses of some Plant Leaves  to  Hydrogen Fluoride
   and Sulfur Dioxide," American Journal of Botany, 43,
   1956.
10. Williams, J.A.,  "A Considerably Improved  Method
    for  Preparing Plastic  Epidermal  Imprints," Botani-
    cal Gazette, 134, 1973.

-------
RATIONALE FOR DETERMINING PRIORITIES AND EXTENT
          OF CLEANUP OF UNCONTROLLED HAZARDOUS
                                        WASTE SITES
                                   WALTER UNTERBERG, Ph.D.
                                         WAYNE L. STONE
                                 Rockwell International Corporation
                                      Newbury Park, California

                                       ANTHONY N. TAFURI
                                U.S. Environmental Protection Agency
                            Municipal Environmental Research Laboratory
                                          Edison, New Jersey
INTRODUCTION

  The Comprehensive Environmental Response, Compen-
sation and Liability Act of 1980, PL 96-510, requires re-
vision of the National Contingency Plan with a section
known as the National Hazardous Substance Response-
Plan and requires that the plan establish procedures and
standards for responding  to releases of hazardous sub-
stances, pollutants,  and contaminants. One aspect of this
new plan is defined as:
  "105 (3) methods and criteria for  determining the ap-
propriate extent of removal, remedy, and other measures
as required by this Act."
  In this paper the authors attempt to develop a rationale
for determining "appropriate extent of removal" for un-
controlled hazardous waste sites. This involves three types
of decisions:
•Determination of Cleanup Priorities—To what area or
 areas to direct the cleanup effort when time or resources
 are  limited and/or the affected area is too extensive to
 clean up the entire release.

•Evaluation of Alternative Cleanup  Methods and Selec-
 tion of Optimum  Methods—Once the priority cleanup
 areas have been selected, (a) to evaluate cleanup alterna-
 tives considering availability and  cost of manpower and
 equipment, effectiveness and speed of deployment, effect
 on  the  environment  and  other applicable parameters;
 and (b) to  select the optimum method or combination
 of methods for the priority cleanup operations.

•Determination of Extent of  Cleanup (How Clean Is
 Clean)—Once the optimum cleanup methods have been
 selected, how far and how long should cleanup proceed?
 The desirable "extent of removal" depends on (a) reach-
 ing  acceptably low levels  of residuals, (b) the (usually
 rising) cost of removal per unit mass of contaminant as
 lower and lower contamination levels are achieved,  (c>
 environmental impact of cleaning methods themselves;
 and other suitable parameters.
   The authors deal, in order,  with these three questions
 and provide decision-making methodologies in these areas
 to assist those charged with directing cleanup operations.
However, "removal" (which is primarily concerned with
cleanup) as addressed here  excludes the initial response
to a hazardous  substance release usually designated as
first-response actions, which protect human populations
against immediate dangers such as fires, explosions, and
ingestion of toxic materials. Likewise  excluded are sub-
sequent remedial actions  that stabilize the environment
after removal.

HAZARDOUS SUBSTANCES

  The  CERCLA legislation, in Section 101 (14), desig-
nates the substances  covered by the Act. More than 600
chemical compounds and elements are now listed while
the exact number is still being decided. The physical be-
havior  of  hazardous substances (HS)  on release can be
classified into 11 categories, (Table I),  including  releases
into water, into air  and onto land. The principal dis-
tinguishing characteristics are density,  vapor pressure,
and reactivity.
  The hazards they present  are categorized and coded in
Table II, and are mainly in the areas of toxicity and com-
bustibility. Figure  1 is a typical page from an alphabet-
ical computation which lists the hundreds of substances
by Chemical Abstract Service (CAS) number, the physical
behavior categories of Table I, and the hazardous effects
coded in Table  II. This provides an initial guide to the
physical nature and danger of any known hazardous sub-
stance.

CLEANUP PRIORITIES

  Even though  first  response actions  have concentrated
primarily on safeguarding the human population,  the haz-
ards to humans are still the paramount consideration even
in the removal phase. This  is  the key issue in responding
to hazardous substance incidents, in contrast to oil spills
where normally no human  lives are endangered. There-
fore, oil spill response procedures are of limited use in this
discussion.
  The  proposed cleanup priority decision methodology
follows the flow chart of Figure 2. Inputs based on the
hazardous substance  release,  the first response  actions,
                                                  188

-------
                                                                                    REMEDIAL RESPONSE     189
                         Table I.
              Physical Behavior of Hazardous
                   Substances on Release

                        Water Spill

  1. Liquids and solids dissolving in water (solubles)
  2. As above, but hazardous gases/vapors result
  3. Insoluble substances lighter than water (floaters)
  4. As above, but hazardous gases/vapors result
  5. Insoluble substances heavier than water (sinkers)
  6. As above, but hazardous gases/vapors result
  7. Liquids or solids that react with water

                        Land Spill

  8. Liquids or solids that are self-reactive
  9. Compressed gas that could be released
 10. Liquids or solids below their boiling/evaporation point
 11. As above, but hazardous gases/vapors result

                         Air Spill

  2. Hazardous gases/vapors from solubles/water
  4. Hazardous gases/vapors from floaters/water
  6. Hazardous gases/vapors from sinkers/water
  8. Self-reactive liquids or solids
  9. Compressed gases expanding
 11. Hazardous gases/vapors from slow evaporators
                        Table II.
          Hazardous Effects Categories and Codes

 a   toxicity data not available
 b   combustible with ignition source (low vapor pressure)
 c   potentially toxic combustion/reaction products
 d   Highly toxic combustion/reaction products
 e   Do not drink contaminated water
 E   Explosion hazard
 f   Do not inhale toxic vapor/dust
 g   Readily flammable (high vapor pressure)
 h   Corrosive
 T   Toxicity hazard
 ()  Parentheses indicate a lower hazard or property value

 Note:  Italicized and capitalized letters indicate serious hazards.
climatic conditions  and  the nature  of the  affected  or
threatened  environmental constituents are entered into
the Hazardous Substance  Incident Profile Form from
which long-  and  short-term damage estimates  are  de-
rived. This form  is dependent on the time and date and
may be filled out  at different times during the cleanup ac-
tion. The form serves as a record of the overall event and
its consequences.
  The generalized environmental response priorities  are
shown on Table III. Normally, the priorities are to guard
against,  in order, human,  animal, ecological, socioeco-
nomic and  social  impacts.  Also  affected constituents
should be protected before threatened constituents. Priori-
ties, however, may  be changed if  special  circumstances
indicate that the situation requires it.
  A numerical Vulnerability Rating  System was devel-
oped to establish cleanup priorities among environmental
constituents. This is illustrated in the worksheet of Figure
3. There are six additive vulnerability parameters, with a
maximum possible score of 100. Parameters 1 through 3
are  hazardous  substance-related   while  parameters   4
through 6 are constituent-related. Further, there are three
subtractive constituent-related  parameters  (numbers   7
through  9) indicative of  the  background contamination
level and assimilative capacity and any self-cleaning abil-
ity.
  The numerical ratings are guides to cleanup priorities
(higher vulnerability ratings mean higher cleanup priority)
subject to two considerations: the priorities among classes
of constituents (axiom 1 on Table III), i.e., whether they
are in  the order A, B, C, etc. as shown on Figure 3; and
the priority of affected  constituents over  threatened con-
stituents (axion 2 on  Table III). The judgment of the user
must make the final priority assignment.

SELECTION OF CLEANUP TECHNIQUES

  The number of hazardous substances  cleanup strate-
gies that  can be applied to different situations is almost
limitless   due to the  complexity  of the  parameters  in-
volved. Hazardous substances  themselves possess a va-
                        Tablelll.
        Generalized Environmental Response Priorities
Axiom 1: Normally protect environmental constituents in the
         following order of priority:
a. Human populations and habitats—human impact
     Cleanup crews
     General population
b. Fauna—animal/ecological impact
     Marine
     Land
     Birds, etc.
c. Flora—ecological impact
     Forests
     Food crops
     Other crops
     Marshes
     Marine, etc.
d. Property—socioeconomic impact
     Industry
     Business
     Municipal
     Fisheries, etc.
e. Aesthetic  and recreational areas—social impact
     Parks
     Beaches
     Wilderness
     Water use and sports areas, etc.

Axiom 2: Protect affected areas before threatened areas.

Axiom 3: Use judgment and common sense in applying Axioms
          1 and 2.

-------
 190    REMEDIAL RESPONSE













COMPOUND
Acenaphthene (83-32-9)
Acenaphthylene (208-96-8)
Acetaldehyde (75-07-0)
Acetic acid (64-19-7)
Acetic anhydride (108-24-7)
Acetone cyanohydrin (75-86-5)
Acetyl bromide (506-96-7)
Acetyl chloride (75-36-5)
Acrolein (107-02-8)
Acrylonitrile (107-13-1)
Adipic acid (124-04-9)
Aldrin (309-00-2)
Allyl alcohol (107-18-6)
Allyl chloride (107-05-1)
Aluminum sulfate (10043-01-3)
Amnonia (7664-41-7)
Ammonium acetate (631-61-8)
Ammonium benzoate (1863-63-4)
Ammonium bicarbonate (1066-33-7)
Ammonium bifluoride (1341-49-7)
Ammonium bisulfite (10192-30-0)
Ammonium carbamate (1111-78-0)
Ammonium carbonate (506-87-6)
Ammonium chloride (12125-02-9)
Ammonium chromate (7788-98-9)
Ammonium citrate, dibasic(3012-65-5
Ammonium di chromate (7789-09-5)
Ammonium fluoborate (13826-83-0)
Ammonium fluoride (12125-01-8)
Ammonium hydroxide (1336-21-6)


WATER SPILL
1
o

v»
O

o>






(X)
(X)
2
1C
CU
N
a.
o
c


CU

o



X E,T
X T
(X)'E,T
X T
X T
X ' T
X
X
(X) 1

X '
(X)
X
X
X
X
X
X
X '
X j
X
X I
x ,
X '
X i
X
X 1
X .

1
E,T
E,T


E,T
E,T

T













T


3 4
-n 1C
— ' Ol
o n
rt -!
v> O-
O
C


at
-a
o
-s
1
X 1
X E,T
X T
1
X 'E,T
X |E,T

1
X E.T
X 'E,T
i








1








5
GO

3
l/l








X

I
X
X

X


X

X
X
X
X
X
X
X
X
X
X
X
X
X
(x)



6
JC
O)
N
-i
a.
o
c
I/I

P)

o
~*


T
E,T
T
T






















7
jO
n>
Ol
£

m








X
X






















LAUD SPILL
8
,-,
CD
Ol
£










X
x






















9
0
o
-1
-s
n>



CL
o

in









x
















10
in

o
m

Ol

o
2
ft-

0
13


X
X
X
X
X
X
X
X


X
X


11
-r-
Ol
N
-I
o.
o
c
I/I

Ol
•o
o
-1


E.T
T
E.T
T
I
E,T
E.T


E,T
E,T




1











1

X T
















NOTES
b, c, e, f
b, c, e_, f
f, g_
skin, eye hazard, f_, %, h
skin, eye hazard, f, g_, F
poison (cyanide), d_, e, f
^violent reaction with water, eye haz-
j ard, skin burns, irritant, £, £, h^
skin, eye hazard, §., £, g_
Poison (cyanide), can explode u,e,f,£
b, f, irritant
skin poison, b, c, e_, f
eye hazard, £, g_
eye hazard, skin poison, f_, g_

f, tu choking gas

e, I

etches glass, e_, f, ^
sulfur dioxide formed with acids, £



e , I

oxidant, £_, £, g_, h^
e, f., li
etches glass e_, f, h.
f , h


                                                      Figure 1.
                                  Typical Page From a Listing of Hazardous Substances
riety  of physical, chemical  and  toxicological  combina-
tions  so as to make a fairly standard cleanup and treat-
ment  methodology, sueh as for oil spills, not appropriate
or applicable.  It is  therefore  impractical to attempt  to
develop  a specific cleanup  methodology for each  haz-
ardous substance and spill situation combination. Instead,
an attempt is made to present for each hazardous  sub-
stance physical behavior release class (Table I) a variety
of spill treatment techniques that may be useful in clean-
ing up  hazardous substances falling within a  particular
class.
  The treatment techniques,  which are listed in  Table IV,
fall under the classifications  of  mechanical,  chemical,
and sorbents, gels and foam methods. The treatment tech-
niques are geared toward both containment  and removal
actions and include the following cleanup strategies:
  (1)  Dilution and dispersion into the environment
  (2)  In-situ treatment
  (3)  Offsite treatment and disposal
  (4)  Treatment on site in an offstream treatment system
                        VULNERABILITY RATINGS OF
                         AFFECTED/THREATENED
                           ENVIRONMENTAL
                           CONSTITUENTS
(Table 3)
                                       (F19. 3)
                  Figure 2.
     Cleanup Priority Decision Flow Chart

-------
                                                                                     REMEDIAL RESPONSE     191
Method
Use/ Application
               Table IV.
Hazardous Substances Cleanup Techniques

               Advantages
A. MECHANICAL TECHNIQUES
 1.  Dispersion



 2.  Dilution (Water)


 3.  Dilution (Air)




 4.  Booms

 5.  Weirs

 6.  Spill Herding
Employ water jet, mechanical
mixer, aerator, etc. to promote
dilution. Should be used where
adequate dilution can be attained.
Consider water use classifica-
tion, and the ability for natural
or forced dilution.
Use blower equipment to intro-
duce clean air to lower hazardous
substance vapors to less than
TLV or lower flammability limit.
Use in calm and sheltered areas.
Need deployment device. Use
in fairly calm waters.
Need a weir and boat. Use in
calm, large waterbody.
Apply chemical herder to pro-
tect shore or other facilities.
               Inexpensive, easily implemented
               Inexpensive and timely
                Can direct air away from
                populated areas
               Can be used over a larger area.
               Many varieties.
               Easily used.

               Useful in rough water.
Disadvantages
Pollution may be spread over
a larger area
Pollution may be spread over a
larger area.

Pollution may be spread over a
larger area. Need large-capacity
blowers.
Cannot be used in rough or
rapidly rapidly moving water.
Cannot be used in rough water.
Not always available.
Not easily obtainable; not 100%
effective.
 7. Dredging
  8. Burial
 9. Dikes (Earth)

 10. Dikes (Foamed
    Polyurethane)
 11. Trenches
 12. Diversion
 13. Dam (Underflow)

 14. Mist Knockdown
 15. Vacuum Skimmer
Uses in hazardous substances
spills is largely untested.
Use only as a temporary meas-
ure to retard spread of spill un-
til recovery, or when spill sub-
stance is harmless.
Create barrier with bulldozer
or earthmoving equipment.
Need trained personnel to con-
struct on hard, dry surfaces.
Construct with bulldozer or
earthmoving equipment.

Use earthmoving equipment to
divert channel and sandbags,
dikes, etc. to block entrance
of pollution.
Create dam and underflow drain
in relatively small stream.
Use fine mist spray to suppress
water-soluble low-lying
hazardous vapors.
Use vacuum tank to generate
draining force for removing
small spills.
B. CHEMICAL TECHNIQUES

 1. Neutralization
                Can effectively remove
                pollutants.
                Easily implemented.
                Material on site; common
                equipment.
                Can hold up to several feet
                of water.
                Construct with common
                equipment.

                Separates contamination.
                Simple to construct.

                Reduces air pollution potential.


                Easily used.
Apply acid or base in powder
or slurry form.
                Effective in restoring pH to
                proper range.
Equipment not readily avail-
able; potential for environmen-
tal damage; generation of large
amounts of contaminated
material.
Not an environmentally sound
practice.
Composition of soil may not be
suitable (possible leakage).
Equipment hard to obtain;
leaks on wet ground.
Move large amounts of material;
highly permeable soil may not
contain spill.
Large amounts of earth must be
moved. Clear area and proper
soil conditions required.

Can only be used in small
streams.
Can create a water pollution
 problem, runoff must be con-
tained.
Equipment not always obtain-
spills only.
Added chemicals may be a
threat to the environment.
Difficulty in determining cor-
rect amount of chemical
addition.

-------
192     REMEDIAL RESPONSE
 2.  Precipitation
 3.  Chelation
Add chemicals to clarify or pre-
cipitate out soluble substances.
Chalating agents are added to
water to bind (and deactivate)
metal ions.
Can be performed in situ or in
field treatment unit.
Effective in removing spilled
metal ions.
C. SORBENTS, GELS AND FOAMS
 1. Activated Carbon
 2. Polyurethane
    Foam

 3. Ion Exchange
 4. Gelling Agent
Implement treatment by (1) drag-
ging AC-filled bag through water,
(2) passing water through AC
beds or columns, or (3) employ-
 ing mobile treatment unit. Used
 mostly for organics.
Apply via foam-laden belts or
shredded foam broadcast over
the spill to absorb pollutants.
Water must be pumped through
bed or column used mostly for
acid, base, metal, or toxic salt spills.
Adds gels through dispersion de-
vice to reduce viscosity (and
immobilize) spill.
Most universal sorbent; read-
ily available.
Can be squeezed and reused.
Effective removal method for spe-
cific pollutants

Effective containment method;
can reduce vapors.
Ultimate reaction, including
paniculate matter formed, may
pose a new threat to environ-
ment.
Misapplication of treatment
may cause environmental dam-
age. Essential metal nutrients
may be inadvertently removed.
Costly if carbon is not regen-
erated.
Bulky and difficult to transport.
A large amount of equipment
must be deployed. Resins must
be regenerated.
Practical use for small volumes
only; difficult to obtain; creates
a disposal problem; heat may be
generated in the reaction.
^^^^^ Environmental
^"^~^^^ Constituents
Vulnerability ^"~"-^-^__^
Parameters — .
1. Susceptibility to this
HS via pathways of:
a. Air (0-4)
b. Surface Water (0-4)
c. Groundwater (0-4)
d. Land Surface (0-4
e. Subsurface (0-4)
2. Proximity to Release Pt. (0-12)
3. Contamination Vector (0-12)
4. Population/Size/Extent (0-16)
5. Intrinsic Values
a. Hater Quality (0-4)
b. Air Quality 0-4)
c. Food Chain 0-4)
d. Ecological (0-4)
e. Economic (0-4)
f. Aesthetic (0-4)
g. Recreational (0-4)
6. Time Restrictions (0-12)
INTERIM TOTALS (100 MAX)
SUBTRACT (for this HS):
7. Assimilative Capacity (0-12)
8. Self-Cleaning Ability (0-12
9. Background HS Level (0-12)
RATING TOTALS
A. Human Populations
and Habitats

Al












A?












A3












A4












AS












A6












A7












A8











B. Fauna

Bl












B?












B3












B4












B5












B6











C. Flora

Cl












C2












C3












C4












C5












C6











D. Property

°1












D2












D3












D4












D5












D6











E. Aestheti
tecreationa

El












E2












3












E4











c and
Areas

E5












6











                                                       FigureS.
                                    Vulnerability Ratings of Environmental Constituents

-------
                                                                                   REMEDIAL RESPONSE     193
                                                      Table V.
                                        Hazardous Substances Cleanup Selections
Physical Behavior Class
(From Table I)
Potential Cleanup Altern-
atives (From Table IV)
   Physical Behavior Class
   (From Table I)
Potential Cleanup Altern-
atives (From Table IV)
 1. Liquids and Solids Dissolving in
   Water (Solubles)
 2. As Above, but Hazardous
   Gases/Vapors Result
 3. Insoluble Substances Lighter
   Than Water (Floaters)
 4. As Above, but Hazardous
   Gases/Vapors Result

 5. Insoluble Substances Heavier
   Than Water (Sinkers)
 6. As Above, but Hazardous
   Gases/Vapors Result

 7. Liquids or Solids that React
   with Water
A.I
A.2
A.12
B.I
B.2
B.3
C.I
C.3
As above, but add
A.3
A. 14
C.2
C.4
A.I
A.2
A.4
A.5
A.6
A.12
A.13
A.15
C.2
C.4
As above, but add
A.3
A.14
A.7

As above, but add
A.3
A.14
A.I
A.2
A.3
A.4
A.5
A.6
A.7
A.12
                                                              8. Liquids or Solids that are
                                                                Self-Reactive
                                                              9. Compressed Gas that Could
                                                                be Released
  10. Liquids or Solids Below Their
     Boiling/Evaporation Point
                                                             11. As Above, but Hazardous
                                                                Gases/Vapors Result
A.13
A.14
A.15
B.I
B.2
B.3
C.I
C.2
C.3
C.4
A.3
A.8
A.9
A. 10
A.ll
A.14
A.15
B.I
C.2
C.4
A.3
A.8
A.9
A. 10
A.ll
A.14
A.15
B.I
A.8
A.9
A.10
A.ll
A.15
B.I
C.2
C.4
As above, but add
A.3
A.14
  The potential cleanup alternatives that may be applic-
able, depending on the particular substance spilled and
spill situation, for each physical behavior class are listed
in Table V. Figure 1  will aid in  determining into which
physical behavior class a particular hazardous substance
falls.
  Once the potential cleanup alternative (or combination
of alternatives) is selected, several  options will probably be
able to be eliminated for favored due to obvious limiting
factors such as available equipment, manpower, financial
resources,  etc. Also,  the spill cleanup advantages/dis-
advantages listed  in Table IV should be  seriously  con-
sidered. If no cleanup selection is feasible or if significant
environmental damage will result from cleanup  actions,
no action at all may be the appropriate response.
DETERMINATION OF ALTERNATIVE
CLEANUP TECHNIQUE COSTS

  After a  potential spill cleanup technique or combina-
tion of techniques has been chosen, costs for each selec-
tion can be determined  from past  experience, on-scene
advisors, cleanup contractors or  any other references at
hand. The costs to be included, where applicable, are:
•Labor
•Capital or rental costs
•Material and supplies

Daily cleanup  costs ($/day) for each  cleanup technique
can be computed by adding the appropriate cost compon-
ents.
                                                      •Packaging and transportation
                                                      •Ultimate disposal

-------
194     REMEDIAL RESPONSE
  For each cleanup  technique  or techniques for which a
daily cleanup cost ($/day) was  calculated, an attempt can
be made to try and  obtain an  estimate of the amount  of
hazardous substance each cleanup technique is capable of
extracting on a mass per  day (Ib/day) basis.  This par-
ameter in most cases cannot be easily identified.  There-
fore, estimates of extraction performances should be made
from past experiences of the OSCs or their advisors, from
cleanup contractors, or from  cleanup equipment  manu-
facturers' data.
  Unit cost ($/lb) is defined as the amount of dollars re-
quired to extract one pound of hazardous substance. This
can be calculated  by dividing the daily cleanup cost rate
($/day) by the extraction performance rate (ibs/day). Al-
ternative cleanup techniques can now be compared on a
unit cost basis  to get an indication of relative  economic
merit.

DETERMINATION OF MOST EFFECTIVE
CLEANUP TECHNIQUE

  In many spill situations, it may not be realistic to choose
cleanup techniques solely on a unit cost ($/lb)  basis. Al-
ternative cleanup techniques can be ranked on a unit cost
basis but the listing  priority should be qualified with cer-
tain cleanup effectiveness criteria that will be affected  by
the peculiarities of a specific spill situation and may out-
weigh any cost considerations. The cleanup effectiveness
criteria include:
•Completeness of removal—Can the  desired  or required
 cleanness levels or standards be attained?
•Rate of removal—Can the substance(s) be cleaned up in
 time to  protect  critical habitats  or reduce evacuation
 times, etc?
•Logistics—Can the  techniques be easily deployed?  Neces-
 sary equipment? Is it available?
•Operational requirements—What levels of labor, skill,
 etc., are required? Is it affected by weather?
•Environmental impact—Will the environment be adverse-
 ly affected?
•Personnel protection—Will response personnel be un-
 duly subjected to hazards?
•Disposal problems—Will large amounts of wastes be gen-
 erated? Will this result in transportation and ultimate dis-
 posal problems?
  After consideration of the preceding cleanup effective-
ness criteria,  the cleanup technique priority list  can be  re-
ordered to obtain the effective cleanup strategy for a par-
ticular hazardous substance incident.

EXTENT OF CLEANUP (HOW CLEAN IS CLEAN?)

  Based on the results of the two previous sections, a spe-
cific strategy will have been selected  to clean up the en-
vironmental constituents having the highest priorities. This
strategy will  consist of one or more cleanup  actions at
the same time,  followed sequentially by other  actions. For
instance,  simultaneous  mechanical  and  chemical  tech-
niques may be  used to extract and  treat  the  hazardous
substance at  the incident site,  followed by transportation
  0.9-
~ 0.7-
1 0.3-
3
   o.e-


   O.L
PUT KEEH.V.
comma)
CUMULATIVE
CLEANUP COSTS
                                       ~l
 STATUTORY
 ENVELOPE
                                            751
                                           " ENVELOPE
                           12      16
                         Time, Weeks
                        Figure 4.
          Control Graph for Cleanup Time and Cost


                           HS   hazardous substance
                            A   larger constant effectiveness
                            B   smaller constant effectiveness
                            C   effectiveness decreases as
                               cost  rises
                                                 ACCEPTABLE
                                                 LEVEL
                      CLEANUP COST

                        Figures.
                Cleanup Effectiveness Curves

                        Table VI.
            Factors Influencing Extent of Cleanup

1.  Acceptably low final contamination levels
2.  Change in effectiveness of cleanup techniques as contamina-
   tion levels decrease
3.  Change in incremental cleanup costs as contamination levels
   decrease
4.  Environmental impact of cleanup techniques themselves
5.  Effect of simultaneous vs. sequential cleanup of different en-
   vironmental constituents
6.  Overall and ultimate  environmental consequences—assimila-
   tive capacity of environment
7.  Public awareness and  political pressures
8.  Budget and time constraints (Superfund: $1 million and 6
   months)

-------
                                                                                  REMEDIAL RESPONSE     195
of the extracted hazardous matter to another location for
disposal. These various techniques may be combined in
simultaneous/sequential/intermittent ways, as needed.
  During the removal or cleanup phase one must initially
project and later keep track of three main parameters:
•Daily  cost of individual cleanup  techniques  and of the
 total cleanup  effort,  including cost of measurements in
  item 3 below
•Daily amount of hazardous substance extracted
•Daily  or periodic change in the hazardous substance re-
 sidual contamination level  in  the environmental  con-
 stituents affected by the incident, based on standardized
 meaningful measurements
  The factors influencing the extent of cleanup during the
removal phase are listed  in Table VI.  Unfortunately, all
factors, except budget and time constraints which are dic-
tated by outside considerations, are difficult to pin down.
  Clearly, the basic goal of the cleanup operation is to re-
duce the level  of toxic substances to an environmentally
safe level, i.e.,  a level that does not present a hazard to
human health  or welfare, including  water supplies and
crops. Unfortunately, there are  virtually  no quantitative
guidelines concerning  acceptable residual levels of toxic
substances  for spill  cleanup  operations.  Many  of the
metals whose salts are considered hazardous occur natural-
ly in soils at low concentrations and are important trace
             elements in the biosphere. As a basic guideline, ideal clean-
             up operations should reduce the levels of toxic substances
             back to the natural levels found in the general area of the
             spill.
               In certain cases, specific numbers for "safe" levels of
             contaminants are given in various regulations, particularly
             for drinking water and substances  regulated under the
             Clean Air Act.
               Although  quantitative guidelines for cleanup standards
             have not  yet been established, it is possible to go through
             the list of toxic substances and identify those that will re-
             quire extensive cleanup because they present severe en-
             vironmental hazards.  Thus polychlorinated  biphenyls,
             polynuclear  armomatic  compounds, various halogenated
             aromatic and aliphatic hydrocarbons, nitrosoamines and
             certain heavy metal salts of mercury, beryllium, and cad-
             mium, among others, would merit special attention.
               From a general consideration of the chemical, physical,
             and toxicological properties of  the toxic substances listed,
             it is  possible to perform a  crude stratification into sub-
             stances that  are likely to present long-term environmental
             hazards and require extensive cleanup and substances with
             higher vapor pressures or greater chemical reactivity that
             present a relatively lower long-term environmental hazard
             and may require less extensive  cleanup. A tentative strat-
             ification  scheme along those lines involving four groups
                                                                                     Date:
               COST PARAMETERS (CP)

  1.   CLEANUP LABOR                                            (0-3)
  2.   EQUIPMENT                                               (0-3)
  3.   SUPPLIES & CHEMICALS (including fuels,  lubes, power)      (0-3)
  4.   CLEANUP IMPACT (allow for persistence and                 (0-3)
                    assimilative capacity)
  5.   AREA USE INTERFERENCE (due to continued cleanup)          (0-3)
                                              TOTALS         ICP:
              BENEFIT PARAMETERS (BP)

  6.   ECONOMIC (normalization of income  from  area)              (0-3)
  7.   ENVIRONMENTAL (protection of ecosystems and habitats)      (0-3)
  8.   AESTHETIC (preservation of visual  quality & scenery)      (0-3)
  9.   RECREATIONAL (normalization of  indoor & outdoor areas)     (0-3)
 10.   PUBLIC (normalization of public access  & use)             (0-3)

                                                             EBP:
TOTALS
                                                       EBP   ZCP:
                                                                                ENVIRONMENTAL CONSTITUENTS
                        A.
                                        CONTINUE CLEANUP?  YES/NO:    	  	  	 	  	  	  	
                                        (YES IF X»0.  NO IF  X«0.  GET MORE FACTS IF  X IS BETWEEN +3 AND -3)
                                                     Figure 6.
                                      Cost/Benefit Analysis of Incremental Cleanup

-------
1%     REMEDIAL RESPONSE
of inorganic and seven groups or organic compounds is
presented in Table VII.
  The factors listed in Table  VI can be condensed into
three indicators  for use in deciding the extent of removal,
or, "How Clean is Clean?"
•Ability to approach  or reach acceptable residual con-
 tamination levels
                       Table YD.
     Classification Scheme for Long-Term Environmental
        Hazards of Inorganic and Organic Compounds
PART 1: INORGANIC MATERIALS

GROUP II: Inorganic metal  salts likely to require extensive
           cleanup. Long-term environmental hazards. Drink-
           ing water hazards. Likely to bioaccumulate.
GROUP 12: Inorganic nonmetal compounds likely to require ex-
           tensive cleanup. Long-term environmental hazards.
           Drinking water hazards. Likely to bioaccumulate.
GROUP  13: Inorganic metal  salts presenting  long-term en-
           vironmental hazards, but  likely to require less ex-
           tensive cleanup than Group II.
GROUP 14: Inorganic materials less likely to present long-term
           environmental hazards requiring extensive cleanup
           operations. May present severe short-term toxic or
           explosion hazards.

PART 2: ORGANIC MATERIALS

GROUP 01: Pesticides,  likely to  present long-term environ-
           mental hazards,  drinking water hazards and bioac-
           cumulation hazards. Extensive  cleanup operations
           likely to be required.
GROUP 02: Nitrosoamines: Highly potent carcinogens,  severe
           drinking water hazard, although  long-term  biode-
           gradation likely.  Extensive cleanup operations like-
           ly to be required.
GROUP 03: Polynuclear  aromatics  (PNAs): Potent carcin-
           ogens, low volatility, long-term environmental haz-
           ard, water and bioaccumulation hazards. Likely to
           require extensive cleanup.
GROUP  04: Halogenated hydrocarbons  and  related  com-
           pounds: Known or likely carcinogens, probable bio-
           accumulation, dkrinking water hazards, likely to re-
           quire extensive cleanup.
GROUP 05: Aromatic phenols, amines, nitro derivatives and re-
           lated compounds: Known or  likely  carcinogens,
           probable bioaccumulation, drinking water hazard.
           Biodegradation possible,  may  present  less long-
           term hazard than Groups 01 through 04.
GROUP 06: Amines, nitroles, aromatics and other compounds
           presenting environmental  hazards, known or  sus-
           pected  carcinogens, drinking  water hazards, but
           more reactive or volatile than previous groups and
           less  likely  to  present long-term  environmental
           hazards.
GROUP 07: Hydrocarbons, acids, esters and other compounds
           that are volatile or reactive enough to be  biode-
           gradable. Present lower long-term hazards and less
           likely  to require  extensive cleanup. May present
           severe short-term explosion or toxic vapor hazard.
•Comparison of costs versus benefits of incremental clean-
 up as a test of whether to continue cleanup
•Time and/or total cleanup cost limitations

Cost Benefit Analysis
  It is desirable  to  reach acceptable residual contamina-
tion levels in the contaminated environmental constituents
and prevent threatened constituents from being affected.
Further, cleanup costs should not exceed the benefits that
they provide. Lastly, time and/or total cleanup cost limita-
tions, statutory or otherwise, should not be exceeded. In
principle, all three conditions should be satisfied simultan-
eously.  In  practice, the connection between  increasing
costs,  decreasing contamination  levels,  and  increasing
benefits  can be tracked on  an elapsed time basis. If all
three conditions  cannot be  satisfied, then a compromise
must be reached. The procedure detailed below is designed
to furnish the user the needed data to make a decision.
  First,  daily cleanup costs and  cumulative committed
weekly cleanup costs are charted, for instance on a graph
like Figure  4. This  gives the overall picture for  removal
costs. Two  rectangular envelopes  are  pre-drawn on this
graph: the  Superfund statutory limitation  of $1 million
and 6 months, and (in dashed lines) the envelope at 75% of
these amounts for advance planning and control purposes.
  Next, a plot of effectiveness of cleanup should be made.
This is  a plot of Residual  HS Level  vs Total Cleanup
Cost, separate curves  and acceptable residual  level lines
being drawn for each  pathway or environmental constit-
uent,  like Figure 5. If the  cleanup effectiveness is con-
stant over the cleanup period,  then the curve becomes an
inclined  straight  line,  showing that equal increments in
cost produce equal decrements in contamination  level. If
the decrements become smaller while cost increments stay
the same, i.e., the curve  becomes  flatter, then the clean-
up effectiveness for the particular pathway drops as lower
contamination levels are reached.
  Third, make  a Cost Benefit Analysis  of Incremental
Cleanup as  shown in Figure 6 which is a form on which
both  cost parameters  and benefit parameters are rated
for the environmental constituents that  are benefitting
from  cleanup. This  analysis is similar to the one used in
another study.'" If the benefit score exceeds the cost score,
then it is useful to continue  cleanup; if the difference is 3
points or less, more facts are  needed for a decision; and
if the cost score exceeds the benefit score, then  cleanup
should be stopped. The analysis should be repeated at later
times to see  if the situation has changed.
  To make the decision on How Clean is Clean, or when to
stop cleanup, at different points in time:
•Determine if the residual HS contamination is at the ac-
 ceptable level for the constituents of interest (Figure 5)
•Make the cost/benefit analysis whether or not to continue
 cleanup for the constituents of interest (Figure 6)
•Determine if the total cleanup costs are within the speci-
 fied cost-time envelope (Figure 4)
  If the residual contamination is at the  acceptable level
within the specified cost-time envelope, cleanup obviously
should be stopped.  If the acceptable level has not been
reached, but the costs of continuing outweigh the benefits,

-------
                                                                               REMEDIAL RESPONSE     197
though time and funds remain, then a closer look must be
taken as to the impact of a nonacceptable residual level. If
funds and time run  out before an acceptable level is
reached and the cost/benefit analysis indicates that clean-
up should continue, strong consideration should be given
to continuing cleanup as part of remedial action. The user
of this rationale must find a solution  which is environ-
mentally,  economically, and legally acceptable.  Often a
compromise is the only possible way.

CONCLUSIONS

  This study was prepared to assist on-scene decision-mak-
ing personnel during removal activities following acciden-
tal releases of hazardous substances, as required by Public
Law 96-510, or CERCLA. Decisions to be made are on
cleanup priorities (what to clean up first?), on the selec-
tion of cleanup techniques (how to do it?), and on the ex-
tent of removal (how clean  is clean?). The section on
Cleanup Priorities contains forms on which to record the
hazardous substance incident profile, and to work out the
vulnerability ratings of all the environmental constituents
affected or  threatened by  the incident. From  these the
cleanup priorities  by environmental constituents  are de-
rived. The section on Selection of Cleanup Techniques
lists various mechanical, chemical and biological tech-
niques and indicates how choices should be made in  par-
ticular cases.  The section  on Extent of Cleanup (How
Clean is Clean?) contains forms and graphs on which to
enter projected and actual cleanup costs, the progress in
cleanup as evidenced by decreasing residual hazardous
substance levels, and cost-benefit analyses of incremental
cleanup actions. The user derives the environmental, eco-
nomic, and legal decision indicators that must be com-
bined to determine the appropriate extent of cleanup.

ACKNOWLEDGMENTS

  This study was prepared as part of  Contract No. 68-
03-3014 with the Environmental Protection Agency.
  Special thanks is given to Mr. Frank Freestone, EPA
Oil and Hazardous Materials Spills Branch, for providing
technical direction for  this project. Appreciation  is also
given to Dr. Roy Clark, Rockwell International, for his aid
in preparing Figure 1 and Tables II and VII.
REFERENCES

1. Versar,  Inc.  "Handbook  for  Oil Spill  Protection
   Cleanup Priorities." EPA-600-8-81-002. USEPA, Cin-
   cinnati, Oh. February 1981.
2. S&D Engineering Services, Inc. "Training Course Out-
   line: Hazardous Material Incident  Cleanup Decision-
   Making." USEPA, Edison, N.J. February 1981.
3. Robinson, J.S.  "Hazardous Chemical Spill Cleanup.
   Noyes Data Corporation." Park Ridge, N.J. 1979.

-------
                   HAZARDOUS SUBSTANCE RESPONSE
                               MANAGEMENT MODEL

                                           J. BILL HANSON
                                        RICHARD STANFORD
                                U.S. Environmental Protection Agency
                              Office of Emergency and Remedial Response
                                           Washington, D.C.
                                       ROBERT W. PEASE, JR.
                                          PAUL J. STOLLER
                                        The MITRE Corporation
                                        Bedford, Massachusetts
INTRODUCTION

  The Environmental Protection Agency  (EPA) is now
using on an interim basis  a Hazardous  Substance Re-
sponse Model  as  a management tool  to lead  govern-
mental project  managers through the activities related to
cleaning uncontrolled releases  of hazardous substances.
Response funds are available primarily from  the Com-
prehensive Environmental Response Compensation and
Liability  Act of  1980, P.O.  96-510  ("CERCLA" or
"Superfund"). Superfund was enacted on December 11,
1980 and establishes broad Federal authority to respond to
releases  or threats of releases  of hazardous substances,
pollutants or contaminants  from vessels  and facilities.
The  Federal  and State governments under the Act may
take response actions whenever  there  is a release or a
substantial threat of a release which may endanger public
health or welfare  or the environment.  The Model  pre-
sents in  Flow Chart form (Figure 1) a  suggested master
network of activities designed to orchestrate  a  cleanup
project in a manner that approximates the process of re-
sponse currently being used on an interim basis by the
EPA. The Model, in its completed form, is currently be-
ing developed  and will include management-related ac-
tivities, community relations, government/agency coordi-
nation, contractor  procurement, contractor monitoring
and technical decision-making.
  The goals  of the  Model are:  (1) to describe  the ac-
tivities necessary to manage cleanup of an uncontrolled
hazardous waste site; and (2) to present this information
in a form that is easy to  understand, and which will
assist those responsible for  the day-to-day management
of a project. In short, it is comprehensive in  scope and
designed  for practical  use.  The  Model is a vehicle in-
corporating the requirements of  CERCLA with the ex-
perience of the people who are  well-versed in  evaluat-
ing, planning and cleaning releases.
  The use of a management model  provides the follow-
ing benefits:
Information Transfer

  The model allows new projects to benefit  from  past
experience and to conform to CERCLA requirements by
identifying the critical decisions in the project which must
be made before succeeding activities can begin. Addi-
tionally, it provides a checklist of items which should be
considered  (but  not  necessarily  performed,  depending
upon  the  circumstances of the  specific project), thus
ensuring that important components of the cleanup are
not overlooked.
Funding Decisions
  The model organizes cleanup operations into discrete
phases  with funding  and decision points between each
phase. Therefore, the performance of a project and the
justifications for continuation come under pre-established
review  by the funding authorities.  This allows project
tracking on a large scale and improved cost estimation
because of sharper definition  and  limitation of subse-
quent activities.

Goal-Oriented Sequencing of Activities
  The model arranges  all site responses into  activities
which accomplish the immediate needs  of protecting the
public health while providing support for the development
of permanent remedial measures. Use of the model helps
to avoid unnecessary or counterproductive activities (and
their associated cost)  and serves to ensure  that neces-
sary activities or data collection are not overlooked.

Public Support and Participation
  The model  serves to facilitate public  support and par-
ticipation  by  laying out  in advance the purposes and
limitations of each operational phase. In this way, pro-
ject managers avoid generating unrealistic expectations.

Project Continuity
  The use of the model will define the  role of State and
Federal decision makers and thereby ensure that the con-
tinuity of a project is not lost if there is a changeover to
personnel. An additional benefit arises from the standardi-
zation of projects and provision of program consistency.
                                                   198

-------
                                                                                      REMEDIAL RESPONSE   199
PHASES OF RESPONSE

  There are two general  types of  removal  response-
immediate removal  and planned removal. An immediate
removal may be taken when action must be taken within
hours or days  to prevent significant  harm to  human life
or health, or  the environment.  This is the type of  re-
sponse which  has been  traditionally undertaken  for  oil
and hazardous substance removal. Planned  removals are
taken to stabilize releases where several weeks or months
are available to  identify responsible  parties, plan  the  re-
sponse and conduct procurement procedures. Planned  re-
movals bridge the gap between immediate response (which
must occur within hours or days) and remedial response
(which can take a year or more to institute).
Phase I—Discovery  and Notification
  Sites requiring response  are identified in several ways.
First, owners and operators  of inactive hazardous waste
sites  were required to  notify EPA by  June  9,  1981.
Second, the  general public can  inform  the National Re-
sponse Center or  the  EPA of  potentially  hazardous
sites.  EPA is  also  using aerial photography  to identify
potential sites.
Phase II—Preliminary Assessment
  Based on  the available  information, preliminary site
assessments  can be conducted  immediately  or deferred
to be included as part of an area-wide search.
  The purpose of the preliminary assessment is to:
•identify the source  and general nature of  hazardous sub-
 stances;
•estimate the potential of the threat;
•initially determine  whether  potential responsible  parties
 will undertake site evaluation and response activities;
•determine whether  immediate measures are required.
                             To make these determinations, the site manager evaluates
                             previous disposal practices, information from generators,
                             aerial  and  other photographs, and  literature  searches.
                             Personal interviews are useful at this  phase.  A perimeter
                             inspection  may be needed to determine the potential for
                             releases. Finally, an on-site inspection may be performed
                             if  the release may  present an immediate threat  to public
                             health and if sophisticated safety equipment is not needed.

                             Phase III—Immediate Removal
                               Immediate removal actions can be initiated at any time
                             during  response to address  acute situations. These are
                             situations where response actions must begin within hours
                             or days to mitigate fire or explosion hazards, to protect
                             drinking water supplies, or otherwise  to prevent or  miti-
                             gate harm to human life or health or to the environment.
                               Immediate removal actions  should be conducted  by
                             the responsible party. Only if the responsible party is not
                             immediately available or cannot  rapidly  respond,  can
                             CERCLA funds be used.
                               By statute, immediate removal actions should be com-
                             pleted  within six months and with less  than $1 million.
                             However, if acute harm still threatens the environment or
                             public  health,  immediate removal  actions may  continue
                             until the situation is stabilized.

                             Phase IV—Evaluation and Determination
                             of Appropriate Response
                               To determine the nature  and extent of the release and
                             the priority of the site for  response, greater  emphasis is
                             given  to on-site sampling and analysis, especially when
                             there  is  an apparent risk  to public health  or  environ-
                             ment. Activities during this phase should include:
                             •assessing amount, types and location of hazardous ma-
                              terials
                 |. DISCOVERY Cr
                   NOTIFICATION
||- PRELIMINARY
  ASSESSMENT
                                                     HI- IMMEDIATE REMOVAL
IV- EVALUATION & DETERMINATION
   OF APPROPRIATE RESPONSE
                                • BASED ON AVAILABLE
                                 INFORMATION
                                • INITIAL DETERMINATION
                                 OF SOURCE ft NATURE OF
                                 POLLUTANTS
                                • SEARCH FOR RESPONSIBLE
                                 PARTY
                    V- PLANNED REMOVAL
                      • PRIORITY SITE OR TO PREVENT
                       RAPIDLY DETERIORATING CONDITIONS
                      • CONSIST OF OPERABLE UNITS
                      • COST LESS THAN 11 MILLION
                      • TIME REQUIRED LESS THAN 6 MONTHS
                    VI-REMEDIAL RESPONSE PLANNING
                     » ACTION NEEDED IN HOURS
                      OR DAYS

                      • PREVENT/MITIGATE
                       IMMEDIATE THREAT TO
                       HUMAN LIFE Er HEALTH
                       OR ACUTE ENVIRONMENTAL
                       HARM

                      • PREVENT/MITIGATE HARM
                       TO REAL OR PERSONAL PROPERTY
 • SITE INSPECTION -
  SURVEY, DETERMINE NATURE
  AND EXTENT OF RELEASE
 • DETERMINATION OF RESPONSIBLE PARTY
 • RANK SITE FOR PRIORITY LIST
                      • PREVENT CONTAMINATION
                       OF DRINKING WATER SUPPLIES


                      VM- REMEDIAL
                          IMPLEMENTATION
 VIM- MONITORING ft
     MAINTENANCE
                                           -o-
                       -o
                                                      ) CONSTRUCTION
                                                • MAINTENANCE & OPERATION
                                                 OF REMEDY
                                                • MONITORING FOR EFFECTIVENESS
                                                 OF REMEDY
                      • STATE/EPA AGREEMENT

                      • REMEDIAL INVESTIGATION
                        - COMPREHENSIVELY DETERMINE
                        NATURE AND EXTENT OF CONTAMINATION
                      • FEASIBILITY STUDY

                        - SELECT COST-EFFECTIVE ALTERNATIVE
                        - COMMUNITY RELATIONS PLAN
                        • CONCEPTUAL DESIGN OF REMEDY

                      • FINAL DESIGN OF REMEDY
                                                      Figure 1.
                                                 Phases of Response

-------
200     REMEDIAL RESPONSE
•determining or documenting immediate threats  to  the
 public or the environment
•reviewing records
•determining sampling protocols for any subsequent re-
 medial investigations
The search for responsibile parties will continue through:
•search of local records
•interviews with local persons
•review of waste disposal records and labels.
  If a responsible party does not exist, cannot be identi-
fied,  will not act  in a timely fashion, or is  likely to be
"judgment-proof" for all or a significant portion of the
costs of the release, the site is considered  for further
fund-financed removal or remedial response.
  At this point a decision is  made to proceed with a
planned  removal  and/or  remedial  response depending
upon the need  for action (as determined by  the immedi-
acy of the threat  and the site priority ranking). The con-
ditions  under which these alternatives are selected and
characteristic activities  of each are  described in detail
below.

Phase V—Planned Removal
  A planned removal action is taken in response to a re-
lease of  hazardous  substances which,  while requiring
some near-term actions to stabilize a threat to public
health or the environment, does not require immediate
removal actions.  Such a situation allows some technical
planning  to more carefully  select  appropriate response
activities.
  Restrictions are placed on planned removal activities in
order to conserve fund monies. All planned removals must
consist of operable units, that is, self-contained compon-
ents which will achieve some interim level of  release miti-
gation or elimination without relying on future response
actions.  Further, planned removals can be taken only at
priority sites or in situations where they are  necessary to
prevent rapidly deteriorating conditions. Finally, to the ex-
tent practicable, planned removal actions should conform
to  the cost-effective analysis required  for full remedial
activities.
Phase VI—Remedial Response Planning
  Remedial actions  are most appropriate for long-term
and costly cleanups. The first step of remedial response
planning is the  remedial investigation. The purpose of the
remedial investigation is  to  determine comprehensively
the extent and  nature of the problem and to support the
development of alternative measures.  Typical  remedial
investigations include:
•geophysical and hydro-geological investigations
•soil sampling and analysis
•hazardous waste characterization
•groundwater, surface water and air monitoring.
The second step  in  remedial  response  planning is  the
feasibility study. The feasibility study uses the results of
the remedial investigation to develop and evaluate alter-
native remedies, assess  the environmental affects of the
release  and  alternative  remedies,  determine  the cost-
effective remedial action, and prepare a conceptual design
of  the approved  action.  A typical  feasibility study in-
cludes:
•definition of response objectives and criteria
•laboratory studies
•development and evaluation of alternative remedial ac-
 tions
•environmental assessment
•selection of an alternative
•conceptual design.
  The  selected  remedial measure  should  meet  public
health  and environment  requirements.  If  the  resulting
remedies are  more expensive than a predetermined cost
guideline, then  other remedies,  including non-cleanup
alternatives, shall be developed and evaluated.
  The third activity in remedial response planning is the
remedial design. In this step the  approved remedial ac-
tion will be clearly defined for implementation. A typical
remedial design may consist of the following elements:
•site response plan
•relocation plan
•engineering drawings and specifications
•contract documents.
  During the  remedial  planning process and before
Phase VII, several technical, institutional and  legal re-
quirements must be met  by the States.  The State must
show that it will be able to contribute its share of the re-
medial response planning and implementation costs, as
well as  future operations and  maintenance costs. The
State must also identify any required off-site disposal fa-
cilities required. Agreements between the State and EPA
are signed prior to initiating Phase VI, VII and VIII ac-
tions. These  agreements set forth  the State's and EPA's
responsibilities, define the scope of the project, and speci-
fy funding. When the State  assumes the lead and funds
are transferred to the State, a cooperative  agreement is
prepared. A  less formal  state agreement  (or memo of
understanding) is signed when EPA maintains the con-
tractual lead.

Phase VII—Remedial Implementation
  Several preparatory activities are necessary before con-
struction  begins.  Final permits must be  obtained and
funding  authorization  is required from the State and
EPA. The State must also demonstrate its  ability to as-
sume the responsibility for Phase VIII activities.
  Remedial implementation—particularly for large pro-
jects—may be segmented. Each segment must be an op-
erable unit and would remedy  or mitigate at least part
of the problem  caused by the uncontrolled site. Project
segmentation allows the EPA Fund Manager to preserve
the size of the Fund and  to balance response actions
across more sites.

Phase VIII—Monitoring and Maintenance
  After  the  remedial  action has been completed,  the
States monitor the effectiveness of the response and pro-
vide any operation and  maintenance necessary for the
continued effectiveness of the permanent remedy.
  Elements of Phase VIII can include:
•operation of treatment and collection systems
•reports of monitoring and maintenance program
•documentation of the determination that the substantial
  danger has been successfully reduced to design criteria.

-------
SURFACE SEALING TO MINIMIZE LEACHATE GENERATION
         AT UNCONTROLLED HAZARDOUS WASTE SITES
                                       DONALD E. BANNING
                                U.S. Environmental Protection Agency
                            Municipal Environmental Research Laboratory
                                           Cincinnati, Ohio
INTRODUCTION

  Many existing technologies, such as those currently be-
ing used for construction, hydrologic investigation, waste-
water treatment, spill cleanup and chemical sampling and
analysis,  can be applied to uncontrolled hazardous waste
sites. The minimization of surface infiltration will, in al-
most all cases, be an integral part of the remedial steps at
those sites where the waste has been buried and the cost of
removal is prohibitive.
  Minimizing surface infiltration typically consists of re-
grading,  diverting surface water runoff and preventing or
eliminating infiltration. The effectiveness of surface seal-
ing depends upon the contribution of surface infiltration
to the total problem at the  site. From a cost  effective-
ness standpoint and ease of applicability, minimizing sur-
face infiltration poses marked advantages over other types
of remedial action unit operations.
  The Solid and Hazardous  Waste  Research Division of
EPA has been involved either directly, through  actual
EPA funding or indirectly through  technically supported
efforts at several sites  where minimizing surface infiltra-
tion has  been implemented.  Two of those  sites are dis-
cussed in this paper.

WINDHAM, CONNECTICUT

  After a site selection process that involved considera-
tion of over 400 sites, Windham, Connecticut, was selected
for remedial work in 1977. The Windham,  Connecticut,
landfill is located in east central Connecticut, is approxi-
mately 25 acres in size and consists of two sections (Figure
1).
  The landfill is located in an area of sand and  gravel
immediately adjacent to the  City of Willimantic water
supply reservoir. The eastern (old) half of the site is  ap-
proximately 10 acres with its lower portion in the ground-
water. The western (new) landfill  is  approximately  15
acres and is above  the groundwater. The site was oper-
ated from approximately 1945 to 1978.
  A series of monitoring points emplaced  through and
around the site defined the area of groundwater contam-
ination and its movement with time  (Figure 2). The mon-
itoring system would also be useful in determining  the
effectiveness of the remedial action after it was  imple-
mented. This monitoring network included:
  o
HARTFORD
           OMANCHESTER
      GLASTONBURY
       CONNECTI  CUT
                        TOWN OF WINDHAM I
                   Figure 1.
Location of the Windham Landfill, Windham, Connecticut,
          from Approximately 1945 to 1978
                   Figure 2.
  Location of Monitoring Points at the Windham Landfill,
              Windham, Connecticut

(1) Suction and pan lysimeters in the refuse above the
   water table and in the surrounding  soils to de-
   termine moisture infiltration
(2) The  emplacement  of piezometers into the water
   table below and around the landfill to determine
   the quantity and quality of leachate being  gen-
   erated and its movement through the subsurface
                                                 201

-------
202    REMEDIAL RESPONSE
  (3)  The establishment of surface water sampling points
  (4)  The placement of suction lysimeters to obtain leach-
      ate in the refuse above the water table
  (5)  The implementation of a bimonthly  sampling pro-
      gram to determine  changes in the  geometry of the
      pollutional plume with time
  Between February and  June 1979, wells,  ponds and ly-
simeters were sampled on a monthly basis to determine
water quality.  The water levels in the wells  and ponds
were measured weekly and the volume of water collected
in the pan lysimeters was measured.
  The leachate, of moderate strength, was generated by
the old fill whereas leachate of somewhat higher strength
was generated by the newer portion of the landfill. Leach-
ate degraded the groundwater down-gradient of the fill.
There was a  significant contrast between the background
water quality measured upgradient of the site and that
of  contaminated  groundwater found  down-gradient of
the landfill.
  An analysis of the results of the weekly  monitoring of
groundwater and surface water elevations  and daily pre-
cipitation data verified significant infiltration was taking
place into the landfill. As much as 85% of the water pass-
ing through the refuse was derived from infiltration of pre-
cipitation.
  The mass loadings  to the Windham Reservoir from the
disposal  site prior to any remedial action are shown in
Table I.
                        Table I.
          Mass Loading of the Windham Reservoir
                by the Windham Landfill*
                 (Before Surface Sealing)
         Old Landfill
         raj/1   Ib/day
New Landfill
mg/l    Ib/day
Total Load
Ib/day   Ib/year
TOC (total
org. carbon)
Total Ions"
Sodium
Chloride
Iron
Manganese
Hydraulic
Load
242 54
4,212 948
85 19
80 18
61.3 14
4.5 1
15,000 gal/day
ground water
663 127
8,736 1677
580 111
510 98
270 52
3.5 0.7
23 ,000 gal/day
ground water
181
2.625
130
116
66
1.7

661,000
958,000
47,600
42,300
24,000
615

 *Based on chemical analyses for March 79 for MP 024 and MP (C26; background con-
 centration for all constituents negligible (MP «8)
"Specific Conductance x 1.56 = Total Ions
  Remedial actions proposed to abate and prevent pollu-
tion from the landfill included:
•Regrading of the landfill to maximize surface water run-
 off and minimize infiltration
•The placement of a 20-mil PVC top seal
•Covering the top seal with  approximately 18 inches of
 final cover
•Revegetation (see Figure 3).
The remedial  actions were designed to  be passive to  in-
sure minimum future maintenance. From the late summer
to the early winter of 1979,  these remedial actions were
implemented at the site and final revegetation was accom-
plished in the spring of 1979.
  Beginning in the fall of  1979, with the installation of
the seal,  bimonthly sampling was  instituted  at selected
                                                               Table II.
                                                 Mass Loading of the Windham Reservoir
                                                        by the Windham Landfill*
                                                         (After Surface Sealing)
                           Constituent

                           TOC
                           Total Ions"
                           Sodium
                           Chloride
                           Iron
                           Manganese
                           Hydraulic
                           Load:
                                                Old Landfill
                                                rag/1   Ib/day
                                                  43.3

                                                2808
                                                  32.2
                                                  73
                                                  14.1
                4.33

               280.61
                3.22
                7.30
                1.41
            0.20   0.02
          12,000 gal/day
          ground water
New Landfill
mg/l Ib/day
3.6
7800
290
312
51.2
0.11
0
0
0
0
0
0
Total Load
Ib/day Ib/ycar
4.33
280.61
3.22
7.30
1.41
0.02
1580.45
102422.65
1175.30
2664.5
514.65
7.3
                                                            (no ground water)
Total*


99.76
89.31
97.53
83.70
97.78
98.81
Averaje -
94%
 *Based on latest chemical analyses for each monitoring Point (March 1981 for MP 124 ind
  May 1981 forMP#26).
 "Specific Conductance x 1.56 = Total Ions.

monitoring points. Results of the sampling showed that the
impact of the site on the groundwater has been mitigated
between November 1979  and March  1981.  The current
mass loadings subsequent to the surface sealing, as well
as, the reductions achieved to date are presented in Table
II.

CHARLES CITY, IOWA

  This site was a chemical  waste disposal dump from  1952
to 1977 by a manufacturer of feed additives and veterin-
ary Pharmaceuticals for  the poultry industry. It is located
in the floodplain of the Cedar River (Figure 4).
  Since  1977, detailed studies undertaken there  revealed
that the site contains over  three million cubic feet of waste
material containing 28 priority pollutants. It lies over a
major water  supply aquifer for municipal and rural areas
of northeastern Iowa. The following  are calculated esti-
mates of some of the major compounds present:
   Estimate of the Major Chemical Components Contained In
         The LaBounty Dump Site, Charles City, la.
                                       Chemical

                                       Arsenic
                                       1,1, 2-Trichloroethane
                                       Nitrobenzene
                                       Orthonitroaniline
                                       Phenol
                                                                    Amount Ib.

                                                                    6,044,000
                                                                       70,000
                                                                      280,000
                                                                     1,500,000
                                                                       27,000
                                         Early in 1979,  the Solid and Hazardous Waste Re-
                                       search  Division (SHWRD)  was requested  by  both the
                                       Office  of Solid Waste  (OSW)  and EPA,  Region VII,
                                       Kansas City,  to implement some form of remedial action
                                       research at the site. To date, ORD effort at this site has
                                       consisted primarily of providing guidance on the place-
                                       ment  of effective  monitoring  system, assisting in the
                                       development  of final  closure plans, which included a cap-
                                       ping operation and conducting an on site waste sampling
                                       study to ascertain the feasibility of in situ stabilization. If
                                       the  capping operation (clay) does not sufficiently reduce
                                       the discharge of contaminates to the Cedar River, the over-
                                       all closure  plan calls  for the construction of an upgrad-
                                       ient cut-off wall to reduce or eliminate horizontal ground-
                                       water flows into the refuse mass.

-------
                                                                               REMEDIAL RESPONSE    203
    ^VEGETATED FINAL COVER
              r-PVC MEMBRANE SEAL
                            REGRAOEO LANDFILL
                      REFUSE
                       Figures.
       Typical Section Through the Windham Landfill
  Between 10/2/79 and 10/9/80,  an elaborate monitor-
ing  system was used to quantitate the discharge of three
pimary/indicator contaminates from the site to the Cedar
River (Table III).
  The surface infiltration minimization plan for the 13.3
acre area of the Charles City Site consisted of a two foot
thick clay cap, ground cover, erosion control at the site
toe, storm sewer rerouting, surface runoff controls and
post installation monitoring. The capping was completed
in November  1980. Monitoring data taken to determine
its effectiveness are presented in Table IV.
  A comparison  of the calculated LaBounty load  con-
tribution to the Cedar River before and after the capping
operation is contained in Table V.
   The wide range of concentrations for the contaminants
before and  after sealing makes it difficult to conclu-
sively  say  whether  or not the capping operation was
effective.
   An alternative conceptual approach was evaluated by
the National Enforcement Investigation Center in Denver,
Colorado. In this analysis, the investigators made a basic
assumption that near-site and on-site rainfall are the pri-
mary driving forces for producing leachate. By this meth-
od, anomalies in  correlated data might indicate effects of
other possible derived forces such as upward leakage from
the Cedar Valley  aquifer or bank recharge from the Cedar
River.
   The analysis suggested that shortly after a rain, a leach-
ate pulse forms  and begins moving toward the Cedar
River. The pulse is characterized by a higher flow volume
                                                                                                   • SEE BELOW
                                                   Figure 4.
                                       Site Location, LaBounty Disposal Site

-------
204     REMEDIAL RESPONSE
                                                     Table HI.
                                               LaBounty Disposal Site
                                               Cedar River Monitoring
                                               (Before Surface Sealing)
                       STATION 11
                                                              STATION 12
                           CALCULATED
                           LABOUNTY CONTRIBUTION
Day-Yr
10/2/79
1/14/79
2/13/79
4/16/80
5/08/80
6/13/80
7/11/80
8/7/80
9/10/80
10/9/80
Note:
Arsenic
Cone. Load
2.40
<2.00
4.30
6.4
7.4
4.2
8.2
4.1
4.0
^2
Cone.
Load
*Measured flow
4.7
<9.5
7.2
24.7
13.5
15.0
24.1
6.2
18.4
6.47
PPb
Ib/day
ONA
Cone.
0.29
12.
0.32
0.049
0.14
0.158
0.07
0.07
0.07
0.14

Load
0.6
56.8
0.5
0.19
0.25
0.56
0.21
0.11
0.32
0.45

at Station 11 used in
1,1,2-TCE Arsenic
Cone. Load Cone. Load
6.1 12.0 20.9 40.8
5.4 25.6 9.2 43.6
9.3 15.7 43.4 73.1
<5 19.3 15.5 59.9
^5 9.1 33.5 60.9
<:5 ^17.8 33.8 21.

-------
                                                                            REMEDIAL RESPONSE    205
                                                 Table V.
                                     Loadings to Cedar River from LaBounty
                                          Before and After Capping
                    Before
           LaBounty Contribution
                  After
       LaBounty Contribution
Arsenic
Load
36.1
34.1-43.6
65.9
35.2
47.4
106
49.3
38.4
102
35.6
ONA
Load
9.7
80.2
11.5
13.9
7.05
25.9
19.6
3.55
11.2
10.6
1,1,2-TCE
Load
11.4
12.3
23.6
3.9-23.2
5.4-14.5
21.5-39
11.7-26.4
4.6-12.2
9.3-32.3
0-16.1
Arsenic
Load
28.2
33.2-36.1
48.0
64.2
82.5
48.0




ONA
Load
5.42
3.11
3.53
4.11
6.65
3.92




1,1,2-TCE
Load
9.0
7.0
5.0-13
3.0
0.0-2.9
0.0-18




     Q= 4x13.3x0.052x1,440 = 3,984 gal/day

  The net effect of the site capping should be to reduce
infiltration through the site by  80%. As a result  mass
loadings/migration due to infiltration will be reduced by
CONCLUSIONS

  A synthetic capping  remedial  action  at  Windham,
Connecticut under rather ideal hydrological conditions
lead to an overall 94.0% reduction of the mass loadings
between March 1979 and March 1981.
  In a complex hydrologic situation 6 months of post re-
medial action monitoring is inadequate to determine effec-
tiveness. Conceptual approaches to predicting the effec-
tiveness of capping operation even in complex hydrologic
settings, are available, but the verification of the extent of
their accuracy remains to be proven.

ACKNOWLEDGEMENTS

  Steven W. Sisk, Hydrologist, EPA, Office of Enforce-
ment, National Enforcement Investigation Center, Den-
ver, Colorado, for his contribution "Hypothetical Assess-
ment of Rainfall Effects  on Leachate Production at the
LaBounty Site, Charles City, Iowa."

-------
     APPLYING TECHNIQUES FOR SOLIDIFICATION AND
          TRANSPORTATION OF RADIOACTIVE WASTES
                             TO HAZARDOUS WASTES

                                          J. W. PHILLIPS
                           Hittman Nuclear and Development Corporation
                                         Columbia, Maryland
INTRODUCTION

  The equipment discussed in this paper was originally
developed in response to the needs of the nuclear power
industry for  processing  and  transporting  radioactive
wastes. Prior to 1979, liquid wastes, primarily in the form
of boric acid and miscellaneous chemical drains, could be
transported for burial in their liquid state. In mid-1979,
the licenses for the three burial sites were revised to pro-
hibit acceptance of liquid  wastes. Utilities whose plants
did not contain  systems  capable of solidifying liquid
wastes were faced with a crucial decision. They could cut
back on the volume of liquid wastes being generated and
treat them with deep bed demineralizers or they could hire
an outside firm to come onto the site and, with portable
equipment,  solidify the liquids  prior to shipment. For
many utilities, the first option was not feasible due to the
large quantities of ion exchange  resins that would be re-
quired to treat the liquid wastes involved.
  Both mobile processing equipment, and equipment that
is permanently installed on-site are discussed in this paper.
The transportation equipment was designed and built to
conform to specific requirements pertaining to radioactive
wastes. The purpose  of the following  discussion is  to
identify how these types of equipment could be modified
and used to accommodate hazardous chemical wastes.

SOLIDIFICATION OF LOW-LEVEL
RADIOACTIVE WASTES

  As stated previously, the solidification  of low-level
radioactive  wastes can either be  handled  with  mobile
equipment,  leased and  operated  by a company  that
specializes  in  this  work or the  equipment  can  be  pur-
chased and installed for operation by a utility. Numerous
current suppliers exist for  both  mobile solidification
services and supply of systems  for  permanent  in-plant
installation.
  The types of wastes being solidified are spent ion ex-
change resins, concentrated solutions of boric acid and
sodium sulfate and  calcium  fluoride  sludges.  Various
lubricating and turbine oils have also been  successfully
solidified. Most of these wastes have, for the past sev-
eral years, been solidified  with  Portland Cement using
various additives to condition the wastes based on their
chemical characteristics. A  urea-formaldehyde (UF) based
material has also been extensively used over the past 5
to 7 years.
  The use of UF has come under increasing attack over
the  last two years  because it is a condensing polymer
and produces liquid as it polymerizes. When the resin
hardens, it shrinks, forcing  liquid  out  of the matrix.
This liquid is very  low in pH as a result  of the acidic
catalyst used to initiate the polymerization process.
  A third product, which is only beginning  to be used, is
a material  produced by Dow  Chemical Company re-
ferred to as the "Dow media." The Dow process is a
three part  process which uses a binder, a promoter and
a catalyst.  For  certain wastes, an extender is also re-
quired. The proportions of materials used, and the chem-
ical formuli of the ingredients, are  proprietary to Dow
Chemical Company.
  Bitumen, or asphalt, has been used  for the solidifica-
tion of radioactive  wastes in Europe  for several years.
While a few systems using bitumen have  been sold in
this country for installation in nuclear power plants, none
are in operation at this time.
  Due  to numerous recent  changes  in the  burial site li-
cense  in  Barn well,  South  Carolina, there has been a
significant increase  in  the number of mobile systems in
use over the last two years. Power plant operators have
turned to mobile solidification services  in order to satisfy
evolving criteria. This  increased use makes it very diffi-
cult to determine the exact number of power plants using
mobile systems at any one time. However, it is estimated
that 60 to 70 percent of the operating units were contract-
ing  for some form  of  mobile solidification service as of
July of this year.
Typical System
  Since almost all solidification of low-level radioactive
wastes is performed using cement, a typical solidification
system currently in use will be described.  The compon-
ents of the system include a disposable container with in-
ternal mixing blades, a mixing head, a mixing motor, a
cement storage hopper, a cement feed system, a dust col-
lector, a pump skid and a control panel.
  When the radiation level of the waste materials is high
enough to require shielding for the operator, the solidi-
fication process is  performed with the  disposable liner
either in a process  shield or in a shielded  shipping cask
(which can be used  for subsequent transport of the waste
                                                  206

-------
                                                                                REMEDIAL RESPONSE     207
to the burial site). After the liner has been placed in its
desired location, the mixing head with mixing motor is
bolted to the liner. A thick rubber gasket between the mix-
ing head and the liner neck provides a seal between the
inside of the liner and the  outside atmosphere. All other
connections to  the  liner are  made  through the mixing
head. A prefilled cement hopper is positioned on a scaf-
fold and connected to the cement inlet valve on the mixing
head via a flexible screw feeder. The cement feed motor is
located on the mixing head. An induced draft dust col-
lector is connected to the vent connection on the mixing
head to catch any cement  dust. A central control panel
contains the necessary stop/start switches, indicators and
control logic to operate the system.
  Mobile systems are leased to a utility for periods rang-
ing  from the duration of a  single  solidification opera-
tion to several years  of service. Operator services are sup-
plied with the equipment. When the duration of the con-
tract is long enough and  the frequency of use is high
enough, operating personnel are permanently located at
the site. Upon termination  of the contract, the equipment
is removed from the site.
            The procedure is quite  different  for  permanently in-
          stalled systems  which are owned,  operated  and main-
          tained by the utility. They are located within a building
          designed especially to afford the operator the maximum
          possible radiation shielding. These shield walls, however,
          tend to make the equipment less accessible. In-plant sys-
          tems  are substantially more automated than  the mobile
          systems.  A simplified process flow diagram  of the in-
          plant solidification system being installed at the three-
          unit Palo  Verde plant west  of Phoenix, Arizona,  for
          Arizona Public Service is shown in Figure 1.
            Aside  from the  degree  of automation involved,  the
          primary difference  between a mobile system  and an in-
          plant system is now and where the mixing of  wastes and
          solidification medium takes place. In an  in-plant system,
          the mixing takes place outside the disposal container, be
          it a 55-gallon drum or a large liner.  In the system shown
          in Figure 1, wastes to be solidified are transferred to the
          waste batch tank, where samples are taken and appropri-
          ate chemical adjustments are made prior to solidification.
          The batch  tank  is equipped with level indicators and an
          electric mixer. The waste transfer  pump transfers  the
             WASTE
             INLET
 WASTE
 BATCH
 TANK
                                    CHEMICAL
                                    ADDITION
                                    TANK
                                                                            BULK
                  METSO
                  STORAGE
                  SILO
BULK
CEMENT
STORAGE
SILO
       WASTE
       FEED
       PUMP
                                                   CHEMICAL
                                                   ADDITION
                                                   PUMP
                                                                      SCREW  CONVEYOR
MIXER
                         FLUSH
                         MODULE
      DISPOSABLE
      CONTAINER
                                                [  TRANSFER CART
                                                  o
                                                   Figure 1.
                                            Simplified Palo Verde PFD

-------
208    REMEDIAL RESPONSE
wastes at a predetermined flow rate from the batch tank
to the waste mixer.
  The cement and Metso* are stored in silos, designed to
hold large quantities of these materials to take advantage
of bulk deliveries. Cement and Metso are discharged from
the silos  at feed rates controlled by horizontal screw con-
veyor. Direct current drive motors are used which allow
the feed rates to be adjusted depending on the desired
mixing ratios of waste, cement and Metso for each par-
ticular waste form. The cement/Metso mixture enters  the
unit from the top at the back end of the mixing screws.
As the cement is moved forward, the pumped wastes  en-
ter the mixing barrel and are mixed with the cement. The
discharge valve is opened and the cement  paste drops into
the disposable container.
  At the end of each cycle, the waste feed system and the
mixer are completely flushed. Extra dry cement added to
the liner, or to the last drum, adsorbs the flush water. The
primary flush module contains an air  accumulator bottle
which provides the motive force for flushing the mixer in
the event of a loss of electrical power.  The mixer also
includes  a  hand crank on the  motor so that it  can  be
cleaned out manually if necessary. After  the container is
filled it is moved to a capping station where a lid is auto-
matically affixed. The  container is checked for  external
contamination, decontaminated if  necessary, and moved
into a shielded storage area to await transportation to the
burial site.

TRANSPORTATION

  Obviously, the  primary design consideration of radio-
active waste shipping casks  are their  radiation shielding
capability. This concern is of little consequence in trans-
porting hazardous chemical  wastes. However, the struc-
tural  integrity  required in a design  that  must  support
thick steel and lead walls, and that conforms to the NRC
regulations, results in  a package whose  ability to with-
stand foreseeable  transportation  accidents is  not easily
matched.
  Several types of shipping containers  exist including
standard truck vans, shielded vans, box casks for hand-
ling small numbers of drums, and large cylindrical casks
for shipping liners or drums. Figure 2 shows a box cask
which holds 12 drums. A roller assembly pulls out from
the end of the box for ease in loading and  unloading
drums. A cylindrical cask which will hold either a 170-ft3
steel liner (6 ft high by 6 ft  in diameter) or 14 SS-gallon
drums is  shown in Figure 3.  The SS-gallon drums used in
the nuclear industry are designed and tested in accordance
with DOT Specification 17H and are certified to hold 840
pounds of waste.
  The  tie down connection  of both  the box  casks and
cylindrical casks are designed to withstand the following
acceleration loadings:
•Front to back     10 g           "Vertical           2g
•Side to side        5 g

•Registered trade name for sodium metasilicate anhydrous, manu-
factured by Diamond Shamrock,  Soda  Products Division. The  full
product name is Metso Beads 2048.
Thus, the tie down connections on a typical shipping cask
weighing 50,000 pounds are designed  to withstand simul-
taneous loads of 500,000 pounds forward, 250,000 pounds
sideways and  100,000 pounds  upwards. Although  the
trailers used to transport the cask cannot withstand these
forces, the strength of the containers insures that the con-
tents will remain undamaged  and the cask  integrity will
not be breached.
                       Figure 2.
           Reusable Cask Which Holds 12 Drums
                        Figure 3.
  Reusable Cask for a 170 ft' Steel Liner or 14-55 Gallon Drurro

-------
                                                                                   REMEDIAL RESPONSE    209
  Drop tests must also  be performed  on the casks. The
cask must be dropped onto an unyielding surface with no
resultant damage  to its  contents or to the cask integrity
when dropped in every orientation. Two drop tests exist:
a one-foot  drop test  and a thirty-foot  drop test. Most
casks  are designed to withstand  the  one-foot test,  al-
though in all probability these casks would not yield in a
drop test of as much  as ten feet. Only specialized casks,
designed to handle "large quantities" of radioactive ma-
terial, will withstand the thirty-foot drop test.


CHEMICAL VERSUS RADIOACTIVE WASTES

  Before the  use  of the cement solidification technique
for toxic and hazardous wastes is examined, the advan-
tages and disadvantages of handling  chemical wastes as
compared to radioactive wastes should be noted. This is
done in Table I.
APPLICATION OF EQUIPMENT TO SPECIFIC USES

  In considering  using the equipment described above
for radioactive wastes for the immobilization of hazardous
wastes, one must  first define the conditions under which
hazardous wastes  are found.  It is  expected that the con-
tents of many of the drums will  be unknown,  lagoons
containing undefined substances will be encountered and
the quantities of contaminated soils and  sludges  will  be
considerable.
  Drums containing solids are not  candidates for solidifi-
cation. The emphasis in this paper is on the solidification
of liquids and pumpable sludges. Containers that are  in-
tact must be opened and the contents inspected. Breached
drums must also be inspected to determine the status of
their contents.  Once  the  containers with  liquids are
identified, their contents must be  transferred to  a batch
tank or a disposable liner depending on the type of system
to be used. This transfer from the dumping grounds to
the solidification area will  probably be the most critical
and most dangerous step in the process.

                       Table I.
       Advantages and Disadvantages of Handling of
     Chemical Wastes Compared to Radioactive Wastes
         Advantages
 1. The material is not  radioactive,
 therefore proximity of the worker to
 the waste, or thicknesses of walls and
 material density, are not a problem.
 2. Contact of the waste with protec-
 tive clothing can be permitted.
 3. Glass and/or plastic viewing de-
 vices can be built into  the system
 wherever deemed appropriate.
        Disadvantages

1. Small exposures to  some wastes
through inhalation or physical contact
could be fatal even in very small doses.
2. Contents of most containers will be
unknown and identification nearly im-
possible.
3. Some of the wastes may be explosive.
4. Procedures  for the transfer of the
liquids to the batch tank or a disposal
liner must be carefully planned.
5. Synergistic effects may result when
different wastes are mixed.
6. Chemical analysis of the contents of
each container will have to be made.
7. Process control plans will have to be
developed for each site.
ADAPTATION OF MOBILE
SOLIDIFICATION SYSTEMS

First Example
  In order to assess how the equipment could be used, two
scenarios  were developed and studied. The first situation
is the best case for a mobile system. It is a small site con-
taining  drummed wastes  and  requires that  no  liquids
leave the  site.  It is  assumed  that the number of drums
present is less than a thousand, with most still intact.
  Before  the actual solidification process can start, it is
recommended that a prefabricated  enclosure be erected
over a prepared  surface. The prepared surface prevents
any  spilled  liquid from further contaminating the sur-
rounding  soil. The  prefabricated enclosure  will protect
the equipment and personnel from  the  elements and es-
tablishes a controlled environment around the process to
minimize airborne hazards.
  Drums  are  brought into the enclosure where they are
opened, inspected and samples taken for  analysis. The
chemical analysis can be best accomplished in a centralized
laboratory equipped to handle toxic and hazardous ma-
terials. It  should be assumed that a particular drum could
be in the enclosure  for as long as  two weeks while the
chemical  analyses are being performed. When  the con-
tents of several drums have been selected as  a combined
batch for  solidification, additional samples must be taken.
Samples of equal size from each drum are combined and
the proper pretreatment, as determined by the  chemical
analysis, is made. Typical pretreatments include pH ad-
justment, addition of emulsifiers or addition of solvents.
  The goal  of the pretreatment process is  to produce a
mixture which will dissolve or remain suspended  in water.
This is  essential  since  the setting of the cement is de-
pendent on the presence of water.  Upon completion of
the pretreatment  process a small quantity, approximately
100 to 400 ml of the mixed wastes, are solidified as a test
sample. Several samples may actually be made using dif-
ferent ratios of waste and cement.
  In the final analysis,  two samples are  selected to repre-
sent a range of waste to cement ratios that will result in
an acceptable product.  These ratios are then used to de-
termine the optimum quantity of wastes to be solidified in
a single operation and to estimate the quantity of cement
necessary to successful solidification. This procedure,  or
process  control program,  is repeated for each  batch  of
wastes that differs significantly from previous batches.
  When the successful  solidification tests are completed,
the actual waste processing can be initiated. Since the
chemical analysis previously performed has identified the
wastes, the appropriate precautions  can be taken regard-
ing proper clothing and  the  necessity for breathing ap-
paratus. The operating personnel, properly dressed, would
then transfer the contents of the drums to the mixing liner
and the designated pretreatment would be accomplished.
  Liquids would  be transferred by pump.  Sludges and
pasteous  solids might have to  be hand transferred. For
large quantities  df  nonpumpable materials,  the drums
might better be emptied into  a mixing tank with the top
opened  or half-opened. The material could then be pre-

-------
210     REMEDIAL RESPONSE
treated and mixed until a slurry is developed that can be
pumped into the disposable liner for solidification.
  Once the waste is in the liner, the operation is identical
to the operation for radioactive wastes. Cement and any
designated dry additives are put into the liner through a
screw feeder at approximately 100  Ib/min.  During this
phase of the operation, a dust collector continuously pulls
air  from  the liner and exhausts it through  a baghouse
filter. Additional  off  gas controls may have  to be  in-
cluded depending  on the type of materials handled, and
gas generation expected due to the  pretreatment opera-
tion or to the reaction of the wastes to the solidification
media.
  There  are obviously certain materials that  do not lend
themselves to  this on-site solidification.  Materials too
hazardous to  risk direct handling could be  handled re-
motely. However, the type of equipment necessary  for
remote handling is not conducive to the limited scope of
this scenario. These materials would  have to be packaged
and sent to a centralized facility.
  With the solidification process complete, the mixing
motor is removed and all other connections  are broken.
The container is capped with either a standard 55-gallon
drum lid  or with a snap tight, nonremovable lid. At this
time the liner is moved to a suitable storage location to
await transportation.

Second Example
  The second scenario also describes the use of a mobile
system and is, again, a small site, but one in which broken
and leaking drums  litter the site causing contaminated
soil and sludges. Damaged and/or leaking drums could be
handled  in a  centralized processing  facility if they  are
considered too dangerous for  on-site processing. These
drums could be repackaged  in  an overpack such as an
83-gallon drum for  transportation.  The basic setup  de-
scribed  in the first  scenario for handling intact drums
could still be used, although precautions would be neces-
sary due  to the high probability of exterior contamina-
tion.  Decontamination facilities could be  provided with
subsequent solidification of the decontamination  solu-
tions.
  Small  quantities of sludge and contaminated soil  are
best handled onsite in loading drums or liners (depending
on  the viscosity of the material), which are  then moved to
the solidification facility. In-drum mixing systems can be
used for soils to keep the motor sizes reasonable.

ADAPTATION OF IN-PLANT
SOLIDIFICATION SYSTEMS

  The significant question regarding toxic and hazardous
wastes is  how  to  handle large uncontrolled sites where
thousands of drums  and millions of gallons  of chemical
wastes have been  indiscriminately dumped.  Cleanup of
these sites will take years. No simple solutions are avail-
able. The costs are going to be high and the  progress slow.
In most  instances complex  processing  facilities built at
the existing dump sites will be the technical, if not the
political,  solution.  The ultimate disposal of  the newly pro-
cessed wastes will depend on the specifics of each site.
   Once the decision is made to process existing wastes by
immobilization in an inert binder such as cement, a de-
termination of the exact nature of the wastes to be pro-
cessed and the support facilities  required must be made.
The forms which these wastes may take include:
•Liquids from lagoons, settling ponds and drums
•Sludges from lagoons, settling ponds and leaking drums
 or other containers
•Contaminated  soils caused  by leaking containers or di-
 rect dumping of liquids and sludges
•Pasteous solids from breached and/or intact containers
•Solids in drums or in other containers or from open con-
 taminated sites

Facility Design
   Using these five waste forms as a basis for facility de-
sign and considering the possible hazardous nature of the
materials, a conceptual facility flow pattern can be made.
This flow pattern,  shown in Figure 4, includes several
features commonly used in the nuclear power industry for
waste handling and  personnel  access control. Depending
on the specifics of the site and the wastes, many of these
operations can be performed remotely. These remote op-
erations are performed  using TV  monitors mounted
throughout the  facility and overhead bridge cranes to
move containers about.
   As shown in Figure 4,  materials  from the site are
brought into the facility at the receiving point, probably by
truck.  The containers are unloaded and  moved into an
inspection  room where the container  is opened and the
contents inspected. If the contents are stable, a decision
can be made to send the container to the storage area for
shipment or to the repackaging area for placement in an
overpack.  In  either case, a  sample may be  taken for
analysis to aid in determining the proper method of ulti-
mate disposal.
   Containers, whose contents  upon inspection are iden-
tified  for  solidification,  are  first sampled.  While the
samples are being analyzed,  the containers are sent  to a
contaminated  storage area  after  appropriate labeling.
Containers with similar contents are identified and sample
solidifications   are   performed   to   develop  specific
process parameters  and  determine necessary additives.
These  containers  are  then retrieved  from storage and
moved into the area with the batch tank. The contents of
each container are transferred to the batch tank by pump
or by  picking  up the container using special handling
equipment and pouring. The batch tank is sealed and the
contents mixed with appropriate solvents, emulsiilers and
chemicals for pH adjustment. At the end of the mixing
cycle a sample can be drawn for verification purposes.
Remote Operations
   This  section discusses a solidification operation that is
controlled remotely and automatically from a central con-
trol  panel or control room. Dry cement feed is fed along
with the waste feed  but enters the mixing device prior to
the wastes. Feed rates are controlled and can be adjusted
if necessary. The cement paste free falls into a disposal

-------
                                                                                REMEDIAL RESPONSE     211
            r
                 UNCONTAMINATED
                 MATERIAL  FLOW
                                                BATCH TANK
                                                & PRETREATMENT
                                            T	T	L	
       —  Personnel Access
       —  Old  Containers  from  Site
       —  New  Containers
       —  Materials
                        CONTROL  ROOM
                                                   Figure 4.
                                 Hazardous Waste Solidification Facility Flow Diagram
container until a preset level is reached. The interface be-
tween the mixer and the container is sealed with appropri-
ate vent  connections.  Ultrasonic level probes are used to
prevent contamination of the instruments.
  When  the container is full, waste feed is terminated by
either diverting it back to the batch tank or by completely
flushing  the lines. The  diversion mode is selected when
additional containers are to be processed within a reason-
able time frame. If the batch tank is empty,  or insuffici-
ent material is available to process another container, the
feed line is flushed. Additional dry cement is fed through
the mixer and  into the container  to  hydrate the flush
water. Once the container is full it moves out of  the fill
position  into the capping station where it is capped and
sealed.
  The solidification   process itself has several  fail-safe
design features built into it. First, the motor for the mixer
is located outside  the solidification area  and contains a
hand crank. This back-up system permits emptying of the
mixer should the process be stopped in mid-stream due
to motor failure or loss of electrical power. Maintenance
on the motor can  also be performed without entering a
contaminated area.
 Second, the system flush is controlled through a flush
module mounted outside the solidification area, again for
maintenance  purposes.  The  flush  water is  kept under
pneumatic pressure at all times so it is available even dur-
ing a loss of electrical  power. This module is  used for
normal operations to ensure  that the proper quantity of
flush water is used and to ensure that the unit is operable
if needed in an emergency.
  Capped containers are inspected and tested for ex-
ternal  contamination and decontaminated if necessary.
With the container certified  as free of contamination it
should be labeled as to its contents  and status and stored
until final transportation for disposal.
REFERENCES

1. Phillips, J., Feizollahi, F., Martineit, R., Bell, W. and
   Stouky, R. "A  Waste Inventory Report for  Reactor
   and  Fuel-Fabrication  Facility  Wastes,"  ONWI-20.
   Prepared for U.S.  Department  of Energy, Office of
   Waste  Management and Battelle Memorial Institute,
   March 1979.
2. Guilbeault, B.D., "The 1979 State-by-State Assessment
   of Low-Level Radioactive Wastes  Shipped to Burial
   Grounds," NUS-3440, Rev. 1.  Prepared for EG&G
   Idaho, Inc., November 1980.

-------
         RENOVATION OF A WOOD TREATING FACILITY
                                      W. LAWRENCE RAMSEY
                                   O'Brien and Gere Engineers, Inc.
                                           Washington, D.C.
                                       RICHARD R. STEIMLE
                          Maryland State Office of Environmental Programs
                                         Baltimore, Maryland
                                       JAMES T. CHACONAS
                                   Maryland Environmental Services
                                         Annapolis, Maryland
INTRODUCTION
  In August 1981, work began on the renovation of an
out-of-service wood  processing facility located on a 130
acre site in St. Mary's County, Maryland. The project is
a cooperative effort between regulatory and service agen-
cies and the  private sector to clean up a significant haz-
ardous waste problem in a fairly commonplace industrial
operation.

BACKGROUND

  In the summer of 1975,  inspectors of the Maryland
WRA determined that the waste practices at a wood treat-
ing facility in  St. Mary's County, Maryland, might be
impacting ground water. Subsequent investigation  indi-
cated  that wastewater from  a creosote and pentachloro-
phenol (penta) wood  preserving operation was being dis-
charged to unlined  "evaporation  lagoons." While the
practice was obviously unacceptable, the regulatory pro-
grams at that time were not designed to handle  ground-
water discharges.
  After much discussion, it was decided that the best re-
vision to this  facility would be to  construct a  recycle/
treatment  facility utilizing  spray  irrigation for excess
wastewaters. Using this  system, most of the penta and
creosote would be recycled. Wastewater containing tan-
nins and lignins in addition to waste preservatives would
be lagooned and disposed of by spray irrigation. The esti-
mated discharge would be about 1,000 gal/day.
  In addition to the ongoing problem, the company would
be required to renovate  the existing lagoons and clean up
the area. In August  of  1977, a State Discharge Permit
was issued requiring  the development of a plan  for site
renovation and requiring the construction of a wastewater
treatment facility. During the summer of 1977, the  State
obtained some composted sewage sludge to  incorporate
into the contaminated soils  to determine if  biodegrada-
tion by land farming was possible.
  When the  test proved successful, the original plan was
modified to  include  on-site  renovation  of soils  contam-
inated by  the  waste  ponds. The idea was  to construct
three spray irrigation  fields. One would be in use, one as
a back-up and the final one to reclaim the contaminated
soils.  Each  year a  portion  of the  contaminated soils
would be mixed with compost, tilled and seeded.  The fol-
lowing year, that site would be rotated into spray irriga-
tion use. Any remaining liquid waste in the ponds would
be treated in the first year with the normal plant waste.

SITE INVESTIGATION

   In 1975, samplings of  waters in a spring  fed "fresh
water  pond"  near  the wood  treating facility  showed
phenolic concentrations of  1.80 mg/1. Analyses of oily
material floating on the pond was determined to contain
77.5 mg/1 of phenolics. There was no evidence of a sur-
face connection between the plant's waste ponds and the
fresh water pond. It was observed at the site that the oily
material was "bubbling"  up from the spring  end of the
pond.
   These observations made it necessary to consider the
imposition of site renovation conditions on the  property
owner. To support  the conditions, a field investigation
was begun.
   On May 6, 1975,  three monitoring wells were installed
in augered soil borings. The boring logs showed that a
coarse to  fine silty sand formation existing from the sur-
face to a depth of 15 to 20 ft. Below this is a sandy, blue/
black  clay which extended to a depth of at least SO ft.
Field investigators noticed strong phenolic odors in soil
samples 9 to  13  ft below the surface. The  monitoring
wells were constructed with 1.25  in.  PVC schedule WO
pipe with 5 ft slotted screens.  The screens were located
approximately 15 to 20 ft below the surface. The water
table was between 1.5 to 3.5 ft from the surface.
   Water  samples obtained  from these  three  wells on
June 11, 1975, showed concentrations of phenols of 1.0f
9.0 and 14.4 mg/1. The  pH of the groundwater ranged
from 5.5  to 5.7  The groundwater flow determined from
water  elevations in the monitoring wells was from the
wastewater holding  ponds to the fresh water pond. The
conclusions of this interdepartmental study was that waste
phenolics  in the ponds had percolated into the ground*
water and then migrated to the fresh water pond.
   The  stream which  drains the fresh  water pond WM
studied by a private consultant. In that study, a water
 sample taken  approximately 900 ft below the pond con-
 tained 0.12 mg/1 phenol. The consultant reported that
few aquatic insects  were found  in this section of the
stream. However, phenol concentrations decrease and the
biomass increases substantially within 0.25 mile from the
                                                   212

-------
                                                                               REMEDIAL RESPONSE    213
first sample point. The data indicate a significant improve-
ment of water quality as the distance from the fresh water
pond increases.
  To support the pending legal actions, a more detailed
hydrogeologic investigation was conducted. On December
7, 1978, WRA investigators sampled ground and surface
waters. A  sample of the groundwater  from one  of the
wells installed in  1975 showed creosote concentrations of
1.6  mg/1 and pentachlorophenol concentrations of  1.34
jig/1. Analyses of liquid samples in the holding  ponds
showed PCP concentrations ranging from 0.62 to 66.06
/ig/1 and creosote concentrations ranging from 0.654 to
70.72 mg/1. Analyses of  the stream samples showed PCP
ranging from 0.01 to 27.0 Mg/1 and creosote  concentra-
tions ranging from   0.001  to  1.769 mg/1.  Pentachloro-
phenol and creosote concentrations in  the fresh water
ponds were 0.01 Mg/1 and 0.308 mg/1 respectively.0'
  Samples taken  at  various monitoring wells on the site
indicated that the phenolic concentrations increased  with
depth  until the water table was intercepted. At the water
table, analyses showed a band of high phenolic concentra-
tion that  indicated   that the  wood preservatives were
"floating" on the groundwater.
  Sampling of shallow dug residential  wells in the  area
failed to substantiate the County  Health Department's
sampling that showed high phenolics. Samples also showed
phenolics in monitoring  wells up the hydrologic gradient
from the facility. It  was proposed that an air  vector was
responsible for carrying vapors from the venting  of the
pressure vessels to these sites.  This theory  will never be
tested as the site is shut down.

TECHNICAL BASIS FOR
THE RENOVATION PLAN

  Renovation of  the wood preserving plant incorporates
basic biological wastewater  treatment processes currently
used in the wood preserving industry.® For this project,
freshly composted  sludge from the Washington, D.C.
Wastewater Treatment Plant was mixed with the creosote
and penta  contaminated soils to accelerate biological de-
gradation.  The composted sludge provides a biologically
active soil  addition containing up to 25 % dead and living
microorganisms.  Bacteria, actinomycetes and fungi com-
prise most  of the active microbial population.(3)
  Since the contaminants are by nature,  bactericides, the
microorganisms introduced to the contaminated soils will
undergo some acclimation.  Following acclimation, how-
ever, creosote reductions exceeding 90% have been  ob-
served in activated sludge and soil irrigation wastewater
treatment processes.  Additionally,  penta reductions ex-
ceeding 99%, during a 50-day period, have been reported
in a bench-scale  study  utilizing activated  sludge treat-
ment/4' Increased soil organic matter has also been shown
to expedite penta degradation within the soil medium.(5)
  The composted sludge to be used for site renovation
will be composted for 21  days and in lieu of the usual cur-
ingprocess will be applied to the site. Incorporation of the
composted sludge into the waste contaminated soils will
not only restore  the microbial population  but will also
create  optimum soil  conditions  for  biological activity.
These  soil conditions include pH,  structure, moisture
and  temperature.  The compost contains approximately
15% limestone and has a pH near neutral. At the planned
application rate  125-130 tons/acre, the composted sludge
should adjust the previously acid soils to a pH more con-
ducive to microbial activity.
  The subsoils on the site range from sandy to clayey.
Incorporation of  the  composted  sludge into  sandy or
droughtly soils increases the moisture  holding capacity of
the soil.  In a clayey or compacted soil, the  incorporated
compost enhances soil structure allowing better water and
gas exchange within the soil and  thus promoting a more
aerobic environment.
  Soil temperature will be affected not only by microbial
activity, but also by its ability to hold solar radiation. The
existing soil color will be darkened  by  the composted
sludge additions. The darker  soil color should maintain
warmer soil temperatures later in the fall and early in  the
spring and consequently enhance biological activity during
colder months of  the year. In addition to microbial  de-
gradation of the creosote and penta contaminants, sun-
light should contribute to the  destruction of other phen-
olics and wood  chemicals particularly in existing lagoon
and "fresh water" pond areas.
  Contaminants which have leached past the zone of mi-
crobial activity in the  soil, are expected to eventually dis-
appear via dilution.  Following seeding,  the composted
amended soils should  be  more than adequate to promote
revegetation of the site.(6)

METHOD OF SITE RESTORATION

  Site restoration for this  facility is  to  be  executed  ac-
cording to plans prepared by Lyon Associates,  Inc. of
Baltimore, Md.  Following general  site  inspection,  all
supply and test wells are to be abandoned and sealed. The
lagoons are to be  drained and spray irrigated in a desig-
nated disposal  area.  Following this procedure all con-
tainers will be cleaned and stored on site for removal. At
the same time the lagoons will be breached to  prevent
further ponding of water.
  Following dewatering, all  contaminated structures in
the pond area such as cement foundations, steel pipes,
etc., will be removed  and hauled to an approved  hazard-
ous waste facility. Proper erosion and sediment controls
will be utilized during the land clearing.
  The "fresh water"  pond will be dewatered by  breach-
ing the embankment. Sediment control will be maintained
downstream from the pond by using soil erosion control
fencing. Since the pond is spring fed, spray irrigation may
be utilized at the discretion of the contractor.
  The earthwork  will involve uniform regrading of  the
lagoon area maintaining slopes less than 1 to 2. The fresh
water pond embankment will also be removed.
  Following regrading, composted sludge will be applied
to the entire site at the rate of 125-130 tons/acre (approxi-
mately 2 in.) and tilled into the top 4 to 6 in. of soil. The
site will then be hydroseeded and the  grass mowed  as re-
quired during the growing season.

-------
214    REMEDIAL RESPONSE
CONCLUSIONS

  When this project was begun five years ago, the State
had in mind several objectives. The first was to demon-
strate the feasibility of reclaiming a  hazardous waste
site. At the time this renovation began,  RCRA and other
laws governing safe disposal of hazardous wastes were
just being discussed; the  techniques applied here were
innovative.

  The second  objective  of the cleanup  was the develop-
ment of low cost, unsophisticated solutions to waste prob-
lems to encourage cooperation and participation  by  in-
dustry  in  cleaning up their  own problems.  Too  often,
over-zealous enforcement  combined with  unclear and
overly restrictive technical  support have caused activities
to occur  in the court room  with technically ill-prepared
judges making decisions  instead of renovation of a  site by
experts.


  Finally, from a management standpoint, the State de-
sired to treat or destroy  hazardous wastes at,  or as close
to, the point  of generation  as possible. This technique
eliminates  possible transportation  accidents and  the
prospect  of  creating two hazardous waste problems  in-
stead of one.
REFERENCES

1. Steimle, et al., WRA Field Investigation Report: "A
   Subsurface  Investigation  of  Creosote  and  Penta-
   chlorophenol Concentrations in the Soils, Hollywood
   Md., 1978."
2. Ramsey,  W.L.,  In-House Memorandum. Water Re-
   sources Administration, Maryland Department of Na-
   tural Resources, 1979.
3. Poincelot, R.P., "The Biochemistry of Composting,"
   in Proc. 1977, National Conference on Composting of
   Municipal Residues and Sludges.
4. Dust, J.V. and  Thompson, W.S.,  1973,  "Pollution
   Control in the Wood  Preserving Industry," Part 4,
   Biological Methods of Treating Wastewater, Forest
   Products J. 23(9):59-66.
5. Kuwatsuka, S., "Degradation of Several Herbicides in
   Soils Under  Different  Conditions,"  Environmental
   Toxicology of Pesticides, F. Matsumura, G.M. Bouch
   and T. Masato, Eds. Academic Press, New York, N.Y.,
   1972, 385-400.
6. Hornick, S.B., et al., "Use of Sewage Sludge Compost
   For Soil Improvement  and Plant Growth," U.S. De-
   partment of Agriculture Science and Education Ad-
   ministration,   Agricultural  Reviews  and  Manuals.
   ARM-NE-6, August, 1979.

-------
                            THE FORT MILLER SITE:
       REMEDIAL PROGRAM FOR SECUREMENT OF AN
         INACTIVE DISPOSAL SITE CONTAINING PCB'S
                                 WARREN V. BLASLAND, JR.
                                     WILLIAM H. BOUCK
                                     EDWARD R. LYNCH
                                    ROBERT K. GOLDMAN
                                  O'Brien & Gere Engineers, Inc.
                                       Syracuse, New York
INTRODUCTION

  In September 1980, the General Electric Company and
the New York State Department of Environmental Con-
servation (NYSDEC) signed the first major agreement in
the nation in which a  corporation agreed to pay for re-
medial action to  clean up abandoned hazardous waste
dump sites where, in the past, it had disposed of its in-
dustrial waste. The agreement involves remedial actions
at seven land disposal  sites in Saratoga, Rensselaer, and
Washington Counties. General Electric has agreed to carry
out all engineering studies and the necessary remedial ac-
tion at four locations and has agreed to pay a percentage
of the cost  of the engineering studies  and  necessary
remedial action at  the three other sites.
  One of the sites included in the agreement is known as
the Fort Miller Site.  It  is located in the Town of Fort Ed-
ward,  Washington  County,   approximately 7.5 miles
south of Hudson Falls,  New York. (Figure  1.)

GE/NYDEC AGREEMENT

Terms of Agreement—Fort Miller Site

  General Electric,  under  the terms of  the agreement,
conducted field investigations at the site to determine the
current condition  of the site, including the hydrogeology,
the areal and vertical extent of waste present, the physical
state of the waste and the means by which wastes have
been released, may be released, have migrated or may mi-
grate from the site.
  At the completion of the field investigations, an engi-
neering report was prepared to include:
•All data from the field investigations
•Identification of alternative remedial programs
•Selection of a recommended remedial program to meet
 established goals. (The goal of the remedial program
 was to abate any significant current and future releases
 of hazardous wastes from the site.)
•Implementation schedule and preliminary plan including
 the identification of all property to which access would
 be required to implement the remedial program
•Program for continuing site maintenance
•Establishment of monitoring plan to evaluate the success
 of the remedial program
  Following approval of the  engineering report  from
NYSDEC, General Electric submitted a final plan for im-
plementation of the recommended  remedial program.
Following approval  of the final plan, General Electric
will complete the construction  elements  of the remedial
program at the site.
  Upon completion  of the  construction  elements of the
remedial program and following certification by NYSDEC
that the construction elements are in accordance with the
approved engineering report and final plan, General Elec-
tric will maintain and monitor the site for a 30 year period.

Current Status

  The Fort Miller Site Engineering Report and final plan
have been approved by NYSDEC. The construction ele-
ments of the recommended  remedial program will be be-
gun in March, 1982 and will be completed during 1982.

FORT MILLER SITE

General
  The Fort Miller Site was  an open municipal/industrial
dump site of 2.5 acres which was reported to have  oper-
                    \  in-

              AREA OF INVESTIGATION
                     Figure 1.
               General Site Location Plan
                                               215

-------
216     REMEDIAL RESPONSE
ated as a burning dump between 1953 and  1965. During
the period in which dumping was taking place at the site,
waste  products  containing  polychlorinated  biphenyls
(PCBs) were deposited  at the site. These  waste products
included  reject  electrical components  which  contained
PCBs.

Background Information

  Prior to developing a field investigation  work plan at
the site, all background  regional and site specific historical
information was reviewed including aerial  photographs of
the site prior to and during the years in which the site was
operated.  The aerial photographs aided in identifying the
manner and mode in which the site was operated. Coun-
ty-wide soil and groundwater surveys, local groundwater
well logs and USGS  topographic mapping  were also re-
viewed.

Field Investigations

  A field investigation  was undertaken to determine the
three-dimensional limits of the  site,  and  to defined  the
ways and  extent by which PCBs were migrating from the
site. The field techniques utilized during the field investi-
gation included:
  (1) Topographic survey
  (2) Magnetometer survey
  (3) Subsurface exploration (test borings and test pit
     trenching)
  (4) Groundwater monitoring
  (5) Surface water monitoring

Topographic Survey

  A topographic survey was conducted at  the site and im-
mediate surrounding  area to determine the location and
elevation  of significant features at the site.  Elevations
ranged from about 215 feet  (USGS  datum) at the  south
end of the site to 160 feet at the toe of the site. Surface
water runoff from the site and adjacent areas flows into
intermittent drainage ways  which ultimately empty into
the Moses Kill, a tributary of the Hudson River.

Magnetometer Survey

  A magnetometer survey was conducted at  the site  to de-
fine the  southern,  western  and northwestern  limits  of
dumping at the site. The steeply sloped northern and eas-
tern limits were determined by visual observation. It was
necessary  to accurately  define the limits to insure that al-
ternative remedial programs considered would encompass
the entire area containing dumped refuse.

Subsurface Exploration

  Subsurface explorations included the drilling of test bor-
ings and the digging of test pit  trenches. Fourteen (#1,
3, 7, 8, 9, 11,  12, 13, 13A,  14, 19,  20, 21, 22) test bor-
ings were  drilled at the site as shown in Figure 2.  Seven
(#1, 3, 9, 13, 19, 20, 22) of the test borings were drilled to
refusal and in two of them (#1,  20), rock core samples
were obtained. In the seven test borings drilled to refusal,
soil samples were  taken at  five-foot intervals. In the re-
maining seven, no samples were  taken and soil descrip-
tion were made from the driller's observation of auger
cuttings as they came from the hole.
  The subsurface  materials encountered at the site were
consistent with those reported in  the county soil survey.
They were divided into three  basic soil types: A medium
stiff to very stiff brown to gray brown clay. Generally this
material, within the limits of the dumping area, extended
to bedrock. Outside the  limits  of the dumping area the
brown to gray clay material was underlain by a medium to
stiff gray varved clay. The brown to gray, medium to stiff
gray varved clay was generally encountered at a depth of
10  to 15 feet below grade.  This material was softer and
moister than the brown to gray brown clay. The bedrock
encountered, underlying the above clay soil, was a gray,
very dense, silty weathered shale. Depth to bedrock under
the site and along the periphery ranged from 23 to 35 feet.
  Test pit trenching operations were conducted to verify
horizontal limits of the site as determined by the magneto-
meter  survey,  to  determined the  vertical limits of the
dumping at the site and to determine the limits of capac-
itor dumping at the site, which if the dumping was isolated
to a smaller area(s), would allow the remedial program to
address a smaller area(s). The location  of the test pit
trenches are shown in Figure 2.
  With the exception  of trench No. 1,  all trenches had
evidence of various quantities of  capacitors or capacitor
paper. Based on these data, it was  decided that the remed-
ial  program would encompass  the entire site.  The hori-
zontal limits of the site defined  by the magnetometer sur-
vey were verified by the test pit  trenching. The vertical
limits of dumping at the site were defined by data obtained
from the trenching operations in conjunction with test bor-
ings installed through  the site to the clay. The estimated
volume of refuse was 20,000 yd3.
  During  the trenching  operations, water was  observed
at the clay/refuse interface in trenches 4, 5, and 6. In each
of these locations  a stainless  steel groundwater observa-
tion well was installed for sampling prior to backfilling.
A cross section through the site is shown in Figure 3.

Groundwater Monitoring

  Groundwater monitoring wells were installed in each of
the fourteen test boring holes. In addition, five (#2,4,5,6,
10) five-foot long  stainless steel wellpoints were installed
at the base  of the dumping area. Three others  (#16, 17,
18) were installed in test pit trenches #4, 5, 6, respectively,
prior  to backfilling.  The  location  of  all groundwater
monitoring wells is shown in Figure 2.
  Several sets of groundwater elevations were taken at the
site. Wells #16, 17, 18, and 21  were shallow monitors (5 to
10 feet below grade) which extended through the dumped
refuse to the clay. The measured  groundwater elevations
in these wells, along with the  visual observations  made
during  the  test pit trenching operations,  indicated the

-------
                                                                                REMEDIAL RESPONSE     217
       V
          LEGEND


      -0.   BORROW UREA TEST BORINGS


      0   TEST BORINGS «/ GROUNOWATER MONITORS


      B   SURFACE WATER SAMPLES        /

      O   WELLPOINT  MONITORS


      A   STREAM SEDIMENT SAMPLES


      O   FLOW MONITOR


     t^^m  TRANCHES

          LIMITS OF REFUSE
                                                    Figure 2.
                                         Fort Miller Site—Existing Site Plan
presence of a local zone of saturation (or perched water
table) at the refuse/clay interface. Discharge of this perch-
ed water table does occur along the exposed faces of the
dumping area.
  Well Nos. 19, 20, 22 are deep monitors (23-25 feet below
grade) which  extend  into the underlying clay. Based  on
groundwater elevations  taken from these wells, the perm-
anent water  table generally  occurs at  a depth of from
3-15  feet below  the  clay. A  profile  of the permanent
groundwater table is shown in Figure 4. Based on the pro-
file, it was determined  that  groundwater moves radially
from the site towards the intermittent streams.
  Groundwater samples were collected from 21 of the 22
wells (#21 was dry). A  strict sampling protocol  was im-
plemented to  eliminate  inadvertent introduction  into the
well (or sample)  of substances which  lead  to inaccurate
results during subsequent laboratory analysis. A quality
control procedure was also implemented to insure accuracy
of the field and laboratory procedures and analyses. The
collected samples were analyzed  for  pH,  conductivity,
PCB, Total Organic  Carbon (TOC) and selected heavy
metals (lead,  chromium, nickel,  copper, selenium and
zinc). For verification purposes, several of the wells  were
resampled and reanalyzed. All groundwater samples  were
filtered prior to analysis.
  Of primary interest to this  field investigative effort was
the concentration of PCB migrating from the site. A gen-
eral summary of the data is as follows:
•No PCB was detected in the background well (#1).
•The concentration of PCB in the wells installed around
 the periphery of the fill area ranged from less than 0.01 to
 3.7 ug/1 averaging0.7 jig/1.
•The concentration of PCB in the wells (#16, 17, 18) in-
 stalled in the perched groundwater table ranged from
 1.23 to 46 ug/1 averaging  19.74 ug/1.

Surface Water Monitoring

  Surface water  runoff  from  the  Fort  Miller Site is
collected in a system of drainage swales which converge,
at the northwest corner of the site, into an unnamed in-
termittent stream,  tributary to the Moses Kill. A V-notch
weir was installed across this stream at the location shown
on Figure 2.
  Flow measurements were taken at the weir location for a
two month period. Also, rainfall intensities were  moni-
tored over the same period of time.
  Stream samples were collected at peak  and base flow
conditions at the  weir and at background locations  and
analyzed for PCB. The concentration of PCB at the weir
location  (#6) (Figure 2) during the peak  and base flow
conditions were 0.30 and  0.80 jig/1 respectively. The con-
centration of PCB at the  background locations (#1, 2, 3,
4, 5) (See Figure 2) during the  peak and base flow con-

-------
 218    REMEDIAL RESPONSE
                         EXISTING
                         GR10E
                                                       400

                                                    SECTION A
                                                    Figure 3.
                                         Existing Site Plan—Cross-Section A
ditions ranged from 0.04-0.58 yg/1 and  0.08-0.96 ug/1,
respectively. Surface water samples were not filtered prior
to analysis.

Air Monitoring

  Prior to the initiation of the Fort Miller Site, it was
acknowledged, because  of  previous air monitoring  con-
ducted by NYSDEC, that volatilization and airborne dis-
persion of PCB from the site was a problem which would
have to be addressed by the proposed remedial program.
For that  reason  air sampling and analysis was  not  con-
ducted at the site.
Safety Protocol

  The implementation of the  various field investigative
techniques in an area known to contain PCB wastes re-
quired the use of a field safety protocol. The safety pro-
tocol implemented was:
•During the drilling of test borings,  the installation  of
 groundwater  monitoring wells, the digging of surface
 trenches and the sampling from groundwater wells, a dual
 carbon respirator,  disposable rubber gloves and  boots
 and a disposable acid-resistant suit were worn.
•During other field investigations, the safety equipment
 listed above was optional, unless a distinct chemical odor
 was noted, in which case the dual carbon respirator was
 worn.
  With the exception of the dual carbon respiratot^new
safety equipment was worn each day (when used) and was
buried at  the site at the end of each day's work. Field
personnel  were instructed to replace filters on  a weekly
basis. A hard hat was worn at all times and a Scott Air
Pack and emergency eyewash station were available on-site
for use if needed.

PCB TRANSPORT MODES

  As stated previously, the goal of the remedial program
is to abate any significant current and future releases of
wastes from the site. The sampling and analyses which
were accomplished indicate that the contaminant of pri-
mary  concern is PCB.  Identified  transport mechanisms
by which  PCB may be released  from the site included
groundwater migration, erosion transport and volatiliza-
tion.

Groundwater Migration

  From the field investigations, it has  been determined
that there  are two groundwater zones at the site, a shallow
perched groundwater zone  at  the clay/refuse  interface
and a  permanent groundwater zone at a depth of 3-20
feet below grade within the dumping area.
  The groundwater discharge flow rates from the shallow
perched groundwater zone and the clay permanent ground-
water zone were calculated to be 500 gal/day and 5,000
gal/day, respectively.

-------
                                                                                REMEDIAL RESPONSE    219
  Results of sampling and analysis from monitoring wells
installed within  the dumping area indicate PCB concen-
tration in the leachate ranging from  1.23 to 46 ug/1. At
the maximum flow rate of 500  gal/day  in  the perched
water table, the peak quantity of PCB discharged from the
site in the leachate flow would be in the range of 0.002
to 0.070 Ib/year. Similarly, measured PCB concentrations
in the permanent groundwater zone at the   periphery of
the site range from 0.01 to 0.37 Ug/1,  indicating a peak
quantity of PCB  discharged  from the site  through  the
groundwater system (maximum flow rate of 5,000 gal/
day) would be on the range of 0.0002 to 0.0056 Ib/year.

