EPA/540/8-91/012
OSWER DIRECTIVE #9355.8-01
MAY 1991
Design And Construction Issues
At Hazardous Wastes Sites
CONFERENCE PROCEEDINGS
HYATT REGENCY AT REUNION, DALLAS, TEXAS
MAY 1-3, 1991
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Office of Emergency and Remedial Response
Harzardous Site Control Division
Washington, D.C. 20460
Printed on Recycled Paper
-------
NOTICE
Development of this document was funded, wholly or in part, by the United States Environmental
Protection Agency. This document has not undergone a formal USEPA peer review. Since this
document is essentially a collection of papers presenting ideas of individual authors, it has not been
reviewed subject to USEPA technical and policy review, and does not meet USEPA standards for
USEPA document publication. The views expressed by individual authors are their own and do not
necessarily reflect the views, policies, or ideas of USEPA. Any mention of trade names, products,
or services does not convey, and should not be interpreted as conveying, official USEPA approval,
endorsement or recommendation.
This document is not intended to and does not constitute any rulemaking, policy or guidance by the
Agency. It is not intended to and cannot be relied upon to create a substantive or procedural right
enforceable by any party. Neither the United States Government nor any of its employees,
contractors, subcontractors or their employees makes any warranty, expressed or implied, or assumes
any legal liability or responsibility for any third party's use of or the results of such use of any
information or procedure disclosed in this report, or represents that its use by such third party would
not infringe on privately owned rights.
ACKNOWLEDGEMENT
The U.S. Environmental Protection Agency (EPA) wishes to thank all of those who participated in
the development of the Agenda, Proceedings Manual, and attended the first "Conference on Design
and Construction Issues at Hazardous Waste Sites", held between May 1-3, 1991 at the Downtown
Hyatt Regency at Reunion in Dallas Texas. The feedback received during and after the conference
was extremely positive; it is EPA's plan to sponsor this conference on an annual or biennial basis over
the next several years until a 'steady-state' in design and construction at hazardous waste sites is
reached.
Several individuals played an important role in making the conference the success it was. In
particular, Kenneth Ayers, William Zobel, and Edward Hanlon of USEPA, and Michael Blackmon
and Chris Fafard of PEER Consultants, and the PEER Consultants Word Processing staff, should be
recognized. We thank these individuals, the Conference Abstract Review Committee, and the authors,
speakers, panel members, and conference participants for a job well done. Their efforts will help
insure that design and construction efforts during hazardous waste site remediation will continue to
see quality improvements in future years.
Paul F. Nadeau, Acting Director
Hazardous Site Control Division
U.S. Environmental Protection Agency
Conference Project Managers
Kenneth W. Ayers, USEPA Edward Hanlon, USEPA
CDR William R. Zobel, USEPA Michael Blackmon, PEER Consultants
Abstract Review Committee Plenary Session Speakers
Robin Anderson, USEPA, Washington, DC Paul F. Nadeau, USEPA, Washington, DC
William Bolen, USEPA, Chicago, IL Timothy Fields, Jr., USEPA,
Walter Graham, USEPA, Philadelphia, PA Washington, DC
Edward Hanlon, USEPA, Washington, DC Robert E. Layton, Jr., USEPA, Dallas, TX
Donald Lynch, USEPA, New York, NY Colonel Wayne J. Scholl, U.S. Army Corps
Brian Peckins, U.S. Army Corps of of Engineers, Washington, DC
Engineers, Washington, DC James W. Poirot, CH2M-HU1, International,
CDR William Zobel, USEPA, Denver, Colorado
Washington, DC
Luncheon Speaker
Donald Brown, Stubbs Overbeck & Associates, Inc., Houston, Texas
ii
-------
PREFACE
CONFERENCE ON DESIGN AND CONSTRUCTION ISSUES
AT HAZARDOUS WASTE SITES
MAY 1-3, 1991, HYATT REGENCY AT REUNION, DALLAS
The first U.S. Environmental Protection Agency (EPA)-sponsored national conference on design and
construction issues at hazardous waste sites occurred between May 1-3, 1991 at the Downtown Hyatt
Regency at Reunion in Dallas Texas. Ninety-five presentations of technical papers with three panel
discussions on technical/policy issues and case studies were held.
Included in this publication are questions and answers from the panel discussions, as well as text of
the technical papers. In some cases, the authors' names and addresses are included at the end of their
respective papers. This 'Conference Proceedings' culminates and memorializes the significant efforts
made at and for this conference.
This national conference was warranted and timely due to the increased complexity of issues related
to this subject area and the growing number of hazardous waste sites entering design and construction.
The conference also had a different intent, agenda and format than other major hazardous waste
conferences. An open exchange of ideas to promote formal and informal discussion of design and
construction issues was planned, in order to encourage national consistency, help develop more
efficient and practical means to move design and construction projects through the pipeline, and
augment EPA's current efforts to revise its Superfund design and construction guidance and policies.
Topics covered a range of issues, including pre-design activities, construction administration and
claims, community relations, health and safety, and government policy. Participants include the U.S.
Department of Energy (DOE), Department of Defense (DOD), Bureau of Reclamation, and Army
Corps of Engineers, as well as EPA, numerous design and construction contractors and State agencies.
EPA wishes to thank all of those who participated in the first "Conference on Design and
Construction Issues at Hazardous Waste Sites". It is EPA's plan to sponsor this conference on a regular
basis over the next several years. The next conference is tentatively planned for early April, 1992 in
Chicago.
Future inquiries regarding this conference and next year's planned conference are encouraged to be
made in writing to the attention of: Kenneth Ayers, Chief, Design and Construction Management
Branch, U.S. Environmental Protection Agency, 401 M Street, SW, Mailcode OS-220W, Washington
DC 20460, or by contacting EPA's Design and Construction Management Branch at (703) 308-8393.
iii
-------
SUMMARY OF QUESTIONS AND RESPONSES FROM THE PANEL SESSIONS
DESIGN AND CONSTRUCTION POLICY PANEL SESSION
Kenneth Ayers (Co-Chair)
Hazardous Site Control Division
Office of Emergency and Remedial Response
USEPA
Charles Schroer (Co-Chair)
Acting Chief, Construction Division
USAGE
James Feeley
Chief, Superfund and Emergency Response Section
Texas Water Commission
Doug Smith
U.S. Department of Energy
John J. Smith
Acting Branch Chief
Remedial Operations and Guidance Branch
USEPA
James Moore
Baltimore District, USAGE
1. QUESTION: What support is available through the Corps of
Engineers or the Bureau of Reclamation for RCRA-
lead actions?
RESPONSE: The Corps and Bureau of Reclamation are available
and have done support of RCRA actions.
2. QUESTION: How are lessons learned at remediation sites in
Texas shared?
RESPONSE: Papers are presented at Conferences such as this.
The State is open to provide any information they
have to interested parties.
3. QUESTION: With regard to Texas State Enforcement actions, how
do you view cost recovery and what documentation is
acceptable with the courts?
IV
-------
RESPONSE: Cost recovery on non-NPL sites has been successful
on solid waste enforcement sites. On NPL sites the
recovery is conducted in conjunction with the EPA.
The State Superfund program has not proceeded far
enough to start recovery.
QUESTION: With the new shift to PRP lead RD/RAs, we need to
evaluate what the PRPs are like; some may be
trusted, some not. This may be determined in the
negotiation process. It should be considered that
the public does not trust the PRPs.
RESPONSE: A draft PRP Oversight Guidance Document has been
prepared and distributed to RPMs. Current EPA
staffing and funding will not be able to handle
oversight of a large number of new sites. The
amount of liability that the EPA will assume in PRP
oversight must be evaluated. Possibly each
enforcement action needs to be evaluated, case by
case, and as little oversight as necessary be used.
Begin with high oversight and, if the PRP is
determined to be performing acceptably, reduce the
amount of oversight. This approach is presented in
the guidance which is still out in draft form,
awaiting feedback. The guidance is flexible, based
on the RPM's evaluation of the PRP performance.
Part of the guidance was to have the PRP provide an
Independent Quality Assurance Team provide QA data
to the RPM for review. This point is still in
contention.
QUESTION: When can we expect an agreement between EPA, Corps
and Bureau of Rec on data validation?
RESPONSE: EPA Region 2 does have an agreement with the Corps,
and a national agreement has been proposed.
Current trends are shifting toward more autonomy to
the EPA Regions, and fewer national agreements are
being signed. EPA Regions and the respective Corps
and Bureau of Reclamation representatives need to
determine, on a Region-specific basis, whether such
agreements are necessary. If so, meetings should
be set, and decisions made, on this issue.
QUESTION:
RESPONSE:
What is the status of "Lessons Learned"?
the direction for distribution?
What is
A computerized "Lessons Learned" system was
developed at the Corps which every field office
could input to and be read at headquarters. The
-------
current status of the system was not available. It
is not currently being used for the Superfund
Program, but it will be adopted.
7. QUESTION: One of the problems many RPMs run into is the
acquisition of property during a Remedial Action.
Many Remedial Actions have been halted to allow
time to acquire a piece of property, property
easement or long-term lease. Under SARA, the law
requires that any property acquisition during a
Remedial Action must be accepted by the State after
the action is complete.
In Pennsylvania, the Commonwealth has refused to
accept any of the acquired properties. Since an
agreement with the Commonwealth has to be in place
prior to the property acquisition, projects in
Pennsylvania are stifled at the time. Is anything
being done to deal with the States' concerns that
they will be liable for any contamination remaining
on the site after the Remedial Action?
RESPONSE: It is a requirement under the law; we can't get
around this. EPA's tentative understanding is that
states may not be considered liable for any
contamination remaining on the site after remedial
action. However, if only an easement is needed
which will expire at the end of the Remedial
Action, State approval is not required. If we are
buying property and the lease actually comes to the
EPA, an agreement for transfer to the State is
required. In regard to the liability issue, in the
Superfund contract the States have agreed to
operate and maintain long-term remediation after
the EPA completes its efforts. We don't understand
or know all the details of why Pennsylvania has
taken this stand at the this time.
8. QUESTION: The States are mainly fearful of "owner liability"
for property which they must take over after the
Remedial Actions. Is there any effort at
Headquarters to relieve the States of this
potential liability in the future?
RESPONSE: As discussed above, EPA's tentative understanding
is that states may not be considered liable for any
contamination remaining on the site after remedial
action. EPA is still investigating this issue.
However, it was never fully understood why
Pennsylvania was not willing to accept the property
VI
-------
acquisitions. This can be discussed with
Headquarters Council to see if there is some
wording which may assist in the negotiations with
Pennsylvania.
9. QUESTION: Do you think the Federal Government could be a
central data-gathering point for "Lessons Learned"
within the Corps, and States as well, and could
there be a publication of this data for those who
need this information?
RESPONSE: This would be the optimum; however, resources are
not available at this time to facilitate this large
an effort. For a computer database of "Lessons
Learned", significant screening of the data to be
input must be done. "Lessons Learned" may become
purely emotional or personal, which are not the
intent of this type of database. Information which
is not clearly worded and analyzed could be
misinterpreted or lead to liability. This will
require mature screening.
RESPONSE: The EPA Design and Construction Management Branch
produces a bimonthly flyer, "RD/RA Update", which
provides current information and "lessons learned"
on RD/RAs.
10. QUESTION:
11,
RESPONSE:
QUESTION:
RESPONSE:
In regard to setting design parameters, I
understand that there is a lack of data available
to make site decisions. When you get into the
construction phase, you need to take the
opportunity to seek the data to verify the design
parameters that you have designed with and make
necessary adjustments. Is the Corps of Engineers
putting into place any mechanism to keep the
Designer involved during construction to verify
design parameters?
Absolutely. An agreement is made with the designer
for involvement throughout construction to discuss
problems, etc.
In regard to AE liability, could this point be
expanded on?
In just the last few years the EPA has gotten
heavily involved in design, and we are now seeing
designs being implemented. EPA's REM contracts had
the standard AE liability clause, which states that
if there is an error or omission that the AE firm
Vll
-------
12,
QUESTION:
RESPONSE:
13. QUESTION:
RESPONSE:
14. QUESTION;
will go back and correct this error or omission at
no cost, and if the error or omission was caused by
negligence and the error caused significant
increased costs the AE firm will be liable for
these costs (rough interpretation). If the EPA
gives the designer definitive direction, more than
likely the EPA is assuming liability for the
affects of that direction.
The guidance given to the AE determines if the AE
is liable for errors. An AE liability clause is
being drafted specifically for the ARCs contracts.
However, the negligence standard, Section 119 set
up for indemnification, may conflict with the
liability clause; it is not a clear-cut issue.
What are the differences between the two liability
clauses?
Section 119 imdemnification addresses third party
liability associated with releases or threatened
releases. AE liability is two party, which focuses
on design errors or omissions. The clause being
drafted for the ARCs contracts is not substantially
different. It clarifies that for a cost
reimbursement contract, if an error or omission
occurs the EPA is only asking for the error or
omission to be corrected at no cost; EPA is not
asking for additional design work to be performed
gratis. Also, it clarifies negligence portions so
there is not a conflict with the negligence
standard, Section 119.
What efforts have been made by the Corps; of
Engineers to involve small, disadvantaged or women-
owned business in your contracts?
Efforts are being made to track small and
disadvantaged business (SDB) contracts and insure
the set-aside levels are met. Future effort will
be made to track the amount of subcontracts which
are awarded to SDBs.
Should private industry be performing site clean-up
with the question of clean-up sufficiency? Could
other mechanisms be used to achieve clean-up?
Could the property be given to the AE firm in
return for the clean-up? The EPA providing funds
to the Corps who then pays an AE does not seem
efficient.
viii
-------
15. QUESTION:
16,
RESPONSE: An "orphan" site is a site which does not have a
viable PRP. This does not necessarily mean there
is not an owner; it means that the PRP does not
have the funds or ability to do the remediation.
The EPA cannot just take possession of the
property. The legal ramifications would be
extensive.
In regard to efficiency, the decision was made to
get the most technically qualified people working
on Superfund site remediations due to the potential
risks posed at these sites. Costs were not the
central factor. The Corps was determined to have
the technically qualified personnel needed.
Efficiency, strictly in regard to profit margin, is
not the whole picture.
In Navy programs there is not enough contract
oversight available. The oversight official cannot
keep up with the amount of data and information
generated by several contractors at several sites.
The contractors then operate with minimal
oversight. I think the Government or the EPA's
time would be better spent on enforcement.
RESPONSE: That is one of the problems focused on by EPA
Administrator William Reilly — enforcement first;
however, resolving this problem will not occur
overnight.
The EPA has had a preponderance of excellent
contractors and encourages initiative by
contractors. It depends on your outlook and how
you want to use a contractor. Overall the products
received by the EPA have been excellent. However,
the EPA would prefer to have 100% enforcement
actions and not spend any of the Superfund.
QUESTION: Any lessons learned from Value Engineering Studies?
RESPONSE: We are very receptive to contractors proposals on
how to better clean up sites. At the Sikes
project, a major revision is underway. Under State
law, however, we cannot share the savings.
There is a Federal value engineering clause in
which savings achieved through a value engineering
study are shared with the contractor. The
Bridgeport, New Jersey site had a value engineering
ix
-------
proposal which was accepted by the government
during construction.
17. QUESTION: I understand there is an interagency agreement
between the EPA and the Corps that states any
contract that is more than $5 million will go to
the Corps. I recommend that this limitation be
extended from $5 million to $10 million. There are
45 ARCs contractors, and you will achieve your
goals more quickly and efficiently.
RESPONSE: There is no dollar value specified in the
interagency agreement. It was strictly a policy
call on EPA's part. Any projected remedial action
of less than $5 million, the Regions have their
choice of using the Corps, bureau of Rec or an ARCs
contractor for either or both Design and
Construction. For contracts between $5 million and
$15 million, the Regions haves the choice of using
an ARCs contractor or Corps of Engineers for design
and implementing the construction through the Corps
or Bureau of Rec. Anything over $15 million is to
be designed and constructed by the Corps of
Engineers or Bureau of Rec.
A policy letter has been drafted to consider
exceptions to this policy on a case-by-case basis.
The Regions would have to make a strong argument to
waive this criteria.
-------
COMMUNITY RELATIONS PANEL SESSION
Melissa Shapiro (Chair)
Office of Emergency and Remedial Response
USEPA
Michael McGaugh
USEPA Region I
Betty Winter
USEPA, Region IV
Karen Martin
Superfund Community Relations Coordinator
USEPA Region V
Louis Barinka
Remedial Project Manager
USEPA Region VI
Betty Williamson
Community Relations Coordinator
USEPA, Region VI
George Hanley, USAGE
Pat Ferrebee
U.S. Navy
1. QUESTION: Regarding your comments about not responding
argument by argument in the Responsiveness Summary,
how do we have a complete Responsiveness Summary if
we have, say, forty arguments in the formal
comments and do not address them individually?
RESPONSE: There are situations where the arguments get so
outside of the reality of the project that they
start creating problems that do not exist, be it a
hypothetical question or whatever. Instead of
being in a response mode where we literally go
through the arguments and respond sentence by
sentence, we got back to the basics of "did we
consider reduction of volume, toxicity and
mobility". We focused on those issues that they
were trying to attack, rather than getting into a
point by point discussion in the response.
XI
-------
QUESTION: My understanding of the Responsiveness Summary is
that you can take the comments and group them into
general points, as opposed to addressing specifics
to the letter.
ANSWER: Correct.
QUESTION: Concerning sending documents out to the TAG group
(Technical Assistant Grant group of the New Bedford
Community Environmental Awareness Group), they were
sent out as draft documents — were they EPA
internally reviewed, or what do you mean by
"draft"? Were they documents that actually came
from the RI consultant?
RESPONSE: The documents the TAG group received from their own
consultant were final documents; the group had
hired this consultant to generate their documents.
The TAG group received from EPA draft RI documents
prepared by EPA's RI consultant.
QUESTION: I assumed that EPA had a consultant doing the
RI/FS; did the TAG people see documents directly
from that consultant, or did EPA internally review
these documents, with their technical people,
before they went to the TAG group?
RESPONSE: EPA reviewed them first, they were draft final
versions.
QUESTION: Did you find, when you first started working with
the Community Group, that they were organized to
the point that they were ready to apply for the
Technical Assistance Grant, or anxious to do so, or
was it something that had to be worked with?
RESPONSE: No, the group had to be formed. Going back to the
initial meeting they had, there was difficulty
because the area was settled with Portuguese with
little English speaking ability. EPA community
relations coordinators went out with bilingual
material trying to generate interest for the group.
At an early meeting with the City of New Bedford,
one woman requested to be the project manager. She
organized sign up sheets which were taken to the
community.
QUESTION: Did you provide assistance or background for the
incorporation process which the Community Group
went through?
XII
-------
RESPONSE: They did that on their own. Problems developed
with attaining Non-Profit Organization status
through the IRS.
7. QUESTION: When initial newspaper articles (regarding a
hazardous waste site) came out, was there any sort
of direct response by the Navy, or did you just
kind of let them go and continue with your
community relations?
RESPONSE: We did a press release, and we also prepared
several fact sheets which we took to the community.
We put them in the post office, library, etc. We
had already set up an information repository — we
had done a few of the preliminary things we would
do for a CR. We did not respond directly to the
newspaper article because they had taken all our
words, distorted them, then devoted several pages
to the distorted version.
8. QUESTION: When your press officer experienced loss of
credibility, was there a communication strategy or
any kind of a CRP in place?
RESPONSE: We did not have a CRP in place. We figured out our
plan as events unfolded. We had never gone out and
done interviews, never prepared a formal CRP. I
can tell you now that the Navy does not like to be
in that kind of a situation. As soon as we realize
contamination on a site, we like to get started
[with community relations]. The sooner you get
started, the better off you are. We learned a
valuable lesson in Mechanicsburg, and if nothing
else, it was worth that experience.
9. QUESTION: Can you describe the situation in Mechanicsburg?
RESPONSE: We had PCBs in the drainage ditch, but that is not
what was described in the paper. Mechanicsburg is
a supply depot, where the U.S. keeps their
strategic supplies on a 700 acre paved site. We
keep strategic reserves of lead, chromium,
manganese, etc. Mechanicsburg is a ship's parts
control center. We have parts of ships in supply
there. It's one of those sites that, in time of
war, ships parts to whoever needs them. One of the
things that was a problem for us when the state
discovered the PCBs was that we could not identify
their source because we had no record of having
stored PCBs on that facility. We finally
Xlll
-------
10. QUESTION:
RESPONSE:
discovered that we had rebuilt transformers there,
and that was the source of PCB contamination. We
did an extensive storm sewer evaluation to try to
track down where the PCBs were coming from, and we
had to replace portions of that storm sewer.
We had talked to the newspaper about other sites,
though, including the site where we had buried
outdated medical supplies from World War II.
That's where the newspaper blew things out of
proportion.
Is this a base where Navy personnel live, and what
kind of Community Relations exist with the base
people?
We informed them first, because people on the base
don't like hearing about a base problem from a
neighbor or friend. They want to know about it
first. They were kept informed through briefings.
11. QUESTION: On the Formerly Used Defense Sites, does the Corps
conduct the community relations plan or is there
some leftover military facility that handles it?
RESPONSE: If it's an active installation, the Corps provides
technical support; the installation prepares; the
CRP. For remediation at active installations,
there is a book called The Commander's Guide to
Installation Restoration — an Army publication
from USATHAMA — that says the responsibility of
the installation commander is to be the paragon of
environmental virtue. The cleanup of a site is his
responsibility, so even if the Corps may be running
everything else at a site, the chairman of the
technical review committee is invariably the
installation commander.
The CRP is usually prepared by active
installations. We do offer to them to do the CRP
and allow them to fine tune it, in which case we
would turn it over to a contractor. The decision
was made a long time ago that, with the number of
sites we have in the Kansas City district, we would
need an enormous number of people to prepares all
the CRPs or even to manage that many contractors.
xiv
-------
HEALTH AND SAFETY PANEL SESSION
Joseph Cocalis (Chair)
Office of Emergency and Remedial Response
USEPA
John Moran
Health and Safety Director, LHSFNA
Sella Burchette
USEPA
Les Murphy
International Association of Fire Fighters
Denny Dobbin
NIEHS
Thomas Donaldson
Omaha Division, USAGE
Ira Nadelman
USAGE
Mary Ann Garrahan
OSHA
Diane Morrell
Ebasco
1. QUESTION: Is a hazardous waste site defined by OSHA?
Example: An office trailer in the support zone.
Do these workers need training?
RESPONSE: There is an internal inconsistency within the
standards. In Section E it is defined that if an
employee is on site regularly, and has no exposure
or potential for exposure, he still needs 24 hours
of training and one day of on-site training. This
includes everybody that enters the boundaries of
the site.
In Paragraph A of the standard: There must be an
exposure from a hazard on-site for 120 to apply.
The policy is, if there is a hazard the standard
applies, if there is no hazard from the site then
the standard will not apply.
xv
-------
COMMENT: Section A states that where exposures are known or
potential then the standards apply if in one of 5
categories. The potential for exposure includes
accidental exposure.
QUESTION: Does 1910.120 have a requirement for one Health and
Safety Plan, rather than 50, one for each
contractor on site. Should there be just one plan
for the entire site?
RESPONSE: That is the way the standard is interpreted.
COMMENT: The contractor should approve the plans for the
subcontractors.
COMMENT: There is no rule saying there has to be just one
health and safety plan (HASP) , but any others
should be just as strict and specific as the prime
contractors, to make life easier. It is the Prime
Contractor's responsibility to oversee the
subcontractor's HASPs.
COMMENT: The contractors write the HASP and the
subcontractors must comply. It is not always the
case where it is one site, one HASP, especially for
very large sites where there are four or five major
contractors. A better way to work it is one
project, one HASP.
COMMENT: During contract solicitation, they normally require
a HASP. Exceptions are when contractors onsite
want a more restrictive plan. They may add to the
minimum OSHA and Corps requirements.
QUESTION: Should contractor's health and safety record be a
requirement for bid specs?
RESPONSE: For a request for proposals, the health and safety
record is a factor. For an IFB it is not a factor.
Contractors at this point should be aware of the
requirements. Some major contractors, for example,
the government, considered this a prequalification.
QUESTION: Congress passed a law about agencies assuming
responsibility for worker training. Of the
1.8 million workers that qualify for this, 53% are
firefighters, who have the highest risk, but are
not adequately trained. Thirty-one percent aro law
enforcement agents, also high risk and one percent
are hazardous waste workers who have a lot of
xvi
-------
training. Doesn't the firefighter agency have a
direct responsibility to provide adequate funding,
as Congress had intended the money to be used, as
in the case of firefighters with high legitimate
claims to be trained rather than for EPA to assume
responsibility.
RESPONSE: The request for training is very competitive. Many
agencies apply, but only a few get it. This agency
has filled their obligation thus far with the money
that is available. If there is more money, there
will be more training. The Hazardous Materials
Transportation Act of 1990 is providing a grant
program at the local level through public sector
agencies. When this grant goes through, there will
be more training from the fire service.
5. QUESTION: If the funding is inadequate, why not go back to
Congress and say there is not enough money for
training, so these people can do their jobs
effectively. It seems as though the money is going
to the wrong people. Educational Institutions
should not be getting the funding because they do
not have the authority to help. People who do the
work should be getting the money.
RESPONSE: The President's budget cut the programs funding in
half. Appropriations are fighting it now.
6. QUESTION: What is to be done with utility companies wanting
to come on site for routine maintenance, main
breaks, etc. Should they have a trained contractor
on site with them? Is it okay to just have a site
safety person out there to make sure that the
utility company complies with the site HASP? How
does OSHA feel about this? Is this a violation to
have them on site?
RESPONSE: OSHA says there must be site representative who is
trained with a trained person monitoring. All
utilities should have properly trained people.
COMMENT: We must coordinate very early on. If you know
there are gas mains and underground utilities,
there is plenty of time before an investigation
begins to contact utilities and make them aware of
what is to come on this site. Be prepared because
there is no excuse to send untrained workers onto
the site to be exposed to what is there.
xvii
-------
COMMENT: These utility people won't be going onto the sites
V*»T-\7 (TFt-On .
very often.
RESPONSE: It is not okay to send untrained personnel on site.
COMMENT: There are contractors available to work for
utilities that have people that are properly
trained. There is no reason for utility companies
to put people's lives in jeopardy by sending
untrained workers onto a hazardous waste site. If
utilities don't have qualified people, there are
contractors who do.
7. QUESTION: What is going on around the country before we come
into town and say we have Superfund site? How are
the communities and responders, gas companies,
firefighters, etc., handled prior to the listing of
Superfund site? As far as training is concerned,
focusing only on Superfund is hitting only on a
limited portion of the market.
RESPONSE: Superfund is such a small portion of what trainers
have to deal with. Many fire chiefs, etc., are
stuck to tradition, not training. When dealing
with them (firefighters) on Superfund sites, you
get the same knee jerk reaction as with various
other kinds of exposures. It is very tough to get
the cooperation needed that you get from the Corps
of Engineers and the EPA.
COMMENT: Concerning the planning step of emergency response,
having this carried out at the remedial action
phase by the construction contractors will not only
entail the fire department but some of the issues
on utility services, etc. It is an issue that is
difficult with the Corps of Engineers in execution.
If it is difficult to affect the execution of the
emergency response plan because the service
providers are not trained or properly equipped to
perform the service. My recommendation for this
conference is that the support of a task force that
would look at the review process prior to design in
terms of the community service based around a
selected site and go through a thorough fact
finding process to delineate the best course of
emergency response that can get plugged into
design.
XVlll
-------
COMMENT:
CONCLUSION:
There has to be cooperation from the government to
recognize that they must get out more information
and get it out in a more timely manner. Also, the
need is the same from emergency responders. In the
past, when looking at any federal documents with
requirements a contractor could assume that these
requirements were being met. So in a situation
dealing with a large union, such as the fire
fighters, you go into a town, coordinating with
town officials and a fire chief. If in fact that
is not the way this needs to proceed, then that
word has got to get out to the contracting and
government community because it is a fairly logical
assumption to make, that if the fire chief says he
can respond there will be a response and there
won't be any question as to whether his people are
trained. From both sides there has to be a clear
view of what all the issues are.
We are all now witnessing EPA, Corps of Engineers,
and experienced trainers all working together for
Health and Safety. One year ago you wouldn't have
seen this. Hopefully, there will be more progress
in the year ahead, especially with training in
emergency response.
xix
-------
TABLE OF CONTENTS
CONFERENCE PROCEEDINGS
Page No.
I. CASE STUDIES
Composite Concrete Liners for Radioactive Wastes
Thomas Ambalam, Kaiser Engineers 2
Fast Tracking Remedial Design at the Cape Fear Wood Preserving Site
Thomas Clark, COM 7
Ambient Air Quality Management at French Limited Superfund Site
Bruce Dumdei, ARCO 23
Remedial Construction at the Industrial Waste Control Site, Fort Smith,
Arkansas
Santanu Ghose, USEPA 78
Bayou Bonfouca Superfund Site Case Study of Selected Issues
Robert Griswold, USEPA 108
Soil Remediation in the New Jersey Pinelands
Edward Hagarty, C.C. Johnson & Malhotra 128
When is a Superfund Remedial Action "Complete"? A Case Study of the
Crystal City Airport RA Implementation and Transition to O&M
Bryon Heineman, USEPA 138
WEDZEB Enterprises Remedial Action: Planning for an Efficient
Remedial Action Completion
Tinka Hyde, USEPA 161
The Landsdowne Radiation Site; Successful Cleanup In A Residential
Setting
Victor Janosik, USEPA 167
Remedial Design Approach and Design Investigations at the Bayou
Bonfouca Site
Kevin Klink, CH2M Hill 174
Value Engineering Studies of the Helen Kramer Landfill Superfund Site
Amy Monti, URS Consultants 202
Remedial Action In and Around Light Industrial Activity at the Denver
Radium Superfund Site
Timothy Rehder, USEPA 229
Streamlining Remedial Design Activities at the Department of Energy's
Monticello Mill Tailings NPL Site
Deborah Richardson, Chem Nuclear 238
xx
-------
TABLE OF CONTENTS
CONFERENCE PROCEEDINGS
Page No.
Construction of a Kaolin Clay Cap for Buried Nuclear Waste
C.J. Schexnayder, Nello L. Teer Co 249
Lessons Learned From Remedial Design of Helen Kramer Landfill
Superfund Site
Vern Singh, URS Consultants 277
Contract Security in Superfund: An Open Dialogue Between Government
and the Remedial Construction Industry
James Steed, Formerly of Texas Water Commission 286
Remedial Design and Construction at the Charles George Landfill
Superfund Site
Robert K. Zaruba, USAGE 292
II. COMMUNITY RELATIONS
Bells and Whistles: Community Relations During Remedial Design and
Remedial Action
Karen Martin, USEPA 308
Effects of Public Input and the Sampling Protocol on the Remedial Design
Process
Raymond Plieness, Bureau of Reclamation 328
III. CONSTRUCTION MANAGEMENT ISSUES
Remedial Design and Construction at the Picillo Farm Site
Mark Allen, Bechtel 335
Remedial and Post-Construction Activities at the Triangle Chemical
Company Site
Roger Brown, Weston 347
Concrete Cover Applications in Lined Drainage Ditch Construction
Camille Costa, Dynamac 358
A Case Study of Change Orders at a Supefund Site: Geneva Industries
Site--Houston Texas
Paul Cravens, Texas Water Commission 376
Transportation and Disposal of Denver Radium Superfund Site Waste
Richard Ehat, U.S. Bureau of Reclamation 390
Cost Estimating Systems for Remedial Action Projects
Gordon M. Evans, USEPA 399
xxi
-------
TABLE OF CONTENTS
CONFERENCE PROCEEDINGS
Page No.
HTW Construction Documentation Report: A Necessary Element in a
Successful Remediation
Heidi Facklam, USAGE 403
Change Orders Can Ruin Your Day: An Analysis of Construction Change
Orders in the Region 6 Superfund Program
Mark File, USEPA 409
Remedial Action Bids and Cost Estimates
Amy Halloran, CH2M Hill 420
RAC to PRP: The Thin Gray Line
Philip Kessack, ACRC 439
COS: An Expert System for the Analysis of Changes Claims
Moonja Park Kim, USACERL 472
The Tunnel Syndrome Solution: Can It Be Applies to Cleanup Projects?
Norman Lovejoy, Kellogg Corporation 487
The First Step for Strategic Environmental Project Management:
Environmental Cleanup Project Contract
James H. Paek, University of Nebraska 500
Permitting Superfund Remedial Actions or Nightmare on NW 57th Place
Lynna Phillips, EBASCO 518
State Oversight at Two Uranium Mill Superfund Sites in Colorado
Donald Simpson, Colorado Department of Health 528
Mobilizing for Remedial Construction Projects
Gary Stillman, Weston 550
Management of Change Order Conditions: A Superfund Case History
Myron Temchin, West HAZMAT, (303) 792-2535 Presented At Conference But Not Published
Construction Disputes on Hazardous Waste Projects
Theodore Trauner, TCS 558
Comparitive Roles of the EPA and the Bureau of Reclamation During the
Construction and Implementation of the Lidgerwood, North Dakota
Superfund Project
Laura Williams, USEPA 570
xxn
-------
TABLE OF CONTENTS
CONFERENCE PROCEEDINGS
Page No.
IV. GROUNDWATER REMEDIATION
Pitfalls of Hydrogeologic Characterization
Steven Acree, USEPA Published, But Not Presented At Conference 589
Areawide Implementation of Groundwater Institutional Controls for
Superfund Sites
David Byro, USEPA 601
Verifying Design Assumptions During Groundwater Remediations
Michael Grain, USAGE 606
Hydrologic Risk Aspects of Hazardous Waste Site Remediations
William Doan, USAGE 622
Design and Construction of the Groundwater Treatment Plant at the
Conservation Chemical Company Site
Peter Harrod, ABB Environment Services 642
The Construction and Operation of the New Lyme Landfill Superfund Site
Groundwater Treatment Facility
Donna Hrko, USAGE 659
Arsenic Removal at the Lidgerwood Water Treatment Plant
Harry Jong, Bureau of Reclamation 668
Successful Program Management for Remedial Design/Remedial Action
James Kilby, Monsanto 673
Advances in Hazardous Waste Alluvial Sampling
Lowell Leach, USEPA 681
A Comprehensive Groundwater Quality Assessment and Corrective Action
Plan for a Single Hydrologic Unit with Multiple Contamination Sources
C.M. Lewis, USDOE 701
A Perspective for NAPL Assessment and Remediation
Mark Mercer, USEPA 735
Optimizing and Executing a Multi-Faceted Remedial Action Plan
Dennis Peek, Geraghty & Miller 748
V. HEALTH AND SAFETY
EPA/Labor Health and Safety Task Force
Joseph Cocalis, USEPA 760
xxm
-------
TABLE OF CONTENTS
CONFERENCE PROCEEDINGS
Page No.
Airborne Exposure at an Acid Sludge Remedial Site
Stephen Davis, IT Corporation 766
An Overview of the NIEHS Superfund Worker Education and Training Grant Program
Denny Dobbin, NIEHS 785
Hazardous Waste Sites: Worker Protection Perspectives
John Moran, LHSFNA 814
Crisis in the Fire Service
Les Murphy, IAFF 827
Worker Protection Standard
Vicki SantOrO, USEPA (201) 321-6740 Presented At Conference But Not Published
USEPA Generic HASP
Vicki SantOrO, USEPA (201) 321-6740 Presented At Conference But Not Published
USEPA Health & Safety Certification
Vicki SantOrO, USEPA (201) 321-6740 Presented At Conference But Not Published
Overview of Hazard Waste Health and Safety Requirements
Rodney Turpin, USEPA (201) 321-6741 Presented At Conference But Not Published
VI. POLICY/MANAGEMENT ISSUES
Superfund — Program Standardization to Accelerate Remedial Design and
Remedial Action at NPL Sites
Shaheer Alvi, USEPA 838
Environmental Protection Agency Indemnification for Remedial Action
Contractors
Kenneth Ayers, USEPA 850
Innovative Design Review and Scheduling Tools: Potential Benefits to
HTW Remedial Projects
Gregg Bridgestock, USACERL 859
Basic Principles of Effective Quality Assurance
David E. Foxx, Foxx & Associates 885
Specifications for Hazardous and Toxic Waste Designs
Gregory Mellema, USAGE 889
Lessons Learned During Remedial Design and Remedial Action Activities
at Superfund Sites
Dev Sachdev, EBASCO 895
XXIV
-------
TABLE OF CONTENTS
CONFERENCE PROCEEDINGS
Page No.
Forecasting Staffing Requirements for Hazardous Waste Cleanup
Robert Salthouse, Logistics Management Institute 907
The Effects of the Davis-Bacon Act on the LaSalle Electrical Utilities
Phase I Remedial Action
David Seely, USEPA 919
Surety Bonds — Superfund Projects
August Spallo, USAGE 938
Remedial Design Schedule Management
Charles F. Wall, EBASCO 970
Remedial Management Strategy
Thomas Whalen, USEPA 1022
Acquisition Selection for Hazardous Waste Remediation
William Zobel, USEPA 1031
VII. PRE-DESIGN ISSUES
The Importance of Pre-Design Studies in Superfund Remediation
Jeffrey Bennett, Malcolm Pirnie, Inc 1042
RI/FS and ERA Impacts on RD/RA at Superfund Sites
William Bolen, USEPA 1060
Excavation/Off-Site Incineration RD/RA - Optimization of the
Planning/Investigation Process Based on the Two NPL Case Studies
John Gorgol, EBASCO 1087
Writing a Record of Decision to Expedite Remedial Action: Lessons from
the Delaware PVC Site
Stephen Johnson, DE DNR 1096
Site Characterization Data Needs for Effective RD and RA
John Moylan, USAGE 1103
New Bedford Harbor, Massachusetts Review of the Remedial
Investigation/Feasibility Study Process and Its Impact on Remedial
Design/Remedial Action
Mark Otis, USAGE 1110
The Pre-Design Technical Summary
Kenneth Skahn, USEPA 1118
xxv
-------
TABLE OF CONTENTS
CONFERENCE PROCEEDINGS
VIII. DESIGN ISSUES
Accelerating the ROD to Remedial Action Process: Sand Creek Industrial
Superfund Site (OU1), Commerce City, Colorado
Brian Pinkowski, USEPA 1125
Remedial Design of Superfund Projects -- What Can Be Done Better?
John Holm, USAGE 1141
Constructability Input to the HTRW Process
James Moore, USAGE 1148
Applications of a Design/Build Advisor Expert System to Environmental
Remediation Projects
Thomas Napier, USAGE 1162
"Conforming Storage Facilities" Remedial Construction Activities
D.M. Velazquez, DLA Published, But Not Presented At Conference 1173
IX. TREATMENT TECHOLOGIES
A New Horizontal Wellbore System For Soil and Groundwater Remediation
Ronald BittO, Eastman Christensen Published, But Not Presented At Conference 1186
Soil Bentonite Backfill Mix Design/Compatability Testing: A Case History
Jane Bolton, USAGE 1203
Remedial Design for Solvent Extraction of PCB Contaminated Soils at
Pinette's Salvage Yard
Steven J. Graham, EBASCO (617) 451-1201 Presented But Not Published At Conference
United Creosoting Company Superfund Site, A Case Study
Deborah Griswold, USEPA 1219
Considerations for Procurement of Innovative Technologies at Superfund
Sites
Edward Hanlon, USEPA 1232
Trial Burn at MOTCO Site, LaMarque, Texas
MaryAnn LaBarre, USEPA 1256
Construction of Groundwater Trenches
Gary Lang, USAGE 1268
European Soil Washing for U.S. Applications
Michael Mann, Geraghty & Miller 1285
XXVI
-------
TABLE OF CONTENTS
CONFERENCE PROCEEDINGS
Page No.
Remedial Design Procedures for RCRA/CERCLA Final Covers
Donald Moses, USAGE 1300
The Challenge of Treating Superfund Soils: Recent Experiences
Carolyn Offutt, USEPA 1330
Tower Chemical: Remedial Design for a Small But Complex NPL Site
Victor Owens, EBASCO 1346
The Importance of Test Fills for the Construction of HTW Caps and Liners
David Ray, USAGE 1360
Nuclear Waste Densification by Dynamic Compaction
Cliff Schexnayder, Nello L. Teer Co Published, But Not Presented At Conference 1382
USEPA Region II Treatability Trailer for Onsite Testing of Soils and
Sludges
William Smith, CDM Published, But Not Presented At Conference 1409
Bioremediation of Toxic Characteristic Sludges with Biological
Liquids/Solids Slurry Treatment
Donald Sherman, RTI Presented At Conference But Not Published
Summary of Issues Affecting Remedial/Removal Incineration Projects
Laurel Staley, USEPA 1442
Remediating TCE Contaminated Soils: A Case Study of a Focused RI/FS
and Vacuum Extraction Treatability Study
Winslow Westervelt, Gannett-Flemming Inc 1458
xxvn
-------
A Comprehensive Groundwater Water Quality Assessment and
Corrective Action Plan for a Single Hydrologic Unit with
Multiple Contamination Sources
(Author(s) and Addresse(s) at end of paper)
Introduction
This paper illustrates a case where numerous sources of contamination
and intermingling plumes exist within a single hydrologic regime. It
attempts to demonstrate that a fundamental knowledge of the
hydrogeology of the area, comprehensive contaminant tracking, and
definition of preferential pathways for contaminant migration can be
more useful to environmental restoration efforts than an intensive
effort at one facility at a time. It proposes a comprehensive method for
groundwater assessment and clean up as an alternative to a site by site
approach.
Background
The Savannah River Site (SRS) is a nuclear weapons complex operated
by Westinghouse Savannah River Company (WSRC) for the U.S.
Department of Energy (DOE). It occupies a three hundred square mile
area in South Carolina which bounds the Savannah River (Figure 1).
The General Separations Area (GSA) is a fifteen square mile area which
lies near the geographic center of the SRS. To the north and west the
GSA is bounded by Upper Three Runs Creek, to the south by Fourmile
Creek and to the east by McQueen Branch (Figure 2). The streams each
ultimately flow into the Savannah River. These streams are the
dominant influence on groundwater flow in the uppermost aquifer
below the GSA. The area bounded by these streams exists as part of a
single hydrogeologic system.
The chemical separations facilities and many waste management
facilities serving the SRS are located in the GSA. More than thirty
separate sites in the GSA have been identified for environmental site
investigation under either RCRA or CERCLA (Figure 3). These include
unlined basins which received waste, shallow land burial sites, coal pile
runoff basins, collapsed underground process sewer lines, and leak and
spill sites.
A map of tritium concentrations in the water table indicates that some
intermingling of plumes has occurred (Figure 4). All of the sites in the
GSA exist in the unsaturated zone above the single hydrologic system of
the GSA. Once contamination from any site migrates through the
unsaturated zone and enters the groundwater it becomes part of that
larger system. Most of the sites are characterized by a potential for
701
-------
both hazardous and radioactive components of soil and groundwater
contamination.
Discussion
An extensive hydrogeologic characterization has been completed for the
F and H Area Seepage Basins (FHSB), which are waste sites in the GSA
(Figure 2). An implementable groundwater remediation plan has also
been prepared for the FHSB in accordance with RCRA and South Carolina
Hazardous Waste Management regulations. Hydrogeologic
investigations and preparation of corrective action plans at the other
facilities in the GSA are in various later stages of preparation.
Schedules for environmental restoration work have been driven largely
by regulatory deadlines.
There are two primary findings of the hydrogeologic assessment and
modeling studies of corrective action options. First, the plumes at FHSB
should not be treated as isolated zones of groundwater. The FHSB
plumes exist as part of the larger hydrogeologic regime of the GSA. The
migration patterns of contaminants have been linked to features and
characteristics of that regional hydrogeologic system. Any groundwater
corrective action plan at the FHSB should take into account effects at
adjacent facilities and effects on nearby streams and wetlands. Second,
the most important corrective action is source control. Preliminary
estimates indicate that groundwater remediation schemes will provide
only minimal additional benefit to groundwater and stream water
quality as compared to the effects of discontinuing discharge of waste to
the basins and to basin closure.
These findings imply that a comprehensive approach to groundwater
assessment and remediation may be the best way to approach
environmental restoration in the GSA. A plan for comprehensive
groundwater quality assessment and corrective action for the General
Separations Areas of SRS has been developed and proposed as an
alternative to groundwater remediation at individual facilities. The
plan will allow for interim actions in areas prioritized according to their
potential risk to human health and the environment. The plan proposes
to treat the entire area as a whole and is based on technical, logistical
and cost/benefit considerations. The main obstacle to implementation
of this environmental restoration program may be regulatory rigidity.
702
-------
F and H Area Seepage Basins
The FHSB received radioactive wastewater, primarily evaporator
overheads, from the F and H Separations Areas from 1955 to 1988. The
basins were designed for slow seepage and migration through the
sediments and shallow groundwater to Fourmile Branch to allow for
decay of the radionuclides present in the feed streams. During
operation they received a combined average of 80 million gallons
annually. The components of the waste stream included tritium,
cadmium, chromium, barium, silver, phosphate, lead, mercury, nitrate,
sodium, Sr-90, Cs-134 and Cs-137.
Past operation of the unlined earthen FHSB for disposal of waste water
has resulted in plumes of groundwater contamination. The plumes
extend from the basins to the wetlands at Fourmile Branch (Figure 4).
The primary contaminants are tritium and nitrate. Concentrations of
mercury, lead, cadmium, radium, and gross alpha above the primary
drinking water standard are present. The pH of water in the plumes
(pH=3.0-4.5) is lower than expected for natural groundwater in the
area. Wetlands areas downgradient of the basins and Fourmile Creek
have also been impacted by discharging plume water.
The main body of contaminated groundwater flows from under the
basins toward Fourmile Branch. Plume water discharges to the creek
and wetlands flanking the creek. Areas of dead and stressed vegetation
are present in the wetlands. Agents in the wetlands soil which are at
levels potentially toxic to trees are pH, nitrate, aluminum, manganese,
zinc, cadmium, and sodium. Aluminum and manganese were not
present in the waste stream. They are thought to have been leached
from subsurface minerals by the low pH plume water. Drought
conditions during 1977 and subsequent years are thought to have
exacerbated the damaging effects of the contaminated water by
concentrating salts and by failing to provide rainwater to dilute the
plume water (Greenwood et al, 1990).
Preliminary studies indicate that flushing with clean water reduces
leachate to non-toxic levels (Loehle, 1990). It is anticipated that natural
rainfall combined with closing the basins should lead to wetlands soil
and ecological recovery. Field and laboratory studies involving planting
natural wetlands vegetation in stressed soil are planned to test this
assertion. Field investigations of the areas suffering vegetation
mortality reveal that reforestation is already underway in H-Area.
703
-------
Secondary succession is occurring in the understory as shrubs and
saplings are beginning to re-colonize the areas.
Fourmile Branch has been impacted by groundwater seepage from the
plumes of contamination. Stream water samples taken downstream of
the basins exhibit higher levels of tritium, nitrate and sodium than
samples taken from upstream of the basins. Tritium concentrations in
the creek exceed the primary drinking water standard. Concentrations
of mobile contaminants in groundwater discharge are diluted by stream
flow. Downstream concentrations of nitrate and sodium are elevated
relative to upstream samples, but do not exceed primary drinking water
standards. No hazardous constituents have been detected in the creek
water (Looney et al, 1988).
Environmental Remediation Activities
Two source control measures have been taken at the basins: 1)
discontinuing their use and, 2) emplacement of low permeability caps.
Use of the basins for waste disposal was discontinued on November 7,
1988. The waste stream which used to be discharged to the basins is;
treated and the effluent is released to Upper Three Runs under an
NPDES permit. Low permeability caps have been emplaced, according to
an approved RCRA closure plan, over the basins to minimize infiltration
of rainwater through the contaminated sludge and soil beneath the
basins.
As required by RCRA, a groundwater remediation plan was developed
to address the plumes at FHSB. Preparation of the remediation plan
included an extensive hydrogeologic characterization, a review of
potential remediation options, and groundwater modeling to assess the
effectiveness of the proposed remediation. The results of these efforts
are summarized in the following sections.
Geology of the GSA
The uppermost aquifer underlying the GSA is comprised of
unconsolidated coastal plain sediments which dip regionally seaward.
The sediments are primarily unconsolidated sands and clays. Generally,
the sandy units function as aquifers and the clays as aquitards. Thin
discontinuous cemented zones are occasionally encountered in core.
Carbonate zones ranging from calcareous muds and sands to silicified
shell hash have been observed in core from the GSA.
704
-------
A detailed study of the lithology and hydrology of the Tertiary
sediments of the south central portion of the GSA has been conducted as
part of the preparation of a groundwater corrective action plan for the
FHSB. Geologic correlations of aquifer and aquitard units were made for
wells in the area based on core/cutting descriptions and geophysical
logs. This information was used to construct lithologic cross sections,
structure contour maps, facies maps and isopach maps of the aquifer
units and confining beds. Hydraulic head data from the monitoring
wells have been compiled and used to construct potentlometric maps of
each of the units and to study the vertical head relationships within and
between the aquifers. These are combined with groundwater
monitoring data to produce a hydrogeologic interpretation which
identifies preferential migration pathways.
An example lithologic cross section depicts the aquifer units, and the
location of screen zones of the monitoring wells in each unit (Figure 5).
The uppermost aquifer is a regulatory term; the uppermost aquifer
includes all aquifer units which are hydraulically connected to the
water table beneath a site. There are three aquifer units in the
uppermost aquifer at the GSA. The aquifer units, their properties and a
review of the formation names and hydrostratigraphic nomenclature at
SRS have been discussed in detail in publications by SRS workers
(Harris et al, 1990; and Aadland, 1990). The three units are commonly
known, from shallower to deeper respectively, as the water table, the
Barnwell/McBean, and the Congaree.
The aquifer units are separated by two leaky aquitards. The two
aquitards are known locally as the Tan Clay and the Green Clay. The
Tan Clay supports the water table and overlies the Barnwell/McBean
unit. The Green Clay separates the the Barnwell/McBean from the
underlying Congaree. Vertical migration through the clays is variable,
depending upon the local thickness and competency of the confining
units. Local discontinuities in the clays are observed to provide
preferential pathways for vertical migration.
The Congaree unit is underlain by the Ellenton Formation which is the
principle confining unit for the uppermost aquifer beneath the GSA.
The Ellenton is a regionally competent aquitard which hydraulically
separates the Tertiary sediments of the uppermost aquifer from the
Cretaceous sediments below.
705
-------
Groundwater Flow in the GSA
Horizontal groundwater flow in the units above the Green clay (the
water table and McBean) is dominated by Upper Three Runs and
Fourmile Creek. The water table map indicates the presence of a
groundwater divide near the geographic center of the GSA (Figure 6 and
Figure 7). North of the divide lateral flow is north toward Upper Three
Runs Creek. South of the divide, flow is generally southward into
Fourmile Creek.
Below the Green Clay, flow in the Congaree is towards Upper Three Runs
across the entire GSA (Figure 8). Fourmile creek is not deep enough to
incise the Green Clay, and therefore exerts no influence on flow in the
Congaree.
Recharge of the uppermost aquifer is from rainfall infiltration through
the unsaturated zones and the aquitard units. In the GSA, the water
table and Barn well units discharge into Upper Three Runs and Fourrnile
Creek. The Congaree discharges into Upper Three Runs.
Preferential Flow Pathways
Preferential contaminant flow pathways have been identified at the F
and H area seepage basins. These preferential pathways are often
associated with mappable geologic features in the sediments below the
basins. Correlations of geophysical logs and core descriptions at
monitoring well clusters indicate offset of beds. These displacements
are mappable and can be illustrated in cross section (Figure 5). These
offsets are interpreted as being the slip surfaces of slumps. A
conceptual diagram depicts a slump feature and the mechanism for
offset of beds in unconsolidated sediment (Figure 9).
These offsets are observed to displace confining units and provide
vertical preferential pathways. The slump feature illustrated
hydrostratigraphic cross section lies directly below the F-Area seepage
basins (Figure 10). Figure 11 represents the same cross section shown
in Figure 5 and depicts contours of concentrations of lead in the
groundwater. The tan clay confining unit is offset providing a
downward flow path for contaminants beneath the basins. This figure
illustrates that the path of contaminant migration is primarily
downward in the location of the slump feature.
706
-------
An example of a slump feature providing a horizontal preferential flow
pathway is shown in Figure 12. This figure shows the location of the
slump features where they offset the Tan Clay, and the concentrations
of tritium in the water table. The offsets coincide with the location of
horizontal preferential contaminant pathways. The offset planes
apparently provide high permeability zones which allow accelerated
flow compared to the adjacent sediments.
It is likely that the individual slump features are associated with
regional trends across the GSA. Geologic features tend to occur as part
of regional patterns. Regional zones of carbonate have been mapped in
areas in the GSA where the most subsurface data exists (Figure 13).
The slump features may be associated with the occurrence of carbonate
zones (one of several potential mechanisms for slumping is the
dissolution of carbonate material and subsequent collapse of overlying
sediments). This possibility is under investigation. A petrographic
study of the carbonates in thin section is ongoing, and more core and
geophysical data from new wells are being used to further map the
occurrence of offsets due to slumping.
Not all of the preferential pathways observed at the FHSB have been
directly linked to slump features. Other preferential pathways may be
related to slump features which have not yet been mapped, or they
may be related to textural heterogeneities in the subsurface such as
coarse grained sand lenses or h?gh permeability zones in the carbonates.
More data is being acquired to investigate the relationship between
geologic features and preferential flow. Work done to design a
corrective action plan supports the notion that the key to designing an
effective groundwater remediation system is to identify and understand
the preferential flow pathways.
Corrective Action Plan
The choice of a groundwater remediation plan to treat the hazardous
constituents was complicated by the fact that a primary constituent of
the plume water is tritium. Tritium is a radioactive isotope of hydrogen
(H-3). There is no implementable treatment for tritium removal from
water. The half life of tritium is relatively short, approximately twelve
years. One reason that the seepage basins were originally used for
disposal of this waste stream was to allow for the decay of tritium as
the water migrated slowly through the ground towards the creek. This
allowed for a smaller amount of tritium to be released to the surface
waters which eventually flow offsite than if the waste water was
707
-------
released directly to a stream. In view of the tritium component in the
plume water, there was reluctance by project environmental
professionals to discharge water extracted from the ground to a surface
stream after treatment to remove chemical constituents. One goal of the
remediation was to minimize migration of tritium into Fourmile Branch.
The chosen corrective action plan was to extract water before it could
flow into Fourmile Branch and the wetlands, neutralize the pH, treat it
for chemical constituents, and then inject the treated water back ini:o
the ground upgradient of the basins. This cycling of groundwater would
restrict tritium from migrating into Fourmile Branch, and allow more
time than the natural system for tritium decay. This system seemed
the most acceptable solution to the problem of controlling the spread of
the plume of hazardous constituents and radionuclides with regard to
existing technology and regulatory constraints.
FHSB Plumes are Part of a Larger Hydrologic System
Groundwater modeling studies of the pump/treat/inject system
provided unexpected results. Particle tracking analyses were employed
to attempt to optimize the design of extraction/injection systems in the
water table and Barnwell/McBean units to maintain hydraulic control of
the plumes. Results indicate that 100% capture of the targeted plurae
water in the water table and Barnwell/McBean aquifers is possible, but
not if extracted water is injected, back inf.o the aquifers as planned.
Preliminary results indicate that if the targeted plume area is
attempted to be controlled and 100% of the extracted water is injected,
less than 20% of the plume water can be stopped from entering
Fourmile Branch. Correspondingly, there will be an increase in the
percentage of plume water that moves down into the Congaree
(Geotrans, 1990a; Geotrans, 1990b).
The explanation for the low efficiency of the extraction/injection
network is that it was conceptualized as a closed system, but in
actuality, there is no mechanism to stop rainwater input into the
system. The primary recharge to the shallow aquifers is infiltration of
rainwater. By extracting water before it discharges and injecting it
upgradient, the mechanism for water to leave the system is removed,
but there is no mechanism to stop rainwater from continuing to enter
the system. As long as 100% of the water which is extracted is injected,
there will be a continual increase in the volume of water in the system.
Since this excess water must somehow leave the system, it escapes by
708
-------
flowing around and through the extraction network or by migrating
downward.
A system is now being designed to optimize control of the portions of
the plume that contain the highest concentrations of contaminants. This
type of system will be designed to target the preferential flow
pathways. A more narrowly focused remediation, which seeks to
optimize control of mass of contaminants, rather than volume of water,
is expected to be more efficient.
The necessity of understanding the whole of the regional hydrogeologic
setting of the GSA is underscored by the results of both the modeling
and the FHSB hydrogeologic characterization. Information on the nature
of the relationship between the plumes of contamination and the
geology, and how those plumes fit into the larger hydrogeologic system
will be needed to properly design a corrective action program at the
FHSB.
Source Control the Most Effective Corrective Action
A two dimensional flow and transport model was run to assess the
effectiveness of a pump/treat/inject system which successfully
prevented 66% of the targeted plume water from entering Fourmile
Creek. Modeling of contaminant levels at a hypothetical monitoring well
downgradient of the extraction network indicate that the
extraction/treatment/injection system would have a negligible effect on
the concentrations of nitrate and tritium (Figure 14). Results of the
modeling indicate that compared to the closure of the basins, the
additional benefit of a post-closure groundwater remediation program
will be minimal.
Based on the transport modeling results, it is clear that the most
significant corrective action has already been accomplished at the FHSB.
The source of contamination has been controlled. Discharge of waste
water to the basins was discontinued in November 1988. The basins
have been physically and chemically stabilized, backfilled, and RCRA
closure caps are being emplaced. Prior to closure of the basins, tritium
concentrations in downgradient monitoring wells increased or remained
at equilibrium. Since closure, concentrations of tritium in monitoring
wells at the basins are declining (Figure 15). The levels of tritium,
nitrate, and other contaminants which discharge to the creek and
wetlands are anticipated to decline similarly, in response to the
termination of discharge and closure of the basins.
709
-------
Results of the transport modeling indicate that source control (stopping
discharge of waste to the basins and covering them with a low
permeability cap) is the most important corrective action. On the whole,
the studies at FHSB suggest that groundwater corrective action
programs will be vastly expensive ($25-30 million) and marginally
beneficial. This implies that environmental dollars and efforts in the
GSA may be better spent on identifying and eliminating or capping
other sources of contamination in the GSA before attempting an
extraction and treatment program at any specific facility.
Another argument for comprehensive assessment and corrective action
for the entire GSA is that although various corrective action scenarios
may be workable, the impact of them on the hydrologic system of the
area as a whole has not been fully assessed. Changes to flow patterns at
the F and H basins resulting from corrective action could complicate the
ongoing groundwater quality assessment and plans for corrective action
at adjacent sites.
Comprehensive Approach to Groundwater Clean-Up
A comprehensive assessment of the groundwater contamination in the
entire GSA would lead to a more efficient and effective approach to
groundwater remediation. There are . at least 30 separate potential
sources of groundwater contamination in the GSA. These are all in
different stages of characterization.
A comprehensive corrective action program for the entire GSA is being
developed. One general failing of the plans to remediate the plumes at
FHSB, is that the effects of implementing an extraction/injection system
on the hydrology of nearby facilities and plumes has not been
adequately defined. A comprehensive hydrogeologic characterization of
the entire GSA will allow for the most technically sound approach to
environmental restoration. It would also provide the most cost
effective approach to corrective action facility design. Designing a
number of plume specific facilities will likely prove to be the most
expensive and inefficient approach in the long term.
One or several large integrated groundwater treatment facilities could
be designed to address groundwater contamination problems in the
entire GSA. In a comprehensive corrective action plan, extraction and
injection well fields could be placed for the best advantage of
groundwater clean-up in general. It seems likely that as
710
-------
characterizations proceed in the GSA that modifications will need to be
made to the original design at FHSB to adjust for other corrective
actions.
Comprehensive Assessment and Corrective Action Plan
The comprehensive assessment of the GSA will be the basis for a
conceptual plan for groundwater remediation of the GSA as a whole.
The comprehensive assessment will include an interpretation of existing
data, a plan for acquisition of additional data, a GSA wide monitoring
program, and conceptual design of any appropriate corrective actions.
Groundwater modeling will be used to simulate the hydrogeologic
system and gauge the effectiveness of proposed corrective action
scenarios. Risk assessment will be used to justify a decision not to
remediate contamination or to quantify the benefits of a proposed
corrective action.
The following sections briefly describe the elements of the
comprehensive assessment plan.
Comprehensive Assessment Strategy Document
This document will outline the strategy for the comprehensive
groundwater assessment and corrective action for the GSA. A project of
this magnitude will require a carefully thought out plan and a great
deal of coordination. A detailed schedule and discussion of each of the
following sub-tasks and how they are interconnected will be included:
* initial hydrogeologic assessment report
* proposed comprehensive monitoring network
* proposed physical tests required to adequately characterize
the multi-aquifer system
* program for unsaturated zone characterization
* stream and wetlands characterization
* preparation of a comprehensive interpretation of the
hydrogeology and groundwater quality in the GSA
* modeling of flow and solute transport in both saturated and
unsaturated zones
* risk assessment
* feasibility studies and innovative technology assessment
* proposed corrective actions
Once completed this document will serve as a guide for managing the
comprehensive assessment of the GSA.
711
-------
GSA Hvdrogeologic Assessment
An initial GSA hydrogeologic assessment report will be completed. This
will be accomplished by assembling all existing monitoring well data,
geologic data, modeling studies and stream and wetlands data. There
are data from approximately 500 monitoring wells in the GSA. This
report will attempt to identify all contaminants which may require
corrective action. It will also identify areas where no data exists or not
enough data exists to make an interpretation.
The report will serve as a guide to planning monitoring well networks
and as a data base for ground water modeling. It will provide plume
maps of pertinent pollutants. Hazardous constituents and radionuclides
will be considered. Lithologic cross sections and figures showing the
extent of contamination in cross section will be provided. An attempt
will be made to identify the likely sources of contamination. The report
will include plots of time trend data. Interpretations will include
whether specific plumes are likely the result of a continuing source or
represent a migrating slug such as may be associated with an old spill.
Vertical and horizontal head relationships will be discussed. Estimates
of hydraulic conductivity, flow rates and other aquifer properties which
can be used in the modeling will be presented and discussed.
The report will include a section which discusses the types of data
which should be acquired in order to better characterize the GSA. A
discussion of the reliability of the existing data will also be included.
Phase 2 Hydrogeologic Assessment
The initial hydrogeologic assessment will be updated and revised to
include geologic and monitoring data acquired during the course of the
comprehensive assessment. The new information will be compiled and
any changes in interpretation will be documented in a Phase 2
assessment report. The report will include a section which discusses the
status of plume delineation and the actions necessary to adequately
characterize the system. This includes identification of areas requiring
further plume assessment wells and monitoring.
712
-------
Monitoring and Data Collection Plan
A plan for monitoring and data gathering will be prepared based on the
initial hydrogeologic assessment. In conjunction with the
comprehensive monitoring well network, aquifer tests and other
physical tests which will aid in the interpretation of data and modeling
will be proposed. Program plans for various field projects including
well installation, coring, slug tests, and aquifer tests will be included.
The locations and depths of proposed wells and a cost estimate will be
included.
A comprehensive sampling and analysis plan will be developed. Each of
the hydrostratigraphic units comprising the uppermost aquifer will be
monitored across the area. The monitoring plan will focus on the
quality of groundwater in the general area. The global plan will
incorporate the regulatory sampling and analysis requirements at
specific facilities, and also track contaminant migration beyond adjacent
facilities to identify intermingling plumes.
A second phase of well installation and field tests may be required. The
In this case, a Phase 2 monitoring and data collection plan will be
produced based on the recommendations of the Phase 2 Hydrogeologic
Assessment.
Unsaturated Zone Characterization
In conjunction with the installation of a comprehensive monitoring well
network, a field study of the unsaturated zone is planned. This will be a
program of field permeability measurements and to collect and analyze
samples to characterize the unsaturated zone in the GSA. The study will
be designed to estimate physical properties of the unsaturated zone.
These values will be needed to model the movement of contaminants
from a source at land surface (or trench or vault bottom) to the water
table for risk assessment. The results of the characterization and the
sampling and analysis techniques utilized will be fully documented.
713
-------
Stream and Wetlands Characterization
Documentation of the impacts of facilities in the GSA on Fourmile
Branch, Upper Three Runs Creek and the wetlands surrounding them
will be prepared. This program will include wetlands delineation,
sampling and analysis of water taken from the wetlands and streams,
sampling and analysis of sediments from the wetlands, and studies of
the biological community. Results and interpretation of the data
collected will be included in a report. All procedures will be fully
documented. Sources of variability in the data will be discussed.
Groundwater Flow and Transport Modeling
Groundwater flow and transport models will be used as input to risk
assessment and to simulate various corrective action scenarios. These
simulations will be used to help select the most appropriate corrective
actions for the GSA. The modeling report will include a discussion of the
match between monitoring data and model simulations of plume shapes.
Estimates of predicted contaminant concentrations through time will be
performed. Documentation of the model, parameters, and assumptions
used will be included.
Risk Assessment
The site characterization based on information discussed in the
hydrogeologic assessment reports and the groundwater modeling will
serve as the basis for a risk assessment. The risk assessment will
include an identification of populations and a hazard evaluation based
on a review of the inherent toxic properties of the primary constituents
of interest. Exposure pathways will be identified and documented.
Modeling will be employed to calculate doses and quantify risk at the
points of exposure. Uncertainties and variabilities in the risk
assessment will be fully documented and discussed.
Feasibility Studies and Innovative Technologies
A literature search of potential groundwater treatment technologies and
innovative techniques for remediation will be conducted. A discussion
of the favorable and unfavorable characteristics of each technology will
be presented. This will be the basis of a program of laboratory and
field studies to test the most promising technologies. A complete
description of each of the test procedures, results and interpretations
will be prepared.
714
-------
Recommendations for Corrective Actions
A document identifying recommendations for future actions based on
the findings of the Comprehensive Groundwater Assessment of the GSA
will be prepared. This document will identify plumes of contamination
which may require corrective action under RCRA or CERCLA.
Contaminants of concern will be identified and estimates of the volume
of contaminated water to be remediated will be included.
If deemed appropriate, a conceptual plan for remediation of the GSA as
a whole will be presented. Various types of corrective action will be
considered. These include:
* extraction wells or trenches, treatment and injection
* extraction wells or trenches, treatment and release to
streams
* containment, extraction, treatment
* in situ treatments
* immobilization technologies
* some combination of techniques
Executive Summary
The executive summary will briefly describe the major elements of the
entire Comprehensive Assessment and Corrective Action Plan for the
GSA. The summary will include results of the assessment and modeling.
It will reiterate the recommendations for corrective action. The
document will address the reliability of the data. The report will
caution about potential circumstances or new data which could change
interpretations and recommendations presented. It will also include an
index to the contents of the other volumes in the Comprehensive
Assessment and Corrective Action of the GSA series.
715
-------
Conclusions
A comprehensive assessment plan has been proposed to provide a
framework for investigating and remediating an area with multiple
sources of groundwater contamination in the most logical, scientific and
cost effective manner. This plan has been devised in response to the
results of a series of studies conducted in preparation of a groundwater
remediation program for several facilities in the General Separations
Areas (GSA), of SRS.
On the whole, the studies imply that environmental dollars and efforts
in the GSA will be better spent on identifying and controlling other
sources of contamination in the GSA before attempting groundwater
remediation at any specific facility. The work done to date suggests
that groundwater corrective action programs will be vastly expensive
($25-30 million) and marginally beneficial, as opposed to source control
actions. Ongoing work also supports the notion that the key to
designing an effective groundwater remediation system is to
understand the hydrogeology of the area and its preferential flow
pathways.
A comprehensive assessment and corrective action plan for the entire
GSA will allow for the assessment of the impact of proposed
groundwater remediation activities on the hydrologic system of the GSA
as a whole. Changes to flow patterns at any one facility resulting from
corrective action which could complicate the ongoing groundwater
quality assessment and plans for corrective action at adjacent sites can
be considered.
716
-------
Disclaimer
This paper was prepared by Westinghouse Savannah River Company
(WSRC) for the United States Department of Energy under Contract No.
DE-ACOP-88SR18035 and is an account of work performed under that
contract. Neither the United States Department of Energy, nor WSRC,
nor any of their employees makes any warranty, expressed or implied,
or assumes any legal liability or responsibility for the accuracy,
completeness, or usefulness, of any information, apparatus, or product
or process disclosed herein or represents that its use will not infringe
privately owned rights. Reference herein to any specific commercial
product, process, or service by trademark, name, manufacturer or
otherwise does not necessarily constitute or imply endorsement,
recommendation, or favoring of same by WSRC or by the United States
Government or any agency thereof. The views and opinions of the
authors expressed herein do not necessarily state or reflect those of the
United States Government or any agency thereof.
Author(s) and Address (es)
Catherine M. Lewis
Westinghouse Savannah River Company
Savannah River Site
Building 704S-57N
Aiken, SC 29808
(803) 557-2848
Martha A. Ebra
Westinghouse Savannah River Company
Savannah River Site
Building 320-4M
Aiken, SC 29808
(803) 725-1795
O. Beth Wheat
Westinghouse Savannah River Company
Savannah River Site
Building 703-H
Aiken, SC 29808
(803) 557-8701
717
-------
References
Aadland, R.K., 1990. Classification of Hydrostratigraphic Units at
Savannah River Site, South Carolina. Savannah River Laboratory,
WSRC-RP-90-987, Westinghouse Savannah River Company, Aiken,
SC
Aadland, R.K., M.K. Harris, C.M. Lewis, T.F. Gaughan, and T.M. Westbrook,
1990. Hydrostratigraphy of General Separations Area (SRS), South
Carolina. Poster Session at Geological Society of America National
Meeting 1990, Dallas., TX
Geotrans Inc, Environmental Consultants, 1990a. Preliminary
Evaluation of Remedial Alternatives for the F-Area Seepage
Basins, Prepared for Westinghouse Savannah River Company,
Contract AX853019, November 1990.
Geotrans Inc, Environmental Consultants, 1990b. Preliminary
Evaluation of Remedial Alternatives for the H-Area Seepage
Basins, Prepared for Westinghouse Savannah River Company,
Contract AX853019, November 1990.
Greenwood, et al, 1990. Assessment of Tree Toxicity Near F and H Area
Seepage Basins, WSRC-RP-90-455.
Harris, M.K. , R.A. Aadland, and T.M. Westbrook, 1990. Lithological and
Hydrological Characteristics of the Tertiary Hydrostratigraphic
Systems of the General Separations Area, Savannah River Site,
South Carolina, Bald Head Conference on Coastal Plains Geology at
Hilton Head, SC, November 6-11, 1990.
Loehle, et al, 1990. Recovery of Contaminated Wetland Soils at SRS by
Natural Rainfall: An Experimental Toxicological Study, WSRC-RD-
90-14.
Looney, et al, 1988. Sampling and Analysis of the F and H Area Seepage
Basins, DPST-88-229.
Westinghouse Savannah River Company, 1990, RCRA Post Closure
Permit Application for the F Area Hazardous Waste Management
Facility, Savannah River Site. Prepared for the U.S. Department of
Energy. Submitted to South Carolina Department of Health and
Environmental Control (SCDHEC) December 3, 1990.
718
-------
References (continued)
Westinghouse Savannah River Company, 1990, RCRA Post Closure
Permit Application for the H Area Hazardous Waste Management
Facility, Savannah River Site. Prepared for the U.S. Department of
Energy. Submitted to South Carolina Department, of Health and
Environmental Control (SCDHEC) December 3, 1990
719
-------
SOUTH
CAROLINA
LOCATION OF THE GENERAL SEPARATIONS AREA,
SAVANNAH RIVER SITE
Figure 1. The Savannah River Site (SRS) is a DOE nuclear weapons
facility. It is located in South Carolina, encompasses approximately 300
square miles and borders the Savannah River. The General Separations
Area is located near the center of SRS.
720
-------
CM
IV
«! J* }>
IJI.1--
=tj * I § =
~c-E-S"
u -o -c
" . u « =
3.^2 "
£^^ * «
-1
-I
-I
-I
-I
-I
-I
-S
I
-I
-I
k
I I I ii i iT"
I ! I I I
~i i i r
721
-------
722
-------
_u
JD
u
- .s
Crt •«•
C "»
o
2 o
E ™
a
•S eo
' •-
a-i
OO T3
=
723
-------
111. Ill ill ill ,|J
3123 El E3 • Oil ffl il ffl BIS H "^ ;;*il
JM-i
CO
I
CO
CO
CD
DC
C_J
cn:
CO
DC
CO
CD
CD
DC
-sC
I
724
-------
U. -o
-------
CO
Q_
\
\
.5
\
\
\
a
726
-------
!
fli
DC
O
UJ
UJ
DC
<
CD
2
O
LU
CJ
<
u_
DC
ID
CO
CJ
I — I
DC
h-
LU
:E
o
LU
(—
O
Q_
in
-------
CA
u
a.
E
3
o
«9
00
2
'•3 T)
V
U V
e £>
o
" »-,
at
E- 2
":!
ov v2i
§ I"
00 -i
£ -;
728
-------
CD
CD
CQ
00
Lu
/
1
r--
oo
CD
oo
u_
/
/
I
1^
CD
CD
C/)
LL
/
/
1
-------
S|
f*'»*g»g«9»*»*»g»g*ggSt»»«»»
C_J
CO
I
CO
CO
CD
DC
C_J>
CO
^:
CD
CD
CO
CD
730
-------
00
LU
CL
o:
I
UJ I—
o cc
CC O
< tr
_i o
Q. 2
731
-------
i
1
m 3
o u
S 8
nun
CO u op
aj — • to
s J5 Sf
e
e o.—
'" 15 oo
t/5 Wl C
U C •-
c o «
o -s -d
N a
732
I
^X« \
5s \ \
0 >>.£
«« « -
o E ^
eg
s £>•*-
"00.
5£ E
l
..
[i. 3 O
-------
PREDICTED CONTAMINANT CONCENTRATIONS
14000
— CONTINUED USE
oao SOURCE CONTROL
10 20 30 40 50 60 70 80 30 100
TIME (yeor-5)
cn
e
tr
LaJ
O
z
o
o
180
160
140
120
100
80
60
40
20
0 -
— CONTINUED USE
ooo SOURCE CONTROL
ooo PUMP, TREAT,
INJECT
0 10 20 30 40 50 60 70 80 90 100
TIME (years)
Figure 14. The predicted contaminant concentrations at a hypothetical
monitoring well downgradient of the F area seepage basins for three
scenarios are compared. The benefit of discontinuing use of the basins
and capping is dramatic, but the additional benefit of a pump, treat and
inject program is negligible. •
733
-------
CD
OQ
CO
E-
CO
D
J
O
a
I 5
1 B
\ \
I I
" 3
it
•*J -0
• PH e
u
a.
CO
n Or
o» ^"^
s «
~ s
s P
IS
PQ
CO
W
E-
CO
o
CO
a CC
S w
IS
S i=
Or
2i
eg
*
T3
I «
oo "
W> CA
•c s
O «J
1 a
E oo
^£ '5 »i
^11
!si
l-io
1 8 S
H< 3
.^.i3
vn -a 3
— c =
ee 'S
|U. §
M U b-
E -S =5
o
Is
734
-------
A Perspective for NAPL Assessment and Remediation
Mark Mercer, PE
Hazardous Site Control Division
Office of Emergency and Remedial Response
U. S. Environmental Protection Agency
Washington, DC
ABSTRACT
This paper compares contaminant mass release rates to the subsurface at facilities receiving
contaminated water and facilities receiving organic fluids (Class One and Class Two sites) with
contaminant mass removal rates possible with ground-water (GW) pump and treat. Mass removal
rates possible with nonaqueous phase liquid (NAPL) plume investigation and remediation are
presented. Containment and cleanup approaches for Class Two sites (NAPL) are discussed. Some
reasons for current GW pump and treat failures are also discussed.
OUTLINE
1.0 Introduction to Issue
2.0 Background
3.0 Different Images of Subsurface Migration Pathways
3.1 Need to Differentiate Waste Disposal Sites
3.1.1 The Nature of Wastes at Hazardous Waste Sites
3.1.2 Class One Sites
3.1.3 Class Two Sites
3.1.4 Historical Techniques for Identification of NAPL Presence
3.2 Comparison of Mass Release Rate and Mass Removal Rate
3.2.1 Mass in Per Year
3.2.1.1 Class One Sites (APL Lagoon)
3.2.1.2 Class Two Sites (NAPL Lagoon)
3.2.2 Mass out Per Year
3.2.2.1 Ground Water Pump and Treat
3.2.2.2 NAPL Pump and Burn with Surfactant Wash
3.3 Relative APL and NAPL Contaminant Mass at a Class Two Site
3.4 Need to Sample All Subsurface Pathways
4.0 Response to Problem of Mass Removal Rate
4.1 Selecting Appropriate Cleanup Approach for Each Class
4.2 Alternatives for Mass Removal Rate Problem
4.3 Containment Versus Removal
4.4 Time Required for Demobilization of NAPL Plume
5.0 Summary
6.0 References
INTRODUCTION
The issue has been raised regarding the effectiveness of ground-water (GW) pumping and treating
for the remediation of abandoned hazardous waste disposal sites. This paper focuses on typical
abandoned commercial hazardous waste disposal sites. Many of the simplifying assumptions possible
for these Superfund sites are inappropriate for small spill sites.
735
-------
The purpose of this paper is to present alternative approaches for characterizing and remediating
subsurface contamination by nonaqueous phase liquids (NAPLs). The opinions are my own and do
not represent the policies of the Agency. For purposes of discussion, it is convenient to group
Superfund sites into two categories based on site characteristics. Hypothetical scenarios are presented
to illustrate key points and do not represent actual field data from individual Superfund sites. I am
soliciting opinions from the technical community on application of these alternative approaches for
evaluating remediation of subsurface contamination. This subject is currently under deliberation at
EPA and further evaluation of these ideas and of additional field work may result in modifications
to Agency policy.
This paper suggests that the problem with GW pump and treat is one of using it on sires with
nonaqueous phase liquid (NAPL) plumes. This problem is limited to NAPL sites as GW pump and
treat works well for sites without NAPL plumes. A concept of a Superfund site with only a dilute
solute plume migrating away is often incorrect; frequently a mobile NAPL plume will be present.
If a NAPL plume is present, and not considered, there will be an underestimation of the mass of
contaminants released and misunderstanding of the subsurface pathways through which the bulk of
the contaminant mass migrates from the site. This paper suggests that sites could be differentiated
as to whether they have APL or NAPL plumes.
The first group of topics discusses an identification of the problem. The mass release rates at a
hypothetical pair of hazardous waste disposal sites are compared. The practice of looking for a GW
concentration of 1, 20, or 33 percent of the equilibrium solubility concentration as an indicator of the
presence of a NAPL plume is reviewed. The mass removal rate possible with GW pumping is
compared to a treatment train of NAPL pumping, secondary recovery, and GW pumping for currently
contaminated waters only. Field data comparing relative proportion of contaminant mass in an
aqueous phase liquid (APL) and NAPL plume are compared for a site. The importance of these
infrequently investigated NAPL migration pathways is raised. The importance of sampling all major
contaminant migration pathways is stated. The differences in sampling depth are compared for APL
and three NAPL migration pathways.
The second group of topics evaluates alternative responses to the problem. Proposed remedial ion and
containment responses are discussed. The relative velocity of hydrophobic and hydrophilic
contaminants in NAPL and APL plumes are compared. The time required for a mobile NAPL plume
to convert into a tail of residual saturation is suggested. Finally, the principal issues are summarized.
2A BACKGROUND
The concern about the effectiveness of GW pump and treat operations first surfaced at the EPA
Office of Emergency and Remedial Response (OERR) in December of 1987 when GW experts from
the EPA lab in Ada, Oklahoma, raised the concern to OERR Headquarter's staff. OERR
commissioned a study of 19 sites to investigate the causes for poor response at GW pump and treat
sites (EPA 1989). This report found that aquifer cleanup progressed as predicted at some sites, but
results at other sites were disappointing.
The study looked at 19 sites where the GW extraction system had been in operation a sufficient length
of time for an assessment of whether the contaminant concentrations were declining as predicted.
The analysis did not distinguish between different types of sites. The study concluded that there was
not a way to anticipate when GW pump and treat would or would not be successful.
Travis and Doty (1990) suggest that the Superfund Program should abandon efforts to remediate GW
to health-based levels. Their opinion is that none of the 19 sites showed any conclusive proof of a
successful remediation or of satisfactory progress in reducing contaminant concentrations. They
736
-------
suggest Superfund should focus on pumping GW for 3 to 5 years (for mass reduction), and then
discontinue GW extraction. GW water pumping is not seen as adequate, and other approaches are not
considered. The problems identified in the 19-site study are presented as cause for giving up on
efforts to remediate contaminated GW. EPA, on the other hand, prefers to use the problems
identified in the remediation of the 19 sites to focus efforts on solutions to the difficulties identified.
3J) DIFFERENT CONCEPTS OF SUBSURFACE MIGRATION PATHWAYS
This paper suggests that the problem with GW pump and treat is one of a misconception of a
hazardous waste disposal facility. The mass of contaminants released is often underestimated and the
subsurface pathways moving the bulk of the contaminant mass away from the site are not recognized.
Typically, the plume expected is the plume that is sampled (and hence remediated). Typically, the
expected plume is a dilute solute (APL) plume. For sites where this concept is correct, GW extraction
can remove contaminant mass at an adequate rate. However, for many abandoned hazardous waste
disposal facilities, the dilute solute plume (APL) represents a small fraction of the contaminant mass
that the facility released to the subsurface. The remaining contaminant mass is in NAPL plumes, both
the migrating mass and the stationary tail. Pumping GW before removing the mobile NAPL mass and
the NAPL mass in the tail will require inordinate timeframes for the extraction of the contaminant
mass.
This paper suggests that site investigators could categorize sites into two classes. Sites in the first class
would continue to receive GW pump and treat remediations. Sites in the second class would have
their NAPL plumes and tails sampled. GW extraction is important at all sites, but at sites in the
second class, it needs to be preceded by extraction of the mobile NAPL (when still present), along
with some effort to extract the stationary NAPL tail (residual saturation) using secondary recovery
techniques. Extracting GW before removing NAPL will draw uncontaminated water over the NAPL
mass and generate more contaminated GW by dissolving the contaminants currently in the NAPL
plume into the water. This process can repeat for thousands of years before depletion of the
contaminant mass.
This paper suggests that simple extraction of contaminated GW is not adequate to remediate sites in
the second class. Approaches proposed in this paper may or may not be sufficient to reach health-
based levels in the currently contaminated area. However, demobilization of the mobile NAPL mass
may sometimes be a viable approach and will help protect currently uncontaminated areas from
exceeding health-based levels in the future. Unfavorable geology can eliminate the possibility of
cleaning up an aquifer (e.g., karst terrane can limit the ability to find the plumes). NAPL sites are
no different than APL sites in this regard; certain geology will present more challenges than the
Superfund Program can address at this point in time. This discussion will focus on sites where the
geology allows a successful remediation. This discussion will focus on abandoned commercial
hazardous waste disposal facilities. These sites receive many truckloads of waste; a small spill will
not present the problem that is discussed in this paper. The smaller contaminant mass may permit
successful removal of the mass by GW extraction. This paper will focus on organic chemical
contaminants; dissolved metals will not be discussed. Additionally, vapor phase transport pathways
exist and can cause contamination of infiltrating rainwater, however, for the sake of focus, vapor
phase transport will not be addressed by this paper.
The total mass released to the subsurface should be compared to the amount accounted for in the
sampled plumes. Information concerning the exact amount of mass released is typically not available.
However, plausible estimates should be made for comparison to the amount found in plumes leaving
the site.
737
-------
The concept of an abandoned hazardous waste disposal site releasing only a dilute solute plume (APL)
that transports contaminants from the facility to the sampling well causes problems with the choice
of subsurface contaminant migration models. For sites with only an APL plume, the APL models
make a realistic attempt to model the situation. For sites where the principal contaminant mass is in
a mobile NAPL plume, the use of APL models produces results inconsistent with the actual
conditions.
3JL NEED TO DIFFERENTIATE WASTE DISPOSAL SITES
Superfund sites could be discussed in two groups. The two groups would be differentiated by the
scale of the subsurface contaminant mass released and the type of subsurface transport pathway(s).
The utility of dividing abandoned hazardous waste disposal sites into two groups is that it allows us
to predict whether GW extraction and treatment will work or whether a faster contaminant removal
method is required. The correct concept of a site is important for both sampling and remediation.
The following paragraphs describe the characteristics of Class One and Class Two sites.
3.1.1 THE NATURE OF WASTES AT HAZARDOUS WASTE SITES
A good concept of a typical abandoned hazardous waste disposal facility begins with the type of waste
in the hazardous waste system. The proportion of solid and pourable waste is important to consider
in understanding how hazardous waste migrates away from abandoned hazardous waste disposal
facilities. Contaminants in solid waste move into the subsurface only after dissolving into the
percolating rain water. This pathway is limited by the low hydraulic loading the rain provides.
Contaminants in liquid waste move into the subsurface as fluid percolating through the pore spaces
in the soil. The hydraulic loading is provided by the waste itself, not the rain.
It is often thought that hazardous wastes are primarily solid materials placed on the land. However,
only 10 to 20 percent are solid wastes; the remaining 80 to 90 percent are pourable wastes (Skinner,
1984). Hence, the concept of a solid material leaching into percolating rain water should only be used
at those type sites. The most significant contaminant loading comes from liquid hazardous waste.
Hazardous liquid waste can be in two forms: it can be water contaminated with a few ppm of
contaminant, or it can be pure organic fluid. Just as the hydraulic loading differed for solids and
liquids, the two types of liquid waste pose two different contaminant loading rates. The 1,000,000
ppm contaminant concentration in pure organic fluid provides much more contaminant mass than
does water contaminated with a few ppm.
The placement of liquids onto the land was commonly practiced before the promulgation of the
Hazardous Waste Regulations in 1980. The Regulations now require treatment to Best Demonstrated
Available Technology (BDAT) standards; the only material that can be land disposed is the irreducible
treatment residual.
3.1.2 CLASS ONE SITES
If water contaminated with a few ppm of organics is placed into a pit, pond, lagoon, or landfill, then
contaminated water will leak out. This, by definition, is called an aqueous phase liquid (APL). The
contaminants leaving the site will only form a primary APL plume. This type of site cannot form a
NAPL plume. A primary plume is one that carries contaminants from the disposal pit to the sampling
point. A secondary plume is one that carries contaminants from a primary plume to the sampling
point. A dilute solute model is appropriate for these sites.
738
-------
3.1.3 CLASS TWO SITES
If organic byproduct fluid is placed into a pit, pond, lagoon, or landfill, then organic fluids will leak
out. This nonwater fluid is, by definition, called nonaqueous phase liquid (NAPL). The contaminants
leaving the site will form a primary NAPL plume, and a secondary APL plume will form from the
NAPL plume. Dilute solute models are not appropriate for these sites since most of the contaminant
mass migrates as a highly saturated NAPL plume. The dilute solute contamination that occurs does
not start at the site; it starts at the interface between the NAPL plume and the water. Hence, the
dilution that occurs between the site and a well, say 200 feet away, is different from the dilution that
occurs for contaminants that move 180 feet in the NAPL plume, and then move 20 feet in a dilute
solute plume.
3.1.4 HISTORICAL TECHNIQUES FOR IDENTIFICATION OF NAPL
Site investigators have used the occurrence of concentrations of 1, 20, and 33 percent of equilibrium
solubility as benchmarks indicating the presence of a NAPL plume (Cherry 1990, Miller 1990). This
practice sets the standard too high. Secondary APL plumes will typically show much lower
concentrations at the actual well position. Individual molecular identities are typically 0.1 to 2 percent
of the NAPL plume. This limits the maximum concentration in GW at the interface between the
NAPL plume and the water to approximately 0.1 to 2 percent of the equilibrium solubility. As the
hydrophobic contaminant migrates from the interface to a distant point (such as the actual well
location and depth), its concentration falls off sharply. Hydrophobic NAPL contaminants exhibit
highly retarded transport velocities in dilute solute plumes (APL). Commercial synthetic organic
chemical production is only 46 years old. Thus, the more hydrophobic contaminants have limits on
the total distance they can travel from the NAPL plume itself. Dispersion of APL transport between
primary NAPL plume and sampling point further reduces the concentration.
Sometimes the different plumes move in different directions. If the wells are placed for a different
direction of travel, then the distances between the NAPL plume and the well may be too great. Depth
of sampling is typically appropriate for dilute solute contaminated GW plumes, and typically a large
vertical distance from sinker and floater plumes.
12 COMPARISON OF MASS RELEASE RATE AND MASS REMOVAL RATE
A simple mass balance can be used to estimate the timeframe required for remediation. The mass
removal rate can be compared to the mass in place to estimate whether the contaminant removal rate
is sufficient to clean up the site in a reasonable timeframe. To illustrate the point, the next
paragraphs present two hypothetical mass release rates and two hypothetical mass extraction rates.
The mass in per year for the example Class One and Class Two sites are compared to the mass
extraction rates possible with GW extraction. The mass in per year for Class Two sites is also
compared to the mass extraction rates possible using a train of three extraction techniques. The first
technique is extraction of highly saturated mobile NAPL mass (where present); the second technique
is a secondary recovery technique to remove the residual saturation of the tail and the residual
saturation left by pumping the mobile NAPL mass; and the final technique is GW extraction of the
mass of contaminant dissolved in GW at the start of remediation (sorbed contaminant mass in
equilibrium with the dissolved concentrations is also included). The amount of GW pumping is much
smaller in this case because most of the NAPL mass has been extracted by the first two techniques.
739
-------
3.2.1 MASS IN PER YEAR
3.2.1.1 Class One Sites (APL Lagoon)
The hypothetical Class One site is an unlined lagoon that has water contaminated to 3 ppm placed into
the lagoon. For both examples, the hydraulic conductivity is assumed to be 0.001 cm/sec. The area
of the Class One lagoon is 10,000 ft2 based on dimensions of 100 feet by 100 feet. The hypothetical
contaminant concentration is 3 ppm.
If this lagoon is kept full; 77,418,000 gallons of water can percolate through the bottom of the unlined
lagoon each year. At 3 ppm, this volume of water will contain 232 gallons of organic contaminant.
This represents 4.2 barrels per year (55-gallon barrels).
The preceding values are thought to be representative of typical sites. Clearly, all of the values will
change from site to site. The reader can vary some of the values and obtain a feel for the range of
possible values for the number of barrels of contaminant that can percolate per year. The number
will vary, however, all Class One sites will have a small estimate of barrels per year.
3.2.1.2 Class Two Sites (NAPL Lagoon)
Historically, the hypothetical Class Two site is an unlined lagoon that has 55-gallon drums and 5,000-
gallon tank truck loads of organic fluids placed in the lagoon. No contaminated water was sent to the
lagoon. This site represents the typical abandoned commercial hazardous waste disposal facility that
accepted waste from more than one factory; hence, the wastes arrived at the site by either tanker
truck or by flatbed truck loaded with up to 80 barrels. For this example, the hydraulic conductivity
is also assumed to be 0.001 cm/sec. The area of the Class Two lagoon is smaller, 100 ft2 based on
dimensions of 10 feet by 10 feet. The contaminants are in pure form (neat); that is to say they are
approximately 1,000,000 ppm in concentration.
If this lagoon is kept full; 774,180 gallons of organic fluid can potentially percolate through the
bottom of the unlined lagoon (this assumes that the ratio of density to viscosity for the organic fluid
is the same as water; clearly the ratio can be higher or lower). Since this fluid is pure organic fluid,
the amount of organic contaminant is the same as the amount of fluid percolating through the bottom
of the lagoon. This 774,180 gallons represents 14,076 barrels per year (55-gallon barrels) This
quantity can be expressed as 155 truckloads of waste (3 truckloads per week).
The preceding values are thought to be representative of typical abandoned commercial hazardous
waste disposal facilities. Clearly, all of the values will change from site to site. As with the Class One
example, the reader can vary some of the values and obtain a feel for the range of possible values for
the number of barrels of contaminant that can percolate per year. The number will vary; however,
all Class Two sites will have a large estimate of barrels per year.
3.2.2 MASS OUT PER YEAR
3.2.2.1 Ground Water Pump and Treat
The previously stated contaminant loading rates can be compared to hypothetical contaminant mass
removal rates possible with extraction of contaminated GW. Clearly, the pumping rate can vary as
can the contaminant concentration in the produced waters; however, a representative rate can be
suggested. The rate discussed here was based on 10 actual sites where contaminated GW is being
extracted (USEPA 1989). The representative GW extraction rate is suggested as 150 million gallons
740
-------
per year. The concentration averaged across all produced waters is 3 ppm (this high concentration
is a favorable assumption as a much higher concentration in the lagoon would be necessary to produce
waters with a 3 ppm concentration). The reader can vary these parameters and evaluate the possible
mass removal rates possible with different pumping rates and different average concentrations.
Removal of 150 million gallons per year at an average concentration of 3 ppm removes 450 gallons
of organic fluid per year. This represents 8 barrels of organic fluid per year.
GW extraction and treatment in this hypothetical example can remove 8 barrels per year. For the
Class One site that releases 4 barrels per year, this approach can remove the contaminant mass in a
similar timeframe to the period that wastes were released to the ground. For Class One sites, GW
extraction is a viable tool. For these sites, GW extraction involves removing the contaminants in the
same form that they were released.
However, for Class Two sites, the contaminant mass removal rate of 8 barrels per year is so much
smaller than the release rate of 14,000 barrels per year that the approach is not viable. For every year
of releases, over a thousand years of GW extraction will be needed to remove the contaminant mass
by pumping GW with 3 ppm contaminant concentration. This approach does not involve extracting
the waste in the form that it was released; it involves extracting a much larger fluid volume with a
much lower contaminant concentration.
When the concept of a Class Two site is confused with a site having only a dilute solute plume, the
extraction of GW is pursued as a viable approach for containment or restoration. Unfortunately, the
concentration reduction over time will be less than anticipated due to the gross understatement of the
mass needing removal.
3.2.2.2 NAPL Pump and Burn with Surfactant Wash
A higher mass removal rate is possible by extracting highly saturated volumes of mobile NAPL. This
will leave a residual saturation in the area where the NAPL mass was pumped. The tail left by the
migrating NAPL plume will also represent a volume of aquifer with residual saturation. Pumping will
not remove this residual saturation; a secondary recovery technique is required. (For sites where the
mobile NAPL plume has moved so far as to have left its entire mass as a tail of residual saturation,
the secondary recovery technique is the first technique to be used, as there is no highly saturated
volume that can be pumped.) The secondary recovery techniques will leave a contaminant
concentration that is typically higher than health-based levels. GW extraction and treatment is
required as a third activity if health-based goals are intended.
Hence, the contaminant mass removal rate suggested is based on a train of three approaches: NAPL
pumping, secondary recovery, and conventional GW pumping of a small volume of GW. As with the
other hypothetical release and removal rates, the following parameters are felt to be representative;
however, they can vary and the reader is encouraged to explore the effect of changing the parameters.
For this hypothetical example, the NAPL extraction rate is set at 5 million gallons per year (20%
NAPL and 80% coproduced water). The surfactant wash (secondary recovery technique) is set at a
rate of 15 million gallons per year. The final phase, GW extraction, is set at the same rate as the GW
extraction alone, 150 million gallons per year. The three phases of extraction are done in sequence
rather than simultaneously. This combined treatment train can produce at an average rate of 286,000
gallons of organic fluid per year. This represents 5,200 barrels per year, or 57 truckloads per year.
This extraction train cannot be applied to Class One sites because they do not have a volume of
residual saturation of NAPL or a highly saturated volume of NAPL. This train can only be used for
741
-------
Class Two sites. At Class Two sites where the mobile NAPL plume has dissipated its volume by
leaving a tail of residual saturation, only the last two steps of the train can be taken. For Class Two
sites with the full train, the extraction rate of 5,200 barrels per year is sufficiently closer to the release
rate of 14,000 barrels per year to permit extraction in possible timeframes. The extraction riay take
longer than the period of releases, but at least it will not be over a thousand times the release period.
This extraction approach involves removing the wastes in the form disposed or as close to it as
possible. Removing the wastes in concentrated form increases the removal rate to the point where
it is closer to the release rate.
33 RELATIVE APL AND NAPL MASS AT A CLASS TWO SITE
This paper suggests that many sites that have had their GW plumes sampled and pumped should have
also had their NAPL plume pathways sampled. The hazardous waste practice is not monolithic.
Rather, there is a considerable distribution of types of approach. NAPL plume sampling is being
done by some site investigators.
The relative importance of the APL versus the NAPL pathways can be seen by considering the rate
of waste loading to the facility and the GW's ability to carry the mass away at concentrations typically
found in GW. A spill site or underground tank leak of 0.05 gallon per hour may or raay not
overwhelm the GW's ability to transport the contaminants away. A characteristic of abandoned
commercial hazardous waste disposal facilities is that of receiving more barrels of organic fluids than
can be transported away dissolved in GW. The ratio of mass in the APL plumes to mass in the NAPL
plumes varies; however, a feel for the scale can be obtained by looking at a site where both the APL
and NAPL plumes have been investigated.
The Hyde Park landfill/lagoon is a facility that received substituted and unsubstituted organic fluids
that were byproducts of a synthetic organic chemical manufacturing facility. Company records show
from 66 to 250 million pounds of non-NAPL waste (solid and hydrophilic liquids) were placed in the
facility. The records also show 93 to 350 million pounds of hydrophobic organic fluids (7 to 27
million gallons of NAPL) were placed in the lagoon (District Court 1980, Morgan 1979, Versa r 1980).
The Remedial Investigation has characterized the magnitude of the APL and the NAPL plumes.
Three thousand eight hundred gallons of hydrophilic and hydrophobic contaminants were found
dissolved in GW. Thirteen million eight hundred thousand gallons of NAPL plume were found
migrating down dip of the aquitard (Conestoga-Rovers 1989a and 1989b). Hence, if a Jiite has
received more mass than can be explained by the contaminants in solution and the sorbed
contaminants that are in equilibrium with those concentrations, it is likely to have had a NAPL plume.
The NAPL plume may consist solely of the immobile residual saturation left by a mobile NAPL plume
that has depleted its highly saturated volume, or a mobile, highly saturated NAPL volume may also
be present. Typical commercial hazardous waste disposal facilities received 2 to 20 truckloads per
day. Allowing 1 truck load in solution and 10 to a 100 truckloads for sorption, the rest of the waste
must be present as mobile or stationary NAPL mass (volatilization will occur).
M WE NEED TO SAMPLE ALL SUBSURFACE PATHWAYS
We need to sample all subsurface pathways in order to design appropriate remedial/containment
measures for Class Two sites. Different pathways flow at different depths, directions, and velocities.
Modeling of these parameters can help focus the sampling effort to intersect these pathways. It is
important to be aware of five classes of migration pathways:
1 Dilute solute plumes from dry landfills
2 Dilute solute plumes from lagoons
742
-------
3 Floater plumes - Light Nonaqueous Phase Liquid (LNAPL)
4 Neutrally buoyant plumes - Neutrally buoyant Nonaqueous Phase Liquid (NNAPL)
5 Sinker plumes - Dense Nonaqueous Phase Liquid (DNAPL)
Detailed discussion of modeling the APL and NAPL plumes is beyond the scope of this paper. The
reader interested in further discussion is directed to OSWER Directive 9285.5-1 (USEPA, 1988).
O RESPONSE TO PROBLEM OF MASS REMOVAL RATE
4J. SELECTING APPROPRIATE CLEANUP APPROACH FOR EACH CLASS
This example suggests that it is better to attempt to remove the wastes in the form they were released,
or as close to that form as possible. Applying GW extraction to a Class Two site is not limited by the
liquid-to-liquid dissolution rate (NAPL to APL); rather, it is limited by the ratio of the total mass in
place to the mass removal rate. A triple unit train may or may not be able to reach health-based
concentrations in a reasonable timeframe; however, it will make progress much faster than GW
extraction alone. At sites where the NAPL mass can be found and extracted, the triple train offers
hope of faster, more efficient remediations. Whether the triple train can satisfy all goals or not is an
issue; however, the first two unit operations can remove mass faster than GW pumping, and the third
unit operation (GW pumping) will be as fast as GW pumping alone. Hence, the triple train will
always put you closer to your goals in a given timeframe than GW pumping alone.
12 ALTERNATIVES FOR MASS REMOVAL RATE PROBLEM
While GW extraction cannot remove the contaminant mass of a Class Two site in a reasonable
timeframe, that is not justification for discontinuing subsurface remedial efforts. The Class One sites
can continue to receive GW extraction as the sole subsurface remedial activity. The Class Two sites
can have both the APL and NAPL plumes sampled and investigated. The NAPL plumes that can be
found can be extracted by pumping of the highly saturated volumes and secondary recovery of much
of the residual saturation tail. This may be enough to demobilize the NAPL plume and improve the
site sufficiently to be considered remediation of the site. In some cases, it will also be appropriate
to follow the first two techniques by conventional GW extraction. For the GW extraction phase to
be able to reach health-based levels, the secondary recovery technique must remove most of the mass.
The degree to which secondary recovery techniques can remove the mass has yet to be demonstrated.
Research is needed on secondary recovery techniques.
If removal of the residual saturated mass is substantially incomplete, further efforts involving
extracting contaminated GW may not be worthwhile. It has been suggested that the oil industry can
only produce 30 to 50 percent of the oil in the ground, and we should expect the same. For a number
of reasons, we may expect better yields.
First, the oil industry deals with very large-scale oil bearing formations; fortunately our plumes are
much smaller.
Second, they are producing fuel at an economic cost near 30 dollars a barrel. The additional costs of
secondary recovery before the 1972 embargo meant that secondary recovery was not utilized. Early
in the oil exploration period, only the easy oil was produced. As oil became more scarce, more costly
deposits were exploited. After the cost jump of 1972, it became practical to practice secondary
recovery. However, the price of 30 dollars a barrel still limited the degree to which it was practical
to produce oil for a profit by secondary recovery techniques.
743
-------
The Superfund Program is involved with protecting human health from exposure to carcinogenic
chemicals. We typically are not limited by a point of diminishing returns dictated by producing oil
for less than 30 dollars a barrel. Our reasons for extracting fluids from the ground are profoundly
different. The point of diminishing returns is also much different.
Third, the oil industry extracts oil from large deposits that are typically averaging 20 percenl oil and
80 percent brine. They try to find domes that trap closer to 100 percent oil to make their efforts more
efficient, but typically they must also harvest less concentrated deposits. Hazardous waste NAPL
plumes are smaller and the bulk of the migrating mass is saturated. Based on Schwille's; (1988)
measurements of residual saturation, we may in some cases be able to remove 75 to 85 percent of the
highly saturated NAPL volume as free fluid. The remaining 15 to 25 percent is the residual saturation
that requires secondary techniques for extraction. The tail of the NAPL plume will be close to
residual saturation and cannot be pumped; hence, it also requires secondary recovery techniques.
When initial saturation is only 20 percent and residual saturation is 5 to 15 percent, only 5 to 15
percent of the pore volume can be freely pumped. With 20 percent initial saturation and removing
10 percent, only 50 percent of the oil can be removed by pumping. At hazardous waste mobile NAPL
plumes, we may be able to freely pump 75 to 85 percent of the highly saturated NAPL volume and
still leave the same residual saturation volume.
The degree to which we can remove the residual saturation by secondary techniques is currently
unknown. We are researching this question at the present. However, it is clear that we can spend
more than 30 dollars a barrel to push the extent of extraction to higher levels than the oil industry is
able to extract economically.
We do feel that there may still be limits to our ability to remove NAPL's from the ground, but it will
be a different limit than for the oil industry producing economical fuel for motor cars.
43 CONTAINMENT VERSUS REMOVAL
It may be better to immobilize the highly saturated NAPL plumes at multiple sites by NAPL pumping
without secondary recovery than to polish a single site to health-based levels. If a mobile DNAPL
plume is present, a GW hydraulic gradient control effort will not stop the DNAPL plume. The
DNAPL plume will flow under the wells and form a new APL plume on the other side. Effective
containment measures at a site require understanding DNAPL pathways.
Hydrophobic contaminants move with a retarded velocity when migrating as a dilute solute (APL),
and an unretarded velocity when in a NAPL plume. Hydrophilic contaminants in APL or NAPL
plumes flow with a more similar velocity. The actual velocities will depend on the density, viscosity,
octanol-water partition coefficient, and aquitard dip to hydraulic gradient comparison. However, the
extreme retardation of compounds with log octanol-water partition coefficients over three suggests
that the NAPL plume will be faster than the APL plume for these compounds. In these cases,
containing the spread of hydrophobics in the fast moving concentrated NAPL plume is more
important than containing the retarded flow of hydrophobics in the dilute plume.
This paper suggests that the contaminant mass leaving Class Two sites will be found in two to three
forms: highly saturated mobile NAPL plume(s) (if present), tail of plume(s) consisting of residual
saturation of NAPL, and a secondary dilute solute plume(s). Historically, during operations of a
hazardous waste pit, pond, lagoon, or wet landfill, the waste organic fluids would be saturated in the
pit. This will cause highly saturated conditions in the porous media surrounding the pit. The NAPL
will displace most, but not all, of the water in the pore spaces. This non-water fluid will move under
a pressure gradient (due to negative buoyancy, hydraulic head, or chemical head). As long as the pit
744
-------
is kept full, the plume will not have a tail. The highly saturated conditions will be present from the
pit to the front of the NAPL plume. After cessation of waste loading to the pit, a tail of residual
saturation will develop between the pit and the mobile NAPL plume. As the mobile NAPL plume
migrates, it leaves some of its volume behind in the form of a tail of residual saturation. At some
point in time, this will deplete the mass in the mobile NAPL plume. At that point, the NAPL mass
will be stationary (further movement of contaminant mass will only occur as a dilute solute plume
(APL)). Clearly, some Class Two sites will have three forms present and some sites will have only two
forms present; it is a matter of time. If we get to the site after a long time, then the NAPL mass will
be a stationary source. If we get to the site shortly after cessation of waste loading, then the mobile,
highly saturated NAPL plume will be large.
It has been hypothesized that the time required for the mobile NAPL plume to become a tail of
residual saturation is less than a year (Cherry 1991). This would suggest that Superfund sites would
generally have a residual saturation volume and a dilute solute plume, but no mobile NAPL plume.
This image would support the idea of using gradient control to contain a site that is too difficult to
remediate. However, this containment approach would not be valid at sites where there was a mobile
DNAPL plume. At these sites, the mobile DNAPL plume will pass under the gradient control well
field and form a new dilute solute plume on the other side. This paper suggests that a much longer
time is required for the depletion of the highly saturated NAPL mass at commercial abandoned
hazardous waste disposal facilities.
4A TIME REQUIRED FOR NATURAL DEMOBILIZATION OF NAPL PLUMES
Whether a mobile NAPL plume is present or not is best determined by sampling; however, a
theoretical discussion can provide insight for determining when to look for a mobile NAPL plume.
For the sake of discussion, DNAPL plumes in simple geology will be discussed. The hypothetical site
has one aquifer with one thick impermeable bottom (aquitard) that has a dip, and it is reasonably
homogeneous and isotropic. The hypothetical site is an unlined pit that has the same mix of
substituted organic fluids poured into it for 10 years at such a rate as to maintain ponded conditions
in the pit at all times. This will give a continuous steady release rate. The height of the fluid in the
pit will change the mass flux out of the pit, but it will not make a large difference in the velocity of
the plume; it will cause a change in the cross sectional area of the plume. The saturated conditions
in the pit cause highly saturated conditions in the plume (previously water filled pores prevent full
saturation).
The density, viscosity, hydraulic conductivity of aquifer, and dip of the aquitard affect the actual
velocity of a DNAPL plume. Clearly, the velocity affects the distance the plume travels each year.
We can imagine three segments with different velocities. The velocity of the unsaturated zone
segment is the highest; since the pressure gradient is greater than one (say 1.5 for TCE), the direction
is downward. The downward migration continues through the saturated zone until reaching the
aquitard. This segment is at a slower velocity, because the pressure gradient is now the difference
between the DNAPL density and the density of water (say 1.5 - 1 = .5). The third segment is the
horizontal migration of the DNAPL plume as it moves down dip. This velocity is the smallest since
the gradient is the negative buoyancy multiplied by the slope of the aquitard dip (and some influence
due to the natural hydraulic gradient which will be ignored).
The first two segments are relatively fast and the third slow; as an approximation, the first two
segments will be considered to require less than 2 months. The 2-month time is small compared to
the 10-year life of the pit and will be ignored. Thus, the simplified model has the DNAPL plume
migrating down dip of the aquitard. The length traversed each year is X, the actual value of X
depends on the parameters discussed above. Using the variable "X" makes the discussion independent
of these parameters. At the cessation of waste loading in 10 years, we can expect the DNAPL plume
to have moved a distance of 10 X.
74
-------
Residual saturation can be 5 to 15 percent of the pore volume (Schwille 1988); for this example, the
value of 15 percent will be assumed. For this example, an initial high saturation value of 85 percent
will be assumed. During the 10 years of waste loading, the plume is highly saturated from the pit to
the tip of the DNAPL plume; there is no tail of residual saturation present. After cessation of waste
loading, a tail of residual saturation will develop between the pit and the mobile plume. Once the
DNAPL plume reaches 5.6 times its highly saturated length, it will consist of a tail of residual
saturation that is not moving as a mobile DNAPL plume. The plume will reach a distance: of 56 X
(lateral spread will reduce total distance).
If it took 10 years to travel a distance of 10 X, then approximately 46 years would be required for
the mobile DNAPL plume to dissipate its volume as a tail of residual saturation (total time from start
56 years). Commercial synthetic organic chemical production has been occurring for the last 46 years.
Hence, the likelihood of investigating a commercial hazardous waste disposal facility with a mobile
DNAPL plume is high. The approach of using gradient control to stop migration on sites where
health-based goals are unattainable is unworkable in that the mobile DNAPL plume would not be
contained. The mobile DNAPL plume would pass by the containment effort and dissolve into GW
on the other side and compromise the effort.
LO SUMMARY
In summation, this paper suggests that it is important to sample the proper depths so as to sample all
contaminant pathways, especially NAPL migration pathways. The concept of the site must fit the site
so that the proper samples are taken, and so that the remedial measures designed for the site actually
fit the site. The proper concept of an abandoned commercial hazardous waste disposal facility is
necessary for extracting contaminant mass within acceptable timeframes, and for implementation of
effective containment approaches.
6,0 REFERENCES
Cherry, J., 1990. Presentation at EPA Washington D. C., Monday, May 7.
Cherry, J., 1991. Personal communication with author at EPA Washington, D. C., Tuesday,
February 5.
Conestoga-Rovers & Associates, 1989a. APL and NAPL Plume Refinement in Overburden, Hyde
Park Requisite Remedial Technology Program. Submitted to USEPA Region II.
Conestoga-Rovers & Associates, 1989b. Bedrock APL and NAPL Plume Refinement, Hyde Park
Requisite Remedial Technology Program. Submitted to USEPA Region II.
Consent Judgment in the United States District Court for the Western District of New York. Civil
Action No. 79-989. EPA Draft (5/14/80)
Miller, D., 1990. Presentation at EPA Washington D. C., OERR Division Directors Office
Morgan, R., Johnson, J., Mason, B., desRosier, P., Librizzi, W. 1979. Initial report: technical
evaluation of Data on Hooker Chemicals and Plastics Corporation Waste Disposal Sites in Niagara
Falls, New York. USEPA
Schwille, F. 1988. Dense Chlorinated Solvents in Porous and Fractured Media - Model Experiments.
Trans, by Pankow, J. F., Lewis Pub., Chelsa, Mi, TD 426.s3813 1988 628.16836 87-29679, ISBN 0-
87371-121-1
746
-------
Skinner. 1984. Banning wastes from land disposal. EPA OSWER Briefing on 1984 Amendments to
RCRA. Dec. 11, 1984, Wash., DC.
Travis C., Doty C. 1990. Can Contaminated Aquifers at Superfund Sites Be Remediated?
Environmental Science and Technology, Department of Energy, Vol. 24, No. 10.
USEPA. 1988. Superfund Exposure Assessment Manual. USEPA,OERR. Contract No. 68-01-6271.
OSWER Directive 9285.5-1, EPA/540/1-88/001
USEPA. 1989. Evaluation of Ground-Water Extraction Remedies, Volumes 1 and 2, EPA, OERR.
Cont. No. 68-W8-0098. EPA/540/2-89/054, EPA/540/2-89/054b.
Versar, 1980. Assessment of Risk Associated with Implementation of Containment and Monitoring
Programs for Hyde Park Landfill Site. EPA, Dioxin Task Force. Cont. # 68-01-5948, Work Order
1, Subtask 7.
747
-------
OPTIMIZING AND EXECUTING
A
MULTI-FACETED REMEDIAL ACTION PLAN
C. DENNIS PEEK, P. E.
Geraghty & Miller, Inc.
14497 N. Dale Mabry
Suite 115
Tampa, FL 33618
(813) 961-1921
INTRODUCTION
The objective of this paper is to discuss the successful
implementation of a multi-faceted, multi-PRP remedial action at
the Seymour Site (former Seymour Recycling Corporation) in
Seymour, Indiana.
The importance of the this site is the rate of progress made
in the implementation of remedial design (RD) and remedial action
(RA) . The Seymour Site is one of the first NPL sites remediated
by the potential responsible party's (PRP's) to reach this point
in remediation. The RA is nearly two years ahead of schedule on
the 58 month schedule of the Consent Decree.
Although the Consent Order and Remedial Action Plan (RAP)
attempted to anticipate every eventuality, the details of program
design and implementation required nearly continuous coordination
and adjustment. This paper discusses techniques used to
accomplish this program in approximately one-fourth of the time
initially projected. Particular focus is on the impact on the
design and construction processes.
An unusually high level of cooperation was achieved by all
parties to the remediation that has enabled the project to reach
this level of clean-up in such a short time frame.
BACKGROUND
The general scope of this project is to remediate the site in
accordance with the Consent Order, Record of Decision (ROD), and
Remedial Action Plan (RAP) through the use of several
technologies. Initial remedial action involved the implementation
748
-------
of a Plume Stabilization project in accordance with the Agreed
Order to reduce the spread of ground-water contamination prior to
finalization of the Consent Decree.
DESCRIPTION OF SITE
The Seymour Site is a 14-acre facility located two miles
southwest of the City of Seymour, Indiana, on land owned by the
City of Seymour in an industrial park at the local airport,
Freeman Field. This facility operated as Seymour Recycling
Corporation, a processing center for waste chemicals, until late
1980. When the facility was closed, over 55,000 drums, 100 bulk
tanks of various sizes, and tank trucks, most containing waste
chemicals, were left on the site. Ten buildings were left
standing and an incinerator had been operated at the facility.
Hazardous substances had leaked into the ground causing
contamination of the soil and the shallow ground water aquifer.
Surface water run-off and the incinerator operation had spread
contamination along the natural drainage ditch leading from the
site, known as Northwest Creek. Vapor emissions, fires, and
noxious odors had been common problems prior to site closure.
The Seymour Project begin as a United States Environmental
Protection Agency (USEPA) Region V Emergency Response action in
1982. The drums, tanks, and some surface soil were removed from
the site and a clay soil layer placed on site. The Remedial
Investigation (RI) was completed in 1985. The Feasibility Study
(FS) was published in 1986.
PROJECT HISTORY
In response to findings of the investigation, a negotiation
among the PRP's, USEPA Region V, and the Indiana Department of
Environmental Management (IDEM) ensued. The result of these
negotiations was a Consent Decree entered in the Indianapolis
Federal District Court in December of 1988, that included 109
PRP's. A Trust Agreement was part of the Consent Decree
establishing the Seymour Site Trust (the Trust) with Monsanto
Agricultural Chemicals as Trustee.
Geraghty & Miller, Inc. (G&M) became involved during
negotiations prior to the Consent Decree. Geraghty & Miller
believed that it was important to begin plume capture quickly
rather than waiting until negotiations were completed. An Agreed
Order between the USEPA and the list of "Generator Defendants"
(PRP's) was signed in January, 1987, prior to the conclusion of
negotiations with the PRP's to allow the implementation of a Plume
Stabilization Project by Geraghty & Miller to reduce the migration
of ground-water contaminants. A Discharge Authorization was
granted by the City of Seymour (a PRP) in October, 1988, to allow
the discharge of pretreated ground water to their public owned
treatment works (POTW). After approval of the Consent Decree, the
Trust retained Geraghty & Miller as prime contractor to implement
749
-------
the RD/RA and fulfill the objectives of the Consent Decree, the
ROD and the RAP.
PROJECT APPROACH
As the Agreed Order dealt only with interim capture and
pretreatment of the ground-water plume, it was specific as to the
approach to recover ground water, pretreat it, and discharge it to
the City of Seymour POTW. This approach was based on a Workplan
prepared in December, 1986, by Geraghty & Miller. This interim
pretreatment was based on the estimated final pretreatment process
so that the preliminary pretreatment plant actually became a pilot
plant to test the expected long-term or final treatment method and
equipment. A treatability study and test period were required to
acclimate the POTW to the pretreated discharge water and ensure
that no adverse effects would result from Site discharge.
The Consent Decree and the RAP were specific as to what
Remedial Action was to be taken from an overall viewpoint but
allowed for the study of various aspects of remediation prior to
final selection of method and configuration. For instance,
although some type of pump and treat system for ground water
treatment was required, the final treatment method was ~~o be
determined based upon information obtained during the Treatability
Study conducted under the Agreed Order. Final recovery well
locations, configurations, and sizing were to be determined after
completion of an aquifer step test, additional rounds of monitor
well sampling and refinement of the ground-water model. These
were some of the numerous aspects to address during the RD/RA
implementation that had an impact on the design and implementation
process.
Site remediation involved a number of different technologies:
ground-water recovery and pretreatment (iron pretreatment, air
stripping, filtration, liquid phase adsorption using granular
activated carbon (GAC)), discharge to the City POTW, and expansion
of the ground water recovery system. Coordination of over twenty
employes from eight offices and four different groups internally
plus coordination with laboratory personnel, several
subcontractors, regulatory personnel, and regulatory consultants
made this project and interesting experience. Extensive
integration of various tasks performed by or under the direction
Geraghty & Miller was required and is ongoing. These tasks are :
• Groundwater Investigation, Sampling, Monitoring
• Plume Stabilization
• Preliminary Pretreatment Plant
18 week Treatability Study
• Groundwater Modeling
750
-------
• Risk Assessment - Water & Air Pathways
• Baseline Air Study
• Past Plant Risk Assessment
• Air Monitoring Programs
• VES Predesign study
• Demolition of On-site Buildings
• Asbestos Removal from Buildings
• Disposal of Hazardous Wastes Stored on Site
• Removal of Contaminated Sediments from Northwest
Creek
• Bioremediation
• VES Installation
• Containment and Disposal of all Site Storm Water
Run-off
• Cap Construction
• Final Pretreatment Plant Design & Construction
• Sewer Line
• Deep Aquifer Wells
• Long-term Cumulative Risk Assessment
Conventional control and monitoring of project activities
plus extensive financial planning and control was and is part of
the overall project.
Of significance is the fact that all the remediation work is
risk-driven. The objective of the remediation is to reduce risk
to a maximum cumulative excess lifetime cancer risk level of 1 x
10~5 at and beyond the site boundaries and of 1 x 10~6 at the
site's Nearest Receptor over a 70 year lifetime exposure.
Basically, water and air pathways must be considered.
DISCUSSION
The actual RD/RA work can be viewed in two significant,
distinct parts, the Plume Stabilization Project and the Remedial
Design/Remedial Action. These parts are discussed individually
751
-------
below. Of significance is the number of different tasks that had
to be integrated to achieve the objective.
PLUME STABILIZATION PROJECT (AGREED ORDER)
In the first part, Geraghty & Miller, under the Agreed Order,
prepared plans and specifications and provided construction
management for a ground water pretreatment plant that was a pilot
operation to study the proposed treatment methods. At this point
in the site investigation, only an approximate value of the volume
of ground water to be recovered and treated could be estimated, up
to 300 gallons per minute. Consequently, the preliminary
pretreatment plant was designed to operate in the range of 100 gpm
but with the intent of being expanded up to 300 gpm. The original
layout basically consisted of one treatment stream from an air
stripper through multi-media filters to a GAC unit. The GAC unit
was sized large enough hydraulically to accommodate up to 300 gpm
easily and up to 600 gpm if required. Adequate floor space was
made available to allow installation of a second air stripper and
multi-media filter sized for up to approximately 200 gpm. This
plant was completed in the third quarter of 1988 and start-up
preparations begin for a late 1988, early 1989 test period.
An aquifer step test was performed and the plant was started
up for an 18-week test and phased treatability study. The data
obtained from the treatability study and test proved the basic
concept of air stripping and GAC adsorption as a viable
pretreatment method and indicated specific areas that required
further refinement for a final plant configuration. After
completion of the study, the plant continued in full operation,
for the purpose of plume interception and stabilization, for over
fifteen months until being shutdown for modifications. The
additional data learned during this fifteen month period was used
to further refine the design and long-term operation goals for the
final plant.
The aquifer step test indicated that the original ground-
water model, developed using data from slug tests performed during
the RI, was inadequate. The ground-water model was replaced with
a new model of substantially larger scope that was based on data
obtained during the aquifer step test and preliminary pretreatment
plant operation. This new model revealed that the plume of
contaminated ground-water was moving approximately twice as fast
as the RI had estimated. The overall impact of this new finding
would later have significant impact on the project. If the plume
had been moving faster than the original model indicated, then the
possibility of a much larger plume existed that may require
significantly larger extraction rates for capture and subsequent
treatment. As overall treatment strategy and plant design may be
impacted by this new development, the importance of obtaining
current data for design work was apparent. Consequently, a new
ground-water investigation was launched to determine the extent of
the plume and to calibrate the model. This data would also be
used to develop a solute transport model of the shallow aqu:.fer.
752
-------
However, RD/RA activity continued, with the understanding that
revisions may be necessary in the work based on the new data.
The new study finally determined that the plume was over one
mile to the northwest of the site. New pumping strategies were
developed and treatment alternatives examined for the farthest
area of the plume. It was determined that an additional recovery
well pumping at 200 gallons per minute (gpm) would be required at
the nose of the plume but the contamination was such that the
discharge could go straight to the POTW without pretreatment.
After confirmation with POTW officials and concurrence with the
agencies on scope, new RD/RA activity for this part of the project
was implemented that continues at this time.
REMEDIAL DESIGN / REMEDIAL ACTION
In the second part, the Trust engaged Geraghty & Miller as
general or prime contractor to remediate the site. Some of the
activities of this part were running in concurrence with the first
part. Remedial design activities for the Consent Decree and RAP
implementation begin in the second quarter of 1989. Specific
objectives of the Trust in addition to satisfying the requirements
of the Consent Decree, ROD, and RAP were:
• Solve environmental problems
• Accelerated schedule - early completion
• Avoid stipulated penalties
• Operate & construct without excess exposure of the
public to hazardous materials
• No lost workday injuries
• Positive community relations
• Operate within budget
The Trust's strategy to achieve these objectives consisted of
the following elements:
• Develop aggressive schedule
• Avoid interruption of engineering
• Develop large bid packages
• Utilize experienced contractors/personnel
• Shorten communication lines
• Team approach between PRP's, contractors, USEPA,
IDEM, City of Seymour
753
-------
• Cost sensitivity
• Community relations
• Flexibility
It was determined very early in the process that time was of
the essence in the implementation of the RD/RA. Delays in
remediation could potentially increase risk through exposure as
contamination could spread. More extensive contamination would
only increase costs and time required to complete remediation. In
any large project, time to execute the work has a significant
bearing on cost, particularly when field activities are underway.
Due to the geographical location of the site, project timing and
thus completion could also be severely affected by weather,
particularly winter and rainy seasons. Another potential source
of delay was a lack of current information. Much of the data
gathered in the RI/FS phase was several years old when RD/RA
activities begin in early 1989.
In order to expedite remediation and to reach the stated
objectives in a timely manner, a fast-track approach was used.
This approach places tasks and the decision making process in a
parallel rather than strictly sequential mode. Multiple
activities occur simultaneously with periodic updates and sharing
of information to review current status. In short, rather than
wait for all data to be collected and analyzed, process design
decisions are made based on preliminary information. The design
is revised as necessary. Rather than wait for the design to be
completely finalized, construction begins with minor changes
occurring as part of the construction process. Rather than wait
on full regulatory approval, work proceeds with the realization
that some changes will probably be required after regulatory
agency review and approval.
In order for this fast-track approach to be successful, some
specific techniques were adopted. From an overall viewpoint,
flexibility and adaptability, the ability to respond quickly to
changes, were key traits that were essential for success. First,
the project RD/RA was broken into clearly distinct phases that
were independent enough so that work could run simultaneously.
Second, internal project communications were improved through the
use of frequent meetings of all key personnel and routine weekly
conference calls among all personnel, including regulatory
agencies. The importance of good communication in this approach
cannot be overemphasized. Third, informal technical reviews with
agency personnel during the engineering process were held to
discuss issues and the overall project direction. Fourth, work
was allowed to proceed based upon verbal approval from agency
personnel rather than waiting for formal, written authorization.
It was accepted that this approach was at risk but it allowed
engineering design to proceed without significant interruptions.
754
-------
First Stage - Site Preremedial
The first stage begin with additional tests and studies being
performed on the site in the areas of bioremediation and vapor
extraction in preparation for the final remediation. Also, during
this period, ground-water sampling continued on the site and a
baseline air monitoring study workplan was prepared and submitted
for approval.
A preliminary design for the soil vapor extraction system
(VES) had been developed during the development of the RAP. The
VES predesign study was conducted to quantify and qualify soil
contamination and soil gases plus obtain data on the soil
permeability over the site so that the design of the VES could be
finalized. In the interim, preliminary design drawings were
developed and submitted to the agencies for review and approval
with the understanding that the design would be modified based on
the results of the VES study. The preliminary drawings were used
to obtain bid pricing and scheduling so that RD/RA work could go
forward.
Second Stage - Initial Remedial Design/RemedialApt^OQ
In the second stage, Geraghty & Miller proceeded with design
and construction of the RD/RA except for the final pretreatment
plant design. By proceeding with the design work, and expecting
that changes would be required, it was possible to substantially
define the scope of work, prescreen and qualify potential
subcontractors, develop a bid package covering the bulk of
remedial construction activities, bid the work, award a contract
to the selected bidder, and begin field work while final data was
still being obtained and evaluated. For bid and design purposes,
this stage of the work was broken down to six phases:
I. Site Civil Work
II. Decontamination Facility
III . Demolition
IV. Vapor Extraction System
V. Sediment Removal
VI. Cap Construction
In fact, the bid package was structured in anticipation of
the changes by establishing unit prices for work expected to
change. Using this approach, valid comparisons between bids could
still be obtained thus keeping the bid process competitive and
effective. Field mobilization could then occur so that
remediation could begin sooner than under a sequential approach.
To expedite work, it was decided to prepare an overall Site Health
755
-------
and Safety Plan for all activities and issue addenda for each
phase of the work. Likewise, workplans for each phase were
individually prepared and submitted for review and approval.
Although the final ground-water pretreatment plant design and
construction was not part of this stage of the work, projections
and estimations were made as to the configuration and size of the
final plant. This information was then used to design the
expansion of the existing pretreatment pilot plant building to
house the decontamination facilities (increasing square footage
from 2640 to 7590) and treatment equipment for use in treatment of
site run-off water captured during construction. The treatment
equipment for processing site run-off water was selected with the
intent of reusing as much equipment as possible in the final
design.
The work executed under this phase consisted of preparation
of plans (health and safety, and work) and specifications for the
expansion of the pretreatment pilot plant building for use in
treatment of run-off water and for decontamination of equipment
and personnel, demolition of ten buildings on site, nearly all of
which contained asbestos, disposal of hazardous wastes stored on
site, removal of contaminated sediments from a nearby creek,
containment and disposal of all site storm water run-off,
installation of the vapor extraction system, application of
nutrients to enhance biodegradation, construction of a twelve acre
RCRA type cap, and expansion of the ground water recovery system.
By the way, this stage of heavy site activity was accomplished
with no recordable accidents or injuries after over 300 days in
the field.
The soil VES design was modified using the predesign study
data and reviewed with the agencies for concurrence before actual
construction begin. An interim review meeting was held at the
site with all affected parties to discuss the study before1 the
design drawings were revised. After acceptance of the design,, the
construction drawings were modified and the scope of work changed
by contract change order. The unit pricing method of this item
netted a cost reduction of over $200,000 because the number of
laterals was reduced.
Third Staye - Final Pretreatment. Plant Design and
Construction
The third stage again used phasing of the work required for
remediation. The phases of this work were:
VII. Final Ground-water Treatment Plant
VIII. Lift Station & Sewer Line Installation
IX. Well and Pipeline Installation
756
-------
In this stage, Geraghty & Miller, using data gathered during
the treatability study, finalized design of the final pretreatment
plant, modified the existing plant and site run-off water
treatment equipment and added new equipment required for the
treatment of iron and increased capacity from 100 gpm to 400 gpm.
A capacity of only 300 gpm (minimum) was required by the Consent
Decree. The final plant capacity was set at the "best guess
estimate" based on all known data at the point of design plus
projections. This data, from the calibrated model, indicated that
the existing two recovery wells would be operated at approximately
140 gpm but that up to an additional 200 gpm may require treatment
in the plant if the deep aquifer was contaminated.
This final plant was constructed and placed on-line in the
first quarter of 1991. The basic plant configuration consists of
large aerator tanks and sodium hypochlorite injection for iron
treatment, a continuous backwash sand filter for removal of iron
precipitate and sludge, parallel air strippers (existing pilot
unit plus a new 300 gpm unit for the expansion) for removal of
volatile organics, and a series granular activated carbon (GAG)
system with two each 20,000 pound GAC vessels for removal of non-
volatile organics and to provide a safe backup for the air
strippers. The plant is fully automated utilizing electronic
control and instrumentation systems with remote monitoring by use
of a computer and modem. A meteorological station installed
during early studies was connected into the plant control system
for data accumulation.
A new sewer lift station and over 2000 feet of 8" double
containment force main is being installed to connect the plant to
the municipal sewer system. That work is ongoing. Plans have
been made for the installation of four deep aquifer monitor wells
that can easily be converted to recovery wells if contamination is
found. The pipelines for these recovery wells have also been
designed so that installation could be quickly implemented.
Capacities for these wells were estimated with a high accuracy
based on the extensively developed model.
Plans for the new recovery well discussed as part of the
Plume Stabilization Project were also prepared as part of this
phase. A new 8000 foot 4" pipeline for that well is currently
under design.
Fourth Stage - Long-Term Operation
The fourth stage of the RD/RA involves the shift from
construction to operation. The emphasis is on long-term
operation, sampling programs, monitoring system performance, and
performing risk assessments. This work is phased as follows:
X. Vapor Extraction System Start-up, Operation and
Maintenance
757
-------
XI. Maintenance Plan for Cap and Site and Security
Plan
XII. Vapor Extraction System Closure Plan
XIII. Ground Water Monitoring
XIV. Cumulative Risk Assessment
The majority of this work is ongoing or under development at
this time. Work continues on these items while completed RD/RA
items, such as the final ground-water pretreatment plant, are
maintained in operation. Certain closure items, such as Operation
and Maintenance Manuals, for the final plant are prepared during
this stage. The long-term cumulative risk assessment is also
being prepared as part of this stage.
CONCLUSIONS
The successful implementation of the RD/RA at the Seymour
Site, using an aggressive, fast-track approach to project
execution, has demonstrated the viability of such an approach to
the remediation of Superfund sites. The keys to success of this
approach are good communication and a cooperative team approach to
the project. All parties to the project (regulatory, PRP's,
consultants, and contractors) must be part of the team and be
willing to operate in a cooperative manner with the common goal of
achieving an effective remediation.
758
-------
V. HEALTH AND SAFETY
759
-------
EPA/Labor Health and Safety Task Force
Joseph C. Cocalis, P.E., CIH and Kenneth W. Ayers, P.E.
U.S. Environmental Protection Agency
401 M Street S.W. (OS-220W)
Washington D.C. 20460
(703) 308-8356
John Moran
Laborers' National Health and Safety Fund
905 16th Street, N.W.
Washington, D.C. 20006-1765
(202) 628-5465
INTRODUCTION
In response to worker protection issues arising from activities at several NPL sites in 1989, Don
Clay, Assistant Administrator of the Office of Solid Waste and Emergency Response, U.S.
Environmental Protection Agency (EPA) established a special EPA/Labor Health and Safety Task
Force. The initial goal of the Task Force was to improve adversary relationships that were
developing between labor unions, the U.S. Army Corps of Engineers (USAGE), and the EPA
Regions. The long term goal of the Task Force is to provide a forum for the discussion of health
and safety issues at Superfund sites.
The task force, focusing only on worker health and safety issues at hazardous waste sites, is
composed of key EPA personnel from the Environmental Response Team and the Hazardous Site
Control Division and personnel representing the three principal construction trade unions involved
in hazardous waste clean-up. The International Association of Firefighters (IAFF) have recently
been included in task force activities. The Occupational Safety and Health Administration
(OSHA) and the USAGE serve as technical advisors to the task force.
Members include:
1. Joe Cocalis (Co-chair, EPA), EPA Hazardous Site Control Division (HSCD),
Design and Construction Management Branch (DCMB).
2. John Moran (Co-chair, Labor), Director Safety and Health, Laborers' National
Health and Safety Fund (LNHSF).
2. David M. Traenor, Director of Research and Education, International Union cf
Operating Engineers (AFLCIO).
3. Donald Elisburg, Executive Director, Occupational Health Foundation, (note: The
Occupational Health Foundation is a technical resource center that is sponsored by
25 Unions)
4. Les Murphy, Director, Hazardous Materials training for Emergency Response
Personnel, IAFF.
5. Vernon McDougall, Teamsters.
760
-------
6. Kenneth W. Ayers, Chief, EPA HSCD DCMB.
7. Rod Turpin, Chief, Safety and Air Surveillance Section, EPA Environmental
Response Team (ERT).
Note: Technical advisors who have attended meetings include: Thomas Donaldson, Robert Stout,
and Reuben Sawdaye (USACE); Maryann Garrahan, Elizabeth Grossman, and Charles
Gordon (OSHA); Charles Reese (LNSHF) and Joe Vita (Teamsters); and Sella Burchette
and William Zobel (EPA).
The Task Force, which meets bimonthly, has dedicated its recent efforts to reviewing site safety
and health issues and all OSWER Superfund safety and health directives and guidelines. The goal
of this review is to share with Labor organizations, actions that EPA is undertaking or anticipates
to improve worker health and safety at Superfund sites. Specific issues the Task Force has
addressed include site characterization (clean versus contaminated areas), training, response to
health and safety inquiries, and communications between the various parties involved with
Superfund activities.
DISCUSSION
1. Clean versus contaminated areas. One of the issues that the Task Force is
investigating is how to designate areas within a Superfund site as "clean"; that is
areas where the OSHA worker protection standard does not apply. Of particular
concern is for the health and safety of untrained workers performing intrusive
operations in designated "clean" areas who uncover unknown pockets of
contamination. The Task Force is supporting the development of design guidelines
and models which will assist the design engineer in estimating the occupational
health risk from existing remedial investigation data. The issue also encompasses
the redesignation of established areas.
a. Design guidelines. Where clean areas are adjacent to exclusion zones, a site
assessment should be the basis for the establishment of "clean" areas. Aerial
photography, topographic analyses, and site historical data are useful
analytical tools, but should be supplemented by sampling and not be the
sole criteria used to make decisions.
A good rule of thumb is to define clean areas as areas with less than three times
background concentration. Where background concentrations are exceeded or
unknown, a site characterization/risk assessment that is reviewed by a competent
person, such as a certified industrial hygienist with site characterization
experience, is recommended. (Reference 1, an Environmental Response Team
draft fact sheet on establishment of work zones, contains additional information on
clean zone designation).
For intrusive operations in the vicinity of contaminated areas, it is often prudent to
require workers to have the 40 hours of training so that they can recognize hazards
and take appropriate corrective action.
b. Modeling. Modeling can be a useful tool for predicting a protective level
of occupational exposure from site data. A USACE - EPA team modified
existing models to assist in a characterization/assessment of the Baird and
761
-------
McGuire Superfund site in Holbrook Massachusetts. The models were part
of an initial attempt to project borehole concentration to the potential for
occupational exposure. The models, which were considered protective, still
require refinement and field testing. HSCD, with the cooperative efforts
of the Task Force, is pursuing contractor model development and
validation.
c. Redesignation of established areas. EPA requires justification for
redesignation of clean areas and major changes to this policy are not
anticipated at this time. Situations will arise where additional information
warrants an investigation into the validity of "clean" zone designations. The
proper mechanism to investigate boundaries is through modifications to the
Health and Safety Plan (HASP). Boundary modifications should be
proposed in writing through the prime contractors' professional staff with
review by the industrial hygienist. In situations where the health and
safety of workers is in question, a conservative approach is necessary and
an interim protective interpretation of boundary lines should be considered.
2. Response to Labor inquiries. The Task Force has strongly endorsed a policy of
open communication, in which all health and safety inquiries receive a prompt and
professional response. Issues the task force is investigating include: Labor
participation in health and safety programs, OSHA inspections, imminent danger,
and other unsafe or unhealthful working conditions.
a. Labor participation. The Task Force is encouraging labor participation in
the health and safety programs at Superfund sites. A Labor representative
should be given the opportunity to accompany the inspector during non-
OSHA inspections and evaluations. Situations that exclude Labor
participation create an atmosphere of distrust, promote the spread of
rumors and are often counterproductive.
b. Health and Safety Enforcement. Inspections for enforcement purposes are
the responsibility of OSHA. The remedial action construction manager is
responsible for enforcing the terms of the contract for day-to-day worker
protection. The construction manager's responsibility include the issuance
of stop work orders in situations where violations of the health and safety
provisions of a contract are violated.
c. Imminent danger. Whenever and as soon as one is made aware of a danger
which could reasonably be expected to cause death or serious physical
harm, that person has the responsibility to immediately notify the affected
employees, and parties with the responsibility and authority to remove the
danger. In situations where an imminent danger exists, both the prime
contractor's site coordinator and the construction manager's on-site
representative have the responsibility and authority to stop all activities or
withdraw employees. If steps are not taken to remove the danger, OSHA
should be immediately contacted.
d. Other than imminent danger. For Federal-lead remedial action projects,
health and safety inquiries should be channeled through the construction
manager, who has the responsibility to notify the prime contractor's site
coordinator (or the responsible party) verbally and in writing of the unsafe
762
-------
or unhealthful condition. For other than Federal-lead projects, the prime
contractor's site coordinator should be notified verbally and in writing of
the unsafe or unhealthful condition.
3. Dissemination of Health and Safety Information. One of the objectives of the Task
Force is to identify problem areas and to disseminate information/instruction to
remedy the problem. Problem areas previous identified include: confusion about
health and safety roles and responsibilities among the numerous parties involved
with remedial activities, the establishment of work zones within a site, and
compliance with various health and safety instructions and regulations. The
establishment of work zones and compliance will be discussed in detail as part of
the Environmental Response Team presentations. The roles and responsibilities
fact sheet which was drafted by EPA's Hazardous Site Control Division in response
to a Task Force request will be discussed in detail here.
a. Roles and Responsibilities Fact Sheet.
(1) Remedial Project Manager (RPM). As the EPA's prime contact or
representative for a site, it is important for the RPM to be a strong
safety and health advocate. The RPM has the responsibility to
coordinate, direct, and review the work of EPA, responsible parties,
other agencies, and contractors to assure compliance with the
National Contingency Plan. As such, the RPM oversees compliance
with health and safety programs. The RPM does not have a direct
line of authority to the prime contractor. The RPM should be
informed of situations where health and safety issues impact overall
project cost, scheduling, technical quality, or public
health/environmental protection. However, the RPM's primary
responsibility is oversight, not action. Items requiring action
should be referred to the appropriate individuals or agencies (i.e.
the construction manager, prime contractor, the State, responsible
party, or OSHA).
(2) Architect Engineer. The architect engineer (AE) is responsible for
the development of specifications for the site health and safety plan
and for the description of minimum requirements for health, safety,
and emergency response during the remedial design. An estimate of
increases hazards over background and the degree of existing
hazard should be specified in the remedial design. During the
design phase, it is the responsibility of the AE to establish
boundaries where 29 CFR 1910.120 applies. The criteria used in
such determination should include remedial investigation data and
the Agency for Toxic Substances and Disease Registry (ATSDR)
Health Assessment.
(3) Construction Manager. The construction manager, usually USAGE,
BUREC or an ARCS contractor under a contractual or interagency
agreement with EPA, or the oversight official for responsible party
remediation, oversees the remedial design and remedial action
health and safety programs. During design, specification, review
and acceptance of the health and safety plan (and program) is a
construction manager/oversight official responsibility. During
763
-------
remedial action, the construction manager/oversight official verifies
compliance with the health and safety plan and with the health and
safety provisions of site-specific contracts. The construction
manager has the authority to suspend unsafe operations and to
require modifications to health and safety plans. Results of
inspections/oversight are reported to the RPM.
(4) Prime Contractor. Implementation of the Health and Safety
Program is the responsibility of the prime contractor for both fund
and enforcement lead projects. The prime contractor's HASP is
mandated by OSHA and/or the construction contract as the legally
enforceable plan on a Superfund site.
(5) Subcontractors. Although subcontractors are responsible for the
health and safety of their own employees, they should structure
their health and safety plans to smoothly interface with the prime
contractors overall site HASP. The prime contractor will review
and approve the subcontractor's HASP (note: the subcontractor's
HASP will have the prime contractors HASP incorporated into it).
b. ERT Fact Sheets. ERT Fact sheets are discussed in other papers from this
session. Areas discussed, in detail include OSHA-EPA relationships,
worker training, the site HASP, and the EPA Health and Safety Program.
4. Emergency response. Most sites are too small to warrant fully staffed on-site
medical and firefighting facilities. Where services can be provided by
surrounding communities, EPA may provide limited training and support to assist
the local community in providing OSHA response specific to hazardous waste, on a
case-by-case basis. An issue the Task force is investigating is how to obtain
agreements early in the remediation process. A fact sheet on this subject will be
distributed later this year.
a. Service upgrades for OSHA compliance. To compensate for OSHA
requirements specific to hazardous waste training and support, EPA may
provide limited training and support to upgrade local service capabilities or,
a case-by-case basis. The amount of training and support that local
firefighting and/or emergency response personnel will require for OSHA
compliance (section q of the worker protection standard, if off-site
responders) depends on site-specific conditions (i.e. off-site training
duration can vary between 24 and 40 hours). Examples of the types of
support that may be provided by EPA to local responders on a case-by-case
basis include on and off-site training, no-cost personal protective
equipment and specialized haz-mat equipment loans, and medical
surveillance.
b. Agreements. As a minimum the emergency response plan should be a
separate section of the site HASP. Agreements, which must be made prior
to site entry are between the party responsible for the HASP and the party
providing the response services (i.e. the AE firm for design operations
involving site entry and the prime contractor for remedial action). Because
failure to secure agreements can result in remedial project delays or work
stoppage, it is important for EPA to solicit early involvement of community
764
-------
relations staff and to address emergency response through pre-design work
plans, etc. This will entail an evaluation of local fire departments,
hospitals, police departments, etc. to provide coordinated services to the
RD and RA. Selection of the provider should be based on an evaluation of
current capabilities, required support levels, response time, jurisdictional
authority, and cost to the Government. This information is often available
from information gathered as part of predesign activities.
Training. The site industrial hygienist (or equivalent position) should make
a copy of the site HASP (to include the emergency response plan) available
and provide on-site training for local firefighting and emergency response
personnel subject to respond to calls at Superfund sites.
CONCLUSIONS
Because of the complex relationships between the many parties involved in Superfund remedial
design and remedial action, health and safety roles and responsibilities are often misdirected,
resulting in ineffective or unresponsive programs. The Health and Safety Task Force is an
effective forum for resolution of issues and communications between the parties involved with
remedial design and remedial action.
DISCLAIMER
This report has undergone a relatively broad initial, but not formal, USEPA peer review.
Therefore it does not necessarily reflect the views or policies of the Agency. It does not constitute
any rulemaking, policy or guidance by the Agency, and cannot be relied upon to create a
substantive or procedural right enforceable by any party. Neither the United States Government
nor any of its employees, contractors, subcontractors or their employees makes any warranty,
expressed or implied, or assumes any legal liability or responsibility for any third party's use or or
the results of such use of any information or procedure disclosed in this report, or represents that
its use by such third party would not infringe on privately owned rights.
We encourage your comments on the utility of this paper and how it might be improved to better
serve the Superfund program's needs. Comments may be forwarded to the attention of Joseph
Cocalis, Design and Construction Management Branch, USEPA, Mailcode OS-220W, Washington
DC 20460. Mr. Cocalis will relay any comments to the attention of the Task Force, where they
will be considered and addressed.
REFERENCES
1. Establishing Work Zones at Uncontrolled Hazardous Waste Sites (in Draft # 9285.2-06fs).
2. Hazardous Waste Operations and Emergency Response: RCRA TSD and Emergency
Response Without Regard to Location (#9285.2-07fs).
3. Hazardous Waste Operations and Emergency Response: Uncontrolled Hazardous Waste
Sites and RCRA Corrective Action (in Draft #9285.2-08fs).
4. Hazardous Waste Operations and Emergency Response: General Information and
Comparison (in Draft, #9285.2-09fs).
5. Hazardous Waste Operations and Emergency Response: Available Guidance (#9285 2-
lOfs).
765
-------
Airborne Exposure Control at an Acid Sludge Remedial Site
Stephen L. Davis, CIH, CSP
IT Corporation
312 Directors Drive
Knoxville, Tennessee 37923
(615) 690-3211
Bhupi Khona
U.S. EPA Region III
841 Chestnut Street
Philadelphia, Pennsylvania 19107
(215) 597-0439
LO INTRODUCTION
Use of a real time air monitoring program including, mobile work area monitoring, and perimeter
monitoring with a centralized alarm function, as a tool to aid in suppression of site emissions is a
relatively recent approach to emission control during site remediation. Experience at this site has
provided useful information that can be applied to subsequent similar remedial projects. Please note
that the opinions in this publication are those of the authors and do not represent any official position
of the U.S. Government.
The site consisted of a lagoon which was used for disposal of sulfonated mineral oil production
wastes, motor oil reclamation wastes, coal fines, and other sludge residues from approximately 1935
to approximately 1975. In the late 1970s, part of the lagoon wall failed, allowing sludge to enter a
nearby creek. This initiated a series of responses culminating in neutralization and stabilization of
the site during 1989 and 1990.
During early stages of this effort in 1983, remedial operations were initiated and consequently
terminated due to significant release of acidic aerosols and/or vapors into the surrounding
environment. In July, 1989, remedial operations were re-initiated and involved primarily excavating
the sludge down to bedrock, mixing the sludge with lime (stabilization) to increase the pH, and
backfilling to the desired grade.
IT Corporation (IT), serving as an independent consultant to the U.S. Army Corps of Engineers
(COE), Omaha District, performed air monitoring consisting of real time air monitoring at the work
face, time-weighted average (TWA) sampling at the work face, and datalogging for six existing air
monitoring instruments at three perimeter locations. IT also provided related health and safety
consulting services. This effort was initiated by the US Environmental Protection Agency (Region
III) remedial project manager in order to apply a different technical approach in an attempt to obtain
additional information.
This publication specifically addresses the use of real time air monitoring and datalogging
instrumentation at this hazardous waste remedial site. Key topics are the airborne concentrations of
hazardous chemicals in the active work area, the use of the perimeter monitoring and logging s /stem
to track and control airborne exposures at the site periphery.
766
-------
2.0 BACKGROUND
IT Corporation was contracted during the second year of the remedial action to provide assessments
(independent of the remedial contractor) of airborne contaminant concentrations and related on site
health and safety practices. IT personnel conducted real time air monitoring for hydrogen chloride
(HC1), sulfur dioxide (SO2), and organic vapors at the work face and installed a datalogging system
to record and store input from six existing perimeter instruments monitoring hydrogen chloride and
sulfur dioxide.
Real Time Air Monitoring
The real time air monitoring for SO2, HC1, organic vapors and a variety of other contaminants was
conducted using direct reading instrumentation at four primary locations: near three perimeter
monitoring stations and at the downwind edge of the work area. The downwind edge of the work
area was assumed to be representative of the worst-case exposure hazard. Monitoring was performed
from an all terrain vehicle to enhance mobility and to transport the equipment required for the
simultaneous measurement of the three primary analytes. Results were documented on field activities
forms and real time air monitoring logs.
The air monitoring for SO2 was conducted using a battery-powered Gastech GX-82 confined space
unit, equipped with a sulfur dioxide electrochemical cell. A supplemental SO2 instrument, U.S.
Industrial Products, Model SO-261, was used for approximately one month. HC1 concentrations were
monitored using a battery-powered Sensidyne SS2000 portable toxic monitor equipped with a HC1
electrochemical cell and a SO2 scrubber to eliminate interferences due to cross-sensitivity to the two
contaminants. Organic vapor concentrations were monitored using a battery-powered Century
Systems portable Organic Vapor Analyzer (OVA), model OVA-128. The direct reading instruments
were calibrated daily using the manufacturer's recommended procedures. Drager detector tubes were
used, when possible, to confirm elevated measurements obtained from direct reading instruments or
to investigate other potential contaminants not detectable by the instruments.
Perimeter Air Monitoring
A data acquisition system (logger) was installed to record measurements from existing HC1 and SO2
instruments located at the three perimeter air monitoring stations. The system consisted of an 8
channel analog connection board and a Toshiba portable computer, model 3200. The data logger
recorded instantaneous readings at 20 second intervals, daily average readings and daily minimum and
maximum readings. The system also provided high and low level alarms and constant display of
readings. These data were stored on the portable computer hard disc and backed up on 3.5 inch discs.
Data logger channels 1 and 2 recorded measurements for HC1 and SO2 instruments, respectively, from
air monitoring station 1 located on the northwest perimeter of the exclusion zone; channels 3 and 4
recorded measurements for HCL and SO2 instruments, respectively, from station 2 located on the
northern perimeter of the exclusion zone; and channels 5 and 6 recorded measurements for HCL and
SO2 instruments, respectively, from station 3 located on the southeast perimeter of the exclusion zone.
M DISCUSSION
Work Face Air Monitoring Results
In general, HC1 was not detected. Readings of 2 to 7 ppm were recorded on 2 days. During these
measurements the SO2 monitor detected elevated concentrations of SO2. Because the HC1 sensor is
cross sensitive to SO2, any leak in the scrubber attachment would cause the instrument to read
767
-------
positive in an atmosphere containing SO2. This was confirmed by using the instrument without the
scrubber for one day. During this day, the sensor repeatedly read in excess of 10 ppm. The detector
tube tests for HC1 were negative, except for one at 0.5 ppm. In summary, these data indicate that
HC1, if present, existed at very low concentrations.
Sulfur dioxide was detected regularly throughout the project. Peaks in excess of 100 ppm, the
National Institute for Occupational Safety and Health's immediately dangerous to life or health value,
were detected in the work area. In addition, airborne concentrations of SO2 in the work area often
appeared to average in excess of the Occupational Safety and Health Administration (OSHA)
permissible exposure limit of 2 ppm. Please note that accurately determining a work place average
was difficult due to accessibility challenges. Drager tube tests for SO2 were generally positive,
although not in exact agreement with instrument readings.
Elevated SO2 readings were strongly associated with the disturbance of black sludge material, in
excavations and in the mixing pits. Elevated readings occurred throughout the project. However,
readings decreased as the sludge excavation and stabilization was completed, and at the project's end,
SO2 was consistently not detectable.
Table 1 presents a summary of SO2 readings. The "site activity" column addresses remedial activities
being performed on-site. The information presented on remedial activities is minimal. If these
activities cannot be clearly determined from daily logs, the entry of Unknown appears;. The
maximum peak reading for each day is also presented. The final column provides a qualitative
estimate of exposure. Entries are, Light (average of recorded readings is less than 5 ppm), Moderate
(average of recorded readings is 5 ppm to 10 ppm), and Heavy (average of recorded readings is
greater than 10 ppm). The assessments in this column are subjective in that they are influenced by
a number of factors, such as the time spent on-site, the distance from the source, etc. This column
is included only to provide a rough estimate of conditions and is not a quantitative measurement.
In general, organic vapor readings were equal to offsite background during the entire project.
Occasional readings of 5 to 10 ppm were obtained. However, these readings were generated from
vehicle exhaust, rather than site contaminants. The source was confirmed by conducting repeated
tests which tracked readings to vehicle exhaust.
Perimeter Results
The perimeter air monitoring results are summarized in Table 2 and illustrated using sample graphs
in Attachment 1. These graphs represent days during which perimeter sensors measured relatively
heavy off gassing. Because both HC1 and SO2 perimeter sensors were calibrated to SO2 and no HC1
was detected in the work area, these data are presented as SO2 concentrations. The data log graphs
display readings which represent SO2 concentrations and time. All daily averages were less than 2
ppm. Maximum (peak) readings ranged from the same as averages to a high of 14 ppm. Any logger
measurement greater than 10 ppm is, however, suspect, as the scale of the perimeter monitors was 1-
10 ppm. There is no verification that the voltage at the remote connection is linear when the
instrument meter exceeds full scale.
The results indicated daily maximum SO2 concentrations increased for most work shifts during mid-
September through October. The frequency of occurrence of peak SO2 readings also increased during
this time period.
A number of factors may have influenced perimeter monitoring instruments and the data logger, such
as cross sensitivity of the sensors to the two contaminants, calibration of instruments and weather
conditions (temperature, humidity, etc.). The effects of these factors are discussed below:
768
-------
(1) Perimeter monitoring instruments occasionally failed during active work shifts. When this
occurred, no electrical signal would be generated in the corresponding sensor wiring until the
sensor was replaced or repaired. In some cases, a failed sensor would be replaced or serviced
within minutes. In other cases, the failed instrument might remain on-line due to lack of an
immediately available replacement sensor. For the most part, failed sensors were replaced
within 30 minutes following failure.
During normal functioning, the perimeter monitors generated 1-5 volts at the remote sensing
jack. This voltage corresponded to instrument readings of 0-10 ppm. When an instrument
failed or was disconnected, no voltage existed in the sensor leads. The datalogger was set so
that zero voltage was interpreted as a negative reading. This setting allowed the low end
alarm to trip so that a failed instrument would not go unnoticed. The negative scale limit was
approximately minus (-) 2.5 ppm. Thus, during any period in which the sensor was
disconnected, the logger recorded a reading of approximately -2.5 ppm. This reading would
be included in the daily average, thereby erroneously decreasing it.
(2) It is normal for perimeter sensors of the type used in this project to exhibit drift or change
in readings over a period of time. This drift is related to the sensor type and conditions of
use. Standard quality control practices generally require calibration adjustments at intervals
of appropriate duration to provide the desired accuracy.
The remedial contractor reported that perimeter sensors were calibrated at weekly intervals.
Occasionally the calibration of the perimeter sensors was checked, by IT and/or the remedial
contractor, with the perimeter sensors at their field locations. This procedure tested the
accuracy of the entire remote sensing system. In this procedure, the instruments were exposed
to a test gas of 5 ppm while in place at the perimeter stations and connected to the remote
sensing system. Ideally, the instrument meters, the datalogger and the strip chart recorders
should all have read 5 ppm.