Erosion Transport

  As discussed previously,  samples of surface water run-
off were  collected from the drainage channels during  a
rainfall event and analyzed for PCB. Although these data
are insufficient to accurately quantify PCB losses, they in-
dicate that the actual transport of wastes through erosion
was on the magnitude of 1.0 Ib/year.

Volatilization

  As discussed previously, volatilization of PCB from the
site was a known phenomena which would have to be
addressed by the recommended remedial program.

ALTERNATIVES

  Based upon a review of the PCB transport modes  and
their significance at the Fort Miller Site, potential secure-
ment alternatives were identified and evaluated. The three
alternatives which were considered in the engineering re-
port were:
•In-place containment of wastes
•Removal of wastes to new on-site secure landfill
•Removal of wastes to off site secure landfill
  Under each of these alternatives there  were a number
of options pertaining to construction materials, surface
drainage, groundwater  control,  leachate collection  and
control of volatilization. The criteria used in evaluating
the alternatives included technical feasibility, ease of im-
plementation, compliance  with regulatory  requirements
and costs.
  Each of the three alternatives considered, if properly
implemented, would satisfy the goal of the Fort Miller  Re-
medial Program. Also, each alternative is compatible with
existing State and Federal regulations and would be like-
ly to receive regulatory agency acceptance.
  Within certain limitations, each alternative is technically
feasible. However, there are certain advantages and  dis-
advantages to each of the  alternatives. The  primary  ad-
vantage of in-place containment was  that existing solid
waste  deposits would not be disturbed. Under the other
two alternatives, the excavation and transfer of the refuse
greatly increases the exposure, and therefore the potential
for release, of PCB to the environment.
  The critical period of exposure would be during  the
excavation phase, when the main PCB deposits now buried
below the surface would be uncovered. An intense rain-
storm during  this period  could result in  a large loss of
PCB to the environment  through erosion. The potential
for volatilization losses during this period would also be
greatly increased during removal to a new on- or off-site
facility. The period of potential exposure includes the
transportation from the site to the secure facility.  Based
on the above and cost estimates, in-place  containment of
wastes was  recommended  for the Fort Miller Remedial
Program.

RECOMMENDED REMEDIAL PROGRAM

  The recommended remedial  program is in-place contain-
ment of wastes,  to include a clay cap with vegetative cov-
er, a relocated surface drainage system, a leachate collec-
tion system and a gas venting system. The  purpose of
these facilities is to isolate the waste materials from the
surrounding environment. The plan also includes pro-
visions for maintenance of the facilities,  and  monitoring
to measure the effectiveness  of the  remedial program.
For  the purpose of describing the remedial program and
monitoring plan, the site will be defined as the vertical
and horizontal area containing dumped refuse.

Site Preparation

  All trees and brush within the area to be capped will be
removed, and stumps will be ground to grade. The  wood
chips will be spread evenly over the site prior to capping.
Tree  stumps within the area to be capped will be chem-
ically treated  to kill  roots and prevent resprouting. To
minimize future vertical displacement of the final cap, the
surface of the existing site will be proof rolled.
  The site  area  will be regraded to reduce the slopes at
the northern and eastern sides. The regrading  will consist
of placing fill at the base of the site to extend to toe of the
slope northward and eastward.
  The material to be used as fill will be on-site clay which
will be excavated from borrow pits located east and north
of the existing drainage swales. The material will be in-
stalled in lifts not to exceed 12 ft at a standard proctor den-
sity of 80-85%.

Final Cover

  Following site preparation, the site will  be covered with
a clay cap, using the same on-site soils from which the
embankments  are to be formed. The cap will consist of a
minimum 3.5 ft thick clay  layer. The  bottom 2 feet  layer
will be installed  in 6" lifts with each lift achieving stand-
ard proctor density of 90%.  The top 1.5 ft  (minimum)
clay layer will be installed in 0.5 ft lifts with each lift
achieving a standard proctor density of 80-85%. The cap
will be covered with a minimum of 0.5 ft of  topsoil and
seeded to promote a grass cover.
  Each winter,  the upper zone of the cap will be sub-
jected to freezing. In frost-susceptible soils, such as clays,
the formation of ice lenses, and the subsequent thawing of
these lenses in the spring, can  leave cracks in the clay ma-

-------
220     REMEDIAL RESPONSE
terial which could raise the average permeability of the soil
mass. The degree  of ice lens formation is related to the
depth of  protective cover,  the availability of the  fresh
water and the size of the pore spaces in the soil.
  At the Fort Miller Site, ice lens formation should not
be a serious  problem  because  the on-site silty clays are
generally  less  frost susceptible than other silty types. In
the Fort  Miller area,  the average depth  of frost pene-
tration is about 2.5 ft. Construction of a minimum 3.5 ft
thick clay cap, covered with 0.5 ft topsoil will leave a com-
pacted clay layer extending 1.5 ft below the average frost
line, minimizing the possibility of cracking the impervious
layer due to annual freeze/thaw cycles.
  In dry summer months, an exposed clay layer would be
subject to shrinkage cracking. The placement of the topsoil
layer along with the vegetative cover will  provide an in-
sulating effect which would prevent evaporation and sub-
sequent shrinkage. If minor cracks do  develop, the rein-
troduction of moisture into the clay will result in  swelling
which will close the cracks.
  The clay cap will be installed on a 5% grade  on top,
and  a 1:6 slope over the  horizontal limits of the site. In
areas outside the limits of the cap, the clay materials will
be installed on a  1:4 slope in fill areas and  a 1:3 in cut
areas.

Surface Drainage

  At the present time, all surface runoff from the site is
collected in a series of drainage channels along the toe of
the site.  The fill placed during regrading operations will
cover the existing channels along the north and east  faces
of the site, and new channels will be constructed at the base
of the proposed slopes.

Groundwater Control System

  The surface drainage channel will be installed to a depth
of a minimum of  ten feet below the toe of the refuse to
control the level of the permanent groundwater table.

Leachate Control System

  After placement of  the final cover, the infiltration of
surface  water into the refuse  layer and the subsequent
leaching of this water through the face of the landfill will
be eliminated. Leachate could continue to discharge  from
the site,  however,  until the local zone of saturation at
the interface of the refuse and underlying clay is drained.
To eliminate this potential uncontrolled discharge, a leach-
ate collection system will be installed around the site.
  The collection system will consist of perforated drain-
age pipe installed along the base of the landfill. The pipe
will be laid in a shallow  trench backfilled with granular
material up to the  elevation of 1-2 ft below the bottom of
the refuse. The  side of the trench downgradient of the
landfill will contain a barrier  of  compacted clay fill to
prevent leachate from migrating beyond the drain.
  The collection system will discharge to a pumping sta-
tion  which will  pump  the  leachate to  an underground
holding tank, the content of  which will  be periodically
pumped to a portable tanker for transport to a permitted
treatment facility. Since the impermeable cap over the site
will eliminate percolation of surface water  into the refuse,
the source of additional leachate will be removed and the
existing leachate should be collected and disposed of with-
in an estimated six months to  one year following cap in-
stallation.

Gas Control System

  In  a  typical municipal landfill,  methane gas is pro-
duced by  the biological degradation of organic materials
under anaerobic conditions. Following the installation of
the cap over the site, the generation of gases from the
refuse could result in pressure-induced stresses which could
damage the cap.
  The length of time the  site has been  closed, along with
the burning of refuse which has been reported, would in-
dicate that most organic materials  have already decom-
posed, and  that  the potential for  future generation of
significant quantities of methane is therefore  quite lim-
ited. However, to ensure  that the cap is not damaged by
gas pressures, the recommended remedial program will in-
clude a passive collection and treatment system.
  The gas collection system will  consist of a network of
perforated pipes  laid in shallow  trenches  excavated into
the surface of the refuse. The  trenches will be backfilled
with gravel and enclosed  at the surface by the final cov-
er material. The gas collection pipes will lead to a common
above-ground vent. The vent will be fitted to an adsor-
bent-type  treatment unit and will remain open to the at-
mosphere.

Site Security

  Site security will include fencing sufficient  to restrict
unauthorized access and warning signs to discourage tres-
passers. The Final Plan showing the recommended alterna-
tives is shown in  Figure 4. A cross-section taken through
the site is shown in Figure 5.

MAINTENANCE PROGRAM

  A major portion of the long-term maintenance effort
will involve the vegetative cover on the completed cap. No
trees, shrubs, brush or  deep  rooting weeds should be
allowed to germinate or establish on the site. If visual ob-
servations made during the performance of other main-
tenance activities indicate that  low growing, deep rooting
weeds have  germinated from the cap,  a herbicidal weed
control program would be initiated. Periodic inspection of
the site  will also reveal any problems of erosion, disease,
or thinning of grasses which would then be corrected. The
grasses on the site will also be moved  periodically as re-
quired.
  The estimated  50,000 gal of leachate will be collected
and treated as  required. The gas collection facilities will
be operating continuously. The adsorbent-type treatment
unit will be replaced as required. All  mechanical equip-

-------
                                                                                REMEDIAL RESPONSE     221
                                                    Figure 4.
                                            Fort Miller Site—Final Plan
ment (i.e., pumps) will be maintained as suggested by the
manufacturer.

MONITORING PROGRAM

  The purposes of undertaking monitoring activities are
to measure the effectiveness of the remedial program.
  A 30 year monitoring program was developed to deter-
mine failure of the remedial program. Failure of the re-
medial program is deemed to have occurred if any of the
following conditions are observed:
  (1) Any portion of the cap is eroded, allowing wastes to
     be carried away by surface runoff
  (2) Surface water percolates through the cap and into
     the refuse, creating a source of leachate
  (3) Waste  material is volatilized through the  cap  in
     excessive quantities
  (4) The groundwater table rises  above the bottom  of
     the refuse, resulting in water saturation of the refuse
     and the possibility of the formation of leachate
  The monitoring  activity  which will  determine if the
above failure modes have occurred, respectively, is as fol-
lows:
  (1) Cap Erosion—Periodic inspection of the cap
  (2) Surface Water Infiltration—Flow is observed in the
     leachate collection system (after the shallow perched
     groundwater has been drained).
  (3)  Volatilization—PCB concentration rise significantly
      above baseline concentration. The baseline concen-
      tration will be established following cap installation.
  (4)  Groundwater  Table  Interrupting  Site—Ground-
      water levels  rise  to intercept the site. The ground-
      water table will  be monitored continuously in the
      vicinity of the toe of the site.

By undertaking the monitoring activities outlined above, it
will be possible to measure the effectiveness of the  re-
medial program and to detect any failure of the system
to abate the release of material from the site.
  Failure of the  remedial program  will not  be deter-
mined by analyzing groundwater or surface water samples
for PCB. The very low PCB  concentration levels in the
groundwater, presently below the refuse layer, may differ
at different sampling locations or differ at the same samp-
ling location at different times for many years.
  A variation in PCB concentration at any downgrad-
ient groundwater  monitoring or surface  water location
could simply be the result of  a  higher  or  lower zone
of PCB  concentration moving slowly past the sampling
point. Because  these PCB concentrations  may  vary con-
siderably and since such variation would  not necessarily
be the result of a failure of the remedial program, it
was not  attempted to define  failure of the remedial pro-
gram by attempting to interpret such results.

-------
222     REMEDIAL RESPONSE
                                                                                              s
                                                                                               /?!"
                                                                                                   EXISTING GB40E
                                                                                                  LEGEND
                                                                                                   * \ *   APPROX.
                                                                                                   \»   REFUSE
                                                                                                       LIMITS
                                                        100

                                                     SECTION A
                                                    Figures.
                                            Final Plan—Cross-Section A
CONCLUSIONS

  The work accomplished under this program and the re-
sults can be briefly summarized as follows:
  (1) Field investigations were  undertaken to determine
     the condition of the Fort Miller site and the physical
     extent of the wastes.  It was determined that there
     are about 20,000 yd3 of refuse  which have been
     dumped in an area of approximately 2.5 acres.
  (2) Sampling and analyses were undertaken to define
     groundwater flow patterns  and  rates, and to esti-
     mate the  magnitude of migration of wastes (PCB)
     from the  immediate dumping area via groundwater
     flow. It has been shown that the loss  of PCB by this
     mechanism is insignificant.
  (3) The primary means by which PCB could be trans-
     ported  from the  dumping  area  are volatilization
     and erosion.
  (4) Three alternative  remedial  programs for the  Fort
     Miller site  were evaluated:  in-place containment,
     removal to a new  on-site landburial  facility and re-
     moval to  an off-site landburial facility. The criteria
     used to evaluate these alternatives include compat-
     ibility with program goals, technical feasibility, ease
      of implementation, compliance with regulatory re-
      quirements and  costs.  In-place containment has
      been designated as the recommended remedial pro-
      gram for the Fort Miller Site.
  (5)  A recommended Fort Miller remedial program has
      been presented. The program is based  on in-place
      containment of the wastes, including a clay cap with
      vegetative  cover, relocated surface drainage system,
      leachate collection system, and gas venting system.
      The plan also  includes provisions for maintenance
      of the facilities and monitoring to measure the pro-
      gram's effectiveness.
  (6)  The preliminary  implementation schedule calls for
      construction activities to be completed during 1982.
REFERENCES

1.  Blasland, W.V., Knowles, G.D., Lynch,  E.R., and
    Goldman, R.K., "Remediation at an Inactive Munic-
    ipal  Disposal  Site Containing PCB's", Proc.  13th
    Mid-Atlantic Industrial  Waste,  June 29, 30,  1981;
    476.

-------
                 ORGANIC LEACHATE EFFECTS  ON THE
                       PERMEABILITY OF CLAY LINERS

                                           D.C. ANDERSON
                                         K.W. BROWN, Ph.D.
                                                J. GREEN
                                         Texas A&M University
                                 Texas Agricultural Experiment Station
                                         College Station, Texas
INTRODUCTION

  Saturated conductivity or permeability is the primary
laboratory measurement made on compacted clay soils to
assess their suitability for use in constructing compacted
clay liners for hazardous waste landfills or surface  im-
poundments. The value obtained for the permeability is
used to judge whether a compacted clay soil liner will pre-
vent the movement of leachates into water bodies below
or adjacent to the disposal facility.
  There is, however, little information available on the
impact of organic fluids, likely to be placed in such con-
finements, on the permeability of the liner. Also, there
has been on simple permeability test method  developed
that could be suitable for use with the range of possible
waste fluids. Furthermore, a study evaluating the permea-
bility of a range of typical clay soils with a spectrum of
potential waste fluids would generate a valuable data base,
useful for hazardous waste  disposal permit applicants,
writers and reviewers. Consequently this study was  un-
dertaken with the following objectives:
•To construct from available information a delineation of
 the physical classes  of fluid-bearing hazardous  waste,
 the leachates they generate, and the predominant fluids
 in these leachates.
•To develop a simple,  inexpensive and rapid method for
 the comparative permeability testing of compacted clay
 soils, that would be suitable for use with a wide range
 of possible waste fluids.
•To evaluate the permeability of a range of typical clay
 soils to a spectrum of potential waste fluids.


PHYSICAL CLASSES OF HAZARDOUS WASTE

  Land disposed hazardous wastes generally fall into the
following four physical classes: aqueous-inorganic,  aque-
ous-organic, organic, and sludges(1) (Table I).
  Aqueous-inorganic is the class of wastes in which water
is the solvent (dominent fluid) and the solutes are mostly
inorganic. Examples of these solutes are inorganic salts,
metals dissolved in inorganic acids  and basic materials
such as caustic soda. Examples of wastes in this category
are brines, electroplating wastes, metal etching wastes and
caustic rinse solutions.
  Aqueous-organic is the class of wastes in which water is
the solvent and the solutes are predominantly organic.
These solutes  are  organic chemicals that are polar or
charged, as is inferred by their water solubilities. Examples
of this class  of waste are wood preserving wastes, water-
based dye wastes, pesticide container  rinse  water  and
ethylene glycol production wastes.
  Organic is that class of wastes in which an organic fluid
is the solvent or dominant fluid and the solutes are other
organic  chemicals  dissolved  in the organic solvent.  Ex-
amples of this class of wastes are oil-based paint wastes,
pesticide manufacturing wastes, spent motor oil, spent
cleaning solvents  and solvent  refining and reprocessing
wastes.
  Sludges represent the  fourth class of wastes. They are
generated  when a waste stream is dewatered, filtered or
treated for  solvent recovery. Sludges are characterized
by high solids content such as that found  in settled matter
or filter cakes and consists largely  of clay minerals,  silt,
precipitates,  fine solids and high molecular weight hydro-
carbons. Examples of this waste are API separator sludge,
storage tank bottoms, treatment plant sludge or any fil-
terable solid from any  production  or pollution control
process.
  Both  economic  and pollution control pressures con-
tinue to mandate solvent recovery and reductions in dis-
charges  of aqueous waste streams. These factors  have
made, and will continue  to make sludges the fastest grow-
ing class of wastes. After placement of sludges in a waste
disposal facility, leachates migrate out of the sludge due
to  gravitational  forces, overburden pressures and hy-
draulic gradients. These leachates  are similar in physical
form to the first three classes of waste shown in Table I.

LEACHATE GENERATED BY HAZARDOUS WASTE

  To determine the effect of a specific waste on the per-
meability of a specific clay  liner,  two unique leachates

                       Table I.
            Physical Classes of Hazardous Wastes
Waste Class
Aqueous-inorganic
Aqueous-organic
Organic
Sludges
Solvent Phase
Water
Water
Organic Fluid
Organic Fluid or
Water
Solute Phase
Inorganic
Organic
Organic
Organic and
Inorganic
                                                    223

-------
 224    REMEDIAL RESPONSE



HYDRAULIC AND
BEARING PRESSURE



INFILTRATION OF
OUTSIDE WATER


FLUID PORTION OF
THE WASTE



HAZARDOUS
WASTE




WATER SOLUBLE
PORTION
OF THE WASTE


PRIMARY LEACHATE
CLAY
LINER



SECONDARY
LEACHATE

i

UNDERLYING '
STRATA
                        Figure 1.
 Sources of Leachate that May Come in Contact with Clay Liners
                        LEACHATE
 [   SOLVENT or FLUID PHASE    |   |  SOLUTE or DISSOLVED PHASE |
   [ORGANIC FLUIDS
ORGANIC
[INORGANIC]
                         Figure 2.
            Composition of the Leachate of a Waste
DISSOLVED
PHASE
PREDOMINANTLY
ORGANIC

PREDOMNANTY
AQUEOUS

INORGANIC
COMPONENTS

ORGANIC
COMPONENTS
                        Figures.
         Primary Leachate Generated at Disposal Sites
FLUID
PHASE
DISSOLVED
PHASE
                        Figure 4.
        Secondary Leachate Generated at Disposal Sites

must  be investigated. These  leachates are the flowable
constituents of the waste and the flowables generated from
percolating water leaching through the waste (Figure 1).
  Flowable  constituents of a waste, hereafter referred to
as the primary leachate, includes both  fluids in the waste,
the solvent and  all components dissolved in these fluids,
the solutes. Depending on the physical class of a waste, its
primary leachate  may  be  aqueous-organic, aqueous-
inorganic or organic. Leachate generated from water per-
colating through a disposal facility is composed of water,
the solvent and all  the components dissolved  or carried
with it  in the water, the solutes. This flowable mixture,
hereafter referred to as the secondary leachate, may be
aqueous-organic or  aqueous-inorganic, depending on the
composition of a waste.
  The predominant fluid or solvent phase of a leachate
may be water or any organic fluid (Figure 2). Corre-
sponding solutes in a leachate are any chemicals that dis-
solve in the solvent phase. As in the physical classes of
wastes,  both  primary  and secondary leachates are di-
vided into a solvent phase (the predominant fluid compon-
ent) and a solute phase  (components  dissolved in the
solvent). (Figures 3 and 4). While the solutes in the leach-
ate may affect the permeability of a clay liner, the solvent
phase will  usually exert a dominating influence on per-
meability.
   Essentially  all available literature describing leachates
generated  by hazardous  waste  disposal  considers only
water as the predominant fluid.(2>3) Water is viewed as the
carrier fluid and the organic chemicals are considered to be
present in only trace quantities. While this is probably the
case at the interface of a secondary leachate  and a water
table, the fluid phase at the interface of a primary leachate
and a clay liner will depend  on  the physical  class of the
waste being disposed.  An organic waste  or sludge with an
organic fluid  phase will most probably expose the clay
liner to the concentrated organic fluids contained in the
waste.
                                                                                       FLUID PHASE
                                                                                                   Pin
                                                    Figures.
                                       Solvent Phase of the Leachate of a Waste

                             FLUIDS IN HAZARDOUS WASTE LEACHATE

                               As  stated  previously, organic  fluids  that have been
                             placed in  waste  impoundments cover the spectrum of
                             chemical species. For the purpose of experimentally evalu-
                             ating the effects organic fluids may have on the permea-
                             bility of clay soils, the fluids have been classified into four
                             groups. These groups are based on the physical and chem-
                             ical properties that will govern their interactions with clay
                             minerals. These properties include acidity, basicity and
                             polarity (Figure 5). All clay liners are initially wetted with
                             water.  Consequently, the  interaction of organic  fluids
                             with clay minerals must be viewed in the context of fully
                             hydrated clay minerals.

-------
                                                                                  REMEDIAL RESPONSE    225
                                                     Table II.
                                     Properties of the Four Organic Fluids and Water

Fluids
Acetic Acid
Aniline
Xylene
Methanol
Water
Temp.
Fluid
Melting
Point
17
-6
-47
-98
0
Range of
State (°C)
Boiling
Point
113
184
139
65
100
Density
at 20°C
g/cm3

1.05
1.02
0.87
0.79
0.98
Viscosity
at 20°C
Centipoise

1.28
4.40
0.81
0.54
1.0
Water Solubility
at 20°C
g/1

00
34.0
0.2
0=
»
Dipole
Moment
(debyes)

1.04
1.55
0.40
1.66
1.83
Dielectric
Cons t an t
at 20°C

6.2
6.9
2.4
31.2
80.4
  Regardless of the type of organic fluid,  if it has a low
viscosity, it will be leachable. Fluids with low viscosity are
"by their very nature" leachable and able to extract or-
ganic components from otherwise dry waste.(4) However,
several other fluid properties effect the resultant soil per-
meability values. Some of these properties are  discussed
with respect to the  four groups  of  organic fluids and
water in the following sections.
Organic Fluids Studied
  Organic fluid representatives of the four classes of or-
ganic fluids (Figure 5) were selected for use in the compar-
ative permeability studies.  The four classes studied were
acidic (acetic acid), basic (aniline), neutral polar (methan-
ol) and neutral nonpolar (xylene) organic fluids. The four
organic fluids and water along with their relevant physi-
cal and chemical properties are shown in Table II.
  Water containing CaSo4 at a concentration of 0.01N was
used to establish  the baseline permeability  of  each soil
core and as  the control  fluid. The calcium salt was se-
lected for use due to its stabilizing effect on permeabil-
ity. A concentration  of 0.01N was selected because it ap-
proximates the salt concentration typically found in soils.
Clay Soils Studied
  Four clay  soils  typical of those used  for the lining  of
hazardous waste landfills and surface impoundments were
selected for these  studies. These clay soils  are referred  to
by the following  names:  noncalcareous  smectite,  cal-
careous smectite, mixed cation kaolinite and mixed cation
illite. The soils are described in Table III.

MATERIALS AND METHODS

  To assess the suitability of compacted  clay soils for the
lining of waste disposal facilities, the primary laboratory
measurement made is saturated conductivity or permea-
bility.  These tests  often take months to complete for two
reasons. Compacted clay  soils often have  permeability
values lower than 1(T8 cm/sec, and it is often necessary to
pass a pore volume of water through a soil before a stable
baseline  permeability  value can be  obtained.  After the
permeability baseline  is  established, the passage  of at
least a pore volume of the organic test fluid may be neces-
sary to fully determine the effects an organic  fluid may
have on the permeability of the compacted clay soil. Con-
sequently a pressurized air source has been used to increase
the hydraulic gradient and reduce the time needed for
testing.15'6'
  In this study, the authors utilized a  pressurized,  air-
induced, elevated, hydraulic gradient in a rapid compara-
tive permeability method for use on compacted clay soils.
A hydraulic gradient of 361.6 (equivalent to a hydraulic
head of 42.2 m of water) was used  for the two smectite
clay soils.  A hydraulic gradient of 61.1  (equivalent to a
hydraulic head of 7 m of water) was used  for  the illitic
and kaolinitic clay soils.

                        Table III.
              Properties of the Four Clay Soils
Clay SoH
Description
% Sand
( 50nm)
% Silt
( 50-2.0nm)
% Clay
( 2.0nm)
Predominant
clay
minerals*
Shrink-Swell
Potential
Corrosivity
(Steel)
Cation Exch.
Capacity
(meg/lOOg)
Total
Alkalinity
(meg/lOOg)
Noncalcareous
Smectite
35-37
26-28
36-38
S
M
K
Very high
High
24.2
3.3
Calcareous
Smectite
7-8
42-44
48-50
S
K
Very high
High
36.8
129.3
Mxd Cation
Kaolinite
39-41
17-18
42
K
M
Mod.
High
8.6
0.8
Mxd Cation
Illite
14-15
38-39
47
I
S
Mod.
N.D.
18.3
4.2
•In order of descending quantity in the soil.
N.D. Not determined
S—Smectite, M—Mica, K-Kaolinite, I—Illite

-------
226     REMEDIAL RESPONSE
  The equation used for calculating the  permeability of
the compacted clay soils was as follows:
K =
         ATH
(1)
where:
K = permeability constant (cm/sec)
V = volume of liquid (cm3)
A = cross-sectional area of liquid flow (cm2)
T = time (sec.)
H = hydraulic gradient
   Compacted soil cores used to evaluate the permeability
of the soils to organic fluids were prepared at or above
the optimum water content. After the clay soils to be test-
ed were compacted into  the  soil  chambers,  they were
mounted in  permeameter  base plates and  fitted with a
fluid chamber and permeameter top plates. Each top plate
was fitted  with a pressure  inlet to connect it to the pres-
surized air source via the  pressure distribution manifold
(Figure 6).
   A moisture trap, pressure regulator and pressure gauge
was placed between the  pressurized air source  and mani-
fold.  The  moisture trap was positioned between the  air
source and regulator to  prevent the buildup of debris  on
the membrane in the regulator. The pressure gauge went
between the regulator and  manifold so that the hydraulic
pressure being applied  to the permeameters could  be
monitored.
   Cutoff valves for the pressure leading to  each permea-
meter were placed between the manifold and permeameter
top plates. These valves allowed the placement or removal
of individual  permeameters  without  depressurizing the
entire test system.
   To avoid channel formation, the compacted clay soils
were seated at low  pressure. By letting 10 cm of the stan-
dard leachate stand on the soils for 48 hours, an effective
seat was obtained for the top few millimeters of the soil.
The thin layer prevented  bulk flow,  thereby  permitting
the rest of the soil to adequately  seal the permeameter
sidewalls at elevated pressures.
   All gaskets used in the permeameters were teflon to pre-
vent their  deterioration and possible  blow out from con-
tact with  the  various organic  fluids. To avoid  leakage
around the hard teflon gaskets, all metal surfaces against
which the  gaskets seated were wiped clean of grit. In gen-
eral,  all permeameter components were found  to with-
stand continuous operational use at  pressures equivalent
to 50m of water.
   Directly under the  soil  chambers  and in the permea-
meter  base plates were  porous stones to permit  seepage
of the leachate to  the outlet in the base plate.  To limit
the extent of leachate mixing after passage through  the
compacted soil, the outlet was fitted  with an  adapter to
0.3 cm inside  diameter  teflon tubing. The use of trans-
lucent teflon  at the outlet on the base  plate provided a
convenient window with which to monitor the expulsion
of trapped air.  Standard leachate  (0.01N CaSo4) was
passed  through all soil chambers  until there were no air
bubbles visible in the outlet tubing. Where soil piping oc-
curred in soil samples, eluded soil clay particles were  vis-
ible both clinging to the inside walls of the outlet tubing
and as a suspension in the collected flow samples.
  Teflon tubing carried the leachate to an automatic frac-
tion collector which collected leachate samples simultane-
ously from ten permeameters at specific time increments.
Since there  was a potential  for volatile losses occurring
during leachate delivery from  the tubing to the sample
bottles in the  fraction collector,  the top  of each sample
bottle was fitted with a long  stem funnel and the fraction
collector was placed in an air tight cooled chamber. Addi-
tionally, the entire test apparatus was fitted into a vented
hood (Figure 6). This extra precaution was taken as insur-
ance against worker injury in the event of accidental spills
or system leaks.
                                                    Figure 6.
                                     Schematic of the Permeability Test Apparatus
                              After the cores were seated in the permeameters at low
                            pressure, the selected air pressure was applied to the fluid
                            chamber until  stable permeability values  were obtained
                            with the standard leachate. At this point, the pressure was
                            released and the permeameters were disassembled to per-
                            mit examination of the core for signs of swelling or de-
                            terioration of any kind.  If the clay had expanded out of
                            its mold, the excess was removed with a straight edge
                            while trying not to smear the surface of the soil. All ma-
                            terial that had  expanded out of the cores was oven dried
                            and  weighed to estimate the  percent swelling that had
                            taken place.
                              With the three soil types that had swollen,  additional
                            standard leachate was passed to ensure that the permea-
                            bility was not  affected by the excess soil removal. This
                            extra step was  not  necessary with the kaolinitic soil since
                            it  experienced  no  swell  after passage of the standard
                            leachate.
                              Next,  the remaining  standard leachate  was removed
                            from the fluid chambers and replaced with the organic
                            fluids. After passage of the simulated primary leachates
                            (organic fluids),  the permeameters  were depressurized,
                            disassembled and the cores dissected to determine if struc-
                            tural changes had taken place in the compacted clay soils.

-------
                                                                               REMEDIAL RESPONSE    227
8  :
  10'
                                                           IOO
               NONCALCAREOUS   SMECTITE   a
               CALCAREOUS  SMECTITE      A
               MIXED  CATION  KAOLINITE    o
               MIXED  CATION ILLITE       •
                               WATER (0.01 N CaSO4)
   0.5    ttO     0.5     I.O     I.5     2.0   2.5    3,0
                   PORE  VOLUMES
                      Figure 7.
  Permeability of the Four Clay Soils to Water (0.01N CaSo4)
                                         II  :
                                                                               NONCALCAREOUS  SMECTITE   a
                                                                               CALCAREOUS  SMECTITE      A
                                                                               MIXED CATION  KAOLINITE    O
                                                                               MIXED CATION ILLITE       •
                                                              05
                                                                    0.0
                                                           0.5     I.O     l!5
                                                             PORE  VOLUMES
                                                                                               2.0    2.5     3.0
                       NONCALCAREOUS   SMECTITE
                       CALCAREOUS  SMECTITE      A
                       MIXED  CATION  KAOLINITE    o
                       MIXED  CATION ILLITE       •
 10'
   0.5    00
OS     10     1.5    2.0    2.5     3.0
  PORE  VOLUMES
                       Figure 9.
      Permeability and Breakthrough Curves of the Four
                  Clay Soils with Aniline

  Permeability values for the clay soils have been plotted
against the cumulative pore volumes of the test fluids that
passed through the  compacted  cores.  The volume of
"fluid out" is divided by the volume of the pore space in
a given core to  obtain the  fraction of a pore  volume
passed at each permeability value. Recorded data values
are shown graphically in Figures 7 through 11.
  In several of the figures representing primary leachate
treated cores, the increase in organic fluid as a percentage
of the incremental  leachate  volume ("fluid out")  has
been depicted across the top of the  permeability graphs.
For  these  breakthrough  curves, the  values  for  "fluid
out" were determined by  one of  two methods depending
on the fluid to be analyzed. Fluids that were insoluble
(xylene) and minimally soluble (aniline) were determined
simply by recording the volume of the organic and aque-
ous  layers  in  the  sample collection  bottles.  The  only
miscible  fluid for which determinations were made  was
methanol.  The percentage of methanol in water  was de-
termined using a thermoconductivity gas chromatograph.

RESULTS AND DISCUSSION
                      Figure 8.
  Permeability of the Four Clay Soils Treated with Acetic Acid
                                           Permeabilities of the four compacted clay soils to the
                                         standard leachate (0.01N CaSo4) are depicted in Figure 7.

-------
228     REMEDIAL RESPONSE
  100
                         NONCALCAREOUS   SMECTITE   a
                         CALCAREOUS  SMECTITE      A
                         MIXED  CATION  KAOLINITE    o
                         MIXED  CATION I LUTE       •

                                    METHANOL  <»*»
            00
0.5     10     1.5
  PORE  VOLUMES
                                       2.0
2.5
                                                    30
                                                                    NONCALCAREOUS  SMECTITE
                                                                    CALCAREOUS  SMECTITE     A
                                                                    MIXED  CATION  KAOLINITE    0
                                                                    MIXED  CATION  ILLITE       •
                                                                        00
                                  0.5     1.0     1.5
                                    PORE  VOLUMES
                                                                                                    2.0    2.5    3.0)
                       Figure 10.
 Permeability of the Four Clay Soils Treated with Methanol and
    the Breakthrough Curve for the Mixed Cation Illite Soil

The vertical dashed line represents the point at which the
permeameters were depressurized for  the final time prior
to  placement of  the  organic  fluids  in the  non-control
permeameters. There appears to have been little effect of
depressurization on the permeability of the control cores.
   Permeability  of the noncalcareous  smectite and mixed
cation kaolinite soils were essentially  constant during the
passage of  approximately  two pore volumes of the stan-
dard leachate. In contrast, the permeability of the calcar-
eous smectite decreased slowly while  the permeability of
the mixed cation illite increased slowly.
   Traditionally, the  permeability  testing  of prospective
clay liners for hazardous waste landfills and surface im-
poundments,  has  used only standard aqueous leachates
(such as 0.01N CaSo4 or CaCL) as the permeant fluid. All
four of the clay soils used in this study, if only evaluated
by this traditional test, would qualify as adequate for lin-
ing hazardous waste disposal facilities on the basis of their
having permeabilities lower than 1 x 10"7 cm/sec.
   All  four  clay soils permeated with acetic acid showed
initial  decreases  in permeability (Figure 8). However, there
was also a  significant amount of soil piping occurring in
these cores  as was shown by the presence of soil particles
both clinging to the inside walls of the outlet tubing and
                                                                 Figure 11.
                                            Permeability and Breakthrough Curves of the Four Clay Soils
                                                             Treated with Xylene

                                         deposited on the bottom of the leachate collection bottles.
                                         In addition, the leachate from these cores was usually
                                         tinted (red, creamy or black) indicating that soil compon-
                                         ents were  dissolved by the  acid. The initial decrease in
                                         permeability is  likely due to the partial dissolution and
                                         subsequent migration  of soil particles. These migrating
                                         particle fragments could then lodge in the fluid conduct-
                                         ing pores,  thus decreasing the crossectional area available
                                         for fluid flow.
                                            Two of the soils treated with acetic acid  (calcareous
                                         smectite and mixed cation kaolinite) showed continuous
                                         permeability decreases throughout  the test period. Both
                                         the noncalcareous smectite and the mixed cation illite
                                         eventually began to rebound from the initial permeability
                                         decreases,  but not until passage of 39% and 62% of a pore
                                         volume respectively. The permeability increases on both
                                         of these soils were probably due  to the progressive soil
                                         piping that eventually cleared the initially clogged pores.
                                            Permeabilities and breakthrough curves for the four clay
                                         soils treated with aniline are given  in Figure 9.  While all
                                         four clay soils showed significant permeability increases,
                                         the calcareous smectite soil showed the least.
                                            Both the noncalcareous smectite  and mixed cation illite
                                         exhibited early breakthrough of aniline with concurrent

-------
                                                                               REMEDIAL RESPONSE    229
permeability increases. There was some indication that the
rate of permeability increase for these two clay soils was
leveling out just above the 1 x 10~7 cm/sec benchmark.
While the permeability of the mixed cation kaolinite soil
eventually exceeded 1 x 10"7 cm/sec, the breakthrough of
aniline and concurrent  permeability  increases  took sig-
nificantly longer to occur. Only the  calcareous smectite
clay maintained a permeability value below 1 x 10~7 cm/
sec.
  Permeabilities and a breakthrough curve (mixed cation
illite) of four clay soils treated with methanol are given in
Figure  10.  All  four  clay soils   eventually  attained
permeabilities substantially in excess of  1  x 10~7 cm/sec.
The permeability of the noncalcareous smectite increased
the fastest and reached the highest permeability value of
the four methanol treated cores. Examination of the meth-
anol treated soil cores revealed  structural rearrangements
of the soil particles  resulting in the  formation of large
pores and cracks visible in the surface of the soils.
  Methanol breakthrough was  monitored on the mixed
cation illite soil. Figure 10 depicts the step by step con-
current increase in the permeability and percent methanol
in the "fluid out"  for the illitic soil. Since no particle  mi;
gration was detected in the methanol treated cores,  soil
piping was ruled out as a mechanism  for the observed
permeability  increases. There was likewise no leveling of
the rate of permeability increases as might have been ex-
pected if the increases were solely due to the lower viscos-
ity and density of methanol compared to water.
  Permeabilities and breakthrough curves of the four clay
soils treated with xylene are given in Figure 11. All four
xylene treated soils showed rapid permeability increases
followed by leveling at a permeability roughly two orders
of magnitude above  their permeability rates  of water.
These permeability increases are far  greater than could be
explained in  terms of the lower viscosity  and density of
xylene compared to  water, indicating that other factors
such as structural changes in the soil are  responsible  for
the large increases observed.

CONCLUSIONS AND RECOMMENDATIONS

  The time needed for the comparative permeability test-
ing of clay soils can be significantly decreased by the  use
of elevated hydraulic gradients. A  hydraulic gradient of
between 50 and 100 seems optimal for such tests in that the
tests can be completed in 2-3 months, while permeability
increases occur  slowly enough to  obtain several points
along a changing permeability curve.
  Three classes of organic fluids (basic, neutral polar and
neutral nonpolar)  may cause substantial increases in  the
permeability of clay liners. Organic acids may also degrade
the effectiveness of clay liners although not as quickly as
the other three organic fluid classes examined.
  Consequently,  permeability  of a clay liner  may  be
effected by the primary leachate (free fluids) of a waste.
Whenever organic liquid bearing wastes are to  be con-
tained in a landfill or surface impoundment by a clay liner,
the permeability of the liner should be evaluated with a
standard leachate  (such as 0.01 N CaSo4), and also with
the primary leachate of a waste.
REFERENCES

1. U.S. EPA,  "Report to  Congress:  Disposal  of Haz-
   ardous Waste", USEPA #SW. 115. Washington, D.C.
   1974.
2. McDougall,  W.J., Fusco,  R.A., and Obrien, R.P.,
   "Containment and Treatment of the Love Canal Land-
   fill Leachate." Presented at the Annual Water Pollu-
   tion Control Federation Meeting held Oct. 11, 1979.
3. Chian, E.S.K.  and  DeWalle,  F.B.,  "Evaluation of
   Leachate Treatment." Volume I and II. USEPA #EPA
   600/2-77-186 a & b. Cincinnati, Ohio. 1977.
4. Walstenholme, R.M., "Disposal of Solvent Waste." In
   the Second Solvent Symposium held by the University
   of Manchester Institute  of  Science and Technology,
   Manchester, England, 1977, 138.
5. Bennett,  J.P.,  "Permeability  of Soils  at Elevated
   Permanent Pressures."  Master's  Thesis  at Colorado
   State University, Fort Collins, Colorado.  1966.
6. Jones,  C.W.,  "Permeability Tests  with the  Perma-
   nent Water Under Pressure." Earth Laboratory Report
   #EM-559. Division of Engineering Laboratories, Com-
   missioner's Office, Denver, Colorado.  1960.

-------
              MULTIATTRIBUTE DECISION MAKING FOR
       REMEDIAL ACTION AT HAZARDOUS WASTE SITES

                                             TERRY H. ESS
                            Hazardous Materials Control Research Institute
                                        Silver Spring, Maryland
                                         CHIA S. SHIH, Ph.D.
                                   University of Texas at San Antonio
                                           San Antonio, Texas
INTRODUCTION

  Many modern technology related problems, especially
those involving man created risks, are extremely complex
and uncertain. Large volumes of data and multiple con-
flicting objectives lead to the requirement for incorporat-
ing subjective judgments into the decision process.  If the
existence of adversary positions  (i.e.,  a politicized con-
flict) is involved, then something other than just intuition
must be relied on  to find generally acceptable solutions.
The problem area of hazardous waste management is just
such a situation. It would be beneficial  to  find a sys-
tematic technique which would allow decision makers to
adequately  address the  complex  issues involved  and de-
velop viable solutions.
  Multiattribute decision analysis  is  an  apparently ef-
fective tool for this type of problem. It is highly  flexible,
incorporates methods  to handle uncertainty, multiple ob-
jectives, etc. and is a well developed technique. However,
it is not without faults.  The most glaring as pointed out
by Rowe(3) is its inability to properly treat the public's sub-
jective perception  of risk. In addition, conventional de-
cision analysis can  become extremely cumbersome  if the
problem being analyzed is complex and no readily  avail-
able means of tree "pruning" exists.
  An obvious answer to this is to somehow integrate multi-
attribute decision analysis with a quantitative risk assess-
ment technique which addresses public perception. The
basic steps entailed in  such a unified approach are deline-
ated below:
  1.  Construct a decision tree for the potential course of
     remedial actions.
  2.  Complete a detailed risk analysis.
     a.  Determine  the  objective  risk of each  decision
        branch. (Note:  According to  some authorities01
       this would be termed modeled risk.)
     b. Determine the appropriate risk referents.
     c. Use an objective risk vs. risk referent comparison
       to determine if  a decision branch requires  modi-
       fication or should be eliminated from considera-
       tion (i.e., pruned).
  3.  Conduct a sensitivity analysis of the risk comparison
     to determine  which branches are  only "marginally"
     acceptable.
  4.  Complete the multiattribute decision analysis with
     the pruned tree.
  5.  Conduct a sensitivity analysis of the "solution.''
This  paper will explore this integrated technique in more
detail.

DECISION TREE CONSTRUCTION

  The  organization  of  decision  and consequence  se-
quences is the first task  in the decision making process.
Generally the following  steps should be utilized to con-
struct a tree:
  1.  Generate an objective hierarchy which terminates in
     the desired attributes and attribute  measurements
     (Figure 1 diagrams  a possible hierarchy for hazard-
     ous waste problems).
  2.  Determine the viable courses of action available.
  3.  Determine the possible chance events (i.e., hazards,
     outcome, etc.) resulting from a decision.
  4.  Arrange the  decision options  and resulting chance
     events in chronological order (Table I provides a
     generalized  structure for  problems  which  largely
     involve risk).
  5.  Evaluate the  specific probabilities for each chance
     event.
  6.  Evaluate the magnitude of each attribute.
  A  number of key areas in  this process require a more
detailed explanation.  First it is necessary to take a closer
look at the meaning  of  "viable courses of action." The
                     Protect biological
                     systems (especially
                     man) in the most ef-
                     ficent acceptable
                     manner

i
1
Acceptable protection Efficent manner


Human
r~
i
i
1
effects Non-human effects Colt
J 	
1$)
1

Lud ti
(yr)
   Risk  Aesthetics  * induced anomalies
                in indicator species
                (animal & plant>
   fatal- acres
   ities  set a-
   etc.   side for
         waste
                       Figure 1.
           Hazardous Waste Objective Hierarchy
                                                   230

-------
                                                                                   REMEDIAL RESPONSE    231
                         Table I.
             Possible Risk Classification Scheme
     Risk Description
       Class of Consequence
Fatalities    Morbidity    Property
                        Damage
I mediate
Catastrophic
Involuntary
Regulated Voluntary
Ordinary
Involuntary
Regulated Voluntary
Delayed
Catastrophic
Involuntary
Regulated Voluntary
Ordianry
Involuntary
Regulated Voluntary


X
X

X
X


X
X

X
X


X
X

X
X


X
X

X
X


X
X

X
X


X
X

X
X
 Decision   Hazard
                  Outcome   Exposure
                                   Conse-
                                   quence
                    Attribute
                    Magnitude
              \^) AJCOf

                       Figure 2.
            Generalized Decision Tree Structure
term implies that some pre-decision tree criteria is used to
eliminate  "non-viable"  alternatives  from  entering into
the decision making process. In the hazardous waste area
the principal criteria to serve this purpose is implementa-
tion  time. The discovery of an  uncontrolled dump  site
for highly toxic substances  would require initial positive
action be implemented in a  minimum of time. The num-
ber of viable options would probably be small. In the case
of planning a  new controlled disposal/storage site this
constraint would be  greatly reduced, therefore, allowing
for the consideration of a much wider scope of options.
  A  keener look is also required in the area of hazard
event space. A quick glance  at  Figure 2 would  cause
some to  think  that  what we  are talking  about is one
"success" event and  a  few  (one  or more)  "failure"
event(s). This is of course far from the truth in anything
but the most simplistic cases. In  most "real-life"  prob-
lems the failure  pathways  shown correspond to  the  top
event of an appropriate fault tree.  A fault tree is  an-
other type  of analytical tree which allows us to  depict the
logical interrelationships between basic events that lead to
a  undesired event (i.e.  the failure  event).  Using fault
trees, Boolean algebra  and  various statistical techniques'6'
it  is possible to determine  failure event probabilities. In
many cases the probabilities determined are only order of
magnitude estimates. The success events in Figure  2 repre-
sent the summation  of all possible  success  events  for a
specific decision. In practical terms the probability of suc-
cess is  equal to  one less the sum of all the  computed
failure  event  probabilities  for a specific  hazard chance
node.
   This  leads us right into  the next area of  decision tree
construction which  requires elaboration,  the significant
potential inaccuracies  in both probability and attribute
assessments. As  indicated  in the prior paragraph even
when "objective" methods are used,  the information  ob-
tained may be only accurate within an order of magnitude.
When intuitive judgments are involved, which is often the
case in these assessments,  then  even more inaccuracies
can be expected.  In fact, the level of accuracy could be so
low that a  single  "correct"  answer cannot  be determined.
This does not negate the value of using quantitative tech-
niques but should caution us to  refrain from making  un-
justifiable claims of accuracy. Because of this, it is almost
as important to  obtain variance  information  about  as-
sessments as it is to  know  their mean.  This variance in-
formation  becomes especially important in the subsequent
step of sensitivity analysis.
                                                                                    Table II.
                                                                         Denney Farm Risk Data Summary
                                                                   Alternative     Joint Prob-   Involuntary   Reg. Vol.
                                                                                   ability    Fatal. Morb.  Fatal. Morb.
                                                                                    (Term)

                                                            1. Leave buried
                                                            2. Install & maintain a
                                                               groundwater monitor-
                                                               ing system
                                    3.  Excavate & store mater-
                                       ial on site
                          .01
                         (long)

                          .9
                         (long}

                         3.3'10'^
                         (long)

                          .45
                         (long)

                          .2
                         (short)

                         3.2>10"5
                         (short)

                           .04
                          (long )

                           .1
                          (short)

                         2.5*10~Z
                          (short)
                                                                      145   1301


                                                                      12    107


                                                                      38    341


                                                                      12    107
                                                                       5


                                                                       7





                                                                       12
45


60
          36
                                    4.  Excavate s transport
                                       liquids and residues
                                       via truck to Syntax
                         same as 3 plus:

                         3. 5*10 "7
                          (short)
                                                                                     (short)

-------
232    REMEDIAL RESPONSE
                        Table ID.
                 Risk Referent Assumptions
  Indirect gain-loss balance:
Class

[public)
Regulated Voluntary
( workers J
Alt
^
2
3/4
3/4
Balance
Marginally unfavorable
Indecisive
Marginally favorable
Favorable
Value
.001
.01
.1
1.0
  Controllability:
Alt
1
2
3/4
Control
Approach
.1
.3
1.0
Degree of
Control
.1
.3
1.0
State of
Implement.

.5
1.0
Basis of
Effect.

.5
1.0
RISK ANALYSIS

  Risk cannot be meaningfully analyzed in an aggregate/3'
It must be separated into classes which are commensurate
with the factors  leading to subjective perception.  A pos-
sible classification scheme for hazardous waste problems is
provided in Table III. This scheme is modified slightly
from that advocated by Rowe in order to specifically fit
the area of hazardous waste. The principal changes are the
deletion of the following risk classes:  identifiable  risk,
voluntary risk, natural risk and man-triggered risk. Since
generally  acceptable solutions are  our interest,  not the
acceptability to specific groups, the  identifiable classifica-
tion is  not  needed. Only the  involuntary and regulated
voluntary classes are required since  the worker's environ-
ment is regulated. Finally, since we can usually classify all
hazardous  waste  problems as man-originated, there is  no
need for the other origination classes.
  After agreeing on an appropriate  classification scheme,
the objective (estimated) magnitude of risk of each class on
each  decision  tree must  be determined.  In  general this
would be formulated in  the following manner for  each
decision branch when a decision tree  format is used:
  Where  j   a specific path in a decision branch
       Pj   - j oint probability along path j
       Ai     a specific risk class
       a^   - the consequence magnitude of risk Ai  on
             path j (i.e. number of fatalities, etc.)
       E     total population exposed to risk Ai
        T - time in years
       R0     Objective risk magnitude
(Note: This provides a measure of risk commensurate with
the manner of data presentation used by Rowe(3).)
  The next step is to determine appropriate risk referents.
The purpose of a risk referent is to serve as the measure of
risk acceptability (incorporating subjective perspectives).
Rowe  has proposed a methodology  of calculating risk
referents which is composed of two steps:
   1. Using historical data as  a base, absolute risk re-
     ferences are determined for each risk class.
   2. These absolute references are modified to fit the spe-
     cific  situation  being analyzed producing risk  re-
     ferents.
A clearer understanding of the purpose of this process can
be gained by examining Figure 3. In this exhibit the risk
referent  calculation steps are related  to the transforma-
tion factors responsible for the subjective perception of
risk. It should be noted that Rowe's assessment  of the ac-
curacy of this method indicated potential variances in ex-
cess of one order of magnitude. Prospect theory(5) provides
a  formal explanation of the psychological rationale be-
hind the determination of risk referents  and gives some
hope for future refinement of this process.
   The final phase of the risk analysis can now be accom-
plished. This is nothing more complicated than a compari-
son of a decision branch's estimated risk (for each class of
risk) with the appropriate risk referent. If the risk does not
exceed the referent by more than one order of magnitude
then the  risk is considered publicly acceptable. This order
of magnitude comparison is used due to the inherent in-
accuracies  in the risk information.  If this criteria cannot
be met then the choice remains to either modify the de-
cision branch (which will  probably effect some or  all of
the non-risk attributes being considered such as cost) or to
eliminate that branch. It should be noted that one of the
modification steps possible when the  difference between
objective and subjective risk is large is to attempt to edu-
cate the  public  to  the  "actual" risk. In many cases this
may be a difficult, costly and time consuming path but in
some cases it may be the only option other than complete
abandonment of a project.

Factors involving type of consequence:
•Voluntary or involuntary (1)
•Discounting of time (1)
•Controllability (2)
Factors involving nature of consequence:
•Position in hierarchy of consequence (1)
•Ordinary of catastrophic (1)
Other factors:
•Magnitude of probability of occurrence
•Propensity for risk taking (2)
(1) Explicitly included in determination of absolute risk reference.
(2) Explicitly included in determination of risk referent.
                        FigureS.
      Transformation Factors Utilization in Risk Referents
MULTIATTRIBUTE DECISION ANALYSIS

  This process can be  broken into three principal parts:
1) the determination of utility functions for each attri-
bute, 2) the development of a multiattribute utility func-
tion and 3) the calculation of expected utilities. The first
two are by far the most complicated. The determination of
an attribute's utility function is a well developed analyst-
decision  maker  interrogation procedure. It normally in-
cludes questioning both the qualitative and quantitative

-------
                                                                                    REMEDIAL RESPONSE    233
                        Table IV.
              Risk Referent Calculation Factors
Risk Classification

ic, fatal
Involuntary, ordinary,
fatal
Involuntary, catastroph-
ic, health effect
Involuntary, ordinary,
health effect
Regulated voluntary,
ordinary, fatal
Regulated voluntary,
nrrtinarv. health r*f?»r*t-
Risk Ref
_7

S'lO-''
5-flO""7
3*10-=
ino-4
6*10-'
Risk
Prop
Fac.
1

.1
.1
.1
1.0
1.0
Proportion.
Derating
Factor
ait- i no i
A J.t j, . UU1
2 .01
Alt 1 .001
2 .01
3/4 .1
Alt 1 .001
2 .01
Alt 1 .001
2 .01
3/4 .1
1.0
1.0
Control.
Factor

Alt 1 . 01
Alt 2 .015
Alt 1 .01
2 .015
3/4 1.0
Alt 1 .01
2 .015
Alt 1 .01
2 .015
3/4 1.0
1.0
1.0
  Note: All risks are treated as immediate

characteristics of the decision makers preferences. The re-
sult is a function which transforms attribute magnitudes
into a cardinal measurement, utility, structured so that the
best course of action is  always the  alternative with the
highest expected utility.
  The  development  of  a multiattribute utility function
basically involves the assessment  of attribute  independ-
ence and then the development of an appropriate mathe-
matical formulation. This formulation  allows the analyst
to combine each attribute's utility function into a  single
utility function. A simplified  summary of the  independ-
ence assumptions and resulting formulations in multiat-
tribute utility theory is provided  in Figure 4.  It should
be noted that the concept  of preferential independence
involves attributes under the condition of certainty while
utility independence  is specifically concerned with uncer-
tainty. This process is developed according to the decision
maker's perception of the  attributes'  relationships, not
according to some established standard rules.  Some gen-
eral  observations about  the independence  perceptions
likely to be held by most  decision makers in hazardous
waste problems seems in order though. First, the risk at-
tributes normally will be both preferential and utility in-
dependent of the other attributes. Second, each risk at-
tribute  would normally be both preferential and  utility
independent of each other.
  The  third part of the procedure, the calculation of
expected utilities, is basically just a mechanical process of
"averaging  out" and "folding back."  Averaging out in-
volves nothing more than the computation of PjUj (p-
joint probability along decision path j; Uj-utility magni-
tude for the jth path) for a decision node. Folding back
entails the elimination of the less desirable paths at a de-
cision node. Unlike most of the other aspects of this inte-
grated analytical procedure, this portion of the process
is purely "objective."

SENSITIVITY ANALYSIS

  Sensitivity analysis is used  at two points in  this inte-
grated procedure: 1) after the risk  analysis and 2) after
completing the decision analysis. A good question at this
point is what is this sensitivity analysis about and why is
it necessary? As expressed previously, a significant level of
inaccuracy  is inherent in  the  type of problems we  are
dealing with. Sensitivity analysis is a post solution investi-
gation that indicates how much trust should  be placed in
solutions when we know that all or most of our parameter
values (probabilities,  attribute assessments, etc.) are not
certain.  Therefore, sensitivity analysis has  a very im-
portant part to  play  in  the overall analysis. Using this
technique after risk analysis allows us to ascertain if any of
the "acceptable" decision branches  is really only margin-
ally  so. There would  be a great deal of doubt if one of
these marginally acceptable branches was  the solution or
a part of the solution chosen during decision analysis. In
that case we would probably be  inclined to look at the
problem with even greater intensity. In addition this  tech-
nique will help us identify options whichJiave no realistic
chance of being accepted. The same type  of function is
provided  by the sensitivity analysis conducted after de-
cision analysis.
   Sensitivity analysis  does not have any rigid rules about
what specific methods to use. Within the  context of  the
problems  we  are addressing,  two  possibilities present
themselves.  The  first is simply  to  change the value of
selected  parameters and see if the "solution" changes.
A more  useful approach  involves the substitution of
variables for selected parameters (normally one at a time),
then solving the  decision  problem  in terms  of the  vari-
able. By changing the variable we can graphically depict
the  effect  it has on the desirability of each  possible
decision.  This  allows  us to  quickly  determine which
parameters  are  significant and therefore require  close
attention.
Independence Definitions:
  Preferential Independence (PI)— attribute X is PI of attribute Y if
  preference for consequence (x.y1) with y' held fixed do not depend on
  the amount of y'.
  Utility Independence (UI)— attribute X is UI of attribute Y if prefer-
  ence for lotteries on (x,y') with y1 fixed do not depend on the amount
  ofy'.
  Additive Independence (AI) — attribute X and Y are AI preferences
  for lotteries (x,y) depend only on the marginal probability distribu-
  tions on x and y.

Utility Formulations:
  IfXiUiXi.i  = 1,2..., n then
  u(x) = EkjUj(x) + kjjUjWUj) +...k,  nnu, (x,)...un(xn)
If jx,xi|  PIX
  <     >  ™AU,i=2,3 .....
                          n and X^IX, then either
  1)1+ ku(x) = III
  2)u(x) = £kiui(Xj)
  IfXiAiXi, i = 1,2,... ,n then
  u(x) = Ekiui(xi)
Source: (1)
                         Figure 4.
           Summary of Multiattribute Utility Theory

-------
234     REMEDIAL RESPONSE
APPLICATION

  In order to illustrate the use of the integrated method-
ology, the case study of Denny Farm 1(1) will be used. In
this case an uncontrolled chemical  dump site containing
TCDD  along  with other substances was  discovered in
Missouri. The problem posed is how best to eliminate the
health risk at and around this site. Four viable alternatives
were suggested: (1) leave the site as is, (2) install and main-
tain a groundwater monitoring  system, (3) excavate the
dump and restore the waste in a  controlled manner on
the current site, and  (4) excavate  the site  and transport
liquids  and residues via truck to  Syntex,  an approved
hazardous waste storage site. Since  joint probabilities for
each chance path have been delineated in  the study, the
simplified decision tree shown in Figure 5 is used. Due to
the limited information provided in the report only four
attributes can  be  used: two classes of human risk, fatali-
ties and morbidity, cost and  lead time.  The  division of
human  impacts between fatalities and morbidity is made
possible by adopting the arbitrary assumption that ten
percent of all possible  harmful human  exposures  de-
termined in the study will result in fatalities. In  reality
this assumption would have to be verified  and revised as
necessary.
  The next step is to proceed with a risk analysis. The risk
data provided by the study is summarized in Table II. Use
of this data requires that it be expressed in terms commen-
surate with the risk referents that will be calculated. This
requires that a time duration in years for long term  risk be
established and that estimates of the total population ex-
posed be determined. For the purposes of  this paper the
long-term risk duration is assumed to be evenly distributed
over a span of 30 years.  Estimates of the total popula-
tion exposed were calculated using data provided in the
study. In some but not all cases this could be reasonably
assumed to be equal  to the maximum exposed numbers
derived in the study. The objective risk was then calculated
using the equation provided in the "Risk Analysis" sec-
tion of this paper.  These risk estimates are summarized
in Table V. With objective risk determined, the next step
was to calculate risk referents. This was accomplished us-
ing the procedure outlined by Rowe.(3) In order to do this
some  reasonable assumptions about the  public's and
worker's perception of the indirect  gain-loss balance and
the controllability of each alternative had to be made (see
Table III). A  summarization  of the  factors used in the
risk referent calculation  is shown  in Table IV. A com-
parison of objective risk  and risk referents is provided in
Table V. A quick glance at this comparison indicates that
none of the proposed alternatives is "acceptable" to the
public and that alternatives 3 and 4 are only marginally
"acceptable" to the workers in terms of fatalities.
  Regretfully, no information was  provided in the study
with which to conduct a sensitivity analysis of the above
risk assessment.  Illustrating  this procedure will require
that some reasonable variance information  be  postulated.
It will be assumed that:

•The  parameters  used in  the  decision tree, joint  proba-
 bilities and attribute magnitudes,  can  vary by ±  50%
                       Table V.
                    Risk Comparison
Alt
1

2
/4
/4
od
Risk classification

Involuntary, catastrophic, health

Involuntary, catastrophic, health
Involuntary, ordinary, health
Involuntary, ordinary, fatal
Involuntary, ordinary, health
Reg voluntary, ordinary, health
Involuntary, ordinary, fatal
Involuntary, ordinary, health
Reg. voluntary, ordinary, fatal
Reg. voluntary, ordinary, health
Objective
Risk
3.3.KT"
3.0<1
7.5'10">-5.0»10'1
7.5«10°-5.0.10J'
7.6'10"'-3.3 10"'
4.5xlo"°-4.0'10'"'
3.»-io"-2.e.ifl"
7.9.10""-5.2-10"J
8. 3vlO"J-7.5'10'Z
1.0-10'°-9.0»10"5
9.0.10''-6.0«10"'
Same as 3
Risk Referent
Range
1.0.1flH-l.
s.o.io"'1-?.
s-O'id'^-s.
3.0«-lo"l-3.
5.0'lo"'-5.
3.0»lo"*-3.
1.0,H>'-1.
6.0«10"3- .
Sane as

°'l°"
o 10"
0 10'
o.io"1
OrtO"
ono"4
one"
06
3

     Reg. voluntary, ordinary,
     health
 within the limits imposed by the exposed population size
 and the maximum possible probability of 1.0.
•The risk referents can vary by ± one order of magnitude.
Using  these assumptions, the  information  calculated In
Table VI indicates the possible range of variance in both
objective risk estimates and risk referents. This provides a
sound  basis for assessing which alternatives have a realis-
tic chance of being acceptable.
  At this  point it is fairly clear that alternatives 1 and 2
will under no foreseeable circumstance approach an accep-
table level of risk so they can be eliminated. It appears
potentially feasible and worthwhile to modify alternatives
3 and  4 by eliminating the risk to  the public of signifi-
cant release of residual TCDD after excavation by either
eliminating most of this residual during the clean-up or by

-------
                                                                                   REMEDIAL RESPONSE     235
                            Human  Risk
                          Fatal Morbid.
Cost   Lead Time
(10 $)   (mo)
                                      Human Risk     Cost  Lead Time
                                     Fatal  Morbid.  (10  S)   (io>)
                                    0

                                    3

                                    45

                                    60

                                    36

                                   108

                                    2

                                    9
                                           0

                                           0

                                           0



                                          .75

                                          .75

                                          .75
 3.5

 3.5

 3.5

 3.5

 3.5

 3.5

 3.5

 3.5
0
0
5
7
4
12
0
3
45
60
36
108
3.0
3.0
3.0
3.0
3.0
3.0
8
8
8
8
8
8
10.5

10.5

10.5

10.5

10.5

10.5

10.5

10.5
                 0


                 3


                 45


                 36


                108






                 0


                 3


                 45


                 36


                108


                 2


                 9
3.0


3.0


3.0


3.0


3.0







3.5


3.5


3.5


3.5


3.5


3.5


3.5
10.5


10.5


10.5


10.5


10.5


10.5


10.5
     Figure 6.
Revised Decision Tree
                        FigureS.
                   Initial Decision Tree
some  form  of  encapsulation.  This  modification  will
cause both the cost and lead time of each alternative to
be increased.  No other risk modifications seem  realistic
with the information given. As indicated in Tables V and
VI, even with the suggested modification alternatives 3 and
4 would still not be acceptable. This would suggest that it
is time to "go back to the drawing board" and assemble
some  other   alternatives.  Possible  other  approaches
would need to include means to either mitigate or eliminate
the public exposure due  to tornados and contaminated
workers. Since no in-depth information is provided by the
study for other approaches or modifications, the heroic
assumption that  alternatives 3 and 4 (modified) are the
best that can  be offered  must be  made to proceed with
the analysis.  In  real life this  is a possible outcome; in
which case someone is stuck in the unenviable position of
explaining this to the public.
  With the risk assessment completed a revised decision
tree including only alternatives 3 and 4 (modified) is all
that is necessary (see Figure 6). Completion of the analysis
requires utility functions for each individual attribute and
a multiattribute utility function. The utility functions used
for each of the attributes are diagramed in Figures 7-10.
Figures 7 and 8,  Fatalities and Morbidity, provide an ex-
ample of risk neutral functions while Figure 9, Costs,  is


0)
O
C
01
^ CP

-------
236    REMEDIAL RESPONSE
  tl U
  U «
  o. a
       1.0-
        .6-
        .4-
                   30
                             60

                          Morbidity
                                      90
                                                120
                                                                    .2-
                                                                                  Lead Time (months)
                        Figure 8.
                 Morbidity Utility Function
                         Figure 10.
                 Lead Time Utility Function
  a u
  u n
  a, a
      1.0-
        .8-
        .6-
        .4-
        .2-
                            Ruman Risk   Cost  Lead  Total
                           Fatal Morbid.  (10 S) Time
                                              (no)
          Relative Heights   .38

                             1
.12

  1
                   1.0       2.0       3.0       4.0

                     Cost  ($ Millions)
                        Figure 9.
                   Cost Utility Function
data. In order to combine the four separate utilities into a
single utility value a  multiattribute utility  function  is
necessary. As in the case of the individual utility functions
above, this function is directly related to the perceptions  of
the decision maker in question. For the purposes of this
paper the use of an additive function seems  reasonable
due to the general observations mentioned  previously and
the small relative difference between the two alternatives
being considered.
  Finally expected utility for each  path is calculated us-
ing the mechanics of decision analysis to arrive at a "solu-
tion." The decision tree  with all calculated values is dia-
gramed in Figure 11. Alternative 3 (modified) is obviously
preferable to 4 (modified) in this  particular case.  Since
both alternatives are so similar it is unlikely that a  sensi-
tivity analysis would show this to change  within any
reasonable limits.


CONCLUSIONS:

  The proposed methodology provides a quantitative tool
that is systematic but flexible; is capable of handling un-
certainty, multiple  conflicting objectives and the subjec-
.25   .25  1.0

 .3    .1   .6


 .3    .1   .6


 .3    .1   .39


 .3    .1   .44


 .3     .1   .10
                              .92    .92
                                                 0  .S


                                                 0  .5


                                                 0  .29


                                                 0  .34


                                                 0   0


                                                 0  .SO


                                                 0  .46
                        Figure 11.
                     Problem Solution

live  judgments of  decision  makers;  and addresses the
highly subjective nature of public risk perception.  With
the prudent use of a  "viable option" criteria and risk
analysis, the number of options that  must be fully con-
sidered can  be effectively limited.  On  the  other hand,
this approach will help pinpoint requirements for consid-
ering a wider scope of alternatives  when necessary. The
process is obviously powerful and, therefore, inappropri-
ate for use with single attribute problems.  Given the
potential complexities inherent in hazardous waste prob-

-------
                                                                             REMEDIAL RESPONSE    237
lems the use of this  technique appears justified.  This
method of problem solving does not try to eliminate the
use of  subjective value judgments, an impossible and
counterproductive task, but does require these judgments
to be openly scrutinized.
  Finally, it should be obvious that this methodology is
by no means perfect. It is an initial attempt to bring to-
gether a number of powerful quantitative tools and apply
them to the area of hazardous waste. A great deal of re-
finement and improvement is still required in this area.

REFERENCES

1. Buchanan, J. et al., "Technical Study and  Remedial
  Action for Denney Farm Site  1, Auroria, Mo.  (Final
   Report)." Ecology and Environment, Inc., 1980.
2.  Keeney, R. & Raiffa, H., Decisions with Multiple Ob-
   jectives: Preferences and Value Tradeoffs,  John Wiley
   & Sons, 1976.
3.  Rowe, W., An Anatomy of Risk, John Wiley & Sons,
   1977.
4.  Shih,  C., "Decision Analysis and Utility Theory,"
   Seminar  on  Risk and Safety Assessment,  HMCRI,
   1981.
5.  Tversky,  A. and Kahneman, D., "The Framing of De-
   cisions and the Psychology of Choice," Science, Jan-
   uary 30, 1981.
6.  Vesely, W. et al., Fault Tree Handbook, U.S. Nuclear
   Regulatory Commission, NUREG-0492,1980.

-------
  RISK ASSESSMENT NEAR UNCONTROLLED HAZARDOUS
             WASTE SITES: ROLE OF MONITORING DATA
                                GLENN E. SCHWEITZER, DIRECTOR
                             Environmental Monitoring Systems Laboratory
                                 U.S.  Environmental Protection Agency
                                           Las Vegas, Nevada
PREVIOUS EXPERIENCE WITH RISK ASSESSMENT

  Many assessments of the risks associated with human
exposure to industrial chemicals have been carried out dur-
ing the  last decade. Most studies have been conducted
by Government agencies, particularly OSHA and EPA or
by industry in response to regulatory requirements. Pub-
lic concern over exposures to carcinogens, in particular,
has been a significant  stimulant in encouraging such as-
sessments.
  Risk assessments have considered a number of technical
approaches to  several  types of exposure  scenarios. For
example, clinical and epidemiological  studies to investi-
gate human health  effects as the pivotal concern  in risk
assessments have been  triggered by reports of worker ex-
posures  in certain   manufacturing facilities. Modelling
techniques, using limited data on chemical properties,
have been employed to predict future risks that  might re-
sult  from  the manufacture and use of newly developed
chemicals. Finally,  materials balance  studies have been
important in estimating environmental discharges from
certain production processes and certain uses as a basis for
risk  assessments of industrial  chemicals such  as vinyl
chloride and formaldehyde.  While monitoring data have
often been used in such assessments, only on a few occa-
sions were the  monitoring data collected  in a systematic
manner  designed to support  authoritative risk  assess-
ments.
  The most widely used methodologies for  assessing chem-
ical risks have emphasized a chemical-by-chemical analy-
sis. One approach  has been to estimate  the toxicity  of
the chemical as a function of dose, usually drawing  on
the results of laboratory experiments. A second  approach
relies on direct  observations of the biological effects  of
the chemical exposure,  effects either  on human popula-
tions or more frequently on biological surrogate popula-
tions. In either approach the dose levels or the  effects to
critical receptor populations are estimated,  with the risk
then determined to be the probability that a person with-
in the exposed  population of concern  will suffer  an ad-
verse effect from the likely exposure.
  The uncertainties in  these methodologies as illustrated
in Figure 1 are  well known. They include the  problems
on extrapolating effects from laboratory animals to man,
in determining  the critical receptor populations  and  their
activity  patterns, and  in estimating dose levels.  Never-
theless, these approaches have been used to support many
regulatory decisions.

NEW FACTORS TO BE CONSIDERED
AT UNCONTROLLED SITES

  While past experience provides a point of departure for
developing methodologies appropriate for assessing risks
near hazardous waste sites, and specifically uncontrolled
sites,  different constraints  and  uncertainties must be
addressed. Many types of chemical mixtures are involved,
including mixtures  of  unknown composition. Thus, the
chemical-by-chemical  approach is not  sufficient to esti-
mate risks. Also, risks to future generations may over-
shadow  the nearer-term risks that have often dominated
past  risk assessments.  This longer time horizon requires
greater attention to the chemical loadings  of environ-
mental compartments which serve as temporary pollutant
reservoirs such as ground water, sediments, and soil. Final-
ly, containment uncertainties require the  taking into ac-
count of unanticipated future chemical  leakages from the
site which over time could dominate concerns over risks.
  Risk assessment methodologies must be flexible in order
to accommodate a wide variety of site-specific charac-
teristics. They should be adaptable to varying time limita-
tions for carrying out  assessments. This time span could
range from a few months (which limits sampling to one
time) to  several years (which permits phased sampling and
also  seasonal sampling to take into account changes in
weather  and hydrological conditions) depending on both
the urgency of the problem and the public perceptions of
the urgency. With regard to budget constraints, different
approaches may be required depending  on the number of
samples  that can be  processed.
  The starting point for any risk assessment should be the
purpose of the assessment. The purpose should shape the
application of the methodology and the rigor of the assess-
ment. In general, risk  assessments are intended to clarify
the degree of hazard and thereby help  determine: (1) the
priority  for conducting detailed environmental evaluations
and for  undertaking preventive or remedial action or (2)
the extent and character of the  preventive  or remedial
action. The former  objective usually requires a less exten-
sive  assessment.  Often intertwined with these objectives
are: (1) liability and related issues raised at legal proceed-
ings  and (2) concerns over the immediate hazard to area
                                                    238

-------
                                                                                   REMEDIAL RESPONSE    239
residents. Indeed, risk assessment will often be used by
several parties for somewhat different purposes, and there
will frequently be a blue in distinguishing between assess-
ments to set priorities and assessments to determine the
magnitude of remedial action.

ADAPTING THE
CHEMICAL-BY-CHEMICAL APPROACH

  The assessment of the risks resulting from exposures to
individual chemicals will undoubtedly continue to be an
important consideration in assessing total risks near haz-
ardous waste sites. Two key issues, however, are the selec-
tion of the chemicals of greatest concern and the aggrega-
tion of the risks posed by individual chemicals.
  Several factors are important in selecting the chemicals
for individual risk assessment.  For example, certain chem-
icals may have been  deposited at the site in very large
quantities and therefore they are  of principal concern
simply because of their volume. Secondly, one or more of
the chemicals known to be present at the site may  be so
toxic that the threat even at low volumes is  obviously a
major concern (e.g., dioxin).  Certain "indicator"  chem-
icals commonly found at waste sites may behave in the en-
vironment in a manner characteristic of many other chem-
icals as well and determining the environmental distribu-
tion of such "indicator" chemicals would be indicative of
broader  contamination problems.  Analysis  of leachate
from the edge of the site or preliminary monitoring close
                                 to the site might identify chemicals that are escaping from
                                 the site. Finally, sampling for a group of chemicals which
                                 are usually analyzed as a package such as the 129 priority
                                 pollutants might result in considerable savings per chem-
                                 ical in analytical costs. All of these factors should have a
                                 bearing on the selection of the chemicals for intensive in-
                                 vestigation.
                                    After the chemicals of interest are determined, the classi-
                                 cal approach of estimating source strengths, pathway load-
                                 ings, location of critical receptor populations, and finally
                                 exposure levels can be pursued (Figure 2).  In this  regard,
                                 pathway loadings, including movement of the chemicals
                                 from one media to  another,  are of considerable  impor-
                                 tance in estimating both current exposures and exposures
                                  Identify Critical Pollutants
                                  Determine Source Strengths
                                  Identify Critical Receptors
                                  Determine Receptor Locations
                                  Identify Critical Pathways
                                  Determine Pathway Loadings
                                  Determine Contamination Patterns
                                  Overlay Population Activity Patterns
                                  Estimate Exposure Levels
                                                            Figure 2.
                                          Classical Approach to Estimating Exposure Levels
                                                       RISK
                                                  	t.	
                                                   Extrapolations
                                                  Population at Risk
                                             Sensitivity/Activity Patterns
                                                  Estimate
                                                     of
                                                   Dose  / Estimate
                                                             of
                                                           Effects
           Measurements
           of chemicals in
           micro-environ-
           ment of human
             receptors
 Measurements
  of chemical
accumulation in
biological surro-
 gates for man
 Human body4	^Medical
   burden           exams
measurements
  Bioassays of
 biological sur-
rogates for man
Epidemiological
   studies
                                t.
                                                      Figure 1.
                                      Extrapolations Required in Determining Risk

-------
240
REMEDIAL RESPONSE
Direct Human Exposure:
     Via Dermal Contact
           & Inhalation
                                            Indirect Human Exposure:
                                              Deposition on Crops &
                                            Ingestion-Bioaccumulation
                                                in Grazing Animals
                          Gaseous &
                       Airborne Particles
                                                                    Surface Runoff
                                                                                                 Human Exposure:
                                                                                                Recreation Contact
                                                                                                  & Water Supply
                                                                                         Recharge
                                                                                            to
                                                                                          Stream
Downward Movement
   in Soil Moisture
                                                     Figure3.
                                       Environmental Pathways from a Waste Site
 of future generations. In addition to the natural pathways
 of air, water, soil and groundwater (Figure 3), man-made
 pathways such  as  sewers,  drainage ditches  and dump
 trucks can be important.
   In most cases, monitoring is the only reliable approach
 to pathway  analysis. However, frequently  models (e.g.,
 air pollution dispersion, hydrological  and aquatic fate
 models) and analyses of the  physical and other proper-
 ties of the chemicals can assist in determining the best loca-
 tions and time  intervals for monitoring activities. Also,
 models can  often provide a framework for snythesizing
 and analyzing monitoring data.
   In recent years, the kinetic  analysis model has become
 increasingly  popular for providing a general framework
 for multi-media monitoring programs. This model empha-
 sizes the determination and use of inter-media transfer co-
 efficients as  a basis for estimating how specific chemicals
 partition among different environmental compartments.
   The interpretation of monitoring data as an indicator of
 exposure of subpopulations near waste sites to specific
 chemicals is more of an art than a science. Exposure levels,
 now and in the near term via air, food and drinking water
 and to a lesser degree through recreational  uses of water
 and through soil contact,  are usually of  priority  con-
 cern.  Determining  such  exposure  levels requires innova-
 tive approaches to  matching the multi-media contamina-
                                                   tion  patterns in the  area with the habits and movement
                                                   of resident populations. Even greater  innovation is re-
                                                   quired with regard to future exposures, for the trends in
                                                   both pathway loadings and population patterns must be es-
                                                   timated.
                                                     With the exposure patterns for individual chemicals in
                                                   hand, estimates can  be  made of the dose that will reach
                                                   the population. A variety of coefficients are needed, how-
                                                   ever, to convert chemical concentrations in the micro-en-
                                                   vironment of the receptor into effective dose.  Inhalation,
                                                   ingestion, and absorption  coefficients  in particular are
                                                   needed for each chemical  of interest.  Finally, after de-
                                                   termining the dose and the toxicity properties of the chem-
                                                   icals, estimates of the health effects can be made.
                                                     Methods do not now exist for aggregating the health
                                                   effects of individual  chemicals that reach human popula-
                                                   tions into  valid causal health effects  descriptions. The
                                                   presence of one or more highly toxic chemicals might total-
                                                   ly dominate, although such a case will probably be the ex-
                                                   ception.  If several chemicals of comparable potency reach
                                                   the receptor at comparable  dose levels,  an additive ap-
                                                   proach may be the  only course although  there is little
                                                   scientific basis for judging whether simple addition will
                                                   overestimate or underestimate the effects.
                                                     Approaches to determining the toxicity of complex mis-
                                                   tures have  been a subject of research for several years in

-------
                                                                                 REMEDIAL RESPONSE     241
relation to discharges of industrial wastewater.  Some of
the existing mutagenic  screening tests and aquatic bio-
assay procedures seem appropriate for classifying the rela-
tive hazard of specific waste samples, particularly in rela-
tion to safe handling in the field or in the laboratory. How-
ever, comparable approaches to estimating risks to a gen-
eral  population  that is  being subjected to a large num-
ber of chemicals are not available.


AN APPROACH TO
COMPARATIVE CONTAMINATION LEVELS

   Given the many uncertainties in attempting  to assess
risks near hazardous waste sites on a chemical-by-chem-
ical basis, a complementary approach of comparing rela-
tive  contamination levels  can often be very helpful.  The
approach is intended to give a  degree of perspective to
the chemical contamination that is present in an area of
environmental concern.  The objective is to  provide a basis
for determining  relative risk although little insight is pro-
vided  as  to absolute  risk. Two  approaches employing
different types of "controls" are suggested below.
   A control area with characteristics similar to the area of
environmental concern  is highly desirable. Of special in-
terest would be a control area near  the waste site—but in-
sulated by geographic features from the direct influence of
the site. This area would be impacted by all of the indus-
trial emissions and effluents that permeate the region as
well as other  common sources of contamination such as
    I Heavily Populated Areas
Sampling sites on Grand Island are
not shown on this map.
               • Air     ^ Drinking Water
               OSoil    • Surface Water &
               • Sewer     Stream Sediment
               (7T3 Biota
                                          Scale in Mile)
                        Figure 4.
               Control Areas at Love Canal
agricultural  chemicals. Thus,  in comparing the contam-
ination near the site with the contamination in the con-
trol area, it should  be possible to attribute any  higher
levels found near the site to the influence of the site itself
and not to the background characteristics of the region.
The  selection  of control  areas  during  investigations of
Love Canal is shown in Figure 4.
  A  second type of control can be provided by national
baseline data.  Such data indicating the levels of ambient
or background contamination  usually  encountered in
different types of demographic  settings can help  clarify
the significance of environmental measurements  near a
hazardous  waste site.  In Table I, an initial  attempt is
made to provide such comparative  data. However,  several
problems were encountered in developing these data, in-
cluding:
•Many environmental measurements reported in the litera-
 ture are not based on sound quality assurance programs
 and  therefore may be unreliable. Data  of "research"
 quality, a relatively high quality level are presented in
 the table. Background data of this quality are in short
 supply.
•Many measurements are made in suspected hot spot areas
 of a highly localized character. Care must be taken to pre-
 sent only ambient  data which are  characteristic of a
 general  area.  Ambient data which are characteristic of
 several types of urban areas are given in the table.
•Data are available on only a few chemicals in most media.
 For  some media, such as sewer sediments, soils and in-
 door air, little representative data are readily available.
  Comparing  data from different  sampling  programs in-
volves considerable uncertainty.  For example,  seldom are
the analytical precision and accuracy of the data reported
and  even  less  often  is  the  representativeness  of  the
sampling approach known. Even when comparing two
areas using data from the same data set questions con-
cerning comparability of data can arise.
  Love Canal investigations involved comparisons of data
from four areas:
(1) the canal itself,
(2) the adjacent Declaration Area,
(3) a nearby study area and
(4) the Control Areas.
For soil and sediment samples, the precision and accuracy
of each data point were considered to be about a factor
of three. Thus, if less than an  order of magnitude sep-
arated two numbers, these two measurements might be
considered to be essentially the same. When  differences
exceeded an order of magnitude,  however,  the measure-
ments probably indicated significant differences in pollu-
tant levels.
  When comparing the sets of data collected in different
areas, statistical techniques play a significant role. How-
ever,  uncertainties in  interpretation  techniques  remain.
Should comparisons be made between maximum  values,
mean values or some other numbers? Is the frequency of
occurrence of a chemical the significant aspect for com-
parison or is the frequency of occurrence overshadowed
by major differences in levels? How are trace quantities to
be handled?  Finally, even this  data set, which is much

-------
242     REMEDIAL RESPONSE
                                                       Table I.
                                        Comparative Nation-Wide Monitoring Data
                                       in Commercial, Industrial and Chemical Cities

                                   Volatile Organics  Measured in Air and Water
                                    Air (ug/m3)
                             Comm     Ind
                                                Chem
                         Surface Water (pg/l)
                      Comm.    Ind.    Chem.
        Chloroform

        1.2-Dichloroethane

        1,1.1 -Trichloroethane

        Carbon Tetrachlonde

        Bromodichloromethane

        Trichloroethene

        Dibromochloromethane

        Bromoform

        Tetrachloroethene

        Chlorobenzene
(1)
ND-0 44
0.3
ND-1 07
ND-0 09

T-7.8
.02-1.1
T-1.9
0.98-2.5
 ND-0 11   097-2.96
                                                                    (2)

                                                                   1-43
                               4-7
                               4-8
                                       1/87
                                        1-9
                                        1-8
                                         Drinking Water (
                                       Comm.    Ind.    Chem.
                                                 (2)

                                       0.4-311    4-93   0.6-86
                                         T      T-0.4    T-6
                                       0.9-29   0.8-28    T-16


                                        T-16     T-17     T-5
                                        2-3     T-1
ND-0 066

 ND-0 03
0.02-1.8
ND-0 45
        Research quality data  If only one value shown, only one detectable sample was reported.
        ND = Nol delected, T - Trace
        (I (Range of mean values. (2) Range of single samples
        •6-8 cities per category
                                              1 25 Inch
                                              Diameter
                                                     6
                                                    Feet
                                                    Deep
     Cores for chemicals that can evaporate quickly
    ' (2 separate samples)

    , Cores for other chemicals
     (5 cores which are mixed together into one sample)


                        Figure 5.
                An Approach to Obtaining
                Representative Soil Samples
larger than the data sets usually available, could  be crit-
icized as being too small to allow authoritative  conclu-
sions.
  When comparing levels between sampling  sites, care is
needed  to  insure that the contamination of the samples
taken at each site  are representative  of the  overall con-
tamination at the site. This  problem is  particularly  acute
                        when sampling soil and sediments. One approach to ob-
                        taining  a representative sample at a soil sampling site is
                        shown in Figure 5. This approach is useful if there is a
                        likelihood that the contamination extends to significant
                        soil depths. Obviously, if only surface contamination is of
                        interest, a different approach would be appropriate.
                          Similarly, in comparing  levels  between  sampling areas
                        involving multiple  sites, the sample sets should  be  repre-
                        sentative of the areas as noted above.  Ideally,  the num-
                        ber of samples taken in each area of interest should be
                        about  the  same. However, given  limited  budgets,  fewer
                        samples will undoubtedly be taken in control areas. A sta-
                        tistical approach to determining the minimum number of
                        samples to be taken within  an area is possible if the stand-
                        ard deviation of the individual samples can be estimated
                        in advance.  When the standard deviation cannot be es-
                        timated in advance, a preliminary sampling program to es-
                        tablish an estimate of this  parameter can  be particularly
                        useful.
                          If the  foregoing  concerns are adequately taken into ac-
                        count  in the design of the monitoring  program, realistic
                        comparisons of levels can be very useful as a basis for es-
                        timating relative risks.  This approach  can be of partic-
                        ular value  in  determining the most important environ-
                        mental pathways. Indeed, the uncertainties associates with
                        this approach  may be of less magnitude than the extrap-
                        olation uncertainties usually encountered in the  chemical-
                        by-chemical approach described above.

                        COUPLING MONITORING DATA
                        WITH POPULATION PATTERNS

                          The tendency in designing monitoring systems is to focus
                        sampling in the immediate areas of suspected contamina-
                        tion and in areas  near population concentrations. In a
                        simple geographical situation involving a single site and a

-------
                                                                                      REMEDIAL RESPONSE     243
                      Urban Area
cc
                     Urban Subarea
              Industrial Complex

              Waste Site
             i
             Suspected Hot Spot
1. Grid entire area (ABCD).
2. Determine sources, likely hot spots, and populations in each square.
3. Obtain monitoring data for each square. Emphasize squares with high populations and
  those with likely hot spots. Determine average concentration in each square.
4. Determine squares with highest level of:
  Average Concentration x Persons Hours per day in Square.
5. Subgrid squares identified in 4. (WXYZ).
6. Repeat 2,3,4 for subsquares.

                       Figure 6.
    Linking Monitoring Data and Population Distribution

relatively small adjacent population, this approach seems
adequate. However, frequently waste sites will  be inter-
mingled with other sources of chemical  pollutants. More
sophisticated approaches may be necessary to design the
most cost-effective  monitoring system  that  adequately
characterizes  the area  while also providing insights as to
which sources are the principal contributors to contamina-
tion levels of concern.
                                                              A systematic approach based on a geographical grid is
                                                            depicted in Figure 6. The concept calls for determining
                                                            the sources, the population characteristics and  the con-
                                                            centration levels in each media of interest  in each grid
                                                            square. Squares of particular concern—due to suspected
                                                            hot spots or high  population densities—are further sub-
                                                            divided, and each subdivision is then investigated in more
                                                            detail.
                                                              This  approach, of relating all parameters of interest to
                                                            a fixed grid, has special  appeal when analyzing the effec-
                                                            tiveness of proposed  remedial action. Computer models
                                                            can  help test  the  sensitivity  of exposure  levels in  each
                                                            square to specific source reduction scenarios.
                                                              Certain  population groups may be of greater concern
                                                            than others. Some groups may be more susceptible biolog-
                                                            ically to adverse chemical effects, or their lifestyles may
                                                            increase their exposures  to chemicals. In Figure 1, some
                                                            of the population characteristics that should be taken into
                                                            account in detailed population studies are listed.

                                                                      Types of Exposed Populations Near Uncontrolled Sites
                                                                                            Population Characteristics
                                                                                   J   General Populations   / Special Populatic
                                                               Distance to Sue
                                                               Distance to natural or
                                                               man-made pathways frc
                                                               Consi
                                                               uildmg Characteristic
                                                               Use of con;
                                                               material frc
                                                               Air circulation
                                                                                       Figure 1.
                                                                               Population Characteristics
                                                              INVESTIGATING A WASTE SITE

                                                                Risk  assessment data requirements  should be a prin-
                                                              cipal  factor in developing  strategies for investigating a
                                                              specific hazardous waste site. A combination of the chem-
                                                              ical-by-chemical approach and  a comparative assessment
                                                              of general contamination of the area  appears to be a prac-
                                                              tical method for addressing risks at the types of sites most
                                                              commonly  encountered.  For large  or  complicated  geo-
                                                              graphical areas, the sampling plan should be tied  to a geo-
                                                              graphic grid to facilitate analyses of contributing sources
                                                              and of exposure problems.  Four activities used to imple-
                                                              ment  such  an approach are shown  in Figure  8  and  dis-
                                                              cussed below. They are:
                                                              •An analysis of current and archival aerial photography.
                                                               Photo  interpretation can be useful in identifying and

-------
244    REMEDIAL RESPONSE
                       Aerial Imagery
                                     Geophysical Investigation

                                             Ground
                                             Water
                                            Pollution
                                             Plume
               rj'JTT . O 0s TH	7^-~ry	P "v ^*T—"^ "0*0 v^-cr-
          Zoning Board-
        County Health Dept,
                                                                On-Site Observations
              Company
            Headquarters
Records
                                                  FigureS.
                                  Techniques for Targetting Monitoring Activities
             OLANDFILL/DUMP
                     D K A S T F.
                                                  Figure 9.
                                            Use of Aerial Imagery to
                                          Identify Possible Waste Sites

-------
                                                                               REMEDIAL RESPONSE     245
                      Figure 10.
  Use of Aerial Imagery to Delineate Problems Near Waste Site

 characterizing the site, pinpointing containment breaches
 and environmental pathways away from the site, and
 delineating nearby populations and valuable natural re-
 sources (see Figures 9 and 10).
•Geophysical surveys.  Using a combination of ground
 based remote sensing  devices (e.g., seismometers, mag-
 netometers,  radar,  resistivity instruments),  insights can
 be gained concerning the boundaries of the site, densities
 within the site, and anomalies in soil and water patterns
 around the site. In the future it may be possible to de-
 lineate subsurface  plume  patterns in certain  types  of
 terrain.
•A review  of previously compiled  Government  and in-
 dustry reports. Most major sites have been investigated
 by State or local Government agencies and, in addition,
 industry has some records for many sites. Of particular
 importance are records on the content of the site and sus-
 pected containment problems.
•On-site inspection. Even cursory on-site inspection of the
 site and of the surrounding area can provide insights as
 to possible problems that  are not  easily obtained from
 examination of records and photographs. Aerial photog-
 raphy can be helpful in guiding an on-site inspection.
  Within this framework, a monitoring program can  be
effectively designed to provide data needed for risk assess-
ment. The program  should probably concentrate on the
presence and  movement of a limited number of specific
chemicals and also screen for a larger set of chemicals in
a variety  of  media.  Certain environmental pathways  of
concern (e.g., locally grown food and drinking water con-
tamination) may be obvious; others will probably warrant
exploratory monitoring. In any event, a preliminary mon-
itoring effort should help limit a more intensive moni-
toring effort  so that within a  limited budget a sufficient
number of samples can be taken in the media and in  the
geographical  areas of most concern to establish statistical
credibility.
  Political pressures to carry out a hurried program should
be resisted and promises to deliver early results avoided.
Four practical considerations argue against tight time con-
straints. First,  a phased approach to sound and  cost-ef-
fective monitoring is recommended. Secondly, monitor-
ing during different  seasons is usually  important to take
into account  weather and runoff conditions. Third, tech-
nical problems inevitably arise in the taking and analyz-
ing of samples with  attendant delays of days  and some-
times weeks or months. Finally, data formatting and in-
terpretation almost always take much  longer than antic-
ipated.
  There is no standard procedure for presenting monitor-
ing data for risk assessment purposes. Both the site char-
acteristics and the specific discoveries at the site will prob-
ably dictate  somewhat unique formats that  present  the
most information in an understandable manner.  The re-
sults will  seldom be  clear cut, and adequate time should
be allowed for debate prior to settling on  an  interpreta-
tion.


BIOLOGICAL MONITORING—
THE TREND OF THE FUTURE?

  The problems in extrapolating from exposure monitor-
ing data  to  effective  dose and then  to human  health
effects have  been noted above.  Biological  monitoring
addresses  some of these problems. In particular, this ap-
proach attempts to  reduce the uncertainties concerning
the biological availability  of  chemicals  in the environ-
ment  and the  integration of exposures  to   chemicals
through multiple environmental  pathways.  It  provides a
method  of predicting human health effects  while also
directly measuring environmental effects.
  As indicated in Figure 11, one type of biological mon-
itoring is  simply to measure the chemicals  that accumu-
late in species indigenous to the local area. This  type of
program that was undertaken  at Love Canal is shown in
Figure 12. The species selected for the program are com-
monly found in many areas of  the  country. The chem-
icals of principal interest were those chemicals that were
known to  have been deposited in the Canal.  The approach
did not attempt to discriminate between the  original chem-
icals that  were taken up by the biota and the metabolites
of these chemicals.
  Biomonitoring was an additional dimension  to increase
the sensitivity of the  overall monitoring program at Love
Canal. It was hypothesized that chemicals not  detected at
significant levels in air, water or soil  might  accumulate to
higher levels  in biota  because of the multiple routes  of

-------
246     REMEDIAL RESPONSE
 Bioaccum ulation
   Whole Body
   Selected Organs and
   Tissues
   Body Fluids
  Btoeffects
   Biosystem Responses
   Pathological Changes
   Physical Condition
                          Figure 11.
          Biomonitoring Using Selected Animal Species
Field Mice:
Carcass
Hair
Dogs
Crayfish
Earthworms
Silver Maple
6
2


6
8
35
12
' 20
33
21
14
32
13
15
10
5
14
Organics
Inorganics
Inorganics in
Hair
Organics in
Whole Body
Organics in
Whole Body
Inorganics in
Leaves
                          Figure 12.
         Biological Monitoring Program at Love Canal

exposure.  At  the same  time  it  was recognized that  re-
liable  uptake  coefficients would be  needed for both the
chemicals  that were  found  and  for  some that were not
found before sound conclusions could be reached.
  A second type of biological monitoring is the measure-
ment of biological responses  to  chemical   contaminants
using either indigenous biological species or species intro-
duced into the area of concern. There are many research
reports indicating that certain species respond in various
and consistent ways  to  different types  of chemical ex-
posures. However, operational use of this technique will
require baseline  data on response/exposure correlations
developed in controlled laboratory experiments. In  the ab-
sence  of  extensive filed  on the  correlations of  such  re-
sponses with different  types of chemical exposures, such
responses cannot be interpreted in a meaningful way.
  Finally, the  most direct approach of all may  be medical
investigations  and human surveillance to identify possible
health impacts on nearby populations. A variety of tech-
niques have been tried  in other types of environments ex-
posed to air pollutants, food contaminants, and  pesticide
applications. These techniques have  ranged from  routine
chemical analyses of blood, urine and breath to investiga-
tions of impacts on responses of the nervous and immuno-
logical systems.
   Biological monitoring techniques are o'nly now begin-
ning  to emerge  from  the research  stage.  Conceptually,
these direct measures of chemical-biological interactions
can provide persuasive evidence of the problems or lack of
problems near hazardous waste sites.  It seems likely that
during the next  few years  they will be increasingly em-
ployed near sites. In the near future, however, they will
complement and  not replace our traditional reliance on
conventional monitoring.

PRACTICAL ASPECTS OF
IMPLEMENTING A MONITORING PROGRAM

   In  conclusion, the monitoring data will only be useful
if the detailed aspects of collecting and analyzing the data
are sound.  Several practical suggestions  in this regard in-
clude:
•At the outset the objectives of the monitoring program
 should be clearly defined. The users of the monitoring
 data should be  satisfied that the data  will be collected,
 analyzed,  and  presented in a manner  which  is respon-
 sive  to their needs.
•Monitoring programs can be  helpful both in identifying
 hot  spots  of near  term  concern and  in assessing the
 longer  term habitability of the area. Each of these ob-
 jectives may require  a  different program  orientation,
 however, as shown in Figure 13. Usually both objectives
 will  be important in varying  degrees, and  the program
 emphasis should be adjusted accordingly.
•The entire array of monitoring opportunities  should be
 reviewed at the outset. Many of the pathways that might
 be considered, particularly in situations where sites are
         Assessing Exposures Near Uncontrolled Sites
Objective

Chemicals of concern

Levels of concern
Likely sources

Pathways of primary
concern

Populations of concern

Assessment emphasis



Monitoring approach
Awning
Hal Spoil

Identify near term
exposure problems
High toxicity chemicals
suspected to be in area
"High" levels
Past and current leakages
at site
Air, drinking water, food
Susceptible individuals
near hot spots
Determine highest inter-
grated exposure levels for
specific groups of indi-
viduals
Limited multimedia sampling
broad areas and intensive
single media monitoring of
suspected hot spots near
populations. Emphasize mon-
itoring and interface of re-
ceptor
Aliening
Chemical Saturation

Determine long term
habitabilily of area
Many chemicals including
degradation products
All levels including trace leveli
Past, current, and future leakages

Also, soil, sediment, biota,
surface water, sewers, ground-
water
Entire population

Document highest and median
concentration levels in individual
media and compare such levels
with control areas
General multimedia sampling of
broad areas with bias toward
natural and man-made path-
ways from the site. Concur-
rent monitoring in nearby con-
trol areas. Monitor along entire
environment pathway
                           Figure 13.
               Dual Approach to Risk Assessment

-------
                                                                                 REMEDIAL RESPONSE    247
 Outdoor Air

 Indoor Air
   Public Buildings
   living Quarters
   Storage Areas

 Food
   Locally Grown
   Supermarket

 Drinking Water
   City System
   Wells/Springs

 Surface Water
 and Sediment
   Streams/Ponds
   Swales/Sinks
   Transportation
   Arteries

 Soil

 Sewer Systems
   Sanitary
   Storm
   Basement Sumps

 Groundwater

 Vegetation
                      Figure 14.
         Framework for Monitoring System Design
 intermingled  with other  industrial  activities  in  popu-
 lated areas are shown in Figure 14.
•Statistical aspects are an important factor in the  design
 of a monitoring program. A statistician on the planning
 team can help insure that adequate consideration is given
 to these aspects both in designing the program and  in
 formatting and interpreting the data.
•A photo interpreter can  also be an important member  of
 the planning team, both in selecting monitoring sites and
 in designing the approach for relating concentration lev-
 els to population activities.
•Before any samples are taken, a complete sampling plan
 should be developed with built-in check points for ad-
 justing the plan as sampling results are obtained. Devia-
 tions from the plan should be resisted other  than  to ad-
 just the plan in response to unexpected data results.
•As a rule of thumb, on the order of 10% of the sampling
 might be for screening to help clarify hypotheses as to
 possible gradients and hot spots around the site, 80% di-
 rected to investigating the hypotheses and 10% reserved
 for supplementary sampling of neglected areas that arise
 late in the program
•The sampling scheme should include provisions to con-
 firm or reject previously reported findings of a  contro-
 versial nature.
•Access to preferred sampling sites is not always possible.
 The sampling plan should be sufficiently flexible to com-
 pensate for such problems.
•Geophysical  investigations can  be very cost-effective in
 targetting groundwater and soil sampling sites.
•A quality assurance program involving surrogate recov-
 eries, inter  and intra laboratory duplicates  and  field
 and laboratory blanks is essential. Each data point should
 be individually validated as  acceptable data,  and  pre-
 cision and accuracy data should be developed for each
 data set. The quality assurance program may account for
 0% to 20% of the monitoring costs.
•Special efforts are needed to minimize holding times be-
 tween sampling and  analysis. However, extended hold-
 ing times beyond two weeks may be unavoidable. In that
 event appropriate storage procedures are particularly im-
 portant to prevent excessive decay of the samples.
•Contaminants  associated with the sampling and analy-
 tical techniques  are  difficult to avoid, and  data  sus-
 pected of such contamination should be considered for
 discarding. Of particular  concern, for example, are ben-
 zene and  toluene when  using  Tenax  traps, methylene
 chloride and phthalates that are present in laboratores,
 isophorone which can be a derivative  of the laboratory
 solvent acetone and  the  high pH  in groundwater asso-
 ciated with grouting of sampling wells that may result in
 artifacts being observed.
•Data formatting and presentation are  of critical impor-
 tance.  Plotting each data point on maps is probably the
 safest way to insure  a totally objective presentation of
 findings.
•Monitoring data may not provide a definitive portrayal of
 pollutant gradients  or pollution  patterns but may only be
 suggestive of general pollutant distribution.  Interpreta-
 tions of the data may be controversial, and efforts should
 be made to isolate criticisms of the quality of the data
 from criticisms of the interpretations.

-------
COSTS OF REMEDIAL ACTIONS AT UNCONTROLLED SITES
                                         HOWARD L. RISHEL
                                        SHEILA M. KENNEDY
                                             SCS Engineers
                                         Long Beach, California
                                       JAMES J. WALSH, P.E.
                                        DENNIS P. GILLESPIE
                                             SCS Engineers
                                          Covington, Kentucky
                                       OSCAR W. ALBRECHT
                                 U.S. Environmental Protection Agency
                                            Cincinnati, Ohio
INTRODUCTION
  During 1980, the U.S. Congress enacted the Compre-
hensive Environmental Response Compensation and Lia-
bility Act (CERCLA, also known as Superfund, P.L. 96-
510) which was proposed to provide funds for the U.S.
Environmental Protection Agency (EPA) to assist in the
mitigation of pollution problems at uncontrolled waste
disposal sites through remedial actions. The responsible
offices within EPA (the Oil and Special Materials Division,
and the Office of Environment) requested  the Office of
Research and Development to provide technical informa-
tion to support this process. As a part of this effort, SCS
Engineers was contracted to review, compile, update and
integrate existing data  on the costs of such remedial ac-
tions,  in terms of discrete  unit operations which could
then be combined to construct  conceptual remedial ac-
tion scenarios.
  This type  of review-and-update approach was  consid-
ered more appropriate than additional conceptual design
efforts because much conceptual design work had already
been done. The design work which exists, however, is scat-
tered, incomplete and inconsistent in methodology; much
of it is out of date and vague either about the methods
used to arrive at a cost figure or about what components
the cost figure included.
  Through the review-and-update approach, a consistent
methodology on the existing data in terms of scope, lo-
cation, time  frame and cost computations  was imposed.
In addition,  the missing details were supplied and results
presented in  a uniform format, with  a minimum  of over-
lap between  the  individual unit operations. The resulting
document presents these data in a framework of  a broad
and consistent methodology, with enhanced detail.
  No  new conceptual design work was done for this pro-
ject.  Where data  were  incomplete, some detailed  in-
formation was provided, but the thrust of this work was
to enhance previously existing conceptual design data and
make  them  more available  and useful  to enforcement
personnel responsible for overseeing the  retrofit opera-
tions.
  Because the document was intended for use in the re-
trofitting of uncontrolled sites, the unit  operations ex-
amined are limited to those appropriate for the clean-up
of closed or abandoned sites.
Methodology
  In developing and characterizing unit  operations for
remedial action at waste disposal sites, hypothetical site
profiles and their associated unit operation profiles were
developed. Costs for each unit operation were then com-
puted at three cost levels: high and low  U.S. averages,
and  the price estimate for a single city, Newark, NJ
(mid-1980 dollars).
SITE PROFILES

  Site profiles,  or hypothetical  disposal  sites,  were de-
veloped for  landfill disposal and for surface impound-
ments (disposal  ponds). Each of these was portrayed at
five scales of daily operation. The resulting site profiles
were configured to conform to uniform sets of design cri-
teria and  environmental conditions. For  both  landfills
and surface impoundments, the  selected scales of opera-
tion  were  developed  in terms of daily input. This em-
phasis on daily input is consistent with the usual view of
landfill  practices and the assumptions that surface im-
poundments are intended as temporary storage, with rela-
tively short retention times.
Landfills
  The size of the five hypothetical landfill disposal sites is
shown in Table  I. The range  of scale sizes was developed
from data presented  in References (2) and (7). The fol-
lowing assumptions were made:
  (1) The surface area for each landfill is square.
  (2) All landfills are cut and cover operations,  with cut
      slopes at a 2:1 ratio and fill slopes at a 3:1 ratio.
  (3) Operation  at each landfill was 260 days/year for
      ten years before the site was closed.
  (4) The  compaction  rate was  0.596  tonnes/com-
      pacted m3.
  The layout of the hypothetical landfill site, without ref-
erence to the scale of operation is shown in Figures 1 and 2.
                                                   248

-------
                                                                                     REMEDIAL RESPONSE    249
  To allow comparison of remedial actions between sites
operating at different scales, the following environmental
conditions were also held constant:
  (1) Ground surface and groundwater gradient are at a
     1% slope.
  (2) Groundwater is 4.0 m below the ground surface.
  (3) Low permeability strata (<10~6  cm/sec)  is 15 m
     from the ground surface.
  (4) Unconsolidated earth materials have a permeability
     of 10 ~5 cm/sec or greater.
Surface Impoundments
  The scales of operation for surface impoundments are
shown in Table II. The range of scale sizes was developed
from data in References (3) and (8). In this case, the scale
of operation is given in terms of the daily volumetric flow
of influent.

                       Table I.
       Size of Five Hypothetical Landfill Disposal Sites
     Weight of      Volume of       Area of        Waste
     Waste Material   Waste Material   Waste Material   to
     Contained      Contained       Contained      Soil
     (Tonnes)       (m3)           (ha)           Ratio
       24,000         40,000         0.99          1:1
       120,000        200,000         3.33          1.5:1
       240,000        400,000         5.41          2:1
       700,000       1,200,000        12.80          3:1
     1,200,000       2,000,000        18.51          4:1
     1 tonne = 1.1 ton
     skip that

     1 tonne = 1.1 tons
     1m3 = 1.31 yd3
     1 ha = 2.47 acre

   The plan of the hypothetical surface impoundment with-
out reference to the scale of operation is shown in Figure
3. The following  design criteria were common to all of the
surface impoundment site profiles:
•The ponds are square and unlined.
•Berms were constructed from soils excavated during pond
  construction, and have 3:1 side slopes.
•The site operated 365 days/year  for  ten years before
  closure or abandonment.
•Sediment was removed  from pond  bottom every  two
  years.
•Wastewater contained 100 mg/1 settleable solids.
•Density of solids was 2 g/ml.
•Sludge is 70% moisture by weight when removed.
•Wastewater was recirculated after allowing three to ten
  days for solids settling.
•Because of short  detention time and sludge on bottom,
  precipitation, evaporation and  percolation  losses  are
  considered  negligible  when  compared to the  volumes
  involved.
•Average percolation <14 1/day/m2 (0.2 gal/day/ft2)
  To help compare remedial actions for  each of the scales
of operation, the following environmental conditions sur-
rounding each pond were developed:
  (1) Ground surface and groundwater gradient are at a
     one percent slope.
GROUND SURFACt ( IX^-^"^ 83
	 r 	 ^V
hi
hz
"•^j 1 1
/%
SLOPE


I)'"
  GROUND «ATER
                               FLOW OF GROUND WATER
                  Low Permeability Strata
             Total volume of refuse tm )
             Total volume of soil (mj)
             Height of landfill above ground surface (m)
             Depth of landfill below ground surface (m)
             Top side of landfill (m)
             Bottom side of landfill (m)
             Side of landfill  at ground surface (ml
             Area of landfill  at ground surface (m )
                        Figure 1.
               Typical Side View of Landfill
                                           FLOW OF
                                           GROUND WATER
                        Figure 2.
               Typical Top View of Landfill

  (2)  Groundwater is 4.0 m below the ground surface.
  (3)  Low permeability strata  (aquiclude or aquitard;
      K<10~6 cm/sec) located  at  15  m  from ground
      surface.
  (4)  Unconsolidated earth materials have a permeability
      of >10~5 cm/sec.

                        Table II.
Scale of Operation for Five Hypothetical Surface Impoundments
      Influent Volume
      of Waste
      
-------
 250    REMEDIAL RESPONSE
                                                        TOTAL
                                                       LENGTH
BERM
           POND  WATER  SURFACE
GROUND  SURFACE
                                                                A
                                                                                FREEBOARD
   GROUND  WATER TABLE
                                                                                                                4 ID
                                                                                                      ism
 Low Permeability Strata (Aquiclude or Aquitard)

 Where:
 •(Total length) = impoundment surface area.
 •Berm side slopes constructed at 3 horizontal to 1 vertical.
 UNIT OPERATION PROFILES
        •Berm top width nominal 2 m, maximum 3 m rs.
        •0.5 m freeboard designed into all ponds.
        •h = pond depth.
                                                      Figures.
                                           Side View of Surface Impoundment
   Thirty-five  unit  operations   were  identified  which
might be used as  part of a remedial action effort at an
uncontrolled hazardous waste disposal site. These opera-
tions are shown in Table III for both impoundments and
landfills. Although there appears to be some overlap be-
tween the  two lists, each operation was configured sep-
arately for the distinctly different design criteria and en-
vironmental conditions assumed in each site profile.
   AH of the unit  operations  shown were taken from the
literature and only  those  for which the  literature  con-
tained adequate conceptual designs and cost  data  were
addressed. In some cases, a part  of the  necessary  data
was missing and was supplied by SCS.
Alternative Unit Operations

   When a pollution  problem exists, a number of unit
operations may be used interchangeably.  The following
list gives some unit operations which may  be used either
conjunctively or as alternatives for each other:
For elimination of contaminated site runoff, and preven-
tion of  precipitation from entering a landfill or closed
impoundment, the following are conjunctive or alterna-
tive unit operations:
   •Contour grading and surface water diversion
   •Surface sealing
   •Revegatation
   •Berm construction/reconstruction
For minimizing leachate formation,  the  following are
conjunctive/alternative unit operations:
   •Contour grading and surface water diversion
   •Surface sealing
   •Grout curtain
   •Sheet piling
   •Slurry trench
                                  •Well extraction
                                  •Well point system
                                  •Chemical fixation
                               For the control of leachate/contaminated  groundwater
                               migration, the following are conjunctive/alternative unit
                               operations:
                                  •Grout curtain
                                  •Grout bottom seal
                                  •Sheet piling
                                  •Well point system
                                  •Well extraction
                                  •Well injection
                                  •Underdrains
                               For gas (methane and other  volatile hydrocarbons), the
                               following are conjunctive/alternative unit operations:
                                  •Perimeter gravel trench
                                  •Gas migration control—passive
                                  •Gas migration control—active
                                COST COMPILATION

                                  Once  a remedial action unit  operation  had been de-
                                fined in  terms  of its intended use (with respect to the
                                landfill or surface impoundment site profiles) and the ex-
                                tent to which it can be used in conjunction with or as an
                                alternative to other unit operations, the cost of the opera-
                                tion can be defined for each scale of operation associated
                                with the site profile. To do  this,  the unit operation is
                                broken down into component requirements. Each com-
                                ponent is further  defined in terms of  sub-components
                                (labor,  materials and equipment) and costs are assigned
                                to each  component  and sub-component.  The assigned
                                costs are in terms of  mid-1980 dollars for U.S. upper and
                                lower cost overages (for the continental 48 states), as well
                                as for the example location of Newark, NJ.

-------
                                                                                 REMEDIAL RESPONSE     251
                      Table III.
        Unit Operations Used as Remedial Actions
Landfills
 1. Contour grading and surface
   water diversion
 2, Surface sealing
 3. Revegetation
 4. Bentonite slurry trench cut-
   off wall
 5. Grout curtain
 6. Sheet piling cutoff wall
 7. Bottom sealing
 8. Drains
 9. Well point system
10. Deep well system
11. Injection
12. Leachate recirculation
13. Chemical fixation
14. Chemical injection
15. Excavation and reburial
16. Ponding
17. Dike construction
18. Perimeter gravel trench vents
19. Treatment of contaminated
   waste
20. Gas migration control—
   passive
21. Gas migration control—
   active

Impoundments

22. Pond  closure  and  contour
   grading of surface
23. Surface sealing of closed im-
   poundments
24. Revegetation
25. Slurry trench cutoff wall
26. Grout curtain
27. Sheet piling cutoff wall
28. Grout bottom seal
29. Toe and underdrains
30. Well point system
31. Well extraction system
32. Well injection system
33. Leachate treatment
34. Berm reconstruction
35. Excavation  and disposal  at
   secure landfill
  After costs are assigned to each component, conceptual
design capital and operating cost estimates  are accumu-
lated and allowances for overhead and contingencies are
applied.  Total and average life cycle costs are then com-
puted for each unit operation.
Estimation of Component Costs

  For the most part, the 1980 Dodge and Means Guides'4'6'
were used to obtain the needed costs. The costs were then
expressed in terms of metric units.
  Regional  adjustments indexes  presented in the Dodge
Guide(4)  were used to modify the metric versions of the
cost estimates for geographical differences. These indexes
were applied to obtain revised material and labor costs for
the U.S. low, U.S. high and Newark, NJ estimates. No
index was applied to equipment costs, since it was assumed
that equipment costs are the same nationwide.
  Because the Dodge Guide"0 and  Means'6' present costs
differently,  assumptions were made so that the  regional
adjustment indexes could be used for both texts. For ex-
ample, in the Means, labor costs were not identified as a
separate entry, but were included as part of installation.
Thus, whenever the Means was used to present costs, the
Dodge Guide Regional Adjustment Index for Labor was
applied to installation costs.
  Frequently, neither guide itemized costs into categories
of labor, material and equipment but simply presented a
"total"  estimate. Depending upon which reference was
used, the following  rules  were applied:  in the Dodge
Guide, if only  a total cost was presented, an average
labor/material index  was applied to the unit cost;  in the
Means, the  "total" costs include an overhead allowance
of 25%.  This allowance was removed before  the labor/
material index was applied. In all cases, costs were adjust-
ed so that overhead allowances were not included at the
subcomponent level.
  As the scale of operation changed, the quantity of any
one component required for a unit operation also changed.
The  cost of each component presented in the unit opera-
tion  conceptual design cost tables typically includes the
sum of costs  for any material, labor or equipment sub-
components. These total costs for each component do not
include overhead and contingencies.
  Once all the components within a unit operation were
costed, the costs were summed, giving a subtotal capital
cost  for the unit operation. This subtotal capital cost was
then used to obtain an overhead allowance (always 25 %),
and  a contingency  allowance (between 10 and 40%, de-
pending upon the  unit operation). The  subtotal  capital
cost  was added to  the overhead and contingency allow-
ances to obtain the estimate of total unit operation capi-
tal cost. (See  example below.)  This method was used for
all scales of unit operation.

Life Cycle Costing
  Once total capital and operating costs were-determined
for lower  and upper U.S. averages and for  Newark, total
and  average life cycle costs were computed  to ensure that
any  subsequent  cost comparisons  of  unit  operations
could be equitably accomplished.
  Although operation and maintenance (O&M) cost esti-
mates are for 1980, as the first year of operation, O&M
component quantity  requirements were estimated to ac-
curately reflect requirements for each of the first ten years
of remedial  operations.  This  ten-year  life of the  con-
ceptual designs means that life cycle evaluation of operat-
ing costs only addressed subsequent inflation and appropri-
ate discounting  of  these O&M component costs  to their
mid-1980 present values.
  It  was further assumed that capital costs would not  be
amortized and discounted, but would be  considered  as
fully incurred in the first year of operation. As a result of
these assumptions, average annual compounding inflation
rates for  electricity and for all other O&M components
were derived using estimates from the April 1980 Survey of
Current Business'". These inflation rates were derived as
shown in Table IV.
  In determining the present values of future expenditures,
the  March  1980 Gross National  Product  Implicit Price
Deflator of 174.51(5) was similarly evaluated in terms of its
1972 base year to estimate an annual general inflation rate
of 7.4%.  To  this, an assumed 4% social time preference
rate  was added  to create a total annual  discount  rate of
11.4%. The life cycle cost methodology  was then fol-
lowed, in which inflated operating costs  were  discounted
to their mid-1980 present values,  and summed with total
capital costs,  to  determine total life cycle  costs over the
ten-year  life  span  of each unit  operation. Average life
cycle costs were then computed by dividing this  total  by
the site profile's daily scale of operation.
  The contract report of this work includes two tables for
each unit operation,  giving the cost of each component,
plus  overhead and contingency  allowances,  O&M costs
and  life  cycle costs  for a medium size  landfill and im-
poundment. One of these unit operations is given as part
of the costing example in Tables V and VI.

-------
252     REMEDIAL RESPONSE
EXAMPLE UNIT OPERATION

  An example of remedial action unit operation can be
taken in the revegetation of a landfill (Unit operation 3,
Table III).  All other unit operations were derived simi-
larly and reported in the same format in the contract re-
port.
  One such unit operation is described in Table V in gen-
eral terms.
  Table VI describes how the vegetation unit  operation
would be achieved in a medium  size landfill operation.
The component requirements  and their  associated costs
when the revegetation  unit  operation  is  applied  to  the
5.41 ha landfill site are  shown in Table VI. In this table
both capital and O&M component requirements are identi-
fied and costed in terms of mid-1980 dollars.
  The costs associated  with  each  component may repre-
sent the sum of various labor, material and equipment
costs incurred in accomplishing that component. For ex-
ample,  costs for the "mulching"  component are for ap-
plying mulched hay over 5.41  ha.  The  labor  costs,  ac-
cording to the 1980 Dodge Guide, are typically $85.20/ha
($34.50/acre). This number  was adjusted  for regional
labor cost differentials  and multiplied by the number of
hectares involved; these labor costs  ranged from  $50 to
$110.  The  value  for  Newark, NJ  was $100.  Material
(hay) and equipment costs were added to these labor cost
estimates seen in Table VI. A similar process was followed
in costing each capital and O&M component.
  Once all of the capital components had been identified
and costed, allowances  for  overhead and contingencies
were added to complete the capital cost portions of these
estimates. In all unit operations, a 25%  overhead allow-
ance was assumed.  This assumption is partly  based on
the fact that the Means construction cost guide(6) also as-
sumes a 25 % allowance for contractor's overhead. A con-
tingency allowance to cover unforeseen cost additions was
also applied to each capital cost subtotal. In general, the
contingency allowance ranged from  10 to 40% depending
on the extent to which the unit operation was expected to
encounter unforeseen difficulties.
  Each of the O&M component costs were escalated for
future inflation and then discounted to their present values
(in mid-1980 dollars). These present values were summed
over a ten-year life cycle of future  site maintenance and
then added  to the capital cost total to determine total life
cycle cost. Average life cycle cost was computed by di-
viding this  total by the  landfill's former daily scale of
operation.

APPLICATION AND LIMITATIONS

  The contract report document,  produced as a result of
this review  and update  work, consists primarily of unit
operation  costs  at  the  component and subcomponent
levels, and average life cycle costs.  By applying this cost
information to  the  report's discussion  of alternative or
conjunctive unit operations, enforcement personnel will be
able  to configure and  cost  complete  remedial  action
scenarios.
  The following is an example intended to show how the
report may be used to help develop an estimated cost for
a remedial action  scenario, involving a combination of
unit operations. The estimated cost for the hypothetical
remedial action is then calculated using the cost data pre-
sented in the report.
Hypothetical Problem
  An abandoned  hazardous waste  site has been investi-
gated  and  is  found  to be  contaminated surface and
groundwater.  It is decided  that the site must  be isolated
by: (a) preventing surface runoff from entering the stored
hazardous waste,  (b)  preventing  groundwater migration
through the site, and (c) implementing a monitoring pro-
gram to confirm the effectiveness of the steps  taken. The
unit operations required are as follows:
•Contour grading and surface water diversion
•Surface sealing
•Bentonite slurry  trench cutoff wall.  Monitoring is  in-
 cluded in unit operation.
  For example purposes, it  is assumed that the hypotheti-
cal site has the following dimensions:
•Surface area  = 4 ha
                        Table IV.
        Derivation of Inflation Rates for O&M Inputs
Type of O&M Input                Electricity

Related Published                 Electric
Cost Index:                      Power

March 1980 Index Value:            305.7

Average Annual Index % Increase
Since Base Year (i.e., 12.75 years
since mid-1967 when index
equaled 100):                     9.160%

Assumed Future Inflation Rate         9.2%
All Other
O&M Inputs

Consumer
Price (CPI-W)

239.9
7.104%

 7.1%
                        Table V.
              Unit Operation 3—Revegetation

Use:
  Revegetation helps to physically stabilize the earth material and re-
  duce infiltration; it also serves to minimize erosion of the cover ma-
  terial by wind and water.
Configuration:
  Revegetation involves first grading the landfill,  covering  it with 1
  suitable, fertile soil, adding soil supplements, and then seeding.
Conjunctive Uses:
  Contour grading and gas migration control systems are used as colt
  components.
Assume:
  1. Entire surface of landfill is revegetated.
  2. 0.6 m of clay and silt loam will be used for landfill cover.
  3. Clay and silt loam are easily accessible;  transportation costs tK
    not included.
  5. Native grasses will be used for seed.

-------
                                                                                REMEDIAL RESPONSE    253
•Site is square, 200 ra each side
•Average depth of bentonite slurry trench cutoff wall
 must be 10 m to impervious material.
•Hydrological  investigation  indicates  that  bentonite
 slurry trench cutoff wall must extend around three sides
 of site  to cut off groundwater migration through  the
 site.
  The procedure for estimating the cost of this hypotheti-
cal remedial action scenario is as follows:
  1. Refer to the  pertinent tables in the report, and list
    the components of the unit operations selected. This
    step is shown in Table VII, first column.
  2. Refer to the price list  in the appendix of the report,
    and determine what units will be required to measure
    the cost of each component. This step is shown in
    Table VII, second column.
  3. Calculate  the number of units of  each cost com-
    ponent required for the site. This step is shown in
    Table VII, third column, for the hypothetical  site
    used in this example.
  4. Refer to the price list  in the appendix of the report,
    and list the unit cost for each cost component re-
    quired. A decision will have to  be made on whether
    to use U.S.  high, U.S. low, or Newark, NY  costs.
    For this example scenario, U.S.  high costs were used,
    and are shown in Table VII, fourth column.
  5. The final cost calculation required multiplication of
    the number  of  units (Step 3  above) by the unit
    cost (step 4 above), as appropriate, and summation
    of the cost components to arrive at a total cost. For
    this example, see Table VII, fifth and sixth columns.
  For the hypothetical remedial action scenario used here
as an example,  the  estimated costs were calculated  as
follows:
    Total capital cost                        $1,277,100
    Total O&M Cost, during 10 years        $  128,700
    Total 10-year life cycle cost              $1,405,800

CONCLUSIONS

  The literature review verified that little has been done in
estimating the cost of hazardous waste cleanup at un-
controlled or abandoned sites. Those sources which  ad-
dressed remedial  responses frequently followed a  case-
study or national, industry-wide approach;  cost informa-
tion, if provided, was too  highly aggregated for costing
separate remedial unit operations. Those sources  which
did develop cost estimates at the unit operation level fre-
quently omitted critical components  or allowed substan-
tial overlap  in  the scope of each unit operation. This
study has attempted to  overcome such  deficiencies, as
well as to quantitatively bound the effects which  scale
economies and regional price differentials are expected to
have on the costs of implementing 35 different remedial
response unit operations.
  the  primary product of  this study has been a costing
methodology  which was  consistently applied to each of
these unit operations.  The resulting cost estimates would
seem to lend  themselves  readily to:  (1) comparing costs
for alternative unit operations which perform  the  same
                        Table VI.
         Costs of Revegetation for Medium Size Landfill
                    (5.41 ha-27,685m3)
Capital Costs
Area Preparation
Excavation, Grading & Recon-
touringof site
Hydroseeding
Mulching
Capital cost (subtotal)
Overhead Allowance (25 %)
Contingency Allowance (10%)
Total Capital Cost
O&M Costs
Grass Mowing (6/yr)
Refertilization (1 time/yr)
Total O&M Cost
Total Life Cycle Cost'
(over 10 years)
Average Life Cycle Cost*
Per ha
Perac

Lower U.S.
3,710
43,820
4,900
1.450
53,880
13,470
5,390
72,740

320
250
570
77,550
14,335
5,796
Dollars
Upper U.S.
6,440
50,790
6,900
2,080
66,210
16,550
6,620
89,380

650
350
1,000
97,810
18,079
7,310

Newark, NJ
5,760
49,050
6,200
1.860
62,870
15,720
6,290
84,880

550
300
850
92,050
17,015
6,880
   •See text for methodology and assumptions

function, or (2) computing combined cost estimates for
unit operations which comprise a complete remedial re-
sponse scenario. The problem with the first type of use is
that such simple cost comparisons do not address techni-
cal differences in the capabilities or efficiencies of alterna-
tive unit operations which accomplish the same goal. Such
differences depend  on both the inherent configuration of
each respective unit  operation,  and the  environmental
setting under which it is actually implemented.
   As comprehensive  as the  site profiles  may  be, they
still represent a conceptual design of an  environmental
setting and are no substitute for actual site conditions
when choosing among alternative unit operations. A sub-
sequent report may address some of these qualitative con-
siderations.
   The  problem with the second  type  of application is
that the simple addition of  unit operation costs, when
configuring a complete remedial response scenario, ig-
nores both real-world influences  on  conceptually derived
costs and the minor joint effects  of scale and agglomera-
tive economies.  More  specifically, some scale economies
may still be enjoyed when multiple unit operations all re-
quire similar component inputs. Agglomerative economies
resulting from  minor component overlap or redundancy
may also  occur when  unit  operations   are  combined
into remedial  response  scenarios. Subsequent research
should address these considerations.
   Because complete remedial action  scenarios for uncon-
trolled sites typically consist of  several unit operations,
much remains  to be  done, even from  a conceptual de-
sign cost perspective,  in identifying the most appropriate
short- and long-term remedial scenarios.  Such an effort
might include a systematic evaluation  to  determine the
most prevalent pollution problems  occurring at  uncon-
trolled landfill  and impoundment sites. Once this set of

-------
254    REMEDIAL RESPONSE
"typical"  pollution cases  has been  determined, likely
remedial action scenarios could be configured using  unit
operations developed in this study.  The resulting com-
posite cost estimates for these  scenarios could be de-
termined using unit operations developed in this study and
then compared to determine the relative and absolute  cost
advantages of each alternative scenario.
REFERENCES

1. Current Business Statistics,  Survey of Current  Busi-
   ness, Vol. 60, No. 4 U.S. Department of Commerce,
   Bureau of Economic Analysis. Apr. 1980.
2. Fred C. Hart Associates,  Inc. Analysis  of the Tech-
   nology,  Prevalence and Economics of Landfill  Dis-
   posal of Solid Waste in the United States—Volume II.
   EPA Contract No.  68-01-4895,  U.S.  Environmental
   Protection Agency, Office of Solid Waste, Washing-
   ton, D.C. 1979.97pp.
3. Geraghty and Miller, Inc. Surface Impoundments and
   Their Effects on Ground Water Quality in the United
   States.  EPA 570/9-78-004,  U.S. Environmental Pro-
   tection  Agency, Office  of Drinking Water, Washing-
   ton, D.C., 1978.
4. McMahon, L.A., "1980 Dodge Guide to Public Works
   and Heavy Construction Costs," McGraw-Hill, New
   York, N.Y., 1979.
5. National Income and Product Tables, Survey of Cur-
   rent Business,  60, No.  4. U.S.  Department of Com-
   merce, Bureau of Economic Analysis, Apr. 1980.
6. Robert  S. Means Company,  Inc., "Building Construc-
   tion Cost Data 1980," 1979.
7. SCS Engineers, "Study of On-Going  and Completed
   Remedial Action Projects: Survey Results and  Recom-
   mended Case Studies," US EPA, Solid and Hazardous
   Waste Research Division, Cincinnati, Ohio, 1980.
8. SCS Engineers, "Surface Impoundment Assessment in
   California."  US  EPA Contract No. 68-01-5137, Of-
   fice of Drinking Water, Washington, D.C. 1980.

-------
                                  FORCED CLEANUP;
           A POLICE ACTION OR A MONEY JUDGMENT?
                                          JAMES C. SCOTT
                                        ROBERT B. PEARCE
                                     Black & Veatch of Michigan
                                           Detroit, Michigan
INTRODUCTION

  In September 1980, Black & Veatch of Michigan, Con-
sulting Engineers, was appointed to  the unique position
of Circuit Court Receiver. Not only did this appointment
come as a surprise, but also as an agent of the court, it
gave the Engineer responsibility of undertaking compli-
ance with a previously entered Court Order.  Included
were certain activities which had not been performed by
the Defendant in accordance  with the direction of the
Court. These actions were in the case  of the State of
Michigan, et al. vs.  Berlin & Farro Liquid Incineration,
Inc., et al.; a case involving an uncontrolled hazardous
waste site in Genesee County, Michigan.
  In addition to making the appointment  of a Receiver,
the Court required the Defendant to deposit a cash bond
to ensure payment to the Engineer for services to be per-
formed. Work was to include isolation, stabilization and
confinement of liquid hazardous waste, with related long-
term monitoring to ensure compliance. The program was
to cover design, methodology for implementation and
performance of the work.
  When the Engineers' recommended program was sub-
mitted to the Court for approval, the Defendant took ac-
tion to recover the bond and transfer jurisdiction in the
case to Federal  District Court. There he was involved in
bankruptcy proceedings. At issue was the clarification of
whether the Circuit Court Order constituted enforcement
action under the State's police powers or constituted a
money judgment against the Defendant.
HISTORY

  In 1971, the Berlin & Farro Liquid Incineration Com-
pany applied for and received from the State of Michigan
a license to haul liquid industrial waste,  as well as an air
quality permit to install a liquid waste incinerator. The in-
cinerator was placed in  operation in late 1972 and the
Company began accumulating wastes at the site.
  Subsequently, community complaints led to an investi-
gation by the Michigan Department of Natural Resources
(DNR).  The inspection revealed  Berlin & Farro was op-
erating in  violation of various State Regulations. Formal
Administrative Complaints were filed against the Com-
pany by the State in 1973.
  In 1974, the DNR Air Quality Division conducted an
intensive investigation of the incinerator and concluded
it could continue to operate on an interim basis provided
particulate control equipment was installed and certain
procedures were  followed during its installation.  By this
time,  Berlin  &  Farro had  accumulated a  substantial
amount of drummed liquid waste, installed several holding
tanks  both above and below ground,  and constructed
two large storage lagoons for liquid waste.
  In March 1974, spills of industrial waste on the site were
documented  by the DNR while investigating compliance
with previous orders. By  May,  a  Consent Order  was
drafted by the Michigan Air Pollution Control Commis-
sion and a Pollution Incident Prevention Plan was sub-
mitted to the Michigan Water Resources Commission.
  In November,  several spills of liquid industrial waste
again  occurred on the property, resulting in contamina-
tion of a County Drain. But,  by October, Berlin & Farro
was making  some progress in implementing its Pollution
Prevention Plan  and the  Water Resources Commission
provided a permit to  allow continued  storage of liquid
waste  on the site.
  DNR investigations conducted early  in  1975  showed
the Company was not continuing to make good faith pro-
gress in meeting air, waste storage and sampling  require-
ments. By August, the Company had accumulated over
2,300 drums (55 gal each) of liquid and solid waste on the
site. An estimated 2.5 million gallons of bulk liquid waste
was also being stored in the lagoons and underground
storage tanks.
  Several meetings were held by the DNR with the State
Attorney General and Berlin  & Farro attorneys in an at-
tempt  to reach some agreement on the air and water
quality problems at the site.  These  efforts  failed and in
September 1975, an Emergency  Order to Cease and De-
sist was issued against the Company by the DNR. This
Administrative Order was to close the incinerator and re-
voke the liquid industrial waste hauling license.
  Ten days after the Emergency Order was  issued, Berlin
& Farro attempted,  but failed to have the Circuit Court
restrain the DNR from enforcing its Order. Two months
later the Company was served by the Department with an
Order for hearing an  administrative complaint. Early in
December, a hearing  took place and  in February 1976,
the hearing examiner  issued recommendations which in-
cluded reinstatement of the liquid waste hauler's license
                                                   255

-------
256    REMEDIAL RESPONSE
and a permanent Cease and Desist Order for operation of
the incinerator.
  The Air Pollution Control  Commission then  acted,
but in a somewhat less positive fashion.  They issued a
permanent Order to close the incinerator,  unless proper
air pollution devices were installed. In April, the DNR
had restored the Berlin & Farro hauling license.
  In June  1976, another spill occurred at the site, again
contaminating the County Drain and a downstream creek.
Various  administrative  and  legal  maneuvers  continued
well  into 1977, when Berlin & Farro was  implicated in
other litigation as having received C-56 from the Hooker
Chemical and Plastics Corporation. DNR personnel then
attempted  to sample soil, water, waste and sediment at
the Berlin  & Farro site.  The Company refused to grant
permission for entry and a search warrant was eventually
obtained. Samples were taken  from the site, drains,  la-
goons and nearby creek bed.
  Results from the survey  revealed that soils from the
lagoon and surrounding areas were contaminated with ex-
tremely high concentrations of heavy metals (lead, nickel,
chromium, copper and  zinc), organochlorines and other
organic compounds. Sediments in the County Drain con-
tained concentrations of organochlorines higher than those
found in the lagoon, with some values exceeding 27,000
mg/kg. Contamination of sediment in the drain and creek
occurred downstream for over 15 miles.
  Administrative efforts continued into 1979 when the
Michigan  Attorney  General  finally  filed  suit  against
Berlin & Farro et al. in Genesee County  Circuit Court.
This litigation led  to a Preliminary Injunctive Order but
the Company  failed to comply with all its terms.  As a re-
sult, in August 1980, the Genesse County  Circuit Court
issued a Contempt Order and fined Berlin  & Farro $250
(maximum under Michigan law). The Court also ordered
appointment of a Judicial Receiver to assure compliance
with the previous Consent Order.
  To appoint a Receiver, the Court requested the State
to prepare a list of qualified firms and/or persons. The
DNR provided the names of three engineering firms and
one private individual, qualification and experience docu-
ments were solicited, and during a September 1980 hear-
ing, the Court ordered the appointment of Black & Veatch
of Michigan. Further, it ordered the defendant to post a
$10,000 cash bond to cover the affairs of the Receiver.

RECENT EVENTS

  Black & Veatch was directed by the Court to develop a
program to assure compliance  with a limited number of
specific items from the Preliminary Injunctive Order. The
items selected  were determined  during the September
hearing and were mutually agreed upon by the State and
Berlin & Farro.
  For the most part, the expected results and to some ex-
tent,  the methodology by which these items were to be
completed  were stipulated  in the Order,  e.g.,  "...shall
erect  a fence,  at least six feet in height (including barbed
wire topping), along the perimeter..." or "...shall deposit
soils  from the Site...on top of the sludges in order to
minimize any potential odors to the extent possible." At
the same time, some vagueness existed, i.e., was it to be a
cyclone fence or stranded wire and how much  cover was
justified to  minimize odor? The answers would have a
significant effect on the cost.
  The Court also directed Black & Veatch to include an
implementation schedule in the program. Prompt enforce-
ment was the goal. While  the program was to provide for
strict compliance, the Engineer was requested to consider
alternatives  which might provide satisfactory compliance
at a lesser cost.  A second hearing was to follow the En-
gineer's Report and provide a form for final negotiation of
method, cost, contractors and  schedule. At this  point,
Berlin & Farro continued to insist on the right to plan and
determine compliance activities.  The Court, on the other
hand, appeared increasingly impatient and less responsive
to the Berlin & Farro demands.
  Black & Veatch submitted a report to the Court  within
three weeks. Since considerable earth work was to be in-
volved, the impending winter made it apparent that rapid
mobilization would be essential to ensure completion prior
to frost. Failure would delay completion for at least six
months.
  But the Engineer apparently was not the only one busy.
Only a few days after filing its Report with the Court,
Black & Veatch was served with  a Temporary Restraining
Order. This notice came from the Federal District  Bank-
ruptcy Court in Detroit, Michigan. A hearing date was set
and the work stopped. Since that time, no real progress has
been  made in resolving the continuing migration of pol-
lutants from the site.  What has occurred has been a vari-
ety of legal and bureaucratic actions aimed at achieving
the long term goals of the participants.

SPECIAL CONCERNS

  Rather than continue with the chronological discussion
of historical facts, attention will  now be focused on some
of the more interesting factors which have evolved  in this
case.  Some will  be typical to  many cases.  Others will
suggest a variety of complicated factors that make imagin-
ative  investigation  and thorough  planning essential to
successful resolution of hazardous waste problems.
Money
  The title  of this  paper makes reference to  A Money
Judgment.  It is  of little consequence whether or not this
case involves a money judgment. Money has and will con-
tinue to be a key issue. It is probable the same will hold
true in most other  cases. Corrective costs are high and
available sources of funds are limited. Defiance may have
played a major role in the inactivity of the Defendant in
this case but the cost of corrective action surely has been
a major concern. Some progress was possible when Com-
pany resources  were used in remedial  activities.  When
cash  distribution to  the  Receiver  became an  issue, the
creditors objected and quickly reduced  the Court's flexi-
bility and enforcement.
  The State frequently claimed that monies could easily be
made available from Company assets (land) and State or

-------
                                                                                 REMEDIAL RESPONSE     257
Federal funds. This unfortunately has not been the case.
The problem with the Company assets has  been compli-
cated by the bankruptcy proceedings and  even further by
a legal technicality which  will be discussed below. Statu-
tory and political factors have also restricted ready alloca-
tion of governmental funds. Even appropriations to the
Superfund are  not likely to be  directly  applied  to  this
project.

Jurisdiction
  Jurisdictional questions have  been far from simple.
What started as an incinerator-air pollution question has
developed into  a water resources concern. Improper  and
uncontrolled storage rather than burning of the materials
became the long-term issue. Action to rescind the incin-
erator license therefore had limited effect on  environment-
al control.
  Eventual litigation  appeared to place the  State  Circuit
Court in a position to enforce permanent cleanup.  The
role of the Federal Bankruptcy Court  interfered with the
effectiveness of this effort and has significantly delayed
control of the pollutants.
  Declaration of a Toxic Substances  Emergency  by the
State of Michigan (only the second such declaration in
the State's history) has shifted the responsibility  for di-
recting actions from the  DNR to the Michigan  Toxics
Substances Control Commission (TSCC).  This has result-
ed in the Commission initiating an entirely new program
for permanent site cleanup. At the same time, it has been
reported by the State that the Circuit Court continues to
claim jurisdiction in the case and that it has indicated any
program developed will be subject to review by  its  Re-
ceiver prior to authorization.
Liability
  When initially considering cleanup  responsibilities, the
TSCC resolved to solicit  the  services  of  a consultant to
develop various alternatives and recommend the  "best"
alternative for a permanent solution at the site. Through
subsequent actions  however, the Commission itself  cre-
ated definition  of the alternatives and had an evaluation
of these performed by an Environmental Protection Agen-
cy consultant.
  An interesting  liability question is raised. In perform-
ing as  a Court  Receiver,  Black &  Veatch  concluded
that their liability was  limited to the execution  of the
Court Ordered tasks in  a responsible and professional
manner. The assignment did not include creation or evalu-
ation of the final design concepts. Such was not the case
in the TSCC plan. Their original plan called for the crea-
tion of concepts which could have had far. reaching lia-
bility implications to the consultant  who created them.
The eventual action, which called for independent evalua-
tion of the TSCC-created alternatives, still carries with it
a far greater liability potential than many firms might be
willing to prudently accept.
  There is a similar concern for  potential future liability
on the part of a prospective  recipient of the hazardous
waste.  One alternative calls for permanent  burial of the
Berlin & Farro materials  at the Hooker Chemical Com-
pany  "vault"  in  western Michigan. While there is  a
recognized presence of some Hooker wastes in this ma-
terial, it includes many hazardous materials derived from
other sources. If the Berlin & Farro waste were accepted,
the leachate  system at the  vault  might collect some of
these other wastes and place Hooker in a position of fu-
ture responsibility for treatment and disposal of materials
it  did  not produce,  a circumstance it would  certainly
not welcome.

Time
  As pointed out in the historical summary, difficulties
in this  case were first identified in 1973. Over eight years
later, little has  been done to correct the environmental
abuse at the site.  Granted, action has  been effective in
halting an escalation of the problems but a real solution
has yet to be  implemented. This situation is extremely dis-
appointing in view  of the major actions taken: state ad-
ministrative orders, Circuit Court Orders, appointment of
a Court Receiver and a State Declared Emergency.
   Initial efforts appear to have been directed toward con-
trol of the facilities while still operational. As problems
became more pronounced,  a shift in strategy seems to
have occurred. The procedures called for in the Court's
Preliminary  Injunctive Order  were  clearly  intended to
secure the site to minimize immediate environmental de-
gradation, rather than toward a permanent solution.
   The more  recent steps, in response to the emergency
declaration by the  TSCC, are  now  directed toward the
original objective of a permanent solution. This direction
will, of course, be accompanied by an increased  cost of
implementation. It would  be  speculative  to   suggest
whether this  vacillation in objectives was premeditated or
circumstantial but it is  probable that it will markedly af-
fect both the time at  which contamination is controlled
and the point at which the case is eventually concluded.
Legal Constraints
   The nature of the action often required in hazardous
waste cases indicates that legal  considerations become of
paramount importance. In the case of Berlin &  Farro,
the State operated for some time under  the  misguided
conception that funds would be reasonably available by
liquidating the site  on which the Company was operating.
Unfortunately, the corporation held  title to only  a small
portion  (approximately ten acres)  with the  remaining
property being held personally by the Company president,
a private  individual and not a  party in  the litigation. As-
cess to the assets was  further complicated by the bank-
ruptcy proceedings.
   Enforcement power does not carry with it the automatic
right of access to  the  property of the Defendant. Con-
fusion and inaccurate public reporting  developed regard-
ing  the  ease with which  investigations could  be  per-
formed. This resulted in several instances of strained re-
lations and implied threats.  At one  point,  the Company
president brandished  a  shotgun to prevent Court  au-
thorized seizure of a Company vehicle.
   Two bonds were posted in this case. The first, a cor-
porate guarantee  bond has been  placed  on the list  of

-------
258     REMEDIAL RESPONSE
debtors to receive its  share  of the  liquidated assets of
Berlin  &  Farro preventing  funds from becoming im-
mediately available. Even the second, a cash bond in the
hands of the Circuit Court, became temporarily unavail-
able for its intended purpose as a result of continued legal
maneuvering.

CONCLUSION

  Certain conclusions can be drawn from the experiences
in this case.  Hopefully they  can be used beneficially' in
other hazardous waste projects.
•A change in certain statutes may be appropriate to better
 define which agencies or authorities have jurisdiction in
 various hazardous waste affairs. The need for protection
 of personal rights and a balance of  powers prevent a
 clearly defined answer to this question.
•Participants  in hazardous waste activities should  give
 careful  consideration to  potential  liability.  Long-term
 risks could well overshadow the long-term benefits.
•A realistic assessment should be made at the onset re-
 garding desired results and the time for accomplishment.
 Wasted effort and lost time will otherwise result.
•The entire hazardous waste  arena  is constantly faced
 with legal entanglements. Lack of attention to this fact
 may well result in a failure to meet the desired objective.
  Finally, the question of A Police Action vs. A  Money
Judgment has yet to be clearly settled. While the bank-
ruptcy Court  eventually directed payment of the Engi-
neer's  fee in  this case,  the Order directing this action
stated  only  "...that  Black &  Veatch be paid...". It did
not provide a general determination of the issue.  Rather
the Court reaffirmed during its hearing that all issues in-
volving disbursement of Company assets must come be-
fore an individual judgment,  an action which certainly
will encumber enforcement activities and make the ulti-
mate decision academic.

-------
                 TECHNICAL AND FINANCIAL ASPECTS
                OF CLOSURE AND POST CLOSURE CARE
                                         JOHN W. THORSEN
                                     Weston Designers/Consultants
                                       West Chester, Pennsylvania
INTRODUCTION

  Federal legislation and regulations have been developed
to protect  public health and environmental quality from
harmful discharges  of hazardous wastes. One of the spe-
cific requirements addressed by Federal law is the need to
provide closure and post closure plans with assurances of
financial capability to implement these plans. Federal law
requires that owners of hazardous waste management fa-
cilities provide for proper closure of. their facility. It also
requires that the owner  of the facility provide for post
closure care, at a site where the waste remains after closure,
for a period of 30 years. The purpose of these requirements
is to assure the  public that, if an incident occurs which
could force an operational hazardous waste management
facility to close, funds will be available to properly close
the facility and assure no environmental or health impacts
occur.
  The regulations also specify that adequate funding must
be available to  assure closure and  post closure care, as
needed, and that such  steps can  be  implemented  and
carried out with no financial burden being incurred by the
public.
  One aspect that is not addressed by this regulatory pro-
gram is post closure care after the owner's responsibility
has been terminated, i.e., what happens to a site after 30
years to assure environmental protection.
TECHNICAL ASPECTS

  Closure is  the period after wastes are no longer ac-
cepted, during which the owners or operators  complete
treatment, storage and disposal operations, dispose or de-
contaminate equipment and apply cover to or cap land-
fills. The purpose of the closure standards is to assure
that all hazardous waste management facilities are closed
in a manner that:
•Minimizes the need for post closure maintenance, and
•Controls,  minimizes or eliminates to the extent  neces-
 sary to protect human health and the environment, post
 closure escape of waste, leachate,  contaminated rainfall
 or waste decomposition products  to ground or surface
 waters  or to the atmosphere.  There are two types of
 closure and post closure  requirements in the Federal
 rules:
 A. General requirements which are applicable to all facil-
    ities, and
 B. Specific technical requirements which are included in
    the facility-specific regulations.
  The closure standards,  specifically  for the  plan and
cost estimate,  assures the owner has considered  the  ac-
tions  and costs that must  be taken to properly close  his
facility. The regulations specify timing for the develop-
ment  and submittal of closure plans.  Closure standards
require. that closure plans had  to be developed by May
19, 1981  for existing hazardous  waste management facil-
ities, or during the development of a plan of operation  for
new facilities. During the interim status period, plans must
be maintained  on-site. These plans need not be submitted
for regulatory agency review until they are submitted either
prior to closure or to obtain a final permit.
  If a facility is to close before a final permit is issued,  the
closure plan must be submitted to the EPA Regional Ad-
ministrator, or authorized State at least 180 days  before
closure is anticipated to take place so the regulatory agen-
cy can review and approve the document and so the requi-
site public participation, with regard to plan actions, can
be accomplished.
  For those facilities in existence on November 19, 1980,
the closure plan must be developed  for all portions of  the
facility that were active on that  date. If some areas were
closed after this date and before the development of  the
plan,  these areas must be properly closed in accordance
with the  closure standards. Partial closure is also  a con-
cept that can be utilized to reduce the financial require-
ments for facilities, generally landfills. This allows an own-
er to  sequentially close areas of his facility. By minimiz-
ing the  active portion  of the facility,  the necessary  ac-
tions needed to meet the closure requirements are less and,
hence, the costs needed to implement the plan are less.
  The regulations also require the closure plans be revised
when there is a change in the plan, including changes in
activities, schedules or estimated cost for closing the facil-
ity. The cost estimates must be updated annually.
  The closure plan is the cornerstone of the closure regula-
tions. Closure  plans must address the following basic ele-
ments including:
•How and when a facility will be closed
•An  estimate  of the  maximum inventory of waste  on-
 hand at any given time
•Decontamination activities
                                                    259

-------
260   REMEDIAL RESPONSE
•Schedule of activities over the closure period
•Cost estimate to implement the plan.
  The schedule for closure must address, at a minimum,
three key dates:
•The final date of waste acceptance
•Dates for completion of specific activities
•The final date of closure completion.
  A schedule would naturally begin with a 180 day notifi-
cation prior to the beginning of closure. The second date
would be the beginning of closure. An important require-
ment is  all treatment and disposal of wastes on-site, must
be accomplished within 90 days of closure. Closure must
be complete within 180 days of the acceptance of the last
waste shipment.  This  180-day requirement may be waived
if it can be shown that, realistically, more than one year is
needed. An  example of this situation would be closure  of
surface  impoundments where stabilization of the  sludge
must take place before cover and vegetation can be placed.
  The final date of completed closure is  the date, after  all
activities have taken place, upon which separate certifi-
cation can be made  by an independent registered  pro-
fessional engineer and by the owner that closure has taken
place in accordance with  the closure plans, and is com-
plete.
  There are facility-specific requirements outlined in the
Federal  regulations and these  are  further  addressed  in
guidance documents  prepared  by the U.S. EPA. These
documents explain the facility-specific  requirements and
suggest  methods for meeting them. They also provide out-
lines for the development of closure plans to meet the gen-
eral and facility-specific requirements.

Post Closure Care

  While closure requirements impact all hazardous waste
management facilities, post closure care requirements im-
pact those only where the wastes remain after closure,
i.e., hazardous waste disposal facilities which include land-
fills, land application  facilities and seepage facilities.  Post
closure  is the period  after closure during which owners,
or  operators of disposal  facilities, must conduct certain
monitoring and maintenance activities.
  Post  closure  care  consists of,  at  least, groundwater
monitoring and  reporting in  accordance with the regula-
tions and maintenance of monitoring and waste contain-
ment systems. In addition, facility security provisions may
be  required  to remain operable during the post closure
care period or a portion thereof. The circumstances, dur-
ing which security would be required, is where an oppor-
tunity exists  for the accidental encroachment on the facility
by people, or animals which could result in adverse health
or environmental impacts.
  The post closure care period  begins upon the  certifica-
tion by an  independent  registered  professional engineer
that closure  has  been completed.  No agency action is  re-
quired to initiate the post closure care period. As with the
closure  requirements, the post  closure care plan must  be
submitted 180 days before closure is anticipated to begin.
  Post closure owner responsibility extends for a  period of
30  years.  Activities must  be  structured  over this 30-year
period to assure there  are  no adverse impacts  resulting
from the site on human health and  the environment.
There is a provision allowing application to the appro-
priate  regulatory  agency  for a  shortening  of the 30-
year timeframe.  Criteria have been established to evaluate
if foreshortening the post closure care  period would still
continue to provide  for no  adverse human health, or en-
vironmental impacts.
  One of the methods EPA has used to attempt to assure
the facility will  be safe after post closure owner respon-
sibility has terminated, is land use controls consisting of a
notice  to the local  land authority, generally a zoning
board. This notice must  include a  survey plan  with the
dimensions  of the landfill cells or other  disposal facility
specifications with respect to permanent benchmarks; it
must be submitted within 90 days after  closure is com-
plete. This survey plan must be developed by a registered
land surveyor and must  include  the type, location and
quantity of hazardous waste disposed  of in the facility.
In addition, the  owner must record a notation on  the deed
for the  property that will notify any potential purchaser
of the past use and current and future  use restrictions
for the facility as a result  of the activities  that have taken
place at the facility.  The  regulations require post closure
use of the facility cannot disturb the integrity of  the  final
cover, liners or other components of the containment sys-
tem.
  As with the closure, the post closure plan is the corner-
stone of the post closure  requirements. Its basic  purpose
is to assure the owner has considered the actions and the
associated costs, to provide for post closure care.
  Some  of the  plan development requirements  are the
same as for post  closure plans: the how and why must
be specified, the activities and frequencies  to assure appro-
priate post closure care must be identified, a schedule must
be developed, and cost  estimates must be provided.  Plan
revisions are allowed and, in fact, needed where the sched-
ule activities, or  cost  estimates change.
  The  plan describes the basic activities that must  take
place during post  closure care. These include monitoring
and  maintenance activities.  Monitoring  of groundwater is
the primary monitoring requirement. Continuation of the
groundwater monitoring system utilized during the active
life of the facility must be continued, sometimes  at a less
frequent rate. In  addition,  inspection  and  maintenance
activities must be conducted to assure the integrity of the
facility. Specifically, the  following activities must be in-
cluded in the inspections and maintained if the need arises:

•Run-off control structures
•Settlement (through maintenance and use of benchmarks)
•Erosion damage
•Gas control facilities
•Leachate management facilities
•Groundwater monitoring system

   After the activities and schedules have been established,
cost estimates must be prepared  to implement the  plan.
Ground rules for cost estimates are briefly addressed in the
section on Financial  Aspects.

-------
                                                                                REMEDIAL RESPONSE   261
FINANCIAL ASPECTS

  Four topics are discussed below under the financial as-
pects of closure and post closure care. These include:
•the basic rules for estimating the costs of  the activities
necessary to implement the closure and post closure plans;
•the financial mechanisms that can be utilized to meet the
financial responsibility requirements;
•the timing for  implementation of this RCRA require-
ment and,
•the current regulatory situation regarding the provision of
financial assurance to the regulatory agencies.
  In developing  estimates to implement the actions out-
lined in  closure  and  post closure plans,  four different
techniques can be used:
•The experience of the owner or operator
•Contractor estimates of the actions
•Cost estimating handbooks
•Work-ups from labor, materials and equipment require-
ments.
  Experience to  date has shown owner/operator  exper-
ience generally provides the least costly of  the different
techniques.
  As indicated above, the estimates to implement the plans
are made in current year dollars and must be updated to
account for inflation. These updates must be annual and
for this purpose a standard tool to gage inflation over the
past year has been selected by the  U.S. EPA. This  is the
"Implicit Price  Deflator  for  Gross National  Product"
which is developed by the Bureau of Economic Analysis
in the U.S. Department of Commerce. This document is
published monthly, and is available from the Bureau of
Economic Analysis.  Further, since it is a regulatory re-
quirement, U.S.  EPA regional personnel  should have
ready access to this information.
  The actual index is developed  in a ratio manner by
dividing the deflator  for  year "X" by the deflator for
year "X-l". This ratio is  then multiplied by the estimate
to result in a revised total dollar figure needed  for the
accomplishment  of the closure and  the post closure re-
quirements. The  financial assurance mechanism must then
be modified to assure that adequate funds are available to
implement the closure and post closure plans.

Basic Cost Estimation Rules

  Basic rules for  estimating closure costs  are outlined
below:
•Costs must be  based on the activities described  in  the
  closure plan
•The estimate must be made when the closure costs  are
  most expensive, i.e., when the maximum extent of opera-
  tions  is the greatest and  there is the largest anticipated
  on-site inventory of waste materials
•Costs can be based on the owner or operator conducting
  the closure activities which allows the owner/operator to
  take advantage  of the depreciated  value of equipment
  used in closing a facility
•All pertinent costs must be identified and  included in the
  cost estimate
•Costs are estimated  in the year of preparation and in-
 flation factors are considered on an annual basis using
 the index described above.
  The ground rules for estimating costs for post closure
activities are only  slightly different  from those  in  the
closure sequence. The largest single difference is the cost
must  be  based  on  contracting for  the provision  of
services needed to implement and carry out the closure
plan. Other ground rules include:
•Cost must be based on the activities described in the post
 closure plan
•Cost  must reflect the care  for the  total area of haz-
ardous waste containment
•Cost  must be complete and  include  all the  pertinent
 costs
•Cost estimate must cover the entire 30 year time period
 required for post closure owner responsibility.

Financial Mechanisms

  As  required by the  hazardous waste regulations, own-
ers  or operators  of hazardous waste management facil-
ities, must establish financial assurance for closure and,
where  necessary,  post closure  to assure the plans can be
implemented at any point in time and the funding needed
for the plan implementation will be available. The follow-
ing options are  available for providing financial assur-
ance for closure and post closure care:
•Trust fund
•Surety bond guaranteeing performance
•Stand-by letter of credit assuring funds
•Combination of the above.
  These  methods are outlined  specifically in  the Fed-
eral Register,  January 12, 1981.(5) States' requirements
may vary from the Federal requirements, but they must be
at least as stringent  as the Federal requirements in provid-
ing financial assurances. Examples of the mechanisms are
also provided in the January 12, 1981, Federal Register.

TIMING OF REGULATIONS

  The regulations were scheduled to be effective on July
13, 1981. However, because of concerns raised regarding
the limited  range of  mechanisms promulgated by U.S.
EPA in January,  the U.S. EPA has extended the  comment
period and,  subsequently, extended the effective date until
October 13, 1981.
  Basic concerns regarding the mechanisms promulgated
in January, centered  around the lack of a self-insurance
mechanism  and a financial test mechanism  to  meet the
financial assurance, and liability requirements as required
by the law.  The concerns expressed by the industrial com-
munity were very specific. The U.S.  EPA has  indicated
it is looking into,  and most likely will  develop, mecha-
nisms  for self-insurance and for a financial test so that a
firm can provide for  self-insurance to meet the financial
assurance, and liability requirements. EPA agreed there
was no point in initiating and acquiring much more costly
mechanisms outside a firm if,  in fact, the ability to insure
oneself was going to be promulgated at some point in the

-------
262   REMEDIAL RESPONSE
future. There are other concerns regarding the need for
separation of, or duplicate closure and post closure mech-
anisms, as well as the ability for a corporate entity to pro-
vide financial assurances  for  multiple sites  using  one
"umbrella" mechanism.
  One of  the basic underlying concerns regarding the
mechanism selected and utilized under this program, re-
volves around tax considerations. If EPA allows insurance
to guarantee closure  and  post-closure funds, then the
funds set aside through insurance premiums would be tax
deductible  as business expense. However, if these funds
were placed in a trust, there would be tax liability to this
money. The regulated community sees this as being an un-
desirable situation since they would  be taxed  for money
they could not utilize in their day-to-day business.
  U.S. EPA  anticipates  the issue will be resolved in  a
fashion so the revised regulations can be published before
the October 13, 1981 deadline.

CONCLUSION

  The depth and breadth of events that must take place in
this one  sub-rpogram of the hazardous waste program
illustrates the complexity of the entire program.  Hence,
great care must be taken in analyzing and meeting the re-
quirements specified by the State and Federal programs.
  The basic purpose of the Federal requirements  for haz-
ardous waste sites is to show that the owner or operator
has adequately  considered,  and planned for the  closure
and post  closure  technical and financial requirements.
The technical requirements are needed for both  the in-
terim  status standards, as well as for obtaining a RCRA
permit.  The  financial  assurance requirements are also
needed for both  interim status and RCRA permits; how-
ever, the implementation has been delayed until at least
October 13, 1981.
  The importance of careful cost estimates for closure and
post closure care cannot be over-stressed.  These  require-
ments can have a direct financial impact on a firm and,
therefore-,  must  be analyzed in conjunction with  the
closure and post closure plans to minimize exposure.
   As specific as these  rules appear to be, because each
 facility is unique, there is a capability for innovative, and
 creative rule interpretation to  allow for minimization of
 exposure under  these  regulations.  Such  decisions may
 often mean the expenditure of  capital costs as opposed to
 long term operation and maintenance costs. Cost analysis
 should be readily employed to  assure the most cost effec-
 tive long term solution for closure and post closure.
 REFERENCES

 1. Burt, R.E., et al., "Draft Guidance for Subpart H of
   the Interim Status Standards for Owners and Operators
   of Hazardous Waste Treatment, Storage and Disposal
   Facilities,   International  Research  and  Technology
   Corporation, Aug. 29, 1980.
 2.  "Hazardous Waste Management System;  Standards
   Applicable to Owners and Operators  of Treatment
   Storage and Disposal Facilities; and Permit Program,"
   Federal Register, 46, No.  24, Feb.  5,  1981,  11126-
   11177.
 3. "Interim Status  Standards  for Owners and  Operators
   of Hazardous Waste Treatment,  Storage and Disposal
   Facilities, Federal Register,  45, No. 98, May 19, 1980,
   33232-33258.
 4.  Severn, R.R., et al.,  "Draft Guidance  for Subpart
   G of the Interim  Status Standards for Owners and Op-
   erators of Hazardous Waste Treatment, Storage and
   Disposal Facilities," International Research and Tech-
   nology Corporation, Oct. 6, 1980.
 5.  "Standards Applicable to  Owners and Operators of
   Hazardous Waste Treatment, Storage  and  Disposal
   Facilities;  Consolidated Permit Regulations," Federal
   Register, 46, No.  7, Jan. 12,  1981, 2802-2882.
6. "Standards  Applicable to Owners  and Operators of
  Hazardous  Waste Treatment,  Storage  and Disposal
  Facilities; Consolidated Permit Regulations," Federal
  Register, 46, No.  Ill, June 10, 1981, 30624.

-------
 PRACTICAL CONSIDERATIONS FOR THE PROTECTION OF
PERSONNEL DURING THE GATHERING, TRANSPORTATION,
STORAGE AND ANALYSIS OF SAMPLES FROM HAZARDOUS
                                        WASTE SITES
                                       GORDON A. ALLCOTT
                                      ROBERT VANDERVORT
                                          Radian Corporation
                                          Salt Lake City, Utah
                                          JOHN V. MESSICK
                                          Radian Corporation
                                             Austin, Texas
 INTRODUCTION

   Uncontrolled hazardous waste sites can present a spec-
 trum  of potential  health,  safety and environmental
 hazards.  Detailed information  regarding  each  site is
 necessary before comprehensive plans can be developed
 and safely implemented to mitigate hazards presented by
 the uncontrolled waste. Acquisition of detailed site spe-
 cific information can involve serious potential risks.
   In this paper, the authors present practical suggestions
 for the protection of investigators in exploring, sampling
 and evaluating uncontrolled hazardous waste sites. Also
 considered are factors related to protection of the public.
 First a generic approach to hazard control planning is pre-
 sented. This is followed by discussion of a composite case
 study based on real experiences in several waste explora-
 tion and analogous hazard situations.

 GENERIC APPROACH TO
 HAZARD CONTROL PLANNING

   Occupational  exposures  to  hazards  associated with
 waste site exploration and evaluation can be controlled or
 avoided. Success  in prevention of adverse exposures  de-
 pends on detailed_planning, training and execution follow-
 ing principles of industrial hygiene, safety and occupa-
 tional medicine. The discussion which follows presents a
 logical sequence  of  assessment and  planning activities
 which has been developed and applied in actual waste site
 projects.


 Preliminary Hazard Assessment
   The first step in assessing potential  hazards at  a waste
 site is to collect  and evaluate all known and available
 information  regarding the site.  Each   site  presents a
 unique matrix of risks.  Among the important variables
 which define relative risk are: (1)  materials deposited at
 the site, (2) type and integrity of hazardous material con-
 tainers, (3) hydrogeological and topographical features of
 the site  and  (4) the  period during which deposits were
 made at the site.  To help assure that  all useful informa-
 tion is acquired, a list of informational categories (Table I)
with some specific examples has been developed. In prac-
tice, Table I can be used as  a  basic checklist  during
documentation, search and site visit activities.
  The second step in preliminary hazard assessment is the
identification of information  which  has relevance to the
specific site but is uncertain or unavailable. The items in
Table I can also be useful to  this  analysis. The objective
here is to identify all potential hazards or gaps in  known
information  which could reasonably affect safety in ac-
tivities to be conducted on or off the site. Caution must
be observed  to prevent undue biasing of this analysis by
facts which  are known about the site. For example, at
some sites a great amount of descriptive information may
be available. The importance  of missing information may
be difficult to appreciate.
  To avoid  overly conservative assessment of potential
hazards, a third closely related  effort should accompany
step two.  This effort should try to logically eliminate
potential hazards or data gaps from  serious consideration
when there is good reason to believe they have  no im-

                       Table I.
            Important Features of Uncontrolled
                 Hazardous Waste Sites
Waste Parameters
Sources/volume/form
 History of waste deposits
  Dates
  Sources of waste
  Type and quantity of waste
 Sources of additional information
Containment/confinement of waste
 Waste containers
  Types, age, condition
 Designed confinements
  Pits, lagoons, cells, trenches, cover
Uncontrolled practices
 Open dumping
 Open burning
 Flooding/evaporation/percolation

Hazardous properties
 Physical and chemical properties
  Flammability, corrosivity, reactivity
 Potential toxicity
  Acute, chronic
Site Characteristics
Topography
Geology
Hydrology
Climatology
Wildlife (reptile, animal, insect)
Ground cover
History of sampling/exploration
Accessibility
Security
Adjacent tenants
Related Considerations
Nearby population centers
Hospital facilities
Ambulance service
Fire district
Utilities available/proximity
Industrial equipment rental
Law enforcement
                                                  263

-------
264     SITE & PERSONNEL SAFETY
portance. For example, a site may be known to contain
only hazardous wastes from building demolition (e.g. as-
bestos) and junk transformers (e.g. PCB's). In this case,
organophosphate pesticides could be eliminated from con-
sideration since there is  no logical  evidence  that  they
would be present at the site. Similarly, serious considera-
tion of buried drums of flammable solvent and other un-
likely hazards could be rejected.
  The fourth  step in  the  preliminary hazard  assessment
is to obtain  descriptions  of the site exploration, waste
sampling and  other activities which are  proposed. A list
of several types of activities which may  be required in
exploring and evaluating a waste site prior to development
of remedial action plans  is given in Table II.  The initial
focus  of this  effort should be  the  equipment ancf  pro-
cedures to be employed on site. Next, this analysis should
consider offsite activities which will be  required to  sup-
port the proposed project. Transportation of waste sam-
ples to a laboratory  and  their analysis  are examples. It
is important that anticipated procedures be described to-
gether with the rationale for their selection.

                            Table II.
                 Site Reconnaissance, Exploration,
                Sampling and Evaluation Activities
 Surveying
 Plat, elevations, grid
 Water table elevation
 Metal detection
 Surface probing
 Marking, staking, flagging

 Access
 Timber and brush removal
 Surface grading
 Road construction
 Water pumping/diversion

 Security
 Fencing, barricades, signing
 Offshift site and equipment protection

 Sampling/Exploration
 Core/well drilling
 Excavation/trenching
 Drum/container sampling
 Gathering of soil, waste, water, air,
  vegetation, biological samples
 Disposal of excess sample material,
  washings, etc.
Sampling Handling/Analysis
 Identification and logging
 Packaging
 Storage
 Aliquoting
 Extraction
 Filtration
 Chemical analysis
 Waste disposal

Inspection/Maintenance
 Personal protective items
 Support system (compressors, gen-
  erators)
 Hygiene facilities
 Exploration equipment

Shipping/Receiving
 Labelling, packaging and shipment per
  DOT regulations
 Items handled
  Equipment
  Supplies
  Samples
  Clothing
  Wastes
  The fifth  step in the preliminary hazard assessment  is
the simultaneous consideration of potential hazards and
the activities proposed to be conducted on and off site.
The purpose of this exercise will be to identify situations
where hazardous exposures of research  personnel, sub-
contractors  or the  public may occur. The interface be-
tween each activity  and each known or accepted potential
hazard must be considered. Prediction of both "probable"
and "worst   case"  exposures  is  very important to this
analysis.  Examples  of  several  "worst  case" incidents
which could  occur both on and off the site are presented in
Table III.
  An attempt should be made to describe the logical con-
sequences of uncontrolled exposure incidents.  Could a fire
or explosion occur? Could personnel be exposed to high
airborne concentrations  of  toxic  contaminants?  Could
personnel be splashed with toxic or corrosive materials?
These are examples of questions which require probable
and worst case answers. Thoughtful organization of data
from this step in the analysis can  help  to highlight im-
portant areas of uncertainty and high hazard situations.

Exposure Control Program Development

  Having identified  probable  and  worst case exposures
which may occur in work on or off the site, plans can be
formulated for protection  of personnel.  Experience has
indicated that a broad range of protective measures may
be  applicable to different sites  or  to distinct activities
planned for a single site. For each waste site scenario, the
objective is to provide a program of protection which of-
fers a reasonable margin of safety to exposed employees.

                        Table ID.
              Examples of Worst Case Incidents
Fire and/or explosion of flammable or combustible solvents or pesticide
 mixtures
Explosion of waste containers containing shock, pressure or heat sensi-
 tive materials
Penetration or rupture  of compressed gas cylinders (buried or at the
 surface) containing toxic materials
Penetration of protective gear by toxic liquids, gases or vapors
Penetration of protective gear by equipment movement, flying debris, or
 contact with sharp objects
Interruption or contamination of supplied breathing air
Excavation and surface cave-ins
Equipment rollovers
In transit leaking or rupture of sample containers
Rupture or leakage of sample containers while in storage
Violent reaction of waste samples with analytical reagents
Medical emergency in hazardous area (e.g. heart attack)

  In waste site  work as in  other  occupational settings,
feasible engineering, administrative and work  practice
exposure controls should be applied to reduce reliance on
personal protective clothing and equipment. Contrary to
the opinions of some, the self contained  "moonsuit" ap-
proach to employee protection should not always be neces-
sary to provide a reasonable margin of safety.
  Job hazard analysis is a  useful approach  to identifying
potential exposures and required controls. Where serious
potential hazards  are  indicated, careful review of the
factors which dictate the  need for  maximum protection
should be made. In some cases,  extreme caution may be
dictated by the  absence of information  which could be
safely acquired with reasonable cost and effort.
  In other situations,  incompatibilities  may arise  from
inappropriate combinations of equipment, procedures and
hazards. Substitution of alternate methods (e.g. a change
in drilling equipment and procedure) or limiting the use of
a method (e.g. no drilling at coordinates where metal de-
tectors indicate the  possible presence of buried drums
and cylinders) may reduce or  eliminate  these incompati-
bilities.  In some situations,  freedom to  substitute  tech-
niques may be limited by legal or contractual constraints.
However, where improved safety may result, the impact of

-------
                                                                           SITE & PERSONNEL SAFETY     265
                                                     Table IV.
                              Sample Schedule of Personal Protective Items by Work Activity
        Work Activity
1
1. Initial survey of site
2. Metal detection survey
3. Site clearing and
preparation
4. Core drilling and core
handling
5. Mound exploration
6. Final survey of site
7. Rescue, repositioning
of support systems
|
X
X
X




X


*
*






*

X
X
X

X

X
X
X



*

X

X
*
X



*

X

X
*
X

X
X
X

X

X
X
X
•
	
X
X
X




X


I 	

*

X

X
*




*

X

*

X

   'Clothing or equipment which may be required based on site conditions.
these constraints should be discussed with the sponsor-
ing group so that reasonable compromises can be achieved.
  When personal protective clothing and equipment are
necessary to supplement  other exposure controls,  they
should  be  selected  in a logical, objective  manner.  De-
cision logics to  select respiratory protection devices are
available from the National Institute for  Occupational
Safety  and Health  (NIOSH),  the American  National
Standards Institute  (ANSI), respirator equipment manu-
facturers and others. Similar decision logics are becom-
ing generally available for  other forms of  protective
clothing and equipment (e.g. gloves, boots, body cover-
ings). Suggestions for protective clothing and equipment
ensembles usage in common waste site work are given in
Table IV.
  Whenever heavy reliance is placed on protective clothing
and equipment,  it must be remembered that  no protec-
tive system is one hundred percent effective. A number of
probing questions must be answered. Will the scheme of
protection afford adequate safety in the event of prob-
able and worst  case circumstances?  Is  the information
being sought worth the risk?  How will erosion  of the
Planned safety margin be detected?
  In addition to selection and use of protective clothing
and equipment,  a variety  of support activities and fa-
cilities are necessary  to  complete the hazard abatement
plan. Categories of support items which may be required
at specific sites are given in Table V. Fortunately, many
field  applicable emergency abatement  and personal hy-
giene support items are now commercially available. Ex-
amples  would be  portable  chemical  toilets and hand
washing stations and  portable eye-wash fountains. These
and  other items can  be  relied on for control of  routine
work hazards and minor emergencies.
  Hazard control planning should also include provisions
for major accidents. Emergency plans should address seri-
ous  injuries, fires, explosions, spills,  releases and other
adverse events. A list  of  several elements which should be
included in emergency planning is given in Table VI.

Control Plan Implementation
  Implementation of  hazard control plans should  involve
a variety of training and rehearsal sessions. Additionally,
physical examinations, clinical tests and other procedures
may be  required. A list of pertinent training and examina-
tion items is found in Table VII.
  On-site, full-time safety and health supervision is highly
recommended where  significant  potencial hazards exist.
No hazardous work  should commence until all  hazard
control  systems and facilities are set up and checked out.

-------
266     SITE & PERSONNEL SAFETY
                             Table V.
       Site Exploration Support Facilities Equipment and Services
                                   Emergency Services
                                   Fire department
                                   Ambulance
                                   Medical care
                                   Law enforcement

                                   Utilities
                                   Electricity
                                   Potable water
                                   Telephone
                                   Sanitary sewer

                                   Site Security
                                   Fences/gates
                                   Lighting
                                   Guard station
                                   Security officers
                                   Intruder alarm

                                   Other
                                   Freight pickup/delivery
                                   Industrial equipment rental
                                   Fuel delivery
Facilities
 Chemical toilets
 Handwashing station
 Protective clothing and
  equipment storage
 First aid station
 Lunch tables/benches
 Temporary office
 Lockers
 Water coolers

Equipment
 Emergency shower(s)
 Eye-wash fountain(s)
 Communications equipment
  Telephone
  Public address system
  Siren/horn
  CB radio
  Voice actuated tape recorder
   and radio communication
  Fire alarm
 SCBA units
 Stretcher
 Electrical generator
 Air compressor
 Clothes washer
 Clothes dryer
 Meteorological station
 Water pumps
 Fire extinguisher(s)
 Each  site exploration procedure should be  tested  in a
 no hazard area with the scheduled hazard control measures
 in  effect. Any problems encountered should be satisfac-
 torily resolved  before work is performed  on  the waste
 site. Work on the site should proceed from areas of rela-
 tively low hazard to areas of greater hazard.

 COMPOSITE CASE STUDY

   This section describes practical lessons which have been
 learned in waste site  and other projects involving analo-
 gous work situations. The information gained from  sev-
 eral projects  has  been combined to  form a  composite
 case  study so  that a broad range  of experience  can be
 discussed as  well  as protecting  provileged  information
 regarding specific waste sites.
 Site Description
   The hypothetical waste site  to be evaluated is  located
 approximately ten  miles  outside a small industrial center
 in  the western United States. The total  site encompasses
 320 acres of basically level land. A marsh extends  from
 within the  site to the east.  A  drainage  ditch defines the
 northern boundary of the property. Vegetation at  the site
 consists of grasses,  reeds, bushes to 4 ft  and trees to 40 ft
 high and 0.5 ft in diameter.
   Waste disposal at the site began in the 1930s. Industrial
 development accompanying World War II increased use of
 the site  during the  1940s. During the period 1940 to 1975,
 the local municipality and  several  nearby industries in-
termittently used  the site. One  local  company provided
records of waste which they deposited there. Other com-
panies used the site at various times; some of these are no
longer in business.

                      Table VI.
             Elements of Emergency Plan
Anticipated emergencies
 Weather extremes
 Fire and explosion
 Work accidents
 Medical emergencies
 Civil disobedience/unauthorized
  entry to site
Established procedures/
responsibilities
 Communications
  Weather service
  Fire department
  Hospital/industrial clinic/
   physician
  Ambulance service
  Law enforcement authority
Rescue/extrication of injured or
ill
 Supervisor and employee action
 Outside support
Fire suppression/control
 Supervisor and employee action
 Outside support
Medical care
 First aid
 Outside medical support
Site evacuation
Site security
Evacuation of adjacent properties
                         Table VII.
            Examinations, Training and Rehearsals
Examinations/Histories
 Hearing
 Vision
 Pulmonary function
 Medical history
 Employment history
 Physical examination
 Clinical tests
Training/Rehearsals
 Emergency and routine communications
 First aid (CPR, EMT)
 Respiratory protection use, limitations, inspection, maintenance
 Emergency equipment (deluge shower, eye-wash)
 Rescue of ill or injured
 Site evacuation
 Use, limitations and care of protective clothing and equipment
 Site surveying, sampling and exploration procedures
 Sample identification, packaging, storage, shipment
 Laboratory procedures
 Disposal of hazardous materials
 Fire fighting

   Physical evidence indicates that  wastes have  been de-
posited in scattered  locations on approximately 100 acres.
Waste has  been buried  on dry ground and submerged in
the marsh  to  depths of 15  ft.  Liquid  wastes have  been
discharged into pits for  evaporation, burning and/or per-
colation. Drummed waste litters an  0.5  acre area.  It is
suspected that similar waste has been piled on the site and
covered with earth to form mounds.

Work Assignment

   The  site will be  evaluated in phases.  Results of  each
phase will  be  considered in planning successive phases.
Phase I work, discussed here, involved the following  tasks
and information:
   (1)  Location of all waste disposal areas
   (2)  Core drilling  of dry land  waste disposal areas and
       adjacent ground using a 50 ft grid pattern

-------
                                                                          SITE & PERSONNEL SAFETY    267
  (3) Core sample analysis by RCRA procedures
  (4) Sampling and analysis of groundwater and marsh
     water
  (5) Excavation of waste mounds with photographs and
     description of their contents
  (6) The small area where drummed waste resides on
     the ground surface is to be roped off and left un-
     disturbed.
Project Chronology
  The following major activities were performed:
  (1) An interdisciplinary project team was assigned and
     project organization established
  (2) Available information was reviewed and a list of de-
     sired information prepared
  (3) A preliminary  reconnaissance  visit to the site was
     conducted including  discussions  with several waste
     generators
  (4) A comprehensive plan to perform the work assign-
     ment was prepared including  an integrated health
     and safety plan
  (5) Required equipment, clothing,  etc. were assembled,
     fitted and tested
  (6) Project employees were examined, trained and re-
     hearsed in planned activities
  (7) A project base was established at the site
  (8) Planned activities and safety  procedures were re-
     hearsed with all project personnel including sub-
     contractors in clean areas at the site
  (9) Preliminary and indepth site work was performed
 (10) Core and water samples were shipped to the labora-
     tory
 (11) Laboratory analyses were performed
 (12) A technical report was prepared.
  In the remaining sections of this  paper,  highlights of
safety and health activities are presented.

Site Reconnaissance
  The site was observed and photographed  from undis-
turbed areas  surrounding the waste site and from well es-
tablished roads through the waste disposal area.  Recon-
naissance personnel  wore  impervious  boots and  gloves,
eye protection,  safety  helmets, and  half-facepiece air-
purifying respirators while in or near waste disposal areas.
No samples of waste or water were gathered or any waste
disturbed: Detector tubes were used to indicate the pos-
sible presence of classes of vapors and gases.
Project Work and Hazard Control Planning
  Project team members including safety and health per-
sonnel worked together to identify  potential hazards,
formulate plans for conducting required work, and inte-
grating safety and health considerations into all phases.
Key elements of the comprehensive safety and health plan
were:
  (1) Job hazard analysis  for  all project activities with
     identification of activity specific control measures
  (2) Consideration of  routine and  worst  case exposure
     possibilities  and  formulation of procedures  for
     normal work and emergency action
  (3)  Specification of field and  laboratory support  ser-
      vices and equipment, facilities and safety supervision
  (4)  Purchasing, assembly and  testing of all protective
      systems before traveling to the site
  (5)  Medical examination and training of in-house  and
      contractor personnel prior to waste site work
  (6)  Establishment of a base facility at the site complete
      with safety, hygiene and emergency systems prior to
      initiating site work
  (7)  Full-time safety and health supervision of all work
  (8)  Rehearsal of all site exploration and sampling work
      in clean areas before entering contaminated areas
  (9)  Rehearsal of all sample gathering, packaging, stor-
      ing,  aliquoting,  analysis and waste disposal  pro-
      cedures prior to handling real samples
 (10)  Rehearsal of emergency actions
 (11)  Post medical examination and debriefing of all pro-
      ject personnel

Description of Base Facility
  A base facility was established on an access road upwind
and equidistant to most areas of  waste exploration work.
The facility included a trailer-mounted emergency shower,
hygiene facilities, stores of protective clothing and equip-
ment, rescue equipment, first aid  supplies, portable power
generators and communications equipment. Lunch tables,
chemical toilet, equipment cleanup, inspection and repair
facilities  were also provided. Direct communication lines
and message relay systems were established with the local
hospital, fire department,  ambulance service and  private
emergency assistance group. The base facility was manned
by an experienced safety  and health  professional. This
person had authority to stop work at any time if unac-
ceptable risks or failures arose.

Preliminary Survey of Site
  Prior to sampling, core drilling and  excavation of the
site, a preliminary survey and mapping was performed
by employees working in pairs wearing impervious suits,
gloves,  boots, safety  helmets, eye protection  and  air-
purifying respirator. Metal detectors were  used to locate
buried metal  (possible waste drums). A narrow metal  rod
was used to probe for "drum graves"  or  other areas of
weak  surface  support. Later, no core  drilling would be
performed where metal was detected or infirm ground  was
discovered. These precautions helped  to reduce the  po-
tential for penetrating a  waste drum with the  core drill
and also potential upsets of the drilling equipment.

Core Drilling and Mound Exploration
  An independent drilling contractor was engaged to core
drill the  site. A two-man team was thoroughly trained in
all sampling and  safety procedures. A geologist sampled,
described and packaged  core  segments. During the ad-
vance into the site, drilling and retreat from the site these
men wore impervious  boots, gloves and air-supplied  hel-
met with vortex  tube cooler. Air-supplied helmets  were
selected  after consideration  of probable and worst  case
core drilling  incidents. This  analysis indicated only a re-

-------
268     SITE & PERSONNEL SAFETY
mote chance for exposure to contaminant concentrations
near an IDLH (immediately dangerous to life and health).
  Compressed air was supplied from a  gasoline fueled,
rotary  air compressor with 100 ftVmin capacity at 100
psig. The compressor was fitted with a high temperature
shutdown device. Inlet air was drawn from an elevated air
intake. Compressed air was regularly checked for compli-
ance with Class D breathing air requirements and continu-
ally filtered for removal of particulates and organics. Con-
densate was removed at  the compressor and at  multiple
drip legs in the pressure hose system. An air hose, 1 in.  in
diameter connected the compressor to two dual outlet pres-
sure/flow regulation and filtration units. Each helmet was
supplied with 100 to 200 ft of air hose (3/8 in. diameter).
  A microphone attached to a  portable recorder was in-
stalled inside the geologists helmet so that descriptions  of
core samples could  be tape  recorded. This microphone
was also connected to a two way radio. The on-site safety
supervisor maintained the air compressor, tested  air qual-
ity, maintained radio contact with the geologist and direct-
ed advance and  retreat activities. He also supervised dis-
robing and  hygiene practices. Air-supplied helmets and
support equipment were thoroughly inspected, cleaned and
repaired as necessary each day.
  The safety supervisor remained in sight of and in radio
contact with core drilling. Air horns were available for
signalling and alarm.  The base camp was equipped with a
self-contained breathing  apparatus  (SCBA) which could
be used  for rescue after calling for and confirming that
outside help was on the way. Back up  SCBA rescue as-
sistance could be on-site in less than 15 min.
  Mound exploration was  performed using the same en-
semble of protective clothing and equipment. A  backhoe
was used to open selected mounds from the upwind side.
Photographs were taken, descriptions recorded and over-
burden replaced.

Groundwater and Marsh Water Sampling

  After all core drilling was complete and slotted screen
casings installed, water samples were obtained from core
holes.  Depth to  water measurements were also made and
later combined  with  survey information to describe the
water table on the site. During water sampling from core
holes the same  protective  ensemble used in  preliminary
survey of the site was employed.
  Marsh water samples were gathered by wading into the
marsh  at the periphery of the waste deposit. Chest waders,
impervious gloves, goggles and half-facepiece air-purifying
respirator were used.  A safety harness and attended life-
line were also worn.

Sample Handling, Transportation, Storage,
Analysis and Disposal

  The  waste and  water  sampling and analysis  program
was planned, critiqued and rehearsed before any samples
were collected. Provisions  for documented chain of cus-
tody were built into the system. Laboratory space  for
sample  storage and  analysis was dedicated prior to site
work.
  Selected segments of core samples  were deposited in
numbered, one quart, glass jars with threaded, Teflon® -
lined closures. Similar jars were used for water samples.
After closure, jars were cleaned of surface contamination
and  additional identification attached.  Soil  and waste
samples were stored in a freezer on site. Water samples
were refrigerated.
  Sample jars were placed in protective plastic mesh and
packed  in ice chests filled  with vermiculite  and "Blue
Ice." Chests were  sealed and marked in  compliance with
Department of Transportation regulations. Samples were
conveyed to the laboratory by Federal Express or as truck
cargo with returning team members.
  At the  laboratory, technicians wearing gloves, apron,
goggles  and half-facepiece  air-purifying  respirators un-
loaded sample chests into freezers and refrigerators. This
ensemble of protection was worn during all work with the
samples in designated, restricted access laboratories. Later
individual samples were removed to a fume hood, opened,
an aliquot withdrawn for analysis and the remainder re-
turned  to storage. All extractions, filtrations, etc. were
performed in  laboratory fume hoods. Diluted wastes were
deposited  in  55-gal  drums  and  disposed as hazardous
waste.

CONCLUSIONS

  Each  waste site project has unique requirements for
personal protection. The basic approach to hazard control
planning presented here should have wide application. As
experience is gained in  this area,  more comprehensive
checklists  and decision logics can be developed to facili-
tate hazard control planning.
  The  need to train personnel,  rehearse  proposed work
and critique work practices and protective systems cannot
be overemphasized.  Well trained  personnel  appreciate
potential  hazards  and the limitations of protective sys-
tems. They will endeavor  to minimize exposure and chal-
lenge to their protective equipment.
  Field and laboratory supervision by qualified safety and
health professionals  is vitally important.  Both work situ-
ations demand full understanding of potential hazards and
protective options. Proper inspection and maintenance of
protective systems can only be assured by qualified per-
sons. Conditions at any site are subject to change. Meth-
ods to quantify changes in  potential  hazard are limited.
Persons who recognize the potential impact of changing
conditions and are willing to halt work  until proper ac-
commodations can be made are indispensable.
  Exploration of  uncontrolled hazardous waste sites is
essential to development of  remedial action plans. It can
be done safely with proper information, planning and ex-
ecution. Each waste  site experience can contribute to im-
proved methods if properly evaluated at its conclusion.

-------
                           SAFETY PROCEDURES FOR
                   HAZARDOUS  MATERIALS CLEANUP

                                      ROBERT W. MELVOLD
                                        STEVEN C. GIBSON
                                        Rockwell International
                                      Newbury Park, California
                                        MICHAEL D. ROYER
                           Municipal Environmental Research Laboratory
                                U.S. Environmental Protection Agency
                                          Edison, New Jersey
INTRODUCTION

  EPA and contractor personnel at hazardous substance
cleanup  operations face the risk  of direct chemical ex-
posure or exposure to  chemical-caused hazards such as
fire, explosion or oxygen depletion. Many of the perils of
entry, assessment and cleanup of hazardous waste sites
have been recognized and addressed by numerous authors,
but  a number of important safety topics have  yet to be
adequately covered.
  This paper describes  a project  that: (1) identifies spe-
cific hazardous substance cleanup tasks  that require im-
proved safety guidance and (2) provides monographs con-
taining the  best available safety guidance for selected
topics. Two of the safety monographs produced as a re-
sult of the project are summarized. The monographs are:
(1) air quality monitors and (2) medical surveillance for
hazardous materials cleanup personnel.

IDENTIFICATION OF SAFETY TOPICS

  The purpose of this project was to identify significant
safety topics related to hazardous materials cleanup that
had not been satisfactorily  addressed  by  previous  in-
vestigators or adequately covered in current or draft EPA
manuals. The topics identified were then reviewed and,
with the EPA Project Officer's  concurrence,  the most
worthy were researched and written up in the form of
monographs.  The approach employed  consisted  of  (1)
conducting  an  exhaustive survey  of  safety literature,
(2) evaluating EPA current or draft safety documents and
(3) soliciting comments from field personnel on potential
safety topics.

Safety Literature Survey
  An extensive literature survey was conducted on safety
procedures, techniques and equipment used in the control
and/or cleanup of hazardous  materials. In addition to
examining pertinent books, report, manuals and  other
printed  materials, there were  numerous personal com-
munications. Staff members of government agencies, trade
organizations, professional associations and private com-
panies, who are knowledgeable in the area of controlling
hazardous materials or cleaning up uncontrolled hazardous
waste sites, were contacted.
EPA Safety Documents Review
  Thirteen EPA draft and published safety guidance re-
ports  were reviewed  to  develop  a  preliminary list  of
safety topics that required additional attention (Table  I).
A comparison matrix was prepared for the documents as
part of the reviewing process and a summary report was
written. Three of the manuals are  clearly more complete
than the others:
  (1)  Personnel Protection and Safety Training  Manual,
      EPA Course No. 165.2 (A8)
  (2)  Hazardous Materials  Incident  Response  Opera-
      tions Training Manual, EPA Course No. 165.5 (A9)
  (3)  Hazardous Waste Site Investigation Training Man-
      ual, (A10).
  These manuals are  collections of a number of mono-
graphs,  reports,—sections of other manuals and ANSI,
OSHA/NIOSH, CFR documents as well as commercial
catalogs that were assembled for EPA training courses.
In fact, several EPA documents (or sections thereof) cited
in Table I can be found in Manuals A8, A9 and A10.
  In addition to having produced the safety documents
cited in Table I, the EPA is involved in other efforts to
establish sound safety procedures for use during hazardous
waste site investigations and hazardous substance clean-
up activities. The EPA signed a Memorandum of Under-
standing (MOU) in  December 1980  with  the USCG,
OSHA and NIOSH. The MOU objectives include develop-
ment of a comprehensive guidance manual  that will  es-
tablish procedures to protect workers involved in investi-
gating and cleaning up  hazardous  waste  sites and  re-
sponding to emergencies involving hazardous substances.
  Another EPA effort that will provide safety  informa-
tion and guidance for worker health involves the prepara-
tion of 29 monographs on "Technical Methods for  In-
vestigating Sites Containing Hazardous Substances." This
work is being sponsored by the EPA's Discovery and  In-
vestigations Branch, Hazardous Sites Control Division.
Final Selection of Safety Monograph Topics
  From the review of the cited  EPA safety manuals/
memoranda  and other available safety literature, a pre-
liminary list was assembled of safety topics to be empha-
sized in the study. The topics were:
•Air quality monitors
                                                  269

-------
270     SITE & PERSONNEL SAFETY
                                                        Table I.
            Summary of Draft EPA Manuals/Procedures Relevant to Safety Procedures in Hazardous Materials Cleanup
                       Document  Title
                                                        Prepared by
                                                                                          Status
          Al.   Agency-wide Policy Order 1440.1
               Respiratory Protection

          A2.   EPA Occupational  Health & Safety
               (OHS) Manual,  Chapters 1-7

          A3.   EPA OHS Manual,  Chapter 8, Lab-
               oratory Use of Toxic Substances

          A4.   EPA OHS Manual,  Chapter 9, Haz-
               ardous Waste Site Investigations
               and Emergency  Spill Responses

          A5.   EPA Safety Manual  for Hazardous
               Waste Site Investigations--
               Draft, 1979

          A6.   Oil & Special  Materials Con-
               trol Division's  Draft Interim
               Safety Procedures
          A7.   Field Health and Safety Manual

          A8.   Personnel Protection  and Safety
                Training Manual  for EPA Course
                No.  165.2
Office of Occupational Health
and Safety (OOHS)

AA for Planning & Management,
OOHS

OOHS
OOHS
OOHS and  National Enforcement
and Investigation Center
(NEIC)

Environmental Response Team
            at the request
of the  Superfund  Implementa-
tion Task Group

U.S. EPA  Region IV

U.S. EPA, Hazardous Response
Support Division  and the
National  Training and Opera-
tional  Technology Center
Sent out  for  review by EPA sub-
units,  10/80

Issued  to EPA employees on
September 12, 1977

Sent out  for  review by EPA sub-
units,  9/80

Sent out  for  review by EPA sub-
units,  9/80
                                Draft, 1979
Circulated  to various EPA sub-
units for comment on January
2, 1981
Published  October 1980

In use on  recurring basis for
training course
          A9.   Hazardous Materials  Incident
                Response Operations  Training
                Manual, EPA Course No.  165.5
          AID.  Hazardous Waste Site Investi-
                gation  Training Manual
          All.  Hazardous Waste Site Safety
                Procedures
          A12.  Memo  on Update of OOHS/ERT
                Activities
          A13.  Environmental Protection Agency
                FY-1980-81, Medical  Monitoring
                Program Guidelines
U.S. Environmental Protection
Agency, Hazardous Response
Support Division and the U.S.
Coast Guard

Field Investigation Team (FIT)
National Project Management
Office, Ecology and Environ-
ment, Inc.  under EPA Contract
No. 68-01-6056

Richard D.  Spear, Chief
Surveillance & Monitoring
Branch, Region II, U.S. EPA

Robert C.  Magor, Ph.D.,
Director,  Office of Occupa-
tional Health & Safety
(PM-273)

OOHS
In use on  recurring basis for
training course
Has been used  in training
courses
Memo of 5/21/80  that outlines
Interim HWS Safety  Procedures
This memo indicates level of
sophistication  that has been
achieved in OOHS  evaluation of
protective equipment  (5/10/79)

Appended to AID
•Medical surveillance
•Criteria for selection of protective clothing
•Personnel and equipment decontamination
•Index of operational and procedural manuals (to simplify
 location of appropriate safety information)
•Criteria for delineating the "hot line", and
•Odor detection for respirator canister failure (detection
 of an odor  typically signals either a leaking facepiece or
 failure of the adsorbent canister.)
  Based on  the results  of the survey of EPA and other
safety literature, it was  determined that a monograph on
air  quality monitors should  be produced and work was
               initiated. Due to time and budget constraints, some topics
               could not be addressed in monographs. To prioritize the
               remaining topics,  comments were solicited  from a selec-
               tion of EPA Regional personnel and private  contractors
               with hazardous material handling and/or clean up experi-
               ence and responsibilities. A synoptic matrix of their com-
               ments  and  suggestions relative to  the proposed mono-
               graph  topics appears in  Table II.  Two of the six sug-
               gested  topics—the index  of operational and  procedural
               manuals, and medical  surveillance—received a consensus
               endorsement.  Personnel and response equipment decon-
               tamination  received somewhat less  support. The remain-

-------
                                                                           SITE & PERSONNEL SAFETY    271
                                                    Table II.
                                        Responses to Suggested Safety Topics

Topic
1
2
3
4
5
6
7
Topics:
A
Has reservations
about training
people to use
odor detection as
method, Desensiti
zation is a prob-
lem. Avoids
problem by using
fit test and
fresh canisters.
Hot line varies
with wind, con-
containment. NFPA
guidelines are
sufficient. Use
of programmable
Q
May be too dan-
gerous ; not too
excited about
this idea.
Every spil 1 is
different so
Also approaches
by government and
private compan-
ies vary. Not too
be nice. idea .
Full range of Uses almost ex-
clothing is cur- clusively throw-
rently employed by away clothing.
TATs. Development Doesn't think
of criteria is a sophisticated
more difficult criteria is re-
than it is worth. quired for his
region.
Must search a Believes this
multitude of mate- to be a good
rial s; should idea . Could in-
avai lable. Sympa-
thetic with
idea.
Uses annual pub-
vice checkups
for response
personnel .
Sample requisi-
tion by EPA staff
can result in
equipment contam-
ination; depot
decontamination
is suggested.
Interested in
idea.
Would like an
and for specific
materials not to
be flushed with
water.
of OHM-TADS and
list by
chemicals .
Doubts a medical
tical . Personnel
would not pay
proper atten-
Not much in-
volvement with
decontamination
and therefore
not much need for
this item.
Would like to see
expenditure for
resources to re-
spond to spil Is.
C
No need for study
of odor detection ;
fool ish to train
people.
Coast Guard HACS
program can be
distances for hot
1 ine and evacua-
tion within half
hour of input . Thi s
fore, redudant .
EXE has program
with NIOSH to de-
velop cri teria ;
therefore, prob-
ably not neces-
sary.
Uses computer
tie-in to OHM-TADS
to get ful 1 [irmt-
Also, TAT has I/?"
document from
4UCTR117. It
might be OK to
develop an index,
thou<|h.
Considers this
that should be
investigated.
Uses soap/water
washdown. Also uses
disposable cloth-
ing, whenever pos-
sible. Believes
further research
of topic is
indicated.
Interested in repre-
ing deteriorating
drums safely.
1 - Odor Detection of Respirator Canister Failure
2 - Criteria for Delineating the "Hot Line"
3 - Criteria for Selection of Protective Clothing
4 - Index of Operational and Procedural Manuals
	 D 	 . E 	
Would like to see Foolish and imprac-
a cross-reference tical. S6me odors
chart or index for deaden the senses.
canisters. Not
excited about odor
detection.
Prefers to use
air moni toring
dispersion model-
ing can oe used for
spi 1 Is.
Assumes worst case
and uses "moon
suit" when hazard
unknown. Rarely
uses anything less
than splash pro-
tect ion , face
mask, hard hat and
Hyqienist makes
dec is ion on cloth-
ing protection.
Supports idea of
an index . Manual s
can be identified
various subject
tabula tions .
Bel ieves thi s
one.
Expressed no com-
ment on this
topic. (Propri-
etary considera-
tions may be
involved. )
Interested in
Also, likes info
on neutral ization
and gas/1 iquid
immobi 1 ization
for field opera-
tions.
test.
Has an SOP that
uses operational
parameters to prag-
matical ly del tneate
hot 1 me; not enthu-
puter program.
The currently used
4 levels of protec-
tion are adequate
not for cleanup.
Decision to down-
grade should be up
to hygipmst/
Only experience wi i 1
properly show the
advantages and dis-
many manual s .
determined by hygien-
ist/toxicologist for
easel ine and
fol lowup.
Kpcommends good med-
ical survei 1 lance
with depot-level de-
contamination and
No other topics
suggested.

r
Subjective; many
different odors,
many simi lar
Sensitization is
a factor. Some
chemicals do have
characteristic
odors ; perhaps
pilot study
called for.
Hot 1 ine is arbi-
trary in field; no
visible contamin-
ation oftentimes.
Safety manual for
s i te investiga-
teams . Not enthus-
iastic .
Uses clothing ap-
appropnate to
TLV, IDLH and
tors. Uses Noyes
chemical data
book , d repet i -
tion of CHRIS.
Considers present
techniques
adequate .
I ikes a modi f ica-
tion of this -
terse, brief ex-
manuals; abstracts
could be used.
tant, but not
often done, for

qui red.
Currently uses d
procedure for
scrub-down in con-
junction with wash
containment . Not
enthusvastic.
Woul d 1 i ke an
index for syno-
nyms and a cost
rate for disposal
of drummed wastes.
G
Not enthusi-
astic about
this idea.
Determination
of perimeter
should be
provided.
People do
not pay at-
tention to
clothing or
its selec-
tion until
it is too
late.
Johns Hopkins
University has
done some
work on this
subject.
Bel leves
search in
this area is
Bel ieves
this topic
requires
further
No other
topics
suggested.
Consensus
Negative
Mixed
Mixed-
Negative
Posi ti ve
Positive
Mixed-
Positive
5 - Medical Surveillance
6 - Personnel & Response Equipment Decontamination
7 - Others (to be suggested by interviewees)
ing topics received mixed reviews, except for odor detec-
tion of respirator canister failure, which received  a uni-
formly negative response.
  With the aid of the interviewees'  comments, the final
list of safety monographs was developed. The list includes:

  (1) Air Quality Monitors
  (2) Medical  Surveillance  for  Hazardous  Materials
     Cleanup
  (3) Selection  of  Personnel  Protective  Clothing  for
     Hazardous Materials Cleanup
  (4) Personnel & Equipment Decontamination
  (5) Index of Operational Manuals for Hazardous Ma-
      terials Cleanup
The "hot line" criteria topic received a positive response,
but  it was not  addressed by this  project  since another
EPA project* is investigating criteria for evacuation dis-
tances for hazardous  substances  derailment  incidents.
The "hot line" is directly related to evacuation distances
and it was  decided to address  the  topic  only once,  in
the other project.
*EPA Contract No. 68-03-3014, Task 10 (Phase II), TMS-2, is develop-
ing a Fire Control Handbook for Hazardous Substances.

-------
272     SITE & PERSONNEL SAFETY
  Two safety  monographs produced  by this effort  and
further described in abridged fashion in the following sec-
tions of the paper  are Air Quality Monitors and Medical
Surveillance for Hazardous Materials Cleanup.
AIR QUALITY MONITORS

  Combustible and/or toxic gases or vapors and oxygen-
deficient atmospheres may be encountered with lethal re-
sults if proper protection is not provided. Hazardous ma-
terials cleanup personnel must ascertain accurately and
efficiently what atmosphere conditions prevail in order to
take  safe and corrective action. This responsibility re-
quires that  response  personnel use portable monitoring
devices that are extremely reliable.
  A search  of available  product literature, however, re-
veals a bewildering array of different  air quality  moni-
tors, employing various principles of operation, packaged
in many sizes and shapes, with a host of operating char-
acteristics and selling  for a variety of prices. To assist re-
sponse personnel in selecting  an appropriate air quality
monitor, an evaluation of commercially available monitors
was conducted. The price of the  air quality monitors was
not a consideration in  the evaluation.
  Only those devices capable of monitoring  hazardous
gases or vapors on a continuous and  quantitative basis
were considered. Thus, detectors that indicate the presence
of  a substance  by  discrete  sampling were  eliminated,
whereas  monitors that possess  the  ability  to  quantita-
tively sense fluctuations  in gas  or vapor concentrations
over a period of time were included.
Evaluation

  A  desk-top analysis was conducted  of commercially
available  devices  for the  monitoring  of  combustible
and/or toxic gases or vapors and oxygen levels.  Thirty
basic design  configurations,  representing models  cur-
rently produced  by  18  different manufacturers, were
evaluated.
  The gas-phase monitoring devices were segregated into
three categories characterized by  the  gas system  being
monitored: (1) oxygen, (2) combustible gas or vapor and
(3) toxic gas or vapor (as represented by H^S and CO). The
toxic gas or vapor category was limited to H2S  and CO
monitors for two reasons. First, many toxic gas monitors
are capable of detecting numerous gases. A complete desk-
top evaluation and comparison of all monitor/gas com-
binations was considered  unworthy of the time and  cost it
would require. Second,  a fair comparison of  toxic gas
monitors could only be made between monitors for which
there was performance data on the same gas(es).
  Unfortunately, there was no gas for which performance
data were available for all toxic  gas monitors. However,
data was available for FL,S and/or CO  for each monitor,
so the toxic gas monitors were evaluated in two groups
depending upon whether  CO or RS monitoring data was
available  for the  instruments.  Those  instruments for
which performance data existed for both CO and H2S were
evaluated in both groups.
  Fifteen instrument parameters were  identified as the
most important for the intended use of the monitors. A
tabulation of the 15  specifications and qualifications, in-
cluding solid-state electronics and the degree of skill re-
quired for operation, for each combustible gas vapor mon-
itor are given in Table  III.  Similar tables were prepared
for the toxic gas/vapor  (IL,S and CO) and oxygen moni-
toring devices.
  The  evaluation was accomplished by  applying to each
specification or qualification a factor that relates the im-
portance of that particular parameter to the features de-
sired in the final product. Thus, rating criteria  were es-
tablished to reflect differences in the various devices and
weighting factors were assigned to each particular specifi-
cation  so that each device was rated in accordance with
the criteria. (See Table IV for evaluation rating criteria.)

Conclusions
  With the device specifications and qualifications identi-
fied and the evaluation criteria established, individual rat-
ings  were determined for each  device  specification. By
summing all the individual ratings for a particular device,
an overall  rating for the device was obtained. Overall
ratings for the top-rated devices  in each of the four de-
vice categories are as follows. Among oxygen monitors,
the Dynamation O2-25,  Energetic Ecolyzer 400 and Bio-
marine  OM325R all  received the  highest ranking  with
identical scores of 245. Ranked second with a score of 235
was the Gastech Model 1641.
  Among combustible gas monitors, the Century OVA-
128 was ranked highest  with a score of  258, followed by
the Dynamation LCD-Combo with a score of 240. Rating
totals in the H2S toxic  gas  monitor category show high
values  of 225  for the HNU's PI-101 and 214 for the In-
terscan Model 1170.  For the CO toxic gas monitor cate-
gory, the Enmet COS-100  was highest rated with 219,
while the Gastech EC-231 was second with 216.
  Presumably, any of the highly rated devices can be used
for the specific monitoring  application  for which it was
intended. However, it must be remembered that the desk-
top evaluation relied  almost exclusively on manufacturer's
product literature and on telephone conversations with the
manufacturers or their representatives. An independent ex-
perimental and/or field evaluation of the various moni-
tors would provide better data on which to base the evalu-
ation.  It is quite possible that monitor ratings developed
from first-hand  field evaluation  could differ from those
derived from  a desk-top  or literature evaluation. More-
over the assembled information is dated and therefore sub-
ject to change with time.

MEDICAL SURVEILLANCE FOR
HAZARDOUS MATERIALS CLEANUP

  A monograph has been produced describing a medical
surveillance program for hazardous materials cleanup per-
sonnel. The program is  a test protocol designed to super-
vise the health and well-being of the hazardous material
cleanup team members.  Medical surveillance programs for
hazardous  materials  cleanup personnel are directed  at

-------
                                                                         SITE & PERSONNEL SAFETY    273
                                                  Table III.
                             Combustible Gas/Vapor Monitors—Operating Specifications
Instrument* Vol umey Setup Temp . Opera t1 no
(Operating Weight Time Range Range (s) Accuracy
Principle) (in3/lbs) (win) {°C) (ppm) (X F.S.)
GasTech Inc.
Johnson
Instrument
Division
Ho del 1 314** 248/8 1 -8 to 0-1 001 +10
(CC) 43 LEL
Toluene
0-500 ppm
Organic
Vapor (OA)
Model 1641**'* 344/8 1.5 0 to 0-100% *5
(CC) 40 LEL
Methane
Dynamation 84/3 <2.5 -9 to 0-100% +2
Inc. LCD 49 LEL
Combo" (CC) Hethane

Enmet Corp. 224/6.5 10 -10 0-50*. *3
CGS-100*" to 50 LEL
(SS) Methane
(OA)
Century Sys- 425/12 1 0 to 0-10 ppm +1
terns Corp. 55 0-100 ppm
OVA-128 , 0-1000 ppm
(GC/FIOT
Infrared 227/9 5 0 to 0 100* +5
Industries, 50 LEL
Inc. IR-711 propane.
(IR)I1 0-1000 ppm
JP-5
International 167/5 0.5 o to 0-100; LEL *5
Sensor Tech- 40 Methane
nology
AG6000** (SS)
Scott Avia- 126/5 Instant 0 to 0-100'. LEL *10
tion/Div. of 43 Hexane
A-T-0 Inc.
Model 0-17 (CC)
Energetics 432/8 1 -18 0-lOOi LEL +1
Science/Oiv. to 40 Methane
of Bee ton
Dickinson
S Co. Eco-
lyzer 400**
(CC)
Appliances 40 Pentane
Co. Model
260" (CC)
Biomarine 561/9 5 < 5 15 0 100% LEL +5
Industries, to 40 Hexane
Inc. Model
902" (CC)
* All devices are of solid-state construction and require
bustion, SS • Solid State, GC/FID = Gas Chromatography/F
** Also equipped to monitor either toxic gases or oxygen.
"* Also equipped to monitor both toxic gases and oxygen.
/ This principle of operation is sensitive to the C-H bond
gases/vapors.
t A/V/F = Audio/Visual/Failsafe; Bat • Battery Check; Del
1 Not determined by independent assessment.
OA - Others available.
Minimum
Detection .
(I F.S.) Alarms'




< 1 A/V/F
(Bat,
Del)



2 A/V/F
(Bat,
Det)
1 A/V/F
(Sat,
Det)

<1 A/V/F
(Bat)


< 1 A/F
(Bat,
Det)
25 A/F
(Bat)



3 A/V/F
(Bat)


5 A/F
(Bat)


1 A/V/F
(Bat,
Det)




(Bat)

(Bat)


average skill on the
lame lonization Dete


and, therefore, the

Detector Check.


Intrln- Calibration
slcally Medium/
Safe Time (min)




5-10




Yes " Methane/
5-10
,
5-10

Yes Hethane/
5-10


Yes Methane/
5-10


JP-5/5-10



Yes Methane/
5-10


Yes Hexane/
5-10


Yes Methane/
5-10





5-10

5-10


part of the operator
ctor, IR = Infrared A


corresponding device




Response
Time,
Interferences (sec




cones, lead
vapors



Argon, 'hel ium 4


{filter will stop
most)
None (while de- 30
tecting total
comb. )

None (while 2
detecting total
comb . )

ing CH bond



H2, Methyl 20
Mercaptans


Poisons - sili- 5-10
ones, lead
apors

one (while de- 15
ecting total
omb. ) Poisons
ilicones, sili-
ates, lead
apors

cones , si 1 icates ,
lead vapors

cones, lead vapors


Operating Principle code
isorption.


will respond to most, but




901 Operating Remote
) Period (hrs) Readout



8 No





6 0-1 V

9 0-2 V


10-14 No



8 0-5 V


58 0-100
mv



10 No



4-6 No



10 No






8-9 No

°


: CC = Catalytic Corn-



not al 1 , combustible




employment qualifying, general health maintenance and
prevention or early detection of harmful effects of haz-
ardous substances. In the case of adversely affected mem-
ber^) of the cleanup team, this may require medical in-
tervention and removal from the cleanup team.<2) The ob-
jective of this monograph is to  describe and  document
the elements that  comprise a reasonable medical surveil-
lance program for hazardous materials cleanup personnel.
The described elements are based on state-of-the-art medi-
cal methodology and  the proposed medical surveillance
program is expected to be tailored to specific applications
as required by the user organization.  Also, medical sur-
veillance programs  should be  designed to  secure  maxi-
mum preventive  benefits  at minimal cost and  incon-
venience.
  A medical advisory board sponsored by the organiza-
tion responsible  for  the hazardous materials  cleanup
should have the  responsibility of developing  a medical
surveillance program  and evaluating  its results. The
board should be  empowered to make recommendations
based on  its review of the  examinations performed  on
personnel.
  Two examinations that are commonly performed'2' are:
•Preplacement
•Periodic
Preplacement/Pre-Employment Examinations
  Preplacement examinations serve an essential function
in health surveillance by providing an historical record of

-------
274     SITE & PERSONNEL SAFETY
previous exposures, information  on the state  of health
prior to joining the team and a baseline for comparisons
with  later health observations.  Preplacement  examina-
tions are used to ensure that workers are physically able
to use personal protective equipment.12' Employment and
medical history and a physical examination consisting of
physical  and biological  monitoring are elements  of the
preplacement examination and should be tailored to the
specific hazards of the job under consideration.
Periodic Examinations
  A periodic examination is a tool that can be used to
detect incipient disease,  physiological  changes, biochemi-
cal deviations or evidences of absorption of toxic agents
and to establish interim  reappraisals.<3) Periodic examina-
tions consist of periodic baseline examinations (compre-
hensive)  and  hazard-oriented surveillance  examinations
(limited).
  The periodic examination provides two primary services:
(1) maintains a continuing record of the general health
status of the team members  and  (2) provides an epidemi-
ological record, followup analyses and/or examination for
possible latent effects of exposure to some hazardous sub-
stance^).
  Periodic examinations would be similar to pre-employ-
ment/preplacement   examinations.  However,  the  em-
phasis would be on possible exposures that had occurred
since the previous examination. An example of the physi-
cal/biological  parameters  that  a  periodic examination
could  investigate'4' for possible exposure  of  a response
team member to selected hazardous substances is shown in
Table V.
  The hazard-oriented  surveillance examination is  de-
veloped to screen for exposures  that may occur during a
hazardous material cleanup operation, such as at an  un-
controlled hazardous waste  site.  Hazardous  materials
cleanup personnel may be exposed to toxic chemicals that
may or may not have been identified from the onset of the
response action.'2' Based on the premise that hazardous
materials  response team  members could be operating in
situations where  the hazards  are unknown, three points
should be made:

  (1) It is not economically feasible to test the  team mem-
      bers for  all hazardous substances  if they were  ex-
      posed to unidentifiable numbers and types of chem-
      icals in question.'2'
  (2) There are no general tests that can be used to screen
      for  organics or inorganics generically. The tests  are
      chemical-specific and the biological pathway  ex-
      amined would be dependent on the type of chemical
      in question.(2)
                                                      Table IV.
                                               Evaluation Rating Criteria
Vol ^/Weight Setup Time
in Ibs Solid State (mm) Sk
0-750 15
751-1500 13
1501-2250 11
2251-3000 9
3001-3750 7
>3750 5
£N '••
16-30 w
31-45 No
46-60
>60
Minimum Detection
°1
0-0. St 10
0.6-1. OS 5









Oxygen
None
Interferents
Interferents
Toxic i Combust
GC/FID
Combustibles
2'< LEL 4








20
•2500 ppm 10
* 2500 ppn- 0
ible Gases*
30
None stipulated, .ery
specific if
1R absorption
muroOroceSSO
(Used bv the
SO)

gas tknown.
with
r 30
HI RAN-


Temperature Range

Lower Upper Operational Accuracy
lill Limit (°C) Limit (°C) Range(s) (* F.S.)
IS 0-10 15 Average <-30 15 >70 >2 Ranges 15 < +2 20
11-20 10 Ability 10 -20 to -30 12 60 to 69 2 Ranges 10 +3 to +5 15
0 21-30 5 -10 to -19 9 50 to 59 1 Range 5 +6 to +10 10
31-60 1 Greater 0 to -9 6 40 to 49 > +10 5
>60 0 Technical >0 0 < 40 Add 5 if factory
Ability 0 allows options
for range(s).
Alarms
H-S and CO Audio
1 ppm 10 Fai Isafe:
1-2 ppm 6 Detector
•> 2 ppm 4 Battery







Interferences

1R absorption 22
Hydrocarbons with sim-
ilar frequencies to
the material being
mom to red
Colorimetnc 22
Strong reducing
agents
Ctic combustion 20
Si 1 1 cones , lead
vapors, (poisons.)
Sol id state 17
Marcaptans, H^
Intrinsi- Calibration Procedure
5 cally Safe Contained gas
5 Y ?fl unnecessary 14
Direct cylinder
5 . attachment to
5 N° ° monitor 12
Calibration gas
introduced wi th
bag and cyl inder 10
Cal ibration gas
introduced with
syringe and
cylinder 8
Response
Time (Sec)
Photo ion izat ion 10 0-10 30
Many organics (mate- 11-20 24
rials with lonization 21-30 18
potential equal to or 31-60 12
less than the energy 61-90 6
level of the UV source) >90 0
Electrochemical 5
Unsaturated hydro-
carbons, C?HrOH, H,,
NO ° £



Calibration gas
introduced with
syringe and
glass container
Cal ibration gas
introduced with
permeation tube
and instrument
Cal ibration
gases plus ex-
ternal computer
for computations
Operating
Period (Hr)
>16 25
12-16 20
8-11 15
4-7 10
- 4 5








Calibration
Time (min)
0-5 6
6 6-10 5


21-30 3




2
Remote Readout
2 Ranges 10
1 Range 5










   Because of tie «ide disparity in product information on Interferents. allowance has been made for Interferents generally associated with
   specific principle o< operation by subtracting tne effect of the Interferents from the manmum allowable (30) according to the following
   schedule   major single interferent. -5. small interferent family. -8;  large interferent family, -10.

-------
                                           SITE & PERSONNEL SAFETY    275
                     Table V.
Biological Parameters Affected by Selected Hazardous Substances4
s-
o
-p c
CO O
• I—
in co
-o. — .
i— c ai
ro <1) -p
c a.  S-
•i- "O
4-J — - Ol
ro S-
Q. E =!
13 rO CO
U X O
O LU Q.
O X
~^ U Hi

fO "O i—
U O ro
•1- •!- =3
-a s_ c
ai 01 c:
S: D. ro
Asbestos x x
Beryllium x x
Carcinogens (other x x
classified)
Lead x x
Ozone x x
p-tert-Butyl toluene x x
Styrene x
Toluene x x
Trichloroethylene x x
Vinyl Chloride x x




c
o
•r- O)
-P U
ro c
U ro
• r- E
<+- s-
•r- o
4_> 4 —
i- i.
QJ C1J
(_> >, D_
ro
C S- >>
ro 1 i-
•r- X ro
O c
•i- -P 0
co co E
>, CD O) r-
(— "^/ r~ -^
Q. LU O Q.
X XX
X XX
X

X
X XX
X X
X
X X
XX X
X XX










>^
en =-,
o en
i — O
0 r—
10 -P O
•r- >, -P
CO 0 >,
to >, <_>
E i— E

x c: -P c
LU 'r- 3 •!-
i- Q. i-
-Q rD oo ^
ro
-1
X

X X

X



X
X
X

01 ^
co O)
ro CO
f~ (O
-P (J •!-<_> C Ol
C '1— E T--I- CO
3 E ro E E ro
O ro t/l ro ro 4-^
CJ •• -p c -P CO ro
CO 3 ro 3 C -C
TD O) i — 1- i — ro D-
O -r— CD | — CD S- CO
OS- h- 0
r- -P E <-> E -C
CQ 00 I3-i- 3O Q_
•r- i- -P S- -r-
QJ E dj QJ d) i_ QJ
+J Ol 00 O OO O C
Oi -c > — - ro • — -.c: •!-
i— O O Q. ^~
D. |— r— (— oo ro
E "O O ro D- O -^
O O CD x CD -C i—
O O 00 O OO Q. ec
,^
CO


X XXX

X

X

X
X
XXX







o
E
rO

•r- Z3 O)
-0 1—00
3 CD ro
S- "O
•I- O •r-
i— E -P
•r- ro Q.
CQ CD Oj
* 	 * Q_
1 	 CO
ro 0. C -0
•P 1— ro ro
0 CD S- Ol
r— CD r— _l




X X

X





X X
Table VI.
Health Effects and Medical Surveillance for
Selected Airborne Contaminants (Partial List)3
Airborne
Contaminant Health Effects
Acetone Irritation of respiratory tract
and mucous membranes;
narcosis; dermatitis
Alumina Lung overload; no fibrosis
Ammonia Irritation or respiratory tract
and mucous membranes
Antimony Gastrointestinal and heart
Pentachloride disorders; respiratory irritation
Asbestos Fibrosis of lung; bronchogenic
carcinoma; mesothelioma;
cancer of stomach, colon and
rectum

Bauxite (may Lung overload
contain silica)
Beryllium Fibrosis of lung; dermatitis


Medical
Surveillance
Interval history


Interval history
Chest X-ray; lung
function
EKG; Chest X-ray;
lung function
Chest X-ray; lung
function; sputum
cytology (OSHA re-
quirement — annual
medical surveillance)
Chest X-ray; lung
function
Chest X-ray; lung
function; skin in-
inspection
Airborne
Contaminant
n-Butyl alcohol


Butyl Cellosolve (2-
Butoxy Ethanol)
Cadmium Oxide


Carbon Dioxide
Carbon Monoxide





Health Effects
Irritation of respiratory tract
and mucous membranes;
narcosis; dermatitis
Irritation of respiratory tract
and mucous membranes;
narcosis; dermatitis; red cell
hemolysis
Pulmonary edema; fibrosis of
lung; kidney damage


Asphyxiation
Asphyxiation; disturbed
consciousness



Medical
Surveillance
interval history


Complete blood count

Chest X-ray; lung
function; cadmium
and protein in urine;
urinalysis
None
Interval exam; car-
carboxyhemoglobin
if intoxication
suspected




-------
 276    SITE & PERSONNEL SAFETY
  (3) Field monitors can be used to evaluate the type and
     concentration of chemicals  on site that the team
     might be exposed to indirectly.
  Personal monitoring and area monitoring (manual and
automatic) are the most common  techniques used to de-
termine an exposure to atmospheric chemicals. However,
both are subject to significant errors. Biological monitor-
ing may be useful to validate the atmospheric monitoring.
Also, exhaled air  analysis can  be used to measure ex-
posure to industrial chemicals.(3> A partial list of potential
health effects, type of medical surveillance and industrial
hygiene considerations required for several specific air-
borne contaminants is given in Table VI.<3) The usual ap-
proach in determining that a chemical insult to the body
has occurred  due  to exposure  to a hazardous material in-
volves medically observing changes, either from the norm
or a previous baseline,  in a person's health.
   During  a  response action,  response  team members
should be carefully observed by the on-scene coordinator,
on-site  industrial hygienist or medical personnel  and fel-
low team members. The first warning signs of toxic ef-
fects from a  hazardous exposure are often observed by a
fellow worker.(5) Finally examinations  on the termination
of employment  are desirable since they  document the
health status then and provide evidence of any  changes
that have  occurred during the employment period. This
does not rule out the possibility that effects may show up
at a later date.

ACKNOWLEDGEMENT

   This project was sponsored by the U.S. EPA's Oil and
Hazardous Materials  Spills Branch, Municipal Environ-
mental  Research  Laboratory-Ci, Edison,  New   Jersey,
under  Contract  No.  68-03-3014 with  Rockwell  Interna-
tional. The content of this publication does not necessarily
reflect the  views or policies of the U.S. Environmental
Protection  Agency, nor  does  mention of  trade names,
commercial products or organizations  imply endorsement
or recommendation by the U.S. Government or Rockwell
International.
  The authors express their appreciation to EMSC staff
members Mr. Fitz Bush, Mr. David Cabrey, Ms.  Patricia
Casey, Dr. George Schneider,  and Ms. Patricia Scofield,
for their  assistance  to the project.  Also,  the  authors
acknowledge the assistance  rendered by Ms. Debra Sibert
and Mr. Eugene Port of Health Science Associates of Los
Alamitos, California.

REFERENCES

1. Pritchard, J. "A Guide  to Industrial Respiratory Pro-
   tection," LA-GG71-M, UC-41, Los Alamos Scientific
   Laboratory, Los Alamos, New Mexico, March 1977.
2. Melius,  J. and Halperin, W.,  "Medical Screening of
   Workers at  Hazardous  Waste Disposal Sites," pre-
   sented at the Society  for Occupational and Environ-
   mental  Health  Conference  on  Hazardous  Waste,
   Washington, D.C., December 7-10,  1980.
3. Galley,  L.J.  and Cralley,  L.V., "Patty's Industrial
   Hygiene and Toxicology, Volume III, Theory and Ra-
   tionale of Industrial Hygiene  Practice," John Wiley
   and Sons, New York, N.Y.,  1979.
4. NIOSH Pilot Study for Development of an Occupa-
   tional Disease Surveillance Method,  published May
   1975.
5. Proctor, N.H. and Hughes, P., "Chemical Hazards of
   the Workplace." J.B.  Lippincott  Company, Phila-
   delphia, Pennsylvania, 1978.

-------
                             INITIAL SITE PERSONNEL
PROTECTION LEVELS BASED ON TOTAL VAPOR READINGS
                                        RODNEY D. TURPIN
                                U.S. Environmental Protection Agency
                                    Environmental Response Team
                                          Edison, New Jersey
INTRODUCTION

  The EPA Environmental Response Team (ERT) was
established in October 1978 to provide technical assistance
to On-Scene Coordinators (OSC), Regional Response
Teams (RRT), EPA  Headquarters  and other  Regional
Offices, as well as other governmental agencies at spills of
hazardous materials and at uncontrolled hazardous waste
sites. In this paper, the author describes the procedures
used by ERT in selecting the appropriate level of personnel
protection during the initial assessment and operational
phase. Personnel protection is  categorized into three levels
and criteria for selection is based on potential total vapor
exposure.
  The objective of using total atmospheric gas/vapor con-
centrations for determining the appropriate personnel pro-
tection level is to provide a numerical criteria for selecting
Level A, B,  or C protection.  Until atmospheric contam-
inants are specifically identified and personnel protection
selected based on lexicological properties, total gas/vapor
concentrations provide a numerical value that can be used
as a guide for selecting personnel protection equipment.
  Although total gas/vapor concentration measurements
are useful to a qualified professional for the selection of
protection equipment, caution should be used in the inter-
pretation of this data. The response of an instrument to
several gas/vapor contaminants does not provide the same
sensitivity as measurements  involving a single contam-
inant. Since total vapor field instruments see all contam-
inants in relation to a specific  calibration gas, the concen-
tration of an unknown environment may be over esti-
mated or under estimated.
  When carcinogens or other highly hazardous substances
are suspected,  the protection level should not be based
solely on  the total gas/vapor  criteria, rather  the level
should be selected on a case-per-case basis weighing heav-
ily on potential exposure and chemical characteristics of
the suspected material.