The majority of "field" checks resulted in instrument and logger readings of 4 to 6 ppm, an
error range of plus or minus 10 percent of the instrument scale. However, some "field" checks
resulted in instrument and logger readings as low as 0 ppm and as high as 10 ppm. These data
indicate that a shorter calibration frequency would be more appropriate.
(3) Perimeter sensors often exhibited daily zero drift that appeared to be temperature dependent.
This drift took the form of a gradual decrease in the baseline reading, during the morning,
with minimum readings occurring in the early afternoon. The baseline readings began a
gradual increase from 1500 to 1700 hours and would typically return to approximately the
original reading by the end of the day. The typical drift was approximately 0.5 ppm. This
drift could not be observed during days in which significant off-gassing was detected, since
the peaks masked any drift.
(4) Transmission from nearby (1-3 feet) hand held radios was observed to cause false readings
on the datalogger. Hand held radios were frequently used near the datalogger to communicate
with field crews during off-gassing and subsequent suppression activities. The magnitude and
direction of radio-induced deflection varied with the type and individual radio units. Radios
at 456.800 MHz caused a negative deflection of 0.3 ppm, radios at 136.4125 MHz with a
private line tone (sub-audible) of 4 Z (136.5 Hz) caused a positive deflection of 0.18 ppm.
Each transmission lasted approximately 5-20 seconds. The number of individual transmissions
during an off-gassing event was varied and is estimated to have been 5-20 transmissions over
769
-------
a 5 minute time period. Due to the magnitude of the deflections in logger readings, the effect
of radio transmissions was probably minimal.
4.0 CONCLUSIONS
Use of a real time air monitoring program including, mobile work area monitoring, and perimeter
monitoring with a centralized alarm function, as a tool to aid in suppression of site emissions is a
relatively recent approach to emission control during site remediation. Experience at this site has
provided useful information that can be applied to subsequent similar remedial projects.
4.1 Supplied air (Level B) personal protective equipment for on-site workers was appropriate in light
of the work area SO2 readings.
4.2 Real time monitoring at the work face can be used as a first indicator of unacceptable off-
gassing if it is possible for the operator to maintain an appropriate location relative to the emission
source. Specific recommendations for this activity include:
Radio communication with the control center,
An all terrain vehicle,
Outer instrument cases which can be kept closed during instrument operation,
Instruments with adjustable, audible and visual alarms, and
Rugged instruments which will operate accurately under adverse conditions.
4.3 Real time perimeter monitoring with an alarm system can be an effective tool in the control of
airborne emissions that pose a potential risk to off-site receptors. Appropriate installation offers the
following:
Instant, unattended alarm function when preset concentrations are exceeded,
Instant, high resolution measurement of elevated readings,
Instant alarm notification of major sensor failure or disconnection,
Instant notification of the effect of emission suppression activities, and
Verification of on-site sensor calibration.
Please note that this type of air monitoring application is limited to contaminants that can be
measured on a real time basis. Unfortunately, there are numerous contaminants that cannot be
measured using this method.
4.4 Of the monitored airborne chemicals, SO2 was the only agent detected in the work area at a
concentration approaching or exceeding the PEL or TLV.
4.5 A number of recommendations can be made for future similar projects based on lessons learned
at this project. The perimeter air monitoring program should include the following:
A computerized, centralized system with continuous display and adjustable high and
low alarms for each channel,
High alarms set at suppression concentrations, low alarms at readings that will ir dicate
sensor disconnection or failure,
Perimeter sensors calibrated in place (at the point of use) on a daily basis or on a cycle
proven to minimize inter-calibration drift, and
Monitoring stations sheltered from direct sunlight and environmental extremes where
possible.
770
-------
5.0 ACKNOWLEDGEMENTS
The authors would like to thank the following U.S. Corps of Engineers personnel:
Larry Janis
Sandra Cotter
Gregory McCleaf
Nancy Flaherty
James Thornton
The authors would also like to thank Melissa Smith for her invaluable technical support and James
Bolden, Mark Brown, Phillip Mitchell and Lawrence Webster for the field work which made this
publication possible.
771
-------
Table 1: Real Time Monitoring for Sulfur Dioxide At The Work Face
Date Site Activity
7/26 Stabilization & bedrock neutr.
27 Unknown
30 IT worked on datalogger system
31 IT worked on datalogger system
8/1 IT worked on datalogger system
2 IT worked on perimeter system
3 IT worked on perimeter system &
health and safety issues
6 Stabilization
7 Stabilization
8 Stabilization
9 Stabilization
10 Moving stabilized material
13 Moving stabilized material
14 Stabilization
15 Stabilization
16 Stabilization
17 No stabilization
20 No stabilization
21 No stabilization
22 IT worked on datalogger system
23 No stabilization
24 No stabilization
27 Stabilization
28 Unknown
29 No stabilization
30 Unknown
31 IT participated in meetings
9/1 Unknown
4 Unknown
5 Unknown
6 Stabilization & spreading
7 Maintenance
10 Stabilization
11 Stabilization
12 Stabilization
13 Stabilization
14 Unknown
15 Unknown
17 No stabilization
18 Stabilization
19 Unknown
20 No stabilization
21 Stabilization
22 Stabilization
24 Stabilization
25 Unknown
26 Unknown
27 Unknown
Maximum Peak
4 ppm
17 ppm
No monitoring
No monitoring
No monitoring
No monitoring
No monitoring
Average
Light1
Moderate2
>100 ppm
>100 ppm
>100 ppm
38 ppm
0 ppm
No monitoring
0 ppm
8 ppm
18 ppm
1 ppm
0 ppm
0 ppm
No monitoring
0 ppm
0 ppm
0 ppm
14 ppm
0 ppm
>100 ppm
No monitoring
0 ppm
0 ppm
0 ppm
39 ppm
No monitoring
30 ppm
10 ppm
>100 ppm
30 ppm
1 ppm
Eqpt. failure
No monitoring
10 ppm
1 1 ppm
2 ppm
24 ppm
43 ppm
>100 ppm
38 ppm
1 ppm
>100 ppm
Heavy3
Heavy
Moderate
Moderate
Light
-
Light
Light
Light
Light
Light
Light
-
Light
Light
Light
Light
Light
Heavy
-
Light
Light
Light
Moderate
-
Light
Light
Heavy
Light
Light
-
-
Light
Light
Light
Light
Heavy
Moderate
Moderate
Light
Heavy
772
-------
28 Unknown
29 No intrusive work
10/1 IT instrument failure
2 Unknown
3 Stabilization
4 No stabilization
5 Stabilization
6 Unknown
7 Unknown, IT in meetings
8 Unknown, IT in meetings
9 Stabilization
10 Stabilization
11 No stabilization, ATV broken
12 No stabilization, ATV broken
15 Stabilization
16 Breaking sludge
17 Stabilization
18 Intrusive work stopped 1030
19 Stabilization
20 Stabilization
21 Stabilization
22 Stabilization
23 Unknown
24 Unknown
25 Stabilization
26 Unknown
29 Unknown
30 Stabilization
31 Moving stabilized material
11/1 Moving stabilized material
2 Moving stabilized material
3 Moving stabilized material
1 ppm
2 ppm
No monitoring
10 ppm
18 ppm
No monitoring
25 ppm
>200 ppm
No monitoring
8 ppm
>200 ppm
45 ppm
No monitoring
No monitoring
105 ppm
70 ppm
45 ppm
No monitoring
30 ppm
57 ppm
64 ppm
34 ppm
3 ppm
37 ppm
5 ppm
0.5 ppm
12 ppm
4 ppm
0 ppm
0 ppm
0 ppm
3 ppm
Light
Light
Light
Light
Moderate
Heavy
Light
Heavy
Heavy
Heavy
Heavy
Moderate
Heavy
Moderate
Moderate
Light
Light
Light
Light
Light
Moderate
Light
Light
Light
Light
Light
Light = average of recorded readings is less than 5 ppm
Moderate = average of recorded readings is 5 ppm to 10 ppm
Heavy = average of recorded readings is greater than 10 ppm
773
-------
Table 2: Perimeter Air Monitoring Results in Parts Per Million
Date Station One
HC1 SO2
07-31-90
08-01-90
08-02-90
08-03-90
08-06-90
08-07-90
08-08-90
08-10-90
08-13-90
08-14-90
08-15-90
08-16-90
08-17-90
08-20-90
08-21-90
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
0.218666
0.228545
0.451147
23.2559
-0.11792
0.028976
0.443268
0.639779
0.484450
0.64539
0.544382
1.05736
0.533688
5.61039
0.538808
0.635496
0.540385
0.566658
0.548493
0.573434
0.453457
4.94167
0.467520
0.576578
0.498662
0.58538
0.568870
0.581381
0.491828
0.558821
0.475263
0.617653
0.428022
6.79416
-0.21958
0.001274
0.337027
0.695016
0.418807
0.808455
0.475654
1.17624
0.445015
5.64917
0.436993
0.55897
0.507235
0.542068
-0.27696
0.493171
0.207691
5.41362
0.451739
0.998285
0.450805
0.561429
0.560707
0.573418
0.550301
0.576066
Station Two
HC1 SO2
-0.02653
0.007459
0.181675
4.84107
-0.21878
0.020885
0.168542
10.4676
-2.11133
0.574055
1.292675
1.46763
0.464795
5.39973
0.312858
0.625608
0.423118
0.499985
0.525347
0.546927
0.458411
3.88458
0.373333
0.629004
0.284010
0.524528
0.630285
0.694142
0.675476
0.691499
0.612320
0.632181
0.007628
10.3078
-0.37336
0.019380
0.181443
5.27453
0.241870
0.759921
0.293881
0.376954
0.242475
4.01638
0.151392
0.327764
0.220187
0.260167
0.426712
0.451359
0.385293
2.33972
0.341133
0.530362
-0.20471
0.468853
0.505975
0.560074
0.610965
0.626207
Station
HC1
1.208898
4.14935
0.911855
5.26882
-0.12829
5.15901
0.052394
0.26115
0.126346
1.067
0.248152
8.22274
0.125985
5.66539
-0.16118
0.116236
-0.03552
0.003776
-0.00555
0.454662
0.593955
3.78823
0.572897
0.762311
0.570381
0.66691
0.653720
0.6629
0.594187
0.600454
Three
SO2
1.061149
13.4155
1.084212
11.1299
-0.34056
''.43604
0.194220
0.709795
0.377092
2.35527
0.638332
9.42208
0.390085
7.34186
0.434448
0.873183
0.565080
0.655714
0.606832
1.68909
C.454818
4.26275
0.344434
0.748656
0.370274
0.680923
0.670684
0.722861
0.699200
0.741739
774
-------
Table 2: Perimeter Air Monitoring Results in Parts Per Million (Continued)
Date Station One Station Two Station Three
HC1 SO2 HC1 SO2 HC1 SO2
08-23-90
08-24-90
08-27-90
08-28-90
08-29-90
08-30-90
08-31-90
09-01-90
09-04-90
09-05-90
09-06-90
09-10-90
09-11-90
09-12-90
09-13-90
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
0.167908
0.433813
0.369186
0.436606
0.361122
0.462934
-0.07811
0.708685
0.582038
5.83666
0.503138
0.814141
0.533452
0.571578
0.377099
0.492256
0.613669
1.25812
0.626169
0.630534
0.599665
1.21665
0.496377
0.689274
0.596304
1.01667
0.629796
1.63974
0.211830
1.63061
0.715023
0.734829
0.487178
0.526338
0.429385
0.547596
0.229015
1.24318
0.374222
5.17254
0.341401
0.817908
0.266057
0.324318
0.327884
0.477168
0.565861
0.965062
0.609145
0.612388
0.536501
1.27323
0.581192
0.671465
0.570512
1.0408
0.593574
2.20533
0.587735
2.69881
0.694002
0.702448
0.649511
0.689921
0.378752
0.675072
0.678973
11.3162
0.591518
5.82542
0.583764
0.856212
0.270274
0.562824
0.272388
0.572793
0.724469
8.91864
0.433282
0.477095
0.311532
4.29643
0.616805
1.18834
0.242190
2.2842
0.387854
2.98588
0.622905
7.14937
0.613176
0.633248
0.359173
0.637136
0.180582
0.865464
-0.22509
11.2104
0.335953
4.77801
0.327647
0.582444
0.188341
0.255483
0.252883
0.523614
0.664847
11.7645
0.238900
0.239699
0.165086
3.44505
0.429819
1.25237
0.270041
7.21022
0.248828
1.31325
0.315686
4.93893
0.587820
0.602651
0.560531
0.584771
0.538100
0.939162
0.093500
1.66008
0.569180
1.03548
0.909965
11.0752
0.407468
0.537233
0.458980
0.546564
0.486005
1.2595
0.562125
0.593961
0.946363
1.98293
0.153495
0.543277
0.366015
2.96037
0.441124
3.20235
0.447746
4.92422
0.663353
0.684825
0.546370
0.625827
0.496318
1.59305
0.601037
1.55757
0.561076
1.6067
0.894283
11.0788
0.513857
1.17897
0.548537
0.794785
0.706176
1.67074
0.742994
0.765523
0.905739
1.6339
0.157642
0.297858
0.318243
2.4897
0.299024
2.14476
0.056875
4.98306
775
-------
Table 2: Perimeter Air Monitoring Results in Parts Per Million (Continued)
Date Station One Station Two Station Three
HC1 SO2 HC1 SO2 HC1 SO2
09-14-90
09-15-90
09-17-90
09-18-90
09-19-90
09-20-90
09-21-90
09-22-90
09-27-90
09-28-90
09-29-90
10-01-90
10-02-90
10-03-90
10-04-90
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
0.632037
1.3665
0.678694
0.689358
0.658989
4.17587
0.567901
0.979477
0.445739
0.679667
0.319884
0.625353
0.502197
0.592521
0.392080
0.970332
0.322738
8.22386
0.160800
0.466969
0.509437
0.934412
-0.03679
0.086942
0.451966
0.82033
0.329090
1.23255
0.433147
0.538248
0.601776
0.665422
0.660285
0.679499
0.691291
3.92223
0.564394
0.768038
0.592316
0.889377
0.529228
0.577498
0.523810
0.627136
0.529538
1.20346
0.305549
6.63425
0.227759
0.570098
0.725175
0.835951
0.764152
0.926013
0.779963
0.834333
0.720115
0.878661
0.684140
0.765003
1.011046
10.2855
0.475876
0.517087
0.457825
3.02186
0.620693
2.50606
0.851918
8.01287
0.637655
0.670237
0.643043
1.98722
0.643417
1.78366
0.613436
10.9986
0.636793
10.7084
0.571423
0.701654
1.338748
11.5491
0.450112
1.98199
1.272049
10.9073
0.844350
11.2116
0.565748
10.3379
0.238004
0.254694
0.876272
5.31377
0.310657
1.06799
0.501377
9.12265
0.381250
0.421774
0.428452
1.00473
0.376264
1.62171
0.599032
12.6445
0.439748
11.5001
0.490644
0.693644
1.231777
11.6919
0.112906
0.852321
1.202671
11.3207
-1.02877
3.67831
0.465359
0.758623
1.134496
2.79069
0.457238
3.71701
0.612096
8.59351
0.801310
11.1004
0.540646
0.599124
0.692534
4.23503
1.225300
10.0934
0.540003
4.74135
0.748594
8.792481
0.438842
0.669437
0.455651
8.96364
0.662610
4.10752
0.015805
2.37276
0.556522
0.679064
0.410863
0.474119
1.166132
2.97085
0.444048
1.16935
0.701284
8.03739
0.917769
10.3246
0.698602
0.306667
0.715351
4.67376
1.289720
1 1 .642
0.571834
4.61632
0.783730
10.8006
0.711938
1.08464
0.791022
10.5988
0.896587
4.3711
O.C55366
1.96524
0.644311
0.901597
776
-------
Table 2: Perimeter Air Monitoring Results in Parts Per Million (Continued)
Date Station One Station Two Station Three
HC1 SO2 HC1 SO2 HC1 SO2
10-05-90
10-06-90
10-07-90
10-08-90
10-09-90
10-10-90
10-11-90
10-12-90
10-15-90
10-16-90
10-17-90
10-18-90
10-19-90
10-20-90
10-21-90
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
0.337549
2.58424
0.144389
6.04446
0.203678
4.24588
0.216242
4.95017
0.369269
0.490225
0.409392
0.603731
0.642055
0.692225
0.574869
0.597795
0.356748
0.571927
0.493103
1.0705
0.441467
5.0764
0.509233
0.613982
0.590917
0.772166
0.485579
1.15056
0.069186
8.18505
0.592207
2.43555
0.552109
9.89101
0.667160
0.76933
0.548916
6.19211
0.389111
0.708028
0.638191
0.813129
0.735746
0.849916
0.709460
0.718229
0.724645
0.844392
0.766151
1.77038
0.647236
4.94936
0.711977
1.03719
0.782786
1.03627
0.522958
0.792444
0.847718
5.9021
0.533152
5.95856
0.854951
11.4171
0.307422
0.459641
0.584957
6.3775
0.651374
10.7252
1.263770
10.6528
0.525351
1.19827
0.297965
0.317629
0.271591
1.36599
0.563084
5.3952
1.178451
13.0279
0.935273
13.6996
-0.30333
0.619428
1.318833
10.8837
0.458957
12.5905
0.273072
3.33339
0.539427
12.23
0.848566
0.9053
0.417435
5.19925
0.824325
11.044
0.820481
10.1462
-1.29730
1.17865
1.177523
1.18569
0.630567
1.35353
0.615993
2.23998
0.359008
11.3586
0.515698
1.52486
0.654874
0.870908
1.701023
12.0178
0.962250
13.8455
0.482702
1.93792
0.461862
0.598889
0.469162
0.56014
0.246155
5.35341
0.163596
0.241926
0.291845
0.510044
0.983460
4.98269
0.648506
1.54515
1.315751
4.16883
0.786316
3.23031
0.286432
6.79763
0.084801
0.196473
0.288101
2.29855
0.387232
1.66324
0.566888
8.07674
0.547457
2.05161
0.536748
0.710884
0.447423
0.639269
0.662276
7.59283
0.193971
0.325377
0.304925
0.811593
1.143544
6.3925
0.875071
1.81553
0.868958
7.18853
0.581671
3.77464
0.511471
6.97684
0.208100
0.370799
0.488409
6.61175
0.745319
3.3339
0.394888
3.03305
777
-------
Table 2: Perimeter Air Monitoring Results in Parts Per Million (Continued)
Date Station One Station Two Station Three
HC1 SO2 HC1 SO2 HC1 SO2
10-22-90
10-23-90
10-24-90
10-25-90
10-30-90
10-31-90
11-01-90
11-02-90
11-03-90
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
Avg
Max
0.396159
4.45959
0.525435
0.632874
0.250841
1.53931
0.300442
0.719591
0.206132
0.792664
0.166417
0.302162
-0.64198
4.70063
-0.56463
-0.28445
-0.59535
-0.34578
0.717306
6.21439
0.760732
0.969035
0.508911
0.719265
0.541110
1.08792
0.484855
0.625406
0.475593
0.521282
0.023151
4.82902
0.084276
0.254457
0.019943
0.24595
1.041768
11.0256
0.437081
0.663212
0.782840
14.4231
0.709762
1.23537
0.357939
1.31845
0.395247
3.87723
0.088870
5.06107
0.062430
0.282891
0.052388
0.250881
0.361557
11.0157
-0.68047
0.33695
1.022063
14.2214
-0.32901
1.6382
0.626638
2.00388
0.471956
2.17151
0.439167
5.08053
0.473396
0.563062
0.443151
0.554115
0.445705
4.8817
1.038122
11.8846
0.488120
5.31435
0.689267
5.40751
0.232645
0.579698
0.376058
0.634779
0.010480
4.76726
0.083017
0.404541
-0.06792
0.341658
0.554291
4.9769
1.391161
13.4662
0.775249
12.5049
1.067826
8.46791
0.257112
1.23587
0.211860
0.619953
0.146103
7.97325
0.184529
0.425997
0 159726
0.409535
778
-------
CD
a
5 -
evi n
« to
o
0)
o
6
O
CO C/3
en 33
C
CM C/J
eo r\i vo ao
^ rj in co
v o in o
to H r- in
o en r» •
f- r» • o
• • o 1-1
o o i
a\ in
f-» vo
r~ in
vo in
to vo
H r>
vo vo
O O I
CO
o
09
if) r>
•
O
• o\ o\
o\ H
f^ -H
•/: < W < £
SONIQV3U
779
-------
5
a
5
i- CM n
O
cr>
C7>
o
•" P
o in
* in
in
o
oo
H
00
o
in
2 •*
JJesl 3= eo
S •
~ O
- in
,si
o
H CM r»
r» vo r*
en vo n
n co in
en CM CM
•H o n
• • •
000
vo r* vo
en o IN
in «t ov
n vo .-i
vo o *r
i H . CM
• o •
o i o
in en *•
CM n ^-
n o o
*r o •
CM en H
eo r> H
• •
o o
V O CM
r» en in
n in CM
H oo r»
in H •
vo • o
• o H
o I
H H DO
H CO CM
iH * O
O> 1^ DO
co o o
n o t»
• • • •
o o o o
CO
n
OV
o
. M
*j ^ m
« 2 *>
u 5 w
w <
O> tM n
vo rt fM
CM VO O)
en o o
vo i-( o»
O (N «t
• • • •
o o o o
4J
a
c x
•H a
< E :t
N «• O
SONIQV3M
780
-------
a
o
a
UJ
I]
§
,- CM n
O
O)
CO
6
8
8
CO
8
o
o
o
c\i
co
8
CO
o.
LU
H in ^
t^ en xo
vo
o o
i n o co
en ^r en
en IN en
in o n
• tH CO CM
I XO O .
. . f\)
o o
•f i-4 CN
co o in
o xo en
rt rH n
vo r~ .
in CM in
o o
H vo co
in H n
•-(no
vo CM t^
vo r-t r--
r» r~ .
000
g
n H in
OHO
•H VO f-
n in o
en r\i .
•* n H
o o
at 10 tT C X
e v > -H ia
< w < S S
t- o T
8
ctl
8
S9NIOV3H
781
-------
UJ
o
i- CJ
I I
* in CO
0
O)
00
T^*
1
o
Y™*
_+
^
'
ff
"V
- .- .
.V.-VM^
«,
...*^>"
•-i
"";.'.'"*•"- IT.
-^"'•A'1
t "r.vlrt;-
•" mtff ;•.«,—.. '
."."'* '
j^aO.WKsST"""'"
_. .u.". "
"... ««/.•:.
" "~ '"" :_ :^.....^
ii
ij
1 1 1 I 1 1 I i I 1 1 1 I
W N ^ O ™ *0 ^ (O if) ^f {ij ^| v« |
&
a «-
e
0
_ o
o
N
»•
_
N
v
_ ^
n
^
«•
•t
N
N
O
»•
«9
P>
m
«
b
*•
^%
_
« N
^ *™
. . i
« N
" O «M
6 «
" O
<•*
Ul
5
0 —
o i-
-s
d>
o
N
IF-
K^
- o
£i
o
^
M
OD
- «
o
o
n o t» t^
^i o co r» *
u en *s- UD to
r-> cc 10 co r» en
O O . H
• • O •
O 0 1 O
en eo '(• *o
CM o en ci> oo
o •«* « N »
CM WJ H in OS N
en IH e» in
in in n •
. . . H
o o n
H n in vo
•or r~ rg en
•-< en CM 05 en
w H in 05 \o
^ * •* n \o •
• en N n
O • • H
0 CJ
in r- rr en
n r> vi) H
CM.O en o r»
. ° H H o) n
O iH O O
r^ t^ «) .
• • • iH
O 0 0
in n cu CN
en n n co
,_, eo CN M en
rj en en H D
^H PC •* O 1*. H
• in n vo
^3 t in •
o o o
4J
a
c C X
« « jj > ^ (0
" S w < S: S
UJ
-------
O
0)
6
CM
CM
3S
o
CSCO
a\ H 01
H o> n
n n n
in o> o
* r-« .
r- o n
o o
CM CM *
n oo CM
CM n n
r- vo vo
oo CM vo
n • •
• o H
O I
n H oo
CM vo r-
o m H
H v o
O CD •
r- T- CM
• • H
H O
n CM c~
n c^ m
oo in oo
co n co
H in •
n rf o
OHO
n
cijvo
o CM
•-1 OO CO
H
in
in
•si
in
00 H *
in o> •»
Ol H *»
CM H CM
CN CM O\
in • r^
. o •
o i o
w ^ vo
r- oo in
in n o
in T in
00 CM H
**• • •
• O H
o o
0 •" u
•rl S**1
u^ M
n) ce (0 O1 C X
C4J >-H (0
P) CM -t-
SDNIQV3H
783
-------
Q
2
§
UJ
^
P""
C/3
UJ
I]
*
O '
i i
i i
i i
i i
i i
i j
i i
i !
i
i i
- CM w * tn «o
O
o>
<§•
in o H n
o to in in
o eo n o\
in H ri 01
H H n H
VO (N • VO
. . O •
O O I O
(N co o at
en in eo r-
en o in r>
N vo in *
^ r» o n
• n • \e
o • o •
O I O
O) VO « H
*• in o in
n en in H
co H u> r*
* r- r- H
• •«* o •
O • • (N
o o
en r* * n
,.4 VO •* UI fM
u n n n r»
CM 33 IN in si r*
r- en o eo
n n H •
• • • n
o o o
CM
O
CO
r» n t»' IN
in ci o co
in in CD M
o in «e H
a\ r» 01 N
•»f ^- n in
• • "i •
0000
vo r* cci CM
("I rH ffi VO
•-1 o» if co H
u 10 vo in N
s in «o o o
cs H • n
« • o •
O O I O
C 01
O 4-1
•H >-,
4-1 rH
ID IS
4J C
o>
a
4J
en
IT c
SONIOV3U
784
-------
An Overview of the
NIEHS Superfund Worker Education and Training Grant Program
Denny Dobbin
Joseph T. Hughes, Jr.
National Institute of Environmental Health Sciences
P.O. Box 12233 MD 18-02
Research Triangle Park, NC 27709-2233
(919) 541-0752
Joyce Reimherr
Katherine Roberts
National Clearinghouse on Occupational and Environmental Health
c/o Workplace Health Fund
815 Sixteenth Street, NW, Suite 301
Washington, DC 10006
(202) 842-7833
INTRODUCTION
The Superfund Amendments and Reauthorization Act of 1986 (SARA) authorized an assistance
program for training and education of workers engaged in activities related to hazardous waste
removal, containment and emergency response. Grant recipients must be non-profit organizations
with demonstrated access to appropriate worker populations and experienced in implementing and
operating worker health and safety education training programs. The National Institute of
Environmental Health Sciences (NIEHS) was given responsibility for establishing and managing this
program.
The scope of training in this area is great since the United States is a major producer of hazardous
materials and waste. The Environmental Protection Agency (EPA) estimates that 57 million metric
tons of hazardous wastes are produced each year. In addition, the Occupational Safety and Health
Administration (OSHA) estimates that 13,600 spills of hazardous materials occur annually outside
fixed facilities and 11,000 spills occur annually within fixed facilities. An estimated 1.2 million
workers are involved with such uncontrolled hazardous material clean-up and emergency response.
During the first three years of the NIEHS Superfund Worker Training Program, the eleven initial
grantees developed curriculum and started training programs throughout the country to help
employers meet OSHA training requirements under 29 CFR 1910.121, Hazardous Waste Operations
& Emergency Response. Over 6,340 safety and health training courses have been delivered to the
target populations identified by Congress reaching approximately 154,241 workers involved in
hazardous waste operations and emergency response. This resulted in almost 3 million contact hours
of classroom presentations and hands-on field exercises.
During 1990, Congress significantly expanded the NIEHS worker training program by allocating an
additional $10 million to support worker training activities. After soliciting new applications through
a December 1989 Federal Register announcement, the NIEHS received 41 new applications with
combined budget requests totalling over $44 million. After a lengthy review by committees of outside
experts and other federal agencies, the NIEHS announced ten new awards in September 1990,
including 5 existing grantees and 5 newly-supported organizations. There are now over 60 individual
institutions in this program. This new support expands the scope of NIEHS-supported training to
785
-------
include workers involved in generating and transporting hazardous materials and wastes, oil spill
cleanup workers and workers involved in the cleanup of nuclear weapons facilities.
BACKGROUND
Hazardous waste workers include workers at active and inactive treatment, storage and disposal sites,
hazardous waste clean-up sites, and emergency response personnel. In addition to actual site workers
and managers, Federal, state and local personnel may be involved with site investigation and remedial
action.
Of the various sites, those involved with hazardous waste clean-up and remedial action pose i:he most
severe health and safety concerns. These sites are characterized by the large variety and number of
substances present, unknown substances and general uncontrolled condition of the site. Among the
many potential hazards at these sites are:
1) Chemical and radiation exposures
2) Biological hazards
3) Fire and explosion hazards
4) Safety and electrical hazards
5) Heat stress and cold exposure
6) Oxygen deficiency and confined spaces
An important component of health and safety programs for hazardous waste workers is appropriate
health and safety education and training. The Superfund Amendments and Reauthorization of 1986
contains important occupational health and safety provisions which address these needs. Seciion 126
required the Occupational Safety and Health Administration to promulgate interim (in 60 days) and
final (within 1 year) standards for the health and safety protection of employees engaged in hazardous
waste operations. These standards must address the following worker protection provisions:
1) Site Analysis
2) Training
3) Medical Surveillance
4) Protective Equipment
5) Engineering Controls
6) Maximum Exposure Limits
7) Information Programs
8) Handling
9) New Technology Program
10) Decontamination Procedures
11) Emergency Response
A minimum level of training for hazardous waste workers and supervisors is specified in Section
126(d). General site workers are required to receive a minimum of 40 hours of initial instruction
off-site and a minimum of three days of actual field experience under the direction of a trained,
experienced supervisor at the time of assignment. Supervisors are required to receive the same
training as general workers and a minimum of eight hours of specialized training on managing
hazardous waste operations.
The Superfund Amendments and Reauthorization Act of 1986 established a program of grants for
training and education of workers who are or may be engaged in activities related to hazardous waste
removal, containment or emergency response. Recipients of these grants were to be nonprofit
organizations with demonstrated ability to reach target worker populations and with demonstrated
786
-------
experience with implementing and operating worker health and safety training and education
programs. Responsibility for administering this grant program was
given to the National Institute of Environmental Health Sciences (NIEHS).
The National Institute of Environmental Health Sciences is part of the National Institutes of Health
(NIH) in the Department of Health & Human Services (HHS); its mission is to support research and
training efforts which increase understanding of the relationship between environmental exposures
and human health effects and disease.
Congress authorized funds for this program for a five-year period beginning in October, 1986. Up
to $10 million may be used for this program in each fiscal year. In the 1989 appropriation the
Congress increased the program to $20 million per year for Fiscal Year 1990 and 1991.
DISCUSSION
Program Description
The NIEHS hazardous waste worker protection program sought grant applications from qualified
nonprofit organizations to develop and administer health and safety education programs for hazardous
waste workers. Target populations for this training are:
(1) Workers at active and inactive hazardous waste treatment, storage and disposal facilities.
(2) Workers engaged in clean-up or remedial action at waste sites.
(3) Emergency response personnel.
(4) State and local personnel engaged in hazardous waste site investigation, remedial action or
clean-up.
Training programs were to satisfy minimum requirements for hazardous waste workers as specified
in Occupational Safety and Health Administration (OSHA) regulations which are or may be
promulgated. Grants were made for curriculum and training materials development and support,
direct student training and support,and training program evaluation. It was intended that the grants
address all of the above elements in order to achieve a fully integrated and effective program.
Training and education programs had to address each of the following elements, at a minimum, for
all workers:
(1) Biology, chemistry, physics and nature of hazardous materials
(2) Industrial toxicology
(3) Safe work practices and general site safety
(4) Engineering controls and hazardous waste operations
(5) Site safety plans, standard operating procedures
(6) Decontamination practices and procedures
(7) Emergency procedures and self rescue
(8) Safe use of field equipment
(9) Handling, storage, and transportation of hazardous wastes
(10) Use, care and limitations of personnel protective clothing and equipment
(11) Safe sampling techniques
(12) Rights and responsibilities of workers under OSHA
In addition to the above education and training, some foremen, supervisors and other general site
workers with additional technical responsibilities were required to provide additional specific training
to include topics such as:
787
-------
(1) Site surveillance
(2) Site safety plan development
(3) Use of special instrumentation for site assessment
(4) Safe use of specialized equipment
(5) Use and decontamination of special personnel protective equipment
(6) Other topics which may be specific for a particular work site
Applicants were to have demonstrated experience with implementing and operating worker health and
safety training programs and to have the ability to reach target populations who are or will be engaged
in hazardous waste removal, containment or emergency response. A major goal of this grant program
is to assist organizations with development of institutional competency to provide appropriate training
and education to hazardous waste workers. Consortia consisting of two or more nonprofit
organizations were encouraged to apply and share grant resources in order to maximize worker group
coverage and to bring together appropriate disciplines and talents. To the maximum extent, training
programs were designed to include a mix of classroom instruction and hands-on demonstration and
instruction which simulates site activities and conditions. It was intended that offsite instruction be
supplemented with onsite training under the direct supervision of trained, experienced personnel.
Full program grants were awarded to organizations with demonstrated past worker health and safety
training and education capability and ability to reach and involve target populations. Grants were
for hazardous waste curriculum development, direct worker training, program evaluation activities,
and related support activities. Grants were made for a five year period with annual renewal based
on availability of funds, determination that grants are achieving training objectives and recipient
submission to NIEHS of copies of all training and educational materials developed under the grant.
Characteristics of Hazardous Waste Worker Training Programs
Hazardous waste worker training programs funded by NIEHS grants have the following
characteristics:
(1) Demonstrated ability to reach and involve target worker populations engaged in ha2:ardous
waste clean-up, containment or emergency response.
(2) Demonstrated past worker safety and health training and education capability.
(3) A Program Director with demonstrated capacity for providing leadership and assuring
productivity of labor education programs. The Program Director shall have responsibility for
general operation of the training program including quality assurance and program eval uation.
(4) Sufficient program staff with demonstrated training experience to assure curriculum
development, training and quality assurance. Availability of appropriate technical expertise
including but not limited to toxicologist and industrial hygienists must be demonstrated.
(5) Availability of appropriate facilities to support described education and training activities.
(6) A specific plan for preparing course curricula, distributing course materials, conducting direct
worker training and conducting program evaluations. The plans include involvement of
appropriate health and safety disciplines.
(7) A Board of Advisors or consultants representing user populations, industry, governmental
agencies, academic institutions or professional associations with interest and expertise in
worker training and hazardous waste operations. The Board is to meet regularly to evaluate
training activities and will provide advise to the Program Director.
(8) Consortia must have specific plans and mechanisms to implement the cooperative
arrangements necessary for program integration and to insure effectiveness. Specific
expertise, facilities or services to be provided by each consortium member must be identified.
788
-------
The Current NIEHS Program
In May 1987, the NIEHS awarded eleven multi-year grants to non-profit organizations for curriculum
development and initial worker training. The NIEHS grantees, which include 5 consortia of
universities and public health organizations, 5 international unions and one municipal fire department,
are charged with adapting existing health and safety information to fit the needs of a wide variety
of exposed hazardous waste workers and emergency responders across the country.
During the first three years of the NIEHS Worker Training Program (FY 1987-89), the NIEHS has
successfully supported eleven primary grantees that represent over sixty different institutions who
have trained over 154,000 workers across the country and presented 6,340 classroom and hands-on
training courses, which have accounted for almost 3 million contact hours of actual training.
Approximately 60% of the NIEHS-supported training was focused on reaching public sector
emergency responders, such as police and firefighters, who constitute the bulk of the target
population identified by Congress in Section 126 of SARA.
In response to an additional Congressional appropriation of $ 10 million for support of worker training
activities related to hazardous waste operations and emergency response, the National Institute of
Environmental Health Sciences (NIEHS) published a Federal Register notice on December 28, 1989
soliciting applications to support direct training activities by non-profit organizations targeted to
employees handling hazardous waste or responding to hazardous materials releases.
With the recent awarding of supplemental funding to five additional non-profit organizations, the
NIEHS is now supporting sixteen separate institutions and consortiums, which involve 58
organizations that are currently conducting training activities throughout the nation. NIEHS grantees
have developed curriculum which is tailored to the educational needs of each of the target populations
identified by Congress. See Appendix A for a annotated list of the current programs.
Quality Assurance
NIEHS has established stringent requirements for the development of quality, state-of-the-art
training programs by the grantees. In addition, NIEHS has pursued a rigorous quality control audit
program.
Under the OSHA standards only general criteria are provided for training and trainers. OSHA has
proposed a Training Program Accreditation Standard which will be under 29 CFR 1910.J21. A final
29 CFR 1910.121 standard is at very best over a year away and probably much more than that. In the
interim there are no criteria which permit the employer of trained personnel or government agencies
to evaluate or judge the acceptability, appropriateness, or quality of training programs, much less the
competence of those so trained. Further, annual refresher training has begun. The quality of such
training faces the very real potential of erosion to the least level of competence of refresher trainees,
a problem exacerbated by the wide range of differences in training programs being provided to meet
the initial training requirements as well as a lack of verification of basic training adequacy.
At a meeting of the NIEHS Worker Training grantees in June 1989, it was recognized that while each
grantee had developed and was delivering quality training programs, a comprehensive "Criteria for
Training Providers" was not only appropriate for these grantees but had merit in providing guidance
to other Federal agencies, State agencies, and private organizations engaged in hazardous waste
operations. As a result, an ad-hoc committee was established to consider the merits of the concept
and to develop a draft document of key issues for consideration.
789
-------
The ad-hoc committee concurred with the merit of concept, developed a draft document, and met
in early January, 1990 to refine the draft. The ad-hoc committee draft was then circulated to NIEHS,
all grantees, and a broad range of external experts for review and comment. A meeting was ihen held
in Washington, D.C. in March, 1990 of the grantees, the external experts, and several Federal agency
representatives. This NIEHS Worker Training Grant Technical Workshop resulted in a document for
use by the worker health and safety training community.
There was general agreement among the participants on a number of issues. Participants reached
agreement that: 1) the time specified for coverage of the topics required under 29 CFR 1910.120 was
inadequate to present a quality training program; 2) Emergency Response personnel should be
covered by the OSHA accreditation regulations; 3) the final OSHA regulations establishing a new
occasional worker category which would require only 24 hours of training for General Hazardous
Waste Site Operations could not be sufficiently detailed to develop a recommended guideline; 4) that
refresher training where mandated by 29 CFR 1910.120 should be covered by the accreditation
procedures and should only be delivered by training providers whose relevant core program i:> already
accredited; 5) and that hands-on training should be an essential element of the generic training
programs and should encompass at least 1/3 (one-third) of the training program hours.
Two major issues emerged during the workshop conference. The OSHA regulations under 1910.120
essentially focus upon these major hazardous materials operations categories: General Hazardous
Waste Operations, RCRA-TSD Operations, and Emergency Response. Each deals with and is faced
with potential exposures to hazardous materials. Yet the setting for each is dramatically and
materially different. Hazardous waste operations, for example, are covered not only by 1910.120 but
generally by the OSHA Construction Standards under 29 CFR 1926. RCRA sites are covered by the
OSHA General Industry Standards under 29 CFR 1910. The work environments, employment
practices, and potential exposures vary dramatically in these different settings. As such, these basic
issues need to be considered when addressing training programs to meet the needs of workers and
employers in these diverse settings.
The second issue relates to emergency response. While there was broad agreement that the Emergency
Response category should be covered by the OSHA proposed Training Accreditation Rules under 29
CFR 1910.121, there was substantial concern about the content and criteria for such draining
programs.
National Clearinghouse on Occupational and Environmental Health (NCOEH)
In order to assist with the broad dissemination of curricula for hazardous waste worker training,
NIEHS supported the creation of the National Clearinghouse on Occupational and Environmental
Health. The Clearinghouse was established through a supplement to the Laborers-Associated General
Contractors grant who sub-contracted with the Workplace Health Fund in Washington, D.C. The
Clearinghouse has created a curriculum guide of training materials and a resource library of health
and environmental information regarding hazardous waste, toxic releases and emergency responses.
The Clearinghouse also publishes a regular newsbrief and serves as a networker between NIEHS
grantees and other organizations concerned with quality worker safety and health training.
(1) Goals of the National Clearinghouse
Specifically the National Clearinghouse is:
a) to assist with organization of technical workshops to facilitate updating and clarifying
the complex and continually evolving knowledge in the field;
730
-------
b) to produce a monthly newsletter for similar purposes and to facilitate information
sharing between grantees;
c) to develop a brochure, portable exhibit and training catalog about the training for
outreach purposes;
d) to collect, archive, and report upon new information;
e) to serve as a central repository for curricula and other information related to
hazardous waste training; and
f) to work with NIEHS grantees to develop protocols and quality controls over the
dissemination of this curricula.
The National Clearinghouse's first major task was to establish protocols and quality control over
dissemination of curricula. The Clearinghouse was assigned the task of collection and distribution
of grant-developed curricula to second round applicants for the next round of training grants, and
for the general public. A major component of this task was to produce a catalog of the curricula.
Based on the interest in the catalog and curricula we have learned that there is a considerable demand
for high quality training and training materials that will address the 1910.120 requirements, and that
are likely to satisfy the criteria for accreditation. NCOEH has answered hundreds of phone calls and
written requests about these materials.
Two other items have greatly contributed to the demands for National Clearinghouse services. The
first is a document developed by NIEHS and its grantees in collaboration with other agencies and
concerned parties titled, Worker Criteria for Worker Health and Safety Training for Hazardous
Waste Operations and Emergency Response. In the absence of an accreditation standard, it has been
this document, even in its preliminary draft form, that many people have turned to--including
OSHA--for guidance as to what constitutes appropriate, quality training of workers in the field of
hazardous waste.
The second has been demand for the Hazardous Materials Training for First Responders curricula
developed and produced by the International Association of Fire Fighters. Inquiries and orders have
been received from a wide range of entities, from heavy industry to municipal fire departments and
LECPs, who find this first responder or awareness level training applicable to their needs.
National Clearinghouse: Communications & Networking
While the distribution of curricula and the "Criteria" publication have been major activities other
activities have also been on-going. These include drafting copy and designs for a program brochure,
developing a training catalog and assembling an exhibit. These are being developed for outreach
purposes. An internal newsletter is published monthly for circulation among grantees.
Additional technical workshops have been held including one clarifying the nature of emergency
response training and a second on development of health effects modules. This latter workshop was
cosponsored with the Association of Occupational and Environmental Health Clinics with whom
NIEHS and the grantees are working to obtain case-based training materials to help teaching about
health effects. A technical workshop on quality training for prevention of work-related injury and
illness associated with hazardous chemical transportation has been suggested.
A computerized data base has been established for collecting, archiving, and circulating abstracts of
relevant pertinent books and documents. The National Clearinghouse has been aided in this regard
by the input of a number of people involved with building labor and health and safety resource
centers including Helen Beal at OSHA, Ruby Tyson, librarian for the AFL-CIO, librarians for
AFSCME, OCAW and NIEHS, and in particular the Labor Occupational Health Program at UC-
79 f
-------
Berkeley whose guidance documents on organizing a labor health library have proven to be an
invaluable resource.
In this second year of its operation, the National Clearinghouse hopes to better respond to the needs
of grantees and put their claims to our attention at least on par with those of the general public. This
will be aided by a streamlined and clarified decision-making process with grantees to have a more
formal and continuous mechanism for input in the form of an Executive Committee to the A dvisory
Board.
This structure reflects a sense of ownership and investment in the National Clearinghouse operation
on the part of grantees, which should help to "institutionalize" the project and nurture both a clear
and official role for these "worker educators." The role of an Advisory Committee has been spelled
out and executive officers proposed for election annually. This clear mechanism of involvement and
decision-making should foster among the labor and university colleagues an on-going mechanism for
cooperation and collaboration -- essential elements in maintaining high quality training.
The National Clearinghouse can carry out outreach on the program's behalf in consultation with
NIEHS and the grantee executive committee. Plans are being made to exhibit, distribute copies of
the program brochure, training catalogs, and other documents produced in conjunction with NIEHS
and its grantees at approved exhibits and meetings. The National Clearinghouse is willing to accept
relevant notices, documents, news or other items for inclusion in the newsletter and looks forward
sometime in the future to producing a quarterly or bimonthly newsletter for a wider audience. The
National Clearinghouse is developing and maintaining mailing lists in order to notify those interested
when new or updated documents, such as the training catalogs and curricula listing, are available.
The National Clearinghouse Tomorrow -- Potential
In the future, the National Clearinghouse will move toward improving and expanding efforts in each
of the above areas. The Workplace Health Fund publications program is growing rapidly, so the Fund
and National Clearinghouse together are increasingly becoming a focal point for distribution of
occupational health and safety literature. The infrastructure built for this purpose can then in turn
support further literature distribution.
The library and information infrastructure being built can eventually serve as an increasingly more
effective vehicle for workers, their leadership and communities to access the information they need
to empower their own efforts to improve both occupational and environmental health and safety.
Because of the centralized Washington, D.C. location the National Clearinghouse can be an efficient
resource center for supporting improved research and research as well as training.
Because of the heavy demand for information and materials experienced and the appreciation
expressed upon finding items in a readily accessible, centralized source, a strong long-range potential
in this project is to foster ongoing collaborative efforts between labor, the academic community, and
government.
CONCLUSION
When Congress passed the Superfund Amendments and Reauthorization Act of 1986 (SARA), ,t gave
NIEHS two major tasks: to develop programs to support basic health research on risks posed to human
health by hazardous waste sites and to support curriculum development and pilot worker training
efforts targeted to employees who are involved in cleaning up hazardous waste sites, handling toxic
materials or responding to hazardous environmental releases.
792
-------
Populations of hazardous waste workers continue to increase including hazardous waste generators,
employees involved in cleanups at Department of Energy (DOE) nuclear facilities, hazardous materials
transportation workers and volunteer firefighters who respond to hazardous releases.
Currently, the NIEHS Worker Training Program is concerned with promoting the development of
quality training curricula, adequately qualified training staff, effective methods for assuring the
competence of trainees in a core of required skills and knowledge, and functional evaluation
procedures for worker training programs.
The NIEHS Superfund Worker Training Program is committed to assuring that the Congressional
mandate for worker protection under SARA (Section 126) is carried out by creating field-tested
models of effective training techniques and skills-based curriculum across the country that are
accessible to a broad cross-section of the hazardous waste workforce.
REFERENCES
Cohen, H.H. & Jensen, R.C., Measuring the Effectiveness of an Industrial Lift Truck Safety Training
Program, Journal of Safety Research, 1984, Vol. 15, No. 3, 125-135.
Cohen, A., Factors in Successful Occupational Safety Programs, Journal of Safety Research, Vol. 9,
No. 4, 1977.
Cohen, A., Smith, M.J. & Anger, W.K., (1979). Self Protective Measures Against Workplace Hazards.
Journal of Safety Research, 11, 121-131.
Deutsch, S.: NIEHS Hazardous Waste Worker Training Grant Program. Year 3 Site Visit Review
Meeting. Research Triangle Park, NC, August 27, 1990.
Heath, E.D., Worker Training and Education in Occupational Safety and Health: A Report on Practice
in Six Industrialized Western Nations. American Journal of Industrial Medicine, 1981; 2, 379-403.
Hopkins et al., Behavioral Procedures for Reducing Worker Exposure to Carcinogens: Final Report,
NIOSH Contract 210-77-0040, University of Kansas, 1981.
Hughes, J.T.: An Assessment of Training Needs for Worker Safety and Health Programs: Hazardous
Waste Operations and Emergency Response. Applied Occupational and Environmental Health, 6(2),
February 1991.
Komacki, J., Heinzmann, A.T., & Lawson, L., Effect of Training and Feedback: Component Analysis
of a Behavioral Safety Program, Journal of Applied Psychology, 1980, Vol. 65, No. 3, 261-270.
Komacki, J.,Barwick, K.D., & Scott, L. R., A Behavioral Approach to Occupational Safety:
Pinpointing and Reinforcing Safe Performance in a Food Manufacturing Plant. Journal of Applied
Psychology, 1978, Vol. 63, No. 4, 434-445.
Maples, T.W., Jacoby, J. A., Johnson, D.E., Ter Harr, G. L., & Buckingham, F. M., Effectiveness of
Employee Training & Motivation Programs in Reducing Exposure to Organic Lead & Lead Alkyls,
American Industrial Hygiene Association Journal, 1982, Vol. 43, No. 9, 692-694.
Moran, J. and Dobbin, D.: Quality Assurance for Worker Health and Safety Training Programs:
Hazardous Waste Operations and Emergency Response. Applied Occupational and Environmental
Health, 6(2), February 1991.
793
-------
National Institute of Environmental Health Sciences: Minimum Criteria for Worker Health and Safety
Training for Hazardous Waste Operations and Emergency Response. Research Triangle Park, NC,
April 24, 1990.
National Institute of Environmental Health Sciences: Guidelines for Site Visit Reviewers. Research
Triangle Park, NC, March, 1990.
National Clearinghouse on Occupational and Environmental Health: Curricula Listing - Hazardous
Materials and Waste Workers Health and Safety Training. Washington, DC, January 9, 1991.
National Institute of Environmental Health Sciences: Chartbook on Worker Safety and Health
Training. Research Triangle Park, NC, August, 1990
National Institute of Environmental Health Sciences: Summary of NIEHS Funded Superfund Worker
Training Grants. Research Triangle Park, NC, September 15, 1990.
National Institute of Environmental Health Sciences: Testimony of National Institute of
Environmental Health Sciences on the Occupational Safety and Health Administrations Proposed
Standard for Accreditation of Hazardous Waste Operations and Emergency Response Training. 29
CFR 1910.121. Washington, DC, January 29, 1991.
National Institute of Environmental Health Sciences, Three Progress Report. Research Triangle Park,
NC, April, 1991.
National Institute of Environmental Health Sciences: Superfund Hazardous Waste Worker Health and
Safety Training Program - Notice of Meeting. Federal Register, December 19, 1986.
National Institute of Environmental Health Sciences: Summary Report of 1990 Site Visit Reviews of
NIEHS Worker Training Grantees. Research Triangle Park, NC, February, 1991.
National Institute of Environmental Health Sciences: Request for Applications for Worker Training
Grants. Research Triangle Park, NC, December 28, 1989.
National Institute of Environmental Health Sciences: Hazardous Waste Worker Training Program -
Final Rule. 42 CFR Part 65, Monday October 22, 1990.
National Institute of Environmental Health Sciences, Request for Applications - Worker Superfund
Training Grant Program. Federal Register, Research Triangle Park, January, 1986.
Occupational Safety and Health Act of 1970, Public Law 91-596, 91st Congress, S. 2193, December
29, 1970, Section 21(c).
P.L. 99-499: Superfund Amendments and Reauthorization Act of 1986. Section 126. Worker
Protection Standards. October 17, 1986.
Robins, T.G., Hugentobler, M.K., Kaminski, M., & Klitzman, S., Implementation of the Federal
Hazard Communication Standard: Does Training Work?, Journal of Occupational Medicine, Vo. 32,
No. 11, November 1990, 1133-1140.
Rockwell, T.H., Safety Performance Measurement, Journal of Industrial Engineering, 10,12-16, 1959.
794
-------
Smith, M.J., Cohen, H.H., Cohen, A., & Cleveland, R.J., Characteristics of Successful Safety
Programs, Journal of Safety Research, Vol. 10, No. 1 (1978), 5-15.
Tarrants, W. E., The Measurement of Safety Performance, 1980, Garland Press, New York.
Vojtecky, Michael A., Workplace Health Education: Principles in Practice, Journal of Occupational
Medicine, Vol. 27, No. 1, January, 1985, p. 29-33.
Whiting, Basil J., Early Worker and Employer Training Initiatives at OSHA, Toxicology and Industrial
Health, Vol. 5, No. 4, 1989, p. 87-95.
Zohar, D., Cohen, A., and Azar, N., Promoting Increased Use of Ear Protectors in Noise Through
Information Feedback, Human Factors, 1980, 22(1), 69-79.
Zohar, D., Promoting the Use of Personal Protective Equipment by Behavior Modification
Techniques, Journal of Safety Research, Vol. 12, No. 2, 1980.
795
-------
NIEHS Superfund Worker Education and Training Grant Program
Appendix A.
SUMMARY OF NIEHS FUNDED SUPERFUND WORKER TRAINING GRANTS
The following is a general summary of the sixteen Superfund
worker training grants supported by NIEHS. Individuals are
encouraged to contact grantees directly for more specific
information about a particular program.
796
-------
NIEHS Superfund worker Education and Training Grant Program
Principal Investigator/Institution;
Marianne Brown
University of California
California Consortium
UCLA Center for Labor Research and Education
1001 Gayley Avenue, Second Floor
Los Angeles, CA 90024
Telephone: 213-825-3877
Fax: 213-825-3731
Other Participating Organizations:
University of California at Berkeley
Labor Occupational Health Program
University of California at Los Angeles
University Extension Program
University of California at Davis
University Extension
University of California at Irvine
Extension Program
University of Southern California
Continuing Education Program
Los Angeles Committee on Occupational
Safety and Health
Target Training Populations:
Superfund site workers; state/county emergency response
personnel; waste transportation personnel; and waste site
assessment workers
Program;
Curricula have been developed for all target populations
involved in handling hazardous waste and emergency response.
New courses have been pilot tested. Courses are delivered
throughout the state of California, with recent expansion
into Nevada, Arizona and Federal Region Nine.
79?
-------
NIEHS Superfund worker Education and Training Grant Program
Principal Investigator/Institution;
David McCormack
International Association of Fire Fighters
1750 New York Avenue, NW
Washington, DC 20006
Telephone: 202-737-8484
Fax; 202-737-8418
Other Participating Organizations;
None
Target Training Populations:
Emergency response personnel and first responders nationwide
Program:
Curricula and training materials are being developed to
training fire fighters nationwide. These could eventually
affect the nation's entire fire service i.e., approximately
one million professional and volunteer fire fighters. The
program places emphasis on improved training to assure that
personal protection is adequate for use by fire fighters in
responding to hazardous substance emergencies. The
materials have been pilot tested in the fire service. The
end products will be disseminated among the fire service
nationwide.
798
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution;
Audrey Gotsch, Ph.D.
University of Medicine
& Dentistry of New Jersey (UDMNJ)
New Jersey/New York Consortium
675 Hoes Lane
Piscataway, NJ 08854-5635
Telephone: 201-463-4500
Fax: 201-463-5231
Other Participating Organizations:
New Jersey Department of Labor
Hunter College, School of Health Sciences
Empire State College
State University of New York
New York Committee for Occupational Safety and Health
Oil, Chemical, and Atomic Workers Union,
Local 8-149
Target Training Populations;
Waste clean-up site workers and supervisors, site assessment
personnel, waste treatment, storage and disposal facility
works and waste transporters. Target personnel for
emergency response personnel that are first responders
include 100,000 police, fire fighters, and emergency medical
technicians in New Jersey.
Program;
Curricula are being developed for all areas of hazardous
waste and emergency response as required by OSHA including
that for first responders. New courses are pilot tested for
both 40 hour clean-up work and first responder courses
including: six hours for first responder awareness; eight
hours for first responder operations; and twenty four hours
for hazmat technicians. In addition, courses of eight hours
for HazMat Emergency Medical Technicians have been prepared
and offered.
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institute ion:
David M. Treanor
International Union of Operating Engineers
1125 Seventeenth Street, NW
Washington, DC 20036
Telephone; 202-429-9100
Fax; 202-429-0316
Other Participating Organizations;
None
Target Training Populations;
Operating Engineers engaged in hazardous waste operations.
Program;
Curricula are being developed and used in training programs
targeted at on-site worker populations of equipment
operators. Emergency response training is included as part
of the curriculum. Trainers from the union locals are
trained in an eighty hour "train-the-trainer's" course. The
trainers return to their local and train workers in forty
hour sessions.
SCO
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution;
Sylvia Krekel
Oil, Chemical & Atomic Workers
255 Union Boulevard
Lakewood, CO 80228
Telephone; 303-987-2229
Fax; 303-987-1967
Other Participating Organizations;
None
Target Training Populations;
Hazardous waste treatment, storage, and disposal facility
workers
Program;
A curricula are being developed and used in training
programs targeted at on-site worker populations of oil,
chemical, and atomic workers. Training emphasis is placed
on treatment, storage, and disposal sites. Emergency
response training is included as part of the curriculum.
Rank and file trainers from OCAW local unions are trained in
a "train-the-trainer's" course then go on to train workers
in eight hour (refresher) and twenty-four hour (basic
training) sessions.
801
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Invest jg at or /Institution;
Charles (Chuck) Levenstein, Ph.D.
University of Lowell
Research Foundation
Northeast Consortium
One University Avenue
Lowell, MA 01854
Telephone; 508-934-4000
Fax; 508-452-5711
Other Participating Organizations;
Boston University School of Public Health
Harvard Educational Resource Center
Tufts University, Center for Environmental
Management
Yale University, Occupational Medicine Program
Massachusetts Coalition for Occupational Safety
and Health
Maine Labor Group for Health
Connecticut Committee for Occupational Safety
and Health
Rhode Island Committee for Occupational Safety
and Health
Target Training Populations;
Waste site clean-up workers; emergency response personnel,
treatment, and disposal facility workers; and waste
transporters
Program;
Curricula are being developed for all areas of hazardous
waste and emergency response, including first responders.
New courses were pilot tested. Courses are delivered in six
New England states.
802
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution;
Frank Martino
International Chemical Workers Union
1655 West Market Street
Akron, OH 44313
Telephone; 216-867-2444
Fax; 216-867-0544
Other Participating Organizations;
United Steel Workers of America
University of Cincinnati
Greater Cincinnati Occupational Health Center
Target Training Populations;
Industrial fire brigades and hazardous waste treatment,
storage, and disposal facility workers
Program;
Curricula are being developed and used in training programs
targeted at on-site worker populations of member of the
International Chemical Workers Union and the United Steel
Workers of America. Training emphasis is placed on
hazardous waste treatment, storage, and disposal site
workers and those workers serving on emergency response
teams or fire brigade teams in plants. Emergency response
training is included as part of both curricula and both are
given in thirty-two hour courses - eight hours more than the
minimum required. A course for workers at nuclear
facilities has also been developed.
803
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution:
Chief Roger Ramsey
Seattle Fire Department
301 Second Avenue South
Seattle, WA 98104
Telephone; 206-386-1481
Fax; 206-386-1669
Other Participating Organizations;
Washington State Fire Training Service
Target Training Populations;
Emergency response personnel and first responders
Program;
Curricula are being developed for emergency response for
first responders. New courses have been pilot tested.
Courses are delivered to Seattle Fire Department Personnel.
The Washington State Fire Training Service delivers the
basic course on recognition and identification of hazardous
materials to fire fighters in other fire departments in
Washington State.
804
-------
NZEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution:
Carol Rice, Ph.D.
University of Cincinnati
Midwest Consortium
College of Medicine
Department of Environmental Health, M.L. 056
3223 Eden Avenue
Cincinnati, OH 45267-0056
Telephone: 513-558-1751
Fax: 513-558-1756
Other Participating Organizations:
Southeast Michigan Coalition on Occupational
Safety and Health
Greater Cincinnati Occupational Health Center
University of Illinois
University of Kentucky
University of Michigan
University of Wisconsin
Murray State University
Michigan State University
Purdue University
Target Training Populations;
Waste site workers and supervisors; treatment, storage, and
disposal site workers; emergency response personnel; and
waste transporters
Program;
Curricula are being jointly developed for all areas of
hazardous waste and emergency response personnel. New
courses have been pilot tested and are now delivered in six
mid-western states.
805
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution:
Higdon Roberts, Ph.D.
University of Alabama Birmingham
Center for Labor Education and Research
University Station
1044 South Eleventh Street
Birmingham, AL 35294
Telephone; 205-934-2101
Fax: 205-975-6247
Other Participating Organizations:
Deep South Educational Resource Center
Target Training Populations:
Heavy equipment operators, laborers, waste transportation
workers, and governmental personnel involved with hazardous
waste sites
Program:
Curriculum have been developed and being used in training
programs targeted at on-site worker populations including
technical personnel, general laborers and equipment
operators and transporters. Emergency response training is
included for general hazardous substance site workers.
Curriculum for RCRA site workers is under development.
Courses are given at locations through-out the southeast.
806
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution;
James (Mitch) Warren
Laborers-AGC Education
and Training Fund
Route 97 and Murdoch Road
PO Box 37
Pomfret Center, CT 06259
Telephone; 203-974-0800
Fax: 203-974-1459
Other Participating Organizations:
None
Target Training Populations:
Skilled construction laborers engaged in hazardous waste
clean-up
Program:
Curricula are being developed for use in training programs
targeted at on-site worker populations of laborers.
Emergency response training is included. Trainers from the
union locals are trained in a 120 hour "train-the-trainer's"
course. The trainers return to local training centers and
train workers in 80 hour sessions. Supervisory courses are
given.
807
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution;
Brian Christopher
Alice Hamilton Occupational Health Center
410 Seventh Street, SE
Washington, DC 20003-2756
Telephone: 202-543-0005
Fax: 202-546-2331
Other Participating Organizations;
Illinois Institute of Technology Research Institute
University of Maryland
Alaska Health Project
North Carolina COSH
AFSCME
Target Training Populations;
The Hamilton Center has targeted state/county/local
governmental workers for awareness training, with most being
identified through AFSCME. Two regional training centers
in the Mid-Atlantic and the Pacific Northwest will conduct
training for all populations covered by OSHA 1910.120.
Program;
Curricula are being adapted to cover all the proposed target
populations, with the addition of courses for oil spill
cleanup workers, which will be developed by the Alaska
Health Project. Most of the training will take place in
Maryland, North Carolina, Illinois & Alaska.
808
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution;
Vernon S. McDougall
International Brotherhood of Teamsters,
Chauffeurs, Warehousemen and Helpers
of America
25 Louisiana Avenue, NW
Washington, DC 20001
Telephone: 202-624-6960
Fax: 202-624-6918
Other Participating Organizations;
None
Target Training Populations;
The Teamsters union proposes to initiate a training program
which focuses on two important worker populations: 1) truck
drivers involved in hazardous waste site cleanup; and 2)
drivers and handlers who are involved in transporting
hazardous materials.
Program;
For cleanup workers, the Teamsters will be adapting the
Laborers/AGC curriculum to be delivered at seven existing
regional training centers— three on the West Coast, two in
the East and two in the Midwest. For transporters of
hazardous materials, a 3 and a half hour awareness course
will be established based on new DOT regulations and
delivered to transportation workers in regional sessions
across the country.
809
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution;
Jeffrey A. MacDonald
George Meany Center for Labor Studies
10000 New Hampshire Avenue
Silver Spring, Maryland 20903
Telephone; 301-431-6400
Fax: 301-434-0371
Other Participating Organizations;
University/College Labor Education Centers
International Chemical Workers Union
Railway Labor Executives Association
AFL-CIO Department of Occupational Safety & Health
Target Training Populations:
The Meany Center is developing a national training program
for railroad workers who are involved in transporting
hazardous materials and hazardous waste. Tiered training
will be targeted to railroad workers who are involved in
both awareness level of spill reporting, as well as actual
response action and cleanup of hazardous materials.
Program:
Regional hazardous materials awareness training will be
conducted by various adjunct university faculty in labor
education programs for railroad workers who may be involved
in emergency responses to hazardous spills and releases. A
longer course for maintenance of way workers and signalmen
who are involved in actual spills cleanups will be developed
and conducted in conjunction with the International Chemical
Workers Union (ICWU).
810
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution;
William K. Borwegen
Service Employees International
Union, AFL-CIO
1313 L Street, NW
Washington, DC 20005
Telephone; 202-898-3200
Fax: 202-898-3491
Other Participating Organizations:
None
Target Training Populations;
A nationwide program is being developed to train highway
workers, sewage and water plant operators and gas utility
workers in first responder awareness and hazardous materials
technician level competency.
Program:
SEIU will be developing regional level hazardous materials
training courses in conjunction with existing NIEHS grantees
on the West Coast, the Midwest and the East Coast. SEIU
proposes to train public sector first responders with an
adapted 8 hour awareness course.
811
-------
NIEHS Superfund Worker Education and Training Grant Program
Principal Investigator/Institution;
Franklin E. Mirer, Ph.D.
International Union, UAW
8000 East Jefferson Avenue
Detroit, Michigan 48214
Telephone; 313-926-5566
Fax; 313-824-5700
Other Participating Organizations;
University of Michigan
Target Training Populations:
Workers in the transportation and metalworking industries
who are engaged in hazardous waste generation operations;
will be targeted for both awareness and technician-level
training. Both general generator site workers and
industrial emergency responders will be targeted in the
Midwest.
Program:
UAW will develop its program based on the work of its
existing joint labor-management hazard communications
program, which supports the development of trained local
union safety and health leaders. Extensive job site
exposure and task analysis for workers involved in hazardous
waste generator operations will be conducted as part of
developing site-specific curricula on hazardous waste
handling and industrial emergency response.
812
-------
NIEHS superfund Worker Education and Training Grant Program
Notice: While the majority of this report has undergone extensive
agency review for other purposes and is consistent with regard to
NIEHS policies, the combined report to EPA's 1991 Conference on
Design and Construction Issues at Hazardous Waste Sites has not
received formal peer review. It is published here as an timely
interim report of information should be available for immediate
use. Please contact Denny Dobbin for further information.
313
-------
Hazardous Waste Sites:
Worker Protection Perspectives
John B. Moran, Director
Occupational Safety and Health
Laborers' Health and Safety Fund of North America
905 - 16th Street N.W.
Washington, DC 20006
(202) 628-2596
Donald £. Elisburg
Occupational Health Foundation
1126 - 16th Street N.W., #413
Washington, DC 20036
(202) 887-1980
I. INTRODUCTION
The Laborers Health and Safety Fund, a jointly trusted labor-management group, has been intimately
involved in hazardous waste activities at the local site level to the Federal level with specific regard
to worker safety and health issues for over two and a half years. This involvement has been on a
national scale involving NPL sites and state designated uncontrolled hazardous waste sites and through
participation as members of the EPA-Labor Task Force on Superfund Safety and Health. Extensive
interactions have occurred with local, state, federal agencies including EPA, OSHA, and the U.S.
Army Corps of Engineers, owners; contractors; construction management firms; LEPC's; Emergency
Responders; and several construction labor unions.
What has emerged in the analysis of several case histories is a rather comprehensive view of the
complexities in the implementation of the regulations mandated by the Superfund Amendment and
Reauthorization Act in a confusing arena involving several regulatory agencies and the contractors,
training providers, LEPC's, community representatives, construction managers, emergency
responders, and workers who are directly involved in remediation activities. The central role of
worker protection in these complex undertakings will be addressed.
II. BACKGROUND
The Superfund Amendment and Reauthorization Act (SARA) passed by the Congress in 1986
established two essential components relevant to worker protection at hazardous waste sites. Title 1
Section 126 essentially established the worker protection and "right to know" initiative and required
OSHA [126(a)] and EPA [126(f)] to promulgate specific regulations to protect workers involved in
hazardous waste operations and emergency response. Likewise, OSHA was mandated to promulgated
regulations to accredit training programs pertinent to the training requirements established by the
Congress within SARA and promulgated by OSHA pursuant to the Title I. OSHA promulgated the
worker protection regulations embodied within 29 CFR 1910.120 as interim final regulations on
December 19, 1986 and subsequently issued final regulations which became effective on March 6,
1990. A correction notice was published on April 13, 1990 and on April 18, 1991. OSHA has only
issued a notice of proposed rulemaking regarding the accreditation of training programs. OSHA has
issued various directives concerning 1910.120 but has not issued a comprehensive compliance
guideline to it's enforcement staff to aid in a uniform hazardous waste site inspection policy.
814
-------
EPA was likewise mandated by Congress under SARA to establish several elements related to
Community Right-to-know, known as Title III. This was addressed in several sections to Title III and
includes establishment of LEPC's (Local Emergency Planning Committees) in many communities
throughout the country. EPA retained of course, the responsibility for directing the nations efforts
to clean-up uncontrolled hazardous waste sites through a process of identification, evaluation,
ranking, listing, and either directly funding removal or remediation efforts or causing such to occur
by PRP's (principal responsible parties). EPA was also required to promulgate worker protection
standards for those places of employment not covered by the OSHA regulations (codified at 40 CFR
Part 311).
The Department of Defense (DOD) and Department of Energy (DOE) also have uncontrolled
hazardous waste sites on the federal facilities for which they are responsible. Such facilities are
exempt, however, from SARA Title III requirements but federal employees are covered by the OSHA
regulations by Executive Order 12196, including those engaged in hazardous waste operations and
emergency response. Non-federal workers are covered by the respective OSHA or EPA regulations.
ATSDR (Agency for Toxic Substances and Disease Registry) was created by SARA and mandated
among other things, to conduct and report Health Hazard Assessments at each uncontrolled hazardous
waste site which EPA listed on the NPL (National Priority List). ATSDR has no such authority for
federal facilities, although MOU's (Memorandums of Understanding) are intended with DOD and
DOE to fill that gap.
Various other federal agencies have defined roles with regard to hazardous materials if one considers
the broad range of uncontrolled hazardous waste sites, transport of hazardous materials, hazardous
materials spills on land and water, hazardous waste disposal, and the like. Agencies potentially
involved in some required manner with uncontrolled hazardous waste sites are:
EPA - Superfund Program Lead Agency.
OSHA - Worker Protection standards and enforcement thereof.
NIOSH - Research in support of OSHA proposed standards, Health Hazard Evaluation
(HHE's), and certification of respiratory protective devices.
USCG - Spills of hazardous materials in waterways.
DOD - Facility sites
DOE - Facility sites
ATSDR - Health Assessment Reports, Public Health Advisories
NIEHS - Worker Training Grant program as established by SARA.
DOT - Transport of hazardous materials.
States - Title III programs, 23 State OSHA's, State "EPA" programs, State Health
Departments (some with ATSDR contracts to develop health assessments).
BLM - (Bureau of Land Management) Federal lands with waste sites.
USAGE - U.S. Army Corps of Engineers serves essentially as hazardous waste site
remediation project managers for the EPA at sites for which EPA is directing
the remediation rather than a PRP.
DOJ - Consent degrees with PRP's.
A "typical" Superfund Site remediation project will involve the following agencies directly:
USEPA - (regional office primarily)
USAGE - (district office primarily)
STATE - (several agencies possible)
Local Community - (LEPC or committee)
ATSDR - (directly or contractor)
815
-------
OSHA - (directly, only if requested acting on a complaint or under a directed
investigation)
A "typical" uncontrolled hazardous waste site also involves the following usually private entities:
site characterization contractor
design contractor(s)
RI/FS contractor(s)
ROD contractor(s)
Prime contractor on the site
Sub-contractors on the site
Workers (organized and unorganized)
Local community emergency responders
Local emergency medical personnel
PRP's
Hazardous waste transporters, if a removal project.
Hazardous waste receivers, if a removal project.
From initial listing for evaluation and ranking to the first "shovel full of dirt" in a remediation action
typically requires 6-8 years.
Each phase requiring activity on the site requires compliance with worker protection regulations.
SARA uniquely establishes the requirement that all workers whether employed by private employers;
local; state, or federal governments; and even volunteer emergency responders be protected in
accordance with established regulations when engaged in hazardous waste operations and emergency
response. While all workers so engaged must be protected, the responsibilities for ensuring such
protection has been nested within a large number of governmental agencies with inherently different
missions and jurisdictions. In waste site activities, these differing jurisdictions result in discrete
boundaries being drawn with OSHA being responsible for site worker protection and EPA responsible
for site activities and public protection. In actual practice, such clear distinctions do not occur as
site worker protection issues are directly linked to public protection issues, for example.
III. CASE HISTORIES
Ouincv Naval Yard: Ouincv MA.
As one element of the massive multi-billion dollar Boston Harbor clean-up project, the ex-Quincy
Naval Yard was selected as the site for a sewage sludge treatment facility. The site had been
previously declared an "uncontrolled hazardous waste site" by the State of Massachusetts thus
requiring compliance with 29 CFR 1910.120. Extensive clean-up and removal had occurred on the
site before the Boston Harbor project element work began. The construction of the treatment facility
required substantial ground work, pipe laying, pile driving, and foundation laying. This work was
defined in the specifications as NOT COVERED by 29 CFR 1910.120 despite the fact that it was still
carried as an uncontrolled hazardous waste site by the State. One excavation event lead to worker
exposures resulting in acute exposure health effects. Despite the insistence of the construction
management firm that no special precautions were necessary and that the contractor would not be
reimbursed for additional worker protection measure instituted, the owner, the Massachusetts Water
Resources Authority, over-ruled the management firm and 1910.120 based worker protection
practices and procedures were utilized in all of the excavation activities in area's previously identified
as contaminated (prior to earlier clean-up) and in such activities where very deep excavations were
required. This specific site was the basis for OSHA's first policy statement indicating that a
816
-------
designated uncontrolled hazardous waste site could have "clean zones" not requiring compliance with
1910.120 (P.K. Clark, OSHA letter to J. Moran dated July 30, 1990).
Bunker Hill: Kellogg. Idaho
One of our nation's largest NPL sites, it evidenced total disregard of the EPA and OSHA requirements
when visited in 1989-1990. Severely contaminated equipment was sold to private parties from the
site, the site was not secured and children played in the areas severely contaminated with lead and
arsenic. Many other serious problems were evident. ATSDR, for example, issued a Public Health
Advisory in 1990. The yards surrounding homes in nearly communities had several inches of the top
soil removed and replaced with clean soil to reduce the potential for further childhood lead poisoning.
Even this work was performed in violation of 1910.120 and indeed, the OSHA Area Director
permitted, once the issue was raised, that such workers could be trained at the OSHA created 24 hour
category months before that final 1910.120 was in place and law.
EPA subsequently issued an order requiring several site activities to remedy the earlier abuses
occurring at the site particularly with regard to the smelter complex including security fencing,
cessation of sales of scrap materials and equipment, and abatement of deteriorated asbestos insulation.
Further, much work on the site was being done under emergency action provisions under EPA
contract thus avoiding compliance with the Congressionally mandated Davis-Bacon Act wage
provisions. A subsequent ruling, specifying compliance requirements by the Department of Labor
was applied nationally.
Newport. Rhode Island
Excavation work in preparation for the construction of a multiple story building in a major urban
area uncovered a partially collapsed large fuel storage tank. Soil samples evidenced 7-9 ppm lead in
the EPTox test (it is, then, a hazardous waste) and 3,000-7,000ppm total lead (to which workers were
exposed as dust and orally due to transfer of dirt from hands and clothing). This contaminated soil
was to be removed. Is this a work area which requires compliance with 1910.120? NO according to
OSHA at that time. The "hazardous waste" debris from the excavation was transported over
community streets to a community owned lot where it was stored.
This example is one of many related to an emerging national problem where "hazardous waste" is
encountered in "normal" construction activities. It is not uncommon and has not been addressed, as
yet, by the regulatory agencies at the federal level.
Arkansas
A PRP incinerator operation worked their laborers 12 hour shifts, sometimes back-to-back, while
wearing Level C and Level B equipment. A site review suggested that a large number of the 1910.120
requirements were not being complied with. Working with the contractor, the key site management
was replaced and compliance with 1910.120 pursued as a top priority. Workers called the area OSHA
office, which was unable to respond except through the Dallas Regional office. They were advised
that response from Dallas would take some time. Months later, OSHA still had not responded,
although activities at the site have improved significantly.
Charles George Landfill: Tvngsboro. Massachusetts
OSHA responded to this NPL site based upon written complaint filed by an employee representative,
after all other avenue's to address worker protection concerns were rebuffed by the prime contractor
817
-------
and the federal agencies involved. Response required more than 4 weeks from the time the complaint
was filed. OSHA found heat stress and confined spaces program inadequate (technically) but stated
that they could not cite the contractor because 1910.120 only requires that there be such elements in
the Site Safety and Health Plan. The report of the investigation took weeks. The closing Conference
with workers and worker representatives was narrowly structured to cover only the organized sub-
contractors not the prime or the whole site despite the 1910.120 requirement that the prime is
responsible for safety and health on the whole site. Nearly all sub-contractors, the prime and the
Corps of Engineers were eventually cited by OSHA. The prime contractor contested the citations and
some were dismissed by the U.S. Department of Labor solicitor in Region I as "unenforceably vague".
Work has essentially concluded but a special study indicated subsidence was more severe that
anticipated and liner life may be as little as four years. Latest concerns from the community involve
excessive methane levels in the collection system, manholes, and from the capped area vents. The
town is now concerned about fire and explosion potential.
During early work, run-off to a stream next to the site occurred as evidenced by a dark colored
sediment. Analysis by EPA indicated, as reported at a community meeting, that the stream was
"relatively safe". No analytical data was provided nor numbers given despite requests. This
heightened worker and community concerns because of the appearance that the Federal agencies were
not being completely open with the community and the workers, many of whom attended the
community meetings.
Nvanza; Ashland. Massachusetts
Requests for information by workers and their representatives was essentially ignored by the prime
contractor, the Corps of Engineers, and EPA. Worker representatives were initially not allowed on
the site even in the support zone. Based upon limited information in the Local Repository, a report
identifying areas of concern was prepared by worker representatives. A worker filed ;i formal
compliant with OSHA, which responded some four weeks later.
The EPA Regional office instituted procedures requiring that all contact with the EPA regarding this
site be in writing. OSHA's attempt to conduct an inspection was rebuffed by the contractor with the
Corps of Engineers initially supporting the contractors position. OSHA had to obtain a federal court
order to enter the site. The contractor then denied right of worker representatives to accompany the
OSHA inspection team. OSHA sought another court order although uncertain as to how to proceed
in the matter. Extensive discussions resulted in an arrangement between OSHA and the Corps of
Engineers with the Corps directing the prime contractor to allow the inspection to occur with
employee representatives accompanying the inspection team.
Issues of worker acute illness when a large number of partially filled drums were uncovered in a
"clean" zone, compliance with many requirements of 1910.120, excessive levels of and the inability
to adequately monitor mercury and dimethyl mercury, results of analysis of samples taken near but
outside the site showing high levels of mercury, and failure to develop a coordinated emergency
response plan with the local community resulted in nearly all sub-contractors, the prime, the Corps,
and EPA being cited by OSHA and closure of the site for over 6 months. The time period
encompassed by this case was over a year in duration.
Baird-McGuire: Holbrook, Massachusetts
This site employees both a ground water treatment and incineration approach to remediation. The
area upon which the treatment plant was to be constructed is immediately contiguous to the
contaminated zone and decon pad but was, none the less, termed clean in the specifications thus not
818
-------
coming under the provisions of 1910.120. Once completed, however, the Plant must be operated
under 1910.120 provisions. Workers and their representatives asked the basis for the "clean"
designation. Nine months later, after that work area was sampled and analyzed for the first time and
the whole site characterization was re-examined and other "clean" areas on the site were declared
contaminated, an answer to this simple question was provided. In the meanwhile, hot zone activity
could not begin because the local emergency responders were not prepared, lacking training and
equipment. During that time the ATSDR Health Assessment Report was released and the incinerator
approach discussed with the local community. Both were received with alarm and concern.
Central to resolution of the zone designation and thus 1910.120 compliance issues at this site was a
definition of what constitutes "reasonable possibility of exposure" to workers. Such is a requirement
under 1910.120(a). An initial meeting with OSHA by worker representatives, workers, the contractor,
and the Corps of Engineers resulted in OSHA's being unable to provide any guidance on this issue
as compliance policy had not been established by OSHA headquarters. Over four months later, OSHA
provided written guidance (P.K. Clark, OSHA letter to J. Moran dated October 3, 1990). In the site
characterization re-evaluation a dispersion model was developed to serve as one tool to address this
issue.
IV. WORKER PROTECTION ISSUES
The few cases briefly highlighted in the previous section serve as the basis for the focus on worker
protection issues specific to activities pertinent to the scope of the OSHA and EPA hazardous waste
operations and emergency response standards. The issues of concern which have arisen in our analysis
and which are discussed below are not one's uniquely occurring at only an individual site rather they
represent issues relevant to all. More importantly perhaps, our initial focus on worker protection
issues has clearly demonstrated that these issues are related and central to nearly all hazardous waste
site activities including the local community.
Responses to information requests:
One of the most serious problems we have observed was the response to requests for information from
workers, worker representatives, and even contractors. Despite the requirements of 1910.120(i) and
1910.120(b)(l)(v) and the broader requirements under the hazard communication regulations
(1910.1200 and 1926.59), there is a great reluctance particularly on the part of Federal agencies to
respond to requests for information. The simplest question, such as the site characterization data
supporting zone designations, has required several months for a response. In other cases, information
requests were simply denied and workers, their representatives, and/or contractors had to seek
information at the local information repository.
The failure to promptly and professionally respond to even the simplest of information requests or
questions was a common failing at each of the hazardous waste sites we have evaluated. The
consequences have been deepening of communication problems between site owners/managers and
contractors, sub-contractors, and workers; development of a higher level of mistrust; greater costs;
and increased concerns from the local communities. It is clear to us that the rights of workers and
their representatives to information pertinent to the potential risks workers face on hazardous waste
sites is poorly understood by most Federal agencies involved and by most contractors as well.
Many have been surprised at the impact of these failures to communicate when worker protection
concerns were voiced and further that the impact has spread beyond the workers on the site to the
local community. Projects have been stopped or delayed, costs have unnecessarily increased, public
and worker confidence has been eroded, contractors have been unnecessarily impacted, and concerns
819
-------
have been raised at headquarters level of agencies involved thus creating substantial additional effort
and frustration at the agencies field and regional levels.
Fundamental to the reasons for the reaction to these issues is the increased awareness principally
though the required 1910.120 training requirements and OSHA's hazard communication standards by
workers and worker representatives of the potential hazards associated with uncontrolled hazardous
waste site work. Safety hazards have long been recognized in the construction industry with little
recognition of the health hazards associated with such work. This has begun to change with the
implementation of the worker right-to-know requirements under the OSHA Hazard Communication
regulations. A similar increased awareness is developing at the public level as well, as a consequence
of the community right-to-know SARA Title III requirements and the Community Relations Plan
required by EPA for NPL site remediation and longer term removal projects.
While hazardous waste site risk assessment has become a sophisticated science and risk management
has reasonably well understood dimensions as embodied within 1910.120, our abilities at effec tive risk
communications with workers, contractors, emergency responders, and local communities is terribly
inadequate. Where a minor risk communication problem can begin and spread to large dimensions
is when worker protection concerns are not adequately addressed when first raised.
Worker representatives:
While responses to requests for information or to questions from worker has been a significant
problem, the problem for representatives of such workers has been even more difficult. Worker
representatives have been denied access to hazardous waste sites on which the workers they represent
work. Access denial has been to the support zone area containing site offices, not just the operating
areas on the site despite the fact that such representatives have the proper training although such is
not required to enter the support zone. Representatives have likewise, been denied access to
information such as Site Safety and Health Plans, Site Characterization Reports, participation in
OSHA walk-around inspections, and the like.
Much of the basis for this problem arises from the apparent fact that most Federal agencies and many
contractors are simply unaware of the rights worker representatives have under the OSHAct to act
in behalf of the workers they represent. Further, when a worker raises a safety and health concern
to his or her representative, that representative not only has a moral and ethical burden to address the
concern, but a legal one as well. Denial of worker representative participation in safety and health
issues is not only a violation of OSHA regulations but such serves to escalate concerns among the
workers.
Training
The interim 29 CFR 1910.120 required a minimum of 40 hours of training off-site and 24 hours site-
specific for uncontrolled hazardous waste site workers. The final 1910.120 regulations added a 24
hour off-site 8 hour on-site worker training category, although such was not specified in SARA as
was the 40 hour requirement. While the on-site distinction between the two categories is blurred at
best in actual reality this is compounded by OSHA's failure to provide compliance guidelines.
Unfortunately, the general trend is toward employment of only 24 hour trained workers despite the
inherent danger in assuming the site to be within OSHA's limits for such workers.
In addition to the dimension of the training issue noted above; owners, site-managers, and contractors
have the added problem that no criteria exists, with regard to training program content, curriculum,
trainer provider requirements, testing, and the like. Those issuing specifications, therefore, have little
820
-------
to require except minimum training hours and the broad list of issues to be addressed as contained
in 1910.120.
The consequences of these issues are that a broad range of worker and supervisor competency and
proficiency exists on hazardous waste sites. In cases where well trained and poorly trained workers
have been employed at the same site, a large disparity exits related to emergency response actions,
personal protective practices, worker practices and procedures, etc. This has resulted in well trained
workers in Level C equipment working alongside poorly trained workers in Level D gear who were
suffering acute irritation responses and poorly trained workers using inappropriate respiratory
protection when faced with an acute exposure situation.
1910.120 Regulations:
The 1910.120 regulations were mandated by Congress in SARA. Initially issued as interim regulations
on Dec. 19, 1986, the final regulations became effective March 6, 1990. Despite the fact that
1910.120 represents a reasonably broad based regulation unlike most other OSHA regulations and that
it establishes regulatory compliance requirements in a unique construction setting, no compliance
guideline or directive yet exists.
This is a particularly relevant issue as much of the 1910.120 standard is in performance language.
Absent specific compliance guidelines or directives enforcement is extremely difficult. For example,
1910.120 requires that a confined space program be part of the site safety and health plan. In an actual
case OSHA was unwilling to cite a contractor for not having a confined space program because the
contractor stated that no worker would enter a confined space thus on such procedure was required
in the Plan. 1910.120 does not offer this option, however. In another instance, the confined space
program was incorrect and indeed a threat to worker safety. Again OSHA refused to cite on the basis
of the fact that, in accordance with 1910.120, the contractor did indeed have a written confined space
program even though it was incorrect. The problem is that while 1910.120 requires several procedural
plans such as for Confined Spaces, the only criteria as to the content of such is contained in references
to the standard which are not enforceable per se.
Similar problems have occurred with the required heat stress program. In that instance a contractors
heat stress program involved workers weighing "in" in the morning when they started work and
weighing "out" at the end of the day. Clearly the purpose of a heat stress management program is to
prevent acute heat illness especially heat stroke which can occur quickly and be life threatening. The
program noted above completely failed to provide adequate protection for the workers but OSHA did
not cite the contractor.
OSHA, while still failing to have a 1910.120 compliance guideline or directive, has been interpreting
the standard on a request-by-request basis. In June, 1990 OSHA issued an interpretation of 1910.120
which stated that hazardous waste sites could have areas which could be designated clean and not,
therefore, require compliance with 1910.120. In October, 1990 OSHA issued a policy statement with
regard to the interpretation of "reasonable possibility of exposure" which keyed that determination
to the definition of "exposed" in the hazard communication standard. That policy however, then went
on for several paragraphs explaining further what "exposed" meant to OSHA. The result is very
confusing and remains so yet such is critical to worker protection requirements in Site Safety and
Health Plans, Specifications, and the like.
At best, there is tremendous confusion in the field especially between site owners, managers,
contractors, and workers with regard to the specific requirements of 1910.120. OSHA's fragmentary
interpretation of the standard presents a less than coherent approach and often raises more questions
821
-------
than it resolves. Further, where OSHA has been requested to assist in such interpretations in the field
response has been poor. In one specific instance OSHA simply refused to provide any guidance
whatever to a group consisting of labor, contractor, and site manager. This presents serious problems
to those writing specifications, managing sites, contractors, and workers as there is no clear, common
basis for decision making.
Enforcement:
Enforcement of the 1910.120 regulations is the responsibility of OSHA. OSHA's failure however to
issue compliance guidelines or directives has resulted in other Federal agencies and State agencies
having to assume this burden. For example, on EPA managed superfund sites, EPA has had to
establish it's interpretation and rules with regard to 1910.120. Likewise States which are managing
superfund site activities have had to respond to these interpretation issues at the state level. At least
two OSHA State Plan States in promulgating their State specific version of 1910.120 dealt with some
of these issues. For example Alaska removed the scope qualifier "reasonable possibility of exposure"
and required that all "in scope" operations comply with the standard. Training issues were clarified
in both the Alaska and Washington regulations. In that regard, Federal OSHA interceded resulting
in Washington changing their regulations to comply with Federal OSHA 1910.120. The new
Commissioner of Labor in Alaska announced in March, 1991 that the State version of 1910.120 would
be rolled back to the Federal Standard.
OSHA's compliance staff is suffering from a lack of guidance on this standard thus enforcement is
vague and not uniform. When specific guidance is requested from field offices, the result is all to
often "no guidance". In other instances, the OSHA area office is unable to provide an inspector when
a complaint has been filled.
Emergency Response:
29CFR1910.120 requires that coordination occur with the local community with regard to emergencies
which might occur on a hazardous waste site. The concept is to link SARA Title I and III at the local
site level. A hazardous waste site contractor may provide for on-site emergency response activities
thus negating the need to call upon the local emergency responders in the event of an emergency. In
most instances, however, the site emergency response is usually an emergency alarm and evacuation
approach with a call to the local emergency response group. Clearly, in that approach, coordination
with local emergency responders and emergency medical care facilities is required in order to be
prepared to respond to worker injury or acute illness events or other site emergencies. In every case
with which we are familiar, the coordination with the local emergency response entity ha:5 been a
source of extreme confusion, problems, and difficulty. In every instance, the local emergency
response group was not properly trained or equipped to respond to an emergency at the site. Further,
coordination with the local emergency response group always occurred very late in the site activity
schedule often resulting in suspension of work at the site until the coordination, training, and
equipping problems could be worked out. Added to this problem is the fact that no criteria exists
from OSHA as to the content of emergency response training for hazardous waste site emergencies
and the further fact that OSHA excluded emergency responders from the 1910.120 training
accreditation proposed rule.
Interagencv Coordination:
The typical EPA superfund site directly involves at least EPA, The Corps of Engineers, and OSHA
from the Federal agency perspective. State and local governments are involved as well, of course.
Federal agency activities specific to hazardous waste sites are normally conducted through the
respective agencies regional or area offices. EPA's mandate is to protect the public health and welfare.
822
-------
The Corps essentially works for EPA as site construction managers and are responsible for the
preparation and issuance of solicitations and the awarding of contracts based upon previous site
specific work such as the RI/FS which is normally conducted by a contractor to EPA. OSHA is
responsible for the enforcement of the site worker protection standards under 1910.120. Each agency
has a separate and distinct mission and area of responsibility. In theory the areas of responsibility
between EPA and OSHA do not overlap nor are there major gaps in site specific responsibilities as
the Corps or other site construction management firm serves as the point where the EPA public and
the OSHA worker areas come together.In reality the relationship between the Corps or the site
manager and EPA is not well defined with regard to safety and health issues particularly where
problems arise or decisions need to be made. As is usually the case in the construction setting, the site
manager acts in behalf of the owner (EPA) to meet schedules and cost criteria. Major issues arise
when changes may be required that effect cost or schedule especially where safety and health issues
are involved which is further exacerbated by the failure of OSHA to provide specific guidance or
assistance. Compounding this problem from EPA's perspective is the requirement that EPA serves
as the principal contact with the local community. When site issues arise which are not effectively
handled, EPA has to deal with the community issues which often arise.
Site Safety and Health Plan Approach:
Hazardous waste site remediation solicitations take, essentially, one of two basic forms. In the first,
zone designations and worker protection criteria such as levels of PPE are specified based upon the
site characterization report, RI/FS, the ROD and similar information. In the second, the information
is provided but the bidding contractor is responsible for specifying the details in the bid response. The
first method is often preferred because it narrows the cost spread in the bid responses and simplifies
the bid review process. However, when changing site conditions or questions regarding the basis for
such decisions arise the resolution to these are frequently time consuming, difficult, and tend to focus
more on the costs and contract modifications paperwork required than on the fundamental worker
protection issues which are involved. In addition, as is the case at Baird-McGuire, where the site
characterization and RI/FS were inadequate with regard to providing all of the site characterization
data in a complete manner the resolution of these issues becomes extremely complex and beyond the
purview of the site manager.
Carcinogens:
Unlike NIOSH and other regulatory agencies, OSHA exposure regulations often, especially for
carcinogens such as asbestos, establish exposure limits at which significant lifetime risk is believed
to be present to those workers exposed. Many hazardous waste sites contain known or suspected
carcinogens. Construction workers, normally unaccustomed to a focus on health concerns as the
construction industry in general sees injury as it major risk, do not understand the less than full
commitment to protecting them from exposures to such materials in waste site operations. Similar
concerns arise at the local community level with regard to the potential risk associated with the
presence of carcinogens on a waste site. This concern is greatly heightened, among workers and the
public, when ATSDR Health Assessments Reports are released related to sites which contain
carcinogens.
This problem is further exacerbated by insensitive site managers who, in discussions with contractors
seeking reimbursement for resources spent in upgrading worker protection, claim the contractor has
been over-protective of workers health. The only acceptable view with regard to exposures to
carcinogens, from a worker and public health protection perspective, is that no preventable exposure
should be allowed to occur.
823
-------
Hierarchy of Control:
The philosophy behind 1910.120 is very poorly understood from one fundamental perspective.
1910.120 essentially requires that worker protection levels be DECREASED as information becomes
available to support such a determination. The 1910.120 approach requires that workers be protected
at least to the Level B ensemble when hazardous materials are known to be present and the level of
exposure is not known or can not be estimated with a high degree of confidence sufficient, for
instance, to comply with the NIOSH Respirator Decision Logic for respiratory protection selection.
In every case with which we are familiar, the opposite has occurred. That is, workers were in a lower
level of protection when exposure problems developed which exceeded the capacity of their protective
ensembles. Subsequently, the level of protection was increased.
V. CONCLUSIONS: LESSONS LEARNED
Based upon over two years of very active and in-depth activity at a number of our nations hazardous
waste sites many of which have involved literally months of effort at a single site to resolve even the
simplest of issues, we offer the following conclusions with regard to current activities pursuant to
29CFR1910.120 and specifically the worker protection aspects of that standard and such work:
1. SARA established a unique and comprehensive approach to worker and public protection
associated with potential exposures arising from hazardous materials including those on
uncontrolled hazardous waste sites. While that unique landmark legislation i;3 fully
encompassing of worker protection, the actual implementation of the intent of Congress is
nested in several federal regulatory agencies whose jurisdictional boundaries are often not
clear and precise on hazardous waste sites. While these agencies have designated regulatory
responsibilities it has become increasingly evident that they were less than able to effectively
communicate with each other. Further, it is evident that no one agency is "in charge" of
hazardous waste operations and, thus, no one agency is "accountable". EPA, through the
Special Task Force on superfund Safety and Health, has recognized this deficiency and
attempted to close this gap. Recent participation by OSHA and the Corps in the Task Force
is a useful emerging aspect of these areas.
2. It is increasingly evident that worker protection on hazardous waste sites is not just one of
many basic items which must be completed on a project check list. Such worker prelection
issues are central to the hazardous waste site actively, are dynamic and are demanding of far
more focused attention and concern than has been evidenced in all of the sites with which we
have been involved. Failure to address worker protection concerns can have far reaching,
costly effects. More than ever before, effective worker protection programs offer the
opportunity for workers to be participants and partners in an important National undertaking.
3. OSHA simply has not taken the 29CFR1910.120 regulations for which it is responsible with
any degree of commitment. The enforcement activity even after three years of 1910.120 is
spotty and confused at best, no doubt due to the lack of compliance guidance. EPA, the
Corps., labor organizations, and contractors have been frustrated by OSHA's lack of response
to issues raised about 1910.120. OSHA has, furthermore, confused the intent of these
regulations by issuing policy statements, Instructions, and local interpretations which have
served to create confusion rather than resolve it. Indeed, many aspects of 1910.120 written
in the 1980's popular "performance" language are unenforceably vague which compounds the
OSHA's compliance staffs difficulties and confuses those seeking to interpret and comply with
the standard.
824
-------
4. Currently no criteria exists upon which routine decision making can be based with regard to
what constitutes reasonable possibility of worker exposures. This deficiency, combined with
the added confusion caused by OSHA which allows clean areas on hazardous waste sites and
permits a lesser trained worker category, creates a significant potential for worker exposures
and confuses any attendant decision making process. This is compounded by the common
approach of increasing the level of worker protection at waste sites as exposures are confirmed
rather than the approach required by the pro-active 1910.120. The recent development of a
modeling approach by EPA and the Corps of Engineers offers the potential for coming to
technical grips with the "reasonable possibility of exposure" issue.
5. The frequent approach of specifying levels of worker protection required in specification
packages is deficient in that, as currently employed, the basis for such decisions is not
presented and can not be verified by the bidder. Subsequent changes based upon emerging
site data is, as a consequence, complex and difficult. The inclusion of requirements for
designers to include site characterization specifics would help resolve this problem.
6. Workers and their representatives have a right to ask questions with regard to worker safety
and health issues. The norm is no response or an incomplete response. When pursed, that such
responses often take weeks or months is totally unacceptable. Workers and their
representatives have a right to a courteous, prompt, and complete response to any questions
raised with regard to worker safety and health issues. Indeed, much of what they frequently
ask should, under 1910.120 and 1926.59, be routinely provided without the need to make a
request.
7. The adequacy of worker and supervisor training programs is presently unknown and no
criteria which such programs must meet is currently being used in specifications for hazardous
waste work. OSHA has delayed its response in this regard despite the SARA mandate. In the
interim, the NIEHS National Workshop Report provides the only guidance and it is essentially
not used. As a consequence, the degree of worker and supervisor proficiency and competency
varies widely resulting in increased risk to many workers and a potential threat to nearby
communities. This is most evident in "open" annual refresher training programs, often
conducted by the NIOSH ERC's, where the broad range of core training proficiency, or lack
thereof, is very evident.
8. All to often site characterization reports, RI/FS reports, and the like are incomplete and do
not contain all of the information pertinent to an effective worker safety and health program.
Frequently, the identification of all contaminants found on the site and the sample locations
are boiled down to "critical contaminants" and "zone boundaries". This information is further
reduced and condensed in the solicitation package. The result is that all to often critical
information is excluded, critical contaminant and sample locations are lost, and the basis for
ongoing site activities is lacking as the focus remains on the few critical contaminants rather
than the full list of known contaminants. A contractor bidding from such a solicitation
package may indeed submit what is believed to be a valid proposal only to find after work
begins that the situation is far different than was believed. In the process and the often
protracted procedures required for changes, worker protection is at risk.
9. Based upon the current OSHA confusing information on clean zones and what constitutes
reasonable possibility of worker exposure, zone designations on hazardous waste sites is very
suspect. Worker protection is a key issue here as the trend portrayed by OSHA is to loosen
such site criteria. The inclusion of clean areas requiring no 1910.120 training (but perhaps
other training), a 24 hour worker training category, and a 40 hour worker training category
825
-------
has already caused great confusion on waste sites which is being compounded by the zones
and possibility of exposure issues.
10. Emergency responders remain a serious issue at hazardous waste sites. Two issues are of
importance. First, appropriate and adequate response to protect workers and the public from
the hazards associated with an on-site emergency. Second, prompt and professional response
to the site in response to the severe injury or illness of a site worker. Both of these issues
present a serious threat.
11. Superfund remediation activities are expensive and time consuming. Yet it appears that once
the remediation work begins, cost control becomes extremely important and is vigorously
enforced by the site manager. In this regard, hazardous waste site work is not unlike the
typical construction contract work; i.e., if it isn't in the contract, you don't get paid for it.
Worker protection suffers under this approach as appropriate worker protection measures can
only be justified in these terms if the problem one was seeking to prevent occurs because the
desired protective approach was not used. The view expressed by at least two site managers
that such issues often represent overconcern for the workers is dangerous: to workers.
12. Hazardous waste site remediation work often involves three Federal agencies but at a
minimum involves at least an environmental government agency and an occupational
governmental agency. These two governmental entities do not share overlap in areas of
responsibility. The occupational entity governs the site and the environmental the area outside
the site. They deal with different regulatory philosophies, differing target populations,
differing risk levels at which they regulate, different enforcement procedures, widely
different enforcement powers, etc., etc,. Yet much of what the environmental agencies require
for site activities affects workers and what is done on the site in response to occupational
regulations impacts the local community. No one is responsible for these overlap areas and for
sorting out the conflicts in worker and public protection which can and do arise. EPA,
through the Special Task, is beginning to address these issues.
13. Worker protection is a key aspect of hazardous waste site work. Properly addressed, a safe,
productive, and cost effective remediation or removal project can occur which assures the
health and wellbeing of the workers involved and the protection of the nearby public.
Improperly addressed, worker confidence and public trust can be seriously eroded resulting
in a wide range of unnecessary complexities and costs. Not withstanding the lack of details
from OSHA which are needed with regard to 1910.120, the technical expertise and resource
materials do largely exist to provide effective worker protection. The results of the EPA Task
Force efforts and changes the Corps is initiating are critically important indicators of recent
progress in addressing these issues. Other hazardous wastes activities would benefit by using
this information.
826
-------
Crisis in the Fire Service
Les Murphy
NIEHS Grant Manager
International Association of Fire Fighters
1750 New York Avenue N.W.
Washington, DC 20006
(202) 737-8484
INTRODUCTION
In addition to the inherent hazards of their work operations fire fighters suffer a number of
administrative and management difficulties. These are a result of the unique characteristics of
employment, perceived obligations to function in all hazardous conditions, and lack of regulatory
support. After all, fire fighters are those people who come when you summon them with a 911 call.
Unfortunately, in Kansas City when the fire fighters responded to a construction site fire it was
considered routine, although typically hazardous, but it turned out to be tragedy for the families of
six fire fighters. Explosives were stored on the site, without any indication of the nature of this class
of hazardous material, and when fire detonated it six fire fighters died. Hazardous waste clean up
sites represent a class of exposure with a high element of the unknown.
Fire fighters are typically municipal employees or non-compensated individuals (volunteers) with an
obligation to function in the same fashion as paid municipal fire fighters. The use of the phrase fire
fighter shall include both compensated and non-compensated personnel.
The Fire Service, through their independent efforts, have developed substantial technical advances
in the safe remediation of emergency situations. The consolidation of these processes and life saving
techniques was greatly advanced by organizations such as the International Association of Fire
Fighters and the National Fire Protection Association. The International Association of Fire Fighters
has developed a number of training aids, such as audio visual training packages, for use as teaching
aids in Hazardous Waste and Material topics. A study conducted by Johns Hopkins University has
verified that over 50,000 fire fighters have been trained in the Tier I, First Responder at the
Awareness level, program. The IAFF Tier II, Operational Level training program has recently been
released and it is anticipated that the training numbers will be as impressive as they are for Tier I.
Regulatory and legislative emphasis has traditionally centered on pro-active topics. Building codes
and safety and health standards are written with the intention of preventing emergency situations.
Dealing with crisis situations has been mostly left to the Fire Service and their own resources.
For many years the traditional role of the fire fighter was to extinguish fires and rescue endangered
people. Gradually their role expanded to include all emergency situations, such as homeowners with
flooded basements and even rescuing cats from trees. Fire department service charges have been
implemented in many communities to discourage the request for frivolous services such as pet rescues.
Preventive measure activities were significantly improved by involvement in pro-active goals such
as Fire inspections, community training and standards development. While the great traditions of the
fire service span over centuries it is only in the past twenty years that the major progress has been
accomplished in the emergency medical care provided to victims of emergencies. Formerly injured
persons were extricated from the scene and transported to medical treatment facilities. Now victims
are rescued, receive medical stabilization at the site, and are transported in vehicles while providing
continuing medical care until arrival at a shock trauma hospital unit. In fact in many communities
827
-------
the crisis is simply a medical emergency and the fire department operates within the framework of
the medical care establishment.
The advances in the fire service were quite supportive of other progressive actions dedicated to
making the work place, home and general environment safer. These apparently parallel ambitions
may now be on a collision course due to the differing concepts involved. EPA developed SARA and
included ample provisions for contingencies that may occur during clean up on a hazardous waste site.
Title I, Section 126, promulgated requirements for health and safety for employees involved in
hazardous waste operations and response to hazardous material incidents. Protection, liaison,
communications, and responses within communities were further promulgated in Title III. In both
of these programs the fire department was an important integral part of the system.
Yet, in a real world situation the implementation of the elements of these plans fall far short of
regulatory intentions. A classic example is the Baird-McGuire Superfund Site in Holbrook,
Massachusetts. This site, under the management of EPA and the Corps of Engineers, failed to
adequately conduct a comprehensive site characterization study. As a result of considerable urging
a recharacterization study was completed and areas previously designated as "clean zone" were
reclassified as contaminated.
Even though EPA's own Health and Safety Audit Guidelines clearly stated the need for evaluation
of local community response capabilities this was not done. The Holbrook Fire Department is fairly
small but has trained all its' personnel at Tier I, Awareness Level. Their statutory responsibility for
Emergency Response includes light rescue, heavy rescue, fire suppression and emergency medical
services. The Holbrook fire department has neither the resources to provide additional train.ing nor
the equipment necessary for a response to the Baird-McGuire site. As the system states "Evaluate the
emergency response and medical resources available for hazardous waste site emergencies".
BACKGROUND
OSHA: In 1970 the Occupational Safety and Health Act (PL91-596) was enacted. The OSHAct
mandated the minimum standards for safety and health performance. Where states provided their
own standards that were at least equivalent to the federal requirements they operated their own
programs. Otherwise the federal programs applied. Two important distinctions in the OSHAct
dramatically affected the Fire Fighters Service.
COVERAGE: The act excluded coverage (requirements) for state, municipal and local government
employees. Subsequently in many states the provisions of the Act did not cover fire fighters.
EMPLOYEES: The act is directed for implementation in the work environment where an
employer/employee relationship exists. Non-compensated (volunteer) individuals do not fall under
the provisions of the Act.
Although the original OSHAct and standards did not have much impact on the fire service subsequent
performance requirements promulgated do dovetail with fire fighter work activities. In addition to
OSHA, the Environmental Protection Agency and US Department of Transportation have promulgated
safety and health standards that affect the fire service. The standards that were written did not
directly address professional fire fighters, and created a new phrase - Emergency Responders. A
needs assessment by OSHA justified the development and implementation of new standards dealing
with job performance elements that closely paralleled those of the fire fighters. However, fire
fighters were not covered by these standards in many states. Standards development resulted in the
publication of the following OSHA standards:
828
-------
29CFR 1910.155 (SUBPART L): Standards were published covering private fire departments
within that class of industry that maintains an on-site first line of defense. Typically they
were known as Fire Brigades but have subsequently been expanded into special teams such
as Hazardous Material Response crews, Chemical Spill response teams, etc. This standard
deals mainly with fire emergency situations.
29CFR1910.1200: The Hazard Communication standard was promulgated to assure that
workers would understand the hazards of the materials they may work with, or be exposed
to. Many states enacted legislation with a similar intent which also protects the general
population and which is commonly known as the "Right to Know" laws.
29CFR1910.120: This standard was developed in response to identified needs and in
collaboration with the Environmental Protection Agency. Commonly referred to as the
HAZWOPER standard, it provided requirements for preparation, protection and clean-up
processes at hazardous waste sites. Paragraph Q addresses Emergency Response activities
which closely parallel the class of activity that may involve fire fighters. In fact, since the
OSHA act, and subsequently OSHA standards did not apply to municipal employees it was
necessary to instruct EPA to develop a similar standard which would apply to municipal fire
fighters. The two standards are identical, it is only the authority for implementation that is
different.
TRAINING: A critical element of the standards is a requirement for training. HAZWOPER, Right
to Know, and Hazard Communication sections all contain detailed descriptions of the training needed.
In the five tiers of competency training required under the HAZWOPER (paragraph Q) standard are
specific measurement levels (ie - hours of training) and refresher training.
FIRST RESPONDER - AWARENESS LEVEL: Shall have sufficient training
FIRST RESPONDER - OPERATIONS LEVEL: Shall have at least eight hours training
HAZARDOUS MATERIAL SPECIALIST: Shall have at least twenty four hours training
HAZARDOUS MATERIAL TECHNICIAN: Shall have at least twenty four hours training
ON SCENE INCIDENT COMMANDER: Shall have at least twenty four hours training
TRAINERS (29CFR1910.120 (q)(7) shall have satisfactorily completed a training course for teaching
the subjects, such as the courses offered by the U. S. National Fire Academy, or they shall
demonstrate competency via academic credentials and experience of an equivalent nature. For
comparison purposes the OSHA Construction Standards require training for asbestos clean-up workers
in a course at an EPA Center or training equivalent to that presented by EPA
(29CFR1926.58(e)(6)(iii).
A proposal is being developed by OSHA whereby training required by the HAZWOPER standard be
via an OSHA certified training course. This would exclude the training required under
29CFR1910.120(q) in that OSHA has announced an intention to NOT review or certify training
courses for First Responders (fire fighters). Of course it is possible that OSHA recognizes the
historical competency in emergency response situations that the fire fighter possesses. Subsequently
the experience and training that fire fighters possess prior to undertaking of the curriculum delineated
in the OSHA standard places them in an advantageous position. Although the Fire Service has utilized
protocols, established SOP's, and generally conducted Emergency Response activities for years the
new emphasis on chemical emergencies is treated as if it is a new concept.
829
-------
EMERGENCY RESPONSE ACTIVITIES
The Fire Service has generally kept current with technology developments in all areas of emergency
response. Consider the following activities that were in place long before the requirements of SARA
and HAZWOPER were promulgated in 1986.
o EMERGENCY MEDICAL CARE: In almost every community the emergency response
medical care provisions have been maintained by the Fire Department. Where the fire
department formerly provided simply transport (ambulance) services they now provide on-site
medical care, in transit medical care and transportation of the victims. Staffing has now
improved from simple ambulance drivers to Emergency Medical Service Technicians (EMT)
to Para-medics and communication systems have developed access to shock trauma units.
o SPECIALIZED RESPONSE UNITS: Many municipal fire departments developed special units
for responding to emergencies such as trench failures and transportation incidents involving
chemicals. Where applicable the municipalities have maintained response units capable on
reacting to waterway incidents, helicopter units, and the like.
o INCIDENT COMMAND SYSTEM: An integral part of the fire service is the System
management of all incidents within the structure of an Incident Command System. It
establishes the protocols and Standard Operating Procedures (SOPs) to be followed in an
emergency response situation. As part of the system it ties in to other activities such as Pre-
incident planning, Recognition and Identification, Training, etc.
o PRE-INCIDENT PLANNING: In a way this might be described in the same vernacular as
the EPA developed site characterization process. It is a substantial effort - fire fighters do
a lot more than just battle fires. Of course this effort is on a community scale. In the
simplest of methods a pre-incident plan might state that all alarms at hospitals or hotels would
involve response by two engine companies and a ladder company. The fire service also has
a process called "pre-firing a building". By simulating various incidents the response protocols
are designed and if the actual incident takes place the plan, personnel and equipment are up
and functioning without delay. System safety processes such as fault tree and failure effect
mode are adaptable to analyzing and developing response protocols. While hazardous waste
sites employ a process plan for clean-up which contains the elements of safety, health and
community protection, the emergency responder must deal with potentials which are saddled
with uncertainties. Pre-incident planning turns possible chaos into a manageable response
activity.
An emergency response is not a simple containment or extinguishment function. Other
considerations are included in the pre-incident plan. The following case scenario will
illustrate the essential need to operate under a pre-plan concept supported with a high level
of data accumulation.
SHERWIN-WILLIAMS FIRE
A huge complex operated by a major paint manufacturing firm caught fire. Although the building
was protected with sprinklers and major fire divisions were provided with standard separations the
fire grew unabated. When the first responding units arrived on the property there was substantial
involvement in the structure and contents. A major component in the fuel was combustible and
flammable liquids in both large and small quantity containers. As the heat caused rupture; of the
containers the liquids spread like a lake, under fire doors and into adjacent fire divisions. The
ignition traveled along with the fuel and soon the total building was involved.
830
-------
Traditionally the fire department will give priority to rescue of endangered personnel, next establish
a containment process and finally enter into the extinguishment phase of the operation. The first two
goals were quickly accomplished but extinguishment posed a major problem. There were enormous
quantities of unburned solvents serving to fuel the continuing fire. The complex was located on a
parcel of land with a major influence on the water shed and next to the body of water that provided
municipal water supplies. Should the fire department apply hose streams to the fire the water would
serve to transport the contaminant into the community water supply; or at least dilute it so it would
enter into the ground, unburned, and affect the water shed properties. Subsequently the decision was
made to let the fire burn out and exercise a containment protocol. Naturally, the final result was the
total loss of the major buildings and stock at this complex and a dollar loss in the vicinity of One
Hundred Million Dollars.
The fire department's sensitivity to issues other than the traditionally steeped process of attacking and
extinguishing fires is a tribute to their flexibility in developing protocols concerned with community
and environmental factors. Where it was common for the fire department to dispatch a pumper to
the scene of a vehicle accident simply to hose down the roadway to remove spilled fuel or other
contaminants they now conduct an evaluation to determine if the contaminant should be washed away
or contained.
The Fire Service has clearly demonstrated a positive reactive progress in adjusting their protocols,
processes and responses in order to keep in step with both fire fighter safety and environmental
concerns. It is somewhat enigmatic that OSHA develops a standard for an industry that it did not
previously regulate, and even after publishing the standard (29CFR1910.120(q)) it is necessary for
EPA to publish the same standard in order that it would apply to most fire fighters. Most large
municipal fire departments have followed the concepts of what OSHA has termed voluntary
compliance. The Voluntary Protection Program (VPP) activity encouraged by OSHA is exactly what
has been happening in many major Fire Departments throughout the country. However, this process
has been going on for years - long before OSHA was even created.
Fire Departments, and Police Departments, have a structured personnel recruitment policy. Even
after selection the fire fighter must undergo extensive training and orientation. During the history
of the fire fighter's employment they are required to continue with refresher and upgrading training.
Promotions frequently involve additional training. With the development of Hazardous Materials
Response Teams members were exposed to frequent detailed training in most, probably all, of the
elements covered in the OSHA, EPA and FEMA prerogatives.
The growth of competency and proficiency of fire fighters has not been in an isolated environment.
Collaboration and assistance by Federal Agencies such as the National Fire Service Academy and
FEMA has been important and valuable. Development of structured training programs with funding
from the National Institute of Environmental Health Sciences(NIEHS) served to codify and structure
many of the training programs being utilized by the various fire departments. Unfortunately, there
are parallel developments by other groups with similar goals but somewhat different approaches.
Many years ago this country determined that standards were essential and organizations, such as The
Bureau of Standards, were created to develop standards for screw threads, nails, lumber sizes, etc.
In the development of standards for Emergency Response we have an emerging science where
different groups do almost the same thing but call it some thing else. In 1984 the Fire Fighters
identified the need to develop a standard for Competency of Responders to Hazardous Material
Incidents and Recommended Practices for Responding to those incidents (NFPA472/NFPA471).
These standards were finalized and approved by NFPA in 1988. The process within NFPA and ANSI
is to continue review and republish new versions every three to five years. Subsequently NFPA 472-
1989 indicates by date the particular version in use. Contrasting that are the OSHA Standards with
831
-------
their review and updating system running years behind changing technology. Consider the National
Electrical Code - OSHA adopted the 1970 issue, which remained in force for the next fifteen years,
even though during that period there were five major reissues of the standard by ANSI/NFPA.
During the period that NFPA was developing NFPA 471 & NFPA 472, NIOSH was developing a
Guidance Manual along the same lines in collaboration with OSHA, USCG and EPA. At the same time
OSHA was developing the HAZWOPER standard. The undeniable value of each of these documents
is only clouded by the use of differing terminology to say the same thing. Illustrations are:
OSHA/EPA FIRE FIGHTERS
Vapor Protective suit Level A
Splash Protective suit with Level B
SCBA
Splash Protective suit Level C*
Exclusion Zone Hot Zone
Decontamination Reduction Warm Zone
Zone
Support Zone Cold Zone
* Level C involves the use of protective clothing and an Air Purifying
Respirator (APR). Fire fighters rarely use APRs.
EMERGENCY RESPONSE TO HAZARDOUS WASTE SITES
Emergency response to an unregulated hazardous waste remediation site is an activity that may be
safely accomplished. Conditions that generally are the basis for classification as a hazardous waste
site are fully investigated. This data is essential to three processes used by the fire department. In
responding to an emergency situation the fire department first completes a Recognition and
Identification (R&I) operation. Extensive training has taken place to develop competency in this
activity. Recognition and Identification is a process whereby the fire department will evaluate the
available data and conditions at the emergency scene. In preparation for this the fire fighters are
trained in a variety of technical subjects such as:
o Vehicle classifications, shapes, markings, etc: The shape of a tank car is an indicator of the
commodity that may be involved. Department of Transportation markings and the UN
Classification System placards will indicate the class of commodity that a vehicle is
transporting.
o NFPA Marking System (NFPA 704) is a system whereby hazardous materials within a fixed
establishment are identified indicating their hazard classification and severity.
o Basic chemistry, reactive qualities of chemicals and warning characteristics of hazardous
materials are categories of learning that are essential to the fire fighter.
832
-------
The list of topics is endless and varies from community to community. Pre-incident planning adds
information to the data matrix which guides the fire fighter in recognition and identification.
Sometimes the name of a company will suggest the product line, such as Xpolsives Inc would
recommend caution in responding to the plant or vehicle emergency. However, AAA Manufacturing
Inc gives little warning as to the product and materials and this data might be available in the pre-
incident plan.
Obviously, the Site Characterization work completed on a Hazardous Waste site is an excellent
resource for the Fire Department Recognition and Identification process. It is essential for pre-
incident planning and Incident Management System (IMS) program implementation that all the data
is available and acted on. Typical emergency response incidents have visible conditions alerting the
fire fighters to the nature of the hazard. Fire, smoke, vapor, overturned tank truck, spilled liquid and
even odor are warning signs. However, a response to a hazardous waste remediation site might be a
medical emergency and the classic warning signs might not be present. Knowledge of the availability
of site characterization studies and the presence of a Site Specific Safety Program will dramatically
influence the fire fighter's decision making abilities.
Another process commonly utilized by the Fire Department is Pre-incident planning. Although it is
not possible to anticipate every class of emergency that may take place in a community the fire
department utilizes a system safety concept to catalogue most of the potential emergency situations.
Information is gathered from various sources and activities to develop the pre-incident plan. EPA
with all their sophistication in developing protocols fails to use the local resources. As the Baird-
McGuire clean up plans grew no effort was made to develop liaison with the Holbrook Fire
Department. A community liaison officer would have gone a long way in ensuring that should a need
arise for emergency assistance that the resources would be suitable for the site. Funding assistance
is available to local communities under Part 310 (CERCLA). In the case at Holbrook a complete pre-
incident plan would have identified the need for additional training and equipment.
Community inspection programs are intended to gather information about the nature of hazard levels
in the plants, stores, etc. Naturally there is an added benefit in ferreting out apparent violations of
local community rules and regulations.
Local rules and regulations frequently require businesses to obtain permits for storage and use of
hazardous materials. For many years this permit system applied mainly to flammable and combustible
liquids. With the increased environmental concerns, and recognition of other hazard classes such as
poison gas, the nature of pre-incident planning escalated. EPA published standards for reporting of
chemicals that have high hazard classifications and the minimum quantities that will trigger the
reporting requirements. The intention of these regulations is to make the information available to the
local Fire Departments and Emergency/Civil Defense organizations.
Liaison with local high risk facilities establishes a communication link. Naturally the fire department
has community maps but in some instances, such as a Health Care Facility, a more detailed site plan
would be needed. These locations might have separate buildings which contain storage of hazardous
materials, even a hazardous waste staging area, and other structures requiring a high priority of rescue
and evacuation. Many of us might think that house numbering is a convenience for visitors,
deliveries, etc; but, it actually is a fire code requirement to assist the emergency responder to quickly
identify the location where an emergency exists. Where a large facility might have only one street
number there is a requirement to number each building on the site.
Many communities have a system of permitting facilities with fire protection devices and alarm
systems to connect to the fire department. Pre-planning will take into consideration the nature of the
protection and locations within a facility. While the facility might have a zone alarm system the signal
833
-------
to the fire department is basically a location alarm. Subsequently the pre-plan may designate the
central location for the alarm panel to be the first area to be reached when responding to an alarm
notice. Obviously if the emergency is quite visible (smoke, fire, etc) this may change immediately
on arrival at the scene. In most areas when the water supply is temporarily cut off for repairs, or
sprinklers are shut off for service the facility is required to report this to the fire department. The
fire department will then place a special tag on the facility plan to indicate a modification in the pre-
plan.
Pre-planning is an integral part of fire department planning. Naturally it includes response protocols,
and a priority system. Staging of equipment is important. When the emergency apparatus arrives at
the scene the pumper would position it self in direct line with the municipal water supply (hydrants)
and Siamese building connections. Staging of the fire apparatus is an important element so that madly
arriving fire engines wouldn't block access and egress from the site. It certainly would be ineffective
if a fire engine blocked the entrance to a site. Every detail is planned in advance to maximize the
effectiveness of the response activities. In an emergency situation time is always the greatest enemy.
A quick response, well planned and executed will minimize the extent of the emergency.
The many elements involved in pre-incident planning are intended to dovetail with the operations
of the fire department. Extensive training is undertaken, on a continuing basis, by all fire fighters
to develop proficiency in emergency response and the pro-active functions. As the emergency starts
to develop the next Fire Department system is implemented. Small or large, any incident needs a
management system. In the fire service this is termed Incident Management System (IMS);
occasionally also known as the Incident Command System (ICS).
The purpose of an Incident Management System is to provide structure and coordination to the
management of emergency incident operations in order to provide for the safety and health cf fire
department members and others involved in the incident. The system consists of four najor
components; each with integral sub-elements with a proven history of effectiveness.
ADMINISTRATION: The overall fire department plan places the management of an IMS in the
Administrative function. An overall administrative activity involves all the day to day operation of
the fire department and includes elements such as recruitment, training, management of benefit
programs, etc. The implementation of the IMS is directed as an administrative requirement.
STRUCTURE: The fire department will develop a plan taking into consideration the size and
complexity of the available resources. It will take into consideration such elements as COMMAND
STRUCTURE, TRAINING, INTERAGENCY COORDINATION and other QUALIFYING
FACTORS. The flexibility of the program is obvious. Larger fire departments with many tiers in
the command structure would have the plan developed to take into account the First Response
command structure and subsequent changes in command as other senior officers might arrive on the
scene. In localities where there may be hazardous waste remediation sites, facilities controlled by
government agencies and other local Emergency or Disaster Councils the interagency coordination
with these groups is worked out in advance.
Many years ago, in the early days of World War II the world's largest Ocean going liner, the
Normandie, was at a pier in New York City undergoing renovation. A welding torch started a small
fire below decks. Unfortunately, there was no IMS in effect and a great deal of time was lost in
coordinating with the Coast Guard and Fire department. There was a lengthy period of time that
discussions took place as to who would have the overall command authority. As a result the ship was
lost and laid on the bottom at the pier for the remainder of the war.
834
-------
The EPA clean up site does not become an Island in the Sky. Although the Site Clean-up program
is fairly structured with an identified command structure the variables that will take place during an
emergency incident must provide for partial transfer of command in pre-identified areas. At the
Browns Ferry Nuclear Plant fire many years ago the local fire department arrived quickly, laid hose
lines and prepared to conduct suppression operations. However, the plant personnel insisted that
water should not be applied to the fire. The fire continued to grow as dialogue became heated
between the fire fighters and plant management. The facility was almost lost, with some significant
radiation contamination possible, when the major decisions were assigned to the fire department and
the incident was terminated without a major nuclear incident.
A natural coordination system that exists in almost every instance is the collaboration with the local
police department in traffic control and community evacuation. A Liaison is established to assure that
requests for assistance are coordinated.
The Incident Command System shall provide a series of supervisory levels that are available for
implementation to create a command structure. Naturally it will be dependent on the size of the
department. The modular sectioning of this structure will allow for application of only those series
of supervisory levels that may be required for a particular incident.
The major system component is at the Operational Level. The operational level consists of those units
that are directly involved in rescue, suppression and other primary missions. Part of the operational
level are the HAZMAT teams, EMS (ambulance) and specialty teams such as a support function for
refilling SCBA bottles. The HAZMAT team has both operational and support functions. The support
function would be involved in decontamination and logistical support. The basic system components
are:
o Operational
o Incident Commander
o Command Staff
o Planning functions
o Logistics
o Communications
o Staging
o Finance
The Incident Commander shall be responsible for the overall coordination and direction of all
activities at the incident scene or the major liaison where management of an incident is controlled by
another agency - such as EPA or the Coast Guard. In any event the Incident Commander is in charge
of all fire department personnel and in a coordinated effort he will direct his personnel. His
command staff will consist of supervisory personnel in charge of operational components and
planning, logistics, communications, staging and finance.
The Fire Department administrative, management and operational activities have been tested and
proven on a daily basis throughout the United States. The system approach works well on small
incidents, such as a home fire, as well at larger incidents at chemical plants. To date the typical
835
-------
response to hazardous waste sites has been for medical assistance. On these sites ordinary injuries
occur and those incidents of heat exhaustion and heat stroke take place as a result of working in
protective clothing.
There is a potential for a major incident at a hazardous waste remediation site and pre-planning is
essential. Most sites consist of ground and water contamination. The remediation process consists
of collecting and disposing of the hazardous chemicals. In this process there is a possibility of
bringing the contaminant to a collection area, increasing concentration levels and thereby increasing
the potential for area involvement.
The Fire Department has one final process in incident management. Small incidents are ended with
the Incident Termination - that being the time when all activities are concluded and they leave the
scene. The Termination Process also includes some post-incident activities. Information is collected
on the incident and processed for various follow up functions. One is simply for debriefing purposes
- the operational teams, etc, will review the incident to evaluate how it went. Improvements and
changes detected as a result of a review of the incident can then be incorporated into the training
process, review of Recognition and Identification, and Pre-Incident planning systems to determine
if they were effective.
In some communities there is a fire department charge for services and ambulance (EMT) services.
The Termination Process will include documenting the incident to assure that charges are made and
submitted. An interesting old law is the Fire Fighter's Rule. In many states the fire fighter who is
injured while responding to an emergency incident cannot sue the individual or entity whose
negligence caused the incident. There are some exceptions to the rule, notably those injuries or
fatalities that may take place as the result of a negligent release of a hazardous substance. In the
Termination Process a compilation would be completed regarding the expenses of dealing with the
incident. Where negligence is of such a nature as to grossly disregard responsibilities the fire
department would be prepared to document the charges and submit a bill to the offending parties.
Fire departments are a public service supported by the tax dollars but irresponsible behavior which
drains the resources of a community must be paid back. Many communities have adopted a rule that
if your fire alarm keeps malfunctioning and sending out false alarms you will be penalized $50,
perhaps more, after the second false alarm.
The system approach used in the fire service is very effective. Protocols are developed for every
category of emergency including those anticipated at a hazardous waste remediation site. "Fools rush
in where heros fear to tread". An entry to an emergency incident can be accomplished safely when
all data is available and a pre-incident plan coupled with an Incident Command System is employed.
Fire fighters are trained, competent experts in their field. Three fire fighters died in a high rise fire
in Philadelphia in February 1991. The unexpected took place in that fire. Elevators didn't work,
water pumps failed or were out of service and the difficulties grew because the building was not in
compliance with local ordinances. It is sad to hear of these stories; explosions in Kansas City, High-
Rise fires in Philadelphia; but it is time for recognition that the fire department is more than a group
of people at the end of a 911 telephone call. A system safety approach deals with probability and
possibility. The most effective program reduces possibility and probability to a low level; but,
incidents should not be unexpected, just unwanted.
836
-------
VI. POLICY/MANAGEMENT ISSUES
837
-------
Superfund-Program Standardization to
Accelerate Remedial Design and Remedial
Action at NPL Sites
M. Shaheer Alvi, PE ,Chief
Contract Management Section
Emergency and Remedial Response Division
U.S. Environmental Protection Agency - Region II
New York, NY 10278
(212)264-2221
Ming Kuo, PhD, PE
ARCS II Program Technical Support Manager
Ebasco Services Incorporated
Lyndhurst, NJ
INTRODUCTION
In response to persistent criticism regarding the slow pace of the Superfund Remedial Program, EPA
has developed the Alternative Remedial Contracting Strategy Contracts (ARCS) to accelerate the
progress of the remedial program and maintain control of project costs while ensuring the protection
of human health and the environment through effective, high quality response actions. Another goal
of the ARCS contracts is the rapid preparation and assembly of bid packages to complete the remedial
design and to expedite remedial actions.
The concept of Superfund Program Standardization (SPS) is to support the attainment of the ARCS
common goals to optimize quality, timeliness and cost-efficiency of the remedial response program.
EPA Region II, under the ARCS II Program, has developed various generic technical documents and
drawings to facilitate the preparation of documents and drawings efficiently and cost-effectively.
The standardized documents will technically provide consistency and uniformity of general
requirements and eliminate duplication and uncertainty thereby resulting in a significant savings of
time and cost in remedial design and remedial actions.
The generic technical documents and drawings were developed based on previous experiences arid the
existing database, files and documents accumulated under both the REM III and ARCS programs.
They utilized the experience gained on previous RD/RA, and the similarities with conditions at
previous sites and combined with a good understanding of the capabilities of remedial technologies.
They serve as a proper technical tool employed to ensure that sites of similar complexity are
remediated in a comparable manner to avoid the fatal flaws impacting remedial response actions.
This paper briefly describes the EPA Region II SPS background and discusses each standardized
documents for RD and RA and other generic documents for RI/FS. In addition, the general
approaches to utilize these standardized generic documents are also presented.
BACKGROUND
In late November 1989, EPA Region II tasked Ebasco Services Incorporated (Ebasco) under ARCS
II Contract to undertake the Program Standardization. The purpose of the program standardization
project was to develop the generic technical documents and drawings which are commonly applicable
838
-------
to the ARCS II Superfund Program. The most useful documents of general applicability and
commonly utilized technologies were selected for this standardization program. The SPS was
conducted in two phases. Higher priority standardized documents were prepared during Phase I
activities. The Phase II activities were put off for future implementation. Table 1 presents the scope
of work and associated standard documents.
Ebasco completed the Phase I activities and submitted all standardized documents to EPA for review
and comment in June 1990. Since then Ebasco has used these draft standardized documents in the
preparation of remedial design and remedial action documents as well as RI/FS reports. These
documents were provided to other ARCS II contractors for appropriate utilization.
DISCUSSION
The standardized documents and drawings provide a unified support basis for preparation of RD
specifications and drawings, RA plans and other RI/FS reports. The major components and
utilization of these generic documents are briefly discussed as follows:
A. REMEDIAL DESIGN SPECIFICATIONS AND DRAWINGS
1. GENERIC REMEDIAL DESIGN SPECIFICATIONS (GRDS)
The GRDS are prepared in accordance with the Construction Specification Institute
(CSI) format which is subdivided into 16 Divisions. These divisions form the
framework of the specifications and contain the technical requirements for the
category of work within each Division. Each Division is then subdivided into three
distinct groupings of related information (i.e. Part 1-General, Part 2-Product and
Part 3-Execution).
The GRDS is designed in a template format so that the boilerplate sections (Part
1-General) with generic description can be easily used in site-specific documents
with minor changes. The standard sections (Part 2-Product and Part 3-Execution)
can be incorporated by filling the site-specific information in the blanks. All
generic documents are available in PC diskettes in order to minimize typing
requirements.
GRDS of Carbon Adsorption Units includes Division 1-General Requirements,
Division 2-Equipment and Division 3-Mechanical as shown in Table 2. The
primary section is Section 11255-Activated Carbon Adsorption Unit which includes
Part 1-General, Part 2-Product and Part 3-Execution. Part 2 describes
specifications for equipment, material, fabrication and accessories. Part 3 includes
specifications for erection/installation, testing and inspections. Section 15010-Basic
Mechanical Requirements provide the support for carbon adsorption unit fabrication
and specifies piping, fitting, hangers/supports, joints, sleeves, cutting and patching.
GRDS of Packed Column Air Stripper includes the primary Section 11230-Packed
Column Air Stripper. Part 2 of the Section specifies the equipment components and
leaves blanks for site-specific dimensions. The major equipment components
include column structures and internals, water distributors, air exhaust ports and
moisture separator, Subpart 2.02 specifies column materials and packing materials.
Subpart 2.03 specifies the fabrication requirements of all column elements and
accessories.
839
-------
TABLE 1
PROGRAM STANDARDIZATION SCOPE OF WORK AND
ASSOCIATED STANDARD DOCUMENTS
PHASE I - COMPLETED
Task
Standard Documents
Remedial Design
la
Ib
Ic
Id
Generic Remedial Design Specification of
Carbon Adsorption
Packed Column Air Stripper
Pumps
Site Work
Remedial Design
2a
2b
2c
2d
2e
Remedial Action
3a
3b
Sc
3d
RI/FS
4
5
6
6a
6b
6c
6d
6e
PHASE II - FUTURE
Remedial Design
7a
7b
7c
7d
7e
7f
7g
Remedial Design
8a
8b
8c
Generic Remedial Design Drawings of
Carbon Adsorption System
Packed Column Air Stripping System
Pump Configuration
Extraction and Reinjection Well Details
Capping, Fence and Gate Details
Generic Health and Safety Plan for
Remedial Action
Generic Quality Assurance Plan for
Remedial Action
Generic Community Relations Plan for
Remedial Action
Generic Bid Evaluation Procedures for
Remedial Action
Generic Work Plan
Generic Field Sampling and Analysis Plan
Generic RI Subcontract Bid Package of
Drilling Services
Survey Services
Removal and Disposal of RI Wastes
Fence and Gate Installation
Cost Estimate Database for Coat Screening
for Feasibility Study
Additional Remedial Design Specification
of
Concrete
Masonry
Metals
Moisture (Dewatering)
Finishes (Painting/Coating)
Reactor/Clarifier/Thickener
Mixing Tank
Additional Remedial Design Drawings of
Butler Building Details
Erosion and Sediment Control Details
Access Roads and Temporary Storage Area
840
-------
TABLE 2
GENERIC REMEDIAL DESIGN SPECIFICATION OF CARBON ADSORPTION UNIT
TABLE OF CONTENT
DIVISION 1 - GENERAL REQUIREMENTS
Section 01005 - Specification Outline
Section 01010 - Summary of Work
Section 01065 - Health and Safety Requirements
Section 01070 - Abbreviation
Section 01080 - Identification Systems
Section 01200 - Project Meeting
Section 01300 - Submittals
Section 01400 - Site-Specific Quality Assurance Plan
Section 01510 - Temporary Utilities
Section 01660 - Testing, Adjusting and Balancing of Systems
Section 01720 - Project Record Documents
Section 01730 - Operation and Maintenance Manuals
Section 01735 - Final Inspection and Acceptance
DIVISION 2 - EQUIPMENT
Section 11255 - Activated Carbon Adsorption Unit
Part 1 - General
1.01 - Summary
1.02 - Related Sections
1.03 - Reference/Regulations
1.04 - System Description
1.05 - Design/Performance Requirements
1.06 - Submittals
1.07 - Quality Assurance
1.08 - Project/Site/Environmental Conditions
1.09 - Maintenance
Part 2 - Products
2.01 - Equipment
2.02 - Materials
2.03 - Fabrications
2.04 - Accessories
Part 3 - Execution
3.01 - Erection/Installation
3.02 - Testing and Inspections
DIVISION 15 - MECHANICAL
Section 15010 - Basic Mechanical Requirements
Part 1 - General
1.01 - Summary
1.02 - Related Sections
1.03 - Conditions
Part 2 - Products (Not Used)
Part 3 - Execution
3.01 - Piping Installation
3.02 - Installation of Fittings
3.03 - Installation of Hangers and Supports
3.04 - Installation of Joints
3.05 - Installation of Sleeves and Escutcheons
841
-------
TABLE 2 (Cont'd)
GENERIC REMEDIAL DESIGN SPECIFICATION OF CARBON ADSORPTION UNIT
TABLE OF CONTENT
Section 15060 - Pipes and Pipe Fittings
Part 1 - General
1.01 - Summary
1.02 - Related Sections
1.03 - Submittals
1.04 - Quality Assurance
1.05 - Project/Site/Environmental Conditions
1.06 - Operating and Maintenance Instructions
Part 2 - Product
2.01 - Material
2.02 - Pipe Insulation
2.03 - Valves
Part 3 - Execution
3.01 - Erection/Installation
3.02 - Testing and Inspections
GRDS of Pumps includes the primary Section 11211-Submersible Pumps, Section
11212-Sump Pumps, Section 11213-Horizontal Centrifugal Pumps, Section
11215-Vertical Turbine Pumps and Section 11216-Sludge Pumps. The key part of
each pump section is Part 2-Products which specifies the requirements of equipment
and accessories with blanks for site-specific information and dimensions.
GRDS of Site Work includes Section 02040-Dust and Vapor Control, Section
02090-Off-Site Transportation and Disposal, Section 02140-Aqueous Waste Hand ling,
Section 02200-Earthwork, Section 02210-Placement of Material and Final Cap,
Section 02220-Asphalt Cutting, Removing and Surfacing, Section 023600-Steel Piling
and Section 02900-Restoration of Site Vegetation. The primary part of site work
specifications is Part 3-Execution which specifies the construction requirements and
procedures. For example, the aqueous waste handling specifies dewatering, off-site
aqueous waste transportation/ disposal and on-site aqueous waste treatment/disposal.
2. GENERIC REMEDIAL DESIGN DRAWINGS (GRDD)
The GRDD are intended to develop an Automatic Computer Aided Design and
Drafting System (ACDD) incorporating the standard details common to most remedial
designs in the acceptable design drawing formats and files. These standardized
drawings were developed based on the existing drawing file of the previous RD/RA
work with any necessary modifications. The GRDDs completed in the Phase I
assignment include the detailed figures for a carbon adsorption system, packed
column air stripping system, pump configurations, extraction and reinjection wells,
capping, and fence/gate details.
As shown in Figures 1 and 2, the GRDD for the carbon adsorption system presents
a typical flow diagram and associated general equipment arrangement for a
two-train-3 vessel operation system. This GRDD shows all configurations of
drainage, compressed air, water inlet/outlet, wastewater and backwash water but
leaves the blanks for site-specific dimensions. The GRDD for the packed column air
stripping system shows all figures for the nozzle, stripping column, water distributor,
842
-------
z 5;
S 2
O O
I
£ > £
* 5 I
§
u> \r
w in
m ffi a. o m a a.
S * Z XE€
g 5
$ W
< d;
(D P=
s£
Jl
o
CE
S 3
S
i
84
-------
I8
in N
T3SS3A
Jo
gs_
".I
ti
o
tf
z.
°c
&
b K
-s
S c
i
g
^
1
§
U4
e
§
s
K
\n
S
S
I
-------
air blower, mist eliminator and electric/control panel. The GRDD for pump
configurations present most components of motor, pump, coupling, suction/inlet,
discharge, and pump base for centrifugal pump, submersible pumps, sump pumps and
vertical turbine pumps. The GRDD for extraction wells and reinjection well shows
details of bottom cap, pump, well screen, gravel filter, bentonite plug, borehole limit,
well riser and valve box. All dimensions are left as blanks for site-specific
information. Typical capping types and details include engineered soil cap,
engineered soil cap with synthetic liner, and pile supported structural cap. Each type
of cap shows the recommended thickness of the uppermost layer and vegetative cover,
drainage layer, impermeable layer, and compacted fill. The GRDD for typical chain
link fence and gate details present all components and standard dimensions for gate
post, concrete base, latch rod, wire fasteners, chain link fabricated mesh, end post and
turnbuckles, etc.
B. REMEDIAL ACTION STANDARDIZATION DOCUMENTS
1. GENERIC HEALTH AND SAFETY PLAN (GHSP) FOR REMEDIAL ACTION
The GHSP is developed to inform site construction personnel of the known hazards
associated with the RA and to ensure that the construction health and safety program
is performed in compliance with Federal, State and local laws including those set
forth by OSHA. The GHSP establishes consistency by listing of sections that are
common to all sites and are sufficiently flexible to enable the development of HSP for
divergent sites and hazards. The GHSP addresses the potential hazards, protective
measures, emergency response procedures, equipment required on site and specific
roles and responsibilities of site personnel.
2. GENERIC QUALITY ASSURANCE PLAN (GQAP) FOR REMEDIAL ACTION
The objective of the GQAP is to ensure implementation of the engineering/design
criteria and specifications in accordance with the contract procedures and
requirements by the contractors. The GQAP will serve to help contractors expedite
the preparation of construction QAP with more consistency and uniformity. The
GQAP provides detailed sections of contaminant migration, decontamination, control
and shipping of hazardous materials, performance verification, field testing,
inspection, deficiency control, sample validity and subcontractor control and
surveillance.
3. GENERIC COMMUNITY RELATIONS PLAN (GCRP) FOR REMEDIAL ACTION
The GCRP addresses site background, community profile/concerns, key issues and
community relations activities prior to and after remedial action. The GCRP serves
to aid construction management in preparation of site-specific CRP to keep local
public well-informed about the remedial action.
4. GENERIC BID EVALUATION PROCEDURES (GBEP) FOR REMEDIAL ACTION
The GBEP provides a reliable and acceptable methodology and evaluation criteria for
procuring a RA contract through (1) one step competitive negotiation turnkey
procedures, (2) two step sealed proposal/bidding and (3) a combination of one step
and two step approaches. The technical evaluation merits include contract
management plans, project experience, sequence of construction and construction
845
-------
schedule. The proposal price is factored into the quality score. A cost/technical score
ratio or total point score forming the basis for recommending contract award.
C. REMEDIAL INVESTIGATION AND FEASIBILITY STUDY DOCUMENTS
1. GENERIC WORK PLAN (GWP) AND GENERIC FIELD SAMPLINGS AND
ANALYSIS PLAN (GFSAP)
The GWP is organized according to the table of contents presented in the EPA's April
1989 RI/FS Guidance (Reference 1). The GFSAP has been developed based upon the
ARCS II Field Technical Guidelines (Reference 2) and EPA Region II Quality
Assurance Manual (Reference 3). Both documents are not intended for use as a
site-specific WP and FSAP, rather, they shall be applied as boilerplate material to
facilitate the WP and FSAP development processes. They shall be edited as necessary
to satisfy the site-specific conditions. The GWP addresses site background, scope of
work, field investigation, feasibility study, project organization and schedule. The
GFSAP addresses general requirements of field sampling and analysis program,
statement of procedures, QA/QC, sampling packaging and shipment and field
changes/corrective actions.
2. GENERIC REMEDIAL INVESTIGATION SUBCONTRACT BID PACKAGES
The generic RI subcontract bid package consists of complete non site-specific RI
subcontract service inquiry documents which have been developed for drilling
services, removal/disposal of RI wastes, survey services and fence/gate installation.
The generic subcontract bid packages are intended to minimize duplication of effort
and uncertainty surrounding the content and format of RFPs.
The generic RI subcontract inquiry contains two major portions, i.e., contractual
requirements and statement of work. The contractual requirements include the
general specifications of instruction to bidder, subcontract agreement, representatives,
certifications and other statement. The statement of work consists of technical
specifications and requires various levels Of site-specification input. A typical outline
of the drilling services solicitation package is presented in Table 3.
The major technical specifications for drilling services include a generic statement of
work of soil borings and monitoring well installation, well development and
decontamination/containment. The survey services technical specifications include
sample and well location survey, topographic survey/mapping and survey report. The
major RI waste removal/disposal technical requirements include sample collection,
waste characterization, manifest form, transport/treatment/ disposal of bulk materials
and drummed materials.
Hazardous and non-hazardous material classification and associated ultimate
disposition are discussed for each RI waste. Waste types, such as F, P, K, etc. and
applicable disposal technologies or landfill types, either RCRA Subtitle D or Subtitle
C are also described. The major fence/gate installation technical specifications
include new fence installation, existing fence repair, existing fence relocation and
fence materials such as ports/rails, fence fabric, tension bars and gates, etc.
A cost estimate database was developed based on Ebasco internal data, published
literature (e.g., EPA's CORA Model) and available vendor information for cost
screening purposes in the preparation of feasibility study. Cost data is presented in
846
-------
TABLE 3
GENERIC DRILLING SERVICES SOLICITATION PACKAGE
OUTLINES OF CONTENT
I. SUBCONTRACT AGREEMENTS
II. STATEMENT OF WORK AND PROPOSAL REQUIREMENTS
A. Project Description
B. Special Conditions
C. Technical Specifications
1. Codes and Standards
2. Soil Boring and Monitoring Well Installation
3. Decontamination
4. Well Development
5. Containment
6. Rejected Borings and Installations
7. Portable Water Supply
8. Record
9. Price Proposal Form
10. Engineer's Control
ATTACHMENTS
A. Health and Safety Plan
B. Quality Control Forms
C. Subcontractor's Medical Surveillance Program
847
-------
a unit cost form, in terms of dollars per unit operation. The report also discusses a variety of factors
influencing these unit costs. The cost database presents cost data for treatment technologies most
commonly applicable to source control and management of migration as shown in Table 4. A
hypothetical case is presented to demonstrate the application of the cost estimate database.
D. UTILIZATION OF PROGRAM STANDARDIZATION DOCUMENTS
In general, the various sections of standardized documents are grouped in three categories: boilerplate
sections, standard sections and explanatory sections. Boilerplate sections contain non-site-specific
text that has been used previously and can be used directly without revision. Standard sections are
designed as a template format having standard sentences and wordings common to most sites with
blanks that need to be filled in or revised to reflect site-specific conditions and specific project
approaches. Explanatory sections identify for the preparer the site-specific information that would
need to be included in the respective sections. An example is usually provided for the explanatory
sections.
The generic remedial design specifications (GRDS) are also facilitated by three types of guides, i.e.,
a general statement, a specific statement and an explanatory statement. Common information is
consolidated in general statement, where as site-specific information is provided in specific statement.
The GRDS is written in the imperative mode and, in some cases, in a streamlined form. The
imperative language is directed to the subcontractor, unless specifically noted otherwise.
The generic remedial design drawings (GRDD) are developed using a computer based model, an
Automatic Computer Aided Design and Drafting system (Auto CADD). All data input is filed into
the CADD so that it can be extracted for graphs and tables with modification. The Auto CADD
standard details, drawings and files can be retrieved for ease of reference and/or modified for new
drawings. Any site-specific data such as dimensions and sizes are not included in the GRDD and will
be provided by design engineers based on site requirements. The filed standard details can be easily
reviewed, updated and revised to reflect the site-specific conditions.
CONCLUSION
The program standardization documents were developed, consolidating the similarities with conditions
at previous sites and taking advantages of experience gained on previous RDs and RA. From an
environmental standard, the GRDD and GRDS have addressed all necessary environmental elements
and are in full compliance with ARARs. From a technical standard, the GRDD and GRDS have met.
all performance standards with high constructi- bility, practicability, clarity, biddability and
acceptability.
These standardized documents can be used as an effective tool to coordinate interaction among all
disciplines involved in the project. They can be used as a basis to ascertain the RD and RA
requirements resulting in minimal review, modifications and revisions. They would compensate for
learning curves and inexperience which in turn would enable the engineers to focus on the
site-specific appropriate, substantive problems. The use of standard documents would avoid time and
cost delays, last minute disagreement and misunderstandings.
All the documents described above are available in Word Perfect format or Auto Computer Aided
Design and Drafting (CADD) format for expeditious adoption to the Region specific site situations.
REFERENCES
1. Guidance for Conducting Remedial Investigations and Feasibility Studies under CERCLA,
EPA OSWER 9355.3-01, April 1989.
2. Field Technical Guidelines, ARCS II Program, EPA Contract 68-W8-0110, June 1989.
3. CERCLA Quality Assurance Manual, EPA Region II, Final Copy, Revision 1, October 1989.
848
-------
TABLE 4
COST ESTIMATE DATABASE FOR COST SCREENING FOR FEASIBILITY STUDY
SOURCE CONTROL AND MANAGEMENT OF MIGRATION TECHNOLOGIES
I. SOURCE CONTROL TECHNOLOGIES
A. No Action
B. Containment
Capping, Vertical Barrier, Excavation
C. Physical Treatment
Mechanical Aeration, Enhanced Volatilization, In -Situ Soil Flushing,
In-Situ Vacuum Extraction
D. Chemical Treatment
Chemical Stabilization and Solidification, Chemical Extraction
E. Thermal Treatment
Incineration, In-Situ Vitrification
F. Biological Treatment
In-Situ Biodegradation
G. Disposal
Off-Site Waste Landfill, On-Site Waste Landfill
II. MANAGEMENT OF MIGRATION TECHNOLOGIES
A. Groundwater Extraction
B. Physical Treatment
Coagulation/Flocculation/Precipitation.AirStripping^larification,
Filtration, Ion Exchange, Carbon Adsorption, Reverse Osmosis,
Sludge Dewatering
C. Chemical Treatment
UV-Chemical Oxidation
D. Biological Treatment
Aerobic Biodegradation, Anaerobic Biodegradation, In-Situ
Biodegration, Powdered Activated Carbon Enhanced Activated
Sludge
E. Discharge
Off-Site Discharge to Publicly Owned Treatment Works (POTW)
849
-------
Environmental Protection Agency
Indemnification
for
Remedial Action Contractors
Kenneth W. Ayers, P.E.
Design and Construction Management Branch
Hazardous Site Control Division
Office of Emergency and Remedial Response
401 M Street S.W.
Mail Code OS-220W
Washington, DC 20460
(703) 308-8393
DISCLAIMER
This report has undergone a broad initial USEPA peer review. However, it does not necessarily
reflect the views or policies of the Agency. It does not constitute any rulemaking, policy or guidance
by the Agency, and cannot be relied upon to create a substantive or procedural right enforceable by
any party. Neither the United States Government nor any of its employees, contractors,
subcontractors or their employees makes any warranty, expressed or implied, or assumes a ay legal
liability or responsibility for any third party's use or or the results of such use of any information or
procedure disclosed in this report, or represents that its use by such third party would not infringe
on privately owned rights.
We encourage your comments on the utility of this paper and how it might be improved to better
serve the Superfund program's needs. Comments may be forwarded to the attention of Kenneth
Ayers, Design and Construction Management Branch, USEPA, Mailcode OS-220W, Washington DC
20460.
INTRODUCTION
As of the last update on February 11, 1991 (56 FR 5598), 1,189 hazardous waste sites had been
incorporated into the National Priorities List (NPL). These 1100+ sites represent the most serious
threats to human health and the environment from uncontrolled hazardous wastes discovered to date.
Remedial investigations and feasibility studies (RI/FS) are currently on-going at over 700 of the sites.
Remedial designs are under development at approximately 200 additional sites. Finally, remedial
actions are under construction at another 225 sites. The cost of this work to date is over $7.4 billion
with an estimated additional $25 billion needed to complete the work at sites presently listed on the
NPL.
To perform this work, EPA relies heavily on assistance from response action contractors (RAC). In
providing the assistance to EPA, these RACs perform site assessment work, conduct RI/FSs, develop
remedial designs, and oversee and implement remedial actions. As with any engineering or
construction activity, there are elements of risk associated with each of these activities. One of the
primary risks associated with work at hazardous waste sites is the accidental and uncontrolled release
of toxic compounds from the site to the surrounding environment.
To provide protection against losses due to claims for damages resulting from their activities, most
firms purchase liability insurance policies which transfer, for a cost, the risks of loss from the
850
-------
company to the insurance underwriter. However, in the hazardous waste field, adequate and
affordable insurance is not available to cover claims for environmental and health damages resulting
from releases caused by work at Superfund sites. To enable contractors to work for the Agency under
the Superfund program, EPA is authorized to provide indemnification (Indemnification is an
agreement whereby one party agrees to reimburse a second party for losses suffered by the second
party) to RACS for negligence against pollution liability claims arising from remediation activities.
BACKGROUND
Section 119 Response Action Contractors, of the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA)(PL 96-510), as amended by the Superfund Amendments
and Reauthorization Act (SARA)(PL 99-499), authorizes EPA to provide indemnification to response
action contractors performing work at NPL or removal sites. Section 119 was added to CERCLA by
Congress as part of the 1986 amendments in response to an outcry from the RAC community for
pollution liability protection. This outcry arose due to the unavailability of pollution liability
insurance from private sector sources.
In defending their lack of participation in this segment of the market, the insurance underwriters
cited a number of reasons for their unwillingness to provide pollution liability coverage. The major
reason was the risk of large claims for "catastrophic" failures resulting in extensive damage to human
health and the environment. Their fear was that these types of failures could easily result in claims
surpassing $100 million per incident. When this fact was coupled with the litigious nature of the
environmental field, many underwriters declined to issue pollution policies.
A second and equally formidable reason cited by the insurance industry was the imposition of strict
liability standards by the courts. Under strict liability, any entity involved in "ultrahazardous"
activities at the site of a release may be held liable for all costs associated with the release without a
judgement of negligence against them. Damages associated with the release may have occurred on
or off the site. The insurance companies feared that in the future strict liability judgements could
render them the only viable "deep pocket" for legal actions stemming from the site.
Finally, many underwriters expressed the fact that reinsurers had withdrawn from the market due to
record losses posted by the industry in the early 1980s. This resulted in a down turn in the industry
with firms declining to underwrite relatively small high risk portions of the insurance market such
as hazardous waste remediation.
In addition to the lack of pollution liability insurance, RACs also cited several other reasons for
indemnification. The first was the technical risks the RACs accept when they work at a Superfund
site. These include:
1) Work with hazardous and toxic compound and mixtures of these compounds,
2) The uncertainty of innovative or untried technologies,
3) The inherent uncertainty associated with underground work, and
4) Political pressures from outside sources.
As with the insurance companies, RACs face the prospect of law suits being brought against them by
third parties for damages associated with their work at a Superfund site. In addition to the potential
for strict liability, negligence, or theories of liability, RACs may be jointly and severably liable. This
85J
-------
means if the RAC is found to be liable for a portion of the damages, the plaintiff may collect the
entire judgement from him.
SECTION 119
Prior to the enactment of SARA, RACs working for EPA were indemnified through EPA's inherent
contracting authority. This was very limited indemnification for third party liability and defense
costs and did not cover gross negligence or willful misconduct.
It was in this uncertain environment that Section 119 of SARA was enacted. While Section 119
attempted to remedy many of the RAC complaints, it did not absolve the RACs of all potential
liabilities. The amendment to CERCLA did provide the following:
1) Exempted RACs from strict liability under all Federal laws for injuries, damages,
costs, and other liabilities related to release of hazardous substances, pollutants, or
contamination, unless RAC was negligent, grossly negligent, or guilty of intentional
misconduct,
2) Established a negligence standard for RAC liability under Federal law,
3) Provided discretionary authority to extend indemnification against pollution liability
for negligence, and
4) Established a funding mechanism.
Equally important is what Section 119 did not do:
1) Pre-empt State strict liability, and
2) Provide coverage for treatment or disposal facilities governed by the Resource
Conservation and Recovery Act (RCRA).
Section 119 also specified the requirements that a RAC must meet to be eligible for EPA
indemnification. The three requirements listed are:
1) Potential liability exceeds or is not covered by adequate insurance available at a fair
and reasonable price,
2) The RAC must have made diligent efforts to obtain pollution lability insurance, and
3) If the RAC is working at more than one facility, it must perform diligent efforts each
time it begins work at a new facility.
The final requirement of Section 119 was that the President (EPA) would promulgate regulations
under the section. Prior to promulgation of the regulations, the President (EPA) would develop
guidelines for the implementation of the requirements of the section.