FACTORS FOR CONSIDERATION

  In utilizing atmospheric gas/vapor concentrations as a
guide for selecting a level of protection, a number of fac-
tors should be considered:
1. The uses, limitations, and  operating characteristics of
  the  monitoring instruments  must be recognized and
  understood. Instruments such as the photoionizer and
  Organic Vapor Analyzer (OVA) do not respond to all
  substances that may be present or may respond differ-
  ently to identical substances when compared to one an-
  other.  Therefore,  experience, knowledge,  and good
  judgment must be used to compliment the data obtained
  with this instrumentation.
2.  Hazardous other than detectable toxic gases/vapors
  such as non-detectable gases  (i.e., phosgene, HCN,
  chlorine, etc.,)  liquid/solid  particulates,  explosives,
  combustibles, radiation, oxygen  deficiency, and other
  harmful conditions may exist in the atmosphere or on
  the site.
3.   The risk to personnel entering an area must be eval-
  uated against the need for entering. Although this as-
  sessment is largely a value judgment,  a conscientious
  assessment of the variables involved  and  the  risk  to
  personnel must  be balanced against this need for site
  entry.
4. The knowledge that carcinogens or other highly toxic
  substances are involved or suspected requires that gross
  levels not be the sole factor in determining the level of
  protection. Other factors which must be taken into con-
  sideration are  exposure,  chemical  characteristics  of
  known/suspected materials, instruments, weather con-
  ditions, etc.
5. Functions which need to be performed on-site must be
  evaluated. Based upon total atmospheric gas/vapor con-
  centrations, Level C protection may  be judged ade-
  quate.  The work  functions to be  performed such  as
  moving drums, opening containers, bulking of materi-
  als and other operations that increase the probability
  of exposure may require a higher level of protection.

CRITERIA

  The criteria for relating Levels of Protection to tptal
atmospheric concentrations are:
  Level C   0-5 ppm above background
  Level B   5-500 ppm above background
  Level A    500-1000 ppm above background

Level A Protection

  Level A protection provides the highest degree of pro-
tection against hazards due to  inhalation, skin and eye
                                                   277

-------
278     SITE & PERSONNEL SAFETY
irritation, if the inherent limitations of the personnel pro-
tective equipment  are not exceeded. Yet Level A protec-
tion does present the highest  potential for heat stress.
Cooling vests/suits have demonstrated some degree of re-
lief to body  temperature build-up. The range of 500 to
1000 ppm total gas/vapor concentrations in air is based on
the following criteria:
•Since Level  A will provide protection against air concen-
 trations greater than 1000 ppm for most substances, an
 operational  restriction  of 1000 ppm  is established as  a
 warning to:
 •Take into consideration analytical instrumentation sen-
   sitivity to wind velocity, humidity, temperature, etc.
 •Evaluate the calibration and/or sensitivity error asso-
   ciated with the instrument(s).
 •Evaluate the need to enter environments greater than
   1000 ppm.
 •Identify the specific  constituents contributing to the
   total concentration and their associated toxic properties.
 •More precisely  determine the concentration of the in-
   dividual constituents  that make  up the total concentra-
   tion.
•A lower limit of 500 ppm (for Level A) total gas/vapor
 concentration in air is selected as the value to upgrade
 from Level B to Level A in  order  to fully protect the
 skin  and/or eyes until qualitative  and quantitative de-
 terminations can be  made  of the  constituent products
 and skin and/or eye hazards excluded.
•The range of 500-1000  ppm total concentration is suffic-
 iently conservative to  provide a  safe margin  of protec-
 tion due to instrument error,  calibration and sensitivity,
 unanticipated transient concentrations, and  protection
 against highly hazardous substances that could account
 for the total concentrations.
   Experience with properly operating portable field equip-
ments  for measuring total gases and/or vapors has demon-
strated levels approaching 500 ppm have not  routinely
been encountered on hazardous waste  sites. High con-
centrations have  only been encountered in closed build-
ings, at openings  to containers or when working in the
spilled contaminants.
   A decision to require  Level A should also consider the
negative aspects  of this high level of protection, i.e.,
higher probability for accidents due to cumbersome equip-
ment,  increased resources needed  and the physical stress
caused by heat   buildup  in  fully  encapsulating  suits.
These  factors need to be carefully evaluated and balanced
against the reasons for utilizing Level A.

Level B Protection

   Level B protection  is the  minimum level of protec-
tion required for initially entering an open site where the
type(s), concentration(s) and presence of airborne gas/
vapors are unknown. This level of protection provides a
high degree of inhalation, skin and eye irritation or ab-
sorption protection.  Although a small portion of the body
(neck and head) are exposed, primarily liquid/solid ma-
terials  have  the  highest potential of  causing  acute or
chronic effects due to exposure of this area. The use of a
hooded chemical resistant jacket would further reduce the
potential for exposure to this area of the body. Although
the potential for heat stress is not as high as Level A, it
can be a major problem. Cooling vests/suits have demon-
strated some degree of relief to body temperature build-up.
  A limit of 500 ppm total atmospheric gas/vapor con-
centration readings on portable field instruments has been
selected as the upper restriction on the use of Level B.
Although Level B personnel protection would provide ade-
quate protection against most  substances at  concentra-
tions higher than 500 ppm, an upper limit of 500 ppm is
selected as the decision point  for  a careful evaluation of
the risks associated with higher concentrations.
  Considerations:
•Analytical instrumentation sensitivity  to wind velocity,
 humidity, temperature, etc.
•Calibration and/or sensitivity error associated with the
 instruments.
•The probability that substance(s) present are cutaneous or
 percutaneous.
•The necessity for entering higher  concentrations in Level
 B.
•The work function to be done and the increased proba-
 bility for exposure.
•Qualitative and quantitative identification of the specific
 components.

Level C Protection

  This level provides the same high degree of skin protec-
tion as  Level B, but lesser inhalation and/or eye protec-
tion. A relatively low (0-5 ppm above  background) am-
bient concentration has been established as the range for
wearing Level C protection equipment.  An upper limit of
total vapor concentration of 5 ppm (above background)
has been selected primarily based on the use of a full-face
air purifying gas mask with canisters and  requirements as
to its constraints and limitations. These are:
•MSHA/NIOSH approved air-purifying devices  should
 only be worn in atmospheres where the substances have
 been identified or  the potential of exposure is highly un-
 likely.
•Substances must have good warning properties.
•Continuous air monitoring must occur  for the atmos-
 pheric contaminants identified.
•Appropriate, approved canisters must be used.
•Sufficient oxygen (at least 19.5% of air at sea level) must
 be present.
  Full-face, air purifying devices will provide respiratory
protection against most environmental vapors greater than
5 ppm; however, until qualitative and quantitative infor-
mation is available about the substances, concentrations
greater than 5 ppm indicate a higher level of respiratory
protection should be used.

PROTECTION EQUIPMENT

  Since exposure factors vary  from one situation to an-
other,  it is impossible to develop a  specific safety pro-
cedure  that will be appropriate for  all situations;  ERT

-------
                                                                          SITE & PERSONNEL SAFETY
                                                  279
limited the selection of protective equipment to three gen-
eral categories. Although total gas/vapor monitoring is the
primary screening mechanism used, the protection levels
indirectly address protection against three possible routes
of exposure (inhalation,  skin absorption/irritation, and
ingestion).

Level A—Personal Protection Equipment

•Positive Pressure SCBA (MSHA/NIOSH approved) op-
 erated in the positive pressure mode
•Totally Encapsulating Suit (boots and gloves attached,
 cooling vest when applicable)
•Gloves—Inner (tight fitting and chemical-resistant)
•Boots—Chemical-protection, steel toes and shank. De-
pending on suit boot construction, worn over suit boot
•Gloves—Outer,  chemical  resistant.  Depending  on suit
 construction, worn over suit gloves. May be replaced with
 tight fitting,  chemical-resistant gloves worn inside suit
 gloves.
•Underwear—cotton, long-John type
•Hard Hat—(under suit)*
•Disposable protective suit, gloves, and boots. (Worn un-
 der or over encapsulating suit)
•Coveralls (under suit)*
•Two-way radio communications
"Optional

Level B—Personal Protective Equipment

•Positive  Pressure  SCBA  (MSHA/NIOSH  approved),
 operated in the positive pressure mode
•Hooded, two-piece chemical-resistant suit (cooling vest
 when applicable)
•Gloves—Inner, tight fitting, chemical-resistant
•Boots—Outer (chemical-resistant, heavy rubber dispos-
 ables)
•Boots—Inner (chemical-resistant, steel toe and shank)
•Two-way radio communications
•Hard Hats*
•Face Shield*
*Optional

Level C—Personal Protective Equipment

•Full-face gas  mask,  air-purifying respirator  (MSHA/
 NIOSH approved)
•Chemical-resistant clothing
•Overalls  and long-sleeved jacket or coveralls; hooded
 two-piece chemical splash suit (when applicable—hooded
 disposable coveralls)
•Gloves—Outer (chemical-protective)	
•Gloves—Inner (tight fitting, chemical-resistant type)
•Cloth or disposable coveralls—inside  chemical protective
 clothing, cooling vest when applicable*
•Escape pack
•Hard Hat* (face-shield, optional)
•Boots—Outer (chemical-protective heavy rubber throw-
 aways)
•Boots—Inner (chemical-protective, steel toe and shank)
•Two-way radio communications
* Optional

ACKNOWLEDGEMENT

  The author expresses his appreciation to all members of
EPA's Environmental Response Team,  especially Tom
Sell, Training Course Director, for their many contribu-
tions and his  indebtedness to David Weitzman, EPA, Of-
fice  of Occupational  Health  and  Safety, Washington,
D.C., Richard Costello,  NIOSH, Hazardous Evaluation
and Technical Assistance Branch,  Cincinnati, Ohio and
David  Dahlstrom and David Schafer, E&E,  ERT-TAT,
Cincinnati, Ohio for their critical review and comments.

-------
             HAZARDOUS SUBSTANCE SITE AMBIENT AIR
CHARACTERIZATION TO EVALUATE ENTRY TEAM SAFETY
                                       MARTIN S. MATHAMEL
                                      Ecology and Environment, Inc.
                                             Chicago, Illinois
INTRODUCTION

  Because of the wide variety of ambient conditions that
may be encountered in investigation and remedial action
activities  on hazardous substance sites,  a means of de-
termining the potential dangers is essential in order to pro-
tect worker health and safety. A number of airborne toxic
chemicals have been found at the boundaries of hazardous
substance sites (Table I).(1)
  The  air emissions from hazardous substance sites are
complex because there is usually no single point source on
a site and because concentrations of toxic chemicals de-
crease  rapidly as they diffuse  and disperse in the atmos-
phere."1 In addition, explosive or oxygen-deficient atmos-
pheres may exist, and radiation may be encountered.
  Depending on  the nature of the site,  the information
necessary to assess the hazard potential may be obtained
by researching existing data sources. However, in many
cases, a variety of equipment may be required to character-
ize and monitor the ambient air, both on-site and at the site
boundaries. The goals of such an ambient air characteriza-
tion scheme are as follows:
  (1) To qualitatively and quantitatively define the  haz-
      ards due to airborne pollutants on hazardous  sub-
      stance sites
  (2) To locate specific areas of a site upon which to focus
      the investigation and remedial action activities
  (3) To monitor the surrounding community for any ad-
      verse environmental impact that work on the site
      may cause
The  data obtained from the air characterization study is
used to develop a personnel protection plan that allows
workers to perform site activities in the lowest level of pro-
tective gear consistent with maintaining  their health and
safety.
  In this paper, the author discusses the equipment that
can be used to define the  hazards associated with hazard-
ous substance sites. The actual selection  of specific per-
sonnel protective gear based on the analytical data  gen-
erated  by the air characterization is beyond the scope of
discussion but several references detailing selection criteria
are included.

PRELIMINARY HAZARD ASSESSMENT
  Before work is initiated at a hazardous substance site,
detailed background research should be conducted to as-
certain what chemicals,  hazards, special requirements,
etc., may be associated with the site. Possible sources of
information are state, county and municipal agencies, in-
cluding the federal and state EPA.
  All site waste generator records should be reviewed, in-
cluding inventories, shipment manifests, and permits. Per-
sonal interviews with site personnel, public officials, and
private citizens may be helpful.
  The most valuable information, of course, is the results
of any air monitoring or sampling that has been perform-
ed.  In addition, data on the waste composition may pro-
vide information on potential airborne pollutants. It may
be desirable to perform an offsite  preliminary  inspec-
tion, since many times a simple visual reconnaissance can
go a long way in terms of defining probable hazards.

                      Table I.
     Examples of Predominant Species Found in Air Near
                Hazardous Waste Sites.(1)

Benzene
Toluene
o-xy lene
tn + p-Xylene
Aceta Idehyde
Benzaldehyde
laopropyl ether
Phenol
Ethyl acetate
Ethyl ether
Dimethyl ether
Naphthalene
Ethyl benzene
Methyl isobutyl ketone
n-Pentanal
Ch loroform
Methylene chloride
Dibromorae thane
Dichloroethane
Tetrach lor oe thane
Tr ich lor oe thane
Trich loroethylene
Tetrachloroethylene
Carbon tet rach lor ide
Vinyl methyl ether
Vinyl isopropyl ether
Vinyl ch lor ide
Ch lorobenzene
Ch loroto luene
Di ch lorotoluene
Tr ich lorobenzene
Tr ich lorotoluene
Bromoxy lene
PCfl'a
Maximum concentration observed at each dunpiite,
ug/n,3
Love Canal
Niagara Falls,
N.Y.
5703
1472
73
140











172
10


1140
73
270
1140
5



240
too
6700
159
74
44


Kin Hue,
Ediaon, N.J.
1550
2600


245
56
120
10
232
35
10
6100

444
38
266
1250
63
57
22
500
13
394
20
5000
13000

50
34

1

50

Elizabeth,
N.J.
234
325
79
225

















218
95




16






Olh«r
10
64
20
41







11J
2(2


1.1




2.0
4.5
I.I



110000
118





100
                                                            Reference 1
                                                  280

-------
                                                                              SITE & PERSONNEL SAFETY  281
  Once the background data are assembled, a preliminary
hazard assessment is performed, generally requiring an in-
terdisciplinary  scientific  approach  by  individuals  with
training in chemical,  biological and radiological  safety
and health and experiences in determining and implement-
ing a personnel protective equipment program. The task is
not entirely quantitative, since many subjective profess-
ional judgments are required.
  Each site is  unique, posing specific problems and re-
quiring individual attention.  For example,  the site may be
known  to contain a hazardous chemical; however, if the
site is properly managed and the hazardous chemical ade-
quately contained, there may be only a  minimal  threat
to worker health and safety.

AIR MONITORING EQUIPMENT

  In many cases, the information obtained from the back-
ground research either is not sufficient to define the site
hazards or indicates that the hazard potential is great
enough to require specific  personnel protective  equip-
ment. Both instances  necessitate the  use of a  variety of
equipment to characterize and monitor the ambient air so
that the appropriate level of personnel protection can be
chosen.
  Continuous monitoring is also required since workers
may encounter hazards such as explosive atmospheres or
high levels of  radiation for which no protective equip-
ment is available. Continuous  monitoring  would also
allow the selection of the minimum amount of protec-
tive equipment consistent with maintaining worker health
and safety.
  The air monitoring  equipment  described in this paper
can be conveniently divided into two categories:
•Direct reading   instruments,  which provide  a  "real
 time"  readout of the concentration of pollutants, and
•Collection  media,  which collect  and concentrate  the
 pollutants for subsequent laboratory analysis.
  A list of equipment which is sufficient  to evaluate en-
try team safety and to determine  and implement a per-
sonnel protection equipment program for  hazardous sub-
stance site workers is given in Table II.
  With the exception  of explosive  and/or oxygen-defic-
ient atmospheres, there is no way to reliably  predict the
health effects of  exposure to hazardous substances based
on the readout of a single direct reading instrument, un-
less the instrument can quantitatively respond to all of the
substances present at any given instant.  This is especially
true for_entry  into  sites where little or no  background
information exists, since the hazards are undefined and on
sites where "dynamic" activities such as  drum opening,
sampling, drilling, etc., are in progress.
  Hazardous substance  sites may contain a number  ot
pollutants in levels below the detection limits of  direct
reading instruments but still can cause synergistic health
effects.  Moreover, many toxic  substances such as particu-
late organics and metals simply cannot be determined by
a direct reading instrument.
  The effects of personnel exposure to these types of sub-
stances must be carefully determined, especially for long-
term site activities involving repeated exposure. For ex-
                        Table II.
     List of Equipment for Ambient Air Characterization.
Hazard
Explosive
 atmosphere

Oxygen-
 deficient
 atmosphere

Toxic
 atmosphere
Radioactivity
Direct Reading

Combustible gas indicator


Oxygen level meter
1. Portable photoionization
detector (PID)
2. Portable flame ioniza-
tion detector (FID) w/gas
chromatograph (GC) option
3. Colorimetric tubes
1. Radiation survey meters
(alpha, beta, gamma)
2. Passive monitors (alarms)
Collection Syste

Not used


Not used
Sampling pumps in con-
junction with absorption
tubes, filters, and im-
pingers


Dosimeters (film badges)
See References 2 and 3 for criteria for use of combustible gas indicators and colorimetric tubes.


ample, gamma radiation is relatively easy to detect in the
field but low-level alpha or beta radiation  is not. The
health effects of human exposure to low levels of alpha
and beta are potentially more severe than exposure to a
much higher level of gamma radiation. Thus, the deter-
mination of the effects of exposure must  be made  by a
health physicist and cannot be determined  solely  by a
single instrument reading.
  As another example, direct reading instruments such as
the photoionization detector (PID) and .the flame ioniza-
tion detector (FID) are very useful in locating "hot spots"
—those areas of a site that contain significant levels of
volatile chemicals. However, in order to  determine the
health effects of~direct exposure, and ultimately personnel
protection  levels,  the  specific chemicals must be identi-
fied.
  For a positive chemical identification, air samples are
collected for laboratory  analysis; the air contaminants
are concentrated on a collection medium by  passing am-
bient air through it. Since this type of air sampling is ex-
pensive, the value of an extensive background research is
apparent.  For sites with insufficient background data, it
may be cost effective to perform work activities using full
protective  gear, while observing personnel for signs of
stress. Once again, the subjective professional judgment of
an individual experienced in  the implementation of a per-
sonnel protective program is essential.

Explosive and Oxygen-Deficient Atmospheres

  Explosive and oxygen-deficient atmospheres are unique
in that the hazards associated with them can be deter-
mined by using direct reading instruments, provided that
the operator is experienced and trained in  field use of
the instruments. For explosive atmospheres a combustible
gas indicator is used.  Although  there are many commer-
cially available units,  most  test  for  the concentration of
explosive gases and vapors by measuring the heat produc-
ed by the  combustion of a test sample. The readout is
usually  expressed  as  a percent of the lower  explosive
limit (% LEL), a  dimensionless quantity  defined as  the
lowest concentration of a gas or vapor by volume in air

-------
282   SITE & PERSONNEL SAFETY
which will  explode or combust when there is an ignition
source present.
  The  National  Institute  for  Occupational  Safety  and
Health (NIOSH) considers the action level for limitation
of work in explosive areas to  be  10%  LEL, unless pre-
cautions are taken to prevent ignition sources. A reading
of 20% LEL is considered IDLH (immediately dangerous
to life  and health)  and  requires evacuation regardless of
what precautions have been taken. Also, work activities
cannot be  performed in atmospheres where the concen-
trations  of gases or vapors exceed the upper  explosive
limit (UEL), which is the  concentration of a gas or vapor
by volume  in air above which an explosion will not occur
if there is  an  ignition source. Evacuation is necessary  if
UEL concentrations are detected, since dilution  by air
can easily create an explosive situation.  Gear is not avail-
able for  protection against the explosion hazard; immed-
iate evacuation is the only recourse.
  Oxygen-deficient atmospheres can be  determined by us-
ing a commercially available oxygen level meter. A typical
unit works on the principle of selective penetration of a
membrane by oxygen with a subsequent reaction with an
electrolytic reagent. The  readout  is generally directly in
percent oxygen (% 02).
  NIOSH  requires that an  external air supply,  usually a
self-contained breathing device (SCBA), be used for work
in atmospheres with less  than 19.5%  oxygen. Severely
oxygen-deficient atmospheres indicate the presence of sig-
nificant  quantities  of pollutants.  Note that  combustible
gas indicators may not function properly in severely oxy-
gen-deficient areas and should be used with an  oxygen
meter.

Radioactivity
  There are three types of radiation that can be encoun-
tered  on  hazardous substance  sites:  alpha, beta  and
gamma.  Human  exposure  to  each type  of  radiation
causes unique health effects, which are related to the re-
ceived dose.
  Sources  of gamma on hazardous substance sites include
hospital  wastes, which may contain materials such as ces-
ium, cobalt,  and radium. Gamma radiation is easily de-
tected using a variety of instrumentation, since it is in the
form of electromagnetic energy and therefore travels rela-
tively long distances. Because gamma  is penetrating ex-
ternal  radiation, there  is no convenient means of pro-
tecting workers from exposure; if  gamma radiation is de-
tected, personnel should  be evacuated immediately  and
appropriate government agencies contacted.
  Alpha and beta radiation, however, travel much shorter
distances than gamma and are therefore much more diffi-
cult to detect, since the monitoring device must be placed
close to  the source. Alpha  sources are common and in-
clude  substances such as  thorium, radium and  uranium.
Biological  laboratory wastes  may contain beta  sources,
including  radioactive carbon,  phosphorus  and  sulfur.
Since alpha and beta radiation are particulates and tend to
adhere to clothes and boots of site  workers, monitoring
devices  that will detect  their presence should be used to
screen personnel as personnel leave the site.
  Although the external hazards  due to alpha and beta
radiation differ, protective clothing is available. The great-
est danger is in the inhalation or ingestion of alpha or beta
particles. An air-purifying respirator fitted with a radio-
nuclide canister will in most cases offer protection against
radioactive particulates. Adequate decontamination facil-
ities must be available, however.
  For  field  work, personnel should  be equipped with a
portable alpha/beta/gamma  survey meter.  Lapel-sized
passive radiation monitors which respond with an audible
alarm  when  the wearer encounters  gamma  are recom-
mended for at least one team member. In addition, a dos-
imeter, which provides an indication of the  total  whole
body  exposure to radiation over  an extended period of
time,  should be worn by all onsite personnel. Dosimeters
are typically distributed to  workers upon site entry and
are worn on the lapel, underneath any protective clothing.

Toxic Atmospheres

  When workers are engaged in activities on a hazardous
substance site,  Federal safety and health standards141 dic-
tate that they be protected against any respiratory danger.
Personnel must wear an SCBA until the ambient air is fully
characterized. It is desirable to use the lowest  level of res-
piratory protection possible (i.e., the air-purifying respir-
ator or no respiratory protection at all), since the addition
of protective gear generally increases worker stress. Thus,
the air monitoring program must provide the data neces-
sary to determine respiratory protection.
  The problem  of cutaneous and percutaneous chemical
hazards must also be addressed. Skin protection must be
worn if there is any possibility of systemic injury, skin irri-
tation, or death resulting from contact  with  an airborne
gas, liquid, or paniculate. Typically, the following infor-
mation regarding  all airborne pollutants is  obtained in
order to specify protective gear:'5'
•The  chemical identity,  the airborne concentration, and
 the physical state of each pollutant
•Permissible exposure limits of each pollutant
•The vapor pressure and equivalent ppm of each pollutant
•Warning properties for a gas or vapor
•Eye irritation potential for each pollutant
•The   LEL,  UEL, and IDLH  concentration  for  each
 pollutant
  One of the methods of determining the need for respir-
atory  protection  is  to compute the  "equivalent  ex-
posure"'61 based on the first three parameters in the table
above. The calculations involved in computing the equiv-
alent exposure are complex and are best left to the indus-
trial hygienist who has experience in determining accep-
table risk  levels. Many factors  in addition to  those above
enter into the selection of specific personnel protection and
are beyond the scope of this discussion.*
  Direct reading instruments such as the PID,  the FID and
colorimetric  tubes are used to locate "hot spots" (areas
of high air contamination) and in some instances provide
•See References 4, 5, and  7 through 14 for information on selecting
 protective equipment and  determining the toxicological  effects of ex-
posure to hazardous substances.

-------
                                                                              SITE & PERSONNEL SAFETY   283
a tentative identification of volatile inorganic and organic
chemicals. Hot spots are important since they define the
areas of the site that may contain significant quantities of
chemicals.
  Air sampling pumps for volatile organics are usually
placed on the hot spots. Direct reading instruments can
also be used to  monitor  the ambient air for dramatic
changes in quality  that  may occur as a result of activ-
ities such  as drilling, drum opening, etc. because these in-
struments provide a real time readout of pollutant levels.
  The PID  is a non-specific vapor/gas detector. It  uses
the principle of photoionization to detect a wide variety
of volatile chemical compounds. The readout is in parts
per million (ppm),  references to  a specific  calibration
standard.  It does not respond  to the common  constit-
uents of air, including methane.
  The flame ionization detector (FID) in conjunction  with
a gas chromatographic (GC) column is used to detect and
to tentatively  identify  volatile organic chemical  com-
pounds. In the "survey mode" it functions as a non-spe-
cific total hydrocarbon detector, with a readout in ppm,
referenced to a specific standard, usually methane.
  In the "GC mode" the  unit can determine the "reten-
tion time" of an unknown chemical compound.  The re-
tention time is the length of time necessary for the chem-
ical to pass through the GC column. By comparing the
retention  time of the unknown with the retention time of
known standards, a tentative qualitative/quantitative iden-
tification  of the unknown can be made.
  Since specific  chemical  standards and  calibrated col-
umns are needed, an idea  of the nature of the unknown
is  needed prior  to  the  GC  analysis.  Thus, the impor-
tance of  background research is  demonstrated,  since  a
prior knowledge of the air  characteristics allows the selec-
tion of analytical  standards for the GC.
  Colorimetric tubes provide a means of qualitative/semi-
quantitative identification of volatile organic and inorgan-
ic chemicals. The sample is drawn into the tube by means
of a hand-operated pump.  Since the reagents in each  tube
react with a specific chemical compound, prior knowledge
of the chemicals expected is necessary for selection of the
tubes. At a  minimum, tubes for hydrocyanic acid, phos-
gene and  hydrogen sulfide should be used on  sites inves-
tigated.
  For positive identification of all the pollutants that may
be present, various  collection media are used in conjunc-
tion with  sampling pumps. Sampling pumps are commer-
cially  available in a number  of configurations,  but for
general   site   characterization,   an   intrinsically   safe
"NIOSH" pump is  recommended. This type of pump is
portable,  battery  operated, and sufficiently lightweight so
that it can be worn by personnel. Typical collection med-
ia  include activated  carbon,  Tenax  GC,  XAD-2,  silica
gel and Florisil tubes, membrane and glass fiber filters,
polyurethane foam  plugs  and impingers  containing se-
lected reagent solutions. Specific applications for  these
media, including  the typical laboratory analysis required
are given in Table III.*

*For a complete listing of methods and collection media for air sampling,
see References 15,16 and 17.
                        Table III.
      Specific Applications for Collection Media Including
             the Required Laboratory Analysis.
  Pollutant

  Volatile organics
  Particulate
   organics

  Pesticides (in-
   cluding PCBs)


  PBBs

  Metals

  Volatile inorganics

  Particulate
   inorganics

  Cyanides
Collection Media

Carbon tubes
Tenax tubes
XAD-2 tubes
Silica gel tubes

Glass fiber filters


Florisil tubes
Polyurethane plugs
Glass fiber filters

Glass fiber filters

Membrane filters

Impingers/reagent solutions

Membrane filters
Glass fiber filters

Filters/impingers
Laboratory Analysis
Gas chromatograph/mass
spectroscopy (GC/MS)
GC/MS


GC/MS
GC/Electron capture


GC/MS

Atomic absorption (AA)

Wet chemical methods

Wet chemical methods


Wet chemical methods
  As previously mentioned, pump placement for volatile
materials can be determined by using the  PID, the FID
and colorimetric tubes to survey the site for hot spots. In
general,  pumps are placed in selected work areas, as well
as near drums, leachate ponds  or streams, surface water
impoundments, lagoons,  areas with visibly contaminated
soils  and in the headspace of groundwater monitoring
wells. Upwind  and downwind sampling points are recom-
mended.
  Once  personnel protection  has been  determined, air
monitoring is still necessary. For example, assume that an
air-purifying respirator has been determined to be ade-
quate for a worker involved in drilling operations. Since
drilling may result in a spontaneous  release of  toxic sub-
stances, a means to determine if the air quality has changed
is necessary to insure that  the respirator will protect the
worker.  In this situation, a FID or a PID can be used to
give a real time readout  of pollutant levels. A combus-
tible gas indicator/oxygen meter combination should also
be  used  to  determine  if  an explosive  situation has been
created.  If the  drilling operations are long term, it would
be desirable for the worker to wear a sampling pump fitted
with a  carbon or Tenax tube  so that his/her total ex-
posure can  be monitored.  Real-time monitoring  should
also be used for drum opening and staging, sampling, ex-
cavating, draining of tanks and lagoons and related re-
medial action activities.

Environmental Monitoring

  Environmental monitoring also can be performed using
the equipment  discussed in  this paper. Of particular con-
cern is  the  impact that site work activities have  on the
surrounding  community,  since  new hazards may  be
created.  In  general, the  highest concentrations of toxic
pollutants will occur directly over the site, close to the
point source. A reasonable  assumption is that the concen-
trations  will decrease as the pollutants  migrate offsite.
Thus, if a complete air monitoring program for personnel
protection has  been  initiated, the values  measured  onsite
can be considered worst case.

-------
284    SITE & PERSONNEL SAFETY
  Several models have been developed to predict the air
emissions from sites.(l8- "• M) However, offsite monitoring
may be required, since in many cases workers spend a good
deal of time at the site boundaries. A common mistake  is
to remove protective gear, as if the site boundary somehow
limits pollutant migration. This type  of site  boundary
monitoring is  an integral part of evaluating overall site
safety.

CONCLUSION

  Entry into a hazardous substance site involves poten-
tial exposure to a number of hazards. It is essential that the
risks associated with these hazards be determined such that
protective measures  can  be taken to maintain worker
health and  safety. A hazard assessment is necessary  and
should precede any site work. Background research for site
specific information as well as on-site air characterization
will supply the quantitative data needed to perform the
assessment.  A personnel  safety plan should be designed
based on these data.
  In addition  to the quantitative data,  many qualitative
judgments  enter into the determination of an effective
and efficient safety  plan. Perhaps  the  most  important
factor is determining if the physiological and  psycholo-
gical stresses caused by wearing protective gear  are great-
er than the risks of direct exposure to hazardous sub-
stances. Therefore,  individuals  with training and exper-
ience in hazard evaluation and the implementation of  a
personnel protection plan are as necessary as the monitor-
ing equipment used to characterize the ambient air.


REFERENCES

  1. Esposito, M. and Bromberg,  S., "Fugitive Organic
    Emissions  from Chemical Waste Dumpsites,"  paper
    presented  at the 74th annual  meeting of Air Pollu-
    tion Control Association, Philadelphia, Pennsylvania,
    June 21-26, 1981.
 2. "Combustible Gas Indicators, A Manual of Recom-
    mended Practice,"   American  Industrial  Hygiene
    Association, Akron, Ohio, 1981.
 3.  "Direct Reading Colorimetric Tubes, A Manual of
    Recommended  Practice,"  American Industrial  Hy-
    giene Association, Akron, Ohio, 1976.
 4.  OSHA  Safety and Health Standards (29 CFR 1910),
    U.S. Department of Labor, Occupational Safety  and
    Health  Administration, OSHA 2206, revised January,
    1976.
 5.  Pritchard,  J.,  "A  Guide  to  Industrial Respiratory
    Protection," Appendix  F, "Joint  NIOSH/OSHA
    Standards Completion Program Respiratory Decision
    Logic," Publication LA-6671-M, Los Alamos Scien-
    tific Laboratory, Los Alamos, New Mexico.
 6. American Conference of Governmental Industrial Hy-
    gienists, "TLVs  Threshold Limit Values for Chem-
    ical Substances and  Physical Agents  in the Work-
    room Environment," Appendices B, C, and D, and
    "Documentation of the Threshold Limit Values for
    Substances  in  the  Workroom Air," ACGIH, Cin-
    cinnati, Ohio, 1978.
 7. Mackison, F., Stricoff, R., and  Partridge,  L., ed-
    itors,  "NIOSH/OSHA  Pocket Guide  to Chemical
    Hazards," U.S. Dept. of Labor, Occupational  Safe-
    ty and Health Administration, August, 1980.
 8. "Methods for Prevention and  Control of  Occupa-
    tional Skin  Disease," Bureau of National Affairs,
    Washington, D.C.
 9. Sansone, E.  and Tewari, Y., "The Permeability of
    Laboratory Gloves to Selected Solvents," American
    Industrial Hygiene Journal, 39, Feb. 1978.
10. Sansone, E.  and Tewari, Y., "Penetration of Pro-
    tective Clothing Materials," American Industrial Hy-
    giene Journal, 39, Nov. 1978.
11. Williams, J., "Permeation  of  Glove  Materials By
    Physiologically Harmful Chemicals," American In-
    dustrial Hygiene Journal, 40, Oct. 1979.
12. Coletta, G.,  Swope, A., et al., "Development of Per-
    formance  Criteria  for  Protective  Clothing  Used
    Against  Carcinogenic Liquids," Occupational Safety
    and   Health  Administration,   Cincinnati,  Ohio,
    October, 1978.
13. Clayton, G.  and Clayton, F., editors,  "Patty's In-
    dustrial Hygiene and Toxicology," Volumes  1 and 2,
    John Wiley and Sons, New York, N.Y., 1978.
14. Sax,  N., ed.,  "Dangerous Properties of Industrial
    Materials,"  Fifth Edition, Van Nostrand Reinhold,
    New York, N.Y.,  1978.
15. "Industrial   Hygiene Field  Operations   Manual,"
    U.S. Department  of Labor, OSHA, Jan.  1980.
16. "NIOSH Manual of Analytical Methods," U.S. De-
    partment of Labor, OSHA, 1974.
17. "Analysis of Organic Air Pollutants by GC/MS,"
    U.S.E.P.A.,  June 1977.
18. Shen T. and Tofflemire, T., "Air Pollution Aspects
    of Land Disposal of Toxic Waste", Proc. of the 1979
    National Conference on Hazardous Material  Risk
    Assessment,  Disposal, and Management, Information
    Transfer, Inc., Silver Spring, Maryland, 1979.
19. Shen, T., "Emission Estimation of Hazardous Or-
    ganic Compounds from Waste Disposal  Sites," paper
    presented at the  73rd annual meeting  of Air Pollu-
    tion  Control Association, Montreal, Quebec,  June
    22-27, 1980.
20. Hwang, S.,  "Land  Disposal  Toxic  Air Emissions
    Evaluation   Guidelines,"  Office  of  Solid  Waste,
    U.S.E.P.A.,  December, 1980.

-------
     SHAKEDOWN AND PERFORMANCE TESTING OF THE
             EPA-ORD MOBILE INCINERATION SYSTEM

                                    JOHN E. BRUGGER, Ph.D.
                                        JAMES J. YEZZI, JR.
                                      FRANK J. FREESTONE
                           Municipal Environmental Research Laboratory
                               U.S. Environmental Protection Agency
                                          Edison, New Jersey
INTRODUCTION

  The EPA's mobile incineration  system was designed
and constructed under the Office of Research and De-
velopment's program to demonstrate the feasibility and
practicality of  destroying/detoxifying  hazardous  sub-
stances and wastes that have been improperly discarded at
landfills, spilled onto soil or stored  in above-ground con-
tainers. The basic purpose of the incinerator is to allow
cleanup to take place totally at the affected site,  thus
avoiding both transport and the ever present institutional
barrier that an area will become a permanent destruction
site.
  The system is purposely  over-designed to accommodate
diverse situations and can be further modified to handle
unique situations. It is expected that private interests will
make use of the plans, specifications and reports to con-
struct  and use versions that are more dedicated to par-
ticular, defined problems where the research  potential is
not required.
  Those skilled in  the art will  readily  determine what
changes should be made and will, undoubtedly, be able to
reduce both construction and operating costs. As an ex-
ample, operating cost savings can be achieved by better
utilization of waste heat.
SYSTEM DESCRIPTION

  The mobile incineration system is mounted on three
45-ft air-ride suspension trailers  for permit-free hauling
over Interstate Highways and improved roadways, as well
as for short-distance travel over unimproved roads or ter-
rain to a suitable  setup  site. The system is self-con-
tained,  except for consumable  supplies  (water,  fuel,
chemicals and power,  which  preferentially will be sup-
plied by commercial lines, but can be furnished by auxili-
ary diesel generators).
  To conform to applicable  regulations  and operating
permit requirements, gaseous  emissions are continuously
monitored or intermittently sampled to the extent required
and  all solid and liquid wastes  are collected, analyzed
and  treated, as  necessary,  before disposal. The system
was  specifically  designed to meet the stringent require-
ments (40-CFR-761) for the  incineration  of polychlori-
nated biphenyls (PCBs), namely, minimum residence time
of 2 sec at 2200 °F at 3% excess oxygen, and attainment
                      Figure 1.
   Mobile Incineration System. Trailer #1 (lower right) showing
   control booth, solid feed mechanism and kiln with ducting to
 Trailer #2 (center) showing secondary combustion chamber and
   flue gas quench with ducting to Trailer #3 showing paniculate
   filter, mass transfer scrubber, induced draft fan with diesel
       drive and base of stack, and diesel-electric unit.

of a combustion efficiency of 99.9% along with such re-
lated requirements  as limitations on the release of par-
ticulates and acid gases, destruction and removal effici-
ency and rapid shutdown in the case of malfunction  or
failure to meet the criteria of 40-CFR-761.
  The principal unit on Trailer No. 1 is a 52-in. ID  by
16 ft long A.P. Green "Kast-O-Lite 30 in.* refractory-
lined kiln having  specially  designed double layer, leaf
type, stainless  steel seals packed with graphite-impreg-
nated packing and rotated by a hydraulically driven motor
with speed and pitch adjusted to regulate the residence
time of solids.  The effective volume is 236 ft3. Firing is
with two tilted 4 in. Maxon burners with proportioned  air
control. Controlled tuyere (excess) air is also provided to
accommodate atomizer injected liquids, pumpable sludges
and ram fed solids.
  The hopper-fed ram has an adjustable stroke and fre-
quency. Liquid and solid wastes will be homogenized sep-
arately to ensure   homogeneous composition and heat
value, since  variations in feed composition  can lead to
highly  objectionable combustion transients.  Liquid and
solid wastes can be fed into the kiln concurrently. Temper-
ing water is injected into the kiln to control temperatures
and maximize throughput.
                                                  285

-------
286    TREATMENT & DISPOSAL


  The first  trailer also has a motor-powered hydraulic
system, forced air blower and the required piping and in-
strumentation. Thermocouples are mounted to record tem-
perature  (the maximum is  1800°F) and an oxygen con-
centration meter monitors the exit gas.
  The second trailer is designed to support the secondary
combustion chamber (SCC) which has the same bore and
refractory as the kiln, but is 36 ft long. The SCC is sur-
rounded with a shroud through which cooling air is passed.
The second trailer also has two Maxon burners and is de-
signed to meet the 40-CFR-761 requirements  for PCB
incineration.
  Gases pass from the  kiln, whose primary purpose is  to
initiate combustion and to insure that hazardous sub-
stances are fully volatilized  from any solids, through "in-
conel" ducting—designed  to handle  misalignment  and
temperature-induced dimensional changes. A swirl plate at
the entrance to the SCC aids development of turbulence
(the design Reynolds number is greater than 30,000). The
temperature and oxygen concentration are continuously
monitored at the end of the SCC. The residence time is
computed from feed and temperature data. The SCC trail-
er has its own air supplies and the necessary instrumenta-
tion and control equipment.
  The SCC terminates in a gas flow measuring venturi
alloy  nozzle that conducts the flue gas to a quench elbow
equipped  with water spray nozzles  that  drop the gas
temperature to approximately  180°F and remove
particulates and acid gases.
  The gas/vapor  is ducted  to a ground-level, baffled
sump and then ducted  upwards to the first element on
the third trailer, namely,  a "CHEAP"  paniculate re-
mover,  also  fabricated from an "Inconel" alloy. The
CHEAF collects particulates  on a sprayed fiberglass bdt
that is pulled through and re-rolled automatically when
the pressure drop (approximately 40 in. W.C.) exceeds a
set-point as the belt becomes  plugged  with  particulates
(down to the 0.3 /* range).
  The gas/vapor then enters a cross-flow caustic mass
transfer unit ("Ceilcote") (MX) packed with "Telerettes"
and scrubbed with  sodium carbonate or caustic soda as
required. Fluid levels in the  quench tank, CHEAF and
MX sumps are controlled,  pH is measured and  makeup
water or alkaline fluid is added as  required. A demister
section is incorporated into the MX. The ISO hp diesd-
driven induced draft  fan can develop the equivalent of
70-90 in. W.C.  for use at mile-high altitude sites and
consists of an "Inconel" shaft and 36  in. impeller with
aSS316Lhousing.
  A controlled reflow  duct connects the exit of the fan
to  the  inlet  of the CHEAF  to  maintain the correct
pressure drop across the CHEAF. The flue gas from the
fan passes  through a  sound suppressor and the stack,
which  has  ports  for  the  flue  gas  sampling that is re-
quired by regulations.  The  nominal gas flow  is 7200
                                                    Figure 2.
                               Field assembly of Mobile Incineration System showing (left to
                               right) Trailer #1 with feed and kiln, Trailer #2 with secondary
                                 combustion chamber and venturi quench and sump, and
                               Trailer 03 with paniculate filter, mass transfer unit (scrubber),
                                open control panel, draft fan and drive, and electric power
                                    supply. (Back-hoe at far left used in assembly.)

-------
                                                                            TREATMENT & DISPOSAL    287
scfm.  The  instrumentation  and  interlock  system  on
Trailer No. 3 is the most complex.
  The system  is designed to accept as feed those sub-
stances cited pursuant to Sections  311 and  307 of the
Clean Water Act, along with materials designated under
the Resource Conservation and Recovery Act. The feed-
ing of inorganic salts, which tend  to flux the kiln  re-
fractory, and of heavy metals, which form volatile oxides
or halides, must be controlled; data gathering on this  as-
pect  is an integral  part  of  the continuing shakedown/
testing.


SYSTEM TESTING

  At the time  this paper was written, shakedown was  es-
sentially  complete,  i.e.,   the  pumps, valves,  switches,
interlocks, etc. have been tested singly and as  a system,
but without firing the system.
  The current test plan has  five major phases, each of
which involves extensive sampling and analysis:
  (1) commercial fuel oil burn,
  (2) burn of fuel oil spiked with iron oxide particles,
  (3) incineration of fuel oil spiked at increasing levels of
     dichlorobenzene,
  (4) combustion of fuel  oil spiked  with increasing con-
     centrations of PCBs (or Aroclors) and
  (5) incineration of an oil-based waste leaching from a
     local landfill.
  Federal and State of New Jersey permits are both  re-
quired, as is assent from the municipality in which  the
test is conducted. Local residents have visited the facility
and will participate in a public hearing.
  The main purpose of the fuel oil burn is  to establish
baseline data.  Extensive gas  sampling, for example, with
the modified method 5 system and/or SASS train, will be
replicated.  The presence  of  polychlorodibenzodioxins
(PCDD) and  polychlorodibenzofurans  (PCDF)  will  be
carried to the limit  of   detection.  All  wastes  will  be
analyzed and  continuous or intermittent measurements
(as required) will be made  on  carbon dioxide, carbon
monoxide,  hydrochloric   acid,   sulfur oxides,  nitrogen
oxides, total hydrocarbons,  polluting organic hazardous
cfreinicals (POHCs), particulates, etc.
  ;the test with iron oxide-spiked fuel oil will assess the
efficiency of the paniculate removal capability of the
system. Many factors were taken into account  in choos-
ing avery fine iron oxide. Subsequently, testing is planned
With metal soaps, metal alkyls and metal aryls.
  la the third phase, orthodichlorobenzene will be added
to fuel oil in concentrations  of  1, 5 and  25% to assess
DREs, chiefly.
  In the fourth phase, based on the  outcome of the pre-
ceding chlorinated aromatic chemical tests, actual samples
of PCBs or  PCB-containing  fluids will be fed at concen-
trations of 0.05, 1,  10 and 20%. A planned fifth phase
will be based  on the effective incinerability of an oily
waste leachate contaminated with priority pollutants.
  In  all these  tests, the system  will be first brought to
temperature, which  requires  16-24 hours,  and then op-
erated as long as is required to collect meaningful samples.
The system will then be idled (not fully cooled down) un-
til the next phase is to be intiated. Excluding tests on a
proposed  leachate,  the test series will  require 45 days.
The data are expected  to yield a definitive array of com-
bustion and  operating conditions that can  be adapted
for use in detoxifying other hazardous substances.
  Since the mobile incineration system is a research unit,
provision  has been made for the incorporation of modi-
fications,  especially in  air pollution control,  should these
be indicated.

ACKNOWLEDGEMENTS

  The basic system was fabricated by MB Associates (now
Tracor-MBA), San  Ramon, California, under EPA con-
tract  68-03-2515. Especial  mention  should  be made of
the contribution of R. Tenzer,  Wm. Mattox, P. Kirke-
gaard, Wm. Ford, D. McKown, and their associates.
  The major part of  the system was delivered in Sep-
tember 1980 to an on-site operating unit in Edison, N.J.,
namely, the  Environmental  Emergency Response  Unit
(EERU) (a contractor-operated  facility), for completing
assembly,  undertaking  shakedown,  conducting tests  and
performing field demonstrations. EERU was operated by
Mason  & Hanger-Silas  Mason Company,  Inc.,  under
Contract No. 68-03-2647 until June 1981. Thanks are due
to M.  Sproul, G. Campbell,  V.  Manolio, F. Brown  and
the Mason & Hanger home office staff (Lexington, KY)
that prepared most of  the P&I diagrams and initiated in-
stallation  of  the instrumentation and control  equipment
and the various lines (electric, hydraulic, fuel, compressed
air).
  EERU  is  now being  operated by IT Enviroscience
under  Contract  No.  68-03-3069.  ITE has  current  re-
sponsibility  for  completion  of  assembly,  shakedown,
field  testing  and  operation.  The  contributions of  C.
Steuwe, C. Pfrommer, S. Anicito, A. Sherman and R.
Miller, on site, and of R.  Novak,  R. Lovell and many
others at ITE headquarters (Knoxville, TN) are acknowl-
edged.

REFERENCES

1. Tenzer, R., Ford, B.,  Mattox, W. and Brugger, J.E.,
   "Characteristics  of  the Mobile Field Use System for
   the Detoxification/Incineration  of Residuals from Oil
   and Hazardous Material Spill Cleanup Operations,"
   J. Haz. Materials, 3, 1979, 61-75.
2. Tenzer, R., Ford, B.,  Mattox, W. and Brugger, J.E.,
   "Mobile System for the Detoxification/Incineration of
   Cleanup Residuals from Hazardous Material Spills,"
   in  Disposal of Oil and Debris Resulting from a Spill
   Cleanup Operation, Am. Soc. for Testing and Mater-
   ials STP 703, 1980, 118-136.
3. Tenzer, R., Mattox,  W., Brugger, J.E.  and Free-
   stone,  F.J.,  "Design and Testing of a Mobile Incin-
   eration System for  Spilled or Waste Hazardous  and
   Toxic Materials," Proc. of the 1980 National Confer-
   ence on Control of Hazardous Material Spills, May
   13-15, Louisville, Kentucky, 1980,467-475.

-------
          HISTORY AND BENCH SCALE STUDIES FOR THE
TREATMENT OF CONTAMINATED GROUNDWATER AT THE
     OTT/STORY CHEMICAL SITE, MUSKEGON, MICHIGAN
                                        STEPHEN C. JAMES
                                U.S. Environmental Protection Agency
                             Solid and Hazardous Waste Research Division
                                           Cincinnati, Ohio

                                    ALAN J. SHUCKROW, Ph.D.
                                        ANDREW P. PAJAK
                                Touhill, Shuckrow and Associates, Inc.
                                       Pittsburgh, Pennsylvania
INTRODUCTION

  Hazardous leachates and contaminated ground and sur-
face  waters are often associated with  unsecured indus-
trial  waste storage and disposal sites.  Numerous prob-
lems  are encountered in the cleanup of such a site.  One
major problem is identifying the most effective treatment
technology for the contaminated stream. Contributing to
this problem are:
  (1) The  inability to characterize completely  the  con-
     taminated stream due to technical and economical
     limitations
  (2) The paucity of information on the effectiveness of
     techniques for treating the broad spectrum of or-
     ganic and inorganic compounds frequently present
     in these streams.
Therefore, the U.S. Environmental Protection Agency be-
gan a project now being conducted by Touhill, Shuck-
row and Associates,  Inc.  (TSA) to evaluate techniques
for concentrating hazardous constituents of aqueous waste
streams. A literature review, desktop evaluation and lab-
oratory bench scale experimental studies form  the basis
for judging the  potential of  numerous candidate tech-
nologies.
  During the course  of this project, a part of the lab-
oratory experimental work was carried out at  Cordova
Chemical Company, present owners of the Ott/Story Site,
using ground water contaminated by prior operations. This
site provided an opportunity to interface research activ-
ities with development of a treatment system for a  high
priority problem site.
  In  this paper, the authors provide a  brief background
on the site history, a method selection of candidate treat-
ment technologies and summary of results of the experi-
mental studies.

SITE HISTORY

  During 1957, Ott Chemical  Company began  produc-
tion of various organic chemicals in Dalton Township,
Michigan, north of the City of Muskegon.'" Many of the
organic chemicals produced used phosgene as a raw ma-
terial. At the time of plant startup, Michigan State Water
Resources Commission permitted Ott Chemical Company
to discharge wastewaters into seepage beds, provided that
certain limitations on the types and amounts of substances
disposed were adhered to and that the discharge was mon-
itored regularly and results reported to the state.
Permit Violations

  Beginning in 1957 and continuing through 1967, permit
violations were evident. Initially,  chlorides and phenols
were detected in monitoring and water supply wells. Ott
Chemical Company attempted to mitigate pollution prob-
lems, by using a groundwater purge system to contain the
contaminated plume on company property, and by re-
ducing contaminant concentrations in their discharges to
the seepage beds. The State permitted discharge of purged
groundwater to a nearby creek, provided there was no im-
pairment of stream's beneficial use.
  Although the remedial action program reduced contam-
inant concentrations going into groundwater and limited
pollution migration, other problems became evident.  Fish
tainting because of purge water discharge was observed in
the nearby creek, and other groundwaters in the area were
polluted by stored or disposed wastes.
  In recognition of the continuing  and spreading problem
and the fact that  remedial actions were  not successful
completely, the State began, in 1965, water quality inves-
tigations to document the extent  of contamination. At
least 40 organic chemicals, some of which are considered
to be hazardous, were identified in the groundwater.
  Remedial actions  continued but in 1968 the State
ordered  discontinuance of direct  discharge of  process
wastewaters to the groundwater. Although cooling water
discharges  were  still permitted to be  discharged to the
seepage  beds,  treatment of process waters prior to dis-
charge to the Muskegon River was required. Nevertheless,
wastes stored on site and spilled in the plant area continued
to exacerbate and  spread  the problem. In 1974, process
wastes and  purged groundwater were connected to the
municipal wastewater treatment system.
                                                 288

-------
                                                                             TREATMENT & DISPOSAL     289
  Up to 1976, attempts to improve waste handling prac-
tices at the site continued. Some of these efforts included:
waste  incineration,  development of  spill  prevention
measures and plans for deep well disposal. However, con-
trol of the  contaminated groundwater plume was  not
accomplished and the pollutants continued to migrate to-
ward residential  water supply wells down-gradient from
the plant site. A small nearby trout system became severe-
ly degraded by contaminated groundwater exfiltration.

Ownership Change

  Solutions  to  the contamination problems  were hin-
dered, to some extent, by operating entity and ownership
changes between 1965 and 1973. In the 1972-1973 period,
Story Chemical Company became owner of the facility.
Ultimately, during 1976,  Story Chemical Company filed
for reorganization in Federal Bankruptcy Court, abandon-
ing the plant site in September 1977. Site abandonment
caused additional concern because of 23,000 pounds of
phosgene gas left in two containers on site.
  In October 1977,  Cordova Chemical Company pur-
chased the facility, and cooperated with the State in site
cleanup,  including  providing  funds  to furnish  an  al-
ternate water supply  for  the area. In  return, the state
agreed to limit the new owner's liability for previous prob-
lems.
  In 1979, groundwater  contamination was  found at a
depth of 30 m, and the contaminated plume was 400 m
wide. Furthermore, wells in the area were deemed unpot-
able and residents were advised to use bottled water for
drinking.
  Ott/Story operations left a number of unresolved con-
cerns. In particular, large volumes of organic chemical mi-
grating through soil and ground and surface water contin-
ued to have an adverse environmental impact.  It was with-
in this context that bench scale studies were conducted to
uncover suitable treatment methods for decontaminating
polluted groundwater.
                      Table I.
               Treatment Process Trains

Process Train Sequence  Ancillary Processes
Biological followed by
carbon sorption
pH adjustment, coagulation,
settling before biological for
heavy metal control. Clarifi-
cation, filtration before carbon
for solids removal
Waste Type

High TOC, low in toxics (to
biomass)
Carbon sorption followed pH adjustment, coagulation,
by biological        settling before carbon to
              reduce solids loading
Biological with
powdered activated
carbon included
Membrane followed by
biological
Stripping followed by
rartron sorption
Biophysical system pro-
ceeded by coagulation and
clarification to minimize
heavy metals and solids
loading

pH adjustment, clarifica-
tion and filtration before mem-
brane to prevent membrane
fouling

pH control before stripping.
Clarification and filtration
before carbon for solids
removal
                    High TOC, high in toxics (to
                    biomass
                    High TOC, high in toxics
Membrane removes high
molecular weight organics and
inorganic ions; biological re-
moves remaining organics

Stripping to remove volatile
organics and reduce organic
load to carbon; carbon to
provide further organic removal
LITERATURE REVIEW/DESKTOP EVALUATION

  Results of the literature and desktop  evaluation have
been reported elsewhere.<2'3> The following list of candi-
date concentration technologies was developed:
BIOLOGICAL TREATMENT
CARBON ADSORPTION
CATALYSIS
CENTRIFUGATION
CHEMICAL PRECIPITATION
CRYSTALLIZATION
DENSITY SEPARATION
DIALYSIS/ELECTRODIALYSIS
DISTILLATION
EVAPORATION
FILTRATION
FLOCCULATION
ION EXCHANGE
RESIN ADSORPTION
REVERSE OSMOSIS
SOLVENT EXTRACTION
STRIPPING (AIR AND STEAM)
ULTRAFILTRATION
The processes that had the greatest range of applicability
included:
BIOLOGICAL TREATMENT
CHEMICAL COAGULATION
CARBON ADSORPTION
RESIN ADSORPTION
STRIPPING
MEMBRANE PROCESSES
  It also was concluded that  these unit processes would
have to be assembled in process trains (containing more
than one of these unit operations) to achieve the degree of
treatment required  at most sites. Five  possible process
trains were formulated (Table  I). These trains do not rep-
resent the only possible configurations; however, they  do
provide broad range applicability to a wide variety of con-
taminants.
  These process combinations were developed with data
from single compound studies.  Actual treatability  of
multi-compound streams generally  will require bench-scale
studies  using actual wastewater. Because of the specific
contaminants at the Ott/Story site, chemical coagulation
and membrane processes were judged to have limited ap-
plication and were not studied in the laboratory.

BENCH SCALE STUDIES OF UNIT PROCESSES

  In laboratory bench scale studies, individual unit pro-
cesses were examined first; then process  trains were eval-
uated.  Wastewater used  in  these studies was contam-
inated  groundwater at the Ott/Story Site collected from
two wells located in the contaminant plume. For a portion
of the studies, an equal volume  composite of water from
these two wells was used;  in other studies the water from
each well was used separately. Analyses of the composite
sample  and water from the separate wells are shown in
Table II. Although the water from  well W-17d has a much
lower total organic carbon (TOC) concentration,  it con-

-------
290     TREATMENT & DISPOSAL
                                                    Table II.
                                      Groundwater Quality at the OTT/Story Site
      Conventional  Pollutants:

        pll

        COD (mri/1)

        TOC (m.'|/l)

        NH3-N (mg/1)

      Ma^or Priority  Pollutants:
             (in ug/1)

        Vinyl chloride

        1, 1-dichloroethylene

        1,1-dichloroethane

        1, 2-dichloroethane

        Benzene

        1,1,2,2-tetrachloroethane

        Toluene
   Composite  Sample
(50%  OW-9  &  50% W-17d)


      9.3  -  12

   5400

    400    -  1500

     64
       ND  -  32,500

      5   -  6,570

       60  -  19,850

      350  -  111,000

       ND    7,370

        5    1,590

        5    5,850
tains many of the same organic priority pollutants as well
OW-9, frequently at similar concentrations.
  Results  of many of these experimental  studies have
been described in detail elsewhere.(2- "• 5|  6) In this  paper,
a brief summary is presented.

Steam Stripping

  Continuous flow, steam stripping experiments were con-
ducted using a composite sample. The TOC concentration
in the stripper bottoms ranged from 300  to 400 mg/1 and
appeared to be independent of the overhead:  feed ratio.
Overhead TOC concentration approached 3500 mg/1 at an
overhead:feed  ratio of 0.06. The average TOC reduction
between feed and bottoms was 34%.
  These results indicate a major constraint associated with
steam stripping, i.e.,  it is necessary to further treat a bot-
tom waste stream having a flow only slightly less than the
feed flow. Because the TOC concentration appeared to be
independent of system flow rates and feed TOC levels over
the ranges examined  and because additional treatment of
the bottoms would be necessary, steam stripping would
likely be a costly,  yet only  moderately effective, concen-
tration process.

Adsorption by Granular Activated Carbon and Resin

  Batch isotherm  studies were used  to  evaluate various
sorbents and operating conditions.  Then continuous flow
granular activated carbon (GAC) studies were conducted
off pH 9.3 to 10.0, using Filtrasorb 300  GAC.
  In continuous flow  studies three  or four glass columns
arranged  in  series with sampling  ports located  at the in-
fluent and effluent ends of each column  were used. Each
column was 122 cm by 2.5 cm (ID) and  contained 91 cm
of GAC. The system was operated in a downflow mode at
a loading rate of approximately 1.35 l/m3s (2 gal/min/ft).
This provided  an  empty bed contact time (EBCT) of
approximately  15 min. per column.
  Influent TOC concentration varied substantially, rang-
ing from 316 to 950 mg/1. Generally, after only 3 to 10 bed
    Well  OW-9


   9.6-  1105

5200   -  8300

1500     2400

  40     163
    ND

    ND - 2600

    ND - <0.1

  >500 - 2000

   600 - 900

    ND - <1

 3,300   5,600
   Me 11 W-17d


   9.5   9.6

1600     2600

 171     730

   7.9   16
    ND  -  5, 240

    ND

    ND    6,570

 3,670    >1000

 >5000     25,000

    ND  -  1, 590

 3,610  -  7000
                  volumes (BVs), TOC removal decreased to 50%. TOC
                  leakage reached 90% after about 200 to 240 BVs were pro-
                  cessed and continued at this level until up to 500 BVs
                  had been processed. As would be expected,  effluents from
                  columns 2 and 3 were intermediate between  columns 1 and
                  4. Column performance data correspond to  batch sorption
                  study data.
                    The difference between  treatabilities of  the composite
                  groundwater and waters from the individual wells is shown
                  in Figure  1. Organic materials in  W-17d (as reflected  by
                  TOC concentration)  were  more readily  sorbed by GAC
                  than was  TOC in the composite  groundwater or OW-9
                  groundwater.
                    Continuous  flow  resin  adsorption  studies were con-
                  ducted on  the composite  sample  using  XE-347 carbon-
                  aceous resin. Again, these  continuous runs  were preceded
                  by batch isotherm studies.
                    Three columns similar to  those used for GAC studies
                  were charged with 792 to  835 cm3 of resin and were op-
                  erated at  loading rates  of 2.95 to 3.79  BV/hr. EBCT
                  ranged from 6 to 20 min.
                    Breakthrough  characteristics  were similar to those of
                  the GAC studies except that XE-347 TOC removal de-
                  clined more rapidly.  TOC  removal diminished to 50%
                  after about 5 BV were processed  and appeared to stabil-
                  ize at about 10% for at least 120 BV.

                  Biological Treatment

                    Several attempts were made to acclimate an  activated
                  sludge culture to raw groundwater. All attempts, how-
                  ever,  were minimally successful.  Neither  a conventional
                  activated sludge nor a commercial microbial culture could
                  be acclimated. A light colored, filamentous biomass which
                  settled poorly appeared to  be indigenous to the water
                  and inhibited a healthy biomass.
                    Approximately 80%  TOC reduction  was achieved;
                  however,  stripping due to aeration appeared to account
                  for about two-thirds  of this removal. Addition of trace
                  elements and nutrients, and pH adjustment to pH 7.0 to
                  7.5 did not aid microbial  acclimation to  raw ground-

-------
                                                                         TREATMENT & DISPOSAL    291
                                   Composite Groundwater (OW9 L Ill/d)
                                   Raw OH9 Groundwater
                  150    200    250

                   TOC Loaded (mg TOC/g GAC)
                     Figure 1.
               TOC Adsorption by GAC
water.  Addition of  powdered  activated carbon (PAC)
at aeration chamber concentrations  of approximately
10,000 mg/1 also did not aid acclimation to raw ground-
water or improve TOC removal or mixed liquor appear-
ance.

BENCH SCALE STUDIES OF PROCESS TRAINS

Granular Activated Carbon/ Aerobic Biological Treatment

 Because of the  apparent  toxicity  of groundwater to
biological processes and the rapid breakthrough of TOC
in adsorption systems, a process train consisting of  gran-
ular activated carbon (GAC) followed by activated sludge
(AS) biological treatment was  investigated. It was  ex-
pected that GAC  would  protect the biological system
from toxic materials while organics leaking from the  GAC
would  be  degraded or stripped  during the  biological
process.
 Carbon pretreatment of raw groundwater  permitted
development of a  culture of aerobic organisms capable
of further treating  GAC effluent. In excess of 95 %  TOC
removal was realized by this process  during  the period
when TOC  removal by the  GAC  exceeded 30%.  After
this initial period,  process train performance declined as
carbon performance declined. Data indicate that:
 (1) Some fraction of TOC which initially is  sorbed by
    carbon begins to leak through the system  after a
    short period of operation
 (2) The fraction of TOC which leaks through the car-
    bon system is not toxic to the biological  system but
    does not appear to  be removed or reduced either
    biologically or by air stripping associated with  AS
    aeration
 ($) Operation of the biological process at hydraulic re-
  , tention times ranging from 4 to 16 hr, with or  with-
    out powdered activated carbon, or with  or without
    Phenobac seems to  have little impact on process
    performance (based upon TOC removal)
  (4)  Overall system performance was maintained at 75-
      85% TOC removal  (effluent TOC of 100 to 185
      mg/1) for about 21 days (46 retention times for bio-
      logical system and 110 BVs for carbon).
  As shown in Table III, organic priority pollutant analy-
ses  conducted during  operation of  the  GAC/AS  Sys-
tem indicated that almost all  of the organic priority
pollutants detected in the raw groundwater were removed
consistently to less  than the level of detection (0.01 mg/1)
by the system. Additionally,  the AS  process completely
removed the  few  organic  priority  pollutants  leaking
through the GAC system even during the phase of oper-
ation when overall TOC removal was declining. Analysis
of biological sludge showed no organic priority pollutants
at a 0.01 mg/1 detection level.

Granular Activated Carbon/Anaerobic Treatment

  A GAC/upflow  anaerobic filter (UAF) system was in-
vestigated using the  composite  groundwater and water
from  well  OW-9  (both high TOC waters). UAF  per-
formance  declined  as  GAC  performance  declined.
Changes in UAF organic loading rate did not  appear to
affect UAF performance.
  Overall,  the performance of the GAC/UAF process
train was slightly poorer than the GAC/AS process train
achieving a maximum TOC removal of 81%  when treat-
ing  the composite groundwater.  During this period TOC
removal averaged 66%  and ranged from 38 to 81%.  The
average TOC removal by the GAC process was  31 % with
a range of 10-46%; the  average TOC removal by the UAF
process was 50% with a range of 12-67%.

GAC/Trickling Filter Treatment

  The GAC/TF process train was investigated because
operation was expected to be easier and less energy inten-
sive than activated sludge  and also less  likely  to induce
stripping of volatile organic compounds.
  An  unseeded trickling filter provided little  or no re-
moval of TOC from GAC pretreated groundwater. A new
filter,  initially seeded with  activated  sludge recycled for
about four weeks, developed a significant biomass on the
filter  media.  Although TOC removal was substantially
better than for  the  unacclimated  system, performance
was poorer than achieved by the GAC/AS process train.
Moreover,  as GAC performance steadily decreased, an-
aerobic conditions developed in the TF.

Granular Activated Carbon/Anaerobic/
Aerobic Biological Treatment

  The GAC/UAF treatment process discussed earlier was
modified by the addition of an  activated  sludge process.
When TOC leakage from  the  GAC process  was con-
siderable,  the aerobic process reduced the UAF effluent
TOC by 50% or more. Otherwise, little additional TOC
removal was  achieved  by  the AS process. The  process
train provided TOC  removals of 86  to 100%. The im-
portance of GAC pretreatment  was demonstrated when,

-------
292    TREATMENT & DISPOSAL
                                                   Table III.
                                TOC and Priority Pollutant Data for Granular Activated
                                    Carbon/Activated Sludge Process Train (mg/l)
Compound



TOC
Total Cyanide
CN
Total Phenol
Methylene chloride
1, 1-Dichloroethene
1, 1-Dichloroethane
Trans-1,2-dichloro-
ethane
Chloroform
1 , 2-Dichloroethane
1,1, 1-Trichloroethane
Trichloroethylene
Benzene
1, 1,2-Trichloroethane
Perchloroethylene
Toluene
Chlorobenzene
Phenol
2-Chlorophenol
2,4-Dichlorophenol
1,2-Di chlorobenzene
Dibutyl phthalate
Raw
Ground-
water
9-16-80
637
NA
NA
NA
2.1
1.6
2.4
0.06

9.8
72
7.6
0.06
1.2
0.11
0.49
2.3
0.23
0.025
0.040
0.010
0.085
ND
GAC
Effl.

9-16-80
380
NA
NA
NA
0.029
ND
ND
ND

ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Raw
Ground-
water
9-23-80
929
0.11
0.05
16
14
0.06
0.17
0.04

0.70
25
0.39
0.03
1.5
0.07
1.9
0.97
0.29
0.028
0.036
0.010
0.077
ND
GAC
Effl.

9-23-80
604
0.21
0.05
0.16
0.01
0.01
0.02
ND

0.06
1.4
0.04
ND
0.02
ND
ND
0.05
ND
ND
ND
ND
ND
ND
AS
Effl.

9-24-80
90
0.23
0.05
0.10
ND
ND
ND
ND

ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.05
GAC
Effl.

10-1-80
770
0.23
0.05
0.10
0.16
ND
ND
ND

ND
0.05
ND
ND
ND
ND
ND
0.01
ND
ND
ND
ND
ND
ND
AS
Effl.

10-1-80
183
0.20
0.05
0.10
ND
ND
ND
ND

ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
      NA—not analyzed
      ND—not detected
      No other priority pollutants detected at 0.01 mg/l detection limit

within one day after carbon pretreatment was terminated,
UAF process  performance  declined  markedly. Results
from this series of tests indicated that:
  (1) As the UAF process  became acclimated, its per-
      formance improved and it  accounted for a  larger
      share of TOC removal.
  (2) As the carbon column performance declined, UAF
      performance also declined.
  (3) Generally, the AS process provided minimal TOC
      removal.  Aerobic biodegradation, stripping or ad-
      sorption onto mixed liquid solids were negligible.
      However, a normal  appearing,  well-settling acti-
      vated sludge biomass, albeit one with a  relatively
      low MLSS (mixed liquor suspended solids) and low
      sludge production, was observed.
  (4) As TOC leakage from the carbon column increased,
      overall system performance declined; however, the
      amount of removal attributable to the aerobic pro-
      cess increased.
  (5)  The overall system, under the operating conditions
      studied,  generally did not perform  as well as the
      GAC/AS train, i.e., it did not maintain low effluent
      TOC levels (<  50 mg/l) for as long as the GAC/AS
      train. However, both  systems appeared to be able to
      maintain  effluent TOC levels below 100 mg/l for
      equivalent duration.

Ozonation Treatment Studies

  The effects of ozone (03) pretreatment  were evaluated
using O/AS and O3/GAC/AS  process trains. Ozonation
conditions were as follows:
•Air served as ozone source.
•03 dose—2g/hr.

-------
                                                                          TREATMENT* DISPOSAL    293
•Oi flow rate—2 liters/minute.
•Batch 63 contact time—1 hour/3 liters of groundwater.
  The process trains involving ozone pretreatment did not
perform as  effectively as the  other trains studied thus,
suggesting little benefit to pre-ozonation (in terms of TOC
removal aside from  that  attributed to stripping  during
ozqriation).  TOC removal by the O3/AS  train  ranged
from  37 to  60% while removal  by  the  O3/GAC/AS
train ranged from 42 to 88 %.
  While little change in TOC concentration may occur as
the result of  ozonation  alone, ozonation might  be ex-
pected to enhance either the biodegradability or adsorp-
tion of organics. However, pre-ozonation did not  appear
to improve the performance nor increase the longevity of
carbon treatment on the basis of TOC breakthrough.
CONCLUSIONS

  Desktop evaluations identified a number of concen-
tration technologies that may  be applicable for treating
hazardous leachates and contaminated ground and sur-
face waters. The most  promising,  broadly applicable
technologies were:
•Adsorption: carbon and resin
•Biological treatment (aerobic and anaerobic)
•Biophysical treatment
•Chemical coagulation
•Membrane processes
•Stripping: air and steam
  These unit operations then were studied separately and
in combinations using groundwater from an actual haz-
ardous waste contamination site.
  Carbon pretreatment of the raw groundwater permitted*
development of both aerobic and anaerobic biological pro-
cesses apparently by removing pollutants toxic to the bio-
logical processes. Highest TOC  removals were achieved
with an  activated carbon/aerobic biological (activated
sludge) system. However, as the  carbon performance de-
clined, so did that of the rest of the system. The GAC/an-
aerobic biological process train performed somewhat more
poorly than the GAC/activated sludge system. Studies of
ozonation did not imply any advantage as a pretreat-
ment operation. The  process of  ozonating  yielded TOC
removals similar to that achieved by air stripping.
ACKNOWLEDGEMENT

  The  work upon which  this paper is  based  was per-
formed by Touhill, Shuckrow and Associates, Inc., under
Environmental Protection Agency Contract #68-03-2766.
  Cordova Chemical Company, the present owner of the
Ott/Story Site, has been extremely cooperative in the con-
duct of this study. Special thanks to Mr. Barrel Cardy of
Cordova Chemical Company for his assistance toward the
completion of this study.

REFERENCES

 1. Klepper, G., "Groundwater Pollution from the Ott/
    Story Chemical Company Operations, Dalton Town-
    ship, Muskegon County, Michigan." Michigan De-
    partment  of  Natural  Resources,  Water  Quality
    Division, June 1981.
 2. Shuckrow,  A.J.,  Pajak, A.P.,  Osheka, J.W. and
    James, S.C., "Bench-Scale Assessment of Technol-
    ogies for  Contaminated  Groundwater Treatment,"
    Proc. US.EPA National Conference of Uncontrolled
    Hazardous Waste Sites, Oct. 15-17, Washington, DC,
    HMCRI, Silver Spring, Md., 1980,184.
 3. Shuckrow,  A.J., Pajak,  A.P.  and James,  S.C.,
    "Concentration Technologies for Hazardous Aqueous
    Waste  Treatment," EPA-600/2-81-019,  U.S.  EPA,
    Cincinnati, Ohio, 1981.
 4. Pajak,  A.P., Shuckrow, A.J., Osheka,  J.W.  and
    James, S.C., "Assessment of Technologies  for Con-
    taminated Groundwater Treatment," Proc. of the In-
    dustrial  Wastes  Symposia, Las  Vegas,  Nevada,
    September 28-October 3,1980.
 5. Pajak,  A.P., Shuckrow, A.J.,  Osheka, J.W. and
    James, S.C., "Concentration of Hazardous Constitu-
    ents of Contaminated Groundwater," Proc. of the
    Twelfth Mid-Atlantic  Industrial Waste Conference,
    Bucknell   University,   Lewisburg,   Pennsylvania,
    July 13-15,  1980.
 6. Shuckrow,  A.J. and Pajak, A.P.  "Bench Scale As-
    sessment of Concentration Technologies for Haz-
    ardous Aqueous Waste Treatment." EPA-600/9-81-
    002 b,  Land Disposal: Hazardous Waste., Proc.  of
    the Seventh Annual Research  Symposium,  Philadel-
    phia, Pennsylvania, March 16-18,1981.

-------
                 DESIGN OF A TREATMENT SYSTEM FOR
                                HAZARDOUS RUN-OFF

                                    PETER B. LEDERMAN, Ph.D.
                                        JOHN W. HAMMOND
                                          Roy F. Weston, Inc.
                                      West Chester, Pennsylvania
                                  JOSEPH P. LAFORNARA, Ph.D.
                                      Emergency Response Team
                                 U.S. Environmental Protection Agency
                                           Edison, New Jersey
INTRODUCTION

  On July 30, 1979,  the State of Pennsylvania requested
that the U.S. Environmental Protection Agency  (EPA)
conduct an on-site assessment  of an abandoned mine
drainage tunnel, known as the Butler Tunnel, from which
an unknown  discharge  was entering  the  Susquehanna
River. Approximately 500 to 1,000 gal/day of oily sludge,
along  with an  assortment  of  cutting oils,  aromatics,
substituted phenols, alkyl resins and short and branched
chain hydrocarbons were being discharged from the tun-
nel. This discharge was traced to an alleged illegal dump-
ing of  millions of gallons of toxic wastes into an  aban-
doned  coal mine through a sewer borehole at Dupont,
Pennsylvania, two miles inland from the river.
  Roy  F.  Weston,  Inc. was retained  by EPA  to  de-
termine alternative approaches  to treatment and/or de-
contamination of the toxic oily material that might be
found  in coal mine  pools  being drained by the tunnel,
as a result of exploratory drilling (by other EPA con-
tractors) at Pittston, Pennsylvania. In addition,  Weston
was to conduct a similar  investigation on  the Butler
Tunnel. For  both  of  these investigations, the  consul-
tant's work was to include physical and chemical char-
acterizations of the contaminated material,  and an evalu-
ation of disposal techniques for treating the wastes.
  The project had four phases:
  Phase I: pertinent available data on mine pool wastes
  were to be collected and reviewed. Analytical protocols
  for characterization of the mine pool wastes were to be
  developed.  Potential alternatives for concentration and
  disposal of the wastes were to be determined.
  Phase 2: data from  treatment  studies previously per-
  formed on the effluent from the Butler Tunnel were to
  be evaluated. Supplemental treatment studies, if neces-
  sary, were  to be  recommended.  Alternative methods
  for  temporarily treating the  effluent were also to  be
  recommended.
  Phase 3: waste samples collected  from the mine polls
  were to be  analyzed.  Cost  estimates were  to be de-
  veloped for feasible treatment alternatives.
  Phase 4: supplemental analytical and treatability studies
  on the Butler Tunnel  effluent  were to be performed,
  which  were necessary  to recommend  a  temporary
  treatment scheme for the tunnel's effluent.
  Reports summarizing the results  of Weston's overall
project as well as an overview of the over all clean-up op-
eration have been presented elsewhere.ll>2)
  The purpose of this paper is to present a description of
the treatment process train which was designed to remove
hazardous pollutants from the Butler Tunnel effluent. The
development  of the design  basis, results of treatability
studies, the design and costs for installation and operation
of the equipment are presented.

CHARACTERIZATION OF TUNNEL DISCHARGE
  Since August  of  1979, when  the Butler Tunnel clean-
up program  was initiated, samples of the Butler Tunnel
discharge were systematically collected and  analyzed for
oil and several classes  of toxic  organic chemicals. These
included  analysis for dichlorobenzenes, disctyephthalate,
alkylphenols, toluene, xylene, ethylbenzene and naphtha-
lene. After a period of preliminary screening of the efflu-
ent, dichlorobenzene (DCB) was chosen as an indicator
parameter in order to reduce the analytical load to a man-
ageable level. In addition to these analytical data, the dis-
charge flow rate from the tunnel to the Susquehanna was
measured.
  Presented  in Figure  1 is the cumulative distribution of
flow rates from -the Butler Tunnel from August 1979 to
April 1980. As indicated, the flow rates during this period
ranged from approximately 1 mgd to greater than 12 mgd.
Statistically, on 20%  of the days the flow  rate was less
than 3.0 mgd, on 40%  of the days it was less than 6.0 mgd
and on 80% of the days it was less than 12.0 mgd.
   Presented in Figures 2 and 3 is a comparison of the oil
and DCB loads (lb/day) discharged versus  the flow rate
(mgd) of the  discharge on the same day. During this per-
iod, the lowest flow which could be accurately measured
was 2 mgd, therefore this flow rate was assumed for days
on which the  flow rate  was reported as "less  than 2 mgd."
A review of Figures 2  and 3 shows that the most signifi-
cant discharges of oil and DCB, in terms of lb/day to the
Susquehanna River, occur when  the  tunnel flow rate is
low. The sensitivity level for oil was estimated to be 1.0
to 1.5mg/l.
                                                   294

-------
                                                                               TREATMENT & DISPOSAL    295
     "O 12.0
     Ol
      5 10.0
      o
             There Were 31 Days Where the Flow:
             Were Only Specified as > 12 mgd
           This Point Represents 16 Days
           Where Rows Were Estimated as*,
           2 to 9 mgd
                         This Point Represents 52 Days
                        ^-Were Rows Were Estimated as
                         t (o 2 mgd
                       Percentile
                        Figure 1.
              Cumulative Distribution of Flows
           Pittston, Pennsylvania (Butler Tunnel)
                 August 1979 to April 1980
  The results of priority pollutant analysis of a sample of
the tunnel effluent are presented in Table I. On the day
that this sample was collected, the discharge was in excess
of 12 mgd. These data confirm that the Butler Tunnel has
been contaminated by various solvents and other organic
species. Review of the data in Table I shows that the base-
neutral extractable fraction represents approximately 90%
of the total organics reported on the table.
TREATMENT REQUIREMENTS

  In general, the analysis of the data for the Butler Tun-
nel discharge confirmed what was  previously  suspected.
The principal contaminants  in  the  discharge are oil and
dissolved organics.
  Regulations governing the quality of this discharge are
found in Chapter 93—Water Quality  Standards, of the
Pennsylvania Code, Title 25—Environmental Resources.
Review of this  chapter indicated that there were no spe-
cific water quality criteria for the principal contaminants
in the Butler Tunnel discharge.  Therefore,  Section 93.8—
Development of Specific Water Quality Criteria for the
Protection of Aquatic Life, requires that 96-hour continu-
ous flow bioassays be run,  using  the  methodologies re-
ferenced in  Subsection 93.8e of Chapter  93.  However,
during the  analytical program,  bioassays  were not  run.
Therefore, it was assumed, based on conversations with
EPA, that treatment of the Butler Tunnel wastes for oil
and  organics,  such as those in Table  I, would be re-
quired to levels which represented reasonable reduction of
the initial concentrations.

TREAT ABILITY INVESTIGATION

  While developing the preliminary treatability plan, vari-
ous  specific  aspects  of the  overall emergency  program
were considered including:

1. In order to successfully treat, for all the contaminants
   in Table I, it would probably be necessary to use acti-
   vated carbon. Alternatives to activated carbon which
   were initially considered, but then rejected, included:
   •Air stripping of the organic  species—while this might
    be reasonably effective on those species  found in the
    volatile fraction in Table I,  it would probably  have
    little effect on the base-neutral extractable  fraction,
    which  represented around 90% of the total organics
    present.
   •A low-level biological system—this  was rejected be-
    cause of the low values  for BOD5 in the raw waste;
    also the  significant area requirements that would be
    needed to treat the probably flow rates required. (The
    available space for on-site treatment was  limited to an
    area of 250 ft x 70' on a  ledge overlooking the mouth
    of the tunnel.)
2. If activated carbon was to be used, then it would be
   necessary to minimize  any fouling problems due to oil,
   or material which periodically sloughed from inside the
   tunnel.  For this reason,  Weston  decided that  solids
   separation facilities were necessary before  the activated
   carbon  system and that it would also be  necessary to

                         Table I.
                Organic Analysis (GS/MS)
                      Butler Tunnel
                     March 20,1980
                  Pittston, Pennsylvania
A. Base-Neutral Extractables (yug/L)
   1,2-Dichlorobenzene
   Naphthalene
   Diethylphthalate
   Bis (2-Ethylhexyl) Phthalate
B. Acid Extractables (/ug/L)
   None detected
C. Volatile Fraction (/zg/L)
   Methylene Chloride
   Trans-l,2-Dichloroethylene
   Chloroform
   1,1,1 -Trichloroethane
   1,1,2-Trichloroethane
   Trichloroethylene
   Benzene
   2-Chloroethylvinyl Ether

D. Pesticides
   None detected
164.0
  7.6
 13.4
192.9
  1.6
  6.5
  1.3
  3.4
 16.7
  6.8
  1.3
  2.1

-------
 296     GEOHYDROLOGY MATH MODELING
     periodically inject chlorine, for disinfection of the ac-
     tivated carbon.
  3.  If activated carbon was going to be used, pH adjust-
     ment of the raw wastewater may be necessary.
    To implement the treatability plan, the following tests
  were run:
  •Activated carbon isotherms.
  •Chlorine demand.
  •Titration curves.
    In order to economically conduct this preliminary work,
  DCB was chosen as an indicator parameter for analysis of
  the isotherms.
    Presented in Figure 4 is an isotherm run on the tunnel
  effluent. As indicated in the figure, activated carbon may
  be used to treat  the Butler Tunnel wastewater  to DCB
  equilibrium concentration of less than lOMg/1- To establish
  chlorine residual of 2 mg/1, a chlorine dosage of 8 mg/1
  was required. To adjust the  pH of the  raw wastewater
  from 4 to neutral,  approximately  40 mg/1  of caustic  (50
  mg/1 as CaCo3) was required.
&
X
ra
•o
1.
                        Flow (mgd)

                          Figure 2.
                          Oil Load
              Pittston, Pennsylvania (Butler Tunnel)
                   August 1979 to April 1980

   PROCESS DESIGN

     As a basis for its concept design, Weston used the re-
   sults of its data analysis and treatability tests and, in addi-
   tion, incorporated the following factors:
   •Based on conversations with  the EPA project officer, it
   was decided  to investigate designs for two  flow rates—
   -The most significant discharges of contaminants from
    the Butler Tunnel have occurred on days when the flow
    rate was 3.0 mgd and 6.0 mgd. The reasons for this are:
   -The construction of facilities much in excess of 6.0 mgd
    might be prohibitive from an economic point-of-view.
   •The limitations on available land area,  for an on-site
   facility.
     Presented in Figure 5  is the general process flow dia-
   gram for the Butler Tunnel treatment system. As indicated
I
I
                        Flow (mgd)

                         Figure 3.
                        DCB Load
             Pittston, Pennsylvania (Butler Tunnel)
                  August 1979 to April 1980
  in  the  diagram, the Butler Tunnel discharge would be
  treated in the following sequence:
  •Water from the tunnel will flow into a wet well through a
   bar screen that will remove  large solids. This  wet well
   will be constructed at the mouth of the Butler Tunnel,
   below the ledge.  The west well will have an emergency
   overflow provision to the Susquehanna River, so that
   flows in excess of the design flow can be bypassed.
  •Water will be pumped from the  wet well to a flow di-
   vision box. The purpose  of the flow division box is to
   divide the  flow before being fed to  several corrugated
   plate  separators,  operating in parallel, that will remove
   suspended solids  and  floatable oils. It will be important
   in the final design of the facility to seiect lilt pumps,
   whose operation will keep  oil/water  emulsions  to a
   minimum.
  •As described above, the corrugated plate separators will
   remove solids and floatable oils; these two streams will be
   sent to  holding tanks prior  to ultimate disposal. Cor-
   rugated plate type separators were selected over other
   types of separators because they will minimize the land
   area required for the unit operation.
  •After treatment by the separators,  the  wastewater will
   go  through a pH adjustment step (to a final pH near
   6.5-7), flow through another flow division box and then
   to parallel trains of activated carbon adsorbers.
  •The preliminary design  contraints for  the carbon ad-
   sorbers were 30 minutes contact  time and 3 gal/min/ft2
   application rate.  Commercially available adsorbers were
   considered first;  however, in order to satisfy  these de-
   sign constraints and  to allow for mobility and accessi-
   bility at the site, more land area was required than was
   available. For this reason, Weston recommended the con-
   struction of four adsorbers—two trains with two stages
   each.  Each  adsorber would  be  compartmented  into
   quadrants with a common centerwell. The compartment-
   alization will allow for backwashing  facilities of reason-
   able capacity and size to be designed.

-------
                                                                   GEOHYDROLOGY MATH MODELING    297
10,000i-
 1,000
   t
   8
   7

   6

   5
           Plttston, Pennsylvania
           Butler Tunnel
           28 March 1980
                   C (1,2-Dichlorobenzene) - v g/L 100

                        Figure 4.
      Carbon Isotherm for 1,2-Dichlorobenzene on Carbon
          Filtersorb 300 in Butler Tunnel Wastewater
•After being treated in the carbon adsorbers,  the flow
 will go to an effluent sump which will be used to store
 backwash water, and also  to allow  for chlorination of
 effluent prior to discharging to the Susquehanna  River.
 Additionally, the operator will be able to chlorinate the
 influent to the  carbon adsorbers periodically,  to mini-
 mize bacterial growth on the adsorbers.

•Water used for  backwash will go to a backwash receiv-
 ing tank. This tank is necessary  to handle the surges in
 the flow during backwash.  After entering the backwash
 receiving tank, the water will be retained so that solids
 and oil can be  separated from  the  bulk of the water.
 These solids  and oil will be separated and sent to the
 sludge and oil holding tanks respectively.

•Oil from the oil holding tank will be pumped on a peri-
 odic basis into tank trucks for transport to an acceptable
 waste oil disposal facility, such as Rollins in Bridgeport,
 New Jersey.

•For the purpose of this design, the consultant has as-
 sumed that solids  to  be disposed of, will also require
 disposal as a hazardous waste. Such disposal is expensive
 (estimates  are  typically $125/ton); therefore,  a centri-
 fuge  has been included to dewater the  solids prior to
 disposal.
  Presented in Figure 6 is a proposed equipment arrange-
ment site plan that Weston developed for the 6.0 mgd de-
sign case. As can be seen by this  figure, the proposed
6.0 mgd facility will occupy all available space.
                                                                               Corrugated
                                                                            Plate Separators
                           Centrifuge
                            To Truck Disposal
                                                                                                              River
                                                                                   Backwash Pump
                                                      Figure 5.
                                      Pittston Butler Tunnel Process Flow Diagram

-------
 298     GEOHYDROLOGY MATH MODELING
                              Carbon
                            Contact Tank
                               (30' ef)
  Carbon
Contact Tank
,   (30' 0)
  Backwash
Receiving Tank
   (40' 0)
                         fi\.— wet Well
        Notes and Assumptions

           1   Available Area is 250' x 100' with 30' Required Setback. Per Northeastern
              Engineering Report - Page 2.
           2   Corrugated Plate Separator Requires 25' x 50' Plan Area.
           3   Trailers are Approximately 12' x 50'. and Allow 1/2 Length of Trailer
              For Mobility and Accessibility.
           4   Backwash Receiving Tank Sized to Receive Backwash from Two Contactor
              Compartments, Each with Surface Area of Approximately 160 ft2,
              and a Backwash Rate of 20 gpm/ft2 for 15 Minutes
           5   Oil Tank Sized for 10.000 Gal. to Allow for Intermittent Tank Truck
              Removal.
           6   Solids Holding Tank Sized for 1,000 Ib/day of Solids (6 mgd, 20 mg/L TSS),
              Two Days Storage Capacity at 1°0 TSS (10,000 mg/L).
           7   Carbon  Contactors  Will Be Installed 10' Apart to Allow for Piping, Etc.
                                                      Figure 6.
                                                Pittston, Butler Tunnel
                                       Proposed Equipment Arrangement Site Plan
COST ESTIMATES
  Estimates  have been made  for  the installed capital
costs  for the major  equipment  included in the treatment
system  (Table  II.) The estimates  of the total installed
capital cost for the 3.0 mgd and Butler Tunnel Treatment
System  are $2,279,400, and for the 6.0 mgd System, the
estimate is $3,434,700. These estimates are based on an
ENR  Index of 3160,  and do  not  include certain direct
costs  associated  with  construction  such as engineering,
construction  management and  startup. These cost  esti-
mates are considered to be accurate to ±30%.
  Estimates of the  operating  costs (Tables  III and IV)
for the Butler  Tunnel Treatment  System  were also de-
veloped. The estimate of the total  operating cost for the
3.0 mgd System is $l,347,000/year, and for the 6.0 mgd
System, the estimate is $2,559,000/year.
  Review of the data in these tables indicates that the car-
bon represents the  most significant cost.  The principal
factors which will affect these costs are:
                                                    •Raw Wastewater Strength: as the concentrations of con-
                                                     taminants in the raw wastewater decrease, the life of an
                                                     existing carbon bed increases. The value of raw waste-
                                                     water concentrations, used in making these estimates of
                                                     operating costs,  was the median of the values for DCB.
                                                     If the concentration of DCB in the Butler Tunnel dis-
                                                     charge decreases, the life of the adsorption  bed would
                                                     correspondingly increase.
                                                    •Allowable Effluent Concentration at Carbon Exhaustion:
                                                     the  effluent concentration  will directly  influence ad-
                                                     sorption capacity of the carbon bed. As shown for the
                                                     carbon isotherm in Figure 4, as the equilibrium concen-
                                                     tration of DCB in solution increases, the Ib DCB ad-
                                                     sorbed/lb carbon  increases. Increasing the  allowable
                                                     final effluent concentration  of DCB at breakthrough
                                                     from 20 /ug/1 to 400 /zg/1, reduces the estimated carbon
                                                     replacement cost by over 70%.
                                                    •Unit Cost of Replacement Carbon: while this may be ob-
                                                     vious, it is a significant point.  Fundamentally,  the cost
                                                     per  pound for replacement of a particular carbon must

-------
                                                                               GEOHYDROLOGY MATH MODELING     299
                          Table II.
            Estimate of Installed Capital Costs for
              Butler Tunnel Treatment System
                    (ENR Index = 3160)
                                                  Estimated $
Item
 1. Life Pump Station
 2. Flow Division Box
 3. Corrugated Plate Oil Separators
 4. pH Adjustment Facility
 5. Flow Division Box
 6. Carbon Adsorption Facility
 7. Sludge Holding Tank
 8. Oil Holding Tank
 9. Chlorination Facility
10; Centrifuge
   Subtotal
11. Miscellaneous1
12. Contingency

   Total1
t. Includes piping, electrical, instrumentation and site work.
2. These estimates do not include certain direct costs associated with construction of the facilities
  such as engineering, construction management and start-up.
                          Table in.
 Estimate of Operating Costs for Butler Tunnel Treatment System
                       (3 mgd Design)
                    Pittston, Pennsylvania
                         (May 1980)
3 mgd
172,400
25,800
115,100
63,900
25,800
876,100
32,400
22,300
73,000
178,900
1,585,700
396,400
297,300
2,279,400
6 mgd
217,200
25,800
276,100
90,900
25,800
1,368,500
46,600
28,700
124,800
185,000
2,389,400
597,300
448,000
3,434,700
1. Labor
  27 manhours per day
  $11 per hour (salary and benefits)
  52 weeks
   7 days per week

2. Maintenance (mechanical, structural, electrical,
  instrument)
  3 % of capital costs other than carbon inventory

3. Electrical
  HPat$0.50/kwh

4. Chemicals
  Caustic: 350 tons/year at $200/ton
  Chlorine: 36.5 tons/year at $400/ton
  Carbon': 833 tons/year at $l,200/ton

5. Sludge and Waste Oil Disposal
  Sludge Disposal: 600 tons/year at $125/ton
  Waste Oil: 21,900 gal/year at $0.83/gal

6. Total Estimated Operating Costs
$  108,100/year
$  42,500/year


$  17,400/year


$l,084,200/year





$  94,400/year




$l,347,000/year
                                              Table IV.
                   Estimate of Operating Costs for Butler Tunnel Treatment System
                                            (6 mgd Design)
                                        Pittston, Pennsylvania
                                             (May 1980)
                   1. Labor
                     27 manhours per day
                     $11 per hour (salary and benefits)
                     52 weeks
                      7 days per week

                   2. Maintenance (mechanical, structural, electrical,
                     instrument)
                     3 % of capital costs other than carbon inventory

                   3. Electrical
                     HP at $0.05/kwh

                   4. Chemicals
                     Caustic: 700 tons/year at $200 ton
                     Chlorine: 75 tons/year at $400/ton
                     Carbon':  1,670 tons/year at $ 1,200/ton

                   5. Sludge and Waste Oil Disposal
                     Sludge Disposal: 1,215 tons/year at $125/ton
                     Waste Oil: 43,800 gal/year at $0.83/gal

                   6. Total Estimated Operating Costs
                                                                                                                         $  108,100/year
                                                     $  62,300/year


                                                     $  26,300/year


                                                     $2,174,000/year





                                                     $  188,300/year




                                                     $2,559,000/year
1. Actual quantities and costs for carbon replacement will depend on allowable effluent concen-
  tration at breakthrough, and the actual adsorption capacity of the carbon which is finally
  selected for use. The value indicated reflects the probably order of magnitude of these costs,
  assuming effluent DCB concentration at breakthrough of less than 50ng/l.
 1. Actual quantities and costs for carbon replacement will depend on allowable effluent concen-
   tration at breakthrough, and the actual adsorption capacity of the carbon which is finally
   selected for use. The value indicated reflects the probable order of magnitude of these costs,
   assuming effluent DCB concentrations at breakthrough of less than 50 lig/1.

 be combined with the adsorption capacity of that carbon
 to, in effect, establish the cost per pound of contaminant
 removed.  In general,  two types  of carbon  are available.
 The first type of carbon  is virgin carbon; the second type
 is reactivated carbon. The  costs and capacities of virgin
 carbon are higher  than  for reactivated carbon. Virgin
 carbon is  available from a greater  number of suppliers
 than is reactivated carbon;  therefore, purchases of virgin
 carbon are done more in a "buyer's market" than are
 purchases  of reactivated  carbon.  Of course,  in either
 case,  unless some  form of service contract is established,
 variations  in spent carbon disposal  costs/resale  value
 must also be considered.

REFERENCES

1.  "Report to On-Scene  Coordinator: Treatability Study
    Major  Pollution Incident  Butler  Tunnel,  Pittston,
    Pennsylvania",  Roy F.  Weston,  Inc., West Chester,
    Pennsylvania, 1980.
2.  Lafornara,  J.P., Hammond,  J.W.,  Lederman,  P.B.
    and Massey, T.I., "The Pittston Story",  Proc.  of 1980
    U.S.  EPA  National  Conference  on Management  of
    Uncontrolled Hazardous Waste Sites, Oct.  15-17, 1980,
    Hazardous   Materials   Control   Research   Institute,
    Silver Spring, Md., 250-254.

-------
         COMPUTER-ENHANCED GEOPHYSICAL SURVEY
                   TECHNIQUES FOR EXPLORATION OF
                            HAZARDOUS WASTE SITES
                                        G.L. McKOWN, Ph.D.
                                       G.A. SANDNESS, Ph.D.
                                 Battelle, Pacific Northwest Laboratory
                                         Richland, Washington
INTRODUCTION

  Collection of chemical and geological data during site
assessment activities is an integral and necessary part of
hazardous waste  site investigations.  Among the first in-
vestigative methods that might be employed at a site are
geophysical surveys. These ex situ studies  are based on
geohydrologic changes induced by waste materials, and
can  provide  a more  complete picture of subsurface
phenomena prior to in situ sampling and  well drilling.
Included among the potential applications are:

Location  and/or Delineation of the Boundaries  of Sus-
pected Burial Sites.
  Many of the techniques used  depend on a change in
density, reflectivity, susceptibility, or permittivity of tar-
get  materials, all of which are  characteristic of  buried
wastes. This  type  of underground  location is readily
accomplished by a magnetometer/radar combination that
has been  successfully employed by Battelle-Northwest in
many locations.  The ability  to  accurately  locate burial
trenches,  individual targets such as  drums, and concen-
trations of landfilled trash/wastes has been demonstrated.
Other techniques may be used if poor resolution is accep-
table.

Location  of Waste Containers.
  At  many  sites,  the possibility exists  that high  con-
centrations of hazardous materials could be encountered
when probing underground. This is  particularly true at
uncontrolled sites and locations where burning/burial has
occurred. To minimize the hazard of drilling or coring
into such deposits,  a survey should be made before drill-
ing  commences. Radar, magnetometry and metal detec-
tion are applicable  tools since the targets do not provide
adequately  defined acoustic  reflection  and resistivity
measurements are too difficult to interpret unless detailed
surveys are conducted.

Location of Underground Pipes and Cables.
  Wells or cores must not be drilled through underground
utility conduits. This  is a problem at  any  site where in-
dustrial operations are concentrated and where insufficient
information on underground utility routes exists.
  Several techniques are available for locating or trac-
ing  utility conduits. Acoustic or radiofrequency genera-
tors can be used to drive a metallic line and the route traced
by detectors from the surface.  Resistivity  measurements
can provide information on nests or concentrated runs of
metallic objects. For  the most general case where either
metallic or nonmetallic conduits are possible, a combina-
tion of magnetometry and radar offers the greatest def-
inition. About the only situation that may not be iden-
tified  with this combination is  plastic gas lines and the
location of these are likely to be documented since plastic
pipe has not been used until relatively recent times.


Location of Discontinuities in Underground Strata.
  Radar or acoustic systems can be employed to detect dis-
continuities, either natural or manmade, in an underlying
stratum or bedrock. Acoustic/seismic methods can be em-
ployed if the voids are large  (on the order of 100 m2 or
greater), whereas radar  can  detect much smaller holes
(on the order of 1 m2 or less). However,  the radar sys-
tem is limited in depth of penetration to  3 to 10 m for
typical lithology.
  In either case, the presence of intervening  strata con-
founds reflection patterns and limits the use of  these meth-
ods. Other techniques may be used in special instances. For
example, magnetometry  is an excellent tool for location
of voids/faults in magnetic bedrock such as  basalt and
intervening layers are acceptable if they are spatially con-
gruent.  Gavimetry  has  similar  applicability,  although
resolution is poorer  and might be used to detect  mine
shafts, limestone caverns and sinkholes that are evident on
the surface.
Location of Water-Bearing Strata.
  Since the dielectric contant of water-bearing materials
differs considerably  from other geologic structures, it
should be possible to use radar systems to map eleva-
tion and slope of the  water table. On surveys to date, the
authors have had only moderate success in delineating such
features. The water table in one case, was quite obvious
in radar scans, but this represents a favorable case (Depth
of 2 to 3 m and salt  water). In other cases, the depth to
ground water has been too great (> 10 m) or the aqui-
fers not sharply defined enough to allow meaningful in-
terpretation. In some instances, the presence of interven-
ing strata or discrete objects (rocks) totally masks the
effect.
                                                   300

-------
                                                                GEOHYDROLOGY MATH MODELING    301
  In favorable cases, survey techniques might be used to
define the limits of a contamination plume. The ideal case
would be highly contaminated sources infiltrating at a high
rate. Electrical resistivity would be most applicable in at-
tempting  to  delineate the plume;  very  low  frequency
(VLF) resistivity techniques probably offer  the  greatest
chance for success.

Location of Abandoned Wells and Drill Sites.
^Occurrence of old wells, abandoned drill holes, burial
pits, sinkholes, and other penetrations of near surface im-
permeable layers provide potential infiltration points for
contaminated surface water. It has been shown that radar
works well in locating such areas of disturbed earth, es-
pecially considering  that backfilling these holes would
rarely have been done according to good engineering prac-
tices.
  Infrared imaging has also been used with less  spectac-
ular success in locating old excavations.  Acoustic holo-
graphy is a relatively new technique that is ideal for plot-
ting depth and characteristics of boreholes once they have
been located.
Location of Solution Channels. Cracks and Fissures.
  Geophysical surveying in this  application is  less cer-
tain, and depends on characteristics of the specific site.
Acoustic/seismic reflection probably has the widest appli-
cation, although interpretation  would be extremely diffi-
cult in complex matrices.
  Radar probably is  not generally applicable,  although
specific instances  may  be amenable to this treatment.
Tracer methods are applicable, but require an  extensive
experimental program.
Location of Objects of Historical Interest.
  Some Army  installations contain sites of great arch-
aeological interest. Although exploration is not an objec-
tive of site survey  programs, geophysical methods pro-
vide aid in minimizing disturbance of archaeological sites
during conduct of on-site exercises.
  Since most discrete objects and structural features  of
archaeological interest are located within the first few
meters of the surface, a radar system that provides under-
ground map views is an excellent tool for these investiga-
tions. The potential for this application was demonstrated
during a radar survey of Fort Monroe for the U.S. Army.
Geophysical Logging.
  Various down-hole analytical techniques may be applic-
able to site assessment programs. The porosity of media
down boreholes can be determined by neutron-epithermal
neutron methods, which in some cases can be used to de-
rive permeabilities. Natural gamma radiation logs can be
used to determine certain lithologic changes and estimates
of geochemistry. Electrical logs provide a more detailed
observation of water strata than can be obtained from in-
spection of cutting returns.
GEOPHYSICAL SURVEY SYSTEM

  Researchers at Battelle-Northwest have  selected metal
detection, magnetometry and  ground-penetrating radar
for inclusion in an integrated survey unit for burial ground
and  other subsurface  investigations. Radar transceivers
are available for operation at frequencies of 100, 300, 500
and 900 mhz.
  The vehicle was specially constructed of nonferrous ma-
terials to facilitate  the use of  a  magnetometer mounted
alongside the radar unit. In addition to transporting the
radar and magnetometer, the  vehicle contains a micro-
computer and a telemetry system which  acquires, pre-
processes, and transmits radar/magnetic data to a larger
computer at the survey site. Recent additions include an
all-terrain Trackster transporter  and a motor home for
transport between  sites  and to  permanently  house the
computer systems.
  A subsurface survey is accomplished in four steps.
1.  A grid covering the study area is marked on the ground
   surface to guide  subsequent  measurements and to pro-
   vide location coordinates.
2.  Manual survey of the grid using hand-held metal de-
   tectors, marking all identified  targets and transferring
   locations to an overlay map.
3.  Operation of  the survey vehicle over the same grid,
   accumulating radar and magnetometer data. The data
   produced by the survey instruments is transmitted to a
   PDF  11-34 computer  where  it  is processed, formatted,
   and stored on either a magnetic tape or  a disk storage
   unit.  Preliminary data manipulation and processing
   functions can be performed on site to  edit and com-
   press the data for  final  storage or to  facilitate data
   monitoring or previewing during a survey. This near
   real-time data handling capability can be of immeas-
   urable value during conduct of a site survey.
4.  Data superposition and analysis is done mainly by a
   unique software package created to filter and  enhance
   output.  Graphic display devices are used first to pro-
   vide vertical  profiles  of each  radar survey run.  The
   profiles are then combined  in  digital form and sliced
   horizontally to generate a map view of a selected depth
   interval in the area  surveyed.  When specialized en-
   hancement programs  are applied, the output appears
   as color photographs  with specific hues  depicting bur-
   ied targets. This system has been  applied to surveys
   conducted in  Idaho, Washington, New York, Virginia,
   Indiana,  Pennsylvania, Utah and Nevada.

DATA PROCESSING AND DISPLAY

  As received and recorded by the computer in the field,
the radar data are not  in an optimal form to be displayed
and  interpreted.  Subsequent digital data  programs can
enhance the quality and usefulness of the data. Several
computer programs have been developed at Battelle to
enhance and display the radar and magnetometer  survey
date.
  Figure 1  is a  flow  chart showing  the  computer  pro-
grams involved in the data recording/processing/display

-------
 302
GEOHYDROLOGY MATH MODELING
                        MACKTCMETHJJ 3ATA
                          jPWAfiO AMD , CALCULATION . REDACTION
                          DOWNWARD j  Of SECOND    TO
                         CC*JTI'.LAT!GN ' ZESlVA'IVES 1   POU
                                             CONTOUR PVOTTISC
                  -.rfjsir' VOOULATIO
                i Pv-GTQGftAPMlC OUTPUT
'SCALING FOR
COlCfi OUTrLT
                         Figure 1.
            Computer Program Flow Chart for Radar
                   and Magnetometer Data

 sequences  for  radar  and  magnetometer  data.  The
 GEOSUR and ISA programs are for recording and manip-
 ulating the data after it has been transmitted to the com-
 puter from the  survey vehicle, and are not  described
 further.
   The stored digital radar  data  can be thought of as a
 three-dimensional  array containing  120 bytes/ping, N
 pings/track, and M tracks. The data processing programs
 REMVH and HYPER operate on two-dimensional radar
 data; that is, they operate on each track in sequence. Pro-
 gram SLICE operates on the entire three-dimensional ar-
 ray to construct a two-dimensional array.

 Program REMVH

   A characteristic of the reflected radar signals is that they
 contain components that are unchanging from ping to ping
 and convey no information about subsurface reflectors.
 These components are due to  a combination of factors:
 the oscillatory  shape of the waveform,  the presence of a
 ground surface reflection, random  reflections from the
 survey vehicle  and  the  radar  unit itself, and electronic
 effects in  the  radar unit. These unwanted  signal com-
 ponents appear in a radar image as alternately bright and
 dark horizontal bands. The left half of Figure 2 illustrates
 this effect. In each image of Figure 2, the horizontal di-
 mension corresponds to distance along the traverse line,
 and the  vertical dimension  is proportional  to the  travel
 time of the radar signal (or to depth).
  The purpose of  program  REMVH is  to  remove the
banding due to stationary signal components.  Let I (1 <  I
 <  120) be the datum index in a given ping, and let  J
(1  < J < N) be the ping index in a given track (N  is the
number of pings in the track). For each fixed value of I,
the program computes a running histogram as it scans the
array  in  the J direction. At a given point, (I, J), in the
                                                  array, the running histogram includes all  data with the
                                                  bounds II = 1-1 to 12 = 1+ 1 and Jl +  J-JA/2 to J2 =
                                                  J + JA/2, where JA is a specified number. At the edges
                                                  of the array, the bounds are as follows: if I = 1, II  = 1; if
                                                  I = 120, II =  120; if J = 1, Jl = 1; and if J = N, 32 =
                                                  N. The banding is removed by computing an average value
                                                  from the data within the bounds of the running histogram,
                                                  then  subtracting the average from  datum I,  J.  How-
                                                  ever, if the row contains values that are substantially high-
                                                  er or lower than the average value, this simple procedure
                                                  can yield local correction values that are either too high or
                                                  too low. This has the effect of introducing a spurious hor-
                                                  izontal oscillation in the data. The histogram provides a
                                                  simple means for excluding a range of high and low val-
                                                  ues. Improved  corrections are typically obtained by ex-
                                                  cluding the highest and lowest 10% of the data from the
                                                  calculation of the average.
                                                    The right iialf of Figure  2  shows the effect of apply-
                                                  ing program REMVH to the raw radar  data.  This ap-
                                                  proach to radar image enhancement improves the detec-
                                                  tability of discrete buried objects or  irregular masses of
                                                  waste material, but it is not effective in cases where it  is
                                                  necessary to detect flat, horizontal interfaces.

                                                  Program HYPER

                                                    The function of this program is to remove hyperbolic
                                                  reflection  patterns from  radar images, or reflection pro-
                                                  files-  Hyperbolic patterns occur because a radar output
                                                  beam has a finite angular width. Reflections are there-
                                                  fore received from a buried object both before and after
                                                  the  sensing transducer passes directly  over it.  It is some-
                                                  times desirable to remove the hyperbolic patterns from the
                                                  radar profiles because they can  increase the difficulty
                                                  of interpreting the data.
                                                  Program SLICE

                                                    Ground  penetrating radar  surveys normally  involve
                                                  cumbersome methods for mapping  detected buried ob-
                                                  jects. The locations of buried objects are usually deter-
                                                  mined visually from a collection of radar profiles, then
                                                  manually transcribed to a map. The  purpose of program
                                                  SLICE is to utilize a digital computer to  accomplish the
                                                  same task.
                                                    Program SLICE is essentially  a routine to construct a
                                                  selected  two-dimensional  data array from the original
                                                  three-dimensional array. A  depth range  is specified in
                                                  terms of data indices which range from 1 to 120. Then for
                                                  each ping  in the  array, a new array element is  derived
                                                  by averaging the ping over the specified depth range. The
                                                  dimensions of  the new array are  N x M,  where N is the
                                                  number of pings per track and M  is the number of tracks.
                                                  Because the original data spacing along a survey track is
                                                  much less  than the track spacing, it is necessary to ex-
                                                  pand the new array in the across-track direction. This is
                                                  accomplished by linear interpolation. Interpolation in both
                                                  directions  allows  the array  size to be completely selec-
                                                  table.

-------
                                                                 GEOHYRDOLOGY MATH MODELING     303
                                                    Figure 2.
                                    Two Forms of a Single Radar Profile Illustrating
                                          Computer Processing Operations
  In cases where the distribution of buried objects is suf-
 ficiently sparse  to allow radar reflections from a wide
 range of depths, a single, thin, horizontal slice will show
 only those objects contained in the corresponding narrow
 depth range. Examples of the output of program SLICE
 are shown in Section 5.

 Program IDAPIC

  Preliminary video monitor and hardcopy displays of raw
 and processed radar data can be made almost in real time
 by means of a data previewing capability included in pro-
gram IDA. Final output products, however, are  made in
photographic form by utilizing a DICOMED D48, black
and white or color, digital film recorder.
  The function of IDAPIC is to transmit IDA formatted
radar data to the film recorder. It provides  options for
file subsetting,  image contrast enhancement, and color-
coded level slicing.

Program PICT

  This is a general purpose program for producing photo-
graphic images  of raster-scan data from  a wide variety
of sources. In particular, it accepts  the file structure of
the output data produced by program SLICE. It  is other-
wise similar  in both  form  and  function  to  program
IDAPIC.

Magnetometer Data

  The data acquisition system on the survey vehicle is de-
signed to transmit either one or two magnetic field values
along with each transmitted radar ping.  Two  magnetic
field values are allowed to order to provide for the pos-
sible future use of two magnetometers. However, only one
has been used in past and current programs. The magneto-
meter data is, therefore, basically a two-dimensional layer
in the three-dimensional IDA data array.
Program MAG

  Numerous methods for analyzing magnetic data have
been described in the literature. Commonly, these methods
involve data manipulation  operations such  as upward
and downward continuation of the magnetic field, calcula-
tion of vertical derivative and reduction to the pole. Pro-
gram MAG performs all three  of these operations.  Its
basic  function is to systematically alter  or enhance the
magnetic field patterns in such a way that they can be more
easily interpreted.

-------
304    GEOHYDROLOGY MATH MODELING
   In a region free  of magnetic materials, the total magnetic
 field anomaly. AB, satisfies Laplace's equation;that is:
        t  82(AB) |  32(AB)
           dy-      d/-
 (1)
 By the method of separation of variables, the solution of this
 differential equation can be expressed as:
 B(x.y./) = £  £  f 7s,
 • z
 where
(2)
      mx
 f    (-—.yM  =a two dimensional Fourier series expansion
     \ x   ' /   of the magnetic  field in the measurement
                plane z = 0

            Lx = MAx

            L6 =NAy

 M and N  are the number of data in the x and y directions,
 respectively, and Ax and Ay are the data spacings in the x and
 y directions, respectively.

   Upward and downward  continuations of a magnetic
 anomaly are calculated by inserting the appropriate values
 of z into the above summation. Vertical derivatives of AB
 are calculated  by  differentiating the exponent within the
 summations.
Program CNT

  Program CNT is a system routine on Battdle labora-
tory computers.  Its function  is to plot contour  mapi
from various kinds of data arrays. Hardcopy output prod-
ucts are typically drawn on a CalComp plotter.

Program M AGP

  It is often useful and desirable to display magnetic data
in the form of a color picture. This is conveniently accom-
plished  by means of program PICT. However, the  num-
ber of magnetic measurements made in a given area is
generally small  in comparison with the number of picture
elements needed to form a pleasing image of the magnetic
field pattern in  that area.  A procedure is needed to effec-
tively increase  the  number of data.  Program  MAGP,
therefore, is basically a linear, two-dimensional interpola-
tion routine to expand the magnetic data file.
  Program PICT divides the total range of magnetic field
values in the expanded file into a discrete  number of lev-
els, assigns a color to each level, and produces a  level-
sliced color picture of the magnetic field values.

EXAMPLES OF OUTPUT

  An example  of essentially  raw radar data is  given in
Figure 3.  This  scan represents  a plot of depth (negative
vertical  axis) versus track length (horizontal axis), with the
ground  surface at the  top of the raster.  Dimensions of
the vertical profile correspond  to a depth of 6 m and a
track length of 30 m. The return from a small spherical
object about 0.2 m in diameter is noted to illustrate the
                                                    Figure3.
                                        A Radar Profile Showing the Reflection
                                               from a Sphere (arrow)

-------
                                                                 GEOHYRDOLOGY MATH MODELING    305
difficulty in interpreting such records for the location of
specific objects.
  The results of an actual survey yield a series of vertical
profiles. Each record represents an excursion by the survey
vehicle along parallel tracks at some spacing. Interpreta-
tion of vertical profiles by visual  analysis and transfer to a
grid map provide information on large underground struc-
tures, especially continuous forms such as pipes, sewers,
wall foundations, etc.
  An example of color-enhanced horizontal slicing illus-
trating the utility of the processing is given in Figure 4.
In this case, slices from 0 to 1.5  m depth have been com-
bined and filtering introduced to enhance the location of
discrete,  small targets.  The number and  distribution  of
such objects are apparent.
  The area shown covers about 3 acres. Interpretation of
some of the gross  artifacts, few of which are  apparent
from records of prior site activity or surface  features, is
shown.

DIRECTIONS FOR NEW DEVELOPMENT

  Several aspects  of Ground Penetrating  Radar  Tech-
nology require further development in order  to enhance
broad scale application of the method. Presented below
are some areas of improvement that Battelle  is  currently
pursuing:
   •Transceiver-antenna design; improved waveshape, anten-
    na coupling,  increased peak power, tunable-frequency
    systems
   •Antenna  arrays and associated  electronics  to improve
    resolution and to provide increased areal coverage cap-
    abilities
   •Collection, transmission  and  manipulation of data  at
    faster rates and with improved reliability
   •Review-refresh memory data systems; comparison of old
    and new data to detect changes in underground systems
   •Portable, handled radar systems for coverage of small,
    remote, rough terrain areas  with surface obstacles such
    as rocks, trees, fences, etc.
   •Target recognition through signature analysis, 3-D data
    overlay systems,  size and  shape  correlation, deconvolu-
    tion of wave forms
   •Integration of GPR data  with acoustic profiling, sonic
    holographic and/or electrical resistivity  techniques to im-
    prove detection of nonmetallic objects
   •Improved, ruggedized transport  system, including auto-
    matic locator systems to define xy coordinates and attach
    angle of the transporter
   •Electromagnetic  characterization of soils, subsurface
    geologic strata, and representative targets
     Success in expanding the  technology will provide an en-
   hanced capability  for investigations of Hazardous Waste
   sites.
               PACKED CLAY
               (OLD BASEBALL DIAMOND)
OLD DIRT ROAD BED
                                                                                   STORM SEWER
                                                            NUMEROUS
                                                            DISCRETE TARGETS
                                                            POSSIBLE OLD
                                                            TUNNEL STRUCTURE
                            GAS LINES
                            OR WATER PIPES
                                                    Figure 4.
                                 Computer-Produced Map View of Radar Reflections at
                                      a Survey Site with Filtering to Reduce Broad
                                                Reflection Patterns

-------
   THE USE OF MATHEMATICAL MODELS TO ASSESS AND
DESIGN REMEDIAL ACTION FOR CHEMICAL WASTE SITES
                                               C.R. COLE
                                         G.L. McKOWN, Ph.D.
                                 Battelle, Pacific Northwest Laboratory
                                         Richland, Washington
 INTRODUCTION

   It is estimated that there may be 30,000-50,000 old dis-
 posal sites and sites contaminated from spills in the United
 States, of which 1,000-2,000 are sufficiently contaminated
 to pose a hazard to the public. Restoration costs are likely
 to exceed $ 1,000,000 per site.
   Faced  with  the complexity and high costs of address-
 ing such a large number of sites, it is clear that formalized
 methods are required to expedite the restoration process
 and identify the most cost-effective means of site remedial
 response. This paper describes a methodology which util-
 izes mathematical models describing fluid flow and con-
 taminant transport. These constructs have been developed
 over a period of nearly two decades and employed to solve
 a variety of surface and groundwater contamination prob-
 lems. They offer the ability to organize, interpret and bet-
 ter understand what is presently occurring, and in ad-
 dition, provide the ability to predict what will occur in the
 foreseeable  future.  These  capabilities  are of  particular
 value in designing remedial action  since, through the use
 of these tools, one can:
 •determine when a specific site is  sufficiently  character-
  ized and  understood such that when implemented the
  selected remediation program will be effective
 •allow one to  design a cost effective site characteriza-
  tion and data gathering effort
 •rapidly and cost effectively assess a wide variety  of re-
  medial options without moving any dirt
 •design a cost effective site surveillance and monitoring
  plan which will maximize  the ability to detect any devia-
  tion from the remediation plan.

 THE SITE RESTORATION PROCESS
   While it is the abandoned landfill or superfund problem
 set has has stimulated the current interest  in site restora-
 tion, two major programs  in the U.S. have  preceded it.
 The U.S. Department of Energy and its predecessors have
 had an active  Decontamination and  Decommissioning
 (D&D) program for many years now. Similarly, the U.S.
 Army Toxic and Hazardous Materials Agency has con-
 ducted its Installation Restoration (IR) Program since the
 mid-seventies. As a result of these undertakings, a formal-
 ized approach to site  restoration has emerged. The ap-
 proach is divided into three phases  : (1) site characteriza-
 tion, (2) site assessment and (3) site remediation.
Site Characterization
  Work performed during site characterization is directed
to the description of the source(s) of contamination and
the surrounding environment. The objective is to enumer-
ate and quantify to the extent possible the physical, chem-
ical, and biological factors which will affect fate and mi-
gration of contaminants. For maximum efficiency, charac-
terization is performed in an integrated manner with as-
sessment so that only necessary data are collected.
  The first element of characterization,  the preliminary
survey, is focused on gathering all existing data on a site.
If data are sufficient, no further characterization efforts
are required.  If the preliminary assessment reveals data
gaps which must be filled before a final  assessment can be
made, then additional characterization  efforts are imple-
mented. Hence, the preliminary assessment is employed to
determine the adequacy of existing data and the optimum
design for any subsequent data collection activities.
  When active characterization is necessary, it is divided
among a number of field  activities. Geophysical survey
techniques can be employed to map deposits. These efforts
are complimented  by field geohydrologic  investigations
which determine the nature and extent of geologic and hy-
drologic features which will affect the movement of water
and contaminants. Finally, sampling and analysis is per-
formed to detail the disposition of contaminants in the en-
vironment and the physical/chemical properties of geolog-
ic media which would  attenuate  movement. All charac-
terization activities may not be required at every site. As
noted previously, the type and extent of activities should
be identified during a preliminary assessment.
  Post-closure characterization can actually be thought of
as two  distinct  activities: (1) surveillance of remedial
action  implementation and (2)  post construction  mon-
itoring.  The former is directed toward  assuring safe per-
formance of work and conduct of that  work in the man-
ner prescribed in the final plan.  The second activity tries
to confirm the  effectiveness of the  remedial action on a
continuous basis and, in so doing, will provide a warn-
ing in the  event of failure and provide for the collection
of a historical record that can be used to determine the
reason for failure.

Site Assessment
   Activities conducted for the assessment are designed to
determine the implications of the available data. In this
                                                    306

-------
                                                                 GEOHYRDOLOGY MATH MODELING    307
context, they are interpretative in nature. The first level
of assessment, preliminary assessment, is conducted us-
ing available data to determine if sufficient information is
on hand to characterize a site and quantify the risk posed
by that site. Two levels of output are possible. The first is
a determination that current data are insufficient. In this
case, the assessment identifies the minimum necessary data
that should be collected to allow conduct of a final assess-
ment. The second possible output is the result of a final
assessment itself.
  The use of the mathematical modeling tools needs to be-
gin in this initial assessment phase. Even if sufficient data
for implementation of a mathematical model are not avail-
able at this stage the organization and approach required
to implement a  mathematical model can  greatly aid the
initial assessment process and improve ones  ability to un-
derstand the system.  When enough existing data are avail-
able to justify the implementation of a mathematical mod-
el in this initial  assessment phase,  the quality of the  as-
sessment and the understanding of the system improves
even more.
  The final assessment is conducted only after sufficient
data and system understanding  are  in hand. Narrowly
defined,  the product of a final assessment is the descrip-
tion of the time frames,  durations and concentrations of
contaminants that various segments  of the environment
will be exposed  to. In this context, the assessment pro-
duct is a prediction  of what continuous monitors would
record if the situation were allowed to remain in its cur-
rent state. When coupled with data on the dose-response
function for the contaminants of concern, the assessment
output is converted  to an  estimate of the risk posed  by
the site. Consequently, the assessment product is essential
to several key decision points:
1. Which sites pose  unacceptable  risks and therefore  re-
  quire remedial action?  ,
2. How extensive must clean-up action be to bring a site
  within the realm of acceptable risk?
When it is determined that remediation is  required and
conceptual approaches are identified, assessment is again
conducted to predict the effectiveness of proposed alterna-
tives and to identify the most cost effective post-closure
monitoring plan.

Site Remediation

  Remediation activities are the major end-product of site
restoration work. They  are  initiated whenever a  final
assessment concludes that  risks posed by a site are un-
acceptable. At that time,  the preliminary work is begun
to identify all feasible alternatives. After a cursory screen-
ing, the  most promising alternatives are  described in a
conceptual framework and input to the post-closure assess-
ment where estimates are made of the effectiveness of each
alternative. Output from the  post-closure assessment is
compared to output from the final site assessment to de-
termine the level  of risk reduction offered by each alterna-
tive. Alternatives not capable of reducing risks to accep-
table levels are  discarded.  Remaining candidates  are
ranked according to  cost. The  ratio of risk reduction to
cost is then  employed to select the most cost-effective
approach.
  The selected alternative is then taken into final design
and construction. Post-closure characterization is con-
ducted to assure that construction is performed in accor-
dance with final designs and to detect any releases  or ac-
tivities which pose a hazard to the public as a result of
remediation.
  The key element of the above approach to site restora-
tion is the assessment activity. When properly conducted,
an  integrated assessment  continues to focus attention on
the most cost-effective configuration for conduct of work.
In the early  stages,  assessment is employed to minimize
the extent of expensive field characterization  work while
insuring  that sufficient data are collected for evaluation.
In the body of the assessment, the objective is to obtain
as accurate a measure of risk posed by a site as possible
so that a determination can be made  as to whether a site
requires  remediation,  and if so, how much.  The post-
closure assessment is directed to quickly ascertaining the
likely effectiveness  of  remedial  action alternatives and
their optimum design before resources are committed as
well as to identify a minimum, yet sufficient,  configura-
tion for  the monitoring network that will provide all the
desired information.
  The strength of the assessment phase is directly related
to the technology employed to accomplish it. Develop-
ments over the last  decade suggest that, when properly
applied,  mathematical models for fluid flow and contam-
inant transport are the best tools currently available for
solving  environmental  problems. They provide the de-
cision maker with the extra insight needed to determine
the optimum  solutions for a given problem.

MATHEMATICAL MODELING—AN OVERVIEW

  It is important to  briefly  discuss in general terms what
a mathematical model is, its value, its limitations and its
range of application in order to illustrate its  value as a
tool in the problem assessment and  remedial  action de-
sign process. Mathematical models,  whether analytical
or numerical, can only be a limited representation of an
actual system or "cartoons of reality.(4) Models allow one
to examine an entire system by incorporating expert under-
standing of a complex system with the generally limited
field data in  a form  which can be tested against field ob-
servations of system response to various driving forces.
Disagreements between model and observation can be used
to guide  further field studies or to cause initial interpreta-
tions of the system to be modified in plausible ways.  ,
  Processes that occur in the actual system are described
by appropriate mathematical equations which correlate the
system's physical and  chemical  conditions in a proper
manner such that as  changes are made to the physical and
chemical parameters describing  the  natural  conditions,
the appropriate response  of the real  system may be pre-
dicted by solving the mathematical equations.
  In order to construct and effectively utilize  a model it
is important to:
(1)  Understand the physical, biological and chemical pro-

-------
308     GEOHYDROLOGY MATH MODELING
   cesses which occur within the system of interest
(2) Identify and determine the important physical, biologi-
   cal and chemical parameters and inputs which allow the
   solution of the mathematical equations which govern
   the physical and chemical processes that occur
(3) Characterize the current  condition of the system and
   identify its boundaries and boundary conditions
(4) Update model parameters as new system data or a bet-
   ter understanding of the system are attained.
Models, in general, are not static; they  must be dynamic
unless the system being modeled is very simple. A model
acts  as a storehouse for all  the pertinent information
assimilated on the system, and or the understanding of the
processes which occur in the system. A model is capable of
learning in the sense  that as the data density and quality
improve, those improvements can be incorporated into
the model to improve its predictive capabilities. As the un-
derstanding of processes advances, the model equations or
parameter relationships can be changed to incorporate this
new knowledge and so improve ability of the model to pre-
dict. Model maintenance is an integral part of any effort
involving the use of models.
   Mathematical  hydrodynamic  and transport  models,
which can be constructed at various levels of complexity or
realism depending on the use to be made of them, attempt
to duplicate the real  system  by a series  of complex equa-
tions that describe fluid flow and material transport. The
advent  of high speed digital computers has made the solu-
                         Table I.
     Characteristics of a Few of the Ground-Water Flow and
     Transport Models Being Employed in Safety Assessment
      of Nuclear Waste Repositories and for Other Ground-
                      Water Studies
 HYDROLOGIC
  PATHSfa)"4)

  VTT
 FE3DGWI8)

 CFEST(b)
Two-dimensional, analytical/numerical method,
homogeneous geology, saturated flow.
Two-dimensional, finite difference numerical
method, heterogeneous geology, saturated flow,
multilayer systems with interaquifer transfer.
Three-dimensional,  finite  element numerical
method, heterogeneous geology, saturated flow.
Three-dimensional,  finite  element numerical
method, heterogeneous  geology, coupled flow,
energy and solute transport.
 CONTAMINANT TRANSPORT
 GETOUT*6'         One-dimensional, analytical method, chain de-
                   cay, single speciation, equilibrium sorption, con-
                   stant leach rate, dispersion.
 MMT<'-2"          One- and two-dimensional, discrete parcel ran-
                   dom walk numerical  method,  chain decay,
                   single  speciation, equilibrium  sorption,  time
                   variant  leach rate,  dispersion, and  arbitrary
                   source distribution.
 CFEST            Three-dimensional,  finite element  numerical
                   method,  heterogeneous geology, coupled flow,
                   energy and solute transport.

 (a) All the hydrologic models operate in a transient or steady state mode
   and have streamline or pathlme options that can predict travel time.
 (b) Code documentation in progress.
tion of these complex sets of equations practicable and
cost effective for:
(1) Developing an understanding of a complex system
(2)  Determining sensitivity of the system to uncertainties
   in data and understanding
(3)  Providing insight  on the  system and making predic-
   tions of system response to various perturbations
(4)  Prediction of system response to various proposed
   management alternatives
(5) Testing the adequacy of various engineering designs.
  The steps involved in modeling are:
(1) Definition of the study objectives
(2) Collection of data on the system
(3) Formulation of a conceptual model of the system
(4)  Translation of the conceptual model and data on the
   system into the  form required for  the selected numer-
   ical model
(5) Calibration and  verification of the numerical model
(6)  Check of the sensitivity of the model to variations in
   input data
(7)  Use as a  predictive or investigative tool to attain the
   study objectives.
  Model output (or predictions) always needs careful eval-
uation because the results are generally only as accurate as
the data and the knowledge of the system upon which the
models are based. Model results should be considered as
a guide to the probable system response. In general, when
properly applied, models are the best tools available for
assistance in understanding and making decisions on com-
plex systems. Modeling alone, however, is no substitute for
routine monitoring in problem areas. A successful man-
agement program must include  both  modeling and con-
tinued monitoring.

Description of a Few of the
Many Available Mathematical Models

  The effort to site nuclear waste repositories under the
sponsorship of the U.S. Department  of Energy has been
responsible  for the identification and development of a
suite  of ground-water hydrologic  and transport models.
The modeling technology involves: (1) hydrologic codes to
define the ground-water flow field and indicate water flow
paths and travel times and (2) transport codes to describe
the movement and  concentrations of  the contaminants in
the flow field and the rates of arrival'of contaminants at
points of biosphere uptake.
  The groundwater hydrologic and transport codes iden-
tified and developed include several levels of complexity.
Code selection and use are determined by the study pur-
pose, the quantity and quality of input data and the pro-
cesses felt to be driving the system. There are hydrologic
codes at four levels and transport codes at three levels as
shown in Table I.
  In  addition to  the saturated flow models there are
various levels of hydrologic codes for modeling unsatur-
ated hydrologic systems. These codes for unsaturated flow
modeling are required  when the contaminant source is
located in the unsaturated hydrologic system. The codtt
which Battelle has had experience applying include:

-------
                                                                  GEOHYRDOLOGY MATH MODELING    309
•UNSAT(9)—one dimensional, finite difference numerical
 method, unsaturated flow, with crop water extraction.
•TRUST*00'1"—one or two dimensional, integrated finite
 difference numerical method, unsaturated flow, deform-
 able porous medium.
  Numerous mathematical models  have been  developed
and applied  to a variety of surface  water contamination
problems. These models include a variety of surface water
hydrodynamic and transport models which have been util-
ized to understand these complex systems as well as predict
future surface water quality. These models include hydro-
dynamic, contaminant transport  and sediment transport
models for river, lake and estuarine systems.  A few of
these models which Battelle has had experience in apply-
ing are given in Table II.

APPLICATIONS OF MATHEMATICAL MODELS

  Surface water and groundwater hydrodynamic and con-
taminant transport models are directly applicable to prob-
lems related to siting new facilities, assessing existing con-
taminated ones as well as evaluating remediation options
when required. The  level of sophistication required for
any effort is dictated by data  availability, major phenom-
ena of importance (i.e., saturated or unsaturated flow and
the level of resolution desired. The usefulness of models as
an aid in these assessment and remediation  processes is
best illustrated by examples of past work with similar pur-
pose. Three such examples are given below.
                      Table II.
    Characteristics of a Few of the Available Surface Water
     Hydrodynamic and Contaminant Transport Codes
HYDROLOGIC
 EXPLORED
 DKWAV<2>


 DWOPERO
Two-dimensional (x-y), dynamic wave,  finite
difference code for modeling tidal or river flow in
estuary or river systems.
One-dimensional, kinematic wave, finite differ-
ence code for modeling river flow and account-
ing for rainfall and runoff contributions.
One-dimensional, dynamic wave, finite differ-
ence code for modeling river flow and account-
ing for rainfall and runoff contributions.
CONTAMINANT TRANSPORT
 EXPLORE         Same as above but also simulates contaminant
                 transport for dissolved contaminants.
 DKWAV          Same as above but also handles contaminated
                 sediment transport both overland and in rivers.
 SERATRAW       Two-dimensional  (x-z), transient,  finite ele-
                 ment code for sediment and contaminant trans-
                 port modeling in rivers, lakes and estuaries.
 FETRAP)          Two-dimensional (x-y), transient, finite element
                 code for sediment and contaminant transport
                 modeling in rivers, lakes and estuaries.
 TODAMO         One-dimensional,  transient, finite element code
                 for sediment and contaminant transport model-
                 ing in rivers, lakes and estuaries.
'The capability of TRUST has been expanded to allow stream lines
 and pathlines to be predicted from TRUST input and output streams.
Kepone in the James River

  As a part of the Kepone mitigation feasibility study,(5) a
mathematical simulation of sediment and Kepone (a highly
chlorinated pesticide) transport in the James River estuary
was conducted by applying the sediment and contaminant
transport model, FETRA, to an 85-km river reach between
Bailey and Burwell Bays. The FETRA code consists  of
three submodels coupled together to take into account sed-
iment-contaminant interaction. The submodels are:
1) sediment  transport model, 2) dissolved  contaminant
transport model,  and 3) particulate  contaminant  (con-
taminants adsorbed by sediment) transport model. Trans-
port of sediment and particulate contaminants is simulated
for each sediment type or size  fraction. The modeling pro-
cedure of the FETRA code involves simulation of sediment
transport. The results are then used to simulate dissolved
and  particulate contaminants by accounting for interac-
tion  with the sediments. Changes in river bed  conditions
are calculated including: 1) river bed elevation change,
2) distributions of ratio  of each bed sediment component
in the bed  and bed armoring, and 3) distribution of con-
taminants within the river bed.
  The FETRA code  was applied to  simulate the migra-
tion  of  sediment and Kepone for three river discharges.
Tidally  influenced depth and  velocity distributions in the
study area were obtained by the unsteady, one-dimensional
code, EXPLORE. These results were used  by the two-
dimensional  code, FETRA, to obtain longitudinal dis-
tributions of sediment and Kepone. Hence the results dis-
cussed here are  cross-sectionally averaged values chang-
ing with tidal flow. Comparison of computed results and
field data  for both sediment  and Kepone concentration
indicates very good agreement, confirming the  validity  of
the model.
  Mathematical simulation of Kepone transport  under
most probable flow conditions yields an estimate of 89.1
kg/yr of Kepone transported  seaward from Burwell Bay.
Of this  89.1 kg of total Kepone, 25.3% is carried by sed-
iment, while 74.7% is in a dissolved phase. Since it is esti-
mated that approximately 9600 kg of Kepone is present
in the top 0.3m of bottom sediment in the river, with this
flushing rate, it may take at  least  108 yr for natural water-
sediment flushing mechanisms alone to clean  the  James
River.
  The effectiveness of ten partial Kepone cleanup activ-
ities  to  reduce Kepone  concentrations in the river was
studied  by simulating ten hypothetical conditions by the
FETRA code. Through this cost effective evaluation pro-
cess, it  was quickly determined that none of  the partial
mitigation  options would yield significant reductions  in
total flushing times. Hence, ineffective expenditures were
avoided.

Assessment of Seepage From
Buried Uranium Mill Tailings

  The advantages of returning uranium mill tailings to the
pits excavated during surface  mining operations are being
more widely  recognized  and utilized to minimize any un-

-------
310     GEOHYRDOLOGY MATH MODELING
desirable  environmental effects.03'  Particularly  notable
among the benefits is the better control of radon gas emis-
sion through appropriate mill tailing burial. This advan-
tage alone warrants burial of the tailings.
  As  plans are made for the burial of the tailings in pits,
(as opposed to using conventional tailings ponds above the
ground  surface), concern arises about  the  potential for
contaminant seepage from the buried tailings. Disposal in
the mined out pits places the tailings nearer ground water
and may increase the potential for ground water  contam-
ination.
  The purpose of this study was to examine potential for
ground  water contamination by seepage from buried tail-
ings under  four alternatives  of  clay liners  and tailings
placement. To accomplish this comparison of alternatives,
laboratory work and numerous measurements were made
on materials typical of the site. These measurements pro-
vided the data on soil characteristics necessary for input
to the hydrologic flow  and transport models which  were
used in this assessment study.
  The four  alternatives  considered to minimize seepage of
contaminants from buried mine tailings included:
1. Placement of saturated tailings in a covered pit with a
  clay liner in the bottom, but no side liners
2. Placement of dewatered  tailings in a covered  pit with
  clay bottom liner, but no side liners
3. Placement of saturated tailings in a  pit having a clay
  bottom liner, no  side  liners, and drains  to  facilitate
  pumping  of drainage  solution from sumps placed above
  the bottom clay liner
4. Placement of saturated tailings in a pit with both bottom
  and side clay liners.
  The assessment involved combined subsurface fluid flow
and contaminant transport modeling of the four alterna-
tives  for  controlling seepage  from  buried  tailings. The
assessment included combined saturated and partially sat-
urated flow and contaminant transport models for two-di-
mensional vertical,  cross sections typical of the tailings
burial pits proposed for use at the site.
  The results obtained from the models were the  contam-
inant flow paths away from the tailings pits, the advancing
contaminant flow fronts for  various sorbed  and  non-
sorbed constituents of major environmental concern, and
the associated quantities of contaminant flow for each of
the alternatives. These results  enabled the comparison of
the environmental consequences of the  four alternatives.
It was also possible to gain considerable insight about com-
bining the beneficial alternatives to obtain the best overall
method  for controlling seepage from the buried pits.
  Although four alternatives were considered in this study,
only the first three alternatives, which utilize a bottom clay
liner but no side liner,  provided important results. All of
the results obtained in this study demonstrated that a side
liner  on the upper  70 to 80%  of the typical 30.5m
side wall height is of very little or no benefit in reducing
contaminant seepage; hence, high side wall liners are un-
necessary and use of such would moreover be economically
wasteful.
  The study demonstrated the very real benefits of reduc-
ing the volume of fluid available to seep from the tailings.
The similarity fo contamination of consequences for sys-
tems that use dewatered tailings or underdrains tends to in-
dicate that the means of reducing the volume available for
seepage is less important than the actual volume reduction
of water that can be realized.
  Based upon what was learned from this study, it was
felt that the best contaminant seepage control would be
provided through combining the desirable features of the
four alternatives considered. The recommendations would
include the  accepted  pit  covering  procedure  already
planned for the  site; specifically, the additional proposed
control alternative was:
1.  To provide a  well constructed bottom clay liner placed
  at least 3.0 m above the regional water table and not
  less than 0.9 m thick
2.  To provide stub clay  side liners continuing part way up
  the pit side wall that  form a continuous saucer-shaped
  bottom and side liner  for the pit
3.  To install a network of gravity  drains and pumping
  sumps with the drains in the tailings sufficiently above
  the bottom clay liner  to provide very effective drainage
  of the tailings
4.  To pump  tailings drainage effluent from  sumps while
  the pit is being filled  with tailings and on an as needed
  basis for the first 6 to 8 months after the filling has been
  completed.
  Additional studies were proposed to determine the ap-
plicability of these side liner recommendations at other
sites and to assure acceptable minimum contaminant seep-
age under other conditions. Specifically,  further study is
needed for various soil  materials and other pit configura-
tions using the combined approach of lining and reduc-
ing tailings water. Such  studies should also be followed by
companion field observation as the recommended man-
agement practices are put into use at the Wyoming site or
elsewhere  to  check, verify, and improve upon the recom-
mended disposal control practices. In  this way, it will be
possible to economically further reduce contaminant seep-
age and better assure the adequate disposal of uranium
tailings in burial pits.

Initial Assessment of Fixed Flue
Gas Desulf urization (FDG) Sludge
as a Landfill

  The first  full-scale application of IU  Conversion Sys-
tems, Inc. (IUCS) fixation process (Poz-O-Tec* process)
for treating  flue gas desulfurization (FGD) sludge is in
operation  at  the Conesville Power Station of the Colum-
bus and Southern Ohio Electric Company.'3' The Poz-O-
Tec® process blends fly ash, bottom ash (if desired), lime,
scrubber sludge and other additives. The Poz-O-Te^
sludge is  landfilled  in  a  surface disposal area where it
hardens to a material with relatively low permeability.
  In  this  two phase project the movement of contam-
inants in the groundwater is being studied in order to pro-
vide a basis on which utilities can make decisions regard-
ing the Poz-O-Tec*  method,  in particular  and  the db-

• Registered Trade Mark.

-------
                                                                 GEOHYRDOLOGY MATH MODELING    311
posal of FGD sludge, in general. The Phase I efforts dealt
with initial site characterization and preliminary assess-
ment, whereas  Phase II will include extensive field mon-
itoring, model application, and a final assessment.
  In Phase I, which will be discussed here, a model data
base was formulated and  laboratory  studies were  con-
ducted to determine saturated and partially-saturated per-
meabilities of sludge and ash materials, and to describe the
composition and distribution of the sludge and ash leach-
ate. The data base and laboratory results were used to im-
plement, calibrate,  and verify  a two-dimensional, finite
difference, hydrologic flow model for the study area. The
model output was coupled with a  technique of  graphing
contaminant arrival distribution to predict the distribution
in time and  space of  contaminants traveling  with  the
groundwater within the study area.
  The flow model was calibrated against a potential sur-
face representative of a  high Muskingum River stage. As
a check on the accuracy of the model  prediction,  the
model was verified against a low river stage potential sur-
face.  In both cases, the model prediction  was in good
agreement with the field-measured potential surface.
  A sensitivity analysis  was conducted to determine the
sensitivity of the  model predictions to changes  in trans-
missivity (T) and recharge rate (Q). The findings indicated
that the model is  fairly  sensitive to changes  in these
parameters; therefore, the T and Q  distributions arrived
at in the calibration process cannot be changed apprec-
iably without significantly altering the model  prediction.
It should be noted, however,  that  once calibrated and
verified, the final  modified transmissivity distribution was
within plus or minus 10% of the initial transmissivity
distribution.
  Hypothetical examples were  used to demonstrate  the
methodology for studying arrival distributions of contam-
inants at the Muskingum River. Attention was  focused
on contaminant migration from two surface water ponds
(Pbz-0-Tec®  and emergency sludge pond) within the Poz-
0-Tec®  disposal  area and not on seepage to the ground-
water over the entire disposal  area. This emphasis was
chosen because:
•Laboratory studies and field  investigation have shown
 the Poz-O-Tec®  material to be  highly  impermeable
 (10s to 107 cm/sec),  implying little, if any,  water  per-
 colates through this material
•Man-made and  natural channelization has been devel-
 oped in the  disposal area which  encourages  immediate
 runoff of precipitation  into the Poz-O-Tec®  pond (i.e.,
 water  does  not  remain  ponded  on  the Poz-O-Tec®
 material)
•The Poz-O-Tec® pond is hydraulically connected to the
 water table, thereby providing a direct path for Poz-
 O-Tec® runoff to move into the groundwater
•Leakage from the emergency sludge pond (used to inter-
 mittently store thickened  unfixed FGD  sludge)  to  the
 groundwater could easily have occurred between January
 1977 when the Poz-O-Tec® operation began and Sep-
 tember 1978, at which time the pond was lined.
  The  results  of the Phase I modeling,  characteriza-
tion and data gathering  efforts were used to provide in-
put to the monitoring system design of Phase  II. At the
conclusion of these Phase I efforts it was not possible to
thoroughly assess the groundwater contamination result-
ing from the Poz-O-Tec® process for three reasons:
•Chemical analysis of groundwater obtained from below
 the Poz-O-Tec®  and ash pond found sulfate to be the
 only  water quality  parameter above  drinking water
 standards (secondary drinking water standard)
•The source  of the sulfate (Poz-O-Tec® ,  or fly ash, or
 both) has not been identified
•If it can be assumed that contaminants enter the ground-
 water in pulses in direct response to rainfall, arrival  dis-
 tribution techniques  have  shown  that their concentra-
 tions will be greatly reduced in the process of reaching
 the Muskingum River.
  Additional monitoring and water quality  analysis of the
Phase II efforts will be necessary before  the  degree of
groundwater protection provided by Poz-O-Tec® can be
determined.

SUMMATION

  The authors have briefly outlined the need and role of
mathematical models  in  the site  characterization,  site
assessment  and  remedial action  design phases of a  site
restoration process for a hazardous chemical waste dis-
posal site. In order to dispell some of the misunderstand-
ing and distrust regarding mathematical models and their
useage, a brief overview was presented along with sum-
mary descriptions of  typical model application efforts.
Some  of the basic concepts  of  mathematical modeling
and how the various kinds of currently available codes  can
be effectively applied  to problems of a type similar to
those of site restoration at  a chemical  waste site have
been shown.
  The key points regarding mathematical models are:
(1) Mathematical models are simply tools and like many
   other tools come in a variety of forms from simple to
   complex and must  be utilized by craftsmen skilled in
   their appropriate use
(2) Models provide a  systematic approach for organiza-
   tion and interpretation of data on complex systems as
   well as a  means for achieving a better  understanding
   of these systems and lastly provide a means  for com-
   paring alternatives or making predictions
(3) Models can  not be used  in lieu of data for assess-
   ment or as a substitute for monitoring
  Mathematical models need to be utilized throughout
the site restoration process from the initial characteriza-
tion phase and continuing through the remedial action de-
sign,  construction  and final  surveillance  monitoring
phases. The  capabilities of mathematical models can be
utilized in the site restoration process to:
(1) Help determine when a site is  sufficiently  character-
   ized and  understood for remediation  to be imple-
   mented when it is required
(2) Aid in the design of cost and time effective data gather-
   ing efforts associated with site characterization
(3) Aid in the rapid and cost effective assessment of:
   •whether remediation is required

-------
 312    GEOHYRDOLOGY MATH MODELING
    •a wide variety of remediation options when it is re-
    quired
 (4) Aid in the design  of a cost effective site surveillance
    and monitoring plan.

                  REFERENCES

 1.  Ahlstrom, S.W.,  Foote, H.P., Arnett,  R.C., Cole,
   C.R. and Serne, R.J.  "Multicomponent Mass Trans-
   port  Model:  Theory and Numerical Implementation
   (Discrete-Parcel-Random-Walk  Version).''   BNWL-
   2127, Pacific Northwest Laboratory, Richland, Wash.,
    1977.
 2.  Baca,  R.G., Waddell, W.W., Cole,  C.R.,  Brand-
   stetter, A.  and Cearlock, D.B.  "EXPLORE-I:  A
   River  Basin  Water Quality Model."  Prepared  for
   U.S. Environmental Proltection Agency by Battelle,
   Pacific Northwest  Laboratories,  Richland,  Wash.,
    1973.
 3.  Bond,  F.W.  and Nelson, R.W. "Modeling the Fixed
   FGD  Sludge Landfill-Conesville,  Ohio  (Phase I)."
   EPRI-CS-1355. Prepared for Electric Power Research
    Institute by Battelle, Pacific Northwest Laboratories,
   Richland, Wash., 1980.
 4.  Chow, V.T.  "An  Introduction to Systems Analysis
   of Hydrological Problems." Proc.  of the Second In-
    ternational Seminar for Hydrology Professors. Logan,
   Utah, 1970.
 5.  Dawson, G.W. "The Feasibility of Mitigating Kepone
   Contamination  in  the  James  River Basin."  EPA-
   440/5-78/004A. Prepared  for  U.S. Environmental
   Protection Agency by Battelle,  Pacific Northwest
   Laboratories, Richland, Wash., 1978.
 6.  DeMier,  W.V., Cloninger, M.O.,  Burkholder, H.C.
   and Liddell,  P.J.  "GETOUT-A Computer Program
   for Predicting Radionuclide Decay Chain  Transport
   Through Geologic  Media." PNL-2970, Pacific North-
   west Laboratories,  Richland, Wash., 1979.
 7.  Fread, D.L. "National Weather Service Operational
   Dynamic Wave Model." Hydrologic Research  Labora-
   tory, National Weather Service, NOAA, Silver Spring,
   Md., 1978.
 8.  Gupta, S.K., Cole, C.R. and Bond, F.W. "AEGIS.
   Finite-Element  Three-Dimensional  Ground-Water
   (FE3DGW) Flow Model—Formulation Program List-
   ings and User's Manual." PNL-2939, Pacific North-
   west Laboratories,  Richland, Wash., 1979.
 9.  Gupta, S.K., Tanji,  K.K., Nielsen,  D.R.,  Biggar,
   J.W., Simmons, C.S. and Maclntyre, J.L. "Field Sim-
   ulation of Soil-Water Movement with Crop Water
   Extraction."  University of California, Department of
   Land, Air and Water Resources, Water Science and
   Engineering Paper No. 4013, 1978.
10. Narasimhan, T.N.  and Witherspoon, P.A.  "Numeri-
   cal Model for Saturated-Unsaturated Flow in Deform-
   able Porous Media 1.  Theory." Water Res. Res.  13,
   1977,657.
11. Narasimhan, T.N. and Witherspoon, P.A. "Numer-
   ical Model for Saturated-Unsaturated Flow in De-
   formable Porous Media 3. Applications." Water Res.
   Res. 14, 1978, 1017.
12. Narasimhan, T.N., Witherspoon, P.A. and Edwards,
   A.L.  "Numerical Model for Saturated-Unsaturated
   Flow in Deformable Porous Media 2. The Algorithm."
   Water Res. Res. 14, 1978, 255.
13.  Nelson, R.W.,  Reisenauer, A.E.  and Gee,  G.W.
   "Model  Assessment  of Alternatives  for  Reducing
   Seepage from  Buried  Uranium  Mill Tailings at the
   Morton Ranch Site  in Central Wyoming."  PNL-
   3378,   Pacific  Northwest   Laboratories,  Richland,
   Wash., 1980.
14.  Nelson, R.W.  and Schur, J.A. "AEGIS.  PATHS
   Groundwater Hydrologic Model." PNL-3162, Pacific
   Northwest Laboratories, Richland, Wash., 1980.
15. Onishi, Y., Serne, R.J., Arnold, E.M., Cowan, C.E.
   and Thompson, F.L. "Critical Review: Radionuclide
   Transport,  Sediment  Transport, and Water Quality
   Mathematical Modeling; and Radionuclide  Adsorp-
   tion/Desorption  Mechanisms."  PNL-2901,  Pacific
   Northwest Laboratories, Richland, Wash. 1980.
16. Onishi, Y. and Wise,  S.E.  "Mathematical Modeling
   of Sediment and Contaminant Transport in the James
   River  Estuary." Proc.  of the Specialty Conference on
   Verification of Mathematical and Physical Models in
   Hydraulic Engineering, ASCE, College Park, Mary-
   land, August 9-11, 1980, 303-310.
17. Onishi, Y. and Wise, S.E. "Finite Element Model for
   Sediment  and  Toxic  Contaminant  Transport in
   Streams." Proc. of the Specialty Conference on Con-
   servation and  Utilization of Water and Energy Re-
   sources,  ASCE, San  Francisco,  California, August
   8-11, 1979,  144-150.
18. Reisenauer, A.E. "AEGIS. Variable Thickness Tran-
   sient Ground-Water Flow Model. Volume 1. Formula-
   tion." PNL-3160-1, Pacific Northwest Laboratories,
   Richland, Wash., 1979.
19. Reisenauer, A.E. "AEGIS. Variable Thickness Tran-
   sient Ground-Water Flow  Model.  Volume  2. User's
   Manual."  PNL-3160-2, Pacific Northwest  Labora-
   tories, Richland, Wash., 1979.
20.  Reisenauer, A.E.  1979.  "AEGIS.  Variable Thick-
   ness  Transient Ground-Water Flow Model. Volume
   3. Program Listings."  PNL-3160-3, Pacific Northwest
   Laboratories, Richland, Wash., 1979.
21.  Washburn, J.F., Kaszeta, F.E., Simmons, C.S. and
   Cole,  C.R.  "AEGIS.  Multicomponent Mass Trans-
   port Model: A Model for Simulating Migration of
   Radionuclides in Ground Water." PNL-3179, Pacific
   Northwest Laboratories, Richland, Wash., 1980.
22. Whelan, G. "Distributed Model for Sediment Yield."
   Master's Thesis Mechanics and Hydraulics, University
   of Iowa, Iowa City, Iowa, 1978.

-------
              PARAMETRIC ANALYSIS OF GEOLOGICAL
                        HAZARDOUS WASTE DISPOSAL
                                     EDUARDO A. FIGUEROA
                                     FRANK L. PARKER, PH.D.
                         Department of Civil and Environmental Engineering
                                         Vanderbilt University
                                         Nashville, Tennessee
INTRODUCTION

  The uncontrolled discharge of hazardous material has
steadily increased not only because of increases in  pop-
ulation,  agriculture and industry, but also because law
and regulations, such as the Clean Air Act (as amended)
and the Marine Protection, Research and Sanctuary Act
(as amended) curtail the discharge of hazardous pollu-
tants in the air and water, leaving the alternative of land
disposal as the last or final disposal practice. And, as can
be seen from past experiences, the best available technol-
ogy has not always resulted in secure landfills. There is
growing evidence of the vulnerability of clay and plastic
liners to the effects of certain chemicals, there is also the
difficulty of installation and susceptibility of plastic liners
to puncture. Moreover there are little data to show that
the lifetime of the plastic liners,  even if  they stay intact,
will exceed the lifetime of the hazardous  wastes they con-
tain.
  Consequently, the authors have done a parametric an-
alysis to see how the waste disposal repository acts as a
total system and to determine what features need to be
improved to guarantee the safety of the disposal facility.
  To study the problem, one has to look at the system as
a whole, where the various parameters interact, and allow-
ing variation in values of the parameters according to the
uncertainty of  these values. To do  this, the authors
adopted a simple conceptual model in which the repository
is located close to the ground surface (less than 10 meters),
the waste is packaged, is leached by groundwater, is  sorb-
ed during its travel underground and finally reaches the
biosphere. Dilution in the biosphere has  not been consid-
ered.  Concentrations in the  biosphere have been  com-
pared to recommended limits.
  In this initial screening  of  geological hazardous waste
disposal sites, one can afford to use simpler models, so
that with a minimum of information they can be satisfac-
torily used to predict the behavior of contaminants under
normal or extreme conditions. A simple one-dimensional
analytical solution to the diffusion-dispersion equation in
porous media was used to show that, in a multiparameter
system characterized by the packaging, the waste disposal
technique, the geology and hydrology of the site, there are
a number of combinations (multidimension response sur-
face) that will result in concentrations released to the bio-
sphere that are less than permissible values, but more im-
portant that there are many parameter combinations that
will not meet release criteria.

ANALYTICAL MODEL

  The  one-dimensional transport equation  for horizon-
tal, single-phase flow  in saturated, unconsolidated and
homogeneous media can be expressed as:
9c      9c   _. 92c   p9s..  n
— = - u—+ D r-7--r-+ kcn
9t      9x     9xz   e9t
(1)
Ogata and Banks2 presented a general solution of the form:

                                   l+i
— = tt[erfc[  !
           L2(0/Pe)1/2
                             erfc
 (2)
where

6  = (u*t)/X

Pe= (u*X)/D
Sorption is implicitly considered as "u" was used as the
velocity of the pollutant, under the assumption that the
mass transfer of solute to sorption sites is rapid and that
the sorption equilibrium is linear, ds/dc = Kd, where Kd
is the distribution coefficient and the term [1  + (Kd/e )]
is known as the retardation factor and gives the relative
velocity of movement of a sorbing solute to the movement
of water.

Input Data

  The study considered the solubility  limit as the initial
concentration of the pollutant in the aquifer and the longi-
tudinal dispersion coefficient with a fixed value of 0.9 m/
day.  The parameters that were varied were: groundwater
velocity 0.1, 0.01,  0.001, and 0.0001 m/day,  partitioning
coefficients of 1, 10, 100, 1000, and 10,000 1/g and dis-
tance to the biosphere of  100,  1000, and 10,000 meters.
The data are summarized in Table I. Biological or chem-
ical degradation were not considered, since the pollutants
under study are "conservative" substances.
                                                  313

-------
  314    GEOHYDROLOGY MATH MODELING
                          Table I.
             Chemical Characteristics of Pollutants
                 Partition
                 coefficient
                 d/g)

                 10M04

                 100-250

                 1-100
                               Permissible
                               ambient cone.
  Solubility
  limit
  (mg/D
                              0.001

                              0.004

                              0.005
 0.012

 0.26

 1.75
Pollutant



AROCLOR 1254

ENDRIN

TOXAPHENE



RESULTS
    Three different pollutants were  studies:  PCB, toxa-
  phene,  and Endrin.  Each  has  different  sorption charac-
  teristics, different solubility limits and different permiss-
  ible water concentrations.
    Aroclor 1254 (PCB), has high partitioning coefficients
  from 10* to 1061/g, low solubility limit of 0.012 mg/1 and
  permissible ambient concentration of 0.001 ug/1.  Endrin
  has partitioning coefficients  of 100  to 250 1/g, solubility
  limit  of 0.26  mg/1,  and permissible ambient  concentra-
  tions of 0.004 ug/1. Toxaphene has partitioning coeffic-
ients of 1 to 1001/g, solubility limit of 1.75 mg/1, and per-
missible ambient concentration of 0.005 ug/1. The ambient
concentration values were those recommended by EPA in
Quality Criteria for Water3 and the solubility values are
those in the treatability manual.(4) Partitioning coefficients
have been obtained from different publications and from a
Vanderbilt University study.<5>
  The results of the analyses are presented in Table II and
in normalized form in Figures 1, 2, and 3, so the analysis
will be valid for any pollutant if we know the initial con-
centration, the release  criteria concentration, and the par-
titioning coefficient.
  At 1000 meters from the source, which is considered as a
realistic distance in humid climates  for the pollutant to be
released to the biosphere in the case of near surface geo-
logical repositories, Aroclor 1254 will reach the release cri-
teria limit with groundwater velocities of 0.1, 0.01, 0.001,
and  0.0001 m/day, at 6.25  x 103,  23.0 x  103, 35.0 x 105
and  37.0 x 103 days,  respectively.  For the same ground-
water flow rates, Endrin will reach the biosphere in 8.7
x 103, 22 x 103, 29 x 103, and 31 x 103 days, and toxaphene
at 22.5 x 103,24 x 103,25 x 103, and 25.5 x 103 days.
  As one can see,  pollutants can be expected to appear in
concentrations higher  than recommended values in about
10
-4




o
O

O


-5
10
  -6
10
  -7
10
           ax,
IIM

  ay
                          Kd   100
Velocity
(m/d)
a 0.
b 0.
c 0.
d 0.
1
01
001
0001
Dist. from
source
Tm)
x
y
z

100
1000
10000

             [dx
          bx'
                                                                                             '
                                                                                                        c
    10*
                         10J
    10"
                                                                        105   TIME   (d)   lo6
                                                       Figure 1.
                                       Time to Reach Ambient Criteria Concentrations

-------
                                                                  GEOHYDROLOGY MATH MODELING     315
                       Table II.
      Time to Reach Ambient Criteria Concentration (d)
Velocity (m/d)
Distance (m)
AROCLOR 1254
ENDRIN
TOXAPHENE
0.1
100
240
200
230
1000
6250
8700
22500
10000
88000
110000
1650000
0.01
100
350
225
250
1000
23000
22000
24000
10000
600000
740000
2400000
Velocity(m/d)
Distance (d)
AROCLOR 1254
ENDRIN
TOXAPHENE
0.001
100
370
240
255
1000
35000
29000
25000
10000
2300000
2 100000
2600000
0.0001
100
390
245
260
1000
37000
31000
25500
10000
3500000
2900000
2700000
 20 years, with high groundwater velocities and  in  100
 years, with low velocities. These times are not sufficiently
 long to prevent contamination of the biosphere from these
 contaminants.
                    Even if one considers the best possible conditions, high
                  retardation factors, low groundwater velocities and high
                  ratios of release concentration over solubility (10-4), the
                  time involved is only 120 years. If one considers low re-
                  tardation factors, high  groundwater velocities and mod-
                  erately low  ratios of release concentration to solubility
                  (10-7), at 1000 meters from the source, the release criter-
                  ion will be reached in a decade.

                  CONCLUSIONS

                    This simple model shows how recommended maximum
                  release concentrations to the biosphere can be reached in
                  short periods of time if the leachate is permitted to reach
                  solubility limit values in the aquifers underlying hazardous
                  waste sites. Natural existing conditions will not secure the
                  pollutants for extensive periods of time. Only in very lim-
                  ited situations, with low concentrations of hazardous ma-
                  terials located in extremely favorable hydrogeological con-
                  ditions, (low precipitation and thick clay formations), will
                  the pollutants be secure in near surface geological repos-
                  itories. If these calculations are correct,.for near surface
                  storage one can assume that one can only substantially de-
                  crease the groundwater velocity with great effort and the
                  distance to the biosphere with equally great effort.
 -4
10'



 o
 0

 0


r5
n-7
        ax
                (c,d)x
ay
                        Kd   1
                                                                    a
Velocity
(m/d)
a 0
b 0
c 0
d 0
.1
.01
.001
.0001
Dist. from
source
(m)
X
y
z

100
1000
10000

                                                                                       b
                                               10H
                             10J
                                                                            time   (d)
                                                     Figure 2.
                                     Time to Reach Ambient Criteria Concentrations

-------
  316

  -4
GEOHYDROLOGY MATH MODELING
                                                                         I  I   I I
10
  o
 O

 O

  -5
10
10
  -6
      |   I  I  I M I j

     axil (b,c,d)x
|   I  I  I I I  I
I    I  I   |  I I I I

 ay  ,(b,c,d)y
                                                                                                       TTT
                                                                                                 aa   bz
                      Kd   10000
                           Velocity
                              (m/d)
                             Dist. from
                                source
                                 (m)
                           a  0.1       x 100
                           b  0.01      y 3000
                           c  0.001     z 10000
                           d  0.0001
                                                                                                      (c,d)z
                          10J
                                      10
                                105   *:^.^  /w\   106
                                                     Figure3.
                                      Time to Reach Ambient Criteria Concentrations
                                                                           time   (d)
     Consequently, the major parameters that  are suscep-
   tible to revision are the solubility limit and the partition
   coefficient.  To decrease the solubility limit  requires an
   alternation  in the form  of the waste. This, of course,
   would differ for specific waste types. One  may conclude
   that, for certain wastes, only total destruction, by incin-
   eration,  for example, is acceptable.  Whereas  for  the
   others, it is possible that encapsulation methods may be
   sufficient. It may also be possible to change by altering
   the natural  chemistry of the site. However, this can only
   be a temporary solution.
     Therefore, it may be necessary to examine alternatives
   to near surface disposal. Disposal at intermediate depths
   (>50 m) would in general, decrease the groundwater flow
   and increase the distanceto the biosphere.

   NOMENCLATURE
   co —initial concentration mg 1 -'
   c  —final concentration mg 1 -'
   D —dispersion coefficient m'd -'
   £ —soil void fraction %
   k  —reaction constant
   n  —order  of reaction
   s  —mass of solute sorbed per unit dry mass of solid
   t  —time
                                                 u  —average velocity m d -'
                                                 x  —distance in flow direction m
                                                 p  —bulk density of soil g m -3
                                                 REFERENCES

                                                 1. Council of Environmental Quality, "Contamination of
                                                    Groundwater by Toxic Organic Chemicals," U.S. Gov-
                                                    ernment  Printing Office,  Washington, D.C.  20402,
                                                    January,  1981.
                                                 2. Ogata, A.,  Banks, R.B.,  "A Solution of the Differ-
                                                    ential Equation of Longitudinal Dispersion in Porous
                                                    Media,"  U.S.  Geological Survey Prof. Paper 411-A,
                                                    Washington, D.C., 1961.
                                                 3. U.S. Environmental Protection Agency, "Quality Cri-
                                                    teria for Water," U.S. EPA, Washington, D.C. 20460,
                                                    1976.
                                                 4. U.S. Environmental Protection Agency, "Treatability
                                                    Manual,  Volume I. Treatability Data,"  EPA-600/8-
                                                    80-042a, July, 1980.
                                                 5. Jaffe, P.R., Parker, F.L., Wilson, D.J., "The Fate of
                                                    Toxic Substances in Rivers,"  Proc.  of the ASCE
                                                    National  Conference of Environmental  Engineering,
                                                    Atlanta, Georgia, July, 1981.

-------
           LOCATING OF GROUNDWATER POLLUTION
                SOURCES FROM LIMITED FIELD DATA

                                     JACK C. HWANG, Ph.D.
                                  ROBERT M. KOERNER, Ph.D.
                                  Department of Civil Engineering
                                          Drexel University
                                     Philadelphia, Pennsylvania
INTRODUCTION

  Conventional methods of locating an unknown ground-
water pollution source require an extensive number  of
observation wells from which sufficient data can hope-
fully be obtained to plot concentration contours for the
identification of the pollution source.(1) Such an operation
usually is quite expensive and furthermore, with limited
field data, results are approximate, at best. However,  by
assuming constant dispersivity of the contaminant and the
steady state  groundwater  flow field,  the mass transport
equation for the contaminant becomes linear. Thus,  to
locate the pollution source is now equivalent to solving
an identification problem of a linear dynamic system.
  The aquifer parameter identification problems or  so
called inverse problems, have been investigated for the past
10-15 years.  Faust and Mercer(2) gave a brief account of
its recent development.  In general,  the calculation pro-
cedure for groundwater flow consists of finding a parame-
ter set (transmissivity, storage coefficient, sources, etc.)
that minimizes deviations between observed and calcu-
lated values of hydraulic head. Thus, for steady state flow
problems, the criterion adopted to define the optimal  set
of parameters is to minimize:
  m
                                               (1)
where Hj and hj are the observed and calculated heads at
node i for a total of m observations. For the unsteady state
flow problems the criterion function is the minimization
of:
                                               (2)
where hjjj.j and h,,,^ are the calculated and the observed
heads made  from O to T at mth discretized  point cor-
responding to time tj, where j = 1,2, ...T.
  Several methods are available in order to achieve the
minimization. Cooley(3) derived the modified Gauss-New-
ton procedures after linearizing  the steady state equation
using a similar technique to that used by Yeh and Tauxe.(4)
Frind and Pinder(4) solved the inverse problem for trans-
missivity by  the Galerkin finite element approach using
steady state hydraulic heads as input data. For the dynam-
ic system, Yeh(6) presents a good review on the methods of
aquifer parameter estimation for unsteady state flow in an
unconfined aquifer. Methods such as quasi-linearization,
maximum  principle and gradient method, influence coef-
ficient  method and  the minimax-linear programing  ap-
proach are presented in detail  for achieving the minimi-
zation.
  The application of the system's approach, however, for
the identification of a groundwater pollution source simi-
lar to those for aquifer parameter identification seems to
have been  overlooked by previous researchers. In fact, if
a steady-state flow field  is assumed, the system control
equation is linear.  The criterion is then the minimization
of the following expression:
       m
  Jt  = S(yi>t-yin)2
       i = 1
(3)
in which y ,t and y-t t are the calculated and observed heads
of concentration made at ith well corresponding to time t.
The matrix equation resulting from the finite element dis-
cretization for the mass transport equation can be written:
  c = Bc + g
                                                                                                      (4)
where the location and the strength of the pollution source
are contained in the vector g. The estimate, g, approaches
the true value of g as Jt approaches a minimum value. The
method adopted  in this study for achieving the minimi-
zation is based on system sensitivity theory which will be
described in the following sections. As written, the paper
presents the theoretical foundations for the study  which
will be presented at a subsequent time.

ANALYTIC FORMULATION

  The  governing equation  of  two-dimensional  mass
transport of a contaminant in groundwater flow can be
written as follows:
3c_
9t
T^+ v -^
3x       3y
(5)
                                                  317

-------
318    GEOHYRDOLOGY MATH MODELING
where
  c     =
  u,v   =
  Dx,Dy =
  q
concentration of the contaminant
velocity components of groundwater
components of dispersivity
source term
  The source term q can be formulated according to the
physical behavior of the pollution sources. Adey and Con-
nor(7) outlined some expressions of the source term includ-
ing sources with time decay, rate, etc. In order not to ob-
scure the trend of thought in a jungle of equations, a sim-
plified, but typical, formulation will be described.
  The pollution source considered here is of a point source
(or sources) with constant strength. Thus, the source term
is as follows:
  q = Eqit?(x-x*i)(y-y*i)                      (6)

where
t?      =  Dirac Delta function
qi      =  strength of the ith source
(x*j,y*i) =  coordinates of the ith source
  The major axes of the  dispersivity  tensor coincide with
the  global axes of  the  aquifer and,  furthermore,  the
velocity field of the groundwater is steady but may vary
spatially.
  With proper  initial  and  boundary conditions  pre-
scribed, the governing equation can be solved numerically
to determine the concentration distribution of the pollut-
ant over a period of time. Using the finite element method,
it can be  shown"" that the Galerkin's  expression for an
aquifer element of depth b and fluid density p is:
                                                          where M, K and A are Jhe assembled matrices of M, K
                                                          and A, respectively, and c and q are assembled vectors of c
                                                          and q, respectively.
                                                            For convenience, we can drop the notation "--" and (8)
                                                          to obtain the following:
                                              c = Be +  g

                                              where

                                              B = -M"1  (K + A)

                                              g = M" q
                                                                                                            (9)



                                                                                                           (10)

                                                                                                           (11)
                                                If one assumes a zero normal gradient of concentration
                                              across the whole boundary, the entries in the vector q all
                                              vanish except those corresponding to source locations. For
                                              example, for a source of strength r located at the j node of
                                              the descretized finite element domain then the entry of the
                                              j th row of vector q is r.
                                                The problem of locating a pollution source is an inverse
                                              problem of that just being described. The concentration is
                                              considered known, in fact, they are given from the concen-
                                              tration measurements  taken at equal time intervals at ob-
                                              servation wells. The problem is to identify the unknown
                                              vector  g which  contains the information of where the
                                              source (or sources) is  located and what is its strength (or
                                              strengths).
c  {(K +  A) c  - Me -  Q} = 0


in which the superscript T indicates transpose and,
c  = concentration vector of nodal unknown
c  = time derivative of c
0  - interpolation function
                                       (7)
                + D
                                                dx  dy
                          dx dy
K =


A = //pb[u+  (f£)T +
              m
M = //pb  if>   dx dy


Q - /   q^ds + Z q± 6  (x -  x*)(y -  y*)
q  =  normal derivative of c prescribed on segment Sj of
      the boundary

  By assembling the element matrices into global matrices
for the whole solution domain and applying the boundary
conditions, one arrives at the final form of finite element
formulation of the problem as follows:
______                           (8)
Me +  (K +  A) c  =  q
                                                         METHOD OF SYSTEM SENSITIVITY

                                                            Let yt be a measurement vector at time t then:
                                                            y, = Fct                                        (12)
                                                         where  F is the location matrix of the observation wells.
                                                         F is an m x n matrix with zero entries except at well nodes:
                                                                                                          (13)
0 ••

0 -

0 .
                                                                           • 0

                                                                          •10
                                                                        m
                                                                                                          (14)
                                                If yt is an estimate of y, and

                                                J, = (y,-y,)TR(y,-yt)
                                              then J, represents the magnitude of error between the es-
                                              timated and  observed concentration at time t. (R is a
                                              weight matrix of the observation  wells.) In particular,
                                              for R  = I (identity matrix) then
                                                     m

                                                     E   (yi
                                                    1=1

-------
                                                               GEOHYDROLOGY MATH MODELING    319
  The system sensitivity theory  provides a  recurrence
formula for g (estimate of g) for  the next time step in a
manner so as to minimize Jt. When Jt approaches a
minimum value, g approaches the real g.
                                            agi is  the  so called trajectory sensitivity vector which
                                           should be obtained from the governing equation of the
                                           system. From equation (9) it can easily be shown that
St+At
  9g
                                                                                                         (23)
in which A is the step size, ranging between O and 1 for
normalized elements.  ^ will be  evaluated through  the
governing equation of the physical system and the mea-
surement matrices.
  For convenience,
  let
     = yt - yt
                                   (17)
                                           Furthermore, if the time derivative  of the concentration
                                           is approximated by a finite difference formulation, i.e.,
                                                Ct+At  " Ct
                                                     At
                                                                                           (24)
                                          the combination of equations (23) and (24) gives the de-
                                          sired  expression of  the  trajectory  sensitivity  vector  in
                                          iterative form:
  thus
  Jt =  et R  et
  and
                                   (18)
                                            3c
                                                                     =  (B
                                                             3c
                                                               t+At
                                                                                           (25)
                                                                          i =  1.2,... n
  9J
  now
  3J
  and
3J
        t    9et  '  3*d
              i =1.2	n    (19)
 (efc  R et) =  2R  et
                                                (20)
(yt  - yt)
                                                (21)
                                                        in which
                                                        a.
                                                                    •*- i  th row
                                           In short, equations (16), (22) and (25) compose the itera-
                                           tive numerical system for obtaining the  converged value
                                           ofg.

                                           OVERVIEW OF COMPUTATIONAL PROCEDURE
    T~
                      ....  n
  thus


  3J
  1.2,  ....  n
                                                (22)
                                             For illustration, a generalized example will be discussed.
                                           Consider a given suspect site that is arbitrarily subdivided
                                           into a finite element  pattern of quadrilateral elements.
                                           The groundwater flow pattern can be in any direction with
                                           respect to this pattern but must be known or, at least, as-
                                           sumed. The  observation wells are located on the grid and
                                           are numerically generated through equation (8). With data
                                           from these wells, the problem becomes one of locating
                                           the pollution source which is arbitrarily shown in Figure
                                           1. (Note  that multi-source problems can also be handled
                                           by this method.)
                                             The computer programs  are written in such a way that
                                           the assembled matrices, M,  K, A can be called  out  for
                                           matrix operation. Arbitrarily select 1/3 of the total nodes

-------
320   GEOHYDROLOGY MATH MODELING
•  Observation wells  X  Source
                        Figure 1.
       Example Problem Showing Typical FEM Grid and
               Source and Observation Wells
and regard them as observation wells. The generated time
series of  concentrations at these nodes then serve as
measurements at observation wells.
  The flow chart of a sample computer program is given
in Figure 2.  The  convergence criterion currently being
used is:

I   (gI+1  - g^/gj-  I 1 io"3                  (26)


i.e., when the relative error in each entry of the g vector is
less or equal to  10~3, it is considered to be a convergent
system.
  During  the  initial iterations,  adjustments  are  needed
for selection of proper step size (A). To big a step size may
cause the process  to diverge.  It, in  general, reflects on
                                                    Supply initial values
                                                   Main Iteration scheme
                                                                                    exit
                                                   Figure 2.
                                  Computer Flow Chart for Program Currently in Use

-------
                                                                 GEOHYROLOGY MATH MODELING   321
the growing residue. The step size should be cut in half if
such a situation occurs. As noted in the introduction, pro-
grams are currently being run using the method in a wide
variety of situations.  Results will be forthcoming shortly.
They are in the form of pollution contours, the focus on
which clearly identifies the pollution source (or sources).

SUMMARY AND CONCLUSIONS

  The general problem of locating a pollution source by
having data from a limited number of observation wells is
quite challenging, yet worthwhile, since its environmental
impact  is  extremely important. One potentially useful
analytic technique for this purpose is the use of the finite
element method (FEM). While a good deal of information
exists on this subject as far as taking a known source and
mapping out its migration, the inverse problem of finding
the source from a limited number of observation points is
more difficult.
  Still more important is that of the available methodolo-
gies,  the application  of a systems approach (as described
herein) is a new solution technique. With a steady-state
flow field assumed, the system  control equation  is linear
and the criterion function can be minimized to  an arbi-
trarily selected value.
  The method of system sensitivity was described in the
paper along with related theoretical  considerations. This
analytic formulation  was the basic objective in  this study.
While work is still ongoing,  the basic programs are de-
veloped (see flow chart of Figure 2) and various problems
are currently being evaluated.
REFERENCES

1.  Roux, P.H. and Althoff, W.F., "Investigation of Or-
   ganic Contamination of Groundwater in South Bruns-
   wick Township, New Jersey," Groundwater, 18, No. 5,
   1980.
2.  Faust, C.R. and Mercer, J.W., "Groundwater Model-
   ing: Recent Developments," Groundwater, 18, No. 6,
   1980.
3.  Cooley, R.L., "A Method of Estimating  Parameters
   and Assessing Reliability for  Models of Steady State
   Groundwater Flow," Water Resource Research,  13,
   No. 2, 1977.
4.  Yeh, W. W-G and Tauxe, G.W.,  "Optimal  Identifi-
   cation  of Aquifer  Diffusivity  Using Quasilineariza-
   tion," Water Resource Research, 7, No. 4, 1981.
5.   Frind, E.G. and  Finder. G.F., "Galerkin  Solution of
   Inverse  Problem  for Aquifer Transmissivity," Water
   Resource Research, 9, No. 5,1973.
6.  Yeh, W. W-G, "Aquifer Parameter Identification,"
   Journal of Hydraulic Division, ASCE, 101, HY9,1975.
7.  Adey, R.  and Brebbia,  C.A. "Numerical  Method in
   Fluid Dynamics," Pentech Press, 1974.
8.  Connor, J.J.  and  Brebbia, C.A.,  "Finute  Element
   Techniques for Fluid Flow," Newnes-Butterworths,
   1976.
9.  Frank,  P.M.,  "Introduction  to System  Sensitivity
   Theory," Academic Press, 1978.

-------
     DIOXIN INVESTIGATIONS IN SOUTHWEST MISSOURI
                                          DANIEL J.HARRIS
                          U.S. Environmental Protection Agency, Region VII
                                   Surveillance and Analysis Division
                                          Kansas City, Kansas
INTRODUCTION

  During 1971, in a well-known incident of waste misman-
agement, three horse arenas in Missouri were sprayed for
dust control with a mixture of used oil and chemical waste
still bottoms. In the ensuing three years, 64 horses died and
ten adults and children became ill as a result of contact
with the contaminated soil.
  It was not until 1974, and after excavation and removal
of  the  arena soils,  that  tetrachlorodibenzo-p-dioxin
(TCDD) was identified as the causative agent. Investiga-
tion by the Missouri Division of Health and the Center for
Disease Control (CDC) traced the source of this substance
to the defunct Northeastern Pharmaceutical and Chemical
Company (NEPACCO), which had leased manufacturing
facilities at a chemical plant in Verona, a small com-
munity in southwest Missouri.
  As a result of an anonymous phone call from a former
NEPACCO  employee,  the  Environmental  Protection
Agency  (EPA) became interested in the past activities of
this company in  1979. The caller made  allegations  con-
cerning other sites where NEPACCO wastes were disposed
of and provided names of prospective interviewees. The
historical record of the company and the  specter of addi-
tional TCDD sites, coupled  with the Agency's burgeon-
ing interest in hazardous wastes, prompted  an  immed-
iate investigation.  This paper summarizes the results of
that ongoing effort.

HISTORY AND BACKGROUND

  In November of 1969, NEPACCO leased a specific man-
ufacturing area of the Hoffman-Taff, Incorporated chem-
ical plant in Verona for the purpose of making hexachlor-
ophene. Hoffman-Taff had  previously used this area in
the manufacture of "Herbicide Orange"  for the military.
As  a  result of curtailed  usage in Viet Nam,  Hoffman-
Taft lost its contract and the facilities sat idle for a num-
ber of months. Shortly after NEPACCO  leased the man-
ufacturing line, the entire plant was acquired by  Syntex
Agribusiness, Inc.
  The beginning processes of the herbicide and hexachlor-
ophene operations were similar, and  apparently the idle
facilities represented an  opportunity  for NEPACCO to
begin  manufacturing with minimum capital expenditure.
Both  processes involved  the intermediate production of
2, 4, 5-trichlorophenol (2, 4, 5-TCP). A major difference
in the two operations, and of central interest to this in-
vestigation, was the subsequent treatment of the 2, 4, 5-
TCP. Hoffman-Taff  produced a technical  grade herbi-
cide which permitted  impurities, including TCDD, to go
forward into the final product. The pharmaceutical grade
hexachlorophene  necessitated distillation of the 2,  4, 5-
TCP, which resulted in a still bottom waste stream with
TCDD concentrations in the range of 350 mg/1.
  Between  approximately April  1970  and  January of
1972, NEPACCO is  reported  to  have produced 328
batches  of 2, 4, 5-TCP.  Shortly thereafter, the company
lost its market and ceased operations as a result of the lim-
ited use restriction imposed on  hexachlorophene by the
Food and Drug Administration (FDA).

EPA INVESTIGATION

  Agency  investigation  of the past activities  of  NE-
PACCO began in October of 1979 and resulted from an
anonymous phone call from a disgruntled employee who
had  been terminated by Syntex; he  had worked in the
NEPACCO area. This individual identified other former
employees who allegedly had knowledge of NEPACCO
operations, waste management  practices, and  disposal
sites in the area.
  In response to this information, a field team spent two
weeks in southwest Missouri conducting interviews with
former NEPACCO and Hoffman-Taff employees, inde-
pendent trash haulers and landowners and in  reconnais-
sance of reputed disposal sites.  A parallel effort of the
investigation included retrieval and review of all available
historical information on the company which encompassed
FDA files, Missouri Department of Natural Resource and
Division of Health files,  court depositions from the horse
arena incident, as well as in-house documents.
  This initial investigation  resulted  in the  identification
of three potential sites which included Baldwin Park in
Aurora, the Syntex trenches west of the Verona plant,
and  the Denney Farm Site, which was sampled by the
Agency  in  April of  1980. As a result  of  the publicity
surrounding the Denney site investigation and the num-
erous contacts between EPA employees and area resi-
dents, other rumors came to the Agency's attention. These
leads were pursued in August of 1980 and resulted in the
identification of additional sites.
                                                   322

-------
                                                                                    CASE HISTORIES     323
  To date, the Agency has interviewed approximately 75
individuals and has identified seven proven or potential
NEPACCO waste disposal sites which are in various stages
of investigation or cleanup. In significance, these sites vary
from major undertakings such as the Denney  Farm Site,
where remedial work is underway, to farms having small
volumes of NEPACCO wastes with  TCDD  concentra-
tions in the parts per trillion (ppt) range.

SITE STATUS

Baldwin Park
  Located in  Aurora  in  Karst topography,  this 170-
acre site was at one time a lead and zinc mining area
with shafts, reportedly, up to 380 ft. deep. Following ex-
haustion of mineral resources, the tract,  or at least a 30-
acre portion of  it,  was used as a  municipal dump by
Aurora, as well as surrounding communities.
  According to interviewees,  large quantities of expended
filter cake material, used by  NEPACCO  for product
purification, was disposed  of in  the  dump along with
municipal trash. In 1973, the site was closed and reclaimed
and given its present name. Although a sample  of the
NEPACCO filter cake discovered at another location was
found to contain TCDD at a level of 8  ppb,  samples of
runoff and municipal and private wells surrounding Bald-
win Park have not revealed detectable levels of NEPACCO
contaminants.  No final disposition of the site has been
made.

Denny Farm Site

  Identified during the October 1979 reconnaissance, the
Denney farm is the largest proven site discovered to date.
The site is in a rural wooded area of Karst topography
on a ridge top approximately one third mile from the near-
est surface water which is a losing stream. Visual evidence
of the existence  of the site included a depression in the
ground about 10 ft. wide and 50 ft. long, with a mound of
soil at the west end.  Interviewees indicated  that between
30 and 150 drums of waste were buried there in 1971. The
drums were  dumped from  the back of a truck,  lay as
they fell, and were in marginal  condition at the time of
burial. Although eyewitnesses were unanimous  in their be-
lief that the trench contained NEPACCO waste, specific
information as to drum contents was completely lacking.
  In the winter of  1980, the Agency decided to mount a
field sampling investigation with the objective of docu-
menting site contents and any lateral migration of these
contents. Although there was no direct evidence to indi-
cate the presence of TCDD, the knowledge that the drums
came from NEPACCO and the  company's  history of
waste mismanagement, dictated that the study plan be de-
signed to provide the highest levels of personnel protec-
tion. Consequently, a draft plan was developed and wide-
ly circulated for comment within  the  Agency as well as
without. Many excellent  suggestions were received and
incorporated into the final plan.
  Preparatory activities undertaken in support of the in-
vestigation included removal of  site vegetation, the erec-
tion of an eight-foot security fence,  personnel baseline
physical examinations, a town meeting in Verona to ex-
plain EPA activities and, finally, a field trip to the site for
the news media immediately prior to imposing access re-
strictions.
  Sampling, which included collection of perimeter soil
samples as well as exposure and sampling of drums, took
place in April of 1980 with the highly capable assistance
and support of the EPA National Emergency Response
Team.  Initial on-site activities included setting up a de-
contamination station for men and equipment, putting up
a wind speed and direction indicator, and the surveying in
of perimeter sample locations.
  Protective clothing and safety gear, depending upon the
particular work task, ranged from disposable splash suits
with rubber boots, gloves and hard hats, to the encapsul-
ating "moon suits" for those individuals working in the
trench  and sampling drum contents. During the time that
the trench was open, the rural fire department was on
site, and aircraft were restricted from the area.
  Working from one side of the trench, a back hoe was
used to excavate down to a depth of 1.5 to 2.0 ft. Manual
shoveling  completed the removal of soil from around the
drums  in preparation for sampling.  Thirteen drums, some
of them crushed and empty, were exposed. Eight  drums
with visually varying contents were  selected for sampling.
Following completion of sampling, the trench was immed-
iately closed and samples packed for shipment.
  Analytical results received in June 1980 ranged from 1,4
ppb TCDD in some drums up to 319 ppm in others.  A
single  composite soil  sample taken in the trench from
around the exposed drums had a TCDD concentration of
72 ppb.
  Following confirmation of TCDD and based upon ex-
pert testimony that migration would occur, and that the
location was neither  suitable  for  hazardous  waste nor
could be made suitable, a temporary cap was put over the
trench  to  eliminate percolation of precipitation. Concur-
rently,  EPA initiated a weekly monitoring program of pri-
vate wells and surface waters in the area, and directed
Ecology and  Environment, Inc.  to prepare an engineer-
ing report evaluating appropriate remedial measures. The
Agency also  entered into discussions  with Syntex Agri-
business.
  Effective September 12,  1980, Syntex,  while denying
knowledge or responsibility for any acts of NEPACCO or
others  in placing wastes  in the trench, entered  into a con-
sent decree to undertake remedial measures at the site.
The company's  site restoration  plan,  comprehensive in
scope and sophisticated in design,  was approved  by the
Agency on February 25, 1981.  Following construction
and stationing of numerous on-site structures and ancillary
equipment, Syntex began removal  of trench contents  on
June 15, 1981. As of the writing of this paper, remedial
work is still underway.

Syntex Trenches

  Located on Syntex  property  west  of the  company's
plant in Verona are  a number of trenches,  now  inac-

-------
324    CASE HISTORIES
tive and  covered, which were ostensibly used  by Hoff-
man-Tate for the disposal of uncontaminated trash. Two
of these  trenches were reportedly open during the time
NEPACCO was active. Some interviewees  have alleged
that NEPACCO waste, both drummed and  in bulk, was
put in these trenches, apparently without the knowledge of
company management. Syntex and the Agency have open-
ed discussions on this potential site and exploratory sampl-
ing is anticipated in the near future.

Farm Site A

  A former NEPACCO employee rented a house on this
farm located a few miles south of Verona. Six 30-gallon
open top drums, originating from the Verona plant and
containing minor volumes of a dark residue,  had been left
in a barn on the farm by the employee. The drums came
to  the attention of the Agency in  late 1979  and were
sampled  in the winter of 1980. Analysis indicated 7 ppb
TCDD. Syntex  subsequently agreed to take the drums
along with limited quantities of hay, corn and potatoes
which were also stored in the barn. A barn dust sample was
collected following removal  of these materials  and final
disposition of this site is awaiting the analytical result.

Water and Wastewater Technical School

  The hexachlorophene operation in Verona also resulted
in  the production  of  a high-strength refractory waste-
water which was initially discharged  to evaporative la-
goons owned and operated by Hoffman-Taft. In Septem-
ber of 1971, Hoffman-Taff refused to  accept anymore of
this wastewater and NEPACCO subsequently entered in-
to a contractual arrangement with the  Water and Waste-
water School in Neosho, Missouri for  acceptance of this
wastewater and for the conduct of treatability  studies.
For convenience, a portion of this wastewater was put in
a small open top  steel tank on  the school grounds for
analysis and the treatability studies.
  As  a result of the publicity surrounding the Agency's
ongoing investigation  in the area, school officials became
suspicious of the asphaltic-like residue remaining in this
tank and collected a sample  in the fall of 1980. Analysis
by  a commercial laboratory indicated a TCDD  concen-
tration of approximately 2 ppm.
  In response to this  finding, EPA  personnel assisted in
a cleanup of this solid material on school grounds which
had accumulated as a  result of students operating a valve
on  the tank. Following confirmation  of TCDD by the
EPA Regional  Laboratory, which also reported  a  resid-
ual of 62 ppb in the spill area soil after cleanup, the tank
and 15 drums of residue mixed  with  attached soil were
removed to a remote bunker on school property for tem-
porary secure storage.
  In addition, Syntex  Agribusiness assisted by putting up
an  eight-foot security fence around the spill area  while
EPA personnel assisted the school in putting a temporary
cap over the area  to  eliminate the  possibility  of migra-
tion of contaminated soil particles. Although a final solu-
tion for  the spill area and the drummed  wastes  has not
been determined, the public health and environment is be-
ing protected in the meantime.

Neosho Digester

  Between October of 1971  and February 1972, 225,000
gallons of NEPACCO wastewater was trucked to Neo-
sho, about 30 miles west of Verona. The waste was placed
in an open top abandoned wastewater treatment plant di-
gester for storage, apparently with the expectation that the
Technical  School,  which operated the plant at the time,
would find some means for treating it.
  After NEPACCO went out of business, the waste re-
mained in the digester. Starting in 1977, after the city of
Neosho resumed  operational control  of  the  wastewater
treatment  plant, the digester was filled in  with rock, soil,
and various other fill materials. In March of 1981, an EPA
team, using a drilling rig, collected a core sample from the
bottom (20  foot depth) of the digester. Analysis of the
sample revealed 2,500 ppm  of 2, 4, 5-TCP and  60  ppb
TCDD. Following  this finding, down gradient  stream and
private well  water  samples were collected. The analytical
results are not available as of the writing of this paper and
disposition of the site is pending.

Farm Site B

  In addition to Baldwin Park, two farms are known to
the Agency as having received the NEPACCO  filter cake.
Although the source of the rumor is not known, there was
a belief among some  local farmers that  the  filter cake,
presumably  because  of  residual hexachlorophene,  was
good for controlling hoof rot in cattle. In an imaginative
approach  to waste disposal, NEPACCO permitted fann-
ers to take this material, which was subsequently spread
in feedlots and gate areas in such a manner that the cat-
tle  would  walk through it. One feedlot sampled in Aug-
ust of 1980 was found to have 0.4 ppb. In addition, 8 ppb
was found in a sample of the filter cake itself, which be-
cause of a leaking drum, was dumped on  the ground and
never spread or mixed. Although the landowner has been
cautioned to stay  away from the two areas, final dispo-
sition of the site is pending.

DISCUSSION

  At this  point in time, all NEPACCO sites are suspect.
The objective of the investigation in southwest Missouri is
to  determine where all NEPACCO wastes, of whatever
description, were disposed of, and to take appropriate re-
medial measures at those sites judged to pose some hazard
to public  health or environment. This objective is being
accomplished  by  a number of parallel paths which in-
clude interviews  of former NEPACCO  employees and
area residents, sampling and technical evaluation of avail-
able documents and information, with the aim of recon-
structing  the  NEPACCO  process  and  developing  esti-
mates of wastes quantities.
   Approximately 35,400 gal. of NEPACCO still bottoms
have been accounted for to date. This figure does not in-

-------
                                                                                    CASE HISTORIES     325
dude the bottoms recovered from the Denney site which
have not been entirely quantified as of the writing of this
paper. Even less quantification is  possible with  other
NEPACCO waste streams, and there is no assurance that
other sites will not surface. Indeed, at this time, the Agen-
cy is aware of rumors and hints of other sites too vague to
be addressed in this paper.
  Some ten years after the event, this investigation  has
proven to be  extremely  resource intensive. Time  has
dulled the memory, some individuals have moved away,
others have died, and still others have expressed concern
about publicity and are reluctant to talk with EPA inves-
tigators. The investigation has been further hampered by
unavoidable delays  in  obtaining NEPACCO records,  by
limited access  to low analytical detection capabilities for
TCDD and finally, by the absence of guidance criteria in
determining environmentally acceptable levels for TCDD.
  The activities of NEPACCO in the handling and dis-
posal of their various waste streams are classic examples
of hazardous waste mismanagement. These practices have
adversely affected the health and welfare of some individ-
uals and have produced a sense of fear and uncertainty in
the minds of others.
  In bright contrast, the residents of southwest Missouri,
and the Agency, are fortunate in having the problem solv-
ing capabilities of Syntex Agribusiness  brought  to bear
upon the legacy left by NEPACCO. The commitment of
Syntex to the Denney Farm Site has resulted in a model
project which is testimony to the company's positive en-
vironmental philosophy.

-------
           THE DENNEY FARM SITE REMEDIAL PROJECT:
                 A MODEL FOR THE SAFE EXCAVATION,
               STORAGE, AND ELIMINATION OF DIOXIN
                                          RAY FORRESTER
                                          Syntax Agribusiness
                                          Springfield, Missouri
INTRODUCTION

  Syntex' project to clean up and permanently dispose of
hazardous wastes, abandoned some 10 years ago by an-
other company, may be considered a model of industrial
governmental cooperation. Its success is due to the efforts
of Syntex' sophisticated multi-disciplinary team and to
the commitment, cooperation, and sense of mutual respect
that developed among Syntex Agribusiness, the United
States Environmental Protection Agency (USEPA) and the
Missouri Department of Natural Resources (MDNR).
  This  joint effort  was particularly important  because
Syntex  did not  generate  the problem but inherited it.
Syntex  neither created nor disposed of the wastes in the
disposal trench,  and the  disposal site is not  on Syntex
property. Syntex found itself involved in this matter only
because it purchased  a  chemical  plant at  Verona,
Missouri, from another company which had earlier leased
part of its facility to the  North Eastern Pharmaceutical
& Chemical Co., Inc.  ("NEPACCO") for the manufac-
ture  of  hexachlorophene.   In  the  summer  of  1971
NEPACCO,  in  violation  of its lease and unknown to
Syntex, arranged with the owner of the Denney farm, lo-
cated in Southwest Missouri, to have barrels of its waste
transported to the farm and disposed of in a trench lo-
cated on the property. It is  this trench and its contents
that Syntex, the  USEPA and the MDNR have been con-
cerned with since.
  In  1979-80 the disposal trench was found by an EPA
investigation team and determined to  contain highly toxic
wastes, including 2, 3, 7, 8-tetrachlorodibenzo-p-dioxin,
commonly known as dioxin.  Syntex, the  EPA  and the
MDNR joined together to develop a plan designed to clean
up safely the toxic wastes found in the disposal trench.
  Unfortunately, the task  was seriously complicated by
the absence of crucial facts such as the size of the trench,
the number of barrels in it and the identity of the wastes in
those barrels. Testing was undertaken to ascertain this in-
formation using ground penetrating  radar, metal detec-
tors,  electromagnetic conductivity and resistivity.  How-
ever, the results, while providing an estimate, were not as
helpful as one might have wished.  For  example,  after
various tests over a number  of months, estimates of the
size of the trench ranged from a rectangular depression
53 feet  long, 10 feet wide  and 10 feet deep, to a left shoe
print shaped trench some 65 feet long, 20 feet wide and
10 to  12 feet deep. In fact, the trench was found to be
approximately 60 feet long, 8  or so feet wide and only 6
feet deep. Estimates of the number of barrels in the trench
ranged from 30 to 150; approximately 90 were found. Per-
haps most difficult was that while there was some informa-
tion on  the range of wastes  generated by NEPACCO,
which included innocuous materials as well as flammable
solvents and dioxin, no one knew their concentrations or
quantities.

TEAM APPROACH

  To develop a plan to solve this problem safely, Syntex
realized early in the project that the expertise of a num-
ber of disciplines was necessary. Needed was expert tox-
icological advice to assess what risks existed at each stage
of the operation and a determination of when protective
clothing  was necessary and what types were advisable to
wear.  Needed also was a determination how best to as-
sure that no materials would escape into the environment
if some untoward event occurred during excavation. State
of the art analytical capability was  crucial for quick,
accurate assessments of the materials found in the trench.
Particularly important was a construction crew that could
perform under conditions made difficult by the presence of
extensive safety equipment and could  operate their ma-
chinery with such delicacy that drums which  had thin
sides due to corrosion would not rupture.
  Finally needed were sophisticated legal and public rela-
tions input to  help  develop the plan  and maintain the
firm's good relations with the  EPA, MDNR and the local
community.
  Therefore, Syntex gathered together a multi-disciplinary
team  composed of  an engineer,  equipment operators,
skilled laborers, a  lexicologist, specialists in environmen-
tal health and safety, analytical chemists, lawyers, public
relations experts and executives in order to devise a plan
that could safely  tackle  the  Denney farm site  disposal
trench.

PLAN DEVELOPMENT

  After months of hard work, Ithis team developed a plan
whose goal was to minimize and protect against possible
dangers to on-site workers, the nearby community and the
environment while simultaneously maintaining the mutual
                                                   326

-------
                                                                                      CASE HISTORIES    327
objectives of the parties involved: to remove and dispose
of hazardous wastes located at the Denney farm site.
  The Denney farm site Final Plan did exactly that. It
established a series of procedures to remove all materials:
liquids, flowable sludges, chemical solids, containers, vis-
ibly contaminated soils and  other soil from the Denney
farm site disposal trench to provide immediate protective
containment of the hazardous wastes excavated and to de-
toxify those wastes  with higher concentrations of  dioxin
and to provide an environmentally safe way simultan-
eously to contain and detoxify any low level contaminated
soil from the trench.

Safety Program

  However, prior to beginning actual work the Final Plan
required every individual who would work at the Denney
farm site trench to  undergo special medical examinations
and to complete a specially designed training program. A
detailed health and safety review was conducted by occu-
pational physicians, toxicologists and hygienists; it focused
on determining if  any worker  had a   particular  sensi-
tivity or susceptibility  to dioxin or other toxic chemicals
believed to be in  the  disposal  trench. Particular atten-
tion was paid to routes of exposure via the respiratory
tract, skin and mucous membranes. Any worker with such
a sensitivity or susceptibility was not permitted to work on
the project. All workers passing the medical exams then
underwent extensive training at the Verona plant prior to
actually beginning work at the disposal trench. During this
training they were instructed on the tasks they were to per-
form at the trench and  practiced safety and emergency
procedures.
  The prospective workers also learned how to use the
various types of required protective clothing. The  most
stringent protective clothing was Category  I equipment
consisting of complete  respiratory, skin and mucous mem-
brane protection.
  Category I protection was used by  employees work-
ing in the disposal trench while the waste from the trench
was being  removed. These  workers used a  supplied air
continuous flow system for maximum working efficiency
consistent with the primary objective of full  assurance of
respiratory safety. They were also protected by a full body
guardian chemical suit (akin to a "moonsuit"), a white cap
safety helmet completely covering the head and neck, and
neoprene  safety boots. This equipment  afforded  com-
plete body protection and allowed the mobility necessary
to remove safely the drums and contaminated soil.
  Category II equipment,  used only outside of the dis-
posal trench, had less stringent protective requirements. It
consisted of a chemical protection suit, safety hat, gloves,
boots and an air purifying respirator.
  Category III equipment was required for the relatively
minor operations. Category HI standard safety equipment
and clothing included a safety hat, disposable Tyvek suit,
coveralls and safety boots.
  After those workers  who would be participating in the
excavation and monitoring activities at the Denney farm
site underwent medical examinations, were specially train-
ed for the safe  handling  of dioxin, learned how to use
their safety equipment, and were schooled in emergency
procedures, construction at the disposal trench was ready
to begin.


SITE CLEANUP

  The first order of business was to expand  the 8-foot-
tall chain  link  fence  already  surrounding  the disposal
trench to accommodate the  additional structures neces-
sary to implement the  Final Plan.  This fence served to
prevent any individuals or animals from accidentally wan-
dering onto the  site. A guard was  also stationed on the
site to dissuade  the curious and add an additional  meas-
ure of safety and security to the area.
  Next,  a Temporary  Work Site  Protection Structure
(the TWSPS)  100-feet-long and 48-feet-wide was erected
over the entire disposal  trench. The  purpose  of this  struc-
ture was to provide  a cover that would prevent rainfall
from  entering the trench  site, reduce the effect of wind
on the handling  of contaminated soil and provide a safe
storage area for containment drums.
  Once the TWSPS was erected over the disposal trench a
Temporary  On-Site Drum Storage Area (TOSDSA) was
constructed to provide a diked, weather protected and se-
cure storage location for all drummed materials and over-
packs. The TOSDSA was constructed with reinforced con-
crete with  an  emphasis on  structural  integrity and leak
proofing.
  Two Microbiological  Degradation Basins (MDBs) were
then constructed in order  to safeguard and treat contam-
inated soil excavated from the Disposal Trench.
  Finally, before excavation  began,  ancillary but impor-
tant additional equipment was placed at the Denney farm
site: an air compressor  to supply air to the workers who
would  actually be in the trench, electrical wires, pipe lines
for water, fuel, nitrogen, a forklift, a generator and a back
hoe.
  An anemometer was installed to determine wind speed.
If the  wind velocity exceeded a certain level, the project
would  be shut down to avoid the possibility of escaping
particles. Further, air sampling devices were  placed about
the site and their filters were assayed regularly  for dioxin;
none was ever found in any of these samples.

Excavation

  The  excavation began by removing the cap and the first
layer  of earth.  Then,  workers carefully began  digging
around the barrels in the trench with shovels. Both opened
and unopened barrels were uncovered. Once a barrel was
found, its contents and any  material that had leaked out
were extracted by a  special vacuum  system  and  trans-
ferred  into clean barrels. The contaminated  barrels, once
emptied, were then carefully  lifted from  the disposal
trench and placed in overpacks. Once the new barrels and
overpacks reached their capacity, they were sealed, labeled
and taken to the  TOSDSA.
  After the barrels were removed, all the visibly contam-
inated soil was removed and placed in 55 gallon drums and

-------
328    CASE HISTORIES
moved to the TOSDSA.  The remaining  soil was  then
placed in the MDBs. This soil will be tested and, if found
contaminated, subjected to treatment.
  Using this method of excavation, removal, and storage
the disposal trench has been  emptied—90 barrels  were
found in it although some of the barrels were broken or
disintegrated so  an exact  count is impossible.  The dis-
posal trench has been  sampled and the samples are being
assayed to determine if the soil is still contaminated.

DIOXIN DEGRADATION

  It is expected that the most contaminated materials will
be amenable to photolysis, which involves breaking down
the nuclear structure of dioxin using light energy, usually
in a specific wavelength range.  Photolysis studies  have
shown that  dioxins may be photolytically degraded by
natural sunlight.  A photolysis treatment plant provides the
photochemical reaction necessary to  destroy the dioxin
compound by using an ultraviolet source  of light in an en-
closed and controlled setting.
  A photolysis treatment system was  developed to  treat
approximately 4300 gallons of an oily dioxin-contaminated
waste NEPACCO had abandoned in a tank at the Verona
plant. The first step was to make certain nothing leaked
from the storage tank. Six years of work preceded a de-
cision to  go forward  with development and use of the
photolysis process primarily because of  its potential for
success and the degree of safety it offered. It is a low  pres-
sure,  low temperature, and fully closed system providing
almost no opportunity of human exposure to the material
to be detoxified as well as virtually no chance of toxic re-
lease to the surrounding environment.
  The system was successful. During the summer of 1980
Syntex extracted  virtually all of the NEPACCO waste
from the tank and destroyed 99% of the dioxin. This sys-
tem will be used to detoxify a portion of the materials at
the Denney farm site.
  Some materials found at the Denney farm site can also
be disposed of through landfills and other EPA approved
methods.  Other material  cannot be so  easily discarded.
Syntex has explored a variety of different technologies to
dispose of these substances such as microbiological degra-
dation which is the molecular degradation of an organic
substance resulting from  the  complex  actions of living
organisms. To  assist this  biological  process, various or-
ganic nutrients  are to be added to treat the contaminated
soil in the MDBs.  Syntex and the EPA will be closely mon-
itoring the dioxin levels  in the  MDBs  to determine the
success of microbiological degradation.

CONCLUSION

  While the project is  not fully complete, the hazardous
wastes from  the Denney farm site  trench are excavated
and  secured,  with progress being made toward their ulti-
mate disposal. The success of this project is the product of
a team composed  of a solid core of individuals with high-
ly developed  expertise. It  is also the result of the  close
cooperation and mutual respect among Syntex, the EPA,
the MDNR and the  local communities, without which this
project would have not been possible.

-------
            A HAZARDOUS WASTE DISPOSAL PROBLEM
                                                  vs
                       A SYSTEMATIC APPROACH FOR
                       IMPOSING ORDER ONTO CHAOS
                                         THOMAS O. DAHL
                                U.S. Environmental Protection Agency
                              National Enforcement Investigations Center
                                           Denver, Colorado
INTRODUCTION

  Occidental Chemical Company, a subsidiary of Hooker
Chemical Corporation  and  its  parent, Occidental Pe-
troleum Corporation, owns and operates an agricultural
fertilizers and pesticides plant in the San Joaquin Valley at
Lathrop, California.  This  plant came  into  national
prominence in 1977 following discovery that a substantial
proportion  of the pesticide workers were sterile. Pre-
sumably, the cause of this sterility was from exposure to
a nematocide, Dibromochloropropane (DBCP)  which
was both  formulated (1957 to July 1977) and manu-
factured (1963 to 1971 and 1974 to 1976) at the plant. Sub-
sequent to this discovery, the State of California banned
the sale and use of DBCP on August  12,  1977. This
initial action was expanded into a nationwide mandatory
suspension of the production and use of DBCP except in
certain limited areas.
  The  combination  of the sterility issue  and certain
Company  documents indicating potential  adverse envir-
onmental effects caused the U.S. Environmental Pro-
tection Agency (EPA) and State of California to investi-
gate the Company's operations. On  December 18,  1979,
the U.S. and State of California filed suit in U.S. District
Court against Occidental. On February 6, 1981, following
over a  year of  technical studies and protracted legal
maneuvering, a consent decree was lodged with the Court
which established the framework for additional necessary
studies and implementation of remedies.
  The  activities  which led ultimately to  the successful
settlement of the case  were  characterized by a series of
investigations  which  first sought to understand the po-
tential  problems, then  defined the actual problems, and
finally designed necessary remedies. The first of these was
a multimedia environmental audit  of the  Company's
operations.


AUDIT OF OPERATIONS

  From September 18 to 28, 1979, investigators from the
U.S.  Government and  State  of  California reviewed
Company records and conducted interviews of employees
at the  Lathrop  facility.  The subjects of  these inquiries
were: past and  present  waste-generating processes and
the disposal practices  employed; in  additional environ-
mental  data  were collected (air,  water,  soil,  ground-
water). Attempts were made to  gain information from
the beginning of operations (1953) to present. Informa-
tion on this  time span was imperative because past dis-
posal  practices could still be causing adverse effects in
the soil and groundwater systems.
  The audit of Company operations in Lathrop, Cali-
fornia determined that past production and disposal prac-
tices relating to liquid and solid wastes were inadequate,
resulting in contamination of the groundwater and soil
in the vicinity of the plant.  Major findings of the audit
included:
•Wet  phosphoric acid operations, commencing in 1953,
 resulted in  the generation of considerable  amounts of
 gypsum (CaSo6) and losses of fluoride. The gypsum was
 stored onsite during drying on permeable soil presumably
 resulting  in percolation of sulfate ions  to the local
 groundwater which was known to be between 2 and 7.6 m
 of ground surface.
•Ammonia,  ammonium  sulfate  and ammonium  phos-
 phate/sulfate  operations resulted  in discharge  of am-
 monium ions to unlined ponds,  presumably resulting in
 percolation to groundwater.
•Sulfuric acid plant operations resulted in onsite burial of
 vanadium pentoxide catalyst causing, at a minimum,
 soil contamination.
•Pesticide  formulating and manufacturing operations,
 some commencing as early as 1957, were characterized
 by use of over 100 active ingredients. Liquid waste dis-
 posal practices until at  least 1976 included  use  of un-
 lined ditches and ponds in permeable soil areas. Further-
 more, until at least 1970  some pesticide waste  solids
 and concentrated liquids were disposed of onsite in shal-
 low trenches.  Presumably these  trenches were either ex-
 posed directly to the groundwater table or very close to
 it. A review of  Company records for selected years in-
 dicated small  percentage losses  of pesticides. However,
 although  these were small in terms of percentages, they
 represented large quantities when considering millions of
 kilograms of pesticides  were formulated and/or manu-
 factured.
•A review of past environmental data available through
 Company records and State of California files provided
 strong indications of environmental contamination at-
 tributable to the facility. Sulfate concentrations in one of
 the Company wells increased from 28 mg/Vin  1960 to
 over 2,000 mg/
-------
 330     CASE HISTORIES
 had  been  detected near  the  facility  in  soil, shallow
 groundwater and process, irrigation and potable wells.
 All these compounds were known to have been handled
 at the facility. Of these, 12 are known animal carcino-
 gens.
•An aerial overflight of the facility conducted as part of
 the audit, using false color infrared imagery, indicated
 dark and discolored areas  to the west of previous dis-
 posal areas. Vegetation appeared to be stressed; this pat-
 tern  could have been caused by leachate migrating from
 the disposal site.


PLAN OF STUDY

   Based on  the audit  findings and State  of California
records indicating probably significant,  albeit not fully
defined, environmental  contamination,  a plan of study
was formulated to address the problem.  This plan of
study, developed jointly by the Company,  the State of
California, and the USEPA National Enforcement In-
vestigations Center, presented a comprehensive approach
for systematically determining whether certain inorganic,
organic and  radiological substances were present  in the
groundwater and soils in the vicinity of the  Company
facility and whether or not these  substances  were attri-
butable to the Company's operations. The study  would
determine whether, and  if so, which mitigation and re-
medial  measures were necessary to prevent an imminent
and substantial endangerment to public health and envir-
onment. Programs  for groundwater analysis,  soil  analy-
sis, private and public well monitoring and groundwater
modeling were designed. The plan of study was divided
into three phases:
   Phase I—That  phase of  the  groundwater' and  soil
analysis program necessary to determine whether any or-
ganic, inorganic or  radiological constituents were present
in groundwater or soils in the vicinity of the Company's
plant. This was to include,  but not be limited to, the
following:
•Define the  vertical and horizontal zone  of the  afore-
 mentioned constituents within the groundwater and soils
•Establish  the direction of movement of the  subsurface
 plumes and their rates of movement
•Develop proposed mitigation  measures if any are de-
 termined to be necessary to protect public health  and
 the environment against imminent and substantial en-
 dangerment from  contamination  of groundwaters  and
 soil resulting from the Company's operations
   Phase II—That phase of the groundwater and  soil
analysis program necessary to:
•Evaluate  the  overall effectiveness of  any  mitigation
 measures, if any were determined to be necessary
•Define any additional remedial measures needed to be in
 compliance with waste discharge requirements and pro-
 tection of public health and the  environment  from an
 imminent and substantial endangerment
  Phase III—That  phase of the groundwater  and  soil
analysis program conducted after the implementation of
mitigation measures, if any were determined to be neces-
sary,  that would evaluate  ongoing compliance  with the
waste discharge requirements and protection of public
health and the environment.

Groundwater Analysis Program
  A total of 14 monitoring sites were selected surround-
ing the  facility (Figure  1).  Eight of these locations were
selected to reflect presumed upgradient  and downgradi-
ent groundwater conditions, the areal confines of the fa-
cility  and locations of past disposal sites. Six additional
sites were selected to reflect conditions radiating out sev-
eral hundred meters from the plant boundary locations.
Sampling at these  sites  would enable determination of
whether and to what degree contaminants would attenu-
ate in the directions of assumed groundwater movement.
Three permanent wells were to be constructed at each
site. The intent was to locate wells at depths of greatest
contamination, background conditions below the leachate
plume and approximately midway in the plume. It was
recognized that additional wells may become necessary if
the geohydrology of  the  area created  several  plumes
rather than one. In fact, there are now 55 monitoring wells
installed at the site (Figure 2).
  One of the principal dilemmas faced by the formulators
of  the plan of  study was what parameter(s) would be
used to  define the depths of the desired  wells. As noted
previously,  numerous  compounds, both inorganic  and
organic, had been handled at the facility, presumably with
varying  attenuation and migration rates. Of prime'con-
cern  were many of the organics with  known  adverse
health effects. After considerable  discussion, it was de-
termined that existing state of the art information could
not afford an exact determination of the best paramet-
er (s).  Intentions were  to initially  examine  groundwater
conditions down to approximately 46 m or background
conditions,  whichever  was greatest. Because this would
mean numerous samples, the issues of expense and time
to analyze indicator parameter(s)  had to be addressed.
Sulfate  was  finally chosen as the  primary indicator be-
cause (1) considerable quantities had  been introduced to
the groundwater by the facility from  the gypsum ponds,
(2) sulfate was believed to be as mobile as any of the or-
ganics and  (3) it could  be monitored rapidly and inex-
pensively. To augment this indicator, additional indicator
analyses were  performed  onsite  for  pH, conductivity,
Redox potential and ammonia nitrogen.
  In order to delineate  areal stratigraphy, particularly at
the well sites, samples were to be collected at representa-
tive depths during drilling. Once all stratigraphy data were
collected from the 14 sites,  three representative bore holes
to 61 m were to be drilled in the study area to collect
undisturbed samples for representative permeability and
porosity measurements.
  Drilling techniques  employed  to accomplish desired
goals included both a dual tube reverse air rotary rig and a
conventional rotary rig. With the dual tube rig, air flows
down the outer annulus of the drill rod  and carries cut-
tings  and groundwater up  the inner tube to ground sur-
face and exits via a cyclone. Since air flows are on the
order of 30  m/sec, an observer at ground surface sees es-

-------
                                                                                       CASE HISTORIES     333
                                                     Figure 1.
                                  Study Area Showing Permanent Wells Sampled During
                                          Phase I and Interim Phase II Study.
                                                (Scale: 1" = 1,100')
sentially an instantaneous representation of cuttings and
groundwater. This procedure enabled a continuous logging
of stratigraphy by an experienced geologist and also af-
forded indicator groundwater sampling at approximately
3 m intervals. Once a depth of desired sampling was
reached, advancement of the drill rod was halted but the
air  was  left on. This  afforded a  representative  water
sample since the rapid  air flow  purged the drill rod of
any contents.  At certain times, when in clay formations
and some sandy material,  plugging occurred and it was
necessary to add a "water mist" to the airstream to lubri-
cate the drill bit.  Representative  groundwater samples
were then obtained  by  stopping the mist approximately
1 m before the desired depth,  advancing the drill rod and
allowing ample time for air purging of any mist water.
 Once desired  groundwater  and  stratigraphy samples
were collected at a  site, conductivity, sulfate,  ammonia
nitrogen and stratigraphy data were plotted against depth.
Three depths were then selected for permanent wells.  Al-
though  the  depths  were ostensibly to represent the pre-
viously  mentioned  criteria of highest sulfate concentra-
tion,  background sulfate and mid-concentration, reality
often dictated  a more detailed  determination.  For ex-
ample, in some instances, there were no sharply defined
curves; consequently, a combination of the data plots, the
existing stratigraphy  and considerations of adjacent site
data had to be used to select depths of permanent wells.
  Once depths  were  selected, 30 cm  bore  holes  were
drilled using a  conventional rotary rig for installation of
permanent wells of 15 cm diameter steel casing.  Because
a major concern was precluding  the introduction of any
contaminants into  the wells,  the rotary drill  contractor
was instructed to only use water as a drilling fluid, if pos-
sible, rather than drilling mud. This proved successful at
all 42 wells installed at the 14 sites (three per site). At the
bottom  of each casing was a concrete plug followed by a
1.5 m stainless  steel screen to afford adequate withdrawal
from the  water-bearing formation. All wells were gravel

-------
334    CASE HISTORIES
                                                    Figure 2.
                                     Phase II. Permanent Well Sampling Locations
                                               (Scale: 1" = 1,100')
 packed to above the screen and then cement grouted to
 the surface to isolate the screened section.  To assure no
 contaminants would be introduced at the surface, the cas-
 ing extended to  0.6 m above  ground surface and was
 fitted with a metal cap which could be locked. Wells were
 developed/cleaned by pumping at least 7,600  with an
 airlift and/or a submersible pump. Purged water was col-
 lected  in tanks  and returned to the Company's  onsite
 hazardous waste tanks for subsequent disposal.
   Groundwater characterization sampling was conducted
 with a  submersible pump  and an air-inflatable packer.
 The pump/packer assembly was lowered to  a point where
 the packer was just above the top of the well screen. The
 packer was then inflated via an air line to the surface to
 isolate the well casing from the screen down.  The pump
 was activated and a volume of water equal to at least 5
 volumes of casing below this screen was purged to the sur-
 face via a Teflon* tube and discharged to hazardous waste
 tanks. Representative samples could then be collected for
desired analyses (Table I). The selection of these parame-
ters was based on compounds handled at the facility in
greatest quantities, based on audit findings  and com-
pounds previously found in groundwater and soil mirnpta
in the area. The analyses selected included 29 pesticide*,
12 metals and other inorganics and a radiological assay.
  Once sampling was completed at a given well, the pump/
packer/Teflon-tube assembly was moved to the next well
and used again. It was assumed that the purging volumei
would be  sufficient to carry away any contaminants left
on the Teflon from  the  previous  well. To verify thto,
quality assurance sampling was periodically conducted by
purging the Teflon with deionized distilled water. Samphi
were collected of the deionized distilled water as  wcU it
the water discharged from the Teflon at the end of purginf.
 •Registered Trade name.

-------
                                                                                         CASE HISTORIES    335
                       Table I.
    Parameters Chosen for Groundwater Characterization
                       Organics
      DBCP
      ETDB (EDB)
      CE-BHC
      LindaneCY-BHC)
      /3-BHC
      A-BHC
      Aldrin
      Toxaphene
      Thiodan I (Endosulfan)
      Chlordane
      DDT
      Perthane
      Methoxychlor
      Ethyl Parathion
      Methyl Parathion
Malathion
Diazinon
Disyston (Disulfoton)
DBF
Dimethoate
Trithion
Dibrom
Methyl Trithion
Dioxathion
Fenthion
Phosdrin (Mevinphos)
Ethion
Sevin
DNBP
                  Inorganics and Metals
       Nitrate                   Fluoride
       Sulfate                   Copper
       pH                      Zinc
       Conductivity and Temperature Chromium
       Vanadium                 Arsenic
       Nickel
                         Other
Oxidation Reduction (Redox) Potential
Radiological Assay (gross alpha scan, radium scan, uranium scan,
 thorium scan and gross beta scan)

Soil Analysis Program
  The audit of the Company's past practices revealed the
existence  of  a number of past disposal areas, including
unlined ponds, conveyance ditches and  burial  sites. Al-
though the locations of some  of the sites were obvious
since they were above ground  and clearly visible, others
such as alleged pesticide burial trenches in the so-called
"Boneyard" area were  not.  No Company person inter-
viewed could identify the number or exact location of the
burial areas.  In approximately  1970, use of the Boneyard
area  had  been discontinued  and subsequent filling and
grading had  rendered visible  identification  impossible.
The only document  available  which appeared to  show
the locations of some of the trenches  was  a blueprint
submitted to the State of California in 1970. Its accuracy
was presumed by the Company to be questionable.
  In order to determine whether previously disposed ma-
terials needed to be escavated and transported to State of
California approved hazardous waste  disposal  sites, a
sampling program had to be designed for the known areas
such  as past  disposal ponds and  ditches. This included
core sampling to approximately 3  to  5  m below grade.
Sampling  locations were  selected representing the geo-
graphical  bounds  of the disposal  sites, the  discharge
points and the lowest points (in ponds) (Figures 3 and 4).
The procedure used was to collect discrete samples at the
surface, 0.6 m, 1.5 m and 3 m. The surface sample  (0-10
cm) was collected  by shovel and  placed into a solvent-
rinsed glass or other appropriately prepared jars for sub-
sequent analysis (Table II).  The remaining samples were
collected using  a hollow stem auger drill  rig and split-
spoon samplers. The sampling procedure included drilling
down  to the  desired depth  and then  advancing  a split-
spoon sampler with the rig to collect the desired soil sample.
Samplers  were  opened  and contents transferred onto a
sheet of Teflon to preclude cross contamination. Samples,
once split by  quartering or  mixing and aliquoting, were
tranferred to  appropriately prepared glass jars for trans-
port to labs  for analysis. The Teflon sheet was  rinsed
with deionized and  distilled water and a 1:1  solution of
hexane and acetone between samples.  The  split-spoon
sampler was cleaned in a similar manner between samples
to preclude cross contamination.

                        Table II.
      Parameters Chosen for Soil Sample Characterization
                        Organics
                                      EDB
                                      DBCP
                                      oe-BHC
                                      Lindane(T-BHC)
                                      /3-BHC
                                      A-BHC
                                      Aldrin
                                      Toxaphene
                                      Thiodan I (Endosulfan)
                                      Chlordane
                                      DDT
                                      Perthane
                                      Methoxychlor
                                      Ethyl Parathion
                                      Methyl Parathion
                                     Vanadium
                                     Nickel
                                     Copper
                                                      Metals
                                                      Other
                                  Malathion
                                  Diazinon
                                  Disyston (Disulfoton)
                                  DEF
                                  Dimethoate
                                  Trithion
                                  Dibrom
                                  Methyl Trithion
                                  Dioxathion
                                  Fenthion
                                  Phosdrin (Mevinphos)
                                  Ethion
                                  Sevin
                                  DNBP
                                  Zinc
                                  Chromium
                                  Arsenic
                             Radiological Assay (gross alpha scan, radium scan, uranium scan,
                              thorium scan and gross beta scan)
                             PH
                               Relocating past disposal trenches/pits presented a most
                            interesting  technical problem. As  noted previously,  the
                            only written evidence of possible locations was a 1970 site
                            map. A reconnaissance of the area  indicated that some of
                            the map reference points could still be located. After con-
                            siderable thought and discussion the following approach
                            was adopted:
                            •Assume the site map was at least generally accurate
                            •Locate and physically  stake out the disposal sites from
                              site map locations
                            •Attempt to locate disposal sites  by cutting exploratory
                              trenches at right angles to the assumed directional lie of
                              the trenches.  This included using  a backhoe and cutting

-------
336    CASE HISTORIES
                                                     Figures.
                               Study Area Showing Soil Coring Locations Sampled During the
                                    Phase I Study Excluding the Excavation Program.
                                                 (Scale: 1" = 460')
  to the water table which was generally 1.8 to 2.4 m be-
  low the ground surface (Figure 5)
 •Once disposal trenches are intersected, rotate the explora-
  tory trenches 90° and commence in directions of the dis-
  posal trench. This procedure was employed to determine
  the length of the  trench as well as to assess,  by visual
  observation  of trench contents, the need for immediate
  total trench excavation versus sampling followed by pos-
  sible excavation
 •Once initial  excavation and sampling activities are com-
  pleted, collect soil core samples with depth adjacent to the
  trenches for parameters previously referenced (Table II)
 to ascertain whether migration of contaminants requir-
 ing further excavation has occurred.
  The approach  detailed above led to the discovery of
four major disposal trenches and a number of pockets of
disposal pits. The discovery of one trench and how it wi*
subsequently addressed is particularly noteworthy. The ex-
ploratory trench intersected the disposal trench and effortl
began to continue exploratory trenches in the direction of
the disposal  trench (north-south). Visual  observation of
the contents (e.g., remnants of drums, bags and acid tank
liners), however, led the investigators  to believe they had
located the major chemical trench referenced in internal

-------
                                                                                    CASE HISTORIES     337
Company records and interviews during the audit. This
led to a decision to mine out the entire trench with the
backhoe rather than proceed with exploratory trenches.
  The approach taken was to first remove  the approxi-
mately 0.6 to 0.9 m of overburden to one side of the trench
and then excavate from that point down to the water table.
Any metal drum remnants, pesticide produce containers
or other  probable  hazardous waste containers, were re-
moved by hand to hazardous waste  bins  or overpack
drums  and hauled away to California-approved Class I
hazardous waste disposal sites. Soil and other trench con-
tents were excavated by backhoe and placed on sheets of
black plastic  for subsequent  sampling and  decisions re-
garding disposal. The location of where the material came
from in  the  original  trench  was catalogued for future
reference.
  While proceeding in this manner, thousands of '/2 and
1 pint glass pesticide bottles,  some broken and empty of
contents, some partly full, or full were discovered. The
labels of these bottles indicated  such  products  as  40%
DDT,  50%  Chlordane, 2,4-D,  2,4,5-T  and  Dieldrin.
Ominously apparent was the fact that many containers
were still below the water table.
  Because of the inherently hazardous  nature of the dis-
posed material,  a decision  was made to excavate below
the water table  and remove the material  with sealed
trucks to Class I disposal sites. Considering the character-
istics of the disposal area and discussions concerning past
recollections of the trench operation, it was  decided that
excavation 0.9 m below the water table should remove any
containers etc.  originally  deposited. The approach for
accomplishing this included use of a "mud  wave."  This
                     •8E   V    f*
                                                   Figure 4.
                                   Study Area Showing Soil Coring Locations During
                                              Excavation Program.
                                               (Scale: I" = 135')

-------
338    CASE HISTORIES
                                                                                                  90    worm
                                                                                               limi «• MO MITI MI1»UL

                                                                                               Mill «*H«I*V
                                                                                                    OIMOMl 1MMH 0" W1
                                                                                               »0*IN« LOC1TIOMI


                                                                                               1IMIL SUHIMI UMIMTMtIM
                                                                                               VAMDIIM PUTOllOC miltl «
                                                                                               H*VI HIM Cifaxe vf
                                                                                               ®AmN»iu*Tf locnic* or w*ti IMU
                                                                                               WT ni •HKN en NOT fwtwifll <
                                                     Figures.
                                                Trench Location Plan
included starting from one end of the trench with a bull-
dozer, cutting  down 0.9 m and  moving  the excavated
material a short distance forward in a wave. Clean fill
was immediately brought in to replace the  excavated ma-
terial and the process was repeated. Water, pushed for-
ward by the wave, was pumped to a truck from a crude
sump carved out of the downstream end of the trench and
taken to the Company's onsite hazardous  waste disposal
tanks. This procedure was  continued until  the trench was
entirely excavated and filled with  clean fill. Soil exposed
at the bottom  of  the cut  by the bulldozer blade while
advancing the wave had a clean appearance, supporting
the belief that the procedure selected would escavate all
previously deposited containers and solid materials. Eval-
uation of potential  migration of contaminants through
the soil  was accomplished by  the previously  described
core sampling program and  the groundwater monitoring
wells.
  Worker safety during excavation was of primary im-
portance to the investigators. Extensive physical examina-
tions were performed on  the workers before and  after
the project. While working at or near the trenches, workers

-------
                                                                                      CASE HISTORIES    339
were required to wear hard hats, safety glasses,  fresh
respirators with multiple purpose cartridges (including
organic vapors),  rubber boots  with steel  toes, rubber
gloves  and cloth  coveralls which  were changed  at least
daily. Cloth coveralls  were  selected over a more imper-
meable suit since actual physical contact, other than with
the rubber gloves, with excavated materials was minimal.
Since summertime  temperatures  in  the  Lathrop area
often exceed 38 °C, the cloth coveralls also posed less risk
to the worker from the standpoint of heat exhaustion.
For those workers who were collecting samples or other-
wise having greatest contact with materials, impermeable
aprons were provided.