852
-------
INTERIM GUIDELINES
OVERVIEW
On October 6, 1987, EPA's Office of Solid Waste and Emergency Response (OSWER) issued OSWER
Directive 9835.5 "EPA Interim Guidance on Indemnification of Superfund Response Action
Contractors under Section 119 of SARA" to establish temporary procedures to provide indemnification
to RACs under the authority of Section 119. The guidelines, issued under the authority of Executive
Order 12580 (52 FR 5923, January 29, 1987 which delegated authority to indemnify RACs from the
President to EPA, were distributed as interim to allow EPA to provide indemnification under Section
119 while proceeding in a deliberate manner to establish final guidance.
The interim guidelines were developed around four key points:
1) The combination of protection from Federal strict liability and RAC indemnification
would provide adequate incentive for contractors to work for the Superfund program,
2) The indemnification would be an adequate substitute for insurance,
3) Indemnification would be an interim measure until the private insurance market
rebounded, and
4) The indemnification did not create a disincentive to the private insurance market.
These points were to also form the basis for the formulation of the final guidelines.
PROVISIONS OF THE INTERIM GUIDELINES
The interim guidance stated that EPA had determined that adequate private insurance was not
available at a fair and reasonable price, thus the Section 119 basic requirement that private sector
insurance be unavailable for RACs to be eligible for EPA indemnification was satisfied.
Additionally, the guidelines established no upper limits for claims under Section 119, prescribed a
$100,000 deductible for each claim filed, and did not establish any term of coverage or "tail" for EPA
indemnification to expire once it was granted. In addition to providing model contract clauses for
indemnification, the guidelines required all contracts incorporating the model clauses under its
authority to be to be modified by mutual agreement of all parties to the contract within 180 days of
promulgation of final guidelines. Requirements for RACs seeking EPA indemnification were also
delineated.
Requirements for RACs seeking indemnification under the interim guidelines included:
1) Written proof of diligent efforts must be provided to EPA within 30 days of contract
award,
2) If insurance was purchased, a copy of the policy must be provided to EPA, and
3) Additional diligent efforts must be performed every twelve (12) months if insurance
was not purchased.
The guidelines also presented mechanisms for EPA indemnification to be granted to RACs working
for States, other Federal agencies, and Potentially Responsible Parties (PRPs). For sites managed for
853
-------
EPA by the US Army Corps of Engineers or other Federal agencies, the contractor working for that
agency would be indemnified by EPA as if the contractor were working directly for EPA.
The exclusion of treatment facilities governed by RCRA regulations was extended to Publicly
Owned Treatment Works (POTWs). Although POTWs were not explicitly excluded from Section 119
coverage, EPA excluded them as a policy decision to be consistent with the intent of the RCRA
exclusion.
FINAL GUIDANCE
CONTENTS
In October 1989, nearly three years after EO 12580 delegated indemnification authority to EPA, the
Agency issued in the Federal Register for public comment Proposed Final Indemnification Guidance
(54 FR 46012, October 31, 1989). When compared to the liberal provisions of the interim guidance,
the proposed final guidance severely restricted the indemnification available to RACs. The proposed
guidance limited the maximum coverage per contract, imposed substantially higher deductibles, and
limited the term of coverage to ten years. The guidance called for a minimum amount of insurance
to be purchased by contractors each year and that this amount increase by 25% each year with the
anticipated result of the private sector eventually providing all pollution lability coverage allowing
EPA to cease offering indemnification. One final provision was that all existing post-SARA
indemnification agreements must be retroactively brought into compliance with the terms of the final
guidance.
Some of the specific points of the proposed guidance are as follows:
1) RACs were covered if found negligent; however, if a mixed judgement (a finding of
both negligence and strict liability) were handed down, the RAC would not be
covered,
2) Maximum coverage for cost reimbursement contracts was set at $50,000,000 per
contract.
3) Deductibles for cost reimbursement contracts were set at $1,000,000 per occurrence
or claim with no aggregate limit,
4) Coverage for fixed price contracts was set on a sliding scale which was to be factored
into the bid evaluation, and
5) A ten year post-completion term was established for all agreements.
Needless-to-say, the response to the proposed guidelines was overwhelming with over two hundred
comments, requiring over 40 pages to document, received. Unfortunately, the comments were
virtually all negative. They stated that the limits were too low, the deductibles too high, the term too
short, and the fixed price proposal unworkable. Based upon this negative feedback, EPA decided to
delay finalizing the proposal and to reconsider some of the elements.
CONSULTATIVE PROCESS
After completing a thorough analysis of the comments and conducting discussions with many of the
interested parties, EPA decided to employ a consultative process to solicit more specific feedback.
Endispute, Inc. was retained to organize and convene a one-day, facilitated session between EPA and
854
-------
a select group of interested and affected organizations. The purpose of this session was to attempt
to clarify the positions held by each party. EPA made it clear to all participants that the use of the
consultative process was not a prelude to a negotiated rule making and was for informational purposes
only.
Prior to convening the meeting, Endispute interviewed members of each organization slated to attend
the session. These interviews were designed to assist Endispute in formatting the meeting to allow
the concerns of all parties to be expressed.
The consultative session was held in Washington, DC on November 19, 1990. Representatives from
the RAC community, insurance brokers and underwriters, other Federal agencies, as well as EPA
were present. The points raised by the participants were essentially the same as those offered in the
written comments to the proposed final guidelines. The meeting did serve to "clear the air" and assure
the RAC community that EPA was aware of their concerns and attempting to address them in the
guidelines.
CURRENT STATUS
Following the consultative session, EPA reconvened its Indemnification Taskforce to revise the
proposed final guidance based upon the written comments and insights from the facilitated session.
The taskforce met routinely over several months and was able to reconcile many of the issues. Several
issues on which the taskforce was not able to reach consensus were elevated to management for
decisions. The final guidelines are currently ready to enter EPA's formal, internal review process and
then will be sent to the Office of Management and Budget (OMB) for final review prior to issuance.
At this time, EPA does not believe that the guidelines will be proposed for additional public comment
prior to becoming effective.
In addition to the final guidelines, an accompanying set of administrative guidelines are being
developed. The purpose of the administrative guidelines will be to provide the specific details and
instructions necessary for EPA staff to interpret and apply the guidelines equitably and consistently
throughout the program and across the regions.
Since the final guidelines have not completed EPA's internal review process, the specific details are
not releasable to the public. However, some of the basic components of the package that will be
forwarded to OMB for review can be discussed:
1) The final guidelines will contain well defined limits to the amount of indemnification
available to RACs on a per contract basis,
2) The deductibles will be on a sliding scale with higher deductibles for higher contract
limits,
3) A definite term of coverage (tail) will be set,
4) The incorporation of indemnification requests in bid evaluations for fixed price
contracts has been dropped, and
5) All post-SARA contracts must be modified to include the provisions of the new
guidelines.
855
-------
POTENTIAL PROBLEMS
The final guidelines could have substantial impacts upon both EPA and the RAC community. First
the potential RAC problems:
1) It is likely that the availability and the limits of EPA indemnification will be greatly
reduced from the uncapped limits currently provided. This reduction will require
RACs to rethink their current operating procedures and their future plans,
2) All RACs with current indemnification agreements must enter into negotiations with
EPA to incorporate the new guidelines into their existing contracts. This will require
time and effort by the RACs and may cause them to rethink their willingness to
continue to work for EPA, and
3) RACs must develop a strategy to deal with any subcontractors that have been
extended indemnification through the RAC's contract since the new limits will include
any pass-through indemnification.
Potential problems for EPA are:
1) The time and resources to negotiate the new guidelines into all existing contracts (this
includes contracts let by the US Army Corps of Engineer, the US Bureau of
Reclamation, and any other Federal Agency acting in behalf of EPA),
2) The impact on the Superfund program if some of the RACs refuse to accept the new
guidelines and their contracts are terminated. This could stop on-going work and
cause a severe shortage of contractors for the short term, and
3) The cost of doing business could increase substantially as RACs seek to protect
themselves as the risks from pollution liability are reallocated.
DEVELOPMENTS OUTSIDE THE FINAL GUIDANCE
SURETY AMENDMENT
Prior to the present construction season, the Superfund program had been experiencing a decline in
the number of bidders or proposers for many of the remedial action projects under solicitation. This
decrease in competition increased the costs of projects, and if not addressed, could ultimately have
impacted the quality of remediation work being performed. In response to this trend, EPA tasked
the US Army Corps of Engineers to explore the issue and provide recommendations for corrective
actions. The US Army Corps of Engineers issued their findings in Hazardous and Toxic Waste
Contracting Problems: A Study of the Contracting Problems Related to Surety Bonding in the HTW
Cleanup Program. The main finding of the study was that fewer firms were competing for Superfund
work due their inability to secure the necessary bonding required for the contracts. The difficulty
in securing bonds was stemming from the sureties' perception of their potential liability to become
the last "deep pocket" for pollution liability claims when providing performance bonds.
EPA attempted to address this issue in two ways. First, meetings were held with representatives of
the surety industry to explain their liability under CERCLA when providing performance bonds and
to try to allay their fears. Second, EPA, acting through the US Army Corps of Engineers, attempted
to reduce the amount of bonding required to adequately protect the Governments's interests by
856
-------
utilizing various contract types and phasing projects. These attempts did not satisfy the surety
industry.
The surety industry approached Members of Congress to amend CERCLA to allow indemnification
to be extended to surety firms providing performance bonds for Superfund work. Congress agreed
with the sureties' arguments and passed an amendment to CERCLA (Section 1 of Public Law 101-
584) in October 1990. The President sign the bill into law on November 15, 1990. This amendment
limited a surety's liability under the bond to the face value of the bond and extended eligibility for
EPA indemnification to surety firms when they elect to complete the contracted Superfund work to
fulfill their obligations under a bond issued to a defaulting contractor.
REVISED US ARMY CORPS OF ENGINEERS APPROVAL PROCEDURES
When potential contractors prepare proposals and bids in response to solicitations for work, they
invest considerable time and money. Additionally, each proposal or bid must be accompanied by a
bid bond which signifies the contractors good faith to perform the specified work and provides the
government with funds to resolicit if the contractor refuses to accept the contract. One problem with
this typical scenario is that for Superfund work RACs face one final hurdle they cannot control. This
hurdle is approval by EPA to extend indemnification to the contractor. In many cases without EPA
indemnification, contractors are unwilling to risk their corporate assets. If the contractor is the
successful proposer or bidder and EPA refuses to extend indemnification, the contractor is forced to
forfeit its bid bond if it refuses the contract due to potential liability.
Since the decision to extend or not to extend indemnification is out of the contractors control, EPA
and the US Army Corps of Engineers have agreed to test a modification to the normal indemnification
approval process to allow a contractor, providing it has met all other requirements of the solicitation,
to refuse a contract if indemnification is not approved and not forfeit its bid bond. This process is
being tested for one solicitation. Based upon the results of this test and the final indemnification
guidance, the process will be continued, modified, or discontinued.
Under current procedures, a contract is awarded and then the contractor performs diligent efforts and
indemnification is granted based upon the results of the diligent efforts. For the test procedures,
potential contractors will be asked to perform diligent efforts prior to contract award. EPA will
evaluate the contractors efforts and determine if indemnification will be offered prior to award of
the contract. If the contractor has met all other requirements of the solicitation and EPA declines to
approve indemnification for the contractor, the contractor will be allowed to withdraw from the
solicitation and not forfeit the bid bond. If indemnification is approved, the contractor will be issued
a letter granting indemnification immediately after the contract is signed.
DILIGENT EFFORTS
EPA has initiated two efforts to improve the diligent efforts process while awaiting the final
indemnification guidelines. The first is the internal EPA approval process. Since the approval of
indemnification has been delegated to the Director of the Hazardous Site Control Division (HSCD),
responses to requests for indemnification and insurance purchases for contractors working directly
for EPA or another Federal agency acting for EPA are now handled directly between HSCD and the
contracting officer for the solicitation/contract. In the past, all correspondence was routed through
the Procurement and Contract Management Division (PCMD). PCMD is now furnished with a copy
of all correspondence. This streamlined approach has reduced the review and approval process by
several weeks. Additionally, HSCD has provided guidance to the field on the minimum information
needed to review and make a decision on extending indemnification to a contractor. This guidance
857
-------
has improved the submissions and allowed HSCD to respond without requesting additional
information.
The other effort underway to improve the diligent efforts process is the development of a Quick
Reference Fact Sheet clearly explaining the process and what is required in the contractor's
submission. This fact sheet will establish consistency across contracts and assist both contracting
officers and contractors in reviewing and preparing requests for indemnification.
INSURANCE
Over the last two years pollution liability insurance has become more available to RACs. Currently,
two underwriters, American Insurance Group (AIG) and Reliance National Insurance (Reliance), are
offering pollution liability insurance. The usual policies offered by these firms are claims-made, one
year policies with no tail; however, several recent policies have offered one or two year tails. The
policies have limits between $1,000,000 and $5,000,000 with deductibles of $100,000. The rates
average approximately $2.50 per $100 of gross receipts for the contract covered.
In an attempt to stimulate the private sector, EPA has approved the purchase of over twenty policies
over the last several years. To date most policies have been site specific; however, recently EPA has
approved the purchase of contract-wide policies for several ARCS contracts. These policies provide
automatic coverage for all work, except remedial actions, performed under the contract. To expand
upon this trend, EPA is currently negotiating with several firms that have multiple ARCS contracts
to purchase a single policy to cover all the firms ARCS contracts. These contract-wide and multi-
contract policies will greatly reduce the cost of insurance premiums.
CONCLUSION
While the final picture of EPA's indemnification process is still unclear, it is certain that the new
guidelines will drastically alter the assignment of risk from pollution liability suits. Until the new
guidelines are finally promulgated along with their accompanying administrative guidance, the final
impacts on the RAC community and the Superfund program can not be determined.
858
-------
Innovative Design Review and Scheduling Tools:
Potential Benefits to HTW Remedial Projects
Gregory W. Bridgestock
U.S. Army Construction Engineering Research Laboratory
P.O. Box 4005
Champaign, IL 61826-4005
(217) 373-6744
Dr. Diego Echeverry
U.S. Army Construction Engineering Research Laboratory
P.O. Box 4005
Champaign, IL 61826-4005
(217) 373-6710
Dr. Simon Kim
U.S. Army Construction Engineering Research Laboratory
P.O. Box 4005
Champaign, IL 61826-4005
(217) 373-7269
859
-------
1 Introduction
A task of great magnitude is facing the United States for
restoring its contaminated sites. Conservative estimates indicate
a cost of hundreds of billions of dollars to accomplish this
remedial work. Furthermore, hazardous and toxic waste (HTW)
remedial projects involve several challenging characteristics:
(1) the hazardous nature of handled materials; (2) the need to
utilize cutting edge restoring technologies; and (3) uncertainty
on the degree of contamination or amount of contaminated
material.
Because of these demanding characteristics, substantial
time delays and cost overruns unfortunately are common
occurrences on HTW remedial projects. This paper will discuss
several specific tools being developed at the U.S. Army
Construction Engineering Research Laboratory (USACERL) to enhance
the management of traditional construction projects and explore
how these tools, if properly adapted, can help decrease the time
and cost growth of HTW remedial projects.
One such tool provides all project team members with
systematic access to customized checklists containing
biddability, constructibility, and operability (BCO) issues which
need to be examined on a project. Since BCO issues comprise 75
percent of the pre-final reviews conducted by the Environmental
Protection Agency (EPA) before construction is initiated on any
HTW remedial project, it appears this system will lend itself
well to helping improve the execution of HTW work.
Other tools being developed at USACERL facilitate the
estimation of construction project durations and the generation
of construction schedules at early design stages. It is believed
that the application of these same concepts to HTW remedial
projects will result in improved time estimation and time control
tools which will translate into cost savings on HTW projects.
2 Background
2.1 Current approach for Design Review of traditional
construction
2.1.1 Problems associated with the Design Review process
Facility acquisition and/or infrastructure revitalization
is a complex design and construction process that involves many
specialists in widely diverse fields. The accomplishment of this
process is further complicated by the unknowns of site and as-
built conditions. These complexities contribute to the develop-
ment of contract documents that cannot be understood, bid,
administered and enforced (biddability) along with the design of
860
-------
facilities that cannot be efficiently built (constructibility)
nor easily operated and maintained (operability).
In an effort to produce quality construction in spite of
the complexities involved in the design/construction process, the
U.S. Army Corps of Engineers established an aggressive manual
design review program, as illustrated in Figure 2.1. This program
includes (1) a technical review and a value engineering review of
the design package performed by the Corps of Engineers, (2) a
biddability review of the contract documents' structure/content
also performed by the Corps of Engineers, (3) a constructibility
review of the design package performed by the Corps of Engineers'
construction field office, (4) a functional review of the design
package performed by the Army agency that will be using the
facility, and (5) an operability/maintainability review of the
design package performed by the military post engineer, who will
be responsible for operating and maintaining the facility.
COMPREHENSIVE DESIGN REVIEW PROGRAM
BIDDABILITY/
CONSTRUCTIBILITY
REVIEW
DESIGN
REVIEW
VALUE
ENGINEERING
REVIEW
FUNCTIONAL
REVIEW
TECHNICAL
REVIEW
OPERABILITY/
MAINTAINABILITY
REVIEW
Fig. 2.1. Areas Covered by the Design Review Program
Even though the design documents pass through these
multiple reviews by various design disciplines during the design
861
-------
process, major deficiencies still manage to be overlooked. It has
been estimated that approximately half of all construction-
contract modifications can be attributed to design deficiencies
[Nigro 87]. The result of these errors and omissions in the plans
and specifications is an increase in the construction cost and
duration of projects as well as user dissatisfaction due to
higher operation and maintenance expenses. Also, since a typical
Corps District Office manages hundreds of designs concurrently,
tracking the status of individual design reviews or checking on
the actions required by individual design comments is an almost
impossible task to accomplish manually.
2.1.2 Current approach for improving the Design Review process
The solution to the problems stated above is to provide the
project team with the expertise needed to eliminate design
deficiencies before they ever reach the construction stage. A
publication prepared by the Construction Industry Institute
(1986) suggests that savings on the order of 6-23% of the
original project estimate are achievable through proper design
review [Publication 3-1]. In an effort to realize this solution,
the Corps of Engineers has developed two systems to improve the
management and performance of the design review process: (l) the
Automated Review Management System (ARMS) and (2) the
Biddability, Constructibility and Operability (BCO) Advisor
system.
ARMS is a minicomputer-resident program that provides
solutions to many of the problems associated with the scheduling
and management of multiple simultaneous reviews on different
projects and with the disposition of the comments generated
[Kirby 88]. ARMS allows design reviewers and managers to both
obtain review assignments and enter review comments in an
electronic format. Workload information, assignment scheduling
information and the ability to retrieve review comments are
available to all level of users. The minicomputer- (or local area
network personal computer) based ARMS interconnects all reviewers
and managers allowing real-time review management and comment
retrieval.
BCO Advisor is a microcomputer-resident program that
addresses the performance of the design review in regard to
biddability, constructibility and operability topics. It is an
automated review guidance checklist system that assists project
design reviewers in performing their task more accurately and
efficiently. The system also facilitates interaction among
project team members and captures "lessons learned" for
application to future projects.
The use of the BCO Advisor system as an integral part of
the Corps' comprehensive design review program will help to
reduce BCO design deficiencies early in the life of a project,
862
-------
which is when correct decisions have the greatest beneficial
impact on the final cost of a facility. By emphasizing BCO issues
during the design and planning process, contractor productivity
will be ensured, construction cost and time growth will be
minimized, unnecessary changes and claims during construction
will be avoided and safe efficient operations by the user will be
ensured.
The BCO Advisor system helps to generate the review com-
ments that are managed by ARMS. BCO Advisor can be used
integrally with ARMS or it can be used as a stand-alone system
(i.e. in cases where ARMS is not being used). Section 5 of this
paper discusses the potential benefits the BCO Advisor system can
provide to the management and execution of HTW remedial projects.
2.2 Current approach for Duration Estimation and Construction
Scheduling
2.2.1 Problems associated with Duration Estimation and
Construction Scheduling
The approach normally followed to estimate overall
construction duration and evaluate contractor submitted schedules
is described below:
- A/E is required to submit a time estimate of construction
contract duration. However, the A/E's expertise resides on the
design phase as opposed to the construction phase.
- Corps construction personnel manually review and evaluate
contractor submitted schedules. This review demands a substantial
time investment of a highly qualified and experienced reviewer.
- weather impact is assessed in a non-standardized manner,
following a manual approach.
This approach requires improvement because time growth of
construction contracts is a common problem.
2.2.2 Current approach for improving the Scheduling practice
Research work is in progress at USA-CERL to develop
enhanced schedule support tools that contribute to a reduction in
construction time growth.
The objectives of these research projects are fourfold:
a. improve the ability to estimate overall construction
duration prior to starting the construction phase
b. provide enhanced tools for evaluating the reasonableness of
contractor submitted schedules
863
-------
c. improve the monitoring and control of schedule progress
d. provide enhanced ability to acquire and represent, in a
reusable form, scheduling information and experience gained
in construction projects in order to apply the lessons
learned to future projects.
It is important to note that there are computer-based
tools, commercially available, that provide some support to
construction scheduling. These are the so called 'Project
Management Systems' (PMS's). PMS's however, provide limited help.
They only provide support for a CPM representation of project
activities, and the capability of producing schedule reports
(bar-charts, arrow and network diagrams, tables). The research
work in progress at USA-CERL goes beyond the abilities of PMS's.
The objective is to develop smarter tools that not only are able
to store project schedule data, but that also incorporate
construction scheduling experience and heuristics. This
development is being accomplished through the utilization of
innovative computer science techniques, namely knowledge-based
systems (KBS) techniques.
The current focus of this research work is on building
construction. However, the concepts developed can be expanded to
other project types, including HTW remedial projects, as
discussed in Section 5 of this paper.
3 BCO Advisor
3.1 Description
To ensure that a comprehensive review of a project is
accomplished, especially by reviewers who have little or no BCO
background or who tend to concentrate on their own area of
expertise, a guide is necessary to direct reviewers through the
complete review process. This guide is typically in the form of
written checklists; however, checklists for conducting BCO
reviews have had a fundamental conflict: ease of use versus
comprehensiveness. An easy to use checklist is short, simple and
requires little time to utilize but, such a checklist cannot be
very detailed nor provide much useful information for detailed
reviews. A comprehensive checklist, on the other hand, can cover
numerous items that should be examined, but this type of list is
difficult to use effectively and also requires considerable time
to review each item on the list.
The BCO Advisor utilizes a knowledge-base system shell
called KnowledgePro. This shell successfully combines two current
technologies, expert systems and hypertext, which are able to
eliminate the previously stated difficulties associated with the
use of hardcopy BCO checklists. This software allows the
864
-------
establishment of checklist interrelationships, and controls the
level and direction of the information presented. Hence, the BCO
Advisor can present various levels of advice and guidance without
excessive or unwanted detail. Also, the hypertext feature
provides a capability to explain terms that are used in the
questions or checklists but only if the user requests these
definitions.
3.2 Development process
Development of the BCO Advisor began in late 1988 and
initially involved the examination of many sources of
information, within the Corps of Engineers as well as private
industry and academia, to determine if they contained relevant
BCO review information. After studying the various methodologies
and checklist sources of BCO review guidance, work began on the
development of a prototype program. Utilizing the expert system
and hypertext technologies offered by KnowledgePro, a basic
framework for the program was developed that reflected review
techniques currently in use by Corps of Engineers' District and
Division offices.
Based upon comments and suggestions from these Corps review
offices and from several user group workshops held at USACERL, a
final system design iteration was undertaken over the last half
of 1990. The system structure and input/output requirements were
finalized and several new requested features were incorporated
into the program. In addition, an extensive data collection
effort was undertaken to build the checklists contained within
the program.
The current program format, as illustrated in Figure 3.1,
classifies review topics according to the type of review being
conducted (i.e. 35% Concept Review or 95% Final Review) along
with a Special Issues Review category.
,3t5%:CONCEPT
'l:fl REVIEW^
95% FINAL,
.REVIEW
SPECIAL ISSUES
REVIEW 4
Fig. 3.1. BCO Advisor Logic Tree
865
-------
The 35% and 95% review categories are divided into seven basic
design disciplines, as illustrated in Figures 3.2 and 3.3. The
disciplines under 95% are further split into their applicable
Construction Specifications Institute (CSI) Divisions due to the
availability of more detailed design information. Each discipline
(35%) or CSI Division (95%) contains its own set of review
guidelines to which the reviewer refers while checking the
contract documents.
35% CONCEPT
REVIEW
ARCHITECTURAL
STRUCTURAL
ELECTRICAL
CIVIL
MECHANICAL
ENVIRONMENTAL
OPERATIONS/
MAINTENANCE
Fig. 3.2. BCD Advisor Logic Tree (35%)
1
ARCHITECTURAL |
GENERAL
REQUIREMENTS
— SITEWORK
— CONCRETE
— MASONRY
— METALS
WOOD AND
PLASTICS
THERMAL AND
— MOISTURE
PROJECTION
f
| STRUCTURAL |
GENERAL
REQUIREMENTS
SITEWORK
CONCRETE
MASONRY
METALS
WOOD AND
PLASTICS
THERMAL AND
MOISTURE
PROTECTION
•"•*
1 1
1 ELECTRICAL ]
GENERAL
REQUIREMENTS
OPERATIONS/
MAINTENANCE
BASIC ELECTRICAL
MATERIALS/METHODS
LIGHTING
COMMUNICATIONS
WINDOWS L * \ \ "ECHANICAL |
— FINISHES
— SPECIALTIES
— EQUIPMENT
— FURNISHINGS
GENERAL
REQUIREMENTS
SITEWORK
CONVEYING
SYSTEMS
GENERAL
REQUIREMENTS
— FIRE PROTECTION
— PLUMBING
1 HVAC
1
| ENVIRONMENTAL]
GENERAL
REQUIREMENTS
— srrewoRK
CONCRETE
MASONRY
WOOD AND
PLASTICS
THERMAL AND
MOISTURE
PROTECTION
DOORS AND
WINDOWS
FINISHES
SPECIALTIES
EQUIPMENT
FURNISHINGS
CONVEYING
SYSTEMS
MECHANICAL
Fig. 3.3. BCO Advisor Logic Tree (95%)
This breakdown reflects the manner in which construction drawings
are normally arranged and distributed to various reviewers. It
also allows for the concurrent review of drawings and
specifications, the typical and most comprehensive approach to
reviewing a particular design project. Only the Special Issues
866
-------
Review, as illustrated in Figure 3.4, uses its own unique
classification of review topics.
SPECIAL ISSUES
REVIEW
Pig. 3.4. BCO Advisor Logic Tree (Special Issues)
These topics are usually very project-specific and are most
likely to be customized to the differing needs of each Corps'
District and Division office. They are provided for experienced
reviewers who do not need to be "led by the hand" through either
the Concept or Final Design Review but require information on BCO
issues encountered on an infrequent basis.
3.3 How BCO Advisor works
Figure 3.5 illustrates how a typical review session has the
reviewer requesting guidelines within a particular review
category from a series of menus.
35% CONCEPT REVIEW
I
1,
SPECIAL ISSUES
REVIEW
DISCIPLINE
CUSTOMIZE
CHECKLIST
Fig. 3.5. BCO Advisor System Structure
867
-------
The guidelines provided by the program are then used as a basis
for checking for deficiencies in the contract documents. The
complete review involves examining the documents following the
guidelines listed under the applicable topics of the BCO Advisor.
Within every checklist the reviewer has the option to export any
relevant guidelines to an output file and to edit those
guidelines into specific review comments pertaining to the
particular project being reviewed. When more than one session is
needed to completely review a set of contract documents, the same
output file can be used, with additional comments merely appended
to that file. The system also allows for cross-checking between
disciplines at the Final Design Review stage, thereby ensuring a
more complete review. Each discipline can query the system for
guidelines that are outside their area of expertise but are
relevant to reduce conflicts among disciplines. Also, this system
allows each reviewer to customize the checklists to fit their own
particular needs.
The system initially asks the user for information about
the project to be reviewed as well as for the name of the file
that will store the comments gathered from the review session.
After this information is entered, the computer is ready to run
the program.
3.3.1 Main Menus
Figure 3.6 shows the first menu encountered by the user
which gives choices for the type of review to be undertaken.
BCO ADVISOR
HJtit type of Kvieu is king conducted?
35x Concept Review
F3 Select F5 Iwluate T! Bit
HViea IS Display KB F8 M FlIHait
Fig. 3.6. Type of Review
The 35 percent Concept Review covers general issues which need to
be caught in the early design stages. The 95 percent review
covers issues found on the final set of plans, specifications,
and bid documents. The Special Issues Review deals with specific
items which must be addressed on a project by project basis.
868
-------
When 35 percent Concept Review is selected from the main
menu another menu appears, illustrated in Figure 3.7, which
outlines the seven disciplines that contain checklist guidance.
BCO mim
35X CONCEPT REIIEH
Htet discipline wiild you likf to wview?
Return to Opening Menu
F5 Mluat*
FS Disrlau K
Fig. 3.7. 35% Concept Review Menu
Choosing any one of these seven disciplines (Architectural,
Civil, Structural, Mechanical, Electrical, Operations/Maintenance
or Environmental) will produce a checklist dealing with that
topic and level of review.
When 95 percent Final Review is selected from the main
menu, the same seven disciplines are displayed as contained under
the 35 percent Concept Review. Picking one of these disciplines
produces another menu, illustrated in Figure 3.8, which has the
appropriate portions of the sixteen category CSI breakdown for
that discipline.
BCO fltUISOR —
95/. FINAL Wiim ARCHITECTURAL
PRINT ALL CHECKLISTS
Hkt CSI division wultl sou like to review?
F3 Select F5 E
FHie* K J
KB F8KS
FIB luit
Fig. 3.8. 95% Final Review CSI Divisions Menu
869
-------
Choosing one of these categories produces the checklist from
which comments can be exported.
The categories under Special Issues Review, as illustrated
in Figure 3.9, are: Life Safety, Security, Schedule, Special
Requirements, Special Facilities, Site/Regional Characteristics,
Site Adaptation and Geotechnical.
BCD flWISOR
SPECIfll ISSUES HMD
What special issue muld you like to review?
T5 Initiate F7 Hit
F6 lisjflajj KB F8DM FIB to
Pig. 3.9. Special Issues Review Menu
Selecting any one of the first three categories of the Specieil
Issues Review menu produces checklists; however, choosing any one
of the last five categories produces another menu with sub-
topics. These sub-topics can be edited to indicate specific
localities or facility types and customized checklists can then
be provided under each sub-topic.
3.3.2 output procedures
The intent of BCO Advisor is to guide and assist project
review team members in producing comments which are sent to the
project design team members for plan, specification and bid
document modification. This intent is fulfilled more through the
structure of the program rather than the content of the
checklists. Each checklist has been prepared as generic as
possible but the capability exists to make them specific to each
project. Each checklist contains information to remind the user
of items to be reviewed. When an applicable issue is found, the
reviewer can export that comment to a file which is printed at
the end of the review period and sent to the designer for
incorporation into the project.
Each checklist has "hypertext" choices located along the
top of each page of comments. The options available within the 35
percent Concept Review checklists (Special Issues Review is
similar) are EXPORT COMMENTS, PRINT CHECKLIST and EDIT CHECKLIST,
870
-------
as illustrated in Figure 3.10.
viev Guidelines
CONCEPT miU: Architectural
iBffOCT COHMENIS
] Jrawinjs should co«pletelg describe the planned scope of trorlc.
i'M Drauinjs should be free Iron anliguities,
£ Essential details should If provided.
3 Drawings should agree within one discipline and with other disciplines.
•B Drawing title descriptions, towns mwbers, and revision nuMbers, should
be consistent Iron sheet to sheet.
fl Pepirewnts in the contract docuNents, specifications, and drawing notes
snould be consistent with each other.
HHe]p
Space Cent.
R Select
F4 yie»
13 Evaluate F7 Elit
16 1'isrfaa KE F8 DM
Pig. 3.10. 35% Concept Review Checklist
The EXPORT COMMENTS option is used to select comments from the
checklists which are then stored in the output file. After a
comment is chosen for export to the output file, the program
allows the user to customize the comment without changing the
generic checklist. If this option is selected, the computer
functions like a word processor. Comments can be overwritten,
added to or annotated. After each comment is edited, the program
asks for a page or sheet number. This information designates
either a page of the specifications or a sheet of the drawings
which the comment refers to. In the next step, the computer
requests a detail or room number of where the problem might
exist. A specifications paragraph number may be used instead of a
detail or room number. Therefore, a reviewer can generate
specific comments based on the generic guidance furnished by the
BCO Advisor system to clearly indicate problems to the designer
which need to be corrected.
The PRINT CHECKLIST option allows the user to obtain a
hardcopy printout of the entire checklist for a particular
discipline being reviewed. Once the checklist is obtained,
comments can be marked and edited manually for later input into
the computer.
The EDIT CHECKLIST option is provided for a user that may
desire to make permanent changes to the checklists. The generic
list may be altered or added to in order to remind a user of
repetitious problems or lessons learned that are specific to
their location. Again, the computer functions as a word processor
to accomplish this task.
An additional hypertext pick, RELATED INFORMATION, is
included at the top of each checklist for the 95 percent Final
871
-------
Review checklist, as illustrated in Figure 3.11.
einey Guidelines
Architectural: Central Seguireufnts
KMIEJ INFORMION
1. Indicate security requireiients of enployees.
2. Shov traffic control during construction,
3. Insure that test methods, Material specifications or other tunuals are
consistent nitli civil OP Military designations as applicalle.
4. Coordinate large scale plans and elevations with snail scale plans.
5, Coordinate luildinj sections vitli elevations.
6. Show efficiency of fire-safety features and egress systen as incorporated
into the building laput.
Fl Help
Space Ci;.t,
Fig. 3.11. 95% Final Review Checklist
This option indicates that other checklists within the 95 percent
review level may contain relevant information to the discipline
being reviewed. The program allows the user to review related
checklists from other disciplines that may be affected by the
design discipline being reviewed. Comments from that checklist
may be exported and edited or the entire checklist may be
printed.
The final option available to the user of BCO Advisor is
the editing of menus. The 35 percent Concept Review, 95 percent
Final Review Architecture, 95 percent Final Review Civil, 95
percent Final Review Structural, 95 percent Final Review
Mechanical, 95 percent Final Review Electrical, 95 percent Final
Review Operations & Maintenance, 95 percent Final Review
Environmental, Special Issues Review, Special Requirements,
Site/Regional Characteristics, Site Adaptation, Special
Facilities and Geotechnical menus can be edited for specific
needs. This feature allows the user to adapt the headings of
menus, along with the checklists contained under those headings,
to their specific requirements. Therefore, the system's expertise
can always remain current for the type of review being performed
by each user.
3.4 Status
Field testing of the prototype system is scheduled to begin
in late April 1991. A user's manual is currently being written
and the system is due to be installed at nine Corps of Engineer
District offices. Field testing will last approximately four
months.
872
-------
3.5 Finalization and Future Efforts
Based on results from the field testing, the full scale BCO
Advisor can be developed into its final format. Fielding strategy
for the system will be completed in 1992 and Corps-wide imple-
mentation is expected in the latter part of that year.
Currently, additional effort is being expended to develop
an environmental compliance module to the BCO Advisor system
(BCO-E). This module will attempt to assure that the project
design complies with all applicable or relevant and appropriate
environmental and public health requirements along with the
utilization of currently accepted environmental control measures
and technologies. It is believed that BCO-E will produce a more
thorough review of project designs for environmental compliance
which will lead to a lessor number of contractor claims and
change orders, less cost growth during construction and the
provision of safety to workers and adjacent personnel. This sytem
will also enhance the efficiency of the review by providing ready
access to appropriate regulations and by allowing a cross-check
of environmental issues between design disciplines.
Section 5 of this paper provides evidence for the
applicable benefit of the BCO-E Advisor system to HTW remedial
work.
4 Computer Assisted Scheduling
4.1 Description
Research progress to present addresses all the issues
introduced earlier in this paper in Section 2.2.2. A tool is in
development to improve the ability to estimate overall
construction duration. A prototype system named CODES
(Construction Duration Estimating System) is in the process of
being validated and tested. Work is also underway to develop a
computer-based construction schedule generator. An initial
prototype (CASCH, for Computer Assisted Scheduling) has been
developed that is able to generate schedules for building
construction. There is also research work being performed to
improve the consideration of weather impact on construction
schedules.
4.2 Development process
As mentioned, the research strategy for computer assisted
scheduling is to generate smarter tools that not only can
represent project data, but also can incorporate and use some
scheduling knowledge. This goal is achievable through the
utilization of innovative computer technologies that allow the
representation of knowledge consisting of: (1) facts, for example
873
-------
'unprotected exterior concreting activities are sensitive to
weather1; and (2) heuristics, like 'if a component covers work to
be inspected, wait until after inspection to install it1.
The acquisition of the scheduling knowledge is therefore of
paramount importance. Several avenues have been pursued to
acquire construction scheduling knowledge. The most relevant ones
are discussed in the following paragraphs.
4.2.1 Knowledge acquisition
A series of structured interviews with experienced
construction schedulers from different construction firms was
conducted. Five construction schedulers from four construction
firms were interviewed during a period of 18 months in order to
acquire construction scheduling knowledge. This knowledge
acquisition process was complemented with input from Corps of
Engineers experienced construction personnel which was acquired
through two workshops and informal communication.
The acquisition of construction knowledge with the above
mentioned schedulers was performed in several different ways,
described in detail in [Echeverry 91]. Only a brief summary of
the knowledge acquisition process is provided in this paper.
Two approaches were utilized to interact with the
experienced schedulers from the private firms: (1) development of
a schedule for an example building for which complete drawings
and specifications were available; and (2) discussion sessions
based on previous construction schedules developed by the
participating schedulers.
Also, a number of publications related to scheduling were
reviewed, listed in [Echeverry 91] and [Steen 91]. This review
complemented the interaction with the schedulers. Especially
relevant information was obtained from a review performed on the
Corps of Engineers Construction Specifications to identify the
sensitivity of construction materials to weather [CEGS 90].
4.2.2 Summary of acquired knowledge
Schedule Production Phases
Two major phases were observed that comprise the schedule
generation process. The first phase consists of the assimilation
and understanding of project information by the scheduler. The
second phase is the actual production of the schedule.
The experience of the scheduler is useful at the
information assimilation phase in identifying project features
that are common (typical) and features that are unique to the
874
-------
project (project specific). Most of the effort in this phase is
spent by the scheduler in examining those unique project features
and determining how they might be installed and procured.
The schedule production phase is accomplished in two steps.
The first one has a qualitative emphasis and includes: (1) a
breakdown of the project construction into activities; (2) a
preliminary logical sequencing of the defined activities; and (3)
a preliminary consideration of activity durations based on
approximate quantities. The second schedule production step
consists of an iterative process of adjusting and refining the
schedule. Issues considered in this step include: (1) procurement
lead times; (2) crew design and productivity estimation; (2)
expected weather impact; (3) owner occupancy requirements; etc.
Activity Sequencing
Through the interaction with the construction schedulers
and the literature review, several key factors that govern
activity sequencing were identified. Table 4.1 provides a summary
of these factors.
GOVERNING FACTOR
GENERAL DESCRIPTION
Physical Relationships Among
Building Components
Building components are spatially restricted, weather
protected or gravity supported by other components.
Activity sequencing has to respond to these inter-
component relationships.
Trade Interaction
Activity sequencing also responds to the different
ways in which the different crews and their processes/tools/
equipment affect each other during the construction phase.
Path Interference
Building components have to be moved around the job-
site in order to be installed. Activity sequence
has to guarantee an interference-free path for the
displacement of any component and its installing crew
and equipment.
Code Regulations
Activity sequencing is also responsive to construction
phase safety considerations, and to inspection/accep-
tance requirements.
Table 4.1. Identified Categories of Activity Sequencing Factors
875
-------
The following examples illustrate the application of the acquired
knowledge:
- Weather sensitive components (dry wall or ceiling tile, for
instance) are installed after the building enclosure is in
place because the enclosure weather protects these
components (the enclosure and the weather sensitive
components are physically related by the 'weather-protects1
relationship).
The slab on grade is installed after the utility pipes are
in place, because the slab on grade covers the utility
pipes.
The finishes of the first floor, or lobby area, are
normally completed after the rest of the building is
finished because this is typically the access area for all
crews and equipment working inside the building. This
circulation of people and equipment can likely damage the
finishes of the access area if they are completed.
Estimation of Preliminary Activity Durations
Heuristic knowledge was acquired to estimate preliminary
building construction durations based on approximate quantities.
For example, it was identified that the pace of progression of
the structural frame erection normally controls the pace of
progression of following work (rough-in work, wall studs, etc.).
This controlling of the pace happens because the frame erection
provides the areas (floors) where most of the work that follows
is performed.
Activity Weather Sensitivity
A review of the Corps of Engineers Guide Specifications,
and of relevant prior studies, is in progress to identify and
compile activity weather sensitivity knowledge. Also, discussions
with experienced field personnel have been performed to
complement this weather sensitivity information. This information
gathering is presently addressing weather limits for which work
is normally interrupted. The effect of reduced productivity
levels because of less than ideal weather circumstances will be
addressed in future research efforts. There are three major areas
where weather sensitivity information acquisition is in progress:
(l)material sensitivity; (2)operation sensitivity (high winds
sensitivity of structural steel erection, for example); and
(3)labor and equipment sensitivity. A current compilation of
results is available in [Steen 91].
876
-------
4.2.3 Prototype systems for scheduling assistance
Part of the acquired construction scheduling knowledge
described in the previous section has been incorporated in the
form of computer systems. Currently these systems are at the
prototype level. Validation and testing of these prototypes is in
progress.
CODES
This is a prototype system for construction duration
estimation. It incorporates knowledge about: (1) commonly found
major activities for building construction (e.g., structural
frame erection, exterior walls installation, etc.); (2)
preliminary duration estimation for these activities; and (3) a
default logic (or precedence relationship) based on common
building construction practice. The objective of CODES is to
assist in performing reasonable estimations of overall
construction duration, based on a few input building parameters
(number of floors, type of frame, type of enclosure, etc.). CODES
is described in more detail in [Sun 91].
Figure 4.1 shows one of the CODES input screens
Figure 4.1. CODES Input Screen for Number of Stories
Figure 4.2 illustrates the output that CODES provides. In
this case, a barchart of major construction activities was
produced for a ten story building with one basement, and a
877
-------
typical area per floor of 10,000 sqft.
ACTIVITY/WEEK 5 10 15 20 25 30 35
40
45
50
55
60
mobilization**
site&foundation ******
erect_frame
erect_roofframe
place_concr_deck
fireproofing
roofing
instal_elevator
rough_in
enclosure
int_finish_fl_bl
int_finish_fl_2
int_finish_fl_3
int_finish_fl_4
int_finish_fl_5
int_finish_fl_6
int_finish_fl_7
int_finish_fl_8
int_finish_fl_9
int_finish_fl_10
int_finish_fl_l
clean_up
demobilization
***********
***********
**
**************************
***********
********************
********
********
********
********
********
********
********
********
********
********
*********
Start date: Tuesday 10/1/1991
Duration: 55 WEEKS, 385 CALENDAR DAYS, 275 WORKING DAYS
Finish date: Monday 10/19/1992
Figure 4.2. Example of CODES Output
CODES is currently able to provide weather related warnings, as
illustrated in Figure 4.3.
Figure 4.3. Example of CODES Weather Warning
878
-------
However, its weather knowledge is limited to Mid-west weather
patterns.
CASCH
This is a prototype system that incorporates knowledge to:
(1) breakdown building construction into activities at three
levels of detail; (2) sequence construction activities responding
to some of the factors summarized in Table 4.1 on page 17; and
(3) estimate preliminary durations for the defined activities.
The objective of CASCH is to assist the planner in developing
building construction schedules in a fraction of the time
required to do manually. The approach is for CASCH to request
information about general building parameters and quantities to
develop a schedule based on common construction practice. The
user then refines and adjusts this result with project specific
features not considered by CASCH. CASCH is described in more
detail in [Echeverry 91].
Figure 4.4 shows an overview of the operation of CASCH.
Input: building scope description
(limited to about 20-25
questions)
CASCH
Specific Building
Systems, Sub-
systems, Components
WHAT
Activities, activity
sequence and
preliminary
durations
HOW
Sequence
justifications
WHY
Figure 4.4. Overview of CASCH's Operation
The input required from the user is reduced to providing building
system types and approximate building quantities. It is relevant
that CASCH not only deduces activity sequence given its knowledge
of activity sequencing, but it also stores the justification of
879
-------
each precedence link that it deduces.
Figures 4.5 and 4.6 illustrate some of the results that can
be obtained with CASCH. They show the activities related to Site
Preparation and Foundation Work, and to Exterior Skin
installation for a six story building with one basement.
NHfflD-Vm-IICUlU
:owtn. rooms. i»irmitto«
«l«K-ltlK.imUUMKII
rarai-wiu-miii-mmiiiioi
nmi-wjiii-nrctiM
Illl-OMMK-niniUinM
arums p|
minus-ttuounoi
i "i H J'n/i 1 fiSilii' PI'.':..' .HSH^*-«-f«,M«^.!»: KIT! Hilji fi|i| ffij Hf g," , , ' - ',i,.
,i, if!j: 'S\ "JI ifili!' '"!:• IP jjSUxfZfr'£*;""•%'«!*Si '^SfM:'! 11 ffl KW '"i ff '" i"'!
1 f,1'!,':',§.14 811KS;;'.:F'^^
Figure 4.5. CASCH Barchart of Site Preparation/Foundation
*TIOr1. ttVCt-5
mTRtOII-UM01IRY-l1l!tAI.UTtOlf-ttm-<
WI|f DOW- INIf HUXTIOir-
wtmmw. nis
'' n<: -:j' •;* ,'ii rtti ftiiiiu [f M|^;i!« iiillv'^ f;F ?! 1 !';|i *'!!ff '™pllf'l,!^ *
'•' f' "ins? H'.!'"' i'ss!i;«*i«-ia;"i"-''!ii SH* I"]!'1:1! If l$T»P'.^';ilWW!»? Pl1^
Figure 4.6. CASCH Barchart of Exterior Skin Installation
880
-------
4.3 Status and projected future progress
Both CODES and CASCH are at the prototype level. CODES is
being enhanced through validation and testing at the present
time. The effort to produce CASCH is being complemented by an
endeavor to address schedule evaluation. This development
responds to the fact that the Corps of Engineers does not dictate
the schedule for the construction projects it manages. The
approach instead is to review and evaluate the reasonableness of
contractor submitted schedules.
There is also a research effort underway to develop a
computerized weather impact evaluation advisor. The planned
approach is to incorporate the acquired weather sensitivity
information into a computer program. This program will also
contain a database of weather data. This advisor is expected to
estimate the number of days lost due to weather impact on
construction operations.
5 Potential application to HTW remedial projects
HTW remedial projects are deemed successful if they achieve
the following results: (1) procurement protests are not
encountered, (2) construction is completed on schedule, (3) a
remedy consistent with the Record of Decision (ROD) is
constructed, (4) a minimal number of change orders are
encountered and (5) constructor claims are identified and
resolved before the completion of construction. The innovative
tools being developed at USACERL for design review and scheduling
have great potential for being very useful in helping to achieve
these five goals on HTW remedial projects.
5.1 8CO Advisor
The Army Corps of Engineers spends approximately $460
million each year in contingency, supervision and administrative
costs for the acquisition of new facilities as well as for
maintenance and repair of existing facilities. This money is
necessary to cover the costs of change orders during construction
produced by design deficiencies or unidentified site conditions.
The reduction of BCO-E related errors and omissions which
develop into change orders during the construction phase of a
project has substantial benefits. If the contingency, supervision
and administrative costs included in a construction budget can be
cut by just one percent due to a decrease in the number of design
deficiencies that reach the construction stage, the Army will
immediately realize a savings of $4.6 million per year. Also, if
the construction contract documents can be easily understood and
a quality design package is produced that can efficiently be
built, there will be fewer project disruptions during
88]
-------
construction. This situation will allow constructors to lower the
contingency included in their bid prices which will translate to
a reduction in initial project costs.
One proof of the potential savings the BCO-E Advisor system
can produce is evident in a study that was conducted in FY89 at
the Army Corps of Engineers' Sacramento District to determine if
ARMS reduced construction modifications due to design errors;. The
study concluded that for FY88 ARMS reduced the Sacramento
District construction modification amount by $1,178,545, a
savings of approximately 5 percent [CESPK 88]. Since BCO-E
Advisor is a system to help generate the comments (it exposes
design deficiencies) that are managed by ARMS, it is reasonable
to assume that at least the same amount of savings can be
additionally achieved through the use of BCO-E Advisor.
This system possesses great potential for providing support
to the review of HTW remedial projects. Before construction
begins, these projects undergo several extensive reviews of many
of the same areas contained within the BCO-E Advisor system. A
biddability review is performed to ensure that the construction
package is free of significant design errors, omissions and
ambiguities so that bidders can respond in a reasonable manner
and at a reasonable cost. A constructibility review is performed
to enhance the "buildability" of a design by evaluating the
technical product being delivered by the designer for accuracy
and completeness along with eliminating impractical and
inefficient construction requirements. An operability review is
performed to determine whether the particular system or remedial
facility will function optimally, as required by the design
documents, and whether it can be maintained in an acceptable
manner. An environmental review is performed to provide assurance
that the design will meet the technical requirements of the ROD
and to provide consistency between the implementation plans and
the current regulatory and policy requirements. Additionally, the
environmental review determines the adequacy of the documents; in
addressing the potential for environmental releases during
construction and the contingency plans, should such releases
occur.
As previously stated in Section 3, the BCO-E Advisor ce.n
easily be customized to the specific needs of each user.
Therefore, biddability, constructibility, operability and
environmental compliance issues can be inserted into the system
which apply specifically to HTW remedial projects. Once the
information is in the system it can be used to guide and assist
reviewers in conducting thorough reviews of HTW remedial
projects. As proven with traditional Corps of Engineers'
construction, this automated review system can provide the same
magnitude of savings on HTW remedial projects.
882
-------
5.2 Construction scheduling assistance
The research work to present in computerized construction
scheduling support has focused on building construction. It is
recognized that HTW remedial projects substantially differ from
traditional building construction. However, the application of a
similar approach to develop tools specifically targeted to
support the scheduling of HTW remedial projects is not only
possible but desirable.
The overall approach followed here to produce improved
construction scheduling tools is to gather experience and
knowledge accumulated in the past (from experienced schedulers,
and literature review), and incorporate part of this knowledge
into a computer platform. This allows the production of
computerized assistants that can take a more relevant role in
project scheduling.
This approach is potentially very advantageous for
supporting HTW remedial project scheduling. HTW remedial projects
normally incorporate innovative technologies and techniques that
make it extra difficult to anticipate durations and produce
construction schedules. A structured effort to accumulate and
store experience gained in scheduling projects that involve
innovative technologies should soon provide a knowledge-base that
contains the gained experience. This could translate into more
accurate HTW remedial project duration estimations and improved
HTW remedial project construction schedules. An added potential
benefit is the increased productivity of the planners and
schedulers that deal with HTW remedial projects.
6 Conclusions
The U.S. Army Construction Engineering Research Laboratory
has developed several innovative systems that are being utilized
by the Army Corps of Engineers to enhance the design review and
scheduling of traditional construction projects. This paper has
attempted to show the promising potential these tools possess for
application to HTW remedial projects. BCO-E Advisor, CODES and
CASCH will produce more thorough design reviews and more accurate
schedules which will reduce time delays and cost overruns on HTW
remedial projects.
7 References
[CEGS 90] "Corps of Engineers Guide Specifications" (CEGS),
in Construction Criteria Base. CD-ROM distributed
by the National Institute of Building Sciences,
Washington DC, April 1989 to December 1990.
883
-------
[CESPK 88]
"Automated Review Management System", CESPK Test
Report, U.S. Army Engineer District, Sacramento,
Engineering Division, Technical Support Branch,
1988.
[Echeverry 91] Echeverry, D., "Factors for Generating Initial
Construction Schedules", Thesis Dissertation
Submitted in partial fulfillment of the PhD
Degree, Civil Engineering Dept., University of
Illinois, 1991.
[Kirby 88]
[Kirby 90]
[Nigro 87]
[Pub 3-1 86]
[Steen 91]
[Sun 91]
Kirby, J.G., Furry, D.A. and Hicks, O.K.,
"Improvements in Design Review Management",
Journal of Construction Engineering and
Management. Vol. 114, No. 1, 69-82, 1988.
Kirby, J.G., Tupas, M.I., Robinson, P.C. and
Bridgestock, G.W., "Concept Development of an
Automated Construction Design Review Advisor",
US Army Construction Engineering Research
Laboratory Interim Report, (ed G.L. Cohen), US
Army Corps of Engineers, pp. 7-18, 31, 1990.
Nigro, W.T., "Contract Documents: A Quality
Control Guide", Architecture, 1987.
Construction Industry Institute Constructibil.Lty
Task Force, "Constructibility, A Primer",
Construction Industry Institute Publication.
publication 3-1, 1-13, 1986.
Steen, S., D. Echeverry, M. Aboushousha and S,.
Kim, "Severe Weather Impact Analysis for Military
Construction Projects", Interim Report, USA-CERL,
1991.
Sun, R., G. Rao, D. Echeverry and S. Kim, "A
Prototype Construction Duration Estimating System
(CODES) for Mid-Rise Building Construction",
Interim Report. USA-CERL. 1991.
884
-------
Basic Principles of Effective Quality Assurance
Mr. David E. Foxx, CEO
PDX Company
A division of d.e. Foxx & Associates, Inc.
324 West Ninth Street
500 Goodall Complex
Cincinnati, Ohio 45202
(513) 621-5522
INTRODUCTION
Typically in designing a QA program application, one must start with concepts and principles derived
from vision, information, and experience. From the resulting concepts and principles, an application
design is created and validated. Originally, the topic of this paper was based on a designed
application for RA efforts. Our present topic, however, must remain at the concept and principle
level because our study is currently in progress. One can test current applications based on these
concepts and principles. Therefore, this discussion focuses on "Basic Principles of Effective Quality
Assurance".
As one begins to plan the management of the quality assurance program associated with a large
project, there are endless potential points of evaluation and actions to be taken. Often, the key in
selecting the right QA plan is determining what must be examined and when.
BACKGROUND
The Hazardous Site Control Division (HSCD) of the EPA's Office of Emergency and Remedial
Response provides support to the regional offices in a variety of areas. Part of this responsibility
includes reviewing regional program activities and developing procedures to improve the overall cost,
quality, and schedule of remedial projects. PDX Company is assisting the HSCD in reviewing
Construction Quality Management concepts and practices with regard to EPA Regions and the
Alternative Remedial Contracting Strategies (ARCS) contractors.
PDX Company has reported to the EPA on the technical quality assurance procedures currently being
used by the private construction industry. Information was gathered based on our experiences as
construction managers, a review of selected current literature, and a collaborative observation of a
limited number of local construction contractors. Future project work with the EPA will include a
broader study of private QA practices versus potential benefits to EPA operations.
PDX Company has spent significant time and effort in applying third party management (including
QA) strategies. This discussion will cover some of the basic concepts and principles which have
helped us to improve the effectiveness of our third party QA process.
ANALYSIS
A. Definition of Terms
The following definitions must be established:
CM -- Construction Management
A project delivery system utilizing an unbiased owner's representative (or agent). Its
objectives are to minimize project time & cost while maintaining quality. Starts
during preconstruction.
GC — General Contractor
The prime or main contractor signed by the owner for construction of the entire
project construction.
885
-------
QA — Quality Assurance
The process by which concerned others verify that compliance has been achieved.
QC -- Quality Control
The management process by which contractors achieve results that comply with the
requirements.
When developing a QA program for a specific project, a set of requirements is identified. From these
requirements, the contractor develops his/her QC plan. The design team or QA team establishes a
QA plan. The contractor implements construction utilizing the QC plan to achieve results that will
comply with the requirements. The QA plan is simultaneously implemented to verify the compliance
of certain processes and results. Simple!
Simple, if all goes as planned. However, such is not the norm. Occasionally, the processes and results
achieved by the QC program are not in compliance with the construction requirements. Hopefully,
QA catches everything missed by QC. This is the expectation of QA, notwithstanding the extra
degree of difficulty and cost for the QA program, to achieve the same quality as the QC program.
QA is faced with the difficulties of being:
Based on auditing/sampling -- QA checks the process rather than controls the process.
Variable in situations encountered -- QA site conditions, project types and locations, teams, workers,
etc., all vary significantly.
Achieved by influence -- QA does not directly control the implementation of construction activities.
Critical In Timing -- Certain key QA processes are timing sensitive.
Given these conditions, one must structure a fully reliable and cost-effective assurance plan. Because
of the endless possibilities, cost effectiveness revolves around knowing where and when to look. Let
us analyze several key principles.
B. Principles of Effective QA
To assist in the development of a fully reliable and cost-effective assurance plan, we have developed
a few key principles:
Principle 1. Essentially, everyone wants to do a good job.
Principle 2. A normalized sampling and review program must be operative.
Principle 3. Prevention collaboration, if managed, enhances the effectiveness of results
management.
Principle 4. Each significant predecessor event must be completed prior to starting its
dependent event.
Principle 5. Certain stress points (Potential Breaches) cause deviation from quality --
Management of the stress points allows quality.
Principle 6. Systems are more effective than personal intervention.
In discussing these principles, we intend to discuss how to maximize known strategies rather than
"what to do". We will not discuss actions for the examination or resolution process.
Before we begin the discussion of the principles, it is important to note that the foundation of a
successful quality program is the QC program. A well developed, well understood, and actionable
Quality Control program should have the following characteristics: self-checking, self-correcting and
886
-------
multi-project improvement memorizing. Most construction QC programs, however, currently need
support from a QA program. With that in mind, let us analyze the aforementioned key principles.
Principle 1. Essentially, everyone wants to do a good job.
As one establishes a QA program or initiates a cause analysis, time is often wasted in chasing
"do not care" attitudes among project members. Most frequently, more substantial causes
exist and are not typically the result of intentional neglect or malice by project members.
Finding and correcting these more substantial causes will typically have a greater benefit
than resolving "do not care" attitudes among project members.
Principle 2. A normalized sampling and review program must be operative.
Two major efforts are associated with the QA process: Judgmental Review and Normalized
Review. The components of each are listed below:
Normalized Review
Judgmental Review
Engineering testing/sampling
Blind sampling
Appearance modeling (mock-ups)
Unscheduled visiting
Inspecting
Preventive action planning
Although this paper primarily addresses the judgmental effort, the effective implementation
of a normalized quality review process is critical.
Principle 3. Prevention Collaboration, if managed, enhances the effectiveness of results management.
In an environment of clear responsibilities and liabilities, one of the strongest assurance tools
is prevention collaboration. Prevention Collaboration requires anticipation and
communication to all team members of future quality dependent events in order to plan
special handling as the situation dictates. If one is able to anticipate situations and facilitate
proper management responses, the results will probably be in compliance and thereby will not
require corrective results management. In the rare situation of uncertain relationships, it
continues to be useful to discuss significant future quality dependent events. However, it is
our general guideline to cautiously agree with and avoid setting the direction on how to
handle these quality situations. This guideline is more important in uncertain relationships.
Principle 4. Each significant predecessor event must be completed prior to starting its dependent
event
Sounds simple, but this principle is often violated, typically to maintain schedule. Such
violations rarely result in time savings; substantial recycle and quality problems typically
result. Violation of this principle is most prevalent as the project moves across the RD/RA
interface. Obviously, this principle is critical on fast-track projects where multiple package
sequencing is heavily utilized.
Principle 5. Certain stress points (Potential Breaches) cause deviation from quality -- Management
of the stress points allows quality.
Often in executing a project, the quality norm for most of the project deliverables will be in
compliance or will quickly be forced into compliance after start-up confusion is eliminated.
Although the project team is maintaining the state of overall project quality performance at
or above compliance, we have found the regular existence of pockets of wide compliance
variation. These pockets are typically responding to predictable events referred to as
"potential quality breach situations." If anticipated and properly managed, potential breaches
are never allowed to become breaches. If improperly managed, potential breaches become
breaches and remain as such until corrected by outside forces or by itself.
The following quality deviations (potential breach situations) are common to most projects:
887
-------
Start-up operations
Key personnel change
Field design changes/interpretations
Significant number of project change orders
Unexpected site conditions
Significant weather changes
Schedule slippages
Financial challenges
Non-compliance material deliveries
Project close-out
O&M start-up
Each project will have specific potential breaches related to the project's deployed technology
juxtaposed with the skills/experiences of the team or other special situations. Once these
potential situations are anticipated, the appropriate management response (counteraction) can
be established and implemented.
Principle 6. Systems are more effective than personal intervention.
As managers, we often rely on our ability to detect and facilitate correction of compliance
problems. This strategy of personal intervention works well on small 5 or 10 person projects.
On most projects where mass execution is in progress, however, personal intervention
methodology is at a great disadvantage to systems methodology.
Therefore, we have concluded that when a quality deviation is detected, there are two
necessary responses: 1) correct the problem and 2) improve the QC system. Obviously the
deviation has to be corrected. But equally critical is determining the real cause of the
deviation and fully correcting the QC system to prevent these types of problems in the future.
It is important to note that a fully functioning, well developed QC system will correct
problems that the personal intervention manager will never see, and will never need to see.
FINDINGS -- Vision of Future OA
Surely, some readers are asking "why does anyone have to verify compliance?" If we have a
competent contractor, should we not expect quality results and should we not be able to save i:he cost
of QA to obtain more RD/RA? In response, let me repeat an often mentioned vision. Today, we use
QA to verify QC. Tomorrow, QC will be able to stand alone.
Upon contractors fully developing QC programs to: 1) qualify the inputs to the construction process
sufficiently in advance to prevent compromise, 2) measure results and feedback that are internal to
the construction process and 3) progress toward producing only quality outputs, QC will no longer
require QA.
CONCLUSIONS
At what point will contractors successfully operate QC without QA when cost and schedule are given
more importance than quality? QC will stand alone when contractors realize that:
1. Quality is not counteractive or subordinate in importance to cost and schedule
performance.
2. Quality is equal in importance to cost and schedule performance.
3. Quality is the means by which to achieve excellence in cost and schedule performance.
The foundation of a successful quality management program is the QC program. Most construction
QC programs, however, currently need support from a QA program. To assist in the development
of a fully reliable and cost effective quality assurance plan, we have provided a few key principles.
These principles provide the guidance by which one can test current applications.
888
-------
Specifications for Hazardous and Toxic Waste Designs
Gregory J. Mellema, P.E.
U.S. Army Corps of Engineers
Omaha District
215 N. 17th Street
Omaha, Nebraska 68102
(402) 221-4707
INTRODUCTION.
An essential component of any design or remedial action is the contract specifications. As technology
for Hazardous and Toxic Waste (HTW) continues to rapidly progress, it is imperative that the contract
specifications be accurate, understandable, and practical. The objective of this paper is to discuss
how specifications are currently prepared and to detail several ways to improve them, especially when
prepared for HTW designs.
Two things can be said about a HTW construction project with certainty: (1) changes will be made
during the course of construction, and (2) the contract manager and the construction contractor will
seldom initially agree on the effect the changes will have upon a project (Cooney, 1989). When the
contract specifications are not properly written, disputes, claims by contractors, and extra costs, due
to controversy, quickly materialize. Many of these conflicts would have never developed if the
specifications had been properly written in a clear, concise, and understandable manner.
In order to prepare a good specification, the specification writer has numerous and assorted sources
of information available. However, one tool that is severely lacking for HTW designs is the
availability of current, accurate, and comprehensive guide specifications. Many HTW design
specifications are prepared from "scratch" since there are no comprehensive guide specifications
available for reference. Numerous HTW specifications are modified from general construction guide
specifications, which do not adequately address HTW concerns. There is a need in the HTW field
for specification uniformity and additional HTW guide specifications. The Corps of Engineers is
currently working to prepare "Guide Specifications" particularly geared for HTW remedial designs.
DISCUSSION
Specifications are written instructions which describe all the technical requirements of a contract.
In general, contract drawings show what work is to be done, while the specifications are written
descriptions of the quality, performance, and workmanship of the final product. In order to write
a clear and comprehensive specification, the designer should have a thorough understanding of the
work to be accomplished, knowledge of the materials and methods to be used, and the ability to
communicate these ideas in an understandable manner.
Perhaps the most difficult job of the engineer/designer is to translate the technical requirements of
the contract into a document that can be understood by engineers, contractors, lawyers, regulatory
agencies, and the public. When a specification is poorly written, claims, disputes, and controversy
will certainly develop very quickly. There are national seminars conducted regularly which discuss
the legal aspects of construction contracts. Many of the topics focus on dispute resolution, claims
against the federal government, and new strategies in construction litigation (Muller, 1991). A large
number of the disputes involve the interpretation of the specifications. It is vital that the contract
specifications are as complete and thorough as possible. The suggestions which follow are intended
889
-------
to provide some of the basic principles of competent specification writing and to highlight :;ome of
the most common problems.
CLARITY. Specifications, whether they are for hazardous waste remedial designs or not, should be
written in the clearest manner possible. Specifications should be written as directions, never
suggestions. Relative terms such as: "reasonable," "best quality," or "in accordance with standard
practice," are indefinite and should not be used (Abbett, 1963). Such phrases leave doubt as to what
work is really required. The expression "as approved by the engineer" can relieve the contractor of
responsibility since the contractor has no way of knowing what the engineer will require. Other
phrases such as, "The contractor shall provide all materials and perform all labor in connection with
each type of construction," "in accordance with these specifications," or "as indicated on the drawings,"
are essentially meaningless and should not be mentioned. Unusual technical jargon should be avoided
if at all possible. It is important to use words that do not have more than one meaning. The
specifications should not repeat, but rather, complement the information already provided on the
drawings.
The phrase "or equal" appears extensively in HTW specifications, as there are countless new and
proprietary products or services available. Basically, this phrase is often inserted into a specification
to allow a substitution of a different product for a specified product (Sprague, 1990). The
Government is particularly interested in generating competition, therefore, if "or equal" is used, the
designer should insure that there are other products available which meet the specifications. In
general, the use of trade names, proprietary items, and preparing a specification by adapting a
manufacturer's description of a product, should be avoided. It is preferable to specify materials or
equipment by preparing a performance specification, or if absolutely necessary, to qualify a
manufacturer's trade name with the words "or equal." The phrase, if used, should also require any
substitutions to be approved prior to use.
Other specifications are not written in a clear manner because of excessive cross-referencing from
paragraph to paragraph, or to HTW laws and regulations, ending up in a confusing run-aiound.
There are many specifications that contain references to standard specifications or regulations which
are unfamiliar or difficult to obtain. Often times when these standard specifications or regulations
are obtained, they are found to be superseded one or more times. All of this forces the contractor
to wade through a maze of papers, searching for the thing he "is to comply with." The following
sentences, taken from a submitted specification to the Corps of Engineers for approval, illustrate this
problem. "The SSHP shall serve as the Accident Prevention Plan (APP) and activity hazard analyses
(Phase Plans), required by F.A.R. Clause 52.236-13, and Paragraphs 01.A.03 through 01.A.06 and
Appendix Y of USAGE EM 385-1-1. Thus a separate APP is not required."
CONCISENESS. During World War II, it was speculated that many specifications could have been
reduced by 50 to 70 percent in length, without losing any of the essentials, by careful editing (Retz,
1943). This assumption is probably as true today as it was in 1943. Specifications should be written
in as much detail as necessary without becoming too verbose. Often times a complete guide
specification is used for a particular type of construction with no regard to the relative importance
of that phase of the work to the job. For example, a comprehensive "grading" specification would
be required and appropriate for the construction of a RCRA landfill cover. However, a complete
"concrete" specification would inappropriate and unnecessary if the only concrete required on the
project was to plug a few abandoned culverts. Usually, these all encompassing guide specifications
are intended to be used in the design and construction of relatively large and complex structures
which require considerable detail.
890
-------
Another area that requires consideration when producing clear and concise specifications is the
elimination of nonessential words. Goodrich provides the following sentences, extracted from actual
contract specifications, which exhibit how the intended meaning is obscured by verbose language.
Paint shall be of such character that it will protect the steel against corrosion without being
injurious to the health of persons drinking the water after the latter has stood in the tank for
three months.
Drain piping in and about the pump room to be supplied by the subcontractor whether
entirely buried in concrete or not.
All material, which, subsequently to the tests at the mill and to its acceptance there, during
manipulation, in the shops under shears, punch, etc., which shows it is not of uniform quality,
as herein specified, and also hard spots, brittleness, cracks and other defects are developed;
such material shall be rejected (p. 108).
REFERENCES FOR WRITING HTW SPECIFICATIONS.
The specification writer is often times an "assembler" of specifications rather than a "writer" of
specifications. The writer often relies upon many diverse and complex sources of information when
putting a specification together. This information is usually recovered from files manually or by
modern computerized data systems. Regardless of how this information is obtained, the specification
writer must ultimately decide which segments are to be included or eliminated. The information
sources which follow are commonly used by designers when preparing HTW contract specifications.
EPA Technical Guidance. The Environmental Protection Agency (EPA) maintains a vast technical
support program that is available to designers and technical personnel of HTW projects. The Office
of Solid Waste and Environmental Response (OSWER) and the Office of Research and Development
(ORD), within EPA, has developed a directory which provides a point of contact for obtaining
technical assistance. The directory is entitled, Technical Support Services for Superfund Site
Remediation and can be obtained by writing to: Technology Innovation Office (OS-10), U.S. EPA,
401 M Street, Washington, D.C. 20460.
The EPA has numerous technical guidance documents, handbooks, and publications available for
designers to utilize for HTW projects. These documents are useful as references for various design
considerations. There are also automated information systems such as electronic bulletin boards, data
bases, and inventory systems which provide information on almost any conceivable question or
problem. Although guidance documents and technical publications do provide a comprehensive
source of information for the specification writer, they are often difficult to translate into
specifications. One of the problems when using guidance manuals or documents to develop a
specification, is that they provide only technical guidance and relatively few design specifics.
Information from Industry. One of the sources available for the engineer to utilize is the vast and
remarkable supply of information and technology from private industry. Most companies are more
than willing to make presentations or send technical information of their product to the engineer.
Many manufacturers will provide test data, sample specifications, or samples of their products as well.
All of this information is important to consider when determining if a material is appropriate for a
particular project (CSI, 1975). The designer should conduct independent lab tests in order to verify
a manufacturer's product performance claims.
Guide Specifications. General construction guide specifications are another source of information
that is extensively utilized by specification writers. The guide specifications are prepared for
891
-------
adaptation to major projects of varying types and different locations in the United States. Certain
requirements have general applicability to all projects, while other requirements must have blanks
filled in; alternative words, phrases or paragraphs to be chosen; or special paragraphs to be added.
One problem with general guide specifications is that they must be continually updated to keep up
with the latest technology and they do not address HTW issues.
Sometimes a previously written specification from another project is often edited and used as a guide
specification. This practice is to be discouraged as each specification should be site specific. Errors
from the previous project may get passed on to the new specifications. The previous project may
have used obsolete technology or cleanup methods.
HTW GUIDE SPECIFICATIONS ARE NEEDED.
Why are HTW guide specifications needed? There are many reasons why an increased emphasis
should be made on producing guide specifications for HTW construction contracts. In a HTW project,
there are additional considerations that must be taken into account, that a "regular" construction
contract does not have. Items such as: dust control, health & safety, site control, duration of site
work, and weather all require careful consideration during design. For example, dust control on a
normal construction project is used primarily to prevent the dust from becoming a nuisance.
However, on a HTW construction project, dust control may be critical, as the dust may be
contaminated and could possibly be transported off the site. HTW guides can help to "flag" the
appropriate HTW considerations to the designer. Guide specifications help to establish the format
to be used, and as far as practical, the specific requirements to be included. Guide specifications are
produced to promote uniformity of construction, provide requirements that have been coordinated
with industry, and serve as convenient work sheets to be marked by the specification writer preparing
project specifications.
HTW guide specifications are a useful tool to the specification writer. The guides provide
information of a general nature on required materials and methods for a project, or several choices
of materials and methods from which selections may be made. Guide specifications will require
careful editing. The specifications are usually written by those considered to be authorities; on the
subject, and are the result of careful analysis of previous projects and industry standards. The guides
usually list almost every practical alternative possible to be covered by that particular specification.
In order for the guide specifications to be most valuable, they must be constantly revised and updated
to keep up with the most recent technology and practical experience.
In addition to providing uniformity, HTW guide specifications also serve as a checklist for designers.
Guide specifications also help to minimize the time required to develop a new specification, thus
helping to reduce "reinventing the wheel." The utilization of guide specifications also help less
experienced engineers put together a comprehensive specification in much less time.
WHAT GUIDE SPECIFICATIONS ARE NEEDED?
The Corps of Engineers has been engaged in formulating and preparing guide specifications that are
specifically written for HTW applications. Currently, the Corps is engaged in the preparation of the
following HTW guide specifications:
(1) Geomembranes
(2) Geonets
(3) Underground Storage Tank Removals
(4) Health and Safety
892
-------
(5) Soil-Bentonite Slurry Walls
(6) Ground Water Monitoring Well
The guide specifications listed above are all in the draft phase at the present time. Others under
consideration for further development are:
(1) Incineration
(2) Solidification/Stabilization
(3) Clay Liners
(4) Clay Covers
(5) Gas Venting Systems
(6) Chemical Quality Data Management
(7) Soil Vapor Extraction Systems
(8) Drum Removal and Handling
The eventual implementation of guide specifications like these will enable the engineer to provide
a better specification for HTW design and construction purposes.
CONCLUSION
Specifications are practically useless unless they are free from the weaknesses and shortcomings that
have been outlined. On the other hand, specifications can fulfill their purpose and be extremely
valuable to the engineer and the contractor alike, if these deficiencies are eliminated. It makes one
wonder why it seems to be common practice to produce drawings of excellent quality, yet our
specifications are especially lacking. It may be that there is a lack of training on the writing of
competent and complete specifications.
In addition, guide specifications may help to bridge the gap from poor specifications to good ones.
It is imperative, in light of today's society, to have specifications that are not only constructible, but
legally defensible as well. The technology in the HTW field is rapidly changing, and it is difficult,
if not impossible to keep up with the latest innovations and techniques for removing or treating
hazardous waste. Guide specifications must be continually updated in order to incorporate lessons
learned from previous cleanup projects and new technology.
A necessary corollary to the writing of good specifications is that of adequate inspection and quality
assurance. It will serve little purpose to have very well-written specifications if they are not strictly
observed and implemented. This will only happen if there are competent inspectors at the project.
Therefore, any engineering organization that neglects to provide for adequate and complete inspection
is shortchanging itself. Such inspections are as important as well prepared design drawings and
specifications.
REFERENCES
Abbett, R.W., (1963), Engineering Contracts and Specifications, John Wiley & Sons, New York, New
York, 461 p.
Construction Specifications Institute, (1975), CSI Manual of Practice, Vol. 1, 208 p.
Cooney, J.A., (1989), "Evaluation of Procedures for Claims Presentation and Resolution", Superfund
'89, Proceeding of 10th National Conference, Hazardous Materials Control Research Institute, pp.
457-458.
893
-------
Goodrich, C.M., (1943), "Some Amusing Malapropisms", Civil Engineering, Vol. 13, No. 2, February,
pp. 107-108.
Muller, F., (1991), "Construction Contracts and Litigation Popular Topics", Civil Engineering NEWS,
March, p. 4.
Retz, R.T., (1943), "Why Specifications?" Civil Engineering, Vol. 13, No. 9, September, pp. 457-458.
Sprague, C.J., (1990), "Are We Equal to "Or Equal?"", Geotechnical Fabrics Report, Vol. 8, No. 3,
May/June, p.64.
U.S. Environmental Protection Agency, (1986), "Superfund Remedial Design and Remedial Action
Guidance", U.S. EPA, PB88-107529, June, 105 p.
894
-------
Lessons Learned During Remedial Design and
Remedial Action Activities at Superfund Sites
Dev R. Sachdev
EBASCO Services Incorporated
160 Chubb Avenue
Lyndhurst, NJ 07071
(201) 460-6434
INTRODUCTION
The Superfund Program has recently entered into an active phase of Remedial Design (RD) and
Remedial Action (RA) at various uncontrolled hazardous waste sites. There is very little history and
experience available to address the remediation of hazardous substances in various media such as soil,
sediments, groundwater and surface waters. In addition, EPA's emphasis to encourage the evaluation
of innovative technologies during feasibility studies and the use of Small Businesses/Small
Disadvantaged Businesses places further pressure on the engineering and construction professionals
to be very watchful in managing the RD and RA Projects and avoid or minimize mistakes, reduce
losses and perform the RD and RA assignments in a cost effective manner. It is with these issues in
mind that we share Ebasco's experiences on a limited number of RD and RA Projects which we have
completed.
After the RI/FS and public input, the Record of Decision (ROD) is signed by the Regional
Administrator which establishes the preferable alternative to remediate the site. If an innovative
technology has been selected in the ROD, the validity of the technology is mostly based on bench
scale treatability studies performed during RI/FS which then have to be supplemented by pilot scale
treatability studies during the design phase. In some cases, it may happen that the pilot scale studies
may provide data which cast doubts on the applicability of the chosen technology and in turn require
revisiting the whole treatment concept.
Whereas the RI/FS provides information about the extent of lateral and vertical contamination at a
site, further site investigations are invariably performed during the design phase to clearly identify
and define the areas of contamination to be remediated and the extent of the contaminant plume. The
information is used to calculate the quantities of the contaminated source materials to be remediated
and to locate the extraction/injection wells for the pump and treat system for groundwater treatment
etc. Engineering analysis of soils is also performed if foundations are to be designed for the building,
air stripping column, incinerator or any other structure. Sometimes, for groundwater remediation,
the pump test is also performed during the RD phase to determine transmissivity, and storage
coefficient to assist in the design of the pump and treat system and the duration of time for which
this system would operate.
In certain cases, public input becomes critical and the preferred alternative advocated by EPA is
modified to take into account local concerns. These and similar other issues impact on the cost and
schedules of the RD work assignment.
As regards the RA assignments, it is extremely important that the drawings, specifications, and
general conditions be well written, straightforward, simple and unambiguous. Evaluation criteria and
schedule of deliverables should be clear and well defined. It is very important that the
communication between the EPA, the state, the contractor and the community should be initiated well
895
-------
ahead of the start of the remedial action and should be continuously maintained for the duration of
the remedial action.
This paper discusses our experiences at several Remedial Design and Remedial Action Projects in
USEPA Region II under the REM III and ARCS II Programs and focuses on the problem areas
encountered at various sites and the corrective actions taken.
BACKGROUND
Ebasco Services Incorporated (Ebasco) by virtue of being a REM HI Contractor (1985-1990) and the
current ARCS II Contractor has worked on several Remedial Design and Remedial Action Projects
in EPA Region II. The number of sites we have completed RDs and RAs is small compared to the
number of sites where we have completed RI/FS work. We have completed Remedial Designs on sites
where the RI/FS was performed by Ebasco and also on those sites where the RI/FS was performed
by other consultants. Similarly, Remedial Action has been completed by Ebasco at sites where the
Remedial Design was performed by either Ebasco or other consultants. We have selected three (3)
projects to illustrate our experiences: Bog Creek Farm, Monmouth County, NJ (RD), Brewster
Wellfield, Putnam County, NY (RD & RA) and Vestal Water Supply Well 1-1, Broome County, NY
(RA). A brief description of these sites is given below;
BOG CREEK FARM SITE (RD)
The Bog Creek Farm Site, 12 acres in area, is a National Priority List (NPL) Superfund site located
in Howell Township, Monmouth County, New Jersey. It is alleged that in 1973 and 1974, paint
wastes, disinfectants and trash were dumped in a disposal area of approximately four acres in the
eastern portion of the site. The major source of contaminants was located in a covered trench that
ran west to east for about 150 feet. The primary source of contamination at the site was due to
volatile and semivolatile organics (i.e., benzene, toluene, and xylene) located beneath the ground
surface in the trench area. The leachate from this trench area contaminated the groundwater between
the trench and a brook. The surface water and sediments of the farm pond and the adjacent bog were
contaminated. The Record of Decision (ROD) for the first operable unit, signed in 1985, specified
the excavation and incineration of the contaminated soils, pond and bog sediments and the on-site
treatment of aqueous wastes. As part of the Remedial Design (RD) Scope of Work, Ebasco developed
technical specifications and drawings for the Thermal Treatment of soils and water treatment
including air stripping and carbon adsorption units. Additional items included in the design package
were dewatering activities associated with excavation of soils and air monitoring due to the on-site
incineration. Ebasco developed the technical specifications and the bidding documents for the
procurement of a general contractor by US Army Corps of Engineers (COE) to provide detailed
design and construction services. A schematic diagram showing various components is shown in
Figure - 1.
The Remedial Action bid package was complex; on-site incineration of the contaminated soils and
buried wastes had to be coordinated with the dewatering activities associated with the excavation of
contaminated soils below water level and the treatment of on-site contaminated water. The discharge
from the water treatment system had to meet the stringent New Jersey requirements. Also the ash
generated from the on-site incinerator (i.e., the cleaned soil product) had to be monitored for the
leachable fraction of heavy metals, such as lead and chromium, to ensure that the incineration
process, which was designed to remove the combustible organic fractions, did not, in fact, generate
new disposal problems at the site.
Ebasco also assisted EPA and the COE in developing the bid evaluation criteria and the strategy for
procuring a general contractor. The procurement strategy had to be compatible with the technical
896
-------
LU
DC
CO
>-
z
LU
5
5
LU
0)
o
CO
ui Q
tSS
CO < <
« 2 c
5 = 0
cc S <
Ul
la
OC U. «5
O O LU
O H
O Z
Ul
5
X
o
m ui 52
LU
OC
OC
LU
X
K-
897
-------
specifications, ensuring that innovative technologies would not be excluded from consideration.
Ebasco was also responsible for the engineering support to COE/EPA during construction activities.
The engineering support primarily included review of the technical submittals by the construction
subcontractor.
BREWSTER WELLFIELD
The Brewster Wellfield, which provides water to approximately 2200 people in the Village of
Brewster, Putnam County, New York had become contaminated with volatile halogenated organics
from a dry well located adjacent to a dry cleaning establishment. The Village, under a demonstration
grant form EPA, installed a full scale stripping column which is successfully providing a water supply
to the Village and meeting the applicable standards. However, source (groundwater and soils) control
measures were not instituted. The RI/FS work was completed in July 1986. A Record of Decision
(ROD) signed in 1986, called for the following actions:
o Continued operation of the existing air stripping column to treat the Village's water supply.
o Design and construction of a groundwater management system consisting of extraction wells,
treatment of extracted water by a new air stripper, and injection of treated water to contain
the plume of contamination and restore groundwater quality.
Ebasco developed the remedial design (RD) which involved detailed plans and specifications for
implementing the selected groundwater management alternative consisting of extraction wells,
treatment of extracted water by air stripping, and reinjection of treated water through eight (8)
injection wells. The contaminated groundwater contained 6,000 ppb of VOCs such as TCE and PCE.
The treatment system included four (4) stainless steel extraction wells, each containing one
submersible pump to extract approximately 12 to 20 gpm, a 35 feet high and two (2) feet diameter
counter current flow air stripping column and appurtenances and eight re-injection wells located
upgradient of the plume and each of 8" diameter. A schematic diagram showing various components
is shown in Figure - 2.
Ebasco prepared the final design drawings, technical specifications and contract documents
incorporating written comments from EPA and the State of New York. A final engineer's cost
estimate including O&M costs was prepared. Ebasco invited bids, selected the construction
subcontractor and provided construction management services to complete the remedial action.
VESTAL WATER SUPPLY WELL 1-1
The Vestal Water Supply Well 1-1 was one of the water supply wells which provided drinking, water
to the Town of Vestal in Broome County, New York. The Vestal Well 1-1 is located on the south
bank of the Susquehanna River. The well was contaminated with volatile organic contaminants
(VOCs) such as Trichloroethane (TCA) and Trichloroethylene (TCE).
Ebasco, under the ARCS II Contract, was contracted by EPA to perform construction management
services. The goal of the assignment was to reinstate Well 1-1 as a potable water supply for Vestal
Water District No. 1. The design involved the installation of an air stripping column, booster pump,
air blower, clearwell and process instrumentation and controls that comprise a 1,000 gpm VOC
removal facility. A related objective was to deplete the contaminated underground plume by
continuously withdrawing it and removing the contaminants through treatment processing. A
schematic diagram showing various components is shown in Figure - 3.
898
-------
III III
SS!
35
S2»
Sg2
P2*
<«•
3
(0
S
Ul
I-
(/>
>-
c/>
°zi
~ UIQC
^50
jUI<
g UJ^Q
u. > <
z uij "^ Z f >
OH< ^ *^
pig DC ^JZ
o- ujs5
>UJUJ
Sfeg
DC 5 c/)
COQ>-
o
DC
o
0
EXTRACTION
WELL NO. 4
899
-------
Ul
§
2
uZ
uT>:<
_ltf)OC
-Jo
jyt<
^i
toK<2
>^
900
-------
Representative Construction Management activities included, preparation of bid package, selecting
the construction subcontractor, directing mobilization activities of the selected subcontractor,
approving subcontractor's shop drawings and incorporating revisions where necessary, overseeing
adherence to the subcontractor's approved health and safety plan and quality assurance and quality
control procedures. The work which was completed included final inspection and certification,
preparation of "As Built" drawings, overseeing of initial start up and trial period performance and
eventually turning over the operation of the system to the State of New York/township of Vestal.
DISCUSSION
In order to ensure the completion of the RD and RA Projects under the Superfund Program on
schedule and within budget, it is imperative that the specifications and drawings and the information
for bidders should be complete and unambiguous and should achieve the following objectives:
Complete Remedial Action on schedule and consistent with the ROD
Minimize change orders
Minimize claims and all settle claims amicably before project close out.
To achieve these objectives, Ebasco invariably performs in-house biddability, constructability and
operability reviews prior to finalizing the bid package. Additionally, a site visit for all the potential
subcontractors is conducted and an effort is made to answer/clarify all of their questions during and
after the site visits. Addenda are issued, as necessary, to ensure that all subcontractors receive the
same information so that their bid documents reflect identical scope of work and competitive prices.
Ebasco follows the Construction Specification Institute (CSI) format for technical specifications and
follows Ebasco Engineering Procedures to prepare design drawings, specifications and cost estimates
for the RD/RA Projects.
In spite of these preventive measures and best intentions, there are always some issues/problems
which require ingenuity, technical and managerial skills and perseverance on the part of site managers
to successfully complete the RD/RA work assignment. Therefore, the primary purpose of this paper
is to share with you our experiences on RD/RA work assignments which we have completed under
REM III and ARCS II Contracts.
1. Good planning and scoping during all phases of a project results in expedited completion as
well as costs savings.
Good planning of RI/FS, RD and RA at a hazardous waste site is accomplished through in-
house brainstorming sessions, scoping meetings with regulatory agencies (EPA, state, COE,
etc.) and avoiding vagueness in the scope of work to be let out to subcontractors. In the case
of the Brewster Wellf ield Site, there were questions by bidders about the levels of protection,
QA/QC, Health and Safety requirements, on-site storage of excavated material, size of
shipping containers and sampling procedures which resulted in bids far higher than the
budgeted amount. The site conditions and all these issues were discussed in negotiating
sessions with the bidders and they were requested to modify/ revise their bids and submit Best
and Final Proposals. These technical clarifications resulted in approximately 30% reduction
in the bid price.
The prebid meetings and site visits with the potential subcontractors should be used as an
opportunity to expand on the scope of work, the QA/QC and Health and Safety requirements,
and other site specific issues which could impact the schedule or cost. Good planning
901
-------
invariably includes assignment of experienced, competent and motivated individuals on these
projects. Good planning should result in;
Absence of bid protests
Minimal change orders
Minimal claims
Project completion on schedule and within budget
2. Sensitivity to local concerns is extremely important for the expeditious and satisfactory
completion of a project.
Non-hiring of the security guards from the local area on a site by a subcontractor resulted in
material pilferage, destruction of equipment, manhandling of security guards and other
incidents. Discussions with local officials and the Mayor revealed a very hostile attitude
towards EPA, Ebasco and the security subcontractor. It was extremely difficult to get the
work done even during daytime. The workers were afraid to go to this site. A number of
meetings were held by EPA and Ebasco with the Mayor, other local officials and the local
union leaders. After a few meetings, the issue was amicably resolved and the security guards
were hired from the local area.
At another site it was decided to provide alternative water supply to the residents of a
community whose water supply was contaminated. The water supply to the affected residents
was to be taken from another town's water supply system. The town would not allow the hook
up unless EPA promised funds for the repairs and upgrading of the existing outmoded water
filtration plant. A number of meetings were held with the town officials and negotiations led
to the satisfactory resolution of the town's demands.
In order to avoid public meetings becoming very volatile and hostile, it is extremely important
to address local concerns and plan a number of meetings with Mayor and other interested
parties prior to the scheduled public meeting.
3. To encourage participation of SBEs and SDBs, extend a helping hand to ensure successful
completion of small projects.
Based on the current understanding between EPA and COE, the ARCS contractors will
provide construction management services on those work assignments where the remediation
cost is less than five (5) million dollars. In this category of RA projects, a large number of
projects would be within the $1 - $2 million range. In order to meet our subcontracting goals,
it is extremely important that the SBEs and SDBs be encouraged to participate in the bidding
phase of those projects. There is an acute shortage of SBEs and SBDs who are m the
hazardous waste remediation business and those who are in this business are not well versed
in the special requirements such as QA/QC, Health and Safety, etc., needed to do this work.
It is therefore our responsibility to show willingness to train them and help them understand
the site specific demands, if any, so that they can appreciate and properly account for various
costs in their bids. In the case of Vestal, the work was allotted to a SBE based on his bid
price. The contractor had no knowledge about the content or interpretation of the QA/QC
and Health and Safety plans or their implementation but he was keen and cooperative to
implement these requirements. Ebasco virtually prepared the QA/QC and Health and Safety
plans for him, and helped him understand and implement these on site. This was a rewarding
902
-------
experience for Ebasco to have encouraged and trained an SEE in his pursuit of hazardous
waste business.
4. During RD, the information generated during RI/FS, if not checked, may lead to problems
during RA.
Vestal Water Supply Well 1-1 was a water supply well providing water to the town of Vestal.
The well became contaminated and was out of service for a number of years. The remedial
action project required pumping water from well 1-1, treating it through an air stripping
column and supplying the treated water to the Vestal Water Supply System. After the
treatment system was constructed and all the equipment such as booster pump, valves, air
blower and process instrumentation were installed, trial run began. It was discovered that the
yield of the well 1-1 after a few weeks dropped approximately 60% below the expected yield
of approximately 1,000 gpm and was down to 400 gpm. The Well 1 -1 was a water supply well
and was previously providing 1,000 gpm to the town's water supply system. The well was
contaminated with volatile organic contaminants and was out of operation for a number of
years. It is presumed that the well screens got clogged during the extended period of non
pumping. During the design phase the yield of the well was not tested and all the equipment
was designed based on the reported yield of approximately 1,000 gpm. Since the yield during
the trial run was hardly 400 gpm, it was decided to redevelop the well. The equipment was
dismantled, the well was redeveloped and it was possible to bring back the yield to within
90% of the original yield. Unnecessary cost, delay in schedule and the over design of the
equipment could have been avoided if the well yield had been tested during RI/FS or RD
phase.
5. Experienced construction supervisor/superintendent can save lots of agony and minimize
claims.
A majority of the remedial actions to be performed by ARCS contractors cost less than $5
million and further, most of the remedial actions in this category cost less than $2 million.
It is very difficult to hire trained construction inspectors/supervisors for such small projects.
We at Ebasco were fortunate in that many of the inspectors/ supervisors on the RA projects
came from the Ebasco Constructors group and therefore, there were very minor problems on
site and there were minimal claims from the subcontractors. Proper documentation at
Brewster WellField site reduced the claims from $31,000 to $5,000. The claims were related
to delays, additional out-of-scope work, and an alleged different scope of work. The Site
Manager was able to deny/settle those claims primarily based on the documentation he had
prepared in his files, responding to the subcontractor's claims immediately and elaborating
all the circumstances which the subcontractor had knowledge of and did not take preventive
measures to control the damage. However, at the same time, it should be our intention to pay
the genuine claims of the subcontractors.
6. Establishment of a credible relationship with the subcontractors always benefits the project.
It is always beneficial for the project and the agencies if there is credible relationship and
feeling of trust between the prime and the subcontractor. A positive attitude and a
willingness to accommodate each others' point of view will certainly result in successful
completion of the project within schedule and with minimal claims. We at Ebasco therefore
go through an extensive training program of the site managers and prepare them for
cooperative and sincere effort on their part to work with a multitude of subcontractors.
903
-------
In addition to these site specific issues, the following are some of the general issues which should be
carefully considered during RD and RA projects.
i How clean is clean
The issue of "How Clean is Clean" has not been finally resolved and always becomes an issue
between EPA, state and the local community. It has been observed that the community wants
the clean up to be performed to the n'th degree and sometimes even below the background
levels. Similarly, states also want to be sure that the clean up levels being agreed to would
adequately protect human health and environmental and there is a tendency to be somewhat
conservative. The remediation of water or soil at the contaminated site to the condition that
existed before contamination took place is a laudable goal, but can we achieve this? What is
the definition of background levels and how many samples should be taken so that the results
are statistically significant? These questions require guidance from EPA, the state and other
agencies involved.
We have seen that if this issue is not resolved early on, the project could linger on for a long
time. We have also learned that the site manager should work closely with the EPA-RPM and
initiate a scoping meeting between EPA, the state, and other interested parties :;uch as
environmental groups, community leaders, etc., to start the debate on clean up levels and
come up with a resolution.
ii Performance vs detailed specifications
Normally when there is a RD work assignment, it is expected that the specifications and
design drawings would be detailed enough so that the construction can proceed. However,
in the case of hazardous waste remediation projects, sometimes it is not possible to do this and
we have to settle with performance specifications because;
EPA's mandate is not to restrict to a specific treatment technology and that the
competition should be open to as many technologies as possible. For example,
"Thermal Treatment" instead of "Incineration" is generally specified in the technical
specifications.
The input concentration of the waste feed materials is not consistent during the
treatment process operation. The contamination levels in soil and water being treated
can vary drastically during a very short period. It will be therefore difficult to design
a system based on a specific well defined contaminant concentration level.
It is better to leave it to the construction subcontractor how he wants to layout his
operations including the laydown area, trailers, equipment, treatment system and
associated appurtenances, etc., rather than show those details on the drawings.
It has therefore been Ebasco's experience to have a mix of both performance and detailed dra.wings
and specifications; performance specifications for treatment processes and detailed specification and
drawings for items such as access roads, buildings, extraction/injection wells, pumps, blowers and
other equipment.
iii Documentation of important decision/agreements
Particularly during the construction phase of a project, there are many verbal discussions
between the engineer and the construction subcontractor, between the engineer and the lead
904
-------
agency and other agencies and many times agreements are reached, verbal orders are given
and implemented. All these agreements should be documented and copies sent to those who
participated in the agreements. Minutes of meetings should be prepared and copies should
be sent to all who participated in the meeting. All telephone conversations should be properly
recorded and copies of telecon sent to the other parties. Any misunderstanding and
misinterpretation of agreement can result in project delays, cost increases and in some cases
lawsuits.
iv Prequalification of subcontractors
It is very important to keep a list of well qualified subcontractors in various areas of expertise
such as drilling, surveying, treatability studies, remediation, etc. A questionnaire is sent by
Ebasco to the interested subcontractors to be completed. The qualifications are carefully
evaluated and, if considered suitable, the subcontractor's name is added to the list.
Recommendations are requested from the site managers on the performance of subcontractors
on their project and the list is updated based on these recommendations. If the performance
of a subcontractor is not satisfactory, his name is removed from the list.
The bid package for the construction contracts should include the following evaluation criteria
in addition to price and other criteria so that well qualified subcontractor is selected and the
project proceeds smoothly;
Prior experience.
Experience of the key personnel who will work on the project
Equipment that will be available on site
Agreements with haulers of hazardous waste material
Agreements with disposal sites which will accept the waste
These criteria should be taken into account during technical evaluation of the proposals. A
well qualified subcontractor will complete the project on time and within budget and would
be ultimately cost effective even if his price was not the lowest.
v. Ensuring competitive bids.
The bid package should be well written, simple, unambiguous and the scope of work should
be described in great detail so that there are no misunderstandings and misinterpretations.
It may be desirable to include aerial photographs and site conditions data in the bid package.
A pre-bid site visit should be conducted. The site visit should be conducted by a person who
is knowledgeable not only about the site but also about the technical and contractual
requirements in the bid package. All questions during the site visit should be answered as
completely as possible and followed by an addendum to all the potential bidders. All
questions answered on the phone to individuals should be also consolidated and sent to all the
potential bidders so that all the bidders have the same information.
vi. Minimizing claims and change orders
905
-------
It is our experience that well written, unambiguous specifications and drawings providing
adequate details go a long way to minimize claims and change orders. The following
suggestions are made in this regard;
Do not leave "open ended" items; choice of materials for example, should not be left
to the subcontractor.
Be sure that the equipment and materials specified are readily available in the market
Clearly specify the deliverables and the schedule, such as weekly reports, QA plans,
shop drawings, as-built drawings from the subcontractor.
Avoid as far as possible specialty (one of a kind) items. They are more expensive and
difficult to obtain. Servicing and parts replacement may also be difficult.
Be certain to verify that all equipment and materials received on site are in
conformance with the specifications.
Maintain logs and documentation of subcontractor's personnel and equipment on site.
Spare and unneeded equipment should be noted to avoid claims later for "stand by"
charges.
CONCLUSION
1. Additional investigations, treatability studies, pump tests during RI/FS or RD are good
investments and can save a substantial amount of money during remedial action.
2. RD and RA at hazardous waste sites should be taken as a cooperative effort among all the
participants including EPA, COE, consultants and the subcontractor.
3. The construction specifications and drawings should be well written, unambiguous, and in
sufficient detail so that nothing is left to the imagination.
4. Good forward planning, selection of a competent and experienced project team will ensure
smooth and successful completion of the RD/RA project.
5. Sensitivity to local concerns is extremely important for the expeditious and satisfactory
completion of the project.
6. Prequalifying the subcontractors and establishing a credible relationship with the construction
subcontractor benefits the project.
7. Early resolution of some of the issues such as "How Clean is Clean" would avoid unnecessary
delays and keep the project focused on clean up goals.
8. Documentation of decision/agreements reached during verbal discussion, telephone
conversations or during project status meetings is very important and will minimize claims.
906
-------
Forecasting Staffing Requirements
for Hazardous Waste Cleanup
Robert W. Salthouse
(Author(s)' Address at end of paper)
INTRODUCTION
The Directorate of Civil Works of the U.S. Army Corps of Engineers (USAGE) needs to be able
to forecast the staffing levels required to establish procedures for cleaning up hazardous waste sites
and supervise the contractors who perform the cleanup. The Corps performs those tasks in support of
the U.S. Environmental Protection Agency (EPA). The forecasts are used to plan for future work and
to report environmental staffing needs to the Office of Management and Budget.
The Logistics Management Institute (LMI) developed a Civil Works' Superfund staffing
requirements model based on a statistical analysis of historic workload data [1]. In developing the
model, we assume that the size and complexity of future cleanup programs will be related to the size
and complexity of past programs. While a wide variety of factors affect staffing levels, we found that
the two most important ones are total project cost and project type or complexity. By dividing the
Corps' programs into different types of work, we can reliably relate total project costs in dollars to
hours worked. The three types of work we use in our model are remedial design, supervision of
remedial construction, and additional technical assistance to the EPA.
We used historical data to determine (1) the relationship between total project cost and hours
expended for various types of work; (2) the distribution of project sizes, durations, and start dates; and
(3) the functional relationship between time spent and work accomplished. Those relationships and
distributions are embodied in a computer program - the Superfund staffing model - that takes
multiyear program dollars as its primary input and produces multiyear forecasts of staffing levels as
its primary outputs.
Because the Corps' Superfund program is relatively new, however, and the volume of historical
project data incorporated in the model is currently very small, we recommend that the prototype model
be used with caution. For that reason, we also recommend that USAGE collect additional project data
annually from its divisions and districts at the same time that it collects the annual inputs for the
staffing model. It can use that additional project data to refine the prototype model.
BACKGROUND
In cases in which EPA is unable to locate a primary responsible party (PRP) for the cleanup and
in cases in which the PRP is unable to pay the cleanup costs because of bankruptcy or for other
reasons, EPA assumes the PRP's role. Those cases are called Federal lead Superfund projects. Instead
of merely monitoring the process to ensure that the cleanup meets EPA standards, EPA must award
the contract and directly supervise the site cleanup.
When Superfund was new, EPA attempted to use its own in-house personnel to supervise the
design and remediation actions. As the number of sites increased, the tasks of engineering, contract
administration, and contract supervision soon overwhelmed EPA's internal staff, which then turned to
other agencies for help. Since both parts of the remedial action stage - engineering and construction
supervision - are similar to the type of work that the USAGE Directorate of Civil Works carries out
in the normal course of its business, EPA turned to the Corps for help in the remediation stages.
907
-------
USACE now aids EPA in three major areas. First, it carries out design and engineering in
house for remediation actions and it supervises architect-engineers who are under contract to perform
such work. Second, it supervises construction companies that perform the actual removal or
remediation. And, third, it provides technical assistance to EPA, an effort that is less intensive than
design or construction but that requires the technical expertise of USACE engineers. Most technical
assistance projects for EPA to date fall into two categories: feasibility studies and hazardous waste
enforcement support.
When EPA assigns a Federal lead Superfund project to USACE for design and construction,
USACE first provides technical assistance by reviewing the feasibility study that decided on the
chosen cleanup technology. When providing hazardous waste enforcement support, on the other hand,
USACE monitors PRP-led cleanup projects. In this role, USACE does not directly supervise the
project because that is the PRP's responsibility. Instead, it "looks over the shoulder" of the PRP and its
contractor to ensure that the project is carried out properly and that the site is cleaned to the desired
levels.
DISCUSSION
Use of Historic Data
The purpose of the Superfund staffing model is to be able to reliably forecast the staffing levels
needed by USACE to support the EPA's Superfund work several years into the future. We base that
forecast on the statistical analysis of historical data. The historic approach is sound if two conditions
hold: past work was performed efficiently and future work will continue to be similar to past work.
Predictive factors developed from historic data that include inefficiently managed project:? will
simply perpetuate those inefficiencies. However, since USACE's costs for design and construction
management services have been shown to be comparable with those of other Federal, state, and local
Government agencies and with large private-sector companies [2], which provide a measure of USAGE
efficiency, we can use properly sampled USACE data to develop predictive factors that reflect general
industry standards. If we assume that USACE carries out Superfund work at the same level of
efficiency as its other work, then historic USACE Superfund data can similarly be used to develop
predictive factors for efficient restoration work.
In addition, we can account for changes in USAGE'S program mix over time by dividing the
workload into different types of work. Thus, when the mix between those types of work shifts, the
model will continue to predict staffing reliably. For example, by separately forecasting staffing needs
for in-house design, design contracted out, construction, and different types of technical assistance, we
can continue to forecast future needs even if the program moves from an emphasis on remedial design
to an emphasis on remedial action (construction) as more Superfund sites move from the planning and
design phase to the cleanup phase.
Within work types, we assume that future work will be similar to past work. However, since
the Superfund program is relatively new, the nature of the work will possibly change in the future.
For example, EPA - USACE's customer - is moving from a reliance on traditional construction
contracting to a greater emphasis on cost-plus, or reimbursable, contracting. Cost-plus contracts
cannot be as closely specified as conventional contracts and require more USACE supervision. We
have attempted to account for that difference by dividing construction work into reimbursable and
nonreimbursable work.
Because the Superfund program is relatively new, the volume of past work is just barely
sufficient for statistical analysis. In addition, much of the available data was incomplete, further
restricting our ability to generate sufficient sample sizes (and, consequently, restricting our ability to
subdivide the work further into different types of design and construction work). For those reasons,
908
-------
USACE will need to continue collecting the data needed to revise and "tune-up" the prototype model,
in addition to acquiring the input data needed by the forecasting model.
In developing a model based on statistics, we must be careful to choose those factors that are the
best predictors of future staffing. The predictive factors must not only perform well statistically, they
must also be practical. That is, they must be relatively easy to collect without having to undertake a
massive annual data call. In addition, the predictive factors must be leading indicators. For example,
program breakage - stops and starts in program scheduling and execution - undoubtedly affect work
hours. However, changes in staffing and program breakage move concurrently; one cannot be used, in
advance, to predict the other. Moreover, program breakage is already contained in the historic data so
that staffing and workload factors developed from those data will include some normal or average
level of breakage.
A wide variety of factors determines and influences staffing levels. Many of those factors,
however, are not useful in forecasting because they move randomly over time. Since we cannot predict
their behavior, we cannot use them to forecast staffing. Some factors may change very slowly so that,
in practice, they have very little effect on staffing changes. Still other factors, while significant, are
strongly correlated to total project cost. That is, such factors exhibit strong collinearity with the total
project cost. For example, longer projects certainly require more hours of work, but they also generally
cost more. Total project cost, therefore, acts as a proxy for length of time. Project complexity is
another signifu nt indicator of staffing requirements, and it is strongly collinear with project type.
Our past experience with USACE staffing models has shown that the two most important
factors are total project cost and project type or complexity. Not only are they good indicators of
staffing required, but they are also easier to use as inputs than many alternate factors.
In practice, we must choose forecasting factors that can be projected into the future. One of the
advantages of total project cost is that a large portion of USACE's Superfund program in any given
year consists of projects that were started in previous years. Therefore, the forecast for the next 2 to 3
years can be based partially on the existing program and partially on a prediction of the future
program.
The forecasting method uses two basic types of predictive factors. First, we must "spread" the
total program cost over a number of years and second, we must relate it to hours worked. While the
forecasting model includes some additional subtleties, those two factors form the backbone of the
predictive methodology.
Spreading the Work
Since total program cost does not translate into workload for a single year only, it is necessary
to spread those program dollars over a number of years. The historic data show that all types of
Superfund work include projects that take anywhere from a few months to 5 years to complete. Thus,
in any given year, USACE is conducting projects that started in the current as well as in the previous
4 years.
In our model, the spreading algorithm takes three factors into account: project start date,
project duration, and the relationship between chronological time and work hours. We developed
spreading factors for three basic types of Superfund work: remedial design, remedial construction, and
technical assistance. Ideally, we would prefer to develop spreading factors for more types of work and
to check that those spreading factors are significantly distinct. However, we did not have large
enough sample sizes to subdivide design, for example, into in-house design and contracted design. In
some cases, we ran along the margins of statistical significance even for only three project types.
Future data collection should permit more sophisticated spreading calculations by including more
project types.
909
-------
Table 1 presents the distribution of project starts over the fiscal year. That factor is important
since even if a project takes only 6 months to complete, it will cross into 2 fiscal years if started at any
time after March of the fiscal year. The data show that start dates were fairly evenly distributed over
the year. (For comparison, the last column in Table 1 shows a perfectly random distribution of start
dates, i.e., the distribution that would result if one project were started per day, with a total of
365 projects.) That is, the distribution shows no particular bias toward starting projects at the
beginning, middle, or end of the fiscal year; projects have a more or less equal chance of starting at any
time.
TABLE 1
DISTRIBUTION OF PROJECT START DATES BY MONTH
Month
October
November
December
January
February
March
April
May
June
July
August
September
Total
Standard deviation
Sample size
Remedial
design
6.45%
3.23
12.90
9.68
6.45
9.68
968
9.68
323
12.89
645
9.68
100%
3 07%
31
Remedial
action
8 1 1 %
10.81
10.81
8.11
5.40
8.11
5.40
10.81
8 11
8 11
811
811
100%
1.73%
37
Technical
assistance
12.73%
3.64
5.45
7.27
1456
10.91
5.45
1091
7.27
727
5.45
9.09
100%
3.19%
55
Random
start date
8.49%
8.21
849
849
773
8.49
8.21
8.49
8.21
8.49
849
8.21
100%
0.22%
-
The second major factor in determining how the total project cost is spread is the distribution of
project durations: the percentage of each type of project that took less than 3 months to complete, the
percentage that took from 3 to 6 months, and so on. Table 2 shows that distribution for the three
Superfund project types. Even though the duration data for remedial design are sparse, the resulting
findings are reasonable: 78 percent of the projects took less than 3 years to complete, while a few have
taken as long as 4 to 5 years. Interestingly, almost half of the construction projects undertaken to date
have taken, or USAGE expects them to take, less than a year to complete.
The third factor that must be taken into account in spreading the total project cost is the
relationship between chronological time and work time. That is, even if a particular project takes
exactly 2 years to complete, we cannot assume that an equal number of staff hours are spent in each of
those 2 years. Figure 1 shows these relationships for the three major types of Superfund work. As the
graph illustrates, technical assistance projects appear to require more hours up front, while
construction work starts more slowly, gathers steam, and then tapers off toward the close of the
project. While these relationships are based on relatively sparse data, they are consistent with our
910
-------
TABLE 2
DISTRIBUTION OF PROJECT LENGTHS BY QUARTER
(Completion date less start date)
1 quarter
2 quarters
3 quarters
4 quarters
1 year
5 quarters
6 quarters
7 quarters
8 quarters
2 years
9 quarters
10 quarters
1 1 quarters
12 quarters
3 years
13 quarters
14 quarters
15 quarters
16 quarters
4 years
17 quarters
18 quarters
19 quarters
20 quarters
5 years
Total
Sample size
Distribution
Remedial design
0.0
00
11 1
00
11 1
11 1
00
11 1
11 1
333
22 2
00
11 1
00
33 3
00
11.1
00
00
11 1
11 1
00
00
00
11 1
100
9
Remedial action
16.7
33
133
133
46 7
200
33
100
33
36.7
33
100
00
00
13.3
00
00
0.0
00
00
3.3
00
00
00
33
100
30
Technical assistance
7 1
00
21 4
17.9
464
10.7
36
00
3.6
17.9
7.1
7 1
00
36
179
00
36
7 1
00
10.7
36
00
0.0
3.6
7.1
100
28
Wote: Numbers may not add because of rounding
911
-------
Hours
worked
(percent)
100 i-
75
50
25
— Remedial design
• Remedial action, construction
— Technical assistance
25 50 75
Chronological time (percent)
100
FIG. 1. WORK HOURS VERSUS CHRONOLOGICAL TIME
912
-------
experience with military design and construction work. Future data will improve the prototype
model, but in practice, the appearance of these relationships will probably change very little.
The model does not require all of the data just described as direct parameters because we
combine them to calculate a set of spreading factors for each project type. Table 3 shows the final
result, which is incorporated in the Superfund model. As shown in Table 2, projects in all types of
work start in program year N and continue for as many as 4 years beyond it. The work accomplished
in the last year for all three project types, however, is a relatively small percentage of the total; the
bulk of the hours are spent in the first 2 years.
TABLE 3
SUPERFUND SPREADING FACTORS
Program year
N
N+1
N + 2
0 + 3
N + 4
Total
Remedial design
30 1 %
365
206
97
3 1
100%
Remedial action
49 3%
41 5
7.5
1.2
05
100%
Technical
assistance
51 8%
31.0
92
56
2.4
100%
Relationship Between Dollars and Hours
We used the statistical technique of simple linear regression to derive the relationship between
workload and staffing for the various project types. Despite the scarcity of data, it was essential to
divide the work into more than three types since we know a priori, for example, that in-house design
should require more staff hours than the supervision of design contracted out. Nevertheless, the
statistical measures of significance for our small samples show the measured coefficients to be
statistically significant.
Workload was measured as program amount for design and as contract amount for construction.
In all cases, we corrected the dollar amounts to FY90 constant dollars to maintain comparability
among years. The basic linear regression equation was as follows:
where
Hours = c + a X Workload 4- e
k
Hours — the dependent variable, i.e., the quantity we want to predict,
c = & constant term that reflects the nonvariable portion of staffing per project,
a = the coefficient of workload, i.e., the weight attached to workload to predict staffing,
Workload = the independent or predictive variable, and
913
-------
e = an error term that accounts for random variation in staffing unaccounted for by workload.
The Superfund staffing model incorporates the results of a number of linear regression
equations. Those results are shown in Table 4. In addition to the constant term and the coefficients,
the table includes two measures of statistical significance - the t-statistic and R2 - plus the sample
size.
TABLE 4
REGRESSION RESULTS - DESIGN AND CONSTRUCTION STAFFING FACTORS
Work phase
Design
In house
A-E
Construction
100% complete
Current
Constant
(hours/
project)
0.0
00
0.0
00
Coefficient
(hours/
$ million)
2,562
1,960
1,458
1,941
t-statistic
8.6
63
9.7
7 7
R2
90%
64
72
67
Sample size
5
5
12
18
Note: A-E = architect-engineer, i e , supervision of design work contracted out
The t-statistic is a statistical indicator that tests for the hypothesis that the coefficient is
significant, that is, the coefficient is nonzero. If the t-statistic is greater than 2, then the probability
that the variable is not zero is at least 95 percent. As Table 4 shows, all of the t-statistics exceed 2
(The t-statistic is the ratio of the coefficient to its standard error, which is a measure of the statistical
variability of that coefficient.)
The R2 is that fraction of the variance of the dependent variable that is explained by the
independent variable. In terms of our model, it is the fraction of staffing explained by the dollar
workload (for each particular type of project). Even though the lowest R2 is 64 percent, each equation
predicts staffing for a single project only. When a large number of projects are combined, as la the
Superfund program, the equations are summed and the variance around a single project becomes far
less important. In mathematical terms, the error term (e) is random; although for one particular
project the error term has the potential to be quite large, the sum of all the ei for terms tends to become
smaller as more and more projects are summed, since the individual errors cancel each other out.
The R2 does indicate, however, that other factors in addition to total program cost influence
staffing. That finding is not unexpected. More data may eventually allow us to split the work types
into smaller subdivisions and increase the predictability of each equation. But it is also likely that the
R2 will not increase materially. Many factors influence staffing and not all of those factors can be built
into a practical model. The coefficients, however, are an unbiased estimator of staffing and on
average, given enough projects, should provide forecasts that are effective for planning purposes,
particularly at the headquarters level.
The measures of statistical significance show that the estimated coefficients are reasonable
predictors. However, the sample sizes were very small in all cases; ideally, the sample sizes should
exceed about 20 for each linear equation. The sample size requirements are based on the central limit
914
-------
theorem as applied to regression equations. Nevertheless, the model should suffice as a prototype
although the need to collect more data in the future to expand the sample sizes and to confirm and
recalibrate the relationships is obvious.
In all cases, linear regressions were calculated for an unconstrained constant, as well as a
constant constrained to zero. In each case, the equation with a zero constant term exhibited the best
significance indicators and so it was adopted for the staffing model.
We explored the effects of economies of scale by trying nonlinear terms - including both
logarithmic and squared terms - in the regression equation. However, the statistical indicators did
not show those additional nonlinear terms to be significant.
We estimated two regression equations for Superfund construction, or remedial action, work.
The first equation, labeled "100% complete" in Table 4, represents all of the completed projects for
which historic data were available. The other equation represents the incomplete, or "Current,"
projects. Total hours were calculated for that set of projects by adjusting for percent complete.1 The
current projects, so adjusted, indicate higher staffing requirements per dollar. While the difference
may be due to the small sample sizes in both cases or to inaccuracies resulting from the adjustment of
hours, it is also conceivable that hours per dollar have increased because of changes in the type of
work, or possibly an increase in cost-plus contracting. Again, while the results are acceptable for use
in the prototype staffing model, the equation needs to be refined with additional data in the future.
Technical Assistance
The third category of Superfund work is technical assistance, which is not directly tied to total
program cost. Therefore, it is not possible to derive a relationship between total program cost and staff
hours. Instead, we found that average hours per project type was a good predictor.
As shown in Table 5, we found that staff hours expended on such projects differed by the type of
work. That is, feasibility studies clustered around an average of 281 hours, while hazardous waste
enforcement support clustered about an average of 1,147 hours. Given the limitations of sample size,
both appeared to be normal distributions with relatively low variances, for which the average is the
unbiased estimator. Almost no data currently exist for any other types of technical assistance work.
TABLE 5
TECHNICAL ASSISTANCE - AVERAGE HOURS WORKED
Project types
Feasibility studies
Hazardous waste enforcement support
All technical assistance
AMPRS
codes
922
923
All
Average
hours
281
1,147
532
Standard
deviation
116
680
547
Sample
size
16
7
24a
Wote: AMPRS = Automated Management Project Reporting System.
a Includes all projects in codes 922 and 923, plus one project code 926, Remedial Investigation/Feasibility Study.
JWe also adjusted incomplete design project hours, but the results were statistically poor.
915
-------
The model, therefore, forecasts staffing for technical assistance projects based on average hours
for the type of work. For technical assistance projects other than feasibility studies and hazardous
waste enforcement, the prototype model uses the average staff hours for all technical assistance
projects.
The Staffing Model
The staffing model takes design program amounts, construction contract amounts, and the
number of technical assistance projects as its primary inputs. All inputs are split into different project
types, such as remedial response and emergency response, whether or not we were able to develop
different factors for those splits. That makes it easier to modify the model's predictive factors in the
future as well as making it easier to audit and to modify the model inputs. An additional input is the
percent of design work that is accomplished in house versus work done by architect-engineers
(contractors) and supervised by USACE. Other inputs include the number of work hours per year for
converting staff hours into work years.
The model first spreads the program inputs, whether dollars or numbers of projects, into
multiple years before applying the regression factors (or average hour factors) to determine staff
hours. The model converts all dollar amounts, input as then-year dollars, into 1990 constant dollars to
preserve the original regression relationships. The coefficient for each project type is multiplied times
the workload after spreading. In addition, the model multiplies the constant times the number of
projects since the constant was determined for a single project.2 The model estimates the number of
projects per year by dividing the workload measure by the average project dollar size (shown in
Table 6). The number of technical assistance projects, of course, is a direct input.
TABLE 6
AVERAGE PROJECT DOLLAR SIZES
Work phase
Design
In house
A-E
All
Construction
100% complete
Current
All
Average
($ million)
1 2
1 4
1 3
48
16.3
11.5
Standard
deviation
1 1
09
0.9
69
15.3
13.7
Sample
size
6
28
34
16
22
38
Placement is estimated by taking a percentage of program amount, after spreading. This is
displayed as a model output and is also used as an input to the calculation of division and district
overhead. The model outputs staffing in work years and placement in dollars after reconverting from
1990 constant dollars into then-year dollars.
2Although the constant terms in the prototype model are all zero, the model retains this calculation in the event i.hat
future data produce nonzero constants.
916
-------
The model also estimates the number of work years of support required for the Superfund
program by Corps of Engineers divisions and districts. Since we could not measure those hours
directly, we adopted the overhead factors used in the Corps of Engineers Resource and Military
Manpower System (CERAMMS) [3]. We assume that the CERAMMS factors, which are based on
design and construction placement, reflect efficient management and will remain substantially the
same for all types of design or construction. The constant terms in the CERAMMS division and
distrk i overhead equations were set to zero, however, since additional Superfund work (or any other
type of work) will add only to the variable portion. The factors are shown in Table 7.
TABLE 7
DIVISION AND DISTRICT STAFFING FACTORS
Placement
type
Design
Construction
Variable factors (hours/5 million)
Division
0
296
District
765
422
CONCLUSIONS
At LMI, we have built other models that forecast staffing needs for USACE's military programs
and for the Defense Environmental Restoration Program. Based on that experience, we found that
historical data are a reasonable guide to future behavior. Relationships based upon these data can be
modified to reflect process changes and efficiency improvements, when appropriate. We also found
that although a great many factors affect staffing levels to some extent, the two most important factors
are total project cost and project type or complexity.
EPA's Superfund efforts are relatively new and USACE's assistance to EPA started in 1983.
For that reason, the small amount of project data limited our ability to analyze the data for
relationships between staffing and a wide variety of factors. However, our previous experience showed
that total project cost and project type were overwhelmingly the most important predictive factors for
staffing.
While our statistical indicators confirm that those predictive factors work as well for the
Superfund, the relatively small sample sizes mean that we have less confidence in the specific values
of the coefficients that we derived for those predictive factors. If future projects continue to be similar
to our sample of completed past projects in nature and labor-intensity, the coefficients will accurately
predict future staffing requirements. If, however, those past projects do not constitute a truly random
sample of "typical" USAGE Superfund <-ork — if, for example, they are all uncharacteristically labor-
intensive — then the resulting forecasts may be too high (if the opposite, then the forecast will be too
low).
One indicator that the Civil Works' Superfund coefficients are not too wide of the mark is that
they are of the same order of magnitude as the coefficients derived from very large sample sizes (and
subsequently validated) for various types of USAGE military work. For example, the supervision of
military construction work for the Army requires about 1,700 hours per $1 million compared with our
coefficients for Superfund construction work of between 1,460 and 1,940 hours per $1 million.
917
-------
We conclude that the USAGE Environmental Restoration Division can use the prototype
Superfund staffing model to produce rough planning estimates and we have recommended that it be so
used. USAGE divisions can also use the model to forecast their own staffing needs, but they must keep
in mind that the model's results will display greater variation at the division level than at the overall
USAGE level. As the number of projects handled by each division grows, the individual variation
among projects will become less important and, therefore, division forecasts will become more precise.
In addition, we have recommended that USAGE gather more Superfund project data as
additional projects are completed. USAGE can use the larger sample sizes that result to rerun the
statistical analyses and to refine the prototype Superfund model.
DISCLAIMER
The report upon which this paper is based was prepared pursuant to U.S. Army Gorps of
Engineers Contract DACW31-90-D-0076. The views expressed here are those of the Logistics
Management Institute at the time of issue but not necessarily those of the Department of the Army.
REFERENCES
[1] Forecasting Staffing Requirements for Hazardous Waste Cleanup. Robert W. Salthouise. LMI
Report CE004R1. February 1991.
[2] Management Costs of DoD Military Construction Projects. Paul F. Dienemann, Joseph S.
Domin, and Evan R. Harrington. LMI Report ML215. April 1983.
Monitoring and Controlling Engineering and Construction Management Cost Performance
Within the Corps of Engineers. William B. Moore, Eric M. Small, and Jeffrey A. Hawkins.
LMI Report AR801R1. December 1988.
Cost-Competitive Construction Management: A Review of Corps of Engineers Construction
Management Costs. William B. Moore and Jeffrey A. Hawkins. LMI Report AR603R3.
June 1990.
[3] Corps of Engineers Resource and Military Manpower System. William B. Moore, Robert W.
Salthouse, Robert A. Hutchinson, Dr. Robert L. Crosslin. LMI Report AR603R1. May, L987
Author(s) and Address(es)
Robert W. Salthouse
Logistics Management Institute
6400 Goldsboro Road
Bethesda, Maryland 20817-5886
(301)320-2000
(Autovon 287-2779/2127)
918
-------
The Effects of the Davis-Bacon Act on the
LaSalle Electrical Utilities Phase I Remedial Action
David P.Seely
U.S. Environmental Protection Agency
230 South Dearborn St.
Mailcode 5HS-11
Chicago, Illinois 60604
(312) 886-7058
Karen Yeates
U.S. Environmental Protection Agency
230 South Dearborn St.
Mailcode 5HS-12
Chicago, Illinois 60604
(312) 886-3873
INTRODUCTION
The Davis-Bacon Act is a federal labor regulation which establishes
minimum wage rates and fringe benefits for workers on federally
assisted projects in excess of $2,000 which are defined as
construction by the USDOL. These minimum wage rates and benefits
are established on a regional basis by the U.S. Department of Labor
(USDOL).
The LaSalle Electrical Utilities Phase I Remedial Action (RA) was
managed as a "State-Lead" project by the State of Illinois under a
cooperative agreement (CA) with the U.S. Environmental Protection
Agency (U.S. EPA), whereby 90% of the project costs are provided by
the Federal government. Since the RA was considered to be
"construction" under 40 CFR Part 33 and was funded through a CA,
the Superfund procurement regulations required that the
construction contractor utilize the Davis-Bacon Act to establish
the wages and fringe benefits for its employees who worked on the
site. The construction contractor was selected through the formal
advertising process. At the time the bids were received, USDOL had
not yet established worker classifications for all the types of
jobs needed on the site. This situation created much confusion
among all parties involved.
This paper summarizes the events which occurred and the resulting
confusion regarding the applicability of, and the liability for
compliance with, the Davis-Bacon Act to the Phase I Remedial
Action. Based on the experience gained from resolving the Davis-
Bacon issue for the LaSalle project, this paper presents three
recommendations to ensure the same issue does not become a problem
for future Superfund remedial actions.
919
-------
BACKGROUND
The LaSalle Electrical Utilities (LEU) Superfund site resulted from
improper wastes management practices by a former manufacturer of
electrical equipment. The Electrical Utilities Company (EUC)
started manufacturing electrical capacitors at the site prior to
World War II and continued until 1981, when it relocated to North
Carolina. By the late 1940s, the company had begun to utilize
polychlorinated biphenyls (PCBs) in its operation. This
manufacturing practice continued until October 1978. In May 1981,
manufacturing operations ceased at the LaSalle plant.
Subsequently, the Illinois Environmental Protection Agency (IEPA),
enforcing Section 34 of the Illinois Environmental Protection Act,
ordered the production areas of the plant to be sealed. The LEU
office building remained in use by a lessee until some time in the
early 1980s. Since that time, the entire facility has been
abandoned.
Information on the waste management practices of the company is
limited. Undocumented reports allege that PCB-contaminated waste
oils may have been applied as a dust suppressant both on and off
the property as late as 1969. Subsequent to the federal regulation
of PCBs, inventory reports document the disposal of PCBs at
approved facilities.
Beginning in September 1975, numerous government agencies,
including the United States Environmental Protection Agency (U.S.
EPA), the IEPA, and the Occupational Safety and Health
Administration (OSHA), conducted various inspections and issued
numerous complaints and orders to the EUC company as a result of
its manufacturing and handling practices. In 1982, a U.S. EPA
Field Investigation Team contractor completed a preliminary
investigation of the site. As a result, the site was proposed to
be listed on the National Priorities List (NPL) under the
Comprehensive Environmental Response, Compensation and Liability
Act of 1980 (CERCLA) , also known as "Superfund". The site was
proposed for the NPL on December 30, 1982 and became final in the
first publication of the NPL on September 9, 1983.
Analysis of site records revealed only one Potentially Responsible
Party, EUC, from which the U.S. EPA could seek reimbursement of
costs associated with the environmental remediation of the site.
EUC, however, was not financially viable and had petitioned for
relief under Chapter 11 of the Bankruptcy Act on September 19,
1983. Therefore, any action taken under CERCLA authorities had to
be financed by the Superfund. The IEPA assumed the role of the
lead agency in investigating and remediating the site and received
Federal funding through a CA since 1983.
920
-------
Remedial Investigation
Between 1983 and 1988, many Immediate Removal Actions (IRAs) were
completed and the Remedial Investigation (RI) was conducted. The
IRAs involved various measures including waste consolidation, drum
removal, site fencing, and containing or encapsulating the
contamination. The RI found: 1) extensive PCB-contaminated soil
both on and off the site, 2) PCB-contaminated structures which also
contained asbestos, 3) groundwater contaminated with volatile
organic compounds and PCBs, and PCB contaminated sewer and stream
sediments. By 1986, the IEPA had collected enough information to
determine that the off-site PCB-contaminated soil posed an
immediate threat to the public health. The U.S. EPA and IEPA
decided to split the site activities into phases and conduct an
operable unit to address the off-site soil contamination.
The first operable unit, or Phase I RI, indicated that off-site PCB
contaminated soil existed in the following areas: along the
shoulders of an adjacent road for about 1000 feet to the north and
approximately 1.2 miles south of the EUC property, the residential
area directly east of the site, the small commercial area south of
the property, and one residence north of the site. The
concentrations of PCBs found in these areas ranged from less than
0.20 parts per million (ppm) to as high as 5800 ppm. The RI also
documented low levels of PCBs inside houses and commercial
buildings. The highest levels detected were 0.58 ug/100cm2 from a
wipe sample and 13 ppm from a sample of vacuum cleaner dust.
Feasibility Study
The Phase I Feasibility Study (FS) included an exposure assessment
to evaluate acceptable cleanup standards. The exposure assessment
determined that a 10~6lifetime risk level corresponded to a soil
concentration range of 0.05 to 0.5 ppm of PCBs in soils. A risk
level of 10 Corresponded to a range of 0.5 to 5 ppm PCBs. At the
time, there were no formally established cleanup standards for PCBs
in soils. U.S. EPA and IEPA considered two draft policies,
combined with the results of the exposure assessment, to select an
appropriate cleanup level. The policies that were considered were
the draft National Toxic Substances Control Act (TSCA) PCB Spill
Cleanup Policy and the U.S. EPA Office of Research and
Development's advisory levels for PCB cleanups at Superfund sites.
After combining available information, the U.S. EPA and the IEPA
selected a cleanup level of 5 ppm PCB in the soil with a minimum of
three inches of clean soil cover. Below 12 inches in depth, a
cleanup level of 10 ppm PCBs would be applied.
The U.S. EPA and the IEPA determined that the structures (homes and
businesses) needed to be cleaned since samples had already
documented low level contamination inside homes, and it was likely
that contaminated particles would be blown or tracked into the
921
-------
structures during excavation. The draft TSCA PCB Spill Policy was
also utilized to establish a cleanup level for the structures where
excavation would occur. The cleanup levels for the structures were
established at 0.5 ug/100 cm2 for high contact areas and 10ug/100
cm for other surfaces.
The Phase I FS evaluated a variety of alternatives for the cleanup
of the site. Five alternatives ( no action, off-site landfill,
off-site incineration, on-site incineration, and on-site storage)
for the cleanup of the contaminated soil and three alternatives (no
action, conventional industrial cleaning, and specialized cleaning
with replacement) for the structural contamination were evaluated
in detail. All of the alternatives for the soil and structures
were put released for public comment.
Record of Decision
On August 29, 1986, the U.S. EPA Regional Administrator signed a
Record of Decision (ROD) selecting on-site incineration for the
cleanup of the contaminated soils and industrial cleaning for all
structures where excavation would occur. The IEPA had also signed
a ROD selecting the same remedy.
Remedial Design
The IEPA also assumed the lead agency role for the remedial design
(RD) and the remedial action (RA) for the LaSalle site. Between
August 1986 and July 1987, design documents were prepared by
Ecology and Environment, Inc. (E&E) for the IEPA. These documents
contained design drawings and technical specifications defining the
requirements for the excavation of the soil, operation of the
incinerator, cleaning of the structures, sampling and analysis, and
various other activities associated with the remedial action (RA).
Procurement
After the design documents were completed, IEPA proceeded to
procure the RA contractor using the two-step formal advertising
procurement process. Since IEPA had certified that their
procurement system complied with the federal procurement
regulations, IEPA completed the RA contractor procurement with
little input from the U.S. EPA, except for U.S. EPA participation
on the review team for evaluating the construction contractor's
technical proposals in the first step of the procurement. The IEPA
completed the procurement and entered into a contract with the
lowest responsive, responsible bidder. The IEPA entered into the
RA contract with the Westinghouse Electric Corporation (also known
as Westinghouse/Haztech, Inc.) on December 1, 1987.
922
-------
Remedial Action
Between December 1987 and June 1988, Westinghouse prepared all the
necessary work plans, sampling plans, safety plans, etc. which
needed to be approved prior to initiating field work. Actual
excavation started in approximately June 1988. During the summer
of 1988, the question of the applicability of the Davis-Bacon Act
to the remedial action arose.
Davis-Bacon Act
The debate on the applicability of the Davis-Bacon developed from
two inquiries. First, the oversight contractor inquired of the
IEPA about receiving certified payrolls from Westinghouse. At this
time the IEPA was not receiving certified payrolls which are
required under the Davis-Bacon Act. Secondly, at the same time, a
local labor union, the International Union of Operating Engineers
(IUOE) Local 150, filed complaints to the USDOL in July, 1988
alleging that Westinghouse was not paying the appropriate wages to
its employees working on the LaSalle project. Settling this issue
went from answering a simple question to what eventually became a
very complicated dispute made up of contradictions by all parties
involved. The participating parties included: Westinghouse, IEPA,
Region V U.S. EPA, HQ EPA, Regional USDOL, HQ USDOL, as well as
many local, state, and federal politicians.
Discussion
A detailed chronology of events which occurred during efforts to
settle the issue, whether the Davis-Bacon Act wages applied to the
LaSalle remedial action and, if so, which party would be
financially responsible to comply with the decision, is provided in
Attachment A.
On July 15, 1988, the IUOE Local 150 requested that the USDOL
investigate Westinghouse's alleged Davis-Bacon wage violations on
the LEU Superfund remediation project. Since that day,
Westinghouse, IEPA, U.S. EPA (regional and headquarter offices),
and the USDOL (regional and headquarter offices) became involved in
a long and complicated dispute over whether the Davis-Bacon Act
applied to the LaSalle remedial action and, if so, which party
would be financially responsible for compliance with the labor
standard. Throughout the dispute, several other interested parties
were involved in one way or another. The interested parties
included: IUOE Local 150, U.S. Senator for Illinois Paul Simon,
U.S. Representative for the district in which the site is located,
J. Dennis Hastert, the Illinois Attorney General, Illinois Senator
Patrick Welch, and both the regional and local press.
Throughout the dispute, Westinghouse maintained the position that
the LaSalle remedial action did not constitute construction since
923
-------
the project only involved excavation of contaminated soil,
incineration of contaminated soil, backfilling of clean soil,
landscaping, and industrial cleaning of houses where excavation
occurred. Therefore, since the project did not involve
construction, the Davis-Bacon Act did not apply. Westinghouse
maintained their position even though the USDOL had made the
decision that the remedial action at the LEU site constituted
construction and therefore the Davis-Bacon Act was applicable.
Westinghouse also maintained the position that neither the Bidding
Documents nor its contract with the IEPA clearly indicated the
project was either considered construction or that the Davis-Bacon
Act applied.
The USDOL made their initial determination that Davis-Bacon applied
and directed the IEPA to amend their contract appropriately and to
require that Westinghouse submit certified payrolls in December,
1988. The USDOL later confirmed its determination of the
applicability of Davis-Bacon but reversed its directive to have
IEPA amend the LaSalle contract in February, 1989. After further
review, the USDOL concluded that the LaSalle contract and related
Bidding Documents adequately referenced the Davis-Bacon Act and,
therefore, the contract did not need amending. The USDOL's
position was construction is defined as not only construction of
structures, but includes such actions as excavating, landscaping,
and earthmoving.
In January, 1989, the U.S. EPA headquarters received a request from
U.S. Representative J.Dennis Hastert for the U.S. EPA's opinion
regarding the LaSalle labor issue. In March, 1989, the U.S. EPA
headquarters responded to the inquiry by concurring with the USDOL
determination that the LaSalle remedial action was construction and
that the Davis-Bacon Act applied to the contract. However, the
U.S. EPA deferred to the USDOL, as the authority for defining
construction and coordinating the administration and enforcement of
the Davis-Bacon Act requirements in CERCLA and agreed that the IEPA
contract with Westinghouse did not need to be amended.
Based on: 1) the USDOL's determination that the Davis-Bacon Act
applied to the project, 2) the USDOL1s opinion that the contract
and related bidding documents contained the appropriate references,
and 3) the U.S. EPA's acceptance of the USDOL's decision, the IEPA
took steps to bring Westinghouse into compliance with the Davis-
Bacon Act. Westinghouse responded to the IEPA directives by
disagreeing with the USDOL finding that Davis-Bacon applied to the
LaSalle project, but agree to comply with the requests for employee
information and submittal of certified payrolls. Westinghouse also
stated, assuming that the Davis-Bacon Act did apply, any additional
costs to Westinghouse for compliance with this labor standard would
be considered to be the lEPA's responsibility.
Between May, 1989 and August, 1990, the IEPA and the USDOL worked
closely together to enforce the Davis-Bacon wage provisions and to
924
-------
bring Westinghouse into compliance. The IEPA began receiving
Westinghouse's certified payrolls and in a joint effort with the
USDOL, performed an intensive audit of the entire Westinghouse
payroll. They jointly assigned Davis-Bacon Wage Decision IL87-13
job classifications and wage rates for all employees who had worked
for Westinghouse at the LaSalle site. However it was apparent that
the three Westinghouse job classifications associated with
operating and maintaining the mobile incinerator could not be
directly equated with IL87-13 because it made no mention of mobile
incinerators. These three classifications were assigned existing
IL87-13 classifications solely on a logical basis.
During this time period, the IEPA had received a copy of a
Westinghouse letter to the IUOE Local 150, dated May 24, 1988, from
the IUOE Local 150. In its letter, Westinghouse enclosed the
agreement it had made with the union, in which Westinghouse
acknowledged that its contract stipulated Davis-Bacon wages be paid
for the LaSalle project and that Westinghouse intended to pay these
wages to its laborers.
The three incinerator classifications and all other Davis-Bacon
classifications were presented to Westinghouse. Westinghouse
continued to reject the USDOL determination that the LaSalle
project was construction and stated their intent to challenge the
USDOL's definition of "construction". Westinghouse stated it would
cooperate in classifying workers because of the lEPA's mandate from
the USDOL and that its cooperation in no way altered its intent to
legally appeal the construction determination. Westinghouse
disagreed with most of the classifications, particularly those
assigned to the three mobile incinerator job categories. It was
agreed between Westinghouse, the IEPA, and the regional USDOL that
the IEPA would submit a Project Wage Determination for the LaSalle
Phase II remedial action to the USDOL-HQ and the subsequent
determination would be utilized to settle this Phase I dispute. It
was also agreed that the LaSalle laborers needed to be interviewed
for their concurrence on their job description.
During its review for the Project Wage Determination for the Phase
II remedial action, the USDOL-HQ was exploring possibilities for
implementing the provisions of the Service Contract Act based on
the determination that the IEPA could be considered an "extension
of the U.S. EPA" through the cooperative agreement, thereby the
LaSalle procurement could be considered to be direct Federal
procurement. The Service Contract Act is also a federal labor
regulation which establishes wages and benefits based on different
criteria than the Davis-Bacon Act utilizes. The Service Contract
Act can only be utilized for direct Federal procurement. Upon
discussions with USDOL, the Region V U.S. EPA raised concerns
regarding the consistency with the Phase I USDOL regional
determination that the Davis-Bacon Act applied. The Region V U.S.
EPA also raised objections to the determination that the IEPA could
be considered an "extension of the U.S. EPA" since the U.S. EPA
925
-------
could not find a legal basis for such a determination, in fact, one
of the provision for receiving a CA is negation of agency
relationship and that neither party could act on behalf of the
other. Upon further review, USDOL-HQ agreed with the Region V
U.S. EPA position and issued the Project Wage determination for the
LaSalle Phase II project based on the Davis-Bacon Act.
After the IEPA had received the USDOL's decision, the Project Wage
Determination for the LaSalle Phase II remedial action was sent to
Westinghouse and their employees for their concurrence.
Westinghouse again disagreed with the lEPA/USDOL's Davis-Bacon
classifications and there was no consensus among the employees who
responded. Because of this response, the IEPA sent the USDOL a
formal, project-specific wage decision request for the LaSalle
Phase I project. The USDOL issued the Phase I determination on
April 17, 1990. Again, Westinghouse protested the Phase I Wage
decision in a letter to the USDOL on May 7, 1990.
Based on the official wage decision for the Phase I project, the
amount of back wages and fringe benefits were calculated for all of
Westinghouse's LaSalle employees.
Throughout the labor dispute, Westinghouse had maintained they
would require a change order for the entire amount of the back
wages if it was determined that the Davis-Bacon Act applied to the
project. Westinghouse had maintained that the Davis-Bacon wages
were not factored into their bid because the project was not
considered to be construction. In addition, Westinghouse argued
that since wage classifications were not assigned for the three
incinerator categories until April 17, 1990, Westinghouse could
neither have formulated the correct bid nor paid the correct wages
for those categories. On the other hand, the U.S. EPA and the IEPA
had consistently maintained that a change order for these wages
would not be approved because the USDOL had concluded that the
contract and the Bidding Documents had included appropriate
references to indicate that the Davis-Bacon wage rates were
applicable.
On August 2, 1990, the USDOL-HQ issued the official labor wage
underpayment documents for the Westinghouse employees for the
LaSalle Phase I remedial action. The documents contained 98 names
and totalled $751,552.04. The underpayment of the three disputed
incinerator job categories totalled approximately $423,000. The
regional USDOL later issued a letter which gave Westinghouse until
08/24/90 to make the wage deficit payments to its employees.
On August 8, 1990, Westinghouse proposed to settle the Davis-Bacon
dispute in two parts. In Part 1, Westinghouse requested
$888,033.00 for the wage deficiency for the three disputed
incinerator job categories. They then proposed to address all
other labor categories in Part 2, after Part 1 was settled. The
proposed Part 1 settlement included the amount of deficient wages,
926
-------
fringe benefits, overhead cost, and general and administrative
(G&A) expenses. In addition, Westinghouse claimed they were
entitled to a profit, but, in the interest of good will, they would
forego the profit.
The IEPA and the U.S. EPA continued to contend that the contract
and the Bidding Documents indicated that the Davis-Bacon Act would
apply to the LaSalle Phase I remediation. However, since the USDOL
indicated that the disputed incinerator job categories could not be
equated to existing Davis-Bacon wage categories, the IEPA concluded
it would be in the best interest of all parties concerned to settle
the wage dispute without resorting to the threatened litigation
from Westinghouse. However, the dispute was further clouded by the
fact that the U.S. EPA Headquarters had published new Superfund
assistance regulations in June, 1990 (40 CFR Part 35 Subpart 0)
which determined that the excavation and incineration of
contaminated soil would be considered a "service" and not defined
as "construction". The language which defined construction was
never reviewed by Region V of the U.S. EPA, and if it had reviewed
the language, Region V of the U.S. EPA would have never concurred
with promulgation of the regulation because of the proposed
definition of construction, especially in light of the two-year
debate on the LaSalle project. The IEPA presented its
recommendations to the U.S. EPA Region V for its concurrence. Both
parties agreed that it would be in the government's best interest
to seek a fair and equitable resolution of the Davis-Bacon dispute
because the ambiguity of the lEPA's contract, the promulgation of
Subpart O, and the lack of consistency, severely weakened any
positions the agencies could have taken if Westinghouse pursued the
matter in court.
The IEPA countered the Westinghouse proposal by offering to pay
Westinghouse the amount resulting from Part 1 if they agreed to
drop their claim regarding Part 2. Westinghouse agreed to drop its
demands regarding Part 2 of their proposed settlement. The only
hurdle to reach a settlement which now remained was an agreement on
appropriate overhead and the G&A expenses. IEPA did not agree with
the original rates which Westinghouse had proposed in their
original proposal. After further negotiations between Westinghouse
and the IEPA, the rates for overhead and G&A were resolved.
On October 18, 1990, the IEPA received a letter from Westinghouse
in which it agreed to settle the entire LaSalle Phase I wage
dispute for $823,243.23. The IEPA then requested the Region V U.S.
EPA to concur with the proposed settlement. On December 12, 1990,
the U.S. EPA issued written concurrence regarding the proposed
settlement. The IEPA then proceeded to amend their contract with
Westinghouse to reflect the agreed upon settlement.
In summary, it quickly became apparent that all parties involved
were confused, or were at least presenting conflicting positions,
regarding the applicability of, and the liability for compliance
927
-------
with, the Davis-Bacon Act with respect to the LaSalle project.
Clear examples of conflicting information being communicated are
given by the following:
1) The applicability of the Davis-Bacon Act was not directly
addressed by the contract or the Bidding Documents and the IEPA was
not requiring certified payrolls from the beginning of the contract
period.
2) Initial decisions made by the regional USDOL determined that the
Davis-Bacon Act applied to the LaSalle project, and the contract
and Bidding Documents needed to be amended to reflect this
decision. The regional USDOL later reversed its determination that
the contract and Bidding Documents were deficient in covering the
Davis-Bacon Act applicability and, therefore, the documents did not
need to be amended.
3) Westinghouse had consistently presented their firm position that
the project did not entail construction and therefore Davis-Bacon
did not apply. They also claimed that if the Davis-Bacon Act did
apply, Westinghouse could not have been aware of this based on the
contract and Bidding Documents and, therefore, Westinghouse would
not be liable for the costs of compliance. However, the
Westinghouse letter to the IUOE Local 150 clearly indicated that
Westinghouse was aware that the contract and Bidding Documents
indicated that the Davis-Bacon Act applied and that Westinghouse
intended to pay the union workers accordingly.
4) The USDOL regional office had consistently expressed the
applicability of the Davis-Bacon Act for the LaSalle Phase I
remedial action, but the USDOL-HQ, in its review for the Project
Wage determination for the LaSalle Phase II project, initially
explored possibilities to apply the provisions of the Service
Contract Act for the LaSalle Phase II project, The Phase II project
was very similar in scope to the LaSalle Phase I project.
5) The regional U.S. EPA had consistently agreed with the regional
USDOL determinations and firmly believed the USDOL was the only
Agency with the authority to make formal determinations. The U.S.
EPA-HQ had responded to an inquiry from U.S. Representative J.
Dennis Hastert and restated that the U.S. EPA accepted the regional
USDOL determination and also deferred to the USDOL as the decision-
making agency with regard to defining construction applicability
and applying labor standards. However, the HQ U.S. EPA had
published new Superfund assistance regulations in June, 1990 which
defined "construction" in terms of Superfund remedial actions and
suggested the Service Contract wages be utilized for projects which
are primarily excavation and incineration of contaminated soil.
These regulations further add to the confusion since the States do
not have the ability to utilize the Service Contract Act, since it
applies to direct Federal procurement only.
928
-------
CONCLUSIONS
As a result of the experience gained from the LaSalle Phase I
remedial action, the following items need to be addressed:
1) First, the definition of "construction" and the determination of
when the Davis-Bacon Act applies to Superfund contracts needs to be
agreed to, by both the USDOL and the U.S. EPA, to enable both the
USDOL (regional and HQ) and the U.S. EPA (regional and HQ) to be
consistent in applying labor standards. This would enable the U.S.
EPA to accurately advise the States which have CAs.
2) Second, if the Service Contract Act applies to some Superfund
remedial actions, the States need appropriate methods made
available to them by the U.S. EPA for administering the Service
Contract Act for "State-Lead" projects. Otherwise, the States will
be restricted to only assuming the lead agency role for remedial
actions where the Davis-Bacon Act applies.
3) For any Fund-financed remedial action, the correct labor
standards need to be identified early so the contracts and Bidding
Documents can clearly state which labor standards apply to the
projects. This is a must for administering Superfund remedial
action contracts.
If these steps are not followed, it is inevitable that the same
disputes are likely to arise on future projects. The experience
gained from resolving the Davis-Bacon issue for the LaSalle project
shows that this was clearly a time-intensive issue which could have
been settled before it had started. By presenting this paper, the
authors hope that experience gained through the LaSalle project
will be utilized to avoid some major labor disputes during future
Superfund remedial actions.
929
-------
REFERENCES
Black & Veatch, 1986, Phased Feasibility Study for Remediation of
PCS Contamination at the LaSalle Electrical Site (Final
Report). Prepared for the Illinois Environmental Protection
Agency.
Black & Veatch, 1988, Feasibility Study for Remediation of
Groundwater and PCB Contamination at the LaSalle Electrical
Utilities Site (Final Report). Prepared for the Illinois
Environmental Protection Agency.
Ecology and Environment, 1987, Supplemental Remedial Investigation
Report for LaSalle Electrical Utilities. LaSalle. Illinois.
Ecology and Environment, 1987, Bidding Documents for LaSalle
Electrical Utilities PCB Abatement Community of LaSalle.
Prepared for the Illinois Environmental Protection Agency.
Illinois Environmental Protection Agency, 1986, Remedial
Investigation, Electrical Utilities Company. LaSalle.
Illinois(draft). Division of Land Pollution Control.
U.S. Environmental Protection Agency, 1986, Record of Decision for
the LaSalle Electrical Utilities Site.
U.S. Environmental Protection Agency, 1988, Record of Decision for
the LaSalle Electrical Utilities Site.
U.S. Environmental Protection Agency, LaSalle Cooperative Agreement
Davis-Bacon Dispute File.
930
-------
ATTACHMENT A
The following a detailed chronological list of the events which
occurred during efforts to settle the issue of whether the Davis-
Bacon Act wages applied to the LaSalle remedial action and, if so,
which party would be financially responsible to comply with the
decision.
07/15/88 The Illinois Union of Operating Engineers (IUOE) Local
150 submitted a complaint to the USDOL requesting an
investigation of Westinghouse's payment practices at the
LaSalle site. The IUOE Local 150 made allegations that
Westinghouse was not complying with the Davis-Bacon Act
requirements stated in their contract with IEPA.
10/21/88 Not seeing results from their complaint to USDOL, the
IUOE Local 150 requested assistance from Senator Paul
Simon's office to urge USDOL to investigate Westinghouse.
10/26/88 IEPA requested advice and guidance from the Region V U.S.
EPA on the Davis-Bacon wage issue for the LEU project.
This request was made in light of the U.S. EPA developing
new regulations regarding construction contracts. It was
lEPA's opinion at this time the Davis-Bacon Act applied
to the LaSalle project.
Also Senator Simon's office responded to the IUOE Local
150 with assurances that USDOL would be contacted on
their behalf.
10/28/88 Westinghouse requested a site-specific wage decision for
the LEU site.
11/01/88 IEPA sent the U.S. EPA a listing of the categories of
laborers for which the wages were in dispute. The IEPA
requested a Federal determination on the correct wage
classification for these laborers.
11/04/88 A copy of the Westinghouse LaSalle contract and excerpts
from the Bidding Documents pertaining to wage rates were
sent to USDOL by IEPA.
11/07/88 U.S. EPA concurred with the lEPA's determination that the
Davis-Bacon Act should be applied to the Phase I RA
construction contract. The U.S. EPA informed the IEPA
the new Superfund regulation (40 CFR, Part 35, Subpart
O) , in light of the LaSalle situation, would better
define when remedial actions constitute "construction".
Once the new regulation had been promulgated and prior to
executing the subsequent Phase II remedial action
contract, the remedial action would need to be reassessed
to determine if it qualifies as "construction". The U.S.
931
-------
EPA also alerted the IEPA that the USDOL was sent the
disputed labor categories to enlist their help in making
the Federal determination requested in the lEPA's letter
dated November 4, 1988.
The U.S. EPA sent the disputed labor categories to the
USDOL for their review. The USDOL was requested to
advise the IEPA on the correct labor classifications. It
was also indicated to the USDOL that there were
inconsistent opinions at the Regional and Headquarters
level of the U.S. EPA on the applicability of the Davis-
Bacon Act at Superfund cleanups where no tangible
construction is present. The USDOL was told the U.S. EPA
was presently attempting to redefine "construction" and
"services" to be consistent with the Federal Acquisition
Regulations in a new Superfund assistance regulations.
11/09/88 USDOL responded to Senator Paul Simon informing him their
regional office had initiated an investigation of
Westinghouse.
12/05/88 Several Westinghouse employees requested the Illinois
Attorney General's office investigate the LaSalle labor
issue.
12/19/88 The USDOL informed the IEPA that the LaSalle project
constituted "construction" within the meaning of the
Davis-Bacon Act and the Davis-Bacon Act did apply. The
USDOL requested the IEPA to fulfill its enforcement
responsibilities which included: amending its contract
with IEPA to include appropriate clauses of the Davis-
Bacon Act and to require that Westinghouse submit
certified payrolls.
01/13/89 Westinghouse requested the USDOL to reconsider its
determination that the Davis-Bacon Act applied to the
LaSalle project. Westinghouse stated its position that
the LaSalle project was covered by the McNamara-0'Kara
Service Contract Act. Westinghouse's position was based
on its argument that work being performed under its
contract with the IEPA was a "service" and not
"construction". It stated that the amount of work being
done which could be similar to construction was
negligible and the overwhelming portion of the work being
performed (treatment and removal of contaminated soil)
was a "service".
01/25/89 U.S. Representative J. Dennis Hastert inquired of the
U.S. EPA Headquarters if the U.S. EPA accepted the USDOL
determination that the Davis-Bacon Act applied to the
LaSalle project.
932
-------
02/13/89 After continuing review, the USDOL confirmed that the
LaSalle project falls within the scope of the Davis-Bacon
Act but stated that the IEPA contract with Westinghouse
incorporated the appropriate clauses of the Davis-Bacon
Act by reference and indicated the contract did not need
to be amended. The USDOL continued their request of the
IEPA to require that certified payrolls be submitted by
Westinghouse.
03/22/89 The U.S. EPA Headquarters responded to U.S.
Representative Hastert stating that the U.S. EPA accepts
the USDOL determination and deferred to USDOL as the
enforcement authority for the Davis-Bacon Act
requirements.
04/26/89 The IEPA informed Westinghouse of the USDOL's decision
and of the U.S. EPA's concurrence that the Davis-Bacon
Act applied to the LaSalle project. The IEPA requested
information on Westinghouse's employees at LaSalle and
required certified payrolls be submitted corresponding to
all work done under their contract.
05/12/89 Westinghouse responded to the IEPA directive stating that
it disagreed with the USDOL finding that Davis-Bacon
applied to the LaSalle project, but would comply with the
requests for employee information and submittal of
certified payrolls. Westinghouse also stated, assuming
that the Davis-Bacon Act did apply, any additional costs
to Westinghouse for compliance with this labor standard
would be considered to be the lEPA's responsibility.
06/20/89 The IEPA informed Westinghouse that, based on the USDOL
determination in its 02/13/89 letter that the contract
included appropriate references to the Davis-Bacon Act,
the IEPA could not accept financial for compliance with
the Davis-Bacon Act.
06/29/89 The U.S. EPA confirmed its concurrence with the USDOL
determination that the Davis-Bacon Act applied to the
LaSalle contract. The U.S. EPA also agreed with the IEPA
that the contract documents adequately covered the Davis-
Bacon Act and that additional costs for compliance should
not be the responsibility of the U.S. EPA or the IEPA.
07/12/89 The IEPA received a copy of a Westinghouse letter from
the IUOE Local 150 in which Westinghouse enclosed the
agreement it made with the union. The agreement stated:
"The wages cited above are set forth in the Davis-Bacon
(US Dept. of Labor) prevailing wage determination, which
is stipulated in Haztech's contract with the IEPA. The
parties agree that in the event the US Dept. of Labor
933
-------
later determines that a higher wage or benefit rated is
applicable to this project, then all the affected amounts
will be paid retroactive to the project starting date."
07/25/89 A meeting was held with representatives of the USDOL,
U.S. EPA, IEPA and Westinghouse in an attempt to classify
Westinghouse's workers in accordance with the Davis-Bacon
Act. Westinghouse still did not accept the determination
that the LaSalle project was construction and stated
their intent to challenge the USDOL's definition of
"construction". Westinghouse stated it would cooperate
in classifying workers because of the lEPA's mandate from
the USDOL and that its cooperation in no way altered its
intent to legally pursue the construction determination.
Many, but not all, laborer wage categories have been
settled. The IEPA agreed to send Westinghouse a list of
the job descriptions and the corresponding Davis-Bacon
categories and wage rates. This list was agreed to in a
joint effort between the USDOL and the IEPA on 07/24/89.