Private/Public Well Monitoring Program
  As part of the effort to ascertain the extent of ground-
water contamination  as well as to assure protection of
the public health, both private and  public wells in  the
plant vicinity were sampled (Figure 6). Parameter coverage
included the Table I list, excluding pH, Redox potential,
conductivity and temperature.
  Wells sampled included all known drinking water wells
within approximately 1.6 km of the facility. In addition,
the Company offered  a one-time  sampling to any  of its
employees who were using private wells, however remote
from the facility.  Procedures employed included  first
conducting a well inventory survey followed by gaining
consent to sample. Sampling of approximately 28 wells
was conducted twice monthly for  the first 3 months  and
then monthly for 2 months. By that time at least four
complete sets of-data were available for each well sampled.
This number of complete samplings was judges necessary
by the State of California to support decisions regarding
possible closures of  wells.  Sampling methods  included
collecting the samples in prescribed containers as close to
the well head as possible. Water lines were flushed for a
period of time judged sufficient to gather a representa-
tive sample.
Groundwater Modeling
  Although  the previously  described  monitoring  of
groundwater wells and examination of soil samples  was
essential  to document the  extent of contamination,  it
was also  recognized that a  groundwater modeling effort
was necessary,  particularly in  designing and  assessing
long term remedies. The purposes of the Phase I model-
ing efforts were to:
•Develop a  preliminary two-dimensional model of the
  study area
•Gain an understanding of the relative importance of the
  factors that influence groundwater movement in the area
•Make a preliminary assessment of any existing and future
  pollutant transport paths
•Define  future groundwater modeling efforts for subse-
  quent phases
  Inputs to the first phase modeling effort included using
data gathered during  the groundwater, soil and private
well monitoring activities  as well as independent data-
gathering efforts. From the ongoing activities data  was
gathered  relating  to  stratigraphy,  piezometric  head
measurements from  the permanent wells, and  soil char-
acteristics (i.e.  porosity, permeability, grain size, analy-
ses). In addition,  data were gathered  or  estimated  for
existing land use, well pumpage and construction charac-
teristics in the area, return flows from major users in the
area,  precipitation recharge,  boundary conditions and
anticipated soil-chemical-water  relationships. Pump tests
were also performed using a combination of existing pro-
duction wells,  permanent groundwater  monitoring wells
and additional piezometers as needed.
  Once the data were  synthesized into the preliminary
model, several  sensitivity  studies were conducted to  de-
termine the importance of:
•Vertical depths of the flow field
•Boundary conditions
•Recharge sources
•Pumping patterns (extent and seasonal patterns)
•Aquifer characteristics
Since an integral objective of the plan of study was to
determine  necessary remedial  measures,  several  model
simulations were conducted to forecast future pollutant
transport pathways  resulting from various remedial  al-
ternatives.

RESULTS

  Implementation  of the  Plan of  Study  substantiated
initial concerns regarding the seriousness of the problem.
Considerable pesticide-contaminated soils were found in
a former disposal  area, the pesticide pond bottom and
the ditch which conveyed pesticides to the pond. Based on
soil core sampling data, decisions were made  regarding
the degree  of  need for  offsite  disposal  at  California
Class I or II-1 sites. In all, approximately 1100 m3 of ma-
terial were  excavated and hauled to Class I sites and 2400
m3 to II-1 sites. Excavated  areas were backfilled with clean
soil to within 1  m of grade. A 0.3 m layer of clay (106 cm/
sec permeability) was deposited above this, and then 0.6 m
of clean soil was deposited. The final grade was sloped to
accelerate runoff from the  areas.
  Groundwater sampling  demonstrated the presence of
numerous pesticides in wells adjacent to past disposal sites.
DBCP was detected in numerous onsite wells, reaching
concentrations  as high as 1.2 mg/5 in one well 61 m deep.
Sampling of domestic wells confirmed the presence of
DBPC in wells  to the north by the Oxy facilities. Concen-
trations were considerably less than  those found in  the
test wells,  (approximately 4 mg/p). Occidental,  upon re-
view of these data,  agreed to provide a new permanent
source of water to the  affected and potentially affected
residents.
  The data collected from the test and domestic wells were
also used to select an ongoing monitoring program  for
February, June and  October of subsequent years, as well
as locations for additional wells to further refine definition
of the areal and vertical extent of the groundwater con-
tamination. Two of the  wells  completed  to over 91 m
yielded no detectable  DBCP. Another well  nest  con-
structed west of the  Libby Owens Ford (LOF)  plant also
showed no detectable concentrations of DBCP. Ground-

-------
340    CASE HISTORIES
water  modeling  efforts  confirmed  the migration  of
ground water to the west, consistent with  assumed  re-
gional travel, but also indicated a strong vector toward
the northeast, a result of heavy groundwater pumpage.
  These physical studies and modeling results were used by
Occidental to design a proposed groundwater purge sys-
tem to cease further migration of contaminants offsite.
Purged groundwater from five extraction wells will  be
treated onsite  in an activated carbon treatment system.
Pilot plant studies have indicated removals of pesticides to
below  detection limits. The Company has proposed in-
jecting treated wastewaters into a deep aquifer (more than
100 m) which already contains high dissolved solids. The
governmental agencies  are currently reviewing this pro-
posal and once a remedial system is finally  approved,  it
must be installed within 180 days.
  The extensive work referenced in  this report was ac-
complished over an approximately two year period. In ex-
cess of 50,000 analyses of groundwater and soil were con-
ducted. It would be misleading to say  that the work is fin-
ished or that it progressed without any differences of opin-
ion as to  necessary studies and  interpretation of results.
However, considering the state  of the art nature of the
work and the extreme difficulty in working in the shadow
of an ongoing court case, the cooperation was remarkable.
The author believes this project served as an excellent ex-
ample of how industry and government can put their col-
lective expertise to good  use in solving complex technical
problems.
                                                                       0 - 0.6 MILES
                                                                       BB  -2.2 MILES—»
                                                                       C -0.2
                                                                      »0fc  - 0.8
                                                    Figure 6.
                                     Study Area Showing Domestic Wells Sampled
                                              During Phase I Study.
                                               (Scale: 1" = 1,100')

-------
                               IMPLICATIONS OF THE
                CHEMICAL CONTROL CORP.  INCIDENT
                                          ADAM M. FINKEL
                                         RICHARD S. GOLOB
                               Hazardous Materials Intelligence Report
                                      World Information Systems
                                       Cambridge, Massachusetts
INTRODUCTION

  On April 21, 1980, a massive hazardous waste explosion
and fire took place at the Chemical Control Corp. above-
ground  storage site  in Elizabeth, New  Jersey,  consum-
ing at least 20,000 of the more than 45,000 drums of
hazardous waste present at the site. An unusual combina-
tion of wind and temperature conditions and the results
of a 12-month pre-fire cleanup, prevented the occurrence
of an unprecedented public health disaster.
  As of September 1981, the post-fire cleanup  had cost
more than $27 million, and the  final  site mitigation had
not yet been completed. At least four major civil and crim-
inal suits have been filed as a result of this incident, and
private damage claims total over $500 million.
  The Chemical Control incident provides  an important
case study of  the problems surrounding  hazardous waste
management and uncontrolled waste sites.  No acute inci-
dent in the U.S. involving hazardous wastes has incurred
a greater response cost and no single acute incident in the
U.S. has involved a larger quantity of hazardous waste or
a more varied collection of waste.
  In this paper the authors  will  focus on the events sur-
rounding this  incident, and their implications for future
waste-site responses and for federal and state policies reg-
ulating these responses. The paper divides the incident in-
to five distinct phases:
(1) the operations of Chemical  Control before  the New
   Jersey Department   of  Environmental  Protection
   (DEP) took any organized action at the site
(2) The  cleanup that  occurred during  the 12 months  be-
   fore the fire
(3) The fire itself
(4) The  intensive response actions during  the year after
   the fire
(5) The work  remaining at the site and the ongoing costs
   and litigation associated with it.

PRE-RECEIVERSHIP

  This phase  began  in the mid-1970s,  when Chemical
Control began accepting hazardous waste from area gen-
erators;  it ended in  January 1979, when DEP  obtained
a court restraining order and, for the first time in history,
took control of a private company. According to the Over-
sight and Investigations Subcommittee of the House of
Representatives Interstate and Foreign Commerce Com-
mittee, DEP solid waste officials knew about the potential
problems at the Chemical Control storage facility at least
by the spring of 1978, at which time the site contained
about 25,000 drums.
  DEP  and the New Jersey Division of Criminal  Jus-
tice (DCJ)  both agree that their  respective agencies  did
not communicate adequately during the pre-receivership
phase because of understaffing. In addition, a DCJ mem-
orandum in 1978 claims that DEP investigators were  not
adequately  trained and,  although aware of the problems
at Chemical Control, did not have the expertise to assess
their severity. Consequently, according to the memoran-
dum,  the investigators treated the site as a landfill and
applied  landfill guidelines  to its regulation. Other state
officials have said that, if DEP regulators had made ade-
quate checks of the state-required manifest forms, they
would have recognized that Chemical Control, licensed as
an incinerator facility only, was accepting large quanti-
ties of physically non-incinerable wastes.
  Since this occurred, New Jersey has developed the  11-
state Interagency Hazardous Waste Strike Force, a model
program for other  enforcement and regulatory personnel
around the country. In addition, New Jersey's status as
the leading  hazardous waste generating state has made its
regulatory agencies the leaders among the 50 states in deal-
ing with these problems.
  The unregulated existence of Chemical Control during
its early history does not appear to be solely a function
of the times, though. Testimony given at an Oversight
and Investigations Subcommittee hearing on December 16,
1980 claimed that,  even  now with the implementation of
a national manifest system under the Resource Conserva-
tion and Recovery Act (RCRA), operators can still main-
tain sites that pose similar threats as  the Chemical Con-
trol facility. The principal witness at the hearing was Har-
old Kauffman, a former official of Duane  Marine Corp.
in Perth Amboy, New Jersey, who had turned state's wit-
ness after being charged with complicity  in  illegal haz-
ardous waste management.
  During his testimony, Kauffman recommended that the
U.S. EPA and state regulatory agencies change the focus
of their oversight of the hazardous waste  industry.  Ac-
cording  to Kauffman, the state and federal governments
emphasize the routine filings of applications, and at pres-
ent, nearly  anyone can pay a fee and become a licensed
                                                   341

-------
342    CASE HISTORIES
hazardous waste hauler or begin the siting process for a
landfill. John Albert, who allegedly assumed ownership
of Chemical Control in 1977 at gunpoint, claims that he
received a DEP hazardous waste hauler's license without
asking for  one.  Albert's other company, Jersey Sanita-
tion,  recently received a renewal  of its  license,  even
though  its principals  had been  convicted  of felonies.
Kauffman said that, because no agency currently moni-
tors the background and safety history of applicants for
hauling and disposing operations, incidents such as Chem-
ical Control will continue to take place.
  Kauffman testified that he did not believe that a com-
pany  with holdings in  the garbage and sanitation indus-
tries should be allowed to enter the hazardous waste in-
dustry, for several reasons:
(1) operators of garbage trucks can easily mix quantities
    of toxic wastes with the municipal garbage and
(2) garbage haulers usually have established relationships
    with municipal dumps, which are less stringently regu-
    lated than RCRA hazardous waste landfills.
Consequently, these haulers can pass on significant quan-
tities  of hazardous  wastes to the municipal dumps. A
subcommittee spokesman noted  that Kauffman's com-
ments were based on the activities of a limited number of
small companies operating in metropolitan New York, and
do not necessarily apply to the major firms that deal with
both solid and hazardous waste.
  Leonard Tinnan,  corporate development director for
the nation's  largest waste management  firm, Chemical
Waste Management Inc. of Long Beach, California, com-
mented that the large national companies with both solid
and hazardous waste operations had largely succeeded in
maintaining separate corporate structures  for their differ-
ent functions. Tinnan said that companies such as Chem-
ical Waste Management and Browning-Ferris  Industries
have gradually separated the organization of solid and haz-
ardous waste operations due to the different backgrounds
needed to manage each operation, as well as due to the
public fears about the practices that occur among smaller
companies.
  In addition, at the subcommittee hearing, Congressman
Albert Gore (Democrat-Tennessee)  said  that, when  a
hauler has  a  tacit agreement with a disposal facility, the
hauler and facility owner could potentially collude to falsi-
fy the national RCRA manifests so that, instead of trans-
porting the wastes out of  state  to another authorized
facility, the hauler  could mix the waste  with household
garbage and dispose of it at a municipal  landfill. Alleg-
edly, through such practices, hazardous wastes which were
supposed to have been incinerated at  the Chemical Con-
trol facility ended up  at the Kin-Buc municipal landfill
in Edison, N.J., currently one of  the nation's largest un-
controlled hazardous waste sites at 220 acres.
  Although one U.S.  regulatory agency—the Interstate
Commerce Commission (ICC)—is currently attempting to
assert its authority  to  make background  checks on haz-
ardous waste haulers,  the ICC is receiving considerable
opposition  from the regulated community. Trade associa-
tions, including the Spill Control Association of America
and the Hazardous Waste Services Association, agree that
a federal  agency should take steps to curtail midnight
dumping,  but believe that  ICC  regulation of hazardous
waste haulers would duplicate  existing  EPA and U.S.
Department of Transportation regulations and would too
closely control the entry of legitimate firms into the haz-
ardous waste  industry on  a geographical or a market-
share basis.

PRE-FIRE CLEANUP

  This phase extended from April 1979, when DEP hired
Coastal Services Inc. to perform preliminary  cleanup
work, until April 1980, when the Chemical Control stor-
age  facility caught  fire. When  Coastal  Services  began
work, the facility contained a maximum of 60,000 drums
and  laboratory packs of chemicals, including polychlori-
nated biphenyls (PCBs), pesticides,  cyanides,  compressed
gases, organic peroxides, high explosives including TNT,
picric  acid and nitroglycerin, bottled hospital and lab-
oratory specimens and radioactive wastes.
  During  the  one-year pre-fire  period, Coastal  Services
spent approximately $1.5 million and  supervised the re-
moval of about  10,000 drums and laboratory packs con-
taining some of the most dangerous wastes. These wastes
were disposed of at the CECOS International, Inc., chem-
ical  landfill near Niagara Falls,  New York. The major-
ity  of the high explosives  and compressed  gases  were
detonated under controlled conditions at a nearby ord-
nance works.
  Although the preliminary cleanup probably helped to
avert a major public health disaster,  the fire did take
place and caused major damage. As a result, the ques-
tion remains as to whether the pre-fire cleanup was con-
ducted in an expedient  manner  and with the proper re-
moval priorities.
  Before discussing the contributions of Coastal Services
to the pre-fire effort, it must be noted that several ob-
servers have said that, at the same time  as Coastal was
removing  wastes, DEP was allowing more wastes to ac-
cumulate at the site. A Congressional staff member said
that a computer analysis of aerial photographs indicated
an  influx of  several thousand  drums after September
1977. According to a  former  Coastal employee, DEP
realized at the early stages of the pre-fire cleanup that, as
a result of the increasing national attention given to the
site, Chemical Control  was virtually guaranteed federal
and state cleanup  funds.  According to  this argument,
DEP decided  to consolidate the  wastes present at several
smaller waste sites in northern New Jersey and store them
at one central location—Chemical Control—where they
would be cleaned up along with other wastes at the site.
  In contrast, Tom Dalton, former vice-president  of
Coastal Services, said that the drums  which were trans-
ported to the Chemical Control  site during  the pre-
fire  cleanup represented the contents of 14 trailers aban-
doned by the  former Chemical Control operators at vari-
ous  places throughout the  city of Elizabeth.  Dalton said
that the contents of these  trailers contained  a maximum
of  1000  drums, and that  reports  of  massive drum ac-
cumulation resulted  from  a misreading of the aerial

-------
                                                                                     CASE HISTORIES     343
photographs, which actually showed  not an increase in
the total number of drums but rather an increase in the
surface area covered by the drums due to the transfer of
drums stacked six and seven high to piles of drums stacked
two and three high. Nevertheless, the Chemical Control
operators used the allegations of drum accumulation from
outside sources in their trial defense in an attempt to les-
sen their proportionate  liability for the waste consumed
in the fire.
  The success of the pre-fire cleanup depended largely on
the success of the DEP effort to identify the original gen-
erators  of the Chemical Control waste  and  encourage
them to reclaim their drums.  In practice, th is effort
took the form of threatening those generators who could
be identified from the drum labels with legal  action in
the massive liability  surrounding the Chemical Control
site if they  did not  come  forward  and reclaim their
wastes.  During the pre-fire  phase, DEP identified several
major generators  who voluntarily removed almost 7900
drums from the site.
  Gregory Heath,  technical  representative  for Peabody
Clean Industry, formerly Coastal  Services, said that, dur-
ing the  pre^fire cleanup, DEP had refused Coastal's sug-
gestion  that  it transport the overpacks and drums de-
signated for  generator reclamation to  an off-site ware-
house for temporary storage. Heath also said that, as a
result, some  of the drums which had  been bulked and
overpacked for reclamation were consumed in the fire.
  While DEP was operating under severe financial con-
straints, as evidenced by the  depletion of the $20 million
New Jersey Spill Compensation Fund  and the subsequent
decision to temporarily raise the chemical and petroleum
feedstock taxes that  supply this  fund,  several  observers
commented that speed rather than economy should have
been the primary objective during the pre-fire phase.

Superfund
  The same observers expressed concern that the current
EPA administration is encouraging slow, deliberate and
economical  action at  waste sites. The  Superfund  In-
terim Removal Guidance, which was sent in early August
to the federal on-scene  coordinators (OSCs),  mandates
that the OSCs not initiate any  long-term remedial action
at a site,  unless it is  among the 400 waste sites on the
national priority list and unless a significant effort has
failed to locate a responsible party capable of  providing
the necessary cleanup funds.
  Under the guidance document,  OSCs  have authority to
begin   so-called  "immediate   removal  actions,"  or
"planned removal  actions,"  in order  to stabilize the site
and prepare it for long-term  remedial work. OSCs must,
however,  receive approval from EPA  headquarters be-
fore authorizing a removal  action, and  EPA  has indi-
cated that it intends to  "establish a  high threshold for
undertaking such actions."
  In addition, the  interim Superfund guidance, which
will remain in effect until the final National Contingency
Plan is  promulgated  sometime  in late 1982, contains six
case histories designed to provide the OSCs with models
of typical response actions.  At least  two of these cases
criticize earlier EPA actions at specific waste  sites  as
overly ambitious and expensive. According to the interim
guidance,  the case history of the Kin-Buc landfill demon-
strates EPA's new policy that removal actions should not
be initiated at waste sites  with continuous chronic re-
leases of hazardous wastes, unless those releases directly
threaten human populations.
  In practice, EPA policy now  prohibits the EPA-funded
use of groundwater or leachate treatment systems unless
alternative approaches, including the "no-response op-
tion," have been considered; in effect,  the interim guid-
ance discourages the massive contaminant removal and
groundwater treatment actions  that EPA and DEP used
at the Chemical Control  site.  Furthermore,  in the  case
history of the PCB storage site at Sharpstown, Maryland,
the new EPA interim guidance suggests that drums should
be removed to an off-site location only when the drums
are actually leaking, and  that deteriorating storage con-
tainers should only  be staged  or overpacked.  A senior
official  at O.K. Materials Co.  of  Findlay, Ohio, the
major post-fire contractor, said  that this "reduction  in
the formerly aggressive posture of EPA" may result  in
greatly increased costs when cleanup is finally initiated
and  possibly in an  increased  incidence of  catastrophic
events at  waste  sites where the federal government has
adopted the "wait and see" approach.
  Many observers believe that the "wait and see" policy
is inappropriate  when evaluating the need for security  at
an uncontrolled  site. The Elizabeth Fire Department, the
New Jersey DCJ and the Federal Bureau of Investigation
have not yet completed their official arson investigation
into  the  Chemical Control incident, but since the fire
occurred during the cool springtime, it seems unlikely that
its origins were totally spontaneous. Following the Chem-
ical Control incident,  at  least  one other significant ha-
zardous waste site fire has occurred; on July  10,  1981,
the fire at the General Disposal  Co.  waste site in Santa
Fe Springs, California, consumed over  10,000  drums  of
waste and resulted in cleanup expenditures of up to $2.7
million. The probable cause of that fire was arson; a local
resident allegedly  tossed  a Molotov cocktail  onto the
site with the intention of starting a small fire to drama-
tize the need for security and eventual cleanup at  the site.
Fire
  The actual  emergency and subsequent response were
characterized  by a  combination  of luck and  effective
action. Two physical factors decreased the severity of the
fire  itself.  First, the large number  of drums and their
high packing  density created an intensely hot fire, over
2000 °F at its center. As a result, the fire itself acted as a
natural incinerator to degrade the hazardous wastes. DEP
analyses during the fire showed high concentration of ben-
zene, up to 10,000 ppm in the  smoke plume, substanti-
ating the theory that many  of  the  substituted  aromatic
compounds were degraded into  simpler molecules during
the fire. Second, repeated  wind shifts and a high ceiling
kept the smoke plume from settling  over a densely popu-
lated area near the site.

-------
344     CASE HISTORIES
  The firefighters,  under  the  supervision of  Elizabeth
Fire Director Joseph Sullivan, were particularly effective
at combatting the acute chemical emergency. More than
a year before the fire, Sullivan had instituted a special
site-specific training program in anticipation of problems
at the Chemical Control site. Firefighters learned about
the location of access routes, the topographic conditions
at the site,  and the  availability  of water supplies  and
utility outlets.
  Sullivan  said that  the  training program was instru-
mental in increasing the speed of the initial response to
the blaze. He said that more than 250 firefighters worked
on the response, which began  at 2254 LT on  April 21,
and that they successfully contained the fire by 0914 LT
the following morning. The land-based firefighting crews
uses  more than 7.2  million gallons  of water  to  control
the fire and two fire boats stationed in the nearby Eliza-
beth River used several million gallons more.
  Heath of Peabody  Clean Industry said, however, that
DEP  should  have ordered the  firefighters to let the fire
burn  and to apply  water  only to control the spread of
the fire  to adjoining properties. He said that this ap-
proach would have created less contaminated water and
degraded more material, resulting in a cheaper and less
dangerous post-fire cleanup. Alden  McLellan, DEP as-
sistant commissioner, agreed  that the "free burn" was a
viable option,  but  cautioned  that  temperatures at the
outer edges of the fire might not have been high  enough
to degrade the most persistent toxic compounds.
  Acting Elizabeth  Fire  Chief Charles  Swody  said in
September 1981 that  his department had learned several
lessons as a result of the fire.  Swody emphasized the need
for fire companies to maintain  an adequate supply of ex-
tra breathing masks because, during  the Chemical Control
fire,   many   off-duty personnel  and  firefighters  from
neighboring  communities  provided  services and  needed
protective clothing.  Swody also said that, to reduce the
amount of worker exposure,  his department might have
placed greater  reliance on remotely-operated  equipment
and relieved  its workers on  a more frequent  basis. He
claimed,  however, that no fire department in the country
has the budget to afford state-of-the-art protective cloth-
ing for each  responder at  a site incident  such as  Chemi-
cal Control. One firefighter is suing the Elizabeth  fire de-
partment for pulmonary disabilities  allegedly  contracted
during the response.


POST-FIRE CLEANUP

Introduction

  This phase began on April 24, 1980, when DEP con-
tracted with O.H. Materials Co. to secure the site against
the immediate  threat  of contaminated  surface  water en-
tering the Elizabeth  River. Also involved was the removal
of hazardous waste  drums and contaminated  soil  from
the site, as well as the treatment of contaminated ground-
water.
  The post-fire cleanup phase lasted  until April  1981,
when the New Jersey Spill Compensation Fund and the
Federal  Water Pollution  Control  Act Section  311 (k)
funds were  withdrawn and O.H. Materials  removed its
groundwater treatment  system.  During  this  one-year
period,  more than  350 million  pounds of contaminated
debris and soil were removed to the CECOS landfill in
Niagara Falls for disposal, more than 252,000 gallons of
incinerable liquid wastes were taken to the  Rollins En-
vironmental Services incinerator  in Logan Township, New
Jersey, for destruction, and about' 36,450 crushed empty
drums  and  3770 intact drums  were removed  to secure
landfill disposal.
  Three basic questions regarding this phase of the re-
sponse deserve attention:  1) Was the approximately $25
million spent during the year justifiable in terms of the
results  obtained? 2) Was  the level of personnel safety
during the post-fire response  efficient for an action in-
volving  such a  large amount and complex  mixture of
waste? and 3) What new techniques and instrumentation
used at Chemical Control could prove valuable at future
responses?

Cost
  During  the post-fire cleanup, O.H.  Materials used  a
team of 71  workers to remove 25,000 barrels,  at a total
cost of  more than  $17 million.  In contrast,  in a similar
one-year period, Coastal  Services used a team of 14 to
remove  10,000 of the most dangerous drums from the site,
at a total cost of $1.5 million.
  A DEP official commented that, in the days following
the fire, DEP did not have time to negotiate  the costs of
the response because of intense  public pressure  to initiate
an immediate and complete cleanup action. DEP selected
O.H. Materials as  the cleanup contractor not only be-
cause it had the requisite capabilities but also because it
offered  the most immediate initial response. The DEP
official  also commented that, in developing  its response
program,  O.H.  did not give  adequate attention to  cost
considerations, encouraged by the relatively limitless fed-
eral funding for this project.
  The DEP official further said that, in the future, state
agencies  should consider  establishing standing contract
relationships with cleanup firms and that such contracts
should have clearly delineated cost parameters. He  cau-
tioned,  however, that such  relationships  tend to form
exclusive ties with one company and generate charges of
favoritism when incidents with a high potential for profit
and recognition occur.
  In response, Robert Graziano, vice president of O.H.
Materials,  said  that O.H. Materials took a  meticulous,
technology-intensive approach in securing and cleaning
up  the Chemical Control  site.  He said that O.H. com-
pleted more than 90% of the tasks outlined in the original
post-fire  contract,  even  though  problems with  on-site
PCBs and deeply buried wastes were encountered during
the work.  In addition, Graziano said that,  prior to the
fire, Coastal Services did not have to deal with the prob-
lems that  O.H. Materials faced with debris, damaged
drums and unlabeled piles of drums.
  As an example of the legitimate cost differences in  vari-
ous incident responses, Graziano contrasted the Chemical
Control incident and response with that at the Seymour,

-------
                                                                                      CASE HISTORIES     345
Indiana waste  site.  Although  Seymour response crews
were faced with a similar number of drums as that  at
the Chemical Control site, they were able to finance the
response to date at approximately 20% of the cost, pri-
marily, according to Graziano, because: 1) the drums were
segregated before the state of the cleanup,  2)  there was
no fire to cause a deterioration in site and  drum condi-
tions and 3) the remote nature of the Seymour  site did
not require as extensive community safety precautions.

Worker Safety
  The National Institute  of Occupational  Safety  and
Health (NIOSH) has not yet completed its  investigation
into the  various cases of worker exposure to hazardous
substances during  the   post-fire  cleanup   at  Chemical
Control. James Melius,  chief of  the  NIOSH  Hazard
Evaluation Section, reported that, in July 1981,  NIOSH
completed  a follow-up  medical  survey among  the ap-
proximately 390 firefighters and  other response workers
at the initial incident. Of the 224 workers  who  had re-
ported no adverse pulmonary symptoms, such  as wheez-
ing, excessive phlegm, and shortness of breath, immedi-
ately after the fire, 110 reported the onset of one or more
of these symptoms in the following year. Melius said this
incidence of delayed effects was  serious  enough  that
NIOSH decided to conduct a detailed  medical study  of
the 390  workers to  determine the  extent  of  incident-
related pulmonary problems.
  During the summer months, NIOSH had  difficulty de-
termining whether certain transient worker illnesses were
due to cyanide poisoning, as suggested by  the odors  at
the site,  or due to heat stress. Three cases of worker in-
capacitation occurred during the early  morning  hours,
however, when heat-stress would  not be expected to be a
problem.  Although  NIOSH photoionization tests were
negative, NIOSH noted that the vapors could have been
acute  for transient periods only and may not have been
present during the test period.
  NIOSH did make several recommendations about work
periods at the Chemical Control  response and at future
actions, including: 1) each thirty-minute work period  in
full breathing apparatus should be followed by a thirty-
minute rest period,  2)  workers  should be acclimatized
to possible heat-stress problems  upon  entering  the site
and 3) the dispensing of diuretic drinks, including coffee,
colas and alcohol, should be discouraged because of the
potential for dehydration.
  Melius also reported  that, as a result of the cases  of
heat stress  among  the  workers, NIOSH  developed a
system for continuously monitoring both  the environ-
mental conditions and  the worker's heat  stress  in the
field. During August 1981, NIOSH tested the system at a
non-chemical site  cleanup in Arizona,  monitoring the
axillary temperature and heart rate of each worker.
  A Coastal official said that Coastal Services walked
off the Chemical Control site because  of  the  insistence
of a DEP official that  Coastal workers wear full  self-
contained breathing apparatus at the  early stages of clean-
up. This observer disagreed because, at that point in the
cleanup, there was a  15-knot wind and a light rain falling,
conditions he  said created little danger of acute concen-
trations of fumes. He said that the SCBA gear severely
impaired the  ability  of  Coastal  workers to  navigate
through the rubble and wet drums during the early stages,
and that such limited mobility actually made conditions
less safe than if  the workers  had been wearing respira-
tors alone.

New Technology
  Several new protocols and technologies, with potential
usefulness at  other  responses,  have emerged  from  the
Chemical Control incident. These new techniques, which
were developed by O.H. Materials, include the following:
  (1) A  special  barrel grappler that remotely handles
      drums and moves them to a staging area, thereby
      replacing the standard method of overpacking each
      drum  individually.   The  grappler  consists of a
      Caterpillar  backhoe  with  a specially-designed  arm
      that can rotate through 360°. IT  Corp.  used this
      device at the Santa Fe Springs post-fire cleanup.
  (2) A  proprietary  compatibility  sampling  protocol,
      which uses a mobile field laboratory to  test both
      for the  toxicity and chemical compatibility of  the
      liquids contained in the  drums. This sampling pro-
      tocol allowed O.H.  to identify those drums which
      could be bulked to reduce the transportation volume
      of the wastes.
  (3) A  specially-designed compatibility  chamber   for
      liquid wastes, consisting  of a 12,000 gal open-top
      vessel  fitted with a  series  of  thermocouples. After
      adding new materials to the chamber, O.H. per-
      sonnel could precisely monitor the heat of reaction
      from a safe distance to determine if the addition of
      the materials  should be slowed due to incompati-
      bility.
  (4) A health and  safety protocol  for moving personnel
      on and  off the  site, consisting  of several clearly
      delineated "hot," "warm," and "cold" zones.
  (5) A proprietary scheme for identifying and repackag-
      ing the  contents  of the  "lab packs." O.H. con-
      structed a portable building on-site for this purpose
      and retained a professional chemist to perform  the
      testing  and repackaging  of the  laboratory chemi-
      cals.
  (6) The first field use of the O.H. Underground Re-
      covery  and Treatment  System  (URTS)  at  a  ha-
      zardous waste site.   The URTS  uses a  pneumatic
      recovery system to  create a  zone  of  depression,
      thereby accelerating the movement of contaminated
      leachate toward the recovery  wells. O.H. was able
      to  use a variety  of techniques in conjunction with
      the URTS,  including:   1)  the high-temperature
      steam-stripping of  solvents  in  the contaminated
      leachate, 2) a clarifier unit to precipitate out cer-
      tain insoluble constituents from the groundwater
      and 3) an activated  carbon treatment unit for final
      "polishing" of the leachate in preparation for dis-
      charing it back onto the site.
  In addition, the consultants at Roy F. Weston, Inc. in
West Chester, Pennsylvania, took advantage of the Chem-

-------
346    CASE HISTORIES
ical Control incident to make the first successful use of
ground  penetrating radar  (GPR)  technology to locate
hazardous  waste  targets in saltwater. The U.S. Coast
Guard had expressed concern that the force of the initial
Chemical Control explosion had thrown chemical drums
into the Elizabeth River and that the water spray during
the subsequent firefighting operation had caused sections
of the river bank containing the drums to slump into the
river. Although previous GPR surveys in saltwater had
failed because the uneven  salinity of the water attenu-
ated the return  signal to an  indeterminate degree,  the
Elizabeth River had a relatively constant salinity and the
Weston researchers were able to use a combination of GPR
and fathometry readings to determine the exact degree of
attenuation at each  point  and calibrate the results  ac-
cordingly.
Ongoing Action
  After a cleanup program lasting  two years and the ex-
penditure of more than $27 million, the Chemical Con-
trol site still remains a candidate for Superfund remedial
action. According to the  former DEP on-scene coordi-
nator, the site has received  56 "points" on the Mitre
ranking model, placing it about 20th among the nation's
problem  sites. DEP  has ranked Chemical  Control  13th
among the 300 New Jersey sites that it wants EPA to con-
sider  for Superfund remedial actions. In early September
1981, the EPA Office of Emergency and Remedial  Re-
sponse received  a detailed request  from DEP  outlining
the remaining work  necessary to clean up  the Chemical
Control site, which DEP  said still  presented an  uncon-
trolled leachate threat.
  The DEP estimates that an additional $700,000 will be
needed to perform the following remedial actions:
  (1) Removal and disposal of 1500 cubic yards of con-
      taminated debris,  including contaminated soil, ply-
      wood and concrete
  (2) Removal of sludge from the groundwater clarifier,
      contaminated spent carbon from the URTS and 75
      crushed drums
  (3) Removal  and  detonation  of  200 pressurized  gas
      cylinders of "uncertain" integrity
  (4) Dismantling of a 10,000-gallon underground steel
      tank found during the cleanup
  (5) Removal and disposal of 60 drums containing PCB
      wastes  concentrated  from settling  tanks installed
      during the cleanup.  Some of these drums contain
      up to 977,000 ppm PCBs, although the majority of
      them contain less than 1000 ppm
  (6)  Removal and disposal of small quantities of miscel-
      laneous material,  including one drum of contami-
      nated clothing, two drums of arsenic pentoxide,  one
      can of potassium cyanide and a 2000-gallon tank
      truck left on-site by the former Chemical Control
      operators.
  (7)  Additionally, the final URTS treatment of the con-
      taminated  groundwater  had  not  been completed
      when the funds were withdrawn.

  At least four civil and criminal cases were filed as a re-
sult of the Chemical Control incident. In one recent ac-
tion, a federal grand jury in Newark, New Jersey, sen-
tenced John Albert and Eugene Cordon, the two men who
allegedly took over operation of Chemical Control in 1577,
to five and three year jail terms respectively for commit-
ting two counts each of mail fraud in connection with the
Chemical Control operation. These sentences represent the
first  successful federal  felony prosecutions against  op-
erators of hazardous waste storage facilities. Since the il-
legal storage took place before the November 19, 1980
effective date  of  RCRA, the U.S. attorneys decided to
invoke the federal fraud laws.
  The  prosecutors  based their indictment on evidence
gathered  from the  investigation  of the acceptance by
Chemical Control of a single cylinder of highly unstable
perchloroisobutylene from a nearby generator, since  the
unusual nature of this small shipment made the tracing
of bank checks and acceptance letters easy  and verifi-
able. Albert and  Conlon were charged with defrauding
DEP by promising to treat various wastes at the Chemi-
cal Control incinerator, when in fact the wastes were
transported to another unlicensed dump site. They were
also charged with defrauding various generators by prom-
ising to properly dispose of their wastes. The following
three suits are still pending:
  (1) A criminal  trial to bring nuisance  charges against
     Albert  and Conlon for maintaining a  hazardous
     situation at Chemical Control. Since New Jersey
     recently amended its criminal code in 1980 to in-
     clude  hazardous waste violations as felonies,  the
     two  Chemical  Control operators  will  only  be
     charged with misdemeanors.
  (2) A civil suit filed by DEP to recover nearly $24 mil-
     lion in cleanup costs  from the Chemical Control
     operators. This suit also seeks penalties under  the
     New Jersey Solid Waste  Act amounting to more
     than $25,000  per  day; DCJ was pessimistic about
     the  ability  of the  insolvent  Chemical Control
     operators to provide even a small  fraction of  the
     sum  needed  for  significant reimbursement.  The
      federal government may also sue to  recover its
     Section 311 (k) fund expenditures.
  (3) A suit against DEP filed by the Loizeaux Builders
      Supply Co., owners of the property adjacent to the
      Chemical Control facility. Loizeaux alleges  that
      DEP forced him out of business by taking over his
      property for use as a drum staging area shortly after
      the fire.

  Perhaps most importantly, the  Chemical Control inci-
dent  contributed to  public awareness of the need for  en-
vironmentally sound disposal facilities. It also had a major
effect  on the  passage  of  New  Jersey's  new "Major
Hazardous Waste Facilities Siting Act,"  signed on Sep-
tember 10, 1981 by Governor Brendan Byrne. The Act
contains  the  following  provisions  encouraged by  the
Chemical Control incident:
  (1) Siting approval shall not be granted to an applicant
     if any party to the application has been convicted
     of a crime  relating to improper hazardous waste
      management during the previous ten years

-------
                                                                                     CASE HISTORIES    347
 (2) All new major facilities shall be totally or partially
    constructed above ground and be designed to allow
    the extraction of 99.9% of all stored or disposed
    waste
 (3) Local  inspectors who supply information on im-
    proper facility management shall be given one-half
    of any penalty collected in the ensuing litigation
 (4) A facility must undergo a formal siting review pro-
     cess if a capacity expansion of more than 50%
     is planned.

Conclusion

 Despite the controversies surrounding the regulation of
Chemical Control, the response to the  chronic site prob-
lem and  the costs  and approaches involved in the  post-
fire response, the  final  cleanup  of the uncontrolled site
will represent the  successful conclusion of  a hazardous
waste project whose difficulty may remain unsurpassed
in future years. The cooperation between the cleanup con-
tractors and the federal and state regulators proves that,
given sufficient time, money and good  fortune, the mis-
takes of the unregulated  past  can be efficiently  and
safely corrected.
  In the future,  researchers,  engineers and  government
officials  need to work closely to develop protocols,  as
well as checks and balances, that can ensure positive re-
sults at site actions where these factors are not so favor-
able.  As the Superfund cleanup program  begins,  the
private sector and government participants in uncontrolled
site actions should have ample opportunity to perfect the
knowledge gained from the Chemical Control incident.

REFERENCES

  The authors have relied extensively on information that
appeared in  the Hazardous  Materials Intelligence Re-
port, an international  weekly newsletter  focusing on the
safe management of hazardous wastes and hazardous ma-
terials and  published  by World  Information Systems,
Harvard Square Station,  P.O.  Box 535,  Cambridge,
Massachusetts 02238.  The Hazardous Materials Intelli-
gence Report began reporting on the Chemical Control
incident  in its Vol. I,  No. 5  issue on 13  June 1980 and,
since then, has provided regular coverage of the cleanup
operation,  legal proceedings and other issues related  to
the event.

-------
      A FAST TRACK APPROACH TO MANAGEMENT AND
                                 IMPACT ASSESSMENT     (PART i)
                                    GREGORY A. VANDERLAAN
                                U.S.  Environmental Protection Agency
                                                Region V
                                            Chicago, Illinois
INTRODUCTION

  The approach discussed here was developed during a
Federally funded emergency cleanup action under author-
ity of Section 311 of the Clean Water Act. The subject of
the effort and  cause for action, the Seymour Recycling
facility,  is quintessential of the many abandoned,  haz-
ardous waste sites found throughout the country. It is a
large site covering an area of 13 acres with a diverse collec-
tion of hazardous compounds stored in a variety of con-
tainers.

BACKGROUND

  Located in the Indiana community of Seymour, approx-
imately  ninety  miles south of Indianapolis, the Seymour
Recycling Center was  established  in 1971 solely to re-
cover methylene chloride used  for industrial  operations
at the Seymour Manufacturing Company. The recycling
center soon expanded its operation to include the reclama-
tion of other industrial chemicals.
  In 1975, the Seymour Recycling Center incorporated
separately from the Seymour Manufacturing  Company.
In 1978, when  the Recycling Center was sold, owner es-
timates  of on-site waste storage included approximately
42,000 55-gallon drums, 100 bulk storage vessels contain-
ing approximately  676,000 gallons of waste  and seven
20,000 gallon rail cars.
  Shortly after the change in  owners, serious environ-
mental problems  began to develop. The incinerator was
operating  poorly and  without  an  emission  control sys-
tem. Micrometeorological  conditions  often  resulted in
fumigation events affecting nearby residents. These occur-
rences prompted the affected people to complain to  local
and  State pollution control  officials,  demanding  that
they intervene  in the operation and force a cleanup to
attain compliance with the appropriate regulations.
  Representatives of the recycling  facility signed a  con-
sent decree in  June, 1978, with the State of Indiana re-
quiring  them to shut down and dismantle the incinera-
tor and  to reduce their drum inventory on a scheduled
basis. Fourteen months after  the consent  decree was
signed,  however,  the  recycling center's  drum inventory
actually  increased by approximately 3,400. This was con-
firmed  through aerial  photographs evaluated by  U.S.
EPA's Environmental Photographic Interpretation Center
in Warrenton, Virginia. According  to  recycling center
officials, the terms of the consent decree were not being
met for several reasons including labor problems, lack of
availability of secure landfills and lack of State approval to
place drums containing solids in non-secure landfills.
  Several  near catastrophic events developed at the re-
cycling center after agreement and implementation of the
consent decree. A major fire occurred in October, 1978,
when a vehicle struck a drum containing lithium aluminum
hydroxide in hexane, igniting it  and several  adjacent
drums. Also, several  major releases occurred on-site in-
cluding an alkanol amine spill, a chromic/nitric acid spill
and an  alleged hexachlorocyclopentadiene spill. Finally,
in the fall, 1979,  after  repeated failures  on the part of
ownership to meet the consent decree  schedule, the State
of Indiana brought  the site  to  our attention for  en-
forcement action.

AGENCY EFFORTS

  The National Enforcement  Investigation Center com-
pleted a field investigation01 designed primarily to identify
the materials on site and to establish the potential for any
toxic materials to adversely affect the local population,
Liquids and soils were sampled at locations within the re-
cycling center, the drainage ditch north of the site, nearby
residential and industrial wells and the city sanitary sewer
downstream  from the recycling  center.  Analyses of the
samples discovered the presence of  a number  of com-
pounds  in the area soils with documented adverse human
health effects (Table I.).

                      Table I.
     Carcinogenic Compounds Found in Area Soils
Compound

Benzene
Chloroform
Methylene chloride
Toluene
Concentration 0*g/l)

        5
       20
      200
     2000
  The soil in the area was found to be very porous, po-
tentially allowing infiltration of contaminants into the
local groundwater system. Although there were no data
documenting groundwater pollution,  EPA believed the
                                                   348

-------
                                                                                       CASE HISTORIES    349
presence of such large quantities of hazardous substances
on the site and the nature of the soils in the area virtually
guaranteed shallow aquifer contamination.
  While this initial field investigation was intended to de-
velop information on which to base litigation proceedings
under the Resource Conservation and Recovery Act of
1976, its findings made clear the need to  seek other more
immediate alternatives  to achieve a site  cleanup. By the
time this investigation was completed in late 1979, the re-
cycling center's management was operating in total disre-
gard of the consent decree and had allowed the drum in-
ventory to approach 60,000.
  As litigation proceedings went  further and  cases were
filed, the recycling center owners officially declared bank-
ruptcy in early 1980. Without  a responsible party to fi-
nance cleanup, government action was now the only means
by which the site could be stabilized.
  In March, 1980, an explosion occurred at the site as a
result of waste materials reacting after escaping from their
decaying containers. The State of Indiana requested assis-
tance from EPA Region V. A Federal On-Scene Coordina-
tor (OSC)  from the Environmental Emergency Section
traveled to the site to determine if the explosion and leak-
ing containers constituted a  discharge or  threat of dis-
charge to navigable waters. It was necessary to make this
determination  to establish Federal authority and gain ac-
cess to emergency cleanup funds under Section 311 of the
Clean Water Act.

EPA Responds
  Shortly after arrival at the recycling center,  the OSC
made an official determination that the abandoned site
represented a threat to  public health due  to potential sur-
face water contamination from site run-off and  probably
ground water  contamination  from infiltration. The OSC
convened a Regional Response Team (RRT) meeting. All
members for the team  agreed that a  substantial  threat of
run off to waters of the U.S.  existed as  long as the con-
ditions of the site were left unchanged.  The U.S. Coast
Guard representative made it  clear to  the response team
membership that, as the 311 fund administrator, removal
of materials from the  site or any activities not directly
associated with containment would be considered inappro-
priate and not eligible for funding.
  The OSC hired a contractor under Coast Guard emer-
gency procurement authority. Contractor activities for the
next thirty days involved drum segregation and staging,
site grading and paving, earth work designed  to direct
surface water runoff around the site and placement of a
runoff lagoon and treatment system to collect runoff from
the site itself  and treat it through a carbon  adsorption
system.

311 Action

  At the conclusion of  this effort which cost $866,000 and
after evaluation of the work completed, EPA realized that
an alternative approach to contractor procurement was
needed to insure a more cost effective use  of Federal
funds.  Placed in the perspective of implementing a site
cleanup, the funds expended did not produce significant
results. It was evident that extension of traditional 311 pro-
curement practices  to address uncontrolled  waste  site
cleanup would mean large expenditures with little cost con-
trol capability. After a thorough review of the 311 effort,
EPA felt it would be possible to achieve a greater level of
effort at less cost by:
(1) removing the constraints on use of the 311 fund for
    removal of materials from the site and
(2) employing  an emergency  procurement approach in-
    corporating limited competition.
   At this time, EPA's policy was to  use Section 311 au-
thority  wherever  possible  at hazardous  waste  sites
threatening surface waters and to undertake whatever was
necessary in effecting cleanup to mitigate the threat or dis-
charge.  Coast  Guard's 311 policy differed.  Their main
concern was to conserve the 311 fund. They believed the
best way to do that  was  to limit emergency activities at
hazardous sites to containment only. The EPA and Coast
Guard  collective experience at  Seymour Recycling indi-
cated that the two policies on  use of the 311  fund were
not only inconsistent but often in diametrical opposition.


Major Cleanup Needed

   In  July 1981, approximately three months after con-
clusion of the initial response, a special congressional ap-
propriation for Section 311 was enacted. An On-Scene Co-
ordinator made a site visit to assess the current conditions
and define the activity phases necessary to provide greater
site stabilization.
   The OSC convened a RRT meeting and  put forward  a
phased plan for complete site cleanup beginning with the
bulk  tanks, progressing through removal  of drums, in-
vestigations  for buried materials, removal of the contam-
inated soil mass and ground water recovery and  treat-
ment. The members of the RRT believed that the relation-
ship between  the  site, groundwater and nearby surface
water could only be established by completing a  ground-
water study. The Regional Response Team recommended
that while the  groundwater study was underway, a per-
manent dike and security fence should be constructed.
Groundwater Monitoring

  The OSC felt that the rapidly deteriorating conditions
at the site would not allow time for implementation of a
standard piezometric and analytic effort and  that the
relationship would need to be established in a two to four
week period. A quick turnaround approach for identify-
ing  the presence of organically contaminated groundwater
plumes entering nearby surface waters did exist and had
been applied successfully in determining rural waste water
treatment needs.(2) The  OSC felt this approach  could be
applied at the Seymour site, yielding results of sufficient
quality and clearly defining the relationship between the
site itself and  nearby surface water (Table II). Details of
the  groundwater study,  the instrumentation used and  re-
sults are discussed in detail in Part II of this paper.

-------
350    CASE HISTORIES
  The results  of the  groundwater effort  were available
within ten days. They confirmed that the~shallow ground-
water aquifer  was contaminated and contaminants were
entering the nearby surface water system.  The bulk stor-
age vessels and drums, all in a general state of disrepair
and decay, were in fact a source of an on-going discharge
to the East Fork of the White River.

                       Table n.
       Information Derived from Groundwater Study

Rate and direction of groundwater flow
Relative level of groundwater contamination
Source of groundwater contamination
Positions of groundwater inflow into nearby stream
  The instruments utilized  during  the  study included
groundwater flowmeters and ultraviolet flourescence.

CONTRACTOR SELECTION

  Having established the site as a source of contamina-
tion,  the OSC requested funds for  the removal of ma-
terials stored in the bulk vessels.  Both the OSC and the
USCG Contracting Officer agreed that negotiated pro-
curement would  provide the best insurance  for obtain-
ing the "cost effective" approach. They also realized,
however, that the normal procurement process covers a
period of time far  beyond  that considered responsive to
this situation.
  A "fast track" procurement process needed to be em-
ployed in  order  to begin removal activities  as  soon as
possible. The system ultimately developed was based on a
brief scope of work and evaluation criteria ranked in order
of priority (Table III).

                       Table m.
            Fast Track Negotiated Procurement

Distribution of solicitations
Review and ranking
Cost negotiations
Selection

  The evaluation criteria  were  distributed  as  part  of
USCG's solicitation package. All contractor proposals re-
ceived were reviewed by a Technical  Evaluation  Team
consisting of representatives from EPA, the Indiana State
Board of Health and USGC. The proposals  were ranked
based  on point totals generated  through review of the
information provided by the contractors  in response to
the Coast Guard solicitation (Table IV).

                       Table IV.
            Solicitation Requested Information.

Overall company approach
Disposal techniques
Safety precautions
Transportation specifics
Equipment to accomplish task
Pricing
  After ranking, the proposals were grouped into three
divisions:
(1)  acceptable,
(2)  unacceptable but could be made acceptable with slight
    change and
(3)  unacceptable.
Negotiations were begun with those contractors grouped
in the second division. Technical and pricing modifica-
tions  to  their proposals were requested. These  contrac-
tors were extremely responsive and modified their pro-
posals for resubmission in approximately one week. These
proposals were evaluated again (Table V) and a selection
was made.

                       Table V.
            Major Criteria Contractor Selection

Most Environmentally Sound Approach
Most Reasonable Site Safety Program
Most Cost Effective Solution

  This entire procurement process required only four and
one-half weeks to complete (Table VI) and realized a sig-
nificant savings of Federal  resources.  Final cost esti-
mates for acceptable proposals ranged from $350,000 to
$4,000,000. After being evaluated on technical and safety
terms, cost effectiveness became  the  pivotal  factor. Re-
moval of the materials in the bulk storage vessels has been
completed. Costs for that effort went far beyond the con-
tractors estimate but remained less  than the next most
cost effective alternative.
CONCLUSION

  This  procurement approach has  also been used for
two other efforts at the Seymour Recycling Center. These
activities involved a detailed hydrogeological investigation
and the other involved removal of all liquids in 55-galIon
drums on the site. Cost estimates before negotiations were
$64,000  to  $400,000 and  $800,000  to  $13,000,000 re-
spectively. The groundwater  effort has been  authorized
but drum cleanup has not because of the limited 311 fund
situation.
                       Table VI.
          Time Frame for Negotiated Procurement
Element
Solicitation Preparation
Contractor Proposal
Proposal Evaluation
Proposed Revisions
Negotiations and Final Evaluation

    Total Time to Contract Award
Time Span
(weeks)

1
2
0.5
1
1

4.5

-------
                                                                               CASE HISTORIES    351
CONCLUSION

  EPA Region V believes that the approaches used dur-
ing the 311 action at the Seymour recycling center pro-
vided valuable information on which  to  base key de-
cisions in a time frame consistent with  the true meaning
of emergency response. Defining the role of groundwater
at an uncontrolled waste site can provide the OSC with in-
valuable information on which to base an immediate or
planned removal decision. The benefits  of the negotiated
procurement process far outweighed the  necessary re-
source commitment  to effect them in terms of potential
gross savings versus the staff time needed to complete in-
tensive review and quick turnaround.
  After experiencing  the  success of  using  these  ap-
proaches, EPA Region V was particularly pleased to see
the development  of a planned removal scenario as  part
of the Superfund effort and are convinced that this ap-
proach, if managed in a timely manner, can achieve sig-
nificant public health protection quickly and cost effec-
tively.
REFERENCES

  1. Forba,  R.W., "Field Investigations of Hazardous
     Waste  Site Seymour Recycling  Company"  U.S.
     EPA,  Office of Enforcement, EPA-330/2-80-010,
     March, 1980.
  2. "Alternative Rural Wastewater Treatment Systems-
     Case Study Number Two" U.S. EPA, Region 5,
     draft Environmental Impact Statement, June, 1979.
          A FAST TRACK APPROACH TO MANAGEMENT
                           AND IMPACT ASSESSMENT (PART ID
                                  WILLIAM B. KERFOOT, Ph.D.
                                         K-V Associates, Inc.
                                       Falmouth, Massachusetts
INTRODUCTION

  On August 15, 1980,  K-V Associates, Inc. was con-
tracted  by the  U.S.  Environmental  Protection  Agency
Region V under a Section 311 action for emergency service
to determine the direction of groundwater flow and loca-
tion of any subsurface plumes of material associated with
chemical waste  materials improperly disposed of at  the
Seymour Recycling Center. The  analytical services  in-
cluded the following tasks:
•Determine the  direction of groundwater flow from  the
 Seymour Recycling site by obtaining flow data with a
 Model 10 Dowser groundwater flow meter from roughly
 20 locations of shallow augered excavations
•With the aid of existing PVC monitoring  wells, deter-
 mine the vertical profile of flow, if possible
•Clean, develop, and pump monitoring wells surrounding
 the site
•Obtain at least 20 samples of groundwater in the direc-
 tion of prevailing flow  and  along the stream shoreline
 with a well-point sampler
•Analyze water  samples in the field for conductance and
 fluorescent aromatic hydrocarbon content
•Map the direction of groundwater flow and positions of
 leachate from the site, indicating locations of inflow into
 stream, if observed.
  The Seymour Recycling Center is located on the per-
iphery of the Freeman Field Industrial Park southwest of
Seymour, Indiana. To the north of the Site a stable stream
flows directly east-west, then bends northward to  enter
Heddy Run which discharges into  the east fork of the
White River.

GROUNDWATER FLOW METER MEASUREMENT

  Although several areas of ponded water were observed
at the surface near the site, subsurface profiling showed
the water to be perched on compacted clay-loam deposits.
Augering through  the soil would  frequently  encounter
water at 8 to 9 foot depths below the surface. The inflows
occurred through  limited  veins  of  sand  penetrating the
sandy-clay deposits. The water was under some pressure
and would  enter rapidly,  fluidizing the sandy loam de-
posits until  eventually stabilizing at a depth of 6 to 7  ft be-
low grade, depending upon grade elevation.
  Initially,  a 3.5  in. diameter  probe with the  ground-
water flow meter  (Model  10 Dowser, K-V Associates,
Inc.) was inserted into the sandy deposit at the base  of
the hole. Attempts at measurement  indicated no apparent
flow. Samples of the soil mixture were removed from the
hole and placed in a 4 in. diameter  soil flume to measure
the capacity of the soil for flow. No measurable flow could
be  detected through the mixture with a two-inch  water
head even though the soil was a sandy consistency. A suf-
ficient content of clay or silt existed to effectively seal off
flow through the material.

-------
352     CASE HISTORIES
                                                          1000    2000    3000    4000    5000    6000    7000 FEET
                                                              CONTOUR INTERVAL 10 FEET
                                                               DATUM IS MEAN SEA LEVEL
                                                                                                                  OUAOIUNCU IOCAUM
                                                             Figure 1.
                                      Location of Seymour Recycling Site Showing Stream Areas
                                           and Discharge to the East Fork of White River
     depth
     O
4 drain



;j:::-;i
Grounc
Flowmet
aga pipa 1O
organic turtaca
aandy loam
sand • clay
*" ~- — — Piciomatr.c Haad— """
madium aand ' j /-\".;l
Corrfmad Aquitar .- . . .
water Well
er Probe Sam
tor meaiuramant


*. -'_• ..•-"'" ,
Point
pier
                           Figure 2.
            Typical Soil Profile Adjacent to the Site
                                                                           9

                                                                           8-
                                                                           7
                                                                    «
                                                                    «      6
                                                                    a  >•

                                                                    i!   -
                                                                    u.  g.   4
                                                                           2-

                                                                           1 •

                                                                           0
                                                                                  Down the  Well Probe
                                                                                  Seymour, Indiene
                                                                                  August  198O
                                                                                                 10

                                                                                             Readout
                                         Fin* slnd
                                         ft. O.a void frwftofl
                                         •lop* • O.37B
                                                                                                            15
                                                                                                                      20
                         Figures.
Model 10 Dowser Calibration Curve for Water Bearing Strata

-------
                                                                                    CASE HISTORIES    353
Svymour. Indian
K-V Allocates.
Filmoulh, Man
Auguit 19BO
i • Sit* Plan
Inc.
•chui«tt*

                       Figure 4.
   Direction and Rate of Groundwater Flow Near Seymour Site
  To avoid blockage of measurement by intrusion of the
 fine material, the hole was  cased down  to  the porous
 strata. The hole was augered to 10 ft and a 10 foot sec-
 tion of 4 in.  diameter drainage pipe installed. The gelled
 sand which had fluidized into the hole was bucketed out,
 allowing only washed sand from the aquifer to intrude up
 the interior of the pipe. The 2 in. down-the-well probe was
 then inserted into the sandy  material at the base of the
 casing, the casing pulled back above  the probe to  avoid
 interference with capillary flow, and the probe oriented to-
 wards north  before measurement. After initial measure-
 ment of flow, the probe was then rotated 180° and the
 measurement repeated to provide  a check on the first de-
 termination. The results of field measurements taken  from
 cased holes surrounding the Seymour site are given in
 Figure 4.
  The  Model 10 Dowser flow meter  was  calibrated for
flow rate against the Seymour sandy aquifer material ob-
tained from cased holes I and J  placed in a 4 in. cylindri-
cal flow chamber. Using a peristaltic pump, water of
known volume was circulated  through  the chamber.  With
an independent measurement  of porosity of the soil, the
rate of movement in feet/day could be calculated. A cal-
ibration curve for the sandy water-bearing strata is plotted
in Figure 3.
GROUNDWATER SAMPLING

  Samples of groundwater were obtained with a Model 12
miniature well-point sampler (K-V Associates, Inc.). Two
procedures  were followed. With wells AA through J,  the
well-point sampler was inserted into hand-augered, cased
holes (with 4 in. PVC draining pipe) and thrust through
the sandy bottom to 2 ft. below the casing.
  With uncased holes  or cornfield sampling,  the  well-
point was  driven directly through  the  soil to approx-
imately four feet below standing  water level (piezometric
head).  A high ratio of striking force to crossectional area
of the  shaft (5/8 in.  diameter) enabled the well-point to
penetrate the sandy  or clayey subsoil. Water  was then
withdrawn  with the sampler  into  a silt trap, washed out,
and refilled. A sample of the second pumping was taken
for analysis by a field fluorometer and later scanning by
a  Perkin-Elmer  Model  204  fluorescence  spectrophoto-
meter.
  The  conductance  of groundwater and surface water
samples was determined with a Model 1484-10 conductiv-
ity meter (Horizon Ecology Co.). Samples were obtained
for cross-calibration in the laboratory.

SURFACE WATER TRANSECT

  On August 20, surface water samples were taken along
the stream course from  west to east,  progressing  up-
stream. These samples were analyzed  immediately  for
fluorescent  aromatic content. The stream flows through a
2 to 3 ft wide channel as it meanders through a 30 ft wide
man-made  ditch between the cornfield and pastureland.
Grass covers  the sides  of the stream bank in unshaded
areas.  The  sandy stream bed is  soft in  numerous loca-
tions, apparently caused by upwelling waters from ground-
water recharge.
  Two increases in fluorescent organics were  observed,
one near Sample 2 at the  opening of the storm drainage
ditch from the Seymour  site and in the vicinity of sur-
face water Sample 7 taken near groundwater Sample R12.

FLUORESCENT ANALYSIS OF WATER SAMPLES

  Fluorescence spectroscopy is well-known as an analyti-
cal tool for  organic analysis because of its exceptional
sensitivity and  selectivity.(1)  The U.S. Coast Guard  Re-
search and Development Center recently conducted a sur-
vey of the luminescence of hazardous materials/2' A large
number of hazardous organic compounds were identified
as possessing useful  fluorescence. Of 113 materials sup-
plied to the Coast Guard study, 96 compounds were fluor-
escent  at room  temperature, although some admittedly
weak.  The stronger  fluorescent  compounds include  the
benzene and naphthalene  derivatives common  to petrol-
eum products, numerous pesticides  and  herbicides, com-
mon solvents such as toluene and  xylene, dye products
such  as aniline, and the  polyaromatic  hydrocarbons
(PAH's) such as anthacene, chrysene, and pyrene.
  Fluorescence has recently found increasing use as a tool
in the  analysis of complex mixtures of organics.  Frank(3)

-------
 354    CASE HISTORIES
•£  300
 a
 u
 x
 UJ
        Room Temperature Fluorescence
        of 95 Toxic and Hazardous  Materials
                     •- •<
       250          300          350          400
               Emission  Wavelength (nm)

                       Figures.
  Excitation-Emission Plotting of Fluorescence of 95 Toxic and
 Hazardous Materials. Region of Sensitivity of Field Fluorometer
                    Indicated by Box
        Room Temperature Fluorescence
        of 95 Toxic and Hazardous  Materials
       250          30O          350          400

               Emission  Wavelength (nm)


                        Figure 6.
   Zones of Sensitivity of Fluorometer Scans of Water Samples
used excitation of samples at 290 run and quantification of
emission at 329 nm as a method of characterizing crank-
case oils in situations not amenable to analysis by  gas
chromatography. Quantification of number 4 oil directly
in water has been studied by Frank*4' for analysis of oil
contamination.  More  recently synchronous  scanning,
accomplished by linking the excitation and emission mon-
ochromators together, has gained in popularity because of
the elimination of scatter interferences.(5>
  Field water samples were analyzed by excitation at  254
nm and emission from 280 nm to 400 nm. Upon return to
the laboratory, the same samples were analyzed by three
different scanning methods: excitation at 290 nm and emis-
sion scan from  310 to 450  nm (a procedure similar to
Frank01 for crankcase oil), and synchronous scans (30  nm
and 50 nm separation). Results from the four methods are
superimposed over a two-dimensional plotting of the peak
locations of fluorescence of hazardous substances deter-
mined previously by the Coast Guard in Figure 6.

Results

  The organic analyses  yielded  the following  informa-
tion:
(1)  Between  cased  holes BB and  D a  substantial con-
    centration of fluorescent aromatic organics occurred.
    All  scanning procedures revealed a  similar pattern
    with the highest  concentrations coinciding with holes
    BB  and D.  The variety of scans also  revealed two
    fluorescence peaks dominating the elevated regions:
Wave Length (nm
Excitation/Emission
290/324
335/365
Class of Organic Compounds
phthalic acids, naphthalene
anthracenes, plasticizers, light oils
    Previous gas chromatographic analyses of well water
    samples had  identified  naphthalene, diethylphalate,
    ethylbenzene, toluene and tetrachloroethylene exterior
    to the storage site.(6)
(2)  Similarly, groundwater samples downflow of the site
    along a transect through the cornfield Rl through R17
    revealed a rise  above background from Samples R7
    through R16.
(3)  Surface water samples from the east-west creek indi-
    cated inflow of fluorescent aromatic organics in the vi-
    cinity of the drainage ditch confluence. The regions of
    infiltration  were strongest in the vicinity of surface
    Samples 2 and  7 coinciding with the drainage ditch
    and the central portion of the subsurface plume.
(4)  During a heavy thunderstorm, a substantial inflow of
    fluorescent  aromatic organics  was observed issuing
    from the drainage ditch to the east-west creek. The in-
    flow  was readily observable as a discernible  pulse en-
    tering Heddy Run.
  Based  on these results, one can conclude that aromatic
organics  from the Seymour  site are entering the east-west
stream-Heddy Run confluence through two pathways:
(1)  storm runoff through the western drainage ditch and
(2)  groundwater infiltration into the east-west stream and
    from a large subsurface plume containing  materials
    originating  from a primary source at the site (as seen
    in holes AA through D)  and a secondary source as in-
    filtration into the aquifer of storm drainage from con-
    taminants in the western  drainage ditch.

Storm Runoff

  On August 19, a  violent thunderstorm occurred at 4:30
p.m., forcing abandonment  of groundwater sampling. In-
stead, the leachate detector  was moved to the confluence
of the east-west stream and  Heddy Run. Samples were re-
moved from the stream from 4:50 p.m. to 6:00 p.m. at 15
min. intervals.  At 5:30 a large surge of fluorescent aro-
matic compounds was observed at the confluence. At 6:15
samples were obtained from the outflow of the west drain-

-------
                                                                         CASE HISTORIES     355
                     5678     9 ^10    11     12
                            \  /'   ,'
                 drainage ditch   /.\
                              v   :
                                          Seymour  Recycling
Seymour. Indiana •  Site Plan
K-V Associates, Inc.
Falmouth, Massachusetts
August   1980
500
                                       Figure?.
                     Sampling Locations and Position of Observed Plume

-------
 356    CASE HISTORIES
                  Surface Water Samples - E-W Stream
                   Field Fluorometer Scanning Analysis
               Relative  Fluorescence
               50 nm  synchronous  scan
                    leachate from storm drainage ditch
          1   23  4   5   6  7   8  9  10  11
                         Figures.
    Fluorescent Aromatic Concentrations Observed in Surface
                 Water of East-West Stream

                 Well-point Samples - Cornfield Transect
100-
90-
80-
70-
60-
S50-
U
• 40-
o
3
I 30-
20-
10-
0











I
Field Fluorometer Scanning Analysis

































..
        17 16  15  14 13 12  11  10  9  8  7  6  5  4  3  2   1

                        R  aeries

                         Figure 9.
  Concentration of Fluorescent Aromatic Found in Groundwater
          from Cornfield Transect Adjacent to Stream
age ditch from the Seymour site and immediately upstream
of the outflow along the east-west stream. Analysis of the
water samples confirmed that the bulk of the discharge was
issuing from the vicinity of the drainage ditch.
STRESSED VEGETATION

  On edges of the cornfields,  trees at various heights oc-
curred and could be seen to be  under vegetative stress. The
position of various dead and  stressed trees were plotted.
Most were in the vicinity of the western drainage ditch,
although a few trees  occurred upstream along the east-
west stream.  The corn plants  in the region would not be
affected by groundwater contamination since their  root
systems are shallow and rest in the sandy loam above the
aquifer.  However, taller trees (  >20 ft) have roots penetrat-
ing into the  lower  strata which  may contain the trans-
ported organics.


0
u
CB
O
CO
4)
O
!Z
0
ra
0
QC



20-
10-
0
h"H" series
• 1 • .1 .. 1
17 16 15 14 13 12 11 10 9 8 7 6 54 3 2 1
100-i


90-


80

70-
60-
50-

40-
30 •
20-
10
O













. 1
Illllll,



























II....
                                                                       Z YXWVUTSBBCCDDABC DE FGH I  J
                                                                                       Station

                                                                                       Figure 10.
                                                                Relative Fluorescence of Groundwater Samples Observed with
                                                                           50 nm Synchronous Scan Procedure

                                                                                      Table I.
                                                                           GWF Directions & Relative Rates
                                                                         Seymour, Indiana      August, 1980

Hole
A
B
C
D
E
F
G
H»
I*
J
p..
AA
BB
CC
DD
R17
R14
R15
Compass
Direction
338°
340°
327°
310°
352°
347°
315°
u
u
344°
Unstable
320°
334°
346°
335°
311°
335°
23°
Rate
(Readout)
3.5
2.5
6.5
2
4.5
4
3.5


6.5

6
2
3.25
2.5
1.5
3.25
2.75
Velocity
(ft/day)'
1.3
1.0
2.5
.8
1.7
1.5
1.3


2.5

2.3
1.8
1.2
1.0
.6
1.3
1.0
  •H and I were unstable in flow characteristics and appeared to be in-
  fluenced by a possible leak in a water pipe or other localized source.
 •*P was  taken following rainfall and showed erratic flow (towards
  83 °E).
•••0.375 x readout = ft/day.
  u = Unstable.

-------
                                                                                  CASE HISTORIES    357
               Relative  Fluorescence
               30 nm synchronous scan
        20-

        10-
           17 16 15 14 13 12 11 1O 9 8  7 6 5 4.3 2  1
100


 90


 SO


 7O


 60-


 50


 4O-


 3O-


 20-


 1O


 O
•lllllL
            ZYXWVUTSBBCCDDABCDEFGH I J

                     Station
                    Figure 11.
Relative Fluorescence of Groundwater Samples Observed with
         30 nm Synchronous Scan Procedure
           Conductance   micromhos/ cm.
      800
      400-
          17 16 15 14 13 12 11 10 9 8 7 6 3 4 3 2 1
     2400

   o
   o
   £2000
   o 1500
   U
     1200
     800-
     400-
         'ZYXWVUTSAABBCCDDABCDEFGHI J

                    Station
                    Figure 12.
        Conductance of Groundwater Samples
STREAM INFILTRATION

  The shallow (0.5-2.5 ft deep) east-west stream formed
hydraulic  continuity with the shallow aquifer. The water
level of the stream on dry days was roughly 5.5 ft below
the grade of the cornfield. This placed the sandy bottom of
the stream at the top of the porous water-bearing strata.
A hole was augered near Station Two to the  aquifer and
the difference in water  head compared by  inserting  a
primed hose between the hole and the shallow stream. The
observed flow of water was rapidly towards  the stream,
clearly indicating infiltration of groundwater  up into the
stream.
  To estimate the rate of inflow, Lee-type lysimeters were
installed in the stream bed to determine the rate of ground-
water infiltration in the vicinity of the plume intersection.
                                                      Site
                                                      1
                                                      2
                                                      3
                                                      4
           Flow Rate
           246 ml/10 min
           145ml/15min
           60 ml/30 min
           72 ml/20 min

              X   =
Infiltration (ftVftVday)
0.67
0.26
0.05
0.10
                                                                                  0.27
  The section under investigation averaged 2.6 ft in width.
With  an estimated  0.27  ftVftVday infiltration  along  a
projected zone of length of 800 feet of stream bed inter-
cepted by the plume, the rate of discharge would be about
560 ftVday or approximately 4200 gal/day.

WELL DEVELOPING

  On August 22, existing observation wells were pumped
for developing. The  deep well occurring close to the drain-
age ditch south of  the indicated position of  well 9 was
well-developed with no fines  apparent  upon successive
pumping. Wells 1,  5,  and 6 were found to penetrate to
10 ft below the top of the casing. Wells 1 and 5 were devel-
oped  by drawdown and  surging with 5  gal/mm centri-
fugal  pump with hose to bottom of the casing. Well 6 was
drawn down to the bottom of the screen  with little or
no apparent inflow. After two hours, only 2 in. of water
had filled the well screen. This well did not appear cap-
able of being developed. Well 5 had sustained slight dam-
age due to earth  operations in the vicinity. A 20° bend
was apparent 2.5 ft below casing top (one ft below grade).

CONCLUSIONS

  From August 16  through August 23,  1980, K-V Asso-
ciates, Inc. investigated  the  extent  of subsurface con-
tamination of groundwater in the immediate vicinity of
the Seymour Recycling  Facility in  Seymour,  Indiana.
Hand augering on  the site revealed  that water-bearing
strata existed at depths of 8-9 ft below grade.
  Thirty shallow observation holes were excavated across
the site area and  used  to measure the direction  and rate
of groundwater flow. After the major direction of ground-
water flow  was determined, samples were taken of the

-------
358     CASE HISTORIES
groundwater  across  the  axis  of flow with  a  portable
well point sampler driven 14 ft below grade to determine
the location of any material moving through the aquifer.
Samples were analyzed for conductivity and petroleum-re-
lated aromatic organics (by UV fluorescence).
  The following conclusions were reached:
•The mean groundwater  flow direction was  found to be
 away from the town of  Seymour and in a Northwester-
ly direction (approximately  316° NW) evidenced by direct
 flow measurement and  the location  of a plume of or-
 ganic leachate.
•A plume of petroleum-related aromatic hydrocarbons
 was observed extending from the  site in a Northwest-
 erly direction discharging  into surface waters at a con-
 fluence of an east-west stream and  Heddy Run South of
 Hangman Crossing.
•The position of the subsurface plume coincided with a
 zone containing numerous dead or dying trees,  partic-
 ularly larger trees with deeper root structures.
•The mean flow  rate of the shallow  aquifer was  meas-
 ured at 1.2 feet per day compared to a previous estimate
 of 0.016 ft/day.  The combined influence of siphoning of
 flow to the nearby streams and restriction of lateral flow
 strata by natural clay deposits probably contribute to the
 more rapid flow  than previously estimated.
•During storm period temporary changes in flow of the
 streams  (direction) were observed in the aquifer flow
 when storm  flooding occurred, further substantiating the
 connection between both water bodies.
•Since the shallow aquifer is intercepted  by the  stream
 bed of Heddy Run and in places  by an  east-west trib-
 utary,  it is possible that the plume may have stabilized
 in position and is no longer moving laterally. Instead the
   90-
   80-
   70-
 0
 £60'
 0
 »50-
 0
   30

   20

   10
               Storm Runoff Sampling
               50 nm Synchronous Scan
         Heddy Run Confluent
                                     Sources
                                    \
III
1
Time
          18   19  20  21  22  26
              «  o  
  «6  «D  
-------
             HAZARDOUS WASTE SITE INVESTIGATION
          SYLVESTER SITE, NASHUA, NEW HAMPSHIRE
                                         JOHN J. GUSHUE
                                   GHR Engineering Corporation
                                    New Bedford, Massachusetts

                                          JOHNE.AYRES
                                   Goldberg-Zoino and Associates
                                 Newton Upper Falls, Massachusetts

                                         ALVINJ. SNYDER
                                 Environmental Resource Associates
                                       Warwick, Rhode Island
INTRODUCTION

 In this paper, the  authors describe the results of an
engineering and hydrogeological investigation of the Syl-
vester Site undertaken by GHR Engineering Corporation
and its principal subcontractor, Goldberg-Zoino and Asso-
ciates (GZA). The investigation began in April  1980 and
was completed in July 1981 with submittal of the final
project report to the  New Hampshire Water Supply and
Pollution Control Commission (NHWS & PCC) and the
U.S. Environmental Protection Agency (U.S. EPA.)(1)
 The investigation of the Sylvester Site and adjoining
properties was designed to:
(1) Define the extent  of soil, surface water and ground-
  water contamination;
(2) Develop an engineering assessment of the hydrogeo-
  logical and hydrological conditions governing contam-
  inant movement from the site; and
(3) Evaluate alternative remedial actions on the basis of
  technical  feasibility,  estimated  costs  and  environ-
  mental implications.
 The evaluation of remedial actions included  a labora-
tory-scale assessment  of the feasibility of removing  con-
taminants from groundwater in an on-site, above-ground
water treatment plant. The final  project  report, includ-
ing recommendations for  remedial action, provided the
basis for the first  Federal/State  cooperative agreement
under Superfund, with $2.3 million being awarded to the
State of New Hampshire in August 1981  for design and
construction of remedial measures at the site.

SITE DESCRIPTION

 The Sylvester Site  is an approximately six-acre open
dump located behind  the C & S Disposal Company gar-
age on Gilson Road in Nashua, New Hampshire. The site
is a former sand pit which was excavated in places to ele-
vations lower than seasonal high groundwater levels. Upon
completion of sand mining operations, the pit was grad-
ually filled  during the 1970s with  various  types of refuse
and waste materials, consisting primarily of construction/
demolition debris, but also including large  amounts of
chemical liquids and sludges. The waste materials extend
into the water table throughout much of the site. Sev-
eral methods were used to dispose of hazardous liquid and
sludge wastes at the site, including:
•Subsurface  discharge of large volumes of liquid wastes
 through 2-3 ft diameter pipe extending into the site from
 the garage building
•Subsurface burial of drums of waste in at least two loca-
 tions on the site
•Surface and subsurface disposal of sludge-like material
 in several areas of the site
•Stockpiling  of 55-gal drums of chemical wastes on the
 surface of the site
  In May  and June 1980, all drums visible  and access-
ible from the surface of the site were removed by Recra
Research, Inc. of Tonawanda,  New York under subcon-
tract to GHR Engineering. A total of 1,314  drums were
transported from the site for final safe disposal in New
York and Ohio. A complete report on the drum removal
work was submitted to the NHWS & PCC in July 1980.(2)
  The  area surrounding the site is primarily residential
(Figure 1), although the land across Gilson Road  to the
south is undeveloped swamp and woodland.  Directly ad-
jacent to the  east and north of the site are two large trailer
parks. The closest residences, in the Rodger's Trailer Park,
are within about  100 ft of the northeast perimeter  of the
site. The Thompson property, which abuts the site  to the
west, is a large, open parcel that is an active  local source
of sand and gravel.
  As indicated on Figure 1, contaminated groundwater
from the site  flows northwesterly toward Lyle Reed Brook,
which is about 680 ft from the site and the Nashua River.
The contaminant plume from the site has already moved to
and beyond Lyle Reed Brook. Fortunately, the residents of
Jensen's Trailer  Park are not  dependent  upon local
groundwater resources for drinking water, since the trailer
park is served by the municipal system.
  Not  so fortunate are residents along Route 111  in the
path of groundwater flow. Their individual drinking water
wells are in  jeopardy,  i.e., they will eventually be con-
                                                 359

-------
360    CASE HISTORIES
laminated as the plume continues to expand. Trace levels
of organic contamination have already been found in  a
ground-water observation well located between Tumble-
brook Road and the Pennichuck Water Department well,
which is not presently in use. In addition to drinking water
supplies, a principal public health concern at the site is the
Lyle Reed Brook into  which contaminants  moving  in
upper groundwater zones are being discharged,  eventually
being carried in surface flows to the Nashua River.
FIELD EXPLORATION AND TESTING PROGRAM

  Upon completion of drum removal work in mid-June
1980, an extensive field exploration and testing program
was carried out to develop the information needed to eval-
uate remedial options for  the site. The complete explor-
ation and testing program was a multi-phased effort that
consisted of the following:
•Subsurface investigations of the disposal site by remote
 sensing  (magnetometer and partial electrical resistivity
 surveys)
•Excavation of 34 test pits within the confines of the dis-
 posal site,  which included soil sampling and analyses,
 groundwater  sampler  installations,  and  stratigraphic
 logging.
•Execution of 15 test borings  with soil sampling,  rock
 drilling and field analyses  for organic chemical  contam-
 ination
•Installation of permanent groundwater sampling systems,
 both  conventional  observation wells   and  gas  drive
 samplers, at 23 locations in the vicinity of the  disposal
 site (Figure 2)
•Installation of a 1.0 ft diameter well and execution of a
 5-day aquifer pumping test.
  The pumping test was conducted to:
(1)  refine estimates of the hydraulic characteristics of the
    aquifer;
                                                          B  indicates private well in jeopardy
                                                                                                Rodger's
                                                                                              Trailer Park
                                                   Figure 1.
                                       The Area Surrounding the Sylvester Site
                                                 (Not to Scale)

-------
                                                                                     CASE HISTORIES    361
(2) observe the impact of pumping on groundwater qual-
   ity; and,
(3) collect a representative sample of the aquifer for use in
   laboratory-scale testing of the effectiveness of several
   alternative water treatment processes in renovating the
   contaminated groundwater.
  Groundwater quality in the vicinity of the Sylvester Site
has been monitored periodically since July 1979, beginning
with the installation  of  a ring  of 10 groundwater  ob-
servation wells around the site by the NHWS & PCC.
During  the GHR/GZA  site investigation, another 27
groundwater observation  wells were added to the monitor-
ing network.  In addition, 42 private drinking water wells
and 3 Pennichuck Water  Department wells (1 production
well, 2  observation wells) have been tested by the U.S.
EPA as part of this site investigation. Surface water qual-
ity in Lyle Reed Brook and the Nashua River has been
monitored  by the  U.S. EPA on  several  occasions since
March 1980.

FINDINGS: CONTAMINATION
LEVELS AND DISTRIBUTION
Types and Amounts of Contaminants

  The site and its associated groundwaters are highly con-
taminated with a large variety of toxic organic and inor-
ganic compounds.  The groundwater contaminant plume
extends from the disposal site to the vicinity of the Penni-
chuck Water Department well between Tumblebrook  and
Trout Roads.
  The most contaminated groundwater zone extends from
the disposal site to  the vicinity of wells #5-D and #M-2 on
the Thompson property;  however,  some mobile volatile
organic compounds have travelled  as far as the Penni-
chuck well. Several of these compounds, including primar-
ily toluene and tetrahydrofuran (THF),  have been con-
sistently detected in Lyle  Reed Brook in the 5 to 10 mg/1
range since July 1980. The less mobile extractable organic
and inorganic compounds known to be present in  the
groundwater contaminant plume have not  been detected in
Lyle Reed Brook to date.
  A tabular summary of the types  and amounts of con-
taminants  detected in groundwater and surface water
monitoring since July 1979 is presented in Table 1.  The
data in this table provide an indication of the extent of
the adverse water quality  impacts of the Sylvester Site and
a general perspective on observed  contaminant migration.
However, the very dynamic  temporal  and spatial char-
acteristics of the contaminant plume and its movement
are oversimplified here. The ranges of the pollutant con-
centrations presented in Table I are the lowest and high-
est levels detected to date at sampling stations grouped as
Mows:
(1) Groundwater monitoring wells nearest  to the down-
   gradient perimeter of the Sylvester Site, including wells
   #2A, #3, #4, #M-1 and #A-4
(2) Groundwater monitoring wells located between the site
   and  Lyle  Reed Brook, including wells  #A-2,  #A-3,
   #M-2 and #5-D. In this area, only well #A-3 has been
   analyzed for extractable organics
  The principal groundwater  contaminants in terms of
concentrations  are  volatile  organic  solvent compounds
which, in total, have  been found  at  levels  as high as
approximately 1,800 mg/1. Concentrations of non-volatile
organic priority pollutants in groundwater have been in the
5 to 15 mg/1 range. No priority pollutant pesticides or
PCBs have been detected in groundwater. Arsenic levels
in groundwater are  elevated (within the 0.20 to 1.2 mg/1
range) throughout much of the study area. Concentra-
tions of total metals in groundwater are within the 100 to
1,000 mg/1 range in the  study area,  with iron and man-
ganese being by far the major inorganic constituents.
  Volatile organic contaminants in groundwater from the
Sylvester Site were first detected in Lyle Reed  Brook in
March 1980 and have been steadily increasing since. Con-
taminant levels  in the brook in September and December
1980 were in the 7 to 10 mg/1 range in the vicinity of Tum-
blebrook and Trout Brook Roads, and  in the 2 to 3 mg/1
range at the Route 111 culvert.  The decreasing concen-
trations along the brook may  reflect the effects of dilu-
tion, aeration, volatilization and biodegradation.
  No contaminants other than  volatile organics have been
detected  in the brook to date.  The Nashua  River  has
been tested by  the U.S.  EPA at a  point approximately
1,000 ft downstream of where  the brook  enters  the river,
and no volatile contaminants  present in  the brook have
been detected in the river at that point.

Contaminant Transport Mechanisms
  The disposal  site  itself is a depression  carved into  the
natural landscape as a  result of sand mining operations.
The subsequent waste disposal activities have left a very
irregular surficial topography within the approximately 6-
acre  area now  enclosed  within a chain link fence. The
ground surface  immediately  outside  the fence consists of
steep  embankments to higher ground  to the  west and
north, and relatively flat areas to the east and south.  All
surface runoff in the immediate vicinity of the site  flows
into the site; therefore, surface runoff  is  not a  pollutant
transport mechanism from the site itself. Surface trans-
port of pollutants in the study area comes into play as con-
taminated groundwater is discharged to Lyle Reed Brook
and then flows in the brook to the Nashua River.
  The characteristics of  the groundwater flow  regime in
the study area will ultimately determine the fate of pollu-
tants  released into the  aquifer underlying the site, and it
is apparent that the principal threats posed by the  Syl-
vester Site are associated with  groundwater. For this  rea-
son,  the hydrogeologic properties of the aquifer,  along
with the characteristics of the flow regime, are of primary
importance.
  Groundwater contours developed from water table mea-
surements made in  December  1980 are shown  in Figure
2. Assuming groundwater flow to be perpendicular to the
contours presented, the principal direction of flow is to the
northwest toward Lyle Reed Brook. Regional flow is  also
in a northwesterly direction toward the Nashua River and
it is assumed that the Nashua River is the  ultimate "sink"
for groundwater leaving the disposal site. A portion of the
flow reaches the river via Lyle Reed Brook and the  bal-

-------
362     CASE HISTORIES
ance remains as groundwater, probably flowing in a path
roughly parallel to but beneath the brook.
  Hydraulic gradients (taken as  the slope of the water
table) across the  site range from less than 0.2% to ap-
proximately 0.8%,  but typically average 0.3 to 0.4%.
Anomalies in  the gradient values as well as the overall
shape of the groundwater contours  were  consistently ob-
served in the vicinity of wells #A-2 and #M-2. Steepening
of the water table slope at this point is potentially indica-
tive of a localized decrease in aquifer transmissivity, poss-
ibly caused by a geological anomaly. Seasonal variations
of the water table elevations observed at the site during the
duration of the current study indicate that seasonal fluc-
tuations will not have a significant impact on the ground-
water flow regime or, correspondingly, on contaminant
migration.
  Based on a combination of pumping test data, soil grain
size analyses and  published information for the soil types
encountered, the  hydraulic  constants of the aquifer were
estimated. The estimated transmissivity of the fine to med-
ium sands underlying most  of the site is 700 to 2,400 ft2/
day, representing a coefficient of horizontal permeability
of 20 to 70 ft/day (based  on a saturated aquifer thick-
ness of 33 ft). For sand and gravel deposits encountered
north and west  of the disposal area, transmissivity is
estimated at 3,300 to 6,600 ftVday; corresponding co-
efficients of permeability are from 100 to 200 ft/day. The
average linear or "transport" velocity of the groundwater
at the site was estimated to be 0.8 ft/day for the fine to
medium sands and approximately 1.6 ft/day in the areas
north and west of the disposal site, where coarser de-
posits would be expected to control groundwater move-
ment.
  Based on the calculated transport velocities, travel times
were estimated between selected points for transport of
contaminated groundwater originating from the Sylvester
Site. A summary of transport times from the midpoint of
the disposal area to relevant points is tabulated below:
Location
Lyle Reed Brook
Pennichuck Well
Route 111
Nashua River
Nashua River (from Pennichuck
well)
Transport Time
3-5 years
4-6 years
5-7 years
7-9 years
2-5 years
  4
             1. XW to Nllhiu Rlv»r   *.
Plum* Llrnlli ^j /
* " 1 I CIS Olipoul
/ Coapany Cerage
\l
Cltson Road * •lA-fi
	 	 — 	 . 	 L 	



^
\
\
i
/
a
O Multllwd Wdl
^ „ • W.llpolnl W«ll
+ Pinplng Will
C Property Ura
                                                     Figure 2.
                                     Monitoring Wells, Groundwater Contours and
                                      Contaminant Plume Limits at Sylvester Site

-------
                                                                                      CASE HISTORIES     363
 Since the calculated travel times were based on a num-
ber of simplifying assumptions, they are considered to be
approximations  only.  Total  volumes  of contaminated
groundwater within various portions of the contaminant
plume were estimated from assumed porosity values and
observed saturated aquifer thickness. Considering the en-
tire plume from the disposal site to Lyle Reed Brook, the
total  volume of affected groundwater is approximately
60 million gallons. Of this total, about 30 million gallons
lie within the immediate confines of the Sylvester Site as
delineated by the fenced area. The total  estimated flow
of groundwater  through the disposal site, considering a
cross-section across the width of the containment plume
at multilevel well #M-1,  is approximately  31,500 gal/day
(with a range of from 13,000 to 65,000 gal/day).

Contaminant Distribution

  At the Sylvester Site,  the wide  area  over which  waste
disposal  occurred has caused  the  formation  of a corres-
pondingly wide contaminant plume in the vicinity of the
source (see  Figure 2).   As the contamination migrates
northwest, the geometry  of the water table contours tends
to counteract the spreading effects of hydrodynamic dis-
persion, thereby limiting the width of the plume. The re-
sult  is a  fairly  sharply  defined  zone of contamination
whose southerly and easterly borders are confined to the
immediate area of the site. As the plume approaches Lyle
Reed Brook, a degree of spreading is inevitable as much
of the contaminated groundwater flows into the brook,
leaving the site as surface water. The balance of the plume
flows beneath and parallel to the brook.
  Interpretations of contaminant distributions at the Syl-
vester Site based on most recent analytical data are shown
in Figures 3  and 4.  Contours  of  pollutant concentra-
tion depicted  on the figures are  approximate,  idealized
and  subject to change with time, but the representations
adequately illustrate significant trends.
  The observed  distribution of volatile organic chemicals
(Figure 3) closely resembles theoretical modeling of the mi-
gration of a contaminant  from a continuous point source
in a uniform flow field. The point source appears to be
centered  in the vicinity of the subsurface leaching trench
found at  the rear of the garage.  From this point,  lines
of equal  concentration form elliptical lobes which  extend
northwest in the direction of groundwater flow. Slight dis-
              10' to Nashua River   \
            \
                                                     Figures.
                                  Distribution of Total Volatile Organics in Groundwater
                                       at Sylvester Site (in PPM, December 1980)

-------
364    CASE HISTORIES
tortions of the contours are observed in the central and
southern portions of disposal  area and are probably the
results of localized dumping incidents.
  Within the main  lobe of the contaminant plume,  con-
centrations  of the major constituent,  tetrahydrofuran
(THF) are often greater than  1,000 mg/1  For the other
principal  volatile  components of the  waste, including
toluened, methylene chloride,  and MIBK,  corresponding
concentrations are generally one to two orders of magni-
tude lower. The distributions of these compounds and the
additional volatiles  detected typically display more irreg-
ularity than the corresponding distribution of THF. Local-
ized "hot spots" or highly contaminated zones were ob-
served for various  compounds in test  pits  and  observa-
tion wells on the premises of the site, probably indicative
of "slugs" of pollutants percolating into the groundwater
from the unsaturated zone.
  The distribution of metallic compounds  (Figure 4) ob-
served in  the vicinity of the  Sylvester  Site significantly
differs from the pattern displayed by the volatiles. Marked
lateral distortion of the theoretically   elliptical concen-
tration contours indicates either  a large source area or
multiple point sources. Highest concentrations  of contam-
inants were observed in the central portions of the disposal
area,  suggesting that the primary source of the metals is
not the liquid wastes from the leaching pit.
  Although significant attenuation was observed with dis-
tance from  the site, concentrations of  total metals sig-
nificantly higher than background values were  detected
up  to 500 ft from the disposal area.  Since metallic com-
pounds are typically less mobile in groundwater than vol-
atiles, this would suggest that the source of the metals is
significantly older than the source of the volatiles.

Summary of the Extent of Contamination

  Before discussing the various remedial alternatives, sev-
eral important findings relative to the extent of the con-
tamination problem should be summarized. While signif-
icant  adverse  environmental  effects, i.e., degradation of
surface  water  and  groundwater quality,  have  already
occurred in the study area, the worst is yet to come.
  The groundwater  contaminant plume  is moving from
the site  toward the Nashua River at an approximate rate
of from 0.8  to 1.6 ft/day. The leading edge of the plume
extends to the vicinity of the unused  Pennichuck Water
            1. 900' to N.ihu. River
       /   *
                             ThOBpson
                             Property
                                                                                                   • lA-I
            \
1
1
1
t CIS Dlipojal
Company Girage
Cllwn Raid I • •*-«



'\\
V
1
1
1
' 1
O Multilevel Well
^ „ • Wellpohil Well
^ Puieplng Well
t Properly Line
                                                    Figure 4.
                                     Distribution of Total Metals in Groundwater at
                                        Sylvester Site (in PPM, October 1980)

-------
                                                                                 CASE HISTORIES      365
                                       Table I.
                  Summary of Types and Amounts of Contaminants
                        in Groundwater and Lyle Reed Brook
COMPOUNDS /PARAMETERS
A. Volatile Orsanics (ppb)
Vinyl Chloride*
Chloroethane*
Methylene Chloride*
1 , 1-Dichloroethane*
t-1 , 2-Dlchloroethylene*
Chloroform*
1 , 2-Dlchloroethane*
1,1. 1-Tr Ichloroechane*
Trlchloroethylene*
Tetrachloroethylene*
Benzene*
Toluene*
Ethyl Benzene*
Xylenes
Tetrahydrof uran (THF)
Methyl Echyl Ketone(MEK)
Methyl Isobutyl Ketone(MIBK)
Acetone
B, Extractable Organlcs (ppb)
2- Chlorophenol*
2 , 4-D imethylphenol*
2-Nitrophenol*
Pentachlorophenol*
Phenol*
o-Cresol
m-Cresol
Benzole Acid
1 , 2-Dichlorobenzene*
1 , 4-Dichlorobenzene*
Naphthalene*
Bls(2-ethylhexyl)Phthalate*
Dl-n-Butyl Phthalate*
Dlethyl Phthalate*
Dimethyl Phthalate*
Pesticides*
Polychlorlnated Blphenyls*
C. Inoraanlca (ppn)
Arsenic*
Barium*
Cadmium*
Chromium(total)
Lead*
Marcury*
S.l.nlum*
Sllv.r*
Coppar
Iron
Managanaaa
Nlckal
Zinc
D. Other Pararatttrt
pH
Specific Conductance (uMHOi)
Total Organic Carbon (ppm)
Chemical Oxyjen Dimand (ppn)
Total Phanollci (ppn)
LEVELS FOUND IS WELLS LEVELS FOUND IN WELLS LEVELS FOUND IN
NEAREST TO THE SITE"' BETWEEN SITE AND BROOKb' LYLE REED BROOKC '

ND to 950
ND to 320
ND to 47,500
ND to 210
ND to 5,000
ND to 1,600
ND to 890
ND to 1,400
ND to 2,700
ND to 1,600
ND to 2,800
< 10 to 100,000
ND to 2,700
SD to 1,000
ND to 4,300
44,000 to 1,200,000
ND to 7,000
ND to 33,250
ND to 48,000

ND to < 10
ND to 40
ND to 15
ND to < 10
ND to 2,700
ND to 1,137
ND to 864
ND to 320
ND to 88
ND to < 10
ND to 45
ND to < 10
ND to < 10
KD to 47
»D to UO
	 ND In this area 	
	 ND In this area 	

.04 to 1.7
< .1 to 2.5
* .005 to .01
.01 to .7
< .01 to .5
< .0002 to < .001
< .01 to < .01
.01 to .02
< .01 to .6
18. to 640.
2.8 to 115.
* .02 to .9
< .003 to 20.5

5,9 to 6.8
250, to 4,092,
12. to 7,600.
711. to 19,300,
.298 to 18. S

	 ND In this area 	
	 ND in this area 	
ND to 122,500
ND to 15
ND to 18,000
ND to 31,000
ND to 7
ND to 2,000
ND to 15,000
ND to 570
ND to 3,400
< 100 to 29,000
ND to 1,100
ND to 1,200
ND to 10,000
4,400 to 1,500.000
ND to b.OOO
ND to 13,000
ND to 36,000

— ND in OW A- 3—
	 ND in OW A- 3 	
— ND in OW A-3 	
— ND in OW A-3—
6,464
1,969
3,672
6,011
— ND In OW A-3 	
	 ND In OW A-3 	
	 ND In OW A-3 	
	 ND in OW A-3 	
	 ND in OW A-3 	
— ND In OW A-3 	
	 ND in OW A-3—
	 ND In OW A-3 	
	 ND in OW A-3 	

.006 to 1.05
.08 to .9
* .002 to «.005
.01 to .19
« .01 to I. 01
«,0002 to '.001
« .01 to < .01
.01 to .01
.04 to .6
16, to 580.
.2 to 120.
.2 to .9
,09 to 9.

5.9 to 6.8
228. to 9,517.
6. to 4,700.
— no data —
« .00} to 25.2
ND • not detected; NC - preeence »uapect«d, but not conflrmid; * - a priority

— ND In brook 	
ND to 11
	 ND in brook 	
ND to 81
	 ND in brook 	
ND to < 10
— ND In brook 	
	 ND in brook 	
	 ND in brook 	
	 ND In brook 	
ND to t 10
SD to 2,700
	 ND in brook 	
ND to 20
ND to 16
< 100 to 7,650
— -ND In brook 	
— NC in brook—
— NC In brook 	

	 ND in brook 	
	 ND in brook 	
	 ND In brook 	
	 ND in brook 	
	 ND in brook 	
	 ND in brook 	
	 ND in brook 	
	 ND In brook 	
	 ND In brook 	
	 ND in brook 	
	 ND in brook 	
	 ND in brook 	
	 ND In brook 	
	 ND In brook 	
	 ND in brook 	
	 ND In brook 	
	 ND In brook 	

-—no data 	
.06
- .005
.01
« .01
» .001
— no data —
« .01
* .1
10.
4.2
.1
.02

— no data —
— no data — -
— no data—
— no data —
— no data —
pollutant,
b> walU bitwxn iltt and brook aril »A-2.  I/A-3, "M-2, and «-D,
c> Brook •ampllnn utatloni rapraiintid In table art: NHDB 04, 04A, 05,  05A, and 06.  Inorganic! data ara
  available only for atatlon NHDB 04A along thin portion of Lyli Hied  Brook.

-------
366    CASE HISTORIES
Department well located between Tumblebrook and Trout
Brook Roads. Contaminants (in groundwater) that have
already moved to and beyond Lyle Reed Brook are, for all
intents and purposes, not recoverable. Thus, the risk of
organic contamination reaching private supply wells along
Route 111 is high regardless of what is done at the site.
  The estimated travel time for groundwater to move from
the Pennichuck well to Route 111  is from  1 to 3  years
(travel time to the Nashua River  is  estimated at from 2 to
5 years).  It is  also possible that contaminated  ground-
water is  moving in bedrock joints. Thus, any  bedrock
wells located downgradient  of  the  disposal site are in
jeopardy.
  The most contaminated zones  extend from the disposal
site to within from 100 to  200  ft of Lyle Reed  Brook.
If not prevented, these zones of highest groundwater con-
tamination will reach Lyle Reed  Brook within a year; the
Pennichuck well in 1 to 4 years; Route 111 in 2 to 6 years;
and  the Nashua River in 3 to 9 years.  Furthermore, if
contaminants are not prevented from leaving the disposal
site, then groundwater contamination between the site and
the Nashua River will  steadily increase  for several  years
and then remain at high levels for an indefinite period.
  Since a major portion of the groundwater plume is dis-
charged to Lyle Reed Brook, pollutant concentrations in
the brook would also increase  for several years before
stabilizing at  high  levels for an  indefinite  period. Of
course, any increase in the pollutant load in the brook will
be reflected in the Nashua River as well.

REMEDIAL ACTION ALTERNATIVES

  During this investigation, several alternative measures
available  for  mitigating the environmental  and  public
health threats posed by the Sylvester Site were considered.
These included:
•No  Action—natural flushing in combination  with fre-
 quent water quality monitoring
•In-place hydrologic isolation of contaminated soil and
 groundwater zones
•Interception and treatment of contaminated groundwater
 in the area between the site and Lyle Reed Brook
•Excavation of the disposal site for secure landfill dis-
 posal either in an existing facility in upstate New York,1
 or in a new facility that would have to be built in the
 Nashua area for that purpose
•Combinations of the above
  The various  alternative remedial action  measures are
briefly discussed below.

No Action (Natural Flushing)

  Under certain conditions, an appropriate means of deal-
ing with  an existing soil  and groundwater contamination
problem  is to take little or no direct preventative action,
thereby relying upon natural processes to dilute, disperse
and  degrade the contaminants  over an extended time
period. In terms of environmental protection, a no action
policy is the  least desireable alternative  due to the high
potential for adverse impacts on surface water and ground-
water. The key unknowns are the magnitude of subse-
quent stream  and groundwater contamination; and the
time-span required before contaminant levels in water
would decrease to "acceptable" levels.  For the Sylvester
Site, the "no action" option was considered to be unac-
ceptable due  to the extremely high contamination levek
in groundwater, the mobility of the contaminant plume
and the  potential for long-term adverse environmental and
public health conditions.

In-Place Hydrologic Isolation

  Hydrologic  isolation options for the  Sylvester  Site in-
clude the installation of a slurry cut-off wall around con-
taminated zones (possibly with bedrock  grouting) and in-
stallation of an impermeable cover over  the contaminated
area.  A slurry cut-off wall is a vertical, essentially im-
permeable barrier installed to prevent groundwater inflow
and outflow from the contaminated zone. A slurry  cut-
off wall installed to surround contaminated zones would
prevent  clean  groundwater moving above rock from be-
coming  contaminated, and would  prevent  further leach-
ing of contaminated groundwater from the isolated area.
  There are two viable options for the location  of the
slurry trench  cut-off wall around the site (see Figure 5).
The first option would isolate only the immediate area of
chemical dumping. The second option would encompass
all of the area within the plume from the site to the edge
of a bedrock  valley which runs north-south roughly par-
allel to Lyle Reed Brook.
  The second slurry wall option would isolate the  bulk of
the contaminant plume, while the first would have no
remedial impact on the portion of the contaminated zone
which has already moved beyond the confines of the site.
In conjunction with the slurry cut-off wall, an imperme-
able surface cap consisting of either impervious soil (i.e.,
clay) or a synthetic liner would be required to prevent in-
filtration of precipitation through the enclosed waste.
  Hydrologic isolation by a combination of a slurry wall
and injection grouting of the bedrock carries the contain-
ment alternative discussed above one step further. In addi-
tion to the cut-off wall within the overburden and the im-
permeable surface cap, fractured bedrock along the desig-
nated perimeter line would be sealed by the injection of
a bentonite/cement grout under high pressure. The en-
suing reduction of rock  permeability would  be  accom-
panied  by a decrease in contaminant migration. At the
Sylvester Site, bedrock has been cored  in only one loca-
tion (boring M-l) and was found to be moderately frac-
tured.
  The slurry  wall/grout curtain alternatives would retard
essentially all  movement of contaminants across the con-
tained area. Over a period of time, however, there is the
possibility of leakage through the cut-off walls. For this
reason,  a monitoring system  would be needed. An intrin-
sic  problem with any hydrologic isolation system, is the
inability to alleviate contamination  which  has  already
moved off-site. The limits of the containment area would
have to be expanded beyond practical proportions to iso-
late all contamination associated with the site.

-------
                                                                                       CASE HISTORIES    367
Groundwater Interception and Treatment

  A preliminary assessment of the technical feasibility and
estimated costs of an above-ground treatment system for
renovating the contaminated groundwater underlying the
Sylvester Site was prepared through a laboratory-scale
treatability  study. This study was  conducted with assis-
tance from  Environmental Resource Associates, Inc.  The
purpose of the laboratory-scale treatment study was to de-
velop a "first-cut"  assessment  of  the treatability of the
highly contaminated groundwater at the site. As such, the
study was confined to a small number of alternative water
treatment methods tested under a limited set of operating
conditions.  Using an integrated batch testing approach,
the pollutant removal effectiveness of  two alternative
treatment process trains were evaluated:
•Air stripping  followed by  neutralization/precipitation
  followed by activated carbon adsorption
•Air stripping  followed by neutralization/precipitation
  followed by biological treatment
  Contaminated groundwater  for  the treatability study
was obtained during the aquifer pumping test conducted
during the period January 28-February 2, 1981. The raw
groundwater sample was collected toward the end of the
pumping in order to base the treatability assessment on
feedwater that would be most representative of the effluent
produced under extended pumping conditions.
  Using the information developed during the laboratory-
scale evaluations, preliminary design and cost estimates for
a full-scale, on-site treatment system were developed. The
conceptual treatment system proposed for the  Sylvester
Site groundwater is illustrated in the block flow diagram
in Figure 6. Longer duration and more operationally flex-
ible pilot-scale testing will be needed prior to final treat-
ment  system design. Among the key uncertainties to be
resolved  in pilot-scale testing area the need for a  metals
removal  step and the optimum  operating conditions  for
the biological and carbon adsorption processes.

Total Removal Alternatives

  For comparison with  the remedial  measures outlined
above, preliminary cost estimates were prepared for  the
so-called total removal options involving excavation of the
disposal  site for secure  landfill disposal. Assuming an
average depth of  removal of  10 ft  over the 6-acre site,
             1.900' to Nashua Rlv.r
                                       OlM-2

• 12



Cllson Road

CCS Disposal
Company Garage



• lA-6






"^A
\
\
1
1
I
/
O Multilevel Wall
0 „ • Wellpoint Well

+ Pumping Well
t Property Line

                                                     Figure 5.
                                       Slurry Wall and Cap Isolation Options for
                                                   Sylvester Site

-------
368     CASE HISTORIES
the costs to excavate,  transport  and dispose of the ma-
terial in an existing secure landfill facility in upstate New
York and to design and construct a new secure landfill
disposal facility in the Nashua area were estimated. The
conditions at the Sylvester Site are such that,  even with
removal of  the disposal site,  an extensive  groundwater
interception and treatment system would be needed as part
of a total removal plan.

RECOMMENDED REMEDIAL ACTION PLAN

  The recommendation of the GHR/GZA Project  Team
was that the final remedial action plan for  the Sylvester
Site include both  in-place hydrologic isolation of contam-
inated  zones and  interception of contaminated ground-
water for on-site  treatment. Specifically, it was recom-
mended that the final plan consist of:
(1) Design and installation of slurry trench cut-off wall
    Option 2 to completely surround approximately 12.5
    acres of the contaminated zone, including  the  6-acre
    disposal site
(2) Design and installation of an impermeable top-seal
    over the isolated zone to prevent infiltration of rain-
    water
                                 (3) Design and installation of a groundwater interception
                                     and treatment system to renovate that portion of the
                                     contaminated aquifer remaining outside the isolation
                                     zone
                                 (4) Installation of 6 to 8 new multilevel groundwater ob-
                                     servation wells to  monitor  the effectiveness of the
                                     slurry wall  system and track the contaminant plume
                                     in the region between Lyle Reed Brook and the Nashua
                                     River
                                    In  addition to the design and implementation of the
                                 slurry wall/treatment system it was recommended that:
                                 (1) Lyle  Reed  Brook  be  posted and  fenced to  prevent
                                     human exposure to the contaminated stream and
                                 (2) city water be provided to those residences along Route
                                     111 between Countryside Drive  and Jensen's Trailer
                                     Park who now rely on private wells for drinking water.
                                    The severly degraded condition of the groundwater in
                                 the area  from the site to Lyle Reed Brook requires that
                                 the total remedial action plan for the site include ground-
                                 water interception for on-site treatment. The groundwater
                                 interceptor wells would be located downgradient of the
                                 selected slurry  wall.  Through  strategic location of the
                                 wells, with respect  to  both areal orientation and depth,
                                 a minimal amount of contaminated groundwater flow will
    WELL  FIELD
  Removal contaminated
  groundwater from
  plume and deliveri
  it to above-ground
  treatment system.
                           VAPOR RECOVERY
                            Reduces levels of
                            organic solvents In
                            spent air from
                            stripping column.
                            Activated carbon
                            used to recover
                            solvents in liquid
                            form.
                          AIR  STRIPPER
Removes volatile
organic solvents
from groundwater.
Utlllies air strip-
ping column to
pass air and water
in countercurrent
flow over a packed
inert media.
pH ADJUSTMENT
METAL REMOVAL

  Adjusts pH of
  groundwater to
  1.5.  Removes
  metals by preci-
  pitation and
  clarification.
                                                                              BIOLOGICAL
Will degrade alcohols
by biological activity
and remove remain-
Ing volatile organlcs
by air stripping.
                                                                                                     CARBON  ADSORB
Removes high
molecular weight
organlcs and
color by adsorp-
tion on activated
carbon beds.
                                                                                                     LEACHING TRENCH
                                                                                                      Covered In-ground
                                                                                                      leeching trenches
                                                                                                      to return treated
                                                                                                      groundwalor to
                                                                                                      aquifer.
                                                       Figure 6.
                                  Block Flow Diagram of Conceptual Treatment System for
                                     Groundwater Renovation at Sylvester Site, Nashua,
                                                    New Hampshire

-------
                                                                                        CASE HISTORIES     369
pass  through the  interception system  to downgradient
areas.
  The principal variables with the interception and treat-
ment scheme are:
(1) the treatment processes to be used,
(2) the degree of pollutant removal desired and
(3) the length of time that the treatment plant would have
   to be operated.
  Based on  aquifer volume calculations and  interceptor
system flow  rates, the treatment periods required to treat
given multiples of the initial contaminated aquifer volume
were estimated. This was done for the two slurry wall op-
tions and  the additional case of interception  and treat-
ment without hydrologic isolation. The results are pro-
vided on Table II.  the effluent from the  treatment plant
would be  recirculated back through  the  aquifer  by dis-
charging into a subsurface leaching trench to be  located
outside the slurry wall.
                       Table II.
   Estimated Treatment Periods Required for Various Design
        Flow Rates and Hydrologic Isolation Options
Remedial
Option

Treatment With-
out Hydrologic
Isolation
Slurry
Wall
Option 1
Slurry
Wall
Option 2
Pump-
ing
Rate
(GPM)

 25
 50
100
 25
 50
100
 25
 50
100
Treatment Periods in Years Re-
quired for Various Aquifer
Volumes
1      3       5
4.3
2.2
1.1
2.1
1.0
0.5
1.0
0.5
0.25
12.9
6.6
3.3
6.3
3.0
1.5
3.0
1.5
0.75
21.5
11.0
5.5
10.5
5.0
2.5
5.0
2.5
1.25
Notes:  1. A 1-volume treatment period represents  time required to
        pump a volume equal to the initial contaminated  aquifer
        volume at given pumping rate.
      2. Treatment periods represent times  required to treat given
        multiples of initial contaminated aquifer volumes.  For treat-
        ment without hydrologic  isolation,  the initial volume is 60
        million gallons; for slurry wall option 1, the initial volume is
        30 million gallons; and, for slurry wall option 2,  the initial
        volume is 15 million gallons.
  The critical aspect of the slurry wall alternative is the
degree of contaminant isolation achieved. The major de-
sign consideration for the  slurry wall will be to formu-
late a slurry mixture that will have not only the necessary
very low permeability, but also the ability to  resist chem-
ical reaction and breakdown by the organic contaminants.
The installation of an impermeable surface over the iso-
lated area will prevent the  build-up of liquids inside the
slurry wall, thereby preventing development of a hydraulic
head to promote leakage. Proper design and installation
of the isolation system should  result  in minimal leakage
through the slurry wall, and any leakage that does occur
would be diluted by large volumes of groundwater.
  It is anticipated that any  leakage  of contaminants
through the bedrock underlying the  recommended isola-
tion area will  be minimal due to absence  of a driving
hydraulic head. However, only an extensive program of
bedrock drilling and permeability testing would reveal the
quantities of flow (if any) from the area through  bed-
rock units. Bedrock grouting was not recommended since
the available data did not justify the expenditures  that
would be necessary.

Cost Effectiveness Analysis

  To evaluate the cost-effectiveness of the recommended
remedial action plan,  the methodology used by the U.S.
Environmental Protection Agency for municipal  waste-
water treatment projects was employed. Using this stand-
ard approach, the various remedial action  alternatives for
the Sylvester Site were evaluated in  terms of their com-
parative present worth.
  To provide a common basis for comparing the costs of
the various treatment system alternatives,  it was assumed
that groundwater treatment would be provided for at least
as long as necessary to treat five multiples of the initial
contaminated aquifer volume (see Table II). This assump-
tion was necessary in order that each alternative slurry
wall/treatment system have an equivalent  remedial bene-
fit (i.e., treat five initial aquifer volumes). This assump-
tion also establishes  the approximate  time  period over
which the treatment plant would have to be  operated under
each alternative,  and thus sets the estimated number  of
years for which operating and maintenance costs would be
incurred.
  Using this methodology, the costs of  the various al-
ternatives were estimated on a comparative present worth
basis; the results are presented in Table III. The present
worth cost comparisons indicate that the recommended re-
medial action plan (Item 5a.) is the  most cost-effective al-
ternative.
  By basing the comparative cost analysis on treating five
multiples of an initial contaminated  aquifer volume,  the
evaluation has  ignored aquifer recharge through rainfall
infiltration.  In actuality, infiltration outside of the slurry
wall and cap system will probably require a 50 gal/min
interception rate under Option 1. As shown in the above
tabulation, even if a  50 gal/min plant is  combined with
slurry wall Option 1 and is operated for only 5 years,  the
recommended Option  2 system still has a lower present
worth. With the Option 2  slurry wall and cap, a 25  gal/
min pumping rate will be sufficient to intercept the con-
taminant plume.
                                          CURRENT STATUS

                                            The recommended remedial plan was selected for imple-
                                          mentation in late August 1981, the design phase of the
                                          slurry wall and cap isolation system was being undertaken
                                          and plans for pilot-scale treatment system testing were be-
                                          ing formulated. Also, an interim (emergency) groundwater
                                          interception  and recirculation system to  reduce contam-
                                          inated groundwater discharges to Lyle Reed Brook was be-
                                          ing  designed.  These design efforts are being conducted

-------
 370    CASE HISTORIES
                                                     Table in.
                                 Summary of Remedial Action Alternatives and Estimated
                                      Costs, Sylvester Site, Nashua, New Hampshire
                         DESCRIPTION OF REMEDIAL ALTERNATIVES
                                                                            ESTIMATED COSTS
                   Total  removal of dump site for secure  landfill
                   disposal out-of-state, and intercept and treat
                   aquifer for 20 years 9 50 gpm (AS-N/P-BIO-GAC)

                   Total  removal of dump site for disposal in new
                   secure landfill in NH, and intercept and treat
                   aquifer for 20 years 9 50 gpm (AS-N/P-BIO-GAC)

                   Aquifer interception and treatment alone for
                   20  years 9 50 gpm (AS-N/P-BIO-GAC)
                   Slurry cut-off wall and clay  cap Option 1 to
                   isolate approximately 7.2 acres, with
                   a.   aquifer interception and  treatment
                       for 10 years 9 25 gpm (AS-BIO-GAC)

                   b.   aquifer interception and  treatment
                       for 10 years 9 25 gpm (AS-N/P-BIO-GAC)
                   c.   aquifer interception and  treatment
                       for 5 years 9 50 gpm (AS-BIO-GAC)

                   d.   aquifer interception and  treatment
                       for 5 years 9 50 gpm (AS-N/P-BIO-GAC)
                   Slurry cut-off wall and clay  cap Option 2 to
                   isolate approximately 12.5 acres, with
                   a.   aquifer interception and  treatment
                       for 5 years f 25 gpm (AS-BIO-GAC)
                   b.   aquifer interception and  treatment
                       for 5 years t 25 gpm (AS-N/P-BIO-GAC)

                   c.   aquifer interception and  treatment
                       for 3 years 9 50 gpm (AS-BIO-GAC)
                   d.   aquifer interception and  treatment
                       for 3 years 9 50 gpm (AS-N/P-BIO-GAC)
                                             $ 19,000,000 to $ 28,500,000



                                             $ 13,000,000 to $ 19,500,000


                                             $  9,000,000 to $ 13,500,000




                                             $  3,420,000 to $  4,850,000


                                             $  5,170,000 to $  7,500,000


                                             $  3,110,000 to $  4,360,000


                                             $  4,690,000 to $  6,740,000




                                             $  2,770,000 to $  3,810,000


                                             $  3,850,000 to $  5,430,000


                                             $  2,720,000 to S  3,670,000


                                             $  3,780,000 to $  5,280,000
                 Notes:   1.  Costs are comparative  present worth values.
                            Present worth of estimated annual treatment plant operating  and
1.

2.



3.



4.

5.
                            maintenance costs were  computed using an interest rate  of  7-3/8
                            percent as provided  by  U.S. EPA, Region I, Boston,  MA.
                            Treatment plant assumed to operate for number of years  necessary
                            to treat 5 multiples of the initial contaminated aquifor volume.

                            Salvage values assumed  to be zero.
                            AS « air stripping;  N/P • neutralization/precipitation  for metals
                            removal; BIO = biological lagoons; GAC = granular activated carbon
                            adsorption.
by GHR Engineering Corporation, again in conjunction
with Goldberg-Zoino and Associates.
ACKNOWLEDGEMENTS

  The successful investigation of the Sylvester Site re-
quired a high level of cooperation and commitment from
many individuals. While it would be inappropriate to iden-
tify all major contributors to this effort, it is appropriate
to acknowledge the guidance and  cooperation received
from  Mr.  Carl Eidam of the U.S. EPA  and Mr. Michael
Donahue of NH WS & PCC.
                                 REFERENCES

                                 1. GHR  Engineering Corporation,  "Hazardous Waste
                                    Site Investigation, Sylvester Site, Gilson Road, Nashua,
                                    New Hampshire," Final Report prepared for the New
                                    Hampshire Water Supply and Pollution Control Com-
                                    mission and the U.S. Environmental Protection Agen-
                                    cy—Region I Laboratory, July 1981.
                                 2. Recra  Research, Inc.,  "The  Characterization of the
                                    Drums of Waste Material Located on the Gilson Road
                                    Hazardous Waste Disposal Site in Nashua, New Hamp-
                                    shire: Operational Summary," prepared for GHR En-
                                    gineering Corporation, July 1980.

-------
  CASE HISTORIES IN HANDLING UNKNOWN HAZARDOUS
                 MATERIALS AT DUMP SITE LOCATIONS
                                        THOMAS F. DALTON
                                         Hazchem Services Inc.
                                         Garwood, New Jersey
INTRODUCTION

  The author's purpose in writing this paper was to pre-
sent a series  of case histories involving  the cleanup of
hazardous dump sites where the type or degree of hazard
or some of the materials handled were unknown.
  Each  of  the three case histories  involved a different
set of circumstances requiring decisions and techniques to
determine as close as possible what was the nature of the
problem, how best to handle or control the material and
then, how to dispose of it.

CASE HISTORY #1
CHEMICAL CONTROL—NEW JERSEY

  This location, at the beginning of the remedial action,
presented a formidable challenge to  state  and cleanup in-
dividuals, mainly  due to the wide  variety of  unknown
chemicals and the extreme  danger  involved in handling
them.
  Initially, there were almost 500,000 laboratory size con-
tainers  and  bottles, indiscriminately  mixed together in
drums,  cartons  or stacked on tables.  They ranged from
unknown chemicals to correctly identified materials.
Classification System
  The breakdown of each container into a specific cate-
gory, not only  for  classification purposes, but also for
disposal, created a situation which called for broad group-
ing with many  divisions. Into each division, were sub-
divisions, which would allow the grouping of like mater-
ials and make it easy to combine those  subdivisions so
they could be disposed of at a secure landfill based upon
the limitations and groupings required by the disposal
company. For example, acids were classified as follows:
  Class A. Inorganic
   1. Fuming (liquid)
  2. Non-fuming (liquid)
  Class B. Organic Acids (liquid)
  Class C. Crystalline
  The disposal protocol called for  taking.Class A items
and placing them  into Group A (compatibility grouping
of a disposal  company) for  disposal of laboratory packs.
Class B items would be placed into Group D and so forth.
It would be possible for grouping into any disposal com-
pany's compatability grouping.  Following the need for
classification was a necessity for record keeping,  samp-
ling and coding.
Record Keeping
  This is probably one of the most tedious requirements
at any dumpsite location. Record keeping at  a  major
cleanup site requires  the services of at least a  full-time
person and the cooperation of everyone.
  All of  the items that were located which were consid-
ered extremely  hazardous  (highly toxic,  very unstable,
explosives or highly reactive) had to be noted in the log
book along  with those items requested by  the state or
federal agencies which were followed with a chain-of-
custody (evidentiary) protocol.

Coding of Drums
  The coding of drums is not a problem  when there are
only several hundred drums or batches of drums with the
same characteristics.  When there are 50,000 to 60,000
drums with many different physical and chemical proper-
ties, then one has to  establish a more sophisticated sys-
tem.
  At Chemical  Control, a six digit  coding was worked
out with  the following design. Starting inside the  build-
ing with #1, then going in a clockwise position, Front #2,
Back #3, Right Side  #4, Left Side #6.  Any materials
from a source off the property would be #6. The  next 5
numbers  were the drum numbers for samples, identifica-
tion and  statistical purposes. In addition, each disposal
location  was coded with a letter beginning with A (for
SCA). This system resulted in a drum so marked that one
could determine exactly where it came from and where it
was  going.  For example:  DRUM  #201109A. This de-
signation meant that  the drum came from the  front of
the building, was drum #1109 and was sent to SCA for
disposal.
  It is important to set up a clear, simple means of review-
ing drums for  statistical study and  also  as a means of
tracking the drums.
  All of  the laboratory pack drums were packed with all
bottles or containers identified and grouped into classes.
Each drum had a list  of contents  which could be verified
by subsequent inspection of the disposal  company. This
made for rapid evaluation by the disposal companies  af-
ter  the lists had been checked  and verified.  Once the
system was set  up, the transportation, checking and dis-
posal preceded quite rapidly.
                                                   371

-------
372    CASE HISTORIES
Sampling
  Sampling was a tedious and difficult project to set up
and follow due to the large numbers of drums and solid
materials that were sampled.  Various  sampling devices
were used, ranging  from a 40 in. by 3/8  in. PVC tube
for drum  samples to turkey basters made of plastic  for
the semi-liquid materials.  There were also composite
samples for use in a CG unit, which were collected in an
8 oz. bottle, then transferred to a 30 ml plastic container.
It was difficult to write  a six digit  number with a wax
crayon on a vial this small, so numbers had to be written
on a plastic bag in which the vial was placed. Solids and
liquids were collected in a variety of ways and coded with
the same numerical system used on the drums.
Unexpected Situations
  There was  always a  surprise or two  each  week at  the
site.  Sometimes there were cyanide drums  or acid drums
leaking, evolution of H2S gas near  the building, drums
containing metallic sodium and sodium anhydride would
explode after a  heavy rainstorm, malodorous  leaks of
multi-colored oil,  greases and plastics periodically found
their way out  of the stacks of drums.
  After a few of these episodes, it became a regular prac-
tice of the local Fire Department to find out if they could
use water or sand and soda ash on a fire and what type of
safety equipment should be utilized.
  An explosion and fire occurred at the site about one year
after the work commenced (and has been described in an-
other paper  in these  Proceedings).  Today  it  has been
cleaned up and no longer is a threat.

CASE HISTORY #2
(PICILLO FARM, RHODE ISLAND)

  This situation was different from Chemical Control be-
cause most of the drums were buried or partially covered
in several large trenches filled with several feet of water.
The area was remote and access to the site itself was lim-
ited. Working with  state officials to determine the best
plan to remove some drums that appeared to be sodium or
aluminum anhydride called for an initial  delay  of six to
eight weeks until the  ground could dry up.
   Sodium anhydride,  when  contacted with water, reacts
violently with a  tremendous heat release.  Hydrogen  gas
is also released  and  a  hot mixture containing  caustic
soda remains. When water is dumped into sodium anhy-
dride,  it is impossible  to control the speed of reaction,
hence  it reacts  in  an explosive-like manner.  For  this
reason, it was necessary  to wait  until the water had dis-
appeared and the weather improved. The hydrogen gas, if
not dissipated, could collect and explode.
  The  plan,  worked out with the State  Department of
Environmental Management, was to remove the drums
from the ground,  sample  and  identify  the  contents,
over-drum, if necessary and store for  future disposition.
  The drums were removed mechanically after the ground
dried out and the surface soil was removed,  if necessary,
sampling was done with a 40 in.  thief,  the sample placed
in plastic sample jars with screw caps. Analysis indicated
both used  and unused sodium anhydride were  in  the
drums.
  Temporary storage was  in a box trailer  with more
permanent storage at an old ammunition silo at the Na-
tional Airguard Air Station.
  There were two options for disposal:
  (1) React  under controlled conditions to end products
     which could be disposed  of  locally. Cost between
     $350-$500/drum.
  (2) Ship to an approved disposal site located in Nevada.
     Cost of disposal  and transportation approximately
     $800-$l,000/drum. This  method  was  ultimately
     selected due to the limited number of  drums and
     disposal out-of-state.
  Later on,  as the monies became available,  a more  ex-
tensive cleanup at Picillo Farm  was initiated by the state
of Rhode Island. Most of the drums were buried or parti-
ally buried in  shallow  cells scraped out by a  bulldozer,
then dirt was pushed over the drums. Due to the geology
of the area, the cleanup had to be done quickly to prevent
both surface and groundwater contamination.
CASE HISTORY #3
(KIN BUC LANDFILL—NEW JERSEY)

  This landfill situation differed from the other two be-
cause all  of  the drums and materials  were completely
covered with soil to a height of 80 ft above grade. The lo-
catioan of the site was also a problem, since there were no
good access roads to the area in which the work had to be
performed.
  Initially, the plan was to treat oily wastes which were
leaching out  periodically and finding their way through
the marshlands to the Raritan River. The USCG and EPA
had  monitored  this situation and decided to attempt to
control  the leachate  which resembled  an  ongoing oil
spill.
  Workers sampled the deposits which were sent for analy-
sis. In the meantime,  since there was some old  technical
data available, it was examined as a logical safety protocol.
GS/MS reports on the old samples  showed 50 to 60 dif-
ferent organics. The atomic absorption  reports indicated
the presence of heavy  metals, especially those highly tosic
to humans and water supply systems:  lead, cadmium,
copper, chromium, zinc, barium and lanadium.
  The old reports also indicated small amounts of PCBs,
along with many benzene ring homologs.  This information
indicated that the  initial work should be done  with full
face respirators, as well as protective suits, gloves, boots.
Later on, as the level of  PCBs  and other phenolic and
naphthenic compounds were quantified, the workers had
to change their safety protocol.
  The first step was to remove debris, set up filter fences
and  install both a disposable  and  a permanent oil spill
boom. Due to the location of some of these boom and fil-
ter barriers, which  had to be monitored daily, small boats
were used to negotiate the marshlands. This was especially
true after  a heavy rain when the leachate volume in-
creased.

-------
                                                                                      CASE HISTORIES    373
 When the volume of leachate began to reach 2 to 3 gal/
min with a well defined black organic layer, skimming into
drums was done in an effort to remove more of the or-
ganic layer of leachate instead of allowing it to course its
way through  the swampy marshlands  into the Raritan
River. The work recently has been accelerated and modi-
fied in an attempt to:
 (1) Concentrate the leachate
 (2) Treat the leachate, in situ
 (3) Consider long term cleanup with either carbon or
     incineration
 With  a potential  for a high flow rate of leachate,  de-
pending on local hydraulic conditions, Kin Buc will be a
problem of immense proportions for years to come. Cre-
ating Kin Buc took many years, cleaning it up will take as
long, if not longer.

SUMMARY
 The three case histories, presented, represent hazardous
dumpsites in which the materials were:
 (1) Above ground
 (2) Partially buried
  (3)  Completely buried
While it is an oversimplification to state that each and
every landfill has to be approached differently,  there is
no  doubt that working  with materials that are above
ground  is easier, although there is an inherent danger of
spills and percolation into groundwater.
  Buried  materials  still  present a strong problem of
aquifer  contamination, however, this  process  may be
slowed down depending on the nature  of  the materials,
the condition of the containers or drums and the percola-
tion rate  of the  soil  surrounding  the material.  Those
drums, partially buried, can exhibit a higher corrosion rate
due to the metal/ground interface which causes  a higher
concentration of electrolytes to occur at the interface.
  Of the three sites mentioned, the last one (Kin Buc) is
still ongoing with  a great deal of work to do. The capping
of the dome-like  dump reduced the percolation through
the infrastructure of the dump. The problem of a varying
effluent occurs with  higher  concentrations  of  organics
and soluble inorganics in the leachate. This produces a
paradoxical situation in that there is less leachate, but of
higher toxicity.

-------
      CONTAMINATION CONTROL AT ROCKY MOUNTAIN
                       ARSENAL, DENVER, COLORADO
                                      JOHN WARDELL, Ph.D.
                                        MARGOT NIELSON
                                           JUDITH WONG
                                U.S. Environmental Protection Agency
                                             Region VIII
                                           Denver, Colorado
INTRODUCTION

  In response to requirements to  contain or clean  up
contamination at Rocky Mountain Arsenal (RMA), Den-
ver, Colorado, the Department of the Army (DA), began
a program to mitigate the contamination problem at that
installation in 1975. The approach initially taken by DA
was to contain contamination at the installation's boun-
daries.
  In 1978, a  pilot  wastewater treatment system began
operations at  the ridrthern arsenal boundary to remove
organic  contamination  from  ground water.  The  pilot
system proved successful.  It has been  expanded to  in-
tercept all contaminated groundwater crossing that boun-
dary. A similar system is also being designed to intercept
all contaminated water crossing the northwest boundary.
This paper describes the history and operation of these
containment systems.

HISTORY OF THE CONTAMINATION PROBLEM

  Established in  1942, RMA occupies 17,000 acres of land
10 miles  northeast  of Denver, Colorado. The northern
and  eastern  borders are  bounded  primarily  by agri-
cultural land. Agricultural lands, an industrial area, and
a  housing  development  are  located  adjacent to  the
southern  boundary.  Residential  areas are located  along
the installation's western boundary (Figure 1).
  Within eight square  miles north of  the northern  ar-
senal  boundary,  the land is primarily agricultural. Popu-
lation density is 10-15 people per square mile. Water is
supplied  by  approximately 90 wells. Thirty percent of
these  wells use water from the aquifer affected by con-
tamination from the arsenal. Irrigation, stock, commer-
cial and domestic wells are affected.
  Constructed to support World War  II, RMA manu-
factured  and  assembled  toxic  chemical end  items and
incendiary munitions. At the end of the war, RMA was
placed on standby  status. A small number of activities
including  demilitarization  of  obsolete hazardous and
toxic  munitions  occurred at the arsenal until the begin-
ning of the Korean War.
  In 1946, parts of the installation were leased to private
industry  for  chemical manufacturing. The major lessee,
Shell  Chemical Company  (Shell), had  used a large por-
tion of these original chemical manufacturing  facilities
since 1952. That company  has made additions to their
leased facilities for manufacturing pesticides.
  The arsenal was selected as the site to construct a fa-
cility to manufacture GB, a highly toxic, non-persistent
nerve agent. This production facility was completed in
1953. Manufacturing of this chemical agent occurred un-
til 1957, but filling of the munitions continued until 1970.
RMA  produced  a biological anticrop agent from  1959-
1962, and emptied cyanogen chloride and phosgene bombs
between 1956-1969.
Unlined Lagoon Used
  Industrial waste effluents  resulting from past DA and
lessee  manufacturing activities were initially discharged
into an unlined  basin immediately north of the original
"plants area" called Basin A. Basin A received these in-
dustrial  wastes until 1957.  When the GB facility was
placed in operation, this basin was enlarged and new un-
lined basins were  constructed northwest of  Basin A
(Figure 2).
  In the summer of 1954, some farmers northwest of the
arsenal complained  that groundwater used for irrigation
had damaged their crops. Precipitation was much less
                      Figure 1.
                 Vicinity Map of RMA
                                                  374

-------
                                                                                     CASE HISTORIES    375
                                 Cm OF DENVER
                      Figure 2.
              Location of Features at RMA
than normal in 1954 and increased pumping from shal-
low irrigation wells occurred to support crop production.
In response to these complaints and subsequent claims
for damage, DA:
 (1) Retained a firm of consulting engineers to investi-
    gate the problem of possible groundwater contam-
    ination described by the farmers.
 (2) Requested the U.S. Geological  Survey (USGS) to
    study water quality on RMA and in adjacent farms.
 (3) Contracted the University  of Colorado to initiate
    plant bioassay, geological and chemical studies to
    determine the identification and source of any con-
    taminants causing crop damage.
 Some stock and irrigation wells were  abandoned be-
cause of high salinity, and compensation was paid to a few
landowners for crop damage. Evidence suggested that
high salinity resulting from groundwater migration from
unlined basins on RMA was the cause of the problem.
New Storage Constructed
 As a result  of this investigating, Basin F was construct-
ed. Built to hold 240 million gallons, it covered  93 acres.
It was lined with an asphalt liner covered with one foot of
soil. Upon completion in 1957, all industrial wastes pro-
duced by DA or Shell at the arsenal were deposited in
Basin F. An attempt was made to transfer residues from
the unlined basins into Basin F (Figure 2).
Contaminants Detected
 In 1974, organic  compounds were detected crossing the
northern arsenal boundary.  Diisopropylmethyl phospho-
nate (DIMP) and  dicyclopentadiene  (DCPD)  were de-
tected in surface water draining from a man-made bog at
the north boundary of the installation. DIMP  is a per-
sistent material produced in small  quantities during pro-
duction of the GB nerve agent. It is not very toxic, but it
is an excellent indicator compound because it originates
from a single activity. DCPD is a chemical used by Shell
in production of some of its pesticides. Though not very
toxic, it has  an extremely  foul odor at extremely small
concentrations making potable water in which it is con-
tained unfit for consumption.
  DA initiated several actions after detecting these com-
pounds. A previously existing groundwater monitoring
program was expanded. Tests for DIMP,  DCPD,  and
several other compounds were added. These latter com-
pounds were chemicals that could have been disposed of as
a result of manufacturing activities at the arsenal. A dike
was  built to stop  off-post discharge of surface water
from the  bog.  The toxicological  impacts  of DIMP  and
DCPD were also evaluated.
  In December 1974, the Colorado Department of Health
(CDH) detected DIMP at less than one /xg/1 in an off-site
well.  The  detection of both DIMP and DCPD in  off-site
surface water and  subsequent  discovery of DIMP in  off-
site groundwater led to issuance of administrative orders
by the CDH  in April 1975. These orders required an im-
mediate stop to off-post  surface  and subsurface  dis-
charge of these contaminants, preparation of a plan to
prevent  their further discharge and implementation of a
water quality monitoring program to demonstrate compli-
ance with the first two requirements.
   Subsequent to issuance of these administrative orders,
other organic  compounds have been  detected  at  the
northern boundary. The most significant of these com-
pounds  is dibromochloropropane  (DBCP),  a soil fumi-
gant manufactured by Shell. DBCP has been reported to
cause male sterility and is a possible carcinogen. A drink-
ing water standard has not yet been  established for  this
chemical.
   In  May 1980, contamination of groundwater off-post
northwest of the arsenal (Figure 1) was detected in wells.
Potable water  sources  contained  between 0.2-0.4  /ug/1
DBCP.
  APPROXIMAl
   OFF-SITE
 CONTAMINATION
 COMME RCE
  CITY
                                 GROUKDUATER FLOWS


                           CITY OF DELIVER
                        Figures.
          Estimated Divide between Northwest and
                North Groundwater Flows

-------
376    CASE HISTORIES
GEOLOGY AND HYDROLOGY

  The bedrock consists of shale and claystone with lenses
of siltstone and sandstone. Alluvium and windblown de-
posits cover this bedrock.  Silty clay, silt, sand and gravel
form these unconsolidated deposits of alluvium. In many
places, wind-deposited silts and sand cover this alluvium.
The alluvium ranges in thickness between 5-80 feet. The
alluvium contains the majority of the highly permeable
materials."'3'61
  Based on previous geohydrologic studies, groundwater
flow is believed to be continuous  through the shallow
RMA aquifer.<4|6) If the alluvium is dry, water flows in the
upper bedrock formation.<3|6) North of the original "plants
area,"  water  flows in  a  northerly,  northwesterly and
northeasterly direction. Water underneath Basin A flows
to either  the northwest  or north boundary  of the ar-
senal(4'6) (Figure 3). Basin A received  large quantities of
potentially hazardous contaminants.
  Groundwater flow across the northern boundary occurs
in both the alluvium and sands in the bedrock. The aquifer
is defined as all sand and gravel,  gravel and sand units
that are unconfined above an impermeable barrier. The
alluvium and sands are hydraulically connected and should
be considered to be one hydrogeologic unit. Groundwater
flow,  however, occurs primarily in  the alluvium because
permeability values in the coarser sand and gravel strata of
the alluvium is much greater than the permeability of the
silty and clayish sands.<5)
  Permeabilities of sand, and sand and gravel units were
estimated to be 3,000 gpd/ft2 (400 ft/day) and 5,000 gpd/
ft2 (668 ft/day), respectively, based on  aquifer tests. Units
containing appreciable quantities of silt had permeabilities
of an order of magnitude less. Units containing clay had
negligible  permeability compared  with permeability  of
sand, and sand and gravel units.7
  Aquifer thickness varied from zero on bedrock highs to
about 40 feet about 1 mile south of the northern  arsenal
boundary.(3i8) Analysis of cross-sections  indicated that
this aquifer is generally continuous.(3|8) The aquifer thick-
ness consists of permeable materials above and below the
water table. The  saturated thickness was  determined  by
calculating the difference between the  potentiometric and
bedrock surface. The potentiometric surface was initially
evaluated  in 1978, before construction and operation of
the contamination containment system. It was  generally
toward the north with an average gradient of 0.006 ft/ft.8
Subsequent measurements taken in 1979, after the contam-
ination  containment system had begun operation, showed
similar  flow characteristics. Gradients between 0.006 to
0.008 ft/ft were observed along the north boundary.0"
  The  clearest  picture of contamination  migration and
distribution affecting the north boundary can be provided
by considering the underlying aquifer as two separate
bodies of groundwater. The most significant body  moves
beneath Basin F toward  the north boundary in a  north-
easterly direction. Contaminants leached from the surface
waste  basins (e.g., Basin A) are carried by this subsurface
flow across the northern arsenal boundary. The other body
of groundwater flows northwesterly parallel and beneath
First Creek. It is generally free of contamination and has
much higher volume than the body of groundwater flow-
ing underneath Basin F. In the vicinity of the north boun-
dary, water table contours of both bodies become parallel
to that boundary. Contamination, therefore, is carried di-
rectly north across the boundary.'3'8'

IMPLEMENTATION OF CONTAMINATION
CONTROL PROCEDURES
Introduction
  Contaminated groundwater has been  detected crossing
both the northern and  northwestern arsenal boundaries
(Figure  2). Contamination  crossing  the northwestern
boundary was detected  in 1980.  Consequently,  contain-
ment strategies there are  only  presently being designed.
The focus of this discussion, therefore, is on containment
strategies undertaken along the northern boundary.
  A two-step approach was implemented. A pilot contain-
ment system was installed  to evaluate the feasibility of the
overall  approach.  Once satisfied that  the  pilot system
could be operated successfully, the containment system
would be extended to  intercept and  treat contaminated
groundwater crossing along  the  entire  affected part of
the northern boundary.
Pilot Containment System
  The pilot containment system was designed to demon-
strate the feasibility  of  containing off-post groundwater
contaminant  migration   across  the  northern  arsenal
boundary. It was placed in operation in  July 1978 and its
performance evaluated in the summer of 1979.(1)
  The pilot containment system is composed of:
  (1)  Dewatering well subsystem
  (2)  Treatment plant
  (3)  Recharge well subsystem
  (4)  Impermeable barrier
  (5)  Monitoring well subsystem
  For a flow schematic of the system  and  plant layout
see Figures 4 and 5 respectively.
                      TREATlffitn SYSTEM
                       Figure 4.
       Schematic Diagram of Pilot Containment System

-------
                                                                                      CASE HISTORIES    377
                                     X
X


A
             S\
                              HITORINC IJELLSXN
HHH
                                   TREATMENT FACILITY
                                       DEUATERIHG WELLS
Premature filter plugging is indicated by a high pressure
drop alarm. Backwashing is done one filter at a time using
filtered water from the  onstream filter for  a pre-estab-
lished time period. The backwashed filter is automatically
returned to service when this operation is finished. The
particle-laden backwash water  is  collected  in  a  sump
where  the particles are  allowed to settle  out.  The  de-
canted water is reprocessed through the treatment plant.
  Pressure is used to move water from the particle filters
to one of two  columns  packed with  20,000 Ibs of acti-
vated carbon (Figure 6). Only one column is used at a
time.  Water flows down through  the  column  and  the
activated carbon packed into the column absorbs the  or-
ganic contaminants. The treated water flows to the  re-
charge wells by gravity for reinjection into the aquifer.
                       Figures.
     Overview of Pilot Containment System (not to scale)
  Flow of contaminated water across the arsenal boun-
 dary is stopped by the impermeable barrier. Dewatering
 wells remove water from the aquifer for treatment in the
 treatment plant. Recharge  wells inject  water  back into
 the aquifer.

 Dewatering Well Subsystem
  This subsystem is composed of six 8-inch diameter wells
 placed within 30-inch diameter gravel-packed holes. The
 wells are screened through the entire alluvial aquifer thick-
 ness. They are approximately 225  feet apart.  Each well
 has been provided  with a flow control  system and sub-
 mersible pump. The pumping system was designed to keep
 a constant head within each well.  Sensors were  used to
 recycle water back into the well instead of to the treatment
 system when the pumping water level fell below a prede-
 termined level.
 Treatment Plant Subsystem
  The  contaminated  water  removed by the  dewatering
 wells is discharged into a sump. This water is then pumped
 through a filter to remove suspended solids. The water is
 passed through  an  activated granular  carbon column to
 remove  organic contaminants  (e.g.,   DIMP,  DCPD,
 DBCP). The treated groundwater is discharged to the re-
 charge well system and then back into the aquifer by grav-
 ity drainage.
  Contaminated water is pumped through the treatment
 system at a controlled rate (design flow is 10,000 gal/hr).
 A  flow control system regulates  flow to the treatment
 system from the sump on the basis of water level in the
 sump. Contaminated water is filtered to remove particles
 before  being passed through the packed carbon column.
 Two filters  are used.  Each  filter contains four feet  of
 filter media. This media is a blend of graded coal and sand.
 The filters are operated in parallel.
 Backwashing is required periodically because the filters
 become plugged with  particulates. Backwashing can  be
 done manually or automatically on a preset time interval.
                                                                    Figure 6.
                                                 Treatment Facility Prefilters (left) and Adsorbers (right)
                                               Spent carbon must periodically be replaced with fresh
                                             activated carbon which is delivered to the treatment fa-
                                             cility in specially designed trailers. Transfer of the carbon
                                             from this trailer to the column is accomplished by filling
                                             the trailer with treated water to slurry the fresh carbon.
                                             The treated water had been stored in the empty column
                                             prior to arrival of the trailer. Spent carbon is transferred
                                             to and from the trailer  as  a  slurry by pressurizing the
                                             column or  the truck  with  compressed  air. The  slurry
                                             water is drained into the sump and then processed through
                                             the treatment system.
                                               A major  consideration in deciding to use carbon ad-
                                             sorption columns was the  ease  of operation and  small
                                             manpower  requirements. Downtime of the treatment fa-
                                             cility was less  than  1% during FY-79. Only  simplistic
                                             problems typical with system start-up were encountered.
                                             Flow  recorders  did  not operate properly, pump  seals
                                             needed replacement and  a faulty solenoid valve required
                                             repair.

                                             Breakthrough

                                               To determine when the carbon in a column requires
                                             replacement, the concentration of DIMP in water being

-------
378     CASE HISTORIES
discharged from a column is monitored. When the DIMP
concentration approaches 50/ig/l, new carbon is ordered.
This level was selected so that the column could still be
used until the trailer containing fresh carbon arrives.
  DIMP was selected as the indicator compound because
previous testing had shown that it was the compound
that "breaks through" the carbon before any of the other
organic compounds.  To monitor  DIMP passing through
the column, samples of  influent  and  effluent  were col-
lected  daily. An  example  of a typical DIMP "break-
through" curve is given in Figure 7.
  Pilot bench studies had suggested that the carbon needed
would be 1.10 Ib carbon/1000  gal of  wastewater. Be-
tween  July  1978  and  June 1979, two columns were
changed. The ratios were 1.90 and 1.29 lb/1000  gal of
wastewater,  respectively. The higher than expected car-
bon usage rates were  attributed to bed siphoning and
subsequent formation of air pockets in  columns packed
with carbon. To overcome this problem a siphon break
was installed.'"
Recharge Well Subsystem
  The recharge well subsystem is composed of twelve 18-
inch diameter wells approximately 100 feet apart. They are
installed within 36-inch diameter gravel-packed  holes.
These  wells  are screened along the full  thickness  of the
aquifer. Treated water is injected  continuously by gravity
unless the level of water in the well rises above a pre-de-
termined level. At that time, a float control valve  closes,
halting flow into  the  well  until the  water level drops
enough for flow to resume.

Impermeable Barrier
  A 1500 foot impermeable bentonite barrier anchored
approximately 2 ft into bedrock was installed between the
dewatering and recharge wells.  It was installed to physi-
cally prevent flow of the groundwater through the aquifer,
isolating upgradient  and downgradient flow  at the treat-
ment system site. This barrier prevents mixing of contam-
inated and treated groundwater.
Monitoring Well Subsystem
  Observation  wells were installed both upgradient and
downgradient of the pilot containment system. They were
small diameter PVC casing screened  within the alluvial
aquifer. Water levels and chemical quality of the ground-
water are monitored periodically at each of these wells.
Evaluation of System Performance
  In June 1979, the performance of the pilot  containment
system was  evaluated to determine its ability to remove
organic contaminants from  the groundwater. If the sys-
tem operated successfully, it would be extended to inter-
cept and treat  contaminated groundwater crossing along
the northern arsenal boundary/"
  The results of the evaluation were:
  (1) The carbon column removed organic contaminants
      from the groundwater. The contaminants could not
      be detected using gas chromatography. Eventually,
      contaminants were detected because the adsorptive
= 500 -
                     REPLACED CARBON
                               REPLACED CARBOI
           10      10      30     40     50      60     9&
                          WEEKS
                       Figure 7.
         Graph of DIMP Breakthrough Curve During
                 Pilot System Operation
      capacity  of the activated  carbon  became  over-
      whelmed. Therefore, successful operation requires
      periodic replacement of the activated carbon.
  (2) Downgradient flow of groundwater was essentially
      unchanged. This latter consideration is as impor-
      tant as the first requirement in a water-scarce area.

Expansion of the North Boundary
Pilot Containment System
  Construction of the expanded containment system was
begun in early 1981. It is scheduled to begin operations in
August of 1981. It consists of 42 additional dewatering re-
charge wells, and monitoring  wells, increased treatment
capacity, and extension of the  bentonite barrier 4500 feet
to the east and 1500 feet to the west of the present pilot
containment system barrier. The  design of this addition
was based upon  the  performance  of  the original pilot
system.<2)
Northwest Boundary Containment System
  As described earlier, DBCP contamination was detected
off site at the northwest arsenal boundary. A containment
system is being designed to address that problem. Present-
ly,  the situation is similar to the  north boundary system
with one exception. Preliminary calculations indicated that
the dewatering wells alone could  effectively  remove con-
taminated groundwater from the aquifer. Two rows of de-
watering wells are planned approximately 800 feet apart.
Therefore, an  impermeable barrier was not included in
design of the system.

SUMMARY AND CONCLUSIONS

  DA has implemented a containment strategy to prevent
off-site migration of  contaminants at  RMA. Initially, a
pilot containment system was constructed along the north
boundary to test the feasibility of this containment ap-

-------
                                                                                    CASE HISTORIES  379
proach. The pilot containment system demonstrated that
the approach:
  (1) Removed organic contaminants of concern to below
     detectable units from the groundwater
  (2) Did not adversely affect the flow and distribution of
     the groundwater downgradient from the system
  As a result, an extension was designed and is being con-
structed to  treat contaminated groundwater along this
boundary. The concept is also being implemented to con-
tain contaminants in groundwater crossing the  north-
west boundary.
  Though this approach works, there are important ques-
tions remaining to be answered:
  (1) The length of time required to run the system must
     be determined. Containment could cost more than
     clean  up in certain situations because of the  costs
     associated in operating a containment system for
     many years. The maintenance associated with op-
     erating a system for many years must also be con-
     sidered.
  (2) The impact on land use options caused by use of a
     containment strategy needs to be addressed. In the
     arsenal's example, sale of land  may be precluded
     unless the  new owners accept responsibility  for op-
     eration of the system.  Use of contaminated land is
     also restricted during the time frame that contamin-
     ants continue to migrate from that land.

 REFERENCES

 1. D'Appolonia Consulting Engineers, Inc., "Evaluation
   of North Boundary Pilot  Containment System, RMA,
   Denver,  Colorado,"  Project  Number  RM79-389,
   July 1979.
2. D'Appolonia Consulting Engineers, Inc., "Conceptual
   Design of the North Boundary Containment System,
   RMA, Denver, Colorado (Expansion)," Project Num-
   ber RM79-480, October 1979.
3. Kolmer, J.R. and Anderson,  G.A., "Part  1— Pilot
   Containment Operation,  Final Environmental Impact
   Statement, Installation Restoration of Rocky Moun-
   tain Arsenal," Department of the Army, Office of the
   Project Manager of Chemical Demilitarization of In-
   stallation Restoration, July 1977.
4. Konikow, L.F., "Hydrogeologic Maps of the Alluvial
   Aquifer In and Adjacent to the Rocky Mountain Ar-
   senal, Colorado," U.S. Geological Survey Open File
   Report 74-342, 1 sheet, 1976.

5. May, J.H., Tompson,  D.W., Law, P.K. and  Wahl,
   R.E.,  "Hydrogeologic  Assessment of Denver  Sands
   Along North Boundary of Rocky Mountain Arsenal,"
   U.S. Army Engineer Waterways Experiment Station,
   January 1980.
            S.G.,  "Digital Model Study of  Diisopro-
   pylmethylphosphonate  (DIMP)  Groundwater Con-
   tamination, Rocky Mountain Arsenal  near Denver,
   Colorado, Progress Report— Phase I,"  U.S. Geologi-
   cal Survey, Denver, Colorado, June 1977.

7. Vispi, M.A., "Report of Findings,  Rocky Mountain
   Arsenal Pumping Test," U.S. Army Engineer Water-
   ways Experiment Station, 1978.

8. Zebell, R.A., "Basin F to the North Boundary, Volume
   I: Geotechnical Definition,  Rocky Mountain Arsenal,
   Denver, Colorado," U.S. Army Engineer Waterways
   Experiment Station, 1979.

-------
    THE EFFECTIVE USE OF RESOURCE RECOVERY IN THE
      CLEANUP OF UNCONTROLLED HAZARDOUS WASTE
        SITES—BASED ON THE CALIFORNIA EXPERIENCE
                                          WILLIAM QUAN
                               California Department of Health Services
                                          Berkeley, California
INTRODUCTION

  One of the most important environmental concerns to-
day is the location and management of uncontrolled haz-
ardous waste sites. Normally, an uncontrolled site is iden-
tified through anonymous tips and/or governmental  sur-
veillance activities. Once the location of such  a site be-
comes widely known there is much pressure exerted by
the news media and the public  on governmental agen-
cies to remove  the hazardous waste from the site as
quickly as possible.
  Very often the most expeditious way of cleaning up a
site is removal of the waste for disposal at a permitted haz-
ardous waste disposal site; this is not necessarily the most
prudent  way of solving the problem. Furthermore, Cal-
ifornia and the rest of the  country are now experiencing
increasing difficulty in siting new hazardous waste  dis-
posal  sites. Concurrently,  permitted  sites are  facing in-
creasing  public pressure to close.  For example,  even after
obtaining  185 permits  the  five-year-old Earthline haz-
ardous waste disposal  site in Wilsonville, Illinois  was
closed as a public nuisance."1
  Therefore, if the limited number of currently operating
hazardous waste disposal sites are filled, the states will be
faced  with frequent incidents of illegal  dumping of haz-
ardous wastes and/or firms going out of business because
there is no facility which can accept their waste.
  In addition, until recently, only surface and ground-
water protection were generally considered in the location
of a hazardous waste disposal site, but now there are  also
concerns about air pollution and  public health effects due
to emissions of hazardous waste  from disposal sites. Re-
cently, several papers'2'3t 4) have addressed the problems of
air quality deterioration and the  control of air emissions
from hazardous waste sites.
  According to a 1980 EPA contracted study(5) of nine un-
controlled hazardous waste sites in the United States, the
remedial action technologies usually  employed  were on-
site containment  and surface/groundwater monitoring,
and waste removal for landfill burial or incineration at
permitted facilities. Only once was resource recovery em-
ployed as part of a cleanup program. In the long run, even
though it usually takes a longer period of time for a site to
be cleaned up through resource recovery, it is probably, in
many instances, the soundest solution. Sometimes it  may
take a joint cooperative effort of everybody concerned for
resource recovery to work. Furthermore, landfill disposal
and/or on-site containment measures may be only tem-
porary solutions.
  Four California case studies of uncontrolled hazardous
waste sites are presented. For the purposes of this paper,
an uncontrolled or abandoned hazardous waste site is de-
fined as  having one or more  of the following charac-
teristrics:
(1)  is no longer used or is inactive and is no longer main-
    tained,
(2)  is a location of unpermitted or improper waste dis-
    posal
(3)  has no known owner, and
(4)  containment and  monitoring techniques are inade-
    quate.
  Hazardous  waste is  defined  in the California Admin-
istrative Code,(6)  as any  "waste material or mixture of
wastes which is toxic, corrosive, flammable, an irritant, a
strong sensitizer or  which generate pressures through de-
composition..."
  Due to requests for confidentiality or to pending litiga-
tions, the names and locations of some of the following
studied sites have been deleted from this paper.


CASE A: CONTRA COSTA COUNTY, CALIFORNIA

  The site was formerly used by its previous owner for the
fabrication of steel.  One of the operations was a galvan-
izing mill, which was used to coat steel pieces with zinc
from a zinc acid solution so as  to protect the steel pieces
from corrosion. Apparently the spent galvanizing solu-
tions were ponded from 1960 until  1975 when the plant
closed. The pond was built in existing marshes where soils
vary from clay to sandy silt. In 1977, the property was
sold. The responsibility for the maintenance of the pond
was then transferred to the new owner.
  During 1977, the pond overflowed. Subsequently, dikes
were constructed to contain future overflows. In the same
year residents registered complaints about the pond and
emissions from the  pond. During 1978, the runoff rate
from the nearby railroad company's fill was periodically
exceeded which led to the pond overflowing once more and
into San Pablo Bay.
  Over the years,  wastes continued to seep from the pond
and some seepage had even reached the Bay although mon-
                                                  380

-------
                                                                                        CASE HISTORIES   381
                                                    Table I.
                                         Summary of Analytical Results for
                                         Contra Costa Steel Plant Lagoon
Parameter
pH
Sulfate
Cadmium
HexavaLent Chromium
Trivalent Chromium
Total Chromium
Lead
Zinc
Total Solids
Units
Units
Mg/Kg(l)
Kg/Kg (1)
Mg/Kg(l)
Mg/KgU)
Mj>/Kg(l)
ME/KR (i)
Hg/Kg(l)
Z (2)
Composite Pond
Water Sample
1.9
15,400 Mg/£
0.62 Mg/U
0.02 Mg/i
3.15 Mg/J.
3.15 Mg/i
0.10 Mg/fc
2,530 Mg/1
—
Pond Bottom
Surface Sedijnent
1.5
51,000
0.85
<0.2
18.6
18.6
190
5,570
51.6
CORE SAWLES
Station 1
0-4" 11-14"
2,1
15,000
0.97
<0,2
37.4
37.4
214
2,910
66.9
2.7
2,980
0.68
r0.2
18.5
18.5
23.9
2,050
83.4
Station 2
0-4" 9-13"
2.4
2,650
< 1.00
<0.2
14.1
14.1
191
416
71.8
3.2
1,260
1.60
<0.2
31.3
31.3
28.1
413
80.5

Station 3
0-4" 13-17"
1.8
109,000
0.88
<0.2
34.2
"34.2
26.3
9,650
57.6
1.9
56,000
4.55
<0.2
45.5
45.5
81.9
15,700
56.9
California
TTLC *
*2 (corrosive)
—
10
50
250
--
50
200
~
EPA
EP Toxic Ity**
i2
—
1.0
—
--
5.0
5.0
—
—
(1) Mg/Kg expressed as Wet Weight

(2) Oven Dry (103°C) Weight Basis

*   TTLC's (Total Threshold Limit Concentration) are from the California Assessment Manual for Hazardous Wastes, p. 66-67.

**  These values are not equivalent to California TTLC but are the same as STLC (Soluble Threshold Limit Concentration) .
                             © Sampling Station
                             Date of Sampling 8/13/80
                     Figure 1.
            Core Sampling Points at Pond
itoring of the ground water, which is approximately three
feet below ground, indicated no contamination.

Remedial Actions

  The remedial actions employed were basically contain-
ment measures, for example,  diking. During these years
there were also proposals to treat the pond wastes for dis-
charge into the sanitary sewage system and San Pablo Bay,
disposal of the waste at a Class I hazardous waste disposal
site(7) or to permit the liquid to evaporate from the pond so
the pond  bottoms and adjacent  contaminated soils could
be excavated and hauled to a Class I site for disposal.
  In 1980, the owners authorized a commercial laboratory
to sample and analyze the pond  water and adjacent soils.
This was done to ascertain the alternatives for deposition
of the pond's waste water and the contaminated soils. As
the analytical report (Table I) indicates, three core samples
(the core samples were collected  at the locations indicated
in Figure 1), a bottom surface sediment sample,  and a
composite water sample were collected and analyzed. Bas-
ically, the results indicate that all samples are considered
hazardous  by  the  California  Department  of  Health
Services per the California Assessment Manual for Haz-
ardous  Wastes® or CAM.
  Subsequently,  the  Department's   California   Waste
Exchange (CWE) became involved. The CWE realized
from the analytical results that the primary chemical com-
pound, zinc sulfate, dissolved in the waste water is used
in the agricultural industry.(9'10)
  Sulfur, assimilated as  the sulfate ion, is considered  a
secondary plant nutrient for protein synthesis  and sulfur
deficiencies exist in California  soils at the eastern foothills

-------
382   CASE HISTORIES
in the San Joaquin Valley, the central coastal range and
the Sacramento Valley. Zinc, a micro-nutrient, needed for
terminal plant growth, is assimilated as zinc ions. As with
sulfur, zinc is often found deficient in California soils.
  A nearby firm,  Veale Tract Farms,  when contacted by
the CWE, expressed an interest in the three million gallons
of pond water for use as a micronutrient on its marginal
soils. Veale Tract Farms proposed to dilute the pond water
to an appropriate level  before application to their soil.
Thus far, of the two proposals received by the owners, one
for hauling and disposal  at a Class I site and the other for
hauling and application at Vale Tract Farms, the latter was
less costly by $700,000.
  As for the pond bottom surface sediment and the con-
taminated soil, they may also be beneficially used by roto-
tilling at a 1:100 ratio with native soils, so as to reduce the
lead content to an acceptable level. In this way all of the
waste on-site could be recycled.
  At this point, the most economic identified remedial ac-
tion technology is still resource recovery. Finally, the Cal-
ifornia  Department of Health Services has estimated that
cleanup of the site through disposal at a Class  I site can
cost as much as $29,000,000.

CASE B: CONCORD, CALIFORNIA

   Prior to 1964 the site was owned and used by a salvage
company operator. In 1964, the site was sold but the sell-
er retained  one half  of the  mineral rights to the eastern
portion of the site where the salvage operation had been
located. Apparently,  the operator had used the site over
the years to dispose of 60 to 70 tons of material from the
nearby  Allied Chemical's Union Collier Plant.  The piles
of material  disposed are believed to have been spent coke
filters used in sulfuric acid production (see Figure 2).
   In 1972 the U.S. Navy acquired the above site as part of
a two mile buffer zone around a munition's area. During
1980, the site earned to the attention of the California De-
          California Department of Health Services

                        Table n.
      Summary of Analytical Data for Solid Waste Piles at
            Allied Chemicals' Union Collier Plant
                         Mg/kg
   Parameter

   PH
   Ag
   As
   Ba
   Bi
   Cd
   Co
   Cr
   Cu
   Fe
   Hg
   Mn
   Mo
   Ni
   Pb
   Sb
   Se
   Sn
   Sr
   Tl
    V
   Zn
   Br
   Rb
    Y
   La
   Ge
   Th
   Te
   Ti
Coke Pile
fl

  3.1
26 ±14
Coke Pile
12

    2.2
 89 ±19
251 ±44
 31±8
3.800

58 ±20
 14±6
 16±8
1,600

  972

  244

36±24
77 ±10
14±4
33±6
37 ±10
19±18
  552
 48 ±8
 1,670

 23 ±16
 4,120

 7,030

 23±6
 48±8
 44±8
                                             Calif.
                                             TTLC
                              50
                            1,000

                              10
                              150
3,500
 200
  50
1,000
  10
   50
  240
  200
              1,550
              113±44
                        Figure 2.
                                                            Note: The analytical error for metal determination is approximately ± 10%. unleu otherwtM
                                                               indicated
partment of Health Services when it was identified as a
site that had received industrial wastes in the U.S. Con-
gress* EckhardtSurvey.
  Preliminary results (see Table  II) on samples collected
and  analyzed  by  the California  Department of Health
Services indicated  that the coke piles were acidic, pH 2.2
to 3.1, and contained high levels of metals: lead,  1600 to
4120 ppm; arsenic, 251 ppm; selenium, 972 to 7,030 ppm;
and tellerium, 552  to 1550 ppm. Since these concentrations
of heavy metals exceeded the Total Threshold Limit Con-
centrations in the Department's CAM for lead and arsenic,
the coke piles are designated as hazardous wastes.
  The site is presently unsecured; there are no fences and
based on preliminary information there may be leaching
of heavy metals into the  surrounding soil. In addition, the
piles of waste have been cited by  the California Regional
Water Quality Control Board (CRWQCB) as a source of
contaminated runoff to tidal and  navigable waters. How-

-------
                                                                                       CASE HISTORIES    383
ever, since the site is not adjacent to a residential area
there appears to be no imminent threat to human health.

Remedial Actions

  During early 1981, the Department requested the present
owner, the U.S. Navy, to perform extensive sampling and
analysis of the coke piles and the adjacent soils to  deter-
mine the average concentration of the  contaminants  in the
coke piles and the extent of soil contamination or  leach-
ing. Additionally, the  U.S. Navy was requested by the
CWE to analyze the coke piles for sulfur.
  Two California oil refineries contacted by the CWE ex-
pressed interest in the coke piles as a low grade fuel  pend-
ing results of the sulfur concentration analyses. The 60-70
tons of coke in the piles is to be blended by the refineries
with their coke so as to produce a low-grade fuel with ac-
ceptable heavy metals  and sulfur concentrations, in this
context, the coke piles would be a very small fraction of
the final fuel.

CASE C: OAKLAND, CALIFORNIA

  The site near the Oakland Coliseum was formerly used
by Pacific Oxygen, which manufactured acetylene. Acety-
lene is produced from a reaction between calcium carbide
arid water.0 1}
CaC
2H20
                     »Ca(OH)2 + 134kcal/mole.
  The by-product or waste is hydrated lime, Ca(OH)2.
  According to  the  local health agency,  the  Alameda
 County Health Department, the site had been inoperative
 since 1976 but prior to that Pacific Oxygen used the site to
 dispose of their  acetylene sludge, CA(OH)2 (See Figure
 3). When the site closed down 1,000-2,000 tons of acety-
 lene sludge had accumulated.
  In August 1979 a "concerned citizen" contacted the De-
partment via the Alameda County Health Department to
report that children had ventured onto the unsecured site,
the chain fence had been knocked down, to play in the ace-
tylene sludge piles. The site also had an open sump 6 ft
x 8 ft x 3 ft deep which was three-quarters full of liquid.
  Subsequently, samples of the acetylene sludge and sump
liquid were collected and analyzed by the California De-
partment of Health Services. The results (see Table III)
indicate that although the acetylene sludge  contain rela-
tively low amounts of toxic heavy  metals,  the pH was
high, 12.3, indicating that the sludge is corrosive and there-
fore a hazardous  waste under California regulation. The
pH of the sump liquid was  11.04; therefore it is a likely
irritant, which makes the liquid also a hazardous waste.
  In April, 1980,  when the City of Oakland  obtained title
to the Pacific Oxygen site, the Department's CWE was
contacted for assistance in recycling the wastes that were
on the site. The CWE determined that the wastes may be
recycled in several ways:
 •as a soil amendment,
•as an agent to precipitate heavy metals from waste waters
  at a sewage treatment plant,
•to formulate a paving material for road construction.
  Eventually a  local paving company,  Gallagher & Burk,
Inc., was located  to recycle  the wastes. The wastes would
be mixed with rock and gravel to improve the sand equiv-
alent of gravel for use as a road base.

Result

  1,000-2,000 tons of lime  sludge and liquids were recy-
cled at no cost to  the City of Oakland. This was the single
largest waste transfer the CWE facilitated  during 1980.
With the sludge and liquids removed, the City of Oakland
                                        California Department of Health Services

                                                    Table III.
                                    Summary of Analytical Results for Pacific Oxygen
                                                      Mg/kg
      Parameter
      pH
      As
      Fe
      Mn
      Ni
      Sr
      Zn
      Ti
      Y
            Sump
            liquid*
            11.04
                5
            1,033
               34
              122
                                     Acetylene Sludge**
                                     Pilel
                                     V below surface
                                     12.3±0.1
                                     823 ±81
                                      44±13
                                     186±18
                                      21 ±7
             Acetylene Sludge**
             Pilel
             1 Vz' below surface
             12.3±0.1
              26±5
             139±19
              20 ±16
              38 ±8
              29 + 7
Calif.
TTLC
                                                          72 + 52
                                                          10 + 5
                                                                                              50
                                                                                      200
                                                                                      200
       •Results were reported on 9/12/79
      "Results were reported on 7/17/80
      Note: Only results for all metals above the detection limit of the X-ray fluorescence spectrometer are reported.

-------
384   CASE HISTORIES
is now proceeding with plans to develop the site into a park
and recreation area.

CASE D: NORTHERN CALIFORNIA

  The owner had apparently been using  the site for un-
authorized storage and treatment (Figures 4 and 5) because
an investigation by the Department  of Health Services'
surveillance  and enforcement staff in mid 1980 revealed
that the owner has never applied for a hazardous waste fa-
cility permit. The owner had been, for a period of time,
hauling in drums of ink waste, ink wash, glues, waste oils
and solvents for  storage until he could  find a recycler.
Otherwise, the waste was hauled to a Class I site.
  Although  the site was used to store flammable waste,
the Department of Health  Services staff  saw no fire ex-
tinguishers and safety showers. Because of these and other
deficiencies the State Fire Marshal  believed that a fire at
the site would cause the closing of an adjacent freeway.
  The above site was also unsecured  and  unpaved.  From
time to time, there had been numerous spills of hazardous
wastes.  In addition, glues, ink waste and  ink wash wastes
were disposed  into  unpermitted  plastic-lined  pits.  An
analysis of contaminated soil by  the  California Depart-
ment of Health Services in mid 1980 showed relatively high
levels of chromium, copper, lead and zinc. Finally, there
was no containment provisions for any spills which could
threaten the pollution of surface and ground waters.

Remedial Actions

  In late 1980, the CRWQCB issued a Cleanup and Abate-
ment Order (CAO) to  the  site owner after  inadequate
responses to informal  requests for cleanup. The order re-
affirmed previous informal requests to have all removed
materials (soils) disposed at a Class  I disposal site. The
second part of the Order required that containment berms
and impermeable surfacing be constructed for all  tank
drum storage and handling areas.
                       Figure 4.
                        Figure 3.
                       Figures.

  Subsequent  inspection by this  Department and other
governmental  agencies in early  1981 indicated that the
above storage/treatment site posed  a  clear and present
threat to public health  and the environment. Drums of
hazardous  waste were  seen deteriorating. Accordingly,
the Department urged the local authority to proceed with
the enforcement and cleanup of the site pursuant to the
California Health and Safety Code.m
  To expedite the identification of the wastes, firms whose
hazardous  waste were  observed  were contacted.  Upon
learning of the situation the Department's CWE provided
names of recyclers who might be of help in the site clean-
up. Eventually, the CWE with the help of recycling con-
tractors, facilitated the recycling of 900 of the 2100 drums
of flammable waste. In addition, 607 drums were recycled.
However,   150 drums of fiberglass  and foaming  resins

-------
                                                                                       CASE HISTORIES   385
were not recycled primarily because the producer insisted
they were cleanup residues and therefore they be disposed
of at a Class I site as originally intended.
 ROADBLOCKS AND REMEDIES

  Once an abandoned or uncontrolled hazardous waste
 site has been identified, an immediate qualitative assess-
 ment of the problem site must be made. Specifically, in-
 formation (i.e.,  the presence of toxic and/or flammable
 vapors and toxic heavy metals) must be quickly gathered to
 ascertain if there is imminent hazard to human health and/
 or the environment. If an imminent hazard exists, then
 governmental agencies working with the responsible own-
 er, if there is one, should immediately construct contain-
 ment measures or even temporary ones, if necessary. Re-
 gardless whether or not an imminent hazard exists, the site
 should be secured from public access.
  Once the problem site becomes public knowledge, the in-
 volved government agencies should educate the media and
 concerned groups about  the  situation. If in the opinion
 of the governmental agencies, resource recovery can be
 used to clean up the site, the media and  all concerned
 groups should be so informed. The benefits of resource re-
 covery over land disposal should be extolled. Education
 and  allowance of input from concerned groups by govern-
iment agencies will help to allay public distrust and anxiety.
  After an uncontrolled site has  been identified,  to assess
 whether an imminent hazard exists, sampling and analysis
 should immediately commence and almost concurrently a
 comprehensive program for sampling and analysis should
 be prepared. At this  point,  the resource recovery staff
 should be consulted for recommendations on the param-
 eters the  sampling  and  analytical  programs should be
 addressed.
  Until recently, the Department's CWE usually became
 aware of a problem site only after sampling and analysis
 of the site had been performed.  Any lack of communica-
 tion  will unnecessarily delay the recycling of the wastes be-
 cause in many instances  of the long turn-around time
 (months) between sampling  and the  completion of the
 chemical analyses and the lengthy time needed in many in-
 stances to locate recyclers. The  Resource Recovery staff
 should be involved as soon as possible. The time to locate
 recyclers may be significantly reduced if a Resource Re-
 covery staff has access to a computer-based waste ex-
 change. Instead of a quarterly catalog of wastes available,
 which is now distributed by most clearinghouse waste ex-
 changes, a daily update would be available from the com-
 puter data base to  all interested parties. Presently, the
 CWE is planning to complement its program through the
 assistance of a computer-based waste exchange.
  In addition, instead of depending on a response to a cat-
 alog  or computer listing of available wastes, the Resource
 Recovery staff should also actively engage in the matching
 of the waste with  potential  recyclers.  Finally, because
 sometimes wastes are recycled over great distances (i.e.,
 1000 miles) there should be attempts to locate recyclers on
 a regional instead of a State basis.
REMEDIAL ACTIONS AND FLOWCHART

  Any remedial action program for an uncontrolled haz-
ardous waste site  should consider the possibility of clean-
up  through resource recovery.  The following proposed
schematic (Figure 6) for maximizing resource recovery of
hazardous  wastes from an uncontrolled site incorporates
the above remedies to roadblocks. Just before efforts  are
initiated to qualitatively identify imminent human health
and environmental  hazards, the Resource Recovery staff
should become involved.  This  early inclusion of the  re-
source recovery staff allows additional time for the loca-
tion of potential recyclers.  It has been the experience of the
CWE that  approximately  2 weeks is frequently necessary
to determine whether there is a potential for resource  re-
covery.
  The resource recovery potential is assured when there
are known and willing recyclers. However, this is not  the
usual case and the potential for resource recovery, in many
instances depends on the vision which is basically a func-
tion of the knowledge and experience in analytical and
industrial chemistry of the  Resource Recovery staff.
  To maximize the resource recovery potential of a waste,
the literature should be reviewed for uses  for the chem-
ical constituents in the waste. Reference texts such as  the
Encyclopedia  of  Chemical  Technology(l3)  Condensed
Chemical Dictionary(l4) Chemical Sources(l5) Directory of
Chemical Producers in the C/.S.,(16) California Manufac-
turers Register^ etc., have been found to be useful  by
the CWE. The first four books identify the uses for a par-
ticular chemical. The last book, which is updated annually,
lists companies by geographic region, industry, and SIC
number and therefore  permits identification of potential
recyclers for the hazardous chemical waste.
  Oftentimes, the waste cannot  be recycled as is, but must
be treated  by  biological, chemical and/or  physical tech-
niques to  reduce  or  remove  certain  chemical  constit-
uents before recycling is possible, economics notwithstand-
ing. If this is the case, then the feasibility of recycling of a
waste will also depend on the treatment costs. In the end a
waste is only recyclable when it is both technologically and
economically feasible.

CONCLUSION

  California has proven through its recent experience that
resource recovery can be an effective tool for the cleanup
of uncontrolled hazardous waste sites. Resource recovery
means recovery of raw materials and energy. In addition,
land is saved for  future generations. As the problems of
siting new hazardous waste disposal sites increases and re-
source recovery becomes more publicized remedial action
programs  for uncontrolled hazardous waste disposal  site
will consider resource recovery as a matter of course.

ACKNOWLEDGEMENTS

  The author wishes to express special thanks to Ms. Alyce
Tom for the typing of this manuscript and  to my colleag-
ues who provided me assistance and information  for this
paper.

-------
386   CASE HISTORIES
                                                     Location of
                                                       Site
                                                     Educate and
                                                  Inform Concerned
                                                      Groups	
                                                     Qualitative
                                                    Identification
                                                        of
                                                      Hazards
               	/ On-site Visit/

               --—-^Historical Information/

                 	/ Qualitative Sampling  /
                   / and Analytical Program/
                                                     Containment
                                                    of Hazards and
                                                    Securing of Site
                                                         _L
  Identification
and Amount of Waste
                                                      Identification
                                                         of
                                                      Treatment
                                                      Technologies
                                                                       Historical Information/

                                                                    ~~Y Comprehensive Sampling /
                                                                       and Analytical Program/
                                                                    •—/ Llterlature /
	I Manufacturers of      /
    / Treatment Technologlces /

     ^Identlfl
     Recycler
                       titled Potential
                                                     Figure 6.
                                       Flowchart for Resource Recovery of Wastes
                                              From an Uncontrolled Site
 BIBLIOGRAPHY

  1.  "A  permitted disposal site is closed down", Chem-
     ical  Week, Vol. 129, No. 4, P. 62, July 22, 1981.
  2.  California Air Resources Board,  "Determination of
     the  Air  Quality  at the Abandoned Waste Disposal
     Site  in Fullerton, California," Nov., 1980.
  3.  Thibodeaux, L.J., "Estimating The Air Emissions of
     Chemicals from Hazardous Waste Landfills,"  /. of
     Hazardous Materials, 4, 1981, 235-244.
  4.  Shen,  T.T., "Control Techniques for Gas Emissions
     from Hazardous Waste Landfills," JPACA, 31,  1981,
     132-135.
  5.  Neely, N., Gillespie,  D.,  Schauf, F., and Walsh, J.,
     Remedial Actions at  Hazardous Waste Sites: Survey
     and  Case Studies, EPA-430-9-81-005, January 1981.
  6.  California Administrative Code, Title  22, Social Se-
     curity  Division 4. Environmental Health, Chapter 30,
     Minimum  Standards  for  Management  of Hazardous
     and  Extreme   Wastes,  Section 66088, Register 79,
     No.  19, 1979.
  7.  Franks, A.L., California State Water Resources Con-
     trol  Board's Waste Discharge Requirements for Non-
     sewerable Waste Disposal to Land, July 1980.
         8.  California Assessment Manual for Hazardous Wastes,
            June 1981.
         9.  Soil Improvement Committee of the California Fer-
            tilizer Association,  Western Fertilizer Handbook, 6th
            Edition, 1980.
        10.  Berg, G.L., 1981 Farm Chemicals Handbook.
        11.  Kirk-Othmer Encyclopedia  of Chemical Technology,
            Third Edition, /, 1978, 203-210.
        12.  California  State  Department  of  Health  Services,
            Health and Safety  Code, Hazardous  Waste Control
            Law, Div. 20, Chapter 6.5.
        13.  Kirk-Othmer Encyclopedia  of Chemical Technology,
            Third Edition, 1978.

        14.  Hawley, G.G., The Condensed Chemical Dictionary,
            9th Edition, 1977.

        15. Directories  Publishing Co., Inc.,  Chem. Sources—
            U.S.A. 1980 Edition.

        16. SRI,  International, 1981 Directory of Chemical Pro-
            ducers United States of America.

        17. Times Mirror  Press,  1981  California Manufacturers
            Register, 34th Edition.

-------
                THE GENERATOR'S LIABILITY FOR PAST
             HAZARDOUS WASTE DISPOSAL PRACTICES
                                         JOSEPH M. MANKO
                                                  AND
                                            MARC E. GOLD
                                  Wolf, Block, Schorr and Solis-Cohen
                                       Philadelphia, Pennsylvania
INTRODUCTION

  One need only pick up any newspaper to be reminded
that improper disposal sites were used by industry as haz-
ardous waste burial grounds over the past decades. Most
new federal, state and local legislation and  regulations
deal prospectively with the disposal of hazardous waste.
They leave for the courts the difficult task of determining
who will bear the legal responsibility for cleaning up the
hidden horrors which threaten public health and natural
resources. In this paper, the authors review the many com-
plex technical and legal issues  involved in assigning and
apportioning  this  liability and provide some  practical
guidance to hazardous waste generators  on  future pre-
cautions.
  In discussing the potential liability of generators for past
hazardous waste disposal practices,  the authors are not
focusing on owners or operators of hazardous  waste treat-
ment, storage or disposal facilities or even the haulers who
brought the waste to such facilities. In most  cases,  their
liability is determined on a statutory basis, under certain
circumstances going back to the Refuse Act of 1899'" and
more recently  the Resource  Conservation and  Recovery
Act ("RCRA")(2) and the Comprehensive Environmental
Response,  Compensation and Liability  Act  ("Super-
fund").(3) What the authors do address is the legal respon-
sibility of a typical hazardous waste generator who hired
someone to haul hazardous waste to an off-site disposal
facility.

BACKGROUND
The "Throw Away Mindset"
  Just a few years ago, before the recent hazardous waste
catastrophes raised the public's consciousness  level, the
industrial mindset with respect to hazardous waste disposal
was generally "out of sight, out of mind."  Most  industries
assigned their purchasing agents the task of arranging for
the disposal of industry's unwanted product: its hazardous
waste. Purchasing agents were,  for the most  part, un-
familiar with the nature of the wastes or  the proper dis-
posal methods required. Faced with relatively insignificant
disposal costs, even for the most toxic wastes and relatively
limited  technology or  interest in recycling such  wastes,
arrangements for the disposal of hazardous  waste were
routinely made, with little consideration as to its ultimate
disposition. Rarely would a generator take the time to es-
tablish and apply transporter or disposer selection criteria
other  than price. It was the atypical generator who even
considered examining the transporter's equipment, inquir-
ing about its reliability or reputation, or even determining
from regulatory authorities whether it had valid permits to
conduct its business or any enforcement proceedings pend-
ing against it.
  Perhaps most important of all,  many generators never
even inquired as to the transporters'  intended  disposal
location and few, if any, ever followed  a waste shipment
to assure that  it reached the proper destination.  Other
than in those rare instances where a generator had  ample
land adjacent  to its  plant on  which  to  dispose  of  its
waste,(4) little care was taken to make certain that hazard-
ous wastes were disposed of properly.

The Love Canal Syndrome
  The tragedies that beset the community of Niagara Falls,
New York, captured the nation's attention and have held
it for the past  several years. It was impossible to escape
reading  or  avoid seeing on television  stories  about the
anachronistically labelled "Love Canal." There followed
soon on the heels of this disclosure the discovery of other
hazardous waste nightmares  such as  the  Valley of the
Drums in Kentucky, the dumping of PCBs along the road-
side in North Carolina, and the frequent midnight dump-
ing of hazardous waste in the back woods of New England.
  Today, every state has its own litany of hazardous waste
horrors—monuments to unbridled industrial  waste  dis-
posal. And as  if these hazardous  waste exposures  to the
soil, surface and groundwaters from these  incidents were
not enough, the fire and explosion of chemical wastes
stored in Elizabeth, New Jersey, lifted into the air for
miles  around  toxic fumes deadly enough to  threaten
millions of people with immediate harm. The  notion be-
came  aware that there were ticking timebombs, on and
above the ground  which made everyone vulnerable to
hazardous waste impacts from a variety of environmental
pathways.

Enter the Government
  As  is so often the case, government rushed  into  the
breach with new legislation and regulations. The narrow
focus of the Toxic Substance Control Act ("TSCA")(5)
with its  ambitious regulatory  program requiring preregis-
tration of new chemical substances and  the lengthy de-
                                                   387

-------
388    LIABILITY, LEGAL & PUBLIC ISSUES
lays in implementing RCRA—the two major federal leg-
islative efforts in 1976—demonstrated the need for further
legal authority to specifically address past problems. Sim-
ilarly, as states prepared to accept delegation of various
federal regulatory programs, the need to adopt new state
hazardous  waste laws and regulate  the siting  of  new
hazardous waste facilities became evident.
  Although hazardous waste legislation and the educa-
tional benefits of the implementation of such programs
will prove  to be helpful  in the  future, there was still no
direct mechanism for providing the requisite funds to clean
up the hazardous waste disposal  sites which  continue to
contaminate the environment. The  issue then shifted to
one of economics which was compounded in large measure
by the insolvency of most illegitimate transporters (i.e.,
the midnight dumper), the nonaccountability through in-
solvency or disappearance of the owners and operators of
improperly run sites, and an overlay  of extremely  complex
legal and scientific problems.
  Congress finally addressed these issues, albeit less clear-
ly than anticipated, by the passage of Superfund which
was  signed into law in  December,  1980. The approach
adopted was to establish  a $1.6  billion fund to be used by
the federal government  to respond to  the most serious
hazardous waste disposal site problems on a priority basis,
88% of which would be raised by a tax on the petroleum
and chemical industries starting April  1, 1981. Since even
this ambitious undertaking would be inadequate to clean
up all such sites, it  was left for the legal system, through
the application of statutory and common law principles,
to develop standards to determine and apportion liability
among the various participants in the hazardous waste dis-
posal chain.

RESPONSIBILITY FOR PRIOR PRACTICES
  Typically, a hazardous waste disposal site will be dis-
covered  where the  transporters involved and site owner
and/or  operator are incapable of responding to  environ-
mental problems, either because they are bankrupt or have
fled  the jurisdiction. Combining  this  fact with the pub-
lic's jaundiced view of industries, as having amassed great
profits  from their  manufacturing activities  while gener-
ating and  improperly  disposing  of tons of hazardous
waste,  quickly results in pressure to place the clean up
responsibility on the generator.(6) Support  for  this ap-
proach  is founded  on the philosophy that industry can
either internalize the clean up costs, adversely impacting
only the generator's shareholders,  or  externalize them
through higher costs of goods to the consumers.  Industry
counters this view by suggesting that the public  bear the
expense  through general tax  revenues, like  some of the
present proposals to clean up the Three Mile Island nuclear
plant, since it was the public at large that created the de-
mand for and presumably benefitted from the products
manufactured.
  As this debate on  the liability issue continues, the public
has grown more distrustful of corporate officials and with
the slow pace at which governmental  officials have been
able to  attack these problems—both technologically and
fiscally—the public's confidence in their ability to rescue
them has also waned. At the same time, scientific methods
have developed the capability to determine down to one
part per trillion the presence of contaminants in the en-
vironment. It is in this  setting that  the battle lines have
been drawn pitting against each other teams of experts to
determine the extent of needed clean up and the question
of who should bear the legal and financial responsibility.
Such determinations will not come swiftly.

COMMON LAW LIABILITY
The General Rule
  Where a generator directly disposes of hazardous waste,
rather than hiring an independent contractor to remove the
waste for proper disposal, his liability is more easily de-
termined under  the various statutory  and common law
theories. Thus,  it will be difficult for a generator of haz-
ardous waste to escape liability where it can be shown that
he dumped the waste at a site which was known or which
should have been known to be improper.  Since most gen-
erators did not dispose of their own wastes, this area will
not be discussed further; instead, the more common pat-
tern will be emphasized: the generators who hired inde-
pendent  contractors to remove  hazardous wastes  from
their plants for off-site disposal.
  The  general common law rule is that an employer is
not liable for the acts  of its independent contractor.*7*
This rule is based on traditional contract principles where
one is  free to shift responsibility to another independent
actor who  is competent to  perform the  task. However,
over the years the courts have carved  out important ex-
ceptions to this rule resulting in significant limitations on
its applicability.

The Exceptions to the General Rule
NEGLIGENTLY HIRED CONTRACTOR. First, a gen-
erator  will be held directly  liable for the  acts of its in-
dependent contractor  which cause physical harm to  third
persons if the generator has been negligent in, or fails to
exercise reasonable care over, the hiring of a skillful con-
tractor.(8) The applicability of this exception will depend
upon the generator's  compliance with an  evolving set of
standards dealing with the degree of investigation of the
transporter's competence, reputation and license status.
  In making an assessment of the independent contractor-
transporter, some predictable questions arise:
   (1) Did the hauler have a prior record of violations?
   (2) Had there been adverse press about the transporter's
     practices?
   (3) Were the transporter's prices significantly lower than
     his competition?
   (4) Was the transporter's equipment in good operating
     condition?
   (5) Did the generator check the transporter's references?
   In short, liability may be established upon this principle
where  it can be shown that  the generator  knew or should
have known of deficiencies  in the transporter's qualifica-
tions. Assuming that  the generator is found to have con-
ducted a sufficient investigation of the independent con-
tractor selected to remove the hazardous waste or that the

-------
                                                                  LIABILITY, LEGAL & PUBLIC ISSUES     389
generator can show that even had such an investigation
been conducted, it would not have disclosed any negative
information, the generator has only addressed the thres-
hold issue in attempting to escape liability.
STRICT LIABILITY. The second exception to the inde-
pendent contractor rule involves principles of strict liabil-
ity for ultrahazardous or abnormally dangerous activ-
ities/9' This theory is premised on the need to find that
the generation of hazardous waste is "abnormally danger-
ous" and involves such a high degree of risk that no mat-
ter who is hired to remove the hazardous waste and  no
matter how extensive the investigation of the independent
contractor may  have been, the generator should be held
liable for any harm caused by the wastes.
  The traditional example of an ultrahazardous activity is
the manufacture of dynamite, which has been found  by
the courts to be so abnormally dangerous that  the manu-
facturer remains liable for harm caused by the dynamite's
explosive  characteristics. Concluding that an activity or
material is ultrahazardous and thereby imposing a stand-
ard of strict liability merely makes a  societal judgement
that while the  conduct or material itself should  not  be
banned, it carries such a high degree of risk that the pro-
ducer must be held absolutely accountable for any result-
ing harm.
  Much of the recent  legislation has attempted to define
retroactively certain activities involving hazardous waste as
"ultrahazardous", or "abnormally dangerous."(10) This
raises a myriad of questions as to whether the waste  be-
comes ultrahazardous at the time it is improperly disposed
of, or whether it retains an ultrahazardous character from
the point of its generation. This area is extremely complex
and raises such  fundamental  issues as  the constitution-
ality of the retroactive application of statutory declara-
tions and the precise factual  determinations necessary to
form the underpinnings of a plaintiff's case.
DIRECT LIABILITY FOR A PECULIAR RISK. If  the
generator properly selected an independent contractor and
principles of strict liability do not apply, he may still be
directly liable for harm caused by the hazardous waste if
it posed a "peculiar risk" against which special precau-
tions should have been specified but  were not.(ll) Here,
much will depend upon the  nature of the waste and  the
specific arrangement between the generator and the inde-
pendent contractor as to the recognized peculiar risks and
the concomitant precautions that were specified.
  Information  about the hazardous nature of the waste
and the precautions necessary to be taken in handling it
should have been communicated to the independent con-
tractor if the generator is to discharge its obligations in
this area. However, some of the many unanswered ques-
tions include determinations  of whether the risk was
''peculiar" and the specified precautions were reasonable.
VICARIOUS LIABILITY. Even assuming the generator is
able to escape direct liability, he still faces the possibility
of;being  found vicariously liable. For example,  even if
special precautions were identified by the generator but  the
independent  contractor  failed to  implement  them and
harm  resulted,  the generator may be held  vicariously
Gable. (12>
  Similarly, the generator may be vicariously liable for a
nuisance or trespass caused by the independent contractor
depending  on the extent of the generator's knowledge or
the foreseeability  of  such results.(13) Until more courts
address these issues, the standards of common law liabil-
ity will still be evolving  and the establishment of direct or
vicarious liability will turn on the particular set of facts.
Possible Defenses to Common Law Liability
  Assuming  that a court would find a generator liable
under any one or more of the common law rules described
above, there are defenses that may be available to avoid
generator liability.
1.  Where negligence is an element of the cause of action
  against  the  generator,  contributory  or comparative
  negligence may be asserted as a defense. For  example,
  where the operator of the disposal facility that wrong-
  fully  accepted the waste sues  the generator for the re-
  sultant harm, the generator could point to the consent
  of the operator or other acts in an attempt to shift the
  liability.
2.  Lack of causation linking  the generator's waste to the
  site or to the harm will most certainly be raised by gen-
  erator-defendants. Here  the generator could show that
  although the waste was improperly disposed of it did not
  create harm  or  even  the likelihood of harm, that the
  likelihood of harm  was not foreseeable, that the waste
  was not a substantive  factor in causing the harm or that
  the waste cannot even be traced back to the  particular
  generator.
3.  The statute of limitations  for the cause of action  may
  already have run. The key question here is when did the
  cause of action against the  generator arise: Was it when
  the disposal took place, the subsequent discovery of the
  waste site, or the subsequent injury? Often the type of
  action brought will define  the length of  the applicable
  statute of limitations and the court's answer to the
  above questions will determine the number of years that
  have elapsed prior to suit.
  Finally,  assuming that liability is established,  there are
significant  questions as to the measure of damages and
the way in which such  damages are to be apportioned.
These issues  may turn on the extent to which cleanup is
necessary and the value of the property before  the injury
as compared with its present  value. In addition,  there
are extremely critical issues to resolve as to the apportion-
ment  of liability among many defendants, methods  of
contribution, and joint  and several  liability. Answers to
each of these questions may dictate litigation posture,
strategies and settlement possibilities.

FEDERAL LAW ENFORCEMENT

  Prior to the adoption of Superfund, the federal  gov-
ernment had proceeded against hazardous waste genera-
tors for past practices primarily under the emergency pro-
vision of RCRA,  Section 7003,(14) which was amended in
1980 to require the government to prove only that an immi-
nent and substantial endangerment to public health and
welfare  may exist.(15) To date, all but one court has held

-------
390    LIABILITY, LEGAL & PUBLIC ISSUES
this provision to be jurisdictional only and the courts have
required the. government to prove its entitlement to relief
under the substantive standards established by the federal
common law of nuisance."" Although the courts have not
yet held hazardous  waste generators  liable under  this
RCRA provision for  their disposal of hazardous waste
through independent contractors, there are likely to be
many cases filed by the federal government on this theory
as well as by affected landowners under similar common
law theories."7'
  Superfund was heralded as the tool which the federal
government would use in  the future to impose strict lia-
bility on hazardous  waste generators  for  past  disposal
practices. Conceptually this is analogous to the standard
of liability in the products liability area.
  Congress relied on the standard of liability established in
Section 311 of the Clean Water Act"8' for establishing
strict liability. It is important to recognize that "hazardous
substances,"  the term used under Superfund, is  very
broadly defined"" to include hazardous pollutants under
Section 112 of the Clean Air Act,'20' hazardous and toxic
substances  under Sections  304, 307 and 311 of the Clean
Water Act/2" toxic substances  under   Section  7  of
TSCA,(22)  and hazardous  waste under Section  3001  of
RCRA.'23'  In  addition,   Section  102  of  Superfund'24'
authorizes the Administrator of EPA to list additional sub-
stances as  hazardous not already  included within  this
broad definition. It is obviously all inclusive.
  The first requirement imposed under Superfund was the
reporting of sites not registered  under RCRA which  had
been used as hazardous waste disposal facilities.'25' Based
on these reports, EPA will  be able to identify and evaluate
thousands of potential sites for priority cleanup.
  Included within the class of persons covered under the
liability section of Superfund are operators, owners, trans-
porters and generators.'2*1 The exposure is quite broad cov-
ering the costs of  removal and  remedial  action  incurred
by the government or other persons consistent with the
provisions  of the  National  Contingency Plan.'27' These
costs would include  not  only  site investigations,  but
measures ranging up to and including  the  temporary or
permanent relocation of residents and the provision of al-
ternate water supplies, as well as reimbursement for dam-
age  to natural  resources  under certain  circumstances.
There is a significant requirement limiting the extent of site
cleanup under Superfund to actions that are cost-effective
which  takes into account  the limits of the fund and the
need to respond to other priority sites. However, that lia-
bility for personal injury claims was specifically deleted in
a compromise effort to pass Superfund.
  The defenses available under Superfund are intentional-
ly very limited, covering releases of hazardous  substances
caused solely by acts of God,  war and a third party not
in a direct or indirect contractual  relationship  with  the
generator. In  order to come within this last defense,  the
generator must show by a preponderance of the evidence
that he used due care, taking into consideration the nature
of the waste, and took precautions against foreseeable
acts  of third parties  and the consequences of such acts.
The  true scope of the Superfund defenses  will have to
await judicial review on a case by case basis before their
availability can be fully evaluated.
  As indicated,  Superfund would appear to foreclose a
generator from asserting many of the common law de-
fenses previously described; however, there is serious ques-
tion as to whether the retroactive application of this statute
will  withstand constitutional challenge. Also, while the
liability provisions of Superfund purport to be immed-
iately effective, many of the statutory prerequisites, such
as revisions to the National Contingency  Plan, the need
for state participation and the establishment of a priority
list, have not yet been implemented, so that a court may
be persuaded that asserting liability under Superfund may
be premature.
   Also, President Carter's Executive Order08' delegating
much of the authority to various federal agencies, primar-
ily EPA, is coming under close scrutiny  by the new ad-
ministration. On the other hand, the government will ar-
gue that Superfund was intended to be used as an immedi-
ate stopgap measure to clean up sites before they create
human disasters and that relief need not await the comple-
tion of the regulatory process or strict adherence to pro-
cedural preconditions.  Obviously, this is another issue
which must await judicial interpretation.

STATE LAW ENFORCEMENT AND
RECENT DEVELOPMENTS

   An example of the changing legal theories of liability
potentially applicable to hazardous waste generators is no
more evident than in Pennsylvania. Until September, 1980,
the  Commonwealth  of Pennsylvania relied primarily on
the Clean Streams Law'29'  which regulates  the pollution of
surface and groundwater essentially through a permit pro-
gram. This law provides for vigorous enforcement, abate-
ment actions  and the  assessment of penalties against,
among others, landowners of contaminated property.
   Pennsylvania courts have held landowners to be strictly
liable for the cleanup of their property even if the damage
was caused by a prior owner.'30' In the context of generator
liability, however, unless it  could  be proven  that  the
generator "discharged" or "permitted the discharge" of
hazardous waste into the surface or groundwater,  the
Clean Streams Law was not an effective means of impos-
ing  liability. In instances where a generator did nothing
more than contract with a transporter for the disposal of
hazardous waste, courts would have to stretch the "per-
mit to discharge" language to establish liability.
   Prior to its amendment in 1980, the Pennsylvania Solid
Waste Management Act  ("Act"),<31) proved quite inef-
fectual since there was no meaningful provision for direct
enforcement against hazardous waste generators and pen-
alties were limited to $300 per day. However,  the 1980
amendments to  this Act'32'  provided  a comprehensive
hazardous waste program including a manifest system,
permits,  licenses for  transporters  (which exceed  the
RCRA requirements for transporters) and a variety of ad-
ministrative, civil and criminal penalties, ranging up to
$500,000 and  20 years imprisonment. However, unless
the  provision declaring certain hazardous waste activi-

-------
                                                                 LIABILITY, LEGAL & PUBLIC ISSUES    391
ties "ultrahazardous" and thereby subject to strict liabil-
ity for any resulting harm (which on its face is retroactive)
includes the act of generating hazardous waste, generators
may still be able to escape direct liability for the prior off-
site disposal of hazardous wastes.(33)
  Perhaps the trend of judicial decisions in this area can
be seen in New Jersey v. Ventron Corp.  (see footnote 6)
which interpreted the New Jersey Spill Compensation and
Control Act(34) strictly and apportioned liability jointly
and severally against two  parent corporations of wholly
owned subsidiaries. While the case is  arguably limited to
its facts and is strictly an  application  of  New Jersey sta-
tutory law, it reflects the judiciary's attitude on hazardous
waste disposal issues (in this case, mercury) that wherever
feasible, the defendant most  financially  able to effect a
remedy is likely to be responsible, since the potential  ad-
verse public impact of improper hazardous waste disposal
is astronomical.

WHAT'S A GENERATOR TO DO?
SOME PRACTICE ADVICE

  Once a generator has been sued, he  obviously will have
to defend himself in court against the various theories of
liability claimed by the plaintiff, many of which have been
described in this paper. Open questions that will invariably
be litigated include the burden of proof  necessary to es-
tablish causation, applicability of joint and several liabil-
ity, bases of apportionment, constitutional limits on retro-
active application of statutes and statutes of limitation.
  Prior to suit,  a generator can attempt to uncover  po-
tential problem areas and take certain actions to be in the
best position in the event litigation is brought.
  The generator can audit its hazardous waste disposal
records as far back as he can (since there may be no time
limitation) to determine whether its waste was hazardous.
This determination can be  accomplished by consulting
with his engineers and chemists, reviewing raw material
records, etc. Next, the generator can review the names of
each of his transporters and disposers  to see whether they
are any longer in business  and whether any are now being
investigated or have already been prosecuted.
  All contract  forms  for any waste disposal activities
should be reviewed  to  determine whether any  "special
precautions" were identified. All  personnel  involved in
hazardous waste disposal activities should be interviewed
and the facts memorialized for future reference.(35)
  Generators should  review their insurance policies to see
whether or not there is any possible coverage should there
be claims  arising out of hazardous waste activities.  The
"sudden and accidental" exclusions in most policies have
now given way to more liberal coverage for environmental
impairment, but questions  as to applicability for past prac-
tices are obviously quite important to any determination
of coverage.  In  addition, there is at least one  reported
case where courts have interpreted insurance  policies and
found cover age.(36)
  The authors can do no more than note the problems
that they can foresee arising over the next few years in
the area of generator liability for hazardous waste prac-
tices.  Very few "hard" answers can be given since very
few have yet been furnished; instead, most answers are
merely arguments for judges and juries, whose decisions
will become "the law." Then and only then will we know
what  consequences will be  meted out to generators of
hazardous waste for their past disposal practices.


REFERENCES AND FOOTNOTES

 1. 33 U.S.C. §407.
 2. 42 U.S.C. §6901, ef. seq.
 3. 42 U.S.C. §9601, et.seq.
 4. A generator with an on-site hazardous waste disposal
   facility was still required to obtain the necessary per-
   mits and approvals and to comply with existing oper-
   ating requirements.  By using plant property, the gener-
   ator only avoided the risks of employing a transporter
   for off-site disposal and the lack of control inherent in
   such practices.
 5. 15 U.S.C. §2601, et. seq.
 6. For example, the court stated in New Jersey v. Ven-
    tion, _____A.2d	, 1 Chem. and Rad.  Waste
   Litigation Rptr. 348 (Super. Ct. Ch. Div., August 29,
    1979):
      As we become more sensitive to our  environment
      and more aware of the impact of pollution on our
      environment, we must demand that the unchecked
      development  of  products which release pollutants
      into our environment be controlled. It does not of-
      fend this  court's sensitivities nor infringe upon a
      manufacturing defendant's constitutional rights to
      impose strict liability  upon a defendant who, dur-
      ing the course of a profit making venture discharges
      into  the  environment a dangerous or  hazardous
      pollutant which  results in damage or harm to the
      public, notwithstanding  an absence of intent or
      negligence on the part of the defendant.
 7. Restatement (Second) of Torts §409.
 8. Restatement (Second) of Torts §411.
 9. Restatement (Second)  of Torts §427A. An "abnorm-
    ally dangerous"  activity  is defined in Restatement
    (Second) of Torts §520 to include the following factors:
      •Existence of a high degree of risk of some harm to
       the person, land or chattels of others;
      •Likelihood that the harm that results from it will
       be great;
      •Inability to eliminate the risk by the  exercise of
       reasonable care;
      •Extent to which the activity is not a matter of com-
       mon usage;
      •Inappropriateness of the activity to the place where
       it is carried on;  and
      •Extent to which its value to the community is out-
       weighed by its dangerous attributes.
10. See Pennsylvania  Solid Waste Management Act, 35
    P.S. §6018.401(b):
        The storage,  transportation, treatment  and  dis-
        posal of hazardous waste are hereby  declared to

-------
392    LIABILITY, LEGAL & PUBLIC ISSUES
        be activities, which subject the person carrying on
        those activities to liability for harm although he
        has  exercised utmost care  to prevent harm, re-
        gardless whether such activities were conducted
        prior to the enactment hereof.
      Also,  in  the legislative history of Super fund it is
      stated that:
        For the purposes of this act, Congress declares
        that manufacture, use, transportation, treatment,
        storage, disposal and release of hazardous sub-
        stance  are ultrahazardous  activities.  126 Cong.
        Rec. S 14972 (daily ed. November 24, 1980).
      Note,  however,  that  the  Pennsylvania language
      does not  include "manufacturer" or "generator",
      whereas Congress did include the  former term in its
      legislative intention.
11.  Restatement (Second) of Torts §413.
12.  Restatement (Second) of Torts §416.
13.  Restatement (Second) of Torts §427B.
14.  42U.S.C. §6973.
15.  The pre-1980 version of RCRA's imminent hazard sec-
    tion required the government to prove that "the hand-
    ling, storage, treatment, transportation or disposal of
    any solid waste or hazardous waste is presenting an
    imminent and substantial endangerment to health or
    the  environment.  Now,  all the government has to
    prove  is  that the listed activities "may present"  such
    an endangerment.
16.  See  United States v. Midwest Solvent Recovery, Inc.,
    484  F.Supp. 138 (N.D. Ind.  1980); United States v.
    Solvents Recovery  Service of New England, 496 F.
    Supp.  1127  (D. Conn. 1980). Contra, United States v.
    Diamond  Shamrock,        F.Supp	(E.D.
    Ohio 1981).
    Further, in  light of the holding in City of Milwaukee
    v. Illinois, 451 U.S	68 L.Ed.2d 114 (Sup. Ct.
    1981), that the federal commonal law of nuisance has
    preempted in the water pollution area, it may also be
    held to be inapplicable to waste disposal matters.
17.  In Swell v.  Petro Processors of Louisiana, Inc., 364
    So. Rptr. 604 (Ct. App. La. 1978), generators who had
    actual  knowledge  of improper waste disposal prac-
    tices were held liable for damages to a neighboring
    landowner's property.
18. 33U.S.C. §1321.
19. 42U.S.C. §9601.
20. 42U.S.C. §7412.
21. 33U.S.C. §§1314, 1317, 1321.
22. 15 U.S.C. §2607.
23. 42 U.S.C. §6921.
24. 42 U.S.C. §9602.
25. 42 U.S.C. §9603.
26. 42 U.S.C. §9607.
27. The National Contingency Plan was developed under
    the Clean Water Act to describe the procedures and
    criteria for responding to oil spills. 40 CFR Part 1510.
    Under Section 105 of Superfund, 42 U.S.C. §9605, it
    is  to be significantly revised to incorporate pro-
    cedures, priorities and methods of responding to re-
    leases of hazardous substances.
28. 46 Fed. Reg. 9901 (January 19,  1981).
29. 35P.S. §691A et. seq.
30. National Wood Preservers Corp.  v.  Pennsylvania
    Department of Environmental Resources, 414 A.2d
    37 (Pa. 1980).
31. 35 P.S. §6001 et. seq.
32. 35 P.S. §6018.101 ef.*??.
33. Future off-site disposal activities by  generators  is
    highly regulated in Pennsylvania and the Act specifi-
    cally cuts off liability  for civil and criminal penalties
    provided a  generator uses a licensed transporter and
    the waste is received by an appropriate permitted stor-
    age,  treatment  or  disposal facility. 35 P.S. §§6018-
    605, 6018-606.
34. 58 N.J.S.A. 10-23.11 et. seq.
35. It  is  important to consider  the discoverability of any
    written memoranda prior to their preparation so that
    appropriate safeguards can be instituted.
36. Lansco Inc. v.  New Jersey Department of Environ-
    mental Protection,  138 N.J.  Super. 275, 350 A.2d
    520 (Ch. Div. 1975) aff'd. per curiam 145 N.J. Super.
    433, 368 A.2d 363 (App. Div. 1975).

-------
      RECOVERING DAMAGES TO NATURAL RESOURCES
                                      UNDER CERCLA

                                       EDWARD YANG, Ph.D.
                                             AMY HORNE
                                           Resources Program
                                      Environmental Law Institute
                                           Washington, D.C.
                                          OSCAR ALBRECHT
                                  Office of Research and Development
                                      U.S. EPA, Cincinnati, Ohio
INTRODUCTION

  Until  recently,  pollution  damages to publicly owned
natural resources  were seldom paid for, due to the dif-
ficulties of establishing "ownership", assessing physical
damage and deciding on monetary values of the resources.
Increased oil and hazardous substances spills with the last
two decades, together with improved knowledge of the
functions of natural resources, have led to the recogni-
tion that these functions warrant compensation for dam-
ages even though they are not explicitly  traded  in the
marketplace. The 1980 Comprehensive Environmental Re-
sponse,  Compensation and Liability  Act,(1) otherwise
known as the Superfund Act, establishes liability for dam-
ages or injury to, destruction of or loss  of natural re-
sources resulting from hazardous substances releases, in-
cluding  the reasonable  costs of assessing  such damages
(section 107 (a)).  Further, in order to facilitate damage
recovery, the Superfund Act establishes a Hazardous Sub-
stances Response Fund which can be used  by the govern-
ment to clean up spills  and resource damage losses re-
cover. Consequently, the federal government has the un-
precedented opportunity to codify some damage assess-
ment techniques by promulgating regulations specifying
analysis procedures (section 301 (c) (1)).
  The economic theory  of  externalities'2'  holds that the
optimal compensation procedure will extract from pol-
luters an amount equivalent to the social value  of the
damage.  Otherwise, disregarding effects of income dis-
tribution, society will lose, due to misallocation of its re-
sources. In this paper, the authors briefly outline the eco-
nomic theory underlying  resource  damage assessment
and highlight  some obstacles to establishing estimation
methods. Further, methods currently employed in legal
proceedings are reviewed and an alternative is considered
that can be applied under the Superfund Act. As such, the
paper's focus is on deriving a monetary measurement  of
the damages rather than assessing physical  damages, such
as identifying  the pollutant, determining the contamina-
tion level and the effects caused by the pollutants.

PLACING MONETARY VALUE ON
DAMAGED NATURAL RESOURCE
  According to economic theory, the appropriate valua-
tion of damages should be based on the demand that ex-
isted for the natural resources before they were injured.
The primary obstacle economists face is the non-market-
goods nature of open-access natural resources, such as
public beaches, state parks, open lakes or rivers. Because
these resources are not traded in the market place there is
no price from which monetary damages can be derived.
  In such  cases, economists resort to the concept of the
willingness of the users to pay for the natural resources.
A person's willingness to pay for a resource is the amount
of money  he or she is willing to accept in place of using
the resource. When travel and equipment expenditures are
incurred, for example, by a user of a natural resource, the
willingness-to-pay figure must be  reduced by these costs,
resulting in a "net" willingness-to-pay.  For example, if a
sport  fisherman is  willing to pay $5.00  for catching a
trout, and the travel and  equipment  expenditures per
trout is $3.00, the net gain to him is $2.00 per trout.
  The value of a natural  resource, therefore, is the will-
ingness-to-pay net  of the  costs  incurred to use the  re-
source, aggregated for all of the users or potential users.
The value of the damage is the  reduction of this aggre-
gated net willingness-to-pay that is caused by the incident.
This measure reflects society's welfare loss due to  the
damage.
  Economic techniques developed to measure net willing-
ness-to-pay(3) generally are expensive and time-consuming.
Moreover, many are  designed  for research rather than
damage recovery. The most widely used technique surveys
users  of the resource and attempts to elicit their willing-
ness to pay. In order to obtain accurate results, the survey
must be a valid and sufficiently large sample of the user
population.  Construction  of such a  sample  often  re-
quires a substantial effort.
  Another economic technique relies on changes in prop-
erty values to estimate  damage  values. However,  the
uses of this latter technique thus far have been mainly
limited to valuing the benefits of clean air.

TECHNIQUES USED IN LEGAL PROCEEDINGS
  In  the  past, the considerable time and expense  re-
quirements of the economic methods discussed above, lim-
ited their  application in  legal  proceedings.  Further,
courts lacked the technical expertise necessary to arrive at
a consistent framework to value the damages. In fact, only
recently have courts begun to hold the polluters  liable for
                                                   393

-------
394     LIABILITY, LEGAL & PUBLIC ISSUES
damages to open-access natural resources, under the con-
cept of public trusteeship.  Nevertheless, courts had to
develop methods that would quantify the damages so that
compensation could be made to the plaintiff. The follow-
ing  subsections  discuss  these  traditional  assessment
methods.

Replacement Costs
  In cases involving injury to a natural resource, replace-
ment costs are sometimes awarded by courts as a com-
ponent  of restoration  damages.<4) For example, a court
may order the defendant to pay for the cost of replacing
a grove of destroyed fruit trees with a comparable type
and number of seedlings. Similarly, a court may award a
plaintiff damages to cover the cost of restocking a pol-
luted stream or lake with fish and other aquatic life. In
these cases,  the  replacement  cost is intended  to com-
pensate for the loss of economic uses of natural resources
and restores the injured party to his or her pre-injury posi-
tion.
   Replacement costs also have been used to determine the
 value of non-economic uses of natural resources. In State
 of Tennessee ex rel. Goodrich v. Riggen,™ the State sought
 to  recover the value of deer killed by negligent use of an
 herbicide. At the trial, the State presented a number of
 expert witnesses testifying as to value of the deer. Surveys
 found that  deer hunters spent $1,035-$ 1,727.47 for each
 deer killed.  Wildlife experts placed  the  recreational,
 aesthetic, economic and  ecological value of a deer at
 $2,500-$5,000.(6>  The  plaintiffs sought  $2,500  per deer
 killed'7' based on the aesthetic, ecological and recreational
 value of the  deer. The court rejected the plaintiffs'  as-
 sessments of the value of the deer and instructed the jury
 to consider instead:

     "the  average cost per deer, cost  affected  by  the
    survival ratio, restocking cost, the fact that deer  re-
    produce themselves, transportation costs for the deer
    and economic benefits which you find the proof the
    State of Tennessee may receive  (sic), including that
    from deer hunters; to reach an amount which would
    fairly compensate the State of Tennessee  for its  al-
    leged loss of deer and unborn fawns."...(8)

  An important fact which  perhaps influenced  the court
is that in a different  part of the State deer were so over-
populated that many would starve without control mea-
sures.'"  The jury  finally awarded the State of Tennessee
$1,119, based on testimony which  asserted that the cost of
catching and  transporting deer from the overpopulated
area was $100 per deer."01 The State lost on appeal.""
  The replacement cost valuation technique was recently
employed by the  plaintiffs in  Commonwealth  of Puerto
Rico vs. SS. Zoe Colocotroni.™ This case arose from an
incident in which an oil tanker dumped 1.5 million gallons
of crude oil  in order to free itself after  running aground
on a reef. The oil came ashore in Bahia Sucia, a Puerto
Rican lagoon  which,  prior  to the spill  was a "healthy,
functioning  estuarial  ecosystem" in which  the angrove
forests  functioned  as  breeding,  feeding and nursery
grounds for fish, shrimp and  benthic organisms as well
as being the primary food-producing agents of the organic
materials available to the aquatic food chain.03'
  The oil spill resulted in the death of more than 92 million
marine animals in a 20 acre area for which no market
value, in  the sense of  lost market profits or revenues,
could be ascribed.04' In addition,  the oil penetrated the
mangrove forests, resulting in a significant increase in
the mangrove mortality of the area.
  More than five years after the incident, the mangrove
community showed a significant reduction in the number
of species of macrobiota,  the  number of molluscs and
crustaceans, and  other marine organisms. The seagrass
community showed a significantly altered pattern of plant
biomass and the number of species and population density
of marine organisms.
  The district court awarded the following damages: for
the more  than  92 million  marine organisms killed, the
defendants were to pay $5,526,583.20, a figures derived by
referring to the prices of marine organisms from biological
supply laboratories.
    "The  lowest  possible replacement  cost  figure is
    $.06 per animal... Accepting the  lowest  replacement
    cost, and attaching damages only to the lost marine
    animals  in the West  Mangrove  area,  we  find the
    damages caused by Defendants to amount to $5,526,-
    583.20.""5'
For damages to the mangrove forests, the court  awarded
damages for the 23 most affected spots in the West Man-
grove area. It found that the best means of reestablishing
these  areas was  by the intensive planting of mangrove and
restoration of this area to its condition before the oil spill.
Evidence  showed that  planting mangroves  costs about
$16,500 per acre, and  that the  five year  monitoring and
fertilization  program would cost $36,000  per  year  or
$180,000 for the five years. The total  damage of the pol-
lution to  the West Mangrove  equalled  $559,000. The
court  also awarded $78,108.80  in cleanup costs  incurred
by the plaintiffs. The total award granted by the district
court  was $6,164,192.09.
Contribution to Productivity
  The contribution of a given resource to ecosystem pro-
ductivity had been used to  place a value  on the damaged
resource, on  at least one occasion. In State of Florida v.
Bruce,(^ the administrative law judge assessed the value
of the detrital production of some felled mangrove trees.
The judge accepted the State's contention  that detrital
value, the accumulated leaves, branches  and seeds which
serve  as an essential element in the estuarine food chain,
should be the basis for calculating damages. Its value is
defined from its contribution, through the food chain, to
the support of recreational and aesthetic  fish species. The
court agreed that lost detrital value could be derived from
the size of the lost canopy cover, which could be found by
aggregating the measure of the cut mangrove tree diame-
ters. Thus, in this instance 2000 square feet of canopy was
lost.  The value for the lost 2000  square feet  was esti-
mated to be $2,760. This sum was derived by assuming a
$4000 per acre year value, which was multiplied by 15,
the number of  years required for a mangrove seedling to
reach maturity.

-------
                                                                  LIABILITY, LEGAL & PUBLIC ISSUES    395
  In addition to showing  that the lost detrital produc-
 tion was a factor to be considered in calculating the value
 of the damages, the State also argued that mangroves
 served many other functions,  such as water cleansing,
 pollution  control,  nutrient uptake  and  assimilation,
 shoreline stabilization, flood prevention, water conserva-
 tion, oxygen production and sequestration of heavy metals
 and other  poisonous  substances. However, compensa-
 tion was not sought for the latter values.

 Willingness-to-pay Survey
  A technique for estimating the value of damages caused
 by a pollution incident by surveying  people's willingness-
 to-pay for  recreational uses of those resources was pre-
 pared by plaintiffs for presentation to a court. Unfortun-
 ately, no decision was reached in  that case as  to  the
 method's legal validity. In the Matter of the Complaint of
 Steuart Transportation Company,(IT>  a claim arose from a
 spill of 250,000 gal of #6 fuel oil into the Chesapeake Bay.
  As a result of the  spill, thousands of waterfowl were
 killed, 27 miles of wetlands were fouled, many   inverte-
 brates were killed, many oyster beds  were extensively pol-
 luted.08'  Both the Commonwealth of  Virginia and  the
 United States filed  suit, the former seeking $731,500 in
 damages to waterfowl and other costs,  the latter seeking
 $487,000 in cleanup costs and $1 million for loss of water-
~fowl.<19) The defendants eventually settled out of court for
 $115,000.(20)
  As part of its evidence as to the value of the damages
 caused, the United States introduced a study which esti-
 mated the lost hunting value resulting from the bird kill.<21)
 An expert  witness applied the  results from a previously
 completed  study, which concerned the hunting  expendi-
 tures made by waterfowl hunters in the seven states lying
 within the Pacific Flyway, to the facts in Steuart.
  In  that study, randomly selected  hunters  were sent a
 questionnaire concerning  household income after taxes,
 number of seasons of experience  in hunting waterfowl,
 hunter's cost per season, hunter's seasonal  bagged Water-
 fowl harvest  and the hunter's estimate  of how much his
 costs would have had to  have risen above present costs
 before he would have decided to totally abstain from wat-
 erfowl hunting  during the  season. Using  a statistical
 formula, the  marginal value of a bagged waterfowl for a
 representative hunter was obtained.

 Lost Business Opportunity
  Damages to open-access  natural resources can injure
 business opportunity if business uses the resources as  an
 input, such as an oyster farm using clean  open-water in
 Skansi v. Humble Oil & Refining Company.m In this
 case,  the defendant argued that the plaintiff should not
 receive more  than his  expected net profit on oysters de-
 stroyed through water pollution. The court disagreed, say-
 ing that the plaintiff had lost not only his net profits but
 also all of the expenses of planting and cultivating, which
 were included in the sales price.
  In Carr v.  United States,™  the court refined  the lost
 revenue valuation technique used in Skansi by  limiting
 the compensation to the net loss sustained by the plaintiff.
This figure was derived by multiplying the estimated fore-
gone yield in bushels by the average price per bushel for
that year and subtracting the costs of tonging and mar-
keting. Since the plaintiff  had not expended  money for
tonging and harvesting, the court was economically cor-
rect in eliminating those costs  from the estimated lost
value.  Consequently, the  plaintiff received  lost profits
plus those expenses that were already incurred.
Discussion

  The above valuation methods, which were actually used
in legal proceedings, all attempt to measure the social
values of the damaged resources. Yet the methods are not
consistent: for the same  damage, different values resulted,
depending  upon which method  was  employed.  These
methods also vary in their  strengths and weaknesses. Re-
placement costs only address the supply side of the re-
sources, ignoring the level of demand for the resources
(i.e., should lost resources  always be replaced  or restored
at whatever cost?).
  If only the supply side is considered, society might re-
place or restore resources whose costs exceed their social
value. The contribution to productivity  method is more
useful when parts of ecosystem are damaged, disrupting
the aggregate  function  of  the ecosystem. However, this
method's usefulness is limited if the functions or products
of the ecosystem are not well defined  and not traded in
the market place.
  The expenditure survey method has  the strongest  eco-
nomics basis, since it directly measures the users' willing-
ness-to-pay for the damaged resources. Its only weaknesses
are that: (1) costly surveys may be necessary and (2) the
accuracy of the results depends on  the sincerity of the
people surveyed and on their interpretation of the ques-
tions. For example, in Steuart Transportation  the answer
to the question of how much the costs of waterfowl hunt-
ing would have to rise above the present cost before the
hunter would decide to abstain  from  hunting may  lead
to inaccurate results if  the hunter is not constrained by
his actual income in answering.
  The lost business opportunity method only applies when
a business  using the open-access resources is  injured by
damages to those resources. Hence,  the method does
not address the value of the resources in public uses.  In
fact, in the context  of  the Superfund Act, such private
damage recovery is excluded from the use of the Fund.
  The applicability of the above methods under Super-
fund is  limited by the potentially large number of spills
and types of natural resource  damage. In order for the
Fund to process the claims expeditiously, the damage valu-
ation methods  have  to  be, as Congress intended, simple
and require minimal administrative cost.