It was agreed by all parties that the IEPA would ssubmit
a Project Wage Determination for the LaSalle Phase II
remedial action and that the USDOL's determination would
be utilized to settle this Phase I dispute. It was also
agreed that the LaSalle laborers needed to be interviewed
for their concurrence on their job description.
07/26/89 The IEPA sent the list of job descriptions and the
corresponding Davis-Bacon categories and wage rates to
Westinghouse.
08/03/89 Westinghouse disagreed with most of the Davis-Bacon job
classifications and/or the wage rates suggested by the
IEPA/USDOL. Westinghouse also maintained their position
of non-agreement with the USDOL's construction and Davis-
Bacon applicability rulings for the LaSalle project.
08/23/89 The IEPA submitted a Project Wage Determination request
for the LaSalle Phase II remedial action to the USDOL in
Washington, D.C. for approval.
10/16/89 The IEPA received the USDOL Project Specific Wage
Determination for the Phase II remedial action.
01/19/90 The IEPA sent a formal Wage Determination questionnaire
for the Phase I project to Westinghouse and to the
laborers for their concurrence.
01/24/90 The IEPA received a request from the Illinois Attorney
General's office for copies of all certified payrolls
received from Westinghouse and the IEPA complied with the
request the same day.
934
-------
02/02/90
02/09/90
03/02/90
04/17/90
05/07/90
05/31/90
06/19/90
06/20/90
Westinghouse disagreed with the lEPA's wage proposals in
a response to the lEPA's questionnaire.
The IEPA received the final response to their
questionnaire from the laborers. There was no consensus
of opinion from the laborers regarding the IEPA wage
proposals.
The IEPA sent a formal request to the USDOL for the Phase
I remedial action which acknowledged that three of the
Westinghouse job categories were not listed in the
previous determinations.
The USDOL issued a job-specific wage decision for the
LaSalle Phase I project.
Westinghouse sent a letter to the USDOL protesting their
04/17/90 LaSalle Phase I wage decision.
The IEPA sent a letter to Westinghouse stating that the
lEPA's position had not change from the position stated
in its 03/02/90 letter to the USDOL.
A meeting was held between representatives of the IEPA,
U.S. EPA, and the USDOL. This was a meeting held in
preparation for a meeting with Westinghouse to be held
the next day. The USDOL stated, after reviewing
payrolls, ledgers, and employee surveys, the back wages
owed to the direct employees of Westinghouse was
approximately $792,000. The possibility of debarment of
Westinghouse was discussed if they refused to comply with
the Davis-Bacon Act. There also were many discussions on
the recent discovery of the fact that the newly published
final Subpart 0 Superfund regulations contained language
on defining construction in conflict with the USDOL
determination. The Subpart 0 regulation was changed
after the Region V U.S. EPA had concurred with it. The
preamble to the regulation states that the operation and
handling of materials and operation of a mobile
incinerator may be considered services. This language
was never agreed to by the Region V U.S. EPA and it had
caught all parties involved by surprise.
A meeting was held between Westinghouse, the IEPA, and
the USDOL. Westinghouse had stated that it had slightly
different figures than those of the USDOL and would work
with the USDOL to resolve the discrepancy. Westinghouse
also stated their overall costs were between $ 1-2
million, including overhead and administration costs.
Westinghouse was unwilling to pay all costs associated
with compliance but was now open to negotiate a
935
-------
settlement. Otherwise, it was prepared to pursue this
matter in court.
07/90 The U.S. EPA Headquarters issued an Engineering Forum
Fact Sheet containing information on the definition of
construction. The fact sheet stated "Burning
contaminated material and treating contaminated water are
services, not construction" and that remedial actions may
"be either construction, service, or both". It indicated
that for construction, Davis-Bacon applies, but does not
apply to a service.
08/02/90 An audit of certified payrolls by the IEPA and the USDOL
was completed and agreements were made between
Westinghouse and the IEPA and USDOL regarding the method
of calculating back wages. Westinghouse still did not
accept or acknowledge that the LaSalle project was a
construction project or that Davis-Bacon applied.
Westinghouse did agree that if Davis-Bacon did apply, the
total back wages and fringe benefits calculated was
correct as of 08/02/90.
The IEPA received the LaSalle underpayment of wages
roster from the USDOL. The roster contained 98 names and
totalled $751,552.04. The underpayment of the three
categories of incinerator workers was approximately
$423,000.00.
08/08/90 The USDOL issued a deadline of 08/24/90 to Westinghouse
for making a decision of payment of the back wages.
08/15/90 The IEPA received a draft letter containing
Westinghouse's request to settle the wage issue in two
parts. Part 1 was to deal with the underpayment of wages
for the three categories of incinerator workers and part
2 was to deal with the underpayment of wages for all
other job categories. Westinghouse had requested
$888,033.00 to settle part 1. The IEPA sought the Region
V U.S. EPA concurrence in proceeding with negotiations.
08/28/90 Due to the negotiations between Westinghouse and the
IEPA, Westinghouse agreed to pay for part 2 of the wage
negotiations if the IEPA agreed to pay Westinghouse
$888,033.45 for part l. This payment included direct
labor as well as fringe benefits and company overhead.
09/05/90 The IEPA received information from Westinghouse for the
purpose of auditing the overhead rates covering the
period of 1986 to 1989.
09/06/90 The Westinghouse overhead rate information was sent to
the U.S. EPA, by the IEPA, as supplemental information in
936
-------
its request for concurrence for paying Westinghouse the
$888,033.45. This amount was to cover the three
incinerator positions not originally included in the
USDOL's Davis-Bacon wage decisions.
10/11/90 The U.S. EPA verbally agreed that the IEPA proceed with
the $888,033.45 payment to Westinghouse for the purpose
of settling the issue without a legal battle.
10/15/90 The IEPA entered into further negotiations with
Westinghouse, in which Westinghouse agreed to use the
U.S. EPA negotiated overhead and G&A rates. This
agreement reduced the settlement figure to $823,243.23.
10/18/90 The IEPA received Westinghouse's formal offer to settle
the entire LaSalle wage dispute for the agreed amount of
$823,243.23 for the three incinerator job categories.
11/28/90 The IEPA requested written concurrence from the U.S. EPA
regarding the proposed settlement for the entire LaSalle
wage dispute.
12/26/90 The U.S. EPA concurred with the IEPA for the settlement
amount of $823,243.23.
937
-------
SURETY BONDS - SUPERFUND PROJECTS
AUGUST V. SPALLO
District Counsel
Kansas City District
U.S. Army Corps of Engineers
601 Federal Office Building
601 East 12th Street
Kansas City, Missouri 64106
(816) 426-3943
TABLE OP CONTENTS
TOPIC PAGE
Introduction 1
Background 2
Discussion 17
Conclusion 29
References 30
938
-------
INTRODUCTION
The purpose of this paper is to discuss and analyze the
subject of surety bond requirements in connection with construction
and service contracts, and particularly surety bond requirements
for contracts related to the Environmental Protection Agency
("EPA") Superfund program. The need to address this subject arises
as a result of complaints by private sector firms, who are
interested in obtaining government contracts related to the
Superfund program, that performance bonds for hazardous and toxic
waste work are not readily available from corporate sureties.
939
-------
BACKGROUND
Remedial actions under the Superfund program are accomplished
by contracts between the United States and private sector
contractors. The work to be performed can be in the nature of
construction, or services, or a combination of both. The Miller
Act1 is a federal statute that requires performance and payment
bonds for any construction contract exceeding $25,000. The Federal
Acquisition Regulation ("FAR") implements the requirements of the
Miller Act.2 The FAR provides that the penal amount of
performance bonds shall be 100 percent of the original contract
price, unless the contracting officer determines that a lesser
amount would be adequate for the protection of the government.3 In
the case of contracts other than for construction ("service
contracts"), government agencies generally shall not require
performance and payment bonds.4 However, performance and payment
bonds for service contracts may be used as permitted by the FAR5,
as follows:
(a) Performance bonds may be required when necessary to
protect the Government's interest. The following
situations may warrant a performance bond:
(1) Government property or funds are to be provided to
the contractor for use in performing the contract or as
partial compensation (as in retention of salvaged material) .
940
-------
(2) A contractor sells assets to or merges with another
concern, and the Government, after recognizing the latter
concern as the successor in interest, desires assurance
that it is financially capable.
(3) Substantial progress payments are made before
delivery of end items starts.
(4) Contracts are for dismantling, demolition, or removal
of improvements.
The contractor is required to furnish all bonds before
receiving a notice to proceed with the work or being allowed to
start work.6 In addition, the government may require additional
performance bond protection when a contract price is increased.
The increase in protection shall generally equal 100 percent of the
increase in contract price. The government may secure additional
protection by directing the contractor to increase the penal amount
of the existing bond or to obtain an additional bond.7 The
performance bond required appears in the FAR.8 The bond provides
that the surety will be liable to the government for the penal sum
of the bond in the event that:
(1) the contractor does not perform and fulfill the
undertakings, covenants, terms, conditions, and
agreements of the contract during the original term of
941
-------
the contract and any extensions thereto that are granted
by the government, with or without notice to the
Surety (ies), and during the life of any guarantee
required under the contract, and
(2) the contractor does not perform and fulfill all the
undertakings, covenants, terms, conditions, and
agreements of any and all duly authorized modifications
to the contract that hereafter are made. Notice of those
modifications to the Surety(ies) are waived.
Key to this discussion is the definition of construction and
service contracts. The Service Contract Act does not define
services except by exclusion. If it is not construction it is
service by default. Conversely, Congressional Conference Committee
reports indicate if it fits under construction, it is not service.
The preference, if any, is clearly toward construction.9 The
Service Contract Act specifically exempts contracts for
construction, alteration and/or repair, including painting and
decorating of public buildings or public works.10 "Construction",
"building" or "work" under the FAR are broadly defined and they
include such things as altering, remodeling, and installation of
items fabricated off-site, painting and decorating, and
transporting materials or supplies to or from a public works,
buildings, or structures. Construction activity is distinguished
from manufacturing, furnishing of materials, or servicing and
942
-------
maintenance work. The terms "building" or "work" include without
limitation, buildings, structures, and improvements of all types,
such as bridges, dams, plants, highways, parkways, streets, levees,
canals, dredging, shoring, drilling, blasting, excavating,
clearing, and landscaping.11
The distinction between construction and service is important
because some contractors perceive that it is in the contractor's
interest to have the contract characterized as a service contract
rather than as a construction contract, in that, contractors
believe that if a contract is designated as a service contract, a
performance bond will not be required. This is somewhat of a
misconception on the part of a contractor since the contracting
officer may require a surety bond on a service contract if a bond
is necessary to protect the government's interests.5 The interest
of contractors in having the work designated as service type work,
rather than construction, is illustrated by a bid protest dealing
with this very subject.12 The facts of the protest are as follows.
The Kansas City District (KCD), solicited proposals under a
competitively negotiated request for proposal (RFP) for the
construction of a transportable incineration system for explosives
contaminated soils at two sites ("Site I and Site II"). Basically,
the work was broken down into three phases. Phase I work consisted
of regulatory requirements and preparatory work efforts for
construction activities. At both sites, Phase II work consisted of
excavation, transportation, handling, incineration, and disposal of
943
-------
approximately 18,000 cubic yards of contaminated soil (TNT & DNT),
as well as treatment of approximately 400,000 gallons of explosives
contaminated water (TNT & DNT). At Site I, Phase III work
consisted of separating, transportation, handling, incineration,
and disposal of approximately 36,356 cubic yards of stockpiled
explosives contaminated soil (TNT & DNT) and debris. Phase III
work was a Government option.
The RFP's SPECIAL CLAUSE (SC)-44. LABOR-ADDITIONAL
REQUIREMENTS, classified the incineration portion of the work as
service, all other work required within 5 feet outside building
lines as building construction, and all other construction not
defined in the RFP as heavy construction. SOLICITATION PROVISIONS
(SP) 37. PERFORMANCE AND PAYMENT BONDS, required submission, for
Phases II and III, a performance bond equal to 100% of the contract
price and payment bond equal to 50% of a contract price of
$1,000,000 or less, 40% if in excess of $1,000,000 but no more than
$5,000,000 and in the amount of $2,500,000 if the contract price
exceeded $5,000,000.
An amendment clarified the classification of work, per phase,
at each work site. It contained revised SC44 which classified all
of Phase II at work Site I as either building or heavy
construction. Phase III work at Site II was classified service for
the incinerator operation, moving of existing stockpiled material,
and disposition of materials. All other work at Site II was
classified either building or heavy construction.
The protest consisted of three separate claims each of which
944
-------
will be addressed individually. The first claim was that KCD's
construction classification was improper. The second claim was
that the penal sums of the performance bond were unnecessarily high
and would unjustifiably restrict competition to only the highest
capitalized firms thereby precluding the protestor from competing
for this project. The third claim was that no performance bonding
should be required for any portion of the work classified as
service. The protestor sought a reclassification of the
construction work at both sites, a reduction of the penal sum of
any required performance bond to less than 100% of contract price,
and elimination of any required performance bond for service
classified work.
In support of its first claim the protestor argued that there
was no construction or excavation of any sort at Site II except for
some insignificant construction to improve roads and construction
of an administrative area. The protestor further argued that the
excavation work at Site I was part of the demolition of the
settlement lagoons and clearly was not Miller Act construction.
Additionally, the protestor asserted that the erection and
operation of the contractor's equipment was not construction work.
The FAR defined construction as "...construction, alteration,
or repair (including dredging, excavation and painting) of
buildings, structures, and other real property."11 The Office of
Management and Budget's Standard Industrial Classification (SIC)
Manual, Part I, Division C, classified as heavy construction, under
SIC Code No. 1629, (1) clearing of land; (2) earth moving, not
945
-------
connected with building construction; (3) industrial incineration
construction; (4) soil compaction services; and (5) kiln
construction. Road construction, except elevated, was classified
under SIC Code No. 1611 as highway and street construction.
Plumbing and electrical work, including telephone/telephone
equipment installation, were each classified as construction
(special trade contractors) under SIC Code Nos. 1711 and 1731
respectively. Structural steel and metal work were each classified
as construction under SIC Code No. 1791. Excavation work was
classified as construction under SIC Code No. 1794. The
installation of conveyor systems and the erection and dismantling
of machinery and other industrial equipment were each classified as
construction under SIC Code No. 1796. Lastly, the building of any
industrial building or warehouse was classified as general building
construction (non residential) under SIC Code No. 1541.
At Site I, the solicitation required the excavation and
removal of explosives contaminated soils and water from several
lagoons. The material would be transported to a holding area built
adjacent to an on-site industrial incineration facility. The
excavation process was to proceed simultaneously with the
incineration process. Following incineration, the excavated areas
were to be backfilled with the cleaned soil, compacted, graded and
seeded. In conjunction with the above, the solicitation
contemplated the installation of an erosion and sediment control
facility, construction of roads for transportation of excavated
materials, construction of an on-site scale for truck weighing,
946
-------
construction of an on-site water treatment facility, substantial
concrete foundation construction to support the incinerator, the
initial on-site erection and subsequent dismantling of the
incinerator requiring structural steel and metal work, the
installation of an on-site conveyor system to convey the soil from
the holding area to the incinerator, and installation of temporary
site utilities including telephones, electrical power, sanitation
waste containment and a water supply system. The entire facility
covered an area of approximately five (5) acres. Therefore, a
substantial amount of the work required at Site I was construction
work.
The work at Site II was similar to the work required at Site
I with the exception of excavation. The explosives contaminated
soil and debris were stockpiled in containers stored under tent-
like storage areas at the work site. In all other respects the
construction requirements of Site II were similar to Site I, which
were substantial.
A substantial portion of the work previously described and
contemplated by the solicitation fit the FAR definition of
"construction" in three specific ways at Site I and in two specific
ways at Site II.
SITE I
(1) the work site, which is real property, would undergo
significant and substantial alteration in as much as four of six
947
-------
existing lagoons which were to be removed in their entirety with
level, grass covered replacements left in their place. The
remaining two lagoons were to be backfilled in a configuration
allowing future utilization as lagoons;
(2) existing explosives contaminated soil and water would be
excavated;
(3) numerous structures would actually be constructed
involving work specifically classified as construction (heavy,
building, or other) by the SIC code manual.
SITE II
(1) A site, which is real property, adjacent to the
incineration work site would undergo significant and substantial
alteration since that area was to be filled with the cleaned soil
following incineration leaving a level, grass covered replacement
in place;
(2) numerous structures were to be constructed involving work
specifically classified as construction (heavy, building, or other)
by the SIC code manual.
The designated construction classification for work at both
sites conformed to the definition of the term "construction" in the
FAR. A service classification for the previously identified work
items was unwarranted.
The responsibility for determining whether a contract should
be considered one principally for construction rests primarily with
the contracting agency which must award, administer and enforce the
948
-------
contract.13 Consequently, the GAO will not disturb a good faith
determination by a contracting officer (CO) that a contract should
be for construction.14 The protestor presented no evidence to
suggest that the CO did not act in good faith in determining that
the subject contract for both sites was considered principally one
for construction. Neither did the protestor show any abuse of
discretion or violation of procurement regulations associated with
the subject solicitation.
The second claim that the performance bond's penal sum (100%)
was unnecessarily high, thereby unjustifiably restricting
competition, is closely linked to the first claim. The protestor
attempted to demonstrate the existence of a disparity between a
high penal sum requirement and a low construction requirement.
This attempt was flawed, however, given the previously shown
substantial construction actually involved at both sites. It
followed, therefore, that no disparity existed between the penal
sum of the bond and the solicitation's construction requirements.
The FAR states that "[t]he penal amount of performance shall
be 100 percent of the original contract price, unless the
contracting officer determines that a lesser amount would be
adequate for the protection of the Government."3 The protestor
argued for an interpretation of this language which overlooked any
requirement on the part of the CO to determine initially that the
penal sum should be set at 100% of the original contract price.
This argument was misplaced. The CO is given a mandate to set the
penal sum at 100% of contract price as the initial course of
949
-------
action. The language is not permissive, it is mandatory. An
exception is carved out providing for the exercise of discretion
where circumstances may warrant a lesser penal sum. As with any
other use of discretion, its exercise will be upheld if it is
devoid of abuse.15 The CO found no reason to determine that, as an
exception to the mandated penal sum of 100% of contract price, a
lesser penal sum would be adequate for the protection of the
Government. In an effort to persuade the CO to make such a
determination, however, the protestor stated that perhaps the CO
believes he did not have the authority to require a penal sum of an
amount less than 100% of contract price. Accordingly, a number of
examples were provided by the protestor to demonstrate the
existence of authority for the CO to make the determination that
the protestor desired. Authority, however, was not an issue in the
case. The CO is provided unequivocal authority by the FAR to make
determinations warranting an exception to the requirement for the
setting of the penal sum at 100% of contract price.3 The protestor
overlooked the fact that the thrust of this particular FAR
provision is to provide the Government with as much protection as
needed. This is borne out by the FAR which authorizes an increase
in required performance bond protection when a contract price is
increased.7 It is not the function of this FAR provision to
facilitate the acquisition of performance bonds by firms that have
exhausted their bonding capacity. Rather the function of the FAR
provision is to provide for the Government's need to have adequate
protection through the proper implementation of its regulations.
950
-------
The fact that a particular contractor may be unable to obtain
bonding does not make the requirement improper if it is otherwise
appropriate.16 In order to protect the United States and all
persons supplying labor and materials under contracts for
construction, the Miller Act1 requires that the contract awardee
furnish performance and payment bonds for all contracts which
exceed $25,000 in amount.17 Although a bond requirement may result
in a restriction on competition, it nevertheless can be a necessary
and proper means of securing to the government the fulfillment of
the contractor's obligation under the contract in appropriate
situations.18 The bonding requirement applied to the procurement
since a substantial amount of construction work would be required
at each work site. In reviewing a challenge to the imposition of
a bonding requirement, GAO looks to see if the requirement is
reasonable and imposed in good faith. The protestor bears the
burden of establishing unreasonableness or bad faith.19 In this
case, the protestor failed to demonstrate that the CD's compliance
with the FAR in setting the penal sum of the required performance
at 100% of the contract price was unreasonable or imposed in bad
faith.
The third claim was that no performance bonding should be
required for work classified as service. The FAR states that
generally, agencies shall not require performance and payment bonds
for other than construction contracts. However, performance and
payment bonds may be used4 as permitted in other sections of the
FAR.5 In related cases the GAO has found that although, as a
951
-------
general rule, in the case of non-construction contracts, agencies
do not generally require bonding, the use of bonding is permissible
where the bonds are needed to protect the government's interest,
regardless of whether the agency's rational comes within the; four
reasons given for requiring a performance bond.20 Bonds may also
be required where the continuous operation of critically needed
services is absolutely necessary.21
In this case, a performance bond was required for work
classified as "service" at Site I for four reasons. First, the
excavation of contaminated soils (construction) and incineration of
same (service) was interwoven into an integrated work effort. As
previously noted, the excavated soil was to be transported to a
holding area near the incinerator. In the event of a prolonged
work stoppage, the availability of a performance bond covering only
the excavation portion of the project would frustrate the principal
contract objective by leaving the Government without any protection
for the accomplishment of that objective, that is, the
decontamination of the excavated soil through incineration.
Second, the problem would be further compounded by the continued
excavation and stockpiling of contaminated soil, with no
incineration, by a surety's replacement contractor, since the
holding area for contaminated soil was not designed to contain
continuously stockpiled quantities of soil. Accordingly, a
spillover situation could have occurred involving a potential risk
that areas outside the holding area would become contaminated as
contaminated rain water runs off which could result in the leaching
952
-------
of contaminants into the ground. The continuous operation of the
incineration services was absolutely necessary in order to guard
against this potential risk. Third, the Government would not begin
receiving delivery of the end item, that is, the decontaminated
soil following incineration, until substantial progress payments
had been made for (1) the completion of Phase I work, and (2)
substantial, if not complete, construction of the entire facility.
Pursuant to the provisions of the FAR, a performance bond for
service work is warranted under these circumstances.20 Fourth,
Government funds were to be provided to the contractor for use in
the performance of the contract. The object of the contract,
incineration, could not commence until completion of construction
of the facility, which would be funded by the Government. Pursuant
to the provisions of the FAR, a performance bond for service work
was warranted under these circumstances.22
A performance bond was likewise justified for work classified
"service" at Site II pursuant to the provisions of the FAR since
Government funds were to be provided to the contractor for use in
performing the contract and substantial progress payments would be
made for Phase I work and construction of the facility before
commencement of delivery of the end item to the government.23
The Comptroller General, in denying the protest, held that the
performance bond requirement was unobjectionable where an agency
determines that a bond is necessary to assure the continuous
operation of the process of excavation and incineration of
contaminated soils, the interruption of which might result in
953
-------
contamination of the surrounding area, and substantial progress
payments would have been made prior to completion of performance.12
The Comptroller General held in a subsequent decision that a
performance bond requirement in a solicitation issued as part of a
cost comparison pursuant to Office of Management and Budget
Circular No. A-76, for facilities maintenance at an academic
institution housing over 1,000 personnel, was unobjectionable where
substantial government-furnished property will be provided to the
contractor for performance of the contract and the services to be
performed are critical to the continuous operation of the
facility.24
The purpose of this background has been to set out the rules
pertaining to performance bond requirements as those rules relate
to Superfund projects. The next section of this paper will discuss
the problem based on the experiences of the Kansas City District
and other Corps elements with this subject.
954
-------
DISCUSSION
The first indication of a problem with contractors' being
unable to obtain corporate surety bonds to guarantee performance of
a contract in connection with cleanup of a Superfund site came to
the attention of KCD in 1988, when bids were opened in response to
an Invitation for Bids ("IFB"). Three of the six bidders submitted
individual surety bid bonds. Since the fee charged by an
individual surety for a bond is greater than the fee charged by a
corporate surety, it seems apparent from an economic standpoint,
that the bidders were unable to obtain bonds from a corporate
surety.
The applicable FAR provision provided that bonds are
acceptable from individual or corporate sureties. Under the
regulation, an individual surety was defined as a person, as
distinguished from a business entity, who is liable for the entire
penal amount of the bond.25 It is the responsibility of the
Contracting Officer to determine whether the proposed individual
sureties are acceptable to the government.26 This was KCD's first
experience with investigating the individual surety and verifying
the assets and liabilities listed by the individual. In this case,
the proposed individual sureties were determined by the Contracting
Officer to be unacceptable. None of the three bidders protested
that determination.
Although the individual sureties were subsequently determined
to be unacceptable, the bids at the time of bid opening were not
955
-------
considered to be non-responsive. The Comptroller General has held
that a completed SF 24 is proper 'on its face1 when it has been
duly executed by two individual sureties (whose affidavits indicate
that, subject to further investigation, they both have net worths
at least equal to the penal amount of the bond), and the completed
SF 24 contains no obvious facial defects, such as the omission of
the penal amount, or the markup or alteration of the bond without
evidence of surety approval.27
Rather, the individual sureties were determined to be
unacceptable as a matter of responsibility, since the accuracy of
information concerning a sureties' financial condition is a natter
of responsibility.28 It is within the broad discretion of the
Contracting Officer to decide what specific finemcial
qualifications to consider in determining responsibility.29 When,
as a result of an investigation, there are serious doubts raised in
the mind of the Contracting Officer concerning the sureties'
financial resources and there is reason to question the business
integrity and credibility of the proposed individual sureties;, the
Comptroller General has held that given that the purpose of the
bonding requirement is to provide the Government with a financial
guarantee, we think it is clear that such information, which
diminishes the likelihood that this guarantee will be enforceable,
may be considered by the agency in determining the sureties'
acceptability.30 Where there is sufficient information to
legitimately cast doubt on the integrity of the sureties, the
Contracting Officer can justify a reasonable basis to question the
956
-------
accuracy of the financial representations and make a determination
of non-responsiblity.31
It is quite well established that, in making a determination
regarding responsibility, the Contracting Officer is vested with a
wide degree of discretion and business judgment and that decision
will not be altered absent a strong showing by the protester that
there was bad faith by the procuring agency or that there was no
reasonable basis for the determination.31 Contracting officials
are presumed to act in good faith and there must be convincing
proof that the agency had a malicious and specific intent to harm
the protester to establish otherwise.32
Often times an individual surety will offer to pledge assets
that are not solely owned by the individual. In cases where an
individual surety is one of several partners in a particular asset
and cannot legally pledge the asset, an agency may reasonably not
consider the value of that asset in determining the surety's net
worth.33
Although a Contracting Officer may contact an individual
surety to obtain additional information concerning listed assets
and liabilities, there is no legal requirement for the Contracting
Officer to make repeated contacts with the individual surety to
verify information, particularly where additional contacts will not
help to remove the doubt surrounding the veracity of the proposed
surety's statement of assets and liabilities.34
There are times when time is critical for award of a contract,
and in such cases an agency is not required to delay award
957
-------
indefinitely while a bidder attempts to cure a responsibility
problem.35
New rules regarding the acceptability of individual sureties
for all types of bonds, except position schedule bonds went into
effect on February 26, 1990. The new rules are much more
comprehensive than the former rule, and cover such subjects as
acceptability of individual sureties, security interests by an
individual surety, acceptability of assets, acceptance of real
property, substitution of assets, release of lien, and exclusion of
individual sureties.36 Since the new rules went into effect, the
number of individual surety bonds submitted to KCD in response to
IFBs and RFPs has decreased significantly.
It is now appropriate to discuss statistical data, and the
perceptions of the contracting industry and surety industry
concerning bonding. The information on these subjects was
developed by the U.S. Army Corps of Engineers, Water Resources
Support Center, Institute for Water Resources, in a study of
contracting problems related to surety bonding in the hazardous and
toxic waste clean-up program ("Corps Study").37 The Corps Study
included an analysis of 24 Superfund contracts awarded by the
Kansas City and Omaha Districts during the years 1987 thru 1989.
The study demonstrated that the ratio of award amount to
government estimate rose from .8 to 1.2. In addition, the ratio of
award amount to government estimate tended to increase with the
size of the project. The type of remedy that was utilized also
affected the award/estimate ratio. Award ratios of 1.3 were
958
-------
observed for the waste containment projects, on the average, as
opposed to .85 on the other extreme for alternative water supply
projects. The remainder of the projects were around the 1.0 area.
The conclusion drawn from this information is that there is a
tendency for large projects to run at a higher ratio of
award/estimate.38
An analysis of the contract data indicated that out of the 24
projects, four contracts involved situations where the apparent low
bidder was not awarded the contract due to an inability to secure
bonding. These four contracts totaled approximately $31 million.
$3.9 million in additional costs were incurred because of the
necessity to utilize the next low bidder. This was an average of
a 14% increase in costs for the four contracts. The ratio of high
bids to low bids has been found to drop from around 2 to 1 in 1987
to 1.3 to 1 in 1989. The range of bids also tends to decrease with
the size of the project. The high-low bid ratio also varies by the
type of project. The collection and disposal of waste products has
a large variation in the ratio of the bids while the waste
containment, innovative technology projects and alternative water
supply products have high-low bid ratios of around 1.2.39
To determine if the bonding issues had contributed to any
reduction in the competition for Superfund projects, the bids for
the 24 projects were examined. The number of bids decreased from
6.2 on the average in early 1987 to 4.6 in late 1989. The number
of bids also tended to lessen somewhat as the size of the project
increased. The latter phenomena is also experienced on all large
959
-------
construction projects. The type of project also influences the
number of bids received. Waste containment projects received the
most bids, followed by alternative water supply and soil waste
water treatment projects. The least number of bids was received on
the innovative technology projects. These projects received an
average of only two bids.40
There is considerable variation in the distribution of
contracts among HTW contractors. In the Kansas City District,
about 400 firms are on the bidders' mailing list for all
construction, including HTW contracts. In 1987 through January
1990, 24 contractors competed in the HTW program, and 14 were
awarded contracts. Five contractors, individually or in
partnership, have received 78% of the HTW contract dollars. Five
of the 14 firms obtained approximately 58% of all the projects.
The firms receiving awards are, for the most part, large firms with
experience in waste handling in general.40
There have not been any Superfund projects that could not be
placed under contract due to the unavailability of bonding. The
study showed, however, that corporate surety participation is
confined to a few companies. Six surety firms bonded 83% of the
total Superfund project dollars, and 71% of the projects were
bonded by five surety firms.41
The perception of the problem in the contracting and surety
firm sector was also studied by the Corps. From the point of view
of the contracting industry, a major problem in the HTW program is
that many contractors competing for contracts are unable to obtain
960
-------
the required surety performance bonds for construction contracts.
Some contractors are unable to secure bonds due to the surety's
perception of liability risk at HTW projects and some contractors
have exhausted their bonding capacity. Contracting firms maintain
close contact with the surety industry and routinely seek
information relative to bond availability. They are aware of the
surety industry's stated reasons for not providing surety bonds.
However, contractors assert that corporate surety decisions on
providing bonding are not uniform. Consequently, bonding may be
provided in some instances based on the surety's relationship to
the contractor rather than on purely objective standards. Remedial
action contractor (RAC) associations point out that there are many
firms that are interested in participating in the HTW cleanup
program, however, only a few are consistently able to meet the
bonding requirements necessary to continually compete for
contracts. Some companies stated that they did not even
participate in bidding on HTW projects for reasons of liability and
the inability to obtain performance surety bonds in the HTW area.42
The RAC associations stated to the Corps Study group that the
number of contractors bidding on HTW treatment projects is fewer
than those bidding on non-hazardous and toxic waste projects, in
part due to the bonding problem. One contracting firm pointed out
that the HTW program is comparatively small in relation to the
entire engineering and construction industry activity in this
country. Many firms reported that they have elected not to
participate in the HTW cleanup program when they experienced
961
-------
difficulties in securing bonds or anticipated complications in that
area. Contractors perceive that the problems in contracting in the
HTW area to some extent are due to the Government's use of
contracting procedures developed for non-HTW construction and
service contracting. HTW work involves a perceived increase in the
possibility of liability in excess of traditional construction
projects. There is also a strong perception in the surety and
insurance industry that the odds of incurring liability given
recent asbestos litigation are much greater than before.
Contracting firms felt that the laws, regulations, standard
Government procurement forms and procedures on HTW contracting
efforts were not totally appropriate.43
The experience of the Omaha and Kansas City Corps Districts
disclosed that there was a small number of bids received on several
HTW projects. According to several HTW organizations interviewed,
including the Hazardous Waste Action Coalition, Environmental
Business Association, Associated General Contractors, National
Solid Waste Management Association and the Remedial Contractors
Institute, the key factor contributing to lower competition for
some HTW projects is the inability of many contractors to secure
bonding. Despite a proven history of competence in doing such
work, strong finances, assets and profitability and sound
leadership and experience in the firm, the Corps Study reports that
the resulting shortage of qualified firms that are able to
consistently arrange surety bonding may be reflected in higher
costs to the government. A restriction on competition, with only
962
-------
four or five final bidders in many cases, may have resulted in
higher contract bids than would otherwise be expected. Several
contractors stated that they do not have the extensive financial
equity necessary to satisfy corporate sureties and secure surety
bonds.44
The Corps Study group was told, by those surety bond firms
that were interviewed, that their concerns are summarized in a
document entitled "Hazardous Wastes and the Surety.45 The sureties
believe that design of any sort is not traditionally a surety bond
activity. Bonding companies perceive that the risk of bonding
design elements of HTW cleanup is even more substantial than what
is faced on normal construction projects. This stems from the view
that the actual knowledge and experience in the area is limited.
Designs may become obsolete very quickly as changes in the HTW
processes evolve and generally there is considerable difference of
opinion among technical experts on design adequacy. Performance
bonds are normally used in construction contracts. In such
instances, the design is fixed and technical interpretations are
more uniform. However, where design elements and construction are
combined in the same contract, bonding problems may arise due to
the increased risk to the surety associated with the unknowns on
HTW project designs.45
Surety firms have stated that the present unfavorable legal
environment, with widespread litigation and large awards, has made
insurance companies very cautious about insuring HTW projects.
Although vocal in their assertions that they not be treated as a
963
-------
substitute for insurance, they fear that by bonding such work they
may in the future be sought out based on a legal theory which would
treat them as if they were insurance. The cause for liability,
such as the appearance of a disease 20 or more years after exposure
to toxic substances, leads to a very uncertain situation for
sureties. According to the surety firms interviewed, toxic tort
litigation features are an important reason for their present
reluctance to participate in the HTW cleanup field. In the toxic
tort arena a very long time period (10 or 20 years) between
exposure and development of injury is typical. Unlike other
prototypical injury situations, toxic liability involves long time
periods between the alleged exposure and the discovery of
damages.46
There is a concern by surety firms that they will be targeted
by third party liability plaintiffs in the event other parties
whose actions may have caused the injury are judgment proof. The
lack of sufficient insurance or indemnification for the HTW
remedial action contractor leads some bond underwriters to be
concerned that the corporate surety based on its providing a surety
performance bond may be adjudicated to fill the insurance void so
that the third party's injury can be compensated. They worry that,
after insurance coverage has lapsed or expired, and perhaps after
decades have passed, the corporate surety firm which provided the
bond may be looked upon by the courts as the insurer of last resort
or a "deep pocket." This unknown risk has led some corporate
sureties to forego involvement in the HTW market. Surety bond
964
-------
producers that have made such a decision indicate that they would
be more likely to participate in the market if the applicability of
SARA indemnification to the surety was clarified. Moreover, that
the performance surety bond be clearly represented as being
intended by the Government solely as a guarantee of performance by
the contractor and not in any way as protection for contractor
caused injuries to third parties.47
"Indemnification" is an agreement whereby one party agrees to
reimburse a second party for losses (in this case liability losses)
suffered by the second party. A recent development in the area of
Indemnification of Superfund contractors may serve to alleviate
some of the concerns that sureties may have in providing
performance bonds for Superfund contracts. This development is in
the form of an amendment to the Superfund Amendments and
Reauthorization Act of 1986 ("SARA"). The amendment adds to the
definition of a response action contractor, any surety who after
October 16, 1990 and before January 1, 1993 provides a bid,
performance, or payment bond to a response action contractor, and
begins activities to meet its obligations under such bond.48 Also
contained in the new legislation is the following language:48
(g) Surety Bonds —
(1) If under the Miller Act, 40 U.S.C. sections
270a-270f, surety bonds are required for any direct
Federal procurement of any response action contract, they
shall be issued in accordance with 40 U.S.C. sections
270a-270d.
(2) If under applicable Federal law surety bonds are
required for any direct Federal procurement of any
response action contract, no right of action shall accrue
on the performance bond issued on such response action
965
-------
contract to or for the use of any person other than the;
obligee named in the bond.
(3) If under applicable Federal law surety bonds are
required for any direct Federal procurement of any
response action contract, unless otherwise provided for
by the procuring agency in the bond, in the event of a
default, the surety's liability on a performance bond
shall be only for the cost of completion of the contract
work in accordance with the plans and specifications
less the balance of funds remaining to be paid under the
contract, up to the penal sum of the bond. The surety
shall in no event be liable on bonds to indemnify or
compensate the obligee for loss or liability arising from
personal injury or property damage whether or not caused
by a breach of the bonded contract.
Although the newest version of the indemnification clause does not
provide any specific reference to the availability of
indemnification for sureties, the term "response action contractor"
is being read in some quarters to encompass sureties that .begin
activities to meet obligations under their bond guarantees.49
966
-------
CONCLUSION
It appears that, in some cases, contractors are having
difficulty obtaining surety bonds for Superfund projects. It
also appears that surety firms are not overly enthusiastic
about issuing bonds for this type of work. On the other hand,
sureties have been willing to issue, and contractors have been
able to obtain bonds for Superfund work, and as a result there
has not been a significant adverse impact on the Corps
Superfund contracting program. It may be that part of the
problem that some contractors have experienced in obtaining
surety bonds is due to their inability to meet the criteria
for financial capability and experience which surety firms
require for issuance of a surety bond. In other cases the
Contractor can meet the surety's financial and experience
requirements, but cannot obtain the necessary bonding because
the contractor has reached the limit of its bonding capacity
with the surety. With respect to the apprehension of the
surety industry that issuance of a bond may expose the surety
to a type of liability that is not intended, by the surety, to
be covered by the bond, the bonding problem discussed herein
may, at least in part, be reduced if the surety industry is
satisfied that the recent Superfund indemnification amendment
will provide the surety with a greater degree of protection
against potential liability under the bond.
967
-------
REFERENCES
1. 40 U.S.C. 270a-270f
2. FAR 28.102
3. FAR 28.102-2(a)(1)
4. FAR 28.103-l(a)
5. FAR 28.103-2
FAR 28.103-3
6. FAR 28.102-1(b)
7. FAR 28.102-2(a)(2)
8. FAR 53.301-25
9. 29 CFR 4.111(b)
10. 29 CFR 4.115 (b)(1)
11. FAR 22.401
12. International Technology Corporation, 90-1 CPD 544
13. 50 Comp. Gen. 807(1971)
14. Abbott-Power Corporation, 77-2 CPD 434
15. Space Services International Corp., 84-2 CPD 430
Wright's Auto Repair and Parts, Inc., 83-2 CPD 34
Triple "P" Services, Inc., 81-2 CPD 436
16. BPOA Industrial Painters, 88-2 CPD 281
17. FAR 28.102-1(3)
18. D.J. Findley, Inc., 86-1 CPD 121
19. IBI Security, Inc., 89-2 CPD 277
20. FAR 28.103-2(3)
Aspen Cleaning Corp., 89-1 CPD 289
PBSI Corp., 87-2 CPD 333
21. Diversified Contract Services, Inc., 89-1 CPD 180
Intermodel Mansgement Ltd., 89-1 CPD 394
22. FAR 28.103-2(a)
23. FAR 28.103-2(3)(1)(3)
24. J & J Maintenance, Inc., 90-2 CPD 35
25. FAR 28.001
26. FAR 28.202-2
27. O.V. Campbell & Sons Industries, Inc., 88-1 CPD 259
28. Northwest Piping, Inc. 89-1 CPD 333
Transcontinental Enterprises, Inc., 87-2 CPD 3
29. Labco Construction, Inc., 89-1 CPD 135
Dunbar and Sullivan Dredging Co., 88-2 CPD 301
30. Ware Window Company; Saleco-Ware Window Company, 89-1
CPD 122
31. Carson and Smith Constructors, Inc., 88-2 CPD 560
Gem Construction Co., Inc., 88-2 CPD 530
32. Ram II General Contractor, Inc., 89-1 CPD 532
33. Aceves Construction and Maintenance, Inc., 89-1 CPD 7
34. Hirt Company, 88-1 CPD 605
Construct Sun, Inc., 89-1 CPD 431
35. Eastern Maintenance and Services, Inc., 88-1 CPD 266
36. FAR 28.203
FAR 28.203-1 through 7
968
-------
37. Hazardous and Toxic Waste (HTW) Contracting Problems,
A Study of the Contracting Problems related to Surety
Bonding in the HTW Cleanup Program Prepared by
U.S. Army Corps of Engineers, Water Resources Support
Center, Institute for Water Resources, July 1990
38. Corps Study, page 18
39. Corps Study, pages 18, 19
40. Corps Study, page 19
41. Corps Study, page 29
42. Corps Study, pages 29, 31
43. Corps Study, page 31
44. Corps Study, page 32
45. Corps Study, page 33
46. Corps Study, page 34
47. Corps Study, pages 35, 36
48. P.L. 101-584, 104 STAT. 2872, November 15, 1990,
Section 119
49. The Bureau of National Affairs, Inc., Environment
Reporter, Volume 21, Number 27, November 2, 1990, page
1247
DISCLAIMER
The views expressed in this paper are solely the
personal views of the author, and should not be
construed as reflecting the views of any other
person, the U.S. Army Corps of Engineers, or any
other agency of the Federal Government.
969
-------
Remedial Design Schedule Management
C.F. Wall, P.E.
Ebasco Services Incorporated
2111 Wilson Boulevard, Suite 1000
Arlington, Virginia 22201
(703) 358-8911
Thomas Whalen, P.E.
U.S. Environmental Protection Agency
401 M Street, S.W.
Mail Code OS-220W
Washington, D.C. 20460
(202) 308-8345
970
-------
NOTICE:
The remedy-specific schedules are
generic in nature and have been
developed with the objective of
demonstrating management approaches
to reducing the overall remedial design
duration. They present reasonable
approximations of the interrelationships
of those activities required to successfully
complete a remedial design. The schedules
and LOE estimates are intended for
training purposes only and should not
be used to develop site-specific schedules.
The schedules and LOE estimates used by the
party contracting for design must reflect
their own experience with similar projects.
971
-------
1.0 INTRODUCTION
The purpose of this paper is to present the results of a study
conducted to quantify the Remedial Design (RD) phase of the
remediation of a Superfund site.
The purpose of the study is two-fold.
1) Produce remedy-specific generic schedules to be used as
tools to assist all parties involved in the development of
Remedial Design (RD) schedules;
2) Provide insight, via use of the RD generic schedules, to
schedule optimization and schedule maintenance; and
resource load the generic schedules to provide additional
guidance to those tasked with planning, producing, or
managing RDs.
1.1 Background
Successful management of a remedial design depends on the perfor-
mance of responsible and qualified architectural or engineering
(A/E) firms, the maintenance of schedules and budgets, and the
rapid resolution of problems. Techniques for establishing good
design management include requirements that a schedule be agreed
to between the contracting party and the designer, that the
schedule be reviewed and updated monthly, and that enforcement of
the schedule by the contracting party be maintained. Of course,
the schedule must be reasonable, must establish obtainable goals,
must contain sufficient detail to permit task control, and must be
based upon a complete scope of work.
There are many reasons for maintenance of a schedule. The schedule
is a tool used to discuss the design contract between the contract-
ing parties and is also the principal tool for exacting control of
contract progress. The schedule also is the basic documentary and
analytical tool for negotiation and settlement of requests for
equitable adjustments, claims and disputes as well as for contract
termination and closeout.
The contracting party has the exclusive responsibility of schedule
enforcement, of explicit approval or rejection of the schedule and
of imposing sanctions for non-compliance. The control of the
schedule is the exclusive responsibility of the designer who also
has responsibility for handling unforeseen conditions and interface
impacts. Schedule revisions may be requested by either party;
however, revisions to the schedule are approved by the contracting
party.
The remedy-specific RD schedules discussed herein are generic in
nature and have been developed with the objective of reducing the
overall remedial design duration. They present reasonable
approximations of the durations and interrelations for those
activities required to successfully complete a remedial design.
972
-------
The generic schedules should be used as a guide during development
of site-specific schedules.
It is hoped that the judicious use of generic design management
schedules will result in a successful project. The hallmark of
a successful design project includes a design resulting in a remedy
consistent with the Record of Decision (ROD), a design that is
completed on time, and a design that is completed within budget.
Of course, the ultimate goal of a design project is the initiation
and successful completion of Remedial Actions.
2.0 APPROACH
The approach used to develop the generic RD schedules consisted of
the following steps:
• Develop a single, generic RD schedule using a commercially
available, computer-based, scheduling software package and
the Standard RD Tasks as a starting point.
• Canvass the ROD summary documents to identify the universe
of technologies being considered for site remediation.
• Develop a series of remedy-specific, generic RD schedules
via brainstorming with a multidisciplinary team of
scientists and engineers with experience in engineering
design and construction, cost and scheduling, and remedial
technologies.
Resource load (Level of Effort only) the generic RD
schedules using the RD experience gained in the Superfund
program and the RD experience of senior engineers and
scientists.
2.1 Scheduling System
Scheduling is the detailed listing of activities that must be
performed to reach defined organizational objectives. In any
contracting arrangement, scheduling is necessary and may range from
a simple agreement to deliver a product on a specified date to an
intricate, multi-activity schedule requiring detailed integration
of activities and resources.
Scheduling serves several purposes. It provides the framework for
discussing any aspect of the contract. It is a principal tool for
monitoring progress and is a prime consideration in negotiating a
contract. And, as stated earlier, it is the benchmark against
which negotiation and settlement of contract adjustments, claims,
and disputes are conducted.
Several types of scheduling systems are available. These include
milestone charts and bar charts which depict the activities to be
scheduled verses time. These schedules do not provide information
about the interrelationships of tasks to be performed. To develop
973
-------
this interrelationship requires a network analysis approach such
as the Critical Path Method (CPM).
CPM is a network of project activities showing both estimates of
time necessary to complete each activity within the project and the
sequential relationship between activities that must be followed
to complete the project.
The overall duration of the project is controlled by the critical
path through the overall schedule. The critical path is the
sequence of activities requiring the longest period of time to
complete. This path is called the critical path because a delay
in the time required to complete this sequence results in a delay
for the completion of the entire project.
The computerized scheduling systems selected for this project are
OPEN PLAN and PRIMAVERA, both microcomputer-based, commercially
available software packages.
2.2 Standard Remedial Design Tasks
Design is a scheme in which means to an end are laid down in an
arrangement such that the elements create a work of art, a machine,
or other man-made structure. The design for a remedial action must
be consistent with the ROD, comply with Superfund program policies
and procedures, minimize change orders, and prevent construction
contractor claims.
Listed in Table 2-1 are 11 standard tasks that may be used in A/E
agreements for RD services. The tasks are intended to provide a
consistent method of reporting design work. While some variations
are anticipated because of the variety of design projects and
differences among the A/E firms providing services for remedial
design, the standard tasks should be used and reporting formats
should be designed to be consistent with this set of standard RD
tasks.
For Federal-lead design, tasks 5, 6, and 7 contain those activities
which may be considered properly chargeable to design as set forth
in the Federal Acquisition Regulations at 48 CFR 15.903(d)(1) (ii) .
The regulation limits the total cumulative contract price for
design services to 6% of the estimated construction costs.
Some specific comments applicable to all standard RD tasks include:
• All standard tasks need not be used for every A/E agree-
ment. Use only those relevant to a specific design.
Flexibility is provided for reporting work associated with
the design tasks. For some A/E agreements, all of the
design work effort may be reported within only two tasks
(preliminary and prefinal/final design). In other A/E
agreements, the size of the project may be such that the
use of two or more sets of design tasks may be appropriate.
974
-------
For example, if the project involves major construction
items such as buildings, and also the design of groundwater
collection systems, it may be appropriate to have one set
of design tasks for the construction work and one set for
the groundwater collection system. In addition, the site
may have one or more operable units, which would require
the use of several sets of design tasks.
Services anticipated during the preliminary design (30%
complete), intermediate design (60% complete), and
prefinal/final design (90% to 100% complete) will depend
upon the Work Plan for the A/E agreement.
• Depending on the magnitude of the A/E agreement, the number
of documents to be submitted at the designated completion
points of design may vary. Also, depending on the size,
complexity, and timing of the specific design effort, there
may not be a need for all three phases of design. Under
these circumstances, communication between the contracting
party and the A/E is required to assure acceptable work
products.
The Standard Remedial Design Tasks are briefly described below:
TASK 1 - PROJECT PLANNING
This task includes work efforts related to the initiation of
a design project after the A/E agreement is executed. The
Project Planning Task is complete when the Work Plan is
approved by the contracting party. For purposes of transfer-
ring necessary data, this task also includes coordination
between the firm that conducted the remedial investiga-
tion/feasibility study (RI/FS) and the lead design firm. In
addition, initial value engineering (VE) screening will be
performed on all projects to identify high-cost, non-industry
standard items and unusual design criteria.
TASK 2 - FIELD DATA ACQUISITION/SAMPLE ANALYSIS
This task consists of the effort required to obtain field
samples and information needed to support the design effort
that was not produced during the RI/FS. It also includes the
analysis and validation of analytic results. This task begins
when any element authorizing field work, as outlined in the
Work Plan, is approved and may end when data validation is
complete.
TASK 3 - TREATABILITY STUDIES
This task includes efforts related to conducting pilot and
bench scale treatability studies during the RD. Specification
and procurement of study contractors, sampling, analytical
testing, data acquisition and validation, and reporting efforts
associated with these tests are included.
975
-------
TASK 4 - DATA EVALUATION
This task includes efforts related to the organization and
evaluation of data that will be used later in the design
effort. It is anticipated that this effort will use both
existing data and data collected and verified in Tasks 2 arid 3.
The data evaluation task usually begins on the day when
validated data are received by the designer and ends when it is
decided that no additional data are required.
TASK 5 - PRELIMINARY DESIGN
This task begins with initial design and ends with the comple-
tion of approximately 30 percent of the total design. It
incorporates work related to the preparation of plans and
specifications, unit process and equipment selections, cost
estimating and client review.
TASK 6 - INTERMEDIATE DESIGN
This task begins at the completion of the preliminary design
phase and ends with the completion of approximately 60 percent
of the total design. Depending on the size, complexity, and
timing of the design effort, this task may be omitted at the
discretion of the contracting party.
TASK 7 - PREFINAL/FINAL DESIGN
The prefinal/final design phase commences at the completion of
the intermediate design effort and is finalized when the entire
design effort has been completed and approved. Prefinal/final
design documents are submitted in two parts.
The prefinal design documents will be at approximately 90 per-
cent completion of design and will incorporate all work efforts
related to the preparation of the plans and specifications,
schedule and cost estimates, and the final technical reviews.
The final design effort incorporates all work efforts related
to the preparation of 100 percent complete plans and specifica-
tions, including the resolution of all client comments.
TASK 8 - DESIGN SUPPORT ACTIVITIES
This task consists of design support effort which is conducted
during one or all of the three phases of the design. Specific
activities are included on the Activity Listing (Table 2-2).
These activities include, Design Analysis, which includes the
analytical work (calculations and analyses) required to support
the preparation of plans and specifications during design.
Also included are the initial and final technical reviews
(constructibility, biddability, etc.).
976
-------
TASK 9 - VALUE ENGINEERING (VE) DURING DESIGN
If the initial VE screening conducted during the project
planning task identifies a potential cost savings, a VE study
will be initiated under this task. Value engineering is a
specialized cost control technique which uses a systematic and
creative approach to identify and to focus on unnecessarily
high cost in a project in order to arrive at a cost saving
without sacrificing the reliability or efficiency of the
project. This task also includes the cost of design revisions
resulting from the VE study. The VE study is performed during
the preliminary design phase.
TASK 10 - COMMUNITY RELATIONS
This task incorporates all work efforts related to the prepara-
tion and implementation of the community relations plan during
the design phase of the project. This task begins at the onset
of project planning and may continue up to the completion of
the design.
The present draft of the Standard RD Tasks includes TASK 11 -
PROJECT COMPLETION AND CLOSEOUT. This task is not considered on
the schedules included in this report. Activities within this
task, although necessary, do not contribute to the initiation of
the Remedial Action activities.
Several post-RD activities are included within the purview of this
study. These activities are listed in the Activity Listing
(Table 2-2). They are necessary to provide an estimate of the
start of onsite Remedial Action activities and reflect the often
accepted statement that design is not complete until construction
is complete.
2.3 Technology Categories
ROD summaries were perused to identify the variety of alternative
technologies being considered to remediate NPL sites. Broad
categories identified include:
Onsite treatment of soils/sludges and surface water
• Onsite containment
• Offsite treatment and/or disposal
Pump & treat groundwater
• In-situ treatment of groundwater
• Extending or upgrading existing water supplies or providing
alternate water supplies
977
-------
Excavation and/or demolition
• Onsite storage/disposal
No action
Brainstorming sessions were held to determine categories of
remediation which could be developed into remedy-specific generic
RD schedules. The resulting categories consisted of:
• Pumping, treatment, and discharge of ground and surface
water and leachate
Civil Engineering activities
Onsite thermal destruction
Treatment of soils and sludges
These categories were further developed, considering simple and
complex cases, into the final generic RD schedules.
3.0 RESULTS
3.1 Schedule Components
The initial step in developing a non-remedy specific, generic RD
schedule was to produce a comprehensive list of activities that
represent the possible sub-elements of each of the RD Standard
Tasks as well as some post-RD tasks which are required to initiate
the RA (Remedial Action). The tasks and the respective activities
that comprise them are presented in Table 2-2. The scope of the
specific activities that comprise each major task are generally
apparent from their titles.
3.2 Assumptions
The assumptions used in developing the schedules typically apply to
all the schedules regardless of the technology applied to remedy
the site. These assumptions include:
A cost-reimbursement, task order type contract, similar to
the EPA's REM and ARCS contracts, is used for the Remedial
Design.
A fixed price type contract for construction will be
awarded to the lowest, responsive, responsible bidder after
the solicitation of sealed bids.
• The Feasibility Study data are sufficient to specify the
Bench and Pilot testing.
978
-------
The contracting party design reviews are conducted parallel
with the continuing design process rather than in series.
The individual activity durations, for each of the remedy-
specific schedules, were selected based on a review of
ongoing RD projects and brainstorming discussions with
consultant and regulatory personnel knowledgeable of the
various cleanup technologies, design requirements, procure-
ment and planning needs.
The 60 percent design submittal is not required for the
"simple" designs.
Formal Value Engineering is not required for the "simple"
designs.
The pilot-scale equipment is available i.e., long-lead
procurement and/or fabrication is not required.
Laboratory analysis is conducted similar to EPA's DQO (Data
Quality Objectives) Level III i.e., full CLP (Contract
Laboratory Program) validation is not required.
Resource requirements do not restrain the duration of an
activity.
3.3 Remedy-Specific Schedules
The ROD review activity described in Section 2.0 resulted in the
selection of nine characteristic remedial design categories that
typified the universe of remedial actions being considered or
implemented at Superfund sites. A general definition of the nature
of each of the categories was developed along with appropriate
assumptions. The required work activities were selected, integrat-
ed with time-phased logic, and given appropriate durations by a
team of design engineers. The resulting activities, durations and
dependencies were then used to generate nine remedy-specific,
generic schedules and associated time-phased logic diagrams. Those
nine characteristic remedies for which generic schedules were
developed and their durations from RD start to 100 percent design
approval are shown in Table 3-1.
In the following discussions each of the nine typical or charac-
teristic remedies is described. It should be noted, as previously
discussed, that a site-specific design may have a combination of
these remedies as the overall project solution. It is assumed, in
that case, that each of the component remedies is worked in
parallel and that the more complex, time-consuming remedy controls
the overall project duration. The major assumptions that were made
in developing the schedules are presented and the results are
discussed.
979
-------
1. Civil Engineering - Simple
This design is for a remedial action that involves
principally civil engineering design. This simple category
will contain such remedies as fencing, groundwater
monitoring, and minor earthwork, demolition or removal
activities.
Scheduling Assumptions:
No treatability studies would be required.
Data gathering activities would include collection of
survey, geotechnical, and chemical analytical data.
The simplicity of the design activity and magnitude of
the design effort would allow elimination of the 60
percent intermediate design submittal.
Figure 3-1 presents this generic schedule in bar chart
format.
2. Civil Engineering - Complex
As with the simple case this design activity is principally
a civil engineering design activity. The complex case may
require more extensive date collection or design effort
such as a RCRA (Resource Conservation and Recovery Act)
cap, extensive or complicated excavation or demolition
activities, or the design of other engineered structures.
Scheduling Assumptions:
o The magnitude of data gathering activities is greater
so that the durations of sampling and analysis are
greater than the simple case.
A 60 percent design submittal is required.
Value Engineering is required.
Figure 3-2 presents this generic schedule in bar chart
format.
3. Pump & Treat - Simple
This design category is for groundwater withdrawal,
treatment and discharge or disposal and surface water/
leachate treatment. The technology categories include
physio-chemical and/or biological treatment of liquids.
Specific technologies may include: air stripping, carbon
adsorption, metals precipitation, ion exchange, multi-media
filtration, aerobic and anaerobic biodegradation, evapora-
tion, and distillation. In this simple case the tech-
980
-------
nologies would be proven for the contaminants of concern
and would be available in "off the shelf" package treatment
units. In addition, the aquifer characteristics would not
be complex, and standard pumping systems would be used.
Scheduling Assumptions:
Bench scale testing without pilot scale treatability
tests would be sufficient for design.
Extensive aquifer testing and collection of chemical
analytical data would not be required.
A 60 percent design submittal would not be required.
Figure 3-3 presents this generic schedule in bar chart
format.
Pump & Treat - Complex
This pump & treat design category is as described in the
Simple Case; however, the aquifer, contaminants, and the
pumping and treatment system design effort is a more
complex, time consuming effort. Innovative water treatment
technologies may be considered.
Scheduling Assumptions:
The complexity of the aquifer system requires extensive
aquifer testing.
The contaminants present and processes selected require
pilot scale testing in addition to bench scale.
• The complexity of the design effort dictates a 60
percent design submittal.
Figure 3-4 presents this generic schedule in bar chart
format.
Qnsite Thermal Destruction
This design category includes, onsite: incineration,
pyrolysis or in-situ vitrification.
Scheduling Assumptions:
Performance type specifications are produced in the
design of the thermal destruction unit.
Detailed design of auxiliary systems (water supply,
electricity, fuel, material handling) is required.
981
-------
• Bench scale treatability and a pilot scale test burn
are required. It is assumed that pilot test burns are
conducted at an existing facility.
Figure 3-5 presents this generic schedule in bar chart
format.
6. Soils/Sludge Treatment - Simple
This design category includes the physical, chemical or
biological treatment or volatilization of soils and
sludges. All non-thermal destruction of solids would be
treated under this category. In this simple case the
process chosen would be a well proven technology for the
contaminants of concern and for the existing site condi-
tions.
Scheduling Assumptions:
• Bench and pilot scale testing programs would be
required, however, they would be of a relatively short
duration.
• The simplicity of design activity and magnitude of the
design effort would allow elimination of the 60%
intermediate design submittal.
• Formal Value Engineering is not required.
Figure 3-6 presents this generic schedule in bar chart
format.
7. Soils/Sludge Treatment - Complex
This design category is similar to the simple case;
however, as a result of complex contaminants and site
conditions, innovative processes requiring extensive
testing and development are required.
Scheduling Assumptions:
• The selected process requires extensive bench and pilot
scale testing.
The design magnitude and complexity dictates the
submittal of a 60 percent design package.
Figure 3-7 presents this generic schedule in bar chart
format.
982
-------
8. Pump S Treat - Simple (Expedited) and
9. Civil Engineering - Simple (Expedited)
Both of these categories were developed for those sites
where the remedial design is simple and straight-forward
and where additional data collection is not required.
Sites where the scope is limited to minor removal actions
or administrative controls would fall into these catego-
ries:
Scheduling Assumptions:
A single contractor performs the RI/FS, the RD, and
construction management.
Additional data collection to support the RD is not
required.
Following not required:
Treatability Studies
Value Engineering
60 percent design submittal
Client agreement at pre-design meeting to initiate some
aspects of design before approval of the Work Plan.
Figures 3-8 and 3-9 present these generic schedules in bar
chart format.
4. 0 APPLICATIONS
Several work applications would find the generic RD schedules a
beneficial tool. The generic schedules can be used for multiple-
site planning by an entity with responsibility for all sites within
a geographic region (for example an EPA Regional Office or the
USAGE Design Office).
This can be accomplished as follows: for a given suite of sites,
peruse the ROD to select the appropriate remedy-specific, generic
RD schedule. Where more than a single remedial alternative is
contemplated for site remediation, the remedy (and generic
schedule) with the longest design duration should be selected.
Then, using the generic schedules selected for each site, a master
milestone schedule can be produced. Figure 4-1 is an example of
a typical milestone schedule. This schedule is anchored by semi-
fixed target dates for a specified event (for example, start of
construction) for each site. These target dates are typically set
by management during overall program planning. The master
milestone schedules are used to track progress, identify problem
areas, and allocate resources to meet management goals.
983
-------
A second use for the remedy-specific, general RD schedules is to
develop an initial site-specific schedule by those parties
responsible for performing the RD or those responsible for managing
or overseeing the RD.
The initial step in the application of the generic RD schedules to
actual sites is a thorough review of available site information,
including RODs, RI/FS reports, and other available data.
This review of site data will permit the selection of the remedy-
specific generic schedule appropriate for the site. Where two or
more remedy categories are applicable to the site (for example,
groundwater treatment and onsite incineration), a base generic
schedule is selected. All other factors being equal, the generic
schedule of longest overall duration is selected as the base
generic schedule, while the remaining applicable schedules
("subsidiary" schedules) are pooled for incorporation into the base
generic schedule. The durations of work activities within each of
the "subsidiary" schedules are compared to the durations of
equivalent activities within the base generic schedule, and the
longest duration for each activity "plugged" into the generic
schedule for the site. The manipulation of work activities to
achieve a site-specific generic schedule is a straightforward task
using the scheduling software currently available.
This series of steps results in a conservative, first-cut RD
schedule. This generic schedule can be used as a basis to
construct a detailed site-specific RD schedule which must satisfy
the interrelationships (predecessor-successor) for each activity
which, for the more complex sites, involve many engineering
disciplines and individual design efforts.
For the party responsible for overseeing the design, the generic
schedule can be used as a basis for negotiations with the A/E
performing the design. Also, the schedule can be used as a check-
off to ensure that all critical elements to the RD are present.
5.0 SCHEDULE OPTIMIZATION
This section presents a discussion of the optimization assumptions
used to develop the generic RD schedules and a discussion of
schedule maintenance.
Many of the assumptions discussed herein can be centered around two
important areas: communication and the sensitivity of activity
durations. The identification of all interested parties for a site
and early frequent communication among them should identify
potential schedule delays and allow sufficient time to resolve
problems.
984
-------
Task activity durations are sensitive to individual site charac-
teristics, the design complexity, and the needs of the owner.
Therefore, it is important that site-specific RD schedules be
developed that reflect these sensitivities and emphasize the need
for communication throughout the progress of the design.
5.1 Generic Schedule optimization Assumptions and Discussion
Assumption; RD Work Assignment issued coincident with close of the
PRP moratorium period.
Discussion; As a fully optimized approach, the generic schedules
show the RD Work Assignment issued, and work commencing, one day
after the ROD is signed. This schedule can be modified to allow
for a moratorium period for negotiations with Potentially Respon-
sible Parties.
To facilitate this approach requires frequent communication,
starting early in the planning process, among the owner, the state
agency(s), and others with responsibility for producing, reviewing,
or managing work. It is critical that all parties recognize the
ultimate goal of the Superfund program (cleaning up contaminated
sites, identified as such on the NPL) and that planning must go
beyond an intermediate goal (e.g., issuing a ROD). Once the effort
is made to identify all concerned parties, data can be distributed
and discussions held to bring about early agreement on those issues
which will comprise the State Cooperative Agreements, Superfund
State Contracts, and Interagency Agreements.
The administrative workload that occurs during the RI/FS - RD
turnover can be reduce by using a single A/E contractor to perform
both assignments, as proposed under ARCS. This approach should
lead to a more efficient design and will instill confidence that
the cleanup is proceeding with the best mix of technical quality
and cost/schedule efficiencies.
Long delays (up to several years) which have occasionally occurred
between ROD signing and initiation of RD activities can present
additional problems to rapid execution of designs. These delays
may cause problems with the reliability of time sensitive data
(e.g., contaminant plume location), the availability of personnel
with knowledge of past site activities, and the potential lack of
consideration in the FS of recently developed technologies. The
resolution of these problems often requires extended durations for
several RD activities.
Assumption; Tasks required to procure treatability testing and
field investigation services, if needed for the RD, are authorized
in the Work Assignment as interim tasks.
Discussion; Considerable schedule savings can be realized by
initiating the time-consuming procurement process prior to formal
Work Plan approval. The key to pursuing this mode of operation is
to develop a thorough understanding of the requirements of the Work
985
-------
Assignment during early activities and to obtain concurrence
regarding the selection approach. Specifically, these exceptions
can be realized by a thorough data review, by definition of clear,
concise and well thought out design criteria, by a pre-design
meeting which presents sufficient detail to permit concurrence with
the design approach and the type and quality of data needed to
initiate the design, and by early input from the Project Delivery
Analysis which may affect the design (e.g., performance vs.
prescriptive specification).
Assumption; Review/Approval durations are optimum.
Discussion; The durations presented in the generic schedules
require that the reviewers have a thorough knowledge of the ROD and
owner requirements and that the reviewers have been kept informed
throughout the progress of the project. A necessary corollary to
this assumption is that the reviewers give prompt attention to the
review package.
The review and approval activities within the generic RD schedules
are the responsibility of the owner. These activities may be
conducted in parallel with other ongoing activities or in series,
whereby subsequent activities do not start until the review is
completed, comments are resolved and approval to proceed is
provided.
Owner-responsible activities used on the generic RD schedules are
classified as follows:
Serial Reviews/Approvals and Parallel Reviews
Review Draft RD Work Plan
Approve Final RD Work Plan30% Design Review
Review Community Relations Plan60% Design Review
Approve Community Relations Plan90% Design Review
Review Draft FSAPTechnical Reviews
Approval Final FSAP
Approve Investigation Contracts
Approve Treatability Contracts
Approve 100% Design
It is desirable that the owner provide the coordination role during
the review process. The owner should collect the review comments
and provide the design with a concise comment package. This method
will also allow the owner to screen and respond to comments which
need not be passed on to the designer.
The actual durations for review activities for any particular site
are a function of the complexity of the site characteristics and
of the design, and also of the administrative requirements of the
owner and the reviewers. The specific review/approval activities
which are the owners responsibility should be clearly and separate-
ly identified on the project schedule. This will reinforce the
responsibilities of all parties to the contract and provide early
986
-------
knowledge of the consequence of allowing these activities to move
onto the Critical Path.
Assumption: All task durations are considered reasonable. The
design is correct.
Discussion; The durations of the activities which comprise the
various generic schedules were selected based on experience.
However, no time was included to compensate for technical or
administrative problems which could arise. This lack of schedule
contingency is considered appropriate for the presentation of
generic schedules in that one does not plan for technical errors
or the lack of proper administrative management. Some potential
delays can be anticipated, such as a decrease in production due to
inclement weather; however, these are site-specific problems and
as such should be considered when developing individual, site-
specific design schedules.
An additional item which can affect the overall RD schedule is the
use of a design which is extremely complex due either to an
innovative approach to the remedy or to a multiplicity of operable
units. This also should be considered during the preparation of
the site-specific schedule by the designer.
For example, the use of innovative technology (as mandated by SARA,
the Superfund Amendment and Reauthorization Act) may obviate a
quick RD schedule by requiring time consuming treatability studies,
including the potential need for long-lead procurement of equip-
ment. Also, a technology that is new, without a record of
performance, may cause a conservative reaction among the interested
parties, leading to lengthened activity durations (e.g., increased
review time).
It is important that early communication be established with all
interested parties concerning the overall site schedule and the
planned design approach. This communication should prevent
potential schedule delays caused by a "change in direction" during
the design sequence or by the need for resolution of comments
originated by uninformed reviewers.
Assumptions: Reports are not on the Critical Path.
Discussion; The results of bench and pilot scale testing, field
data evaluation, and design analysis are communicated to those
requiring the information in a timely manner. There are not
planned "comment periods" during which work is suspended while a
report is undergoing review. Reports provide formal documentation
of data and decisions but are not on the critical path.
Assumption; Work Assignment for A/E support during construction
is in place prior to approval of 100 percent design.
Discussion: The concept of working to a "total" remediation
schedule for a single site (RI through completion of RA) in an
987
-------
efficient manner necessitates the early identification of an A/E
firm to provide engineering support to the owner during construc-
tion. This will assure that the RA will not be delayed due to lack
of engineering support and also permits timely support to the owner
for several pre-award activities including conducting the pre-bid
conference and evaluation of the RA bidders.
It is incumbent upon the owner to initiate the activities necessary
to procure an A/E firm to support the RA task. It is probably most
efficient to use the services of the A/E firm performing the design
for this effort.
Assumption; Additional time required to incorporate significant
design changes as a result of Value Engineering or other technical
reviews (Biddability, Constructibility, etc.) is not represented
in the generic schedules.
Discussion; The resolution of technical comments should take place
within the design cycle of performance/review for the identified
submittal stages (30%-60%-90%/100%) with the stipulation that all
comments be resolve prior to submittal of the final design package.
The impact to the RD schedule of a Value Engineering change is
implicit to the VE decision process.
Assumption; Sufficient planning is performed to avoid schedule
delays due to lack of adequate funding.
Discussion; Funding estimates are prepared for each phase of a
remediation project. As a project matures, additional site data
is collected and the work is more clearly focused on the ultimate
remedy. This evolutionary process requires that the budget
estimate for a site be modified to reflect the evolution of the
project. The milestones where the need for revised funding
estimates may occur include submittal of Work Plans for the RI/FS
or RD, construction cost estimates that are prepared for the
various design stages and the construction bid. A significant
increase in required funding and the necessary authorization
process at any of these milestones has the potential to delay the
project.
Although the reallocation of funds to meet the needs of a single
site can be difficult, the suggestion presented in this document
of early and frequent communication among the interested parties
can reduce as much as possible these potential delays.
Other optimizing assumptions will probably come to light as more
experience is gained by the industry. Several, assumptions which
were not included in developing the generic schedules are discussed
below.
Early RD Start; Although there may be programmatic procedural
problems with this assumption it is included to illustrate a major
overall schedule reduction potential. In this scenario the RD Work
Assignment would be issued at approximately the same time as the
988
-------
finalization of the Feasibility Study (FS), allowing the initiation
of planning and data gathering activities prior to the ROD
signature. The risk of this course of action is a change in the
selected remedial alternative resulting from the ROD process. This
approach is most applicable for those sites where the selected
remedy is unequivocal. Issuing the RD Work Assignment at FS
finalization could accelerate the RA start by months.
Performance Specification; Experience is showing that some RD
alternatives (such as onsite thermal destruction) are best handled
using a performance type specification that allows the use of
alternative processes as long as the performance criteria are met.
There are two schedule reductions that can result from a perfor-
mance-based approach.
Bench/Pilot Scale Testing - In a performance-based approach
which allows alternative technologies, extensive testing
may have limited additional value to the vendor. The
vendor in many cases will have sufficient experience and
prior operating data on the process to be able to cost the
system as long as good waste characterization data are
available, thereby eliminating the need for a testing
program.
Eliminate 60 Percent Design Submittal - A performance-based
procurement will have fewer design drawings and specifica-
tions, thereby making the definition of a logical 60
percent design break difficult to conceive.
Contracting Strategies
• When time is of the essence or when innovative, state of
the art designs are to be implemented, a site-specific,
fixed price type contract should not be used to conduct the
remedial design because of the time required to complete
this type of procurement and the inflexibility of such a
contract.
If the project delivery analysis reveals that the cir-
cumstances are not appropriate to the solicitation of
sealed bids for construction, then competitive proposals
should be requested and a fixed-price or cost-reimbursement
type contract, or combination thereof, should be emplaced.
Appropriate circumstances may include, for example, the
construction and operation of a remediation technique for
which there is no past experience.
6.0 SCHEDULE MAINTENANCE
Several approaches used to optimize Remedial Design schedules were
discussed in Section 5.0. It must be emphasized, however, that
preparing a schedule will not "make it happen." In order to be
successful the schedule must first be "doable." In essence, the
work breakdown structure must be in sufficient detail to identify
989
-------
critical work elements, the logic or precedence of activities must
be correct, and the duration of each of the individual activities
must be sufficient to accomplish the work with the resources
available. Once an optimized schedule is agreed to by the RD
contractor and the contracting party, maintenance of the schedule
is dependent on a number of key elements that become especially
critical because of the optimizing assumptions.
Some of the more important areas that are discussed here include:
• Communications
Project Delivery Analysis
• Basis of Design Report
• Reviews
• RI/FS-RD-RA Transition Planning
Cost Estimating/Funding
6.1 Communications
In the optimized RD numerous concurrent activities will be
occurring with parallel and concurrent review steps. In this mode
of operation, with fewer defined "stop and check" points, the
greatest danger to schedule maintenance can be having to redo work
that has been completed without the concurrence or understanding
of the owner. Regular project communications involving the
appropriate decision makers or their representatives are necessary
to eliminate false starts or misdirected activities. The communi-
cations or reporting plan must however, strike a balance between
keeping decision makers informed and imposing a paperwork burden.
6.2 Project Delivery Analysis
The Project Delivery Analysis (PDA) is the development of the
contracting strategy for the completion of the remedial action and
includes:
1) The number and scope of RA Contracts
2) Contract Types
Lump Sum
• Unit Price
Cost Reimbursable
3) Contracting Procedures
IFB (Invitation for Bids)
RFP (Request for Proposal)
Pre-qualification
990
-------
4) Design Approach
Detailed Design (prescriptive specification)
Performance Specification
The decision made in the Project Delivery Analysis must be well
conceived and involve the decision makers. PDA must be done as
part of RD scoping as the decisions made will dictate the scope and
complexity of design. It is possible that the delivery approach
can not be finalized without additional data and this must be
reflected in the schedule.
6.3 Basis of Design Report
The Standard Tasks for Remedial Design make reference to a Basis
of Design Report (BODR). The optimized schedule with parallel
review tasks will only succeed if there are no surprises. The
objective of the BODR, therefore is to document the criteria for
design and clearly establish the design decisions upon which
subsequent analyses should be based. If the basis of design is
firmly established, subsequent design reviews should not reveal the
need for significant changes in the design approach with resultant
schedule delays.
6.4 Design Reviews
Optimized schedules are predicated on parallel design reviews while
subsequent design steps continue. The inherent risk associated
with such an approach is the potential for redesign resulting from
review. This risk can be minimized in two ways. First, as was
discussed previously, if good communications have been established
the review should present no surprises to the reviewing authority
and there should be no resultant schedule delays. Second, the
review procedure can be enhanced both in its timeliness and its
thoroughness by using a panel approach similar to that used for
Value Engineering. A detailed format needs to be developed for
such an approach but in essence it would involve assembling a panel
to accomplish a team review, including a design review presentation
and resolution of comments, prior to adjourning the panel.
6.5 RI/FS-RD-RA Transition
This RD schedule optimization exercise has emphasized the impor-
tance of not allowing the transition between remedial stages result
in work stoppages. Major schedule optimization can be accomplished
through total project scheduling and overlap of the remedial
stages. Many RD standard tasks can start prior to ROD signing and
RA planning can start prior to 100% design approval.
6.6 Cost Estimating
One major impediment to schedule maintenance that frequently
results in schedule slippage is the identification of costs in
excess of program budgets, requiring reallocation of funds. This
991
-------
can occur both in planning and construction and will be even more
important when working to an optimized schedule. This potential
schedule maintenance problem underscores the importance of cost
control and cost estimate updating.
7.0 OPTIMIZATION IMPACT - EXAMPLE SCHEDULES
As part of the RD Schedule Management assignment seven generic
schedules were developed for a range of remedial alternatives.
Those schedules contained several optimizing assumptions that have
been described in Section 5.0. Two expedited schedules were also
prepared for simple design assignments, with little or no addition-
al data gathering requirements.
In order to illustrate additional optimization or schedule
reduction alternatives, the onsite thermal destruction generic
schedule has been further optimized with assumptions that may have
application at a specific site. To further illustrate the impact
of schedule optimization we have taken that same generic schedule
and eliminated virtually all optimizing assumptions in order to
demonstrate the overall duration of the non-optimized schedule.
7.1 Fully Optimized Onsite Thermal Destruction
A fully optimized Onsite Thermal Destruction generic schedule has
been developed and is presented as Figure 7-1. In addition to the
optimizing assumptions that were built into the original schedule,
the following four schedule reduction alternatives have been
incorporated into this example schedule:
Early RD Start; Although there may be programmatic procedural
problems and limitations due to current policy guidance with this
assumption, it was incorporated to illustrate a major overall
schedule reduction potential. In this scenario the RD Work
Assignment would be issued at approximately the same time as the
finalization of the FS, allowing the initiation of planning and
data gathering activities prior to ROD signature. The risk of this
course of action is a change in the selected remedial alternative
resulting from the ROD process. This approach is most applicable
for those sites where the selected remedy is unequivocal. Issuing
the RD Work Assignment at FS finalization could accelerate the RA
start by as much as five months. In the fully optimized schedule
appended here it reduces that time by 15 weeks.
Performance Specification; Experience is showing that the onsite
thermal destruction alternative is likely to be a performance type
specification that would allow alternative processes as long as the
performance criteria were met. There are two schedule reductions
that can result from a performance-based approach.
Bench/Pilot Scale Testing - In a performance-based approach
which allows alternative technologies, extensive testing
may have limited additional value to the vendor. The
vendor in many cases will have sufficient experience and
992
-------
prior operating data on the process to be able to cost the
system with good waste characterization data. Therefore
the pilot scale activities have been eliminated from the
program. This schedule reduction reduces the bench/pilot
scale program by eight weeks and takes this program off the
critical path.
Eliminate 60 percent Design Submittal - A performance-based
procurement will have fewer design drawings and specifica-
tions, thereby making the definition of a logical 60
percent design break difficult to conceive. Therefore, the
60 percent design submittal is eliminated. In so doing,
the overall design time has been reduced without entirely
eliminating the time that was included in the 60 percent
design step. This schedule reduction reduces the overall
schedule by four weeks.
Mobile Laboratory Data; The generic schedule includes laboratory
turnaround time as well as Level III validation time. The fully
optimized schedule has been reduced by two weeks by placing a
mobile laboratory onsite and utilizing Level II data for design
purposes.
Summary; The three schedule reduction alternatives described above
have reduced the overall schedule (RD start to RA start) by
12 weeks. Assuming that RD can start prior to ROD signing, the ROD
to RA start could be reduced by an additional 15 weeks.
7.2 Non-Optimized Schedule
In order to illustrate the schedule impact of eliminating virtually
all schedule reduction/optimization approaches incorporated into
the generic schedule, a non-optimized schedule has been prepared
(Figure 7-2) with the following deviations from the generic
schedule:
No Interim Authorization; Only Project Planning and Community
Relations Planning is authorized in the RD Work Assignment. No
other work proceeds until these plans are approved. This extends
the schedule by 11 weeks.
Review Schedule;
Draft Work Plan Review has been extended to four weeks.
• 30% Design Review is a four-week Serial Review.
• 60% Design Review is a four-week Serial Review.
• 90% Design Review is a six-week Serial Review.
100% design approval has been extended to three weeks.
All of these review considerations extend the schedule a total of
17 weeks.
993
-------
Value Engineering; In the non-optimized schedule VE-driven
redesign activity is included, increasing the 60 percent design
activity by two weeks.
Summary; The non-optimized Onsite Thermal Destruction Schedule
takes 97 weeks from ROD signing to RA start as compared to 67 weeks
for the generic schedule and 40 weeks for the fully optimized
schedule.
8.0 RESOURCE LOADING OF THE GENERIC RD SCHEDULES
The contractors working on the Superfund Program have considerable
experience planning and conducting Remedial Designs for hazardous
waste sites. This experience was used to resource load the generic
schedules. Two approaches were used. Firstly, a data base was
assembled of RDs conducted under the USEPA's REM III Program.
These data were categorized by technology. Secondly, a group of
engineers, scientists, and managers were assembled to brainstorm
specific Level of Effort ranges for each activity in each of the
nine generic schedules. These personnel had experience in
technical areas pertinent to remedial design, including planning,
treatability studies, field data collection, basic engineering
design, technologies available to remedy hazardous waste sites, and
contracting mechanisms.
8.1 Assumptions
The resource-loading activity was accomplished within pre-defined
boundaries so that some reasonable quantification could be
achieved. The general assumptions are discussed in this section.
Assumptions specific to a particular schedule are discussed in
Section 8.2 (Results).
The generic RD schedules previously developed (see Table 3-1) were
not modified during the resource-loading exercise. Activity inter-
relationships and durations were fixed.
The typical RD assignment was a turnover (intra-company) from an
RI/FS work assignment. A cost-reimbursement, task order contract
is used for the remedial design.
All review comments will be consolidated by the lead agency
(contracting party) and transmitted to the RD contractor in a
single package and within the allotted schedule.
A range of job hours is selected for each scheduled activity. A
typical RD assignment is expected to fall within this range.
The resource-loading activity is limited to LOE i.e., job hour
estimates. No attempt was made to estimate other direct costs or
subcontractor costs. These costs however, are identified by
category in the summary tables. Also, the program management
activity LOE is not included in these estimates. This activity
includes cost/schedule control, progress reporting, problem
994
-------
solving, contractual modification justification, subcontractor
control, invoicing, and other general management functions required
to run task-order contracts. These costs can be estimated as a
percent of the total task LOE and may vary depending on factors
such as work complexity, the total number of active tasks in the
contract and the RD contractors corporate management structure.
8.2 Results
The results of this resource-loading activity are presented in the
following subsections.
8.2.1 Remedial Design Experience Matrix
Data were collected from within the USEPA REM III Program to
summarize current remedial design experience. Data sources
included monthly progress reports, individual RD work assignment
work plans, and interviews with site managers.
Many of the RDs were not conducted within the template of the
Standard RD Tasks; therefore, a subjective evaluation of each
project was completed to correlate the actual project task
structure with the Standard RD Tasks. This evaluation relied
heavily on the site manager interviews. Project RD LOE experience
was used as one source of data during the brainstorming session.
8.2.2 Resource Loading the RD Schedules
The nine remedy-specific generic Remedial Design schedules were
resource loaded using a brainstorming technique. Drawing from the
REM III Program, corporate, and personal experiences and the
activity durations within the generic schedules, the team assigned
a range of LOE to each activity in each of the nine schedules. The
first schedule addressed was Pump and Treat - Complex as the team
experience was greatest in this technology. The LOE ranges for
this schedule were used as a template to select appropriate levels
of effort for activities in the other eight schedules. Therefore,
modifications to an activity LOE among the several schedules was,
of necessity, supported by sound technical reasoning. This
approach also resulted in some activities having the same LOE range
for all generic schedules.
The following paragraphs discuss loading of each of the 9 generic
schedules.
Pump and Treat - Complex
Assumptions used to load the activities are presented for each
standard task in the following paragraphs.
Task 1 - Project Planning
Three technical experts (civil engineering, hydrogeology, and
chemical process engineering) are needed to support the work plan
995
-------
preparations. The contracting party will consolidate comments to
maximize efficiency of review and comment resolution efforts.
Task 2 - Field Data Acquisition/Data Analysis
Four technical specifications are required (drilling/well installa-
tion, laboratory analytical services, surveying, waste disposal).
The field data collection effort is 6 weeks in duration and
includes a 2-week pumping test. Twenty samples are analyzed and
DQO Level III validation is used.
Task 3 - Treatabilitv Studies
For contracting and evaluation purposes assume three separate
innovative technologies are potentially viable treatment options.
One contract modification is issued. One person is needed at the
site periodically to oversee the pilot test programs.
Task 4 - Data Evaluation and
Task 8 - Design Support Activities
One of the early deliverables from these tasks is the Basis of
Design Report. It is estimated that five criteria categories are
addressed in this report. They are: site elements (civil)
criteria, hydrogeologic criteria, process design criteria, health
and safety criteria, and environmental criteria.
It was assumed that six permits would be required including NPDES
(National Pollution Discharge Elimination System), air, wetlands,
erosion and sedimentation control, and local municipality. The RA
contractor will acquire the building and construction permits.
The final technical design reviews (constructibility, biddability,
operability, environmental, and claims prevention) are included
under this task.
The Operations and Maintenance Manual is, at this stage, a detailed
"specification" to guide the contractor. The manual is completed
by the RA contractor during start-up operations.
Tasks 5. 6. and 7 - Design
It was determined that ESSENTIALLY there should be no difference
in LOE between prescriptive and performance specifications. Most
site designs will require both using prescriptive specifications
for site-specific requirements such as earthwork, and using
performance specifications for many of the innovative technologies
which have limited performance histories.
Three design packages are delivered for review: preliminary,
intermediate, and pre-final/final.
996
-------
Task 9 - Value Engineering
The level of effort for VE during design is taken from the USEPA
guidance document for performance of VE during remedial design.
Task 10 - Community Relations
This task is essentially an extension of community relations (CR)
activities conducted during the pre-design (RI/FS) phase. LOE is
typically a function of schedule duration. Activities include
revision of an existing CR plan, one public meeting, and continued
CR support through the start of construction.
Task 11 - Project Completion and Closeout
Activities and associated LOE required for this task were assumed
to be included in "Program Management".
Summary
The total estimated LOE for the Pump and Treat - Complex generic
RD schedule is 8,350 to 11,149. With a schedule of 13 months (to
approved of 100% design), this loading is equivalent to 4-1/3 -
5-1/2 people full-time.
Pump and Treat - simple
The Task 2 field data acquisition is set at 6 weeks with 10 samples
collected and analyzed. Also, it is assumed that a pumping test
is not required. The design tasks LOE is estimated at one-third
of the complex design. An intermediate design submitted and formal
value engineering are not included in this design. The LOE
required to obtain permits and site access is held constant for all
cases. Permit requirements are typically tied to very specific
data acquisition and reporting formats irrespective of the
complexity of the design.
Summary
The total estimated LOE for the Pump and Treat - Simple generic RD
schedule is 3372 to 4691. With a schedule of 10 months (to
approval of 100 percent design), this loading is equivalent to 2
to 3 people full-time.
Pump and Treat - Simple (Expedited)
The expedited schedule assumes no additional field data collection
is required to complete the design. A portable, "off-the-shelf"
treatment system will be selected. The treatment system vendor
will supply much of the design analysis.
The product of the design tasks will be a package consisting of
twenty specifications (civil, chemical, and mechanical) and five
997
-------
drawings (site plan, general arrangement, P&ID (piping & instrumen-
tation diagram), electrical, and a process diagram).
Summary
The total estimated LOE for the Pump and Treat - Simple (Expedited)
generic RD schedule is 1,641 to 2,225. With a 4 month schedule (to
approval of 100% design), this loading is equivalent to 2-1/2 to
3-1/2 people full-time.
Treatment of Soils and Sludge - Complex
The field data acquisition activities require five specifications.
In addition to those identified previously, the services of a
geotechnical laboratory are also required.
The average NPL site is 10 acres in area. Assume the field data
collection requires 5 weeks and includes the collection of 300
samples, all but 30 are analyzed using an on-site laboratory.
Assume that one technology of a very complex nature will be studied
under the treatability task.
The design criteria to be considered include civil, process
engineering, health and safety, and environmental. The design
components were estimated using a large, east coast Superfund
project as a template. This project design package included 50
specifications and 33 drawings.
Summary
The total estimated LOE for the Treatment of Soils and Sludge -
Complex generic RD schedule is 10,570 to 13,823. With a 17 month
schedule (to approval of 100% design), this loading is equivalent
to 4 to 5 people full-time.
Treatment of Soils and Sludge - Simple
The site for which this category is considered appropriate is
assumed to be one acre in area. Fifty samples are taken during the
field investigation of which 10 are sent to an off-site analytical
laboratory. Design criteria and design activities are similar to
the complex category; however, LOE is considerably reduced due to
the reduction in complexity. As with the other "simple" catego-
ries, the intermediate design submittal and value engineering are
not required.
Summary
The total estimated LOE for the Treatment of Soils and Sludge -
Simple generic RD schedule is 4,406 to 5860. With a 9 month
schedule (to approval of 100% design), this loading is equivalent
to 3 to 4 people full-time.
998
-------
Civil Engineering - Complex
The model used for this design category was a large, east coast
Superfund site which included several activities: soil excavation,
water treatment, a slurry wall, and building decontamination. The
actual LOE for this site was reduced to "remove" the Pump and Treat
aspect from consideration.
Field data collection activities are assumed to be similar to those
required in the Soils/Sludge - Complex category. Similar design
criteria are also considered. An intermediate design submittal and
formal value engineering are included in this category.
Summary
The total estimated LOE for the Civil Engineering - Complex generic
RD schedule is 10,420 to 13,605. With a 12 month schedule (to
approval of 100% design), this loading is equivalent to 5-3/4 - 7-
1/4 people full-time.
Civil Engineering - Simple
The field data acquisition consists of installing 3 shallow
monitoring wells and excavating several test pits. Ten samples are
analyzed at an off-site laboratory. Four design criteria are
considered in developing the Basis of Design (civil, hydrogeologic,
environmental, and health and safety).
The design is straight forward with 20 specifications and 5
drawings required for the procurement package. The design reviews
are performed by a single person (rather than a team) and the
operability review is not performed.
Summary
The total estimated LOE for the Civil Engineering - Simple generic
RD schedule is 3,146 to 4,227. With a 9 month schedule (to
approval of 100% design), this loading is equivalent to 2-1/4 - 3
people full-time.
Civil Engineering - Simple (Expedited)
In this generic category there are no field data collection
activities and no laboratory analysis. The Basis of Design Report
is issued during activity 0103 (Define Design Criteria). The
design activities are simple and uncomplicated with minimal
institutional concerns.
Summary
The total estimated LOE for the Civil Engineering - Simple
(Expedited) generic RD schedule is 1,641 to 2210. With a 4 month
schedule (to approval of 100% design), this loading is equivalent
to 2-1/2 to 3-1/2 people full-time.
999
-------
On-site Thermal Destruction
The LOE for a generic design for on-site thermal destruction was
estimated by first determining the limiting size of a site which
could be remediated under ARCS. A typical unit cost of $750 per
cubic yard for incineration is assumed and a programmatic liiaita-
tion exists of $15 million for Construction under ARCS. Therei'ore,
the maximum excavation permitted is 20,000 cubic yards using ARCS
as the contracting mechanism.
An existing Superfund incineration project with a required quantity
of excavation reasonably close to this limit was selected as the
template for the generic design.
Some water treatment will be necessary for incineration of sludges
(treating effluent of the dewatering effort). Treatability studies
are required at the bench-scale for the water treatment arid at
bench and pilot-scales for the material to be incinerated. Five
specifications are needed to conduct field data collection
activities.
The LOE to support the field data collection activities is assumed
to be similar to that required for the Treatment of Soils/Sludge -
Simple category. A typical site one Acre in extent and with a
required depth of excavation of 10 feet satisfies the area and
volume assumptions presented here and under the Soils/Sludge -
Simple category.
Four design criteria are considered: civil, process (including
also the electro-mechanical criteria), environmental, and health
and safety.
The design activities are similar to the complex categories
previously described and include formal VE and an intermediate
design submittal.
Summary
The total estimated LOE for the On-Site Thermal Destruction generic
RD schedule is 9,851 to 12,939. With a 12 month schedule (to
approval of 100% design), this loading is equivalent to 5-1/2 - 7
people full-time.
9.0 COST CONSIDERATIONS
The preceeding sections of this report present estimates of level
of effort (job hour) requirements for the nine remedy-specific,
generic remedial design schedules. An estimate of cost can be
developed by determining the distribution of the various profes-
sional/technical classifications required and applying the
appropriate salary value to calculate the cost of services.
However, this is incomplete because the LOE estimates presented
herein are for technical production and did not include LOE
required for program management services (including cost/schedule
1000
-------
control and management reporting). Also, some RDs require a
significant expenditure of funds for other direct cost (including
subcontracting for field and laboratory data collection). other
direct costs (ODCs) are very site-specific and are not presented
in this document.
10.0 CONCLUSIONS (RESOURCE LOADING ACTIVITY)
1. The resource data presented in this report, when combined with
the remedy-specific, generic RD schedules, are an excellent
tool for:
initiating planning for RD work assignments; and
use as an aid to review and provide constructive criticism
by RPMs to those producing site-specific RD schedules.
2. The user must be aware that the resource reports and graphs,
like the generic RD schedules they compliment, are not
substitutes for site-specific schedules and budgets developed
by individual task managers.
3. All resource estimates presented in this document are based on
actual work assignment data and the personal experience of
individual engineers, scientists, and managers.
4. All resource estimates are presented within the format of the
Standard Remedial Design Tasks.
5. Table 10-1 presents a comparison of LOE for each standard Task
for each of the nine generic RD schedules.
11.0 RECOMMENDATIONS FOR USERS
The following recommendations are offered for consideration by
users to further enhance the usefulness of the concept of a generic
RD schedule.
The approach presented in this manual should be used by all
parties to an RD work assignment. They will then have a
common starting point from which project-specific discus-
sions and eventually a site-specific schedule and budget
can be developed.
Develop and implement a schedule tracking system to monitor
progress on Remedial Designs at all sites. This system
will provide the contracting party with a real-time measure
of predicted vs. actual activity relative to the baseline
schedule.
To maximize cost and technical efficiencies and to become
aware of and correct possible deficiencies, initiate the
technical reviews (biddability, constructibility, environ-
mental, claims prevention, operability) during intermediate
1001
-------
design. For similar reasons, a VE screening should be
initiated early in the project schedule and a formal VE
review, if deemed appropriate, should be conducted during
intermediate design.
For those sites whose RD will be conducted outside the
limits of the assumptions presented in this manual, obtain
specific information about duration requirements and
current practice for procurement, interagency agreements,
owner reviews, etc., which may effect the start or overall
duration of a Remedial Design.
For those sites where early RA starts are required to
protect the health and safety of the public or for other
reasons, the RD/RA schedule can be organized to allow for
early RD completion and RA implementation on the simplest
operable units first. This would allow earlier RA starts
while simultaneously proceeding with design on the more
complex operable units.
For any site, the same A/E firm should be used to conduct
the RI/FS, the RD, and the construction management. This
project management concept reduces procurement delays,
reduces time required for internal quality control, and
improves contractor accountability.
The standard tasks for remedial design services are
intended to provide a consistent method of reporting design
work. They should be used to the maximum extent possible
within the constraints of site-specific or other criteria.
1002
-------
12.0 REFERENCES
U.S. Environmental Protection Agency, undated. Management of
Construction in the Superfund Program.
Certo, S.C. 1980. Principles of Modern Management. W.M.C. Brown
Co. Publishers.
U.S. Environmental Protection Agency, June 1986. Superfund
Remedial Design and Remedial Action Guidance, OSWER Directive
9355.0-4A.
U.S. Environmental Protection Agency, June 19, 1987. Draft
Standard Remedial Design Tasks.
U.S. Environmental Protection Agency, September 1986. Superfund
Treatment Technologies: A Vendor Inventory, EPA Report No. 540/2-
86/004(f).
OPEN PLAN software package, WST Corporation, Houston, Texas.
PRIMAVERA software package, Primavera Systems, Inc., Bala Cynwyd,
Pennsylvania.
1003
-------
TABLE 2-1
STANDARD TASKS FOR REMEDIAL DESIGN (RD)
1. PROJECT PLANNING
2. FIELD DATA ACQUISITION/SAMPLE ANALYSIS
3. TREATABILITY STUDIES
4. DATA EVALUATION
5. PRELIMINARY DESIGN (30% COMPLETE)
I
6. INTERMEDIATE DESIGN (60% COMPLETE)
7. PREFINAL/FINAL DESIGN (90%/100% COMPLETE)
8. DESIGN SUPPORT ACTIVITIES
I
9. VALUE ENGINEERING (VE) DURING DESIGN
10. COMMUNITY RELATIONS
11. PROJECT COMPLETION CLOSEOUT
1004
-------
TABLE 2-2
GENERIC SCHEDULE TASKS/ACTIVITIES
TASK1
PROJECT PLANNING
OBTAIN SITE ACCESS
DATA REVIEWS
DEFINE DESIGN CRITERIA
INITIAL SFTEVISFT
PRE-DESIQN MEETING
PREPARE DRAFT RD WORK PLAN
REVIEW DRAFT RD WORK PLAN
FINALIZE RD WORK PLAN
APPROVE FINAL RD WORK PLAN
PREPARE HEALTH & SAFETY PLAN
PREPARE DRAFT FIELD SAMPLING & ANALYSIS PLAN
REVIEW DRAFT FIELD SAMPLING & ANALYSIS PLAN
FINALIZE FIELD SAMPLING & ANALYSIS PLAN
APPROVE FINAL FIELD SAMPLING & ANALYSIS PLAN
TASK 2
DATA COLLECTION/SAMPLE ANALYSIS
PREPARE FIELD SAMPLING & ANALYSIS SPECIFICATION
REVIEW AND APPROVE SPECIFICATION
ISSUE INQUIRY
RECEIVE FIELD SAMPLING & ANALYSIS BIDS
EVALUATE FIELD SAMPLING & ANALYSIS BIDS
APPROVAL/ISSUE CONTRACT
FIELD SAMPLING/SURVEYS
LAB ANALYSIS
DATA VALIDATION
TASK 3
TREATABILITY STUDIES
PREPARE TREATABILITY SPECIFICATION
REVIEW & APPROVE TREATABILITY SPECIFICATION
ISSUE INQUIRY TREATABILITY SPECIFICATION
RECEIVE TREATABILITY BIDS
EVALUATE TREATABILITY BIDS
APPROVAL/ISSUE TREATABILITY CONTRACT
BENCH-SCALE PROGRAM
BENCH-SCALE REPORT
PILOT-SCALE PROGRAM
PILOT-SCALE REPORT
TASK 4
DATA EVALUATION
INITIATE DESIGN CRITERIA EVALUATION
1005
-------
TABLE 2-2 (Cont'd)
GENERIC SCHEDULE TASKS/ACTIVITIES
TASKS
PRELIMINARY DESIGN
PROCESS/EQUIPMENT SELECTION
PRELIMINARY DRAWINGS & SPECIFICATION - 30% DESIGN
PRELIMINARY SCHEDULE/ESTIMATE - 30% DESIGN
PARALLEL REVIEW • 30% DESIGN
SERIAL REVIEW - 30% DESIGN
TASK 6
INTERMEDIATE DESIGN
PRELIMINARY EQUIPMENT SELECTION
DRAWINGS & SPECIFICATION • 60% DESIGN
SCHEDULE/ESTIMATE - 60% DESIGN
PARALLEL REVIEW - 60% DESIGN
SERIAL REVIEW • 60% DESIGN
TASK 7
PRE-FINAL/FINAL DESIGN
FINAL EQUIPMENT SELECTION
DRAWINGS & SPECIFICATIONS • 90% DESIGN
FINAL SCHEDULE/ESTIMATE 90% DESIGN
SERIAL REVIEW • 90% DESIGN
FINALIZE DRAWINGS & SPECIFICATIONS
APPROVAL 100% DESIGN
TASKS
DESIGN SUPPORT
FINALIZE DESIGN CRfTERIA
PERMITS, APPROVALS & SITE ACCESS
SITE SAFETY PLAN SPECIFICATION
QUALITY ASSURANCE PROJECT OUTLINE
OPERATION & MAINTENANCE MANUAL
DESIGN ANALYSIS
INITIATE PROJECT DELIVERY ANALYSIS
FINAL PROJECT DELIVERY ANALYSIS
INniALTECHNICAL REVIEWS • 60% DESIGN
FINAL TECHNICAL REVIEWS • 60% DESIGN
TASK 9
VALUE ENGINEERING
VALUE ENGINEERING STUDY
1006
-------
TABLE 2-2 (Cont'd)
GENERIC SCHEDULE TASKS/ACTIVITIES
TASK 10
COMMUNITY RELATIONS
REVISE COMMUNfTY RELATIONS PLAN
REVIEW COMMUNITY RELATIONS PLAN
FINALIZE COMMUNITY RELATIONS PLAN
PUBLIC MEETING RD COMMUNITY RELATIONS PLAN
APPROVE COMMUNITY RELATIONS PLAN
COMMUNITY RELATIONS SUPPORT
PRE-BID
ROD SIGNING
ROD SIGNED
POST-BID
POST RD ACTIVITIES
PREPARATION OF ARCHITECT/ENGINEER SCOPE
ISSUE ARCHITECT/ENGINEER WORK ASSIGNMENT
ARCHITECT/ENGINEER SERVICES DURING CONSTRUCTION
PRE-SOLICITATION NOTICE
ISSUE/RECEIVE BIDS
EVALUATE BIDS/AWARD CONTRACT
NOTICE TO PROCEED/MOBILIZATION
START SITE CLEANUP
1007
-------
TABLE 3-1
DURATIONS: REMEDY-SPECIFIC
GENERIC REMEDIAL DESIGN SCHEDULES
REMEDY
TOTAL
DURATION
1. CIVIL ENGINEERING - SIMPLE
9 MONTHS
2. CIVIL ENGINEERING - COMPLEX
12 MONTHS
3. PUMP AND TREAT - SIMPLE
10 MONTHS
4. PUMP AND TREAT - COMPLEX
13 MONTHS
5. ON-SITE THERMAL DESTRUCTION
12 MONTHS
6. SOILS/SLUDGE - SIMPLE
9 MONTHS
7. SOILS/SLUDGE - COMPLEX
17 MONTHS
8. PUMP AND TREAT - SIMPLE (EXPEDITED)
4 MONTHS
9. CIVIL ENGINEERING - SIMPLE (EXPEDITED)
4 MONTHS
1008
-------
Ill
Z UJ ff
o£ fo
S
CO
CO
H-
0
CC
D
CC
§
V?
op
*" C/3
Hco
CD >
^cr
O
u.
LL
UJ
LL
O
_J
LU
TOEATM
ILS AND
V)
N
N
o
N
R
8
CC
UJ
LU
UJ
S
W
S
i
i
i
i
i
8
i
i
1
s
M
N
§
i
1
8
I
1
g
N
i
i
O
I
j1
n
in
1009
-------
K O O
••*'<
«• u 2
a?2 a
la ~ "
CJ>
LU LU
CD CO
CD LLJ
CJ
LU
CO
1
CD
-d HH
cz?
cc
D_
cc
:2I LU
LU LU
CD CD
CC ~Z.
LU
LU
CO CJ>
o o o.
•~i m CM
o o o.
*i CM CM
O O O
•« »i CM
O O O
»< nj
o •w o
o *-•
Ct]
of
o o
a
en
s
I
5 fi z
i-t u. (B
CJ
s
M
y
OJ
-t->
O
0
13 c
c c
D e
«n
BO.
J I
1010
-------
1011
-------
1012
-------
1013
-------
1014
-------
1015
-------
coo
.. *
.. « o
*i u 2
{.i..
09 I O
LLJ
co
CJ CJ>
LU X
CD LU
a.
CD CD
(— i CJ
I
CD
LU
CD CD
cr a
CL ID
CO
CO
LU
CD
CO
CD
CC LL.
CD
LLJ
CO
LU
LLJ
cr
ce
*-l f* *"•
o o o
•*-i M1 »i
O O O
O O Q
CO
QJ
I Si
z
O
>*>4
u
LO
CO
LU
to o
§ i
I 2
1016
5
i—«
CO
•-I CO
3 ^
CO
•o
0
T5 C
C C
0 ID
a a
j i
-------
e
t-S-
< "
u o
.* a
U Z
>- CD.
O LU
CD CD
•<: LLJ
D_
x
CD LLJ
LU 00
CD C_J>
CC
Q_ LU
CD
cr
LU
cr
LU
CO D_
ID ZD
Q_
o •** o.
»•< o **
O «i O.
U
ae
o o o
in
o>
§
S I
LU
5 a
S
S
to
£ S
g S
1017
9
•0 C
C C
9 «
Oi-i
9 a
J I
-------
< <•
u o
0 I O
CD
O
LU
"ZL D_
CD X
UJ
CJ
LU LU
I— CO
CD -=£
cr cj
d.
Q.
LU
CD
cr: cc
i—i CD
LU
CO
*
ID
u
a:
O
C3
to
OJ
S3
CO
^^ ~— ^_
* i_t_ S
CO
CO
TS
C
•0 C
C C
9 C
0 Q.
J I
TTTTF
-------
CC
S
a
v
in
&
u
**
in
CL
II
.*•_
in
a.
u
S
in
a
a
-B* >
n
o
II
in
2
5 lii
£
m
I S
=)
§
a ^
u • t
c »! Si ,
a ' •
1019
-------
1020
-------
ill
85-*
Sue
.. *
LU
CJ
LU
CD
D_
CD
o o
CD
CJ
LU
CD
CE
CL
CJ
ZD
CE
LU
CD
cr
CO
LU
O
cr
LU
LU
•
en
LU
I—•
H—I
CO
w I
CD
000.
- o ru
o •*« o.
o o o.
•n ^ C\J
O O O.
«H «n ru
o o o
o o o a
o
c
V
H- CC
I
5
§B „ *
5 5 a
CD ^ S M (/)
Z X f Tt B» W
« B to p Ly o
o i-
g c
5 - 3 SJ i 2 S
3 >- p 03 > « y
H o o •< LU z z
OT § < 5 < 3 S
Q ^ y »- LU t:
1 2 5 2 E 5
_ CO
50 O LU
Z i—' "i
S H- »-
S 2 2
!U * P
a z - 5 a
u. a LU 3
* « 3 I
LU « S
5 > u
U
at
O
»*
u.
en
QJ
C
c c
0 C
DO.
J I
1021
-------
Remedial Management Strategy
Thomas A. Whalen, P.E.
Construction Management Consultant
Design and Construction Management Branch
U.S. Environmental Protection Agency
Washington, D.C. 20460
703-308-8345
PRE-DESIGN PLANNING
There is a fascination to watching a scientific proposal to reduce, control, or eliminate risks to human
health and the environment contained in the Record of Decision (ROD) emerge through the aid of
engineering to a remedial design on paper. And indeed it is, for the conception of the remedy can
involve as much a leap of the imagination, and as much a synthesis of experience and knowledge as
any scientist is required to formulate a hypothesis. And once that remedy is chosen in the ROD, by
the RPM, as scientist, it must be analyzed by the engineer, as designer, in a rigorous practical
application of the knowledge of pure science.
It is, however, impossible to go directly from the ROD into remedial design (RD). Unfortunately,
a phase prior to design has often been overlooked. That is, "The Preliminary or Design Report Phase"
between planning and design which is described in ASCE's Manual No. 45, "Consulting Engineering;
a guide for the Engagement of Engineering Services." In the Superfund program this phase is called
the pre-design planning phase. During this phase the ROD and supporting documents should be
converted to a scope for RD and remedial action (RA) by expressing EPA's technical and managerial
requirements.
The Pre-Design Technical Summary (PDTS) and Remedial Management Strategy (RMS), completed
during the pre-design planning phase, form the link between the scientific site assessment and the
engineered solution. The PDTS expresses EPA's technical requirements; the RMS contains the
managerial requirements. Therefore, the PDTS and RMS should constitute the complete project
definition, including realistic objectives. The objectives must be quantified; the requirements must
be clearly stated.
The PDTS is a comprehensive compilation of technical information to ensure that the designer fully
understands the technical objectives of the RA. A separate guidance document explains the
preparation and content of the PDTS. The RMS contains an analysis of the major management
considerations required to achieve the goals of the ROD in a timely manner. Preparation of the PDTS
and RMS are essential for the smooth progression of a project through remedial design and remedial
action.
The PDTS should be completed prior to negotiations with the PRPs whereas the RMS should be
completed after the negotiations. It is recommended that the initial RMS be completed by the RPM
if the project is fund lead. (The RMS would also be a useful analysis for the PRP). The RMS should
be considered an iterative document to be finalized by the party that contracts for design prior to
preparing the scope of work for the design.
1022
-------
PROJECT REQUIREMENTS
Innovation is a mandate in the Superfund program. Innovation and project complexity involve cost,
time, and performance risks because of the lack of precedent. The RMS should consider the
allocation of these risks as well as in what degree and where there shall be compromises before the
design is initiated. The terms of the compromises - including inexperience, overly restrictive
technical or managerial requirements, pressures of deadlines and economy in cost - vary the shape
of the project to be designed. The PDTS and RMS must, therefore, contain wise and carefully
selected technical and managerial requirements.
Unfortunately, compromise implies a degree of failure. It is then the responsibility of the designer
to obviate failure within the context of the technical and managerial requirements articulated in the
PDTS and RMS. It is, however, impossible for any design to be "the logical outcome of the
requirements" simply because, the requirements being in conflict, their logical outcome is an
impossibility.
The content of the RMS should, of course, be modified depending on the complexity of the RD and
RA. For simple projects many of the requirements need not be addressed the content and level of
detail are left to be the discretion of the analyst.
DISCUSSION
The managerial requirements stated in the RMS should encourage efficiency and cost-effectiveness
in remedial design, remedial action, operation and maintenance. The requirements should also control
and manage RA risks within reasonable limits. As a minimum, the RMS should contain an analysis
of the project's managerial goals and constraints as stated in the ROD and a critical strategy for RD
and RA as well as EPA policy and guidance.
I. Develop a plan for communications, co-ordination, and organization of all the parties
involved in the project, including procedures for rapidly resolving conflicts.
A. Contracting Party. The RMS should state which organization will contract for RD
(PRP, State, EPA, USAGE or USER) and the RA (PRP, State, ARCS, USAGE or
USER).
B. Communications. The best way to assure a rapid response to conflicts is with a
communication matrix. This matrix should show the procedural flow of information
such as submittals, memoranda, documents, and approvals. These communications
procedures should be agreed to by all parties before the RD begins.
II. Provide a reasonable estimate of the duration and resources needed for design; schedule and
budget projections for remedial action; and cost and schedule control procedures.
A. Funding. Funding considerations are of particular concern in the development of a
management strategy, particularly if the project is a multi-year effort. The strategy
must address the availability of funds including the State cost share and obligations
during future years. The RMS should include budget planning projections based on
the proposed project schedule, contract packages, and the contingent liability for
increased cost during RA.
1023
-------
B. Resources. An analysis must be made to determine the special technical qualifications
required for the work, the workload and availability of the resources required, and
the level of interest of qualified designers and RA contractors.
C. Schedules. Successful management of the project depends on the performance of
responsible and qualified managers and contractors, the maintenance of schedules and
budgets, and the rapid resolution of problems. Techniques for good management
include requirements that a schedule be agreed to between the contracting party and
the contractor, that the schedule be reviewed and updated monthly, and that
enforcement of the schedule by the contracting party be maintained. Of course, the
schedule must be reasonable, must establish obtainable goals, must contain sufficient
detail to permit task control, and must be based upon a complete scope of work.
There are many reasons for development and maintenance of a schedule. The
schedule is a tool used to discuss the RD or RA contract between the parties to the
contract and is also the principal tool for exacting control of contract progress. The
schedule also is the basic documentary and analytical tool for negotiation and
settlement of requests for equitable adjustments, claims, and disputes as well as for
contract termination and closeout.
The party that contracts for RD or RA has the exclusive responsibility of schedule
enforcement, of explicit approval or rejection of the schedule and of imposing
sanctions for non-compliance. The control of the schedule is the exclusive
responsibility of the RD or RA contractor who also has responsibility for handling
unforeseen conditions and interface impacts. Schedule revisions may be requested by
either party; however, revisions to the schedule must be approved by the contracting
party.
III. Develop a Remedial Delivery Analysis.
The Remedial Delivery Analysis (RDA) is the development of the contracting strategy for the
completion of the RD and RA. The decisions made in the Remedial Delivery Analys,is must
be well conceived and involve the decision makers. The initial RDA should be done as part
of the RMS as the decisions made will dictate the scope and complexity of the design. It is
most likely that the RA contracting strategy can not be finalized without additional data,
obtained during the early design tasks, and this should be reflected in the design schedule.
Although the party contracting for RA will have its own objectives and priorities as to cost,
time, and quality, it may often look to the designer as an advisor to recommend the RA
contracting strategy deemed most suited to the project. If the party contracting for RA relies
on the designer and later encounters problems in the selected contracting strategy, it may
blame the designer. In those situations where the party contracting for RA takes the initiative
and mandates a given delivery strategy, the party contracting for RD and RA should explicitly
set forth recommendations for the role of the designer during RD and RA.
A. Design Approach
B. The number and scope of RD and RA Contracts
C. RA Procurement Methods
D. RD and RA Contract types
1024
-------
IV. Determine the number of remedial design and remedial action contracts.
A. Phasing and fast-tracking. Since EPA has a preference for quick action, an important
item to be evaluated in an RMS is the potential for phasing or fast-tracking the
project. These approaches will allow the RA to be implemented sooner than if all of
the steps were treated as a single design and remedial action.
B. Equipment. The ROD may specify a process or remedy that requires special or
proprietary material and equipment, particularly if a new or innovative technology is
recommended. In these instances, it is important to evaluate the delivery schedule for
such material and equipment. This would include the time necessary to review shop
drawings, do performance testing, and for shipping requirements. If these processes
are anticipated to take a long time, consideration should be given to purchasing the
material and equipment under a separate contract to ensure its timely delivery to the
site.
C. Use of or Rights in Patents. There are at least two occasions when the contracting
party may be obligated to pay a royalty for the use of or for rights in patents:
1. The remedial design includes a patented product, apparatus, or process, or
2. A patented product, apparatus or process may be necessary for the proper
performance of a contract.
Royalties for the use of or for rights in patents, are generally allowable costs
within the limits of the principles and procedures contained in the Federal
Acquisition Regulations and EPA's Regulations Governing Cooperative
Agreements.
D. Advertising. When considering when to award the RA contract, especially small
projects, the best time to advertise the RA must be evaluated including the seasons of
year when the work will occur, the geographic location, and other contractors working
at the site.
E. Remedy classification. The remedy should be classified into one or more of EPA's
characteristic remedies. Each remedy may be a separate design or a comprehensive
design may contain a combination of these remedies. In that case, each of the
component remedies is worked in parallel and the more complex, time-consuming
remedy will control the overall project duration.
1. Civil Engineering.
2. Pump & Treat.
3. On-site Thermal Destruction.
4. Soils and Sludge Treatment.
F. Noncompetitive Procurement. The Competition in Contracting Act of 1984 (CICA)
provides for the use of "other than full and open competition" for some acquisitions.
The term "noncompetitive" is often used to mean other than full and open competition.
1025
-------
This means not only sole source acquisitions, but also those situations where an agency
is permitted to limit the number of sources solicited.
G. RA Procurement Methods. Two primary competitive methods procedures may be
used for the procurement of supplies, services, and RA. These are the solicitation of
sealed bids (formal advertising method) and the request for competitive proposals
(competitive negotiation method).
In determining the appropriate competitive procedures to be used, a public agency
should determine:
1. The time available for the solicitation, submission, and evaluation of offers;
2. If the award will be made on the basis of price, other factors or a
combination;
3. If it is necessary to conduct discussions with the responding source about their
offers; and
4. If there is a reasonable expectation of receiving more than one offer,
H. RD and RA Contract Types. The enormous scale and complexity of public acquisition
has necessitated the development of a wide variety of contract types. The term
"contract type" has several different connotations. Often it is used to indicate the
various methods of pricing arrangements, of which there are two basic types: fixed-
price contracts and cost-reimbursement contracts.
V. Assure a quality design which anticipates potential problems, is in sufficient detail to solicit
reasonable offers for remedial action and ensures function, efficiency and economy.
A. Responsibility of the Contracting Party. It is the responsibility of the party that
contracts for design to:
1. Prepare a complete, detailed scope of work for design.
2. Communicate project objectives and critical need dates.
3. Select qualified professionals, identify special expertise needed and authorize
formation of multidiscipline design teams.
4. Establish design criteria and requirements.
5. Provide adequate schedule and budget for design.
6. Require the designer to implement quality assurance, quality control, and peer
review programs.
7. Provide timely reviews and approvals.
8. Stress completeness, timeliness, and professional presentation of submittals.
1026
-------
9. Assure that value engineering, biddability, constructibility, operability, claims
prevention, and environmental reviews of the design are conducted.
10. Be prepared to coordinate, negotiate, and resolve conflicts in a timely manner.
B. Designer Responsibility. Design is a professional service, as defined by State law,
which is required to be performed or approved by a person licensed, registered, or
certified to provide such a service. Because the A-E, as the designer, offers
professional services on the basis of its fitness to act in the line of work for which it
is employed, the A-E has a duty to avoid negligence; provide an implied warranty for
the design; and fulfill specified contractual requirements. The A-E is responsible for
providing professional quality work that meets the standard of care, skill, and
diligence that one in the profession would ordinarily exercise under similar
circumstances.
When a modification to a RA contract is required because of an error
or deficiency in the design, the party that contracted for the design
must consider the extent to which the A-E may be reasonably liable.
C. Risk Management. The RMS should contain an analysis of the potential risks
associated with the project including financial, schedule and technical risks. This
evaluation should include a review of the degree of certainty regarding the estimates
of the types and quantity of work that needs to be done as well as cost estimates.
While the party contracting for RA may wish to shift a significant amount of risk to
the RA contractor, an inordinate or inequitable transfer of risk may impact the
project in terms of increased cost. These increased costs may result from less
competition among RA contractors who may be unwilling or unable to provide an
offer, increased contract modifications because of unknown or unanticipated
subsurface conditions, claims based on conduct of the party contracting for RA, or
schedule delays.
The RMS should contain an analysis of the basis for the method of managing the risks
associated with the project. This includes decisions regarding the method of
procurement, type of contract, availability, types, and amounts of insurance required
by contractors, the availability and amount of bonding required, indemnification and
liquidated damages.
D. Design versus Performance Specifications. Frequently an RA contractor will
encounter difficulties in performing under the specifications or drawings. Generally,
defective specifications are defined as those specifications which contain errors,
conflicts, or omissions which prevent performance completely or in the manner
contemplated by the parties to the contract. The most common defective
specifications are clear errors in the contract documents or conflicts between
provisions. Other common deficiencies include: errors or omissions of important
facts or dates, and conflicts between the specifications and drawings. Many of these
problems may be the pecuniary liability of the designer.
The party contracting for RA impliedly warrants that the RA contractor will be able
to fulfill its responsibilities, as set out in the specifications, if "design" specifications
are provided which precisely states how the contract is to be performed. If the RA
contractor makes a good faith effort to follow the design specifications, but is unable
1027
-------
to comply because the contract documents are inadequate or do not contain the
required or necessary details to complete the item specified, the contracting party
bears the risk of loss.
In contrast, if the party contracting for RA allows the contractor discretion in how to
meet the contract obligations by providing "performance" specifications and no
explicit statement of how to design or build the item is offered by the contracting
party, the inability to complete the contract is borne by the RA contractor. If the RA
contractor has undertaken an impossible task, meets technological problems, or cannot
complete performance because of its lack of experience, the contractor and not the
contracting party, bears the risk of loss.
E. Project Quality. Quality is conformance to the requirements that meet the project's
needs and expectations. Of course, to achieve those needs and expectations, they must
be clearly stated at the beginning of the task as they cannot be misunderstood.
Quality neither depends on, nor is achieved through, multiple reviews.
VI. Develop a well defined scope of work for RD.
It is expected that the RD will be consistent with the Record of Decision (ROD), will comply
with Superfund program policies and procedures, will minimize RA contract modifications,
and will prevent RA contractor claims.
EPA's 11 standard tasks that should be used in architectural or engineering (A-E) agreements
for RD. The tasks are intended to provide a consistent method of reporting design work.
While some variations are anticipated because of the variety of design projects and differences
among the A-E firms, the standard tasks should be used and reporting formats should be
amended to be consistent with this set of standard RD tasks.
The standard tasks for RD are:
1. Project Planning. This task includes work efforts related to the initiation of a design
project after the A-E agreement is executed.
2. Field Data Acquisition and Sample Analysis. This task consists of the effort required
to obtain specific field samples and information needed during the design effort that
was not produced during the RI and FS.
3. Treatability Studies. This task includes work efforts related to conducting pilot and
bench scale treatability studies during remedial design.
4. Data Evaluation. This task includes efforts related to the organization and evaluation
of data that will be used later in the design effort.
5. Preliminary Design. This task begins with initial design and ends with the completion
of approximately 30 percent of the total design.
6. Intermediate Design. This task begins at the completion of the preliminary design
phase and ends with the completion of approximately 60 percent of the total design.
Depending on the size, complexity, and timing of the design effort, this task may be
omitted at the discretion of the contracting party.
1028
-------
7. Prefinal/Final Design. The prefinal/final design phase commences at the completion
of the intermediate design effort and is finished when the entire design effort has
been completed.
8. Design Support Activities. This task consists of design support effort which is
conducted during one or all of the three phases of design.
9. VE During Design. If the initial VE screening conducted during the project planning
task identifies a potential cost savings, a VE study will be initiated under this task.
Value engineering is a specialized cost control technique which uses a systematic and
creative approach to identify and to focus on unnecessarily high cost in a project in
order to arrive at a cost saving without sacrificing the reliability or efficiency of the
project.
10. Community Relations. This task incorporates all work efforts related to the
preparation and implementation of the community relations plan during the design
phase of the project.
11. Project Completion and Closeout. This task includes efforts related to the support of
project completion and closeout activities in both the technical and financial area as
well as in the file maintenance and record indexing area.
VII. Expect that the RA contract documents be free of potential errors, conflicts, omissions,
ambiguities, and misrepresentations and establish a system to administer, interpret and manage
those contracts.
A. Design Reviews. It is the responsibility of the party that contracts for design to assure
that the design reviews and approvals are conducted. These activities may be
conducted in parallel with other ongoing design activities or in series, whereby
subsequent activities do not start until the review is completed, comments are
resolved, and approval to proceed is provided.
The designer has a professional responsibility regarding the impact and liability of the
comments on the design and must communicate this to the contracting party. The
review of the plans and specifications, by the party that contracted for design,
generally is for administrative purposes only. That is, the review is to assess the
likelihood that the project will achieve its remediation purposes and that its
performance and operations requirements have been identified. The structural,
mechanical, and electric aspects of the plans and specifications need not to be
reviewed in detail by the party that contracted for design. The acceptance of plans
and specifications by the party that contracted for design does not relieve the designer
of its professional liability for the adequacy of the design.
B. Principal Purpose of the RA Contracts. RA means those actions consistent with the
permanent remedy taken instead of, or in addition to, removal action in the event of
a release or threatened release of hazardous substance into the environment, to prevent
or minimize the release of hazardous substances so that they do not migrate to cause
substantial danger to present or future health or welfare or the environment. It should
be noted that not all of the activities contemplated as "Remedial Actions" are in the
nature of "construction". Some can be considered to be the performance of "services."
1029
-------
The distinction between construction contracts and service contracts for remedial
action can be quite difficult when determining whether DBA applies. Instances may
arise in which, for the convenience of the contracting party, instead of awarding
separate RA contracts for construction work subject to DBA and for services of a
different type to be performed by service employees, the contract may include
separate specifications for each type of work in a single contract calling for the
performance of both types of work. For example, offers may be solicited for a pump
and treat system or an incinerator, as well as their operation and maintenance. The
installation as well as associated excavation, hauling and landscaping may be
considered to be construction covered by DBA; whereas, operation and maintenance
is a service and not covered by DBA.
VIII. Provide for the inspection of the implementation and completion of the remedial action by
qualified persons to ascertain compliance of the remedy with the ROD and its project
performance requirements.
A. Engineering Support during RA. The contract for A-E support during RA should be
in place prior to approval of 100 percent design. The concept of working to a "total"
remediation schedule for a single site (RI through completion of RA) in an efficient
manner necessitates the early identification of an A-E to provide engineering support
during RA. This will assure that the RA will not be delayed due to lack of
engineering support and also permits timely support to the party contracting for RA
for several pre-award activities including the pre-offer conference and evaluation of
the RA offerers.
It is incumbent upon the party that will contract for RA to initiate the activities
necessary to identify the A-E to support the RA task early. It usually is most
efficient to use the services of the A-E firm performing the design for this effort.
B. Remedial action quality assurance requirements. A requirement for an RA contractor
developed quality assurance plan should be included in the statement of work or
specifications for each RA. The development and application of a site-specific
quality assurance plan will help to ensure that all the components of the completed
project or RA have been completed to meet or exceed design criteria, plans, and
specifications. Quality assurance includes inspections, verifications, audits, and
evaluations of materials and workmanship necessary to determine and document the
quality of the project.
C. Project Performance. Project performance includes project start-up, systems testing
under the various possible operating conditions, acceptance or rejection, warranties,
operation and maintenance manuals, and organizational responsibilities. The project
must conform to its applicable performance and operations requirements.
1030
-------
ACQUISITION SELECTION FOR HAZARDOUS WASTE REMEDIATION
William R. Zobel, P.E.
U.S. Environmental Protection Agency
Hazardous Site Control Division
401 M Street S.W.
Washington, D.C. 20460
(703) 308-8354
INTRODUCTION
Engineers, by the nature of their education, strive for precision and accuracy in their work. The
engineer typically selects building materials through design calculations that point to a specific
amount of a specific item. This process often leads the engineer to the preparation of detailed
specifications and drawings suitable for fixed priced type contracts. There are times when it is not
advantageous to the client to obtain the specificity needed to develop such a contract. Additional
field investigations may be cost-prohibitive or time constraints placed on the client may require
expeditious action. In these cases the engineer needs to develop an acquisition strategy based on the
amount of information available for their design. The engineer would then select the appropriate type
of specification and contract to meet the needs of the project.
The intent of this paper is to discuss the various options available to the engineer in selecting an
acquisition strategy for hazardous waste remediation. Most hazardous waste remediations will use
well defined specifications and a fixed price contract. Hazardous waste remediations provide more
design uncertainty than other civil engineering projects. For this reason the engineer should consider
various acquisition methods early in their design.
BACKGROUND
Traditional Government Construction Contracts
The construction of buildings, roads and dams come to mind when thinking of traditional
Government construction. These projects are usually well-defined. They use a sealed bid
procurement process with design specifications and a fixed price (lump sum or unit price) contract
for construction. These contracts allocate a substantial amount of risk for increased costs, delays and
non-performance on the contractor.(1) In return the contractor adds a contingency for unexpected
work into their offer. A well defined project gives the contractor a better means to define the
possible risk. The amount of contingency a contractor includes in the offer relates directly to the
perceived risk.
Procurement Procedures
The Invitation for Bid (IFB) procedure is the Government's common procurement method for most
civil engineering projects. This procedure is the easiest for the Government to administer and
insures a competitive process. The IFB procedures tie both the Government's and the bidder's hands
at the time of bid opening. Bids opened in public limit both the Government's and the contractor's
options after the opening. The Government determines that the bids are responsive and responsible.
(Responsiveness is a determination that the contractor has met the requirements of the IFB while
responsibility is the determination that the contractor can perform the specified work.)(2) After this
determination, the Government's only option is to award the contract to the bidder with the lowest
1031
-------
responsive and responsible bid. The procedure also limits the bidders' options. A bidder has no
chance to adjust, clarify or correct the bid after the opening. Essentially the Invitation for Bid
procedure is a one shot deal. The prospective bidder must consider the potential risk and include in
the bid price(s) a contingency for that risk.
The Two-Step Sealed Bidding process provides a mechanism for negotiation on the technical aspects
of a project while retaining the competitive nature of the sealed bid. In the first step of this process
the Government issues a Request for Technical Proposals (RFTP) describing the project requirements.
Bidders in turn respond to the Government with a technical proposal explaining their approach to the
project. The Government then reviews the proposals, determines if they satisfy the minimum RFTP
requirements, and possibly clarifies the proposals with the bidders. The second step is the submission
of sealed bids by the bidders whose technical proposals meet the Government's minimum
requirements. The Government opens the bids publicly and awards the contract to the lowest
responsive and responsible bidder.
The two-step sealed bid offers several advantages. First the two-step seal bid accords the
Government a method to review proposals that, under the strict specification adherence of the sealed
bid method, would not be considered equal. The process assures the Government that the
procurement is competitive through the submission of sealed bids during the second step of the
process. The two-step sealed bid permits the Government a means to collect technical information
without the use of research and development contracts. The Government can use this information
to help in future solicitations using the sealed bid method.(2)
The two-step sealed bid has disadvantages as well. Preparation and review of technical proposals is
time-consuming and costly to both the bidders and the Government. Final submissions are based on
the least costly design to assure bidders that they remain competitive. The Government does riot have
the flexibility the select other than the lowest bid although another package may be technically
superior. The flexibility and latitude allowed the contracting officer opens ground for bid protests
and contract disputes.(2)
The Request for Proposal (RFP) procedures are similar to the two-step sealed bidding process in that
they permit the Government and the offerer a means to discuss the project during the procurement
process. Unlike the two-step sealed bidding procedure, price can be discussed during negotiations
using a request for proposal. The offerer typically submits a technical proposal including a price.
The Government then evaluates the technical proposal, and discusses the proposal with the offerer.
Through these discussions the Government determines if the technical proposal meets the minimum
requirements of the specifications. Offerers who meet the minimum requirements are deemed
technically acceptable. The Government requests a best and final offer from those proposers.
Typically the Government selects the offer with the lowest price judged to be technically acceptable.
One advantage in using the negotiated procurement procedure is that it allows the Government
discretion in selecting a successful offerer. The Government, through a source selection plan,
determines evaluation factors, the relative importance of the factors and the importance of the cost
differentials of the offers. Government evaluators use weighted evaluation factors to help in selecting
the best offer. Inclusion of the factors in relative order in the RFP informs potential offerers of the
areas considered critical by the Government.
The Government also has the prerogative to use a tradeoff analysis rather than select the lew price
or highest rated proposal. Using the tradeoff analysis the Government can select a proposal with the
best balance of technical merit and cost. Cost consideration usually occurs in one of two methods:
total points or dollar per point. The total point method includes cost as an evaluation criterion. The
lowest price receives all the points available as determined in the criterion. The other offered costs
1032
-------
receive points proportionally to their costs as compared to the lowest cost. The dollar per point
method divides the offered cost by the point score of the other criteria in the evaluation.(3)
A strict adherence to either method may lead to a selection that is not in the best interest of the
Government. A point score leads to the belief that the evaluation is a precise operation when in fact
the scores show only a relative relationship between the offers. Offers with similar point scores can
vary significantly in technical merit and price. When the Government performs a tradeoff analysis,
the contracting officer reviews the point scores to determine whether a point differential represents
a signficant difference in technical merit. This review may lead the Government to select the
proposal with the "greatest value" (FAR 15.605) rather than cost alone. The application of the
tradeoff analysis to the final scores before award allows the Government to select the best overall
offer.(3)
The Request for Proposal process allows the offerer the opportunity to limit their risk by further
defining their proposed actions within the technical specifications of the project. The Government
has the option under the RFP process to negotiate with the offerers on the technical and financial
aspects of the project if it chooses, allowing offerers to seek clarification on the technical aspects of
the project that could reduce their risk and subsequent offer.
Specification Type
Contract specifications can be classified into three types: design, performance and functional. While
specifications can be classified into these types, few if any specifications are purely one type. Most
often a specification is primarily one type with components of the others included. Traditional
Government construction contracts are a combination of design and performance specification.
A design specification provides specific detail/instruction on the methods and/or materials to be used
to accomplish a task of work. Design specifications set out materials, tolerances, measurements,
quality control, inspection requirements and other specific information. The use of design
specifications for a specific process or product is desirable. Design specifications are advantageous
in that they provide the Government assurance that it gets exactly what it wants. The disadvantage
of the design specification is that the Government accepts all responsibility for a product that does
not function as desired provided the contractor performed within the design specification.
Performance specifications describe the desired result and general approach rather than a specific
process or design characteristic(s). Performance specifications require contractors to select the method
they feel will best meet the Government's requirements. The construction contractor is generally
responsible for the detailed design, construction and final achievement of the product. By using
performance specifications the contractor will share in the risk and responsibility for final project
performance. The contractor generally controls the detailed design and construction process while
the Government inspects and approves the final project. The contractor is responsible for meeting
the requirements of the Government's performance specifications.
Performance specifications generally allow more than one approach in meeting the required end
result, suggesting that this type of specification is not appropriate for sealed bidding. The variation
of approaches selected by potential bidders cannot be assessed in a sealed bidding process. When
prepared to restrict the bidder's options to only those methods that will meet the Government's needs,
performance specifications can be used with the sealed bid.(2) When writing performance
specifications for use with a sealed bid, the engineer must provide a set of criteria that limits
prospective bidders' procedurial options such that all offers are equal.
Functional specifications state only the final or ultimate objective of the desired product. Functional
specifications can be considered a performance specification that does not address any approach or
1033
-------
process in meeting the product. A functional specification describes the minimum characteristics
needed to achieve the objective. The use of a functional specification assumes an attainable objective.
Because of the variety of approaches that can be offered under this type of specification, the
Government must be prepared for the possibility of a lengthy and costly evaluation process.
Risk or responsibility for meeting the project goals varys significantly depending upon the type of
specification. The Government, as author of the project specification, can delegate the risk involved.
By calling for specific materials, equipment or methods in a design specification, the Government
assures itself that the desired work occurs. The Government bears a large portion of the responsibility
and risk for performance since it provides detailed instruction to the contractor. Performance and
functional specifications require the contractors to select methods they feel will meet the needs of the
Government while staying within their cost constriants. Methods used are not a concern of the
Government provided the results meet the requirements of the contract. Performance and functional
specifications tranfer control and risk from the Government to the contractor.(5)
Contract Types Available to U.S. Government
The Federal Acquisition Regulations (FAR) define the system that the United States Government
must use to obtain contractual services. There are four general contract types available under the
FAR: Fixed Price, Indefinite Quantity, Time and Material and Cost Reimbursement. Fixed price
contracts can be divided into four sub-types:
Firm Fixed Price - FAR 16.202 Firm fixed price may be sealed bid or negotiated. This type
of fixed priced contract can be used when defined design or performance specificalions are
available. The contract price is not subject to change despite contractor performance costs.
This type of contract places all the financial risk on the contractor while it places the least
amount of administrative burden on the contracting officer.
Unit Price - FAR 16.2 & 12.403 (c). Unit price may be sealed bid or negotiated. This type
is for construction only. The required quantity of a specific unit can be undetermined, but
a reasonable estimate is known and reasonably definite design or performance specifications
are available for the units. The contract includes a "variation in estimated quantities" clause
to allow equitable adjustment between target quantity and actual quantity delivered. This
provision reduces the contractor's fixed price per unit due to equitable adjustment based upon
actual performance. It places the burden of providing for accurate recording of quantities
delivered on the contracting officer.
Fixed Price Incentive - FAR 16.204 & 16.403. Fixed price incentive can only be used with
negotiated procurement. It is selected when cost uncertainties exist, but there is potential for
cost reduction and/or performance improvement by giving the contractor a degree of cost
responsibility and a positive profit incentive. Profit is earned, or lost, based upon the
relationship of the contractor's final negotiated cost to total target cost. The incentives are
placed on cost.
Fixed Price with Award Fee - FAR 16.402-2. Fixed price with award fee is for sealed bids
only. The contract is firm fixed price at the start based on definitive specifications, but the
contract allows payment of an additional fee or portions thereof for exceptional performance.
The performance must be objectively measurable (ie, "exceptional" versus minimum
requirements of contract). The contract must provide clear and unambiguous evaluation
criteria. The deletion of a work item requires a compensating deletion in award fee.
1034
-------
Indefinite quantity is the second type of contract allowed in the FAR. It may be used with a sealed
bid or negotiated procurement. The Government would select an indefinite quantity contract where
it is impossible to determine in advance the precise quantities of supplies or services that will be
needed by designated activities during a definite contract period. The method of ordering work must
be stated as well as minimum/maximum orders allowable during a specified period. Regulations
require the development of a fixed unit price schedule (Schedule of Work) before award that provides
a basis of cost for items to be ordered. The contract contains estimated quantities used for bid
evaluation. There are two sub-types of indefinite quantity contracts.
Requirements - FAR 16.503. In a requirements type contract the Government is not obligated
to place any minimum orders. The contract obligates the Government to order from
successful contractor and no other source for all supplies and services described in the
contract. The contractor has the legal right and duty to provide the supplies or services
determined by the Government's need and not by a fixed quantity.
Indefinite Quantity - FAR 16.504. In an indefinite quantity type contract a stated minimum
shall be ordered by the Government during the contract period. The contract also must
specify a maximum amount to be ordered. The regulations limit the use of this type of
contract to commercial or commercial type items that the Government needs on a recurring
basis.
Time and Materials type contracts are defined in FAR 16.601. Time and materials contracts may be
sealed bid or negotiated procurements. The Government selects this type of contract when it is not
possible at time of contract preparation to estimate the scope (extent or duration) of work required
with high degree of accuracy. The contract calls for provision of direct labor hours at an hourly rate
and provision of materials at a designated cost. The contract contains estimated quantities used for
bid evaluation purposes. Time and materials contracts require the use of time and cost standards
applicable to particular work items and appropriate surveillance by government personnel. Funding
is obligated to each work order prepared under the contract.
Cost Reimbursement contracts can be used for negotiated procurement only. The total award fee plus
base fee cannot exceed the statutory limits as indicated in FAR 15.903(d). This type of contract is
very costly to administer and requires the contractor to have an adequate accounting system. This
type of contract can be used only when the nature of the work or the unreliability of the cost estimate
makes it impossible to use another contract type. Two sub-types of cost reimbursement contracts are:
Cost Plus Incentive Fee, FAR 16.404-1. Cost plus incentive fee is utilized when development
has a high probability that it is feasible and positive profit incentives for contractor
management can be negotiated. The performance incentives must be clearly spelled out and
objectively measurable. The contract must contain target cost, target fee, minimum and
maximum fees, and fee adjustment formula. The fee adjustment is made upon completion
of the contract and based on the end results, not their cause. Cost plus incentive fee contracts
are suitable for research and development projects.
Cost Plus Award Fee, FAR 16.404-2. Cost plus award fee contracts are very effective in cases
where it is impossible to write a contract specification containing a precise description of the
work expected to be performed. The Government uses a cost plus award fee contract when
contract completion is feasible, incentives are desired but contractor performance is not
susceptible to finite measurement. This contract sub-type provides for subjective evaluation
of contractor performance. The Government determines the fee to be paid and the
determination is not subject to dispute. A cost plus award fee contract must contain clear and
unambiguous evaluation criteria to determine award fee. The Federal Acquisition Regulations
1035
-------
permit the Government a variety of choices in selection of contract type. The Government
must decide where it wishes to place its resources and risk in the completion of a project. The
fixed price contracts force the Government to do a thorough investigation and design prior
to solicitation. The benefit of this work is a contract that minimizes risk allocation to the
Government and has the lowest price at the time of solicitation. The other types of
contracting allow an expedited solicitation while placing greater demands on the Government
in contract administration, risk allocation and potential cost.
Hazardous Waste Categories
The Design and Construction Management Branch, Hazardous Site Control Division, U.S.
Environmental Protection Agency, has developed categories for Superf und remediations for discussion
purposes. These categories are used in subsequent discussions of remediations and possible acquisition
strategies.
Civil Engineering: The simple civil engineering projects contain such remedies as fencing,
groundwater monitoring, and minor earthwork, demolition or removal activities. The complex civil
engineering projects may require more extensive construction effort such as a Resource Conservation
and Recovery Act (RCRA) cap, extensive or complicated excavation or demolition activities, or the
construction of other engineered structures.
Pump & Treat: This category is for groundwater withdrawal, treatment and discharge or disposal and
surface water or leachate treatment. The technology categories include physio-chemical and
biological treatment of liquids. Specific technologies include: air stripping, carbon adsorption, metals
precipitation, ion exchange, multi-media filtration, aerobic and anaerobic biodegradation,
evaporation, and distillation. In the simple projects the technologies would be proven for the
contaminants of concern and would be available in "off the shelf" package treatment units. In
addition, the aquifer characteristics would not be complex, and standard pumping systems would be
used. In a complex pump and treat project, the aquifer, contaminants, and the pumping and
treatment system design effort is a more difficult, time consuming effort such as innovative water
treatment technologies.
Soils and Sludge Treatment: This category includes the physical, chemical or biological treatment or
volatilization of soils and sludge. All non-thermal destruction of solids would be treated under this
category. In the simple project the process chosen would be a well proven technology for the
contaminants of concern and for the existing site conditions. A complex project would include
innovative processes requiring extensive testing and development.
On-site Thermal Destruction: This category includes on-site incineration, pyrolysis and in-situ
vitrification.
DISCUSSION
Hazardous waste remediation does not fit the mold of the typical Government construction project.
The Government spends considerable effort to define a project in its solicitation package. Hazardous
waste sites consist primarily of abandoned buried waste with little or no record of the location.
Sampling during the remedial investigation and feasibility study is directed toward remedy selection,
not design. Many sites require additional sampling and engineering investigation activities so that the
Government can prepare a detailed solicitation package. This effort often conflicts with the
neighboring community's desire for action at the site.
103B
-------
Several design options are available to the Government and its engineer. Most engineers prefer to
devote the time and effort to define a project to the best of their ability. Solicitation packages
generated because of this process place a substantial amount of risk on the contractor. The engineer
ultimately must make the decision regarding adequate design information. The engineer's goal is to
minimize risk to the Government. The engineer must consider the costs for additional investigation
activities versus the potential construction cost savings resulting from better project definition. This
decision is further complicated in hazardous waste remediations by pressure applied from the project
manager and community.
Acceptance of risk by the Government permits the engineer to produce a less-defined package and
use a non-conventional acquisition strategy. This acceptance by the Government often results in
expediting the start of construction activities. It does not guarantee the early completion of a project.
Circumstances at hazardous waste sites often make it uneconomical to investigate the site that would
allow the engineer to produce a well-defined project. In these cases the engineer must recognize this
limitation and modify the acquisition strategy accordingly. Assessing the potential Government risk
is key to these efforts.
Prior to issuing a remedial design assignment, the U.S. Environmental Protection Agency recommends
that the project manager develop predesign technical and remedial management summaries.(4) These
summaries are more thought process than formal documents. The summaries focuses the project
manager to address major components of the remedial design and remedial action. The predesign
technical summary deals with site information including availability of data, selected remedy,
technical approach, applicable or relevant and appropriate requirements (ARARs), health and safety
concerns, and any unresolved issues. The remedial management strategy focuses on the
implementation of remedial design and remedial action activities. This includes consideration of
phasing and/or expediting portions of the remedy. An acquisition strategy is an end product of the
remedial management strategy.
Discussion of the manner in which a project manager or engineer selects the various components will
begin by addressing the simplest type of hazardous waste remediation, the simple civil engineering
project. These projects differ little from any other Government civil engineering project. The
simplest remedies do not deal with hazardous waste. Examples might be alternate water supply
systems or installation of cap material over a site without disturbing the existing soils. Additional
field investigation is minimal and the project can be well-defined. Design work can commence with
the goal of developing a design specification for a fixed price contract procured through an invitation
for bid.
Complex civil engineering projects are actions such as contaminated soils excavation, slurry wall
construction, and building decontamination or dismantling. These project require additional field
investigations prior to commencement of design activities. Health and safety and quality control and
assurance plans become part of the contractor's required submittals. If the project can be
well-defined, the use of design specifications, fixed price contract and an invitation for bid is
preferred. Alternatives should be considered when the engineer cannot adequately define the project
due to unknowns or time constraints. These alternatives include performance specifications,
indefinite delivery or time and materials contracts, and negotiated procurements. Examples where
these may be appropriate would be: 1) expediting building decontamination or dismantling because
of potential risk or interest from the community, 2) expediting soils excavation where contaminants
are known but additional sampling would be required to determine the amount of material involved,
3) wanting to obtain recommendations from private industry for approaches to slurry wall
construction, 4) wanting to discuss proposed construction plan with offerers prior to contract award
when working in residential areas requiring good community coordination.
1037
-------
RECOMMENDED ACQUISITION STRATEGIES FOR
HAZARDOUS WASTE REMEDIATION
REMEDIATION CATEGORY
Specification
Procurement
Contract
SIMPLE CIVIL ENGINEERING
Design Invitation for Bid
Fixed Price
COMPLEX CIVIL ENGINEERING
Design
Performance
SIMPLE PUMP AND TREAT
Design
Two-Step Bid
Request for Proposal
Invitation for Bid
Fixed Price
Indefinite Quantity
Time and Materials
Fixed Price
COMPLEX PUMP AND TREAT
Design
Performance
Two-Step Bid
Request for Proposal
Fixed Price
Indefinite Quantity
Time and Materials
Cost Reimbursement
SIMPLE SOILS AND SLUDGE TREATMENT
Design Invitation for Bid
Fixed Price
COMPLEX SOILS AND SLUDGE TREATMENT
Design
Performance
Functional
Two-Step Bid
Request for Proposal
Fixed Price
Indefinite Quantity
Time and Materials
Cost Reimbursement
ON-SITE THERMAL DESTRUCTION
Performance Request for Proposal
Functional
Fixed Price
Indefinite Quantity
Time and Materials
Cost Reimbursement
1038
-------
Simple pump and treat remediations are those where the movement of the plume has been restricted
or the waste is easily treated. This permits the development of a solicitation package with minimal
additional field investigation and treatment technology is readily available. A package treatment plant
may be a viable option for this remedy. Specifications can be design-based and a fixed price contract
can be procured with a sealed bid. The use of alternative strategies would have minimal impact in
expediting this type of project.
More complex pump and treat remedies require greater flexibility. Additional well drilling may be
necessary to define the plume. To avoid the need for contract modification, renegotiation, or
resolicitation, an indefinite delivery on time and materials contract may be favored over a fixed price
contract.
Some ground water contaminants may require innovative technologies and treatabililty studies. This
work can be conducted by a research and development contract followed by design and construction
contracts. The alternative is to enter into a negotiated procurement and request that offerers
demonstrate that their proposals meet the project requirements. Specification used in the solicitation
would be performance- or perhaps functional-based. Considering treatment alternatives does not
preclude the use of a fixed price contract. A fixed price incentive contract provides the Government
a method to select a technology with cost uncertainties. If the contractor can improve the treatment
performance the cost savings are shared.
Simple soils and sludge treatment projects are those in which the extent of contamination has been
defined and the treatment technology is proven. Minimal additional field work is needed prior to
commencing design. Remediation activities can be easily defined in design- and performance-based
specifications. An indefinite delivery or time and materials contract can be used in lieu of a fixed
price contract if quantity definition is a problem.
Complex soils and sludge treatments and on-site thermal destruction require extensive design and
construction activities. These projects involve substantial risk to both the contractor and the
Government. A negotiated procurement gives the Government the opportunity to evaluate each
offerer's approach to the project. Factors the Government might consider include the technology,
work health and safety plan, involvement and protection of the community, quality control and
assurance measures, and previous experience in similar type work. This process allows the selection
of a contractor that best fits the Government's needs. Specifications for these remedies are
performance- and functional-based allowing the Government to consider a range of approaches to
the remediation. Quantity definition is often a problem. The amount of data collection needed for
design is dependent on the Government's willingness to share in the risk allocation. With known
contaminants and technologies, remediation can be expedited through the use of an indefinite delivery
or time and materials contract. When considering new and innovative technologies, the Government
may wish to further share in the risk and use a cost reimbursement contract. Cost reimbursement
contracts require substantial Government contract management. The advantage of their use in new
and innovative technologies is that they provide a means for the Government to enhance its
knowledge base. Costs involved are actual, not those determined by a contractor trying to consider
all possible contingencies during the solicitation. Modification to the technology during the contract
is easier to accomplish with a cost reimbursement contract.
CONCLUSION
All hazardous waste remediations do not fit the traditional Government construction project mold.
It is often difficult to adequately determine quantities or inappropriate to provide a detailed design.
The Government has many options available within the limits of the Federal Acquisition Regulations
to select an acquisition strategy for hazardous waste remediations. The key to success is the early
1039
-------
selection of a strategy based on the amount of risk the Government wishes to accept. Risk transferred
to the contractor impacts directly on the addition of contingency to the offered price.
The fixed price, invitation for bid, design specification approach is acceptable for simple hazardous
waste projects . As complexity increases, however, this approach becomes undesirable. Procurement
of a complex hazardous waste project must be flexible. The Government and the contractor must
have a clear understanding of their responsibilities at a complex remediation if the project is to be
successful.
A better approach is the evaluation of each hazardous waste project early in its development and the
selection of the proper contract, procurement and specification type. The goal of this procedure
should be to select the method that best balances the needs of the Government with the risk of the
contractor. There is no reason the Government and the contractors cannot be partners rather than
adversaries. Contractor contingency can be reduced by making them feel they are partners in the
project, by reducing the risk through negotiation of complex projects, and by sharing risk.
DISCLAIMER
This report has undergone a relatively broad initial, but not formal, U.S. Environmental Protection
Agency peer review. Therefore it does not necessarily reflect the views or policies of the Agency.
It does not constitute any rulemaking, policy or guidance by the Agency, and cannot be relied upon
to create a substantive or procedural right enforceable by any party. Neither the United States
Government nor any of its employees, contractors, subcontractors or their employees makes any
warranty, expressed or implied, or assumes any legal liability or responsibility for any third party's
use or the results of such use of any information or procedure disclosed in this report, or represents
that its use by such third party would not infringe on privately owned rights.
We encourage your comments on the utility of this paper and how it might be improved to better
serve the Superfund program's needs. Comments may be forwarded to the attention of Kenneth W.
Ayers, Design and Construction Management Branch, U.S. Environmental Protection Agency,
Mailcode OS-220 W, Washington, D.C., 20460.
REFERENCES
1. Cibinic, John Jr. and Nash John Jr., Administration of Government Contracts (Second
Edition, Second Printing), Government Contracts Program, George Washington University,
Washington, D.C., 1986.
2. Cibinic, John Jr. and Nash John Jr., Formation of Government Contracts (Second Edition),
Government Contracts Program, George Washington University, Washington, D.C., 1986.
3. Taylor, John C., "Tradeoff Analysis in Negotiated Procurement Procedures for Construction
(Are the Additional Points Worth the Additional Dollars)", Unpublished paper, United States
Environmental Protection Agency, Washington, D.C., 1988.
4. United States Environmental Protection Agency, Guidance on Expediting Remedial Design
and Remedial Action, Office of Solid Waste and Emergency Response Directive 9355.5-02,
Office of Emergency and remedial Response, Washington, D.C., 1990.
5. United States Navy, Student Guide for Construction Contract Administration and
Management, Naval School, Civil Engineer Corps Officers, Port Hueneme, California, 1988.
1040
-------
VII. PRE DESIGN ISSUES
1041
-------
The Importance of Pre-Design Studies
In Superfund Remediation
Jeffrey R. Bennett
Malcolm Pirnie, Inc.
2 Corporate Park Drive
White Plains, NY 10602
(914) 694-2100
William McCabe
U.S. Environmental Protection Agency
Region II
26 Federal Plaza
Room 29102
New York, NY 10278
(212) 264-0276
Richard W. McCollum
Chief, Superfund Section
U.S. Army Corps of Engineers
Kansas City District
601 East 12th Street
Kansas City, MO 64106-2896
(816) 426-5332
INTRODUCTION
Prior to initiation of the remedial design for the Marathon Battery Superfund Project, a number of
complex design and construction issues had to be resolved. This paper focuses on the importance of
studies performed after the RI/FS and the ROD were finalized to resolve issues vital to completion
of the final remedial design. These included:
• A supplemental sampling program to better define and delineate areas of contamination
• Application of geostatistical analysis methods to evaluate the accuracy and significance of the
sampling data and provide a rational, scientific basis for delineating areas for remediation
• Evaluation of the technical feasibility and development of specifications for solidifica-
tion/fixation of dredged materials
• Archeological investigations to determine the historical significance of areas to be impacted
during construction and identify those areas requiring mitigation, and
• Comparison of transport options for movement of construction materials and stabilized waste.
The results of these pre-design efforts substantially affected final design by reducing the proposed
remediation zones, developing a balanced transportation plan which will reduce both project costs and
local impacts of construction, cataloguing methods and areas for mitigating the impact of construction
1042
-------
on potentially significant archeological sites, and identifying areas requiring only partial mitigation.
The effectiveness of these studies in achieving their stated goals at the Marathon site point up the
need to assess the adequacy of existing data and evaluating practical alternatives to achieving ROD
objectives. Since many of these issues often can not be resolved within the relatively short time
frames and limited budgets characteristic of the RI/FS Process, design engineers and regulatory
officials must be aware of the critical role additional pre-design studies can play in developing a cost-
effective remedial design for a hazardous waste site.
BACKGROUND
Project Location and Description - The Marathon Battery Superfund Site lies on the east bank of the
Hudson River, across from West Point and approximately 40 miles north of New York City in Putnam
County. The portion of the site addressed in this paper includes part of the river near Cold Spring,
NY, and the three wetland areas of East and West Foundry Coves and East Foundry Cove Marsh.
Water flows from the Hudson River through a 70-foot passage under a railroad trestle into the 30-
acre East Foundry Cove, which is an active fisheries spawning area. Flow then proceeds through a
channel and dike system that connects Foundry Cove to Constitution Marsh, a sensitive 281-acre
wildlife sanctuary operated by the Audubon Society. The residential and business districts of the
town of Cold Spring lie close by to the north, and the site also includes a portion of a significant na-
tional historic site, the Old Foundry, to the east. (Figure 1)
Between 1952 and 1979, when the facility became inactive, the Marathon Battery plant produced
nickel-cadmium batteries, and released treated and untreated production wastewater into East
Foundry Cove Marsh and the Hudson River. Concentrations of cadmium, nickel and cobalt in the
marsh sediment at the outfall reached levels as high as 171,000 ppm, 156,000 and 6000 ppm, respec-
tively, with concentrations gradually decreasing in East Foundry Cove and Constitution Marsh. Heavy
metal wastes are concentrated primarily in surface sediments from 0 to 14 inches deep. The
contaminated areas at the site, both exposed and submerged, are affected by 0 to 40 feet of tidally-
influenced water. The amount of sediment-bound cadmium was estimated in the RI/FS at 50 metric
tons.
To allow for widely varying environmental features and pollutant levels, the site was subdivided into
three operable units: Area I consists of East Foundry Cove Marsh and Constitution Marsh, Area II
encompasses the former manufacturing site and surrounding grounds, and Area III includes East and
West Foundry Coves and the Hudson River near the Cold Spring pier. East Foundry Cove Marsh in
Area I is partially isolated from West Foundry Cove and the Hudson River (Area III) by a railroad
bed to the west. Remediation of the adjacent Areas I and HI will be implemented concurrently.
In-depth (Stages 1 A, IB and 2) archaeological studies were also conducted near proposed staging and
treatment areas. The project's environmental and archaeological concerns, along with the impacts of
engineering alternatives and the cost of remediation dictated the application of value engineering
(VE) studies. In what is believed to be the first application of VE to a Superfund site, the Area I
remedial site design was optimized for a potential multi-million dollar savings in construction costs.
A comparable value engineering review of Area III will also be implemented.
Project Responsibility - Remediation overview of the site is being directed by the U.S. Environmen-
tal Protection Agency (Region II), with project management delegated to the Kansas City District of
the Corps of Engineers under an interagency agreement. After completion of the RI/FS by others
and issuance of the Record Of Decision (ROD), Malcolm Pirnie, Inc., a consulting environmental
1043
-------
engineering firm based in White Plains, NY, was contracted by the Corps to complete pre-design
investigations and remedial designs for Areas I and III, along with the Area II plant site.
Pre-Design Recommendations - The ROD recommended that in Area I earthen dikes be constructed
around East Foundry Cove Marsh to isolate it from the tidal flushing of the Hudson River. The diked
marsh would then be flooded and hydraulically dredged, with contaminated sediment pumped to large
settling tanks for mechanical thickening. The VE evaluation of the project subsequently produced
a recommendation which eliminated the need for flooding the marsh and mechanical thickening, and
included diking, dewatering and mechanically excavating the marsh, and stockpiling and treating the
excavated sediment.
The remediation outlined in the ROD for Area III included dredging one foot of contaminated
sediment from East Foundry Cove and the Cold Spring Pier area, constructing a series of lagoons for
dewatering the dredged materials, solidification/stabilization so heavy metals could not leach to the
environment and final disposal of the fixated product in an off-site landfill. Initial pre-design studies,
along with the post-ROD decision by the sponsoring agencies to remediate Areas I and III
concurrently, led to serious consideration of a revised plan for Area III which included hydraulically
isolating East Foundry Cove from the river and Constitution Marsh, hydraulically dredging
contaminated sediment from the cove and pond to the diked East Foundry Cove Marsh (in effect,
using these diked marsh areas as dewatering and sedimentation basins), excavating the cove and pond
dredge spoils from the diked area, treating contaminated sediments and water from the dewatering
area, and off-site landfilling of the fixated product. Contaminated sediments in the pier area would
be handled similarly.
DISCUSSION
To prepare for the RD phase, several predesign investigations were initiated which had dramatic
effects on the ultimate remedial design of the site.
• Sampling - At a site the size of Marathon Battery, the soil and sediment sampling program
is vital for assessing the specific area and depth of contamination, as well as for defining the volume
of contaminated material to be removed, treated and transported. In previous studies, sampling was
used to delineate broad areas of contamination. However, these efforts produced only general levels
of contamination in generalized areas and lacked clarity in terms of the exact depth and lateral extent
of required remediation. Also, these data could not be verified because sampling sites were not
accurately located. A comprehensive sampling program was developed to augment the RI/FS, which
included soil and sediment for Areas I and III, vegetation in Area I and soil in Area II.
To insure accuracy and allow for replicable results, sampling points were located on gridded
site maps and either staked or located electronically, depending on the nature of the site. All samples
were analyzed by a Corps-validated independent lab for cadmium, nickel and cobalt, which were used
in the battery plant, and lead, which had been found in previous sediments samples and was thought
to originate from remnants of the Civil War-era foundry within the site. Cadmium was selected as
the key indicator for development of cleanup plans since that was the most toxic metal, as well as the
metal present in the consistently highest concentrations.
In Area I, 324 sediment samples were collected from 61 staked sampling points located on a
100-foot grid of the 14-acre marsh. (Figure 2) Since the area's dense vegetation — including roots
ranging from very fibrous to large and yam-like — precluded obtaining continuous samples, six-inch
samples were retrieved from depths up to three feet using a custom-made stainless steel hand-
1044
-------
operated corer. Cadmium contamination levels ranged as high as 91,700 mg/kg, with most of the
marsh in the hundreds of ppm range. Using these sampling data, excavation plans were initially
developed requiring removal of from one to four feet of sediment, depending on the proximity to the
tidal channel and outfall, with the deepest excavation to be adjacent to the tidal channel.
In Area III, near the Cold Spring pier, the river area with the highest suspected level of
contamination, gridded sampling points were located through electronic positioning to a tenth of a
foot. A total of 111 samples were taken initially from 49 river locations (on a 150-foot grid within
300 feet from shore) at depths from 0 to 24 inches around the pier, as well as grab samples from six
locations beneath the pier, with four additional samples taken at the adjacent beach. The 150-foot
grid was expanded to a 300-foot grid away from the pier.
In East Foundry Cove and Pond, 67 samples were taken from the 0 to 24 inch depth at 39
staked positions, including 30 sampling locations in East Foundry Cove and 9 locations in East
Foundry Pond. These were located with an electronic positioning system based on a 250-foot grid.
In West Foundry Cove, 26 samples were taken.
Two-phased depositional studies were also implemented for West Foundry Cove to assess the
ROD's findings regarding the depositional nature of the Cove. In the first phase, samples from West
Foundry Cove were analyzed for cesium as well as cadmium and lead to date the sediment and
confirm the depositional characteristics of certain key sub-areas. This was critical in determining
whether any remaining contamination would migrate or remain stationary. [A by-product of
aboveground testing of nuclear devices, cesium concentrations can be correlated with post-1953
testing activities to date materials.] The second phase of the study, scheduled to begin shortly,
involves contaminant flux studies to determine if contamination from West Foundry Cove might be
transferred to East Foundry Cove after remediation. While observations determined that some mixing
does occur, the net effect was confirmed as depositional.
Cadmium contamination was concentrated primarily in the top four inches of sediment in East
Foundry Cove and East Foundry Pond, with cadmium levels in the upper sediment layer of the pond
reaching 3520 ppm. Of the locations tested below 12 inch depth in East Foundry Cove, only one in
the southern part of the Cove had a detectable amount of cadmium (875 ppm), compared with 145
ppm in the overlying (0 to 4 inch) interval at this location. In East Foundry Pond, only two locations
tested below 12 inch depth had detectable amounts of cadmium, which correspond with the highest
measured concentrations from 0 to 4 inch depth. (Figure 3)
Except for two locations adjacent to the Pier and within the boat club marina where
concentrations increased with depth, the Cold Spring pier area also revealed decreasing trends of
cadmium with depth. In addition, 21 sampling locations near the Cold Spring pier area from 0 to 4
inch depth and 35 locations from 12 to 24 inch depth had undetectable cadmium concentrations. The
highest levels of contamination was found in quiescent zones adjacent to the pier, although the pier
structure itself had low to medium concentrations of cadmium. While only low (38.8 ppm) levels of
contamination were found on the beach at from 0 to 12 inch depth, the high degree of direct human
exposure does warrant excavation of contaminated material in this area.
• Geostatistical Analysis - Since the goal of the sampling effort at the Marathon Battery site
was to determine the concentration and distribution of heavy metals on the site, geostatistical analysis,
a technique originally developed to evaluate concentrations and distribution of ore deposits for the
mining industry, was implemented. Geostatistical analysis used the sampling data to model the spatial
relationships of the levels and extent of cadmium contamination. It was also expected to offset the
1045
-------
variation in sampling data expected from non-static areas such as East Foundry Cove and the Hudson
River in the vicinity of the Cold Spring Pier, where tidal flows may alter sampling results.
This type of statistical modeling of contamination concentrations is relatively new to the
environmental field. The model used for the Marathon Battery project, the INSITE System Version
4.0. by Geostat Systems International, Inc., is recognized by regulatory agencies and health depart-
ments across the country. It provides a broader overview of sampling results than similar programs
by producing concentration contours as well as statistical confidence levels, providing more control
over the statistical parameters that are applied.
Accurate geostatistical modeling has several important requirements and stages. It is vital to
use known accurate sampling coordinates, and samples must undergo stringent quality assurance to
insure the accuracy and viability of data. After meeting these two requirements, a basic data analysis
of the sampling results from the Marathon site was performed to preview general results and trends.
The complete data base was then used to produce a variogram or graphic display of the level of error
of the spatial properties and qualities of data. The variogram was used to fine tune the modeling
program by graphing the expected variance in error of a projected value over distance and direction,
and exploring, checking and validating assumptions regarding site hydrology and contaminant
transport mechanisms. The results of the variogram analysis were first compared and contrasted to
those of the exploratory data analysis and to baseline data, and then directly incorporated into the
geostatistical estimation process of "kriging," which employs geostatistical computation methods to
determine the variance of a group of data points as a function of distance from those points.
Used for parameters exhibiting spatial correlation, kriging incorporates the quantification of
the correlation structure by the variogram to estimate the values of parameters at unsampled locations
and to calculate the corresponding estimation variance for the interpolated values. The variance is
a quantification of the lack of data supplying the unknown parameter values. Simply put, kriging
uses a search ellipse drawn between three to six samples to create 'working averages' for locations
between those sampling locations and to develop levels of confidence for these estimates. To orient
results to the physical characteristics of the site, the extrapolated data is then plotted as blocks or con-
tours. (Figure 4) As output, each block in the representational model contains a kriged estimate of
contaminant concentration and a measure of the kriging error of estimation. Using this graphic
display along with the known level of error, one can determine those areas with relatively high
concentrations of contamination, and can evaluate the cost impacts of removing different levels of
contamination.
From the sampling data collected at the Marathon site, the geostatistical model kriged
concentrations of cadmium (in ppm) between the actual sampling points for every 50 feet in five
levels of concentration: 0-5 ppm, 5-10 ppm, 10-50 ppm, 50-100 ppm, and 100-1000 ppm, graphing
each level in a different color as well as determining estimated levels of error for these calculations.
In the kriging process, an estimate with a relatively high level of error not consistent with
other sampling findings could drive the model falsely. As a result, it would require clarification
either from existing data or through additional sampling to determine whether the high reading was
a true hot spot of contamination or an anomaly. (This potential requirement for additional data also
points up the importance of strict locational control of all sampling to verify or deny the data
necessary to redirect the geostatistical model.)
After the first kriging, additional sampling was conducted to explain isolated high values (with
an error of greater than 2.0) which were an order of magnitude higher than adjacent points. Since
1046
-------
these sampling data could be driving the statistical model to indicate a larger than necessary area of
contamination, the cost of remediating the larger areas far outweighed the expense of additional
sampling.
The supplemental sampling was extremely effective: 10 additional samples produced 27 new
blocks of estimated concentrations. More importantly, the new data and estimates showed drastically
reduced concentrations of blocks registering above 100 ppm in the pier area (which subsequently
increased the number of blocks in the 50-100 ppm category). (Figure 5)
By using the kriging methodology it was possible to employ a larger, 'coarser', sampling grid
in the initial sampling effort. 'Questionable' or 'problem' areas could then be identified via kriging,
and resolved via limited, focused additional sampling. The net result was a program that saved
significant time and expense for field and analytical efforts and developed critical contamination data
far more efficiently.
Determining the level of remediation to be accomplished in Area III (cove and river
sediments) was complicated by the fact that the ROD gave no specific numerical contaminant
thresholds to be remediated, but, rather, specified that the site be remediated to a depth of 12 inches
which would expect to remove "up to 95 percent" of the existing contamination. With no absolute
definition of cleanup levels, the geostatistical analysis provided additional guidance in determining
the areas and amounts of remediation. As a result, the design team had a rational, logical and
replicable data base that provided a firm scientific basis for establishing cleanup standards.
According to the ROD, one foot of sediment was to be dredged from East Foundry Cove.
Interpretation of the krig contours indicated cadmium concentrations generally decreased as one nears
the water channels in the East Foundry Cove, with measured values of cadmium virtually
undetectable in these channels.
These data suggested that cadmium is concentrated in the central depositional area of the Cove
and scoured from areas subject to erosion during rising and falling tides. This depositional theory
was confirmed by separately kriging the bathymetric data for specific locations in Area III (performed
along tracklines spaced at 50 foot intervals) and the sampling data, and then overlaying the two.
Kriging the sampling data showed that the increased depth of the subsurface depressions
corresponded to lower cadmium levels. The correlation of the two krigs confirmed the scouring
action of tidal flow and water movement, and the concentration of contamination in the delta. As
a result of these studies, it now appears feasible to limit sediment removal activities to those specific
portions of East Foundry Cove with the highest measured levels of contamination as opposed to the
recommendation in the ROD of remediating the entire Cove. This will result in cost savings of up
to SW-million by eliminating 16,000 cubic yards of sediment from the remediation process. Thus, a
small investment in additional studies yielded very large returns.
• Archaeology - Since the Marathon Battery site includes the West Point Foundry, a national
historical site dating from the Civil War, extensive Stage 1A and IB cultural resource surveys and a
Stage 2 archeological field investigation were conducted at areas which could be impacted by remedial
activities.
The hazardous nature of the site coupled with tight schedule requirements necessitated
alteration of standard archeological practices. Temporary protective enclosures were erected so the
archeological work could continue through severe Northeastern winter weather. (Figure 6) And
because the archeological investigations were being performed at a hazardous waste site, all on-site
1047
-------
project personnel people were required to complete health and safety training. Handling methods also
needed to be modified.
To minimize exposure of the archaeologists to the hazardous environment and to the winter
elements, advanced Rolliometric photographic techniques were used to document excavated areas.
(Figure 7) This customized highly calibrated, computer coordinated photographic method accurately
documents physical features to scale, and eliminates the traditional method of scale drawings by hand
and physical measurements of strata and subsurface structures typically completed at the archeological
site. Instead, this specialized photographic method produces photos and maps showing the sizes and
locations of subsurface archeological sites that can be evaluated off-site with a scale ruler to
determine their relative relationships, size and locations.
Three areas of archeological importance have been investigated to date. Initial field studies
were completed under the aegis of the Corps of Engineers, and more detailed investigations were
subsequently funded under the U.S. EPA's Alternative Remedial Contract Strategy (ARCS) program.
These investigations have added significant new information to the history and development of the
National Historic foundry site, including valuable data on metallurgy in the later 1800's, and provided
new insights into the industrial and technological makeup of the foundry as a critical Civil War
defense establishment.
The results of investigations in the foundry worker housing area, which overlaps the site of
the proposed haul road, radically changed the existing perception of the ethno/economic status of
workers at the old foundry. Based on artifacts and toys found at the site, archaeologists concluded
that educated upper class European skilled workers were present among the work force along with
the unskilled English and Irish workers previously known to have been employed. The foundry
proofing area, the site of the proposed equipment staging and waste stabilization areas, was discovered
to be the site of a unique gun testing platform. Covered by four feet of post-Civil War/20th century
industrial fill, the platform is the only known testing area in existence that was used for proofing the
famous "Parrott Cannon" used during the Civil War. This site will require some degree of avoidance
or protection during remediation. Additional studies used remote sensing surveys with electronic
instrumentation as well as a historic records search to investigate the cove and river near the pier.
Early records as well as the on-site survey and magnetic anomalies revealed evidence of what may
be sunken ships or barges.
The result of these archeological studies could have wide-ranging impacts on the remedial
design. While no remedial action is expected in West Foundry Cove and remediation in the pier area
is not expected to impact historical resources, historical sites in East Foundry Cove may impact the
schedule for remediation and require action prior to dredging, with further measures such as
avoidance or unearthing the site currently under investigation.
• Solidification/Stabilization - The ROD specified stabilizing/fixating the treated material to
tie up heavy metals to allow safe disposal of the material. Separate studies were conducted in Areas
I and III to determine the technical feasibility of solidification/stabilization on sediments from zthe
site. (Figure 8)
In Area I, bench-scale and pilot plant treatability studies, with the latter organized as
demonstrations by five pre-qualified vendors, tested two types of fixation/solidification processes
on representative Area I samples of contaminated marsh sediment with known cadmium concentra-
tions. The vendors used between four and six cubic feet of highly contaminated marsh sediment
(average concentrations: 4,700 mg/kg) in the on-site demonstrations to produce either a solid concrete
1048
-------
product or a soil-like soft product. After both 7- and 28-day curing periods, samples were sent to
a Corps-certified laboratory for Toxicity Characteristic Leaching Procedure (TCLP) toxicity testing.
After a total 60-day curing period, additional samples underwent toxicity and long-term biodegrada-
tion testing to evaluate the long-term stability of the treated material and its potential for heavy
metals leaching from the formed matrix in light of the substantial quantities of large organic material
in the marsh sediment, including yam-sized roots from cattails. All five vendors were able to produce
a non-hazardous substance after the initial 7-day curing, although the 'soft' product failed the criteria
for classification as a non-hazardous substance after the 28-day curing period.
After it was determined that solidification could be successfully achieved to adequately bind
heavy metals for treated material from Area I to pass TCLP testing, additional studies were initiated
using sediment from Area III to develop an optimum generic solidification/fixation process that
would be incorporated into bid specifications to stipulate the required weight and volume of the end
product, and to help develop basic design parameters for other parts of the remedial design. This
generic formula would provide essential design data on the amount of dewatering required, the ability
of the stabilization process to solidify a wide range of sediments, the volume of resulting solidified
material expected, and the optimum fixating agents.
Since cadmium exhibits varying degrees of adsorption based on particle size which would
affect the solidification formulations, the two distinct types of sediment from Area III were utilized:
"coarse" (with more than 90 percent sand), and "fine," which constituted the majority of the sediment
(90 percent or more silt and clay). The laboratory experiments used five gallon sediment samples
from three site areas — the pier area, East Foundry Cove and East Foundry Pond. Prior to initiating
laboratory work to develop the generic solidification formula and determining the total volume of
solidified material, questions were raised as to whether all the sediment would require solidification
as hazardous waste. As a result, untreated samples were analyzed for total cadmium and lead, and
were also subjected to TCLP analysis. After graphing in-situ cadmium concentration against TCLP
leachate values from the samples, it was determined that the fine sediments tended to "hold" cadmium
and lead and exhibited much lower TCLP results than coarse sediments. As a result, it is likely that
substantial amounts of sediment from the pier area, pond and cove will not be classified as hazardous
waste and will require only dewatering and land filling instead of the more costly solidification/fix-
ation. Actual amounts will depend on the mixing and dilution that occurs, and can only be
determined during actual dredging.
In East Foundry Cove and Pond, various ratios of sediment were tested with six fixation
additives using different levels of water contents to develop optimum generic formula. Findings from
this study are currently being expanded using one cubic yard of a blended mix of sediment from the
three areas to test the selected formula on a larger scale. This fixation formula will then be used on
all dredged stockpiled sediment materials which fail to meet TCLP restrictions.
An additional study is currently evaluating the potential for thermal reduction of marsh
sediment, which is characterized by heavily contaminated organic peat moss. The economic and
technical feasibility of mobile on-site incineration is being evaluated for its potential to reduce the
peat content and ultimately the volume of material to be treated and disposed. This testing is ex-
tremely site-specific and could also have major impacts of remedial design and final project cost.
• Transport Options - With approximately 150,000 cubic yards of dredged material to be trans-
ported, materials handling is a major issue that must be addressed during design. Remediation will
require transport of quantities of clean sand and gravel as well as heavy equipment into the site, and
the transport of fixated dredged material to an ultimate disposal site. In an initial pre-design
1049
-------
transportation study, several alternatives were evaluated to determine the most efficient method of
transporting large quantities of materials, personnel and equipment to and from the site.
Truck transport along local roads and/or a haul road, water and rail transport were evaluated
for cost, efficiency and impact on the local population and infrastructure. The Marathon site has no
easy public access. The narrow local roads built in the early 1800's pass through the Village of Cold
Spring's Historic District, known for its 150-year old buildings, antique shops and tourist attractions.
Since prolonged truck traffic could affect the structural integrity of these buildings, and would
severely restrict local traffic throughout the work day along with raising sensitive public concerns and
giving the project unwanted visibility, the local road option was eliminated.
Water traffic via barges was eliminated since barge entry to the cove was restricted by the
railroad crossing and this alternative would limit options for ultimate disposal of the fixated material
or mandate double handling for transport to the final disposal site.
The rail spur option, the easiest to construct from an engineering aspect, could link to a
commuter/freight line which runs adjacent to the site. Built directly on an old rail bed used in the
1800's, the spur could be used for transport of fixated material. (Figure 9) However, further study
determined that rail transport alone would greatly restrict site access since equipment shipments would
merit only "secondary" scheduling priority, and could not be delivered on an as-needed basis. In
addition, rail transport of new construction materials (i.e., earth, gravel, etc.) was found to require
specialized facilities for unloading. This construction was determined to be difficult and costly due
to the constricted site and numerous archeological concerns.
The use of a separate haul road would remove truck traffic from the center of town but may
also require additional archeological mitigation since it passes through a portion of the Old Foundry
historical site. In addition to these archeological concerns, the local geography of steep embankments
would require terracing and switchbacks to reduce the grade and improve stability. Despite these
requirements, extensive economic analysis points to the haul road as the favored option. Construction
priorities are now being evaluated to determine the measures necessary to mitigate impacts on
archeological areas; an archeological data recovery program appears likely.
CONCLUSION
The final remedial design for the Marathon Battery Site is expected to be completed during
the summer of 1991. In completing the design for this complex site, pre-design studies have been
vital in redefining the area and type of remediation as well as in identifying other steps critical to
successful completion of the project.
In undertaking future remedial designs, it is important to remember that the process of
developing an RI/FS and ROD for a site deals with "big picture" issues. It cannot be expected to
answer all of the questions which must be addressed before a detailed remedial design can be
completed. The experience we have discussed in this paper makes a strong case for planning adequate
time and budget for performance of additional site investigations in preparation for the remedial
design. These additional investigations can pay big dividends by defining the actual contamination
in a way that the RI/FS typically does not, and by exploring viable alternatives for achieving the
intent of the ROD. The result is a more complete, practical and cost effective remediation.
1050
-------
WEST POINT FOUNDRY
HISTORIC DISTRICT
FORMER
BATTERY
PLANT
FAIR LAWN
HISTORIC DISTRICT
SPRING
HISTORIC DISTRICT
EAST FOUNDRY
COVE MARSH
COLD SPRING
BEACH
COLD SPRING
WEST FOUNDRY
COVE
CONSTITUTION
MARSH
EAST FOUNDRY
COVE
CONSTITUTION
ISLAND
LEGEND
AREAS OF CONTAMINATION
FIGURE 1
The Marathon Battery site lies on the Hudson
River adjacent to Cold Spring, New York.
Contaminated areas are shaded.
1051
-------
FIGURE 2
Supplemental pre-design sampling points in East
Foundry Cove Marsh were located on a 100 foot
grid.
1052
-------
FIGURE 3
Targeted sediment sampling better defined
contaminated areas and significantly reduced
remediation in East Foundry Cove.
1053
-------
•a
c
ea
if
, = s
rf\ r\ 00
S a -s
P
o ,
-------
V
ZONE V
ZONE
1
11
III
IV
V
VI
AREA
(SO FT)
737,500
132,500
42,500
287,500
42,500
107,500
IN SITU VOLUME
OF SEDIMENTS
FOR 1 FOOT DEPTH
(CU YDS)
27,300
4,900
1,600
10.600
1,600
4,000
50,000 cu yds
100 0 100 200
SCALE IN FEET
FIGURE 5
The supplemental sampling program revealed
distinct zones of contamination in East Foundry
Cove.
1055
-------
FIGURE 6
Temporary protective enclosures allowed
archeologists to work in severe winter weather.
1056
-------
FIGURE 7
Advanced Rolliometric photography methods were
used at the site to document excavated areas.
1057
-------
oo
W
{*
D
o
*-t
CU
OC
O
o
10
in
fe c
f O
00 "M
C CO
'*-" .—
co £
*-• -13
41
o
1)
CO £
4_> <>9
o ^
«!
tC 'U
U "»3
I2,
£ o> 2J
'« u "O
S O < !u 8
W a. a,
1058
-------
FIGURE 9
The proposed design incorporates areas for
dewatering and fixation of dredged material along
with a rail siding for off-site transport.
1059
-------
RI/FS and ERA Impacts on
RD/RA at Superfund Sites
William J. Bolen
U.S. Environmental Protection Agency
230 South Dearborn Street
Chicago, Illinois 60604
(312) 353-6316
James A. Werling, Jr.
ICF Kaiser Engineers
Robinson Plaza 2 Suite 200
Pittsburgh, Pennsylvania 15205
(412) 788-9200
Earl H. Brown, Jr.
ICF Kaiser Engineers
Robinson Plaza 2 Suite 200
Pittsburgh, Pennsylvania 15205
(412) 788-9200
INTRODUCTION
This paper discusses two concepts used during the Remedial Investigation/ Feasibility Study (RI/FS)
process to expedite the Remedial Design/Remedial Action (RD/RA) at a National Priorities List
(NPL) site: the interactive or phased RI/FS and the Expedited Response Action (ERA).
The site, located in the City of Columbus, Bartholomew County in central Indiana, was previously
a small metal plating operation. Existing data collected by the State health department and the
Indiana Department of Environmental Management (IDEM) indicated that soil near the site had been
contaminated with various metals and cyanide.
A fast tracked approach for site remediation was facilitated by successful use of the interactive
process to guide the RI, risk assessment, and FS phases of the project. The current National
Contingency Plan (NCP) and Recent USEPA guidance (EPA 1988) introduced the interactive or
phased methodology as a dynamic and flexible process that enables an effective RI/FS to be
performed in a timely and cost efficient manner. The premise of the phased approach is that the level
of investigation and analysis required in an RI/FS is determined by constant adjustment of
investigation goals as new data is obtained. The goal is to collect sufficient data to support an
informed decision regarding which remedy appears to most appropriate for the site - not the
unobtainable objective of eliminating all uncertainty associated with decision making. Although
explicit in the current NCP and EPA RI/FS guidance, experience has shown that interactive approach
is sometimes employed ineffectively due to a reluctance to deal with uncertainty. This paper
demonstrates that the interactive RI/FS process is a sound approach to conducting and RI/FS.
Key to the interactive process is the early formulation of remedial objectives. The remedial
objectives formed through the interactive approach permits innovative methods, such as the ERA,
to be employed during an RI/FS to expedite a RD/RA. The ERA that was performed at the site
during implementation of the RI/FS, was ultimately consistent with the final remedy, greatly
streamlined the decision process in the FS and simplified implementation of the RD/RA.
1060
-------
BACKGROUND
Site History and Description
The Tri-State Plating Site is located at 1716 Keller Avenue in Columbus, Bartholomew County,
Indiana in a residential and small business area. The site encompasses an area of approximately 130
feet by 120 feet and formerly contained a main electroplating process building with two attached
sheds, a storage building located immediately northwest of the main building, and an open yard
approximately 60 feet by 100 feet adjacent to the north side of the main building. The site plan and
onsite monitoring well locations are shown on Figure 1. The area surrounding the site and off-site
monitoring wells are shown in Figure 2.
Metal plating operations were performed at the site for over 40 years until it was closed in May, 1984.
The plating activities consisted of three major phases. In the first phase, the item was cleaned with
an alkaline cleaning solution and wire brushes. This process occurred primarily outside the buildings
in the north yard, with the spent solutions being dumped onto the ground. After cleaning, the item
was rinsed in water, then hydrochloric acid, then water. This process occurred in several tanks
located inside the building. The second phase of the operation consisted of the item being placed in
a vat containing a nickel solution, followed by another rinse, then placed into a "black" chrome
solution vat. The solution in this phase also contained cyanide, cadmium, copper, lead and other
metals. Finally the third phase consisted of sanding, polishing, inspection and boxing for shipment.
During the plating operations chemicals from the solutions being used were allowed to drip on the
floors while transferring between tanks. The floors were subsequently washed down and the
chemicals were discharged to the local combined storm/sanitary sewer system. These discharges
finally resulted in a "shock" load to the local POTW that interrupted the biological treatment processes,
and ultimately led to the plating operations being closed permanently.
Numerous site investigations were performed prior to the RI, to collect site specific data to
characterizing contamination at the site. The major compounds found in the soils were detected at
maximum concentrations as follows:
Pre-RI Data Phase I/II RI Data
Cadmium l,600mg/kg 56.7 mg/kg
Chromium (total) 52,000 mg/kg 16,400 mg/kg
Copper 7,200 mg/kg 80 mg/kg
Lead 170,000 mg/kg 156 mg/kg
Nickel 2,400 mg/kg 46.3 mg/kg
Cyanide 300 mg/kg 46 mg/kg
Geology
The City of Columbus is situated on the southwestern flank of the Cincinnati Arch and the
northeastern rim of the Illinois Basin (Hill 1981). This location is characterized by Mississippian and
Devonian sedimentary rock formations which dip gently westward. Beneath the study area,
1061
-------
_l
2 <^ !
\ o sS •
« 8 i* i
s is •
" £ si '
UJ £^ i
t 8 50
« Q- yz
« —
§ f 1 s *
1 1 " * (
/
SIDEWALK ^
/
r
_j
o
-•*
1 <
i U
U
Z ,
U
U.
U
»- ,
"
• ^ ^P'?:-, " ";. .-TT) ;.;::< >;• . lr '
1 I < l^';"w^^^:i^:^?
5j P w'^ >::;ii_P"^rn4fi ~i-f~v' -^TC^^
Ji 1 « F'h "^3U--T^% G4^'iT>^--V
' > 3 r ".*-.. «T^.—? I iV^I — Til. "£ " 4>J • . • xT^i
=1 o $, j?^ - 7j-slg^s3r ^ j£feLo~ " vi\ wsi
?S P fe| ^.'itzr-^^^S'^ilKr4 •^r--'' i VLX
• 1 ? 85 |j"-WtiM^--?l ^'-M^V
is « Si" H •" ji». n..-5l£aLSt=t:li^
- - =- ^-H •• { H**ida::3p333s3 •
. •!'•• ^^T" — S*-jrrtX_ rL*-.i 5 .- -
^ - , '«i — wf3.-?^ ^E j^-TKi-y t
» I i " -^iJS?*ff 1
1 2 - _" 1 i- ..; ! ,1*^1 T i "* •. — 5
S J 5
D z « a
g fe 8" w |
U w O j
^ f S
C UJ ff U
i! i IS I
-I UJ _J UJ
S 5 - S S
< ??
" ^-^^ fflTT^s^ 1 15 '
^-^ ®"?^v «V,
\J
o
g
*)
£ u,
•—I o S
^" s» i
— oz 8 < » i
55 •> »
P ^^
^_ »-
^^"^ w
**^ w
I '
e :
* -' ! *""
I? i \ •
i 2 » >• \ •-
5 ®® Si * \ —^
0 — ' J 0 " I 1
^ SS I \ S J
r \ „ «— V - ,«=.
-------
5 lo
?§•
§la
>-
k.
8 s
n
§
85
U £ O
* t- »-
U. UJ Z
®
"3?
3 OE
-------
limestone and dolomite units of the Middle Devonian Muskatotuck Group occur at depths ranging
from 95 to 135 feet. East and west of the study area, the Upper Devonian New Albany <;hale is
encountered at shallower depths. The juxtaposition of the older Muskatotuck group exposed between
younger formations creates a wide bedrock valley beneath Columbus. The bedrock valley is filled
by stratified sand and gravel outwash deposits laid down during the Wisconsinian Glaciation. A
generalized geohydrologic section of the study area is presented in Figure 3.
Dolomite underlies the site at a depth of 119 feet. The outwash in the vicinity of the site consists of
alternating sequences of sand and gravelly sand. Thicknesses of individual layers range from a few
inches to a few feet. In the vicinity of the site, the outwash deposits are covered by 3 to 8 feet of
silty and clayey sediments of recent alluvial origin.
Major drainage systems in the area include the Driftwood River, Flatrock River and the East Fork
of the White River. Haw Creek, a tributary of the Flatrock
River flows northeast to southwest through the study area, passing approximately 500 feet easl of the
site.
Hydrology
Groundwater in the study area is unconfined and found approximately 20 feet below the ground
surface at an elevation of approximately 614 ft. above Mean Sea Level (MSL). Local hydraulic
gradients are flat ranging from 0.0015 ft/ft to 0.002 ft/ft. The aquifer is highly permeable with
hydraulic conductivities ranging from 1700 gpd/ft sq (8 x 10"2 cm/sec) to 5900 gpd/ft 2 (2.8 x 10"1
cm/sec) (Watkins and Heisel 1982). Despite the low gradients, rather high flow velocities ranging
from 1.6 to 3.2 feet per day occur in the study area.
The stratified sand and gravel deposits are an abundant source of potable water. The City of
Columbus secondary wellfield, consisting of nine wells drilled to 110 ft., is located 800 feet northeast
of the site. The secondary wellfield is usually pumped 500,000 gallons per day (gpd) during winter
months, 2,000,000 to 3,000,000 gpd during spring and 5,000,000 gpd during the summer. These rates
are on the increase due to the recent drought that has affected the midwest and other parts of the
country.
The aquifer is also used for industrial purposes. A significant amount of water is withdrawn from
the aquifer by several industries located 250 to 800 feet south of the site. The closest industry
withdraws approximately from 300 to 500 gpm about 260 days per year.
Under normal conditions, regional groundwater flow is influenced by the heavy pumping demands
from the public well field and industrial users. Near the site, influences from the well field are
minimal, however flow direction does appear to be influenced by an industrial well located south of
the site. A water table contour map, constructed from monitoring well and piezometer water level
measurements in April, 1990 is shown in Figure 4. Although not observed during the investigation,
it is expected that the natural regional flow when no pumping occurs is southeastward towards Haw
Creek.
DISCUSSION
A fast tracked approach for site remediation was facilitated by successful use of the interactive
process to guide the remedial investigation (RI), risk assessment, and feasibility study (FS) phases of
the project towards ultimate site remediation. Key to the success of the interactive approach
1064
-------
WEST
700
650-
600-
600
550-
500
k550
500
Verticol Dotum is Meon Seo Levtl
SOURCE! Watkins, F.A. and Heiset, J.E. Electrical • Analog •
Model Study of Water Resources of the Columbus Area. Bartholoniew
County, Indiana, U.S. Geological Survey and Indiana Department of
Natural Resources, Water-Supply Paper, 1981.
HORIZONTAL
••• I
I 2
SCALE IN MILES
VERTICAL
1065
50
SCALE IN FEET
100
FIGURE 3
GENERALIZED
GEOHYDROLOGIC
SECTION
TRI-STATE PLATING SITE
-------
* *• 2
555
2 5 c
J r a
ai - m
m
o
i
g
1066
w o o c w
55R-8
r " ce m
m o X u o
" c "
S =H
o •» » o
O ID*
Iff
-------
was the early establishment of preliminary remedial objectives. As site characterization progressed,
it was realized that early formulation of preliminary remedial objectives and cleanup action levels
could be supported by the initial data base. The preliminary remedial objectives developed through
the interactive process helped to overcome several severe data gaps and assisted in guiding site
characterization towards a selection and implementation of the final remedy. It also led to the
performance of an ERA. The following discussion summarizes each component of the project from
the aspect of how the results were interactively used to guide the next step toward the ultimate site
goal of groundwater remediation.
Phase I RI
The purpose of the Phase I RI was to characterize contamination in several media with the intention
that several exposure pathways would be evaluated. Phase I included the collection of building wipe
samples, surface and subsurface soil samples, sewer soil samples and groundwater samples. The
samples were analyzed for volatile organics, extractable organics, metals, cyanide and hexavalent
chromium. Based on the results of past IDEM groundwater sampling, it was initially assumed that
a groundwater contamination problem was probably not associated with the site. Therefore, the initial
RI groundwater investigation was minimized, developed basically to encompass the site with four
shallow wells (MW-1, 2, 3, and 4) to collect samples to confirm this assumption.
The Phase I groundwater investigation provided data from four newly installed wells plus two
additional industrial wells located south of the site. Water level data indicated that wells MW-1 and
MW-2 were upgradient and MW-3 and MW-4 were downgradient of the main process building. The
analytical data revealed groundwater contamination at the downgradient site boundary in MW-4. The
groundwater was contaminated primarily with metals, specifically high concentrations of chromium
in excess of Federal Maximum Contaminant Levels (MCLs). The total and hexavalent chromium
concentrations at MW-4 were 1,620 and 1,600 ug/1, respectively. MW-1 also had elevated levels of
chromium (total 46 ug/1, hexavalent 50 ug/1). Detectable levels of chromium were not found in MW-
3 or MW-2 on the east side of the site nor in the industrial wells sampled downgradient of the site.
Traces of phthalates were also detected but these low levels were attributed to typical laboratory
contamination from plastic laboratory equipment. Other organic compounds were not detected.
The Phase I results indicated that an additional phase of investigation was necessary to delineate the
extent of the groundwater problem. Prior to developing a Phase II work plan, it was decided that a
preliminary public health evaluation should be conducted to evaluate risks, so that the appropriate
extent of additional investigations be determined. This preliminary PHE was performed during the
Phase II Work Plan development.
Preliminary PHE
The preliminary PHE examined risks in several exposure pathways, including direct contact with
building interiors and soils, incidental ingestion of soils, inhalation of chromium contaminated
particulates and ingestion of contaminated groundwater. The preliminary PHE found that risks in
most pathways were within acceptable levels. The deteriorating buildings, however, presented a
direct contact and physical hazard and the groundwater exposure route remained a potential problem.
A comparison to applicable standards and criteria was made for the contaminants found in the
groundwater at the site. CrVI is a non-carcinogen by ingestion. The MCL for total chromium in
drinking water is 50 ug/1. Two onsite wells had CrVI concentrations of 50 and 1,600 ug/1. Potential
risk through ingestion of groundwater containing Cr VI were evaluated using standard techniques
recommended by EPA (EPA 1987). Risks were examined for two current and future-use exposure
scenarios; (1) groundwater reaching a domestic well and (2) groundwater reaching Haw Creek. For
1067
-------
both the average and plausible maximum concentration the hazard index was calculated to exceed 1,
indicating a potential for adverse non-carcinogenic health effects.
A simplified dilution model was employed to assess the future leaching of chromium into the
groundwater. The future groundwater concentration resulting from leaching from soils was estimated
to be 70 ug/1. Since the phase I RI data showed groundwater concentrations to be significantly
higher, it was probable that soil concentrations somewhere under the main process building were
releasing significant levels of contaminants.
Subsequently, a revised work plan was developed to establish a sampling plan for addressing the
groundwater contamination problem. The PHE also led to the ERA.
Engineering Evaluation/Cost Analysis (EE/CA)
The initial approach towards site remediation was planned along the usual RI-FS-ROD-RD-RA
sequence used on many superfund projects. However as site characterization progressed, it was
realized that the preliminary remedial objectives developed through the interactive process indicated
a need to address contaminated soils as a source of contamination.
Based on the preliminary PHE it was determined that although there was no immediate threat, it was
possible that some time in the near or distant future this threat could be realized. A mechanism that
would address the contaminated source and allow a cleanup that would protect public health, and at
the same time potentially reduce investigative dollar expenditures was extremely desirable.
Exploration of the available programs led to the decision to attempt a non time-critical removal effort
under an Expedited Response Action (ERA). The first step under this mechanism is the EE/CA
report.
The EE/CA is very similar to a focused feasibility study. Once the decision to conduct the Expedited
Response Action (ERA) has been made, basic guidelines for determining the suitability of an ERA
must be met before initiation of an EE/CA. The criteria include the following:
• That a threat exists sufficient to meet the removal criteria as specified in the National
Contingency Plan (NCP).
• The existing threat does not warrant a time-critical removal action to immediately mitigate
the threat.
• The ERA is consistent with the final remedy and attains or exceeds applicable or relevant and
appropriate public health and environmental requirements.
• The remedy can be accomplished within the statutory limits of $2 million for cleanup costs
and 12 months for completion of the ERA.
If these criteria are met, the remedial contract A/E firm may begin generation of the EE/CA report.
At that time, in USEPA Region 5, the large majority of EE/CA's are prepared by A/E firms under
the Removal Program. Depending on the complexity of the site and any additional data that must be
obtained prior to the selection of the recommended removal action alternative, EE/CA's generally are
completed within two months. The cost of preparation of the EE/CA is typically $100,000. The
Removal Program in Region 5 has overseen the preparation of EE/CA's by A/E firms that range from
a minimum cost of $40,000 to those that exceed $150,000 and have taken as short as four weeks to
in excess of three months for preparation and submittal of the final document. The EE/CA cost
1068
-------
analysis level of effort was greatly reduced by using the generated output of the CORA cost model
supplemented with the A/E firm's further analysis and input. A final EE/CA for Agency review was
completed within two weeks at a cost of less than $10,000. This is a significant cost and time savings
as compared to previous EE/CA submittals in Region 5.
The identification of potential removal action alternatives based on CORA and augmented by the A/E
firms judgement and experience indicated that the site remediation should be broken into two
categories; soil removal alternatives and building removal alternatives. Both of these source control
alternatives were measured against the "no action" alternative, which is used as a baseline against
which the adequacy of other alternatives can be measured. Under this "no-action" alternative,
adequacy of other alternatives can be measured. Under this "no-action" alternative, no funds are
expended for monitoring, control or cleanup of the contaminated soil. Based on the standard selection
criteria, the recommended removal alternative was off-site landfill with building decontamination
and demolition. This alternative was the most technically reliable for eliminating the potential
migration of contaminants to groundwater. It was the most costly option at $970,000 but it eliminated
wastes currently in place near residential areas while providing the least environmental impact to the
site area.
1989 Expedited Response Action
The performance of the ERA was an innovative approach which expedited site cleanup, greatly
streamlined the decision process in the FS and simplified implementation of the RD/RA.
The concentration of site related contaminants encountered at the Tri-State Plating Site were
considered sufficient cause to warrant a removal action as set forth in Paragraph (b)(2) of Part 300.65
of the NCP. Criteria for implementation of removal actions in the NCP include:
• Actual or potential exposure to nearby populations, animals, or food chain from hazardous
substances or pollutants or contaminants;
• Actual or potential contamination of drinking water supplies or sensitive ecosystem;
• High levels of hazardous substances or pollutants or contaminants in soils largely at or near
the surface, that may migrate; and,
• The absence of other Federal or state response and enforcement mechanisms to respond to the
site.
The contaminated and deteriorating structures and the amount of contamination remaining in
subsurface soils continued to pose potential threats to the public and environment near the site. Based
on the EE/CA source control removal action was proposed to prevent contaminants remaining in the
subsurface soils from continually leaching to groundwater. The detailed plans, Technical
specifications and Health and Safety Procedures for the ERA were documented in the "Contract
Documents for Construction of the Tri-State Plating Site Expedited Response Action" released in
December, 1988.
Under the direction of USEPA, the REM IV Team planned a remedial action designed to remove
materials at the site that could cause harmful levels of contamination to be released to ground water.
The plan called for the removal of the contaminated structures and soils averaging more than 57
mg/kg total chromium. This action level was based on a leaching model which estimated the average
level of contamination which could remain in soil without resulting in groundwater concentrations
1069
-------
in excess of the MCL for chromium of 50 ug/1. The leaching model was documented in a Technical
Memorandum to USEPA (Brown and Buchovecky, 1988).
The removal program started in February, 1989 with an extensive soils investigation to delineate soils
in excess of the cleanup action level. The sampling program involved the drilling of 18 exploratory
soil borings and the collection and analysis of over 300 soils for total chromium analysis. The data
from this and previous investigations were used to specify the limits of contaminated soil excavation.
Also, the interior surfaces of the main process building were grit blasted to remove surface
contamination and then all structures were demolished and removed to a special waste landfill for
disposal under an IDEM special waste permit. Approximately 2800 cu. yds. of contaminated soils
were then excavated and transported under a USEPA generator number to a RCRA-compliant
hazardous waste landfill. The excavation was subsequently backfilled with laboratory verified clean
soil, compacted, regraded, and revegetated.
Phase II Remedial Investigation
The Phase II investigation was conducted concurrently with the ERA. The primary objective of the
Phase IIRI was to identify residual contamination problems following the removal action and evaluate
their potential impact on human health and the environment. Field Investigation activities were
conducted in parallel with the ERA and included installation of 7 additional monitoring wells and
additional sampling of subsurface soils, and groundwater. Groundwater sampling was performed
before and after the ERA. The results of the Phase II RI investigation and a base line public health
evaluation were summarized in the RI report released on November 22, 1989. Applicable data from
Phase I and II were used in site evaluation. A summary of the significant RI findings is provided
below.
The Phase II RI indicated that residual soil contamination consisted primarily of elevated levels of
chromium. Although residual soil concentrations higher than background were present onsite, the
geometric mean concentration was reduced to well below the 57 mg/kg ERA action level. Only about
10 percent of the 213 samples in the data base contained concentrations above the action level. These
occurred primarily in deeper soils left in place after the removal action. Maximum residual
concentrations in these soils ranged up to 195 mg/kg which was more than two orders of magnitude
less than the maximum concentration present prior to the removal action (52,000 m/kg). Independent
conformational sampling from the ERA indicated similar results as shown below.
Western Central Eastern
13 Foot 20 Foot 16 Foot All
Excavation Excavation Excavation Samples
No. Samples 6 6 6 18
Geometric Mean (mg/kg) 23 45 24 27
Maximum Value (mg/kg) 41 156 49.2 156
According to the baseline risk assessment presented in the RI report, human health risks resulting
from potential exposure to residual soil contamination were extremely low and did not pose a
significant threat to health. The low levels of soil contamination remaining in the saturated zone,
approximately 20 feet below the ground surface did not represent a significant direct contact health
threat. In addition, since the source of contamination in the unsaturated zone had been substantially
1070
-------
reduced, the contaminant levels in the saturated zone would diminish with time as groundwater
flushed contaminants from the soil.
The groundwater results from Phase II indicated a pattern similar to that observed in Phase I, both
before and after the ERA. Groundwater contamination consisted primarily of elevated levels of
chromium. Where chromium levels were highest, the form of chromium was mostly the more toxic
CrVI form. The background concentration of total chromium was 8.9 ug/1. Hexavalent chromium
was not detected in the background well. The concentrations detected in MW-1 were slightly higher
than Phase I (60.8 ug/1 for total chromium). The concentrations in MW-4 for total and CrVI were
1810 ug/1 and 1890 ug/1, respectively. The lower total chromium result is believed to be the result
of inter-laboratory variability. Several other wells had total chromium values that exceeded
background. A shallow downgradient monitoring well, MW-6, had a total chromium concentration
of 28.4 ug/1, indicating that the plume is migrating toward the south, but was being diluted as it
spreads out.
Following the ERA, groundwater concentrations at MW-4 did not change significantly indicating
either that insufficient time had passed for residual contamination to be flushed from the aquifer or
that a significant amount of CrVI remained adsorbed in the aquifer beneath the site. In addition,
elevated CrVI concentrations appeared for the first time at the two other shallow wells on the
southern boundary of the site, MW-3 and MW-3A.
Groundwater chromium concentrations observed at the site continued to exceeded the MCL of 50
ug/1. Transport modeling done to estimate potential future groundwater chromium concentrations
at key off-site receptor areas such as the Columbus city well field indicated limited and remote
possibilities that ground water contamination could have a negative affect on human health or the
environment. However, risk evaluations indicated that ingestion of contaminated groundwater
continued to pose a potential threat to human health for the hypothetical case that a potable well was
installed onsite or in the path of the contaminant plume and eventual discharge to Haw Creek
remained a possibility.
The presence of elevated groundwater concentrations indicated that a feasibility study to evaluate the
cost effectiveness of various remedial actions was warranted. However, the RI revealed several
significant data gaps concerning the extent and migration potential of groundwater contamination.
Because contamination was detected in so few monitoring wells, the extent and mass of contamination
to be addressed was uncertain. In addition, existing data yielded a wide range in aquifer parameters
(eg. hydraulic conductivity, ground water velocity, initial mass, chemical distribution coefficient,
hydrodynamic dispersion coefficients) which would make the conceptual design and projected
cleanup times uncertain. There was also some question as to whether the ERA had completely
eliminated the threat posed by leaching of subsurface vadose zone soils.
These data gaps did not detract from the need to evaluate various groundwater extraction and
treatment alternatives. Therefore a decision was made to proceed with the FS despite the known data
gaps. The data gaps would be addressed by a pilot ground water pump and treat test to be performed
in parallel with the FS.
ERA Pump Test and Verification Sampling
Because these data gaps were recognized at the inception of the FS, additional groundwater sampling
and an aquifer pump test were performed concurrently. The objectives of these investigations were
to provide site specific hydrogeologic data for ground water modeling, verify the effectiveness of the
ERA, demonstrate that aquifer restoration times could be shortened by removal of a quantity of
1071
-------
contaminated water from the aquifer and provide information on pumping rates required to create
an adequate capture zone.
The aquifer pump test included the following investigation activities:
• Installation of an 8 inch extraction well capable of pumping at 300 gpm;
• Performance of a 72 hour sustained rate pump test to determine aquifer hydraulic
conductivity and zone of influence;
• Collection of daily discharge samples to monitor the contaminant removal rate and to
determine if discharge would exceed the permit requirements for discharge to the POTW;
• Collection of daily groundwater samples from MW-4 to determine the relationship between
ground water contaminant concentrations and the number of pore volumes removed; and
• Collection of a final round of groundwater sampling of all study area wells to determine the
extent of contaminant removal.
The pump test was conducted continuously at 310 gpm for 12 days between November 11 and
November 22, 1989. The results of the test and additional sampling were presented and discussed in
the Remedial Action Report for the site issued in January, 1990. Analytical results from the pump
test are presented in Table 1. The significant findings of the pump test are summarized below.
Evaluation of drawdown data to determine aquifer parameters revealed hydraulic conductivities
ranging from 2006 gpd/ft2 to 3370 gpd/ft2 and a specific yield ranging from .04 to .24. Maximum
drawdown near the well screen was 4.04 feet
1072
-------
TABLE 1
PUMP TEST ANALYTICAL RESULTS
TRI STATE PLATING SITE
DAILY GROUNDWATER SAMPLES
Date
11/10/89
11/11/89
11/12/89
11/13/89
11/14/89
11/16/89
11/17/89
11/19/89
11/20/89
11/22/89
Dupl.
11/14/89
FB-1
M-4
Total
Chromium
(Filtered)
(ug/1)
153
480
123
607
796
990
343
160
151
131
808
ND
M-4
Cr(6+)
(Unfiltered)
(ug/1)
NA
NA
83.9
593
564
1100
367
156
149
141
868
ND
NA = Not Analyzed
ND = Not Detected
Det. Limit = 10 ug/1
DAILY DISCHARGE COMPOSITE SAMPLES
Date
11/10
11/11
11/12
11/14
11/15
Volume
0-1 000 gal
1000-28,000 gal
28,000-226,200 gal
250,000-685,000 gal
685,000- 1,000,000 gal
Cr
ug/1
2800
197
179
185
170
Cd
ug/1
ND
ND
ND
ND
ND
Ni
ug/1
114
ND
ND
ND
ND
Pb
ug/1
117
ND
ND
ND
ND
PH
ug/1
7.69
7.62
7.56
7.55
7.57
Tss
mg/1
57
125
NA
NA
NA
Activity
Well Development
Devel. & Step Test
72 Hour Test
72 Hour Test
72 hour test
Notes: All samples unfiltered
Discharge after 11/11/89 was clear. No TSS samples collected
Detection Limits Cd = 5 ug/1 Discharge Limits
Cr= 10
Ni- 11
Pb-41
Cd - 1200 ug/1
Cr * 5000
Ni = 5000
Pb = 600
TSS * 250 mg/1
pH » 6-11 units
1073
-------
after 72 hours and a distance drawdown graph indicated a zone of influence approximately 2000 feet
in diameter. These results were in good agreement with published data for this aquifer and it was
concluded that a hydraulic conductivity between 2000 and 3000 gpd/ft 2 and a specific yield of 0.2
could be used with confidence in further site specific calculations.
Daily discharge composite samples indicated that from 6 to 7 pounds of chromium were removed
from the aquifer during the 12 day test. During this time, chromium concentrations in the discharge
decreased from 2800 ug/1 to a stable low concentration of about 180 ug/1. The POTW discharge limit
of 5000 ug/1 was never exceeded during well development and testing. Maximum nickel and lead
concentrations were 114 and 177 ug/1, respectively, and decreased to non-detectable levels during the
remainder of the test. These values were also well below discharge limits. Cadmium, another
contaminant regulated by the discharge permit, was not detected in discharge samples.
Hexavalent and total chromium concentrations in daily groundwater samples from MW-4 indicated
considerable fluctuation during the pump test. CrVI concentrations increased from a low of 84 ug/1
to a maximum concentration of 1100 ug/1 3 days into the test. Concentrations subsequently decreased
to a value of about 140 ug/1. This trend indicated movement of a contaminant pulse past MW-4 in
a short period of time which suggested much more rapid movement of chromium than was previously
assumed. During the RI and FS, moderate adsorption was assumed because of the steady
concentrations observed at MW-4. Based on the pump test results, it was judged that the distribution
coefficient (Kd) of 20 ml/g assumed in the RI and FS was too high and that a Kd of 2 ml/g or less
was more representative.
Additional groundwater sampling was conducted approximately three weeks after the conclusion of
the Pump test. Hexavalent chromium concentrations in all onsite wells, including MW-4, were not
only below the 50 ug/1 action level proposed in the FS but below detection levels as well. However,
CrVI was found for the first time at high concentrations (400 ug/1) in the downgradient shallow well,
MW-6. Resampling by IDEM in March, 1990 found low levels of CrVI (13 ug/1) in MW-4, verifying
that the pump test had reduced onsite contamination to low levels. However CrVI was apparently no
longer present at MW-6 by this time. These findings appeared to substantiate the idea that chromium
contamination was moving faster than was previously assumed.
Feasibility Study
Because of the ERA, the analysis and decision process in the FS was greatly simplified. Not only had
the ERA removed a significant potential source of groundwater contamination, but risks due to direct
contact, incidental ingestion of soils and inhalation in several potential future exposure scenarios were
reduced to acceptable levels. Therefore the FS needed to consider only one contaminated media-
groundwater. The remedial objective developed in the FS was to remove and treat CrVI
contamination and to restore the effected part of the aquifer to levels less than the MCL of 50 ug/1.
The Feasibility Study developed and evaluated a range of alternatives for groundwater restoration by
natural attenuation and by active groundwater extraction. The Alternatives developed were as
follows:
Alternative No 1: No Action
Alternative No 2: Monitoring
Alternative No 3: Groundwater Extraction/ Discharge to POTW
Alternative No 4: Groundwater Extraction/Onsite Treatment/Discharge to Haw Creek
1074
-------
For alternatives achieving aquifer restoration by active extraction, several alternative pumping rates
were evaluated to vary the restoration times. The Feasibility Study was released for public and State
comment on January 22, 1990. Significant conclusions of the FS are summarized below.
A comparative analysis indicated that active restoration alternatives were more favorable in terms of
overall protectiveness, compliance with Applicable, Relevant and Appropriate Regulations (ARARs),
time required to implement cleanup, long term effectiveness, and reduction of toxicity, volume and
mobility of contamination. Natural attenuation under the No Action and Monitoring Alternatives was
estimated to take more than 40 years, during which time, the plume would continue to present a
potential hazard to anyone installing a well in the path of the plume. By comparison, active
restoration would prevent further migration of the plume and would take from 5 to 13 years to
achieve. The active restoration alternatives were, however, more expensive and difficult to
implement.
The FS established that the no action alternative was not acceptable and that groundwater extraction
using pumping wells and discharge to the POTW was a viable and cost effective alternative. The
aquifer pump test verified that discharge to the POTW was feasible and provided information on a
suitable pumping well design and required pumping rates.
Because contamination was detected in so few monitoring wells, the ground water contamination
plume addressed in the FS was simulated using a contaminant transport model. Use of the model
introduced elements of uncertainty in conceptual components of the alternatives such as the actual
number, and location extraction wells required to achieve cleanup. In addition, the model
incorporated several assumed parameters (eg. initial mass, chemical distribution coefficient,
hydrodynamic dispersion coefficients) that were highly uncertain. For example, using high and low
range variables in the model during the RI gave natural cleanup times ranging from 18 to 1500 years.
The pump test results collected concurrently with the FS indicated a need to revise certain conclusions
reached in the FS. The potentially lower Kd indicated by the pump test data, suggested that
contaminants had moved faster and spread farther than the model used in the FS indicated.
Therefore, additional monitoring wells were possibly needed to determine the presence of
contamination in downgradient areas. Previous estimates of the mass of contamination in the aquifer
were also based on a Kd of 20. Therefore revision of clean-up times and exposure point
concentrations was also necessary. There also remained some question as to whether the ERA removal
action had completely eliminated the threat posed by leaching of subsurface vadose zone soils. As
a result of these findings, predesign investigations were recommended prior to design and
implementation of the Final Remedial Alternative.
Pre Design Investigation
The Pre-Design Investigation activities were conducted to collect additional data necessary for
preparation of an RD/RA. Specific objectives of the pre design activities were as follows:
• Provide a delineation of the current groundwater plume.
• Perform a qualitative evaluation of chemical fate tendencies of CrVI to provide preliminary
pumping rate and scheduled for the extraction well(s).
• Provide the location and specification for each existing and proposed extraction well required
to effectively achieve the cleanup goals of USEPA and IDEM.
1075
-------
The activities performed during the Pre-Design Investigation included installation of 5 additional
downgradient monitoring wells and 3 piezometers to refine the extent of contamination, and collecting
two rounds of groundwater samples from selected monitoring wells. IDEM collected and analyzed
samples from selected wells to provide further information on the concentration trends over time.
In addition to submitting the samples for CrVI and total metals analysis, samples were also analyzed
for alkalinity, sulfate, and chloride to identify anions that might compete with dichromate in
exchange reactions with the soil. Also, the Eh, pH and dissolved oxygen contents were measured in
the field to determine the redox potential of the samples. Unfiltered total metals at MW-4, -6 and -
11 were also collected to determine the difference between dissolved and particulate concentrations.
The additional predesign investigation wells and the three piezometers installed southeast of the
industry located downgradient of the site provided more information on water table gradients. It had
previously been assumed that groundwater flowed to the southeast towards Haw Creek. Using, water
level data from the shallowest wells, the water table contour map shown in Figure 4 was constructed.
The water table contour map clearly established that groundwater in the vicinity of the site flowed
in a southerly direction rather than southeast towards Haw Creek. The bowed pattern of the water
table contours south of the site indicated the influence of the industrial well on water levels in the
area.
The water table contour map suggested the possibility that contamination from the site may be moving
off site in a southerly direction. The predesign sampling CrVI groundwater results are summarized
in Figure 5. The sampling results indicate that CrVI contamination was moving off site to the south.
A narrow finger of high concentration in excess of 1000 ug/1 appeared to extend south-southeast from
MW-4 through MW-6 to P-l. Detectable CrVI contamination was found as far south as P--3. In
addition, CrVI contamination reappeared again in high concentrations at MW-3 and at MW-6 after
having been absent from these wells in the previous sampling round. The new well (MW-11) located
south of the site between MW-6 and 8 contained 86 ug/1.
A plot of CrVI concentrations at MW-4 over time using all RI data is presented in Figure 6. The
gradual reappearance of contamination at MW-4 after it had been removed to near detection levels
during the pump test, suggested that an source of groundwater contamination remained and that CrVI
was gradually being reintroduced into the aquifer through some unknown mechanism. Three possible
release mechanisms were postulated:
• Leaching of residual chromium contamination from vadose zone soils;
• Reestablishment of equilibrium by chromium desorption after the cessation of the pump test;
and
• Presence of fixed trivalent chromium (CrIII) below the water table and slow oxidation of this
material to hexavalent chromium.
Poor correlation between groundwater concentrations and monthly total rainfall amounts as shown
in Figure 6 plus the fact that soil chromium concentrations had been drastically reduced during the
ERA indicated that this mechanism was a remote possibility. Because of the relatively long time
interval required for chromium concentrations to be reestablished to pre pump test levels, chromium
desorption from soils below the water table at a rates much less than removal rates during aquifer
pumping was also regarded as unlikely.
The most plausible explanation was that a substantial reservoir of the less soluble and mobile CrIII
ion had accumulated in the aquifer beneath the Tri-State Plating site. In the absence of chromium
1076
-------
loadings from the plant operations and in the presence of manganese dioxide or other natural
oxidizing agent, oxidation of CrIII to CrVI was occurring. This mechanism might explain why it took
so long for CrVI to reappear after the pump test and also the consistent concentrations detected at
MW-4 prior to the test. Field evidence that this reaction was occurring, however, was limited.
Adsorption is the dominant fate controlling CrVI mobility. However, the divalent dichromate anion,
rather than the hexavalent metal cation is the species that is most strongly adsorbed (EPRI 1985). As
a consequence, the activity of other anions such as carbonate, bicarbonate, sulfate and silicic acid
strongly compete with dichromate for adsorption sites. Standard water quality analyses collected
during the predesign investigation indicated moderately high groundwater alkalinity concentrations
of about 300 mg/1. Given the potentially low density of binding sites in the sand and gravel aquifer
and the high concentration of competing carbonate/bicarbonate anion, CrVI adsorption may be
limited. This agreed with the tentative conclusions about the mobility of CrVI reached following the
pump test. Consequently, contaminant migration velocities approaching that of groundwater, or
nearly 2 ft/day might be qualitatively expected. This conclusion, in turn supported the contention
that oxidation of trivalent chromium rather than desorption of CrVI was the current source of
groundwater contamination.
The predesign data confirmed that groundwater contamination had spread farther and faster than
anticipated in the FS. In addition, the predesign data clearly demonstrated that a residual source of
chromium contamination remained onsite. Based upon the low retardation of CrVI indicated by the
pump test and the high groundwater velocities at the site, it was concluded that chromium was being
generated at a slow but rather constant rate and was migrating off-site rapidly.
These characteristics suggested that, rather than implementing the groundwater remedial alternative
with objective of restoring the entire effected area to MCLs, the remedial response objective could
be met by operating the onsite well, designed and constructed during the pump test, to prevent any
additional off-site migration. The contamination that had already moved beyond the capture zone
of the well would probably flush from the aquifer faster than additional wells could be installed.
Remedial Design and Monitoring
Based upon the data assembled to date, it appeared that the existing on-site extraction well could be
used effectively to prevent further off-site migration. The extraction well did not need to be
operated continuously to be effective. Analysis of Figure 6 indicated that, once a currently
contaminated pore volume was removed, approximately 4 months would pass before groundwater
concentrations again exceeded 50 ug/1. Based on the pump test results, it appeared that a
contaminated pore volume could be removed in approximately 2 weeks by operating the well at 200-
300 gpm. Pump test data also indicated that operation of the pump at these rates would provide a
capture zone that would prevent chromium migration off-site. Therefore, to be conservative, it was
recommended that pump operation be scheduled on a quarterly (3 month) basis with the pump
operating for the first three weeks of each quarter. After the first three weeks, the pump would be
turned off to allow CrVI concentrations to increase to 50 ug/1. Quarterly operation of the extraction
well would insure that CrVI concentrations exceeding 50 ug/1 would not leave the site area.
Based on current information, it was not possible to estimate how long the onsite extraction well
would have to be operated. This information depended on the source and mechanism of CrVI release.
The source of the contamination was suspected to be slow oxidation of CrIII in the aquifer below the
site. Although not as likely, vadose zone leaching or slow desorption of adsorbed Cr (VI) could be
contributing factors. Because only a year had passed since the removal action, and less than a year
had passed since the pump test, these conclusions were based on limited monitoring data and were
1077
-------
-t o e. 11
? a> c =
> o —
-• 2
o o.
» r
H
o
woo
.'?••
M — O
-1-1
S2
ni
IS
1078
-------
7 -
6 -
1 5-
4 -
3 -
2 -
1 -
Monthly
Precipitation
II 1 \
1 I I (
3500
3000 -
2500 -
Cr VI Concentrations
AtMW-4
19-Nov-87 06-Jun-88 23-Dec-88 11-Jul-89 27-Jan-90 15-Aug-90
n 1 1 1 1 1 : —r
19-Nov-87 06-Jun-88 23-Dec-88 11-Jul-89 27-Jan-90 15-Aug-90
FIGURE 6
CrVI Concentrations At MW-4
And Precipitation Data
1079
-------
therefore difficult to confirm. Because these characteristics were still poorly defined, the source
longevity was difficult to estimate.
It was decided that onsite monitoring be implemented to determine the decrease in source strength
so that clean-up times could be estimated. It is recommended that the extraction well be operated in
conjunction with the onsite source monitoring for at least 2 years in order to provide an adequate data
base for planning future courses of action.
On-site Monitoring: Continued monitoring of onsite wells was specified to re-evaluate each of the
possibilities discussed above. In addition, onsite monitoring would allow the effectiveness of the
extraction well to be evaluated. It was recommended that five of the eight onsite monitoring wells
(MW-1, 3, 3A, 4, 4B) be sampled on a quarterly (3 month) basis. The remaining onsite wells (MW-2,
IB and 4C) had never shown any evidence of contamination.
The sampling would coincide with the extraction well operation cycle such that each quarterly sample
is collected at the end of the quarter just prior to pump operation in the next quarter. This insured
that samples were representative of the "equilibrium" concentration for each quarter. These
equilibrium concentrations would be plotted with time to determine any trend showing the decrease
in source strength. Quarterly monitoring data would also be compared with monthly precipitation
data from Columbui Utilities to evaluate the possibility that contamination is the result of vadose zone
leaching.
Quarterly onsite samples would be analyzed for CrVI and filtered and unfiltered total chromium. The
first two rounds of unfiltered quarterly samples would also be analyzed for manganese dioxide. The
presence of manganese dioxide would indicate whether CrIH oxidation i* occurring. If manganese
dioxide is absent or present in low concentrations then it will be unlikely that oxidation of a CrIH
reservoir is the source of contamination.
If initial quarterh monitoring data suggested that CrIH oxidation is the source of current
contamination, then it would be advantageous to consider applying an oxidant to the site to speed up
oxidation and mobilization of CrVI. Evaluation of the use of an oxidant would involve bench scale
testing of subsurface soil samples to determine an optimum oxidant and application method.
Quarterly monitoring would be conducted for at least two years to provide an adequate data base for
planning future actions. If groundwater concentration trends indicate a steady decline it will be
possible to project cleanup times by extrapolation. If, after a period of time, no significant decrease
is noted, additional onsite subsurface investigations of soil concentrations below the water table may
be necessary to further investigate the nature of the source. This would involve drilling of several
test borings to collect soil samples for analysis.
Off-Site Monitoring: Concentrations significantly above the FS cleanup action level of 50 ug/1 are
present in the ground water at distances of over 600 feet south of the site. Because the downgradient
limit of this contamination is undefined, specification of additional off-site extraction wells was
premature without further monitoring and evaluation.
However, it was realized that once the off-site migration was reduced by the extraction well, the
down gradient contamination may flush from the aquifer rather rapidly because of the high migration
velocities. It was recommended that prior to installing additional off-site monitoring or extraction
wells, the onsite well be operated for at least two years and the down gradient wells be monitored to
determine the natural flushing rate. If the natural flushing rate is high, then clean up levels may be
reached in a reasonable time frame and it would not be necessary to install additional extraction wells.
To meet the off-site monitoring objectives noted above, it was recommended that the piezometers
south of the site be replaced with monitoring wells. These wells along with MW-6, 6B, 7, and 10
would be monitored on a quarterly basis. All other off-site wells would be sampled annually. During
this period monitoring of Haw Creek would be performed to insure that the plume will have no
impact on surface water. At least 4 surface water sampling stations would be established and
monitored on a monthly basis. Samples would be analyzed for CrVI.
1080
-------
In the event that natural flushing progresses slower than expected, further plume delineation south
of the site may be required before a downgradient extraction system could be specified.
CONCLUSIONS
Current Status
The Remedial Designs/Remedial Action contract was awarded to a Region III Alternative Remedial
Contracting Strategy (ARCS) firm in October 1990. As of April 1991, the Remedial Design is 100%
complete and the Remedial Action has been initiated. Because of the expedited approach taken on
this site, inception to RA implementation required 3 years. This period is much quicker than the
typical superfund project. The success is due to several factors including the sound interactive RI/FS
approach and the innovative methods (eg. ERA) that were employed.
The expedited Remedial Design can be attributed to the strategies employed during the course of the
Remedial Design. As a consequence the contractor was primarily tasked to design a pump house and
the associated piping from the existing extraction well to the city sewer system. The other aspects
of a typical Remedial Design, and Remedial Action such as source removal, design of the extraction
system, aquifer parameter delineations, placement of extraction well(s), etc., had already been
completed through innovative techniques employed early on in the RI/FS process.
The expedited Remedial Design, which entailed preparation of a work plan, design of the selected
remedy, preparation of plans and specifications, bidding documents, site closure and operation and
maintenance plan, cost estimates, quality assurance plans, and health and safety plan has been
completed at a cost to the agency of less than $100,000.
The remedial Action will be initiated in June 1991 after procurement of bids and award of the
contract. The total remedial action is anticipated to be completed in two years, after pumping has
been initiated and at a cost of approximately $300,000.
Impacts of the Interactive RI/FS Approach
The Tri-State Plating Project included the standard components usually seen in the typical Superfund
project. It included a remedial investigation, a baseline risk assessment, a pump and treat pilot test,
a feasibility study, and a predesign investigation. These components are the major tasks described
in "Guidance for Performing Remedial Investigations and Feasibility Studies Under CERCLA". With
minor variations, this sequence of operations is gaining acceptance as the state of the practice in
hazardous waste site remediation. The aspect that made the Tri-State Plating project particularly
successful, was the interactive way in which the results of one component were used to direct
activities in the others. Although the interactive or phased approach is explicit in the current NCP
and EPA RI/FS guidance, we believe that it is seldom used as effectively as it was at the Tri State
Plating Site to expedite investigations and site clean up.
Key to the success of the interactive approach at the Tri-State Plating site was the early establishment
of remedial objectives. Early in the RI stage, a preliminary public health evaluation identified excess
health risks associated with chromium in groundwater. A remedial objective was established to
prevent the release of contaminants to groundwater and to eventually implement a groundwater clean
up. Despite the existence of data gaps at each stage along the way, the driving remedial objective
enabled the subsequent investigation activities to focus on chromium contamination and determining
site factors relevant to groundwater remedial actions. By specifying the remedial objective early,
project decision makers were equipped with decision making tool to evaluate whether the data gaps
effected the overall approach. In each instance, the decision was that the data gaps were important
but could be answered by appropriate investigation in the next phase.
Often, the interactive approach is not effective because there is a perception that no decisions
concerning remedial objectives can be made until a comprehensive RI is performed. Upon
completion of an initial RI and presentation of the results in an RI report, data gaps are usually
identified. Frequently, discussion of remedial objectives, development of possible remedial
-------
alternatives and other decision activities are postponed pending completion of additional phases of
the RI to address the data gaps. Based on our experience at the Tri-State Site, we feel that this
perception and course of action is unwarranted. Available data can always be used to begin
formulating a course of remedial action for the site.
Site assessment uncertainties will always exist at each stage in the RI/FS process. However these
uncertainties should not hinder the development of preliminary objectives and development of
preliminary response objectives. RI data at each stage may not be fully descriptive of the site, but
any data collected provides some information that can be used to begin the formulation of remedial
objectives.
Impacts of the ERA
The ERA portion of the project is not typical of most superfund sites. However, it was developed as
a direct result of the interactive approach discussed above and had a significant impact on the final
outcome. In order to remove the source of contamination, the former metal plating building and
contaminated subsurface soils were removed during an ERA even before the FS was started. After
completion of the ERA, the FS was greatly simplified in that the only remaining media of concern
was groundwater itself. The selected remedial action to address groundwater contamination, pump
and treat, was further simplified by conducting an aquifer pump test in parallel with the FS This
resulted in a preliminary technical evaluation of the pump and treat design prior to actual initiation
of the remedial design (RD) phase. The remedial action (RA) was also expedited because the actual
test well was designed so that it would be part of the final remedy. Thus, in many instances, the
RI/FS and Expedited Response Action which was performed on this site served as pre-design
activities for the final remedial action.
Use of the ERA approach is now being replaced with a new concept called the Interim Rod. The
Interim Rod can be used to effectively serve the same function as the ERA at the Tri-State Plating
Site.
Interim RODs
The USEPA now intends to address situations that dictate the need to take quick action either to (1)
Protect human health and the environment from an imminent threat in the short term, while a final
remedial action is being developed or (2) institute temporary measures to stabilize the site or operable
unit and/or prevent further migration or degradation by conducting interim action Record of
Decisions in lieu of expedited response actions. An interim action, like an ERA, is limited in scope
and only addresses areas and/or media for remediation and will be followed up by a final operable
unit Record of Decision. Interim actions may be implemented for a completely separate operable unit
or may be a component of the final ROD, dependent upon the reasons for conducting the action (i.e.,
removing soils to eliminate the source of contamination of groundwater versus providing a temporary
alternate water supply and sealing wells that are pumping from a contaminated aquifer).
Since an interim action may be conducted during any phase of the Remedial Investigation/Feasibility
Study to mitigate the more immediate threats, there may not be sufficient time to prepare a
comprehensive RI Report of FS Report.
In fact, preparation of an RI/FS report is not required for an interim action. However, for the
purpose of fulfilling the NCP's Administrative Record requirements, there must be documentation
that supports the rationale for the action. A summation of site data collected during field
investigation should be sufficient to document a problem in need of response; in addition, a short
analysis of what remedial alternatives were considered, which ones were rejected, and the basis for
the evaluation (as is done in a focused FS) should be summarized to support the selected action. The
Interim action decision documentation are outlined in Tables 2 and 3.
1082
-------
TABLE 2
DOCUMENTING INTERIM ACTION DECISIONS
OUTLINE FOR THE PROPOSED PLAN
The Interim Action Proposed Plan should include the following information:
1. Site Description: This section should focus on site characteristics addressed by the limited
action.
2. Scope and Role of Operable Unit: This section of the document should specify how the
interim response action fits into the overall site strategy. The point should be made that, to
the extent possible, the interim action will be consistent with any planned future actions.
3. Summary of Site Risks: This section should provide the rationale for taking a limited action.
This should be supported by facts that indicate the action is necessary to stabilize the site,
prevent further degradation, or that the action can accomplish significant risk reduction
quickly. The information should relate only to the limited scope of the action. Qualitative
risk information may be presented if quantitative details are not yet available, which will
often be the case.
4. Summary of Alternatives: A very limited number of alternatives should be analyzed for
interim actions; in some cases, only one plan of action will be appropriate to consider. The
alternative descriptions should reflect the pertinent Applicable and Relevant and Appropriate
Regulations (ARARs) associated with the action. ARARs are important for the following
aspects of an interim action: any portion of the remedy that is final, materials that are treated
or managed off-site, and any release that will occur during implementation. Requirements
are not applicable or relevant and appropriate if they are outside the scope of the interim
action.
5. Evaluation of Alternatives and the Preferred Alternative: The comparative analysis should
be conducted in relation to the limited role and scope of the remedy. Criteria that are not
pertinent to the selection of interim actions (e.g., long-term effectiveness of a temporary cap)
need not be addressed in detail. Rather, their irrelevance to the remedy decision should be
noted.
6. Statutory Findings: The findings should be discussed in terms of the limited scope of the
action.
1083
-------
TABLE 3
DOCUMENTING INTERIM ACTION DECISION
OUTLINE FOR THE ROD
The ROD, documenting the selection of an interim action remedy, should contain the following
modifications.
1. Declaration:
• Statutory Determinations: The declaration statement should read as follows:
This interim action is protective of human health and the environment, complies with
(or waives Federal and State applicable or relevant and appropriate requirements) for
this limited-scope action, and is cost-effective. This action is interim and is not
intended to utilize permanent solutions and alternative treatment (or resource
recovery) technologies to the maximum extent practicable for this [interim
action/operable unit]. Because this action does not constitute the final remedy for the
[site/operable unit], the statutory preference for remedies that employ treatment
[although partially addressed in this remedy] that reduces toxicity, mobility, or volume
as a principal element will be addressed by the final response action. Subsequent
actions are planned to address fully the threats posed by the conditions at this
[site/operable unit]. Because this remedy will result in hazardous substances
remaining on site above health-based levels, a review will be conducted within five
years after commencement of the remedial action as EPA continues to develop final
remedial alternatives for the [site/operable unit]. The review will be conducted to
ensure that the remedy continues to provide adequate protection of human health and
the environment. Because this is an interim action ROD, review of this site and of
this remedy will be continuing as part of the development of the final remedy for the
[site/operable unit].
2. Decision Summary
• Scope and Role of Operable Unit: This section provides the rationale for taking the
limited action. To the extent that information is available, the section should detail
how the response action fits into the overall site strategy. This section should state
that the interim action will be consistent with any planned future actions, to the
extent possible.
• Site Characteristics: This section should focus on the description of those site or
operable unit characteristics to be addressed by the interim remedy.
• Summary of Site Risks: This section should focus on risks addressed by the interim
action and should provide the rationale for the limited scope of the action. The
rationale can be supported by facts that indicate that temporary action is necessary to
stabilize the site or portion of the site, prevent further environmental degradation, or
achieve significant risk reduction quickly while a final remedial solution is being
developed. Qualitative risk information may be presented if quantitative risk
information is not yet available, which often will be the case. The more specific
findings of the baseline risk assessment should be included in the subsequent final
action ROD for the operable unit and the ultimate cleanup objectives (i.e., acceptable
exposure levels) for the site or operable unit.
• Description of Alternatives: This section should describe the limited alternatives that
were considered for the interim action (generally three or fewer). Only those
requirements that are applicable or relevant and appropriate requirements (ARARs)
1084
-------
to the limited-scope interim action should be incorporated into the description of
alternatives.
Summary of Comparative Analysis of Alternatives: The comparative analysis should
be presented in light of the limited scope of the action. Evaluation criteria not
relevant to the evaluation of interim actions need not be addressed in detail. Rather,
their irrelevance to the decision should be noted briefly.
Statutory Determinations: The interim action should protect human health and the
environment from the exposure pathway or threat it is addressing and the waste
material being managed. The ARARs discussion should focus only on those ARARs
specific to the interim action (e.g., residuals management during implementation).
The discussion under "utilization of permanent solutions and treatment to the
maximum extent practicable" should indicate that the interim action is not designed
or expected to be final, but that the selected remedy represents the best balance of
tradeoffs among alternatives with respect to pertinent criteria, given the limited scope
of the action. The discussion under the preference for treatment section should note
that the preference will be addressed in the final decision document for the site or
final operable unit.
1085
-------
REFERENCES
Brown, E., and E. Buckovecky, 1988. Cleanup Levels for CrVI in Soil. Memorandum to William
Bolen, RPM.
CH2M Hill, ICF Technology, July, 1988. Final Engineering Evaluation/Cost Analysis-Tri-State
Plating Site.
CH2M Hill, ICF Technology, 1989. Remedial Investigation Report. Tri-State Plating Site.
CH2M Hill, ICF Technology, 1990. Public Comment Feasibility Study Report. Tri-State Plating Site.
EPRI, 1986. Geochemical Behavior of Chromium Species. Electric Power Research Institute. EPRI-
EA-4544.
Hill, J.R., 1988. The Geology of Indiana: A General Summary.
U.S. Environmental Protection Agency, 1987. Integrated Risk Information System (IRIS).
Environmental Criteria and Assessment Office, Cincinnati,
Ohio.
U. S. Environmental Protection Agency, 1988. Guidance for Conducting Remedial Investigations and
Feasibility Studies Under CERCLA.
U.S. Environmental Protection Agency, September, 1988. Cost of Remedial Action Model-Users
Manual. Office of Solid Waste and Emergency Response, Washington, DC.
108R
-------
Excavation/Off-Site Incineration RD/RA -
Optimization of the Planning/Investigation Process
Based on Two NPL Site Case Studies
John F. Gorgol, P.E.
Supervising Engineer
Ebasco Enviromental,
A Division of Ebasco Services Incorporated
One Oxford Valley, Suite #414
Langhorne, PA 19047-1829
(215) 752-0212
William Pencola
Senior Cost and Schedule Engineer
Ebasco Services Incorporated
INTRODUCTION
The purpose of this paper is to address the question of how much site investigation work is justified
prior to implementation of remedial actions involving excavation and off-site treatment/disposal.
This question is addressed by analyzing two recently completed NPL site remediations (case studies)
which consisted of excavation and off-site incineration of solid waste materials. Each case study is
analyzed by comparing actual total costs for site characterization and remediation with estimated total
costs for two hypothetical sensitivity cases in which the amount of sampling and analysis for further
site characterization prior to implementation of the remedial action is varied. The results presented
in this paper could be used to assist Project Managers responsible for similar projects in deciding if
sufficient site characterization data (especially the horizontal and vertical extent of contamination)
exist for the cost-effective procurement of the remediation contractor. If sufficient data were not
collected during the Remedial Investigation (RI), additional data could be collected during the
Remedial Design (RD) phase.
It is important to consider the performance of sampling and analysis during the RD for excavation
and off-site treatment remediations because total overall costs could be higher in cases of insufficient
or excessive information concerning site characterization. Insufficient site characterization data could
lead to higher remediation costs due to the perception of higher waste quantity uncertainties by the
bidding remediation subcontractors and/or failure to be able to take full advantage of potential
volume discounts offered by treatment/disposal facilities. The collection of excessive site
characterization data could result in the needless expenditure of funds for sampling and analyses with
no significant cost savings during the remedial action. Although it is clear that the optimum amount
of sampling and analysis for site characterization is based on a cost/benefit analysis between
characterization costs and improved information benefits, there currently is no general guidance on
how to establish the optimum for a given site.
BACKGROUND
In this section background information is provided for each of the case studies. In addition, the
approach taken to analyze each of the case studies is presented and discussed.
Site A is located in a rural area near a large National Forest. The entire site area covers approximately
45 acres including the specific areas of concern which cover roughly 3 acres. For a 50-year period
1087
-------
a wood-tar, waste material was generated from a process for production of charcoal. The process
involved heating wood in the absence of oxygen to produce charcoal, methanol, acetic acid and wood-
tar. The wood-tar was deposited onto the ground where it eventually formed several surface
impoundments and other areas throughout the site. The composition of the wood-tar is such that
although it is not classified as a RCRA waste, it contains elevated levels of phenols and polynuclear
aromatic hydrocarbons (PAHs). The Record of Decision (ROD) for the site specified excavation and
off-site incineration of the wood-tar in order to reduce potential risks to human health.
Site B is located on an approximately 10-acre property in an industrial/residential area. The on-site
facility was used for operations involving copper recovery from scrap wire. A chemical process for
removal of wire insulation produced a waste material which consisted primarily of elemental carbon.
The black carbon waste material contained percent levels of copper and lead. Carbon waste samples
also exhibited elevated levels of tetrachloro- ethylene, and polychlorinated biphenyls (PCBs). The
carbon waste material was placed on the ground surface in a pile near the center of the site property.
The ROD for the carbon waste Operable Unit specified excavation and off-site incineration of the
carbon waste pile to prevent the further spread of contamination to groundwater and to eliminate
human health risks associated with other potential pathways.
It is stated in the RI/FS and RD guidance documents that accuracies for cost estimates should be
order of magnitude estimates at the Feasibility Study stage (i.e., +50/-30 percent) and within +15/-10
percent at the final design stage. These are useful theoretical benchmarks, yet for many remedial
alternatives it is difficult or impossible to determine overall cost estimate accuracies. Overall cost
estimate accuracy depends on the accuracies of a large number of site-specific factors which are often
unique. Even if it was possible to definitively assess overall cost estimate accuracies, it is likely that
optimum final design cost estimate accuracy depends on site-specific factors. For excavation arid off-
site treatment remediations the primary factor affecting overall cost estimate accuracy is the estimate
of waste quantity. To a great extent, the degree of site characterization to improve the accuracy of
waste volume estimates becomes a site-specific judgment call by the cognizant Project Manager. The
primary factor affecting a Project Manager's ability to effectively make this decision is the amount
of experience this individual has gained on previous similar projects. The approach taken in this
paper is to analyze two case studies so that the results could assist Project Managers involved with
future projects with similar characteristics.
DISCUSSION
A minimal amount of intrusive investigative effort was expended during the RI/FS and RD phases
to define waste quantities for each of the actual case studies. Therefore, the two hypothetical
sensitivity cases considered for each site involved increasing levels of site characterization. The
hypothetical cases were developed assuming more sampling and less uncertainty (better estimates)
regarding waste quantity.
The overall approach taken was to estimate incremental costs for additional site characterization and
compare them with the corresponding incremental cost savings which may be achieved during the
remedial action. Incremental costs for additional site characterization consisted primarily of costs for
performance of test pits and completion of relevant laboratory analyses. Incremental cost savings
during the RA phase included anticipated reductions in the unit prices ($/ton of waste) bid by the
remediation contractor. Anticipated reductions in unit prices would result from two factors; quantity
discounts offered by treatment facilities and bid restructuring based on perception of waste quantity
uncertainty. It has been assumed that if remediation contractors believe that there is a high
probability that the actual waste quantity will significantly exceed the estimated waste quantity there
will be a tendency to increase unit prices for waste disposal with or without corresponding reductions
in other bid line items (e.g., lump sum amount for mobilization).
1088
-------
A summary of the assumptions regarding site characterization and remediation parameters for Site
A study cases is shown in Table 1. Case Al represents the actual case study for Site A. Cases A2 and
A3 represent the two hypothetical sensitivity cases considered for Site A. The incremental total
characterization cost for Case Al is for surface measurements of the extent of the tar deposits based
on visual observations. Incremental total characterization costs for Cases A2 and A3 are based on
collection of an additional sample for every 2,000 and 1,000 ft2 respectively.
The total waste quantity remediated was 2,300 tons; however, the final design estimate was only 700
tons (Case Al). This low estimate was due to the significant quantity of tar which had migrated via
subterranean movement from the original source areas. It was assumed that as a result of the
additional sampling the final design estimate would have improved to 1,600 and 2,200 tons for Cases
A2 and A3 respectively.
The adjusted remediation bid unit prices shown in Table 1 represent the expected line item values
in the winning bid (lowest responsive bid). These adjusted remediation bid unit prices decrease with
increased levels of site characterization due to quantity discounts offered by the treatment facility and
an adjustment by the remediation contractor based on the perception that actual waste quantities will
be significantly greater than estimated quantities. The effects of the quantity discounts are
represented by the Base Remediation Bid values in Table 1. The adjustments by the remediation
contractor are represented by the Adjustment to Remediation Bid Due to Uncertainty values in Table
1. The Weighted Average Unit Price Following Negotiation values shown in Table 1 are the average
unit prices which will be actually paid to the remedial contractor at the completion of the
remediation. These unit prices differ from the bid prices for Cases Al and A2 because it has been
assumed that following the discovery of the additional waste during the course of the remedial action
it will be possible to negotiate a unit price discount for the majority of the "extra" waste. To quantify
this effect it has been assumed that the Variation in Estimated Quantity clause in the Federal
Acquisition Regulation (52.212-11) applies which states that price negotiations can be initiated when
the actual quantity exceeds 115 percent of the estimated quantity. The Weighted Average Unit Prices
Following Negotiations multiplied by the actual waste quantity (2,300 tons) yields the total costs
associated with the remediation of waste line item. These costs are computed and compared with the
value for Case A3. In order to obtain relative or incremental costs, the cost for Case A3 was set to
zero and the A3 cost was subtracted from the total costs for Cases Al and A2,
A summary of the assumptions regarding site characterization and remediation parameters for Site
B study cases is shown in Table 2. Case Bl represents the actual case study for Site B while Cases B2
and B3 are hypothetical sensitivity cases. Once again, the incremental total characterization cost for
Case Bl is low because the carbon waste quantity estimate was based on surface measurements and
the assumption that the carbon waste was placed on the surface of a relatively flat area.
The total waste quantity remediated was 1,300 tons; however, the final design estimate was only 760
tons (Case Bl). This low estimate was due to numerous unexpected field conditions including: the
presence of additional carbon waste below the ground surface; the increased density of the waste due
to constant heavy rains during the remediation; and a significant increase in weight due to the
presence of large rock fragments mixed with the carbon waste at the bottom of the waste pile. It was
assumed that as a result of the additional sampling the final design estimate would have improved to
1,000 and 1,200 tons for Cases B2 and B3 respectively. The development of the Incremental Total
Remediation Costs (Table 2) by analyzing the anticipated remediation bid values was performed as
it was for Site A. It was assumed that only a small quantity discount would be realized between Case
Bl and Case B2; and that no further quantity discount would be realized between Case B2 and Case
B3. Following the start of the remedial action it was not possible to negotiate a quantity discount with
the remediation contractor in Case Bl. Therefore, it was assumed the negotiations would also not be
possible for Cases B2 and B3.
1089
-------
TABLE 1
SITE "A" - SUMMARY OF SITE CHARACTERIZATION
AND REMEDIATION PARAMETERS FOR STUDY CASES
CHARACTERIZATION PARAMETERS
Area of Concern (SF)
Sampling Frequency (# of Test Pits/SF)
Total Number of Ter Pits
Number of Samples
(TCL Semi-Volatile Organics)
Incremental Total Characterization Cost ($)
REMEDIATION PARAMETERS
Actual Waste Quantity (Tons)
Final Design Waste Quantity (Tons)
Base Remediation Bid - Unit Price ($/Ton)
Adjustment to Remediation Bid
Due to Uncertainty ($/Ton)
Adjusted Remediation Bid - Unit Price ($/Ton)
Weighted Average Unit Price
Following Negotiations ($/Ton)
Incremental Total Remediation Cost ($)
CASEA1
128,000
0
0
0
3,000
2,300
700
1,043
86
1,129
1,046
182,000
CASEA2 CASE A3
128,000
1/2000
64
64
79,000
992
987
46,000
128,000
1/1000
128
128
144,000
2,300 2,300
1,600 2,200
972 967
20 0
967
967
0
1090
-------
TABLE 2
SITE "B" - SUMMARY OF SITE CHARACTERIZATION
AND REMEDIATION PARAMETERS FOR STUDY CASES
CHARACTERIZATION PARAMETERS
Area of Concern (SF)
Sampling Frequency (# of Test Pits/SF)
Total Number of Test Pits
Number of Samples
(TCL Volatiles and PCB/Pesticides)
CASE Bl
23,000
0
0
0
23,000 23,000
1/2300 1/500
10 46
5 23
Incremental Total Characterization Cost ($)
5,000 15,000 35,000
REMEDIATION PARAMETERS
Actual Waste Quantity (Tons)
Fina
Base
Adju
Due
Adju
Weij
Folk
Incre
Design Waste Quantity (Tons)
Remediation Bid - Unit Price ($/Ton)
stment to Remediation Bid
to Uncertainty ($/Tonj
sted Remediation Bid - Unit Price ($/Ton)
;hted Average Unit Price
>wing Negotiations ($/Ton)
mental Total Remediation Cost ($)
1,300 1,300 1,300
760 1,000 1,200
1,300 1,200 1,200
30 10 0
1,330 1,210 1,200
1,330 1,210 1,200
169,000 13,000 0
1091
-------
RESULTS AND CONCLUSIONS
• The tradeoff between incremental site characterization costs and incremental remediation
costs for Site A is shown in Figure 1. The three study cases for Site A are located on a
dimensionless abscissa in order of increasing levels of site characterization. The cost points
are connected with smooth lines for illustrative purposes, however in practice these lines are
discontinuous. Shaded bands indicate a range of values which may occur based on the use of
a uniform probability distribution of values surronding the data points indicated. This figure
illustrates that the total cost (i.e., sum of the two curves) for Case A2 is lower than the total
cost for either Case Al or Case A3 by $19,000 to $60,000. This figure illustrates that there
are potentially significant total cost impacts for cases distant from the site characterization
optimum.
• The results for Site B are shown in Figure 2. Conclusions similar to those for Site A can be
drawn from these results. The total cost (i.e., sum of the two curves) for Case B2 is lower
than the total cost for either Case Bl or Case B3 by $7,000 to $146,000.
• In general, it appears that it is better to perform too much sampling rather than not enough.
When waste quantity estimates are very low the potential for severe cost increases exist,
especially if subsequent negotiations with the remediation subcontractor are less than
successful.
• The waste quantity discount structures used in this study were the prices encountered for the
specified waste materials during the specific time periods of these remedial actions. Waste
quantity discounts depend on: nature of the waste material, market conditions, absolute waste
quantities, etc. All of these factors should be considered for each site- specific situation.
• It may not be possible to perform cost-justifiable site characterization during the RD phase
due to time constraints or other conditions. In these cases it may be possible to partially
recover potential savings resulting from waste quantity discounts by structuring the bid
pricing form to request prices for a variety of possible waste quantities.
• Although not included in the above analysis, an important consideration may be the costs
associated with increasing previously authorized expenditure levels while the RA is in
progress. Both the administrative costs associated with making changes and the opportunity
costs associated with the incremental funds required could be significant.
• A secondary conclusion from the above data involves the establishment of realistic
contingencies for RAs of this type. The 8 percent and 10 percent contingencies for change
orders/claims recommended in the RD guidance for contracts below $2M and above $2M
respectively may be inadequate.
• The conclusions of this paper are that the degree of site characterization can be important for
excavation and off-site treatment. However, this may not be true for other types of
remediations such as excavation and on-site treatment. Bid prices may not be as sensitive to
estimated waste quantities for other types of RAs.
DISCLAIMER
The work described in this paper was not funded by the U.S. Environmental Protection Agency. The
contents do not necessarily reflect the views of the Agency and no official endorsement should be
inferred.
1092
-------
P
§
O
Q
LU
UJ
CC
1
f-
CO
eg
S1SOO
o spuesnoqi)
1093
-------
GO
%
LU
55
DC
O
LL
C/5
<
<
O
O
I
LU
2
LU
DC
O
CM
LU
DC
D
CD
LL
1
O
h-
Q
LU
LU
DC
P
s
O
1
DC
2
S
jo gpuesnom)
S1SOO
1094
-------
REFERENCES
1. Federal Acquisition Regulation (FAR) 52.212-7, FAC84-51, September 20, 1989
2. Richardson, T.L., P. Dappen and M.C. Ray, Estimated Versus Final Costs on Hazardous and
Toxic Waste Remediation Projects. Cost & Economics, pp 230-235.
3. Schroeder, B.R., Cost Inaccuracies in Superfund Projects: Strategies for Building Better
Estimates, pp 236-240.
4. USEPA, Guidance on Expediting Remedial Design and Remedial Action, EPA/540/G-90/006
OSWER Directive 9355.5-02, August 1990.
5. USEPA, Guidance for Conducting Remedial Investigations and Feasibility Studies under
CERCLA, OSWER Directive 9355.01, Interim Final, October 1988
6. USEPA, Superfund Remedial Design and Remedial Action Guidance, OSWER 9355.0-4A, June
1986.
1095
-------
Writing a Record of Decision to Expedite Remedial Action:
Lessons from the Delaware City PVC Project
Stephen F. Johnson
Delaware Department of Natural Resources and Environmental Control
- Superfund Branch
715 Grantham Lane
New Castle, DE 19720
(302) 323-4540
INTRODUCTION
Fortune has not been kind to the Delaware City PVC Superfund Project Remedial Action phase.
Entering its second year of construction, the project is behind schedule, over budget and completion
is like a distant star — motion towards it is hardly perceptible.
Any construction project of this scope and complexity will contain hidden problems and frustrations
that even the best engineer cannot anticipate. For instance, it has been one of the rainiest years ever
in northern Delaware; personnel have drifted in and out creating a lack of continuity. Yet there are
specific difficulties the project shares with other Superfund projects that can be attributed solely to
the administrative process. It is ultimately constructive to consider the problems on this project, to
attend to the administrative aspects we do control, and learn from past mistakes. A useful place to
look for this purpose is the remedy selection process and its record of decision (ROD). The ROD is
a convenient window into the mind set that governed the site investigation, feasibility study and
remedy selection. It is the one document that best memorializes the conceptual frame work of the
Superfund project. In the case of Delaware City PVC, many of the problems encountered in
implementation can be traced to the ROD.
The 1985 ROD is typical of its era. It contains strengths and innovations in the recovery and re-use
of resources. In some respects these innovations overshadowed the project's fundamental
weaknesses—a sketchy remedial investigation, a conceptually limited feasibility study, and
unspecified goals. Yet it has no lack of detail. Perhaps it is enough to note that the ROD specifies
well locations, diameters and pumping rates, but not soil cleanup goals. As the project proceeded,
amoeba-like, it divided in two. One effort was to meet the requirements of the ROD, the other was
to do something to improve the environment. There is surprisingly little overlap.
It is easy enough to second guess a five year old document. This paper proposes to go beyond fault
finding to a critical examination of the decision process as it actually occurred for this site. My
purpose is to show how many of the delays encountered in remedial action can be traced to the ROD
and suggest improvements to the decision documentation that will expedite remedial action on future
projects. The conclusion is consistent with the October 1990 Clean Sites proposal "Improving Remedy
Selection: An Explicit and Interactive Process for the Superfund Program". I hope to suggest
additions to those recommendations.
BACKGROUND
The Delaware City PVC Superfund Site began with a facility which has manufactured poly vinyl
chloride (PVC) since 1966. It is located in the Atlantic Coastal Plain near a major estuary. The plant
is part of an industrial complex which includes a refinery, a coal fired power station and numerous
chemical companies.
1096
-------
PVC is one of the most common and valuable polymer products and is found in a wide range of
products from containers and tubular goods to medical equipment. Worldwide production is in excess
of 11 million tons annually or about a quarter of all plastic production (Braun, 101). Relatively
speaking, PVC is environmentally friendly. It is produced from petroleum hydrocarbons derived
from natural gas which were once considered waste. The by-products of its manufacture are reusable.
PVC products are readily recycled but may be safely incinerated or landfilled.
The Delaware City facility produces about 130 tons a day of PVC resin which is then shipped to
producers to make into goods. The plant specializes in emulsion grade dispersion resin which is clear
in color and is used in medical products such as disposable gloves and syringes, tubing and fittings.
The plant is a major producer of this high quality resin. Off-grade batches are sold for rougher use,
e.g., automotive bumpers.
The plant process is typical of PVC manufacture. Vinyl chloride monomer (VCM) arrives by rail car
from Baton Rouge. It is processed in 3,000 gallon batches with de-ionized water, heat, pressure and
hydrogen peroxide as a catalyst. Polymerization occurs at 180 degrees F and 180 psi after four to six
hours in a stainless steel reactor. Unpolymerized vinyl chloride monomer boils off as the pressure is
dropped and is reclaimed for use in the next batch. The reactor vessel is then cleaned and a new
batch begun. Leaving the reactor, the PVC resembles white latex paint. It is dried by spraying into
heated air and shipped as pellets. The plant has eight reactors and operates continuously.
Environmental problems at the site originated with the handling of waste water, sludge, and disposal
of off-grade resin batches. To maintain the quality of the product, it is necessary to clean the
reactors after each batch. Until the mid 1980s reactor cleaning was performed by hand using organic
solvents. Process water and cleaning water ran through an unlined drainage ditch to a pair of
concrete aeration lagoons. There it was stripped and treated by bio-degradation. Retention time in
the lagoons was about one day. After treatment the water was discharged to a stream under a
National Pollution Discharge Elimination System permit. Storm water from the plant also ran
through ditches and collected in unlined reservoir ponds. It was pumped to the aeration lagoons for
treatment prior to discharge.
Solid wastes containing traces of solvents and vinyl chloride originated from on-site disposal of
off-grade resin batches, sludge, and solids deposited in the ditches. Every two years or so the
concrete lagoons were drained and bottom sludge removed by drag line. The sludge was buried on
site. One area containing about 25,000 cubic yards of buried sludge was capped in 1979. By this time,
EPA, the State and the responsible party all recognized the risk of ground water contamination from
the site.
The waste water treatment system has been improved in the last five years. Most importantly,
solvents are no longer used in reactor cleaning. That is now accomplished with high pressure water
wash. Also, a primary clarifier was added to the system. It eliminates solids and reclaims
unpolymerized VCM. These improvements were made independently of the Superfund project.
The leaky waste water system and solid waste burial inevitably resulted in ground water
contamination. Ethylene dichloride (EDC), trichloroethylene (TCE) and vinyl chloride monomer
(VCM) were noted in a residential well 300 yards down gradient from the plant in 1982. The
responsible parties undertook an informal remedial investigation (RI) on their own. In 1984 they
signed a consent order with EPA and the State of Delaware for a feasibility study (FS) and
implementation of remedial action plans. Note that the ROD that was eventually reached under this
agreement was not subject to the re- authorization act of 1986 (SARA). The FS began in 1984 with
an updating of the field investigation and was finished in 1986. The ROD was signed later that year.
An amendment to the consent order was added in 1987. It updated the language, required work plans
1097
-------
for design and established a schedule of deliverables. However, it did not provide stipulated penalties
for non-compliance.
The consent order of 1984 recognized two "Areas of Work" that eventually became operable units
(OUs). The plant and some adjacent property had been sold in 1981. By the time design work began
in 1987, the original owner retained nothing at the site except the liability. Complicating the matter,
it had been taken over by a third company. The project was divided between the current and i'ormer
owners of the plant. Each one proceeded in RD/RA with its own design consultants. In every
important way the site became two distinct Superfund projects. While this division of work was
administratively convenient, it has caused technical headaches in design, implementation and
especially in establishing soil cleanup goals. It also affected the human relations aspect of the project.
The responsible parties perceived that their interests were diverging and potentially in conflict. An
atmosphere of distrust affected communications and opportunities to avoid duplication were lost.
Operable unit 1 (OU 1) is the plant itself with emphasis on the waste water treatment system but also
including soil contamination from solid waste burial and sludge in the drainage ditches. The greater
part of operable unit 2 (OU 2) is ground water recovery and treatment but includes solid waste and
contaminated soil on a plot of land which was not transferred with the plant in 1981.
FINDINGS
The ROD addresses two broad areas—ground water recovery and source control. The hydrogeology
of the area presents some unusual problems and was the focus of most of the attention. About a
thousand feet inland of the site lies a pleistiocene buried river valley, a thick and highly transmissive
sand and gravel. The contaminat plume from the site moves inland perpendicularly to this valley and
divides, a branch moving in each direction along the valley. Recovery wells straddle the valley to
both ends of the plume. The recovery rate will be about 450 gallons per minute. Even so, a single
pass through of the plume is projected to take a minimum of eight years of around the clock
operation. Several pore volume flushes are thought necessary to achieve the ground water cleanup
goals.
Recovered ground water is pumped back to a site adjacent to the plant through three miles of
pipeline. The ROD called for reuse of the ground water resource as make-up and cooling water for
the plant. The plant has always used purchased utility water consuming some $70 thousand worth a
year. However, since the plant no longer uses EDC, it did not want to re-introduce it to the system.
Although the ground water was lower in dissolved solids than utility water, the ramifications of
handling deionization resin beds contaminated with EDC, TCE and VCM were unknown. The
question also arose of responsibility for spills or leaks from the ground water delivery system.
Therefore the plant declined to take the water and the other party put in an air stripper to treat the
water prior to surface discharge. The possibility remains of using the stripped water in the plant if
the two parties can reach terms. Ironically, however, re-use of the recovered ground water in the
plant was regarded by the State as an important innovation of the ROD at the time it was written.
Unlike ground water recovery, source control was shared by both operable units. The sources on OU
2 were old disposal and storage areas. Their contribution to ground water contamination was not
quantified in the RI, but they are a diminishing source. By contrast, the leaking lagoons, unlined
ditches and earthen storm water basins on the plant were continuous sources. The principal effort
at source control was lining these surface impoundments. In actual practice, the plant's elimination
of EDC and installation of a primary clarifier decreases contamination significantly before the water
ever reaches the treatment system. A secondary source on the plant was the contaminated soils in
the ditches and impoundments.
1098
-------
Soil cleanup goals. The ROD is silent on the matter of soil cleanup goals except to say that in the
work areas that became OU 1, acceptable levels will be determined at the design stage. The design
document for the drainage ditches, off-grade batch pits and storm reservoir ponds concludes that the
source of contamination is the sludge itself, not the soil, and that sludge and soil can be distinguished
from each other visually. It provides for excavation of the sludge down to the soil interface with an
additional six inches of soil taken out for good measure. This material, except for sludge that was
recoverable for re-use, was to be sent to a RCRA facility for disposal according to the ROD. Any
soil excavated for construction purposes was to be kept on site for fill material. The regulatory
agencies agreed to using the visual criterion to determine soil sample points for analysis.
Chemical analysis of samples from the OU 2 area, which had been used as storage for the resin
product, showed that the buried white waste material is not necessarily contaminated. Further testing
showed only weak correlation between visually identified waste resin and the three contaminants of
concern. There was also a period of confusion over analysis methods and detection levels. In
convoluted fashion these revelations led to acceptance of level of 2-4 ppm on OU 2 for the three
contaminants. The responsible party at OU1 briefly established a level of 5-8 ppb, the detection
limits in soil, for the same contaminants. The reasoning for trying to achieve this level of residual
contamination was obscure but apparently the party believed that EPA required it and it was feasible.
During excavation, it became clear that the 5-8 ppb criterion was not practical. EPA established
250-500 ppb for both units in order for construction to proceed. This level is thought to be
reasonably conservative, but it is at best an administrative compromise since the total quantity of
contamination in the ground and its impact on ground water remain unknown. On both operable
units, soil contamination has proven much more extensive than was determined in the informal RI.
Whereas Clean Sites recommends the development of national standards for selected contaminants,
at Delaware City PVC there was inconsistency for a time from one side of the fence to the other.
This can be attributed directly to the ROD postponing the important decision on cleanup goals to the
design stage and then the division of the design between the operable units.
Disposal. Disposal became the most contentious issue area of the project. Without knowledge of the
extent of contamination in soil on the site or of its contribution to the ground water problem, the
ROD specified disposal of "unrecoverable material" in a RCRA hazardous waste management facility
(HWMF) for the area that became OU 1. While the responsible party for OU 1 was shipping 4,000
cubic yards of soil for disposal at a cost of $1.2 million, on the other side of the fence at OU 2,
contaminated soils and resin were being scraped together in a pile and capped according to the ROD's
selected remedy. The fill is not lined however, so the contamination was to remain in contact with
soil above the water table.
Why the ROD selected two distinctly different remedies-- simple capping and hazardous waste
disposal—for the same contaminants is unclear. The description of the capping operation in the ROD
did not involve extensive earth moving, only grading of the area to be capped. In construction
however, considerable bulldozing and consolidation of soils has occurred, so it would not seem that
movement of soils about the site was the issue. If one party had responsibility for all of the
contaminated solid waste in both work areas, the logic of consolidating it in a single capped fill would
have been more apparent. The ROD did not list such consolidation as a considered alternative. It did
discuss the excavation and removal of soils and sludges on the OU 2 work area, but this was rejected
because it was much more expensive than capping which offered "comparable protection".
The RI identified only the sludge pits, impoundments and storage areas as sources, not the widespread
low level soil contamination. Consequently, data on the extent of soil contamination in the plant was
never developed. The ROD recorded the decision to dispose in a HWMF without knowledge of the
potential volume or environmental risk of the contaminated soil.
1099
-------
Liners and tanks. The considered remedies for the plant waste water treatment system were 'fix-up'
solutions from the inception of the FS. The new system would look just like the old except that its
bottom would be sealed. An alternative of replacing the leaking concrete aeration lagoons with above
ground tanks was not considered in the FS and therefore not mentioned in the ROD. After design
work was nearly complete on lining the lagoons and other impoundments, the responsible party made
a verbal proposal to replace all or some of them with above ground tanks. In the four years elapsed
since the ROD was signed, this had become its nationwide corporate policy. Experience with tanks
at other facilities had been favorable both economically and environmentally. Even though the initial
cost was higher, reliability was better than lined earthen impoundments and maintenance was easier.
However, since this approach was not mentioned in the ROD, it presented an administrative problem
for the EPA and the State agency. Could such a drastic change be accommodated in the language of
the existing ROD? What unknowns would be entailed in re-opening the ROD? Both agencies feared
a loss of momentum on the project if the ROD were re-opened just as construction was finally about
to begin. The responsible party perceived this as inflexibility. The end result was discouragement
with the tank proposal and pushing ahead to implement the liner remedy.
The surprise came when the first excavation for a new storm water collection pond was made.
Groundwater was encountered about 3 feet above its anticipated depth. The lined impoundments all
had to be re-designed to be shallower and still maintain volume. A third storm water pond was added
to make up the difference in volume. Costs increased as the square footage to be lined increased.
Space became a problem; there was barely enough room on the property for all the impoundments.
In retrospect, the proposal to use tanks where possible deserved greater attention. During construction
there were many problems with lining the impoundments. They are not simple basins. The liners are
penetrated by piping and must tie in with cement flumes and gates. Also, lining requires long periods
of favorable weather. Tankage would have been simpler to construct and would have allowed for air
emissions control in the future. Redesign of the lined impoundments and trouble with installation in
the poor weather have been the chief factors in the schedule overrun.
Ditches and pipes. Waste water ran from the reactors and cooling towers to the treatment system
through two unlined ditches called North and South. North Ditch also drains rain water from about
4 acres and South Ditch from 20 acres. Sludge in the bottom of the ditches was recognized as a
potential source of ground water contamination. All alternatives considered in the ROD involved
excavation of contaminated sludge and soil. The selected alternative was to line the ditches with a
single layer of polyethylene protected by a foot of clay with soil and sod on top. An alternative
proposal which was screened out in the FS was to use piping instead of a ditch. This was rejected
because of the possibility of solids build-up from the used process water.
In construction, the contractor proceeded with excavating and lining North Ditch while the
impoundments were being re-designed as described above. Problems arose early when it was
discovered that visible PVC resin was not just confined to the bottom of the ditch but also spread
below the soil surface beyond its present banks. Excavation for the ditch alone created more soil for
RCRA HWMF disposal than was anticipated in the FS for the entire project.
Last summer was one of the wettest in Delaware's history. Repeatedly the delicate grading of the
ditch was washed away before the lining could be installed. One afternoon shower could ruin a
week's work. The expanded excavation area was filled, compacted and re-graded several times before
it could finally be lined. Experience with rain during construction showed that the clay cover would
be subject to erosion by water running through the ditch. Consequently, a concrete bed with sealed
joints was placed on top of the liner. (A discussion of this lining system is found elsewhere in these
proceedings.)
1100
-------
Since the ROD, the plant has upgraded its treatment system to include a primary clarifier. The
process water now loses most of its solids in the clarifier and is piped directly to the aeration basins.
It is apparent that the ditch does not drain the ground it passes through but receives the rain water
it carries from a culvert at its head. In other words, a pipe would have sufficed. However, the
selected remedy was implemented as specified in the ROD. Now rain water runs 300 feet through
a state-of-the-art double lined ditch.
CONCLUSIONS AND RECOMMENDATIONS
The Delaware City PVC ROD has faults of both omission and commission. It was not specific with
regard to environmental objectives (soil cleanup goals) and it was overly specific on remedy selection.
Both sets of faults originate in an inadequate RI/FS. Two other developments exacerbated the
problems. One was the division of the site into OUs on administrative/legal instead of technical
grounds, the other was the perception that the ROD could not be changed in the face of new
information and improving technology.
The RI was strong and detailed on ground water recovery issues. However, it failed to identify
possible sources of ground water contamination fully. Without this knowledge, and without an
adequate understanding of contaminant transfer from soil to ground water, there was no technical
basis for setting soil cleanup levels. Consequently, the decision was put off.
Among Clean Sites' recommendations is "establishing site cleanup objectives and setting cleanup levels
before developing remedial alternatives". A mandate of this nature would have prevented postponing
establishing cleanup levels to the design stage where it was further complicated by the division into
operable units.
The FS with regard to the plant waste water system was too narrow in scope. The production process
was not examined for opportunities to replace solvents or remove and recycle the VCM from the
waste water stream. The FS was not informed of developments in the industry such as solids removal,
and the switch to above ground tanks. While the ROD cannot easily incorporate remedies not in the
FS, it can provide for contingencies. The Delaware City PVC ROD actually contains a good example
of this practice. For this site, the preferred treatment for ground water was use in the plant.
However, subsequent developments favored air stripping before use. This contingency was discussed
in the ROD and adopted with an Explanation of Significant Differences. Regrettably a wider range
of approaches was not considered for the waste water treatment system; there was no discussion in
the ROD of above ground tanks, the elimination of solvents, use of the clarifier or piping. The
mention of these technologies as meeting minimum requirements, as in the case of the air stripper,
would have removed the administrative obstacle of "re-opening the ROD".
Finally, for contingencies that cannot be anticipated, regulators should acknowledge the time lag
between remedy selection and implementation. In the case of Delaware City PVC it was nearly four
years. Responsible parties and remedial project managers need to continue the search for quality
improvement in the design stage and have the flexibility to adapt to better technology. For these
older RODs the need is to simplify the re-opening exercise and make everyone familiar with it. New
RODs should be clear on objectives, that is, where the project is going, but less prescriptive on how
it gets there.
The Superfund process needs the pivot of a firm decision at the conclusion of the FS to propel the
project into design and implementation. Yet we must differentiate between indecision on goals and
the flexibility to achieve good engineering. When the ROD refocuses on environmental objectives,
it will bring out the best performance from the designers. Reaching the ROD will be like lighting
a beacon, not putting on blinders.
1101
-------
DISCLAIMER
The opinions expressed in this paper are those of the author and do not necessarily reflect the position
of the Delaware Department of Natural Resources and Environmental Control.
REFERENCES
BRAUN D., "Thermal Degradation of Poly(vinyl chloride)" in Developments in Polymer Degradation,
3, p. 101 (1983).
CLEAN SITES INCORPORATED, Improving Remedy Selection: An Explicit and Interactive Process
for the Superfund Program October, 1990.
1102
-------
Site Characterization Data Needs for
Effective RD and RA
John E. Moylan
U.S. Army Corps of Engineers, Kansas City District
700 Federal Building
601 E. 12th Street
Kansas City, Missouri 64106
(816) 426-3455
INTRODUCTION
As the number of Superfund sites in the Remedial Design (RD) and Remedial Action (RA) phases
has increased, the adverse impacts of inadequate design data have become apparent. This paper
addresses: types of data most often found to be inadequate and/or have the greatest impact on
effective RD and RA, examples of why these data are needed, data needs for particular remediation
features, and suggested ways to improve site data collection and presentation. Most design problems
that result in schedule slippage and both RD and RA cost overruns result from inadequate site
characterization data. Those data gaps affect not only the high tech treatment processes but also the
more mundane aspects of remediation.
Work through the RI/FS phase is generally the domain of scientists, while engineers have the
functional lead during RD and RA. Often the engineers have very little involvement during the
problem definition or RI/FS phase and the scientists have insufficient follow up in the RD and RA
phases. As a result, too many Records of Decision (ROD) and consent decrees are accomplished
which dictate remedies which are ineffective or marginally effective, much more costly than
anticipated, or impossible to implement. Also many design engineers are accustomed to working from
a clearly defined problem, unlike those found at most subsurface and ground water contamination
sites. Therefore, it is imperative that the scientific disciplines be available through both the design
and remediation periods to better define site conditions and to interpret those conditions for the
designer in order for him or her to assure the adequacy and implementation of the design.
A few examples of problems associated with incomplete site characterization data and/or full
appreciation of site conditions are listed below:
(1) Soils properties and their handling characteristics are often poorly evaluated or even ignored
when considering various technologies. This is especially true for thermal treatment.
(2) Volatile emissions during excavation and handling of contaminated soils are often not
anticipated.
(3) Lack of information on temporal and spatial variations in contaminant loading in ground
water remediation decisions can lead to inefficient designs.
(4) No pre-ROD consideration of availability of utilities resulting in underestimation of costs.
(5) Poor understanding of the impermeability of slurry wall key layer leading to unacceptable
leakage.
1103
-------
(6) Ground water treatment processes which focus on the contaminants of interest but ignore total
ground water chemistry, especially the anions and cations present, will impact the
effectiveness of the treatment process.
(7) Solvent extraction of explosives from soil is feasible, however, the unrecognized instability
of the residue can be disastrous.
(8) Cap designs which utilize the cost effectiveness of geosynthetics but require slopes on which
geosynthetics are not stable or caps which require the use of low permeability clays but don't
evaluate the availability of suitable clay borrow material can be impractical to construct or
very costly.
BACKGROUND
Four major categories of site characterization data can be identified as needed for complete site
characterization to effectively remediate subsurface contamination, including source remediation.
These data categories are:
(1) Site Data
(2) Geochemical Data
(3) Geotechnical Data
(4) Hydrogeological Data
The term "geochemical" is used rather than the more narrow "chemical" term in order to emphasize
the importance of our understanding the chemical processes operating in the geological environment
in order to implement effective remediations. The importance of quality analytical chemistry is
already well understood and appreciated, however, our understanding of ongoing chemical processes
needs improvement. The following paragraphs identify some commonly overlooked data requirements
and include examples of problems resulting from the data gaps.
Site Data needs are often overlooked in the pre-ROD/consent decree phase and even well into design.
Unforeseen cost increases, time delays, and contract modifications can and do result. Some common
data needs include:
(1) Topographic Surveys - The need should be readily apparent, however, this aspect is often
overlooked. In some instances, available general topographic mapping is used without
verification. Consequently during RA, excavation or fill overruns or underruns or impossible
site drainage are discovered which require contract modifications. Property boundary surveys
and adequate horizontal and vertical controls are also included in this category.
(2) Utility Availability - Water, gas, power, and sewer services required for remedy
implementation must be identified. In addition, leaking industrial sewer lines might be
contamination sources and previously unidentified utility lines crossing a remediation site
can cause contract shutdown pending their relocation or protection.
(3) Borrow Availability - In some areas suitable borrow is scarce. The costs of trucking suitable
material from a distant borrow pit will add significant cost and transportation problems if not
recognized. As an example, a 50-acre cap with an average of 3 feet of soil, requires almost
250,000 cubic yards or approximately 14,000 truckloads of suitable earth borrow material.
1104
-------
(4) Transportation Network - The proximity of suitable roadways and/or rail lines is important
to remedies requiring the transportation of heavy equipment and earth materials into the site
or contaminated or treated wastes from the site. Local opposition to frequent heavy truck
traffic and damage to streets and roads, especially through residential areas, must be
anticipated.
Geochemical Data collection can often be improved upon to more confidently select effective
remedies and better effect quality RD and RA. Some examples include:
(1) Multiple Sampling Rounds - In too many cases, remediation decisions are made which are
based on single or poorly timed, multiple ground water sampling rounds. Time allowed for
RD often doesn't provide for seasonal sampling. As a result, chemical loading may exceed
treatment plant capability, or the plant may be overdesigned, or the operating plan is not
optimized to accommodate variations in loading.
(2) Anion/Cation Analysis - These analyses are inexpensive, yet if they are overlooked in ROD
preparation, the designed treatment train may be either more expensive than anticipated or
ineffective if not detected during RD. Eh, pH, and TOC are other chemical parameters which
can affect effective RD.
Geotechnical Data must be gathered for many types of remedies, both for purposes directly related
to the remedial process and for design auxiliary to the actual remedial process, such as building
foundation design or excavations.
(1) Soil Moisture Content - The natural moisture content of site soils, especially fine-grained
soils, is valuable information both in the pre-ROD and RD phases. As examples, the moisture
content of contaminated soil to undergo thermal treatment affects fuel consumption and the
moisture content of a fine-grained foundation soil can be an indicator of the soil's strength
and consolidation characteristics.
(2) Atterberg Limits - These parameters define the plasticity of fine-grained soils, give the
geotechnical designer an early indication of the strength of that soil, especially when
evaluated with moisture content, and can be an indicator of contaminated soil handling and
processing characteristics. The test is relatively inexpensive but the results can be very useful.
(3) Soil Strength Parameters - Generally not needed prior to the RD phase. Some design features
requiring soil strength testing include structure or building foundations, significant
excavations, dredging, and slurry wall trenches. Blow counts from Standard Penetration Tests
can be used for an early indication of soil strength.
(4) Gradations - Some representative gradation or particle size distribution analyses done in the
RI/FS phase can be very helpful in estimating approximate permeability and for designing
efficient monitoring wells. Gradations are required for the design of such things as collection
drains and withdrawal wells and in evaluating soils handling and processing characteristics.
(5) Excavatability - While there is no one test or set of tests to define this design parameter,
valuations and judgments should be made in the pre-ROD phase concerning excavatability
when excavations of any kind are required in the remedy. Excavatability includes such
factors as whether the material can be machine excavated, the necessity for blasting, the
existence of large boulders, the need for dewatering, etc.
1105
-------
(6) Landfill Settlement - Remediations often include capping an existing landfill and perhaps
incorporating a gas collection and venting system. Many such landfills are still settling with
attendant surface disruption capable of adversely impacting the effectiveness of the cap and
vent system. Carefully surveyed settlement data collected throughout the RI/FS phase is
invaluable for remedy selection and as design data. Settlement data collection should continue
through RD and RA and into the operations and maintenance phase if displacements are
continuing and significant.
Hydrogeological Data is routinely collected both during the RI/FS and RD phases. However, several
aspects will be discussed which are sometimes slighted but can be very important to selection of an
effective remedy and to proper design and implementation.
(1) Multiple Water Levels - In order to understand the hydrogeological character of the site in
sufficient detail to select an effective remedy, it is important that enough water levels be
obtained to define both the vertical and horizontal flow directions seasonally and as they
respond to both natural and manmade recharge and discharge. We are working at a ground
water contamination site in the Plains States at which the regional flow is severely distorted
locally by irrigation pumping during several months of the year.
(2) Detailed Stratigraphy - In too many cases, stratigraphic detail has not been well developed due
to poor sample recovery often coupled with too infrequent sampling intervals, lack of
geophysical logs, improper sampler selection, field geologists poorly trained in logging
methods, or combinations of the above. Even relatively minor variations in lithology have a
strong influence on contaminant migration and plume development. This is an important
factor during pre-ROD, RD, RA, and even into the operation and maintenance phase of both
ground water and vadose zone remediation.
(3) Secondary Porosity Features - Joints, defoliation planes, bedding planes, root holes, etc:., often
strongly influence the overall gross permeability of bedrock materials and fine-grained soils,
especially clays. In too many cases these features are not targeted during site exploration and
if they are, the vertical features are difficult to intercept and analyze. Careful consideration
of these features is warranted during the RI/FS phase and remedy selection for problems such
as contaminated bedrock aquifers, multiple stacked aquifers, and slurry walls keyed into an
"impermeable" layer. For sites such as these, additional characterization will also be needed
during RD.
The various types of site characterization data discussed in this paper are not needed or at least not
to the same degree for all features of site remediation. The following remediation features were
considered:
(1) Withdrawal & injection wells (8) Landfills
(2) Internal drains (9) Thermal treatment
(3) Slurry walls (10) Soil washing
(4) Slurry wall key layer (11) Excavations
(5) Caps (12) Dredging
(6) Chemical Stabilization (13) Vapor extraction
(7) Ground water treatment
Table 1 presents a summary of site characterization data determined to be useful or needed for
remediation. The table also suggests in which phase or phases of the remediation process it is
advantageous to acquire the data.
1106
-------
O
C
0
Ul
cr
ID
r—
Ul
U
"Z.
O
h-
o
U)
Ul
CC
Ul
_J
CD
t-
UOJ4OD4X3
J A
6ui6D9jn
UOJ4DADOX3
6UJMSOM
110$
4U9UU4O9JJ,
IDLUJ9U 1
SIIIJ-PUOT
4U9lU4D9J_l
J940M
punojQ
uoj4DZjiiqD4.S
IOOIUU91JO
sdo'i
^
^©X IIDM
s||O/y\
f, JJD|$
SUJOJQ
IOUJ9J.UI
SIOM
UOJ4.O9fU| ^
i M ri jpit 1 1 i *i
1 /»» w««i^<^^> | yyi
J—
o
V
ro
— *
ro
•>
—
«•»
ro
•
ro
—
ro
ro
..
ro
-•"
Topographic
Surveys
—
^ ^
_
—
-*-
J
Eo
4- >
Z3<
—
—
CM
CM
OO
LO
O>
DD<
0400
—
—
•v
—
—
—
_
C
O
4^
O
4-
O-*
wo
,_l ~
DO
^.
»
ro
—
•«T
ro
—
—
^-
_
*r
—
^.
ro
^
0>
_c
a
o
i/>
II
ro
—
«T
ro
—
—
—
^r
—
^j.
ro
•>
C
O
Is
xw
c>
2o
CC
«
^>
ro
—
ro
— *
ro
ro
—
ro
*
ro
— *
ro
—
ro
CM
—
ro
__
~~
L
4-
W4-
I!
= c
oo
LOO
UI9L|OO99
—
—
J-^
ro
•
rO
—
ro
CM
ro
ro
_
ro
*—
O>
U
a>
^
LW
0)4-
tl
<_j
CM
ro
».
c\
CM
CM
CM
CM
ro
CM
rsj
CM
Son Strength
Parameters
—
—
^
ro
•>
CM
—
ro
—
ro
ro
ro
—
ro
Gradations
—
—
_j_
—
—
__
—
4-
"B
o
4-
o
0
o
X
Ul
040Q
DO1UMO94O90
T
»
ro
—
ro
*
ro
—
4-
C
CD
L 0?
*T~~~*
XH-
oa>
-JOO
^]-
•l
rr
—
ro
*
—
v.
1
ro
—
—
T
"J
ro"
*
ro
—
T
ro
4-
O
aC
J)
|I
2_J
—
ro
»
—
—
ro
ro
—
Detailed
Stratigraphy
ro
»
ro
»
ro
ro
»
ro
—
ro
(0
D
4-
0
*-
te
C<0
oo
OL
a>o
IOCL
0400 |OOj6o|
-09DOJp^H
c
4-
c
o
E
-p
c
o
c
o
4-
o
L
. Q)
4|~V
LJ.
CC O
o> o>
o
4-
o
o
o
_9
o
o
4-
w
c
Jr
1 —
•o
a>
•o
c
CC £>
c c
— o
0 4-
4- O
O 0)
X? =
o
X? 0
4- O
1i Q
1107
-------
DISCUSSION
Site characterization should be an iterative process beginning with preliminary assessment and site
investigation, continuing through RI/FS, RD and RA, and in some cases, into operation and
maintenance. We identify information needs, collect data, analyze the data, identify additional
information needs, collect new data, evaluate the new data and re-evaluate old data, etc. This process
operates in varying degrees in the investigation, design, and remediation of contaminated sites. The
effectiveness and degree to which it is utilized is dependent on the commitment of the client or
program manager to the iterative process, the technical competence of those performing the work,
and the timely input of the appropriate technical specialist.
There is a saying which states "We see what we know." In retrospect, the wisdom of this saying is
apparent in the hazardous waste remediation program as we have progressed from it's infancy to what
it is today. A very simple example is what we would see in an aquifer sample as it comes from the
sample tube. Ten years ago, the hydrogeologist might have seen sand with reasonably high
permeability and a likely plume migration pathway. Today's more knowledgeable hydrogeologist
sees a cross-bedded, clean, medium-grained sand with thin, clayey sand interbeds and flecks of
organic carbon. He or she sees the sand as a plume pathway but recognizes the cross-beds as potential
downward DNAPL migration routes, wonders about the impact of the organic carbon, and sees
adsorbtion potential in the clays of the interbeds. A geotechnical engineer tasked with designing a
slurry wall sees a sand not likely to have significant slurry losses, he or she visualizes what the
gradation of the mixed clean and clayey sands might be and how suitable that mixed material might
be for trench backfill, and recognizes that the calcareous powder on the tip of the drive shoe together
with less than full sampler penetration likely represents the presence of boulders. All have looked
at the same sample but see it differently based on their knowledge and experience.
Our challenge is to gather and report as much of the necessary data as possible to satisfy the needs
of all of the specialists involved in the identification, evaluation, design, and remediation of
contaminated sites. Furthermore, we must attempt to accomplish this task in a cost and time effective
manner. In the case of subsurface contamination problems, exploration (drilling, sampling, and well
installation) is one of the most costly and time consuming activities. It is incumbent upon us to
maximize the amount of information obtained from each hole and to utilize the field staff to gather
as much surface site information as possible at the time they are in the field gathering subsurface
data. It is much less costly to anticipate what the likely future data needs are and to collect some of
those data at the RI phase than to have to remobilize to the field and drill and sample new holes in
the early RD phase to gather that data. That is redundant and costly in time and money. As an
example, a few Atterberg Limit or gradation tests from a chemical sampling or monitoring well hole
takes almost no time and adds very little cost while the benefits are significant.
There are steps we can take to achieve the broader site characterization which we now know to be
needed. The process must include a means of recognizing the total data needs and evaluating the risks
of something less then full site characterization. The first step involves bringing together experienced
representatives of the multiple disciplines involved in the total RI through RA process to discus;s and
summarize their data needs and to explore methods of collecting and reporting these data in the most
cost effective manner. The end product might be a site characterization summary with a checklist
of data needs for various contamination scenarios and likely or selected remedies. Quantative or at
least semi-quantative contingencies for specific data gaps associated with particular remedial features
should be included. The contingencies would give decision makers an idea of potential cost impacts
caused by incomplete site characterization and help them better evaluate the benefit to cost ratio of
additional investigation.
The invisible walls separating the various disciplines must be lowered and communication encouraged.
The walls are caused by a number of things: technical jargon, professional jealousy, lack of
1108
-------
understanding of the role of the other team members, physical separation, poor leadership, the
pressure to meet short deadlines causing narrow vision, etc. All must communicate in terms
understood by others and be sure their needs are understood. Environmental problem solving has
brought together specialists who have not had long working relationships and roles are still being
defined.
Perhaps we can learn from the civil engineering profession. Over the past 60 to 70 years, the
engineering geology specialty has developed and matured to serve that profession. The engineering
geologist uses his or her knowledge of geological processes to paint a clear and concise picture of
geological site conditions related to the work. The geologist must develop an understanding of basic
civil engineering and the physical properties of earth materials in order to recognize and evaluate
those geological features which will affect the proposed project. Their reports must be understood
by the civil engineer to be effective. The need for a similar specialty discipline(s) is apparent in the
area of subsurface nation investigation and remediation, especially the site characterization aspects.
There is indication that the specialty may be developing, however, we should recognize the need and
actively work to promote its maturation. The very effective transition into this area of work by
several firms specializing in engineering geology and the closely related geotechnical engineering
branch of civil engineering reflects the applicability of the applied science approach. A few
universities offer engineering geology or geological engineering and some have done a good job of
modifying their curricula to focus on environmental applications.
In order to obtain the complete site characterization so very important to the evaluation and
remediation of hazardous waste sites in the most cost effective manner, site characterization specialists
are needed. These specialists should have solid foundations in geology, hydrogeology, or chemistry,
and training in the basics of civil, chemical, and environmental engineering and the other scientific
disciplines mentioned. The additional training may be either formal course work or on-the-job.
Their function would be to investigate, evaluate, and report site conditions in light of the needs of
the decision maker and the designer. The challenge is great in that the growth in the environmental
field has been explosive. The most broad based and knowledgeable people must be made available
to train and support the many bright but inexperienced people so that they know in order to be able
to see and report.
CONCLUSIONS
Site characterization is a very important factor in the identification, evaluation, remedy selection,
design, and remediation of subsurface contamination sites. Proper characterization is required to
define the health risk, select and effect a remedy, and to assure cost effectiveness. As more sites are
remediated, the need for more complete site characterization to meet the stated goals becomes more
apparent. The task requires satisfying the needs of multiple disciplines in the most effective manner.
Our challenge is to assure recognition of the need by the client or program manager and to develop
specialists knowledgeable of those needs and capable of adequately characterizing the site in a cost
effective and understandable manner. The EPA RPM and regional technical specialists are in a
position to be leaders in encouraging and assuring the necessary integrated iterative approach for
satisfying the data needs for effective RD and RA.
1109
-------
New Bedford Harbor, Massachusetts
Review of the Remedial Investigation/Feasibility Study Process
and its Impact on Remedial Design/Remedial Action
Mark J. Otis
U.S. Army Corps of Engineers
424 Trapelo Road
Waltham, Massachusetts 02254
(617) 647-8895
and
Mary C. Sanderson
U.S. Environmental Protection Agency
J.F. Kennedy Building
Boston, Massachusetts 02203
(617)573-5711
INTRODUCTION
New Bedford Harbor is located in southeastern Massachusetts and consists of over 17,000 acres of
estuary, harbor and bay. Bottom sediments are contaminated with polychlorinated biphenyls (PCBs)
and heavy metals, with PCB levels exceeding 100,000 parts per million (ppm) in some spots. The site
was placed on the National Priority List in 1982 and numerous investigations and studies have been
carried out since that time. The site was divided into operable units in the fall of 1989 and a Record
of Decision (ROD) was signed for the "hot spot" area in April 1990 which calls for dredging and
incinerating of approximately 10,000 cubic yards of the most highly contaminated sediments.
Remedial design for the hot spot is underway. The Feasibility Study for the remainder of the site was
released in August 1990 and the ROD is scheduled for mid 1991.
The Army Corps of Engineers is responsible for remedial design and remedial actions at this site and
has also been extensively involved in the Remedial Investigation/Feasibility Study (RI/FS) process
through the performance of an Engineering Feasibility Study (EFS) and Pilot Study which evaluated
dredging and dredged material disposal methods. The Pilot Study, which involved on-site dredging
and disposal of contaminated sediments, introduced the local community, state and other groups to
the technical aspects of the project at an early stage.
The New Bedford harbor site is unique in both its physical features as well as in the technical and
political/institutional challenges associated with its remediation. Numerous decisions made during
the RI/FS stage will effect the remedial actions and deserve to be reviewed for consideration at
similar large, complex sites. These include the decision to perform the extensive studies which
focused on dredging and dredged material disposal, the participation of the state/local community in
the early stages of the project and the decision to divide the project into operable units. This paper
reviews the RI/FS period and discusses the extensive evaluations performed by the Corps of
Engineers and their impact on the ongoing remedial design work, as well as the eventual remedial
actions.
1110
-------
BACKGROUND
SITE DESCRIPTION
New Bedford, Massachusetts is a port city located in southeastern Massachusetts (Figure 1) where site
investigations conducted in the late 1970's found PCB contamination in various locations throughout
the harbor. Further investigations identified two electrical capacitor manufacturers as major users
of PCBs from the time their operations commenced in the late 1940's until 1977, when EPA banned
the use of PCBs. These industries discharged wastewaters containing PCBs directly into the harbor
and indirectly via the municipal wastewater treatment system. (1)
Additional field studies carried out since the late 1970's have shown PCB concentrations in marine
sediment to range from a few ppm to over 100,000 ppm. Water column concentrations were found
in excess of federal ambient water quality criteria. Fish and shellfish PCB concentrations were found
in excess of the U.S. Food and Drug Administration tolerance limit of 2 ppm for edible tissue. In
addition to PCBs, heavy metals (notably cadmium, chromium, copper, lead) were found in the
sediment in concentrations ranging from a few ppm to over 5,000 ppm. (2)
As shown in Figure 1, the site is divided into three geographical areas, the Acushnet River Estuary,
the Lower Harbor and Upper Buzzards Bay. The estuary is an area of approximately 187 acres which
is bordered by the Wood Street Bridge to the north and the Coggeshall Street Bridge to the south.
Contamination is highest in this portion of the site with PCB levels in the sediments generally greater
than 50 ppm and exceeding 100,000 ppm in the hot spot which is located at the northern end of the
estuary. Metals concentrations reach 5,000 ppm in this portion of the site.
The Lower Harbor area consists of approximately 750 acres which extends from the Coggeshall Street
Bridge south to the Hurricane Barrier at the harbor entrance. Sediment PCB concentrations are lower
in this area and range from below detection to approximately 100 ppm. Metals levels are also reduced
with a maximum level of approximately 3000 ppm.
The Upper Buzzards Bay portion of the site extends south from the hurricane barrier, encompassing
an area of approximately 16,000 acres. Sediment PCB and metals concentrations are considerably
lower in this portion of the site but several localized areas near sewer and stormwater outfalls have
sediment PCB concentrations that exceed 50 ppm. (2)
REMEDIAL INVESTIGATIONS/FEASIBILITY STUDIES
New Bedford Harbor was added to the National Priorities List in July 1982. This resulted in EPA
performing a comprehensive assessment of the PCB problem in New Bedford and led to a Feasibility
Study of remedial action alternatives for the Acushnet River Estuary portion of the site. This FS was
completed in August 1984 and presented five clean-up options for the estuary portion of the site.
Four of these options involved dredging and on-site containment of the contaminated sediments.
EPA received extensive comments on these options from other federal, state and local officials,
potentially responsible parties, and the general public. Many of these comments concerned the ability
of a dredge to remove the contaminants, the environmental impacts of dredging, and the long term
effects of onsite containment of contaminated sediments. EPA decided that additional study was
necessary and had the Corps of Engineers perform extensive evaluations of dredging and dredged
material disposal alternatives for the estuary portion of the site.
1111
-------
iBoston
Upper
Buzzard's
Bay
Not To Scale
Figure 1
New Bedford Harbor, Massachusetts
1112
-------
Subsequent to the initiation of this work by the Corps of Engineers, EPA began work on an additional
FS to provide a range of remedial alternatives for the entire site. This effort was expanded when the
site was divided into operable units in the spring of 1989. The first operable unit addressed the 5 acre
"hot spot" located in the northern portion of the site. A Feasibility Study was prepared for this area
as well as for the remainder of the site. The results of the Corps of Engineers work were incorporated
into and forms a critical component of these studies.
CORPS OF ENGINEERS STUDIES
The Corps of Engineers was initially requested to perform an Engineering Feasibility Study of
dredging and disposal alternatives. A major emphasis of the EPS was placed on evaluating the
conceptual design of dredging and disposal alternatives, their implementability, and their potential
for contaminant releases. The scope of the effort included field data collection activities, literature
reviews, laboratory and bench scale studies, engineering and economic analyses, and analytical and
numerical modeling techniques to assess engineering feasibility and to develop conceptual alternatives.
The objectives addressed in the EFS involved:
* developing a baseline characterization of the Acushnet River Estuary through sediment
sampling, hydrographic and topographical surveys and measurements of the hydrodynamics
and ongoing sediment/chemical transport,
* assessing the magnitude and migration potential of contaminant releases due to resuspension
of sediments during proposed dredging operations,
* performing laboratory and bench scale testing developed specifically for dredged material to
gather technical data needed for predicting the behavior of the dredged sediments if placed
in either confined disposal facilities or contained aquatic disposal sites, and
* combining the technically feasible dredging and disposal technologies into implementable
alternatives and providing concept design cost estimates for each implementable alternative.
Early in the course of the EFS, the Corps recommended and EPA recognized the benefits of including
a field evaluation of dredging and disposal alternatives to supplement the laboratory and modeling
efforts of the EFS. This was particularly appropriate for the evaluation of dredging technologies,
which are difficult to simulate or model and whose performance is highly dependent on site specific
factors or conditions. (3)
A pilot project was performed in the Acushnet River Estuary during 1988 and 1989. The project
evaluated the effectiveness of three types of hydraulic dredges, a confined disposal facility and a
contained aquatic disposal cell. The confined disposal facility was a diked retention basin constructed
on the New Bedford shoreline. Contained aquatic disposal involved dredging a cell or pit in the
harbor bottom, filling this cell with contaminated sediment then capping the cell with clean sediment.
Data generated as part of the EFS were used to design the components of the pilot project, to estimate
contaminant release to surface water and groundwater during the pilot project, and to provide the
basis for the monitoring and evaluation program for the project.
CAPPING VERSUS DREDGING
The merits of capping the contaminated sediments in place versus dredging with onshore/shoreline
containment has been raised as an issue repeatedly at this site. In particular, the principally
1113
-------
responsible parties (PRPs) prefer a capping alternative and have presented a complete remedial
alternative using capping for EPA's consideration. Several factors considered in weighing the capping
alternative against other remedial alternatives include the following:
* the impacts of capping on the bathymetry of the shallow water estuary,
* the overall advantages and disadvantages of a containment versus a removal alternative, and
* the extensive long-term restrictions necessitated by the capping alternative, and the affiliated
operation and maintenance requirements.
The EPS and Pilot Study evaluated contained aquatic disposal which includes a capping component.
The information obtained during these studies was used in our evaluation of the PRP
capping proposal and in the development of capping alternatives which appeared in the FS for the
Estuary, Lower Harbor/Bay.
DISCUSSION
RESULTS OF CORPS OF ENGINEERS STUDIES
The EPS resulted in the conceptual design of several cleanup alternatives for the estuary portion of
the site. These alternatives were evaluated for their implementability and potential for contaminant
release. Contaminant release estimates were provided for each alternative as well as for the various
components of the alternatives. The information was also used in the design of the Pilot Study. The
Pilot Study consisted of the on-site evaluation of three types of hydraulic dredges (cutterhead,
horizontal auger, and Matchbox) along with two disposal alternatives (confined disposal facilities,
contained aquatic disposal). The study was conducted in the estuary portion of the site and involved
the removal of approximately 10,000 cubic yards of sediment. (3) The activities were intensively
monitored with the focus to:
* determine the dredge's ability to remove the contaminated sediment from the harbor,
* determine the sediment resuspension and contaminant release caused by the dredging
operation,
* determine the movement of contamination away from the immediate vicinity of the dredging
operation, and
* evaluate contaminate release associated with the disposal activities.
Monitoring for impacts to water quality throughout the harbor during dredging operations was also
a critical component of the Pilot Study in addition to the monitoring to address the techni-cal
objectives of the study. Physical, chemical and biological monitoring techniques were utilized before,
during and after the dredging operations. The monitoring found only localized impacts that were
attributable to operational or meteorological events.
The major technical finding of the pilot study are outlined below.
1114
-------
* The dredges could remove the contaminated sediment while minimizing overdredging. Initial
PCB levels of 200-500 ppm were reduced to approximately 10 ppm with the removal of
approximately an 18 inch layer of sediment.
* Contaminant release can be restricted to the immediate vicinity of the dredging operation.
Levels of total suspended solids and PCB in the water column returned to background levels
within 500 feet of the dredging operation.
* Dredge operating techniques were developed to meet the objectives of minimizing
overdredging and contaminant release.
* Monitoring techniques were developed and implemented that obtained data to address the
technical objectives of the study and provided assurance that operations were not degrading
conditions throughout the harbor.
The information developed from these studies was incorporated into the EPA Feasibility Study for
both the "hot spot" and the Estuary, Lower Harbor/Bay portions of the site. The input enhanced the
presentation of the operational and cost aspects of the alternatives and provided contaminant release
estimates. The site specific nature of the data generated through the pilot study increased our
confidence in these numbers and significantly decreased the unknowns as we move into remedial
design/remedial action.
APPLICATION OF STUDY RESULTS TO REMEDIAL DESIGN
Much of the information obtained from the Pilot Study will be directly applicable to the remedial
design for the "hot spot" operable unit, as well as for the remedial design and action for the Estuary,
Lower Harbor/Bay portion of the site. Major components that are being directly applied to the "hot
spot" remedial design include:
* A cutterhead dredge was selected during the pilot study as the piece of equipment best suited
for work in New Bedford Harbor. This dredge will be specified for use in the hot spot along
with specific operating procedures developed during the study.
* Sampling procedures and monitoring protocols developed and implemented during the pilot
study will be utilized to monitor water quality conditions throughout the harbor during hot
spot remediation. Monitoring will be conducted by a separate government contractor.
* The experience gained in constructing the confined disposal facility will facilitate any future
CDF designs associated with the remedial action for the Estuary, Lower Harbor/Bay.
The overall cost of the remedial design will be reduced, along with any uncertainty over the
effectiveness of these procedures.
STATE/LOCAL COMMUNITY INVOLVEMENT
Considerable concern and opposition was voiced with the release of the August 1984 Feasibility Study
which proposed alternatives that included dredging. As mentioned previously, these concerns focused
on the ability of dredges to remove the contaminated sediments and the environmental impacts
associated with the operations. The studies performed by the Corps of Engineers were designed to
address the technical questions, but an equally important decision was the involvement of the other
1115
-------
federal, state and local agencies in the process leading to a Record of Decision. A project group
headed by EPA was formed and met monthly over the course of the study period to discuss project
progress and to allow input into decisions being made in the course of the project. Numerous detailed
technical presentations were made as information was obtained through the course of the studies. The
group played an important role in the planning and implementation of the pilot study. The group was
also exposed to many technical issues which may not have surfaced until the remedial design phase
of the project. These include:
* the construction of disposal facilities along the shoreline,
* contaminant levels within the effluent discharged from these disposal facilities,
* contaminant release associated with dredging and disposal operations, and
* appropriate monitoring techniques and action levels.
The pilot study also allowed the project group to view the construction activities that would be
associated with fullscale remediation. "Open houses" were held for the local community workgroup
and other interested individuals to view the work. The study highlighted the operational constraints
that effect our ability to address the technical concerns highlighted above. As we move into remedial
design/remedial action, the experience of the pilot study and the information gained from it should
provide a firm foundation for proceeding with the remedial design phase of the process.
OPERABLE UNITS
The decision made in the spring of 1989 to divide the project into operable units was also critical.
The first operable unit involves the hot spot which is a 5 acre area in the northern end of the
Acushnet River Estuary which contains approximately 45% of the PCBs present on the site. The
remainder of the site includes over 1000 acres with widely varying PCB levels. The hot spot provides
the opportunity to address a large percentage of the contamination in a relatively small area. This
approach has accelerated the remedial design/remedial action schedule. Reducing the time between
the completion of the pilot study and the start of remedial activities is an important point in terms
of public perception. It will also allow the pilot study site and facilities to be utilized in the remedial
action, thereby reducing the cost of the design and construction effort. Further phasing of the hot
spot design and remediation will allow for quicker implementation of the remedy. The first phase
will address site preparation, allowing site activities to begin prior to the completion of the design of
the complex water treatment, incineration, and ash handling portions of the remedy.
CONCLUSION
Sites like New Bedford Harbor are complex both technically and administratively. Technical
challenges at the site include the physical features, widespread contamination and its unconfined
nature. Administrative challenges result from the communities effected, the numerous state and
federal agencies with a regulatory role and the unique nature of the site. As more sites like this are
identified, lessons learned at New Bedford can be applied to allow for a less complex process leading
to site remediation. The major points to emphasize include:
* The advantage of specific studies, preferably pilot studies to address the site specific concerns
regarding both the effectiveness and impacts of remedial action.
1116
-------
* The involvement of the other groups (state, local communities) in the process from the very
early stages.
* The step by step approach to a large complex site proceeding from studies to discrete operable
units to expedite the remediation process yet to allow a learning process as the project
proceeds.
Numerous reports prepared for the New Bedford Harbor site address the questions of dredging and
disposal methods and their effectiveness. The information may be applicable to ongoing work at
other sites. Copies of these reports are available from the authors.
REFERENCES
1) E.C. Jordan, 1989 "Draft Final Hot Spot Feasibility Study, New Bedford Harbor" Portland,
Maine
2) E.C. Jordan, 1990, "Feasibility Study of Remedial Alternatives for the Estuary and Lower
Harbor/Bay, New Bedford Harbor, Massachusetts" Portland, Maine
3) Averett, Daniel E., Otis, Mark J. 1990, "New Bedford Harbor Superfund Project, Acushnet
River Estuary Engineering Feasibility Study of Dredging and Dredged Material Disposal
Alternatives; Report 12, Executive Summary," Technical Report EL-88-15, U.S. Army
Engineer Waterways Experiment Station, Vicksburg, MS
1117
-------
The Pre-Design Technical Summary
Kenneth R. Skahn
Design and Construction Management Branch
U.S. Environmental Protection Agency
Mailcode OS-220W
Washington, D.C. 20460
(703) 308-8355
INTRODUCTION
The Pre-Design Technical Summary (PDTS) is a compilation of available site information prepared
by the remedial project manager (RPM) to provide the designer with a clear understanding of the
technical objectives of the remedial action. Guidance is being developed in the Design and
Construction Branch on preparation of the PDTS. This paper will provide a summary of that
guidance.
The objective of developing a PDTS is to provide a smooth transition from the Record of Decision
(ROD) into the design process. The preparation and use of the PDTS should ensure that the designer
will understand the technical objectives of the design as well as provide the designer with an up-to-
date inventory of all available information that may be pertinent to the design. The PDTS also will
serve the RPM as the initial building block for developing a comprehensive statement of work for the
remedial design.
At a minimum the PDTS should accomplish the following:
• define initial site conditions:
• describe the selected remedy;
• summarize available data;
• identify applicable regulatory requirements; and,
• state all known unresolved issues.
The Remedial Investigation/Feasibility Study (RI/FS) and ROD will be the sources for much of the
information to be summarized or referenced in the PDTS. However, the guidance will identify
additional site-specific information that may be known to the RPM or RI/FS contractor that is not
included in the RI/FS or ROD but should be included in the PDTS.
BACKGROUND
Remedial designers, including ARCS (Alternative Remedial Contract Strategy) firms, the USAGE
(U.S. Army Corps of Engineers), and the USER (U.S. Bureau of Reclamation), recently stated the
need for a document that provides a concise summary of all significant site-specific information used
when transitioning from the ROD into remedial design. The Superfund Remedial Design and
Remedial Action Guidance1 manual issued in June, 1986, called for a "Pre-Design Report" to be
prepared by the lead RI/FS party and provided to the lead design party. The stated objective of the
Pre-Design Report is "to describe the engineering parameters and institutional concerns of the selected
1118
-------
remedy, and package all pertinent information for effectively transferring the project to the lead
design party." The RD/RA guidance manual, however, provides little description on what
information the Pre-Design Report should contain, and it is likely that few Pre-Design Reports were
ever prepared and used.
The U.S. Air Force has seen the need for this type of site-specific transition document and now
prepares a "Requirements and Management Plan (RAMP)2" prior to negotiation of the design contract
for new construction projects. The RAMP addresses such topics as project design information, site-
specific requirements, environmental issues, access information, and long-range base planning.
The Design and Construction Management Branch began developing Pre-Design Technical Summary
guidance because it was apparent that the summarization of site-specific information would serve
several significant purposes. The PDTS will serve the RPM both as a building block in developing
a comprehensive design statement of work and by ensuring that the designer fully understands the
objectives of the remedial action. The PDTS will serve the designer by providing an up-to-date
inventory of data. Use of the PDTS also should alert the RPM and designer to data gaps and help to
avoid delays by identifying, early on, any potential road blocks such as property access and
acquisition needs, permits to be obtained, or unresolved issues. The document also could prove to be
an invaluable source of information that can be used to maintain continuity in the event there is a
change in RPMs or if there is a significant delay between issuance of the ROD and start of design.
COLLECTION OF THE PDTS INFORMATION
For Fund-lead projects (i.e., those projects financed by Superfund) it will be the responsibility of the
RPM to either collect or oversee the collection of the Pre-Design Technical Summary information.
For potential responsible party (PRP) lead sites, the PRP can be required to collect the PDTS
information before finalization of the Administrative Order of Consent (AOC). The PRP would be
responsible for collecting and submitting the PDTS information to the RPM for review and approval;
the information would then be used to develop the Statement of Work to be included in the AOC.
The collection of PDTS information is equally important for a PRP-lead site in that it will ensure that
all parties involved in the AOC (as well as the PRP's designer) fully understand the objectives and
scope of the remedial design and remedial action.
Collection of the PDTS information should begin before or shortly after the ROD is signed. For
Fund-lead projects, it may be useful for the RPM to arrange a meeting with experienced regional
staff, the RI/FS contractor, and state and local officials familiar with the site to discuss and
collectively develop most of the statements.
The PDTS information should be kept brief, using bullet points and tables to present data. Supporting
information can be referenced or included as attachments. The information can be compiled simply
(e.g., a checklist) or as a more detailed formal document, depending mainly on the complexity of the
site. The sources of information to be included in the PDTS should be well documented.
CONTENT OF THE PDTS
A draft guidance document3 has been developed by the Design and Construction Management Branch;
the guidance includes an outline of the information to be addressed by the PDTS. Each outline
element is fully explained and examples are often provided. For simple design projects, many of the
items need not be addressed--the content should be modified according to the complexity of the
RD/RA.
The outline provided in the draft PDTS guidance document is as follows:
1119
-------
PRE-DESIGN TECHNICAL SUMMARY
I. Site Conditions
A. Site description
1. Site history and current status
2. Chemical, physical, and geological characteristics of site
3. Proximity to homes and schools/land and groundwater use
surrounding site
4. Basis for property lines on drawings
5. Likely future use of site
B. Real estate issues
1. Real estate requirements assessment
2. Restrictions or special agreements on easements or access roads
C. Availability of utilities
1. Location and availability
2. Existing agreements or conditions
II. Selected Remedy
A. Description of selected remedy
B. Selected cleanup levels
III. Availability of Data
A. Physical/chemical data collected to date
B. Data retrieval
IV. Technology/Design Approach
A. Waste characterization
B. Treatment scheme
1. Schematic diagram
2. Pre-treatment requirements
3. Treatment design criteria
1120
-------
C. Long-term monitoring requirements
D. Sole source or first time usage of a technology and innovative/SITE
technology
E. Treatability study
F. Special design limitations
G. Flexibility in design
H. Schedule constraints that could impact rate of treatment or unit size
I. Confirmation monitoring
V. Materials
A. Volume estimation and basis of calculations
B. Spatial requirements, staging, etc.
C. Durability of materials
D. Materials/equipment availability
E. Mixed materials
VI. ARARS/Permits/State Involvement
A. ARARs list
B. On-site versus off-site waste management
C. Permits for off-site actions/land use restrictions
D. Extent of State involvement
VII. Unresolved Issues
VIII. Health and Safety Concerns
IX. Other Concerns
A. Community relations activities
B. Confidential information
C. Other RD/RA requirements
X. Appendix
1121
-------
A. Bibliography--existing site information
B. References
STATUS OF THE PDTS GUIDANCE
Initial development of the PDTS guidance document began with a meeting of a work group comprised
of representatives from USAGE, USBR, several design firms, and the Design and Construction
Management Branch. This group met to discuss the types of information that should be addressed
in a PDTS; i.e., the major site or design related data or information that was often inadequately stated
or not provided when projects were turned over to the designer. A guidance document incorporating
the suggestions of the work group was drafted. A draft of the PDTS guidance was sent to Regional
Superfund Branch Chiefs for review in late November 1990. Comments have been received, and the
guidance is being revised in consideration of those comments.
The PDTS guidance will not be issued as a "stand-alone" document but will be incorporated into a
more comprehensive guidance document pertaining to "scoping remedial design" that also is being
prepared by the Design and Construction Management Branch. This new document will include
guidance on developing the Remedial Management Strategy (addressing contracting strategies, phasing
alternatives, funding constraints, and roles of participants), preparing statements of work, establishing
schedules, and cost estimating. Drafts of the "scoping remedial design" guidance document will be
reviewed by an existing work group that currently is revising the 1986 Superfund Remedial Design
and Remedial Action Guidance. A draft of the "Scoping Remedial Design" guidance is scheduled to
be prepared by September, 1991.
SITES WHERE PREPARATION OF A PDTS HAS BEEN REQUIRED
Although PDTS guidance is still in the developmental phase, a Pre-Design Technical Summary was
prepared for a site in Region VII—the Groundwater/Surface Water Operable Unit, Galena Subsite,
Cherokee County, Kansas4. The PDTS was prepared by the RI/FS contractor under the direction of
the RPM. The RPM found the document to be very useful in that it provided the designer (USAGE),
which had no prior knowledge of the site, with detailed information as to what EPA wanted to
accomplish at the site. The PDTS proved to be a valuable source of much of the information needed
to begin the design.
Another PDTS is being prepared in Region VI in response to a requirement in an Administrative
Order of Consent (AOC)5. The AOC requires the PRPs to prepare and submit a PDTS to EPA for
review and approval. The RPM made minor modifications to the text of the draft PDTS guidance
to reflect the fact that the PRPs will be preparing the PDTS. The modified guidance was then made
an attachment to the AOC.