AN ALTERNATIVE SCHEME UNDER SUPERFUND

  The damage  valuation  method under the Superfund
damage assessment procedure  must  strive for a balance
between conceptual validity and procedural simplicity. The
underlying concept is that a relationship exists between
the costs  of conducting the damage  valuation and the

-------
396     LIABILITY, LEGAL & PUBLIC ISSUES
benefits  of   accurately  identifying  the  damage   in
monetary terms. Because the social costs of miscalculating
the value of a unit of damage— the benefit of accurate
valuation— may be small in small releases,'24' the  use  of
expensive valuation methods can not always be justified.
Realizing this, Congress divided the damage  assessment
regulations into two types:
     "Such regulations  shall specify  (A) standard pro-
     cedures for simplified assessment  requiring minimal
     field, observation, including establishing measures  of
     damages  based on units of discharge or release  or
     units of affected areas and (B) alternative protocols
     for conducting assessments in individual cases to de-
     termine the type  and extent of short and long-term
     injury, destruction, or loss." (CERCLA) section 301
The Use of Value Tables for Small Releases
  When the units of discharge or of affected resources are
small, Superfund can adopt value tables along the lines of
those used by the States of Virginia, Florida, Washington
and California.'23' In general, these tables set the values of
fish and game (except  California,  which includes  other
natural resources, such as sand and  flatworms) so that
damages can be recovered accordingly.  In  Virginia,  the
court accepted the  table as evidence of the value of the
lost fish in Commonwealth  ex rel State Water  Control
Board v. Weaver Mirror Co.  No.  4722 (Franklin Co. Vir-
ginia Circuit Court, June 2).
  Florida promulgated its table of values as regulations. <26)
Subsequently, the Florida Supreme Court accepted  these
fish values in State Dept. of Pollution Control v. Interna-
tional Paper Company. It held that the values were proper
and relevant. (27)
  The State of Washington published its Guidelines For
Evaluating Fish Kill Damages and Computing Fish Kill
Damage Claims (hereinafter Guidelines), which bases the
value of fish on its production cost at the state fish hatch-
eries. California used its table to assess compensation in
State of California v.  Alice B.  Copeland  Vincilone,m
which arose after the defendant  conducted leveling and
filling activities in "Big  Hole,"  a complex  of island,
water channels and land masses along the Colorado River.
  Alaska also uses a table to assess pollution damage,  but
with a  different approach. Alaska's  tables consist of a
schedule of penalties  for  oil discharges, which  vary ac-
cording to: (1) whether the oil enters a freshwater, brack-
ish, or saltwater environment or public land (2)  the toxi-
city, degradability and dispersal characteristics of the oil
and (3) the sensitivity and productivity  of  the receiving
environment.'291  This  approach,  although simpler  than
the previous tables to administer,  in that  does not require
counting the damaged units, may  not reflect the actual
value of the damage.
  Although these value tables only cover  a  small  por-
tion of the resources that can be damaged by hazardous
waste  releases,  they  certainly  can be expanded under
Superfund to include  groundwater,  wetland  and  other
natural resources. These tables can also be constructed on
a regional basis so for a given resource to reflect regional
difference in values. However, several weaknesses in the
existing value tables requires correction if they are to be
used for damage assessment under Superfund. The con-
struction of the values should be made more consistent
and be based on sound economics. Several examples will
help illustrate this need.
  The fish values used in Virginia for example, are con-
structed from the average purchase price of the fish. This
is valid as long as the price is below the consumer's willing-
ness-to-pay for the fish. When the demand for the fish is
low the cost of replenishing the fish, including transporta-
tion and labor costs, may be an excessive compensation to
award.
  This criticism also applies to parts of the Guidelines of
the State of Washington, which depends on replacement or
restoration costs. The  Florida value table was  derived
from a mixture of hatchery costs, expenditures and com-
mercial values. Expenditures of the recreational fishermen
cannot be used as the  value, since they do not represent
willingness-to-pay for the fish. Again, commercial  value
is also not accurate  because the price includes other costs
of harvesting.
  It is also questionable  whether the price of the  com-
mercial fish can be applied to species that are also in de-
mand for recreational  fishing; there is no market where
open-access fish are allocated between commercial and
recreational fisherman.
  The California table sums several types of values: re-
placement costs, use value  and psychic value. This ap-
proach reflects the  complexity of natural  resources that
have multi-uses. However, the summation of the values
may result in double-accounting, since these values are
not always complementary.  For example,  the bird that a
hunter wants cannot be used for bird watcher.
  Learning from the past experiences, value tables under
the Superfund can be  constructed more consistently and
accurately. Consistency can be gained by employing the
best components of the state schemes and  applying them,
to the extent suitable, to all small releases. Accuracy can
be improved by basing the techniques on more solid eco-
nomics. The economies of scale can allow careful construc-
tion of the values through willingness-to-pay surveys and
other techniques presently used in economics research. It
is likely that more than one technique will be used, due to
the different types of  the resources (e.g.,  fish, wetland,
groundwater, lake and river).
  When using surveys, the samples should be taken on a
regional basis, so that  regional differences in uses will be
reflected in the values.  Although simpler, value tables cor-
responding to units of discharge may be  difficult to de-
fend,  due to their theoretical weakness that the  specific1
resource itself and its uses are not considered.

Use of Economic Techniques for Large Releases
  When  a hazardous waste release is  large  enough that
costs of conducting a case-by-case valuation can be justi-
fied, several economic techniques and approaches can be
used,  including willingness-to-pay surveys,  travel cost,
household production and revealed preferences. Although
these techniques are cumbersome, they can be simplified in

-------
                                                                LIABILITY, LEGAL & PUBLIC ISSUES    397
terms of procedure and understandability with minimal
loss of analytic rigor. Their chief advantage is that they
are all based on widely accepted economic principles.
  This does not mean there is always consensus on how
the techniques are used. In fact, disputes do arise from dif-
ference  in the empirical  estimation procedures.  It is ad-
visable, therefore, to  determine which  estimation pro-
cedures appear to have the support of most economists
practicing in this area.(30) The procedure should be adhered
to as much as  possible to  minimize  variation in the
values resulting from different estimation procedures.
  Another advantage of the case-by-case techniques is that
the values are generally  derived from the sample of the
people who actually or  potentially use the resources in
question.  The resulting  values,  unlike  those of value
tables, cannot be  repudiated  by the argument that the
values do not reflect the users of the specific damaged
resources.
  The main objective to the case-by-case  approach is its
high  dollar and time costs. But with the status of rebut-
table assumption in court that is granted by the Superfund,
the techniques can be greatly simplified  to reduce cost
and time. Elegance and rigor  would have to be partially
foregone; the confidence interval of the estimates would
have to be increased. This often means small sample sizes
and short survey periods. Such sacrifices are worth the
benefits of bringing sound valuation techniques into the
effort to protect  the  nation's  diminishing natural re-
sources.
 FOOTNOTES

  1. P.L. 96-510, 945 Stat. 2767 (Dec. 11, 1980)
  2. Externality theory is concerned with divergences be-
    tween private and social cost. For example, Firm A
    emits smoke during its production process which soils
    laundry cleaned by Firm B. The marginal social cost of
    Firm A's activity  is greater than the marginal private
    cost and, consequently, the firm is producing more
    than is socially optimal. According to A.C. Pigou, the
    correct policy would be to levy a tax equal to the dif-
    ference between marginal social and marginal private
    costs. This  tax would induce the firm to produce the
    "right" socially efficient output—that at which price
    equals marginal social cost.
  3. See Freeman, A.M. Ill, "The Benefits of Environ-
    mental Improvement"  (Baltimore: Johns  Hopkins
    University Press,  1979) for a discussion of the tech-
    niques.
  4. See  e g Commonwealth of Puerto Rico v. S.S. Zoe
<   Colocotroni, 456, f. Supp. 1327 (1978)
  5. Shelby Law No. 3 (1947)
 i Assignments of Error and Brief of Appellant, at 19.
 I. Michael J.  Bean,  "Law  and  Wildlife and Emerging
    Body Wildlife Law,"  Environmental Law Reported,
    Vol  7 March 1977, p. 50020.
 8.  Assignments of Error and Brief of Appellant, at 8.
 9.  John J. Tomanson, counsel for appellant, personal
    communication, May 20, 1981.
10.  Jim Allison, counsel for appellant, personal commun-
    ication, May 21, 1981.
11.  Assignments of Error and Brief of Appellant, at 28.

12.  628 F. 2d 652 (First Cir. 1980).
13.  456 F. Supp. 1327 (1978), at 1339.
14.  Id. at 1344.
15.  Id. at 1345.
16.  DER Case No. 80-1481.
17.  495 Fed. Supp. 38 (D. Va. 1980).
18.  James  D.  Range and Millicent  A. Feller, "Con-
    gressional Perspectives on the Need for Estimating En-
    vironmental Damage from Oil and Hazardous Waste
    Spills, "Proceedings of the 1979 U.S. Fish and Wild-
    life Service Pollution Response   Workshop,"  St.
    Petersburg,  Florida:  U.S. Department of Interior,
    September 1979.
19.  General Accounting  Office,  Total  Costs  Resulting
    from Two Major Oil Spills,  Washington, D.C.; Gov-
    ernment Printing Office, CED-77-71, June 1,1977.
20.  "$115,000 Awarded in Oil  Spill Suit,"  Chesapeake
    Citizen Report, Vol. 3, no.2, March-April, 1981.
21.  Gardner M. Brown, Jr. and Judd Hammack, "Com-
    monwealth  v. Steuart—Economic  Valuation  of
    Waterfowl," (unpub. paper) May 5,1977.

22.  176 So. 2d 236 (La. Ct. Ap. 1965).
23.  136 F. Supp. 527 (D. Va. 1955).
24.  Small releases  can be  defined on  the basis of either
    units of discharge or units of affected resources.
25.  See Bruce C. Rashkow, "An Analysis of Alternatives
    to the Traditional Approach to Damage Assessment,"
    unpublished paper,  Dept. of Justice, Washington,
    D.C. 205-30.
26.  In a case brought  by commercial fishermen for  dam-
    ages  resulting  from the Santa Barbara oil spill,  the
    plaintiffs had difficulty establishing  a base figure of
    average commercial fish yield. And  catch figures
    normally vary widely  and in addition are  only sep-
    arately reported, making them uncertain. Determining
    lost profits with certainty was further dampened in
    this case by additional damage the fish caused by  un-
    related flooding, Philip E. Sorenson, Environmental
    Damage in Economics and Law: The case of the Santa
    Barbara Oil Spill,  Proceedings, Annual Meeting of
    the  Southern  Economics  Association,   Atlanta,
    Georgia, November 1976, pp. 115-119.
27.  Fla. Stat. Ann. 403.141 (3) (1973) Doct. No.  15156,
    California District Court, Riverside Superior Court.
28.  Fla., 329 So. 2d 5 (1976).
29.  As 46.03.758.
30.  For example,  the Department of Interior  could es-
    tablish a panel of economists expert in the area of
    damage assessment.

-------
          SITE CONTAMINATION AND LIABILITY AUDITS
                           IN THE ERA OF SUPERFUND
                                      JOHN J. HOUSMAN, JR.
                                       DAVID I. BRANDWEIN
                                          DENNIS F. UNITES
                            TRC Environmental Consultants, Incorporated
                                       Wethersfield, Connecticut
INTRODUCTION

  Industrial managers responsible for corporate real es-
tate or environmental affairs are often not fully aware of
the legal and  financial ramifications attending contami-
nated properties.  With  rapidly  escalating  government
efforts to locate and remedy abandoned or uncontrolled
hazardous waste sites, the need  to inventory corporate
real estate, examine relative contamination potentials, de-
fine  potential  liability  exposure  and  institute control
mechanisms is being recognized.
  In this paper, the authors briefly outline the statutory
and  common  law initiatives and  various liability  path-
ways. However,  the focus is on  the presentation of a
structured multidisciplinary methodology, the Site Con-
tamination and Liability Audit, designed to identify  site
contamination and  liability problems  and  establish a
framework for their control. Examples are used to demon-
strate the need for such audits and to illustrate the audit
process.
  The emphasis in the  paper is placed on the property
screening aspects or early  phases of the  audit process.
The  property  screening  phases are designed to obtain a
reasonable qualitative assessment of possible site contam-
ination problems with a minimum of effort and expense.
Site screening  aspects are of particular value in an era of
increased corporate merger activity. Careful screening for
site contamination problems can prevent the acquisition
of severe liabilities.
FORCING STATUTORY AND COMMON LAW
INITIATIVES PREDATING SUPERFUND

  Prior  to enactment of the Comprehensive  Environ-
mental Response, Compensation, and  Liability Act of
1980 (Superfund), persons involved as defendants in lit-
igation related to inactive  or abandoned contaminated
hazardous waste  sites entered the courts under  a variety
of statutory  and  common law provisions. The types of
actions which were initiated can be divided into the follow-
ing broad areas:

  (1) Enforcement under Federal and State statutory pro-
     visions
  (2) Emergency  powers provided by Federal and State
     statutes
  (3) Citizen actions taken pursuant to Federal and State
     statutes
  (4) Actions initiated by government pursuant to Fed-
     eral and State common law
  (5) Private suits brought by injured parties pursuant to
     Federal and State laws

Enforcement Provisions under Federal
and State Environmental Laws
  All  Federal  and State  environmental  laws have  en-
forcement provisions making violation of specific sections
of the law either civil or  criminal  violations,  depending
on  the  circumstances.  Prior  to  Superfund,  the most
widely used enforcement powers were those under RCRA,
the Clean  Water  Act, the  Refuse Act  and their state
counterparts.
  The Clean Air Act, Safe  Drinking Water Act, Toxic
Substances Control Act and their state equivalents have
seen more limited use. Since most of these statutory
provisions  only address activities conducted after enact-
ment dates, effective use for abandoned or inactive waste
sites has been  limited. Successful government  actions
have most  often occurred where the site has involved  on-
going disposal activity.

Use of Statutory Emergency Powers
  Most  recent environmental  statutes give Federal  and
State officials broad authority if an environmental or hu-
man health hazard is imminent. However, in using these
powers, the government must first demonstrate  that an
emergency exists.  Federal and State courts have varied
widely in their interpretation of what constitutes an  im-
minent hazard situation. Once this burden of proof is met,
the government has a wide  range  of remedies in equity
available under Section 7003 of RCRA, Section 504 of
the Clean  Water Act, Section 303  of the Clean Air  Act,
Section  1431  of the Safe  Drinking Water Act, Section 7
of the Toxic Substances Control Act, various state equiva-
lents of these statutes and a number of State general
environmental policy laws. These remedies have included
actions such as:
  (1)  Order immediate restraint of an activity
  (2)  Require determination of the extent of contamina-
      tion or environmental and health damage
  (3)  Require development of remedial plans
                                                   398

-------
                                                                 LIABILITY, LEGAL & PUBLIC ISSUES     399
 (4) Require implementation of containment or removal
     measures
 (5) Require installation of monitoring networks
 (6) Order permanent restoration of the site
 (7) Order installation of permanent alternative  water
     supplies
 (8) Require posting  of  financial security through  a
     variety of mechanisms
Citizen Actions Pursuant to Federal and
State Statute
 Each of the Federal environmental laws enacted  in the
last decade and most comparable State laws contain so-
called "citizen suit" provisions. Generally, under such
provisions, any person or group of persons may commence
a civil suit against any party allegedly  in violation of any
requirement of the act in question.  These provisions also
generally provide  for citizen suit against  the administer-
ing regulatory agency for failure to perform an act or duty
under the law which is not discretionary with that agency.
  Persons commencing such actions must first notify the
respective regulatory agencies and alleged  violators  of the
intention to bring action.  No  separate  action may  be
brought under these provisions if EPA or a state has be-
gun and is diligently pursuing the same action.

Government Common Law Actions
  Federal and State agencies have  used common law
principles in seeking relief against abandoned  or uncon-
trolled hazardous  waste sites. The common  law theories
used have included nuisance, negligence, trespass, restitu-
tion and strict  liability.
  EPA has been  particularly interested  in imposing  a
joint, several  and strict liability standard on  hazardous
waste generators,  transporters  and facility  owners and
operators. For the most part, the courts have  been sym-
pathetic to this approach. Typically,  Federal  and State
government attorneys have used the "shotgun" approach
in  seeking relief under every  possible avenue, citing com-
mon law theories along with the various statutory en-
vironmental authorities available.
Private Common Law Actions
  Plaintiffs have  brought suit  under the common law
tort theories of negligence, nuisance,  strict liability, tres-
pass  and restitution. For  the most  part, these private
parties  have  found  it  difficult to muster the evidence
necessary to satisfy  the strict burden of proof require-
ments of the common  law tort theories. The  govern-
ment, owing to their greater level of resources and clearer
statutory authority to enter property for the  gathering of
evidence, has been  more  successful  in  obtaining  judg-
ments pursuant to the common law tort theories.

INITIATIVES AVAILABLE UNDER SUPERFUND

 The legal remedies available prior to the enactment of
the Comprehensive  Environmental  Response, Compen-
sation and Liability  Act of 1980 (Superfund) pale  as in-
significant before  the awesome  arsenal of authority now
available to government pursuant to the new Superfund
legislation. Essentially,  the government  now  has a free
rein in its apparent power to address  the problems pre-
sented by inactive, abandoned or uncontrolled hazardous
waste sites. Major sections of the new law of particular
importance to this discussion include:
  (1) Section 103(a)—requires that the  National Response
      Center be notified of the release of any reportable
      quantity of a hazardous substance.
  (2) Section 103 (c)—requires that by  June 9, 1981, any
      person owning or operating, or who at the time of
      disposal owned or operated, or  who acting as an
      transporter selected storage, treatment and disposal
      facilities for hazardous substances, must  have re-
      ported  to EPA the existence of such  facilities,
      hazardous substances received, and known, suspect-
      ed or likely releases of hazardous substances from
      known  facilities   (RCRA  permitted  or  interim
      status facilities are excluded).
  (3) Section 104—grants  the  government authority to
      respond to hazardous substance  contamination in-
      cidences or suspected incidences by directly employ-
      ing  investigation,  remedial,  or removal  actions
      (government discretion in these matters is implied).
  (4) Section 107(a)—assigns strict financial liability for
      response, remedial action, removal and destruction
      or damage to natural resources  to facility owners
      and operators,  past  owners and operators, trans-
      porters selecting facilities for disposal and persons
      contracting for disposal at such facilities.
  While clearly possessing the power to act, the govern-
ment will most likely begin exercising  its new authority
slowly.  A myriad of complicated  plans, administrative
procedures and regulations required by the law remain to
be developed.
  As the government acts, the regulated community and
potential defendants are not expected  to sit back while
federal and  state response  teams descend upon them in-
curring large financial charges on their behalf. Structur-
ally,  the law provides  numerous opportunities for po-
tential defendants  to question and challenge government
actions. However,  before  the  proper  questions  can be
asked and meaningful challenges mounted, affected par-
ties must come to  fully  understand the scope and magni-
tude of their specific site contamination  problems.
  The Site  Contamination and  Liability Audit  process
presented herein has  been designed to provide industry
executives and middle level managers with a flexible out-
line of the actions needed to find and  define  their prob-
lems and the potential liabilities presented. After problem
definition and  understanding,  informed  corporate  re-
sponse can be fashioned.

SITE CONTAMINATION AND LIABILITY
AUDIT PROCESS

  The Site Contamination and Liability Audit is a logical
stepwise procedure designed to find hazardous substance
site contamination problems, define these  problems and
set the stage  for informed  corporate response. The audit

-------
400      LIABILITY, LEGAL & PUBLIC ISSUES
                                Table I.
     Site Contamination and Liability Audit Phased Structure
Screening
Phases
Emergency
Action
Phase

Detailed Site
Investigation
and Remedial
Phases
              Phase 1   Initial Property Inventory
              Phase 2   Classification and Identification of Potential
                         Problem Properties
              Phase 3   Preliminary Field Screening
              Phase 4   Prioritization of Problem Properties

              Phase 5   Immediate Emergency Stop Action Response
              Phase 6   Detailed Site Field Investigation
              Phase 7   Definition of Remedial Strategies, Risk and Financial
                         Liability Assessment and Remedial Cost Effectiveness
              Phase 8   Selection of Preferred Remedial Strategy
              Phase 9   Implementation of Remedial Action
              Phase 10  Certification of Performance and Addressing Future
                         Potential Liability Issues
                               Table II.
          Elements of Phase 1, Initial Property Inventory

Objective       Gain understanding of the number, location and use of all properties under
              corporate control (own, rent or lease) or planned property acquisitions.

                                               Plant or
                                               Facility Level
                                               Plant or Fac. Mgr.
                                               Plant Engineer
                                               Environmental
                                               Engineer
Info, and Data   •Topographic map of site area (U.S.G.S. Quadrangle Map).
Req'd. for      'Site layout map showing property boundaries, acreage, building locations and
Initial          adjacent property owners
Inventory       •Period of time property under control
              •Prior use of property, if known
              •Activity summary (e.g., property is manufacturing  location for chlorine gas)
              •SIC (Standard Industrial Code) designation(s)
              •Number of employees
              •Number of production employees
              •Summary of present environmental regulatory status of  site (e.g., RCRA
               Part A permit application filed  for sludge storage lagoon, no present en-
               forcement problems)
              •Clearly indicate status of known contamination problems

              All of the required information for this phase should be readily available to or
              from the designated responsible individuals

              If a large number of properties  are involved, data  processing may be re-
              quired
Assignment of
Responsibility






Corp. Level
Director of
Envir. Affairs
Director of
Property Mgi
Environmental
Counsel

Division Level
Director of
Envir. Affairs
Director of
Property Mgt
Environmental
Counsel
Chief Engineer
Info, and
Data Sources

Equip, and
Services
Needed

Other
Considera-
tions
              A questionnaire or data form may aid in obtaining uniform quality of re-
              sponse from all locations
                               Table III.
       Elements of Phase 2, Classification and Identification
                   of Potential Problem Properties
Objective
Assignment of
Responsibility
Info, and
Data Req'd.,
Aciion to be
Taken
Info, and
Data Sources

Equip, and
Service*
Needed

Oihef
Consideration*
              Identify those properties where present or historic activity gives cause to
              suspect possible site contamination
              Corporate Level
              Coordinator and
              support staff
Division Level
Coordinator and
support staff
Plant or Facility Level
Plant or Facility Mgr and
and support staff
              All Phase 1 data is reviewed by corporate and divisional coordinators. All
              properties are classified: (t) known site contamination; (2) site contamina-
              tion likely or suspect; (3) data for site not complete; and (4) site contamina-
              tion unlikely.
              •Obtain missing data for Category 3 sites

              All data  should be in hand as a result of Phae 1. Responsible  individuals
              provide missing data for Category 3 sites

              If a large number of properties are involved, data processing may be re-
              quired


              A rating or scoring system ma> aid m the categorization of properties. All in-
              dividuals involved should be counselled as to the importance of the project
                                                                                  Table IV.
                                                            Elements of Phase 3, Preliminary Field Screening

                                                  Objective        Using simple and  inexpensive field  investigation  techniques, confirm and
                                                                 qualify site contamination problems

                                                  Assignment of    Corporate and division coordinators assign a site investigation team to work
                                                  Responsibility    directly with plant or facility personnel. The team should consist of a Hydro-
                                                                 geologist, Environmental Scientist and an Environmental Engineer. The learn is
                                                                 responsible for the design of each site screening program and for assuring that
                                                                 the program is properly carried out. At least one team member should be on-
                                                                 site during any field work

                                                  Site            Research—Prior to field work the following data should be obtained and re-
                                                  Screening        viewed:
                                                  Tasks          'Phase 1 and Phase 2 file
                                                                 •Published and unpublished geologic-hydrogeologic reports and maps
                                                                 •Expanded present and  historic site activities report from plant or facility
                                                                 personnel
                                                                 •Copies of any RCRA notifications or permit applications
                                                                 •Copies of any Superfund notifications
                                                                 •Available site maps and surveys
                                                                 •Present and historic aerial photographs of the site area
                                                                 •Waste generation and disposition inventory from plant or  facility personnel
                                                                 (present and historic)
                                                                 •Obtain any available environmental monitoring data
                                                                 Field Program Design—Based on a review of the above data, a one or two day
                                                                 site specific field "prospecting" program  is designed. The following activities
                                                                 might typically be included:
                                                                 •Conduct structured interview of facility personnel followed  by  detailed
                                                                 picture taking tour of site
                                                                 •Water quality survey of standing water bodies, streams,  drainage ditches,
                                                                 seeps, monitor wells (if present) and surface impoundments
                                                                 •Shallow soil sampling survey noting evidence of contamination (odor, dis-
                                                                 coloration, etc.)
                                                                 •Vegetation survey noting species present and evidence of death or stress
                                                                 •Site access survey noting accessibility to general public (particularly children)
                                                                 and means taken to control access

                                                  Info, and        The  responsible plant or facility personnel should be  able  to provide or
                                                  Data Sources     obtain all of the  research data required. Telephone assistance and advice can
                                                                 be provided by the site investigation team
                                                                                Equipment
                                                                                Needed
                                                                                Other
                                                                                Considera-
                                                                                tions
                                                                hand anger
                                                                post hole digger
                                                                shovel
                                                                surface water samplers
                                                                well bailers or pumps
                                                                sample containers
                                                                100 ft steel tape
                                                                   35 mm camera
                                                                   pH meter and probe
                                                                   conductivity meter and probe
                                                                   portable contaminant test
                                                                   kits (HACH type)
                                                                   reagents
                                                                   distilled water
                                                                Enlisting the aid of responsible individuals most knowledgeable concerning
                                                                the site is strongly advised
program is  flexible and can  be adopted  in  whole  or in
part as the needs in given situation demand.
   Depending  upon  the  goals  of a  particular audit, cor-
porate officials  may  require  a  simple problem potential
screening inventory of the properties  under their control
and any planned acquisitions or a  full  scale  site investi-
gation of the type,  source and extent of contamination,
definition  of  possible remedial  measures  and costs,  plan
for  implementing   remedial  work,  environmental  per-
formance monitoring and liability control strategy.
   Phases of the  audit process  which  begins  with general
property  inventory  steps  and  progresses  through  de-
tailed  site   specific   investigation   and   remedial   action
decision  making are  shown  in  Tables  I to XI.  The ele-
ments  of  each  phase  are described  in detail,  with  at-
tention  to  objectives, assignment  of  responsibilities,  in-
formation sources and data needs, equipment needs, out-
side services required and other considerations.

CASE HISTORIES—UNINVESTIGATED
LAND ACQUISITION
Asbestos Contamination
   This case history clearly illustrates  the need to conduct
careful  screening  audits  of  properties  before  purchase.

-------
                                                                                       LIABILITY, LEGAL & PUBLIC ISSUES      401
Late in  1978, a  developer purchased an abandoned Con-
necticut gasket factory and surrounding acreage intending
to develop  a  shopping center  on the parcel  (see Figures
1 and  2). During the early stages  of  construction  it  was
discovered  that  the  site was  dotted  with asbestos  sludge
pits and asbestos gasket waste piles.
   Concerned for the health of workers and area residents,
local health officials contacted  State  and Federal officials.
Teams of inspectors from the Connecticut Department of
Environmental   Protection  (DEP),  U.S.  Environmental
Protection  Agency  (EPA) and  the  Occupational  Safety
and  Health  Administration  (OSHA)  descended  on  the
site. Construction was immediately  halted.
   After several days of negotiation,  the developer  signed
a DEP consent order. THe twenty compliance conditions
specified in the  order involved  everything  from inform-
ing  workers  of  the  hazards  associated  with  asbestos,  to
the  consolidation and  proper  on-site disposal  of  all  as-
bestos wastes. Construction  delays and  compliance costs
added $900,000  to development  costs. If a simple and in-
                                                                expensive  site contamination screening  audit  had been
                                                                conducted, the developer  would  have avoided  the prop-
                                                                erty and its problems.


                                                                Coal Tar Contamination
                                                                   A  Pennsylvania  utility was notified by  the  U.S.  EPA
                                                                that  an oily  discharge  from  one  of  its properties  was
                                                                entering  an  important  trout  stream.  The  utility  con-
                                                                tracted with TRC to investigate the source  and nature of
                                                                the material and to  delineate the extent of contamination.
                                                                   A  review of the land use  in the site area  showed  that
                                                                coal  gas had  been manufactured on the site by  the previ-
                                                                ous owner  and that the oily  substance  seeping  into  the
                                                                stream  was coal  tar.  Aerial  photographs dating back to
                                                                1939 were used   to identify suspect areas  and help  de-
                                                                sign  the drilling  and  sampling program.  The boundaries
                                                                of  the  contaminated area  have  been  identified  and  the
                                                                feasibility  of  various  containment and  removal alterna-
                                                                tives is being studied.
                              Table V.
    Elements of Phase 4, Prioritization of Problem Properties
Objective
 Assignment of
 Responsibility

 Rating and
 Prioritization
 Tasks
Rating and
Prioritization
Info, and
Data Sources
Other
Considera-
tions
From field screening data develop preliminary estimate of the magnitude of
potential contamination and risk presented by each site and rate properties ac-
cordingly. Priority for site specific investigation and remedial action is estab-
lished

Corporate and divisional coordinators, site investigation team, and corporate
or outside counsel

Rating Potential for Health and Environmental Damage

Site rating methodologies have been developed  by: 1) LeOrand;' ' 2) JRB
Associates;'2' and 3) the Mitre Corporation.'3' The methods consider po-
tential receptors, pathways for contaminant migration, contaminant character-
istics, and site engineering characteristics. These methods vary in specific data
requirements and emphasis. Preference should be given to the development
of a rating methodology synthesized from these approaches that is simple to
apply and requires only that data already in hand as a result of Phase 1
through 3 efforts. Using the method developed, each site is evaluated and given
a numerical rating.

Rating Liability/Risk Potential
Close approximations of the financial liabilities presented  by a site cannot be
made until detailed site investigation work is completed and remedial action
option costing developed. However, a qualitative assessment of liability risk
potential can be made by answering the following questions:
•Is the property presently a source of off-site contamination?
•Has off-site environmental or property damage been documented?
•Are environmental regulatory enforcement or legal actions in progress or
 pending?
•Has a "Superfund" notification been filed for the site?
•Does the site contain poorly maintained discrete accumulations of hazardous
 wastes (i.e., drums, waste piles, sludge, etc.)?
•Is the property accessible to the general public?
•Are any private legal actions in progress or pending?
•Has environmental impairment  liability insurance for  the site ever been
 denied?

Based on the answers  to these questions,  the potential financial liability is
rated High Risk, Moderate Risk or Low Risk.  Three or more yes answers
should place a site in the high risk category.

Prioritization for Further Action
Sites with high health and environmental damage scores and a high risk lia-
bility rating would obviously be priority candidates for immediate emergency
stop action response and detailed site investigation

Phase 1 through 3, should have provided all the information and data needed
for Phase 4 prioritization. Follow up telephone conferences may be required
for clarifying data and answering liability risk questions

The value of legal counsel at this stage is of critical importance. The cor-
porate legal staff should be immediately briefed concerning all high score/high
risk sites
                              Table VI.
Elements of Phase 5, Immediate Emergency Stop Action Response

Objective      Prior to the start of and during the detailed field investigation problems may
             be uncovered that require immediate emergency remedial action. Rapidly con-
             ceived and implemented remedial action is taken to mitigate imminent public
             endangerment or potential catastrophic environmental release.

Assignment of   Corporate and divisional coordinators, site investigation team,  facility per-
Responsibility   sonnel, corporate legal staff and outside contractors develop and implement
             emergency remedial measures
Typical Stop
Action
Remedial
Measures
Equipment and
Services
Needed

Other
Considera-
tions
Actions taken will be site specific, responding to the special problems pre-
sented at each site. Typically, these measures might include:
General Response
•Immediately notify and brief all local emergency response officials
•Immediately provide site security—fencing, warning signs and 24-hour secur-
 ity personnel
Drums and Tanks
•Identify contents and quantity, determine structural integrity, transfer con-
 tents if necessary, label, number and properly stage
•As soon as practical, remove all containerized material for proper disposal
Pits, Ponds, Lagoons, Surface Impoundments and Waste Piles
•Identify contents and stored volumes, determine engineering specifics and
 structural integrity of any liners or containment structures
•Take measures as necessary to eliminate or control contaminant release and
 improve integrity of containment, such as:
 1. cease discharge of waste to facility
 2. remove waste for proper disposal
 3. transfer material to secure temporary storage
 4. install top liner
 5. increase dike or berm elevation and
 6. stabilize waste  physically and chemically in situ by fixation and/or
   solidification
Ground and Surface Water Contamination
Before any effective long-term program  to eliminate or control ground and
surface water contamination can be designed and  implemented,  extensive
site  investigations must be performed.  However, the following stop gap
measures can be instituted to provide some degree of temporary control.
•Identify obvious pathways of contamination such as drainage ditches, seeps,
 outflow pipes, etc. and seal them off
•Divert surface drainage away from contaminated areas
•Install combinations of cutoff walls, collection ditches and trenches to con-
centrate and collect strong leachates for treatment and proper disposal

To be determined on case specific basis
All emergency remedial work should be conducted by trained, experienced
and properly equipped personnel or outside contractors.  Work must be
planned in detail and contracts carefully drafted. Contract provisions should
include proper insurance, attention to regulatory detail and adequate emer-
gency procedures and contingencies

-------
402        LIABILITY,  LEGAL & PUBLIC  ISSUES
                                                                                   Table VII.
                                                        Elements of Phase 6, Detailed Site Field Investigation
Object)', e         A complete field investigation is designed and implemented to include samp-
                 ling and analysis  of  wastes and monitoring, sampling  and  analysis of area
                 soils, surface waters and groundwaters. The nature and extent of contamina-
                 tion, present impacts and potential future impacts are determined.


Assignment of    This investigation will need to be conducted by outside contractors selected
Responsibility    by the  corporate, divisional and plant coordinators, together with the ap-
                 propriate legal staff


Detailed          Preparation Tasks
Field            •Assemble and review all data collected through previous phases
Investigation      «Thc importance  of site historical data cannot be overslressed. Prior to con-
Tasks             ducting expensive field work, the record  must be clear concerning property
                  chain of custody, hazardous  materials use, and hazardous waste generation
                  and disposition.  Techniques for  obtaining  missing  information  include:
                  searching land records; obtaining historic air photos and maps; reviewing old
                  industrial directories and commerce reports; interviewing local historians;
                  and interviewing employees  of long standing,  former employees and em-
                  ployees of past operations
                 •Prepare request for proposals for detailed site investigation
                 •Review proposals,  interview  finalists and select environmental consulting
                  contractor
                 Typical Field Tasks
                 •Complete a photogrammetric survey of the site area. Maps  of a scale  I" =
                  20' or  1"  = 50' with a 1' or 2' contour interval are usually required
                 •Using  the site maps and aerial photographs, prepare a detailed surface and
                  subsurface exploratory program plan. The plan considers what data is already
                  known and provides for the stepwise collection of the remaining data needed
                  to meet the Phase 6 objective. The program would typically consist of the
                  following:
                  I.  Sampling and analysis of discrete waste accumulations, for which chemi-
                     cal analyses do not already exist
                  2.  Based  on waste characteristics and analyses,  select contaminant paramet-
                     ers, contamination indicator parameters, and ground and surface  water
                     quality parameters for subsequent monitoring, sampling and analysis work
                  3.  Conduct surface water sampling and analysis program for any standing or
                     flowing water bodies immediately upgradient, on, or immediately down-
                     gradient of the site
Equipment and
Services
Needed

Other
Considera-
tions
                 4. Complete test pit and shallow soil boring program, logging and sampling
                    materials encountered
                 5. Install piezometers and/or groundwater monitoring equipment in selected
                    shallow borings and test pits which intersect the water table
                 6. Conduct deep drilling and boring program, collecting split spoon samples
                    at 5' intervals or change in Uthology
                 7. Install piezometers and/or groundwater monitoring equipment in selected
                    borings and drill holes
                 8. Perform pump tests on selected well installations to aid in defining site
                    hydrogeologic characteristics
                 9. Conduct geophysical survey (eletrical resistivity and/or seismic) to confirm
                    and augment subsurface data and
                 10. Define surrounding property use, present natural resource use and po-
                    tential future nature resource use
                 Data Reporting and Evaluation
                 •Even if a field investigation program is well designed and carried out, sur-
                 prise discoveries will require  modification of the plan as work proceeds. This
                 is the norm rather than the exception
                 •At the conclusion of the investigation sufficient data will be in hand to de-
                 termine the extent and nature of contamination, present impacts and po-
                 tential future impacts. Typical outputs might include:
                  1. Results of waste, soil, surface water and groundwater sample analyses
                 2. Maps showing discrete waste accumulations, types and volumes
                 3. Maps presenting surface water quality data and flow information
                 4. Soil contamination maps and cross sections
                 5. Geologic (bedrock and surficial geology) maps and cross sections
                 6. Results from field and laboratory  permeability tests, transmissivity cal-
                    culations
                 7. Hydrogeologic maps, cross sections and flow  nets and
                 8. Groundwater contamination maps, cross sections and flow nets

                 To be determined on case specific basis
The contractor  should be a full service environmental consulting firm with
site contamination investigation experience and staff expertise in the fields of
Geology, Hydrogeology, Chemistry, Environmental  Engineering, Industrial
Engineering, Chemical Engineering, Hydrology and Environmental Science
                                                                                   Table VIII.
                                                Elements of Phase 7, Definition of Remedial Strategies, Risk and
                                                 Financial Liability Assessment and Remedial Cost Effectiveness
Objective        Several removal, containment and combination scenarios are developed  to
                 remedy the environmental and related health problems presented by the site.
                 The costs of  implementing each  scenario arc  calculated. Relative cost ef-
                 fectiveness is then determined for  each case in terms of implementation dol-
                 lars paid out versus estimated financial liability risks remaining after imple-
                 mentation.


Assignment of    It is recommended that the contractor on board for Phase 6 be retained for the
Responsibility     Phase 7 efforts, working closely with the corporate, divisional and facility co-
                 ordinators and assigned legal staff.

Remedial         Removal Scenario
Scenario         *ln cases where discrete  waste accumulations of known type and volume are
Development      the sources or potential  sources of significant environmental and/or health
                 damage, removal and proper disposal at permitted treatment and disposal
                 facilities is highly recommended
                 •Costs of removal and  proper disposal are easily determined by obtaining
                 quotes from reputable disposal contractors
                 •This scenario assumes that any residual contamination left after removal is
                 insignificant and of no future  consequence
                 Containment Sctnanos

                 •Containment  strategies seek to control the release of hazardous contaminants
                 to  the environment. Control is achieved by  isolation, partial isolation,
                 contaminant  collection,  physical  and chemical stabilization  and alteration
                 of ground and surface water flow to prevent off site contaminant discharge
                 •In  a  given situation one or any combination  of containment strategies
                 might be used
                 •Unit costs are determined using conceptual designs and  standard engineer-
                 ing costing procedures
                 •Containment  scenarios  assume that long-term  maintenance and  monitoring
                 »ill be required to assure conunued performance of the containment goals
                •Monitoring  and maintenance costs  for an assumed period (usually 20-30
                 years) are computed, adjusting for inflation, and added  to scenario imple-
                 mentation costs
                Combination Scenarios
                •In a given situation removal and any combination of containment strategic!
                 might be used

Cost            Costs
Effectiveness      •Costs, as computed above, are  taken as given for each particular remedial
                 scenario
                •Cost effectiveness is determined by employing an event/consequences analysis
                Eveni/Consequenct Analysis
                 •The event/consequence  analysis is a  semi-quantitative  con effectiveness
                  assessment  of the degree of risk control provided by implementing a re-
                  medial scenario of known cost
                •Each remedial scenario is evaluated in relation to the answers to the follow-
                 ing questions:
                 I. How might the strategy fail?
                 2. What is the probability of failure?
                 3. Are obvious or planned fall back strategies available to detect and correct
                     failure situations?
                 4. What financial risk is exposed as a result of failure in terms of resource
                     restitution, replacement, awards for health and property damage, and en-
                     forcement penalties?
                 5. It the financial risk manageable; or should a remedial scenario requiring
                     larger initial investment, but posing substantially lower long-term financial
                     risk be employed?

Other           It is recommended that an insurance consultant who is  expert in enviroo-
Considera-      mental liabilitiy matters be retained to review con effectiveness issues related
tions             to financial liability risks

-------
                                                                                                      LIABILITY, LEGAL & PUBLIC ISSUES       403
                                   Table IX.
  Elements of Phase 8, Selection of Preferred Remedial Strategy

Objective        After careful examination  of the pros and cons attending each  remedial
                scenario developed through Phase 7 efforts, a preferred strategy is selected for
                implementation. The alternatives considered and the rationale for selection of
                the preferred strategy are presented to the appropriate regulatory agencies for
                review, comment,  negotiation and final concurrence or approval

Assignment of    Corporate, divisional and facility coordinators working with assigned legal
Responsibility     staff and outside consultants

Info. Req'd.      The data needed for presentation of  the strategy selection process should be
                in Data on hand  as a result of Phase 6 and 7 tasks.  This  information will
                include: detailed field investigation reports and maps; remedial scenario con-
                ceptual designs and costing data; and cost effectiveness and financial liability
                risk evaluations
                A statutory and regulatory compliance brief should be prepared by  the legal
                staff to determine compliance issues and status related to implementation of
                the preferred remedial strategy. Details concerning the need for any regulatory
                permits and steps to  be taken to maintain compliance throughout remedial ac-
                tion implementation  should be addressed

Other           The importance of  regulatory agency concurrence and  approval of the re-
Considera-       medial strategy cannot be overstressed. While in certain cases approvals are
lions            not required, obtaining consent is still recommended. This reduces the poten-
                tial for costly regulatory agency second guessing during and after  remedial
                work.
                The legal staff must  be convinced that any instrument (letter of approval, con-
                sent agreement, compliance order, permit, etc.) issued by the appropriate
                regulatory agencies is both understandable and conclusive.


                                   Table X.
     Elements of Phase 9, Implementation of Remedial Action

Objective        The engineering,  schedule, logistics  and all  contractual arrangements  are
                finalized. Any required permits are obtained. Remedial work is initiated and
                completed.

Assignment of    Corporate,  divisional and facility coordinators work  with assigned legal
Responsibility    staff, outside consultants, clean up contractors, construction contractors and
                disposal contractors.

Contract        Most corporations have standard construction and service contract  formats.
Considera-       These formats often do not consider  contract conditions needed for  work re-
tions            lated to the handling, transport and disposal of hazardous substances. Topic
                areas  to be considered when drafting site remedial work contracts are listed
                below.
  1. Definitions                            20.
  2. Overall Project Management             21.
  3. Cleanup Operations Management         22.
  4.Construction Management
  i. Security                               23.
  6. Emergency Procedures                  24.
  7. Contingency Plans                      25.
  8. Personnel Safety and Training            26.
  9. Safety Equipment                       27
 10. Utilities                               28.
 II. Materials Handling                     29.
 12. Transportation and Permits              30.
 13. Treatment and Disposal Methods         31.
 14. Treatment and Disposal Permits          32.
 15. Record Keeping and Reporting            33.
 16. Insurance                             34,
 17. Performance Bonds                     35.
 18. Indemnification                        36
 19. Subcontracting Authority                37
Inspection of Work
Sampling and Analysis Protocol
Services to be Performed,
Responsibilities
Facilities to be Provided
Materials to be Provided
Equipment to be Provided
Personnel to be Provided
Project Start and Completion Dates
Project Milestones
Compensation to Contractor
Method and Schedule of Payment
Rate Schedule
Amendment Clause
Payment of Taxes
Title and Disposition of Equipment
Release and Discharge
Termination
Force Majeure
                                                                                  Figure 1.
                                                     Fitzgerald Gasket Plant, Torrington, Connecticut, 1976
                             Figure 2.
Shopping Mall On Former Fitzgerald Gasket Property,
               Torrington, Connecticut, 1980
                                                                                Table XI.
                                                   Elements of Phase 10, Certification of Performance and
                                                          Addressing Future Potential Liability Issues
 Objective        Confirm that remedial work has been completed as planned  and establish
                mechanisms to control any future potential site liability
 Assignment of    Corporate, divisional and facility coordinators work with assigned legal staff,
 Responsibility    outside consultants and regulatory agencies
 Certification     *A licensed professional engineer certification of as-built plans and specifica-
 of Performance    tions for the remedial work is obtained
                •The  certification is  reviewed by the appropriate regulatory agencies, in-
                 spections conducted and conclusive regulatory sign-off obtained via:
                 1. Letters of approval
                 2. Release of bonds
                 3  Statements of compliance with stipulated agreements
                 4. Statements of compliance with regulatory compliance order conditions and
                 5] Statements of compliance with specific regulations
                                              Control of       "Any  required long-term monitoring and maintenance is initiated to detect
                                              Remaining       and control future liability
                                              Liability         «The availability  and  economic feasibility of obtaining environmental  im-
                                                               pairment liability insurance, self insurance, or other forms of indemnification
                                                               is investigated
                                                              •Nonsudden environmental impairment insurance is now offered by the  fol-
                                                               lowing firms:
                                                               1. Shand Morahan          5. Swett and Crawford Management
                                                               2. Alexander and Alexander  6. Travelers
                                                               3. American Intern'l Group   7. Aetna
                                                               4. Stewart-Smith           8. U.S. Insurance—Lloyds of London

                                              Property         "Property transactions involving hazardous substance site  contamination
                                              Transfer          must  consider the responsibilities of the seller, buyer, lessor and lessee with
                                                               relations to liability, site maintenance and site monitoring
                                                              •Responsibilities and liabilities can be shared or assumed by one party
                                                              •Contracts and  sales  agreements  must be carefully  worded  by  attorneys
                                                               working closely with technical experts

-------
404    LIABILITY, LEGAL & PUBLIC ISSUES
CASE HISTORY—INVESTIGATION AND
CLEANUP BEFORE PROPERTY PURCHASE
PCB Contamination
  TRC  performed  a  contamination evaluation  for a
pharmaceutical  manufacturer.  The need  for  the study
arose when the potential purchases of the site required
that  the present owner certify that the site was not con-
taminated.  The  TRC  program included  a number of
shallow borings to determine whether soil or unsaturated
zone contamination  had occurred through spills.  A test
well  was drilled to determine the quality of site ground-
water. Because numerous industrial sources existed in the
area, routine  analyses  were keyed to  an organic chemi-
cal produced by the previous owner. These routine analy-
ses showed that some product spills had  occurred but
that the level of contamination was not significant.
  Analyses for PCBs, however, showed that several spills
had  occurred around the site.  While levels did not  ex-
ceed  "regulated" amounts  it was decided to delineate
and remove the contaminated soils to avoid future prob-
lems.

CONCLUSIONS

  In response to a clear public mandate, statutory  and
regulatory initiatives conceived to seek out, control  and
assign liability for  hazardous substance and hazardous
waste site.contamination problems are now in place. The
site contamination and liability audit process presented
herein has outlined methods for  locating,  investigating
and controlling hazardous site contamination problems.
  The property screening  phases are the most valuable
aspect of  the audit process. Screening identifies  hidden
problems  before time can make them worse. More im-
portantly,  however, is that  a low cost investment in prop-
erty screening prior to purchase can prevent the acquisi-
tion of liabilities of vast dimension.
REFERENCES

1. LeGrand, H.E., "A Standardized System for Evaluat-
   ing Waste-Disposal Sites," National Well Water As-
   sociation Publication, 1980.
2. Kufs, C., et al., "Methodology for Rating the Hazard
   Potential of Waste Disposal  Sites," JRB Associates,
   Inc., McLean, Virginia, May 1980.
3. Chang, S., Barrett, K.,  Haus, S., and Platt, A. "Site
   Ranking  Model  for Determining  Remedial  Action
   Priorities Among Uncontrolled Hazardous Substances
   Facilities,"  Working Draft, The Mitre Corporation,
   Metrek Division, McLean, Virginia, June 1981.
4. Anonymous,   "Insuring  Against  Liability  Under
   RCRA," Chemical Week, August 19, 1981.

-------
        INSTITUTIONAL LEARNING IN A BUREAUCRACY:
   THE SUPERFUND COMMUNITY RELATIONS PROGRAM

                                         STEVEN A. COHEN
                                          Columbia University
                                         New York, New York
                                      THOMAS G. INGERSOLL
                                        West Virginia University
                                      Morgantown, West Virginia
                                           JAMES R. JANIS
                                           ICF, Incorporated
                                           Washington, D.C.
INTRODUCTION

  With the passage of the Comprehensive Environmental
Response, Compensation, and Liability Act in December
1980, the federal government acquired funding and au-
thority to clean up the multitude of abandoned hazardous
waste dumps threatening public health and the environ-
ment across  the  nation. Primary responsibility  for ad-
ministering the Act's provisions was assigned to the U.S.
Environmental Protection Agency (EPA). EPA had, how-
ever, begun planning  its  program—the  "Superfund"
program—some six months in  advance, recognizing that
the  widespread publicity given cases like Love Canal
made the need for such legislation clear.
  In this paper,  the authors analyze the planning and
development of one component of EPA's Superfund pro-
gram—community relations—as an illustration of a gen-
eral model for institutional learning in a bureaucracy.
The model is presented first in outline form. Then, in the
third section, the development of the  Superfund com-
munity relations  program is described  to illustrate the
model. In the  fourth section of this  paper, the authors
analyze  the  structure of the community  relations  pro-
gram, in  its present form, in order to relate the bureau-
cratic learning process  to organizational  aspects of the
performance of program functions. The adequacy of the
model to the learning process in  other  kinds of institu-
tions is commented upon in the conclusion.
  The distinguishing feature of the development of the
Superfund community relations program, as a case  of
institutional  learning, is that it constituted more a re-
vision of an  institutional memory  than the acquisition  of
new knowledge. EPA, as a  ten-year-old institution, was
not only  well-acquainted with the  kinds  of functions that
comprise community relations, but in fact, had a signifi-
cant bias toward their  inclusion  in programs aimed  at
the  protection and  enhancement of the environment.
Thus, the learning that took  place in  the evolution  of
Superfund community  relations  did not begin with a
tabula rasa; rather, it refined and redirected familiar func-
tions. But it did so in a political climate that was becom-
ing unreceptive to the value of those functions.
A Schema for Institutional Learning

  The  learning process  illustrated  by the  Superfund
community relations  program can  be represented by
the following steps:
•The recognition of a problem to which the institution
 must respond.
'Research  on the nature of the problem and its implica-
 tions for institutional organization and performance
•Development  of policy based upon the  research con-
 ducted
•The development of means for disseminating the policy
•The acquisition of programmatic means for implement-
 ing the policy
•Monitoring and evaluation of the program implemented

  The learning process must not  be equated with the re-
search step. Rather, it should be regarded as all those ac-
tions that  lead to the incorporation of relatively novel
functions in the on-going activities of the institution. The
learning that takes  place,  in other words, is  constituted
by  the alteration of  institutional behavior and the ac-
tions leading up to that alteration.  Data-gathering, sub-
sequent analysis  and the  broadening  understanding of
the staff of the institution are only steps in the process.
These steps may serve primarily as instruments useful in
changing policy rather than being valuable in their own
right as the acquisition of knowledge.
  The alteration of institutional behavior is the result of
changes in three areas as a result of the learning process:
•The institution's self-conception.
•The institution's operations.
•The program in question and its results.
These factors require some explanation.
  Corresponding to (or  produced  by) the institution's
memory is a  conception  of its role  and significance
that is  passed down through the staff. This self-concep-
tion may  or may not accord with the conception  that
others—e.g. the public or members of sister institutions—
have of the institution. In any event, the self-conception
will dictate, to some extent, the manner in which  the in-
stitution's  functions are performed.
                                                  405

-------
406    LIABILITY, LEGAL & PUBLIC ISSUES
  On a more formal level, the institution's policy will be
codified in certain explicit standard operating procedures.
A learning process will alter policy,  alter standard  op-
erating procedures in turn  and,  ultimately, alter the ac-
tivities of the institution.
  Finally, the program under whose  auspices  the learn-
ing process begins will be  directly  affected, even in its
structure and operations, by that  process. These should be
consequences  for the results obtained by the program,
that is, the success (or failure) in various aspects of pro-
gram activities.
  The development of the  Superfund Community Rela-
tions program illustrates this learning process, and the
factors changed through it,  against the  backdrop of a
prior institutional memory.

COMMUNITY RELATIONS POLICY
DEVELOPMENT

   In April 1980 EPA established the Office of Analysis
and  Program  Development (OAPD) to provide  pre-
implementation  planning  and  legislative  support  for
the anticipated Superfund program. By mid-1980, OAPD
had established the theoretical parameters of the commun-
 ity relations program. The  office conceived of it as a co-
 ordinated program of federal government interaction with
a local community where that community is the subject of
a federal action. Community Relations was considered to
include local media and press relations, public participa-
tion, local government and local interest group relations
and public information/education.
   With the degree of public, congressional and media in-
terest in the pending Superfund legislation, OAPD recog-
nized that its planning process  had to emphasize effec-
tive response. In emergency situations, an effective re-
sponse was considered one that  ends an existent state of
emergency and permits a return  to substantially  pre-
emergency conditions. In proposed remedial action, an ef-
fective response was  considered  one that would permit
the  Agency to maintain its credibility while eliminating
the  environmental  hazard.  In either case,  the Agency
recognized that its response must  not only remove the
peril to the environment and to human  health but also
remove the perceived threat to human and ecological well-
being as well.
   With this  understanding as background, OAPD pro-
posed that the primary goal  of  the community relations
program  be to facilitate  the implementation of cost-
effective solutions to hazardous waste problems by en-
suring adequate communication  between government and
local communities  during  response actions.  It was as-
sumed  from  the beginning  that hazardous  waste  inci-
dents had  the potential to  generate extremely emotional
reactions  in  local communities. Thus, government ac-
tions  in  this extremely volatile  political environment
would have to be explained with great thoroughness and
care if they were to be fully understood by the commun-
ity.  In order  to achieve the goal of ensuring adequate
communication, a number of secondary goals were ad-
vanced:
•Ensuring the local community a  meaningful voice in
 those implementation  decisions  that  the community
 considers most important
•Establishing a program of public information and media
 relations that is appropriate to the degree of interest and
 concern about the site
•Anticipate potential conflicts and attempt to avoid con-
 flict whenever possible
•Establish standard operating procedures  that ensured
 extensive  interaction  between  local government  of-
 ficials and federal/state officials.
  OAPD generated an issue paper  which outlined these
goals and the resources needed to reach them for  a pro-
gram of the scope envisioned by  Congress. It was  antici-
pated that the Agency would be expected to have the or-
ganizational and technical capacity  to address commun-
ity concerns. This would only be possible if, from the out-
set, community relations were declared a  major program
element in Superfund. OAPD reasoned that waiting un-
til political pressures forced community involvement on
EPA would probably result in the loss of the opportunity
to manage community relations to facilitate rather than
impede cleanup actions.
  Although  the  Superfund program was to have many
elements which would be new to EPA implementation,
the Agency was not without experience in emergency re-
sponse to hazardous waste incidents. Much valuable  ex-
perience had been gained over almost a decade in the im-
plementation of Section 311  of the Clean Water Act. In
order to avoid serious  blunders in the  new  program,
OAPD would have to do a lot of learning in a hurry. To
do so, it let a contract to ICF Incorporated to take a seri-
ous  look at several hazardous waste emergency opera-
tions which had been conducted either under Section 311
or under state auspices in order to analyze the degree to
which community relations efforts had been effective. In
short, the purpose was to analyze what had been accom-
plished at these sites, what had been done well, where the
Agency had failed to provide  for effective community/
government  interactions and generate the lessons  which
would be applicable to the greatly expanded scope of the
Superfund program.
  The  analysis was based on  the assumption that with
knowledge of what had actually taken place in hazardous
waste emergencies across the country, a more realistic
program could be designed—one that was more under-
standing of, and responsive to,  the needs of the  public
and the capabilities  of government. In addition, by look-
ing at the community relations techniques  which had been
employed in the past in these situations, and seeing which
had worked well and which had not, EPA  would learn
what techniques to use in the future or to avoid. The re-
sult, it was hoped, would be a program which would more
effectively—and efficiently—involve citizens in their gov-
ernment's response  to these environmental problems and
inform them of actions underway in their community.
  The methodology employed was  to learn as much as
possible about the cases through  newspaper accounts and
interviews with officials in Washington—the staff of local
congressmen and EPA personnel—and then to travel to

-------
                                                                LIABILITY, LEGAL & PUBLIC ISSUES    407
the site for at least a week of intensive interviewing. Of-
ficials in  state  environmental agencies and Regional
EPA offices were consulted at length. Interviews at the
site  were conducted with  citizens who were actively in-
volved in  the problem, civic leaders, local government
authorities and environmentalists—as well as with people
living in the vicinity who had taken  no interest or ex-
pressed  no concern.  The  completed studies were en-
visioned as background material for the development of a
handbook that would assist EPA personnel  responsible
for implementing the Superfund program.
  By August 1980, it was clear that the researchers were
breaking new ground in the understanding of community
relations at the four sites under review, but it was equally
evident that a  more complete  picture of community re-
lations  around the country was required.  These initial
studies demonstrated a wide range of citizen concern with
and involvement in hazardous waste emergency opera-
tions; thus EPA decided to expand  the study to include
sites in each of^the ten Regions. Site selection was based
on  input   from  311  emergency operations  personnel
throughout the country, with  the final selection geared
toward sites at which Superfund operations might be ini-
tiated. The cases analyzed in this second round brought
the total number of cases in the study to 21.
  Even  as the  results of the  expanded series  of  case
studies were being analyzed, EPA began moving into the
field with  draft guidance to the Regions on the probable
extent of  a community relations program under Super-
fund. The first draft of a guidance document was sent  to
the Regions based on a preliminary analysis of the case
studies and the Superfund headquarters organization re-
quested  that the Regions comment on the effectiveness  of
the proposed guidelines. The draft manual suffered from
the fluidity of planning regarding the shape of the final
program. Although the legislation had now been passed,
the new Administration was not yet established  and the
shape of the final program was in some doubt. By the end
of February, however, Regional comments had been re-
ceived and catalogued by headquarters  and Michael  B.
Cook, EPA's Deputy Assistant Administrator for Haz-
ardous Emergency Response, issued interim  community
relations guidance for site cleanup.
  The interim guidance document established the policy
by which Superfund implementation would be guided  in
the community relations area.  This policy requested that
the Regions adhere to the following principles:
•informing the local community about Agency actions
•empathizing with local concerns, learning about the lo-
  cal community
•avoiding the generation of unrealistic expectations
•being open and forthright with information
•anticipating the formation of ad hoc citizen groups
•coordinating actions with local officials
•assigning  community relations coordinators whenever
  possible
•using a variety of participatory techniques
•considering the establishment of citizen advisory com-
  mittees at sites and spills  having a high degree of citizen
  concern
•providing adequate training for Regional staff.
  Operationally, each of these principles had been derived
from the 21 case studies which had been undertaken by
ICF and each was factored into the newly designed re-
medial and emergency response activity schedules for the
Superfund program.
  The guidance document for community relations during
remedial response activity  was divided into discrete re-
sponse phases and informed by a determination,  on the
part of the EPA on-scene coordinator (OSC), of the de-
gree of citizen concern at a site, combined with the ex-
tent of environmental damage  and the complexity of the
proposed technical solution. The determination of the de-
gree of citizen concern was to be based, at least in part, on
the results of interviews conducted by the OSC at the site
among local officials and concerned citizens.
  The methodology which  had been employed by ICF in
conducting the case studies had been found to  be so ef-
fective in ferreting out the concerns of the local commun-
ity that the interim guidance incorporated that methodol-
ogy into the schema for Superfund response at  all re-
medial actions. In addition, the value  to the Agency of
the ICF case studies was exemplified in the requirement,
placed in the interim guidance, for the development of a
Community Relations Plan for each site at which remedi-
al action was to be taken. The plan  was to be  designed
and executed as an integral part of the remedial response,
not simply appended to the response as an afterthought.
  During  emergency response  actions,  the  interim guid-
ance document noted the importance of keeping  the com-
munity informed of the Agency's actions and the need for
the establishment of a  close  working  relationship  with
local officials. The same principles which mandated the
generation of a community relations plan during remedial
response prompted the inclusion of a similar plan in those
emergency situations which might be followed by remedial
action. As a result, the community relations  component
of the Superfund program had become a significant ele-
ment of all responses which would be taken by the Agency.
   By the summer of 1981, the National Contingency Plan
(NCP), first developed  for the implementation  of the
Clean  Water Act, was ready for publication in the Fed-
eral Register. The newly revised NCP, as  mandated by
CERCLA, was to contain the  final guidance  for the im-
plementation of the Superfund program.  It would out-
line, in detail, the steps which  could be taken at remedial
and removal operations under the authority of the Act.
   Unlike  earlier versions of the NCP, the newly issued
guidance  fully incorporated the Community  Relations
function into the operation of a response. Even the format
of the  NCP points to  this incorporation. The text of
the new NCP mandates  the inclusion of all of  the  com-
munity relations  techniques, including the methodology
developed for ICF's case studies, the use of the  commun-
ity concern and degree  of technical difficulty factors in
determining  the  level of response necessary for an ef-
fective community relations program, and the inclusion of
Community Relations Plans in  all response actions.

-------
408    LIABILITY, LEGAL & PUBLIC ISSUES
  Annex XI of the NCP, titled "Community Relations,"
specifies the  mechanisms to be used, the design of the
community relations plans, and the nature of the commun-
ity relations program. The Annex states that:
     "All community relations functions will be fully in-
     tegrated  within the operational units responding un-
     der provisions of the Plan, and all aspects of response
     operations must include provision for the execution of
     community relations programs."

THE ROLE OF EPA HEADQUARTERS, REGIONS,
AND PRIVATE CONTRACTORS IN
IMPLEMENTING COMMUNITY RELATIONS
PROGRAMS

  The  basic  thrust of  all  efforts  in headquarters,  Re-
gions and  by contractors was to implement community
relations  programs in  the field.   Superfund  resources
would not be used for a nationwide education campaign,
but instead, would be directed to support response actions.
Headquarters
  Headquarters' role included the lead on policy formula-
tion, reviewing community relations plans, training, pro-
gram evaluation and resource analysis. Headquarters pro-
vided informational materials on the Superfund program,
expert personnel for temporary community relations work
at critical sites, overall contract management and other
support services in the field.

Regions
  Regions would take the lead on implementing commun-
ity relations programs, developing  site-specific public in-
formation materials, supervising regional community re-
lations  subcontractors  and developing community rela-
tions plans for headquarters.  The  community  relations
load unit was a matter of Regional discretion.
Contractors
  Contractors  would  provide  support services  to head-
quarters and regional community  relations program ac-
tivities. At headquarters, this would primarily be analytic
work in support of guidance and policy  revisions. In the
Regions, contractor work would be staff and background
work to permit the Regions to stretch their own person-
nel  to  allow  coverage at as many  sites as possible.  One
essential principle  was  to be maintained throughout the
program: contractors would never represent, or even ap-
pear to represent the Agency to the public, other govern-
ment officials, or the media (i.e., to anyone).


IMPLEMENTING COMMUNITY RELATIONS

Organizational Strategy: Headquarters
  One of the four branches in  the original design for the
Superfund  DAA staff office was the "Community Gov-
ernmental Relations Branch." In the original  proposal,
the Branch had three sections: (1) Community Relations,
(2)  Public Information, and  (3) Government  Relations.
Typically, minimal staffing for a section would be three
professionals,  one  clerical  person,  and a  supervisor.
When staffed, the Branch would have approximately 15
members. The original organizational proposal stated
that:
    "The general functions of this Branch would be to:
    (1)  develop public participation/information policy
    and guidance, (2) carry out a coordinated program of
    public education and participation, (3) provide com-
    munity relations support  to on-scene coordinators at
    critical abandoned waste sites,  and (4) coordinate
    the collection and dissemination of technical informa-
    tion and public education materials."

  The  proposal was considered by  the  Superfund Na-
tional Program Manager.  He decided not to establish a
distinct  organizational unit to carry out this function. In-
stead, community relations policy development and co-
ordination was lodged in the Policy  Analysis Branch of
the Deputy Assistant Administrator  Staff office in the
Superfund headquarters organization and each operating
division was  held responsible  for ensuring  the  imple-
mentation of community relations policy within  its own
functional area.
  The rationale for this decision was that community re-
lations was to be an integral  part of the  Superfund pro-
gram: a normal, routine part of doing business. Just as
there would be no distinct organizational  unit established
to monitor the health and safety of staff on-site, there
would also be no need to establish  a distinct organiza-
tional unit to handle community relations.
  This approach has certain intrinsic  strengths and weak-
nesses. On the positive side, it helps ensure that commun-
ity  relations could  be a  fully integrated component  of
response. Those organizations making technical decisions
would  also be  responsible for ensuring that local com-
munities would be informed of and involved in response
actions. This type of organizational structure ensures that
community relations programs will not develop a "life of
their own," and will be given priority only in connection
with central technical functions. A more significant ad-
vantage of the  integrated organization strategy is that a
difficult-to-explain, vulnerable function  is  given  lower
visibility and  protected  from attack from OMB, Con-
gress and others.
  In early 1981, when the Reagan Administration sub-
mitted the FY '82 budget, almost every public participa-
tion organizational unit in EPA was eliminated. It is pos-
sible that a distinct organizational  unit for Superfund
Community Relations would have suffered a similar fate.
  Nevertheless,  a distinct organizational unit for com-
munity  relations could bring significant benefits.  Such an
arrangement allows expertise  to be accumulated and then
concentrated in a single place. It also  ensures that a super-
visory-level individual would be held responsible for de-
veloping and implementing the community relations pro-

-------
                                                                LIABILITY, LEGAL & PUBLIC ISSUES    409
grams.  In the Superfund program, community relations
was a  major responsibility of  the  headquarters  policy
analysis branch chief,  but was only  one responsibility
among many.
  The Superfund  headquarters  community  relations or-
ganization was informal. At the staff level,  Steve Cohen
was responsible for developing policy and ensuring its
overall  coordination and implementation. Each operating
division in the Office of Emergency and Remedial Re-
sponse had a single individual responsible for ensuring the
implementation of community relations policy in Regional
office programs. These individuals met for  a short time
as an "Informal Working  Group," to work with Cohen
and develop a program to implement community rela-
tions policy. This informal group met several times be-
fore its collapse. Cohen found it more efficient to work
individually  with  each  division's  lead community rela-
tions person, rather than convene work group meetings.
Decisions reached by the group proved difficult to imple-
ment due to the fact that each community relations
"coordinator" had  to  clear all tasks with his  or her
formal supervisor.
Regions
  Regional   office  Superfund  organizations  evolved
throughout 1980 and 1981. During that time, community
relations was not organized according to any single  pat-
tern. In  mid-1980,  headquarters  asked  each Regional
Administrator to name a Regional Superfund coordinator.
These coordinators,  in most cases, came to  head the Re-
gion's Superfund organizational unit. In late 1980, head-
quarters requested that each Superfund coordinator name
a Community Relations Coordinator.  These community
relations  coordinators  were responsible for  developing
community relations plans and ensuring the implementa-
tion of those plans.  In  some Regions, plan  development
and implementation was assigned to the Public Affairs or
External Relations Office.

DEVELOPING AND IMPLEMENTING
COMMUNITY RELATIONS PLANS

  The Interim Community Relations Guidance issued on
February 25, 1981, required the submission  of commun-
ity relations plans. Because of the shortage  of EPA  per-
sonnel, the headquarters Superfund office planned to pro-
vide contract resources to the regions to assist in plan im-
plementation. Throughout the spring and early summer of
1980, the Hazardous Site  Control Division (OERR) re-
quested, received and reviewed 30 community relations
plans. Although some  of  these plans were quite good,
overall:
•The majority of  the plans did not contain a sufficient
 quantity of the high quality information needed to make
 an independent judgment regarding the adequacy of the
 plans
•Very little attention was given to the technical complex-
 ity of sites (as required in the Guidance)
•Most of the plans followed headquarters guidelines for
 cost estimates, but it was difficult to determine whether
 or not the level of community relations activity recom-
 mended was desirable given the lack of background in-
 formation
In order to develop consistent community relations plans,
it was decided to also provide contractor assistance to the
regions for developing these plans.
  The specific contractor assistance scheme which began
in August  1981, followed early in the Superfund  pro-
gram provided the regions  with staff assistance  from
Architecture and Engineering (A&E)  firms during tech-
nical  work on-site, and  from the Technical Assistance
Team (TAT) contractor. ICF Incorporated, the firm that
developed much of the community  relations  program
guidance material, was contracted to:
•Work with A&E  firms to develop  community  relations
 capability in these firms
•Develop community relations plans for regional office
 approval
•Monitor contractor performance in the field

FACILITATING INSTITUTIONAL LEARNING

  The entire effort at community  relations  policy de-
velopment and program implementation was an  effort to
facilitate rapid  institutional learning. The  political dif-
ficulties that EPA found itself in at a large number of
hazardous  waste sites  around the  country (e.g.,  Love
Canal, New York; Stringfellow,  California; Memphis,
Tennessee; Valley of the Drums, Kentucky; and Jackson
Township,  New Jersey) convinced Agency management
that new standard operating  procedures had to be de-
veloped and instituted  to forestall some of this opposi-
tion.  It would be impossible to carry forth the  mandate
of  "cost-effective" remedial actions if citizen protest
caused delay and redesign at scores of waste sites.
  The first stage in institutional  learning is to develop
new institutional  routines  or standard  operating  pro-
cedures.  The ambitious research program conducted to
develop  community  relations  policy  included  the  case
studies at 21  hazardous waste sites, interviews with per-
sonnel in every EPA  Region,  and  considerable  brain-
storming and analysis. Once the regional procedures were
developed and communicated to the Regions in  the form
of the February 25 Guidance and the Community Rela-
tions  Handbook, it was then necessary to provide a set of
incentives to secure program implementation.
  The incentives  provided fall under three  categories:
(1) leadership, (2) training expertise and (3) staff-contract
resources.
Leadership
  In a hierarchy such as  EPA, consistent messages from
the top, and  from headquarters to  the  Regions,  are
necessary to facilitate institutional learning and organiza-
tional change. At  the first headquarters meeting  of the
Superfund  Regional Coordinators  and at most  subse-

-------
410    LIABILITY, LEGAL & PUBLIC ISSUES
quent meetings, headquarters staff and the Superfund
National Program Manager  stressed the importance  of
Community Relations  in  Superfund.  This  message was
reinforced by a number of memos to the Regions.
  Regional Superfund  staff members were ready for and
in most cases welcomed this message. The field personnel
had first-hand knowledge of the nature of Superfund's
community  relations problem, and were  eager to take
steps to avoid these problems. Many Regional staff mem-
bers had already  dealt with scared and  angry citizens
and had been shouted at and  harassed at public meet-
ings. The Superfund coordinators were fully aware of the
fact that Superfund and hazardous waste issues were un-
like anything else the Agency had ever  faced. The vola-
tility of the issue  was particularly obvious. This meant
that the Regions were  receptive to this new message and
such receptivity made  the task of institutional learning
considerably easier.
Training/Expertise
   A training or  "briefing" program was  developed  in
spring of 1980 and fielded in  the fall of that year. The
purpose of the program was to brief Regional staff on
community relations issues and practices. In  addition, a
HQ Public Affairs Assist Team was proposed to assist
the  Regions in  implementing Community  Relations  at
critical  sites and  will hopefully be established some day
as  part  of the  Emergency  Response Team  in  Edison,
New Jersey.
  Finally, ICF Incorporated briefed the A&E and TAT
contractors on the lessons learned in the case studies. This
training program provided a means of disseminating new
organizational routines. It is, of course, too early to judge
the success of those efforts.
Contract Staff Resources
  Even if the Regions had the  disposition and the ex-
pertise to implement the Superfund community relations
without resources, all the good  intentions in  the world
will not result in program implementation. Accordingly,
headquarters attempted to  provide contractor commun-
ity relations planning assistance  to the  field to pay for
and plan implementation. It was  hoped that provision of
these  resources would reinforce the importance of com-
munity  relations  and stimulate the Regions to allocate
some of their own resources to this function.

CONCLUSION

  Change is a viable bureaucratic alternative. The in-
clusion of an effective Community Relations component
within the response capability of the Superfund program
demonstrates this viability.  The entire program has been
affected by this inclusion, such that the knowledge gained
during program implementation can be applied to  the
understanding of community relations effectiveness and
bureaucratic operations, as  well.  Had the community re-
lations function not  been integrated into the  operating
divisions  of the  program, this  institutional learning
could not have occurred.

-------
                 THE COMMUNITY HAZARDOUS WASTE
                             COORDINATOR PROGRAM
                                            BETSY GOGGIN
                                          ANN RAPPAPORT
                           Department of Environmental Quality Engineering
                                         Boston, Massachusetts
INTRODUCTION

  All the resources required to solve the hazardous waste
problem cannot be found in any one level of government.
No one, not state government, not the federal govern-
ment,  not industry, not any other group in society can
solve the hazardous waste problem alone. This premise is
the basis for the community hazardous waste coordinator
program.
  The Massachusetts Department of Environmental Qual-
ity Engineering (DEQE)  cannot do the whole job. The
Department's staff realized that in working toward the
resolution of the hazardous waste problem it must rely on
the interest and expertise of the people who are potentially
affected, that is, all the citizens of Massachusetts. It was
the recognition of this that led the Department to asking
the chief elected officials  in Massachusetts' 351 cities and
towns  to appoint community  hazardous waste coordin-
ators. The problem of past improper waste handling at sev-
eral locations led to this request.
  When the Division of Hazardous Waste was established
within the Department in January 1980, it  faced two
immediate uncontrolled sites problems, a huge list of sites
potentially containing hazardous waste, limited staff funds
and the legacy of  poor  communications  between  state
government and communities  during  several  hazardous
waste  incidents in the  past. Having a hazardous waste
coordinator in each community was seen as  a step to-
ward the resolution of both of these.
  The Division asked the community coordinators to as-
sist the Department in  gathering the necessary historical
information on the potential hazardous waste sites and to
coordinate the activities of municipal officials and citi-
zens in this  effort, because community officials and citi-
zens know their town better than anyone else and impor-
tant site-specific  information is readily available to them.
The Division also asked the coordinators to serve as the
community  contact persons for the Department in any
hazardous waste incident  and  to  be responsible for en-
suring  that the appropriate municipal agencies and citi-
zens are kept involved and informed.
  Beyond the need for assistance with these particular un-
controlled sites-problems, the Department was convinced
that the coordinators also could provide a service to the
Massachusetts hazardous  waste management  program.
The coordinators could become advocates in the commun-
ities for responsible waste handling.
  When the Department asked for the coordinator ap-
pointments, their role was described as coordinating local
groups and agencies  in responding to hazardous waste
problems and serving as a  lead community representa-
tive in all hazardous waste management activities. A pre-
vious  knowledge of hazardous waste was not required for
the position but a strong interest and the ability to  man-
age people and projects were strongly recommended. This
approach led to the large variety in the professional back-
grounds of the coordinators.

COORDINATOR RESPONSIBILITIES

  The specific responsibilities of a coordinator vary de-
pending on the needs of the community and the interest
and expertise of the particular coordinator. The materials
sent to the chief elected officials and to the coordinators
themselves outlined the range of activities involved in pre-
venting, detecting  and responding to hazardous waste
problems:
(1) Serving as primary community contact for the De-
   partment and developing community contacts so that
   they can relay important hazardous waste problems or
   questions to the DEQE regional hazardous waste co-
   ordinators.
(2) Learning which citizens/agencies/organizations within
   the community are responsible for or involved in haz-
   ardous waste activities:
   •Fire Department issues licenses for storage of  flam-
    mable materials above and below ground and also has
    primary responsibility for  responding to hazardous
    materials/waste spills
   •Police Department has  the ability to patrol the com-
    munity around the clock and has the authority  to ar-
    rest individuals engaged in criminal activities
   •Board of Health has broad powers to protect public
    health, as well as specific directives including the ap-
    proval of sites for hazardous waste  treatment/dis-
    posal facilities and receiving information from DEQE
    regarding hazardous waste handled within commun-
    ity boundaries
   •Department  of  Public Works or Town  Engineer
    knows the locations of sewers, drain outfalls, etc.
                                                   411

-------
412     LIABILITY, LEGAL & PUBLIC ISSUES
   •Conservation Commission has knowledge of wetlands
    and other natural resources in the community
   •Industrial  Development   Commission,   Planning
    Agency, and/or Chamber of Commerce know about
    the types of industry currently in the community and
    about which industries may  be planning to come in-
    to town—including  hazardous  materials  users and
    disposers. These agencies can also be very helpful in
    encouraging  the  participation  and cooperation  of
    local industries
   • Water Department Superintendent knows the sources
    of groundwater in the community, as well as sources
    of drinking water, recharge zones, etc.
   •Many citizens are knowledgeable in fields relating to
    hazardous waste
(3) Organizing local resources:
   •Contacting and developing local resources
   •Forming a hazardous waste  committee or task force
    to assess the hazardous waste/materials situation in
    the community and to choose activities to encourage
    good  hazardous waste management in the commun-
    ity
   •Working  through  the established authorities  to  en-
    sure proper hazardous waste  management  in the com-
    munity
(4) Educating citizens about:
    •The threat to public health and environment caused
     by hazardous wastes which are mismanaged
   •How  hazardous wastes  can  be sensibly handled by
    using the available technologies for  properly  treat-
    ing, recycling and disposing of hazardous wastes
(5) Preventing and detecting hazardous waste problems:
    a. Assessing the hazardous waste/materials situation
       in the community by identifying:
      •All of the  hazardous materials in  the community
       by  categories,  classes,  volumes,   locations and
       type/age of containment. In addition to industrial
       users, this  survey would include facilities such as
       hospitals, dry-cleaners, research laboratories and
       gasoline stations
      • What goes into the community landfill. Although
       it  is illegal, significant  quantities of hazardous
       waste, particularly that generated  by small firms,
       are deposited in landfills (with or without  the
       operator's approval)
      •Old abandoned dump sites by checking old maps
       and photographs and by talking to older commun-
       ity residents
      •Transportation routes for hazardous materials in-
       cluding truck routes and railroad corridors.  To
       the extent  possible, the frequency and timing of
       hazardous  materials shipments, as well as  the
       types of materials shipped, should be identified
      • Where underground gasoline storage tanks are lo-
       cated and beginning as a program  to require own-
       ers to test them for leakage
   b. Mapping out local water supplies. Becoming famil-
      iar with  the network  of aquifers,  surface waters,
      recharge zones,  and wetlands in the community,
      focusing on  those that feed drinking water sources.
      Determining where water supplies are with respect
      to potential  sources  of contamination.  Finding
      maps which already exist.
   c. Assisting the Department and the U.S. EPA in in-
      vestigating suspected hazardous waste dump sites
      by gathering the necessary historical information:
      •A complete site-use history
       *site uses over the years: industrial; residential;
        waste disposal
       "•owners of  the  site (past and present) property
        maps, deed descriptions and restrictions, etc.
       *chemicals manufactured or possibly dumped at
        the site
      •Types of waste suspected on site
       *nature of industrial  process  that generated the
        waste
      •Evidence of pollution resulting  from the site
      •Location  of nearest  surface  water, including
       changes in surface water elevations  (presence of
       dams/diversions, culverting   of  streams)  and
       changes in direction of flow by natural or man-
       made causes
      •Direction of groundwater  flow in relation to the
       site now and in the past
      •Location of nearest dwelling
      •Location of nearest drinking water supply
      •Type and permeability of the soil
   d. If an uncontrolled hazardous waste site is found in
      the community,  participating in DEQE's  and the
      community's decision-making process  to determine
      what kind of action should be taken. The range of
      choices at some sites includes:
      •Removing the material for disposal at a secure
       chemical landfill or treatment  at an  environmen-
       tally sound facility (this is feasible only in a very
       limited number of instances)
      •Designing and constructing a secure  containment
       for the contamination at the site
      •Bringing in an alternative source of water supply
       or treating a water supply to remove contamina-
       tion
   The Department believes that making these decisions
must be a public process. Unless the potentially affected
parties and the community understand the  options  and
help decide on a particular solution, the particular situa-
tion will be extremely difficult to resolve.
   e. Participating in the negotiations between the devel-
      oper and the community, if a location for an en-
      vironmentally sound hazardous waste  treatment,
      storage,  or disposal  facility is  sound in the com-
      munity.
(6) Ensuring that the community is prepared for a haz-
   ardous waste/materials emergency:
   •Learning about the resources available to respond to
     hazardous waste/materials emergencies
   •Working  with the  agencies responsible—fire dept.,
     civil defense dept.—to  ensure that the appropriate
     people are prepared to  respond  effectively and ef-
     ficiently

-------
                                                                 LIABILITY, LEGAL & PUBLIC ISSUES    413
TRAINING

  At this time, 83 percent or 290 of the 351 cities and
towns in Massachusetts have coordinators.  A volunteer
steering committee of the coordinators has been formed;
the Division and the steering committee are currently plan-
ning a training program which will assist the coordina-
tors in performing the above activities.
  The training program will provide the coordinators with
information on: enforcement and legal authorities, emer-
gency response procedures,  how to  detect  and  prevent
future problems and  the siting of treatment,  storage, and
disposal facilities.  The  training sessions will provide  a
forum for the exchange  of this information between pro-
fessionals  involved or interested  in different aspects  of
the hazardous waste problem.  The coordinators  will be
able to use the information in  their communities to en-
courage responsible hazardous waste management decis-
ions  and avoid repeating past mistakes.  Mistakes to  be
avoided  include:  building  a  subdivision   down-grad-
ient from the town dump, confusion during  a hazardous
materials spill and enforcement actions which were not
well  formulated because municipal  officials, including
police officers, did not understand their jurisdictions.


PERFORMANCE

  A  number of communities already have a  good under-
standing of sound hazardous waste management activ-
ities. For example, not  long ago a developer applied to
a Massachusetts city to build a shopping center. The center
was to be located above a likely source of  future water
supplies. The town agreed that  the center could be built
but stipulated that a system to collect the rain water run-
off from the parking lot must be constructed and that no
small hazardous waste generators, including  hair dressers
or dry cleaners, could be located in the center.
  One of the most important resources for the coordina-
tors' training program are the  coordinators themselves.
They are a  diverse group of people having many areas of
expertise: health officers, conservation commission mem-
bers, firemen, policemen, lawyers, hydrogeologists and
engineers. Many of them are already developing their own
emergency contingency plans or are mapping water  sup-
plies in relation to potential contamination sources.
  One coordinator has given two lectures in  the past sev-
eral months for other coordinators on how to map ground
water flow  and potential sources of chemical contamina-
tion.  He has also assisted one of Massachusetts'  Water-
shed  Associations in  establishing a regional  program for
community coordinators.
  Another  coordinator,  an environmental engineer, has
done extensive emergency response planning for a large
chemical company in Massachusetts. He has participated
oh the coordinators steering committee and has provided
valuable emergency response planning information to the
rest of the group.
  A third coordinator is a self-educated hazardous waste
expert. Since he was appointed- last fall,  he has  read as
much as possible and attended  many seminars and lec-
tures. He has become an important resource for other co-
ordinators and a dedicated advocate for the development
of the training program.
  During the  past several  months, many of the coordi-
nators have given the Department their ideas on how they
think the coordinator training program should be devel-
oped:
•legal authority within the municipality
•the general  concepts—still a  little confused  between
 "wastes" and materials
•specific suggestions as  to organizing the municipal of-
 ficials and  departments in order to  get better coopera-
 tion
•how to inspire citizen responsibility for hazardous waste
 management
•how neighboring towns can share information
•how to set up industrial-community seminars
•more about hydrogeology and its relationship to chem-
 ical contamination
•possibly a listing of types of generators and the kinds of
 waste they are likely to produce
•how to anticipate problems and deal with them when
 they occur
•how to effectively handle emergency incidents—do's and
 don'ts.


DEPARTMENT'S ROLE

  The  community hazardous waste coordinator program
has provided a number  of challenges to the  Department.
During the  past eighteen months when the coordinator
program was conceived  and  organized, the Department
was  undergoing  a  number  of  other  major  changes.
Changes involving organization, the delegation of respon-
sibilities and finding the resources to run the mandated
hazardous waste program. This was  a difficult time to
attempt to develop a new program. Finding the resources
required for the training program has been one of the most
difficult tasks.
  In the time of tightening budgets, it is often difficult to
fund the existing work load, let alone a new program. The
Department is  using in-kind assistance and resources out-
side  the agency  to develop the program including  local
colleges and universities, business and industry, environ-
mental education programs sponsored by civic, profession-
al, and educational organizations and other government
groups  such as  regional planning agencies and  federal
agencies. Specifically, Tufts  University, Boston Univer-
sity,  the Monsanto Company, Massachusetts Cooperative
Extension Service, Nashua  River Watershed Association,
Lincoln-Filene Center for Citizen Participation, Metropol-
itan Area Planning Council are among the organizations
that  have been actively involved in the program. Essen-
tial,  however,  is the  involvement of full time staff mem-
bers  during the program's formative stages. A number of
staff people also are providing part time assistance in spe-
cific technical areas.
  Funding a program using outside resources and in-kind
services creates the difficult problem  of coordinating all
these activities and maintaining the spirit of cooperation

-------
414    LIABILITY, LEGAL & PUBLIC ISSUES
and enthusiasm. Whether this funding approach is viable
remains to be seen.
  The unconventional nature of the community coordina-
tor program caused a certain amount of confusion within
the Department. It is certainly an atypical endeavor for a
regulatory agency. Although  the program was a  logical
step for the Department's key decision makers due to their
long-standing commitment to involving the public in agen-
cy decisions, it was not readily accepted by all the staff
members. Nearly a year of discussion and delays ensued
before the concept  gained acceptance. Some staff mem-
bers  held the  traditional view that  government officials
do their jobs well and that citizens need only be involved
minimally, if at all.
   If the Department had anticipated the reluctance of the
staff to accept the coordinator program, some of the re-
sulting  delays might have  been avoided.  The presenta-
tion  of the program could  have been altered to alleviate
some staff concerns;  in particular,  it was found that a
simple, direct statement of the objectives  was the most
effective vehicle. The reasons were:
(1) To provide a solution to the serious communications
   problems between  DEQE and municipal governments
   which occurred in several hazardous waste incidents
   such as  those in Acton,  Bedford, Lunenburg, Nor-
   wood and Woburn.
(2) To let the public and local officials know that DEQE
   has extremely limited resources and cannot solve the
   hazardous waste problem alone.
(3) To provide a person in each community who is both
   concerned and knowledgeable about the problem of
   hazardous waste to help avoid repeating past mistakes
   such as:
   •building a subdivision downgradient from the town
     dump
    •siting an industrial park on an aquifer or watershed
   •panic/confusion during a hazardous materials spill
   •enforcement actions which  were not well formulated
     because municipal officials, including police officers,
     did not know the proper procedures
(4) To provide an advocate in each community for edu-
     cating  municipal officials, industry and the public
   about:
   •the variety of sound waste management options avail-
     able
   •the problem  of  continued emotional response to
     hazardous wastes
   •protecting groundwater resources
   •promoting responsible waste management practices
     within the community
    •procedures for response to a hazardous waste/ma-
    terials emergency
    •enforcement  activities  at  the Federal, State and
    municipal levels
    •detecting  potential  uncontrolled  hazardous  waste
    sites, especially abandoned municipal dump sites
CONCLUSIONS

  After reviewing the reasons why the Department asked
for the coordinators to be appointed, the staff agreed with
the intent of the program and became much more willing
to assist in its development. Perhaps an even more per-
suasive factor for the staff was the community coordina-
tors'  demonstration  of  their  interest  and  expertise.
Through their activities, it became apparent that local
citizens  do know their communities best and that impor-
tant community specific information is readily available to
them.
  In the long  term,  the benefits of having coordinators
should far outweigh the difficulties involved in getting the
program started. Today's limited budgets and the gen-
eral movement toward reducing the  size of government
make the coordinator's program even more relevant than
when it was conceived.
  The community coordinators are presently  looking into
the possibility  of forming a non-profit association. Such
an association would give the coordinators an independent
base outside of the Department and perhaps generate some
resources. Given the tenuous political support for environ-
mental issues in government, an association could provide
the coordinators with continuity they might  otherwise
lack.
  The coordinator program  is a reflection of the axiom
that  "hazardous waste is  everybody's problem." As the
coordinators become more and more involved in prevent-
ing and resolving specific hazardous waste problems, they
will be  a group of people actively taking responsibility
for problems created by mismanaged hazardous wastes.
  The community hazardous waste coordinator program
is still in the experimental stage. The Department's staff
is learning  an immense amount with  each step that the
program takes and hopes that the coordinators are learn-
ing enough from the program to keep them actively in-
volved.  The Department's staff encourages other agencies
to become involved in similar programs, are eager to learn
from  the experiences of others, and are optimistic about
the benefits of this type of public participation.

-------
              CITIZEN/GOVERNMENT INTERACTION AT
                                 TOXIC WASTE SITES:
                          LESSONS FROM LOVE CANAL
                                        L. GARDNER SHAW
                                                  and
                                       LESTER W. MILBRATH
                                     Environmental Studies Center
                                State University of New York at Buffalo
                                           Buffalo, New York
INTRODUCTION

  It is ironic, at least to environmentalists, that the effort
of William T. Love to develop hydroelectric power—a
form of energy  conversion that  is considered  environ-
mentally benign—should eventually result in one of the
most traumatic environmental tragedies of the 20th cen-
tury. Love's  grand  scheme to develop a new industrial
city utilizing  hydroelectric  power projected the construc-
tion of a canal to carry water from the upper  Niagara
River  to the generation point. After nearly 1000 cubic
yards of the  canal had been excavated in 1895,  the pro-
ject was abandoned.
  In the  1930's the abandoned  trench  had become  a
dumpsite; beginning in the 1940's it was  used by the
Hooker Chemical Company, the  City of Niagara Falls,
and according to recent assertions, the U.S. Army.  The
main user, however,  was Hooker Chemical which for over
a decade dumped hundreds of tons of toxic chemicals,
most of them enclosed in  55-gallon drums, into the un-
finished canal. When the canal was full, Hooker claims
that it was sealed with a clay cap, an accepted procedure
for such dumps at that time.
  During the late 1940's  and early 1950's  the City of
Niagara Falls expanded into the Love Canal area.  Over
700 homes and a school were built in the vicinity of the
canal during the late  1950's.
  Beginning in the late 1960's_and throughout the '70's,
the metal drums that had contained the chemicals de-
veloped holes due to corrosion and began to leak. Toxic
leachate rose to the  surface, especially after heavy rains.
During the 1970's it was discovered that leachate had mi-
grated horizontally  reaching  the  basements of nearby
homes and,  even later, evidence was uncovered that the
leachate had traveled several blocks.
  Chemicals  and drums coming to the  surface and the
appearance of numerous health symptoms, led a few local
residents to  press government agencies to investigate the
problem and take remedial action. Early  citizen requests
were ineffectual; local governments were not equipped
for the problems that citizens were asking them to  deal
with.
  In late  1976, the  N.Y.  Department of Environmental
Conservation (DEC), the City of  Niagara Falls and U.S.
Environmental Protection Agency (EPA) began to in-
vestigate water quality problems in the area. Meanwhile,
the State Department of Health began the examination of
possible health effects  associated with exposure to toxic
wastes. In April 1978,  Commissioner of Health, Robert
D. Whalen, determined that a health hazard existed in
the  Love Canal area and ordered the Niagara County
Health Department to cover exposed chemicals and install
a fence around the area. He also ordered health studies of
residents living adjacent to the canal.
  In May  1978,  EPA announced that air samples  it had
taken from homes adjacent to the canal contained dan-
gerously high levels  of toxic  vapors.  The  State Health
Department studies began the following montluand_found_
high levels of toluene,  chloroform, benzene and ehloro-
toluene.
  On August 2, 1978,  Commissioner Whalen issued an
order declaring  a health  emergency, recommending  a
delay in opening the 99th Street School, and recommend-
ing the evacuation of pregnant women and children under
two years of age from the immediate canal vicinity.
  Governor Carey visited the Love Canal area on August
7, and, at a  public  meeting, announced that the state
would purchase homes  affected by the Love  Canal chemi-
cals. Two days later  it  was confirmed that 239 homes in
two "rings" around the canal would be purchased. Resi-
dents within these first  two rings were offered temporary
relocation  at  state expense, pending purchase  of their
homes. Most of the residents affected by  this decision
were evacuated to temporary housing.
  During August, the State DEC had begun working with
the City of Niagara Falls and Conestoga-Rovers, the con-
sulting firm it had retained to review a site  cleanup plan.
The safety and health of workers and nearby residents be-
came an integral and controversial aspect of the interim
construction plan. The scope of the safety plan included:
•Security and communication provisions
•Personal hygiene and workers' safety
•On-site monitoring and sampling
•Emergency  evacuation provisions (this provision later
became quite controversial).
  Also, during August, the State Department of Health
took blood samples from over 2,800 Love Canal area resi-
dents. With such a large number, there were long waiting
lines which led to complaints about the planning and or-
ganization of the health study.  Following  the sampling,
the  residents eagerly, and later impatiently, awaited the
results. By August 1978, Love Canal had become a highly
                                                   415

-------
416    LIABILITY, LEGAL & PUBLIC ISSUES
publicized and  politicized case marked by an extraordi-
narily high degree of citizen/government interaction.

ISSUES

  By the end of August 1978, citizen activity in the Love
Canal area was directed toward a number of issues:
•How many homes should be purchased by the state and
which ones should they be?
•Which  residents  should be temporarily relocated, at
whose expense and on the basis of what criteria?
•What adjustments would be made  in the property  tax
liability of citizens whose homes had become less valu-
able and  which, if any,  homeowners would receive  the
adjustment?
•What provision would be made for the safety of nearby
residents  during remedial construction; who would  ap-
prove of those  provisions and who  would supervise and
enforce them?
•What negative health  effects  were attributable  to  ex-
posure to toxic wastes  at Love Canal; how would  the
types and levels of toxic substances be determined; how
would  health effects be determined and how would such
information be reported, and to whom?
ACTORS

  Virtually all  levels  and branches of government  that
might have had some possible connection with the Love
Canal case entered the picture at some point. The com-
plexity and frequency of governmental involvement fed a
sense of apprehension and  urgency  to  the  media,  and
through them to the public. On the other side, the media
attention and  the  public clamor brought  about greater
governmental involvement.

Government Agencies

  The first agencies involved were the Regional Office of
the New York State Department of Environmental Con-
servation, the U.S.  Environmental Protection  Agency,
and the Engineering and Planning Offices of the City of
Niagara Falls.  These agencies had become involved prior
to 1978  in an investigation of the causes of a water pollu-
tion problem noted in the Love Canal region. When the
source  was traced to  the Love Canal dumpsite, connec-
tions began to be made with the  complaints of local  resi-
dents of the surfacing of buried  chemicals and the onset
of unusual health symptoms. At this  point the  State De-
partment of Health entered the stage.
  Local  governmental officials  were,  not surprisingly,
reluctant to act in  a way that  might  attract unfavorable
attention to the area or to create animosity between the
people in the city and the chemical industry. The economy
of Niagara Falls relies on two industries: chemicals and
tourism—both of which could  be injured by what  was
beginning to transpire at Love Canal. Indeed, during all of
the early  activity that followed  the events of  August
1978, the City of Niagara  Falls was involved in negotia-
tions with Hooker Chemical Company over the latter's
prospective construction of a major office building in the
city.
  Two days after the Health  Commissioner's August 2
order was issued, New York Governor Carey created an
Interagency  Task  Force  to deal with  the  problems as-
sociated with the Love Canal waste site. The Task Force
functioned at both the state and local level. The state level
Task  Force, headquartered in Albany, consisted of the
heads of  the departments  of Health (DOH), Environ-
mental  Conservation  (DEC),  Transportation  (DOT),
Social Services, Insurance,  Banking,  Housing and Com-
munity  Renewal,  Equalization  and Assessment,  and
Disaster  Preparedness  (ODP). An  on-scene  "working
task force" had  offices located at first in the 99th Street
School and  later in homes, adjacent to  the canal,  that
had been  purchased by the state. This  local Task Force
consisted primarily of representatives  of the Departments
of Transportation, Health,  Environmental Conservation,
and Social Services. The on-site coordinator was Michael
Cuddy of the Department of Transportation. Throughout
this period most  of the governmental  activity at the Love
Canal was performed by agencies that were members of
the Task Force;  hence, most citizen interaction was  with
the Task Force or with the governor and/or  his repre-
sentatives.
  The State Department  of Transportation was selected
as lead agency in the Task Force because of its  experi-
ence with  supervising construction  activity and its back-
ground in dealing with property purchases and relocation
of families. There is a natural tendency for agencies com-
ing into a new situation to define a problem in terms
that are  within  their range of experience  and  compe-
tence; this experience also  will shape the way that  they
envision and pursue solutions to the problem they are con-
fronting. In  this  case, hardly  any agency had experience
closely relevant to the problems that had to  be tackled at
the Love Canal site. Experience  with highway construc-
tion does  not help very  much  in dealing  with an appre-
hensive public fearful of explosions and toxic gas releases.
A background in  negotiation  with homeowners on the
purchase of property via eminent domain  to build a high-
way does not necessarily equip an agency  with a working
model to deal with the purchase of homes  from owners
who are all too anxious to leave an area for fear  of  their
health.
  The primary federal agencies to become involved  were
EPA, FDAA (now the Federal Emergency  Management
Agency, FEMA),  and  the  Center  for  Disease Control;
other agencies also  became  involved  (e.g.,  HUD and
HEW—now HHS). FDAA became an actor  early when in
August  1978 Governor  Carey  began  seeking   federal
financial assistance; its successor, FEMA, ultimately was
a source of federal aid to both the city and the state. The
Center for  Disease Control  was  involved in  planning
health studies of residents and  was responsible for setting
up and managing long range health studies.
  To a great extent, the politicization of  the Love Canal
case was as  much  a product of citizens' perceptions and
misperceptions, apprehensions and misapprehensions, as it
was of ineffective or inappropriate activity on the part of
governmental agencies. Numerous agencies were involved,

-------
                                                                 LIABILITY, LEGAL & PUBLIC ISSUES    417
each perceiving the problem and their own  role differ-
ently. Each tended to function in the manner for which
its experience had prepared it. Because so much learn-
ing had to take place by all parties, it is little wonder that
at critical junctures public passions rose because citizen's
perceptions of agency performance did not mesh well with
their expectations of what agencies should be doing.
  Public appreciation of agency performance  was highest
when citizens perceived  agency personnel as being open,
straightforward and evenhanded. Agencies received low
marks  from citizens, and citizen tension and  anger grew
when agencies appeared to be concealing information,
soft-pedaling bad  news  or treating citizens in  a con-
descending or inequitable fashion.
  Citizens seemed  particularly sensitive  to continuity, or
lack  of continuity, of agency personnel, often expressing
frustration with  the  removal of officials that  they had
come to know and learned to work with. Many viewed it
as particularly salutory that Michael Cuddy was the co-
ordinator of  the local Task Force through nearly all of
the period that the state government was involved.

Political Involvement

  Governmental activity at that  site was stimulated and
facilitated by several concerned legislators who attempted,
through constituent service and  legislation to deal with
the various problems at the site. These included U.S. Con-
gressman John LaFalce, who represents that  district and
State Legislators Senator Daly and Assemblymen Murphy
and Pillitera.

Citizen Activity

  Much of the citizen activity in response to  events at
the Love Canal coalesced around the Love Canal Home-
owners Association (LCHA). During the summer of 1978,
Lois Gibbs became more prominent in homeowners ac-
tivities, and she was ultimately elected president of the
Association.  Her leadership was strengthened  by being
invited to attend a meeting at  the White House on Aug-
ust 9, that also included state  officials,  members of sev-
eral federal agencies and a representative of the President.
Mrs. Gibbs believes that state  and federal officials made
a deliberate move at this time to  recognize her leadership
because they  felt that she was someone they  could work
with.
  In the first stages, LCHA members tended  more to re-
act to governmental action rather than develop a strategy
to influence governmental officials. Shouting and react-
ing emotionally toward officials during public meetings
were modes of activity that proved to be effective in
gaining attention and  sympathy  in  the short  run,  but
proved to be less productive techniques in the long run.
As time went on,  the LCHA leadership learned to  fit
strategies to situations much more effectively.
  In addition to the Homeowners Association, several
groups emerged with smaller memberships and somewhat
different concerns. These included  the Concerned Area
Residents, the LaSalle Development Renters Association,
the "93rd Street Group", and the Ecumenical Task Force.
The first  two were composed  primarily of residents of
government-subsidized housing adjacent to the  canal.
Many  of  the  former were senior citizens; many of the
latter were from racial minority groups,  primarily black.
The Ecumenical Task Force is not really a citizens group;
rather  it is a local arm of Church World Services, an in-
terdenominational cooperative  relief  and  development
agency.

COMMUNICATIONS

Sources of Information

  The  public agencies used a full range of communica-
tion techniques. In many cases,  the device was a press re-
lease or a press  conference; several of the statements
from the governor's office and from the state Task Force
were of this variety. Public meetings also were utilized,
but with uneven  success. Unfortunately, these meetings
tended to generate more heat than light. Many residents
felt that the high  emotions that characterized these meet-
ings resulted because such meetings followed the style of
press  releases  which did not afford an  opportunity for
questions and clarification.
  In other instances, communication was more limited in
scope and more private.  For  example, many of the con-
sultations regarding the construction safety plan and the
tax abatement legislation appear to have been with only a
few members of the LCHA.  Also, when the results of a
chromosome study were available in May 1980, a serious
effort  was made  (before the  study was made public) to
communicate the results to the people who were tested.
  The  Task Force working at  the Love Canal site util-
ized a weekly newsletter, and generally speaking, the meet-
ings of the Task Force  included representatives of the
homeowners and  renters associations  and were often at-
tended by other interested citizens as  well.  The meetings
provided a regular source of information about activity
taking place on the local level. In addition to the news-
letter and word-of-mouth communications, the actions of
the Task Force also were reported regularly by the local
media.

Types of Information

  The  information that needed to be communicated to
the public could  be classified as either: a) progress re-
ports, b) policy statements or c) technical data or reports.
Government agencies were most successful in communi-
cating  on their progress.  While communicating the ex-
istence of new policy positions did  not prove difficult,
conveying the substance of policy did. Questions abound-
ed, for example,  regarding the  intent of the  state  in the
relocation of persons with health problems.  It also was
unclear to many what role the FDAA intended to play in
early August 1978.
  Most troublesome, however,  was  the communication
of information about health  and environmental studies.
Obtaining results from them seemed agonizingly slow. On
the other  hand,  early release of a study from the State
Department of Health elicited  citizen complaints about

-------
418    LIABILITY, LEGAL & PUBLIC ISSUES
lack of interpretive material or standards. One of the most
frustrating incidents for citizens was  an academic and
scientific  debate  that  ensued after the release  of the
chromosome study in May 1980.


ROLES OF CITIZEN GROUPS

  The events  at  Love Canal suggest  several important
questions  regarding the roles of citizen groups in such
settings: (1) What role should citizens  play in the policy
process? (2) How should citizen involvement be  timed?
(3) How can diverse citizen viewpoints be represented most
equitably  and effectively?  (4) How can damaging con-
flicts and  controversies among groups be avoided or miti-
gated?
  Most citizens are  neither  lexicologists,  nor are they
public officials with experience in making or implement-
ing public policies; yet,  citizens may  become  quite so-
phisticated about both technical  and political matters re-
lated  to  toxic waste  disposal.   However,  citizens are
resident-experts on their own communities and are best
equipped  to read their own concerns  and preferences.
Seeking, or accepting, citizens' suggestions on various as-
pects  of a safety and evacuation plan could be a per-
fectly  reasonable  action. On the  other  hand,  citizens
should not be expected to interpret  complex scientific in-
formation couched in sophisticated terminology.
  When citizens  are  given  some formal  role  in  public
policy making, it raises the question of the  extent to
which different citizens'  views should  be represented in
that forum. A concern expressed frequently by members
of other citizen groups was that the Love Canal  Home-
owners Association was afforded more attention and ex-
ercised greater influence on decisions, than citizens from
these  other groups felt was  appropriate. Yet, most of the
influence  exerted by  the LCHA was  through informal
channels;  the  organization  did not  sit  as a voting mem-
ber of any policy making bodies. Because citizen  organi-
zations were not  given a formal role in decision  making
at Love Canal, it avoided  some difficult problems that
formal representation  might engender. The  reaction of
citizen organizations  to the perceived influence of LCHA
was more a reaction  to what these groups  perceived as
substantive inequities rather than  procedural inequities.
  There are two important  exceptions to the previous
comments on formal representation. First, the LCHA was
afforded  permanent representation  at  meetings  of the
local Task Force  very early in the latter's existence. Ob-
jections by renters groups that their interests were not ade-
quately being represented by the homeowners led  quickly
to their inclusion as well.
  Second, a Love Canal Revitalization Agency was con-
stituted with the intention  of having three representa-
tives of the general community as well as representatives
of the Niagara Falls  and nearby Wheatfield Municipal
governments.  Disagreement arose over whether  specific
citizens groups should be represented on the agency or, as
the local government officials felt, the  three seats should
be  occupied  by representatives  of the "community-at-
large." This position  was partly based  on the  view that
the revitalization agency would be administering the pur-
chase of the last group of homes, as well as the reloca-
tion  of  renters,  placing  representatives  of   those
groups (homeowners and  renters) in a  conflict of  inter-
est. The view of the municipal officials finally prevailed on
this issue, although one of the citizen members of the re-
vitalization agency, Bill Wagoner, remained associated
with several of the citizen groups.
EFFECTIVENESS OF CITIZEN/GOVERNMENT
INTERACTION

  The effects of citizen/government interaction on policy
can be viewed from two perspectives: (1) the effect of that
interaction on  policy makers and policy processes and
(2)  the  effect on policy decisions. With regard to the
former, citizen  activity seemed to drive some agencies to
defensive, self-protective postures. This may have been
partly a function of the experience of the particular of-
ficials in dealing with the public. Some of the medical re-
searchers and other scientists involved in the events at Love
Canal were  very reticent about  talking with the public.
On the  other hand, officials in the State Department of
Transportation  (especially those active at the local level),
and local EPA officials were much more at ease.
  The Love  Canal case alerted the public to the dangers of
improper disposal of toxic  waste. Citizen participation in
those events was instrumental in bringing out new di-
mensions of the problem. Through their persistence, citi-
zens at Love Canal interjected themselves into the formal
or informal decision  making structure of many govern-
mental agencies. Citizens'  organizations came to  be re-
garded as parties that had to be consulted before many
types of decisions could be taken.
  In addition to the general effect of shaping perceptions
of the dangers of improper disposal of toxic waste and of
general policies that must be pursued to avoid such prob-
lems, citizen participation at Love  Canal influenced a
number of specific policy decisions. The most prominent
was the decision by Governor Carey to have the state pur-
chase homes in the Love Canal area; later there was a com-
mitment by  the federal government  to provide financial
aid for the relocation of the balance of the affected resi-
dents. It is doubtful that these decisions would have been
made this way without the  citizens'  vigorous and per-
sistent demands for help. The decision to purchase homes
in the outer ring is  the most important achievement of
Love Canal citizen efforts.
  State  legislation providing property tax relief for Love
Canal homeowners,  and federal legislation providing a
capacity for responding to toxic waste emergencies (e.g.,
the so-called Superfund) followed hearings in which Love
Canal residents participated. The occasions for the hear-
ings were themselves a response to pressure exerted by
citizens.  In  addition  to the  policies that were initiated,
numerous policies were revised or modified as a result of
citizen/government interaction. The safety and evacuation
plans and temporary relocation policies are illustrations
of this.

-------
                                                                     LIABILITY, LEGAL & PUBLIC ISSUES 419
IMPLICATIONS

  Many of the implications of the Love Canal case for
governmental response at future hazardous waste sites re-
late to the style in which  agency officials interact  with
the public. The importance of openness and straightfor-
wardness has already been  mentioned. There were, how-
ever, several  other techniques  that had a bearing on the
nature of citizen/government  interaction  and which,
therefore, could be expected to influence the effectiveness
of hazardous waste response efforts elsewhere.
  Citizens are apt to entertain somewhat naive concep-
tions of the role and capability of government, particu-
larly in the early stages of a critical situation.  As citizens
gain experience and sophistication they are more likely to
differentiate the roles of various levels of government and
of agencies  within  a given level or  recognize  the  con-
straints within which public officials must operate. Be-
fore this  sophistication  can  be acquired, a series  of
dashed hoped can reduce the credibility of some agencies,
and government in general, to the  point where effective
interaction with the public becomes  almost impossible.
Such  problems might be avoided if  officials encourage
limited, but realistic, expectations of agency capabilities.
  An early  problem  the Task Force seems to  have ex-
perienced was the release of information by lower echelon
members  of  the Task Force that contradicted other in-
formation being  released.  Reassignment of some  staff
members  to other duties may have helped alleviate that
problem.  However, the reassignments contributed to the
problem  of lack of continuity of personnel, cited previ-
ously.  Release of inconsistent or contradictory  informa-
tion not  only reduces governmental credibility, but can
lead to anxiety heightening uncertainty among citizens as
well. On the other  hand, rigid enforcement of  a "party
line," or  imposition of a "gag rule," on staff members
may have the effect of hiding agency activity from legiti-
mate  public  scrutiny or creating an  impression  of se-
cretiveness and lack of candor on the part of the agency.
This need not be an impossible dilemma; the dual goals
of coordination and candor  seem attainable.
  Promptness and orderliness in providing information is
important to agency credibility, to  say nothing of  its
benefits  for  relieving uncertainty  among  citizens.  Es-
pecially when vital issues such as adverse health  effects
are involved, citizens demand—and deserve—to  know
where they stand. Information may not be available as
soon as the public wants  it; therefore, it is important that
citizens be helped to understand from the beginning how
long it will take for  such  information to become  avail-
able. Even the orderly and accurate release of informa-
tion may lead citizens to feel that they face serious risks;
disorderly communication can only exacerbate that prob-
lem.
  Much of the confusion  and anxiety  generated by the
conflicting statements from government agencies and re-
searchers in  May 1980 could have been avoided had EPA
officials been able to release chromosome study results in
a more orderly fashion—as had been  their original in-
tention. The necessity of  going  to the public with test
results in  order to pre-empt a  press leak, precluded thor-
ough evaluation of the results of the study and created an
aura of importance around the study that was inappropri-
ate given its nature and scope.
  When citizens are already divided by conflicting goals,
the potential for serious  intergroup conflict can  be  in-
creased by a communication practice that appears to show
favoritism toward one group. There may be any number of
legitimate reasons why an agency may wish to encourage
or support a particular group but an appearance of pre-
ferential treatment may raise  fears that  other legitimate
interests are going to be overlooked. A tradeoff is  in-
volved in that dealing  with numerous  groups  may  be
more  cumbersome—and frustrating—than dealing with
one  or two; yet,  the   diversity  of  interests  may  be
capable of effective expression only through several dis-
crete groups.
  Evenhanded  treatment of groups may  become even
more of an issue where,  as in the Love Canal area, cleav-
ages of interest regarding government policy toward the
toxic waste problem overlap other social cleavages in the
community—in this case, race and economic status. The
renters tended to be black, older, and  of lower income
and they did not see  their interests as identical  to those
of the homeowners.
  Perceived preferential treatment may also lead to a view
that agencies are pursuing a "divide and conquer" strat-
egy. The extent to which this perception was created  by
malicious intent as opposed to bureaucratic ineptitude is
unclear. Perhaps the most frequently  cited example of a
"divide and conquer" strategy being employed by gov-
ernment in  the Love  Canal case related to the issue of
tax abatement. An official in the state Task  Force is  re-
ported to have told residents on  the west side of 93rd
Street that they were,  in effect, "sold out" by the Home-
owners Association in discussions on  the boundary lines
for tax abatement.
  Although circumstances frequently dictate  the use of
public meetings, such  assemblages have widely recognized
drawbacks,   especially  when  potentially emotional  or
highly technical subjects are on the agenda.  People who
are under great anxiety and  stress will naturally press
their case in such public meetings and the meetings may
turn into chaos. The  emotional concerns of  citizens will
feed upon each other and almost any response by public
officials is likely to be perceived as inadequate.
  In addition,  such meetings  are generally poor forums
for the display of technical or scientific  virtuosity on one
hand and for the unembellished reporting of raw data on
the other. Data must be accompanied by interpretation,
although this should not be taken to suggest  that citizens
are dullards. Many of the citizens active at  Love Canal
became  sophisticated  in  their  understanding  of   the
language and concepts of toxic chemical wastes.
  In some cases, an effective alternative to the large pub-
lic meeting  is private meetings of agency officials with
the leaders  of  various  citizen groups.  At such  "public
consultations," it is possible to provide an indepth pre-
sentation of information or explanation of policy, and to
share concerns  about  how the material might be received
by the public, in advance of a large meeting.

-------
420    LIABILITY, LEGAL & PUBLIC ISSUES
  Another useful method for improving communication
between citizens and  government is for  the citizens  to
acquire their own experts. In the Love Canal case one ex-
pert  was  funded by the state government and another
volunteered her  services  to the LCHA. Citizen-oriented
expert  consultations have greater credibility than "gov-
ernment experts" and can significantly improve under-
standing by citizens.
  Because there  was no "model" for citizen/government
information  at  toxic  waste emergency sites, decisions
about informing and  involving people  in policy making
tended to  occur on an ad hoc basis at Love Canal.  As is
typical when clear policies of equity have not been worked
out, it was in no  small  degree a case of the squeaking
wheel getting the grease. Unfortunately (from the stand-
point of citizen/government relations) the perception by
citizens that they needed to be vocal  and contentious in
order to gain attention led  to resentment and frustration
in many cases, and helped to create a pattern of adversari-
al relationships.  While adversarial relationships may be
unavoidable in such situations, more careful attention to
the nature and timing of citizen involvement might help
to avoid their more counter-productive aspects.

-------