CONCLUSION
The purpose of developing a PDTS is to provide the designer with a clear understanding of the
technical goals and objectives to be achieved by the remedial design. The PDTS also will serve to aid
the RPM in developing a comprehensive statement of work for design.
The intent is not to place an added burden on the RPM but to ensure that the information provided
by the RPM to the designer is as complete as possible and that the resulting design effort will be as
free from misunderstanding as the RPM can make it.
1122
-------
DISCLAIMER
This report has undergone a relatively broad initial, but not formal, USEPA peer review. Therefore,
it does not necessarily reflect the views or policies of the Agency. It does not constitute any
rulemaking, policy or guidance by the Agency and cannot be relied upon to create a substantive or
procedural right enforceable by any party. Neither the United States Government nor any of its
employees, contractors, subcontractors or their employees makes any warranty, expressed or implied,
or assumes any legal liability or responsibility for any third party's use or the results of such use of
any information or procedure disclosed in this report, or represents that its use by such third party
would not infringe on privately owned rights.
We encourage your comments on the utility of this paper and how it might be improved to better
serve the Superfund program's needs. Comments may be forwarded to the attention of Kenneth W.
Ayers, Design and Construction Management Branch, USEPA, Mailcode OS-220W, Washington DC
20460.
REFERENCES
1) USEPA, Superfund Remedial Design and Remedial Action Guidance. OSWER Directive
9355.0-4A, June, 1986, pages 2-6,7,8.
2) U.S. Dept. of the Air Force, Construction Technical Letter (CTL) 90-1: Management of the
MILCON Planning and Execution Process. March 6, 1990.
3) USEPA, Guidance for Preparation of a Pre-Design Technical Summary (Draft). November
27, 1990.
4) USEPA, Predesign Technical Summary for the Groundwater/Surface Water Operable Unit
(Draft). Galena Subsite. Cherokee County. Kansas. June, 1990.
5) USEPA, Memorandum, Subject: Region 6 Example of How to Incorporate the Pre-Design
Technical Summary into an Administrative Order, (From David A. Weeks to Ed Hanlon),
March 25, 1991.
1123
-------
VIII. DESIGN ISSUES
1124
-------
ACCELERATING THE ROD TO REMEDIAL ACTION PROCESS:
sand Creek Industrial Superfund site (OU1), Commerce City,
Colorado
Authors:
Brian Pinkowski, EPA (Principal Author)
Bruce Hanna, URS Consultants, Inc.
Mikkel Anderson, Brown and Caldwell Consultants
(formerly with URS Consultants, Inc.)
May, 1991
INTRODUCTION
A goal of the United States Environmental Protection Agency (EPA)
has been to reduce the length of the average Remedial Design and
Remedial Action (RD/RA) in the Superfund site cleanup process.
This paper compares the cost and duration of RD efforts from other
Superfund sites to the RD for the Sand Creek site.
The Sand Creek Superfund Industrial site (Sand Creek) is located in
Commerce City, Colorado, a suburb north of Denver. (Figure 1) .
The site and surrounding area are primarily occupied by trucking
firms, petroleum and chemical supply/production companies,
warehouses, and small businesses. There is a small residential
population in the study area which is adjacent to the northeast
border of the site. The portion of the site for which the RD
effort has been completed was a former pesticide and herbicide
manufacturing facility.
EPA Region VIII commitments required that the RD for the site be
completed within nine months of the Record of Decision (ROD).
Facilitating the task was the Region's decision to waive
negotiations with the Potentially Responsible Party (PRP) due to
lack of financial viability. The RD was for an incineration/
demolition/ and soil vacuum extraction (SVE) remedy expected to
costs $7-8 million. The RD package was completed in six months.
The RD effort, accomplished with URS Consultants, Inc. URS (the
ARCS contractor), included nearly $500,000 of additional field work
not originally provided for in the Remedial Investigation and
Feasibility Study (RI/FS).
The Sand Creek RD was completed within six months of the ROD
signing and ranks within the fastest 20% of the 437 completed RDs
across the nation. The intent of this paper is to discuss the
planning, scheduling, and implementation of the Sand Creek RD
effort in comparison with current EPA guidance for streamlining the
RD/RA process as provided in OSWER Guidance.1
1125
-------
Colorado
Denver
Site Vicinity
URS
CONSULTANTS
1126
USEPA
Sand Creek
Remedial Design
Site Location Map
Figure No. 1
-------
BACKGROUND
The Sand Creek Superfund site comprises approximately 480 acres and
contains four known contamination source areas; The Colorado
Organic Chemical Company property (OU1), the L.C. Corporation acid
pits (OU2), the 48th and Holly landfill (OU3), and the area-wide
ground water contamination associated with the Sand Creek
Industrial Superfund site area (OU4). (Figure 2).
The OU1 area was used to manufacture pesticides from 1960 to about
1968, under the name of Times Chemical. Since 1968, when a fire
destroyed three of the buildings on the site, several health
agencies have found unacceptable conditions at the plant. These
have included unsatisfactory waste management practices and worker
safety conditions, violations in storage and handling of flammable
liquids, and soil containing high levels of thermally-altered
pesticides and other chemicals. A second fire occurred at the
plant in 1977. In 1984, in response to an EPA order, the Colorado
Organic Chemical Company removed waste drums and contaminated soil
and fenced-off the area, including an area just north and east of
the Colorado Organic Chemical Company property, which has been
affected by contaminated surface runoff.
The primary contaminants found in the OU1 area are:
Arsenic
Chromium
Dieldrin
Heptachlor
Chlordane
2-4 D
4,4 DDT
The remedy to be designed by URS was for the following: the
excavation and off-site incineration of approximately 1,000 cubic
yards of soil contaminated with greater than 1,000 ppm of HOCs
generally composed of 2-4 D; the demolition of the contaminated
buildings and structures for off-site disposal of the debris; and
vacuum extraction of the VOC-contaminated subsurface soils as a
ground water contamination source control measure.
The EPA's guidance for expediting RD/RA work suggests the
development of a remedial management strategy document to specify
project goals and determine project phasing. The Sand Creek
project utilized the ROD for the description of project goals and
the ARCS work assignment Work Plan to determine project phasing.
The specific language of the Statement of Work provided to URS
prior to development of the Work Plan was designed to provide for
maximum contractor flexibility. The EPA Remedial Project Manager's
role was to oversee the development of the RD package and to
expedite the administrative review and approval process for the
1127
-------
1
V
t-
-------
numerous documents produced in the RD effort.
The specific tasks which were provided to URS in the Statement of
Work are as follows:
It is critical that the RD efforts anticipated under tasks #1, #2,
and #3 be completed by March 15, 1990. It will also be necessary
for URS to begin work for tasks #1, #2, and #3 during development
of the work plan for this work assignment.
1. Prepare design specifications for excavation and off-site
incineration of those soils contaminated with greater than 1000 ppm
concentrations of HOCs. This activity will include soil sampling
to determine the extent of those soils with concentrations above
industrial use action levels and those soils contaminated with
greater than or equal to 1,000 ppm HOCs. Air and airborne
particulate monitoring before, during, and after excavation
activities will be necessary to assess potential impacts on the
surrounding area. Design of an air monitoring plan will be
necessary for this task.
2. Prepare design specifications for remediation of those soils
approximately five (5) feet below the soil surface, using vacuum
extraction technology. The soil contamination to be remediated
with vacuum extraction is primarily from volatile organic
compounds. This task will include design of the treatment system
for the extracted gasses, as well as design of an air monitoring
plan for the vacuum extraction remedial actions.
3. Prepare design specifications for demolition and off-site
disposal of the buildings and possibly the storage and formulation
tanks on the property. This activity will include sampling the
buildings and tanks to determine the type of disposal unit
necessary for the debris.
4. URS shall provide assistance to the EPA's community relations
efforts as needed. This is likely to take the form of providing
assistance at public informational meetings, and providing
photographs of remedial actions similar to that which will be
designed under this work assignment.
Task 1 included additional soil sampling because the RI/FS for the
site covered the entire 480 acres and did not focus primarily on
the OU1 area. One of the results of the RI/FS was to divide the
site into operable units. The information in the site-wide RI/FS
was sufficient to identify the COC area (OU1) as the area of
immediate concern due to the severity of the contamination. The
site-wide RI/FS was also adequate to select a remedy for the OU1
area, but lacked sufficient detail necessary to proceed to RD. As
an example, soil incineration appeared warranted, but limits of the
excavation had not been delineated.
1129
-------
The OU1 Area (shown in Figure 3) is in the northwest part of the
Sand Creek site and is in a zone of low moisture and moderate
climate at the north edge of the Denver Metropolitan area, in a
political subdivision called Commerce City. The site is situated
on a series of low soil benches grading toward Sand Creek to the
North. The soil is generally sandy, silty with some clay lenses
and contains some cemented outcroppings. It is bounded on the
north by the Colorado and Eastern Railroad tracks and on the east
by Dahlia Street. The south boundary abuts property owned by
Asamera Oil Company and is approximated by a fence line. The
western boundary is a fence separating the site from a gravel
processing facility. A large warehouse under separate ownership is
on the site and has been occupied during the period the RI) was
prepared.
Several other buildings, tanks and pads, are located on the site
which were used in the manufacture of pesticides and herbicides by
Colorado Organic Chemical Company (COC) and its predecessors. One
building is occupied by the former owner of COC and is being used
as an industrial real estate office. Most buildings show moderate-
to-high levels of contamination.
Surface soils contain a variety of chemical products and byproducts
including pesticides, herbicides and small amounts of thermally-
altered products including dioxin. Evidence of compliance with
earlier cleanup orders is apparent where the top few inches of soil
were removed after the 1977 fire. Some poorly drained areas showed
high concentrations of HOCs.
Subsurface soils show some high concentrations of VOCs, semi
volatiles including tentatively identified compounds, other organic
compounds and metals. A zone immediately above the groundwater is
heavily contaminated with petroleum residues, and in some parts of
the site, a free-phase material floats on the groundwater surface.
Groundwater is found in a relatively complex system 15 to 45 feet
below surface. The groundwater contains much of the same
contamination as is found in the subsurface soils and the plume of
floating material.
Access to most of the site is controlled by a locked fence.
During the course of the RD, access to the site by the EPA and the
contractor was limited to the sampling activities, surveying and
the vacuum extraction treatability study. Periodic management
visits were conducted for quality assurance and supervisory
purposes.
The principal participants in the RD are the EPA (Region VIII
Superfund Branch), the Colorado Department of Health, URS and its
subcontractors, Brown and Caldwell Consultants (Sampling and
Analysis), Datum Exploration (drilling), Groundwater Technology,
1130
-------
c
&D
'35
c ^
QC
I
ac
O
m
ac
HI
Q.
O
QC
Q.
I
•
I
I
Ul
ff
O
cc
Q.
•a
c
eo
o
a>
0.
c
8-
u
o
O_J
1131
-------
Inc. (vacuum extraction pilot testing), and Shannon and Wilson
(geotechnical testing).
DISCUSSION
To facilitate the eventual contracting for RA, the Sand Creek RD
was divided into the four separate parallel tasks as shown in the
Statement of Work. These were:
1. Excavation and incineration of approximately 1,000
square yards of soil.
2. Vapor extraction of the sub-surface soil.
3. Demolition and disposal of the tanks and
contaminated structures.
4. Air monitoring before and during the RA.
Task number 4 combines the requirements for air monitoring
specified in the first three tasks. Note that the original scope
of work included a fourth task of community relations assistance,
which was deleted from the work assignment during the development
of the Work Plan. At that time, the fourth task of air monitoring
was substituted.
Rather than preparing one large set of plans and specifications for
all tasks, the four tasks were deemed too diverse to attract
sufficiently competitive bids for a single contract. Each was
developed into a separate set of contract documents for that
specific task. This was developed with the assumption that the RA
would be assigned to either the U.S. Army Corps of Engineers or an
EPA prime contractor, who would most likely oversee the remedial
work but subcontract some or most of the specialized tasks.
Superfund guidance specifies the employment of a multi-step process
to be followed in a typical RD:
Work Planning;
Data Acquisition;
Sample Analysis/Validation;
Data Evaluation;
Treatability Study;
Preliminary Design - 30%;
Intermediate Design - 60%;
Pre-final/Final Design - 90-100%; and
Post-Remedial Design Support.
As the work assignment was received in September, 1989, and the
deadline for completion of the RD was scheduled for March 15, 1990,
some elements of the work were required to be started almost
1132
-------
immediately. Figure 4 is a composite bar chart schedule for a RD,
showing the progression and interrelationship of the design
elements. It also contrasts the "fast track" schedule pursued at
Sand Creek with a "normal" RD. One of the major differences
between the two is the parallel approval steps at milestones where
work does not halt to await approvals. At Sand Creek, the
contractor was in close contact with the EPA and others in the
approval system to identify items of the design deliverable which
were likely to be modified. Phone conversations were frequent and
face-to-face meetings occurred weekly. Such interaction is vital
to the success of an accelerated schedule.
It was determined that the field sampling and analysis elements had
the longest lead times and that preparatory work could begin on
certain design elements before the laboratory results were
completed. Therefore, the Sampling and Analysis Plan (SAP) and the
Quality Assurance Project Plan (QAPjP) proceeded apace with (and
somewhat ahead of) the Work Plan. To gain more control over the
schedule of receipt of laboratory data, it was decided that non-CLP
laboratories would be used to the greatest extent. Laboratory costs
would therefore be part of the project budget, rather than
accounted for separately, as is the typical practice.
By the time the Work Plan was submitted in late October,
mobilization activities for the field sampling effort were ongoing.
The sampling work began on November 1, 1989, and was essentially
complete one month later. For the most part, weather remained
favorable during this period, and the work was completed without
incident.
Following the field work, over-lapping of design tasks went into
effect. As the analysis and data evaluation were performed, the
design of the excavation, vacuum extraction, demolition and air
monitoring were progressing.
The successful completion of design depended upon the results of
the testing to provide scope, areas, quantities, and difficulty of
remediation. It was therefore planned that the early stages of the
design should be developed with a great degree of flexibility to
accommodate unforeseen requirements and variances from early
assumptions. The final design data report was not published until
March, almost concurrent with the 90 - 100% final design delivery.
During the analysis, preparation and evaluation of the data,
however, the contractor, his subcontractors and the EPA worked in
close communication so that the trends and preliminary conclusions
shaped the design. Where further data was required to verify
conclusions or to fill gaps, decisions were made to rapidly acquire
samples and perform quick turn-around analyses. Discussions took
place prior to the end of the Work Plan development, allowing a
realistic budget to be developed which anticipated unexpected field
sampling results.
1133
-------
; 1
: ' «
g
0*
w
O
g
.. |
: 1
I
; i
i i
'IS
o
. 1
: g
* V5
•i:
1
j
Illlllllllllll
Y/////2
•E*
Project Planning (RD)
Ov
I
ll
1
|
|
III
l^
1 Conduct Data Acquisition
,1
= ^
1
11
'1
I8
1 Sample Analysis/Validation
s
11
=
=
S
1 Data Evaluation
1
I Perform Treatability Study
5i
|
^
S=
|
i
i
I Preliminary Design 30%
|
11
ll
Intermediate Design 60%
1
5t
|
=
=
II
1
1 PreFinal/Final Design 90 - 100%
A
ss
t
1
1
=
=
l!
ws —
I Post Remedial Design Support
1134
-------
Similarly, work proceeded on the plans and specifications for the
soil vacuum extraction system without waiting for the final vacuum
extraction pilot test report, which was also delivered in March.
The progress of the test and its early findings were communicated
to the design team prior to completion of the pilot test report,
allowing the newly acquired information to be promptly integrated
into the design. The level of advance planning taking place during
Work Plan development allowed the contractor to incorporate last
minute information while minimizing the risk of repeating efforts.
The schedule contained in the Work Plan for the RD is shown in the
bar chart (Figure 4). Actual performance is also shown in hatched
lines. As shown, there are no major variances except that the RA
did not follow immediately after the design as was planned. This
was due to factors beyond the control of the project participants
and relating to the State Superfund contract for the State of
Colorado's 10% share of RA costs. The RA work assignment is now
under way.
Whenever a project is described as "schedule driven" or "fast
track," it is particularly important to recognize the presence of
two distinct classes of needed information: what you know is
missing (the known unknowns) and what you haven't thought of or
can't yet conceive of needing (the unknown unknowns). The planning
process must prepare for each and retain sufficient flexibility to
accommodate a reasonable response to the intermediate findings.
Unexpected Problem I
For Sand Creek OU1, the SAP is a good example. The planning
team considered three primary data needs: the existing data,
data needed to support the range of design options and those
data that might alter the entire scope of the RD, such as an
unexpected dioxin discovery. The first category of data needs
seems fairly obvious, but in reality, the longer existing data
has been in the files, the more suspect it becomes. Primary
data such as boring logs, lab analysis reports, and Quality
Control (QC) runs get collated into summary reports with all
the customary typos and interpretive biases. A case in point
at Sand Creek was the Task l, halogenated soil removal, which,
after significant retrospection, turned out to be based on one
grab sample under a dripping tank tap. Unfortunately, the
tank was long gone and the exact sample location unrecoverable
because sample locations had not been surveyed. With total
unit costs of potentially up to $2,600 per cubic yard for
incineration alone, precise quantities of contaminated soil
requiring incineration are very important. The project team
knew that the data point strongly suggested a problem (2-4 D
at 1.5% by weight), but also needed to better define the
boundaries of contamination to control excavation (known
unknown). Consequently, the plan had the highest sample
density in this vicinity, but had additional areas of
1135
-------
increased coverage surrounding the hot spot to be sure the
project team would not miss the area which approximates the
action level for the soils and 1,000 ppm HOCs contour, the
limit of excavation. In this way, the sampling approach weis
tailored to the design data needs for each area of the sites.
When the design process was nearing completion, sampling
yielded two unsuspected results. The initial data showed much
lower levels of halogenated substances, none above 1,000 ppm.
Two explanations seemed possible, either the initial data
point identifying a problem was erroneous or the problem was
smaller than the first sampling grid. At this point, the team
also knew that there was not a problem too large for the
chosen remediation option. A supplemental sampling on an even
finer grid was devised and executed, ultimately locating a
small pocket of contaminated soil reguiring off-site thermal
destruction.
Unexpected Problem II
The site sampling efforts prior to the RD also revealed a
major unknown. Although it was generally known that the site
had been screened for dioxins (a common micro-contaminant in
some phenoxy pesticides) no existing data confirmed the
contaminants' presence. Since this was suspect result for a
pesticide facility, a large confirmation sampling effort was
undertaken. A coarse grid surface sampling was accomplished
and samples analyzed using guantitative techniques for the
2,3,7,8 isomer. This effort showed the chemical to be present
below action levels and located on the uphill edge of the
site, substantially away from the area of manufacture.
However, the incinerators targeted for the site's soils were
not licensed for wastes containing any dioxin.
Since the ROD had no named remedy for dioxin contaminated
wastes and the dioxin hits were on the fringes of the Operable
Unit boundary, the dioxin soils issue was not included in the
RD/RA effort and it will be addressed by the PRPs as a
separate effort. The solution was a classic "work around."
This preserved the integrity of the ROD's logic, the schedule
and budget.
Unexpected Problem III
Another unknown occurred at the installation of the vacuum
extraction pilot test equipment. To everyone's surprise, a
free-phase hydrocarbon layer (commingled with pesticide
contamination) previously unreported or detected, was present
in one of the observation wells with an unknown lateral
extent. After some review, it was concluded that the presence
of the large amount of hydrocarbon would threaten the
economics of vapor extraction (Task 2) by competing with the
1138
-------
targeted volatile halocarbons for space on the activate
carbon. Further, the impact would be on operation costs, i.e.
increased carbon filter change out rates, and not on the
constructed size of the extraction units. Impact on the other
tasks appeared minimal because construction staging mandated
that the site be cleared first (Task 3) for access for Task 2,
vapor extraction. Task 1, the soil removal, need not be
effected either way. Therefore, by proceeding to completion
on all design tasks, the only significant impact of the free-
phase hydrocarbon on the ultimate timing of the soil vacuum
extraction task: either immediately after Task 3 completion
or after removal of the free-phase hydrocarbon.
Unexpected Problem IV
Schedule evaluation was a constant task and was reevaluated at
each new discovery. Not all discoveries were in the field.
Early planning of the RA indicated two procedural tasks with
potentially significant impacts on the schedule:
subcontractor procurement and laboratory data turnaround. In
the case of procurement, the Federal Acquisition Regulations
were mandatory. Certain bidding steps and approvals are
specified. By scheduling these in detail, the team identified
several instances where procurement was the critical path.
Early emphasis was placed on subcontractor bidding for
drilling services. With the URS procurement staff working
closely with the EPA contracting officers, procurement efforts
met or exceeded scheduling needs.
Planning of the schedule also revealed a potentially fatal
flaw in laboratory analyses turnaround. Although the EPA
Contract Lab Program was initially targeted to handle the
sample flow, careful examination of the total data package
needed indicated that numerous requests would have to go
through the EPA's Special Analytical Services (SAS) which
requires a deliberate bidding process among program contract
laboratories.
Charting out the time necessary to procure laboratories under
SAS, it was quickly evident that the project schedule would be
heavily impacted. An alternative of using a combination of
EPA Regular Analytical Services and URS team lab
subcontractors offered the best apparent schedule. Although
this combination of services put more analytical costs in the
project budget, the project stayed within the authorized
funding and met the scheduling objectives.
CONCLUSIONS
The authors feel that the Sand Creek OU1 RD demonstrates several
significant conclusions.
1137
-------
1) Remedial Designs can be accomplished very quickly in
situations where the physical size and scope of the
remediation is at least within the order of magnitude of the
size and scope of the options chosen in the ROD.
2) Careful planning and scheduling of all aspects of the effort
is important and should include paying continuous attention to
updates throughout the RD.
3) Any design sampling efforts should be focused on objectives
oriented to the needs of the designers. This is addressing
the known unknowns.
4) Expect the unexpected. Do not be surprised if previous
sampling results cannot be exactly duplicated. Have in place
the communication pathways, the technical resources and budget
contingencies to react quickly to surprises. View each in
terms of its potential impact on the project's chosen remedies
and on the schedule. Decide if the issue can wait to be
addressed at a later phase. Move forward on what, is
unaffected.
5) Conduct frequent team meetings with all active contractor,
State, local government, and EPA staff. Keep this limited to
the key players.
While the authors recognize the Sand Creek GUI's technical
challenges may be uncomplicated when compared to some other
superfund designs, the lessons learned seem universal: divide the
project into manageable units. Adopt a reasonable sequence of
remediation events. Conduct detailed planning and scheduling.
Continuously monitor schedule performance against plan. Treat
scheduling as a key objective. Be prepared to work through or
around inevitable surprises. These concepts are not new. Anyone
familiar with conventional construction management will see the
similarity. The results of Sand Creek RD demonstrate that the
techniques described can be successfully applied to remediation
design despite the large amount of technical uncertainty that
usually accompanies remediation efforts.
1138
-------
REFERENCES
1 OSWER Directive 9355.5-02 (EPA/540/G-90/006, Guidance on
Expediting Remedial Design and Remedial Action).
2 EPA Introduction to Remedial Design Schedule Management, EPA
Course held in Washington, DC, June 1989.
1139
-------
coc
EPA
HOCs
OU1
OU2
OU3
OU4
PRP
QAPJP
RD/RA
RI/FS
ROD
Sand Creek
SAP
SAS
VOCs
LIST OF ACRONYMS
Colorado Organic Chemical Company
United States Environmental Protection Agency
Halogenated Organic Compounds
Colorado Organic Chemical Company Property
L.C. Corporation Acid Pits
48th and Holly Landfill
Area-Wide Groundwater Contamination at Sand Creek
Industrial Site
Polenticilly Responsible Party
Quality Assurance Project Plan
Remedial Action/Remedial Design
Remedial Investigation and Feasibility Study
Record of Design
Sand Creek Superfund Industrial Site
Sampling and Analysis Plan
Special Analytical Services
Volatile Organic Compounds
1140
-------
Remedial Design of Superfund Projects:
What Can Be Done Better?
John D. Holm, P.E.
U. S. Army Corps of Engineers
Kansas City District
601 East 12th Street
Kansas City, Missouri 64106-2896
(816) 426-5655
INTRODUCTION
The remedial design phase is a critical component of the Superfund process. Remedial design follows
the Remedial Investigation (RI) and Feasibility Study (FS) components of the Superfund process and
builds upon the knowledge base established by those activities. The purpose of the design process is
to produce plans and specifications that can be implemented by a construction contractor. A
successful design should produce a remedial action consistent with the goals stated by the Record of
Decision (ROD). The design process is subject to a wide variety of factors that often influence the
direction of the design, and ultimately the success of the remediation. The intent of this paper is to
generally identify these factors and present some thoughts on how to improve the remedial design
'phase of a Superfund project.
BACKGROUND
What is a successful project? To some it may be completing the project within the designated
schedule, to others it may be achieving a predetermined level of quality in the design, while others
may judge the design by the total cost of the project. A successful project can generally be defined
as a project completed:
within the allocated time period
within the budgeted cost
at the proper performance or specification level
These goals appear straightforward, but in reality a Superfund project is evaluated by many different
groups, each of which may have different concepts of success. One government agency may evaluate
success based on achieving set goals within an established schedule and budget. Another government
agency may judge success based upon adherence to all applicable regulations, or by the design
resulting in minimal changes to any administrative decision making process.
The local community may evaluate success based on an elimination of health threats, real or
perceived, to their community with minimal disruption to their daily lives and with little regard for
the financial cost to the project. The designers may view success as achieving a comprehensive design
that results in a minimum of change orders or claims. Potentially responsible parties may view success
as cleaning up a site with minimal cost and regulatory interference or by a reduction in environmental
liability. A construction contractor may view success as completing a project ahead of schedule while
maximizing profits. Obviously, some of these concepts of success may be in conflict with each other.
By the time a Superfund project moves into remedial design it has been through an entire sequence
of investigations, studies, reports, and meetings with intense technical, administrative, public, and
legal scrutiny. The process culminates in the preparation of a ROD which describes both the site and
the chosen remediation methods. Unfortunately, all these studies and investigations do not guarantee
a thorough enough knowledge of the site to effectively complete design. As the remedial design
1141
-------
begins a variety of decisions have to be made about items such as scheduling and funding. Oftentimes
these decisions are made based on past experience, on guidelines, or on political or administrative
realities.
DISCUSSION
Design Factors
As the designer begins the task of design, questions often arise as various issues develop. Sometimes
these issues can drastically affect the direction the designers must take. The scope of the project may
begin to change as new factors such as additional site investigation information or new regulatory
requirements generate problems not envisioned by the FS or the ROD. Schedules and budgets can be
affected and soon the project schedule begins to slip, or expenses begin to overrun programmed
budgets. Decisions then have to be made; obligate more money, extend the schedule, find a solution,
ignore the problem, etc.
The factors that arise during a Superfund remedial design may be grouped into one of several general
categories:
Technical
Administrative
Budgetary
Political
TECHNICAL - Technical factors are probably the most easily defined, but are often not fully
resolved due to scheduling, budget, or technical limitations. Examples of such factors might be the
inability to obtain representative soil samples for design, air emissions that may need tight controls
during the remedial action, or groundwater levels that may interfere with excavation and treatment
of soils. The factors are generally resolvable; however, additional time and money may be required.
Unfortunately, there is a real tendency to leave these factors to the construction contract or if
resolution during design will result in a schedule slippage.
ADMINISTRATIVE - Administrative factors are often less easily defined and tend to be kept hidden
from public scrutiny. For example, a project exists for which the ROD was prepared prior to SARA
and which requires solidification of organic waste. Current technical knowledge indicates that
stabilization of organic wastes is often not effective. However, because re-examining the selected
remedy will require reopening the ROD, the design is proceeding utilizing solidification of the
organic waste as the primary remedy.
BUDGETARY - By the time a Superfund project moves into the design phase there have generally
been years of studies performed and hundreds of thousands of dollars spent. After that investment,
everyone wants to see something accomplished. As a result, the design is oftentimes expected to be
completed in a short time frame and on a programmed budget that may not fully account for the
complexities, or realities, of the project.
POLITICAL - Politics have a very real influence on the Superfund design process. Designers are
often under substantial pressure to produce results. The results are often measured in very simplistic
terms (i.e., 'bean counts'). These bean counts often begin to take on a life of their own to the
detriment of the project. The local community may also have a substantial impact on what is done.
For example, the EPA had agreed to perform various emissions control activities on a particular site
during design because of local concerns. As the initial work was accomplished, the analytical results
clearly showed no health problems existed and the emissions control activities were not necessary.
However, because of the promise made to the local citizens, the emissions control activities proceeded
at a cost of many tens of thousands of dollars and substantial wasted effort.
1142
-------
Personnel and Approach to Designs
The Superfund program has resulted in the creation of a huge new industry. Challenges and
opportunities abound. Unfortunately, the number of experienced personnel fall short of the demand.
The experience level of the average hazardous waste professional is low. The older, more experienced
personnel are few and far between.
Design standards do not exist or are very subjective. The work is often poorly defined. As a
profession we have forgotten, or have never learned, a lot of the basics associated with sound design
and construction. We have become enamored with powerful analytical tools, voluminous reports,
endless studies, and detailed schedules at the expense of appropriate design methodologies and sound
engineering judgement. Designers should not forget that it is the basics, such as soil properties or
groundwater elevations, that can drastically impact construction methods and efficiencies.
Construction experience is a must for designers and for managers. Designers must be learning from
their mistakes. One of the best means for a designer to observe problems with a design is to follow
its progress in the field and talk to the people administering or performing the work. No matter how
thoroughly one reviews documents a continuing presence on the site will be revealing to the designer.
It is important for project engineers and managers to be involved during construction. If designers
sit in offices and never see the problems being encountered in the real construction world, many
valuable lessons will not be learned.
Interaction of Designers
The Superfund program is challenging in that a wide variety of disciplines such as geologists,
geotechnical engineers, mechanical engineers, project managers, chemists, industrial hygienists,
chemical engineers, environmental engineers, toxicologists, civil engineers, and electrical engineers
are involved. On any given project each discipline will have to interface with many of the other
disciplines. Each participant in the design process needs to have an understanding of what functions
the various disciplines perform. Failure to communicate with the other participants deprives the
project of a fully functional design team. The most dangerous person on any project is the "lone
ranger" who either believes he knows everything or is not willing to communicate with the other
members.
Level of Design Effort
There seems to be a concept held by some people that design is little more than photocopying a
previous project, changing the names, and sending it out for a contractor to perform. This may be
true on some jobs, such as small underground storage tanks, where the work is repetitive. As projects
become larger, more complex, and less standard the concept of photocopy design becomes less
realistic.
Design takes time. Most Superfund projects spend years in the RI, FS, and ROD stage; then design
is stipulated to be performed within a nine month schedule. Design schedules need to be carefully
considered and must take into account project complexities and the number of unknowns. Setting
design schedules solely to meet administrative needs with little consideration of actual design concerns
will generally create problems later in the design process.
Types of Specifications
Specifications are generally written as either a performance based specification or as a detailed
specification, or some combination thereof. In very simple terms, performance based specifications
state the end product desired and give the contractor flexibility on how to achieve those results. A
1143
-------
detailed specification gives the contractor clear instructions on what is wanted, how it will be built,
what it will look like, etc.
There are advantages and disadvantages to both. With the performance spec the contractor must be
provided sufficient information and direction that the job can be adequately evaluated. The
contractor has much greater latitude in determining his strategy, equipment, and personnel. If the
contractor's method doesn't work then the contractor is responsible for finding an alternative. With
a detailed specification, sufficient design information must be available for the designer to assure that
the project is constructable. If the contractor builds the design as specified and it doesn't work, it
is the designers responsibility, not the contractors. Detailed specifications will provide a specific
product; however, a greater responsibility is borne by the designer.
Construction claims are sure to follow when the design fails to adequately address the site conditions
and the remediation technology. Because of the complexities in hazardous waste work, and the
expense of obtaining information, even the less complicated jobs can have many unknowns.
However^ matters are often made worse by not providing adequate information on basic site
conditions such as soil densities, moisture contents, or water table elevations. It is very possible that
a $1,000 saved during the design by not performing moisture contents on soil samples may cost the
project millions of dollars during construction because the contractor can prove he had no reason to
anticipate a moisture problem. Change orders, changed site conditions, and construction claims can
send the best scheduled and budgeted project into a tailspin that will lead to failure in terms of
budget and schedule.
It is also important for projects to be technically evaluated after completion by the designers as well
as independent reviewers. Many of the remedies being installed on these projects are complex.
Failure to evaluate the performance of the system is shortsighted and hampers our ability as designers
to learn from previous projects.
DESIGN AND CONSTRUCTION EXAMPLES
Superfund Remedial Design
The project involves the excavation of metals contaminated sediments from a marsh with stabilization
and disposal of the contaminated sediments and restoration of the marsh. The ROD specified the
construction of a dike around the perimeter of the contaminated marsh. The diked area was to be
flooded with several feet of water and a small floating dredge used for removal of the contaminated
sediments.
During technical review of the ROD it was pointed out that construction of the dike could be difficult
due to stability concerns with the weak sediments and that the RI did not support the assumed
thickness of underlying soft sediment. Comments were also generated by technical reviewers that
dredging may not be the best alternative due to the material properties of the sediment and the: dense
root mat which overlies the marsh sediments. Also, very large quantities of water would require
treatment if the sediments were dredged.
During the design investigation it became apparent that sediments were much thicker, and much
weaker, than assumed by the RI/FS. Based on this the designers, early in the design, suggested an
alternative that would have reduced the size of the dike by dewatering the marsh instead of flooding.
Excavation of the marsh could then be performed using mechanical equipment instead of a dredge
thereby accounting for the sediment and root mat concerns and eliminating a substantial quantity of
water treatment.
EPA's initial position was that this did not conform to the remedy described in the ROD; therefore,
the design was directed to proceed as originally conceived. Early construction estimates began to
1144
-------
show that construction costs were going to be substantially higher than planned. A value engineering
(VE) study was performed and the resulting recommendations were the same as those previously
suggested by the designers. Due to the potential cost savings, EPA then re-evaluated the ROD and
determined these changes could be allowed. This is a case were administrative concerns initially
overshadowed technical realities and resulted in time delays.
Suoerfund Project During Construction
The project is a former landfill that is being remediated by capping and installation of a slurry wall
and an internal drainage system for containment and hydraulic gradient control. During design a VE
study was conducted which suggested several changes to various components of the design, one being
the use of a roller compacted concrete (RCC) wall along the side of the landfill bordering a stream.
Because of schedule commitments, no design investigations were performed for the foundation of the
RCC wall.
Initial preparatory work by the construction contractor suggested that the RCC wall may be more
difficult to construct than originally envisioned due to poor subsurface conditions. It appeared that
deeper excavations may be required to find a suitable subgrade material. Substantial drilling efforts
were initiated to better define the existing foundation conditions. Evaluations are currently underway
to determine the impact on the design. Construction is being held up during this investigation and
evaluation. If an adequate site investigation had been performed during design the construction
delays arising from this problem, and the construction costs associated with the delays, might have
been avoided.
Department of Defense Construction
The project consisted of the cleanup of an explosives contaminated lagoon at a military ammunition
plant. Design was fast-tracked to meet a funding deadline. Time and funding for site investigations
was not provided. Early in construction a high groundwater table was encountered resulting in
dewatering and drainage features having to be added to the project. The specifications stated that the
contractor is responsible for handling all water; however, no information was provided to the
contractor that water may have been that near the surface. The contractor is claiming a cost due to
project delay attributed to defective specifications because the water table was not shown; therefore,
there was no reason to suspect this problem would occur. This problem could have been avoided by
the installation of a few piezometers during design, at a minimal cost.
Superfund Design Investigation
This project consists of a former creosote plant and nearby bayou containing creosote contaminated
sediments. Part of the remedial alternative for this site consists of the excavation of contaminated
sediments from the bayou and incineration on-site. At the start of design it was assumed that the
contaminated length of bayou had been adequately characterized during the RI.
During the initial design investigation minimal confirmatory borings were conducted which revealed
substantially different conditions. Subsequent boring programs have delineated a greater lateral and
vertical extent of contamination with a resulting large increase in the quantity of material to be
incinerated. Each time a drill rig has been mobilized to the site additional design needs have been
identified; unfortunately, budget limitations always seemed to preclude doing all the required
investigations. This resulted in increased mobilization costs and probably hampered design efforts.
However, the fact remains that several investigations were performed and the scheduling and funding
were provided to accommodate most the designer's needs.
1145
-------
Superfund Project Construction Delay
This is a project where a former gravel pit has been contaminated with a variety of organic
compounds, including PCBs. The chosen remedial alternative provided for excavation of the
contaminated sediments with on-site incineration. The specifications forbid the excavation of any
sediments from the lagoon until a trial burn is completed utilizing PCB waste; therefore, it was
necessary for the contractor to import PCB waste from another source to spike a trial burn sample.
Even though the State was an active participant during both pre-design and design, the State would
not let the contractor bring PCB waste to the site because the incinerator was not a permitted facility.
Resolution of this issue has caused a one year delay and a claim for many millions of dollars. This
is the type of issue that should have been resolved during design by the regulators, not after the
contractor has set up a very expensive piece of equipment that is forced to sit idle while the issue is
resolved. Ironically, if the claim is paid, the State will be funding ten percent of the cost of the delay.
Department of Defense Petroleum Cleanup
The project consists of the excavation of soil contaminated by low levels of PCB. The extent and
level of contamination were poorly defined. Additional investigations were very limited due to
budget and schedule considerations. The specifications were written requiring the contractor to
obtain state permits for disposal of low level PCB contaminated soils based on the assumption that any
of several nearby landfills would accept the waste.
Permits were granted by the state for disposal of the contaminated soils; however, the disposal
facilities refused to accept the waste because of concerns regarding liability associated with the PCBs.
Disposal of the soils was ultimately accomplished at a hazardous waste disposal facility at much
greater expense. The lesson to be learned from this is that assumptions about the availability of
disposal facilities should have been verified during design, based on discussion with the facilities, and
not on an assumption that a permit issued by a regulatory agency will make it automatically acceptable
to a disposal facility.
CONCLUSION
How will the next generation view our efforts in the hazardous waste cleanup arena. Will they
measure success by the schedules that were met? By all the bean counts having been counted? By
the amount of money that was spent? Or by the efficiency and quality with which sites were
remediated that posed a threat to the environment?
The one goal that everyone should have in the Superfund process is that of cleaning up the
environment. The purpose of this entire program is not to generate reports or create employment
opportunities, it is to remediate sites which pose hazards to our environment. These remediation;; can
be accomplished better by maintaining the quality of designs as we push to meet schedules and
budgets.
Better designs can be provided by:
(1) Establishing schedules and budgets consistent with the needs of the project. Often it seems
the primary emphasis in this program is placed on schedules and budgets. Quality seems to
have become a distant third. Quality, budget, and schedule all need to be weighted equally
if the best possible remedial action is to occur. Schedules have to be established based on site
conditions and adjusted, when needed, to address the realities of the site.
(2) Improving the communication between disciplines and organizations. Projects are lacking in
quality not because of a lack of technical or managerial input but because the variety of
1146
-------
disciplines and organizations involved in the design are not communicating. Team efforts are
required on these projects.
(3) Adequate designs need to be prepared based on realistic data. On almost every site it is
imperative that we provide the contractors with a reasonable amount of pertinent information
so that the contractor can competitively bid and reliably construct a project. This information
may be basic, such as groundwater elevations over a period of time or a more comprehensive
soil classification program, yet such information may be essential to the contractor in
determining the construction methods. How can we expect a contractor to develop an
understanding of a site in a few short weeks during bid preparation that the designers may
have had years to attain?
(4) Increase the design experience of the industry by learning from projects during and after
construction.
Ultimately, personnel are the most important key to a successful design. Without the right people,
and the right amount of communication, the project will not achieve the maximum level of success
possible.
1147
-------
CONSTRUCTABILITY INPUT
TO THE
HTRW PROCESS
James P. Moore, P.E.
U.S. Army Corps of Engineers
Northeastern Resident Office
Tobyhanna Army Depot, Box 48
Tobyhanna, PA 18466-5048
(717) 894-7052
INTRODUCTION:
Constructability has historically been viewed as the
process of involving those agencies and persons who supervise
and administer construction contracts in the latter portions of
a projects' design stage. Such involvement normally occurs at
a time when all of the actual design work and technical
specifications are complete, and the final contracting package,
including general and administrative conditions, is being
assembled. Little time, attention, or money was allocated to
this phase, and the potential benefits of an expanded role for
constructability were seldom realized. Fortunately, the
definition, role, and relative importance of constructability
input has changed.
We currently think of "constructability" as generically
consisting of three elements:
a. Biddability - the ease with which the contract
documents can be understood, bid, administered, and enforced.
b. Constructability - the ease with which a designed
project can be built.
c. Operability - the ease with which the resultant
facility can be operated and maintained.
For the purpose of this discussion, I will consider two
additional elements which fall under the "constructability11
umbrella:
d. Feedback - providing data on the efficacy of the
selected remedy, during the remedial action, to the designer
and to the cognizant regulatory agencies.
e. Lessons Learned - a formalized process of reporting
problems encountered, and solutions found, for various remedial
action alternatives, for the purpose of assisting others in
future selections.
Note that in all cases, the purpose of constructability
input is to facilitate or make planning, design, construction,
or operations easier.
In the Corps of Engineers, we have made numerous attempts
to capitalize on the benefits of constructability, for both our
military and civil works programs.. We also seek to adapt these
efforts to the missions and projects which we execute. In
Superfund, we found a civil works program which has all of the
essential ingredients for optimizing these benefits for
1148
-------
us and for our customer, the Environmental Protection
Agency (EPA). Further enhancement of the constructability
process is essential to the continued success of Superfund and
other Hazardous, Toxic, and Radioactive Waste (HTRW) Programs;
that expanded role should also benefit our more traditional
design and construction missions.
BACKGROUND:
The Corps has been involved in the EPA's Superfund Process
since 1981. Our Baltimore District performed the very first
Superfund Cleanup at Lehigh Electric, a PCS site in NE
Pennsylvania, utilizing a design prepared by the Omaha
District. Since that time, our role has expanded on a national
scale, and we have developed entire organizations, procedures,
and reporting requirements to support this mission. In the
meantime, our traditional military construction role is
changing, as we in the Department of Defense move to clean up
our own environmental problems under the DERP Program.
Our efforts in these HTW programs, especially in Superfund,
have taught us some valuable lessons about constructability
review and input.
The first of those lessons is timing. When we began
working in Superfund, our standard procedure, adopted from our
military construction program, was to allow for BCO
(Biddability, Constructability, and Operability Review) at the
concept (30%) and final (90/100%) stages of design. With
Superfund, BCO is most beneficial when it begins at the very
outset of planning or design. In our role as EPA's
design/construction agent, this usually begins with a design
assignment; the Record of Decision (ROD) and all previous
investigatory data, such as the Remedial Investigation/
Feasibility Study (RI/FS), are usually provided as guidance
documents. At this point, our design and project managers have
found it is most beneficial to make preliminary contact with
the cognizant construction personnel, and solicit their input
throughout the entire process. Depending on the phasing of a
particular remedial design/action, we have even been able to
provide input at the EPA Regional level, when subsequent RI/FS
or ROD work is underway. Furthermore, by providing Corps
services in the more preliminary stages of EPA's Superfund
Process, and by writing/publishing comprehensive closeout and
"lessons learned" reports, we hope to provide even more
effective and economical remedies to environmental problems in
the future.
1149
-------
In the traditional BCO Process, both funding and scope were
necessarily limited. It generally consisted of sending the
proposed plans and specifications to the cognizant Resident or
Area Office, where someone familiar with the post or facility
in question would check them for major conflicts with existing
conditions. Design assumptions and parameters, codes and
procedures used, and even the content of the technical and
administrative specifications were usually not reviewed, as
they were dictated by some previously established guidelines.
Furthermore, we held to our traditional "turf" as construction
managers, and did not attempt to question or challenge any
design assumptions, fearing that this was beyond our areas of
responsibility and expertise.
Because of the unique nature of Superfund, these
traditional barriers to comprehensive BCO Reviews are removed.
Our agreement with the EPA provides for reimbursement of the
costs we actually incur in performing all of our activities,
including BCO reviews. This gives us the flexibility to
perform detailed analyses, when required, to achieve the most
practicable and economical remedial action package. It also
allows us to adapt our review for phased remedial action
projects, or for those on which we have less than full
assignment for remedial design and remedial action. In fact,
under a Technical Assistance Assignment, we can be tasked to
perform only BCO functions. In other words, reimbursable
funding allows us to expend the appropriate level of BCO
effort, without undue cost to our customer.
The second unique feature of Superfund is the nature of the
process itself, and the expansion of BCO scope which that
provides. Our normal projects follow predetermined time lines,
where planning, design, funding, and construction are
accomplished in accordance with well established regulations
and codes. They traditionally utilize existing systems, proven
technology, and standard contracting methods. The Superfund
Process does not fit into any of these traditional categories,
as the projects are not tied to any predetermined time lines
and methods; in fact, most Superfund Projects are constantly
being reevaluated and reprioritized for funding and action by
the EPA. The very act of performing remedial design and clean
up efforts often produces the need to perform such
reevaluation, as many of the current RA technologies and
decisions are emperically based. Within this framework, the
concepts of BCO must always be at work. Furthermore, because
there are often no easily defined lines between the activities
which we historically classify as planning, design, or
1150
-------
construction, Superfund permits those of us in the
traditional construction management role to expand our BCO
input into all phases of a project. This type of continuum is
consistent with recent efforts within the Corps of Engineers to
provide "cradle to grave" project management, thereby providing
our customers with better, more economical, and timely
products.
DISCUSSION:
To illustrate cpnstructability input, we can use the
following hypothetical situation:
Project: Blue Moondog Chemical
Description: 100 acre site, containing a 5 acre
lagoon, a 20 acre land disposal area, 15
treatment buildings, and miscellaneous
storage and operating areas
Principal Contaminants: The lagoon sediments contain
PCB and heavy metals; the
liquids contain TCE, DCE, and
benzene. The land disposal
area soils exhibit metals, PCB,
and VOC contamination. The
treatment and storage buildings
contain vats, drums and sumps
containing uncharacterized
liquids, solids and sludges.
During the first Building RA,
it was discovered that there
was asbestos pipe and boiler
insulation, and that it was
contaminated with UDMH
(unsymmetrical dimethylhydrazine).
Remedial Action Status:
In 1985, the EPA conducted emergency removal of 150 drums,
some of which had spontaneously erupted and burned. Between
1986 and 1987, a Group of Potentially Responsible Parties
(PRP's) pumped the lagoon liquids and arranged for their
off-site disposal. A previous Corps Fund-Lead contract, in
1987, resulted in the characterization and disposal of the
residual contaminants in 7 of the 15 buildings; those 7
buildings were also dismantled and the rubble was placed in an
on-site landfill cell for future action. The Corps contract
was suspended and ultimately terminated when it was
1151
-------
discovered that the remaining buildings contained asbestos
pipe and boiler insulation, most of which was also heavily
contaminated by the chemical treatment process previously
performed at Moondog, and with UDMH, a previously unidentified
contaminant.
Regulatory Status:
The EPA and state regulators are presently discussing a ROD
for the lagoon sediments and land disposal area soils. Future
operable units will address potential groundwater
contamination, both on and off the site. The PRP Group has
disbanded, and no viable PRP's are anticipated for future
actions. The Corps is reevaluating the ROD requirements and
design criteria to finish the building remedies and
dismantlement; they have also received a design assignment to
perform predesign activities for low temperature thermal
treatment of the soils and building rubble. The EPA and state
regulators are also concerned about the long term operation cind
maintenance of any groundwater treatment systems which may be
required, and about recent and proposed changes in ARARS
affecting air emissions and landfill requirements.
Miscellaneous: While the original PRP Group has dissolved,
several other PRP's have hired a consultant firm to monitor the
site investigation and cleanup activities. The local citizen
group is extremely active and anxious for a final remedy.
Local, State and Congressional interest is high, as the
remediation of Moondog will clear the way for the development
of an industrial park on adjacent lands.
In this scenario, one can obviously see the need and
opportunity for constructability input. What is not so
apparent is that all of the site participants, ranging from the
EPA and State representatives to the PRP Group, can and should
be considered a part of the constructability process. There
are several reasons for this:
a. They may have specialized knowledge about the site,
such as past operating practices.
b. They may have some chemical and physical data,
developed during previous studies or remedial actions, which
can benefit ongoing work.
c. They may be direct participants, advisors,
protagonists, or antagonists in the ROD or Consent Decree
processes.
1152
-------
d. They may be providing funds or technical support for
the project.
e. They may be the ultimate decision-maker.
t. They may be the ultimate operator of any long-term
cleanup facility and/or assume caretaker status on the project.
g. They may simply have some good ideas about how the site
should be remediated.
Having identified the players, let us now consider how they
can facilitate or hinder the constructability process. From
the EPA's perspective, the most efficient, economical, and
expeditious remedies for Moondog will occur if they can advance
all of the operable units at the proper time and at the proper
pace. In order to maintain this momentum, they will be
considering items such as when the Corps will complete their
investigations and resume work on the buildings; when the
State will promulgate final ARARS for air emissions and
landfill requirements; how and when the PRP's might again
figure into the remedial action or cost recovery; and, most
importantly, how all of these actions will affect the
surrounding community. As a participant in the
constructability process, the EPA Project Manager has the
primary role and the authority to ensure that the database of
knowledge about the site, including relevant economic,
political, and social concerns is constantly updated and
available to all of the other participants. By doing so, the
EPA will receive current, clear, and timely recommendations on
which to base their decisions.
The State's perspective, while similar to that of the EPA,
will involve potentially long term commitments for operation
and maintenance at the site; the impacts of their proposed air
and landfill regulations on Moondog and on other sites within
the State; and the precedents which their actions, decisions,
and agreement might set for future Superfund Sites and for
their own HTW Programs. Timely and comprehensive answers, and
review/approval actions, for all state related items are the
most important constructability input they will provide. These
include anything from water quality certifications, to hauling
permits for oversize loads, to air emissions permits for the
proposed thermal process. Coordination among the various
State agencies; with local governments, emergency responders,
other regulators, and potential disposal facilities, are also
essential contributions by the State. All of these inputs have
a direct and substantial impact on how the remedial action will
be performed, thereby touching on all three elements of "BCO"
previously discussed. Perhaps the most important of these from
the State's perspective (at least for Moondog) is the "0" for
1153
-------
operability. One can easily envision that the design of a
sophisticated groundwater treatment system may initially be
justified, based on the cleanup levels which can be achieved.
However, if the State is not willing or able to provide the
funding and staffing for long term operation of such a
facility, a less sophisticated system may be in order. Given
our present bias for "best available technology", and our
relatively limited experience with many of these treatment
systems, this may be one of the single largest constructability
issues for the next decade.
The PRP groups have an obvious interest in the cost and
efficacy of the site cleanup. However, they also have concerns
about the residual liability which may be incurred by their
participation (or their refusal to participate) in the process.
Those of us in the Government's service sometimes overlook or
fail to fully appreciate what motivates PRP's actions,
especially their potential role in constructability. By
recognizing that the technical and financial concerns expressed
by PRP's during remedial design/action may be the same tests by
which cost recovery is ultimately adjudicated, we can make more
intelligent economic choices. Furthermore, with an increase in
projects where PRP groups actively perform some or all of the
remedial actions, like Moondog, a higher degree of cooperation
and sharing of information is essential. Our BCO Review of
their proposed remedial action documents should be as thorough
as if we are doing the work as a Fund-Lead Project, but should
recognize and respect those areas where they are entitled to
latitude.
Finally, let us examine the role which the Corps of
Engineers plays in this scenario. As EPA's design and
construction agent, and in our role as the Federal Engineer, we
have a responsibility to provide our customer with quality
products; delivered on time and within budget; and to perform
this work in a manner which protects the remedial action
workers and the surrounding community. To those ends, our
constructability actions involve the collection and
consideration of as much information as we can possibly obtain,
as early as we can get it. In our hypothetical example, we
would read all of the relevant documents generated by the EPA's
emergency removal action and the PRP's lagoon work. Data on
the processes used at Moondog would also be important to us,
especially as we designed and executed the building and
process dismantlement/removal. By reviewing all of this data,
and considering what we learned during our first attempt on the
1154
-------
building/process contract, we would have a better handle on the
scope of work for a second dismantlement contract, and a more
focused pre-design for the thermal process. In the interim, we
would also be generating and distributing our own information
to the EPA and the State, for their use in providing future
direction to us, and in formulating future ROD'S, Work Plans,
etc. With due consideration for any legal implications, we
would also share information with PRP's and their consultants.
Chronologically, our involvement at Moondog might have been as
follows:
a. EPA signed an agreement with the Corps to provide
document review and oversight of the PRP's lagoon closure. The
Corps was involved prior to the signing/lodging of the EPA/PRP
Consent Decree. Although not a party to the Consent Decree,
the Corps/EPA Agency status was made known to the PRP's prior
to their signing the Consent Decree. The Corps performed BCO
reviews of the PRP's design, work plan, contract
plans/specifications, health and safety plans and other
pre-work documents. The scope of these reviews was decided
between the EPA's Regional Project Manager and the Corps Design
and Construction District Representatives.
b. The Corps performed on-site inspections and oversight
of PRP remedial actions. The scope of Corps involvement, and
reporting requirements, were contained in a Work Plan, Budget
Estimate, and MOU between the cognizant Corps District and EPA
Region. The Corps provided full-time on-site inspection, and
split/analyzed field samples with the PRP's contractor(s).
Documents and data generated by this involvement, save those
which were designated as proprietary by the PRP or their
contractors/consultants, were used in future Corps
design/construction decisions.
c. After the EPA assigned the Building Operable Unit to
the cognizant Corps Design District, the appropriate
Construction District was consulted and involved with the scope
of work development for the design (Architect-Engineer)
contract and with the Acquisition Planning for both the design
and construction processes. The design assignment was made
under one of the Design District's Indefinite Delivery-Type
Contracts (IDTC); the remedial action contract was a fixed
price, competitively negotiated instrument. The IDTC Design
Engineer was furnished with all of the relevant pre-design
information, including the ROD, RI/FS, documents from the
Lagoon RA, etc. The design work consisted of developing a set
of Request for Proposal (RFP) documents, consisting of Plans,
Specifications, and a Solicitation package. On-board reviews
of the design work were coordinated at the 30, 60 and 90%
1155
-------
levels; Construction District personnel were afforded the
opportunity to review documents and provide input at each of
these steps. Among the more important considerations were what
to ask for in the RFP (i.e. technical factors such as a Work
Plan, proposed schedule, health/safety plan, etc.) and how to
evaluate and score these factors, along with bid price, in
selecting a contractor.
d. With the completed RFP "on the street" Construction
District representatives participated as members of the
Selection Board. The successful contractor was awarded a $6.2
M fixed price contract, based on his choice of a dismantlement
method which promised to minimize air emissions and potential
worker exposure; the contractor also had superior (in-house)
sampling and analysis capabilities.
e. After the appropriate award and pre-construction
submittals/approvals were completed, on-site work began on the
Building Operable Unit. Work on the first 7 buildings
proceeded on schedule during the first year of the scheduled 2
year contract. Periodic sampling of the sump wastes revealed
that they closely approximated those liquids which the PRP's
found in the Lagoons. However, upon discovering that Buildings
8 through 15 contained (contaminated) asbestos pipe and boiler
insulation, which was not revealed by the previous
investigations or by design activities, the contractor was
suspended. Because of the contractor's high extended overhead
costs and the lack of approved sampling, analytical, and
disposal methods for the contaminated asbestos, the contract
was terminated for convenience. Construction and Design
representatives researched all of the design documents for
potential A/E liability by the Corps A/E, and advised the EPA
of any similar liability potential for the previous RI/FS
contractors. After the technical and regulatory questions are
resolved, design of Building Operable Unit, Phase II, will get
underway, utilizing yet another IDTC contractor. In drafting
the scope of work, Design personnel will rely heavily on input
from construction records from Phase I, and on the chemical
data generated by the PRP Lagoon RA, to carefully analyze and
characterize the remaining structures, debris, and
contaminants. Samples of the building rubble, previously
placed in on-site landfill cells, will also be analyzed for the
presence of friable and/or contaminated asbestos, and for
indicia of the contaminants found in the sumps and lagoons.
f. Assuming that the future design and remedial action
work proceeds without incident, the Construction District will
prepare closeout reports for each of the operable units for
which they received design/construction assignments. These
reports will chronologically document both the physical
remediation activities and the certified
1156
-------
chemical data which were generated throughout. "Lessons Learned"
reports and input will also be generated and disseminated via
electronic data bases, papers and presentations, and briefings.
g. Long term action at Moondog will include State
operation of the groundwater treatment system, and the recovery
of costs from PRP's. Documents generated during RD and RA will
figure heavily in the recovery process, as all alleged costs
and actions will be analyzed, in detail, under the most
critical of all circumstances: Hindsight!
As described in this hypothetical situation,
constructability in the HTRW Process seeks to provide relevant
input at all stages of the process, and to facilitate the
transition between these phases. For the RA Phase, where
construction personnel are the focus of the process, our
constructability actions include the feedback of information to
cognizant regulators and designers, to ensure that the RA is
proceeding as planned. At the conclusion of RA,
constructability includes the accurate and complete
documentation of the project, to prove that the designed
objectives have been accomplished, and to provide a history of
the project with a "lessons learned" emphasis.
Unfortunately our experience in achieving these goals for
all HTRW Projects has been less than perfect. This is
especially true for projects like the Moondog example, where
many phases and parties are involved. Among the problems which
we find inhibit constructability input, or decrease its
effectiveness are:
a. All of the site historical data is seldom available to
construction personnel.
b. There is seldom any complete institutional knowledge of
a site, either from a historical or regulatory standpoint.
c. We are seldom asked to participate in any portion of
the RI/FS or ROD development.
d. We are not usually staffed or funded to fully
participate in all aspects of the RD.
e. The RA phase has historically uncovered additional
quantities and types of contaminants, thereby resulting in
delays, cost increases, claims, changes, and incomplete RA's.
f. We have not placed enough emphasis on providing timely
feedback, or to preparing and disseminating our "lessons
learned" reports.
1157
-------
However, the Superfund Program, by virtue of the way it is
funded, and because the workload has somewhat stabilized,
provides us with an opportunity to overcome some of these
problems. We have noted an increased constructability
awareness on the part of the scientists and regulators who
normally control the RI/FS and ROD phases of these projects,
and a mutual understanding among all of the players about their
roles and interactions. Most people now realize that the best
RI/FS, ROD, or RD is worthless if the resultant RA cannot be
bid, constructed, and/or operated in the fashion which it was
intended. For our part, those of us in construction management
now realize that unless we make a conscious effort to track and
participate in upcoming project development, and unless we keep
our regulators and designers intimately involved during the RA
phase, we cannot hope to improve the end product. We also
recognize our duty to tell the design and construction
community about our experiences, especially those procedures
and processes which do not work, so that others can avoid
repeating our mistakes.
In order to maximize the benefits of constructability input
for HTRW Projects, the following are suggested:
a. Regulations/standard practices should require a BCO
review at least at the draft stage of every phase of a project
(i.e. draft RI/FS, ROD, RD, etc.). As the proposed action
becomes more definitized, construction involvement should
increase, with BCO review effort ranging from one week for
small RI/FS documents to one year (equivalent) for BCO on a
large RD effort.
b. Funding and staffing should be programmed in advance,
for BCO input at each phase of the project.
c. BCO comment format and reporting times should be agreed
to with the agency and person who will draft the document(s) in
question. A time frame for review and comment should be agreed
and adhered to. The process must also include written
responses to the comments; an "on-board" session to discuss
questions/disagreements; and a corporate approach to resolving
any remaining problems.
d. When site characterization nears completion and cleemup
alternatives are being evaluated, an active search for "lessons
learned" should be conducted. Any findings, both pro and con,
should be included in the subsequent report. The ROD should
also reflect and account for these efforts.
e. All selection and design decisions for long-term
remedies (groundwater pump/treat, etc.) should consider the
element of operability. Technical difficulty, cost, staffing,
and decommissioning of the systems should be addressed.
1158
-------
f. Ongoing efforts by EPA, the Corps, States, and other
Government agencies and private firms, to share available data,
and to enhance discourse between various HTRW elements, should
be encouraged. EPA's SITE Program is a good example of this.
g. Guidance documents for the HTRW process should address
the creation of project milestones for constructability input,
feedback, and "lessons learned".
h. Innovative contracting strategies should be developed
based on site specific criteria and timing. For instance, if
site knowledge is limited and some type of immediate RA is
necessary prior to the completion of RD, a cost reimbursable
form of contract might be required. This allows the
construction manager to better handle the unknowns and to
provide feedback to the ongoing design process. It also helps
us to be more responsive to customer requests and criteria
changes.
i. Construction personnel should be required to
periodically brief and/or report to regulators and design
personnel on the status of all ongoing RA7s.
j. Closeout reporting formats should be standardized to
facilitate use by regulators in "delisting" NPL sites, and to
feed existing data bases on HTRW remedial alternatives. Any
"lessons learned" should receive the widest possible
dissemination, perhaps via programs like SITE.
CASE HISTORIES:
The following are examples of HTRW projects where
constructability input, or the lack thereof, has been a factor
in our ability to implement the selected remedy:
a. Lackawanna Refuse Site: This $25M RA, conducted in
1987-88, consisted of the excavation, sampling, on-site
analysis, and disposal (or backfill) of 114,000 CY of
potentially contaminated refuse and 8,500 drums. In performing
this remedy, we utilized a unit-price competitively bid
contract format to allow us some flexibility in quantity
variation. Because of this, and by using a special form of the
variations clause, we were able to hold the bid price for
disposal, despite a six-fold increase in the estimate of
contaminated refuse encountered. More importantly, this
allowed us to continue work without any suspension or
interruption; we were therefore able to meet the required
"Land Ban" Disposal Date on 8 November 1988. This project also
included many ideas, generated by the EPA, Corps, State, and
Design A/E personnel which have become "constructability" input
standards for other large HTRW projects in the Baltimore
1159
-------
District: Time lapse video surveillance of work activities;
selection and periodic verification of Key Indicator Compounds
as indicia of contamination; and the use of on-site
laboratories for clean/dirty determinations. One item which
remains at this site is the long-term collection and treatment
of leachate. The original RA contract contained provisions for
the design and construction of an on-site treatment plant.
This requirement was subsequently deleted when it became
apparent that it would not be cost effective or practical for
the State to operate and maintain the plant. The EPA and the
state are currently negotiating with a publicly owned treatment
works.
b. Lansdowne Radiation Site: This site was a duplex
residential structure and surrounding grounds which were
contaminated by Radium 226. In order to advance the RA and
obtain the most technically and cost effective remedy, we
advertised this as a "Request for Proposal" contract with
unit-prices. The successful contractor proposed an innovative
method of dismantling the contaminated structure without having
to use a secondary containment. Other constructability inputs
were: a payment item based on the weight of contaminated
rubble and soil, coupled with the contractor's arrangement with
the disposal facility to pay based on volume, ensured the most
economical and compact handling and transport system; on-site
contractor and Government quality assurance (QA) laboratories
allowed for quick turn-around of analyses and clean/dirty
determinations; the presence of QA personnel, and their
development of a real-time method for determining/predicting
soil contamination, allowed for continued funding and execution
of the project despite a four-fold increase in the amount of
contaminated soil.
c. Bruin Lagoon Site: Our first attempt at remediating
this site, an acidic sludge lagoon (approx. 73,000 CY) was
suspended when an uncontrolled and uncharacterized release
occurred. During the redesign of the project, construction
personnel provided input on methods to predict and control any
future releases, and to provide real-time acceptance criteria
for the neutralized and stabilized sludge.
d. Heleva Landfill Site: This site was originally
envisioned as being 22 acres in size and contained by the
existing physical boundaries. Although we did place test pit
provisions in the contract to accurately define the landfill
limits after the contract was awarded, we were forced to make
other adjustments in grading, drainage, and fencing when we
discovered that the actual limits were 25% larger, and not
constrained by the obvious physical limits of the site. More
extensive investigation prior to the advertisement and award of
1160
-------
this contract may have been useful. Heleva was also one of our
first attempts at using a fully synthetic cap and flow zone
system, coupled with a minimum vegetative soil cover, to limit
settlement. To date, this system has proven to be very
effective and required minimal maintenance.
e. Tyson's Dump Site: This PRP-lead propect used vacuum
extraction for the removal of volatile organics from soils and
sludges in previously closed lagoons. Incomplete and untimely
coordination between the Corps, the EPA, and the PRP led to a
number of disagreements about the Corps' role and authorities
on this site. Corps involvement prior to and during the
negotiation of the Consent Decree may have mitigated these
problems. On the positive side, constructability input at this
site did produce a method for baselining and subsequently
measuring the efficacy of this alternative technology at
various points in time, as opposed to waiting until the soil
long-term cleanup levels were projected to be achieved, some 2
years after the start of RA.
CONCLUSIONS:
The role of constructability input, especially on HTRW
projects, is expanding. There are a number of reasons for
this, principally: remedial action is generally the most
expensive phase of the remediation process; and many remedial
designs are emergent technology which are emperically based on
limited data from other remedial actions.
To maximize the benefits of constructability, opportunity,
funding, and staffing are required. Programs like Superfund
provide many of these essential elements and, because of their
unique nature, are not burdened by some of the traditional
barriers to the constructability process. Personnel involved
in HTRW programs, in general, are coming to the realization
that a continuum of involvement by scientific, design, and
construction personnel, from the RI/FS through project
closeout/O&M, is essential. In the Corps of Engineers we hope
to continue that trend in our Superfund work, and in our other
HTRW missions.
ACKNOWLEDGEMENTS:
I would like to thank the EPA Headquarters Staff for this
opportunity to present my views on constructability. I would
also like to thank the staff at EPA Region III, the
Pennsylvania Department of Environmental Resources, and the
Corps Missouri River Division and Omaha District, for allowing
those of us in construction management to fully participate in
the remediation process.
1161
-------
Applications of a Design/Build Advisor Expert System to
Environmental Remediation Projects.
Thomas R. Napier
U.S. Army Corps of Engineers
Construction Engineering Research Laboratory
CECER-FSC
P.O. Box 4005
Champaign, IL 61824-4005
217/373-7263
INTRODUCTION
The U.S. Army Corps of Engineers Construction Engineering Research Laboratory (USACERL) is
currently developing a knowledge-based expert system for Design/Build construction — a non-
traditional approach to the design, contracting, and construction of facilities. A "DESIGN/BUILD
ADVISOR" will provide expert-based guidance to support project planning and execution by those
who may not have a great deal of first-hand personal experience. The DESIGN/BUILD ADVISOR
provides step-by-step procedural guidance and advice in an interactive menu-driven environment.
The relative newness of environmental remediation construction and limited expertise in this field
strongly suggests that similar type of advisory system would be applicable to environmental
remediation projects. Significant benefits may be achieved. The overall system architecture of the
DESIGN/BUILD ADVISOR would be compatible to environmental construction applications.
The DESIGN/BUILD ADVISOR was developed for application to facility design and construction.
However, this paper focuses on the overall system architecture and its decision support capability for
construction-related issues. Those expert in environmental and hazardous and toxic waste fields can
then visualize the applications of a similar "advisor-type" system to environmental remediation
projects.
BACKGROUND
The Design/Build approach is by no means a new method of constructing facilities. However, it is
not universally practiced within the design and construction community, and project execution differs
widely among facility owners and contractors. Federal agencies' practice of Design/Build
construction also differs from private practice. While the Design/Build approach is used within the
U.S. Army Corps of Engineers (USAGE), it is by far the exception rather than the rule. USAGE
personnel are not nearly as well versed in Design/Build as they are in the conventional deslgn-bid-
build practice. Design/Build practices within USAGE differ, as do project results.
Congress has instructed the Defense services to explore "Alternative Construction Methods" (including
the Design/Build approach), which means that Design/Build will be applied more widely within
USAGE. It became clear that further guidance was necessary to support USAGE Districts in
conducting Design/Build projects.
USACERL had previously developed a guidance document for Headquarters, USAGE (HQUSACE)
on the Design/Build approach applied to Army facilities. This Architectural and Engineering
Instructions (AEI) provides general guidance and in that regard is quite useful1. By necessity,
however, it could not always address specific conditions surrounding any given project. Interpretation
1162
-------
by project personnel is still necessary. It became evident that an additional "advisory" capability was
still necessary to provide guidance more specifically tailored to a given Design/Build project.
The DESIGN/BUILD ADVISOR was initiated to support the planning and execution of a
design/build project through experiences, guidance, and advice collected from knowledgeable sources,
i.e. experts. It will provide a step-by-step "roadmap" of the process, generalized advice for each step,
and project-specific advice for decisions which require design/build expertise. The DESIGN/BUILD
ADVISOR will not substitute for expertise, nor would it usurp an individual's professional judgement.
The DESIGN/BUILD ADVISOR will provide advice which project personnel can then incorporate
into their decision-making process.
DISCUSSION
Application of Knowledge-Based Expert Systems.
Under contract with USACERL, the University of Illinois Department of Civil Engineering, with
support from the University of California Department of Civil Engineering, developed a prototype
DESIGN/BUILD ADVISOR2. USACERL personnel provided expert knowledge and directed
university personnel to other expert sources. USACERL personnel completed the substantive content
of this system. The following describes the application of knowledge-based expert system technology
to this project.
Three major elements should be recognized in the DESIGN/BUILD ADVISOR'S development: a
Process Model of the Design/Build approach, Activity Performance Descriptions, and Knowledge
Representation.
A Process Model was created to formalize representation of phases, activities, sub-activities, and
decisions involved with a Design/Build project. The model presents a chronological sequence of steps
in a hierarchical structure. The process consists of a number of phases. Each phase consists of a
number of activities. Each activity is dependent upon sub-activities. Sub-activities are defined to
the lowest, most detailed level useful to accomplishing the activity. The process model also identifies
the relationships and dependencies among activities. These include activities contributing to the
performance of a higher level activity, activities affecting the subject activity, and activities affected
by the subject activity. The process model identifies points where domain-specific expert knowledge
will contribute to decision making.
An Activity Performance Description provides a definition, description, and information on
accomplishing each activity. This information is drawn from facts, experiences, and advice compiled
from expert sources in the Design/Build knowledge domain. The Activity Performance Description
for each activity portrayed in the DESIGN/BUILD ADVISOR Process Model includes the following:
Activity description.
Purpose or objective of the activity.
Sub-steps.
Super-steps.
Steps upon which the activity is dependent.
Steps impacted by the activity.
Activities or decisions upon which the subject decision is dependent.
Input required to perform the activity.
Production or output of the activity.
Schedules, deadlines, or routing.
Forms used during the activity.
1163
-------
General suggestions.
Cautions.
Knowledge Representation is achieved through an object-oriented approach. Each activity or
decision is represents an object, with which attributes are associated. Data, information, and
heuristics gathered from expert input are codified and entered in a knowledge base. Rulesets are then
developed for each activity. The following information was obtained for the DESIGN/BUILD
ADVISOR.
Requirement for expert knowledge.
Description of the results or outputs of the decision.
Factors considered when making the decision.
Inputs required for each factor.
Determination of how decisions are made.
Hierarchy or criticality among factors and inputs.
The three elements described above are fundamental to knowledge-based expert system planning.
A similar approach would seem to be equally appropriate for environmental remediation projects.
Functional Description of the DESIGN/BUILD ADVISOR.
There were several fundamental requirements that had to be addressed when developing the
DESIGN/BUILD ADVISOR. These involve the contents of the system, users of the system, and
mechanics of implementing the system within USAGE.
Requirements for the system's contents and advice were gathered through experience and exposure
to USAGE Design/Build projects. This included input from HQUSACE, USAGE District, and
USAGE field personnel involved with Design/Build projects, as well as first-hand experience by
USACERL personnel. Recurring issues, questions, and problems involved the following general
topics.
General USAGE policy and procedures relative to Design/Build projects.
Selection of projects suitable for a Design/Build approach.
Development Scope and Statement of Work descriptions for contracted Architects/Engineers
(A/Es) and other services.
Contents and development of solicitation documents (Request for Proposal) for Design/Build
projects; technical specifications, instructions to offerers, proposal submittal requirements,
and other provisions.
Certain features of the contract award process (proposal evaluation and source selection
procedures).
Certain features of construction contract administration.
It was also determined that a representation of the complete military facilities' design and construction
process was unnecessary. The system should focus only on those areas which the Design/Build
approach presented considerations and problems not normally encountered in conventional practices.
1164
-------
The primary users of this system would be USAGE project managers at the District level; those
directly responsible for managing facility design, contracting, and construction activities. Project
management personnel would apply the system to all phases of a Design/Build project. Similar use
could be in a review and oversight capacity at the USAGE Division level. Technical personnel (i.e.
the engineering disciplines) may also apply the system to the development of specifications and other
engineering criteria. However, the system would not be a rigorous engineering design or analysis tool.
HQUSACE personnel may use the system when selecting projects for a Design/Build approach.
The system had to be compatible with the USAGE automation environment. A 286 DOS-based
microcomputer was determined to be the appropriate platform for the system, although a 386 DOS
delivery environment is preferable. Software would have to to run at multiple sites at a reasonable
cost for each site.
Finally, it was decided that knowledge would have to be represented in a object-based environment.
This approach allows the development of logical knowledge packages (nodes) that can resemble the
natural logic of experts in the field. An object-oriented approach also expedites updates to the
resident knowledge. Objects and their associate attributes can be amended independently, without
necessitating the reprogramming of all rulesets associated with interrelated objects.
The DESIGN/BUILD ADVISOR is designed to allow the user to navigate through the system and seek
advice selectively. Information is displayed in a two-tiered system architecture.
The Interactive Index (first tier) uses a multi-level mapping concept. The user may enter the system
at any of the five phases described for Design/Build projects. Activities involved with each of the
phases are displayed in menu format. The user then selects the activity for which information or
advice is sought. Information on the selected activity appears in menu format. It is generally
procedural in nature and includes the following, as applicable to the subject activity.
Description of the activity.
Cautions about the activity.
General suggestions for performing the activity.
Steps immediately preceding the activity.
Steps immediately following the activity.
Schedule and routing information.
Other activities affecting the activity.
Other activities affected by the activity.
Forms, reports, and other documentation involved.
List sub-activities.
Decision Advice.
The user may select any of the options listed, upon which the relevant information appears in text.
Selection of the "List Sub-Activities" option invokes and additional menu of more detailed activities.
The user then repeats the sequence, selecting the desired sub-activity, then the desired information,
as described above. In some cases, Hypertext™ explanations are imbedded in the system for selected
items. When selected, these provide additional explanations, references, or information that must
be considered when performing the subject activity.
There are 186 activities and sub-activities defined within the five phases of the Design/Build process.
The user selects the information he/she requires directly. It is not necessary to progress through a
lengthy sequence of activities or the finest level of detail.
1165
-------
Project-Specific Advice (the second tier) is provided for those decision points which require an
expertise in Design/Build that may not ordinarily be present in a USAGE office. A "Specific Advice"
option appears in the information menu for those decision points. Heuristics and rulescts are
maintained in this tier. The system obtains input interactively from the user by requesting
information appropriate to the activity or decision. Given the users input, the inference engine (the
expert system) triggers rulesets for that activity based on the combination of attributes represented
by the user's input. This second level of information provides guidance that is not possible in static
media such as guidance documents.
Once again, it seems reasonable to assume that a similar approach could be applied to an advisory
system on environmental remediation projects.
Example Application of the DESIGN/BUILD ADVISOR.
The following example describes the logic, sequence, and information involved when consulting the
DESIGN/BUILD ADVISOR. This example involves selecting a facility acquisition approach for a
military construction project. This example may also have parallels in the environmental remediation
field.
HQUSACE or a USAGE District (a USAGE construction agent, also referred to as Field Operating
Activity (FOA)) may consider whether or not it would be advantageous to construct a facility using
the Design/Build Approach. USAGE personnel may apply the DESIGN/BUILD ADVISOR in the
following manner. This example focuses on a sub-activity entitled "Decision: Select Procurement
Approach".
For the purposes of this paper, all information is presented in text format for clarity and brevity, and
to facilitate explanations. Text was also edited for clarity and is not necessarily verbatim as it appears
on the screen. Menus appear in italics. The user's selection from the menu then appears in bold and
are highlighted with an asterisk (*).
Upon entering the system, five Design/Build project phases are displayed to the user. These are:
* Phase 1: Identify Facilities for Design/Build Approach.
Phase 2: Conduct Pre-Design Activities.
Phase 3: Develop and Administer Request for Proposal (RFP).
Phase 4: Perform Proposal Evaluation.
Phase 5: Administer Construction Contract.
Two activities are displayed under Phase 1.
Identify Facility Requirements.
* Determine Facility Procurement Approach.
Two sub-activities appear:
Review Directive.
* Decision: Select Procurement Approach.
The advice for the "Review Directive" option, in summary, instructs the user to consult the project's
design authorization directive (transmitted from HQUSACE) for 1) explicit instruction to implement
a Design/Build approach, or 2) other project instructions, special objectives, or other unusual
conditions that would necessitate or strongly suggest using Design/Build as a means of achieving the
1166
-------
stated objectives. Finding no explicit or implicit instructions contained in the directive, the user
would then judge the most advantageous approach for completing the project; conventional design-
bid-build approach, or a Design/Build approach.
The following menu would appears for the "Decision: Select Procurement Approach" option.
Description of the activity.
Cautions about the activity.
General suggestions for performing the activity.
Steps immediately preceding the activity.
Steps immediately following the activity.
Schedule and routing information.
Other activities affecting the activity.
Other activities affected by the activity.
Forms, reports, and other documentation involved.
List sub-activities.
Decision advice.
The user may select any of these options for further information. The information contained for each
of these options is as follows.
Description of the activity. The conventional design-bid-build or Source-Selection
Design/Build procurement approaches are considered at the outset of the project; one approach
must be selected prior to initiating design work. This activity presents decision rationale for
considering the factors critical to selecting the procurement approach. The decision rationale
applies to both HQUSACE and FOA levels.
Cautions about the activity. The design and construction community must be capable and
willing to enter into a competitive Design/Build arrangement. The US ACE construction agent
should have a reasonable level of confidence that an acceptable number of offerers will
participate. The project must be of sufficient scope and contract amount to attract offerors.
Project requirements must not be so cumbersome or restrictive that potential offerors are
discouraged from participating. However, project requirements cannot be so ill-defined that
offerors will be uncertain as to the Government's requirements, or the Government is vulnerable
to receiving an unsatisfactory facility. Specification development, proposal evaluation, and
design review/approval are activities conducted in a different fashion than traditional US ACE
practices; the USAGE construction agent must be adaptable to these practices. Approval to
initiate a negotiated Source Selection procurement must be pursued per FAR part 15 and other
established procurement regulations.
General suggestions for performing the activity. The USAGE construction agent or
contracted A/E services should be familiar with the availability of design and construction
services and Design/Build activity in the project's locale. Facilities that more closely resemble
facilities in the commercial construction market are generally better candidates for a
Design/Build approach. Consider the suitability of design, engineering, and construction
criteria normally observed in the commercial market. Facilities that are unique within the Army
may be less appropriate candidates. Severe time constraints generally favor a Design/Build
approach over the conventional design-bid-build process, and may sometimes be the only
feasible option.
1167
-------
Steps immediately preceding the subject activity. "ReviewDirective";identifyanyexplicit
directive or implied or indirect project requirements necessitating or strongly suggesting
preference toward a Design/Build approach.
Steps immediately following the subject activity. Phase 2, "ConductPreDesignActivities".
Schedule and routing information. There are no specific schedule requirements. This
decision should be concluded as quickly as possible to maintain progress relative to a
conventional project.
Other activities affecting the subject activity. " Review Directive"; identifyany explicit
directive or implied or indirect requirements necessitating or strongly suggesting preference
toward a Design/Build approach.
Other activities affected by the subject activity. Phases 2 through 5, in their entirety.
Forms, reports, and other documentation involved. Approval to initiate a negotiated Source
Selection procurement must be pursued per FAR part 15 and other established procurement
regulations. Submit request in memorandum format through USAGE Division, Construction
Division, to HQUSACE, CEMP-C.
List Sub-Activities. Initiate Design/Build procedures.
Initiate conventional procedures.
Decision advice. Yes.
No.
"Decision advice" only appears if there is conditional advice resident in the knowledge base. If the
user selects the "yes" option, the system invokes rules and heuristics resident in the knowledge base
for this phase. The user is queried for information that reflects specific project conditions. In the
case of "Decision: Select Procurement Approach", three groups of questions are asked. The system
analyzes the user's inputs for each group, provides advice, then proceeds to the next group. A
conclusion statement aggregates the advice synthesized from each of the three groups. This query is
as follows:
Are there any site or security requirements that would prevent the general design and
construction community from participating in this project ?
* 1) No.
2) Yes.
What are the current conditions of the local/regional construction economy ?
* 1) Relatively inactive; favorable to owners.
2) Average.
3) Very active; favorable to contractors.
What level of capability and interest can be anticipated for this project; how evident is
Design/Build practice in the local/regional design and construction market ?
1) Design/Build is practiced and evident.
* 2) Design/Build is practiced on occasion.
3) Design/Build is rarely practiced.
1168
-------
How capable and receptive is the FOA to adjust practices to a Design/Build approach ?
1) Very capable/receptive, experienced and well versed in Design/Build.
* 2) Moderately capable/receptive, some experience in Design/Build.
3) Limited capability or experience.
Given this input, the following advice is offered.
Project conditions appear favorable to a Design/Build approach. A soft construction economy
suggests designers and contractors are seeking additional avenues for business. Evidence of
Design/Build practice suggests interest, presence of necessary design and construction services,
and local/regional Design/Build experience. Moderate FOA receptiveness/experience
enhances chances of successful project execution, but proper execution must be ensured
throughout the project's progress. A conventional design-bid-build approach would also be
appropriate.
If the response to the first question would have been "yes", the system would have advised that the
Design/Build option not be pursued and would not continue any further. If the responses to all of
the other three questions would have indicated unfavorable conditions, the system would have advised
against further consideration of the Design/Build option, and would not continue any further. This
first group of questions is the only group where negative responses would be "fatal" to the
Design/Build option.
The second group of questions is as follows:
What is the facility type; how common is this facility type in the commercial construction
market ?
1) Very common facility type.
* 2) Both common and Army-unique features are present.
3) Unique facility to the Army; no commercial counterpart.
What is the scope of the project (in contract amount) ?
1) Large (roughly $20M or more)
* 2) Moderate (roughly $5-20M)
3) Small (roughly S5M or less)
To what extent can commercial design, engineering, and construction criteria, specifications,
and detailing be used for this facility in leu of standard US ACE or Army-specific criteria ?
1) Commercial/industry criteria will be suitable for the project.
* 2) Commercial/industry criteria may be suitable for the project; some Army-
specific criteria may be necessary.
3) Only Army-specific criteria is suitable for the project.
Given this input, the following advice is offered.
Project conditions are very favorable to a Design/Build approach. A common facility type
suggests that there is sufficient familiarity and expertise with the facility type present in the
commercial construction market. The project scope is adequate to attract participation in the
project, although care must be taken not to discourage potential offers by inadvertently
1169
-------
imposing cumbersome or restrictive project conditions. The use of commercial/industry
criteria is more consistent with private commercial practices and enhances participation by
potential offerers; the application of Army-specific criteria should be reviewed as the project
progresses. A conventional design-bid-build approach would also be appropriate.
The third group of questions is as follows.
What are the time constraints for design and construction (time to Beneficial Occupancy Date)
relative to a conventional military construction project ?
1) Less time to BOD than a conventional project; 24-30 months or less.
* 2) Comparable to a conventional project; 30-36 months.
3) More time to BOD as a conventional project; 36 months or more.
Are there any existing design/construction documents readily available for this facility type?
* 1) Yes.
2) No.
Given this input, the following advice is offered.
Project conditions are favorable to a Design/Build approach. The time available is ample for
a Design/Build approach, but gives it no particular advantage over conventional design and
construction practices. Existing documents may be helpful to either a Design/Build or a
conventional design approach. A conventional design-bid-build approach would also be
appropriate.
Summary advice on "Decision: Select Procurement Approach" is as follows:
Project conditions are favorable to a Design/Build approach. There may be a potential
advantage over the conventional design-bid-build process. Capability, interest, and
Design/Build experience appear to be present in the local/regional construction community.
This must be verified as the project progresses. The project description appears to be
consistent with commercial/industry design, engineering, and construction practices, which
enhances the chance for successful project execution. This must be verified as the project
progresses. The time available to BOD is ample for a Design/Build approach, but gives it no
particular advantage over conventional design and construction practices. A conventional
design-bid-build approach would also be appropriate.
The user would consider this advice and act according his/her own best judgement. Given this
advice, the user should have a fairly high level of confidence that a Design/Build approach can
successfully be implemented to the advantage of the project. The system provides information that
the user may not ordinarily have at his/her disposal. If, for whatever reasons, however, the user is
still reluctant to commit to a Design/Build approach, he/she can also feel comfortable that
conventional design-bid-build practices would be appropriate.
The example illustrated above represents the typical case in facility design and construction. Either
a Design/Build approach or conventional design-bid-build practices could result in a successfully
completed project. However, unfamiliarity with Design/Build practices and absence of project-
specific guidance would generally steer project management personnel away from non-traditional
practices. As a result, the opportunity to achieve positive results is frequently lost. Reinforcement
1170
-------
from the DESIGN/BUILD ADVISOR should enable USAGE to take advantage of more of these
opportunities.
The DESIGN/BUILD ADVISOR provides advise in a similar fashion for the remaining four project
phases. Phase 2: Pre-Design Activities provides advise on activities and procedures encountered prior
to the development of construction documents (design drawings and specifications). As these
activities are generally similar to conventional design and construction practices, the majority of the
advice is procedural and non-conditional.
Phase 3: Request for Proposal (RFP) Development and Administration provides advice on the
development of the solicitation documents for a Design/Build project. The content and composition
of the RFP differs considerably from conventional construction documents. Therefore, more project
conditional advice is provided. Project conditions dictate the preferred composition of drawings,
sources of criteria, and content of specifications, as well as various procurement and contract award
provisions.
Phase 4: Preform Proposal Evaluation provides advice on Design/Build contractor selection, i.e.
Source Selection procedures. Much of this advice is procedural and can be conveyed in a general,
non-conditional fashion. Development of contractor selection criteria depends upon project
conditions and must be addressed by project-conditional advice. However, as the development of this
material must actually take place during RFP development, advice for certain Phase 4 activities is
contained in Phase 3 for consideration at that time. Advice provided under Phase 3 and Phase 4
activities clarifies these relationships to the system's user.
Phase 5: Administer Construction Contract provides advice on the completion of construction
documents by the contractor. Although this activity differs in sequence from conventional design
practices, it is executed in similar fashion to a conventional project. Non-conditional advice is
appropriate. Once construction documents are approved for construction, the remainder of the
construction process is administered in a similar fashion to a conventional project where non-
conditional advice is likewise appropriate.
Applicability to Environmental Remediation Projects.
There are parallels between Design/Build construction and environmental remediation projects that
suggest a similar advisory-type would be applicable. Environmental remediation is a "non-traditional"
field in that the state of knowledge has not yet matured into standard or widely accepted practice.
There is not yet extensive first-hand experience or a widespread base of expertise. Experiences are
not widely disseminated. Yet, remediation projects must still be conducted with a high levels of skill,
quality, and performance. Process information will be necessary for project planning and strategic
decision making. Generalized information, guidance, and advice will be necessary for procedural and
technical issues. Conditions encountered on a case-by-case basis will necessitate project-specific
advice.
A "strategic planning" phase for remediation projects may parallel the project selection phase of the
DESIGN/BUILD ADVISOR. In a facility construction context, this issue is not overly complex.
Initial planning decisions for remediation projects, however, will be considerably more complex.
Selection of a contract method, for example, will have profound affects on the remainder of the
project. Some project requirements may be well enough defined that a firm-fixed contract will be
appropriate. Most often, at least in facility construction, this approach is selected by default rather
than through consideration of the circumstances. If the scope of services cannot be accurately
defined, or if there are unknown conditions to the project, a deluge of costly and time consuming
1171
-------
contract modifications will be forthcoming. Perhaps a cost-plus type of contract would be better
suited — but at a price of considerably more intense contract monitoring. Project planning personnel
must be able to identify the relevant conditions present for the project, associate project conditions
with advantages and disadvantages of alternative contracting methods, predict the affects the selected
method will have on the remainder of the project's duration, and select the method best suited the
project at hand.
An advisory-type system similar to the DESIGN/BUILD ADVISOR would be able to provide advice
on based on project-specific conditions. The different contracting options would be defined.
Conditions that would suggest advantages and disadvantages of each option would be defined. Rules
would be developed for each combination of conditions. Advice statements would be crafted
accordingly. User input would be solicited to identify the conditions present at any specific project.
The expert system would then invoke the rules and advice consistent with the conditions described
by the system's user. Although the numbers of options, conditions, and combinations are likely to
be many times those involved with the example above, the principles and basic structure will be the
same.
The process model for the DESIGN/BUILD ADVISOR was created as a frame for the expert system.
A secondary use, though one more visible to the system's user, is as a process guide and "roadmap"
through the Design/Build process. Although procedures for remediation projects may be well
established, a consistent process guide may enhance training and familiarization among project
management personnel.
Another possible application of an advisory-type system would be for the development of engineering
requirements, criteria, and specifications for project contract documents. Where a precise description
of methods, materials, or techniques can be made, these may be included in project specifications.
However, if such descriptions cannot be made, methods are as of yet unknown, or a number of
alternative methods may achieve the same results, a performance approach to specifying input and
output requirements may be more advantageous. The nature of the project, existence of criteria,
sources of criteria, and required results contribute to the specifier's decisions. Once again, the
complexities of environmental-related criteria will likely exceed those of building construction.
However the principles and applications could be the same. Decision factors, inputs, rules, and advice
can be created to assist in the composition of criteria and specifications.
Finally, this paper recommends consideration of an additional feature not currently part of the
DESIGN/BUILD ADVISOR. Building design and construction professionals struggle wilh the
problem of contract modifications necessitated by unknown conditions, inaccuracies or ambiguities
in project scope, criteria changes during the project, and other change conditions. The nature of
environmental remediation projects exacerbates this problem severely. Expert system technology may
facilitate management of this problem area. The potential for changes may be so great, arid the
number of conditions and possible resolutions may be so numerous that comprehensive and
meaningful rules and advice may be difficult to develop. However, as experience is gained and
documented over time, and change conditions can be anticipated and modeled with greater
confidence, an advisory-type system may contribute significantly to the management of this problem.
CONCLUSION
In summary, an advisory-type system is being developed to support project management personnel
in the planning and execution of the Design/Build method of facility acquisition. A prototype
DESIGN/BUILD ADVISOR has proven that the system works, is useable for the intended purposes,
and provides valid advice. Current work involves reinforcing the material presently in the system,
i.e. text revisions and editing, and addition of expert-based knowledge. Inclusion of a database to
1172
-------
store project information and document decisions as a project progresses is being incorporated into
the system.
The overall system architecture of the DESIGN/BUILD ADVISOR would be compatible with
environmental remediation projects. Project phases and steps would be formalized in a process model.
Rules would be synthesized from domain-specific expert input into general and project-specific
advice. An interactive menu-driven environment would generate advice based on input from the
system's user.
Significant benefits can be achieved with the application of an advisory-type expert system to
environmental remediation projects. Individuals' capabilities will be enhanced through access to a
knowledge base founded on expert input, which will broaden with additional project experience. The
primary benefit would be in the improvement of the quality of decision making and, therefore, the
probability of successful project execution.
REFERENCES
1. Architectural and Engineering Instruction (AEI), "One-Step and Two-Step Facility
Acquisition for Military Construction: Project Selection and Implementation Procedures",
Headquarters, U.S. Army Corps of Engineers, CEMP-EA; Published by USACERL as
Technical Report P-90/23, Thomas R. Napier, Steven R. Freiberg, August, 1990.
2. Draft Final Report "Knowledge Based Expert System for Design/Build Project Planning",
James H. Garrett, Department of Civil Engineering, University of Illinois at
Urbana/Champaign, Anthony D. Songer and C. William Ibbs, Department of Civil
Engineering, University of California, Berkeley, CA, February, 1990.
1173
-------
"CONFORMING STORAGE FACILITIES'
REMEDIAL CONSTRUCTION ACTIVITIES
D.M. VELAZQUEZ
DEFENSE LOGISTICS AGENCY
Office Symbol: DLA-WIC
Cameron Station, Alexandria. VA 22306-6100
Telephone: (703) 274-6385
INTRODUCTION:
Hazardous waste (HW) remediation has become an increasing complex
issue. In 1980 the Department of Defense (DoD) consolidated the
responsibility for disposal of HW generated by DoD activities under
one agency Defense Logistics Agency (DLA).™ DLA recognized the
importance of studying and developing safe disposal methods for HW
however, the problem of safe storage of HW generated on a daily basis
remained. The Conforming Storage Fa.cility (CSF) program came into
existence as a result of this need. This program constructs storage
facilities which allow temporary storage of HW until proper disposal
is possible.
BACKGROUND:
In 1981 DLA, thru the Defense Reutilization and Marketing Service
(DRMS), embarked on an ambitious program to construct CSF's are most
Defense Reutilization and Marketing Office (DRMO) locations worldwide.
These facilities are "conforming" because they are designated to
conform with the Resource Conservation and Recovery Act (RCEA) Each CSF
requires a RCRA Part B permit before they may be built and operated.
CSF are facilities for the temporary storage1** of HW until proper
disposal is possible. The host installation (owner) of the CSF
•^lj£nm'fc«»d fox-
C2> KGBI.A. ±<* «.n Aerfe to j>z-ovl
-------
submits a RCRA Part B permit application or modification which
provides all the information requirements necessary in order to
determine compliance. The CSF's comply with the Environmental
Protection Agency (EPA) permitting regulations outlined in the Code of
Federal Regulations Title 40, Part 270 (40CFR 270). These regulations
establish the provisions for the Hazardous Waste Permit Program under
Subtitle C of the Solid waste Disposal Act (42 USC 3551), as amended
by RCRA. These regulations cover basic EPA permitting requirements
such as application, standard permit conditions, monitoring, and
reporting requirements. The RCRA permit program has separate
additional regulations that contain technical requirements 40 CFR 264,
266, and 267. These regulations are used by permit issuing
authorities to determine what requirements must be in place in the
permit if they are issued. The CSF design incorporates the applicable
technical requirements.
CSF Design:
CSF's store almost all hazardous property generated by the
military services. te> The CSF design adopts a modular concept to
provide flexibility in adjusting the size to meet site specific
storage needs, separation of flammable and nonflammable areas, and an
interior spill containment system.*7" Inside the CSF there is a
staging area, storage modules, and a covered load/unload area. HW are
off-loaded from the delivery carrier and inspected within the staging
area. After inspection the HW are placed in the storage modules
according to its classification. The spill containment system
consists of a leveled floor within the storage modules accessed by a
ramp from the higher elevation of the staging area and central
corridor. If a leak were to occur it will be confined to the
immediate storage area. The storage containers are maintained
elevated from the floor by placing them on pallet racks or shelves to
facilitate the clean up of spills. The building has a perimeter
curbs, entrance ramps, and raised thresholds for all emergency
personnel door exits to prevent the escape of interior spills to the
outside. CSF's are located within a fenced compound which does not
allow for the unknowing or unauthorized entry. Each facility is
equipped with spill and fire alarm systems, telephones, fire
protection system, emergency showers and eye washes.
CSF Operating Procedures:
DoD and DLA installations are responsible for compliance with
environmental and other pertinent laws and regulations. To ensure
environmental compliance the DRMO's and generators carry out the
<=&•&• ji ox- !.<»•• o*" l»mK«tx-on
»m*>l«» of
1175
-------
following turn-in procedures:
1. Preplan, schedule, and coordinate hazardous property
turn-ins.
2. Process turn-ins of hazardous property as follows:
a. Identification - of hazardous property.
b. Packaging - nonleaking and safe containers.
c. Labeling - to comply with established environmental,
safety and transportation laws and regulations.
d. Disposal turn-in document.
Hazardous waste is disposed of by the use of commercial disposal
service contracts. Contracts are awarded to contractors which are
considered responsive and responsible as outlined in the Federal
Acquisition Regulations and who are licensed by EPA. Licensing also
includes permits of the contractors' disposal facilities.
CSF Design and Construction Program Management:
DLA delegated the specific design and construction management
responsibility if all CSF project to DRMS. DLA continues to oversee
the entire CSF program and provides the planning and programming
guidance and policies.
CSF Construction Funding:
Due to uncertainties associated with obtaining site approval,
design completion, and receipt of the RCRA permit, Congress has
approved a single-line-item (block) funding for CSF's construction
projects in fiscal year (FY) 87. This approach means that the funds
are not earmarked for a specific project but can be utilized where
needed. The fiscal year assigned to CSF project represents the funds
we propose to use for construction. Projects are funded for
construction as they receive the RCRA Part B permit, but not to exceed
the appropriated amount. DLA has received $40.3 million between FY 87
and FY 91 for the construction of CSF projects. DLA estimated an
additional *75 million is required to complete this program.
DISCUSSION:
Even with the best intensions in mind the construction of CSF's
has been a slow process and has suffered several setbacks. This
program has (and still is) been under close scrutiny by Congress, DoD
IG, EPA and the general public.
Congress has imposed the following restrictions on the CSF
program:
1. All CSF construction projects require their RCRA Part B
1176
-------
permit prior to authorizing construction funding.
2. Congress requires a notification of intention to proceed
with construction. There is a 21 day waiting period for this
notification.
The DoD IQ was concerned with the quality of the requirements data
initially used to justify the need and size of the CSFs and the
exorbitant cost to construct these facilities. The DoD IG recommended
DRMS revalidate the need for and size of all CSFs and reevaluate the
standard design. The revalidation and redesign efforts were initiated
to reduce the construction costs and avoid duplicity and unnecessary
construction of storage facilities. The revalidation effort'*1
evaluates the need and size of existing CSF projects. This is a three
phase effort:
Phase I: DRMS reviews the generation data and determines the
sizing and type of facility required. A revalidation package and
questionnaire is prepared and submitted to the User and Host for
review and comment.
Pha.ee II: The User and Host review the revalidation package
and answer the questionnaire.
Phase III: DRMS reviews the User and Host response and
determines if the CSF is properly sized and make the final
determination on size and facility type.
This process is now & standard operating procedure in evaluating
existing requirements and developing and sizing new requirements. The
monetary benefits attributable to the revalidation effort are
anticipated cost reductions in the amount of $26.5 million.
The redesign effort**" evaluates the CSF standard design in an
effort to reduce project costs and eliminate excessive systems safety
criteria. This effort was divided into four phases:
Pha.ee I: Review of proposed changes to the CSF standard
design.
Phase II: Cost comparison between the CSF standard design and
the revised CSF standard design.
Phase III: Incorporate the approved changes which are
economically feasible to the CSF standard design.
The monetary benefits attributable to the redesign effort are
anticipated cost reductions in the amount of $16.5 million.
Jon p«>oo*<*« flow erfemx-fc.
p>z-oc*«<» flow c=Iamz-fc.
1177
-------
The EPA permit approval process takes 2-3 years. In addition,
permit review and approval has a low priority with the regulatory
agencies.
Another complication is the "NIMBY" syndrome (Not In My Back
Yard). The public, has a misconception of the purpose of a CSF.
Great alarm is sounded during public hearings, near by neighbors of
the CSFs often believe that the CSFs process or hold controlled
substances or fear massive contamination of one kind or another. This
causes CSF relocation or design changes beyond anything anticipated or
required by law.
CONCLUSION:
Conforming storage facilities are an effective and safe solution
for the temporary storage of hazardous wastes until they can
eliminated properly and permanently. Effective use of CSFs will
require the education of the public.
REFERENCES:
Defense Environmental Quality Program Policy Memorandum (DEQPPM)
80-5 dated 13 May 80.
Resource Conservation and fiecovery Act - USC Title 42 Section
6901.
Code of Federal Regulations:
40 CFR 264
40 CFR 267
40 CFR 266
40 CFR 270
Solid Waste Disposal Act - USC Title 42 Section 3551
Defense Reutilization and Marketing Manual - DoD 4160.21-M
Mar 1990
1178
-------
( PREPARE PERMIT^
\ APPLICATION __J
.SUHMIT APPUCAT1ON
TO REGULATOR
REGULATOR
REVIEW
RECEIVE NOTICE OF
DEFICIENCY (NOD)
PREPARE/SUBMIT
NOD RESPONSE
REGULATOR
REVIEW
•YES
'NO
REGULATOR
PREPARES
DRAFT PERMIT
COMMENT PERIOD
(45 DAYS)
-YES
•YES
—
NO
NO —
PUBLIC HEARINGS
(30 DAYS)
PERMIT ISSUED J
PERMIT PROCESS FLOWCHART
APPENDIX A
1179
-------
MATERIALS ASSIGNED TO DOD
COMPONENTS FOB DISPOSAL
DoD components shall be responsible for disposal of the following
categories of hazardous materials which have not been assigned to DLA:
1. Toxieological, biological, and lethal chemical warfare
materials which, by U.S. law, must be destroyed. Disposal of the
by-products of such material is the responsibility of the DoD
component with the assistance from DLA.
2. Material which cannot be disposed of in its present form
due to military regulations, e.g., consecrated religious items and
cryptographic equipment.
3. Municipal type garbage, trash, and refuse resulting from
residential, institutional, commercial, agricultural, and community
activities, which the facility engineer or public works office
routinely collect.
4. Contractor generated materials which are the contractor's
responsibility for disposal under the terms of the contract.
5. Sludges resulting from municipal type wastewater treatment
facilities.
6. Sludges and residues generated as a result of industrial
plant processes or operations.
7. Refuse and other discarded materials which result from
mining, dredging, construction, and demolition operations.
8. Unique wastes and residues of a non-recurring nature which
research and development experimental programs generate.
APPENDIX B
1180
-------
CONFORMING STORAGE FACILITY
BASIS OF DESIGN
1.0 INTRODUCTION
1.1 Purpose: To provide site adapted design drawings
specifications for the Defense Reutilization and Marketing Service
(DEMS) Conforming Storage Facilities (CSF).
and
1.2 Directive Authorization: The
accordance with the following:
standard
was in
1. Letter,
Services Assistance.
design
DPDS-L, 2 April 1984, subject: Engineering
2. Memorandum of Understanding between the U.S. Army Corps
of Engineers and the Defense Logistics Agency (DLA), 16 July 1985,
subject: Support of the DLA Environmental Protection Program.
1.3 Criteria: Project Development Brochure I,
February 1987.
revised
1.4 Project Description: The function of the CSF is to provide
for a safe, long term (in excess of 90 days) storage of hazardous
waste and excess hazardous materials in accordance with the Resource
Conservation and Recovery Act, Toxics Substance Control Act and
applicable design criteria.
for:
needs.
A modular concept for the facility was adopted to provide
a. Expansion as required to meet site specific storage
b. Two-hour fire rated separation of
flammable
materials.
c. Segregated containment of accidental spills and
leakages of hazardous materials in accordance with state and federal
requirements.
Staging Area: The staging area consists of the material
handling area inside the CSF. Hazardous materials are off loaded from
the delivery carrier, inspected within the staging area and then
placed in the proper storage module or closet. Hazardous material is
ot to be stored overnight in the staging area. Containment within
the staging area is achieved by perimeter curb and ramp loading down
from the exterior cargo door and personnel door. The emergency
eyewash/shower and other equipment necessary in handling hazardous
materials are stored in this area.
1181
-------
Covered Load/Unload Area: The covered load/unload area is a
pre-engineered metal building. It will have an open front to allow
vehicles to back up to the exterior overhead door of the facility for
delivery and pick up.
Fire Suppression Systems: The standard design provides for
automatic sprinkler system protection of all areas of the building
except the electrical room.
APPENDIX C
1182
-------
s
f KUVUiW GENERATIONS
I DATA
I)
SIZE
REQUIREMENTS
SELECT STORAGE
OPTION
REVALUATION PACKAGE
TO HOST/USER FOR
REVIEW
(Phase II)
YES
SEED
FINALIZED
TO REDESIGN
SIZE REQUIREMENTS
REEVALUATED
REVALIDATION FLOWCHART
APPENDIX D
1183
-------
INVALIDATION
COMPLETE
REVISE
DD FORM 1391
DECREASED/SAME
INCREASE
AUTHOREE
REDESIGN
CEHND PREPARES P & S
FOR STTE ADAPTION
REDESIGN FLOWCHART
APPENDIX E
1184
-------
IX. TREATMENT TECHNOLOGIES
1185
-------
A New Horizontal Wellbore System
For Soil and Groundwater Remediation
by
Ronald Bitto, Haraldur Karlsson and Gary E. Jacques
Eastman Christensen Environmental Systems, a Baker Hughes company
Houston, Texas
(Author(s)' Address at end of paper)
Introduction
This paper will describe the development of an innovative drilling system for installing
horizontal wells for soil and groundwater remediation. The paper will suggest specific applications
for the system. Detailed technical specifications and a summary of a four-well field testing program
also will be presented.
Background: Potential Applications
Over the last decade, horizontal drilling technology has been developed and applied in the
petroleum industry for oil and gas production and in civil engineering projects for utility and
pipeline installation. The oil industry has drilled more than 2,000 horizontal wellbores since 1980.
This experience has helped service companies develop new drilling technology and has helped oil
companies gain a better understanding of how to use horizontal wells for petroleum production. 1
In 1989, the authors initiated a research project to identify potential applications for
horizontal drilling in the environmental industry. This study indicated that many environmental
problems can be solved more efficiently with horizontal wells than with traditional vertical wells.
For example, there are numerous "common sense" applications for horizontal drilling,
including capture of contaminated groundwater or leachate from beneath lagoons, landfills,
buildings, storage tanks, refineries, and chemical plants. (Figure 1). Similarly, horizontal wells
may be used to recover spilled product which has pooled under tanks and processing facilities. In
these cases it is difficult to place vertical wells to perform sampling or remediation.
In other situations, where vertical wells now are used to extract polluted groundwater for
treatment, horizontal wells can offer significant advantages. (Figure 2). By placing a long
horizontal section through the contaminant plume, a single horizontal well may replace many
vertical wells, while also reducing clean-up time. ^
Soil gas extraction is another important potential application for horizontal wells. Figure 3
shows how pairs of horizontal wells can be drilled at different depths. The lower well could be
used to injecti air, while the upper well could be used to extract the air stream along with volatile
organic compounds that have been stripped from the soil. 3
A similar well configuration could be employed with the lower air in the saturated zone.
Air forced into the lower well would bubble through the aquifer, and help remove volatile organic
compounds in a sparging effect that acts like an in-situ air stripper.
Other forms of in-situ remediation also may benefit from horizontal drilling technology.
For example, horizontal wells might be used to convey microbes and/or nutrients for
1186
-------
bioremediation of underground contaminants. (See Figure 4). Likewise, horizontal wells might
make it possible to chemically treat heavy metals in place without incurring the expense and
hazards of digging up contaminated soils.
Horizontal wells also could be applied at landfills and other areas where a barrier must be
installed to keep pollutants from migrating into the groundwater (Figure 5). A series of horizontal
wells beneath a landfill or a lagoon, for example, could be used to place a pressure curtain of
pumped air or water, or a floor of grout, epoxy or cement to contain the potentially harmful
leachate.
Another potential application for horizontal wells is remediation of contamination in
fractured bedrock. Petroleum production in vertically fractured reservoirs has been enhanced
significantly by the installation of horizontal wells which intersect several fracture planes.
Likewise, horizontal wells drilled to cross multiple bedrock fractures at a high angle could
improve product or contaminant recovery where vertical wells have proven ineffective.
Horizontal wells also have the potential of providing improved recovery of dense, non-
aqueous phase liquids (DNAPLs) from aquifers. DNAPLs tend to sink through porous media
until they encounter a low-porosity layer. At this point, the DNAPLs pool along the horizontal
boundary. Because horizontal wells can be installed parallel to bedding planes, the cleanup can be
accomplished more effectively than with vertical extraction wells.
During our technical review, many potential users requested the capability to take samples
of soil gas, soil, and bedrock from beneath landfills , lagoons, tanks, and buildings. In these
situations vertical methods are either impossible, inconvenient, or pose a threat to the environment
by providing contaminants a pathway into the aquifer. Horizontal wells can be applied to handle
the majority of these sampling needs.
Major Design Considerations
Our technical study also determined the industry's preferences for horizontal well
construction and the geologic strata to be drilled, as well as requirements for well depth, overall
length and borehole directional accuracy. Other considerations such as availability of suitable
drilling rigs, site space limitations, and the acceptable operating schedules also were investigated.
These efforts resulted in the general systems specifications listed in Table 1.
Major considerations in designing the drilling system included:
-Placement of horizontal sections at depths ranging from 20 ft. to more than 500 ft.
-Installation of horizontal lengths of more than 500 ft.
—Drilling in very unconsolidated formations
—Effective completion of the wells with a minimum 4-inch OD screen
—Operation with a minimum rig crew
-Use of non-contaminating drilling fluids
-Personnel safety and protection of surface environment from contamination
1187
-------
Custom Drilling System
After reviewing the available contract drilling service and hardware, the project team
concluded that technology is not available within the water well and monitoring well industry to
conduct horizontal drilling operations. In addition, mining and civil engineering technology also
does not meet the specific requirements of horizontal drilling in environmental applications.
Because of these factors, an entirely new drilling system, including downhole technology and a
custom slant drilling rig would have to be designed and built for horizontal drilling in the
environmental industry. The project team developed the concept by adapting advanced oilfield
technology. The rig and downhole tools were designed to work as a system to drill to horizontal
on a 100-ft. radius (Figures 6 and 7).
Important features of the drilling rig that resulted from this effort include:
-Capability to slant the rig mast from vertical to 60 degrees in 15-degree increments.
Figure 8 shows how this capability enables the drilling system to place the horizontal section at any
depth in this range.
—The rig is hydraulically operated for precise, automated control from a single driller's
console. Rated at 2,000 ft. for vertical drilling, the unit has a hoisting capacity of 70,000 Ib. and
30,000 Ib. of push down capability. This gives it ample power for handling the system's dual drill
string which may encounter significant torque and drag during horizontal drilling.
—Pipe handling is accomplished with a hydraulic pipe-handling arm and two hydraulic top
drives: one for the casing and one for the drill pipe. In addition, a power tong make-up and break-
out unit is incorporated for making/breaking connections. Casing tongs are provided to hold the
well casing when required.
-The drilling unit's fluid system - with mud pumps, fluid tanks, solid control equipment
and a grouting machine — is included in a single trailer. The circulation takes place in a closed loop
and requires no earthen mud pits. At the conclusion of the job, drilling fluids and cuttings can be
placed in drums for disposal.
-Rig operations requires only a 3-man crew per shift, with a project engineer supervising
the job.
—Pipe storage, rig-site office and electrical generator are incorporated in a third trailer. The
site office includes a computer and can be used as a laboratory as needed. The generator provides
power for lights used for night-time drilling as well as for the solids control equipment. A small
crane, mounted on the trailer, is used to move drill pipe and casing.
-All three trailers that comprise the drilling unit can be transported without special permits
on highways in the contiguous 48 states.
Downhole Drilling Equipment
Like the drilling rig, the downhole system also had to be specifically engineered to solve
the unique problems associated with horizontal drilling in shallow, unconsolidated formations.
The downhole drilling assembly is comprised of a dual drill string; a hydraulic downhole motor, an
expanding drill bit; and a toolface indicator/inclination measurement device. (See Figure 9).
1188
-------
The unique drilling assembly was designed to address the problems of drilling horizontally
through unconsolidated and heterogeneous formations found near the surface. Such strata make it
difficult to maintain hole integrity, even in vertical drilling. In horizontal drilling, there is an even
greater risk of hole collapse. This is especially true in environmental drilling applications where
most drilling fluid additives are avoided. In such conditions, the horizontal hole could be lost
when the assembly is changed or during installation of completion hardware.
A new casing-while-drilling method was developed to solve this problem. An inner string
of 2-7/8" drill pipe pushes the high density polyethylene (HDPE) casing/well screen into place.
This protects the hole from cave in during drilling and installs the well casing at the same time.
(HDPE was chosen because of its unique physical properties including strength, flexibility and
resistance to damage from a broad range of chemical contaminants.) The casing is centralized in
the hole to permit cementing and effective well completion. Once the well is drilled to total depth,
the inner drilling assembly is withdrawn from the hole and the casing is left in place.
Downhole power and the ability to guide the hole are provided by a steerable downhole
hydraulic motor. The motor is based on oilfield positive displacement moineau motor concept
which converts the hydraulic energy of the pumped drilling fluid into mechanical energy (speed
and torque) that rotates the bit. Refinement of this concept resulted in a specially-designed multi-
lobed motor that is about one-fifth the length of oilfield tools. Flow rates range from 150 to 300
gpm, generate up to 40 hp at the bit.
Directional drilling is accomplished by placing the motor in an eccentric position in relation
to the hole axis by installing stabilizer rings at two points on the motor housing. (Figure 10).
These eccentric stabilizers are positionally matched with the concentric stabilizers in the lowest joint
of outer casing. By orienting the direction of the bit offset (also called "toolface"), the hole can be
steered. The configuration of the drilling assembly is designed to turn the borehole at a constant
rate which can be precisely calculated (See Figure 11). The two stabilizers and the bit gauge serve
as tangency points that define a constant radius arc along which the assembly will drill. Build rate
can be controlled by varying the eccentricity of the inner stabilizers. The system drills a straight
course through regularly adjustments of the toolface from side to side. 4
The downhole drilling system features an expanding bit which drills a hole that is large
enough to permit the casing to be installed during drilling. The bit used in the curved section drills
a 12-1/4" hole for installation of 10-3/4" OD casing, and the bit used in the horizontal section drills
8-5/8" hole to permit running a 6-5/8" OD casing/well screen, and providing space for gravel
packing around the screen. Initial bits used with the system were drag-type bits with
hydraulically-spread wings and tungsten carbide cutting surfaces. Other drill bits developed for the
system include roller-cone bit technology for drilling harder formations and glacial till.
The toolface indicator system is a mud-pulse telemetry system which measures inclination
from vertical and toolface orientation, and transmits the measurements to the surface via pressure
pulses in the drilling fluid. These are detected at the surface by a pressure transducer, whose
readings are interpreted by a surface control computer. The toolface indicator sensors are located
1189
-------
just 12 ft. above the drill bit. The system gives operators the ability to monitor the drill bit's
position and wellbore trajectory every 60 seconds. The TFI therefore saves a significant amount of
time that would be required for single shot surveys, while eliminating the complication arid risk
associated with electric wireline steering tool devices commonly used in petroleum drilling and
river-crossing applications.
Drilling Process
Before drilling begins, wells are carefully engineered to meet the specific objectives of the
project. Site characterization studies, including monitor well data, are reviewed to determine the
size and three-dimensional position of the contaminant plume. Groundwater flow and contaminant
migration characteristics are analyzed to assure proper well placement. Next, surface location and
operational factors are considered. Then, the depth and direction of the horizontal wellbore, screen
length, development and pumping methods are determined.
The rig is moved onto location and aligned to drill the horizontal wellbore in the desired
direction. The angle of the rig's mast is adjusted to drill the horizontal section at the proper depth.
A 14" hole is augered 5 to 10 ft. into the soil and a 12-3/4" conductor is set and cemented
in place to provide a controlled conduit for the drilling fluid.
A straight drilling assembly is lowered in the hole to drill to the required depth so that the
100 ft. radius curve will reach horizontal at the desired vertical depth. Once this depth is reached,
the assembly is withdrawn and the curve drilling assembly is picked up and run into the hole.
The curve is drilled in a 12-1/4" hole and cased at the same time with 10-3/4" casing. The
assembly is oriented in the proper direction using the toolface indicator and by holding orientation
at the surface. The same survey tool is used to track the progress of the assembly. After the 20 ft.
lengths of dual drill string are drilled into the hole, both components of the dual drill string are
added simultaneously with the mechanized pipe handling system in the rig mast. Once the curve
reaches horizontal, the inner assembly is withdrawn leaving the 10-3/4" HDPE in place.
A cementing plug is then run into the hole to seal the end of the casing and to allow the
cement grout to be circulated through the drill string to fill the annular space between the casing and
the hole wall. Once the desired amount of cement is in place, the drill pipe is withdrawn from the
well and the grout is allowed to set. The grout will provide structural support to the casing and
will prevent the migration of contaminants from one zone to another along the outside of the
casing.
As mentioned above, the system uses an 8-5/8" bit to drill the horizontal section. A 6-5/8"
OD HDPE screen is pulled into the lateral wellbore by the drilling assembly as the well is drilled.
The system is steerable for course corrections and adjustments to the horizontal section. Steering
capability is provided by the hydraulic downhole motor, by stabilizers on the casing, and by
survey instrumentation.
Formation evaluation will be accomplished at desired intervals using a core, soil or gas
sampler, which are being developed. Drilling is stopped and the inner assembly consisting of the
bit, motor, and drill pipe is retrieved from inside the slotted liner. The bit and motor are replaced
1190
-------
by the sampling tool and run into the hole. The sampling tool is then drilled into the formation for
the required depth and samples are retrieved. Shelby-tube and soil gas sampling devices also are in
development.
Drilling continues with the horizontal drilling assembly until the desired displacement is
achieved. The inner drilling assembly is then retrieved leaving the 6-inch screen in place.
A combination plug running tool, wash sub is run into the ID of the 6-5/8" casing, and a
plug is placed at the bottom of the screen. The screen is then washed by circulating fluid through
the inner string and out through the nozzles of the wash sub. These nozzles are aimed radially
outward to clean the screen to remove any drill cuttings plugging the screen slots or remaining in
the wellbore. Once the hole is clean, the wash sub is removed and the string is run back into the
hole for the filter packing procedure, should a filter be required between the screen and the
wellbore.
Filter packing is performed using low density materials, placed in a uniform layer around
the screen by circulating it through the drill pipe and into the annulus, thus filling the volume
between the screen and the wellbore.
Once the filter packing is complete, a submersible pump can be lowered into the well to
complete the development. Typical well construction is shown in Figure 12.
An alternative completion method involves using well screen in the horizontal section
which has an additional layer of fine mesh well screen to provide san control, in lieu of the gravel
packing.
A variety of other completion methods are being investigated. In cases where the
horizontal section is placed in bedrock, the well can be drilled without the outer casing string, and
the desired production hardware, for example stainless steel or wire-wrapped screens, can then be
run.
Field Test Objectives
The prototype horizontal wellbore system underwent its first field trials in the summer and
fall of 1990, southeast of Houston, Texas. The objectives of the field test were to:
—Test the functionality of the surface equipment, including rig system components and
circulating system.
—Drill a 45 ft. vertical hole to demonstrate casing-while drilling operations; to test the
functionality of the expanding drill bit; and to gain experience making hole in the target formation.
—Drill a horizontal hole with approximately 400 ft. of departure from the wellhead. This
included drilling from a 45 degree slanted rig position and building the hole's inclination along a
100 ft. radius. This curved section would be drilled in 12-1/4" hole and cased in 10-3/4" HOPE
casing, which would then be cemented in place. Then the smaller drilling assembly would be used
to drill the horizontal section and install the 6-5/8" liner simultaneously.
-Complete the horizontal section by pumping HDPE gravel packing material into the
annulus between the casing and the hole wall.
1191
-------
—Drill a second horizontal well, at a true vertical depth of 30 ft, with a horizontal section
exceeding 100 ft.
During test well drilling, project engineers would monitor the performance of system
components, noting areas for improvement.
Field Test Preparation
To prepare for the test, a vertical surface hole 12 ft. deep was augered and the 12-3/4"
conductor was set and cemented in place. A slanted conductor was installed at 45 degrees, close to
the vertical hole and positioned so the rig would not have to be moved to drill through it. Once this
slanted conductor was cemented in place, an unstabilized rotary assembly with a roller cone bit was
used to drill the cement plug and approximately four feet of the formation.
Vertical Test Well
The vertical hole was drilled with a bottomhole assembly comprised of the 8-5/8"
expanding bit, a 4-3/4" drilling motor placed concentrically in the casing; and the 6-5/8" well
casing. The hole was drilled to 60 ft. in one hour, at a flow rate of 150 gpm. The casing easily
ran into the hole, demonstrating that the motor/expanding bit concept could successfully be
applied. Formation was a fine, unconsolidated sand, interspersed with clay stringers. Pockets of
gravel also were encountered.
Directional Test Well
The first borehole drilled from a slanted conductor demonstrated the directional drilling
capabilities of the downhole system. Drilling parameters and operating procedures were varied to
test directional results.
After the vertical hole was drilled, the rig mast was tilted to 45 degrees in preparation for
drilling the horizontal hole. Then the 6-3/4" motor assembly was made up and inserted in the
plastic casing, and together they were lowered into the conductor.
After orienting toolface to high side (for maximum angle build), drilling circulation was
begun at 200 gpm, and the motor stalled almost immediately. It was surmised that this problem
was caused by the condition of the conductor pipe, which still contained some cement which had
not been drilled out. The assembly was retrieved from the hole along with the casing and a stiff
assembly, including a 12-1/4" bit and two stabilizers, was used to drill from the conductor (12 ft.
MD) to 16 ft. MD, providing a straight pathway for the curve-drilling assembly to enter the
formation.
The curve drilling assembly with casing was run into the hole. The motor was started with
a flow rate of 150 gpm, and the assembly was worked up and down until it ran smoothly into the
hole. Drilling commenced at 4 ft./minute. Because there was no identifiable torque created by the
motor, it is likely that the formation was being jetted away ahead of the bit. The formation was an
unconsolidated, very fine sand.
The assembly drilled to 42 ft., but dropped angle at the rate of 0.58 deg./ft. Below 42 ft.
MD, the penetration rate increased to 3.5 ft./minute, but the hole continued to drop angle at .27
degrees/ft, over the next joint to 62 ft. MD.
1192
-------
On the next joint, the flow rate was reduced to 150 gpm and penetration rate dropped to 2.5
ft./minute. Over this hole section, the assembly began to build angle at the rate of 0.43 degrees/ft.
Because the reduced flow rate appeared to help regain control over the angle build, it was
concluded that the fluid was washing the hole diameter. To reduce these effects, the inner string
was tripped out of the hole. The outer-facing bit nozzles were plugged and the forward facing
nozzles were replaced with larger nozzles.
The assembly was placed back inside the casing and drilling was commenced with 150
gpm of circulation. ROP of 3-4 ft./minute was achieved. The assembly built angle at 0.36
degrees/ft. (159 ft. radius).
On the next joint (102 to 122 ft. MD), flow rate was increased to 200 gpm, to improve hole
cleaning. Penetration rate increased to 4 ft./minute, and build rate increased to 0.54 degrees/ft.
(106 ft. radius).
At measured depth of 136 ft (96 ft True Vertical Depth, TVD), the hole had achieved 52.2
degrees of inclination. (See Figure 13).
Changing the bit nozzles had significantly improved the directional performance of the
bottomhole assembly. Armed with this knowledge, the project team decided to start a new well
with a newly-installed slanted conductor.
Casing from the slanted well would be pulled from the hole for re-use on the second
attempt, after installing new HDPE connections using fusion welding techniques.
Horizontal Test Well #1
A second slanted conductor was augered into place approximately 8 ft north of the first one
and cemented into place. After moving the rig, the stabilized rotary drilling assembly was used to
drill out the cement plug and establish contact with the formation.
The curve-drilling BHA used on this borehole varied from that used on the directional well
in that: a) the outside bit nozzles were plugged and two large nozzles were used at the nose of the
bit. This would result in no hydraulic horsepower at the bit, and less hole enlargement, and b) an
increased bit deflection (caused by greater eccentricity of the stabilizers on the motor body) was
used, resulting in an assembly with a theoretical 90-ft turning radius (compared to the 100 ft radius
used on the directional well).
As in the slant well, it was difficult to build angle in the soft formation immediately below
the conductor. The well dropped angle slightly as the first joint was drilled, then held angle to
approximately 63 ft MD. Then the assembly began building angle steadily, reaching 80 degrees of
inclination at 150 ft TD (87 ft TVD), the end of the 10-3/4" casing section.
Due to the low flow rate, pulse heights from the TFI tool had been adjusted to improve the
strength of the signal. This system performed impeccably while drilling the curved section.
Some hole drag and compressive buckling of the casing were experienced during the
drilling of the curve, possibly due to clay and gravel stringers or to some spiraling of the hole. The
drilling assembly was pulled easily from the casing string, and the casing did not move.
1193
-------
The curved casing was cemented into place through the drill pipe by setting a cement plug,
dropping a dart, then pumping cement until it came out the annulus. Once the cement had cured, a
downhole motor-driven milling assembly was used to mill out the plug and retrieve it. A ring left
in the hole was retrieved in one try with a specially-built fishing tool.
After a cleanup trip, the project team was ready to drill the horizontal section. The
downhole system comprised of an 8-5/8" expandable bit, 4-3/4" drilling motor, TFI measuring
device and 2-7/8" drill pipe was run into the hole along with the 6-5/8" HDPE slotted screen.
Once on bottom, the assembly began to drill immediately with no stalling or sticking. At
the flow rate of 150 gpm, the system drilled at 2-3 ft/minute. It was found that pump rates have a
significant affect on hole inclination. When flow rate was increased to 250 gpm to improve hole
cleaning, inclination dropped by 8 degrees while drilling one 20 ft joint.
By orienting the toolface upwards and holding pump rate steady at 150-175 gpm, angle
was built to horizontal and maintained until 400 ft of total departure was achieved. (See Figure
14). The project team believed they could drill further, but drilling was stopped because all test
objectives had been met. Once total depth was reached, the drilling assembly was withdrawn from
the hole.
Completion
One technical objective of the field test was to prove that a slotted casing could be drilled in
place using the dual string drilling technique. This operation was successfully performed with
slotted casing used from surface to total depth.
Several days after drilling was completed, a gravel packing procedure was attempted on the
well. First a plug was set in the bottom of the well, and a wash sub, run on the drill pipe, was
used to clean the well slots (which were 0.020" in width) only in the horizontal section. Pumps
and seals were configured to reverse-circulate 1/8" HDPE pellets into the annulus between the well
screen and the formation. When pumping commenced, it was found that the hole wall had bridged
into the casing somewhere in the curve about the horizontal section, preventing gravel packing
material from reaching the bottom of the hole. Work continues toward perfecting this; gravel
packing technique. Future gravel packing operations, for example, probably will circulate through
the drill pipe and use slotted screen only in the zones of interest.
The project team also has investigated completion techniques that are less complicated than
the gravel-packing method. Specifically, a new completion string, combining a fine mesh stainless
screen with the HDPE slotted casing, has been designed since the initial field test. This system
should provide adequate sand control in most situations.
Horizontal Wellbore #2
Based on the results of the intial test program some system components were modified and
a second horizontal wellbore was planned. This test well would place approximately 130 ft of
horizontal screen at a target depth of 30 ft. An additional objective would be to fully test all
equipment to be included in the commercial drilling system, some of which were not available
when the first horizontal test well was drilled.
1194
-------
Once the system was assembled, the rig mast was slanted to 60° from vertical and the
9 ft surface hole was augered with a special slant augering assembly. The surface casing was
then cemented in place.
The hole beneath the surface pipe was drilled out approximately 3 ft using the augering
assembly, then the curve drilling assembly, with casing, was run into the hole. After some initial
difficulty in beginning the kickoff (which was corrected by adjusting drilling parameters), the
curve was drilled according to plan, reaching horizontal at a depth of 30 ft below the surface. This
operation took 4 -1/2 hours.
The casing was cemented by pumping cement through the drill pipe then waiting for the
cement to harden before retrieving the casing plug from the hole. Once the cement was cured, an
expanding bit with a tri-cone pilot bit was used to mill out cement remaining at the casing shoe.
Then the horizontal section was drilled, using the same assembly utilized on the previous
horizontal well. A horizontal section of 129 ft was drilled in three hours, for an average
penetration rate of 43 ft/hr. Slotted casing was installed as the hole was drilled.
This second horizontal well also demonstrated the effectiveness of the hydraulic pipe
handling system, which manipulated the dual drill string safely and efficiently.
Figure 15 is a plot of this second horizontal well.
Conclusion
In conclusion, a new horizontal drilling and sampling system has been designed and built
to meet the special requirements of the environmental industry. A prototype system has been
successfully field tested, refined and introduced for commercial use. We believe there will be
many applications for the new system as the environmental industry begins to remediate
contaminated soil and groundwater.
In the future, other technologies are likely to be added to the horizontal wellbore system.
These innovations could include methods for obtaining undisturbed formation samples and
containerized gas samples beneath landfills and buildings; geophysical logging services adapted for
horizontal data acquisition; and completion technology to isolate zones along the horizontal well for
selective sampling and completion.
References
1. H. Karlsson, R. Bitto, "Worldwide Experience Shows Horizontal Wells Success,"
WORLD OIL, March 1989.
2. D. Langseth, A. Smith, "Hydraulic Performance of Horizontal Wells," paper presented at
Superfund 90 Conference, Washington, D.C., November 26-28, 1990.
3. D.S. Kaback, B.B. Looney, J.C. Corey, L.M . Wright III, and J. Steele, "Horizontal
Wells for In-Situ Remediation of Groundwater and Soils," paper presented at the National Water
Well Association Outdoor Action Conference, Orlando, FL, May 22-25, 1989.
1195
-------
4. H. Karlsson, T. Brassfield, V. Krueger, "Performance Drilling Optimization," paper
SPEC/IADC 13474, presented in New Orleans Louisiana at the March 1985 Drilling Conference,
sponsored by the Society of Petroleum Engineers and the International Association of Drilling
Contractors.
5. H. Karlsson, R. Bitto, "New Horizontal Wellbore System for Monitor and Remedial
Wells," paper presented at Superfund '90 Conference, Wshington, DC, November 26-28, 1990.
Authors and Address:
Ronald Bitto, Haraldur Karlsson and Gary E. Jacques
Eastman Christensen Environmental Systems, a Baker Hughes Company
15311 Vantage Parkway West, Suite 320
P.O. Box 670968
Houston, Texas 77267
(713) 442-4895
1196
-------
TABLE 1: SYSTEM SPECIFICATIONS
Depth of Horizontal Section:
Horizontal Length:
Screen Size in the Horizontal Section:
Casing Size in the Curve Section:
Casing and Screen Material:
Horizontal Placement Accuracy:
Pumping Specifications:
Seal Specifications:
18 ft to 500+ft below surface
More than 500 ft
6-inch nominal (6-5/8" OD)
10-inch nominal (10-3/4" OD)
High density polyethylene pipe
True vertical depth: +/- 2 degrees
Azimuth: +/- 2 degrees
Submersible pump ahead of screen
Sand pack or other filter
INJECTION
7VACUUM
"lv^^
Figure 3: Soil Gas Extraction. VOC's are stripped from
soil using parallel horizontal wells.
Figure 4: In-Situ Remediation. Horizontal well efficiently
conveys bioremediation materials to plume.
1197 Figure 5: Barrier to Transport of Contaminants.
Horizontal wells beneath a landfill protect groundwater,
-------
Aulomrtrd
Sl»nt Rig
TARGET ZONE
-n^yyn^^ ,• i^«>t^»v^7*>y T*7T!*^*TTr*r*!*.T, '.**.J*-
Finer Pack
Figure 6: Horizontal Wellbore System
Horizontal Wellbore System Drilling Rig
Mud Cfeantr
Offk*
PIp« Trtdtr
VIEW AT REAR OF TRAILERS
Figure 7: Surface Equipment Package
1198
-------
wsst/f/ss'M/'msMss////m//ms/s/////s/fss'ssm//s/f&//fm/'^^^
18HTVO
End of Cum
(60 n from Wellhead)
-y* lOSttTVD
End of Curv» ** |0 300+ ft TVD
(10ontrom Wetlwad)
Figure 8: Depth Capability. Unit can place horizontal
sections at any specified depth.
10-3/4" O.D. Casing
Grouted In Curv*
Hydraulic
Motor Expanding
Drill Bit
Casing Tool Face
C«ntrallzer» Indicator Stabilizer*
Figure 9: Downhole equipment permits simultaenous drilling
and casing operations.
1199
-------
12" Conductor
PIpeRipe
Production
, Tubing
10" HOPE Casing
12.1/4" Hole
Cement/
Grout Seal
to Surface
Not to Seal*
Submersible Pump 6" HOPE Slotted Liner
Figure 10: Typical horizontal well construction
Plot of directional test well
a."
0)
a"
"o"
o -
"u «
a>
0) »
3
u •
II
V
See* i 2 00
Figure 11: Directional test well.
1200
-------
Detail of lower drilling assembly
Concentric Upper
Casing Stabilizer
Concentric Lower
Casing Stabilizer
• To drill (he curve, eccentric stabilizers on the
motor create bit offset and result in an assembly
which will build angle.
Figure 12: Curve Drilling Process
Bit Offset
Calculation of Buildup Rates
Figure 13: Calculation of Build-up Rates
120.1
-------
Plot of horizontal test well
3 n
f «0
j
I w
v
» • a w ri too t» IM n
SOO(« 1 : 12.50 »*U.I t-rtU. « M.M or^H ,
~~i—i—i—i—i—i—i—r
n» ro JM m
—i
4»
Figure 14: Vertical section of first horizontal
test well,at 100 ft TVD.
20
10
0
-9 10
= 2°
% 30
O 40
5 50
60
10 0 10 30 50
70 90 110 130 150 170 190 210
Vertical Section
Figure 15: Vertical section of second horizontal
test well, at 30 ft TVD.
1202
-------
Soil-Bentonite Backfill Mix Design/Compatibility Testing:
A Case History
Jane M. Bolton, E.I.T., David L. Jaros, P.E.,
Gordon G. Lewis, James M. Zeitinger, P.G.
U.S. Army Corps of Engineers
Omaha District
215 No. 17th Street
Omaha, Nebraska 68102
(402)221-4169
INTRODUCTION
Soil-bentonite slurry trenches have been used in the U.S. as subsurface groundwater barriers since
the 1940's (D'Appolonia, 1980). Construction consists of excavating the trench (typically 2-5 feet
wide, keyed 3-5 feet into an impermeable formation such as rock or clay) while pumping in bentonite
slurry to support the side walls. As slurry leaks into voids in the trench wall soils, clay particles build
up in layers on the trench walls, forming a thin low permeability filter cake. The trench is then
backfilled with a mixture of soil and bentonite, called the soil-bentonite backfill material. Backfilling
with material of the proper consistency (unit weight about 15 pounds per cubic foot (pcf) greater than
the slurry unit weight, with a concrete slump of 2 to 6 inches) does not substantially destroy the filter
cake (D'Appolonia, 1980; Millet and Perez, 1981). Permeability of the completed trench is a function
of both the filter cake and the soil-bentonite backfill material. The term "bentonite" is defined in the
U.S. Environmental Protection Agency (USEPA) slurry trench design guidance document as a soil
composed of at least 90 percent montmorillonite clay (JRB Associates, 1984). Many geotechnical
textbooks, such as Lambe and Whitman (1969), define bentonite as montmorillonite clay containing
primarily sodium as the exchangeable ions in its crystal structure. This paper utilizes the USEPA
guidance document definition of bentonite.
The presence of chemical contaminants in soil and/or groundwater may significantly alter the rate
of water movement through a soil-bentonite slurry trench (D'Appolonia, 1980; JRB Associates, 1984;
Zappi et al., 1989b; Ayres et al., 1983). For example, calcium in soil or groundwater will displace
some of the sodium ions in bentonite. This results in reduced swelling and increased permeability,
not desirable for a groundwater barrier. While the effects of other individual chemicals have been
studied and documented, the effect of multiple contaminants, which frequently exist at hazardous and
toxic waste (HTW) sites, is largely unknown.
This paper presents a general overview of the Corps of Engineers Missouri River Division Laboratory
(MRDL) mix design/compatibility testing methodology, while discussing in detail the testing program
undertaken for the Lime Settling Basins (LSB) site at the Rocky Mountain Arsenal (RMA), Commerce
City, Colorado. Objectives of the LSB testing program are to determine the optimum soil-bentonite
backfill material mix design (soil and percent bentonite) necessary to achieve an in-place slurry trench
permeability of 1 x 10-7 centimeters per second (cm/sec) or less, and to determine whether
contaminants present in soil and groundwater at the LSB site will cause changes in soil-bentonite
backfill permeability over time.
1203
-------
BACKGROUND
Site History. During the 1940's and 1950's, wastewater from production of Army agents was routinely
treated prior to discharge to unlined evaporation ponds. This treatment involved the addition of lime
to the wastewater to precipitate metals, principally mercury and arsenic. Wastewater produced in the
South Plants was channeled into the LSB prior to gravity drainage to Basin A, an evaporation pond
just to the north. The precipitation process produced a lime sludge that contained elevated levels of
heavy metals, arsenic and mercury. Subsequent discharge of wastewater from production of pesticides
resulted in the addition of pesticides to the LSB sludge. The LSB were removed from service in 1957.
Studies have been conducted to characterize the nature and extent of contamination in the soil, sludge,
and ground water in the vicinity of the LSB. The studies revealed the soil, sludge, and ground water
contain elevated levels of organochlorine pesticides, organosulfer compounds, arsenic, mercury, and
Inductively Coupled Plasma (ICP) metals (cadmium, chromium, copper, lead, and zinc).
Site Geology. Bedrock beneath the LSB area is the Cretaceous-Tertiary Denver Formation. The
Denver Formation in the vicinity of the LSB consists of claystone and sandstone. The claystone is
generally soft to moderately hard, brown to gray, and is occasionally silty. A thick, fine-grained
sandstone lense is present in the northern section of the LSB area. The Denver Formation bedrock
lies at depths of 13.0 to 33.0 feet below the surface in the LSB area. The local slope of the bedrock
subcrop is about two degrees to the north-northeast. The dip of the Denver Formation has not been
determined, but it is probably the same as the regional dip, about one degree or less to the southeast.
The overburden in the LSB area consists of Recent fill and Quaternary eluvial and alluvial deposits.
The thickness ranges between 13.5 and 27.5 feet. Recent fill is present almost throughout the entire
area and consists mostly of sludge removed from the LSB. The fill thickness ranges from 3 to 10 feet.
The eluvial and alluvial materials consist mostly of poorly graded, silty, fine-grained sand with
moderate amounts of sandy, silty clay and minor amounts of clayey sand, sandy clay, silty clay, and
lean clay.
The contaminated aquifer is within the overburden and the material is essentially the same as that
described above. The majority of groundwater movement occurs in unconsolidated, fine-grained
sand and/or silty, fine-grained sand and clayey, fine-grained sand. The thickness of the aquifer
ranges from 9.5 to 21.0 feet. The aquifer is unconfined and overlies the top of bedrock.
Contamination. Soil contamination in the LSB consists of raw materials, such as mustard agent
production-related compounds; manufacturing by-products, such as volatile aromatic solvents; and
degradation products from the synthesis of pesticides. Organochlorine pesticides that have been
detected are dieldrin, aldrin, endrin and isodrin. Other contaminants detected were organosulphur
compounds of chlorophenylmethyl sulfide, chlorophenylmethyl sulfoxide, and chlorophenylmethyl
sulfone. DDT was also detected in an isolated area. Volatile organic compounds consist of
chloroform, benzene, and chlorobenzene. The most prevalent metals are arsenic and mercury.
Elevated concentrations of copper, lead, zinc, cadmium, and chromium were also detected.
Groundwater contaminants in the unconfined aquifer include volatile organic compounds, aromatics,
metals, and organochlorine pesticides.
Arsenic, mercury, chromium, and copper are metals that have been detected in the ground water.
Decision Document Summary. The Interim Response Action for the LSB consists of moving the lime
sludges currently located around the basins into the basins, a 360-degree subsurface groundwater
barrier (slurry trench) around the basins to prevent migration of contaminated groundwater, a
groundwater extraction system inside the isolation cell to maintain an inward hydraulic gradient, and
1204
-------
a soil and vegetative cover over the cell to reduce infiltration of rainwater (Wood ward-Clyde
Consultants, 1990).
Pre-Design Field Investigations. Field investigations were conducted during June and July 1990.
Investigations consisted of: electro-magnetic surveys for locating buried metallic objects (none were
found); exploratory drilling and soil sampling in the LSB area; slug tests for hydraulic conductivity
analysis; groundwater and tap water sampling; and bulk soil sampling of borrow areas. All
investigations except the borrow investigations were conducted in level B personal protective
equipment.
A total of 30 borings were drilled for this investigation. Nineteen borings were drilled along the
alignment of the proposed slurry cutoff trench to identify the subsurface materials and to determine
the consistency, density, and moisture content of the overburden; and also to determine the depth and
characteristics of the claystone bedrock for design of the base of the proposed slurry trench. Eight
borings were drilled outside the slurry trench area to further define the extent of the lime sludge
material. Three wells were installed inside the slurry trench area for slug tests to determine the
hydraulic conductivity of the overburden aquifer. Split-spoon samples were taken from all borings
for geotechnical analyses, compatibility testing, and chemical analyses. All drill holes were backfilled
with cement grout after completion.
Development of Laboratory Testing Methodology. In developing the MRDL's test equipment and
procedures, various references were researched including work done by David J. D'Appolonia (1980),
the U.S. Army Corps of Engineers Waterways Experiment Station (WES) (Zappi et al., 1989a, 1989b),
the USEPA (JRB Associates, 1984), Dr. David Daniel (Daniel et al., 1984), and Goldberg-Zoino &
Associates (GZA) (Ayres et al., 1983). The MRDL procedures were patterned after the work done
in 1981 by GZA during design and construction of the Gilson Road Superfund Site cutoff wall.
Procedural and equipment modifications were made at the MRDL based on early trial runs to address
site specific conditions and speed up the overall test process. However, the basic concept of
optimizing the mix design prior to long term compatibility testing was adhered to.
In reviewing the literature, there appeared to be no consensus on which type of permeameter, fixed
wall or flexible wall, produced more realistic results. Each type of permeameter has its advantages
and disadvantages and both can yield grossly misleading results under certain circumstances. Based
on ease of operation and relatively expedient and reproducable results, fixed wall permeameters were
selected for the mix design optimization phase. The flexible wall permeameter was selected for the
long term compatibility phase because of its ability to accurately model various anticipated field stress
conditions.
The equipment was designed and built at the MRDL with input from USAGE engineers, technicians,
and shop personnel. To prevent degradation of test equipment, anodized aluminum base and top caps,
brass stones, stainless steel valves, teflon tubing, and glass burrettes were used. This allowed for
multiple use of most of the equipment components after decontamination of the system prior to
testing.
Backfill Soil Selection
To obtain a low permeability (typically 1 x 10-7 cm/sec or less is specified for completed
soil-bentonite slurry trenches), soil with an appreciable amount of fines is necessary for the
soil-bentonite backfill.
The USEPA recommends the following gradation criteria for backfill soils: maximum particle size
of 5 inches, 65-100 percent passing 3/8 inch sieve, 35-85 percent passing the U.S. standard sieve #20,
1205
-------
and 20-50 percent passing the U.S. standard #200 sieve. Plastic fines are preferred but not necessary
(JRB Associates, 1984).
Soils excavated from the trench may be utilized for the backfill soil. This practice saves the time and
money of locating, purchasing, developing, and hauling borrow soil to the site as well as disposal of
the excavated soil. However, if the in situ soil is not suitable (for example coarse gravel I or is
contaminated (as is often the case at HTW sites) imported borrow soil may be the only viable option.
Due to contamination of the in situ soil, the work plan called for testing of both in situ soil and a
borrow source. Originally, a clay borrow area used in previous remediation projects at RMA was
suggested. However, the clay borrow area is located in a bald eagle habitat which is closed to traffic
from November 1 to April 1 and the amount of clay soil remaining is limited. Therefore stockpiles
of soil excavated from the Lower Derby Dam spillway construction at the Arsenal were selected as
the primary borrow soil. Soil from the clay borrow area would be used as a source of fines only, if
necessary, to blend with either in situ or random fill borrow soil to achieve a low permeability.
Soil samples from several of the borings along the trench centerline were to be blended to form one
composite in situ sample for mix design optimization and compatibility testing. During blending,
however, the reddish brown soil developed a yellow staining over approximately 30 percent of the
surface over one night. At that point Corps personnel decided not to consider the in situ soil for use
in the trench or further testing because of potential field handling problems.
Figure 1 shows the grain size distribution and Atterberg limits for the random fill and clay borrow
soils. The random fill soil contains more fines than EPA recommends. This is not considered to be
a problem since a finer soil will make a low permeability easier to obtain.
Bentonite Selection
General. To obtain a general idea of the effect of site contaminants on bentonite, samples of the
following four bentonites were obtained for this study:
S-5 Natural, Black Hills Bentonite, Rapid City, SD
BH-Natural, H&H Bentonite, Grand Junction, CO
Bara-kade 90 SP, NL Baroid, Houston, TX
Bara-kade 90, NL Baroid, Houston, TX (treated)
The Corps of Engineers' slurry trench guide specification requires use of premium-grade, ultrafine,
natural sodium cation-based montmorillonite powders (Wyoming-type bentonite) that conforms to
American Petroleum Institute (API) Specification 13A, Sections 5, 12 and 13 (API, 1990).
However, most commercially available bentonite is treated and conforms to Section 4, not 5 of API
Specification 13A. Bara-kade 90 is the only bentonite studied which is treated and therefore
conforms to Section 4 of API Specification 13A. Bara-kade 90 is the same bentonite as Bara-kade
90 SP, but one-quarter pound of a polymer is added per ton of bentonite to produce Bara-kade 90
(Anderson, 1991).
Free Swell Tests (McCandless and Bodocsi, 1987). "Free swell" is the increase in volume of a soil
from a loose dry powder form when it is poured into water, expressed as a percentage of the original
(dry) volume. Two grams (2.2 cubic centimeters) of bentonite is slowly poured into 100 milliliters
(ml) of water, and the volume of settled solids is recorded after 2 and 24 hours. For this study, two
tests were performed for each bentonite; one using tap water from the Arsenal and one using
contaminated groundwater from the site. Table 1 shows results of the free swell tests. Percent
1206
-------
1HOI3M AB M3SUVOD IN3DH3d
S ? S
S S §
1HOI3M A8 M3NIJ iN30M3d
1207
-------
24-hour swell is the percentage of the "final" (24 hour) swell achieved after 2 hours (tap water
samples). Percent tap water swell is the percentage, at the given time, of the tap water sample swell
achieved by the groundwater sample. Contaminants decreased the percent swell of all the bentonites,
with Bara-kade 90 exhibiting the greatest decrease (about 50 percent). S-5 takes longer than the others
to achieve "final" swell with both tap water and groundwater. The free swell behavior of BH-Katural
and Bara-kade 90 SP is very similar, with Bara-kade 90 SP showing a slightly higher percent 24-hour
swell after 2 hours and percent tap water swell with groundwater.
Filter Cake Compatibility Tests (D'Appolonia, 1980). As stated previously, the filter cake is an
important component of a completed slurry trench. Filter cake permeabilities may be as low as 10-9
cm/sec (Xanthakos, 1979). For this reason filter cake compatibility tests, in addition to free swell
tests, were used to evaluate bentonite performance. Slurry from each bentonite (prepared using RMA
tap water) was placed in fixed wall permeameters. Slurry was forced through filter paper overlying
a porous stone at the bottom of the chamber by a chamber pressure of 10 pounds per square inch (psi)
for 24 hours. During this time a filter cake of approximately one-half inch formed on the filter
paper. The bentonite slurry was removed with a vacuum bulb and immediately replaced with either
RMA tap water or contaminated groundwater (one of each for each bentonite, for a total of eight
tests). Water was forced through the filter cakes by a chamber pressure of 2-3 psi. The volume of
effluent was measured two or three times a day for two to five days and the permeability was
calculated.
The USEPA recommends the following properties for bentonite slurries: viscosity (measured with
a Marsh funnel) greater than 40 seconds, unit weight around 65 pcf, pH between 7 and 10, and a
bentonite content of 4 to 8 percent (JRB Associates, 1984). Millet and Perez (1981) recommend;
viscosity greater than 40 seconds, unit weight around 65 pcf, and pH between 6.5 and 10.
D'Appolonia (1980) recommends; viscosity greater than 40 seconds, and bentonite content of 5 to 7
percent. In this filter cake study all bentonite slurries were prepared with 6 percent bentonite by
weight.
Marsh funnel viscosity, unit weight, and pH were measured for each slurry and are listed in Table
2. Properties of all slurries lie within the recommended ranges.
Figures 2 and 3 show results of filter cake compatibility tests. Some filter cakes formed cracks upon
initiation of the flow phase of testing. After test completion, cutting the filter cakes into quarters
revealed the cracks extended most or all the way through the filter cakes. However, presence of
cracks did not appear to affect the permeability of the filter cakes. All bentonites except Bara-kade
90 SP exhibit a slight downward trend in permeability over time. Bara-kade 90 shows the least
variation in permeability between tap water and groundwater. The reason for the drop in
permeability of Black Hills S-5 (tap water) between 1390 and 1770 minutes is not known.
Selection. The original work plan called for selecting the bentonite which showed the least variation
in filter cake permeability and percent swell between tap water and groundwater for use during
further testing.
However, the bentonite which exhibited the least variation in filter cake permeability (Bara-kade 90)
exhibited the most variation in percent swell. The designers eliminated Black Hills S-5 due to the
drop in filter cake permeability in tap water between 1390 and 1770 minutes and Bara-kade 90 due
to the large difference in percent swell between tap water and groundwater. Bara-kade 90SP was
chosen because it shows slightly less variation in both percent swell and filter cake permeability
between tap water and groundwater than BH-Natural and it shows a slight increasing trend in filter
cake permeability over time. A 6 percent Bara-kade 90SP bentonite (by weight) slurry was used in
all subsequent testing.
1208
-------
Table 1.
Free Swell Test Results
Bentonite
Black Hills
S-5
H&H Bentonite
BH-Natural
NL Baroid
Bara-Kade 90
NL Baroid
Bara-Kade 90SP
Bentonite
Black Hills
S-5
H&H Bentonite
BH-Natural
NL Baroid
Bara-Kade 90
NL Baroid
Bara-Kade 90SP
Tap % 24- Ground
Water Hour Water
Time % Swell Swell % Swell
2 hr. 530 73 445
24 hr. 720 490
2 hr. 785 91 560
24 hr. 855 560
2 hr. 785 83 400
24 hr. 945 400
2 hr. 765 94 560
24 hr. 810 560
Table 2.
Bentonite Slurry Properties
Filter Cake Compatibility Tests
Marsh Funnel
Viscosity
("seconds) Density (pcf)
1. 48 64.9
2. 48
3. 48
1. 52 65.0
2. 51
3. 52
1. 61 65.1
2. 62
3. 64
4. 64
1. 44 65.1
2. 44
3. 44
% Tap
Water
Swell
83
68
71
65
51
42
73
69
pH
8.7
8.8
9.5
9.1
1209
-------
o
LU
(/)
X
2
o
>-
h-
cr
LU
0.
Figure 2
Filter Cake Compatibility Test Results
BLACK HILLS S-5
GROUNDWATER
TAP WATER
1234
TIME (X 1000 MIN)
BARA-KADE 90
o
LU
00
X
^
o
CD
<
cc.
LU
Q_
1234
TIME (X 1000 MIN)
1210
-------
o
LU
O
_J
I—I
03
LU
CC
LU
1 E
9 E
8 E
7 E
6 E
5 E
4 E
3 E
1 E
9 E
~r
-8
Q
— ft
-8
-8
-8
-8
-9
Figure 3
Filter Cake Compatibility Test Results
BARA-KADE 90 SP
0
/
"X
/"
s
\
2345
TIME (X 1000 MIN)
7
H&H BH NAT
o
LU
I/)
o
>-
I—I
_J
t—H
LU
CC
LU
Q-
GROUNDWATER
TAP WATER
1234
TIME (X 1000 MIN)
1211
-------
Mix Design Optimization
General. The purpose of this phase of testing is to determine the most economical mix of soil, dry
bentonite, and bentonite slurry which will produce an in-place slurry trench permeability less than
or equal to Ix 10-7 cm/sec. Because mixing and placing operations are less controlled in the field
than in the laboratory, the designers specified a maximum laboratory permeability of 5 x 10-8 cm/sec
for evaluation purposes.
Since borrow soil is available nearby at RMA, bentonite is the highest cost item. The HTW testing
technical advisor assumed at some point it would be less expensive to decrease the permeability of
soil-bentonite backfill material by adding additional fines (from a clay borrow area), rather than
additional bentonite, to the random fill borrow soil. The "upper limit" bentonite content was set as
4 percent dry bentonite. Bentonite slurry is then added to the mixture to achieve a (concrete) slump
between 4 and 6 inches.
Procedure. The work plan called for preparation of three samples of backfill soil with the addition
of 0, 2, and 4 percent dry bentonite by weight. Bentonite slurry with a Marsh funnel viscosity of
about 40 seconds is added to each sample to achieve a (concrete) slump of 4 to 6 inches. If fixed wall
permeameter tests of 48 to 72 hours duration did not measure a permeability less than or equal to 5
x 10-8, clay borrow soil would be added to the random fill borrow soil to produce samples with
approximately 10 percent greater fines content than the random fill borrow soil. The procedure
(addition of dry bentonite and bentonite slurry, fixed wall permeameter tests) would be repeated. If
measured permeabilities were still greater than 5 x 10-8 cm/sec, additional clay borrow soil would
be added to produce samples with approximately 20 percent greater fines content than the random
fill borrow soil. If measured permeabilities (after addition of dry bentonite and bentonite slurry)
were still greater than 5 x 10-8 cm/sec, additional clay borrow soil would be added to produce
samples with approximately 30 percent greater fines content than the random fill borrow soil.
Testing. The HTW testing technical advisor intended carrying out these tests in duplicate, using RMA
tap water as the only permeant. The project designers misunderstood and requested one set of tests
be performed using RMA tap water as permeant and one set be performed with contaminated
groundwater as the permeant. In the first tests performed a few of the permeameters emptied of
permeant over one night. The head pressures were 2 psi and the initial permeant volumes were
approximately 200 ml. Examination revealed these specimens appeared to have contracted (specimens
pulled approximately one-eighth of an inch away from the permeameter), pointing to a physical
change as a result of some reaction with the permeant. To prevent preferential flow of permeant
between the permeameter wall and the sample, the permeameters had been coated with a bentonite
paste (approximately 17 percent bentonite and 83 percent water by weight). The bentonite paste wall
coatings were not evident at this point. These conditions occurred more frequently in the specimens
permeated with contaminated groundwater, but also appeared in tap water permeated specimens as
well. It was initially suggested that these failures may have been due to some lattice collapse in the
bentonite resulting from ion exchange. The same or a similar process might possibly cause the cracks
observed during filter cake compatibility tests.
The HTW testing technical advisor suggested attempting to discover the cause of the rapid permeant
loss. In the interest of proceeding with testing, the advisor suggested, and the designers concurred,
a triaxial permeability test be conducted using a 2 percent dry bentonite mix. Since the random fill
borrow soil contains 51 percent fines and little difference exists in the grain size distributions of the
two borrow soils (Figure 1), the addition of fines from the clay borrow soil would likely have a
negligible effect on the permeability of the mix. Early results from a successful fixed wall
permeability test indicated a permeability of approximately 5 x 10-8 cm/sec for a 2 percent dry
bentonite mix.
1212
-------
While the triaxial test was being started, an investigation of the failed fixed wall tests was undertaken.
Two paste coated jars, one filled with tap water and the other with contaminated groundwater were
prepared. Several days of exposure to the liquids resulted in the tap water having a more detrimental
effect on the paste than the groundwater. This was in contrast to the greater frequency of failed
groundwater permeated fixed wall tests. Next, one still intact fixed wall test specimen was allowed
to flow until the entire volume of permeant passed through it. Several hours later it appeared
identical to the failed test specimens; the sample appeared to contract and the bentonite paste coating
was missing.
This (very limited) investigation suggested that due to high permeability, cracking of the specimen,
leakage along the permeameter walls, or a combination of the factors, permeant was forced through
and/or around the specimen. Continued pressure application with no permeant caused drying of both
the specimen and the bentonite paste. (The paste has a high water content (500 percent)). Drying
could cause specimen shrinkage and give the appearance of a physical change due to some chemical
reaction.
The HTW testing technical advisor thought not enough time was allowed between specimen set up and
the start of flow. Persons at WES familiar with this type of testing concurred. All future fixed wall
soil-bentonite backfill permeability testing will be run after incrementing the applied head pressures
slowly over the course of several days.
Triaxial Permeameter Test Results. Figure 4 shows the results of the triaxial permeameter
optimization test. The average permeability, approximately 4 x 10-8 cm/sec, is lower than the
specified maximum of 5 x 10-8 cm/sec. Therefore the optimum mix design is 2 percent dry bentonite
by weight and bentonite slurry added to the random fill borrow soil.
D'Appolonia (1980) recommends the following properties for soil-bentonite backfill material: slump
between 2 and 6 inches, unit weight at least 15 pcf greater than the slurry unit weight, water content
between 25 and 35 percent, minimum bentonite content of 1 percent, and a minimum fines content
of 20 percent. Millet and Perez (1981) recommend a slump of 4 to 6 inches and a bentonite content
of 2 to 4 percent. The USEPA recommends a bentonite content of I to 2 percent, water content of
25 to 35 percent, fines content of 20 to 60 percent, slump of 2 to 7 inches, and a unit weight at least
15 pcf greater than the slurry unit weight (JRB Associates, 1984). Table 3 lists physical properties
of the triaxial permeameter specimen. All properties lie within the recommended ranges except water
content. The reason for the high water content and it's effect on long-term permeability (if any) is
not known.
Long Term Compatibility Tests
Flexible Wall Permeameter Equipment. The basic components of MRDL's flexible wall permeameter
setup are: 1) Six modified triaxial permeameter cells, each consisting of anodized aluminum top and
bottom cell bases, a clear lucite cylinder, anodized aluminum top and bottom specimen caps and brass
porous stones; 2) Separate inflow and outlow glass burettes for flow quantity measurements; 3) Three
pressure regulators with associated pressure gauges to control and monitor cell pressure, inflow, and
outflow pore pressures; and 4) A stainless steel control panel with appropriate stainless steel valves,
teflon tubing and spill containment tray. The LSB testing program utilizes air as a pressure source.
For some permeant liquids, an inert gas (such as nitrogen) should be the pressure source to minimize
biodegredation within the liquid.
Procedure. The test procedure can be broken down into six steps. The first step consists of forming
a cylindrical specimen approximately 2.8 inches in diameter by 2.0 inches high out of the selected soil
bentonite mix from the mix design optimization phase. This is done by using the bottom specimen
1213
-------
Figure 4
o
Ld
O
CQ
<
UJ
^
cc
LU
a.
1
9
8
7
6
5
4
3
2
1
9
E-7
E-8
E-8
E-8
E-8
E-8
E-8
E-8
E-8
E-8
E-9
0
Triaxial Optimization Test
Borrow SoiI and 2% Dry Bentonite
3456
TIME (X 1000 WIN)
8
1214
-------
Table 3.
Physical Properties - Triaxial Optimization Test
Total Percent Bentonite
Slump
Wet Density
Dry Density
Saturation
Void Ratio
Water Content
4.2 percent
6.125 inches
112 pcf
71.5 pcf
100 percent
1.35
56.6 percent
Property
Table 4.
Physical Properties - Compatibility Tests
2% Dry Bentonite 4% Dry Bentonite
Specimen 1
Specimen 2
Specimen 3
Total Percent
Bentonite 3.7
Slump (inches) 4.5
Wet Density (pcf) 109
Dry Density (pcf) 73
Saturation (%) 100
Void Ratio 1.31
Hater Content (%) 49.3
3.7
4.5
108
72
100
1.35
50.0
6.0
5.75
104.5
67
100
1.52
55.9
1215
-------
cap and a latex membrane sleeve within a perforated plastic cylinder as a specimen raold.
Soil-bentonite backfill material is carefully spooned into the mold in two lifts and rodded lightly to
produce a homogeneous low density mass. After taking the necessary specimen measurements and
weights, top cap is set and the cell is assembled. Step 2 consists of filling the inflow and outflow
burettes and porewater lines with site tap water and the chamber with deaired water after making the
appropriate connections to the control panel. Step 3 consists of backpressure saturating the specimen.
Step 4 consists of consolidating the specimen to simulate field stress conditions. Step 5 consists of
flow initiation from bottom to top within the specimen using a relatively low hydraulic gradient (e.g.
28). Inflow and outflow quantities are monitored until the rate of inflow equals the rate of outflow
for at least 5 consecutive daily readings. In addition, at least one pore volume of water must flow
through the specimen prior to introducing site (contaminated) groundwater. As with tap water,
groundwater inflow and outflow are monitored and the test is run until at least two pore volumes of
groundwater pass through the specimen. The final step consists of removing the specimen, obtaining
final weights, measurements, moisture contents etc. Three test conditions are being evaluated: two
specimens of the "optimum" mix design of 2 percent dry bentonite and bentonite slurry added to the
random fill borrow soil and one specimen with 4 percent dry bentonite and bentonite slurry added
to the borrow soil. After one pore volume of tap water passes through the samples, two of them (one
optimum mix sample and the 4 percent dry bentonite sample) will be leached with contaminated
groundwater. Results of the two tests using groundwater as the permeant can be compared to see
whether a backfill with a higher bentonite content reduces changes in backfill permeability over time.
Occasional sampling and chemical analysis of the effluent permeant is done to determine the
effectiveness of the soil-bentonite backfill material in preventing migration of contaminants through
the specimen. It is recommended that the flow phase of the tests be run at least two months to
provide meaningful results concerning the effects of the groundwater on the soil-bentonite backfill
material.
Testing. Long-term compatibility testing began in early March 1991. Presently the first pore volume
of RMA tap water is flowing through the specimens. MRDL personnel anticipate beginning
groundwater permeation (for two of the samples) sometime during the week of April 1, 1991.
Therefore, the effect of site contaminants on the permeability of the soil-bentonite backfill material
is not known at this time. Tap water permeabilities are averaging between 4 x 10-8 cm/sec and 5 x
10-8 cm/sec, similar to values obtained during the mix design optimization phase. Table 4 lists
physical properties of the test specimens. Water contents are higher than recommended values for (as
yet) unknown reasons.
The small volume of effluent to be produced precludes performing a wide range of chemical testing.
Sodium, calcium, and total organic carbon tests will be performed after each pore volume has moved
through the samples. An increase in the amount of sodium and a decrease in the amount of calcium
in the permeameter effluent could indicate displacement of sodium ions in bentonite by calcium ions
from the groundwater.
CONCLUSIONS
The following list of conclusions is to be considered incomplete due to the ongoing compatibility tests.
General Testing Methodology
(1) When designing soil-bentonite slurry trenches through highly contaminated areas, at least one
uncontaminated imported borrow soil should be investigated and tested for use in the
soil-bentonite backfill material. If the in situ soil contains too many contaminants for use,
mix design and compatibility testing of the borrow soil can continue without delay.
1216
-------
(2) Due to the variability of commercially available bentonites, several should be evaluated for
suitablility with site tap water and contaminated groundwater. The evaluation process should
include both free swell and filter cake compatibility tests.
(3) When soils used in soil-bentonite backfill material contain a significant amount of fines,
addition of fines during optimization testing as planned in this study may not be necessary.
(4) During rigid wall permeameter testing the applied head pressure should be incremented slowly
over several days.
LSB Backfill Mix Design
(1) Addition of 2 percent dry bentonite and enough bentonite slurry to achieve a concrete slump
between 4 and 6 inches to the borrow soil produces a soil-bentonite backfill material with a
laboratory permeability less than 5 x 10-8 cm/sec.
DISCLAIMER
This paper is not intended to address every conceivable HTW site condition or all possible applications
of soil-bentonite backfill mix design and/or compatibility testing. Mentioned commercial products
are not the only products of their kind available. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
REFERENCES
American Petroleum Institute, (1990), Specification ISA; Specification for Drilling-Fluid Materials,
July 1, 31 p.
Anderson, Jim, (1991), NL Baroid, Personal Communication with J.M. Bolton, U.S. Army Corps of
Engineers, March 21.
Ayres, J.E., Lager, D.C. and Barvenik, M.J., (1983), "Design of Soil-Bentonite Backfill Mix for the
First Environmental Protection Agency Superfund Cutoff Wall", Proceedings of the Fourth National
Symposium on Aquifer Restoration and Groundwater Monitoring.
Daniel, D.E., Boynton, S.J. and Foreman, D.E., (1984), "Permeability Testing with Flexible-Wall
Permeameters", Geotechnical Testing Journal, GTJODJ, Vol. 7, No. 3, September, pp. 113-121.
D'Appolonia, D.J., (1980), "Soil-Bentonite Slurry Trench Cutoffs", Journal of the Geotechnical
Engineering Division, American Society of Civil Engineers, Vol. 106, No. GT4, April, pp. 399-417.
JRB Associates, (1984), Slurry Trench Construction for Pollution Migration Control, U.S.
Environmental Protection Agency, EPA-540/2-84-001, February, 237 p.
Lambe, T.W. and Whitman, R.V., (1969), Soil Mechanics, John Wiley & Sons, New York, New York,
522 p.
McCandless, R.M. and A. Bodocsi, (1987), Investigation of Slurry Cut-Off Wall Design and
Construction Methods for Containing Hazardous Waste, U.S. Environmental Protection Agency,
EPA-600/2-87/063.
1217
-------
Millet, R.A. and Perez, J., (1981), "Current USA Practice: Slurry Wall Specifications", Journal of the
Geotechnical Engineering Division, American Society of Civil Engineers, Vol. 107, No. GTS, August,
pp.1041-1056.
Woodward-Clyde Consultants, (1990), Draft Final Decision Document for the Interim Response
Action at the Lime Settling Basins, Rocky Mountain Arsenal, Contract No.
DAAA15-88-D-0022/0002, Version 3.1, February, 50 p.
Xanthakos, P.P., (1979), Slurry Walls, McGraw-Hill Book Company, New York, New York, 621 p.
Zappi, M.E., Shafer, R.A. and Adrian, D.D., (1989), "A Laboratory Evaluation of the Compatibility
of Ninth Avenue Superfund Site Ground Water with Two Soil-Bentonite Slurry Wall Backfill
Mixtures", U.S. Army Corps of Engineers Waterways Experiment Station, Draft Report, July, 108 p.
Zappi, M.E., Schafer, R.A. and Adrian, D.D., (1989), "Compatibility of Soil- Bentonite Slurry Wall
Backfill Mixtures With Contaminated Groundwater" Proceedings of the 10th National Conference,
The Hazardous Materials Control Research Institute, Superfund '89, Washington, D.C., November
27-29, pp. 519-525.
12.18
-------
United Creosoting Company Superfund Site:
A Case Study
Deborah D. Griswold
U.S. Environmental Protection Agency
Region 6
1445 Ross Ave.
Mailcode 6H-SC
Dallas, Texas 75202-2733
INTRODUCTION
In 1983 the United Creosoting Company (UCC) site was proposed for inclusion on the National
Priorities List (NPL). With this action the Environmental Protection Agency (EPA) made it possible
for Superfund money to be spent on the remediation of this complex wood preserving site. This paper
will discuss the many challenges the site has posed as it has progressed through the Superfund
"pipeline".
Complicating work on this project is the fact that this site entirely encompasses a residential
subdivision and two industrial businesses. From the beginning residents have requested a complete
buyout of the subdivision. The fact that EPA is unable to comply with their request makes
community relations a challenge. Because of this, communication with the community has been a
high priority. Intensive community relations efforts, both in the past and those planned for the
future, will be discussed.
An innovative technology, available from only one vendor, was selected in 1989 as the method of
remediation for the site. Because of this procurement will be different from what can be considered
the norm. The combination sole source and competitive bid contracting strategy, proposed by the
Texas Water Commission (TWC) to procure services to remediate the site, will also be examined.
BACKGROUND
SITE LOCATION AND HISTORY
The UCC site is located 40 miles north of Houston in the city of Conroe, Texas. Approximately
13,000 people currently live within a two-mile radius of the site. The site is occupied by two
industrial properties and a residential subdivision (Figure 1).
UCC operated as a wood preserving facility from 1946 through the summer of 1972, when it was
abandoned. Formed lumber, such as telephone poles and railroad ties, were treated in a two-step
process by the pressurized addition of pentachlorophenol (PCP) and creosote. The pressure cylinders
were rinsed and wastewater routed to one of the two process waste ponds located onsite.
In the late 1970's the property was divided and sold to several entities. At some time the pit used for
tank bottoms and other residues was covered with soil. Shortly thereafter a portion of the site was
developed into a residential community and the rest became a light industrial area.
During the summer of 1980 surface soils and pond backfill from one of the industrial properties were
donated to the County by the property owner for use on improvements to several Conroe roads. The
soil had been moved and stockpiled by the owner to allow for the installation of paving on his
1219
-------
D
»., t
d,
•H -H
b W
1220
-------
property. After citizens living along one of the roads complained of headaches, burns, respiratory
problems, and damage to vegetation, the contaminated soils were excavated from the roads and
disposed of, and investigations were initiated. The site was proposed for the NPL in September 1983.
Field work for the Remedial Investigation was conducted in December 1984 and August 1985. The
Feasibility Study was completed in May 1986, and a Record of Decision (ROD) signed September
1986. During the Remedial Investigation and Feasibility Study phase it was concluded that seven
properties, six with houses on them, were directly in the way of any future excavation of the old
waste pond. The only method known at the time which could address the contamination at the site
was incineration. As there were no offsite incinerators which would accept the dioxin contaminated
waste, the incineration would have to be conducted onsite. This information was presented to the
community along with the idea of a temporary solution, with final solutions to be evaluated as
technologies developed. The community strongly opposed the incineration proposal. EPA selected
the temporary remedy, which included the following:
o Purchase seven properties located above and adjacent to a former pond area;
o Permanently relocate the persons living in the six houses located on those properties;
o Demolish the six houses;
o Consolidate soils contaminated above health-based levels and visibly contaminated
soils in the pond area;
o Construct a temporary cap over the pond area;
o Evaluate innovative technologies as possible permanent remedies, and;
o Natural attenuation of the ground water contamination.
EPA promised in the ROD to re-evaluate this remedy in five years if no innovative technologies
became available. The health based action levels selected in this ROD are listed in Table 1.
TABLE 1
SOIL ACTION LEVELS FROM THE 1986 ROD
COMPOUND CONCENTRATION
Total Polyaromatic Hydrocarbons (PAHs) 100 mg/kg
Pentachlorophenol 150 mg/kg
Tetra-Dioxin 1 ug/kg
Penta-Dioxin 5 ug/kg
Hexa-Dioxin 25 ug/kg
Hepta-Dioxin 1000 ug/kg
Through the support of EPA initiatives, potential technologies for treatment of creosote by-products
did become available shortly after the ROD was signed. This led to a biological treatment bench scale
study in 1988 and a critical fluid extraction study in 1989. Biological treatment was satisfactory at
degrading the PAHs, but was not sufficiently successful at destroying the dioxins to the necessary
extent. The critical fluid extraction process, on the other hand, showed satisfactory results for both
1221
-------
PAHs and dioxins. Based on an evaluation which considered the results from these studies, and other
proposed remedies, critical fluid extraction was selected as the remediation method for the site in
September 1989.
The second ROD stipulates the following:
o Sample the residential area to better delineate all soils falling above the target action
levels.
o Excavate soils from residential and commercial portions of the site that are above the
action level and treat via critical fluid extraction.
o Dispose of the organic concentrate from the extraction process by off-site
incineration.
o As the action levels and treatment standards for K001 contaminated soils are met,
rebury treated soils on the appropriate portion of the site.
The selection of this remedy initiated a second operable unit and precluded the necessity of
consolidating and temporarily capping the soils from the pond areas. New action levels were set in
the ROD by using the most current risk assessment guidance. These new levels are listed in Table 2.
The new guidance allows for use of benzo(a)pyrene (BAP) equivalents and 2,3,7,8-
tetrachlorodibenzodioxin (TCDD) equivalents for estimating the carcinogenicity of other PAHs and
isomers of dioxin and furans, respectively. The changing of the action levels caused considerable
confusion among the residents and special efforts were made to communicate the meaning of the
changes and the reasons for them to the community.
The purchase of the 7 properties as specified in the first ROD, by the Federal Emergency
Management Agency (FEMA), was completed by the transfer of titles to EPA in August 1990.
Several factors contributed to the long duration in getting the houses purchased. Initially the State
and EPA could not agree on who would hold the titles to the properties once they were purchased.
Eventually it was decided that the Federal government would take the titles until the remedial action
was completed, after which time they would be transferred to the state of Texas. Another problem
arose when several of the houses were appraised at a lower value than the mortgage on them. This
necessitated special procedures by FEMA to allow the purchase of these houses at more than their
appraised value. Finally, an Internal Revenue Service (IRS) tax lean on one house led to delays until
negotiations between FEMA and the IRS settled the matter.
Once purchase of the properties was accomplished the house demolition could commence and a Notice
to Proceed for this work was issued by TWC in October 1990. This interim remedial action work
included demolition of six houses, and erection of a fence around the now vacant lots. The
demolition activities were originally designed and bid to be entirely non-hazardous work. Demolition
of the houses was completed in December 1990.
In January 1990 the additional sampling stipulated in the 1989 ROD was performed as a focused site
investigation. The impact of excavation on local air quality was also evaluated during this effort, and
from this study, is expected to be insignificant.
TWC initiated the design phase for the final remedy in January 1991. A Design Concept
Memorandum is in the process of being finalized at this time. This memorandum will outline basic
design decisions and options in an effort to minimize redesign time due to changes in direction in
future design deliverables. The design of the final remedy is scheduled to be completed January 1992.
1222
-------
TABLE 2
SOIL ACTION LEVELS FROM THE 1989 ROD
COMPOUND
RESIDENTIAL
SOIL ACTION
LEVEL (PPM)
INDUSTRIAL
SOIL ACTION
LEVEL (PPM)
APPLICABILITY
carcinogenic PAHs
(in BAP equivalents)
non-carcinogenic PAHs
carcinogenic dioxins
.33
2000
.001
40
2000
.02
surface soils
surface and
subsurface soils
surface soils
and furans (in
2,3,7,8-TCDD equivalents)
PCP
150
150
surface and
subsurface soils
COMMUNITY RELATIONS BACKGROUND
This site represents an extraordinary challenge because of the active residential community located
within the site and on top of some of the waste. Community relations has been a major consideration
from very early in the project. The community has been vocal in asserting its concerns and has been
able to generate significant media and Congressional involvement. The resident's primary aspiration
is to have the entire subdivision of nearly 100 residences and 28 vacant lots bought by the
government. However, under Superfund, there are only two circumstances when EPA may purchase
property: (1) when the purchase of the property is necessary to physically implement the remedial
action, or (2) when the final remediation for the site cannot otherwise eliminate long term health
dangers.
Complicating the buyout issue recently has been the fact that EPA has recently been ordered by
Congress to "buyout" a similar site in Texas. Neither site meets the two circumstances listed above
which would warrant a buyout. Nevertheless, EPA must purchase the other subdivision but cannot,
according to policy, purchase the one in Conroe. A site related lawsuit between some of the residents
and TWC serves to further complicate community relations.
In the past, area citizens have been kept informed of activities at the site through the extensive use
of community relations meetings. From 1983, when the site was proposed for the NPL, to the signing
of the second ROD in 1989, nine meetings were held with the residents. These have included
informal meetings with the homeowners association, work shops, open houses, and when necessary,
formal public meetings. In addition, press releases and direct mailings to the community have been
employed to update concerned citizens about site activities. A chronology of past and future major
milestones and community relations activities is shown in Figure 2.
It was decided at the beginning of the focused site investigation that even these extensive measures
could be improved on in order to increase public understanding and cooperation. Prior to starting
fieldwork for the focused site investigation, an open house was held to inform the residents of the
1223
-------
§
H
CO
CM
ft
E 8
8
o
0
H
E-t
to
cn
n
en
H
01
oo
CO
oo
oo
n
oo
CO
1
i >
OH
§
Q
H
CO
S
§
Q
H
2
O
H
CO
Q
O
CO
H
CO
O
H
CO
Q
§
OJ
Q
W
O
OH
CO
W
CO
D
§
H
W
CO
1
* 1
* I
* 1
* I
8
6
S
CO
o
H
§
EH
H
55
O
O
CO
C5
CO
O
TJ
<0
P
0)
O
£
•o
0)
W
o
I
B
1224
-------
upcoming sampling event and to give them an opportunity to have input on the locations of the
samples. Samples were to be taken primarily in residential yards, necessitating the obtaining of
approximately 50 access agreements over a short period of time. The Remedial Project Manager
obtained access agreements by going door-to-door in the community. Contact was attempted at each
residence in the community, regardless of whether access was needed, in order to give residents a
personal update on what was occurring in the neighborhood. In cases where the resident was not the
property owner access was obtained from both, and both were given a personal update on the project.
Beginning concurrently with the January 1990 fieldwork EPA began mailing monthly site updates to
the community. These monthly mailings generally included the following type of information
regarding the site.
o Status updates on the various phases of work on the site.
o Common questions and EPA answers.
o Requests for input from the community on specific topics.
o Explanations of how to interpret data presented to the community.
o Schedule of upcoming activities.
o Contacts for additional information and repositories locations.
Another effort to become more accessible to the public was the implementation of a 1 -800 phone line
with an answering machine for 24-hour service. In each mailing the community was reminded of this
toll free number. This number has since been made available for use on all Region 6 Superfund sites.
After results from the sampling became available EPA prepared a set of data packages geared
specifically to the residents. These packages presented the results of the sampling on each resident's,
and their immediate neighbor's, yards in an easy to follow format, both in a table and graphically.
Accompanying these packages was a letter telling the resident whether the data indicated their yard
was eligible for replacement during implementation of the permanent remedy. Shortly after these
packages were mailed a work shop was held to inform residents on the impact of the information on
the community and to allow residents the opportunity to discuss their data package results with
representatives from EPA and TWC.
Once the data had been thoroughly interpreted and the air modelling completed an open house was
held to present the final report for the focused site investigation. At this meeting large maps showing
proposed excavation were available for review. Before this meeting was held the final reports had
been sent to the repositories for access by the community.
Prior to community meetings to discuss the focused site investigation all TWC and EPA attendees
were thoroughly briefed on the status of the project and recent developments. This preparation went
as far as the development of a list of questions expected from the people attending, and responses to
them. These questions were generated from various sources, such as issues raised in the media, and
calls and letters from the residents. Responses to the questions were developed cooperatively by EPA
and TWC. Some examples of the most major concerns, and the EPA responses, are on Table 3.
Following the open house to present the final report for the focused site investigation the questions
and responses were refined and sent to the community in the monthly site update. These questions
and responses were culled from monthly mailings, the questions and answers developed in preparation
for community meetings, and questions asked during the community meetings themselves.
1225
-------
S- I-
~?
o .
0 CU
ll
EPA RESPONSES
"1 o
^3
sz
>- i*
UJ
u
1
z
UJ
ISI
1—
u
cu
CO
c
I/I
cu
c
CU
£
o
u
Response: Adding the cost of residential buyout to
remediation does not make the overall cost less expensive
A)
.c
V
*J
co
II" 1
^ T!
1-5
£E
a c
£l
ui-2
II
O) ...
22
O
*j 3
IU V4.
tl
|l
-1
"" £
ii
.*'
TJ Q-
cy ^
L.
W E .-
cu P °
EPA recognizes that property values may have been depr
Superfund law does not provide for EPA to pay for ecoi
homeowners. Economic damages are normally the concern of
o
.c
CO
L. ••;
The remedial actions planned for this site should, howevc
stigma associated with being a Superfund site from the coi
CU O — • J=
CO •*-» CO +-*
CO C -•
J= ._ — CO
U ?***- V
<5 <5 •=
S-Sji E
xSf fc
iSc"
(U O>
Under Superfund, there are two circumstances when EPA
property: (1) when the purchase of the property is
physically implement the remedial action, or (2) wh
remediation for the site cannot otherwise eliminate Ion
4)
.c
QJ
5
a
f.
\
i
\
V
'at
$£
ll
5
» "
fc£
|S
|8
S*
LU X
l|
li
"8
•M
O
$
Cj
c
CO
4-»
u
u
i
cu
4-*
«*-
0
L.
0)
4^
cu
CO
cu
cu
o
c
in
1
CL.
UJ
CO
£_
o
en
c
CO
CO CU CO
3 -C -
^ Q-
- CO
co £ TJ
above exist at either United Creosoting or the other simil
EPA did not propose a buyout at either subdivision. The
other site results from a line item, which was not ac
request, included in the Agency's appropriations bill.
t- C • V CU 0)
O CO >s 4-» — ' CU
4-> o .n "8
^j 4j ••- a> co 3
§? 3 OJ O-"o
5 E i- cu c
O CO j_i ,
Health professionals at the EPA, ATSDR, and the Texas
Health have reviewed data collected from the site and .
immediate health risk from contaminants is not present in '
Because areas of neighborhood properties have levels
contamination above concentrations that would be potential
for a lifetime of exposure, the remedy EPA selected for the
replacement of these soils.
.
'vi
1
^
4J
^
g
„ »
X
1
1
1
^
4^
1
£
1226
-------
< a> a)
O- JC -
UJ 3 *-• I—
1227
-------
INNOVATIVE TECHNOLOGY BACKGROUND
The technology selected for this site is available only from C.F. Systems (CFS), the patent holder.
CFS is not planning on making the technology available to other manufacturers. Because of this TWC
will be unable to procure the remediation of the site through a single competitive bid, as is the norm.
Instead, at least a part of the design and remediation will have to be secured using noncompetitive
procurement methods. Federal cooperative agreement procurement regulations allow noncompetitive
procurement under four circumstances:
1) the technology is available from only one vendor,
2) an emergency exists which will not permit a delay resulting from procurement,
3) the award official authorizes it, or
4) after solicitation competition is determined to be inadequate.Condition one is met for
the treatment technology. A cost analysis must always be performed when
noncompetitive procurement methods are used.
DISCUSSION
COMMUNITY RELATIONS TECHNIQUES
Although public meetings are necessary during the ROD public comment period, informal meetings,
such as work shops and open houses can be a more effective tool for informing the community. Work
shops are run very similarly to public meetings, except they are much more informal. Although a
formal transcript is not generated, an informal summary is usually developed either from personal
notes or a tape recording. Generally, the work shop focuses on a specific topic, such as a recent
report made available in the repositories. Presentations are kept short in duration, minimizing
background information which has been presented in the past, and focusing instead on specific issues
or future activities. A significant portion of the work shop is dedicated to responding to questions,
which are taken informally from the attendees without the use of a microphone (which many people
find threatening). Open discussion is encouraged, leading to a more conversational approach to
addressing questions.
Open houses are just what the name implies. Posters and informative handouts are made available
for attendees to peruse at their own pace. No formal presentations are made, however, representatives
from all involved government agencies are available to respond to questions.
PROPOSED COMMUNITY RELATIONS EFFORTS
From past experience Region 6 has discovered that the more opportunity the community is given for
input the less likely they are to try and block efforts by EPA to move forward with the remedy. This
is because their concerns are addressed early, before it becomes difficult to change direction due to
their input. TWC has proposed several ways to continue with the expanded community relations
efforts.
At this time TWC intends to make the design deliverables available to the community. This means
the design concept memorandum, 30%, 60%, and 95% complete designs will be sent to the repositories
rather than the final design only, as is typically done. Meetings will be held with the community to
discuss these documents following their delivery to the repositories. The community will be allowed
to voice any concerns they may have on the content or direction of the design. The intent of doing
1228
-------
this is to reduce the likelihood of major community disagreements with the final design. If the
community is not given the opportunity to have input early in the design their comments could lead
to revisions in the final design. Or, if their concerns are not addressed, it could lead the community
to look for to sources, such as their congressmen, in order to have their desired changes made. Many
people who may not have had major concerns with the design may decide to disagree with it simply
because they were not allowed input by not being contacted until the design was final. TWC intends
to continue the use of work shops and open houses during the remedial design and remedial action
to keep the public informed.
TWC will continue the frequent, regular mailings initiated by EPA during the focused site
investigation. This effort will continue through the remedial design and remedial action process. By
mailing updates regularly to the community they are informed that progress is being made. It also
serves to remind them of their responsibility to stay informed on the direction of the project.
A visitor's center is proposed for the site during the remedial action. This center would provide the
community a place to pick up the most recent information available on the site, leave comments, and
possibly even see a video of the remediation process. This center is expected to reduce the feeling
of being denied access to the project which may result from the increased security for certain areas
during remediation. It will also encourage community involvement and self-education.
PROPOSED CONTRACTING METHOD
The treatment technology for this site is available from only one source, leading to the need to use
a different approach for contracting and procurement at this site than is normally used. Typically
Superfund remedial actions are procured using a detailed set of plans and specifications developed
by the design engineer. These plans and specifications are used to invite bids for the project and
award of the contract is to the lowest bidder, leading to a competitive procurement. For this project,
however, TWC's design engineer will prepare a set of plans and specifications for all of the work
except the treatment of contaminated soils. These plans and specifications will be used to
competitively procure a contractor (henceforth called the major contractor) to conduct the site
preparation, excavation, materials handling, site restoration, etc. By splitting out the treatment
portion of the contract TWC intends to maximize the amount of the contract being competitively
procured.
To secure the treatment of the contaminated soils, TWC proposes to contract separately with CFS.
The first contract with CFS will be to design the system needed for the site. This work will be
performed concurrently with the design of the competitively procured work. During the design phase
CFS will provide the specific parameters needed for the soils to be processed through the treatment
system. These parameters will be placed into the competitive contract specifications as the conditions
of the soil necessary for CFS to accept them for treatment. The major contractor will be required to
verify that these conditions are being met. This dual contracting for remedial design will mean
coordinating between the design engineer and CFS to produce a biddable design.
During remedial action CFS will be contracted with to provide and operate the treatment system to
within an agreed to set of criteria. The major contractor will excavate, stage, and pretreat soil as
specified in the contract prior to turning it over to CFS. CFS will then treat the soil to the treatment
standards and turn it back over to the major contractor. The major contractor will then place onsite
the treated soil onsite as specified and restore the excavated areas. The current design engineer will
be contracted with to provide oversight of both the major contractor and CFS.
An alternative to having CFS contract directly with the TWC would have been to make them a
mandatory subcontractor to the prime contractor. This would have been accomplished by negotiating
1229
-------
a fixed price with CFS to be inserted in all of the submitted bids. One reason the contracts are not
being set up this way is the potential legal issues over forcing a contractor to take a mandatory
subcontractor. Another reason is that this would delay TWC from having any contractual commitment
with CFS until after the remedial action is procured. This issue is crucial as some of CFS's equipment
could take a year to receive from the time it is ordered from the manufacturer. By contracting with
CFS separately, to produce the design and perform the treatment, CFS should be able to begin
procuring the equipment necessary for the system significantly ahead of the time the competitively
bid contract is signed. Setting the contracts up this way will increase the amount of oversight
necessary during remedial action, but will also give TWC more direct control over CFS.
CONCLUSIONS
EPA and TWC have found at this site that more extensive community relations activities have
improved the progress of the project. It is possible to work with the community and gain their trust
and consent, even if they continue to maintain different goals from the Agency. As the project
managers and the residents get to know one another on a personal basis all parties becomes more
comfortable in communicating with each other. The project managers are less defensive when
discussing work at the site and the residents learn to trust the managers on a one-to-one basis.
Frequent personal interaction between the project managers and the community, such as door-to-door
contact, accelerates the gaining of this trust.
Work shops and open houses have been found to be effective tools when communicating with the
community. Both of these type of community meetings offer advantages over public meetings. There
is lower pressure on the government representatives and less posturing by all participants. Thorough
preparation for the meetings increases the meetings' effectiveness and the consistency of responses
from different representatives at the meeting. Frequent mailings which contain requests for input
keep the community informed and at the same time make them a part of the process.
Community relations can be especially effective if the government representatives involved have good
people skills, which is the case for the United Creosoting site. The progress with the community at
this site is going to be continued through extensive interaction, such as the use of regular mailings,
frequent work shops on the intermediate design documents, and proactive efforts during the remedial
action.
It is recommended that during the community relations process in remedial design the effected
community is encouraged to take responsibility for their involvement. Let the community know early
on that after the remedial action starts it is too late to make changes which could have been handled
during the remedial design. On the other hand, remember that having the flexibility to change plans
can improve relations and build trust in the community by showing them their best interests are being
considered, if possible.
Procuring innovative technologies can be uncomplicated for State-lead projects, due to the
straightforward requirements of the procurement regulations for cooperative agreements. By splitting
the contracts in the way described the amount of the sole source contract is minimized. This should
reduce the costs of the project by maximizing the amount of work to be competitively procured.
Also, as discussed, by keeping the contracts separate TWC will be allowed more direct control of the
soils treatment contract and CFS can initiate purchase of equipment earlier. A potential disadvantage
is that by procuring the work in this way the contracts will have to be written such that the necessary
interaction between the major contractor and CFS is clearly defined. Otherwise, conflicts could arise
between the two contractors.
1230
-------
DISCLAIMER
This paper was prepared for presentation at the May 1991, Conference on Design and Construction
Issues at Hazardous Waste Sites sponsored by the United States Environmental Protection Agency's
(U.S.EPA) Office of Emergency and Remedial Response. This paper reflects the opinions of the
authors only. This paper does not contain either regional or national policy and should not be
construed as such.
REFERENCES
Roy F. Weston. 1985. Final Site Investigation Report, United Creosoting Company Site, Conroe,
Texas, Volume I. Prepared for the Texas Water Commission in cooperation with the U.S.
Environmental Protection Agency. December.
Roy F. Weston. 1986. Final Feasibility Study Report, United Creosoting Company Site, Conroe,
Texas, Volume I. Prepared for the Texas Water Commission in cooperation with the U.S.
Environmental Protection Agency. May.
U.S. Environmental Protection Agency, Region 6. 1986. Record of Decision, Remedial Alternative
Selection, United Creosoting Company. Signed September 30.
Roy F. Weston. 1989. United Creosoting Super fund Site, Feasibility Study Amendment, Preferred
Alternatives Analysis. Prepared for the Texas Water Commission in cooperation with the U.S.
Environmental Protection Agency. September.
U.S. Environmental Protection Agency, Region 6. 1989. Record of Decision for United Creosoting
Site, Conroe, Montgomery County, Texas. Signed September 29.
Roy F. Weston. 1990. Data Evaluation Report, Focused Site Investigation, United Creosoting, Conroe,
Texas. Prepared for the U.S. Environmental Protection Agency. July.
Roy F. Weston. 1991. United Creosoting super fund Site, Conroe, Texas, Interim Remedial Action
Residential Housing Demolition, Final Report. Prepared for the Texas Water Commission in
cooperation with the U.S. Environmental Protection Agency. March.
40 Code of Federal Regulations. 1990. Part 35 - State and Local Assistance, Subpart O - Cooperative
Agreements and Super fund State Contracts for Super fund Response Actions, Procurement Requirements
Under a Cooperative Agreement.
1231
-------
Considerations For Procurement of Innovative Technologies
at Superfund Sites
Edward J. Hanlon
Design and Construction Management Branch
U.S. Environmental Protection Agency
Mailcode OS-220W
401 M Street S.W.
Washington, D.C. 20460
703-308-8352
L INTRODUCTION
A discussion of issues related to potential institutional barriers associated with the procurement of
innovative or patented technologies at Superfund sites would be useful to government and private
sector employees. This paper explores applicable requirements of Superfund-specific regulations,
Federal procurement regulations, including the Federal Acquisition Regulations (FAR) and the U.S.
Environmental Protection Agency (EPA) Acquisition Regulation (EPA AR,) and certain State-specific
procurement regulations. Requirements for competition and sole-source procurement are
summarized.
Pre- and post-Record of Decision (ROD) solutions to innovative technology procurement barriers,
including use of non-inhibiting Record of Decision wording and consideration of early design
'prequalification' of potential vendors, are included. Pros and cons of contract method and type,
including whether sealed bid or negotiated procurement is preferred, are discussed. A brief
discussion of PRP-lead issues is provided. Where possible, site-specific examples are provided.
2. BACKGROUND
The Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (CERCLA,)
as amended by the Superfund Amendments and Reauthorization Act of 1986 (SARA) requires that
EPA give strong preference towards achieving protective remedies through the use of treatment
technologies that significantly reduce the toxicity, mobility, and/or volume of hazardous waste.1
SARA specifically supported the selection of innovative technologies by allowing the selection of an
alternative remedial action in a Superfund ROD regardless of whether or not such an action has shown
to be successful at any other facility or site.2 SARA also directed EPA to use up to $10 million per
year through 1991 to establish an "Alternative or Innovative Treatment Technology Research and
Demonstration Program."3
The recently updated National Contingency Plan (NCP) identifies EPA's expectation that innovative
technology remedies be considered when they offer "the potential for comparable or superior
treatment performance or implementability, fewer or lesser adverse impacts than other available
approaches, or lower costs for similar levels of performance than demonstrated technologies '4. The
NCP encourages the development of technologies that have not yet been proven in practice in order
to promote the development of new treatment methods for hazardous substances.5 The EPA
Administrator, as well as Congress's Office of Technology Assessment, also stressed that EPA improve
the promotion and use of innovative technologies in the Superfund Program, and reduce the
institutional barriers which make implementation of these technologies difficult6'7.
1232
-------
EPA's Office of Research and Development (ORD) conducts the Superfund Innovative Technology
Evaluation (SITE) program was organized to maximize the use of alternatives to land disposal in
Superfund through field-scale demonstration and evaluation of innovative technologies which offer
some advantage over existing technologies. ORD defined alternative technologies as those alternatives
to current procedures and practices categorized as follows: a) "available alternative technology" - fully
proven and in routine commercial or private use; b) "innovative alternative technology" - fully
developed technology for which performance or cost information is incomplete, thus hindering
routine use at Superfund sites; and c) "emerging alternative technology" - an alternative technology
at a stage where lab testing has been completed and pilot-scale work is now necessary.8
The SITE program fosters commercialization of innovative technologies through two sub-categories
of testing: a) "Demonstration" — at full or pilot scale; or b) "Emerging" — at lab scale. In both cases,
technology developers provide and operate the technology and EPA conducts sampling and analyses9.
Currently 31 technologies are participating in SITE'S Emerging Technologies program, and range from
electoacoustical decontamination to bench and pilot studies of a laser-stimulated photochemical
oxidation process.8 ORD, in conjunction with EPA's Office of Solid Waste and Emergency
Response's Technology Innovation Office (TIO), also conducts conferences to help introduce
promising international technologies through technical paper and poster displays, and showcase SITE
and other domestic innovative technologies.10 Three conferences of this sort have been conducted
to date. TIO also has historical information regarding where and when innovative treatment
technologies have been conducted. Requests for information from the SITE program may be made
by calling (703) 308-8800.
Through 1989, EPA has selected innovative technologies in 37% of Superfund source control Records
of Decision (RODs) which selected treatment technologies. Of these, vacuum extraction (12%,)
bioremediation (8%,) thermal desorption (5%,) in-situ soil flushing (4%,) soil washing (3%,) chemical
extraction (2%,) chemical treatment (2%,) and in-situ vitrification (1%) were selected. Incineration
and solidification/stabilization, which are considered non-innovative, account for another 35% and
25% of the RODs selecting treatment technologies, respectively. Innovative technologies have been
selected more frequently in recent years (52% of the FY-89 RODs involving source control treatment
were innovative technologies)11.
3. DISCUSSION
Due to the general unknowns associated with Superfund sites (e.g., difficulties associated with
properly characterizing the nature and extent of contamination and health risks,) and the general need
to move quickly with implementing remedial actions to protect human health and the environment,
Superfund construction and operation and maintenance (O&M) projects may generally be considered
more likely to experience problems and changes than non-Superfund construction projects. Design
and construction of innovative technology Superfund remedies (RD/RAs) may be more likely to fail
(in terms of non-success in meeting performance/remediation goals and remedial objectives) as non-
innovative Superfund RD/RAs. This is because these technologies, by definition, have not been fully
demonstrated on a number of sites and thus have incomplete performance or cost information, and
few, if any, vendors have sufficiently proven their expertise through implementation.
EPA and the States often work together to manage the remediation of wastes at Superfund sites.
Procedures for the management of projects of both the State and Federal project managers are well
described in EPA's "Superfund Federal-Lead Remedial Management Handbook" and "Superfund
State-Lead Remedial Management Handbook;"12'13 these handbooks should be used as a guide by
project managers when developing strategies to address the issues outlined below.
1233
-------
3.1. Competition
3.1.1. General
The term "noncompetitive" is often used to mean other than full and open competition. This means
not only sole source acquisitions, but also those situations where an agency is permitted to limit the
number of sources solicited. Executive Order 12352, signed by the President on March 17, 1982,
requires agencies of the Federal government to "establish criteria for enhancing effective competition
and limiting non-competitive actions"14. The EPA Administrator has also emphasized the need to
broaden competition where possible in contracting the Agency has involvement in.15 This direction
from the EPA Administrator resulted in an EPA Order entitled Contracting at EPA, in which it is
made clear that procurement strategies which broaden contractor resources available for a particular
function should be favored.16
The Competition in Contracting Act of 198417 (CICA) further specifies requirements for enhancing
competition. CICA also provides for the use of "other than full and open competition" for some
acquisitions. Although not statutorily defined, CICA lists seven different procurement options that
would allow for "other than full and open competition," as follows: 1) only one source available; 2)
unusual or compelling urgency; 3) necessary to maintain a particular service for national security; 4)
international agreement; 5) a statute authorizes a brand name or specific source; 6) national security
would be breached if not done so; or 7) head of agency determines the need, and notifies Congress
30 days prior to the procurement.19
FAR Parts 6.303 and 6.304 require "Justification and Approval" by an appropriate agency employee
(normally a contracting officer) to use one of these options. This justification, as required by statute,
must include: a) a description of the agency's needs; b) identification and discussion of the need for
the option used; c) a determination that the anticipated cost will be fair and reasonable; d) a
description of a market survey conducted or the reasons why one was not conducted; e) a listing of
the sources, if any, which expressed in writing an interest in the procurement; and f) a statement of
the actions, if any, the agency may take to remove or overcome any barrier to competition before a
subsequent procurement for such needs.14 For innovative technology justifications, a brief
description of the technology, how the equipment would be used, why there is a need for sole source
procurement, and a reference back to the ROD, might all be warranted in addition to the above.
3.1.2. FAR Requirements
FAR Part 36.209 notes that "no contract for construction of a project shall be awarded to the firm
which designed it" or provided a 'significant contribution' to the design without approval of the
appropriate agency officials. FAR Part 9.5 discusses general prohibitions against allowing contractors
to perform work for which it received an unfair advantage during procurement.
The "Buy American" Act (41 USC 10) was issued in response to concerns that a significant amount
of Federal funding was being used to purchase foreign materials, and hence help other countries
become competitive in the United States. As a result of this Act, FAR Part 52.225-5 requires that
contractors will use only "domestic construction materials" when constructing a project under Federal
procurement. A "domestic construction material" must pass a two-part test: a) manufactured in the
U.S.; and b) cost of domestic components must exceed the cost of all components. As defined,
construction material is made up of components (e.g., a transformer is a construction material; the
piping, container, electrical circuits, etc. are components;) a component does not include labor or
manufacturing costs. If the use of domestic construction material would unreasonably increase the
price, or would be impracticable, 'Buy American' restrictions would not apply.19 If a promising
international innovative technology were the technology of choice at a site, the project personnel
1234
-------
should solicit government contracting officer assistance to determine whether the FAR's 'Buy
American' requirements apply, and how to address these requirements.
3.1.3. Environmental Protection Agency Acquisition Regulation (EPAAR) Requirements
EPA's Procurement and Contracts Management Division (PCMD) has made several efforts to help
eliminate constraints to the procurement of treatment technologies. 48 CFR Part 1536 of the EPAAR
was recently added by PCMD to clarify the applicability of FAR Part 36.209. Under this rule,
subcontractors performing treatability studies are not prohibited from being awarded the construction
contract for a project. Other subcontractors are also not prohibited from being awarded the
construction contract for a project unless their work substantially affected the course of the design.
Prime contractors of the design and subcontractors whose work substantially affects the course of the
design must receive prior approval by the responsible Associate Director of PCMD under EPA's
Office of Administration and Resources Management before they can be awarded the contract.
3.1.4. State-Lead Requirements
The June 1990 40 CFR Part 35, Subpart O regulations (EPA Grants Regulations)20 establish
administrative requirements for CERCLA-funded Cooperative Agreements and Superfund State
Contracts for States, political subdivisions thereof, and Federally recognized Indian Tribes. It
discusses EPA's allowable procurement procedures for state-lead remedial actions. Part 35.6555 notes
that the state "must conduct all procurement actions in a manner providing maximum full and open
competition," and, under (a)(6,) that specifying only a brand name product without allowing "an
equal" product to be offered is considered an inappropriate restriction on competition. However,
(c)(l)(iii) notes that specifications may be written where competition may be justifiably restricted if
the material, product or service is necessary to promote the use of innovative technologies in a
procurement. If noncompetitive procurement is conducted using such justification, and assuming a
"small purchase exemption" can not be conducted for an innovative technology item [under $25,000,
see Part 35.6565(a)], Part 35.6565(d) requirements apply, and a cost/price/profit analysis in
accordance with Part 35.6585 is required.
40 CFR Part 31.6 (EPA Grants Regulations) note that the Director of EPA's Grants Administration
Division is authorized to approve exceptions from non-statutory provisions of the Subpart O
regulations on a case by case basis. Such "deviations," as allowed in Part 35.6025, might include a
cost/price/profit analysis.
3.2. Sole Source Procurement
3.2.1. General
Sole source procurement is the broadest and potentially the most utilized exception used to justify
"other than full and open competition" under CICA. As noted previously, any agency using this
justification must reasonably show that only one source, and no others, will satisfy the agency's need.
Adequate efforts must be made to ensure that sole source is required. It is improper for an agency
to rely on the sole source contractor for technical advice and expertise; agencies should independently
evaluate technical criteria and make their own decisions. Sole source determinations by agencies have
been overturned when the facts have indicated that other sources could have satisfactorily met the
Government's stated needs14.
Government contracting officers (CO's) frequently require protracted negotiations with the technical
staff on a project to make clear that conducting a sole source procurement is warranted. In one case
negotiations took over a year to reach agreement to use sole source procurement. These up-front
1235
-------
delays do not appear to be offset by time savings in the proposal evaluation process.21 The reluctance
of the CD's to sole source is primarily based on and justifiable due to the overarching statutory and
regulatory emphasis for competition.
Activities such as market surveys and cost and profit analyses should be conducted in a proper and
complete manner by agencies before making the decision to use sole source procurement.
Comparability is a common element to be considered, among others; the reader should solicit the
available reference documents available on how to do these surveys and analyses during project
planning.22 Literature reviews, value engineering, and pre-design brainstorming sessions should also
be conducted. If this effort shows that other contractors may reasonably be able to meet the
government's needs, possibly through minor adjustments to the plans/specifications and/or remedial
action objectives and remediation goals, the government should not use sole source procurement.
With regard to commercial availability of an item (which could be considered a "standard product")
required under the terms of the contract, there is ordinarily an implied government warranty that
such items will be commercially available. Thus, if a sole source supplier is out of business at the
time of award, the government would, thus, likely be liable for ramifications resulting from having
required a non-procurable specification.14
The lead agency for RD could consider publishing a notice in the Commerce Business Daily (CBD)
near the end of design to advertise that the government is considering sole source procurement for
an item, and possibly publishing the justification the agency prepared to use for the sole source
procurement. If vendors other than the sole source express an interest and can provide an 'equal'
performance for the government's needs, the lead agency could then reconsider the sole source
contracting mechanism and/or strategy being taken.
Competitive procurement can also be made for the entire project, with only the technology's "black
box' (e.g., patented item) being a sole-source procurement. Any responsive and responsible contractor
would be considered competitive, as long as one of the subcontractors is the sole source vendor. If,
for example, five proposals from different prime contractors are received that all identify a certain
subcontractor for implementing the innovative technology portion of the contract, a sole source
procurement occurs. In a related manner, this occurred in EPA Region 3 for an RA at the Alladin
Plating site, where all potential bidders identified the same Treatment, Storage and Disposal (TSD)
subcontractor for offsite disposal of excavated hazardous wastes. The EPA Region 3 contracting
officer determined that even though the RD did not require a specific offsite TSD, the low bidder
was required to submit cost and pricing information in order to make a determination that the costs
were fair and reasonable.23
3.2.2. Patent issues
3.2.2.1 General
Patents for innovative technologies will periodically play a role in selecting and implementing
remedies in Superfund. For example, remedies involving soil vapor extraction, in-situ vitrification,
thermal desorption, bioremediation, chemical extraction, and chemical treatment have been patented
wholly or in part by the process, technology and/or specific component. Although a relatively new
factor in the Superfund Program, patent rights have been a long-standing concern of the Federal
Government. FAR Part 27 , "Patents, Data, and Copyrights," is written in a manner that protects the
mutual interest of the contractor and the government, and encourages the contractor to develop and
patent innovative technologies. When new technologies are conceived, a contractor may elect to retain
the title to an invention. If the new technology is conceived in performance of a government
1236
-------
contract, the government retains a royalty free, non-exclusive, irrevocable license for use of the
invention.
Ownership of, or rights in, a patent does not by itself qualify a prospective contractor for sole source
treatment. In fact, the U.S. Comptroller General adopted a position in 1958 disfavoring preferential
treatment of patentees or licensees. The contracting party (government) may, however, acquire a
patent license prior to entering into a contract; this might put an unfair advantage or disadvantage
to unlicensed sources during procurement. Notice of such a license should be placed by the
government in the solicitation to advise offerers (potential contractors for the construction who have
placed a proposal to the government's Request for Proposal (RFP,) or a bid to the government's
Invitation for Bids (IFB)) that if the offerer has not received a license, their bid will be increased by
the royalty the government is obligated to pay24.
EPA's Superfund Program prefers options for obtaining license rights to use patents. Instructing
contractors to risk patent infringement or initiating a patent challenge are the least desirable options,
since this would be contrary to EPA's and the Federal government's policy of creating an environment
favorable to the development of new and innovative technologies. However, in limited cases,
decisions to not obtaining license rights to use patents might be necessary, even though such decisions
might risk patent infringement or initiate a patent challenge. Prior to selection of a patented
technology for use, EPA should consider the necessity and reasonableness of the royalty, the cost for
use of the patent, and the options to provide for competitive procurement, if any25.
A strategy involving formation of a team comprised of government contracting officers, technical
representatives knowledgeable of the technology and legal personnel knowledgeable in patent law
would be an effective approach for properly dealing with patents in Superfund. Once the patent
holder and patent validity are determined by a patent lawyer, it would be up to the team to determine
what would cause or not cause a project to infringe on the patent, and consider whether and how to
obtain a license for its use.
3.2.2.2. Infringement, Royalties and Licenses
Infringement of a patent consists of an unlicensed making, using or selling a patented invention. If
a patent is infringed by or on behalf of the government, a patent owner's sole remedy is under 28
USC 1498 against the government in the U.S. Claims Court for "reasonable and entire" compensation.
The government does not take the property, strictly speaking, and the government's contractors
cannot be enjoined from using a patented invention. The government generally uses previous case
law to determine "reasonableness;" the royalty generally should not exceed the lowest rate at which
the licensor has offered or licensed a public or private entity. To ensure that the work of a contractor
is not enjoined by reason of patent infringement, a FAR "authorization and consent" clause should
be invoked by the government. The government may also shift the financial burden for patent
infringement to the contractor by including a FAR patent indemnity clause in the contract. Use of
this clause is limited to construction or service contracts and to contracts for supplies24.
Prior to selecting a patented product, apparatus, or process for the remedial response, on which a
royalty must be paid, the contracting party should consider: a) the necessity and reasonableness of the
royalty; b) the royalty in any cost-effective analysis and as an evaluation factor in any analysis of the
bid or proposal; c) the use of performance type specifications for competitive procurement of a
royalty-free product, apparatus or process; and d) the use of bid or proposal alternatives to each
proposed patented product, apparatus, of process on which a royalty must be paid26.
The following determinations regarding infringement should be made as soon as possible prior to 'start
up' of the patented technology, process, or item (i.e., the technology, process or item mechanically
1237
-------
begins the treatment process for which it was designed and constructed,) or if treatability studies on
the patented technology are planned: a) clear infringement of a patent (negotiate a license agree ment
generally at the patentee's royalty fee, if determined to be reasonable); b) clearly no infringement
(conduct procurement without further ado); or c) a gray area regarding infringement (negotiate a
license agreement, possibly for less than the patentee's specified royalty fee.)
EPA has planned for obtaining license rights to patented treatment technologies. Basic ordering
agreements (BOA) for treatment technologies are recommended, and EPA has developed a
standardized or model BOA for this purpose.27 The model BOA provides the terms and clauses for
agreements to obtain license rights for treatment technologies in Superfund.
After taking into account the foregoing suggestions, it may be desirable to negotiate with the patent
owner to receive a license for patent use. Royalties for existing patents are generally considered
allowable costs as long as the costs are reasonable. In order to help ensure national consistency, any
government agency carrying out a Fund-lead Superfund remediation that is planning to negotiate for
and receive a license for patent use of an innovative technology at a Superfund site should contact
EPA's Design and Construction Management Branch prior to initiating the negotiations.
3.2.2.3. Federal Acquisition Regulation
CO's have the responsibility to apply FAR Part 27 to any Superfund project. For Federal-lead
construction project RFP's and/or IFB's involving procurement of patented technologies, certain FAR
clauses should generally be invoked in the RFP/IFB clauses and/or specifications. This is
recommended in order to provide the maximum allowable assurances the government can give to
potential offerers or bidders that the government would, under certain circumstances, assume the
liability associated with potential patent infringement and/or authorizes the use of a patent.
These clauses are: 1) FAR 52.227-1, "Authorization and Consent," paragraphs (a) and (b) only (no
Alternates); 2) FAR 52.227-2, "Notice and Assistance Regarding Patent and Copyright Infringement,"
paragraphs (a,) (b,) and (c); and 3) FAR 52-227-4, entitled "Patent Indemnity-Construction
Contracts." FAR 52-227-4 Regarding FAR 52.227-4, the present single paragraph of this section
should be designated (a,) and 'Alternate I' of this section should be designated (b) in the clause use
in the RFP/IFB.
FAR 52.227.3, "Patent Indemnity," should not be invoked in construction contracts, since FAR 52-
227-4, "Patent Indemnity-Construction Contracts," applies and should be used. EPA Regions, with
the Director of EPA's Procurement and Contracts Management Division approval, may invoke FAR
52.227-5, "Waiver of Indemnity," into an RFP/IFB, providing that the patents are identified by
number. If -5 is used, it may not be necessary to also invoke -4, since -4 uses a description of the
patented technology, and -5 identifies the patent by number. Government CO's should provide
direction on this matter.
FAR 52.227-4 and/or 5 are inserted into construction contracts in order to provide protection to
contractors if they will infringe a patent when carrying out the construction according to
specification. It should be noted that a waiver of indemnity may not necessarily cover the contractor
from all lawsuit costs if a patent is infringed, since the government can only provide contractor
protection to the extent that is authorized by statute and regulation. If only the authorization/consent
and indemnity clauses were invoked (without the waiver of indemnity clause,) costs for infringement
would likely be borne by the government due to FAR 52.227-1.
1238
-------
3.2.2.4. State-Lead Projects
Part 35.6565(d) of the Subpart O EPA Grants Regulation notes that noncompetitive proposals may be
procured if the item desired by the government is known to be only available from a single source,
or after solicitation of a number of known sources is shown to be noncompetitive. In this situation,
this section notes that the state must request use of sole source for this item from the EPA Regional
award official (usually either the Superfund Division Director or the Regional Administrator) and
provide a justification for its use, as well as conduct a cost or price analysis, and a profit analysis
giving consideration to the establishment of a fair and reasonable profit, in accordance with the
requirements set forth in part 35.6585.zo
State procurement laws and regulations may also have additional requirements in order to sole source.
An investigation into this, including discussions with Superfund Program managers, Counsel and the
Federal and state agency's Grants Administration Divisions and contract specialists, should be
conducted.
3.3. Acquisition Planning
3.3.1. General
As noted previously, Superfund remediation projects might be considered more prone to problems
and changes than non-Superfund construction projects, since they have incomplete performance or
cost information, and few, if any, vendors have sufficiently proven their expertise in implementing
these technologies. When dealing with high risk procurement, it generally is worthwhile to spend
additional effort in the planning stages prior to procurement to ensure that the best possible strategies
are considered and utilized.
3.3.2. Pre-Record of Decision (ROD) and ROD
3.3.2.1. RI/FS Treatability Studies
The NCP identifies EPA's emphasis on the need to perform treatability studies early in the remedial
process. It notes that since innovative technologies may not have been as thoroughly demonstrated
as other technologies, treatability studies during the Remedial Investigation and Feasibility Study
(RI/FS) may be necessary to provide an appropriate evaluation of these technologies. The goal is to,
"through good science and engineering, establish the probable effectiveness of innovative
technologies." If treatability studies are conducted, EPA can eliminate those innovative technologies
which have little potential for performing well at specific sites28. It is especially important to conduct
treatability studies, and where appropriate, pilot-scale testing of innovative technologies during the
RI/FS, in order to better understand a technology's advantages and disadvantages. These studies and
tests will also provide important information with which a proper detailed analysis of a remedial
alternative against the 'nine criteria' may be conducted during the FS.5 The nine criteria encompass
statutory requirements and include other gauges of the overall feasibility of remedial alternatives.
Analyses performed pursuant to the nine criteria (e.g., reduction of toxicity, mobility, or volume
through treatment; cost; implementability;...) concludes with selection of a remedy that meets the
statutory mandates.29
An inventory of treatability study vendors has been prepared and continually updated through EPA's
Office of Research and Development (ORD.) This can be used to gather information regarding the
availability of vendors to conduct a particular treatability study for a specified technology. In
addition, treatability studies conducted to date on particular technologies have been gathered by ORD
for use by the Superfund program and general public. Benjamen Blaney, Kenneth Dostal or Joan
1239
-------
Colson of EPA's Risk Reduction Engineering Laboratory in Cincinnati Ohio may be contacted for
more information regarding these documents, at (513) 569-7406. These reports have been
computerized through EPA's ATTIC program. Although some difficulty regarding data retrieval has
been reported, information on ATTIC can be gathered by calling (202) 382-5747.
EPA's Off-Site policy and RCRA (40 CFR 261.4) regulation allows up to 1000kg of waste to be
brought off-site to a non-permitted facility for treatability testing without obtaining permits; seiparate
facilities for separate tests can each receive up to 1000kg of waste. After testing is done, the EPA
project manager may authorize the residuals to be returned to the site and stored until the RA begins.
Some concerns have been raised that 1000kg of wastes (approximately three drums) may not be
sufficient to conduct an adequate treatability study. 40 CFR 261.4 allows the EPA Regional
Administrator to authorize an additional 500kg of wastes to be transported offsite for these purposes.
It may even be possible to bring more than 1000kg at one time to a non-permitted offsite treatability
facility by: a) immediately beginning 'treatability testing' on up to 1000 kg of the wastes; and b)
storing up to 1000kg of the wastes on the property of the treatability facility according to RCRA
storage requirements. Storing wastes on a transportable tanker, truck, etc. at a separate facilii y in a
manner complying with the RCRA waste transportation requirements might also be an option to bring
more wastes offsite for these purposes. Concurrence on these and potentially other options should
be received from the appropriate EPA Regional RCRA and/or RCRA Authorized State regulatory
contact.
If the waste is considered acutely toxic, a treatability exclusion may not be allowable. Also, a 45 day
waiting period may be required to allow for a treatability exclusion, unless the State in which the
treatability tests will be conducted is delegated RCRA and has waived this requirement.
3.3.2.2. Forward Planning
Complete forward planning activities must be conducted prior to the initial RI sampling; these should
include historical gathering of data regarding what contamination was dumped at the site or caused
the site to be listed on the NPL. Properly conducted early rounds of sampling could then reveal,
through experienced and best engineering judgement, what two or three remedies would likely be
most successful at the site, including whether conditions would favor use of an innovative technology.
As early as possible during the RI, forward-thinking government and private engineers and scientists,
with strong, field-tested experience in hazardous waste design and construction and well versed with
lessons learned in both procurement/contracting and technical issues, should be solicited for their
judgement on all of these decisions.
These efforts would result in more pilot studies of innovative technologies being conducted during
RIs, and might prevent losses in time and money due to non-implementable RODs. More accurate
cost and implementability estimates can be made during the FS, and a stronger technical database can
be developed to help scope any additional design investigations that might be required to properly
procure an RA contractor and construct the technology.
3.3.2.3. Community and Public Input
Community input as it relates to innovative technologies should not be put off until the formal public
comment period, since more time may be needed to understand the advantages of the technology.
Any uncertainties and short-term impacts, including mitigating measures, should be presented to the
community. On-site, pilot scale treatability studies should be coordinated with the community prior
to starting work.5 In addition, it is recommended that the formal public comment period be used to
provide commercial interests with an opportunity to comment on the government's plan for sole
source procurement, if applicable. This strategy of providing a period of time for comment before
1240
-------
remedy selection, and of properly responding to comments received in the responsiveness summary,
provides for enhanced 'due process,' and stronger justification for sole source procurement might be
realized. Also, the potential for future claims might also be lessened.
3.3.2.4. Cost Estimating
It is recommended that Feasibility Studies should develop order of magnitude cost estimates for RA
alternatives which have a desired accuracy of +50 percent to -30 percent.30 Properly conducted pilot
studies can generate the data needed to estimate the RA and O&M costs of the technology within this
desired accuracy. In addition, the potential license cost to construct and operate a patented
technology should be considered during FS alternative analyses.
3.3.2.5. Interim Action RODs
The NCP identifies that interim actions may be undertaken at a site to address a pressing problem
which will worsen if not addressed quickly. Examples of interim remedies include construction of
temporary caps to control or reduce exposures, or on-site containment structures into which highly
mobile and toxic contaminants may be placed. An interim remedy must be followed by a final
remedy which provides long-term protection of human health and the environment and fully
addresses the principal threats and the statutory preference for treatment remedies.18'31
The concept of addressing contamination on an interim basis is not a new idea. In European
countries, highly mobile and toxic soils and media found at abandoned waste sites are commonly
excavated and placed into conveniently located containment structures. These materials, once
contained, are then studied in a methodical manner to determine which technology would best treat
the waste. Should an innovative technology be considered but fail or not perform satisfactorily,
another technology or approach is considered. Since the wastes are contained, considering innovative
means to deal with the waste need not necessarily result in a worsening of the problem if failure
during testing occurs32. Interim remedies, particularly temporary caps over highly mobile surface soil
contamination, should be considered and used more often in the Superfund Program; such actions
might further the use of innovative technologies at sites.
3.3.2.6. Contingency RODs
When selecting innovative technology remedies with uncertainties for success during remediation, and
a pilot scale treatability studies are proposed during design, proven, non-innovative technologies
could be included in the Proposed Remedial Action Plan (PRAP) and ROD as contingent remedies.
If two different innovative technologies appear to be equivalent during FS evaluations, one may be
identified as the selected remedy and the other as a contingent remedy. Information contemplated
by the ROD but developed after its issuance may encourage the lead Agency to select the contingent
remedy.X9 The PRAP should and the ROD must identify the preferred alternative or selected remedy
and the contingency remedy. In the FS, both remedies should be featured in the Alternatives
Evaluation section as able to fulfill the statutory requirements of Section 121 of CERCLA. An
"Explanation of Significant Differences" (ESD) should be issued and made available to the public if
the contingent remedy will be implemented during RD, RA, or O&M.33
The 'two-headed' ROD option helps move innovative technology projects through the pipeline
quicker since, if the innovative technology pilot study or the construction/O&M fails to meet the
performance goals identified in the ROD, design of the contingent remedy can immediately begin
without the need to reopen the ROD and solicit additional public comment. Further, parallel designs
of both the selected and contingent remedy might also be considered beneficial, in order to ward off
the potential loss of time should the selected remedy fail.
1241
-------
Caution is recommended when selecting and implementing contingent remedies. A remedy should
only be selected if there is strong reason and justification that it will be successful. The EPA Region
should also not automatically begin implementing a contingent remedy if the selected remedy is not
initially meeting the ROD's performance goals during it's RD treatability study or RA; adjustments
to the selected remedy's RD treatability study or RA should be attempted before abandoning the
remedy as non-implementable.
3.3.2.7. Non-Inhibitory ROD Language
The selected remedy in a ROD can appropriately or inappropriately narrow the scope of technologies
available and able to treat wastes at a site. The generic type of technology or treatment family can
be described when choosing a remedy. Specific process options within those categories should be
described if there is confidence that those options will be used. For example, an alternative can be
described as employing thermal destruction rather than rotary kiln incineration if other than rotary
kiln thermal processes are potentially usable.34 With this expansion of potential remedies which could
be used, a performance-based design could then be prepared. This might be preferable since any
advancements or expansions of the number of specific technologies in the generic treatment family
since the ROD was signed can be considered.35
However, certain drawbacks may exist with choosing generic RODs. An ESD might potentially be
required when the decision for a specific technology to implement a generic ROD is made. Also,
Applicable or Relevant and Appropriate Requirements (ARARs) for the RA are generally considered
'frozen' at the time of ROD signing; ARARs promulgated after that time should not be required
provided such ARARs could have been identified before the ROD was signed. If a component of a
remedy is not identified at the time of ROD signing (e.g., a particular form of thermal treatment such
as rotary kiln incineration,) requirements in effect when the component is later identified during RD
or at time of RA contract award will be used to determine ARARs.36 Thus, for example, if new
RCRA treatment standard requirements were placed on rotary kiln incinerators in 1989, but a 1987
ROD identified that rotary kiln would be used, only the RCRA requirements for rotary kiln treatment
in 1987 would need to be met. However, if a thermal treatment ROD were identified in 1987, and
the decision to used rotary kiln were made in 1990, the 1989 requirements must be met.
In addition, the NCP identified the need for better accuracy in and stronger reliability of RI/FS cost
analyses.37 Generic alternatives generally cannot have a detailed cost analysis, since the specific
remedy is not identified; less certainty in the overall cost of the remedy would result, and inaccurate
RA cost planning might occur.
Generic remedies or technologies can maximize competition and potentially prevent bid protests or
claims during RD/RA. These benefits are especially important when choosing innovative
technologies as the sole remedy, since, in general, few vendors or companies will have experience in
implementing them, and competition is limited. With the above concerns in mind, it is encouraged
that EPA Regions consider generic remedies during remedy selection.
3.3.2.8. Sole Source
In certain cases, only one technology, process or potentially only one vendor can and will be
considered/determined able to address the risks at a site before finalization of the ROD (e.g., in-situ
vitrification.) In this situation, the PRAP and ROD should clearly specify that that technology,
process or particular vendor's material, product, or service is the only available item that can properly
address the risks at the site. The rationale for such focus must be clearly provided in the PRAF and
ROD; a complete cost and market analyses and other activities identified under Section 3.2.1. of this
paper should be conducted during the FS to justify such a decision.
1242
-------
The FS should consider the availability of and ability to procure necessary equipment and specialists,
specifically during the implementability and cost analyses of alternatives. As noted previously, the
public could provide an early review and comment on the sole source option in the PRAP, and the
ROD would address the public's concerns in a responsiveness summary. A strong rationale for
conducting a sole source procurement would then be available, and a detailed design specification
using sole source to procure the particular item might then be possible without further need to
brainstorm and consider other procurement options during design. This rationale would serve as the
basis of the "Justification and Approval" effort required by Federal and state government contracting
officers who will utilize "other than full and open competition" during RA procurement.
3.3.3. Pre-Design Planning
3.3.3.1. Pre-Design Technical Summary (PDTS) and Remedial Management Strategy (RMS)
"The Preliminary or Design Report Phase" is customary between the planning and design phases of
engineering projects.38 During this phase in Superfund the ROD and supporting documents are
converted to a statement of work (SOW) for RD/RA by expressing EPA's technical and managerial
requirements. The Pre-Design Technical Summary (PDTS) and Remedial Management Strategy
(RMS,) completed during the pre-design planning phase, link the scientific site assessment and the
engineered solution. The PDTS is a comprehensive compilation of technical information to ensure
that the designer fully understands the technical objectives of the RA. The RMS identifies the
number and type of procurement methods, and types of contracts and specifications applicable to the
remedy39. The actual decisions regarding which procurement strategy and type of contract and
specification to be prepared will be proposed by the design contractor, and reviewed by, discussed
with and approved by the lead and support agencies.
The current EPA policy for pre-design planning is that the lead agency is responsible for
brainstorming and developing a 'project delivery strategy' which will be folded into the SOW for
RD/RA. RMS and PDTS concepts are part of that strategy - they need not be formally prepared, but
the thought process identified in both documents must be completed prior to the SOW. Preferably,
this thinking occurs during the FS, specifically during the implementability and detailed cost analysis
of alternatives evaluation. If an innovative technology alternative's design, construction or O&M will
have significant technical difficulties or unknowns, will pose substantial risk for success, or will
create a procurement nightmare, ramifications therein should be considered and balanced against its
benefits and those of other alternatives prior to its selection as the remedy.
3.3.3.2. Design vs. Performance Specifications.
All specifications must be as clear, complete and definite as possible, as well as not be unduly
restrictive. They must contain the essential physical characteristics and functions required to meet
the minimum needs of EPA, not the maximum desired.40 The party contracting for RA warrants that
the RA contractor will be able to fulfill its responsibilities if it makes a good faith effort to follow
"design" specifications which precisely state how the contract is to be performed. If the RA
contractor fails to comply because the contract documents are inadequate, the contracting party bears
the risk of loss. In contrast, if the party contracting for RA allows the contractor discretion in how
to meet the contract obligations by providing "performance" specifications and no explicit statement
of how to design or build the item is provided by the contracting party, the inability to complete the
contract is borne by the RA contractor. If the RA contractor has undertaken an impossible task,
meets technological problems, or cannot complete performance because of its lack of experience, the
contractor and not the contracting party, bears the risk of loss.41
1243
-------
Performance specifications generally encourage innovation and competition and allow contractors
flexibility in approaching a design and construction item which has intentionally not been precisely
designed. Unless: a) a technology can be efficiently and properly designed to ensure little risk of
failure; b) competition is reasonably expected; and/or c) a sole source procurement is planned,
performance specifications for the procurement of innovative technologies is recommended.
3.3.3.3. Contract Type and Method To Be Used
3.3.3.3.1. General
There are several key references to help determine the proper contract method and type to be used
during procurement of hazardous waste projects. The nature of the project, the degree of risk willing
to be accepted, and level of 'known unknowns and unknown unknowns' are discussed, and excellent
comparative analyses regarding the pros and cons of each, are provided.14>19i39'4°i41'42'43
Two primary contract methods may be used for the procurement of supplies, services, and RA. These
are the solicitation of sealed bids (formal advertising method) and the request for competitive
proposals (competitive negotiation method.) The term "contract type" has several different
connotations. Often it is used to indicate the various methods of pricing arrangements, of which there
are two basic types: fixed-price contracts and cost-reimbursement contracts. In considering the
appropriate competitive procedures to be used, a public agency should determine: a) the time available
for the solicitation, submission, and evaluation of offers; b) if the award will be made on the basis
of price, other factors or a combination; c) if it is necessary to conduct discussions with the
responding source about their offers; and d) if there is a reasonable expectation of receiving more
than one offer.41
The FAR permits the government a variety of choices in selection of contract type. The Government
decides where it wishes to place its resources and risk in the completion of a project. Fixed price
contracts force the Government to do a thorough investigation and design prior to solicitation; these
contracts minimize risk allocation to the Government and have the lowest price at the time of
solicitation. The other types of contracting allow an expedited solicitation while placing greater
demands on the Government in contract administration, risk allocation and potential cost.43
The following is a brief and generalized overview of the applicability of specific contract types and
methods for innovative technology procurement, and is based on certain references.14'19
3.3.3.3.2. Contract Type To Be Used (FAR Part 16)
3.3.3.3.2.1. Fixed Price
Due to the lack of proven cost data, firm fixed-price (lump-sum) specifications for innovative
technologies may generally not be in the government's best interests. This type should only be used
when the specifications and costs can be tightly defined.
3.3.3.3.2.2. Unit Price
Under unit price contracts, the government estimates quantities and pays on the actual costs. Due to
the unknowns associated with these technologies, this type is generally recommended since some cost
risk is shifted away from the contractor.
1244
-------
3.3.3.3.2.3. Cost Reimbursement
Under cost reimbursement contracts, the government also shares in the risk, and provides a means for
the government to enhance its knowledge base. Costs involved are actual, not those determined by
a contractor trying to consider all possible contingencies during the solicitation. Modification to the
technology during the contract is easier to accomplish. These contracts require substantial government
construction contract management43 in the form of heavy oversight, in order to assure that the costs
are 'actual.'
3.3.3.3.2.4. Indefinite Delivery
For this type of contract the maximums and minimums for each order are set. Although it may be
preferable for service procurement, this contract type might not be preferable for innovative
technology procurement.
3.3.3.3.3. Contract Method To Be Used (FAR Part 13)
3.3.3.3.3.1. Small Purchase
If the cost for procuring a technology or an item is under $25,000, less formal justification for sole
source is required.
3.3.3.3.3.2. Sealed Bidding
The sealed bidding method is time consuming and the contract is awarded based on price; no
discussion with the offerers is necessary. Quality, price, and business reputation usually cannot be
bargained for. As such, it is generally not recommended for innovative technologies.
3.3.3.3.3.3. Negotiation (RFP)
Negotiation is involved in most procurement methods other than sealed bidding, and is generally
recommended for innovative technology procurement. Bids must be responsive, but can be
negotiated. Performance specifications for this method are preferred. The government identifies
which offerer is in the 'competitive range,' and negotiations commence to award to the firm with the
best combination of factors identified in the RFP and their proposal. For innovative projects, key
factors include experience, personnel qualifications, past performance, cost, and technical excellence.
Selection should be based on competence, cost, and ability/experience with other similar projects.
A key advantage with negotiated procurement is that it allows the Government discretion in selecting
a successful offerer. The Government, through a source selection plan, determines evaluation factors,
relative importance of the factors and importance of the cost differentials of the offers. Government
evaluators use weighted evaluation factors as a guide in selecting the best offer. Inclusion of these
factors in relative order in the RFP informs potential offerers of the areas considered critical by the
Government. The offerer can limit its risk by further defining its proposed actions within the
specifications, and seeking clarification on technical issues which could reduce their risk and
subsequent offer.43
1245
-------
3.3.3.3.4. Modified Two-Step Sealed-Bid through Prequalification of Vendors and/or RA
Constructors
3.3.3.3.4.1. General
Consideration should be given towards 'prequalification' of potential vendors and/or constructors for
RA of an innovative technology; certain technologies with multiple possible vendors (e.g., chemical
extraction and soil washing) might best be procured this way. For instance, at the beginning of
design, an announcement can be made in the Commerce Business Daily (CBD) calling for the
prequalification of vendors by conducting pilot-scale studies on wastes at the site over a given period
of time during design. Information regarding the remedy, cleanup goals, type and concentrations of
contaminated media, and other pertinent information should be provided. In order to encourage
competition, the government might pay some of the costs for the pilot studies (e.g., lab testing.) A
reasonable amount of time should be provided to those who might conduct such studies. All vendors
who performed satisfactorily (e.g., met the cleanup goal) would then be considered 'prequalified'.
The government would conduct discussions after prequalification to solicit criteria the vendors feel
should be put in the plans and specifications on which they would bid for the project. At this time
the government should consider asking for plans and specifications of the treatability systems used
by the vendors. After a solicitation for this information, the government would carefully assess the
information from the pilot studies and discussions with vendors, and prepare a set of plans and
specifications on which competitive sealed bids would be made. The treatment process would be a
performance specification, for which the low bid would be awarded the project. The Chemical
Control site in New Jersey used this approach for procurement of a vendor to perform
solidification/stabilization; although this is a non-innovative remedy, the procedure is applicable to
innovative projects.44
An alternative to the above is to conduct the CBD solicitation at the completion of design. In some
respects, competition would be enhanced since additional time for new vendors to come into the
marketplace is provided. However, conducting the call for vendors after the design is completed
might unduly restrict competition. As such, the government should consider conducting two CBD
solicitations: one at the beginning of design as discussed above, and one at the end of design. The
second solicitation would be for sealed bids, and allow companies who did not attempt to prequalify
to bid. These companies would be provided samples of the waste to be treated, and required to
submit information in the form of pilot or bench study data, and/or plans and specifications for their
process in sufficient detail to allow the government to make a judgement that that process would have
reasonable chance for success in meeting the performance goals. If one of these companies were the
low bidder, they would be awarded the contract, possibly on a contingent basis. If no pilot study data
were submitted with the bid, the contractor would construct their process at no charge to the
government to full scale at the site. If the process could not meet performance goals, the contractor
would demobilize at no charge to the government; the next lowest bidder with pilot study data would
be awarded the contract.
3.3.3.3.4.2. Treatability Studies
EPA is limited in the number of treatability studies it can perform at a site. Competition would likely
be increased by using a prequalifying method which provides samples of site wastes to prequalified
vendors who can prove they can treat the waste at their facility. It is likely that vendors will invest
in a test during design rather than RI/FS, since the RFP for a specific technology is forthcoming.
Depending on the need for design data, the results of vendor treatability study data may or may not
be incorporated into the RD specifications. If the data is not needed, independent vendor tests could
occur at the same time as design activities, so as not to delay the project. Prequalification
1246
-------
requirements could include permitting at the vendors facility, ownership of pilot or full-scale
equipment, a proper QA/QC plan, and provisions for residual disposal. Providing vendors with the
opportunity to conduct these tests might make it less likely that a bid protest will occur if a
treatability vendor wins the RA contract;21 the number of responsive and responsible bidders capable
of meeting the treatment goals would have been narrowed in a justifiable manner, and protests from
those not capable of meeting such goals can be determined to be non-responsive.
3.3.3.3.5. Service Contract using Competitive Proposals
A fixed price combination of lump sum and unit price may be an option for innovative technologies.
A service contract may be procured using competitive proposals, and bonds would not be required.
Evaluation criteria might weigh technical concerns at 60%, with price weighing at 40%. Construction
specifications for soil excavation would be written; since 'construction' is occurring, wage rates
subject to the Davis Bacon Act would apply. Service specifications, with the principal purpose being
to treat contaminated soils using a mobile treatment unit, would be used; unit price per cubic yard
treated would be the measurement and payment basis. The government would pay if the treatment
goal were achieved.45
3.3.4. Remedial Design (RD)
3.3.4.1. General
As noted previously, the actual decisions regarding which procurement strategy and type of contract
and specification to be prepared will be proposed by the design contractor in their RD Work Plan.
The firm would use all information gathered to date to assist in developing this strategy. The RD
workplan is the first major design deliverable, provided soon after the design contract is awarded to
the firm, and is reviewed by, discussed with and approved by the lead and support agencies.
This effort, the design field investigation, or value engineering efforts might result in a decision to
expand the procurement to a more generic category if it was convincingly determined that other
technologies might also achieve the ROD's remediation goals. If an inappropriately narrow ROD has
been issued, the EPA Region should consider preparing a documentation of non-significant
differences, an ESD or a "ROD Amendment" as early as possible during the design phase to prevent
major disruptions to the project schedule or cost.
3.3.4.2. Data Gathering
Accurate data on heat transfer, mixing, separation, etc. gathered during design, or even the RI/FS,
might provide for better design reliability and greater confidence, thus likely lessening an offerer's
potential bid contingencies to cover unknowns and reducing the overall cost of the RA. Interviews
with a number of potential vendors and/or construction firms who might be candidates for the
construction of the innovative technology soon after the ROD might help guide the direction of the
RI/FS and/or design data gathering effort. Among other things, information regarding what
engineering or investigatory data would be needed to bid the project should be discussed.
Information regarding the availability of data, including what physical/chemical data collected to
date, and how it can be retrieved, should be identified. Materials information, particularly volume
estimation with a basis of calculations, should be provided.46 In general, four major categories of site
characterization data are needed to effectively remediate subsurface contamination, including source
remediation. These data categories include site data, geochemical data, geotechnical data, and
hydrogeological data.47
1247
-------
3.3.4.3. Treatability Studies
As noted previously, it is recommended to conduct pilot-scale treatability studies of innovative
technologies prior to finalization of the ROD. During the design phase, in order to make more
realistic judgments regarding construction costs, and to help in deciding what risk exists regarding
whether the technology will meet the performance goals set in the ROD, it might be useful to scale-
up the pilot scale treatability study.
For example, at the Wide Beach site in New York State, a ROD for "chemical treatment" was signed,
and a treatability study during the RD using potassium polyethylene glycolate (KPEG) dechlorination
was conducted. An RFP for PCB dechlorination was advertised, but required a demonstrated
technology which has proven it could treat wastes similar to those at the site. The prime contractor
awarded the RA had KPEG as the PCB dechlorination process and used the same vendor who did the
treatability studies as a subcontractor. The selected vendor went directly from pilot scale to full scale
on-site remediation (from a 40 gallon pilot reactor to eight 3000 gal reactors.) Although the project
is considered successful, a potentially significant cost savings might have been realized if the designer
had scaled up and fully tested one of the 3000 gal reactors. As noted previously, this data could have
provided more accurate data on heat transfer, mixing, separation, etc. to the offerers and/or bidders
for the RA. This data would have provided better design reliability and greater confidence, thus
likely lessening an offerer's potential bid contingencies to cover unknowns and reducing the overall
cost of the RA.48
It is possible that further treatability studies beyond those conducted during the RI/FS may not be
required in design; verification testing at the start of actual site cleanup may suffice.35 However, it
should be carefully investigated whether RI/FS treatability studies are sufficient to properly design
the remedy, provide sufficient information to potential offerers and/or bidders, and provide for
competitive procurement, as discussed previously.
3.3.4.4. RFP and/or IFB Instructions to Offerers and Clauses
Throughout the remedial pipeline but particularly near the end of design, design contractors and
government contracting officials should critically evaluate the risk of innovative technology
procurement success and failure to the government, design firm, and construction contractor. This
assessment of risk should play an important role in determining what instructions to offerers and
clauses will be inserted into the RFP and/or IFB. The government has an obligation to inform the
potential RA construction firm of known 'unknowns' of the project in the specifications. Special
consideration should be given to inserting and/or reinforcing the following clauses and/or instructions
if innovative technologies are being procured: a) Patents, Data, and Copyrights; b) claims and change
order procedures; c) termination for convenience; d) variation in quantity; e) change in site
conditions; f) certification of performance; g) suspension of work; h) measurement and payment; and
i) default.
3.3.4.5. RD Claims Review
A "claims prevention" review should be conducted as part of the prefinal design review to eliminate
conflicts, inconsistencies, ambiguities, errors, omissions or other identifiable problems in the plans,
specifications and contract documents that may become the source of change orders and claims. This
review should attempt to eliminate unduly restrictive specifications and review "brand name or equal
specifications" to assure that salient characteristics to be met are specified.49 Several key papers were
presented on claims and change orders between May 1-3, 1991 in Dallas TX at EPA's 'Design and
Construction Issues at Hazardous Waste Sites' national conference; these papers should be referenced
for more information regarding how to prevent these issues.
1248
-------
3.3.5. Remedial Action (RA) Documentation
Heidi Facklam of the U.S. Army Corps of Engineers (USAGE) has reported that RD/RA's need to
be properly documented and evaluated; in particular, construction records, conditions and activities
should be recorded and preserved in a readily accessible form. Data regarding construction
modifications and changed conditions, long term performance monitoring and site maintenance, and
baseline information for design of repair/modifications in case of failure should be systematically
gathered and prepared jointly by the design and construction staff. The knowledge gained and lessons
learned during the construction process would provide valuable insight for future construction
projects. Documentation reports for this type of information have been required for nearly one
hundred years for USAGE engineering structures.50
Due to the inherent unknowns associated with innovative technology implementation at Superfund
sites, a standardized and routine documentation effort similar to that required for USAGE projects
would provide a vital service by eventually lessening the procurement risks associated with such
technologies. With the availability of such standardized and readily accessible reports, actual cost data
could be analyzed, designs could be improved and RA change orders minimized. In the absence of
a specific national guidance and/or policy for such documentation, it is recommended that USAGE'S
documentation requirements as outlined in Ms. Facklam's report be followed immediately.
3.4. Enforcement Considerations
The following four considerations are provided regarding Potentially Responsible Parties (PRPs) and
innovative technology RODs: (1) PRP concerns generally focus on cost and continued liability in the
event of remedy failure or implementability problems. If a treatment remedy fails or costs are
relatively high compared to other arguably effective remedies, PRPs will attempt to argue that EPA
is not entitled to full cost recovery. It is therefore important to conduct treatability studies during
the RI/FS stage. (2) Contingent RODs can improve or detract from the lead agencies negotiating
position, depending on the contingencies involved. It is therefore important to clearly identify the
expected performance levels for the innovative technology in the ROD, or negotiation delays will
result. (3) When practicable, contingent RODs for two innovative technologies could provide an
opportunity to generate design-specific data related to the performance of the technology prior to the
final specification of the technology to be implemented. This might allow PRPs to achieve
performance requirements without necessarily being required to implement the most expensive
remedy. However, costs associated with the RD treatability testing of a non-selected innovative
technology contingent remedy may be challenged in cost recovery.5 (4) As noted previously, generic
alternatives generally cannot have a detailed cost analysis; less certainty in the overall cost of the
remedy would result, and potential difficulties in settlement negotiations with PRPs interested
primarily in the 'bottom line' (costs) might occur.
4. CONCLUSION
4.1 Summary
EPA's Superfund 90-Day Study6 makes clear that better ways to enhance the development of
innovative technologies are needed. EPA's PCMD has made several efforts to help eliminate
constraints to the procurement of treatment technologies, and has pledged to continue to work with
the Superfund Program to explore ways to expand the use of innovative technology.51 EPA's Office
of Solid Waste and Emergency Response, and in particular the SITE Program and the Remedial
Operations and Guidance Branch of the Hazardous Site Control Division, have also made progress
towards this goal. These efforts should and will continue.
1249
-------
The following constitutes a summarization of considerations which might help reduce constraints to
the procurement of innovative treatment technologies:
A) Spend additional efforts in the planning stages prior to innovative technology procurement
to ensure that the best possible strategies are considered and utilized. A team of those
associated with the site should meet soon after the ROD to brainstorm; the team should
include: a) government contracting officers, project managers, legal counsel and technical
representatives; b) government contractors (including RI/FS and design) and construction
representatives; and c) potential offerers, bidders and/or vendors.
B) Continue to sponsor national conferences on a yearly basis which help introduce promising
international technologies through technical paper and poster displays, and showcase SITE and
other domestic innovative technologies. On an annual or biennial basis, continue to conduct
a national conference geared towards design and construction issues at hazardous waste sites,
in order to have an open exchange of ideas and promote formal and informal discussion of
design and construction issues. PRPs, private organizations such as the Hazardous Waste
Action Coalition and the American Council of Engineering Consultants, States, Federal
agencies, and private construction firms, vendors, consultants, corporations, and individuals
should all be actively solicited for their participation and insight. These conferences will
encourage national consistency, help develop more efficient and practical means to move
innovative technology projects through the pipeline, and augment EPA's current efforts to
revise its Superfund remediation guidance and policies.
C) Increase involvement of top engineering colleges and graduate schools in the research and
development of new and improved innovative technologies, particularly in the civil,
environmental, chemical and mechanical disciplines. Many of the graduates of these schools
join those organizations leading the effort in hazardous site remediation; their efforts can
strongly influence the regulated community. In addition, as students, they comprise an
excellent form of relatively 'cheap labor.'
D) Utilize performance specifications vs. design specifications when feasible since they
encourage innovation and competition and allow contractors flexibility when approaching a
design and construction item.
E) Interview a number of potential vendors and/or construction firms who might be candidates
for the construction of the innovative technology soon after the ROD. Develop a checklist
of items to be asked, including what specifications should be performance vs. design, what
contracting type and method are recommended, what engineering or investigatory data would
be needed to bid the project, etc. Use the interviews to help guide the direction of the design
and/or construction.
F) Increase the emphasis on the use and development of national innovative technology databases
of treatability studies, treatability study vendors, and post construction reports. These
databases should be user-friendly and accessible to anyone.
G) Increase the consideration and use of interim remedy temporary containment options and/or
Regional facilities which address certain forms of contamination or provide certain types of
treatment.
1250
-------
4.2 Concluding Comments
It has been eminently stated that "innovation is a mandate in the Superfund program. Innovation and
project complexity involve cost, time, and performance risks because of the lack of precedent...there
shall be compromises...The terms of the compromises - including inexperience, overly restrictive
technical or managerial requirements, pressures of deadlines and economy in cost - vary the shape
of the project to be designed...wise and carefully selected technical and managerial requirements
(must be set..).Unfortunately, compromise implies a degree of failure. It is then the responsibility
of the designer to obviate failure within the context of the technical and managerial requirements
articulated (by the government...) It is, however, impossible for any design to be 'the logical outcome
of the requirements' simply because, the requirements being in conflict, their logical outcome is an
impossibility."41
A Physics Professor commenced his first thermodynamics lecture by rewording the three thermo laws:
1) You can't win; 2) You can't break even; 3) You can't get out of the game,52 At one time or
another, those with experience in Superfund might feel this Professor has unwittingly and neatly
described the Program. Since 'we can't get out of the game,' early and well reasoned procurement
planning can speed the development and success ratio of innovative technologies at Superfund sites.
We might 'win' or at least 'break even' more frequently, and continue to improve the methods used
in Superfund to provide protection of human health and the environment.
5. DISCLAIMER
This report has not undergone a formal USEPA peer review. The views expressed by this author are
his own and do not necessarily reflect the views, policies, or ideas of USEPA. This document does
not constitute any rulemaking, policy or guidance by the Agency, and cannot be relied upon to create
a substantive or procedural right enforceable by any party. Any mention of trade names, products,
or services does not convey, and should not be interpreted as conveying, official USEPA approval,
endorsement or recommendation. This document is not intended to and does not constitute any
rulemaking, policy or guidance by the Agency. It is not intended to and cannot be relied upon to
create a substantive or procedural right enforceable by any party. Neither the United States
Government nor any of its employees, contractors, subcontractors or their employees makes any
warranty, expressed or implied, or assumes any legal liability or responsibility for any third party's
use of or the results of such use of any information or procedure disclosed in this report, or represents
that its use by such third party would not infringe on privately owned rights.
Your comments on the utility of this paper and how it might be improved to better serve the
Superfund program's needs are encouraged. Comments may be forwarded to the attention of Kenneth
Ayers, Design and Construction Management Branch, USEPA, Mailcode OS-220W, Washington DC
20460.
6. REFERENCES
I) Comprehensive Environmental Response. Compensation, and Liability Act of 1980. as
amended by the Superfund Amendments and Reauthorization Act of J986 (SARA,) P.L. 96-
510, Section 121(b)(l.)
2) SARA, P.L. 96-510, Section 121(b)(2.)
3) SARA, P.L. 96-510, Section 31 l(b.)
1251
-------
4) "National Oil and Hazardous Substances Pollution Contingency Plan" (NCP,) 40 CFR Part
300.430(a)(l)(iii)(E,) Federal Register Vol. 55, No. 46, March 8, 1990, page 8846.
5) "Advancing the Use of Treatment Technologies for Superf und Remedies," Memorandum from
Henry L. Longest, Director, USEPA/OSWER/OERR, and Bruce Diamond, Director,
USEPA/OSWER/OWPE, to USEPA Regional Superf und Division Directors, OSWER Directive
#9355.0-2, February 21, 1991.
6) (a) William Reilly, USEPA Administrator, A Management Review of the Superfund Program
(Superfund 90 Day Study,) June 1989, Chapter 4; and (b) A Management Review of the
Superfund Program -Implementation Plan. EPA/540/8-89/009, September 1989, pages 96-
116.
7) U.S. Congress, Office of Technology Assessment, "Coming Clean: Superfund Problems Can
Be Solved," OTA-ITE-433 (Washington DC, U.S. Government Printing Office, October 1989,
pages 181-190.
8) USEPA/OSWER, "The Superfund Innovative Technology Program: Technology Profiles,"
EPA/540/5-90/006, November 1990.
9) USEPA/OSWER, "The Superfund Innovative Technology Program: Technology Profiles,"
EPA/540/5-88/003, November 1988.
10) USEPA/OSWER, "Forum on Innovative Hazardous Waste Treatment Technologies: Domestic
and International - Technical Papers," EPA/540/2-89/056, September 1989.
11) USEPA/OSWER, "Innovative Treatment Technologies: Semi-Annual Status Report - First
Issue," EPA/540/2-91/001, January 1991.
12) "Superfund Federal-Lead Remedial Management Handbook," USEPA/OSWER/OERR,
EPA/540/G-87/001, December 1986.
13) "Superfund State-Lead Remedial Management Handbook," USEPA/OSWER/OERR,
EPA/540/G-87/002, December 1986.
14) Formation of Government Contracts. 2nd Edition, Cibinic and Nash, George Washington
University Government Contracts Program, Washington D.C., 1986.
15) "Contracting at EPA," Memorandum from William K. Reilly, Administrator, EPA, to All
Agency Personnel, April 17, 1990.
16) Contracting at EPA. Advance Copy of EPA Order (no classification number yet,) available
from Linda Garczynski, EPA/OSWER/CORAS, 11/27/90.
17) "Competition in Contracting Act of 1984," incorporated into P.L 98-369, July 18, 1984.
18) NCP, March 8, 1990, Preamble wording, Part 300.430(a)(l,) "Program Management
Principles," pages 8702-5.
19) "Student Guide for Construction Contract Administration and Management," United States
Navy, Naval School, Civil Engineer Corps Officer, Port Hueneme, California, 93043, April
1990.
1252
-------
20) "Cooperative Agreements and Superfund State Contracts for Superfund Response Actions;
Final Rule," 40 CFR Part 35, Subpart O, Federal Register Vol. 55, No. 108, June 5, 1990.
21) "Additional Comments on Procurement of Innovative Technology," Memorandum from Linda
Galer, USEPA/OSWER to Ed Hanlon, USEPA/OSWER, May 9, 1990.
22) "Introduction to Procurement Under Superfund," Unpublished Training Manual, Thomas
Whalen (principal author,) EPA/OSWER/OERR/HSCD/DCMB, 1987, Chapter 8.
23) Personal telephone conversation with Sidney Ozer, EPA Region III Superfund Contracting
Officer, March 28, 1990.
24) Draft Fact Sheet, "Use Of or Rights In Patents," Thomas Whalen,
EPA/OSWER/OERR/HSCD/DCMB, undated.
25) Memorandum from Henry L. Longest II, Director, EPA's Office of Emergency and Remedial
Response, to Charles E. Findley, Director, EPA Region 10 Hazardous Waste Division, "Patents
for Innovative Treatment Technologies," September 30, 1987.
26) Memorandum from John T. Rhett, EPA Deputy Assistant Administrator for Water Program
Operations, and Frances E. Phillips, EPA Associate General Counsel for Grants, Contracts and
General Administration, to EPA Regional Administrators, "Royalties for Use of or for Rights
in Patents," Construction Grants Program Requirements Memorandum # 79-2, November 13,
1978.
27) "Memorandum of Understanding Related to Basic Ordering Agreements (BOA) For Treatment
Technologies," Memorandum from David J. O'Connor, Director of EPA Procurement and
Contract Management Division (PCMD) to Henry L. Longest II, Director of EPA's Office of
Emergency and Remedial Response, January 23, 1990
28) NCP, March 8, 1990, Preamble wording, Part 300.430(e,) "Feasibility Study," page 8714.
29) NCP, March 8, 1990, Preamble wording, Detailed Analysis of Alternatives section, pages
8719-8723.
30) Interim Final Guidance for Conducting Remedial Investigations and Feasibility Studies under
CERCLA (RI/FS Guidance.) USEPA/OSWER, EPA/540/G-89/004, October 1988, page 6-4.
31) Fact Sheet, "Guide to Developing Superfund No Action, Interim Action, and Contingency
Remedy RODs," USEPA/OSWER/OERR/HSCD, OSWER Publication #9355.3 -02FS-3, April
1991.
32) Personal conversation with Michael J. Mann, P.E., Vice President of Engineering, Geraghty
& Miller, Inc., 3820 Northdale Boulevard, Suite 200, Tampa FL 33624; in Dallas TX, May
2, 1991.
33) ROD Guidance, USEPA/OSWER, EPA/540/G-89/007, July 1989, pages 8-10, 9-17, and 9-
18.
34) Interim Final Guidance on Preparing Superfund Decision Documents: The Proposed Plan and
Record of Decision (ROD Guidance.) USEPA/OSWER, EPA/540/G-89/007, July 1989, page
6-20.
1253
-------
35) "Request for Review of Memo on Procurement of Innovative Technology," Memorandum
from Linda Galer, USEPA/OSWER to Ed Hanlon, USEPA/OSWER, April 23, 1990.
36) NCP, March 8, 1990, Preamble wording, ARARs section, pages 8747 and 8757-58.
37) NCP, March 8, 1990, Preamble wording, RI/FS section, pages 8722 and 8726-8730.
38) Draft Fact Sheet, "Classification of Engineering Services - ASCE Manual No. 45 Applied to
RI/FS, RD, and RA," Thomas Whalen, EPA/OSWER/OERR/HSCD/DCMB, undated.
39) "Guidance on Expediting RD and RA," USEPA/OSWER, EPA/540/G-90/006, August 1990.
40) Undated Draft "The Problems With Public Procurement Practices, Design Specifications,
Fixed-Price Competition, and Other Stuff," Tom Whalen,
EPA/OSWER/OERR/HSCD/DCMB.
41) "Remedial Management Strategy," Thomas Whalen, P.E., U.S. Environmental Protection
Agency, Hazardous Site Control Division, 401 M Street S.W., Washington, D.C. 20460, as
presented May 2, 1991 in Dallas TX at EPA's 'Design and Construction Issues at Ha2.ardous
Waste Sites' national conference.
42) "Tradeoff Analysis in Negotiated Procurement Procedures for Construction (Are the
Additional Points Worth the Additional Dollars,)" John C. Taylor, Unpublished paper, United
States Environmental Protection Agency, OSWER/OERR/HSCD/DCMB, Washington, D.C.,
1988.
43) "Acquisition Selection For Hazardous Waste Remediation," William R. Zobel, P.E., U.S.
Environmental Protection Agency, Hazardous Site Control Division, 401 M Street S.W.,
Washington, D.C. 20460, as presented May 2, 1991 in Dallas TX at EPA's 'Design and
Construction Issues at Hazardous Waste Sites' national conference.
44) Personal telephone conversation with Rick Schwarz, EPA Region II Superfund Project
Manager, November, 1989.
45) Undated Draft "Service Contract Procured Using Competitive Proposals," Tom Whalen,
USEPA/OSWER/DCMB.
46) Draft "Guidance for Preparation of a Pre-Design Technical Summary,"
USEPA/OSWER/OERR/HSCD/DCMB, November 27, 1990.
47) "Site Characterization Data Needs for Effective RD and RA," John E. Moylan, U.S Army
Corps of Engineers, Missouri River Division, Kansas City, MO, as presented May 2, 1991 in
Dallas TX at EPA's 'Design and Construction Issues at Hazardous Waste Sites' national
conference.
48) Personal telephone conversation with Herb King, EPA Region II Superfund Project Manager,
April 11, 1990.
49) "Management of Construction in the Superfund Program," Unpublished Training Manual,
Thomas Whalen (principal author,) EPA/OSWER/OERR/HSCD/DCMB, 1987.
1254
-------
50) "HTW Construction Documentation Report: A Necessary Element in a Successful
Remediation," Heidi L. Facklam, P.E., U.S. Army Corps of Engineers, Missouri River
Division, Omaha, NE, as presented May 2, 1991 in Dallas TX at EPA's 'Design and
Construction Issues at Hazardous Waste Sites' national conference.
51) "Constraints to the Procurement of Treatment Technologies," Memorandum from Belle. N.
Davis, Director, Policy and Management Support Staff, to David J. O'Connor, Director, EPA's
Procurement and Contracts Management Division, January 29, 1990.
52) Physics lecture, Dr. Robert Meers, Agricultural Engineering Department, Rutgers University,
October 1977.
1255
-------
Trial Burn at MOTCO Superfund Site
LaMarque, Texas
Mary Ann E. LaBarre
U.S. Environmental Protection Agency
Region 6 (6H-ET)
1445 Ross Avenue Dallas, Texas 75202
(214) 655-6735
Philip C. Schwindt, P.E.
U.S. Environmental Protection Agency
Region 6 (6E-SC)
1445 Ross Avenue Dallas, Texas 75202
(214)-655-6486
Alexis W. Lemmon, Jr., P.E.
Metcalf & Eddy
700 South Illinois Avenue, Suite A103
Oak Ridge, Tennessee 37830
(615)-482-0036
Submitted for U.S. EPA Conference on Design and
Construction Issues at Hazardous Waste Sites
Dallas, Texas May 1-3
INTRODUCTION
At the MOTCO Superfund Site, the MOTCO Trust Group and the U.S. Environmental Protection
Agency (EPA), have begun incinerating 11 to 15 million gallons of waste, consisting of waste oils, and
industrial process wastes including styrene tars, vinyl chloride, and small concentrations of PCBs,
mercury and lead. The purpose of this paper is to discuss the initial trial burn and the results.
The discussion will also include operational difficulties and potential concerns of an incinerator.
BACKGROUND
In May 1990, the EPA approved a plan for conducting trial burns of hazardous waste in two
incinerators called Hybrid Thermal Treatment System (HTTS) units, constructed onsite. One unit,
HTTS-2, will be used to process solid material, sludges, aqueous waste, and organic liquids. The
second unit, HTTS-3, is processing aqueous waste and organic liquids.
In early operations, the incinerators were tested using uncontaminated dirt, water, and oil. On May
23, ] 990 waste was introduced into HTTS-3 unit to begin to bring the incinerator up to full operation.
IT Corporation conducted three pretests on July 4-5, July 25-26, and September 6. The trial burn
for the HTTS-3 unit started October 9 and was completed on October 12. The HTTS-3 unit is
continuing to burn waste at conditions based on the operating parameters demonstrated as safe during
the pretests.
The results of the initial pretest conducted by the MOTCO Trust Group on July 4 and 5 met the
performance standards for the Destruction Removal Efficiencies (DREs) of 99.9999% for carbon
tetrachloride and 1,1,2 trichloroethane, and 99.99% for napthalene. The emissions of particulates or
solid particles did, however, exceed the performance standard of an allowable concentration of 0.08
1256
-------
grains/dry standard cubic foot during the first two pretests. Therefore, in August, IT installed a
Hydro-Sonic Super-Sub steam assembly to increase the particulate removal. Results from the pretest
conducted on September 6 show the particulate level met the performance standard.
Trial burn results were received by the EPA on February 27, 1991. Incineration of all onsite waste
material is expected to take at least 14 months after the trial burn of HTTS-2. Delisted ash from the
incineration process will be disposed of onsite, and, after the project is completed, the process
equipment will be dismantled and removed from the site. An impervious clay cap will be constructed
onsite over the delisted ash and covered with a layer of topsoil. The area will then be graded and
seeded, and a security fence will be installed. The MOTCO Trust Group, with EPA oversight, will
monitor the property for at least 30 years to ensure site safety and protection of human health and
the environment.
DISCUSSION
DESIGN AND OPERATION (OF THE HTTS-3)
OVERALL CONFIGURATION
The HTTS-3, a liquids incineration, consists fundamentally of the following functional components:
(1) the waste and fuel preparation and feed system;
(2) the combustion chamber;
(3) the quench chamber;
(4) the gas conditioning system;
(5) the dual Hydrosonic scrubber units;
(6) the induced draft fan;
(7) the stack.
In addition, auxiliary equipment required for supplying, recycling, conditioning and purging the
quench/scrubber liquids contributes to the overall functioning of the gas cleaning system.
The inter-relationships of the various components of HTTS-3 are shown schematically in Figure 1.
Also shown in their approximate locations are the various points of sampling of the flow streams of
the incineration process.
A portion of the fuel and liquid waste feeds are pumped to burners/injectors in the upper section of
the HTTS-3 burner chamber. Both primary and secondary combustion air are introduced into this
section of the HTTS-3 combustion chamber. Near the top of the bottom combustion chamber,
additional waste oil and aqueous waste can be introduced. The minimum 2-second retention time
would be computed on the basis of the remaining chamber volume, starting somewhat below the last
(vertical) point of introduction of waste, and the actual volumetric flow rate of the combustion gases.
The combustion gases then flow in sequence through the gas cleaning system, consisting of the quench
chamber, the gas conditioning system, and the Hydrosonic scrubber, and through the induced-draft
fan and up the stack. Caustic is added at certain points in the gas cleaning system so that the acid
1257
-------
o oo
£ Z
o
Lu i— i
O I—
-H
O
3 D-
O
—1 C5
O _l
1-1 O.
LU
O
on
to
CO
i
ex
1258
-------
gases, HC1 and SO2, as well as particulates are removed from the combustion gas stream before it is
discharged to the atmosphere.
Many additional supporting items of equipment needed for introducing fuel, waste, and combustion
air, and for removing various wastes generated are an inherent part of the overall system. The digital
electronic control system, supplied with instantaneous information by various sensors, measuring
intensive and extensive parameters, are used for observing and controlling the system operation.
OPERATIONAL DIFFICULTIES
Observed difficulties in operation can be considered to stem primarily from two sources. First, the
waste oil feed was not sufficiently characterized from a process standpoint. This was in spite of the
multitude of samples taken and analyzed during the Site Investigation (SI) and Remedial Investigation
(RI) phases of this project. Characteristics such as viscosity, viscosity index, surface tension, and
polymerization potential, important to operation of the incineration system, are generally not
measured or considered significant when the evaluation of risk is of primary concern. And,
follow-up testing to define these additional characteristics was not performed prior to system design.
Second, the combustion characteristics of the wastes were not measured or evaluated. This aspect
proved to be important because of the distribution of particulate sizes passing through or generated
during the combustion process.
An early difficulty in the operation of the incineration system was an observed inability to feed
sufficient quantities of waste oil through the waste oil burners. This was due to the undersizing of
the motor and pump used for this purpose. Lack of understanding of the viscosity characteristics of
the waste oil led to this event. Even though the waste oil feed was heated to lower its viscosity, a
several factor increase in pump size and drive horse power was needed to achieve satisfactory
operation.
The heating of the waste oil feed to achieve lowered viscosities resulted in another problem, again,
at least partially due to inadequate characterization of the waste oil. The problem manifested itself
as plugging of lines, valves, burner nozzles, etc. This plugging was attributed to polymerizing of
components of the waste oil into highly viscous, adhering materials which would coat surfaces and
plug flow-line components. The lack of recognition of this potential difficulty probably goes back
to inadequate SI and RI sampling techniques for obtaining representative samples of highly volatile
materials such as styrene.
Waste characterization did not identify the process problems that would be caused by the presence
in the waste oil of millions of tiny, floating plastic beads. These beads rapidly clogged filters, valves,
etc., and just as rapidly shut down the waste oil injection system. The procedure for rectifying these
plugging problems was very time-consuming and required disassembling, cleaning out the various
components, and then reassembling the waste oil feed system.
During the pretest leading up to the trial burn for the HTTS-3, it was observed that particulate
loadings in the stack were somewhat above the 0.08 grains/standard cubic foot regulatory standard.
After consideration of the possible causes for this and discussions with the Hydrosonics unit
manufacturer, the site remediation contractor concluded that the size distribution of particulates
generated was the cause of poor performance of the gas cleaning system. There was a much higher
concentration of less than -0.5-micron-sized particles than might be expected. Thus, a so-called
Supersub component, for which space had been provided in the original design, was installed and
implemented. The particulate loadings then decreased into the acceptable range.
1259
-------
POTENTIAL CONCERNS
Concerns at the MOTCO site relate to the potential for human health and environmental risks for
off-site areas and for safety and human health risks for personnel working onsite. Releases of
potentially toxic and hazardous materials can originate from exposed onsite materials or from the
dispersion of emissions and residues resulting from materials processing and incinerator operation.
The potential concerns and the mechanisms for assuring control are summarized in Table 1.
The adherence to documented and approved protocols for site operations will satisfy both off-site and
onsite protection and control purposes. So, the concern then is to assure that the incineration process
itself will provide the destruction of the waste hazardous materials to a high degree of effectiveness
and that any potentially hazardous materials in the emissions are effectively controlled. Such
assurance is the objective of the trial burn.
Table 1. Potential Concerns of Site Emissions
Source
Incinerator Stack
Medium
Incinerator Residues
Site Fugitive Emissions
Gas/Vapor
Particulates
Incinerator Ash
Gas/Vapor
Particulates
Assurance of Control
Incinerator operating conditions and
gas composition
Incinerator operating conditions and
plume opacity
EP Tox or TCLP
Perimeter Monitoring
Perimeter Monitoring
Emissions from the stack are the primary concern. These emissions could have small concentrations
of hazardous materials vapors or have small amounts of particulates. These uncollected particles could
have adsorbed hazardous vapors and contain toxic metals. Thus, the stack sampling, to be discussed
later, has the objective of measuring the quantities of potentially harmful materials emitted to assure
that the emission levels are satisfactory from two standpoints: First, the DREs must be at least as high
as those specified by the regulations for the toxic and hazardous organic materials being treated. And,
second, the particulate emissions must be lower than the limits imposed by air pollution control
regulations.
As a potential portion of the stack emissions, acid gases formed in the combustion process also must
be controlled, as specified by regulations. The acid gases generated in the incineration include HC1,
SO2 and NOX.
Ultimately, the dispersion in the air of all potentially harmful emissions must be sufficient, before
reaching any receptor location, to achieve extremely low concentration levels. These extremely low
concentration levels are those needed to assure negligible risk to humans and the environment. Thus,
the emission values measured and achieved during the trial burn testing and controlled thereafter by
specifying the operating conditions to be those under which they were achieved are those which
provide the assurance of negligible risk.
All other residues from the site operations and incinerator operation ultimately end up in the
incinerator ash, since all other residues will be processed by the incinerator. The incinerator ash is
1260
-------
analyzed by TCLP (Toxicity Characteristic Leaching Procedure) or by methods shown to be
equivalent or representative to show its suitability for disposal back to the site.
Thus, all potential concerns are resolved by: 1) performing the trial burn to establish suitable
operating conditions, 2) assuring negligible risk through dispersion calculations and risk assessment,
and 3) analyzing the ash to show its acceptable nature.
OPERATING CONDITIONS
As indicated previously, the objective of the trial burn tests is to verify the performance of the
incinerator system. That is, the operating parameters demonstrated in the trial burn must be shown
to provide the DREs required and, also, to provide sufficient control of other potentially harmful
materials generated in the combustion (incineration) process. These other potentially harmful
materials are the acid gases and particulate matter.
Time, temperature, and oxygen concentration are the regulatory conditions which are specified and
which must be met to assure adequate DREs. Nominally, these values must be at a minimum: 2
seconds, 2012°F and 3 percent, respectively. In addition, there must be sufficient turbulence in the
combustion chamber to provide the intimate mixing of the combustion gases, thus assuring intimate
molecular level contact of oxidizer and organic species. Continuous verification of the efficacy of
the organic destruction process is provided by the measurement of CO levels in the stack. Low levels,
0 to 10 ppm, indicate a highly efficient combustion process. A limit of 100 ppm (as a 1-hr rolling
average) is used as a cut-off value.
Temperature, oxygen, and CO levels are intensive variables and thus can be directly measured.
Residence time, on the other hand, must be computed from two extensive values, combustion gas flow
rate and combustion volume. Combustion volume is generally taken as that volume where the gas
temperature and oxygen are at the required levels to achieve rapid organics destruction.
The performance achieved in the trial burn of the HTTS-3 unit is summarized in Table 2. These
conditions form the basis for the permissible operating conditions for site materials remediation.
1261
-------
Table 2
Comparison of Regulatory Requirements with Operating Conditions
Trial Burn
Performance Regulatory Operating
Characteristic Requirement Conditions
Combustion Chamber >2012°F 2080,2079,2080°F
Temperature
Total Heat Release 64.6, 68.5, 55.0
x 106 Btu/hr
Pressure Drop Across 42.5, 41.3,
Gas Cleaning System 42.2 in. H2O
Stack gas Flow Rates 43,772, 44,638, 44,451 acfm
Retention Time >2 sec 3.95, 3.87, 3.87 sec
(Based on 2880 cu ft
combustion volume)
CO (maximum) <120 ppm 3, 0, 0 ppm
(Based on 99.9% @12.0% CO2
combustion efficiency)
Oxygen Concentration >3.0% 4.1, 3.9, 3.9%
ACHIEVEMENT OF PERFORMANCE STANDARDS
During the conduct of the trial burn for IT Corporation's HTTS-3 at the MOTCO Site, sampling of
the waste feed streams, sampling of the scrubber influent and effluent, sampling of the incinerator
ash and sampling of the stack gases were performed. Three (3) replicate sampling runs were required
to be conducted for each different process operating condition. An operating condition is defined
as the same waste stream, feed rate, temperature and excess oxygen condition. If any of these are
changed, a new operating condition is defined.
For some incinerators, the owner/operator may want to change one or more of the following: 1)
increase the feed rate, 2) change a waste stream, 3) lower the combustion chamber temperature or 4)
increase the oxygen or combustion air flow to the incinerator. Any one or more of these changes will
constitute a new operating condition, thus requiring a separate set of three (3) replicate
sampling runs. IT Corporation decided to conduct the trial burn under one operating condition. All
three (3) replicate runs were required to meet the performance standards of the RCRA Regulations.
These performance standards are shown in Figures 2, 3 and 4.
1262
-------
Figure 2. Destruction & Removal Efficiency for each POHC
w - w
out
ORE = x 100
where:
Win = mass feed rate of one POHC in the waste stream feeding the incinerator, and
Wout = mass emission rate of the same POHC present in exhaust stack prior to release
to the atmosphere.
The ORE for each POHC must be > 99.99% for RCRA and 99.9999% for TSCA
Figure 3. Particulate Emission Rate
P — P V *4
e~ m 21-Y
where,
Pc = corrected particulate concentration in the stack, gr/dscf,
Pm » measured particulate concentration in the stack, gr/dscf and
Y = measured concentration of oxygen in stack gas, % using the Orsat method.
The Particulate Concentration, Pc must be < 0.08 gr/dscf
Figure 4. Hydrogen Chloride (HC1) Emissions
An incinerator burning chlorinated waste and producing stack emissions of
more than 4 Ib/hour of HC1 must control the HC1 emissions such that the rate
of emission is no greater than the larger of either 4 Ib/hour or 1% of the HC1
in the stack gas measured prior to its entering any air pollution control
equipment.
The HCI removal efficiency of the APC device must be > 99%
Sufficient waste was available in order to be able to complete all three (3) sampling runs for the
specific operating condition approved in the Trial Burn Plan. Since each sampling run took six (6)
hours to complete, only one (1) run was completed each day.
The results of the trial burn which were reported by IT Corporation are presented in Table 3 and
Table 4. It should be noted that the data presented in this paper in Table 3 and Table 4 have not been
validated by EPA, Region 6.
It was agreed upon in the Trial Burn Plan that carbon tetrachloride and trichloroethane would be used
as surrogates for demonstrating the destruction and removal efficiency for PCBs. The DRE required
was 99.9999%. Since naphthalene was a constituent in the waste and it is a solid at room temperature,
it was also selected as a compound to demonstrate the DRE of 99.99%
1263
-------
If ultimately verified, the data shown in Table 3 indicate that the HTTS-3 incinerator met the DRE
requirements for the three organic compounds and met the particulate emission concentration of 0.08
gr/dscf. The gaseous pollutants, CO, O2, NOX and SO2 all met the stated objectives. In addition the
opacity of the plume and the combustion efficiency of the incinerator met the stated objectives.
Table 4 shows the metal removal efficiencies for the spiked metals. The emission rates of dioxins,
furans and PCBs are also shown in Table 4. The removal efficiency of HC1 is shown in Table 5. The
removal efficiency exceeded 99.0% for the HC1 generated in the combustion gases.
CONCLUSIONS
Despite the operational and weather-related difficulties encountered, the trial burn runs necessary
for proving the performance of the HTTS-3 were completed successfully. Currently, the results
presented by the Trial Burn Report are being validated by EPA.
Operational problems encountered might have been mitigated somewhat by better physical and
chemical characterization of the feed materials prior to system design; the characteristics would
include items such as: viscosity, viscosity index, surface tension, and polymerization potential.
Better design of the gas cleaning system might have been achieved if the particle size distribution and
amounts generated during combustion were evaluated earlier.
Perimeter monitoring assures that site emissions are being sufficiently controlled.
REFERENCES
- IT Corporation, May 30, 1990, Trial Burn Plan
- IT Corporation, February 1991, Trial Burn Report for HTTS-3, Report Volumes 1-3
- 40 CFR paragraph 264.343 Performance Standards
1264
-------
TABLE 3 PERFORMANCE SUMMARY FOR TRIAL BURN
PARAMETER
UNITS OBJECTIVE RUN 1 RUN 2
RUN 3
ORE - CARBON
TETRACHLORIDE
DRE -1,1,2 TRI-
CHLOROETHANE
DRE-
NAPHTHALENE
HC1 REMOVAL
EFFICIENCY
HC1 EMISSIONS
PARTICIPATE
MATTER (a)
CARBON
MONOXIDE (a.b.c)
OXIDES OF
NITROGEN
SULFUR
DIOXIDE
VISIBLE
EMISSIONS
COMBUSTION
EFFICIENCY
% > 99.9999 > 99.99993 > 99.99994 > 99.99996
% > 99.9999 > 99.999996 > 99.999997 > 99.999994
> 99.99 > 99.998 > 99.998 > 99.998
>99 > 99.90 > 99.92 > 99.94
Ibs/hr
gr/dscf
ppm
Ibs/hr
% opacity
<4
<0.08
<0.90
0.073
<0.79
0.049
<0.71
0.059
<100
<10.3
Ibs/hr <31.17
<20
>99.9
5.9
0.7
99.99
0
5.2
3.8
10
99.99
5.1
1.3
10
99.99
1265
-------
TABLE 4
PERFORMANCE SUMMARY FOR TRIAL BURN
PARAMETER
METAL REMOVAL
EFFICIENCIES:
As
Be
Cd
Cr
Pb
DIOXIN/FURANS
2,3,7,8 TCDD
TOTAL TCDD
TOTAL PCDD
2,3,7,8 TCDF
TOTAL TCDF
TOTAL PCDF
TOTAL PCB
EMISSIONS
UNITS OBJECTIVE
%
%
%
%
%
ng/m3 < 10
ng/m3 < 10
ng/m3 < 10
ng/m3 < 10
ng/m3 < 10
ng/m3 < 10
Ib/hr
RUN 1
>98.4
>94.8
>90.3
96.7
88.4
<0.2
<0.3
<1.3
<0.2
<2.6
<5.0
< 1.4E-04
RUN 2
>97.7
>94.9
>90.0
96.6
91.8
<0.2
<0.3
<1.3
<0.2
<2.6
<5.0
< 1.4E-04
RUN 3
>96.7
>93.9
>86.7
95.3
84.7
<0.2
<0.4
<1.4
<0.2
<2.7
<5.2
< 1.5E-04
(a) Corrected to seven percent oxygen, dry basis
(b) One hour rolling average.
(c) Dry basis
1266
-------
TABLE 5 HC1 REMOVAL EFFICIENCIES
HC1 Generated
HC1 Emission Rate
Units
g/hr
Ib/hr
g/hr
Ib/hr
RUN1
4.28E*05
943
<409
<0.90
RUN 2
4.66E-K)5
1027
<359
<0.79
RUN 3
5.0IE-K)5
1105
<322
<0.71
HC1 Removal Efficiency percent > 99.90 > 99.92 > 99.94
1267
-------
Construction of Groundwater Extraction Trenches
Gary J. Lang
U. S. Army Corps of Engineers
911th Tactical Airlift Group
Box 193
Building 210, Room 108
Greater Pittsburgh International Airport
Pittsburgh, PA 15231
(412) 269-8134
INTRODUCTION
At the Millcreek Superfund Site, located in Erie County Pennsylvania, the remedial activity was
divided into three (3) phases: I) ground water extraction trenches and collection sumps, II) treatment
plant and pumps and piping necessary to transfer ground water from sumps to plant, and HI) closure
cap and flood retention basin. Each phase was to be performed via separate contract. The contract
mechanism for Phase I of the remedial activity, the ground water extraction trenches and collection
sumps, was a negotiated delivery order under a pre-placed remedial action contract. During
pre-award negotiations with the contractor, a continuous trenching machine was chosen as the means
for installing the extraction trenches in view of the potential cost savings and the attractive safety
aspects from limited confined space entry. The trenching machine selected by the contractor was
capable of excavating the trench, installing piping to a depth of approximately twenty (20) feet, and
backfilling with select granular material all in one operation. This paper addresses the operation of
the trenching machine and the problems/experiences associated with this relatively innovative
trenching technique.
BACKGROUND
The Millcreek site is an 84.5-acre tract of land located in Millcreek Township, Erie County,
Pennsylvania, which is situated in the northwest corner of the state along the southern shore of Lake
Erie. The site is adjacent to a highly developed residential and commercial area within the Township
of Millcreek. The topography is relatively flat, with sparse vegetative growth in the central portion
of the site. A wetland of approximately four acres lies southeast of the site, and the eastern edge of
the site lies within the 100-year floodplain of Marshall's Run, an intermittent stream bordering the
east side of the site. The average fill depth on-site is approximately seven feet, and the depth to
ground water on the site varies from zero to several feet.
The site was once a 75-acre freshwater wetland. Between 1941 and 1981, all but 4 acres were filled
with foundry sand and industrial and municipal waste, including drums of solvents, waste oils,
polyester resins, ink wastes, caustics, paint wastes, slag, construction and demolition debris, including
creosote-treated railroad ties, and municipal refuse.
The Pennsylvania Department of Environmental Resources (PADER) first advised the landfill
operator to cease operations in August 1980. In July 1982, at the request of PADER, five monitoring
wells were installed by the Millcreek Township on the Township's 4-acre parcel of land. A hazard
ranking score was determined after a United States Environmental Protection Agency (USEPA)
Technical Assistance Team performed a site assessment in August 1982. USEPA Region Ill's
Remedial Investigation, completed in 1985, discovered extensive soil, sediment, and surface water
contamination. The major classes of compounds detected included:
1268
-------
— Volatile organic compounds (VOC's) such as vinyl chloride; trichloroethylene; 1,2 -
dichloroethylene (acetylene dichloride); 1,1,1 - trichloroethane (methyl chloroform); 1,2 -
dichloroethane (ethylene dichloride); and 1,1 - dicholorethylene (vinylidene chloride) in the
ground water.
— semi-volatile organic chemicals such as bis-(2-ethylhexyl) phthalate, naphthalene and benzo
(a) pyrene in on-site fill materials;
— Polychlorinated biphenyls (PCB's) in the fill and in some sediment samples, and;
— lead in the fill.
In addition to the contaminants listed, numerous other metals, polynuclear aromatic hydrocarbons
(PNAs) and phthalates were detected in the fill materials.
On May 1, 1986 the USEPA issued a Record of Decision (ROD) which proposed remedial actions for
the site based on the Remedial Investigation/Feasibility Study (RI/FS). In 1989, a pre-design study
was completed in which remedial actions were recommended to:
— prevent the air dispersion and off-site transport of contaminants;
— prevent direct contact with contaminants by humans and wildlife; and
— reduce soil, sediment, surface water and ground water contaminant concentrations to levels
acceptable to the USEPA and the PADER.
The selected remedial actions for the site included:
— consolidation of contaminated soils and sediments under a soil cap;
— site grading/placing a vegetated soil cover over low-level contaminated soils;
— construction of surface water management basins and ditches;
— installation of additional monitoring wells; and
— extraction and treatment of contaminated groun water.
As stated previously in the Introduction, this paper deals exclusively with the installation of the
system that was designed to extract the contaminated ground water. The groundwater extraction
system was designed to remove contaminated ground water down gradient of the site contamination.
The extraction of contaminated ground water would prevent continued off-site migration of the
contaminants and would possibly capture some contaminants already down gradient of the site.
Extensive ground water modeling was performed during the Remedial Clean-up Treatability Study
to simulate steady state flow through the shallow water-bearing zone beneath the
Millcreek Site and to model movement of contaminants in this ground water. The Prickett-Lonnquist
Aquifer Simulation Model (PLASM), (Prickett and Lonnquist, 1971) was used to simulate
two-dimensional flow of ground water collection alternatives. A second model, the RANDOM
WALK Mass Transport Model (Prickett, et al., 1981), was used to simulate horizontal movement of
the contaminant plume on the site.
1269
-------
The groundwater modeling provided data on the required location of the collection trenches and the
volume of water to be removed from each of the trenches. From the modeling results, it was
determined five (5) trenches would be required and the optimum location would be the northeast
corner of the site. The results also indicated extracted flows of 8016, 2759, 14067, 16936 and .'(8621
gpd for Trenches 1 through 5 respectively, for a total volume of 80,339 gpd. With the exception of
Trench 3, water levels in the trenches were lowered to an elevation of 700.00 feet, which was
approximately fourteen (14) feet below existing grade. The water level in Trench 3 was lowered to
699.00 feet. The trench system was designed to extract flows at rates above these levels, and the
hydraulics of the systems are more than adequate to accommodate this intent.
Prior to selection of the collection trenches, several alternatives were considered during the initial
design stage. A site-wide network of extraction/recharge wells was eliminated from consideration
as a remedial alternative due to the low potential yield of the contaminated aquifer. Modeling
indicated that the pumping of individual wells at a rate of 24 gpm as listed in the ROD would result
in required differential heads in excess of 80 feet. Field investigations substantiated this with data
indicating that only low sustained yields (less than 5 gpm per well in recent field tests) could be
produced from individual on-site wells.
A series of well-point systems was also considered, but this alternative was abandoned in light of the
superior long term reliability of the collection trenches. Major factors that contributed to the
selection of the trenches over the well-point system were: 1) the trench system provides a continuous
capture over the length of each trench, 2) extraction velocities from the trenches are significantly
lower, thereby reducing the potential for siltation, and 3) less mechanical equipment is required for
the trench system, thereby reducing maintenance costs and downtime resulting from mechanical
failures.
DISCUSSION
DESIGN CONSIDERATIONS
The collection trenches were designed so that contaminated ground water could be extracted and
treated rather than migrate from the site. Trenches 1, 3 and 5 located along the northern edge of the
site were to be installed to the top of the underlying glacial till layers at depths of approximately 24,
26 and 24 feet respectively. Trenches 2 and 4 along the eastern edge of the site were to be installed
to the interface between the coarse and fine sediments at depths of approximately 22 and 20 feet
respectively.
Each collection trench, as originally designed, consisted of the following items: a 200-foot section
of 6-inch diameter slotted polyvinyl chloride (PVC) Schedule 80 pipe; a solid PVC Schedule
80 clean-out section, which added approximately 30 feet of additional piping, including a flushing
riser; a 4-foot diameter precast concrete collection sump for future installation of duplex submersible
pumps (under Phase II of the project); a shut-off plug valve with valve box; a piezometer
located approximately midway between the flushing riser and the collection sump; and a two -stage
granular filter pack in the trench.
The pipe diameter of 6 inches was specified to facilitate periodic cleaning of the system. Although
a 4-inch diameter line could have been specified since it could be cleaned by standard sewer flushing
jets, the 6-inch line provided extra assurance against any flushing problems.
Schedule 80 slotted PVC pipe was specified because it has excellent chemical resistance, it is cspable
of withstanding the loading at the depth of trench required, and because of the variation of slot sizes
available, it could be used in conjunction with the two-stage granular filter pack to provide excellent
drainage capacity. The width of the slots was specified to be 0.020 inches.
1270
-------
The two-stage granular filter pack was designed to work in conjunction with the slotted pipe with
no requirement for a filter fabric to control silting. This would eliminate the chance of clogging of
the filter fabric. The gradation of the primary sand pack around the pipe was specified to provide
a granular material that would not contaminate the pipe, i.e., it would have particles large enough to
be contained outside the pipe and not slip through the 0.020-inch slots in the pipe.
The specified gradation for the primary sand pack was as follows:
Percentage by Weight Passing
Sieve Designation Square-Mesh Sieves
No. 10 100
No. 40 0-5
The specified gradation of the secondary sand pack was based on the existing soil conditions at the
site. The specified gradation for the secondary sand pack was as follows:
Percentage by Weight Passing
Sieve Description Square-Mesh Sieves
No. 4 98-100
No. 10 75-90
No. 20 40-60
No. 40 12-40
No. 60 0-20
Top-of-cover elevations for monitoring wells and flushing risers, and top-of-slab elevations for
collections sumps were based on existing grade.
CONVENTIONAL METHOD VS. TRENCHING MACHINE
Prior to award of the contract, the selected contractor was requested to provide bid proposals on two
types of methods of installation, the conventional method and a method utilizing a trenching machine.
The conventional method employs tight sheeting and dewatering, with confined space entry required
to work in the trench. The trenching machine incorporates trench excavation, pipe insertion and
placing of select granular fill material in the same operation. The apparent advantages and
disadvantages of each method are described below:
Conventional Method - provides for clear controlled inspection of backfill procedures during
construction and thus provides a more consistent final product. This controlled inspection also
provides the necessary data for a conventional quality assurance/quality control program. The method
does, however, require major excavation, tight sheeting and dewatering and the employment of
confined space entry techniques whenever workers are inside the excavation. The proposed cost for
this method was $3.62 million.
Trenching Machine Method - does not require a sheeted trench. With the exception of the sump
installation, all work can be performed on the surface, thereby minimizing the safety hazards
associated with confined space entry. It also eliminates the need for large excavations, which, in turn,
should result in a time savings. However, this method does not provide the opportunity for visual
inspection of the backfill and therefore, there is the potential of undetected bridging of the backfill
1271
-------
material leading to a gap in the filter pack and, ultimately, a silting problem. The proposed cost for
this method was $2.46 million.
In light of the potential cost savings, the minimization of safety hazards and the push for innovative
technologies, the trenching method was selected. Inherent with this method were several design
changes, as listed below:
1. Change in material composition of the pipe from slotted Schedule 80 PVC to slotted
corrugated high-density polyethylene drainage tubing conforming to AASHTO M252 with a geotextile
filter sock. The diameter of the tubing remained as originally designed, i.e., 6 inches.
2. Change in granular filter pack from two-stage to single stage. Single stage sand pack was of
the same gradation of that specified for the secondary sand pack of the two-stage sand pack.
3. Change in trench width from 30 inches to 14 inches.
CONSTRUCTION SEQUENCE
The initial phase in constructing the trenches, as is the case in almost any construction operation, was
clearing and grubbing. All refuse, except salvageable timber, was chipped and stockpiled at a
designated area on site. This stockpile area was secured by a permanent chain-link fence installed
by the contractor. Salvageable timber was stored separately on site for inspection by the property
owner at a later date. The areas cleared for installation of each trench were approximately 75 feet
wide and 350 feet long.
Control points for each trench were installed at the outermost edges of the cleared areas. The cleared
trench areas were secured by erecting a snow fence around the perimeters of each area. Silt fence was
installed along the northern and eastern boundaries of the site to control sediment runoff from the
cleared areas.
The contractor elected to construct one complete trench at a time, although the specifications allowed
concurrent construction. The sequence described in the following paragraphs applies to the
construction of one complete trench.
The next step was the installation of the concrete collection sump, which required dewatering and
shoring to depths of approximately 25 feet. Localized dewatering at the first sump area was
attempted by installing a shallow trench upgradient using the trenching machine. This dewatering
approach proved ineffective, and after several modifications, the upgradient trench was abandoned
in favor of a well-point system. The ground water removed through the well-point system was
pumped to a holding tank and then transferred via another pumping system to a ground water disposal
area designated by the Government and permanently secured via a chain-link fence.
The top elevation of each sump was established at 714.5 feet mean sea level (MSL) to permit
consistent grading throughout the entire site. The existing ground surrounding the collection sumps
and trenches was eventually graded to an elevation of 714 feet MSL.
The excavation for the collection sump was accomplished using a track-mounted excavator. The
excavated soil, since it was considered contaminated, was placed in dump trucks or front-end loaders
and was transported to the designated excess soil storage area. This area was within the chain-link
fence enclosure for the chipped debris stockpile previously discussed.
1272
-------
A standard Occupational Health and Safety Administration (OSHA) approved trench shield was used
to shore the collection sump excavation. The steel trench shield was similar to that used for utility
trenches. The dimensions of the trench shield were 24 feet long by 6.5 feet wide by 10 feet high.
Once the trench shield was in place inside the excavation, 1-inch thick by 24-feet high by 10-feet
wide steel plates were driven down the outside of the trench shield to below grade, using a vibratory
hammer. These steel plates were braced to secure the excavation.
Prior to placement of the precast concrete sections of the collection sump, a layer of crushed stone
bedding, 12 inches thick, was placed in the excavation to support the basin. Each section of the sump
was placed in the excavation using a track-mounted excavator. The top section of the sump contained
pipe sleeves for electrical service and piping to be installed under Phase II of the project. Material
was then partially backfilled around the lowest section of the sump in preparation for placement of
the plug valve and piping accessories.
Upon completion of the collection sump and prior to placement of any additional backfill, a
preassembled unit consisting of a five-foot section of solid Schedule 80 PVC pipe, a six-inch plug
valve, and a one-foot section of solid Schedule 80 PVC pipe was lowered into the excavation. The
free end of the longer section of pipe was attached to the collection sump using a gasketed flexible
coupling similar to that used in sanitary sewer construction. The shorter section of pipe was for
connecting the slotted polyethylene drainage tubing to the valve assembly. The valve itself rested on
a three-foot square concrete pad. A valve stem was attached to the valve and extended to the ground
surface to allow for operation of the valve. Eventually, when the granular material was placed around
the valve assembly, a valve box was installed to protect the valve. The backfilling around the sump
and valve assembly was not performed until the drainage tubing had been connected to the valve
assembly and the trenching machine had placed enough tubing to eliminate the possibility of conflicts
between the backfilling operation and the trenching operation.
After installation of the valve assembly, a shallow bench was excavated along the entire length of the
trench to accommodate the maximum digging depth of the trenching machine. Since the trenching
machine could dig to a depth of approximately 20 feet, and the trenches were as deep as 26 feet, the
benches were necessary to compensate for the difference in depth. The benches were approximately
16 feet wide to accommodate the width of the trenching machine.
The trenching machine used at the Millcreek site was a 1984 Steenbergen/Hollanddrain Trencher,
Model BSY-Super-S-375. It had a 375 horsepower engine and was capable of digging a trench up to
36 inches wide and as deep as 20 feet plus. In 1984, the machine, without extras, cost approximately
$570,000.00.
After excavation of the shallow bench, the trenching machine was positioned along the trench line
and the drainage tubing was snaked through the top of a boot attached to the digging mechanism.
The tubing exited out the bottom of the boot and the leading end was connected to the short section
of PVC of the valve assembly, using a watertight, flexible rubber coupling. The excavation of the
trench and the placement of the drainage tubing was now ready to begin.
The trenching machine excavated the trench to the required depth and grade, laid the tubing at the
specified depths and evenly distributed the select granular material around the tubing all in one
operation. Both the tubing and granular material were fed through the boot attached to the digging
mechanism. The tubing was fed from a large spool at the rear of the trenching machine. The digging
mechanism was similar to that used on conventional trenching machines, but larger and more
powerful.
1273
-------
The digging mechanism could be disconnected from the boot whenever necessary to reposition the
machine or to remove obstacles. The tubing was fed through a large conduit in the center of the boot,
which separated the tubing from the granular material while inside the boot. The conduit was curved
at the bottom of the boot to facilitate laying the pipe on a horizontal plane. A dual laser guidance
system was employed to insure accurate depths and to maintain uniform slopes to within 15/100 of
a foot. The granular material was placed in the hopper portion of the boot, i.e., that portion outside
of the conduit through which the tubing was fed. Loading of the hopper was accomplished with
front-end loaders or excavators. The granular material was gravity placed from the boot and was
distributed under, around and over the tubing.
Material excavated by the trenching machine was deposited alongside the trench. This material was
removed daily with the use of a front-end loader and was transported to the designated excess soil
disposal area on site.
Note: As the trenching machine was excavating and placing the first trench, it became obvious the
further away the machine moved from the dewatered sump area, the more difficult it was for the
machine to excavate and place the tubing. Finally, the tubing broke, and it was decided the same type
of dewatering performed at the sump area had to be performed along the entire trench line to permit
operation of the trench machine as intended. Therefore, a well-point system was installed upgradient
of the trench that ran the entire length of the trench. After installation of the well-point system and
the subsequent dewatering, the trenching machine worked much better and was able to excavate and
place all five trenches. See Problems/Analysis for more discussion relative to dewatering.
Near the completion of the trench excavation, i.e., at the end of the trench where the flushing riser
was to be installed, the tubing was curved upwards at a gradual rise to avoid a 45-degree connection,
or elbow, which could not be accommodated by the trenching machine. The tubing was cut arid the
trenching machine was driven away from the trench area. The area where the tubing curved upward
was excavated using an excavator/backhoe to expose the tubing, and the trench shield that was
previously used to install the collection sump was placed around the tubing. A section of solid
high-density polyethylene pipe was attached to the tubing using a flexible coupling. This section of
solid pipe acted as both the lower portion of the flushing riser and as a transition between the flexible
tubing and the section of solid Schedule 80 PVC pipe that was the final section of the flushing riser.
The PVC pipe was connected to the solid high density polyethylene pipe with a flexible coupling also.
Once all the connections were made, the trench shield was removed and backfilling around the
flushing riser was performed.
At this point, additional select granular material was backfilled into the open trench to bring the top
elevation of granular material to approximately 31/2 feet below grade. A layer of filter cloth was
then placed on top of the granular material to filter out sediments and provide structural support for
the clay backfill that was specified to be placed on top of the granular material. The clay material
was then placed on the filter cloth in 8-inch lifts and was compacted with a dozer. Final thickness
of the clay material was 30 inches. Concurrent with this operation was the backfilling of the shallow
bench excavation. Once the backfilling of the clay material and the bench excavation was complete,
the site was graded to facilitate proper drainage. A drilling crew then installed the piezometer
approximately midway between the collection sump and the flushing riser, taking soil samples to
insure the piezometer was within the confines of the trench. Finally, six inches of topsoil was placed
on top of the disturbed areas, and these areas were seeded, fertilized and mulched.
PROBLEMS/ANALYSIS
A major problem associated with the use of the trenching machine at the Millcreek site was the
machine's inability to trench through in-situ soil without requiring the entire length of trench to be
1274
-------
dewatered to a depth equal to or greater than the bottom elevation of the trench. Visual
classifications of the material at the trenches ranged from very loose gravel, sand and silt to medium
dense gravel, sand and silt to very dense gravel, sand and silt. Blow counts experienced during test
borings ranged from 1/6 inches to 50/2 inches, with the overwhelming majority less than 15/6 inches.
The borings also indicated the site was nearly saturated just below the surface.
During the course of construction, another contractor that specializes in placing trenches using a
trenching machine was contacted. The contractor's representative stated there had been cases in the
past when the trenching machine could not place the tubing without dewatering. The frequency of
this occurrence, though, was less than 1% of all projects. No definite reason for the trenching
machine's failure to perform was provided.
After considerable analysis of the experiences at the Millcreek site, the most logical reasoning behind
the trenching machine's failure to perform without dewatering was the excessive hydrostatic pressure
created by the high water table and the mixture of in-situ silty materials. The mixture of soils and
ground water created enough pressure at the bottom of the trenching machine boot that it pinched
the tubing against the side of the curved section of conduit and prohibited the tubing from being
placed without excessive resistance. This same hydrostatic pressure also displaced the granular
material intended to encompass the tubing, thereby contaminating the sand filter pack.
A well-point system installed along the entire length of trench on the upgradient side eliminated the
hydrostatic pressure problem and did permit the installation of the trenches as intended via the
trenching machine. Yet the well-point solution negated one of the supposed benefits of the trenching
machine, i.e., the installation of a subdrainage system without the need for dewatering.
Another difficulty encountered during the installation of the trenches was "untrenchable" material.
Untrenchable material was defined as material that could not be excavated with the trenching
machine. During negotiations, it was agreed the contractor would not be liable for costs associated
with removing untrenchable material, and that any untrenchable material would be considered a
differing site condition and a modification to the contract would be executed to compensate the
contractor. Through the course of construction, untrenchable material was encountered in four of
the five trenches. The untrenchable material was glacial till that was at a higher elevation than what
was expected from interpretations of the boring logs. Since the intent of the design was for the
trenches to be constructed just above the glacial till, the bottom elevations of the trenches were raised
just enough to clear the glacial till.
Once it became evident during the initial trenching operations that the glacial till was at elevations
that were higher than anticipated, the Government directed the contractor to drill test borings along
the projected locations of the trenches that had yet to be excavated to pinpoint, if possible, the top
elevations of the glacial till. This approach proved invaluable in that it did accurately locate the
glacial till, and it enabled the contractor to adjust the trenching machine depth to avoid the
untrenchable material, thereby eliminating potential impacts and delay costs.
Due to either the untrenchable material or the excessive hydrostatic pressure at the bottom of the
trenching machine, there were several instances when the tubing was crushed, stretched or broken.
This occurred on one occasion even after the extensive well-point dewatering system was installed.
The operators of the trenching machine knew the tubing was damaged on the basis of the reaction
of the tubing and trenching machine itself.
When the tubing became damaged, the contractor had to implement a construction procedure similar
to that employed in installing the flushing riser. This procedure included excavating with a
conventional excavator/backhoe to expose the damaged tubing, installing the trench shield, cutting
1275
-------
away the damaged tubing, connecting the undamaged tubing to the leading end of the tubing that was
protruding from the bottom of the trenching machine boot, using a flexible coupling, and
concurrently removing the trench shield while backfilling with select granular material.
Obviously, this procedure was time-consuming, hazardous, and costly, and it was in the best interest
of all parties to avoid, as much as possible, creating situations that could exacerbate the damage to
the tubing. This rationale was the basis for the Government's directive to drill test borings along the
trench lines in an attempt to ascertain the exact locations of the till material.
After construction of the trenches was completed, a series of pump tests were conducted and another
problem surfaced. Several of the trenches exhibited an abnormally high hydraulic gradient between
the collection sump and the piezometers. The contractor was directed to redrill some of the
piezometers to insure they were within the confines of the trenches. Redrilling and the associated
soil sampling indicated the original piezometers were located within the trenches, but the trenches
themselves were partially contaminated with in-situ materials. One theory on how this siltation
occurred is that during backfilling through the trenching machine boot, the discharged granular
material, since it was discharged solely through the force of gravity, began bridging and created gaps
which were filled by in-situ materials once dewatering was discontinued. A theoretical h solution to
this problem is to attach an external vibrator to the trenching machine boot which would consolidate
the granular material enough to minimize or eliminate any bridging within the backfill. This
approach was not used on this site and it is not known whether this would effectively eliminate the
bridging problem.
Another problem associated with the trench system was the valve stem. During backfilling operations,
one of the valve stems was dislodged from its seat on the plug valve which rendered the valve
inoperable. This unfortunate occurrence will eventually result in some repair and/or replacement
work, but the extent is unknown at this time because the contractor is currently seeking approval to
abandon the plug valve and install a knife-gate valve within the collection sump. In hindsight, a
separate manhole for the plug valve or a manhole large enough to accommodate the plug valve and
the future duplex submersible pumps would have eliminated this problem and would have provided
a means of accessing the valve for future maintenance or replacement.
CONCLUSION
Under compatible subsurface conditions, ground water extraction trenches can be installed more
safely and cost effectively by using a continuous trenching machine in lieu of a conventional
trenching method. The key issue is the compatibility of the subsurface conditions. It is imperative
that the designer conduct a thorough investigation and analysis of the subsurface conditions before
specifying the trenching machine as the method for installing collection trenches. Several
recommendations for owners/designers contemplating the use of a trenching machine are listed below:
1. Drill test borings along the entire length of trench to determine whether any of the in-situ
material within the trench line is untrenchable, i.e., too dense to be excavated by the trenching
machine.
2. Analyze the drill logs to ascertain whether dewatering of the site is required prior to
trenching. The experience at the Millcreek site shows that a site containing intermixed sands, silts
and gravels of varying densities, as opposed to a site with more uniform materials, may not be
conducive to use of the trenching machine without extensive dewatering. However, even with
extensive analysis, it may not be possible to determine whether or not the trenching machine could
work without dewatering. The only true measure would be to conduct a pilot test, using the trenching
machine at the site. The cost of this approach may discourage owners from selecting this method of
1276
-------
trench construction, but the potential cost savings associated with the trenching machine could justify
the additional design costs. Also, the contract could be worded to place some of the risk on the
contractor by making it the contractor's responsibility for dewatering the site, regardless of the
method used. Furthermore, even if extensive dewatering is required, the potential cost savings and
reduction in safety hazards achieved by eliminating the need for massive excavations, sheeting and
confined space entry techniques may still justify the use of the trenching machine.
3. Specify means of insuring consolidation of the granular filter pack material to minimize or
eliminate siltation within the trench. The trench width is extremely narrow (14") and it is imperative
that the trench backfill be kept as clean as possible, since there is little room for error. The use of
external vibrators is a possibility, as well as specifying drilling of test borings through the trench as
soon as portions of the trench are placed, and prior to discontinuing dewatering, if it is required.
This approach may provide the on-site construction managers with some assurance that no bridging
has occurred, and in the event it has, it allows the contractor a chance to correct any deficiencies prior
to final backfilling of the trench.
4. Allow sufficient time between contracts in the event the remedial activity is broken down into
separate phased contracts. With the trenching machine method, there is no opportunity for visual
inspection of the backfill and drainage tubing, and, therefore, there is the potential for extensive
corrective construction in the event portions of the trench are found to be deficient. Specifying
operating tests/inspections such as dye tests, in-line video surveillance, etc. during construction may
minimize impacts and conflicts with follow-on contractors since the deficiencies, if any, could be
positively identified while the trench construction contractor is still on-site.
REFERENCES
1. Final Design Analysis, Ground Water Extraction System, Millcreek Superfund Site, Erie
County, Pennsylvania, Malcolm-Pirnie, Inc., Buffalo, New York, July 1989.
2. The Merck Index, llth Edition, Merck & Company, Inc., Rahway, New Jersey, 1989.
3. Cap and Flood Retention Basin Design, Safety Health and Emergency Response Plan,
Millcreek Superfund Site, Erie County, Pennsylvania, Malcolm-Pirnie, Inc., Buffalo, New
York, January 1991.
4. Exploratory Soil Borings Investigations, Millcreek Superfund Site, Erie County, Pennsylvania,
Malcolm-Pirnie, Inc., Buffalo, New York, August 1989.
5. Cap Construction and Flood Retention Basin, 35% Submittal Design Analysis Report,
Millcreek Superfund Site, Erie County, Pennsylvania, Malcolm-Pirnie, Inc., Buffalo, New
York, March 1991.
6. Trench Construction Plan, Millcreek Superfund Site, Erie County, Pennsylvania, IT
Corporation, Monroeville, Pennsylvania, February 1990.
7. Letter dated March 5, 1991 from Ground Water Control, Inc. of Jacksonville, Florida to IT
Corporation of Monroeville, Pennsylvania.
8. Brochure entitled Ground Water Control Environmental Services, Ground Water Control, Inc.,
Jacksonville, Florida.
1277
-------
11
1 I I I I ,
'fill ,!l
«
r
} H1 i I it!! 111 i 1'I .i IS li
-------
I
UJ CC CL
Ci O UI
i- cn Q
Q KH UJ
LU LU CC Z
cn CL ~
M M O <£
> CL Z r
LU w LU
CC U X CC
> cn
LU CL D CL
K E U. D
o cn
Z CC UJ
« U. X LU
(- I
LU Z I-
Cl O U.
-i LU
Z CL t- >
c r u _i
x o u
>- _i _i o
z " u _i
O CC t-i 0.
LU |-
LU I- CC LU
X
-------
n
•
O
X
o
(-
LJ
i£
in
fc
1
1
o
O
I
-c
•*«»
o
0
1
tJ
I
.6
o
15 fe
*
-------
o
LU
X.
(fl
Id
Q
O
Q
Q
UJ
m
X
o
z
UJ
cc
y
<
o
to
h-
o
c/)
UJ
o
o
UJ
in
LJ
ca
1281
-------
u
UJ
UJ
Q
Q.
GO
o
UJ
_j
_j
O
o
d)
128?
-------
1283
-------
o
I
u
(-
LU
Y
cn
sr
u
i
u
z
UJ
o:
x
o
13
O
o:
u
til
03
1281
-------
EUROPEAN SOIL WASHING PQR IJ^S* APPLICATIONS
Michael J. Mann, P.E.
Geraghty & Miller, Inc.
14497 North Dale Mabry Hwy
Suite 115
Tampa, FL 33618
(813) 961-1921
The purpose of this paper is to present the details of the
introduction of a new soil treatment technology to the U.S. market.
For the purposes of this presentation, I would like to introduce a
concept of three tiers of contaminated soil treatment; traditional
treatment technologies, alternative treatment technologies, and
emerging treatment technologies. Traditional treatment consists of
landfilling, incineration, and stabilization. Alternative
technologies consist of low- temperature thermal treatment,
bioremediation, vapor extraction, and physical screening and
separation to achieve volume reduction. . .the essence of soil
washing. Emerging technologies currently include in-situ
vitrification, RF processes, dechlorination, and possibly some
extraction techniques. This paper focuses on the alternative
technologies. One of the most important lessons we have learned
over the past decade is that no single technology provides a broad
enough capability to solve all the soil situations that we
encounter - - the key to feasible and cost-effective site solutions
is the ability to optimize the use of reasonable alternatives in a
site-specific matrix of use.
The EPA has recognized this need and particularly with SARA,
emphasized the importance of "on-site" treatment technologies.
This policy was initially stimulated through the development of the
SITES program and most recently expanded by the formation of the
Technology Innovation Office (TIO) .
Still, all technologies have their limitations. The
limitations that are most commonly encountered are:
The volume of soil is too big or too small.
The contaminants species and/or concentration
is not process compatible.
Organics and inorganics cannot be handled in
the same treatment train.
The process has little or no commercial
operations experience.
128
5
-------
This presentation is intended to provide a description of a
commercial soil-washing facility operating in Holland for the past
seven years and to demonstrate how many limitations can be overcome
with this system.
BACKGROUND
About the same time as the EPA began an active review of
European technologies, Geraghty & Miller spent about one year
evaluating various soil treatment facilities operating in The
Netherlands, Germany, France, Italy, and the U.K. This search led
us to meet the operational group of Heidemij, headquartered in
Arnhem, The Netherlands. Heidemij is an environmental consulting,
management, and remediation firm over 100 years old and the market
leader in The Netherlands in soil washing and bioremediation.
Heidemij has strong research roots in the Dutch university system
and has applied that resource to real field implementation.
Heidemij currently operates bench scale, pilot, and commercial
soil-washing facilities in Holland, and last year treated more than
150,000 tons of contaminated soil. The USEPA has visited the
Heidemij facilities on many occasions and has prepared papers
providing technology comparisons.
This background led Geraghty & Miller to establish a Joint
Venture with Heidemij and, together, we are actively marketing the
capability, conducting treatability studies, and performing in-
field trials.
;D2SC0SSION
The objective of the Geraghty & Miller Joint Venture is to
contract, own, and operate mobile treatment equipment to manage
contaminated soils with a wide range of soil properties and
contaminant types. The first venture process offered in the U.S.
is soil washing. Soil washing provides a practical method whereby
the entire soil volume can be understood to separate clean
materials from contaminated fractions, and then to direct
appropriate treatment at the contaminated portion. The process
depends on the ability to effect substantial volume reductions and
then to place "clean" soil back on site or to effect beneficial
reuse in construction grade materials meeting applicable
specifications.
1286
-------
PROCESS DESCRIPTION
PARTICLE SIZE/CONTAMINANT RELATIONSHIP
The Heidemij Soil Wash Process is based upon the fact that a
discrete relationship exists between soil particle size and
contaminant residence. The nature of this phenomenon is a result
of many factors, including the manner in which the waste was
disposed, the site soil matrix, the specific contaminants, the soil
cation exchange capacity, particle zeta potential, and dynamic
stresses placed on the materials at the site. The first step in
evaluating the potential application of soil washing at a
particular site is to quantify this particle size/contaminant
relationship. It is not necessary to understand all the geochemical
forces on the material, but simply to perform a standard sieving
analysis and to analyze target fractions. Generally, remedial site
soils will exist in five primary "fractions":
Gross Oversize. This material is >8" and
consists of concrete rubble, tree stumps and
branches, scrap steel, and tires.
Oversize. Material in this fraction is
>2"(500mm) but <8". This fraction will consist
of gravel, cobbles, shredded wood, and
slags.
Large, Coarse-Grained Soils. This material is
in the range of 1/4" to 2" and is composed of
sands and gravels.
Coarse-Grained Soils. This material resides
in the range of 40-60 microns up to 1/4" and
is sand.
Fine-Grained Materials. Clays and silts with
an average particle size of less than 40-60
microns.
Once these particle size fractions have been identified and
quantified, the "percent finer" particle size distribution curve is
constructed. Each resulting target fraction is then analyzed
chemically for appropriate contaminants. The selection of the
analytical menu, of course, will be dependent upon existing
information, the history of the site, and understanding of the
contaminants of concern. The worst case, where no information
exists, will require a full quantitation of each of the particle-
size fractions. This analytical work does not need to be conducted
with the extensive QA/QC that we have grown used to on
investigation projects. Level III data (in accordance with the
EPA's Draft Treatability Study Guidance Document) is acceptable at
this point. The data is reviewed and then an overlay of the data on
1287
-------
the particle-size distribution curve is prepared. The understanding
of this step is the real key to soil washing, for in most cases;, at
least one of the fractions will not be contaminated. The challenge
and capability of the soil wash system is to separate the
uncontaminated fraction(s), and then to direct appropriate
treatment at the contaminated fractions.
PROCESS OVERVIEW
The process is constructed completely of standard, proven
equipment, most of which has been used for decades in the mining
business. The waste pile is excavated and a working pile is
created. The Gross Oversize and Oversize fractions are separated
individually using mechanical screening techniques, while the
coarse and fine-grained split is obtained with the creative use of
hydrocyclones. If required, the coarse-grained materials (the
sands and gravels) are treated by froth-flotation techniques. The
fine-grained materials are more difficult to treat and will be
handled by dewatering, biological, or extraction processes.
The basic soil-wash treatment plant is modular, and easily
transportable. The plant is extremely flexible and can be
configured to handle a very wide range of needs from simple volume
reduction to sophisticated treatment trains. The "basic" plant has
a throughput capacity of 20 tons per hour (tph) and in a full
treatment mode requires about 1.5 acres of laydown space. On a
typical site, the facility area will be graded, a liner placed on
the plant area, and run-on and run-off controls provided. The
plant does not require any special foundation or support work. All
equipment is on engineered skids with quick disconnects and
flexible hosing connections as a basic design feature. If the
remedial site is extremely remote, and roads need to be built into
the area of contamination, then that clearly expands the scope of
the mobilization activities. The plant's primary utility
requirements are water and electrical power. Water is completely
recycled in the system, and therefore no discharge is required, but
make-up water at the rate of approximately 25 gallons per minute
(gpm) is necessary. The 20 tph plant has approximately 1,000
connected horsepower and can operate from an organic mobile
generator if commercial 440, 3-phase power is not available.
The soil wash system can be used on a very wide range of
contaminant species, including heavy metals, semi-volatile
organics, including PCBs and pesticides. If volatile organics are
included in the waste stream, the material will either be pre-
treated by removing the VOCs with a thermal screw, or the entire
system may be operated in an enclosed working space with complete
air emissions control.
1288
-------
Remember, the plant consists of four major sub-systems:
Screening
Separation
Froth Flotation
Sludge Management
A schematic diagram of the plant is attached as Figure 1. Also
remember that the plant will be generating three residual products
that will be managed:
1. Oversize and Gross Oversize material (usually clean)
2. Clean sand (to be beneficially reused)
3. A sludge cake to be appropriately disposed at a permitted
TSD (this is where the contaminants finally reside)
THE SECRET IS TO RECYCLE THE OVERSIZE, REUSE THE CLEAN
SOIL, AND TO KEEP THE SLUDGE CAKE VOLUME AS SMALL AS
POSSIBLE.
Each of the sub-systems will now be explained.
Screening
As mentioned above, a working pile is excavated in the field.
This working pile must first be screened to remove the Gross
Oversize fraction. This will normally be accomplished using a
hopper mounted with a vibrating Grizzly. If annoying hopper
blockage results, it may be necessary to substitute a Kombi screen
or Trommel screen to provide a more uninterrupted step. Gross
Oversize material is periodically removed from the hopper area and
staged for recycling. The "fall through", or the material that is
now <8", is conveyed to the next mechanical screening unit, which
will generally consist of a double decked vibrated screen with
stacking conveyors. The double-decked screen will have two flow
paths: l)an oversize material that is >2" and, 2) a fall-through
that is directed by conveyor to the wet-screening unit.
Wet screening is applied to the stream of soil <2". High-
pressure water nozzles attack the influent stream, breaking up
small clods, dropping out pea-sized gravel, and forming the slurry
that is now pumped to the Separation Sub-system.
Separation
The heart of the Heidemij soil wash system, and the area where
extensive experience has been developed, is the creative use of
hydrocyclones. Conceptually, the use of hydrocyclones is simple:
the influent soil/water slurry is pumped to the cyclone and the
slurry enters tangentially. In the cyclone, open to atmospheric
1289
-------
pressure, the coarse-grained sands are spun out of the bottom,
while the fine-grained materials and water are ejected from the top
of the unit.
Several details need to be pointed out regarding the special
use of the hydrocyclones in this system. First, the cyclones have
field-adjustable cone and barrel components such that the "cut-
point" interface between coarse and fine-grained materials can be
modified consistent with treatment needs. This is extremely
important in achieving the smallest volume of sludge cake requiring
off-site disposal. Secondly, the hydrocyclones can be arranged in
many flow-path configurations depending upon the interface needs
and the goal of minimizing coarse-grained carryover into the fines.
Depending upon the soil to be treated, it may also be
beneficial to utilize gravity separators on either or both of the
coarse/fine fractions. Typical applications might include the
removal of a floating organic layer or, at the other end of the
density spectrum, dropping lead out from the soil-treatment stream.
Coarse Fraction Treatment
The underflow from the hydrocyclones contains the coarse-
grained materials. When treatment is required for this fraction,
it is accomplished using proven air flotation treatment units;.
The first important decision that must be made in this sub-
system is the selection of a surfactant. The selection, made from
scores of alternatives, has one objective: the surfactant, when
contacted properly with the contaminant/soil mass, reduces the
surface tension binding the contaminant to the sand and allows the
contaminants to "float" into a healthy froth which is then removed
from the surface of the air-flotation tank. The selection of the
appropriate surfactant is made during the treatability study at the
bench-scale level.
The air-flotation tank is a long, rectangular tank that is
mixed with the use of mechanical aerators and diffused air.
Retention time is typically about 30 minutes, but can be adjxasted
on the treatment unit.
The flotation units require operator experience to obtain
optimal performance. Primary control parameters are surfactant
dosing, slurry flow rate, air flow rate, and the height of the
overflow weir.
Two streams, the overflow froth, and the underflow sand, are
the effluents from the treatment unit. The froth is concentrated
and usually directed to the sludge management belt filter press
where it is dewatered into a 50-60% solids cake. If, however, the
contaminants from the coarse and fine-grained fractions are not
compatible, then it may not be wise to send the froth to the filter
1290
-------
press, but to manage it separately. The underflow from the
flotation unit (the sand) is now directed to sand dewatering
screens - the dry sand represents the "clean" material that will be
reused, the water is recycled back to the wet screening section.
Sludge Management
The overflow from the hydrocyclone, consisting of fine-grained
materials and water is now pumped to the sludge management sub-
system. As mentioned earlier, the fines represent the most
difficult fraction to treat, as a result of complex binding and
attachment dynamics and mechanisms. If the distribution of fines to
coarse is favorable, it is feasible to simply treat the fines
similar to a wastewater sludge by polymer addition, sedimentation,
thickening, and dewatering. If the fines/coarse ratio is not that
favorable, it may be necessary to consider more sophisticated
treatment. Of course, this upgraded treatment will depend upon the
contaminants of concern, but it may include biological degradation
or metals extraction.
In the primary case (simple treatment), the hydrocyclone
overflow is pumped to the sedimentation area, currently consisting
of banked Lamella clarifiers. An appropriate polymer has been
selected in lab jar testing, and is dosed prior to introduction to
the Lamella. The clarified solids are directed to a sludge
thickener, while the water overflow is returned to the wet
screening area for reuse. The thickened solids are then pumped to
the belt filter press, or, more accurately, a pressurized belt
filter press. This unit is one of the most important in the entire
process in terms of selection. A 15-20% solids influent is
converted into a 50-60% dry solids filter cake. This cake contains
the target contaminants and therefore must be managed by disposal
at a properly permitted off-site disposal facility, depending upon
the specific contaminants and their status in regard to current
land bans.
Residuals Management
The important decision that must be made in selecting a soil-
wash system is the manner in which the residuals from the treatment
system will be managed. Remember, there are three primary
residuals to be handled:
The Oversize and Gross Oversize Material
The Clean Coarse-Grained Material (The Sand)
The Fine-Grained Material (The Sludge Cake)
For the oversize material, efforts will be taken to reuse the
material. Wood and wood products can be shredded, in many areas
this material can be used as a supplemental fuel in co-generation
facilities. Steel scrap can be sold to mini-mills, and concrete
rubble can be crushed for use as aggregate in concrete production.
1291
-------
The clean sand can be used as select backfill, and can usually
be returned directly to the area of excavation. If the site
conditions do not require the area of excavation to be regracled,
the clean material can be used as a construction grade material for
other development uses on site, such as roadways or concrete. In
some states, with California leading the way, this "clean" material
can be sold for off-site uses after meeting certain criteria.
The fine-grained materials, recall that here is where the
contaminants reside, will require disposal off-site at a permitted
RCRA Treatment, Disposal, or Storage Facility (TSDF). When the job
is initially scoped we will make solid determinations regarding the
type of disposal or treatment facility that will be required for
the specific fine-grained residuals from the site. This scoping
decision will usually be limited to a decision between a hazardous
waste landfill or a fixed-base incinerator. This decision will
hinge upon the determination as to the status of the specific
waste(s) with regard to the Land Disposal Restrictions (LDRs),
commonly known as the land bans.
QUALITY CONTROL SAMPLING AND ANALYSIS
Naturally, any decisions in both the selection, qualification,
handling, and disposal of treated residuals will be made using
analytically quantified information. The specific parameters to be
quantified, and the analytical methods to be employed will be made
on a site-specific basis. This decision will be made after an
understanding of the previous work performed, the nature of the
regulatory requirements at the site, and the client/contractor
strategy to be followed.
In most cases, routine quality analyses will be performed on
the project site relying on GC and AA techniques. Periodic
sampling and analyses will be performed on the treated residuals to
verify product quality and the compliance with treatment
objectives.
OPERATIONS AND STAFFING
The soil wash plant is relatively easy to operate. The
flexibility of the plant is such that it need not be kept running
24 hours per day, as is the case with an incinerator, for example.
Currently, the Dutch operate the plant on a 5 day per week/2 shift
per day basis. Preventive and routine maintenance is performed on
Saturday and the plant is shut down on Sunday. Since only pumps,
conveyors, and support equipment are operated, the air flotation
unit is the only sub-system that requires any extraordinary care.
If schedule or production requires, however, 7 days per week/3
shifts per day schedules can be worked.
1292
-------
The field operation is headed up by a Plant Manager, who is
supported by a Plant Engineer, Site Safety Officer, and a
mechanical/electrical technician, the four of whom work the day
shift. The shift crews (two or three depending upon production
requirements) each consist of a Shift Foreman, a flotation unit
operator, a belt filter press operator, and two laborers. All
plant personnel are trained in the requirements of OSHA 1910.120
and all participate in the routine medical monitoring program.
Since this venture represents the use of a new technology to
the U.S., the plant operations staff will be supplemented by
trained and experienced operators from Holland during the first
year of operation.
THE REGULATORY SITUATION
The success of soil washing will be measured by the ability of
the system to meet specific treatability/cleanup standards.
Projects will be regulated, in most cases, by either CERCLA, RCRA,
or specific state law. In the case of CERCLA, no specific permit
is required, but all the normal requirements of a permit must be
documented and met. When the soil-washing remedy is specified in
the Record of Decision (ROD)...as it has been in seven RODs as of
Mid-April, 1991...the permits form no barrier to implementation.
RCRA projects have recently become much more flexible to the
use of innovative technologies through the Corrective Action
Program. An owner/operator can apply for a temporary permit to use
an innovative method for 180 days and renew for another 180-day
period. (Most projects can be completed in this one year period.)
States are also moving ahead rapidly to implement practical
remedial projects. The State of California, for example, is
promulgating policies to permit the treatment and incorporation of
treated residual materials into asphaltic and construction grade
materials.
TREATABILITY STUDIES
Every project will commence with a treatability study. The
purpose of the study is to understand the particle size/contaminant
relationship, to confirm a process for the treatment of the waste
of concern, and to price the service. The treatability study
consists of four phases:
Phase I: "Representative" samples are collected from the site.
This determination of representativeness is important to the client
and contractor since this agreement is the basis of treatment and
pricing decisions. Where possible, we believe that it is very
useful for the client and the contractor to participate mutually in
this "representativeness" decision. The samples are managed with
proper controls, and can be analyzed at the client's facility if
1293
-------
the proper staff and resources are available, at the Geraghty &
Miller Treatability Laboratory in Tampa, Florida, or at the
Heidemij Treatability Laboratory in Waalwijk, The Netherlands. The
analyses to be performed include, first, the sieve analysis and the
construction of the percent Finer Curve. Then, the target
particle-size fraction samples are chemically analyzed for the
required contaminant menu. This phase usually takes about four
weeks and costs between $3,000 and $5,000 depending upon the
analytical requirements. The Phase I results represent a good
"Go/No Go" point, for this information will allow a reasonable
decision to be made regarding the feasibility of soil washing.
Phase II: The next step is to perform bench-scale investigations
to confirm specific unit operations. Specifically, screening,
hydrocycloning, air flotation, and filter pressing studies will be
conducted to select treatment units, and to determine surfactant,
polymer, flow rate, and throughput requirements. This phase of the
treatability study will be conducted in The Netherlands. In this
phase of work, direct equipment and professional support will be
provided by the Mineral Processing staff and the extensive facility
at the Technical University of Delft (The Netherlands). This is a
long-term, funded relationship between Heidemij and TUD that has
proven invaluable in keeping the team at the forefront of soil
treatment. This study will generally take about four weeks to
conduct, will result in the confirmation of a process flow diagram,
confirmation of treatment capabilities, and will cost $15,000 to
$25,000 depending upon the nature of the soil to be treated and the
resulting process treatment train.
Phase III: When necessary, a pilot treatment plant will be
tailored from existing plants at two locations to run the specified
treatment train with actual site soils. The pilot plant facilities
consist of the full range of required treatment units and have the
capacity to run studies at the level of one ton per hour. While
these studies will be normally conducted in Holland, the USEPA has
anticipated the need to ship soils out of the U.S. and has provided
guidelines and requirements in 40 CFR 263. (PCB materials cannot be
shipped out of the U.S.) The scope of the pilot study and the
location where it will be conducted depend directly on the size and
complexity of the project. Where a site situation matches closely
to current experience, it may not be necessary to even conduct a
pilot level study. The team can, where necessary, assemble a pilot
treatment facility at the U.S. site. The cost of the pilot study
involves so many variables, that no good guidelines can be given
without understanding the specific site requirements.
Phase IV: After the completion of the required studies, a report
will be prepared documenting the investigation activities; and
providing conclusions regarding the findings. The report will
provide the confirmed process flow diagram and the general
specifications for the actual facility. The report will commit to
a unit treatment price and specify any particular contractual
1294
-------
qualifications. The document is intended to provide all the
technical information required to negotiate a services agreement.
COSTS
Comparison costs of other forms of on-site treatment are shown
in Table 1. A summary of the unit treatment price, broken down by
major cost components, and at several different volume points, is
presented in Table 2.
KEYS TO A SUCCESSFUL PROJECT
What makes a successful remediation project? Of course, many
things, but for soil washing here are the key issues to consider:
1. Begin with an open relationship between client and contractor.
One thing is certain....the project understanding we start out
with will certainly change during the conduct of the work. It
is extremely important that a relationship of reasonable trust
exists at the beginning of the job and be nurtured through the
ensuing work.
2. The size of the job should be considered, since on-site
technology applications are directly dependent upon volume as
an economic fact. For a soil washing job to compete on a
project where all "normal" remedial alternatives are open, a
volume of more than 20,000 tons is required. On projects
where "normal" alternatives are limited by unusual site
conditions or wastes, then that minimum volume may decrease.
3. The particle size/contaminant relationship is central to the
selection of soil washing. The better the natural
distribution of coarse and fine-grained materials, the more
economical soil washing becomes. Remember, soil washing is
not a set, rigid treatment train, but is modified specifically
for the actual wastes to be treated. Also, keep in mind that
very substantial volume reductions can be obtained by
understanding the particle size/contaminant relationship and
merely screening and separating wastes for the most
appropriate treatment.
4. The understanding of the regulatory situation is very
important. The EPA is in strong support of innovative, on-
site technologies. BUT, that does not mean that any special
consideration or permitting support emerges from this
supporting position. The position of the State regulators is
very important in selecting on-site approaches, and this
position must be factored directly into the client's remedial
strategy.
1295
-------
BENEFITS OF SOIL WASHING
The benefits of soil washing are substantial and are:
• The system is exceptionally cost-effective since it can focus
treatment only on the appropriate fractions, rather than
treating the entire waste stream.
The system can treat both organics and inorganics in the same
treatment stream.
The soil washing system is a true volume reduction option and
directly supports the recycle and reuse of site materials.
The system is consistent with the current EPA directives and
policies requiring on-site, innovative treatment.
Since there is no air emission or wastewater discharge, the
system is easier to permit than traditional remedial
alternatives.
WHAT DO YOU NEED TO DO TO GET STARTED?
Please contact Mr. Michael J. Mann, Mr. Jack Peabody, or Ms.
Jill Besch at (800) 676-1921 to discuss your specific site
situation. We will be happy to provide direct information
regarding your needs, arrange a site visit, if appropriate, and
respond in writing to requests for proposal. As stated above, each
site requires a treatability study, a study that can be tailored to
the needs of your project, conducted in a staged process, and by
using existing site information.
1296
-------
1297
-------
TABLE 1
COMPARISON OF
ON-SITE REMEDIAL TECHNOLOGIES
Incineration
In-situ vitrification
Cost/yd
600-2000
350-400
Low temperature thermal 200-250
treatment
Chemical treatment, (solvent 250-300
extraction, BEST, KPEG)
Soil washing 150-200
Bioremediation 75-100
Stabilization/solidification 20-100
Vapor extraction/soil venting 2-5
Waste Handled
Organic Inorganic
yes
yes
yes
yes
no
yes
no
no
Permitting
RCRA, air
and NPDES
Land ban
restrictions
air
NPDES
yes
yes
yes
yes
no
yes
none
none
Land ban
restrictions
yes
no
air
1298
-------
Item
TABLE 2
SOIL WASHING COSTS
($ Cubic Yard)
20,000
TOTALS
$213
Volume (cubic yards)
40,000 60,000
$ 147
$ 123
100,000
Capital Depreciation
MOB/demob
Labor
Back-up
Chemicals
Maintenance
Equipment upgrade
Safety equipment
Utilities
Material handling
Management/engineering
Overhead
Process testing
Off-site disposal
Site preparation
65
5
25
3
15
8
12
3
6
5
20
9
22
15
0
38
3
15
2
15
4
9
3
6
5
13
8
11
15
0
28
2
12
2
15
4
8
3
6
5
10
5
8
15
0
20
2
9
1
15
3
7
3
6
5
8
3
4
15
0
$ 101
1299
-------
REMEDIAL DESIGN PROCEDURES FOR RCRA/CERCLA FINAL COVERS
Donald D. Moses, F.E.
Chief, Hazardous and Toxic Waste Geotechnical Section
U.S. Army Corps of Engineers
Omaha District
215 North 17th Street
Omaha, Nebraska 68102
(402) 221-3077
I. INTRODUCTION
The design of RCRA/CERCLA final covers requires a systematic process that
begins with the collection of predesign data and ends with a set of plans and
specifications for construction of the project. The remedial design
procedures presented in this paper were developed over the past several years
and are based upon the following experiences:
* Performance of in-house designs
* Reviews of Architect-Engineer (AE) designs
* Lessons learned from construction
* Training Short Courses:
- Clay Liners and Covers for Waste Disposal Facilities/University
of Texas at Austin
- Designing with Geosynthetics/Drexel University
- HELP Model/U.S. Army Corps of Engineers
- Etc.
* Seminars
- EPA: Design and Construction of RCRA/CERCLA Final Covers
(Summer 1990)
- 4th GRI Seminar: Landfill Closures (Dec 1990)
- Etc.
* Supplier presentations
* EPA Technical Guidance Documents
* and from the many technical references noted herein
The target audience for this paper is the project engineer or technical
manager who is responsible for producing plans and specifications for the
construction of a final cover.
II. PREDESIGN REQUIREMENTS
In orier to proceed from the Record of Decision (ROD) into preparing plans and
specifications for the construction of a final cover, it is normally necessary
to conduct predesign surveys and investigations to fill data gaps. The
existing data base available from the Remedial Investigation (RI), the
Feasibility Study (FS) and any other available documents must be reviewed
before scoping a predesign effort. The following predesign information is
normally required to design a final cover:
1300
-------
A. Field Surveys and Record Searches.
1. Topographic Surveys. Topographic surveys of the project
site are required at one foot contour intervals. One foot contour intervals
are desirable for final covers because of the thin component layers (24" clay
barrier) and flat slopes involved with the design. The topographic mapping
should be accurate within +0.1 foot in all dimensions. The topographic
survey should be mapped on the Computer Aided Design and Drafting (CADD)
system. All surface features such as utilities, ponds, fences, trees,
streams, ditches, water, etc., should be delineated on the mapping. The
topographic mapping needs to be referenced to the horizontal and vertical
control used to perform the survey. The surveys should also identify the
coordinates and elevations of existing wells, drill holes, piezometers and any
predesign activities such as trenching and borings.
2. Aerial Photography (Historic Photo Chronology). A record
search should be made to obtain a chronology of historic aerial photographs.
Historic aerial photographs can be used to help identify the nature and extent
of the landfill. This information is used to help define the limits of the
final cover and aid in the cover design.
3. Horizontal and Vertical Control. Three permanent control
monuments need to be established. The monuments should be strategically
located to be used for, but not destroyed by, new site construction. The
monuments should be assigned state plane coordinates and/or be tied to the
horizontal grid used in previous studies. The vertical datum should be sea
level elevations. The control monuments should be described and tied to
references.
4. Boundary Surveys and Property Search. Boundary surveys
shall be performed for all properties or parcels within the project
construction and access limits. The boundary survey traverses should be tied
to the sites horizontal control. All corners and evidence shall be identified
on a traverse plat. A property search is also required which identifies the
property owners of all affected and adjacent parcels of land. This
information is used to prepare right-of-way and construction limit drawings.
5. Monitoring Baseline Surveys. For some projects, it is
desirable to perform surveys to establish the baseline to monitor design
concerns such as settlement and slope movement.
6. Utilities Search. All on-site above and below ground
utilities need to be identified and located on plan sheets. The utility
search should consist of an on-site inspection, a flat file search and
contacts with utility companies. The location of all on-site utilities should
include horizontal alignment, depth or height, type, sizes and the Utility
company contact and telephone number.
7. As-built Drawings Search. Rarely are design or as-built
drawings available for CERCLA sites. However; it is prudent to conduct a
record search for any design, operational or as-built information that can
help identify the nature and extent of the landfill or contaminated area.
1301
-------
B. Geological Subsurface Investigations. After the existing data
base within the RI/FS and all other available documents regarding subsurface
information has been reviewed, the geological subsurface investigations can
then be scoped. The purposes of conducting geological subsurfaces
investigations during the predesign phase are described as follows:
1. Define Limits of Landfill.
It is imperative to properly define the limits of the landfill which is to be
covered. In many situations, the extent of the landfill is not clearly
visible due to cover soil placed over the landfill in the past and
subsequently overgrown by vegetation. The limits of the landfill can be
tentatively defined by first conducting an electromagnetic conductivity (EM)
survey (if the landfill is suspected to contain metals) or a soil gas survey
(if the landfill is suspected to contained organic material). Test pits
should then be excavated around the perimeter of the suspected landfill area
to verify landfill boundaries as estimated by the EM or soil gas surveys. The
test pits shall be excavated radially away from the landfill until the
boundary is identified. Historic aerial photos can aid in trying to delineate
the landfill boundary. The horizontal coordinates and vertical elevations of
the EM, soil gas surveys, and trenching need to be surveyed and recorded.
2. Characterize Site and Borrow Area Soils.
The geotechnical characteristics of the soil at the site and at potential
borrow sources need to be determined. Soil characteristics are required to
determine the suitability of the material for use in the various layers of the
cover system and for use in the settlement and stability analyses. The soil
characteristics are determined by drilling (or trenching), sampling and
testing the material. For certain soil parameters, the cone penetrometer,
standard penetration test, vane shear test and other in-situ tests can be used
to estimate to soil properties.
3. Material Excavatability.
When a project feature such as a leachate collection trench requires
excavation, profile sheets should be included in the drawings informing the
contractor of subsurface conditions and excavation limits. The information
required includes soil or rock type, water table and leachate levels, soil
moisture content, soil horizons and bedrock profiles, rock hardness and
rippability.
4. Methane and Landfill Gas Presence.
Soil gas surveys on the ground or landfill surface and soil gas probes can be
installed into the landfill proper to investigate the presence of methane and
other volatile organic vapors within the landfill.
5. Landfill Composition.
Sometimes it is necessary to excavate actual landfill refuse in order to
minimize earthwork or to locate leachate collection trenches. If this is the
case, then the landfill composition should be determined during the predesign
effort in order to inform the contractor of subsurface conditions.
1302
-------
6. Leachate Levels.
Landfills are normally quite pervious and can have large void spaces creating
significant amounts of leachate. Landfills have their own distinct internal
drainage characteristics. During predesign activities, the surface of the
landfill should be inspected in order to locate leachate seepage exit areas.
Leachate seeps should be surveyed and mapped. Piezometers can be installed in
landfill to identify leachate gradients and flow paths. This information is
used to design and locate leachate collection systems.
7. Define Ground Water Conditions.
Observations wells and piezometers are normally installed during the RI/FS
process. It is normally necessary to define water levels, gradients, water
chemistry prior to constructing a final cover to define baseline conditions.
C. Laboratory Geotechnical Testing Requirements. The following
geotechnical tests are normally required to assess the suitability of borrow
sources for use in the cover layers and to perform the stability and
settlement analyses:
* Soil Classifications
Mechanical Analysis
Hydrometer Analysis
Atterberg Limits
* Moisture Content
* Standard (or Modified) Proctor Compaction Test
* Permeability Tests
* Density Tests
* Dispersive Clay Tests (Borrow Material)
* Soil Strength (Shear Tests)
* Consolidation Tests (for Settlement Analysis)
* Direct Shear Tests for all Final Cover Interfaces
D. Chemical Data Quality Management.
Chemical testing is required to determine if there is a presence of methane or
other volatile organic vapors. Chemical testing is also required to ensure
that borrow sources are not contaminated (TCLP test) and to determine ground
water and leachate chemistry.
E. Map Data Base.
USGS Quadrangle Maps in both the 7.5 minute series and the 1:250,000 scale are
useful design aids. Separates can be obtained from the USGS for both these
map types and can be used to make site and location maps.
F. Field Pilot Studies,
Test fills can be conducted as a component of predesign or as part of
construction to verify or determine design assumptions. The landfill refuse
material can be preloaded to obtain short term settlement values or stability
parameters. Test fills can be used to determine or verify construction
methods such as the placement of select fill on the geosynthetics. Tests
fills can be used to design or verify the stability of the final cover layer
interfaces under worst-case conditions. Test fills can be used to determine
1303
-------
the minimum random fill thickness required to provide a firm foundation to
allow proper compaction of the low-permeability clay layer. ConstructabiLity
and safety issues such as the placement of random fill on steep landfill
sideslopes can be assessed with test fills. The large scale field performance
of final cover components such as the in-situ (large scale) permeability of
the clay barrier layer can be verified with a test fill.
III. DESIGN CONSIDERATIONS
The EPA Technical Guidance Document titled Final Covers on Hazardous Waste
Landfills and Surface Impoundments, dated July 1989 (EPA/530-SW-89-047) <1>
provides design guidance on final covers for hazardous waste units. The
design guidance presented in that document satisfies the requirements of 40
CFR 264 and 265 Subparts G (closure and post-closure), K (surface
impoundments), and N (landfills). The EPA emphasizes that their
recommendations are guidance only and not regulations. The EPA also
acknowledges that other final cover designs may be acceptable, depending upon
site-specific conditions and upon a determination by the Agency that an
alternative design adequately fulfills the regulatory requirements. The
following design considerations adhere to the EPA's recommendations and
reflect additional design requirements.
A. Final Cover System Component Layers.
1. Vegetative Cover. The top layer in a final cover is the
vegetative cover. The primary purpose of the vegetative cover is to resist
wind and water erosion. The vegetative cover minimizes the infiltration of
surface water into the lower layers of the cover system and maximizes
evapotranspiration. The vegetative cover also functions in the long term to
enhance aesthetics and to promote a self-sustaining ecosystem on top of the
cover <2>.
The EPA <1> recommends that the vegetative cover meet the following
design specifications:
* Locally adapted perennial plants
* Resistant to drought and temperature extremes
* Roots that will not disrupt the low-permeability layer
* Capable of thriving in low-nutrient soil with minimum
nutrient addition
* Sufficient plant density to minimize cover soil erosion to
no more than 2 tons/acre/year (5.5 MT/ha/yr), calculated
using the USDA Universal Soil Loss Equation <3>
* Capable of surviving and functioning with little or no
maintenance
The final cover should be built on slopes no steeper than IV : 3H for
maintenance purposes. Equipment necessary to plant and maintain vegetation
cannot operate safely on steeper slopes <2>. This minimum slope
recommendation compares to a slope of 2V on 5H which the Corps of Engineers
uses as the steepest slope that can be conveniently traversed with
conventional mowing equipment <4>.
It is important to note that in many cases, landfill gas must be contained and
controlled to prevent gases from migrating into the root zone of the
vegetative cover and killing the plants.
1304
-------
The EPA has developed "expert systems" which are computer programs that
"mimic" the knowledge and decision making processes of a human expert. An
expert system titled Vegetative Cover Advisor (Veg Cov) analyzes the
properties of the topsoil and subsoil, examines the appropriateness of plant
species, preforms use analysis, examines general requirements, and writes a
conclusion report. This system can be used to verify or aid in the design of
the vegetative cover, topsoil, and select fill.
For sites located in arid regions or final covers with steep sideslopes
(steeper than 1V:4H), an armor system can be used as an alternative to
vegetative cover <2>. Alternative designs could consist of cobbles, gabion
structures, concrete caps, asphaltic cement caps and chemical sealant caps.
2. Soil Cover (Top Soil). Below the vegetative cover is top
soil which is required to support the vegetative cover. The top soil shall
have a minimum thickness of 6 inches and shall be representative of soils in
the site vicinity that produce heavy growths of crops, grass or other
vegetation. The top soil must be free of contamination. The final top slope,
after allowance for settling and subsidence, should have a slope of at least 3
percent, but not greater than 5 percent in order to facilitate runoff while
minimizing erosion <1>.
For cover sideslopes greater than 5 percent, erosion caused from surface
runoff is likely to occur unless erosion controls such as terraces, gabion
structures, riprap, or erosion control mats are designed. As stated
previously, the EPA <1> recommends that slopes and vegetative cover be
designed to prevent the formation of erosion rills and gullies such that total
erosion is limited to less than 2.0 tons/acre/year as determined by the
Universal Soil Loss Equation. In addition, temporary erosion control measures
are required during construction and post construction until permanent
vegetation is in place.
3. Select Fill. Below the six inch thick top soil layer is the
select fill layer. The purpose of the select fill is to provide a soil that
is capable of sustaining the vegetative cover through dry periods and protect
the underlying geosynthetics and clay barrier layer from the elements (frost
penetration and desication). The select fill also provides water holding
capacity to attenuate rainfall infiltration to the underlying drainage layer.
When designing clay barrier cover systems, the thickness of the select
fill should be a minimum of 24 inches (including 6" of top soil) or equal to
the maximum frost depth, whichever depth is greatest. The select fill must
also be free of contamination The select fill material should be of medium
textured soils such as loam soils for both function and constructability.
Loam soils are capable of supporting the root systems of the vegetative cover
and providing water holding capacity. Sandy soils are undesirable because the
material has low water retention and loses nutrients by leaching.
Cohesionless sands and silts are also undesirable because these materials have
been known to cause severe clogging of underlying geotextile filters. Clayey
soil types are more fertile than sandy soils but cohesive soils, especially
highly plastic clays tend to pond water, and are difficult to place on the
underlying geosynthetics. The best materials that are cohesive but not highly
plastic and include SM-SC (sandy silt-sandy clay), SC (sandy clay), ML-CL
(silt-lean clay) and CL (lean clay) as classified according to the Unified
Soil Classification System. The maximum particle size should not exceed 3/8
inch so as not to puncture or damage the geotextile. It should be noted that
1305
-------
the determination of the select fill soil type will ultimately depend upon the
availability of economical borrow sources.
Constructability issues are critical when placing the select fill on the
geosynthetics. Specifying soil types that are not highly plastic clays
assures the select fill material can be placed in homogeneous layers. The
select fill material should be placed starting at the toe of the cover working
up the slope and parallel to the toe. The first layer of select fill should
be placed in a thick loose lift of 15"-18" in depth. Equipment should not be
driven or pulled directly on any underlying geosynthetics. Equipment is
allowed on areas underlain by the geotextile only after the first layer of
select fill has been placed. The select fill should not be dropped or duirped
onto the geosynthetics from a height greater than 12 inches. The select fill
should be placed onto the geosynthetics by dropping (not pushing) the fill
from a small tracked-dozer similar to a Caterpillar Model D4 or D6. To
protect the geosynthetics, achieve a stable structure, and to enhance the
soil's ability to support the vegetative cover, select fill should be
compacted with only minimum effort. Generally, traffic compaction using
placement equipment is sufficient. Select fill should be placed when the
geosynthetics are fully contracted (i.e. during cooler periods of the day) to
prevent excessive thermal stresses in the geosynthetics. This is more
critical for polyethylene products which have a relatively high coefficient of
thermal expansion. The geosynthetics must be anchored at the top of the slope
before placement of select fill. The select fill should not be stockpiled, on
the final cover in heights greater than 24 inches. The exposed geosynthetics
should be covered as quickly as possible to reduce the potential for damage
from ultraviolet radiation, wind, temperature extremes, and on-going
construction activities <5>.
A test section should be constructed before full scale placement of select
fill is allowed on the geosynthetics. The contractor should demonstrate that
their placement method and equipment used will not damage the geosynthetics.
4. Filter Layer. A filter layer is normally required between
the select fill soil cover and the underlying drainage layer. The filter
layer insures consistent drainage properties by preventing migration of fine
graded soil particles into the void spaces of the drainage layer below. The
filter layer consists of either a geotextile or a series of graded granular
materials.
a. Geotextile Filter Fabric Alternative. Adequate
performance of the filter layer depends on several factors. Once the select
fill soil type is specified, the geotextile is chosen based upon the following
design criteria: 1) the fabric must retain the soil (retention criteria); 2)
the fabric must allow surface infiltration to permeate through the fabric into
the drainage layer (permeability criteria); and 3) the fabric must not clog
over time (clogging criteria). In addition, 4) the fabric must survive the
installation process, placement of select fill upon it, and long term loading
from the select fill surcharge (survivability criteria), and 5) the fabric
must be compatible with surface water (compatibility criteria) <6> <7>. The
fabric could also be designed to withstand a tensile force (tensile criteria)
since the material is normally tied into an anchor trench and could be secured
by drainage benches. Design references and procedures for these criteria are
presented as follows:
1) Retention Criteria. To prevent the migration of soil
particles from the select fill into the drainage layer, the voids in the
1306
-------
geotextile filter must be small enough to retain the soil on the top side of
the fabric. It is the coarser soil fraction that must be initially retained.
The coarser soil fraction eventually blocks the finer sized particles <6>.
Koerner <6> presents several approaches to determine apparent opening size
(AOS) of the fabric based upon the particle size distribution of the soil to
be retained. The most conservative method presented by Koerner is after
Giroud <8>. Giroud predicts the apparent opening size of the geotextile based
upon the following soil characteristics: relative density; coefficient of
uniformity and the soil particle size corresponding to 50% finer. Giroud's
method as represented by Koerner is displayed in table 2.14 on page 122 of
reference <6>. Giroud's method is applicable to cohesionless soil types.
When the select fill is highly cohesive and consists of primarily clay
materials, the above referenced relationships are not applicable because the
particle size of clay (0.002 mm) is far smaller than the apparent opening size
of any geotextile. For cohesive soils, the Omaha District uses a geotextile
filter with an Apparent Opening Size (AOS) no finer than the U.S. Standard
Sieve No. 100 and no coarser than the U.S. Standard Sieve No. 70 to separate
the select fill from the drainage layer <9>.
2) Permeability criteria. <6> <8> The geotextile filter
must have an Apparent Opening Size fine enough to retain the select fill but
yet open or permeable enough to allow surface water infiltration to pass
through the filter into the underlying drainage layer. Therefore, it is
necessary to determine the cross-plane permeability (k) of the fabric. In
addition, since geotextiles deform under load the thickness (t) of the fabric
is accounted for in a term called permittivity (Y) where: Y=k/t
The permeability and permittivity values of the filter are determined by using
ASTM Method D4491, Water Permeability of Geotextiles by Permittivity. The
values of these parameters for geotextiles range over several orders of
magnitude as presented below <6>:
Permittivity (Y)-from 0.02 to 2.2/seconds
Permeability (k)=from 0.0008 to 0.23 cm/s
The flow rate through the geotextile as measured by its permittivity (Y), is
selected to be greater than the flow from the select fill times a factor of
safety, usually 10 or greater <7>. The flow rate through the select fill can
be obtained from the HELP model or from the site-specific water balance. The
coefficient of permeability (k) of the geotextile can also be checked to
verify the k value of the fabric is greater than the k value of the soil.
3) Clogging Criteria <6>. The filter fabric becomes
clogged when the soil particles embed within the fabric structure. Clogging
of the filter fabric prevents surface water infiltration from being able to
enter the drainage layer. Koerner <6> states that the likelihood of complete
soil clogging of geotextile filters can be prevented by:
* avoiding cohesionless sands and silts as select fill;
* avoiding gap-graded particle size distributions in
select fill; and
* avoiding high hydraulic gradients.
If these situations cannot be avoided, then the specified select fill material
and geotextile filter can be tested together using either a gradient ratio
test <12> or long term flow test <13>.
1307
-------
4) Survivability Criteria. The geotextile filter must
survive the installation process to perform effectively <7>. The geotextile
must be durable enough to withstand a Caterpillar Model D6 working on loose
lifts thicknesses of 15 inches (see paragraph III. A.3 Select Fill). The
Specifications for Geotextiles developed by Task Force #25 (AASHTO-ABC-ARBTA)
<11> specifies the physical properties required for various degrees of
survivability. For final covers as presented above, the geotextile requires a
"High" degree of fabric survivability with the following minimum property
values: Grab strength - 180 Ibs; Puncture Strength - 75 Ibs, Burst Strength
=290 psi and Trap Tear - 50 Ibs.
5) Compatibility Criteria. The compatibility between
surface water infiltration and the geotextile filter is generally not critical
and does not normally require compatibility testing (EPA 9090).
6) Tensile Criteria. The geotextile fabric can be
designed to withstand tensile forces if the material is tied into an anchor
trench and secured by drainage benches. A heavy woven fabric with a high
modulus of elasticity should be specified for the filter material if the
fabric is to be designed to withstand tensile forces. In addition, the fabric
would have to be sewn in the field in lieu of just overlapping the material.
b. Graded Granular Filter Alternative. A series of
graded granular filter layers can be used as an alternative to a geotextile
filter. The granular layers must be graded both to prevent piping and to
maintain permeability. Criteria for granular filter layers can be found in
Cedergren <14> and the U.S. Army Corps of Engineers, Engineering Manual for
Seepage Analysis and Control for Dams <15>.
5. Drainage Layer. The primary functions of the drainage layer
are to intercept water that infiltrates the select fill and then convey the
water out from beneath the cover. The drainage layer should be designed to
minimize the amount and residence time of water being in contact with the low-
permeability layer, thus decreasing the potential for leachate generation.
The drainage layer must slope to an exit drain and discharge away from the toe
of the cover. The drainage layer can consist of either a geonet or 12 inches
of granular material <1>.
a. Geonet Alternative.
1) EPA Recommendations. The EPA <1> identifies the
following parameters which should be addressed in assessing geonet drainage
materials:
* hydraulic transmissivity (the rate at which liquid can be
removed) should be no less than 3 X 10" meters squared per
second.
* compressibility (the ability to maintain open pore space and
thus transmissivity, under expected overburden);
* deformation characteristics (the ability to conform to
changes in shape of the surrounding materials);
* mechanical compatibility with the FML (the tendency for the
drainage material and the FML to deform each other);
* useful life of the system; and
* ability to resist physical, chemical and biological
clogging.
1308
-------
2) Koerner's Design-by-Function Concept. Koerner <6>
<16> presents the following design criteria which reflects in part the EPA
requirements addressed above:
* Compatibility. As with the geotextile, the
compatibility between surface water infiltration and the geonet is generally
not critical and does not normally require compatibility testing (EPA 9090).
* Crush Strength. The geonet must be able to withstand
normal pressures from the dead load of the select fill material and live loads
from construction and maintenance activities. In order to avoid rib "lay-
down" and/or creep deformation, the normal pressure capability of the geonet
must be increased by a factor of safety over the design value. The
Geosynthetic Research Institutes Test Method GN-1 <18> can be used to
determine the allowable normal pressure. The normal stress on a geonet from
the select fill for a cover design is light as compared to a landfill bottom
liner and usually should not be a critical cover design parameter.
* Flow Capability. <6> The geonet must be able to
convey the designed flow rate which is determined from the HELP Model or from
the site-specific water balance. The allowable flow rate is the quantity for
which the geonet can convey by planar flow or by it's transmissivity. The
transmissivity of the geonet is determined by using ASTM D4716. The
laboratory test flow should reflect the proper normal load and hydraulic
gradient. Since laboratory tests yields the ultimate flow rate, Koerner <6>
recommends that the laboratory flow rate be reduced before use in design.
Koerner reduces the ultimate flow rate to reflect items not adequately
assessed in the laboratory test. The items include preliminary factors of
safety adjustments for the following: elastic deformation, or intrusion of
the adjacent geosynthetics into the geonet's core space; creep deformation of
the geonet and adjacent geosynthetics into the geonet's core space; chemical
clogging and/or precipitation of chemicals in the geonet's core space; and
biological clogging in the geonet's core space. A final factor of safety is
also used where the required flow rate must be less than the allowable flow
rate. Refer to Koerner <6>, pages 350 to 352 for the preliminary factor of
safety values and further discussion.
* Minimum Slope. The EPA <1> states for gravel drainage
layers, particularly where unusually long slopes are required, slopes greater
than three percent may be necessary. This concern is especially true for
geonets because of the thin depth of the material.
b. Granular Material Alternative. The EPA <1> also
describes a 12 inch minimum thickness drainage layer alternative. The 12 inch
thickness allows for both transport of drainage and protection of the low-
permeability geomembrane barrier (FML) during construction. Slopes of 3% or
greater are recommended. The EPA specifies the granular drainage material
have a hydraulic conductivity of no less than 1 x 10 cm/sec and a hydraulic
transmissivity no less than 3 x 10" meters squared per second at the time of
installation. The granular material should be no coarser than 3/8 inch, and
classified as SP and consist of smooth rounded particles. A hydrologic and
hydraulic analysis is still required to verify the layer adequacy.
6. Low-permeability Layer. As per the EPA <1>, the final cover
system is required by 40 CFR 264.228, 264.310, and 265.310 to provide long-
term minimization of migration of liquids through the closed land disposal
1309
-------
unit and to have a permeability less than or equal to the permeability of the
bottom liner system or natural subsoils present. The EPA has interpreted this
to mean that the cover should contain a geomembrane (FML)/soil composite layer
similar in concept (but not necessarily identical construction materials) to
the composite bottom liner detailed in "Minimum Technology Guidance on Double
Liner Systems for Landfills and Surface Impoundments -- Design, Construction
and Operation" (EPA). The two components (FML and soil) of the low-
permeability layer recommended in this document are considered to function as
one system. They should be designed, constructed, and operated to maximize
removal of water by the overlying drainage layer and to minimize infiltration
of water into the waste. The low-permeability layer should require little or
no maintenance during and after the post-closure period.
a. Geomembrane Barrier Component (FML).
1) EPA Recommendations <1>, The FML component of the
low-permeability layer is located above the clay barrier. The EPA recommends
the FML component have the following characteristics:
* The FML should be located below the maximum depth of frost
penetration.
* The FML should be at least 20 mils (0.5 mm) in thickness,
but some units and/or some FML materials may require a
greater thickness to prevent failure under potential stress
of the post-closure care period, or during construction.
The Agency recognizes that some types of FMLs must be
thicker to accommodate unique seamability requirements, or
to increase long-term durability (e.g., increase resistance
to puncture). It should be noted that the Corps of
Engineers Missouri River Division <19> recommends the FML be
a minimum of 40 mils based upon seamability (burn outs) and
survivability during construction.
* The surface of the FML should have a minimum 3 percent slope
after allowance for settlement.
* There should be no surface unevenness, local depressions, or
small mounding that create depressions capable of containing
or otherwise impeding the rapid flow and drainage of
infiltrating water.
* The Agency recommends the use of material and seam
specifications such as those in "Lining of Waste Contaironent
and other Impoundment Facilities" (EPA, 1987).
* The FML should be protected by an overlying drainage layer
of at least 30 cm (12 in.) of soil material no coarser than
3/8-inch (0.95-mm) particle size, Unified Soil
Classification System (USCS) SP sand, free of rock,
fractured stone, debris, cobbles, rubbish, roots, and sudden
changes in grade (slope) that may impair the FML. The
overlying drainage layer should suffice as bedding in most
cases, but care should be taken that any included drainage
pipes are not placed in a way that will damage the FML.
* The FML should be in direct contact with the underlying
compacted soil component and should be installed on a
smoothed soil surface.
* The number of penetrations of the FML by designed structures
(e.g., gas vents) should be minimized. Where penetrations
are necessary, the FML should be sealed securely around the
structure.
1310
-------
* Bridging or similar stressed conditions in the FML should be
avoided by providing slack allowances for temperature-
induced shrinkage of the FML during installation and during
the period prior to placement of the protective layer or
drainage layer.
* Slack should not be excessive to the extent that folds are
created that later may crack.
2) Koerner's Design-by-Function Concept. Koerner <6>
<16> presents the following FML design criteria:
* FML Compatibility and Material Selection. As with the
other geosynthetics, the compatibility between surface water infiltration and
the FML is generally not critical and does not normally require compatibility
testing (EPA 9090). The FML is located on top of the clay barrier layer which
consists of uncontaminated fill material. The compatibility between the
underlying soil is also generally not critical. The FML does need to be
chemically compatible with landfill gases.
Recent Omaha District designs have specified VLDPE over HDPE as the
geomembrane material type. VLDPE is easy to install, has higher friction
properties and conforms better to surface topographic changes such as the
drainage terraces than does HDPE.
* Vapor Transmission (Water and Methane). Water vapor
transmission is determined by ASTM D96. The corresponding coefficient of
permeability (k) can be then be determined for the geomembrane type and
thickness. When the landfill cover is also used to contain methane gas,
methane vapor rates can be evaluated by ASTM D1434 and D814.
* Biaxial Stresses via Subsidence. The FML will need to
withstand stress induced into the material from differential settlement. The
allowable stress is determined from GM4 Three Dimensional Geomembrane Tension
Test <18>. The required stress is determined after the settlement amount has
been estimated and is dependent upon the following parameters: the unit weight
of cover soil; the height of cover soil; the depth of differential settlement;
the width of settlement depression; and the thickness of the liner. Refer to
Koerner <18>, page 39 for further discussion.
* Planer Stresses via Tension. The FML would have to
withstand tensile forces if the coefficient of friction of the upper layer
(geonet/FML or geotextile/FML) is greater than the FML/clay interface. The
tensile force in this case would also be dependent upon the length of slope
and width. Refer to Koerner <18>, page 40 for further discussion.
b. Clay Barrier Component.
1) EPA Recommendations <1>. The clay barrier layer is
located directly below the geomembrane. The EPA recommends the clay barrier
layer have the following characteristics:
* The soil should be at least 24 inches of compacted,
low-permeability soil with an in-place saturated hydraulic conductivity of 1 X
10" cm/sec or less.
* The compated soil must be free of clods, rock,
fractured stone, debris, cobble, rubbish, and roots, etc., that would increase
1311
-------
the hydraulic conductivity or serve to promote preferential water flow paths.
* The upper surface of the compacted soil (which is in
contact with the FML) should have a minimum slope of 3 percent after allowance
for settlement.
* The soil layer should be constructed so that it will
be entirely below the maximum depth of frost penetration upon completion of
the cover system.
2) Design Considerations.
* Composite Action. The clay barrier layer is located
directly below the geomembrane (FML) to create a composite liner system. The
advantage of the composite liner design is that by putting a fine grain
material beneath the FML, the impact of imperfections or holes in the FML can
be reduced by many orders of magnitude <2>. In order to achieve composite
action, the FML must be direct contact with the clay barrier layer.
* Permeability Requirement. As stated previously, the
EPA <1> recommends a low-permeability soil with an in-place saturated
hydraulic conductivity of 1 X 10 cm/sec or less. In addition to meeting the
permeability requirement, the Omaha District has been specifying the clay
liner be constructed of materials classified (as per ASTM D 2487) as either
CL, CH or SC having a plasticity index (PI) of not less than 15. Daniel <36>
warns that clays with a high PI may be a constructability problem. The clayer
layer should not contain debris, roots, organic or frozen materials, stones or
clods having a maximum dimension larger than one inch.
* Thickness Requirement. As previously stated, the EPA
<1> recommends the clay barrier layer be at least 24 inches thick. Daniel
<36> presents the relationship between the hydraulic conductivity versus the
thickness of the liner for both good and excellent construction methods. The
relationship indicates that the 24 inches is an absolute minimum thickness and
greater depths should be considered. The minimum thickness of 24 inches is
based upon constructability considerations and the ability to provide
uniformity in overall permeability.
* Frost Depth Requirement. The drainage layer, the FML
and the clay barrier layer should all be located below the maximum frost depth
penetration. Freeze-thaw cycles adversely increases the permeability
characteristic of the clay barrier layer. Freeze-thaw cycles could also
effect the interface friction between the clay/FML contact and other
interfaces.
* Desication Cracking. Desication cracking adversely
increases the permeability characteristic of the clay barrier layer. The
potential for desication of clay materials depends upon the following fact.ors:
Clay-size particle content, the soil properties such as plastic limit, liquid
limit and plastic index, depth of soil cover, clay barrier moisture content
and compaction history, climate, moisture content and the soil type of the
adjacent random fill.
* Settlement. Daniel <36> recommends not placing a
permanent low-permeability cover on unstabilized waste that will undergo large
settlements. Daniel recommends interim fill to preconsolidate the refuse
before the final cover is placed. Richardson <35> recommends an allowable
subsidence (differential strain between inflection points) of no more than 1%
1312
-------
for the clay layer. Daniel <36> emphasizes the need for composite liners for
covers placed on waste that will undergo settlement.
7. Gas Collector and Removal System. Degradation of solid
organic waste materials in a landfill generates gases, primarily of which is
carbon dioxide (CCO and methane (CH,). The carbon dioxide is heavier than
air and will move downward. The methane however, being lighter than air will
move upward and collect at the bottom of low-permeability geomembrane (FML)
barrier <2>. The potential impacts of gas generation are as follows <30>:
* Explosion hazard. Methane gas can migrate laterally
and vertically and has caused explosions in structures adjacent to and on
landfills.
* Vegetation distress. Landfill gases must be
controlled before they penetrate into the vegetative cover layer. If
uncontrolled, the gases could distress the vegetation resulting in subsequent
erosion of the cover <22>.
* Odor. Landfill gases generate nuisance odors. Odor
becomes a design parameter if the landfill is located adjacent to any existing
or potential developments. Nuisance odors can be a public perception issue
and can effect property values.
* Physical disruption of cover components. Landfill
gases if not properly controlled, can generate uplift forces against FML and
clay low-permeability layers. Uplift forces can disrupt the clay layer and
stretch and bubble the FML which adversely results in increased permeability
properties of the layers.
* Toxic Vapors. Landfill gases can be toxic. Toxicity
is a design parameter when determing venting or treatment requirements.
a. Gas Migration <30>. After a final cover is placed,
gas production can occur at high rates for years and can continue at lesser
rates for centuries. Gas migration occurs by two processes. Convection is
flow induced by pressure gradients formed by gas production in layers
surrounded by low permeability or saturated layers. Convection is also
induced by buoyancy forces since methane is lighter than air. Diffusion is
flow induced by concentration gradients formed by production of methane and
carbon dioxide at concentrations greater than in the surrounding air. Gas
migration rates are affected by the type and age of refuse material, the final
cover design, refuse temperature and moisture content. Vertical or lateral
migration paths for gas movement are influenced by the final cover design and
the presence of migration corridors and or barriers. Migration corridors
include sand and gravel lenses, void spaces, cracks, fissures, utility
conduits, drain culverts and buried lines. Barriers to gas migration include
clay deposits and high and perched water tables. Saturated or frozen surface
layers promote lateral migration of landfill gases.
b. EPA Recommendations. The function of a gas collection
system is to protect the structural integrity of the final cover from uplift
forces from the gas pressure and to protect the environment and public from
the hazardous effects of the gas. The EPA <1> offers the following design
recommendations based upon engineering judgment for a gas vent layer:
* The layer should be a minimum of 30 cm (12 in.) thick
and should be located between the low-permeability soil liner and the waste
layer.
1313
-------
* Materials used in construction of the gas vent layer
should be coarse-grained, porous materials such as those used in the drainage
layer.
* Geosynthetic materials may be substituted for granular
materials in the vent layer if equivalent performance can be shown.
* Venting to an exterior collection point for disposal
or treatment should be provided by means such as horizontal perforated pipes,
patterned laterally throughout the gas vent layer, which channel gases to
vertical risers.
* The number of vertical risers through the cover should
be minimized and located at high points in the cross-section, and designed to
prevent water infiltration through and around them.
c. Gas Control Systems. Gas control systems consist of a
collection, conveyence and outlet component. Gas control systems are designed
to be either passively vented to the atmosphere or as an active system where
the landfill gas is mechanically extracted to the surface. At the landfill
surface, the gases are either dispersed into the atmosphere, collected or
treated. All components of the gas control system must consist of materials
that are compatible with methane gas. Alternative gas control systems are
described below:
1) Passive Blanket and Liner Systems. A continuous
blanket gas collection system consisting of either 12 inches of granular fill
or of a geosynthetic material is located below the clay barrier layer. Filter
layers may be required above and below the continuous blanket. Linear
trenches excavated into the refuse backfilled with granular material can also
be used as a collector component. The granular or geosynthetic gas collection
material should have a permeability coefficent (k) of 1 cm/sec or greater
<36>. Gases are conveyed or removed from the granular blanket or trench
collector in horizontal perforated pipes which are connected to vertical
outlet vent pipes. Continuous blanket and trench gas control systems are
normally passive where the gas is forced through the system by pressure
gradients and buoyancy forces. The thickness of the select fill overburden
must be chosen such that the soil weight exceeds the anticipated gas pressure
<32>. The vertical outlet vent pipes for passive systems need to be located
at the highest elevation of the gas collection blanket to allow maximum
evacuation of the gas <1>. The vent pipes should be anchored to the barrier
layer (FML) in a way that ensures watertightness but allows for some movement
should there be differential settlement (see penetration discussion). The
number of vent pipes should be minimized and are normally spaced abouth 200
feet apart (1 per acre) <36>. The vent pipe depth should be minimized to
avoid stress concentrations at the boot connection. Linear gas collection
systems should only be used for verly low expected gas production rates <30>.
When the refuse material does contain suspecting gas producing material but
off-site migration of the gas is not a specific concern, the passive
continuous blanket gas control system can be used to protect the integrity of
the final cover.
Feeney <32> describes a constructability problem which the Omaha District has
also encountered at several projects. Passive systems relying on granular
blanket (or trench) collector systems may not function until the geomembrane
is completely covered with soil. Prior to placement of the geomembrane,
landfill gas exits the landfill through the path of least resistance. The
path of least resistant is sometimes through the cover soils and clay barrier
layer and not through the gas control system. When the geomembrane is placed,
gas collects under the material where a bubble can form and the geomembrane is
1314
-------
damaged or eventually ruptures. Feeney recommends a method of avoiding this
problem is to leave temporary vents in the geomembrane so that the landfill
gas is dissipated rather than allowed to collect beneath the liner. The vents
should be progressively sealed immediately prior to the placement of soil
cover over the vent.
2) Well Extraction Systems. Gas extraction wells
(perforated vertical collection pipes) can be drilled and placed penetrating
to the bottom of the refuse. Gas extraction wells can be either active or
passive. Active extraction wells used in conjunction with barriers create
negative pressure zones to extract gas <30>. Wells are useful for layered
landfills where vertical migration is impeded. Active gas control systems are
more effective than passive blanket systems. Active well systems with
perimeter barriers should be considered when there is nearby development and
the off-site (lateral) migration of methane gas is either an environmental or
safety concern and when the refuse material is highly organic and will
generate large amounts of gas. Gas monitoring stations should be located
outside the perimeter of the landfill situated between any development or area
of concern. Gas monitoring stations can be used in conjunction with any gas
control system active or passive. Gas monitoring stations <30> should be
spaced every 1000 feet and be able to detect 25% of the lower explosive limit
of methane. The monitoring stations are similiar to piezometers and extend to
the maximum refuse depth.
The boot connection detail is critical for wells that penetrate completely or
deeply into the refuse material. The well itself being ridsid and anchored
into firmer and more compact material will not settle as much as the landfill
surface. This differential movement will create stress concentrating at the
boot connection and can cause the FML to tear away from the rent pipe (see
penetration discussion).
8. Random Fill. Random fill is placed directly on the refuse
material covering the entire aerial extent of the landfill. Random fill is
used to bring the cover to proper grade and elevation reflecting the
settlement and stability analyses and drainage and minimum fill requirements.
Prior to the placement of the random fill material, the landfill surface must
be cleared of vegetative cover and proof-rolled. In certain circumstances,
limited excavation and reshaping the landfill surface can minimize the volume
of random fill material required which could result in substantial cost
savings. Excavation into the landfill material requires specific safety
considerations and is normally avoided if possible. Random fill can be either
cohesionless or cohesive depending of the availability of materials.
Materials which are unsuitable for use as random fill include debris, roots,
brush, sod, organic or frozen materials and soils classified (according to
ASTM 2487) as either MH, PT, OH and OL. The random fill is placed in lift
thicknesses of 8 inches for cohesive materials and 12 inches for cohesionless
materials. Density of the random fill is controlled by the standard procter
test (ASTM D 698) for cohesive soils and the relative density test (ASTM D
4253) for cohesionless materials. Specific density requirements are not used
for the bottom two layers of random fill placed. A procedure specification is
used for the bottom two layers identifying a minimum number of passes of
compaction equipment. A specific density in the first few lifts may not be
possible due to a soft and compressable landfill surface. The random fill
layer must have a minimum thickness to provide a firm foundation to allow
adequate compaction of the low-permeability clay layer. A test fill may be
required to determine this thickness. The Omaha District has been specifying
1315
-------
that the measurement and payment method for random fill be by the ton. This
assures that the contractor is payed for all fill placement noting the refuse
consolidation and foundation settlement are likely to occur.
9. Refuse or Contaminated Material. A final cover can be
placed over landfill refuse material or contaminated natural soils. Landfill
refuse materials can consist of municipal or industrial wastes or be
construction debris. The nature and extent of the waste material
significantly effects the final cover design. The settlement and stability
analyses, the gas and leachate control systems are all effected by the
landfill composition. A final cover over natural soils that are contaminated
is easier to design than a cover over a landfill. Traditional soil
geotechnical sampling and testing can be used to characterize the properties
of the soil required for design. Whereas, the geotechnics of landfill
materials are normally highly variable within an individual site and the
geotechnical properties of waste materials are very difficult to quantify.
10. Optional Layers.
a. Biotic Protection Barrier. The Omaha District has no
experience with the operation and effectiveness of constructed biotic barrier
layers. To reiterate EPA <1> guidance documents, plant roots or burrowing,
animals may disrupt the integrity of the drainage and low-permeability layers.
Physical barriers, such as layers of cobbles or coarse gravel beneath the
select fill, and chemical barriers, have been proposed to discourage or reduce
the threat of biointrusion. Long term monitoring and evaluation of
constructed final covers is required in various locations of the country to
assess actual damage to the drainage and low-permeability layers from
biointrusion.
b. Geogrid Reinforcement <33>. The geosynthetic liner
layer interfaces normally control the design sideslopes of a cover. The
interface friction angles between adjacent geosynthetics or between the
geosynthetics and adjacent soil can range between 8° to 25° <33>. Cover
sideslopes of 1V:4H (14°) and steeper could readily have stability problems at
the cover layer interfaces. Geogrids can be used to reinforce soils to
provide stability to cover sideslopes.
c. Geocomposite Alternatives. There is a wide range of
geocomposite materials available where various geosynthetic layers are
factory-bonded together in one unit. Koerner <6> describes the various forms
of geocomposites noting that the type of geocomposite is controlled by the
function required, such as; seperation, reinforcement, filtration, drainage,
and moisture barriers.
B. Settlement Analysis. Without the proper design considerations,
settlement of the landfill and the underlying natural foundation material can
damage or compromise the integrity of the final cover <25>. Excessive
differential settlement could cause the following failure scenarios:
* Severe cracking of the clay barrier resulting in the loss of the
impermeable characteristic of the layer.
* Steepened sideslopes resulting in slope stability failures.
* Induced tensile stresses in the FML and other geosynthetics.
* Stress concentrations at the penetration connections (i.e. gas
vent boots to FML) resulting in the shearing or tearing of the
FML.
1316
-------
* The flatter (3% to 5%) landfill slopes can change significantly
with time, negating careful contouring and drainage provisions.
The result could be failure of the drainage layer or vegetative
cover.
* Disruption of the leachate or gas collection systems.
The major mechanisms of refuse settlement are as follows: <27> <28>
* The mechanical consolidation or void ratio reduction by
distortions, bending, crushing, and material reorientation.
* Raveling or the movement at fines into large voids.
* Physical-chemical changes from corrosion, oxidation and
combustion.
* Bio-chemical decomposition from fermentation and decay.
The refuse settles from both its own weight and the final cover components.
If the natural foundation material under the waste fill is composed of clayey
soil types, the foundation consolidation will contribute to the overall
settlement of the final cover. Traditional settlement analyses based upon on-
site soil characteristics and loading conditions can be used to estimate the
foundation component of the settlement of the final cover. It is important to
note that many clean-up sites have a combination of remedial technologies.
Ground water pump and treatment systems are often coupled with a RCRA/CERCLA
final cover. In this case, the effects of the ground water extraction system
on foundation settlement must also be determined.
The factors affecting the magnitude of settlement are many and are often
influenced by each other <27>. These factors include:
* refuse type or characterization (i.e. construction debris vs.
municipal wastes)
* refuse density or void ratio
* content of decomposable materials
* waste fill depths
* weight of final cover components
* stress history (landfill operational history)
* leachate levels
* environmental factors such as moisture content, temperature, and
gases present
* water table location
Sowers <28>, Yen and Scanlon <29> and others have developed methods to
estimate the settlement refuse of material. Mechanical settlement occurs
rapidly and is complete in essentially a month <28> and is a function of
compression index (related to the void ratio) of the refuse material and the
consolidation pressures. The combination of mechanical secondary compression,
physical/chemical action, and bio-chemical decay causes settlement to continue
with time. The rate of this secondary settlement is a function of the
secondary compression. Predesign information, such as historic settlement
surveys of the landfill surface is extremely useful in verifying design
assumptions.
When either settlement of the waste fill or foundation is critical, pre-
loading or surcharging can be used to preconsolidate. After time, the
1317
-------
surcharge fill can be reshaped and the final cover components completed.
Large scale pilot tests may be necessary.
C. Stability Considerations. Final covers over landfills or
contaminated soils must remain stable through the 30 year design life and
beyond. All portions of the system must be stable including the natural
foundation materials below and beyond the landfill, the refuse material and
the multi- layered components of the cover. Slope stability failures could be
catastrophic both economically and environmentally.
Stability analysis of landfills and final covers are complicated in part by
the following issues:
* The geological conditions at any site are unique.
* The geotechnics of the landfill materials are normally
highly variable within an individual site and also vary with time. In
addition, geotechnical investigations of landfills are rarely undertaken and
quantification of the geotechnical properties of waste materials is very
difficult <20>.
* Design procedures and guidance have not kept pace with the
rapid development of the wide variety of new materials used in cover designs.
* There are sometimes a lack of adequate test data and test
methods to confidently allow the use of new materials.
The following stability issues should be addressed in a cover design:
1) Cover Component Interfaces. During the past two years over
two dozen cover failures have occurred in the United States as a result of
surface sliding on geomembrane or other low friction interfaces of the cover
system <33>. The geosynthetic liner layer interfaces normally control the
design sideslopes of a cover rather than the stability of the waste fill mass
or foundation. The frictional resistance of all layer interfaces must bi2
analyzed. The controlling interfaces will likely be the geonet/FML
geonet/geotextile or FML/clay. Inclusion of a geotextile bedding beneath the
drainage layer can be used to increase friction values and to prevent
intrusion, by deformation, of the FML into the net or grid of the drainage
layer <1>. Geocomposite systems and textured geomembranes can be used to
improvement frictional resistance. The sliding resistance of the interface
layers must take into account long term creep of the geosynthetics . The
stability of the material interfaces should be designed based upon frictional
resistance between the material interfaces plus a factor of safety. The
geosynthetic components of final covers should not be designed in tension.
With proper detailing and material selection, the stabilizing effects of
anchor trenches and drainage benches can add to margin of safety and prevent
localized and long term failures.
It is imperative that design analyses be based upon friction values that are
specifically determined for each project using samples of actual materials and
reflecting representative placement, loading, and wetting conditions <23>.
For example, interfaces between the geomembranes and compacted clay may be
critical, and their shearing resistance may also be extremely sensitive to the
compaction conditions <23>. Clay barriers are compacted at high moisture
contents being wet of the line of optimum. Soils compacted to the wet of
optimum have lower friction values than the same material compacted with
lesser water contents. Another issue is the effect of a film of water that
1318
-------
develops between the geomembrane and the clay interface such that there is a
possibility that increased pore pressures could result with a corresponding
reduction strength. The interface wetting effects, consolidation conditions,
grid orientations, and the surface texture and cleanliness of geomembranes may
all affect frictional resistance <23>.
2. Waste Fill Mass and Foundation Stability. After the slopes
are selected based upon layer interface friction requirements, the overall
stability of the waste fill mass and foundation need to be analyzed.
Traditional slope stability methods can be used to assess foundation
stability. It is difficult to determine the geotechnical characteristics of
refuse material. Observation of existing slopes of the refuse and back
calculating available strengths have been used in determining the slope
stability of landfill masses <24>. Seismic considerations should be addressed
where applicable in the slope stability evaluation.
3. Other Stability Issues. Boschuk <22> in his review of more
than 20 cover failures, identifies several other stability issues not
described above. The effects of seepage forces resulting from infiltration in
the select fill layer needs to be considered in cover designs. The effects of
desiccation cracking and the corresponding transfer in load to the
geosynthetics is a possible stability concern. Boschuk <22> continues noting
that static shear strength parameters do not address seismic conditions,
freeze/thaw effects, long-term rainfall events, biological and soil clogging,
construction stress, and long-term stress relaxation and creep and stress
transfer in geotextiles. Gas uplift forces under geomembranes can further
reduce stability. Leachate trapped under the low-permeability layers of the
final cover can create a hydrostatic pressure head which can reduce stability
or fail the cover. Tension cracks in the select fill can allow a direct path
for surface runoff to infiltrate the soil in sufficient quantities where
hydrostatic pressures build up leading to instability of the soil cover.
D. Grading Requirements. The grading plan(s) for the final cover can
be developed after the following design considerations have been completed:
* Topographic mapping of the landfill area and beyond is
available.
* The limits of the landfill have been defined.
* After considering health and safety requirements, can the
landfill material be partially graded to minimize random
fill?
* Minimum fill requirements and layer thicknesses have been
determined.
* The maximum or design sideslope has been determined based
upon the stability analyses.
* The initial settlement analyses has been estimated.
* The drainage terraces have been sized, spaced and sloped to
drain.
* Gabion drop structures have been sized and located.
Development of the grading plan is an iterative process where settlement is a
function of fill height but fill height is not known until the final grading
plan is complete. The grading plan should be well defined by horizontal and
vertical control such that the cover grades can be staked in the field without
any scaling from the drawings. The final slopes must reflect minimum grade
requirements of 3-5% (after settlement) to accommodate both internal and
surface drainage requirements. The final slopes must reflect the stability
1319
-------
analysis. Maximum slopes should not exceed grades steeper than IV: 3H to
assure maintenance safety. The grading plan should also identify perimeter
ditches. Grading plans are normally developed for the top of random fill, the
top of the clay barrier layer and for the top of top soil. Each of these
layers should be surveyed after construction. Development of the grading plan
of the random fill must meet the above criteria while minimizing fill
quantities.
E. Hydrologic and Hydraulic Design Considerations.
1. HELP Modeling. After the final cover layers have been
tentatively selected, the Hydrologic Evaluation of Landfill Performance (HELP
II) Model <10> can be used to assess the amount of infiltration which would
penetrate into the refuse material. The Model also predicts amounts of
surface runoff, subsurface drainage and leachate that results from operation
of the final cover. The program models the effects of hydrologic processes
including precipitation, surface storage, runoff, infiltration, percolation,
evapotranspiration, soil moisture storage, and lateral drainage using a quasi-
two-dimensional approach <10> <31>.
2. Cap Internal Drainage.
a. Infiltration Drainage. The final cover's internal
drainage system consists of a drainage layer, a perforated pipe collection and
conveyance system and exit or toe drains. Perforated collection pipes with
point source outlets should be used instead of a continuous outlet at the toe
of the final cover.
b. Leachate Control. As stated in the predesign
discussion, landfills are normally quite pervious and can have significant
amounts of leachate. Leachate seeps exiting from the landfill surface need to
be identified and located during predesign activities. A leachate collection
blanket being either granular fill or a geonet coupled with a conveyance pipe
and outlet is required to control leachate levels. Uncontrolled leachate
levels can build-up hydrostatic pressures behind the low-permeability layers
resulting in decreased stability of the cover system or failure.
3. Final Cover Surface Drainage and Erosion Control. Proper
design of a final cover includes assuring that surface runoff is drained off
of the cover in a manner where erosion of the cover materials is controlled.
Erosion of the final cover is controlled by the vegetative cover (discussed
previously), drainage terraces and armored drop structures. The surface
sdrainage system must be capable of conveying runoff across the cover without
creating rills and gullies. The erosion control features should be designed
so that little long term maintenance is required. In non-level terrain,
diversion structures should be installed to prevent the run-on of surface
water onto the cover <1>. Temporary erosion control measures during
construction such as silt fences and straw bales is integral to any design.
a. Terraces. Slopes greater than 5 percent, are likely
to promote erosion unless controls are included in the design <1>, Terraces
are used to reduce erosion, reduce sediment content in runoff water, intercept
and conduct surface runoff of a nonerosive velocity to a stable outlet or cirop
structure. The Omaha District has been specifying terraces that are 10 feet
wide with a reverse slope of 1V:10H being one foot deep. A hydrologic
evaluation is required to determine the surface runoff from the cover. The
terrace should have enough capacity to control the design runoff event (25
1320
-------
year to 100 year rainfall frequency). The maximum flow velocity must be
analyzed. Flow velocity is dependent upon channel slope and discharge
quantities. The maximum nonerosive flow velocity for average soils is 2 feet
per second. Riprap, erosion control mats and gabions can be used to armor the
sideslope and bottom of terraces in order to resist erosive flow velocities.
The length of drainage terraces is controlled by capacity and the nonerosive
velocity requirement. The maximum spacing between terraces can be determined
with the Universal soil loss equation <3>. The Soil Conservation Service <34>
has developed methods to determine the maximum vertical spacing between
terraces.
b. Drop Structures. All terraces must have adequate
outlets <34>. Terraces normally discharge into central collection ditches or
drop structures that drop down the steep sideslope of a cover. Depending upon
the gradient of the cover sideslope, the drop structure will be constructed of
either erosion control mats, riprap or gabions. As with the terraces, the
drop structures have to be hydraulically sized and designed. A stilling basin
at the bottom of the drop structure being at the toe of the final cover will
be required to dissipate flow velocities in order to discharge the surface
runoff off-site. Drop structures or perimeter ditches may also be required at
the abutment contacts if surface runoff is directed from off-site towards the
final cover.
c. )ff-site Discharge. It is important to note, that if
a final cover functions as designed, there will be an increase in both the
total volume and the peak discharge of surface runoff leaving the site. The
impact to the receiving stream of increasing runoff volumes and peak
discharges off of the final cover should be a design consideration.
d. Floodplain Considerations. Several issues should be
considered if a portion of the final cover is located in a floodplain. First,
does the fill material of the final cover encroach into the zoned 100 year
floodway such that river stages are raised over one foot and flood damages are
induced? Second, could streambank erosion attack the fill material of the
final cover such that streambank erosion control measures are needed?
F. Borrow Areas. The availability of on-site borrow materials should
be evaluated in the Feasibility Study or Predesipn stages. On-site borrow
will normally result in substantial cost savings over off-site materials.
Off-site materials must normally be purchased by the contractor and hauled to
the site. On-site borrow avoids both of these costs. In addition, hauling
large quantities of materials to the project location normally stresses
transporation routes and is usually a public concern. If on-site borrow is
availabile, predesign investigations are requried to map the area and define
the nature and extent of the borrow source. A borrow area grading plan is
required in the plans along with profiles showing excavation limits and
subsurface features. Haul roads from the borrow site to the landfill location
must also be assessed. Borrow areas can be used to mitigate wetlands or other
environmental resources.
G. Cover Penetrations. Penetrations through the flexible geomembrane
by rigid and relatively fixed gas vents, drainage pipes, leachate collection
clean-out risers, piezomenters, monitoring wells and other structures should
be minimized. Where a penetration is necessary, it is essential to obtain a
secure, liquid-tight seal between the structuure and the geomembrane to
prevent leakage of water around the structure <1>. The connection of the
1321
-------
flexible geomembrane to fixed and rigid structures must also be flexible
because differential settlement or any downslope movement of the cover soil
will create stress at the connection resulting in either the streching or
tearing of the geomembrane. The geomembrane boots that are currently used to
connect the geomembrane to structures do not allow for such movement. As
Jaros <5> notes, for CERCLA landfills, where the location, rate, and magnitude
of differential settlements are unknown, additional emphasis is required in
designing more flexible connections.
H. Instrumentation requirements for Post Closure Monitoring <35>.
The monitoring time frame for a RCRA closure is 30 years. Key monitoring
parameters include groundwater, leachate generation, air quality, gas lateral
migration, settlement, slope stability, surface erosion, biotic intrusion and
cover effectiveness. It is necessary to incorporate the proper
instrumentation into the cover design and construction in order to monitor
these parameters of concern. Baseline conditions must be measured either
prior to or immediately after construction depending upon the parameter of
concern. Consistent and accurate record keeping during the post closure
period is essential.
Ground Water monitoring wells are normally placed both up and down gradient
from the landfill and final cover. Baseline index parameters are taken prior
to construction of the final cover. The ground water is sampled and monitored
during the post closure period. It may also be necessary to abandon or raise
existing monitoring wells where fill material will cover the wells.
Leachate seep discharge areas should be monitored at the collection discharge
outlets for flow quantity versus time. Leachate seep discharge should
decrease with time unless there is a failure in the low-permeability liner
system. The concentration of leachate with time can also be monitored.
Piezometers can be installed to monitored leachate levels beneath the final
cover.
Landfill gas concentrations should be monitored for both the underground
lateral movement of the gas and for air quality at the vent outlet locations.
Regarding the underground lateral movement of gas, gas monitoring stations
should be located around the perimeter of the landfill between any development
or area of concern. The lower explosive limit of gas is the parameter of
concern. The monitoring stations should be in place prior to placement of the
low-permeabiliity layers. The sampling frequency should be at least twice a.
day when the FML is being placed or when the ground is frozen. Air quality
should be monitored on the final cover surface for toxic landfill gases
emitted from the vent system. The contaminatnt levels of methane and other
landfill gasses should be monitored with time and compared to the threshold
limit values of the contaminants. For passive systems, internal gas pressures
may be a parameter of concern. Pressure cells can be used to measure gas
pressures.
Subsidence is a critical parameter to monitor because of the unseen damage
differential settlement can cause to the clay barrier, the FML and other
geosynthetics, penetration connections, drainage provisions, slope stability
and to the leachate and gas collection systems. Normally settlement markers
are installed on the final cover above the FML to monitor surface settlement.
Methods are available to monitor foundation settlement and refuse settlement,
if required. However, instrumentation needed to monitor these parameters
requires intrusive effects into the refuse and cover penetrations.
1322
-------
Slope stability can be monitored with visual inspection, surface movement
markers and possibly inclinometers. Inclinometers are used to measure
horizontal movements with depth. Installation of a inclinometer would require
a cover penetration and should be used only where stability considerations are
critical. Movement markers should be located on the steepest slopes of the
final cover and surveyed annually to the nearest one-hundreth of a foot.
The vegetative cover, drainage terraces, ditches and drop structures should be
inspected annually in order to assure that there are no formations of erosion
rills and gullies. The final cover surface should also be inspected for
biotic intrusion and volunteer vegetation.
The effectiveness of the final cover is dependent upon the long-term operation
of the drainage system and low-permeability layers. The outlets of the
drainage system can be monitored. Lysimeters can be installed below the low-
permeability layers to spot monitor leakage. Piezometers monitoring leachate
levels should drop with time. The ground water wells (water quality)
ultimately monitor the overall effectiveness of the final cover.
IV. SPECIAL FEATURES
There are many other features that must be addressed during the design of a
final cover. These features are an integral component of a final cover
design. Items such as the acquistion of construction easements and project
right-of-way are critical and time consuming. These special features are
identified below:
A. Decontamination Facilities
B. Access Routes (Road and Rail)
1. Video tape of existing roads
2. Traffic regulation requirements
a. Load limits on public access routes
b. Highway safety
3. Cap perimeter road
4. Maintenance requirements during construction
5. Road surface rehabilitation requirements after construction
6. Access requirements after construction
C. Staging Areas
1. Support facilities (i.e. construction trailers, lay-down
areas, etc.)
2. Parking areas
D. Security Fencing
E. Utilities
1. Existing location and availability
2. Utility relocation considerations
F. Easements and Right-Of-Way Requirements
G. Phasing Requirements (Order of Work)
H. Operation and Maintenance Requirements
I. Demolition (if required)
J. Material Handling
K. Chemical Quality Data Management
L. Health & Safety
1323
-------
M. Project Camera
N. Pre and Post Construction Aerial Photos
0. Disposal of Cleared, Grubbed and Demolished Material (Hazardous
or Not)
V. POTENTIAL LIST OF DRAWINGS
The following is a list of drawings that would normally be included in a set
of plans for the construction of a final cover:
Cover Sheet
Index of Drawings
Vicinity Map (Large Scale and State map)
Location Map (Small Scale-Nearest Town to Project)
Existing Site Conditions Including Utilities
General Plan
Contractor Access Plan
Horizontal and Vertical Control
Demolition Plan (If Required)
Safety Work Zone Plan (Site Control Plan)
Cap Initial Grading Plan
Cap Low Permeability Clay Liner Grading Plan
Cap Final Grading Plan
Erosion Control Plan (Temporary)
Cap Cross Sections
Gabion Channel Cross Sections and Details
Cap Detail Drawings; Anchor Trench, Collection Pipes and Toe Drains.
Gas Vent, Settlement Monument, Benchmark and Penetration Details
Wash-Down Area Cross-Sections and Details
Monitoring Well Details
Leachate Control Plan and Details
Piezometer Details
Chain Link Fence Details
Borrow Area Grading Plan, Sections and Soil Test Data
Boring Location Plan
Record of Borings (Geological Profile Sheets)
New Utility Drawings (If Required)
New Access Road Profiles and Sections (If Required)
Project Right-of-way Map
VI. POTENTIAL LIST OF SPECIFICATION SECTIONS The following is a list of
specification sections that would normally be included in a set of
specifications for the construction of a final cover.
DIVISION 1 GENERAL REQUIREMENTS
01100 Special Clauses
01200 Warranty of Construction
01201 On-Site Camera
01300 Environmental Protection
01401 Safety, Health and Emergency Response
01402 Chemical Quality Management
01500 Decontamination and Disposal
01501 Summary of Work
01600 Temporary Utilities and Controls
01610 Support Facilities
01620 Security
1324
-------
01700 Measurement and Payment (Optional-Can Be In Technical
Specifications)
DIVISION 2 SITE WORK
02050 Demolition (If Required)
02060 Well Abandonment (If Required)
02100 Clearing and Grubbing
02150 Hazardous Material Excavation and Handling (If Required)
02210 Grading
02215 Geotextile Filter
02220 Test Fill Sections
02222 Wire Mesh Gabions
02243 Crushed Rock Surfacing (If Required)
02244 Low Permeability Clay Liner
02246 Flexible Membrane Liner for Cap Systems (FML)
02248 Cap System Drainage Layer (Gravel Option)
02250 High Density Polyethylene Drainage Net (Geonet Option)
02251 Geogrid Reinforcement Material (If Required)
02252 Gas Venting System (If Required)
02420 Temporary Erosion and Sediment Controls
02435 Permanent Surface Water Controls
02444 Chain Link Security Fence and Gates
02475 Sodding (If Required)
02480 Seeding
02600 Roadways and Parking Areas
02900 Site Maintenance
02910 Monitoring Wells
02913 Demobilization and Project Close Out
C2915 Piezometers
02920 Post-Construction Maintenance Activities
VII. DESIGN ANALYSIS. A design analysis in prepared to document design
assumptions and procedures for all project features.
VIII. QUANTITY TABULATION SHEET
Item
1. Seeding
2. Top Soil
3. Select Fill
4. Filter Layer
a. Geotextile Fabric
Alternative
b. Graded Granular
Layer Alternative
5. Drainage Layer
a. Geonet Alternative
b. Gravel Layer
Alternative
Measurement
Acres
Cubic Yards or Tons
Cubic Yards or Tons
Square Yards
Tons
Square Yards
Tons
1325
-------
6. Synthetic Barrier
Layer
7. Clay Barrier
Layer
8. Gas Control System
a. Collection System
1) Granular blanket option
2) Granular trench option
3) Geosynthetic material option
4) Wells option
b. Pipe Conveyence
c. Vertical Vent Pipes
d. Treatment System
9. Random Fill
10. Clearing
11. Proof-rolling ILandfill Surface
12. Landfill Excavation or reshaping
13. Decontamination Facility
14. Security Fencing
15. Operation and Maintenance
16. All Other Items
Square Yards
Cubic Yards
Tons
Tons
Tons
Linear Feet
Linear Feet
Lump Sum/Each
Lump Sum
Tons
Acres
Acres
Cubic Yards
Lump Sum
Linear Feet
Lump Sum
1326
-------
REFERENCES
<1> Environmental Protection Agency. July 1989. Technical Guidance
Document: Final Covers on Hazardous Waste Landfills and Surface Impoundments.
EPA/530-SW-89-047.
<2> Environmental Protection Agency. August 1989. Seminar Publication:
Requirements for Hazardous Waste Landfill Design, Construction, and Closure.
<3> Goldman Etal. 1986. Erosion & Sediment Handbook. McGraw Hill
<4> U.S. Army Corps of Engineers. 1978. Engineering Manual: Design and
Construction of Levees.
<5> Geosynthetic Research Institute. Proceedings of the 4th GRI Seminar.
CERCLA Landfill Closure; Construction Considerations. David L. Jaros.
<6> Koerner, R. M. Designing with Geosynthetics, 2nd Edition. Prentice
Hall, Englewood Cliffs, NJ 1990.
<7> Geosynthetic Research Institute. Proceedings of the 4th GRI Seminar.
Geotextiles in Landfill Closures. Barry R. Christopher.
<8> Proceeding from 2nd International Conference on Geotextiles. August
1982. Filter criteria for Geotextiles. J. P. Giroud.
<9> Guide Specification. Section 02215. Geotextile Filter
<10> Hydrologic Evaluation of Landfill Performance (HELP II) Model
(Schroeder, and others 1988)
<11> AASHTO-ABC-ARBTA. Task Force #25. Specifications for Geotextiles
<12> Proceedings from 2nd International Conference on Geotextiles.
Evaluation of U.S. Army Corps of Engineers Gradient Ratio Test for Geotextile
performance. Haliburton, T.A. and Wood, P.O.
<13> Proceedings from 2nd International Conference on Geotextiles.
Laboratory Studies on Long-Term Drainage Capability of Geotextiles. Koerner,
R.M. and Ko, F.K.
<14> Cedergren. H.R., Seepage, Drainage and Flow Nets. Second Edition. John
Wiley and Sons.
<15> U.S. Army Corps of Engineers. Engineering Manual for Seepage Analysis
and Control for Dams.
<16> Environmental Protection Agency. July-August 1990. Seminar on Design
and Construction of RCRA/CERCLA Final Covers. Geosynthetic Design for
Landfill covers. Robert M. Koerner
<17> Giroud, J. P. and J. E. Fluet. "A Short Course in Landfill Lining
Systems: Design and Installation". Presented through the Lehigh University
Office of Continuing Education. May 19-20, 1988.
<18> Geosynthetic Research Institute. GRI Test Methods & Standards.
1327
-------
<19> Conversation with David Jaros, U. S. Army Corps of Engineers, Missouri
River Division, Omaha, NE.
<20> ASTM 1070. Geotechnics of Waste Fills. Geotechnics of Waste Fill.
Arvid 0. Landua and Jack I. Clark.
<21> Geosynthetic Research Institute. Proceedings of the 4th GRI Seminar.
Composite Lining System Design Issues. Richard T Von Pein and Sangeeta
Prasad.
<22> Geotechnical Fabrics Report. March 1991. Landfill Covers an
Engineering Perspective. John Boschuk Jr.
<23> ASTM 1070. Geotechnics of Waste Fills. Stability Considerations in the
Design and Construction of Lined Waste Repositories. James K. Mitchell,
Raymond B. Seed, and H. Bolton Seed.
<24> ASTM 1070. Geotechnics of Waste Fills. Evaluation of the Stability of
Sanitary Landfills. Sukhmander Singh and Bruce J. Murphy.
<25> Environmental Protection Agency. Settlement and Cover Subsidence of
Hazardous Waste Landfills. W. L. Murphy and P. A. Gilbert. U. S. Army
Engineers Waterways Experiment Station.
<26> ASTM 1070. Geotechnics of Waste Fills. Settlement and Engineering
considerations in Landfill and Final Cover Design. Derek V. Morris and Calvin
E. Woods.
<27> ASTM 1070. Geotechnics of Waste fills. Settlement of Municipal Waste.
Tuncer B. Edil, Valeri J. Rangvette and William W. Wuellner.
<28> Sowers, G. F. Settlement of Waste Disposal Fills. Proceedings 8th
International Conference on Soil Mechanics and Foundation Engineering, Moscow.
1973.
<29> Yen, B. C. and Scanlon B., Sanitary Landfill Settlement Rates, Journal
of the Geotechnical Engineering Division, ASCE. May 1975.
<30> Environmental Protection Agency. July-August 1990. Seminar on Design
and Construction of RCRA/CERCLA Final Covers. Gas Management Systems. Paul R.
Schroeder.
<31> U.S. Army Corps of Engineers. final Revised Design Analysis. Delaware
Sand and Gravel Superfund Site. May 1989. Revised August 1989.
<32> Geosynthetic Research Institute. Proceedings of the 4th GRI Seminar.
Geosynthetics in Landfill Closures Design Considerations. Michael T. Feeney.
<33> Geosynthetic Research Institute. Proceedings of the 4th GRI Seminar.
Geogrid Reinforcement in Landfill Closures. R. G. Carroll, Jr. and Vicky
Chourey-Curtis.
<34> Soil Conservation Service. Engineering Standard. Terrace. 600-1. March
1983.
1328
-------
<35> Environmental Protection Agency. July-August 1990. Seminar on Design
and Construction of RCRA/CERCLA Final Covers. Post Closure Monitoring.
Gregory N. Richardson.
<36> Environmental Protection Agency. July-August 1990. Seminar on Design
and Construction of RCRA/CERCLA Final Covers. Critical Factors in Soils
Design for Covers. David E. Daniel.
1329
-------
The Challenge of Treating Superf und Soils: Recent Experiences
Carolyn K. Offutt
U.S. Environmental Protection Agency
Office of Solid Waste and Emergency Response
Washington, D.C. 20460
Joan O'Neill Knapp
CDM FEDERAL PROGRAMS CORPORATION
Fairfax, Virginia 22033
(Author(s)' Address at end of paper)
The treatment of Superf und soils is a challenging technical issue that is currently being addressed by
a variety of different groups and programs within EPA. The Superfund Amendments and
Reauthorization Act (SARA) of 1986 specifies a preference for permanent treatment of waste sources
such as soil and there is a further preference for the selection and use of innovative technologies to
accomplish permanent reduction in toxicity, mobility, or volume.
This paper covers five related areas presenting a total picture of this challenging issue including:
Why is treating soils a Superfund priority?
What unique considerations of these soils make their treatment challenging?
What technologies will be effective at treating Superfund soils?
What are the considerations for selecting treatment technologies for Superfund sites?
What technology transfer mechanisms exist regarding soil treatment technologies?
Superfund soils have unique physical characteristics compared to the characteristics and requirements
for the treatment/disposal of other industrial process wastes. The need to treat these contaminated
soils has led to interesting research and demonstrations of treatment technologies.
This paper discusses currently available and innovative treatment technologies including low
temperature thermal desorption, chemical extraction, bioremediation, soil washing, stabilization, and
high temperature thermal treatment. In addition to a summary of how each technology is employed,
both the applicability as well as the problems experienced with each technology are summarized and
supplemented with examples from recent and ongoing Superfund treatment experiences. Sponsors
of completed or ongoing treatability tests will be asked to submit data for the data base being
developed for an EPA project.
INTRODUCTION
Section 3004(m) of the Resource Conservation and Recovery Act (RCRA) mandates that the EPA
require treatment of hazardous wastes prior to land disposal. Known as the "Land Disposal
Restrictions" (LDRs), these regulations may apply to hazardous industrial process wastes as well as
contaminated soil, sludge and debris from Superfund and RCRA facilities that are destined for land
disposal.
The 1989 Superfund Management Review (also known as the 90-Day Study) by the Office of Solid
Waste and Emergency Response (OSWER) acknowledged that Superfund response actions may not be
able to meet existing RCRA treatment standards based on "best demonstrated available technology"
(BDAT) under the LDRs. The existing LDR regulations may limit the potential treatment
technologies available for Superfund clean-ups, with technologies such as soil washing, stabilization,
1330
-------
and biological treatment, being precluded because they may not meet the highest level of performance
required by LDRs. In contrast, the 90-Day Study encouraged the greater use of innovative
technologies and urged the reduction of non-technical barriers, such as regulatory and policy
constraints, that inhibit the use of treatment technologies, while preserving the intent and spirit of
applicable RCRA regulations.
OSWER recognized the potential limitation on treatment technologies for Superfund actions and
developed a process to use LDR treatability variances for soil and debris. Guidance was issued to the
Regions through the Superfund LDR Guide 6A, "Obtaining a Soil and Debris Treatability Variance
for Remedial Actions," (OSWER Directive 9347.3-06FS) in July 1989 and revised in September 1990
(1). Superfund LDR Guide 6B, "Obtaining a Soil and Debris Treatability Variance for Removal
Actions," (OSWER Directive 9347.3-07FS) was issued in December 1989 and revised in September
1990 (2). These guides describe the treatability variance process, include alternate treatment levels
to be obtained under treatability variances, and identify treatment technologies which have achieved
the recommended levels.
A memorandum issued on November 30, 1989 by OSWER entitled the "Analysis of Treatability Data
for Soil and Debris: Evaluation of Land Ban Impact on Use of Superfund Treatment Technologies,"
(OSWER Directive 9380.3-04) provides support for decisions by the Regions to use treatability
variances, when appropriate (3). The analysis identifies some of the key technical considerations to
be evaluated in obtaining a treatability variance.
OSWER recognizes that the use of treatability variances represents an interim approach and is actively
in the process of acquiring additional data for developing separate treatment standards for
contaminated soil and debris.
The collection of data which supports the development of regulations for contaminated soil and debris
is a joint effort by the OSWER's Office of Emergency and Remedial Response (OERR), Office of
Solid Waste (OSW), and Technology Innovation Office (TIO), and the Office of Research and
Development (ORD) Risk Reduction Engineering Laboratory (RREL). The initial data collection
effort by the OERR that produced the data for the development of the treatability variance levels also
identified the types of data needed to develop treatment standards for soil. These initial data are
summarized in the "Summary of Treatment Technology Effectiveness for Contaminated Soil" (4). This
paper describes both the conclusions drawn by OERR to date as well as the unique considerations of
soil treatment which the Superfund program is investigating further. Ongoing research activities are
also described.
ANALYSIS OF TREATMENT EFFECTIVENESS
OERR launched an extensive effort in 1987 and 1988 to collect existing data on the treatment of soil,
sludge, debris, and related environmental media. The results from several hundred studies were
collected and reviewed.
All applicable treatment information from the best documented studies was extracted, loaded into a
data base, and analyzed to determine the effectiveness of technologies to treat different chemical
groups (4).
Based on this analysis, a number of technologies commonly used in the Superfund program provide
substantial reduction in mobility and toxicity of wastes as required in Section 121 of the Superfund
Amendments and Reauthorization Act (SARA) of 1986. For example:
1331
-------
o Thermal destruction has been effective on all organic compounds, usually
accomplishing well over 99% reduction of organics.
o Although the data indicate that PCBs, dioxins, furans, and other aromatic compounds
have been dechlorinated to approximately 80%, more recent data indicate that removal
efficiencies may approach 99.9%.
o Bioremediation successfully treats many halogenated aliphatic compounds,
non-halogenated aromatics, heterocyclics, and other polar compounds with removal
efficiencies in excess of 99%.
o Removal efficiencies for low temperature thermal desorption have been demonstrated
with averages up to 99% for non-polar halogenated aromatics and with treatment
often exceeding 90% for other polar organics.
o Soil washing and chemical extraction data on organic compounds indicate average
removal efficiencies of approximately 90% for polar non-halogenated organics and
99% for halogenated aromatics, with treatment often exceeding 90% for polynuclear
aromatics. The soil washing process, with optimized solvent selection, has
demonstrated removal efficiencies often exceeding 90% for volatile and non-volatile
metals.
o Immobilization can achieve average reductions in mobility of 93% for volatile metals,
with reductions in mobility often exceeding 90% for non-volatile metals.
Immobilization processes, while not actually destroying the organic compounds,
reduce the mobility of contaminants an average of 99% for polynuclear aromatic
compounds. Immobilization may not effectively stabilize some organic compounds,
such as volatile organics, and the long-term effectiveness of immobilization of
organics is under evaluation.
CONCLUSIONS REGARDING SOIL TREATMENT TECHNOLOGY EFFECTIVENESS
Contaminated soils can be treated via three basic mechanisms: (1) destruction of the contaminants
through alteration to a less toxic compound; e.g., thermal destruction, dechlorination, bioremediation;
(2) physical transfer and concentration of the contaminants to another waste stream for subsequent
treatment or recovery; e.g., low temperature thermal desorption and chemical extraction, soil washing;
and (3) permanent bonding of the contaminants within a stabilized matrix to prevent future leaching;
e.g., immobilization and vitrification. In general, the destruction technologies are effective in
reducing the toxicity of many organic contaminants. The physical transfer technologies reduce the
toxicity and often the volume of selected organic and inorganic contaminants. While the bonding
technologies are most effective at reducing the mobility and, therefore, the toxicity of inorganic
contaminants, some increasing effectiveness is being demonstrated on selected organic contaminants
as well. Figure 1 presents a summary of these basic conceptual conclusions. A more detailed
discussion follows.
The technologies that have been widely demonstrated on soils are thermal destruction for organic
contaminants and immobilization for inorganic contaminants. While these two technologies may be
highly effective in treating particular classes of compounds, neither provides an ideal solution to
complex mixtures of organic and inorganic contaminants, which are common at Superfund sites. The
inherent difficulty in treating contaminants in a soil matrix, where waste conveyance and mixing are
in themselves complicated unit operations, contributes to the need to find special solutions. Other
1332
-------
5
CO
•M:
I
a
'8
P
05
«*^
-3
s >-
co O
3
CO
a>
co
•<—i
•g
O
•ZJ
o
£
X
CO
CO
I
"c3
U
•
CO
4)
1
w
lonstrated
e
0>
Q
S P
P
"Sff
ntial Effe
re Data R
o> o
^•^ ^25
PI o
c
4)
•**
A
£
H
B
CA
.§
1333
-------
issues, such as landfill capacity and cost, cross-media impacts, and natural resource conservation, also
support the need to develop and use
alternative and innovative treatment technologies for contaminated soil.
Because of EPA's ultimate goal of developing LDRs for contaminated soil and debris, this study
evaluates a number of treatment options that are applicable to excavated soils. In-situ soil techniques,
such as some types of bioremediation, soil vapor extraction, in-situ immobilization, and combined
ground water and vadose zone soil treatment were not included in the scope of this evaluation.
In-situ techniques should also be considered when researching remediation measures i'or a
contaminated soil problem. When in-situ technologies are used at Superfund sites, the LDRs may not
be applicable because the waste has not been excavated and subsequently "placed" in a landfill or other
RCRA unit.
Based upon the data collected and evaluated by OERR from more than 200 soil treatment tests,
conclusions were developed regarding the effectiveness of six soil treatment technology groups for
each of eleven contaminant treatability groups. For destruction and physical transfer technologies
applied to organic contaminants, the removal efficiency was analyzed. This evaluation factor was
replaced by the reduction in mobility for the following technologies: immobilization, chemical
extraction, and soil washing. The principles of operation and the effectiveness of treatment on
organic and inorganic contaminants are presented below.
THERMAL DESTRUCTION
Principle of Operation
o Thermal destruction uses high temperatures to incinerate and destroy hazardous
wastes, usually by converting the contaminants to carbon dioxide, water, and other
combustion products in the presence of oxygen.
Effectiveness on Organics
o This technology has been proven effective on all organic compounds, usually
accomplishing well over 99% removal.
o Thermal destruction technologies are equally effective on halogenated,
non-halogenated, nitrated, aliphatic, aromatic, and polynuclear compounds.
o Incineration of nitrated compounds such as trinitrotoluene (TNT) may generate large
quantities of nitrous oxides.
Effectiveness on Inorganics
o Thermal destruction is not an effective technology for treating soils contaminated with
high concentrations of some metals.
o High concentrations of volatile metal compounds (lead) present a significant emissions
problem, which cannot be effectively contained by conventional scrubbers or
electrostatic precipitators due to the small particle size of metal-containing
particulates.
o Non-volatile metals (copper) tend to remain in the soil when exposed to thermal
destruction; however, they may slag and foul the equipment.
1334
-------
DECHLORINATION
Principle of Operation
Chemical dechlorination is a process that involves the removal of chlorine atoms from
chlorinated aromatic molecules by alkali metals, glycoxides, and hydrogen and
hydroxyl radicals. This destruction process converts the more toxic compounds into
less toxic products. The transformation of contaminants within the soil produces
compounds that are more readily degradable. An evaluation of the end products is
necessary to determine whether further treatment is required.
Effectiveness on Organics
PCBs, dioxins, furans, and other aromatic compounds (such as pentachlorophenol)
have been dechlorinated to approximately 80% removal, with more recent data
indicating that removal efficiencies may approach 99.9%.
Recent limited laboratory data have confirmed the applicability to other halogenated
compounds including straight-chain aliphatics (such as tetrachloroethene). The
removal of chlorine from aliphatics generally involves the removal of hydrogen.
Recently acquired data for halogenated cyclic aliphatics (such as dieldrin) indicate
that
dechlorination will be effective on these compounds as well.
When non-halogenated compounds or lower molecular weight halogenated compounds
are subjected to this process, volatilization may occur.
Effectiveness on Inorganics
Dechlorination is not designed to treat metals. High concentrations of reactive metals
(such as aluminum), under very alkaline conditions can increase the chemical
requirements and may affect the dechlorination process.
BIOREMEDIATION
Principle of Operation
Bioremediation is a destruction process that uses soil microorganisms including
bacteria, fungi, and yeasts to chemically degrade organic contaminants.
Effectiveness on Organics
Bioremediation appears to successfully treat many halogenated aliphatic compounds
(1,1-dichloroethane), non-halogenated aromatics (benzene), heterocyclics (pyridine),
and other polar compounds (phenol) with removal efficiencies in excess of 99%;
however, the high removal implied by the available data may be a result of
volatilization in addition to bioremediation.
More complex halogenated (p,p'-DDT), nitrated (triazine), and polynuclear aromatic
(phenanthrene) compounds exhibited lower removal efficiencies, ranging from
approximately 50% to 87%.
1335
-------
o Poly-halogenated compounds may be toxic to many microorganisms.
Effectiveness on Inorganics
o Bioremediation is not effective on metals.
o Metal salts may be inhibitory or toxic to many microorganisms.
LOW TEMPERATURE THERMAL DESORPTION
Principle of Operation
o Low temperature thermal desorption is a physical transfer process that uses air, heat,
and/or mechanical agitation to volatilize contaminants into a gas stream, where the
contaminants are then subjected to further treatment. The degree of volatility of the
compound
rather than the type of substituted group is the limiting factor in this process.
Effectiveness on Organics
o Removal efficiencies have been demonstrated by these units at bench, pilot, arid full
scales, ranging from approximately 65% for polynuclear aromatics (naphthalene), to
82% for other polar organics (acetone) and 99% for non-polar halogenated aromatics
(chlorobenzene).
Effectiveness on Inorganics
o Low temperature thermal desorption is not generally effective on metals.
o Only mercury has the potential to be volatilized at the operating temperatures of this
technology.
CHEMICAL EXTRACTION AND SOIL WASHING
Principle of Operation
o Chemical extraction and soil washing are physical transfer processes in which
contaminants are disassociated from the soil, becoming dissolved or suspended in a
liquid solvent. This liquid waste stream then undergoes subsequent treatment to
remove the contaminants and the solvent is recycled, if possible.
o Soil washing uses water as the solvent to separate the clay particles, which contain the
majority of the contaminants, from the sand fraction.
o Chemical extraction processes use a solvent which separates the contaminants from the
soil particles and dissolves the contaminant in the solvent.
Effectiveness on Organics
o The majority of the available soil washing data on organic compounds indicates
removal efficiencies of approximately 90% for polar non-halogenated organics
1336
-------
(phenol) to 99% for halogenated aromatics (chlorobenzene), with lower values of
approximately 71% for PCBs to 82% for polynuclear aromatics (anthracene).
The reported effectiveness for these compounds could be due in part to volatilization
for compounds with higher vapor pressures (such as acetone).
This process is least effective for some of the less volatile and less water soluble
aromatic compounds.
Effectiveness on Inorganics
The chemical extraction process, with optimized solvent selection, has demonstrated
removal efficiencies of 85% to 89% for volatile metals (lead) and non-volatile metals
(copper), respectively.
IMMOBILIZATION
Principle of Operation
Immobilization processes reduce the mobility of contaminants by stabilizing them
within the soil matrix, without causing significant contaminant destruction or transfer
to another medium.
Volatile organics will often volatilize during treatment, therefore an effort should be
made to drive off these compounds in conjunction with an emission control system.
Effectiveness on Organics
Reductions in mobility for organics range from 61% for halogenated phenols
(pentachlorophenol) to 99% for polynuclear aromatic compounds (anthracene).
Immobilization is also effective (84% reduction) on halogenated aliphatics
(1,2-dichloroethane).
Some organic mobility reductions of the more volatile compounds may actually be
removals as a direct result of volatilization during the exothermic mixing process and
throughout the curing period.
The immobilization of organics is currently under investigation, including an
evaluation of the applicability of analytical protocols (EP, TCLP, total analysis) for
predicting long-term effectiveness of immobilization of organics. The preliminary
available data indicate that significant bonding takes place between some organic
contaminants and certain organophilic species in the binding matrix; however,
immobilization may not effectively stabilize some organic compounds, such as volatile
organics.
Effectiveness on Inorganics
Immobilization can accomplish reductions in mobility of 81% for non-volatile metals
(nickel) to 93% for volatile metals (lead).
1337
-------
The effectiveness of the six technologies to treat soil was classified as having demonstrated
effectiveness, potential effectiveness, or no expected effectiveness for the eleven contaminant groups
(Figure 2). The ratings were based on removal efficiency, scale of operation, and potential for
adverse effects as follows:
o Demonstrated Effectiveness: A significant percentage of the data, at least 20%, is
from pilot or full scale operations, the average removal efficiency for all of the data
exceeds 90%, and there are at least ten data pairs.
o Potential Effectiveness: The average removal efficiency for all of the data exceeds
70%.
o No Expected Effectiveness: The average removal efficiency for all of the data is less
than 70% and no interference from the contaminants in the soil is expected.
o No Expected Effectiveness: Potential adverse effects to the environment or the
treatment process may occur. For example, high concentrations of metals may
interfere with biological treatment.
In some cases, a different rating was selected when additional qualitative information and engineering
judgment warranted. Two ratings were selected if the compounds within a treatability group were
so variable that a range of conclusions could be drawn for a particular technology.
Although some of the data upon which the analysis is based have limited quality assurance (QA)
information, the data, nevertheless, do indicate potential effectiveness (at least 90% to 99% reduction
of concentration or mobility of hazardous constituents) of treatment technologies to treat Super fund
wastes. Some reductions in organic concentrations or organic mobility of more volatile compounds
may actually represent the removal of those compounds as a direct result of volatilization.
Technologies where this is most likely to occur include dechlorination, bioremediation, soil washing,
or immobilization, and consideration of appropriate emission controls is required. Percentage removal
reductions (removal efficiencies) are not always a good measure of effectiveness, especially when high
concentrations remain in the residuals. Some of the performance observations are based upon a
relatively small number of data points and may not extrapolate well to the broad array of soils
requiring treatment.
QUANTIFYING TECHNOLOGY EFFECTIVENESS AND LIMITATIONS
TECHNOLOGY LIMITATIONS
A variety of potential limitations to the effective treatment of Superfund wastes were identified in
the analyses of data from OERR's original survey. The EPA offices of OERR, OSW, TIO, and ORD
are now working together to identify technology limitations and their impact on technology
effectiveness.
The data suggest that the treatment of soil and debris with organic contamination, by technologies
other than thermal destruction, will not be able to consistently achieve BOAT standards previously
developed for industrial process wastes. The difficulty in treating soil and debris is a direct result
of the levels of contaminants, the types/combinations of contaminants, the type of matrix, particle
size, and other physical and chemical characteristics of the soil and debris.
1338
-------
If
n
9
1
a\
II
tM
9
5|
O
*» w i|t
I i
O
O
.1
§2
ill
.•
i
i
IS
tutu
?u«< w
o>
-*-»
C3
.s
3
Is
O
U
IL,
C
O)
£
H
O)
4^
w
• *N
•e
o>
^
OX)
1339
-------
The residual concentrations in contaminated soil treated by technologies other than thermal
destruction is highly dependent upon the concentrations in the untreated soil. Therefore, when
evaluating technologies other than thermal destruction, the ability of those technologies to treat high
concentrations of organics should be considered. The number and types of contaminants must also
be carefully screened. Organic and inorganic contaminants may require different treatment
technologies, thus requiring a treatment train. In some cases, different technologies may be necessary
for soils and sludges.
In addition, the distribution of contaminants often is also very non-homogeneous and is dependent
on patterns of contaminant deposition and transport.
The complex nature of solid waste matrices, such as contaminated soil from a Superfund site, severely
complicates the treatment process. Soil is a non-homogeneous living medium, and the proportion of
clay, organic matter, silt, sand, debris, and other constituents can affect the treatability of a
contaminated soil. For example, the complex bonding forces that are exhibited by various soil
fractions, particularly clays and organic matter, can be difficult to counteract and can affect the
treatability of contaminated soil. To further complicate these circumstances, the age of many of these
sites has allowed significant opportunity for environmental weathering of the contaminants and the
medium.
Collectively, these conditions make the treatment of contaminated soil, weathered contaminated
("old") sludge, and debris a formidable technical challenge. EPA intends to quantify the effects of
these factors, and the approach is to analyze the existing treatment data for the effects of these
factors. Specific parameters affecting performance will be identified from existing data; parameters
include: soil morphology (particle size distribution), clay content, permeability, total organic carbon,
cation exchange capacity and as many as twenty other parameters. Differences in treatment
performance among different technologies, contaminants and soil and debris types will be
investigated.
SUPERFUND DATA COLLECTION AND RESEARCH APPROACH
EPA is in the process of developing the final regulations for contaminated soil and debris, and the
Superfund program has a second important goal--timely and thorough technology transfer. The
initiatives EPA has taken involve collecting all existing information on the treatment of soil and
debris to supplement the first data collection effort and conducting experimental tests, when
necessary, to better understand the process (Figure 3). The EPA OERR, ORD, OSW, and TIO are
working together in these efforts due to the complexity of effectively treating soil and debris.
Discussion of the initiatives follows.
Existing Data Collection
The targets for existing soil and debris treatment data include recent EPA remedial/removal actions,
Department of Defense (DOD) and Department of Energy (DOE) actions, Superfund Innovative
Technology Evaluation (SITE) program demonstrations, underground storage tank (UST) corrective
actions, and activities conducted by private research organizations and vendors. The information that
is being requested includes data on performance as well as other information important for technology
transfer. Parameters of interest include: contaminants treated, scale of the test, measured contaminant
concentrations before and after treatment, quality control (QC) protocols, design and operating
parameters of the treatment system, methods to improve performance and problems encountered in
treatment. The information that is collected will be entered in the Superfund Soil Data Management
System, (DMS) designed specifically for storing and managing this information.
1340
-------
GUIDANCE
DOCUMENT &
OSWER
DIRECTIVE
ORIGINAL SUPERFUND
SOIL TREATMENT
DATA BASE
ALTERNATE
TREATABILITY
VARIANCE
LEVELS
EXISTING
DATA
WEATHERED
SLUDGE
REVISED
DATA BASE
TREATMENT
TESTS
VARIABILITY
TECHNOLOGY
TRANSFER
Figure 3. Superfund Data Collection and Research Approach
1341
-------
Soil Treatment Tests
The treatment tests that are being performed are tests on soils, contaminants, and technologies that
lacked adequate treatment performance data where the technologies would be available for treating
contaminated soil and debris (CSD). Twelve treatment tests are currently planned on eight different
Superf und soils representing different soil types, contaminant types and concentrations, and treatment
technologies.
Because the variability of the soil matrix may have significant effects on the ability of a technology
to perform, EPA is especially interested in testing the effects of soil morphology or composition on
treatment technology performance. Preliminary data indicate that clayey soils are treated less
effectively than silty or sandy soils by some technologies. To evaluate this finding, pilot-scale
treatability tests will be conducted on three different soil types - sandy, silty, and clayey from the
eight different Superfund sites. Data generated from these treatability tests and the available
treatment data will be used to investigate the effect of soil type on treatment effectiveness.
The technologies that will be tested include slurry bioremediation, low temperature thermal
desorption, chemical extraction, soil washing, and stabilization. The technologies will be applied to
different types of contaminants as well. Soils with significant levels of poly nuclear aromatic
hydrocarbons (PAHs), pentachlorophenols, volatile organics, PCBs and metals, will be tested. The
stabilization technology may be tested as both a primary technology and as a secondary treatment
process for residuals.
The treatability tests will be conducted according to the EPA "Quality Assurance Program Plan for
Characterization Sampling and Treatment Tests Conducted for the Contaminated Soil and Debris
Program" (5) and site specific Sampling and Analysis Plans. The individual sampling plans specify
holding times, analytical methods, chain of custody, and quality control measures, such as blanks and
spikes. The tests will include measurements of contaminant concentrations before and after
treatment, and measurements of the waste characteristics that affect the performance of soil treatment
technologies. Examples of waste characteristics that affect treatment performance include but are not
limited to moisture content, oxidation/reduction potential, and particle size distribution; the
parameters that affect performance are listed in the QA Program Plan.
OERR recognizes that much of the soil and debris from Superfund sites contains mixtures of
contaminants and that individual contaminants may need to be treated differently. Treatment trains
may be utilized in these cases. EPA wants to know the types of technologies applied to mixtures of
contaminants and the effectiveness of the system. The major source of this type of data will be from
existing treatability data, however, several of the planned treatment tests may also involve treatment
trains. The treatment trains used in the tests will be a technology for treating the organic
contaminants followed by stabilization to treat the inorganics (metals) remaining in the soil residues.
Debris Treatment
Parallel with the effort to collect data on soil is an effort to collect existing information on the
characterization and treatment of debris. The first data collection effort obtained very limited data
on debris treatment. The studies indicated that debris could constitute as much as fifty percent of
the contaminated media, such as might be found at a wood preserving site. OERR also recognized
that the sampling procedures used to provide representative samples of debris contamination were not
well documented. Recognizing the importance of debris, EPA has implemented a comprehensive
review of debris sampling, analysis and treatment approaches. Some characteristics of debris that may
1342
-------
affect treatment include permeability and destructibility. The potential treatment technologies that
have been identified for debris to date are destruction, extraction, and immobilization.
Weathered Contaminated Sludge
The OERR data survey identified the existence of large quantities of weathered contaminated or "old"
sludges on Superfund sites. These sludges have aged or weathered, and are different than typical
RCRA sludges. The data on "old" sludge indicated that sludges are not consistently defined in the
literature. Furthermore, these sludges, when identified, had higher concentrations of contaminants
than soils, and as a result, did not meet treatability variance levels as frequently as soil. Of the OERR
survey data, 55% of the sludge treatment tests met variance levels, while 78% of the soil treatment
tests met variance levels. These results indicate that weathered contaminated sludge may require
separate treatment standards. In order to quantify the treatability of sludges for regulatory
development purposes, more data will be collected on the characteristics and treatability of sludges.
Existing data will be collected as part of the data collection effort, and characterization tests will be
conducted on sludges from Superfund sites to obtain the physical and chemical characteristics of
weathered contaminated sludge. A focused symposium will also be convened to discuss this timely
topic and to compile the experiences of others who have dealt with these wastes.
Variability
An additional factor which influences treatment performance is homogenization of the waste,
whether through materials handling, preprocessing, and or mixing within the treatment system. The
previous OERR data survey indicated that the degree of homogenization achieved can have important
effects on treatment performance and therefore the issue is being evaluated in the current research
approach.
A critical element in soil treatment is materials handling. Special approaches to waste transfer
throughout the treatment system are particularly important for solids and viscous sludges, where
traditional conveyance methods are frequently ineffective. Slugs of material or debris tend to jam
treatment equipment, resulting in breakage, downtime, and the potential for uncontrolled releases to
the environment.
The preprocessing of waste to maximize homogeneity and modify the waste characteristics is also
important to successful treatment technology operation. Any treatment technology will operate most
efficiently and cost effectively when it is designed and utilized to treat a homogeneous waste with
a narrow range of physical/chemical characteristics. If contaminant types and concentrations, waste
viscosity, BTU content, moisture content, acidity, alkalinity, etc., vary widely, control of the system
can be difficult and costly to maintain. Many of these waste
characteristics can be modified and improved with appropriate preprocessing.
In addition, the most effective technology performance is achieved when the soil particle size is small
and the maximum amount of surface area is exposed. This condition facilitates adequate contact
between the contaminant sorption sites and the driving force of the technology (i.e., microorganism,
solvent, warm air, etc.). The key to achieving this contact, and subsequent contaminant destruction,
transfer to another medium, or bonding, is often achieved only through significant mixing, either
before entering or within the treatment unit.
Materials handling, preprocessing, and mixing technologies with potential application to contaminated
soil are currently in use in industries such as construction, agriculture, and mining. All of these
industries routinely handle large quantities of soil or rock. The use of technologies from these
1343
-------
industries should be considered during all soil remediation activities. Materials handling,
preprocessing, and treatment unit mixing techniques should also be incorporated in treatability testing
programs.
The results of such tests will better define the range of waste characteristics which the full-scale
technology will have to treat.
To further investigate this important issue, EPA is performing mixing studies performed on various
uncontaminated soils. The tests are designed to quantify the mixing of soil and test the effects of soil
homogeneity on treatment performance. A selection of soil types, mixing equipment scales, and
moisture contents, representative of different treatment technologies, will be combined to provide
a matrix of samples commonly encountered during treatment. Mixing experiments will be conducted
on three types of uncontaminated soil (clayey, silty, and sandy) at three mixer scales (bench, pilot,
and full) and at three moisture contents (field dry, liquid limit, and plastic limit) to establish trends
in the degree of mixing as a function of soil type, scale, and moisture content, representative of
different treatment technologies. Similarly, treatment and mixing tests will be performed on
contaminated soil at the pilot scale on a select set of samples from this matrix. Data generated from
these tests could be used to establish a correlation between treatment effectiveness and degree of
mixing.
CONCLUSIONS
EPA has launched a comprehensive and aggressive effort to facilitate technology transfer and to
develop LDR regulations based upon best demonstrated available technologies for treating soil and
debris. The technical issues that need to be considered in the development of LDR regulations for
soil and debris have been identified and are being investigated in research programs and by analyses
of existing data.
Timely and complete technology transfer is an important EPA Superfund goal and in addition to
collecting data and developing land disposal restriction regulations for contaminated soil and debris.
Therefore, EPA will continue to seek and evaluate all treatment results, and evaluate the results for
both regulatory development and technology transfer. In this vein, the data and conclusions
presented in this paper represent the most current information available in the Superfund program.
EPA recognizes that with each additional treatment test performed, more valuable information will
be generated regardless of whether the test was successful or not.
It is important that the research, remediation, and vendor experts have an opportunity to participate
in the EPA Superfund technology transfer activities as well as in the development of the land disposal
restriction regulations for contaminated soil and debris. Two options exist for this participation.
First, EPA requests that all available information on the treatment of contaminated soil, sludges, and
debris be forwarded to EPA or to CDM FEDERAL PROGRAMS CORPORATION. Second, public
participation in the regulatory development process through response to upcoming Federal Register
Notices is also encouraged.
The data, experience, and opinions of members of the hazardous waste treatment community, will
be valuable additions to the crucial technology transfer and regulatory development efforts.
Participation in this process is strongly encouraged and will be greatly appreciated. Please send all
1344
-------
available information and any comments or suggestions to EPA OERR or to CDM FEDERAL
PROGRAMS CORPORATION at the following addresses:
Carolyn K. Offutt/Richard Troast
Hazardous Site Control Division (OS-220)
U.S. Environmental Protection Agency
401 M. Street, S.W.
Washington, D.C. 20460
(703)308-8330/308-8323
Patricia Lafornara
Releases Control Branch
U.S. Environmental Protection Agency
2890 Woodbridge Avenue
Edison, NJ 08837-3679
(908) 906-6988
Joan O'Neill Knapp
CDM FEDERAL PROGRAMS CORPORATION
13135 Lee Jackson Memorial Highway
Suite 200
Fairfax, VA 22033
(703)968-0900
REFERENCES
1. U.S. Environmental Protection Agency. Superfund LDR Guide #6A, "Obtaining a Soil and
Debris Treatability Variance for Remedial Actions." OSWER Directive 9347.3-06FS,
Washington, D.C., 1989, Revised 1990.
2. U.S. Environmental Protection Agency. Superfund LDR Guide #6B, "Obtaining a Soil and
Debris Treatability Variance for Removal Actions," OSWER Directive 9347.3-07FS,
Washington, D.C., 1989, Revised 1990.
3. U.S. Environmental Protection Agency. November 30, 1989, Memorandum on "Analysis of
Treatability Data for Soil and Debris: Evaluation of Land Ban Impact on Use of Superfund
Treatment Technologies." OSWER Directive 9380.3-04, in response to Superfund Management
Review: Recommendation 34A, Washington, D.C., 1989.
4. U.S. Environmental Protection Agency. Summary of Treatment Technology Effectiveness for
Contaminated Soil. EPA/540/2-89/053, Washington, D.C., 1990.
5. U.S. Environmental Protection Agency, Office of Solid Waste. Quality Assurance Project Plan
for Characterization Sampling and Treatment Tests for the Contaminated Soil and Debris
(CSD) Program. Washington, D.C., 1990.
1345
-------
Tower Chemical: Remedial Design
For
A Small But Complex NPL Site
Victor H. Owens, P.E.
Remedial Design Manager
Ebasco Services Incorporated
145 Technology Park
Norcross, GA 30092
(404) 662-2316
Natalie A. Ellington
Remedial Project Manager
U.S. EPA, Region IV
345 Courtland Street, N.E.
Atlanta, GA 30365
(404) 347-2643
Beverly Houston
Section Chief
U.S. EPA, Region IV
345 Courtland Street, N.E.
Atlanta, GA 30365
(404) 347-2643
INTRODUCTION
The goal of this paper is to describe one NPL remedial design project and share the circumstances
surrounding a few key issues that arose during the effort. The design project was for the Tower
Chemical site near Clermont, Florida (near Orlando). The design phase was completed in August
1990, under the REM III contract and will be entering a competitive bidding phase for Remedial
Action (RA) in mid-1991, under the ARCS IV contract.
Several unanticipated aspects of the design resulted in a bumpy ride to completion, only a few of
which will be discussed in this paper. Two technical issues and one contracting issue are described
in some detail, and conclusions regarding their resolution are provided. The technical issues were:
1) accommodate a seven-fold increase in the estimate of contaminated soil requiring remediation, and;
2) provide a biddable design document which takes into consideration the limited data used to
determine the quantity of contaminated soil. The contracting issue was one of whether or not to
divide the design into two parts, a water treatment system contract document and a thermal treatment
system contract document.
BACKGROUND
The Tower Chemical Company site is an abandoned pesticide manufacturing facility located near
Clermont, Florida (see Figure 1). From 1957 to 1981, manufacture of pesticides resulted in disposal
of residues that contaminated soil and groundwater with various contaminants including DDT,
dicofol, xylenes, chromium, nickel and lead. Site investigations conducted by the United States
Environmental Protection Agency (EPA) and Florida Department of Environmental Regulation
(FDER) resulted in the site being included in the National Priority List in 1981. In 1983, an
1346
-------
v V 0 L/U S I A
M I N 0 L E
Tower Chemical Site
8 ° 8 16 24
Graphic Scale in Miles
Figure 1
SITE LOCATION MAP
1347
-------
Immediate Removal Measure (IRM) was conducted by the EPA that consisted of contaminated soil
excavation, buried drum removal and pond water treatment. The Remedial Investigation and
Feasibility Study (RI/FS) were completed in 1987 by NUS Corporation, and the Record of Decision
(ROD) was signed in July of that year. The ROD specified excavation and incineration of
contaminated soil, with pump and treatment of contaminated shallow groundwater, in addition to
other activities. Ebasco Services Incorporated (Ebasco) was tasked by the EPA in January 1988, to
prepare a Remedial Design (RD) for contaminated groundwater. In May 1990, the design was
expanded to include all site remediation tasks with some of the site preparation activities being
implemented by the EPA Emergency Response Group, including security fence construction,
contaminated soil excavation and backfilling, and soil testing. In late July 1990, at the request of the
EPA, all site preparation tasks were incorporated into the design scope. Specifications and drawings
were revised to include all soil clean-up activities in conjunction with the implementation of the
groundwater extraction and treatment system. In August 1990, the design was submitted to the EPA,
revised pursuant to review comments, and resubmitted to the EPA.
SITE DESCRIPTION
The main facility consists of a production building, a small utility building, an office, and two
disposal areas: a burn/burial area for solid wastes and a percolation/evaporation pond for acidic
wastewaters. Figure 2 shows the existing site conditions. The site is relatively flat with only about
five feet of relief. Surface water drains into lower areas which eventually drain into an unnamed
stream north of the site. The stream, in turn, flows into the Gourd Neck area of Lake Apopka. The
lake and nearby swamps and wetlands provide an important natural habitat for local wildlife,
including nesting bald eagles.
Groundwater in the vicinity of the Tower site occurs in the unconfined Surficial Aquifer and the
confined Floridan Aquifer. The Surficial Aquifer extends over most of the site and is composed
mainly of quartz sand with varying amounts of clay and silt. Groundwater in the Floridan Aquifer
flows through solution channels and joint systems in the limestone. The Floridan Aquifer is the major
potable drinking water source in central Florida and many local residents have potable water wells
screened in the Floridan. Wells screened in the Surficial Aquifer are not used for domestic water
supplies.
The Surficial Aquifer, in the area of the Tower Chemical Company site, flows generally to the
northeast, towards the unnamed creek. The water table ranges from 0 to 5 feet below the land
surface. Horizontal groundwater velocity is estimated to be less than two feet per year over most of
the site, but localized areas can exhibit a horizontal velocity of 10 feet per year due to steep
groundwater gradients.
The Floridan Aquifer, in the site area, is poorly confined by the overlying Hawthorne Formation
which is laterally discontinuous across the main facility due to the presence of relict sinkholes.
Groundwater in the Floridan Aquifer moves rapidly through solution channels in a northeasterly
direction. The top of the Floridan Aquifer ranges between 54 and 188 feet below the land surface,
with the potentiometric surface between 2 feet above to 10 feet below the land surface.
PREVIOUS SITE RESPONSE ACTIONS
Three Immediate Removal Measures (IRM's) were conducted at the site, following the closure of the
Tower Chemical Company. The first IRM was conducted in 1981, at a nearby spray irrigation field,
under the lead of FDER. The second and third IRMs were conducted in 1983 and 1988, by the EPA,
1343
-------
APPRQX PROPERTY L
9 MONITOR WELL
^^ BUILDING. STRUCTURE.
I I OR CONCRETE PAD
a BUILDING OCCUPIED
BY LOCAL RESIDENT
^, CONTAMINATION PLUME
, ' " ~ OUTLINE Of RELICT SINKHOLE
Figure 2
EXISTING SITE CONDITIONS
1349
-------
at the main facility site. Since it was determined from the results of the RI that the media present
in the spray irrigationfields would not require further remediation, the most significant IRMs
impacting the RD were those conducted at the main facility.
In 1985, the Centers for Disease Control, Agency for Toxic Substance and Disease Registry
(CDC/ATSDR) determined that a potential threat to public health existed at the site due to the
potential for exposure to wastes at the main facility. Field studies identified a 2,275 square foot area
that comprised what is now referred to as the burn/burial area. This area was excavated to an average
depth of eight feet below the surface at which point pesticide concentrations significantly decreased.
At a depth of five feet, approximately 70 empty drums and two partially filled drums were unearthed.
All of these excavated materials were shipped to the Chemical Waste Management facility in Emelle,
Alabama for disposal.
Simultaneous with the excavation activities, water was pumped from the percolation/ evaporation
pond just west of the burn/burial area. This water was treated onsite for DDT and dicofol using
activated carbon absorption and pH adjustment, to levels which complied with existing laws. Once
the water level in the percolation/evaporation pond was lowered sufficiently, excavation of the
contaminated sediments began. The sediments were dewatered and bulked with the excavated soil
from the burn/burial area before being shipped offsite.
Also affecting the design approach were site activities occurring during the design that increased
contaminated soil quantities and changed site conditions. In 1988, the EPA demolished two storage
tanks near the main facility containing hazardous wastes. Approximately 500 cubic yards of
contaminated soil were excavated from beneath the tanks and moved within the fenced area of the
site, along with the rubble from the tank foundation demolition.
DESIGN BACKGROUND
In January 1988, Ebasco was tasked by Region IV EPA, under the REM III Contract, to design a
groundwater extraction and treatment system for the Tower Chemical site. This design was to be
based on existing data contained in the RI Report. After evaluation of data suitability, the EPA
halted the design effort in August 1988, to install additional wells and conduct pump tests. The
design effort was restarted in January 1989, incorporating the additional data. At that time, the EPA
also increased the design work scope to include design specifications and drawings for incineration
of contaminated soil. Excavation, testing, backfilling and other miscellaneous site work were being
designed and provided by another EPA Contractor and were not part of the Ebasco design. In
August 1989, the design was expanded to include a confirming field sampling program to assess the
leachability of pesticides and incinerability data. These data would be used to refine soil thermal
treatment. In January 1990, a 60% RD was submitted by Ebasco to the EPA for review.
In April 1990, the Ebasco design scope was increased to include that site work proposed for another
EPA Contractor. In May 1990, Ebasco was tasked to prepare a design that included all phases of
remediation activity onsite, including contaminated soil excavation and backfilling.
In support of the RD, the EPA conducted groundwater pumping tests in late 1988, in the Surficial
Aquifer, to determine the hydrogeologic properties of the site. One pump test was conducted within
the backfilled waste pond, and one pump test was conducted within the burn/burial area. It was
determined that the relict sinkhole of unknown dimensions beneath the waste pond discovered during
the RI required further definition before an adequate groundwater recovery system could be designed.
In late 1988 and early 1989, the EPA collected soil samples and groundwater samples to determine the
extent and levels of contamination in both media to help define critical parameters for the remedial
1350
-------
design. Additional wells were installed to determine the edge of the groundwater plume, and
additional surveys were conducted to delineate the extent of the relict sinkhole beneath the waste
pond.
In January 1990, Ebasco collected groundwater elevation data and performed slug tests on the new
monitor wells installed by the EPA to support groundwater remediation design. Soil samples were
collected and analyzed for properties useful in preparing bids for thermal treatment. A leaching study
was completed that simulated the flushing of contaminants from sinkhole sediments.
DESIGN DESCRIPTION
The completed design and contract package consists of performance specifications, detailed
specifications, site data, drawings and schedule requirements to obtain and conduct RA services at
the Tower Chemical site. A subcontract for excavation and thermal treatment of soil, and a separate
subcontract for site work with groundwater extraction and treatment was prepared. These subcon-
tracts are to be awarded and managed by a construction manager, who is under direction of the EPA
Contracting Officer. A general overview of the two subcontract documents resulting from the RD
is provided in the following:
THERMAL TREATMENT SYSTEM (TTS) SUBCONTRACT
The TTS Subcontract includes incinerator setup, trial burn, soil incineration, treated soil verification
testing, maintenance of soil stockpile, contaminated soil excavation, treated soil backfill and TTS site
preparation including construction and operation of a retention pond, and all necessary provisions in
support thereof. Approximately 9000 cubic yards of contaminated soil in six different areas of the
site require excavation, incineration, and disposal. An area of the site has been designated as the TTS
work area and is to be used according to the needs of the Subcontractor (see Figure 3). The TTS
Subcontractor will be responsible for all thermal treatment of waste, maintaining and minimizing the
contaminated work area, providing security for the immediate TTS area, providing power and utilities
as needed, pretreating process, excavation or decontamination water for on-site treatment by others,
and setting up and maintaining decontamination facilities for TTS operations, equipment and
personnel. As part of site operations, the TTS Subcontractor will manage water disposal in the
retention pond. Water from excavations, decontamination and processing of soil may be directed to
the pond provided pretreatment requirements are met and pond capacity/water treatment capacity
are not exceeded. The TTS Subcontractor will provide all hardware and controls necessary to convey
the water from the retention pond to the WTS .
It is expected to take 21 months to prepare for the trial burn plan, obtain EPA approval of the plan,
mobilize, set up, shake down and conduct the trial burn prior to starting full production burning.
Thermal treatment is expected to take approximately six additional months at 4.5 tons per hour and
25% down-time. It is possible that TTS operations will be completed early if greater incinerator
capacity or less down-time is achieved.
The TTS Subcontractor will be required to collect and analyze soil samples to verify contaminated soil
excavation completion. The construction manager will collect intermittent companion samples for
verification analyses through the EPA-Contract Laboratory Program (CLP). Operations that may
produce contaminated wastewater, such as excavation or sampling, will not commence until the water
treatment system is functional and can accept the water. Work covered by the TTS Subcontract will
be conducted in two phases. Phase One includes mobilization; excavation and treatment of the soil
from excavations near the building; and treatment of the soil excavated during preparation of the TTS
area (approximately 1,000 yd3 of soil). Phase Two includes excavation and treatment of all remaining
soil and demobilization.
1351
-------
Figure 3
ITS SITE CONFIGURATION
1352
-------
WATER TREATMENT SYSTEM (WTS) SUBCONTRACT
A plume of contaminated groundwater extending across the site and covering approximately 10 acres
will be extracted using 22 wells and treated to meet EPA-approved discharge criteria (see Table 1).
A portion of the treated water will be discharged to a nearby stream while some of the treated water
will be reinjected (see Figure 4). The WTS Subcontract includes construction of roads, and
decontamination facilities for WTS equipment and personnel, grading of the site to promote proper
drainage; installation of wells, piping hardware and controls for groundwater extraction and injection;
construction of a building to house the system; and management and operation of a water treatment
system. The WTS Subcontract includes responsibility for procuring and managing site perimeter
security, as well as arranging for the installation of any utilities, offices, or other support required
by the WTS Subcontractor to operate the system. The WTS Subcontractor will commence with the
installation of the water treatment unit and building prior to the mobilization of the TTS Subcontrac-
tor. Upon completion of the WTS construction and shakedown of the system, the TTS Subcontractor
can begin excavation of the contaminated soil.
The WTS Subcontractor will work concurrent with the TTS Subcontractor once the WTS is operational.
The groundwater extraction system will be installed in two phases. Phase One consists of installation
of roads, wells, piping, controls and other hardware outside of the contaminated soil excavation area.
Upon completion of soil treatment and backfilling, Phase Two of groundwater extraction system
installation will be completed followed by one year operation by the WTS Subcontractor.
DISCUSSION
IN-SITU SOIL FLUSHING
A sampling program conducted by the EPA in November 1988 revealed pesticide-contaminated soil
quantities up to seven times the quantities previously estimated. Instead of 5000 cubic yards of
contaminated soil, the quantity was now approximately 34,000 cubic yards. The new soil data roughly
defined the extent of soil contamination as shown in Figure 5. It is worth noting that the sampling
points used to revise the estimated value were not surveyed nor located using scaled site maps or
drawings. Although the majority of soil requiring remediation was based on some subjective
estimates, it was clearly within the confines of the backfilled relict sinkhole. Depths of soil
contamination appeared to be at least 18 feet below surface, and possibly deeper. In order to excavate
these contaminated soil, it was expected that dewatering rates of several hundred to over a thousand
gallons per minute would be needed. Treated water discharge criteria were required to meet Florida
Class III surface water contaminant levels or Maximum Concentration Levels (MCLs), whichever were
less. However, transporting a water treatment system to the site capable of meeting discharge
requirements and handling large flow volumes was not desirable. Long term groundwater treatment
capacity was not anticipated to exceed 125 gpm and the cost of incineration for the unexpected soil
volume combined with rather large WTS requirements for dewatering effluent would increase initial
remediation cost estimates by a factor of nearly 8.
With the Agency's concurrence, Ebasco decided to explore alternatives to complete excavation of the
contaminated soil that would still achieve all clean-up goals and adhere to the intent of the ROD. The
ideal alternative needed to meet three criteria: 1) avoid significant cost associated with major
dewatering of the sinkhole; 2) utilize only the WTS capacity proposed for on-site groundwater
remediation, and; 3) achieve cleanup of the soil within a reasonable period of time.
After observing that the key soil contaminant, dicofol, was also present in the groundwater plume,
Ebasco proposed in-situ extraction or "flushing" the dicofol. A conceptual diagram of the in-situ
1353
-------
Table 1
Tower Chemical Site Clean-up Criteria
TREATED WATER DISCHARGE CRITERIA
Maximum Surface
Observed Discharge
Parameter Concentration (ug/L) Criteria (iig/L)
Arsenic 10 50
Barium 190 1000
Cadmium 5 0.7
Chromium 710 11
Copper 170 6.5
Iron 9300 300
Lead 51 1.3
Manganese 750
Nickel 420 88
Sodium 270.000 160,000
Zinc 63,000 30
Cyanide 0.02 5
Benzene 8 1
Chlorobenzilate 9 100
Ethylbenzene 420 453
Toluene 14 175
Trichloroethene 6 5
Xylene 1,700 400
Phenol 37 256
Dicofol 1,400 0.08
DDT BDL 0.1
DDE BDL 0.1
DDD BDL 0.1
TARGET GROUNDUATER CLEANUP LEVELS
Indicator Target Groundwater Cleanup
Contaminant Level (ug/L)
Arsenic 50
Nickel 350
Chromium 50
Alpha-BHC 0.05
Chloroform 5
DDT 0.10
Chlorobenzilate 10.0
Dicofol 0.08
Xylene 400
TARGET SOIL CLEANUP LEVELS
Indicator Target Soil Cleanup
Contaminant Levet (mg/kg)
Copper 7,500
Lead 100
Arsenic 5
Dicofol 5
Chlorobenzilate 24
DDT 35
Xylene 50
1354
-------
CRUSHED STONE ROAD
OUTLINE OF RELICT SINKHOLE
CONTAMINATION PLUME
Figure 4
WTS SITE CONFIGURATION
1355
-------
Figure 5
SOIL EXCAVATION
1356
-------
flushing approach is shown in Figure 6. This approach utilized the apparent hydraulic connection
to the Floridan Aquifer caused by the relict sinkhole, and could be adjusted so as to not overwhelm
the original design WTS capacity. Total excavation quantity would be reduced from 34,000 cubic
yards to 9,000 cubic yards and the costs maintained at a level similar to the initial remediation cost
estimates. An initial calculation of the required flushing time was performed based on measured soil
dicofol concentrations, dicofol solubility in water and a simplified trans-port model. The calculations
and concept were formalized and presented to the EPA as a viable alternative to excavation and
thermal treatment.
Although the technical reviews of the in-situ flushing approach determined that this alternative was
feasible, there was still the unknown variable regarding the actual extraction rate of contaminants
from the soil. To resolve this issue, a bench-scale desorption rate study designed to measure dicofol
leaching rates was subcontracted by Ebasco. The purpose of the study was to obtain "quick and dirty"
data to eliminate some of the uncertainties related to the rate of leachability of dicofol. The leaching
study focused on measuring the difference in dicofol concentrations at inlet and outlet of soil columns
and at different flow rates expected both near and at the projected extent of the extraction well cone
of influence. Samples of soil below the water table in the relict sinkhole where high dicofol concen-
trations were expected were collected and sent to the laboratory responsible for the leaching study.
Three bulk samples were collected from three different locations in the contaminated area, but initial
characterizations by the laboratory indicated that none of the samples contained dicofol concentrations
that exceeded the clean-up criteria. Nevertheless, the study was conducted by spiking the soil
samples with dicofol and measuring the rate at which that dicofol was removed from the soil. Study
results indicated that original assumptions used during calculations were slightly optimistic, but the
soil flushing would achieve the required clean-up levels within ten years and at a fraction of the cost
necessary for excavation and incineration.
However, interpretation of results from the study assumed that the spiked medium would desorb
dicofol at the same rate as the naturally acclimated soil onsite. Schedule and budget allocated for
completion of this RD did not allow for an additional field sampling effort or subsequent leaching
studies. The RD was completed with the qualification that the leachate calculations were based on
an artificially contaminated medium. Ebasco's evaluation of the study and the results acknowledged
the potentially non-representative nature of that test, but, if further studies were required by the
EPA, they would have to be obtained during the planning phase of the RA.
CONTAMINATED SOIL LOCATION
It is likely that the unresolved questions about the dicofol leachability that remained at the conclusion
of the design could have been avoided if contaminated soil samples had been found at the site.
However, it was concluded that the actual location of the contaminated soil was probably not as
depicted in Figure 5. The basis for this conclusion was that three random soil samples collected
within the relict sinkhole area, all from different locations but within the prescribed contaminated
area, all showed contaminant levels below the clean-up criteria. In addition to providing an
inconclusive evaluation of the leaching study, the assessment of the soil characterization provided an
uncomfortable level of confidence in the estimated volume of soil requiring thermal treatment and
the most effective configuration for the in-situ soil flushing wells.
Uncertainty in the actual contaminated soil quantity propagated to other aspects of the design.
Therefore, the EPA concluded with Ebasco that confirmational soil contamination data would be
useful, but would be obtained during the initial planning of the RA under the ARCS IV Program.
To avoid the necessity of changing specifications and drawings for the WTS to accommodate any
changes caused by a changed contaminated soil quantity, the WTS was designed to be modular. If the
1357
-------
CONTAMINATED SOIL
TO BE EXCAVATED
CONTAMINATED SOIL REMEDIATED
BY GROUNDWATER FLUSHING
SUFJFICJAL SAND/SiLTY
SAND AQUIFER
K = 40 ft/do/
xxxxxxxxx
HAWTHORNE CLAY
XXXXXXXX X
FLORIDAN AQUIFER LIMESTONE
CLEAN SAND IN
RELICT SINKHOLE
K = 1500 ft/day (ASSUMED)
LEGEND
H WASTE BURIED
JJ BENEATH GROUNDWATER
J FLUSHING WELL
GROUNDWATER
- FLOW DIRECTION
Figure 6
IN-SITU SOIL FLUSHING
1358
-------
contaminated soil quantity decreased dramatically, an additional optional WTS train would not be
requested. On the other hand, additional capacity would be available for installation if the
construction manager determined that the WTS capacity was likely to be exceeded during any phase
of construction.
Ebasco was generally tasked under the ARCS IV Program to support the EPA's remedial construction
manager. By collecting additional soil data and amending the design under the ARCS IV Program,
the problems associated with the uncertain quantity of soil for thermal treatment were addressed.
However, coordinating the work of two separate contractors with interdependent schedules was still
anticipated to cause some difficulty for the construction manager.
TWO-CONTRACT APPROACH
The use of two subcontracts to perform the site remediation evolved initially because of early RD
scope of work requirements. At EPA direction, specifications for obtaining soil thermal treatment
services were prepared as a separate item with other (non-Ebasco) EPA contractors preparing the
remaining design documents necessary for all soil remediation. As the RD proceeded, the EPA
increased the scope of work under the existing Ebasco RD assignment to include all aspects of site
remediation. At that time a decision was made jointly with Ebasco and the EPA to maintain the
design as two separate contract documents. Unwanted side effects expected from a two-contract
approach were generally related to the difficulties of coordinating two contractors with inter-related
schedules. As shown in Figures 3 and 4, the remedial construction was divided into two phases for
each subcontract, or a total of four phases. The WTS Subcontractor will be the first Subcontractor
onsite and the last to demobilize. With careful management of the RA, however, it was expected that
the potential benefits will outweigh any additional problems.
With two separate contract documents, there was expected to be substantial cost savings. For example,
with one contract document and award including both TTS and WTS, it was considered likely that a
TTS Subcontractor would have to subcontract the WTS (or vice versa) and there would be a fee on
fee charge. The fee on fee for either the TTS or WTS was estimated to be far greater than any
additional construction management costs associated with handling two subcontracts. Additionally,
by reducing the contract value using a two-contract approach, it was anticipated that bonding and
insurance would be easier to obtain by bidders. Also, there are some benefits expected during the
procurement process. Although both contracts were scheduled to be awarded simultaneously, having
two smaller, more manageable pieces to negotiate was considered an advantage. Work phasing did
not require simultaneous contract award and therefore a delay in the TTS procurement would not
necessarily delay the overall project completion.
CONCLUSIONS
During the development of the Tower Chemical RD, various issues arose that presented difficulties.
Resolving these issues resulted in a RD substantially different than the one originally planned at the
start of the design effort. From Ebasco's perspective, the design scope started as a groundwater pump
and treat, progressed to include partial soil RD, and finally encompassed the entire site, including soil
excavation and site development. When soil remediation was added and the ensuing dewatering
requirements became essentially infeasible, it was necessary for the EPA to find satisfactory site
remediation using a slightly more innovative approach. In-situ soil flushing provided the means for
remediating the bulk of the soil while maintaining control of RA costs. The two-contract approach
allowed the flexibility of staggered contract awards and avoided duplication of costs. Finally,
ambiguous definition of soil contamination will be refined at the start of RA activities and will
require minimal amendments to the design documents.
DISCLAIMER
The opinions and views expressed in this paper are those of the authors and do not necessarily
represent the opinions or views of the United States Environmental Protection Agency. Any questions
or comments regarding the content of this paper should be addressed to the authors.
1359
-------
The Importance of Test Fills for the
Construction of HTW caps and Liners
David P. Ray, P.E.
U.S. Army Corps of Engineers
Omaha District
215 N. 17th Street
ATTN: CEMRO-ED-GA
Omaha, NE 68102-4978
(402) 221-4493
1. Introduction. Construction problems related to site and
material unknowns at hazardous waste sites, can cause serious
schedule delays and often result in expensive project modifi-
cations. One way to better define various project unknowns
is to specify and construct a test fill. A test fill con-
sists of the construction of a structure which simulates a
full-scale cap or liner system, including all associated com-
ponents, using the materials, equipment, and processes which
are specified for the project site. This paper will describe
the usefulness of test fill construction as well as present
the difference in rationale used to generate effective test
fill specifications used to define design goals.
2. Background. The concept of using a test fill to verify
the adequacy of project materials and placement methods was
made popular during the era of large earthen embankment dams
and extensive levee systems construction. The Corps of Engi-
neers implemented test fill construction as a design tool to
verify project specifications before beginning construction
of a full-scale project.
The construction of cap and liner systems for landfills
and other HTW-related sites require the placement of large
amounts of compacted soil fill, geosynthetics, and topsoil.
material, most often times at sites regarded less than desir-
able for construction. In order to verify design assump-
tions, determine adequacy of construction materials and
placement procedures, and to better define various
site-specific unknowns, it is imperative that a test fill be
constructed and evaluated before full-scale construction be-
gins. The test fill serves to reduce the potential for
costly delays due to construction problems and helps assure
that an adequate cap/liner system will be built.
3. Cap/Liner System Test Fill Design. The designer must
first identify the goals to be achieved by constructing the
proposed test fill. The primary goal is to verify the over-
all constructability of the cap/liner system, that is, can
the specified materials be placed according to the project:
specifications with the proposed construction equipment. The
other goal is to insure that the final cap/liner system will
function as designed. The most important function of the cap
system is to retard moisture migration into underlying waste
layers. The most important function of the bottom liner sys-
tem is to prevent leachate from migrating into the ground wa-
ter. The test fill program is used to verify if the
specified compacted permeability of the low permeability
13bO
-------
clay component of a cap/liner system can be achieved con-
sistently when placed according to project specifications.
In order to insure that the design goals of the test
fill are achieved, the designer must clearly specify proper
QA/QC procedures for the test fill. The importance of an ef-
fective QA/QC plan for the test fill cannot be understated.
A list of guidelines of some of the variables which should be
monitored or controlled during the test fill condition are:
(1) Full characterization of all materials from borrow
areas proposed for use in both the test fill and the large
scale project. In-situ moistures, Atterberg limits, and
moisture, density relationships should be established as ap-
propriate for each material type to be placed.
(2) All soil and/or additives placed in the test fill
should be uniformly distributed to maintain homogenity of ma-
terial for each lift placed. No large diameter (greater than
2-inch diameter) rocks, rubbish, debris, or organic material
should be used.
(3) Specified water contents should be maintained dur-
ing placement and the same moisture conditioning methods
should be used for the full-scale project. It is preferable
to maintain moisture contents above optimum value.
(4) All placement, moisture conditioning, and compac-
tion equipment used on the test fill should be as specified
for the full-scale project. (See Table 1 for typical.equip-
ment applicability for each phase of construction.)
(5) The maximum specified clod size of material should
be maintained and the effectiveness of the construction
equipment to achieve this should be verified.
(6) The maximum loose lift thickness of material placed
should be as specified as well as for the compacted layer
thickness.
(7) Compaction and placement equipment traffic patterns
should be as specified or otherwise monitored. The number of
equipment passes used to compact each layer should be
documented.
(8) The effectiveness of compaction equipment in re-
1361
-------
striated areas should be verified and control measures
taken to insure similar areas in the full-scale project will
be properly compacted.
(9) The test fill should be constructed and maintained
so as to reduce chances of either saturation of subgrade soil
during rainfall events or desication due to drying of
subgrade soil.
(10) Special precautions should be taken in order to
insure that side slopes, layer penetrations (for soil test-
ing) , and damaged soil layers are sufficiently compacted and
sealed.
(11) The specifications should include moisture and
density test frequencies to verify uniform compaction effort
is being achieved.
(12) The test fill should be constructed to the steep-
est slope anticipated for the full-scale project.
(13) As a minimum, the test fill for a cap/liner
project should be constructed to the typical dimensions, shown
in Figure l and the cross section shown in Figure 2. ^ '
(14) A test fill for a cap/liner project should be con-
structed to facilitate field permeability testing. When rei-
quired, the construction of an under drainage system should
be as shown in Figure 2.
(15) Laboratory testing should include permeability
testing of low permeability clay layers.
Another key operation to consider during construction of
the test fill is the placement of the geosynthetic materials.
The test fill offers a great opportunity for the construction
crew to develop the site-specific expertise in placement and
seaming of material in the field with close QA/QC inspection
within the project specification guidelines. This
small-scale operation will serve to familiarize all parties
on the construction/oversight team of what will be expected
of them and how well placement methods will perform. A full
suite of QA/QC testing should be performed according to
project specifications on each layer of the geosynthetic ma-
terial as it is placed. An effort should be made after test
fill construction to verify the geosynthetic material sur-
vivability. In order to function as designed, the material
1362
-------
must "survive or remain undamaged" through all phases of con-
struction. It must be determined if any materials were dam-
aged during construction activities, particularly during the
placement of large amounts of fill material over the
geosynthetic materials.
After completion of the cap/liner test fill, the
specifications should include tests which will demonstrate
that the compacted permeabilities of the clay layer is within
the limits specified. A series of permeability tests run on
undisturbed samples from the compacted clay layer can be used
to verify the uniformity of the in-situ material and indicate
how well the clay layers will perform. However, research has
indicated that laboratory permeabilities may vary from the
actual field permeabilities by as much as an order of mag-
nitude.
Undisturbed samples of the compacted clay layer compo-
nent can be inspected in order to determine how well the lift
layers bonded. Lift layer bond has been determined to be a
key factor for construction of effective low permeability
soil layers.
The best method of verifying in-situ permeability of the
cap/liner system test fill is to pond water over the surface
and collect seepage with an underdrain system and supplement
this information with data from surface infiltration ' ' .
Other popular options for determining in-situ permeability
are the use of the single and double, infiltrometers * ',
sealed double-ring infiltrometers ^ ', and the borehole
method developed by Boutwell and other methods specified in
Daniel's paper, Earthen Liners for Land Disposal Facilities
listed in the reference summary.
An innovative verification test procedure currently used
involves the use of a simulated rainfall event. This test
involves setting up a system which can deliver a measured
amount of water at a set flow rate to the surface of the test
fill. During the design rainfall event, the designer should
measure any slippage of geosynthetic layers built on critical
side slopes by monitoring exposed portions of geosynthetic
layers aligned with control markers and paint lines. l '
Another aspect of this test is to monitor the discharge pipe
of the underdrain layer in order to determine the effective-
ness of drainage collection layers, as well as determine
in-situ permeability after water has ponded on the liner sur-
face.
1363
-------
Once it has been established that the in-situ permeabil-
ity achieved in the test fill is satisfactory, then a set of
index properties can be established for use on the full-scale
cap/liner system. The index properties are defined as the
factors which can be measured in the field by direct testing
within the QA/QC program to verify full-scale material place-
ment. The EPA.recommends measuring the following properties
as a minimum: ^ '
(1) Hydraulic conductivity (undisturbed samples);
(2) In-place density and soil moisture content;
(3) Maximum clod size;
(4) Particle grain size distribution;
(5) Atterberg limits.
Other factors also include:
(1) Field moisture content during and after field placement;
(2) Loose lift and compacted layer thickness;
(3) Number of passes of specified construction equipment.
4. Waste Pile Test Fills. The use of soil and sludge
solidification/stabilization techniques are becoming increas-
ingly important in order to comply with federal, state, and
local regulations on placement of contaminated soils in waste
piles and landfills. The biggest challenge facing the de-
signer of a large solidification/stabilization project is in
determining an effective and economical waste soil mix design
which will result in a material which complies with the
various placement regulations. An important variation of the
cap/liner test fill concept is to provide design information
for placement of solidified/stabilized material within a
waste pile. Although this is a variation of a pilot scale
study, the designer can gain invaluable information concern--
ing effective mix designs, placement methods and liner mate--
rial survivability during test fill placement. Besides com-
plying with the regulations for placing material in a waste
pile such as free liquid content from the Paint Filter Test
(EPA 9095), leachability, minimum soil strength, etc., the
solidified/stabilized material should exhibit a compacted
1364
-------
soil strength great enough to insure final waste pile stabil-
ity, equipment trafficability, control settlement of the
placed material (and thus the final cap over the waste pile),
and minimize or control leachate production due to pile
overburden stresses.
To guarantee that the waste materials in the final waste
pile are stabilized within placement regulations, exhibit
needed design properties, and are placed so as not to damage
any portion of the liner, a set of index properties can be
derived from a test fill to determine a method specification
for the final fill placement. The test fill consists first
of the proper construction of the waste pile liner at the
proposed site. The test fill should be located on the por-
tion of the completed waste pile liner in which the stabi-
lized waste can be placed to the dimensions outlined in Fig-
ure 1 and where waste material can be placed on at least one
berm side slope, if applicable and practical. In order to
insure the integrity of the underlying liner material, it is
recommended that the initial lift of stabilized/solidified
waste material be on the order of from 1.5 to 3 feet thick,
with subsequent varying loose lift thicknesses. Once the
initial lift is established and compacted to a point which
will allow equipment trafficability, then the controlled fill
procedure can begin.
The objectives of the test fill are to establish the in-
dex properties which can be used to develop a method specifi-
cation for monitoring full-scale placement in the waste pile.
The overall objectives of the test fill are as follows:
(1) Observe and evaluate trafficability and
constructability of the waste materials;
(2) Obtain settlement and consolidation data to
evaluate long-term stability;
(3) Determine in-place density and unconfined compres-
sion strength data to evaluate slope stability;
(4) Observe liner material during and after test fill
construction to verify survivability of liner materials;
(5) Determine compliance of placed material with
various placement regulations.
During construction of the test fill, the following pa-
1385
-------
rameters and operations should be measured and recorded in
the field in order to help determine index properties that
will be most critical for waste pile construction:
(1) Descriptions of material types and/or mix designs
used during placement;
(2) Moisture content, compacted density, unconfined
compressive strength, and other strength data relating to
construction trafficability, such as cone index testing;
(3) Material placement, traffic patterns, and grading
and spreading patterns;
(4) Lift thicknesses;
(5) Number of passes of compaction equipment;
Upon completion of the test fill the results of the
various field tests should be analyzed in order to identify
the critical index properties which should be monitored dur-
ing full-scale construction. These properties will insure
the stabilized waste is placed so as to achieve the overall
full-scale waste pile construction objectives.
Typical plots comparing field test results are shown in
Figures 3 through 7, with summary data in Tables 2 and 3.
These plots were generated from the waste pile test fill for
the Basin "F" Interim Action project at Rocky Mountain Arse-
nal and are typical of test results from a test fill. After
analysis of that test fill data, it was recommended that Ba-
sin "F" contaminated sludge be mixed with on-site soil at a
one-to-one mix ratio and that each lift be compacted with
four passes of the compactor. The method specification for
waste material placement included the following index proper-
ties: 1) All solidified material was to pass the Paint Fil-
ter Test; 2) The minimum unconfined compressive strength of
the compacted material was eight psi and the maximum
long-term strength was 25 psi; 3) The minimum cone index
trafficability value was 150 psi; 4) The minimum percent of
compaction was 80% of the standard proctor maximum. The test
fill verified the importance of the index properties in ful-
filling the objectives of the test fill which were:
(1) Reduce to a minimum the amount of leachate produc-
tion from the waste material placed by placing all material
within Paint Filter Test requirements;
1386
-------
(2) Insure slope stability for the finished waste pile
and cover system by placing solidified waste at an unconfined
strength of at least 8 psi;
(3) Not to solidify the waste material so as to be able
to remove it for later soil incineration;
(4) To solidify waste material to insure equipment
trafficability for constructability within the waste pile by
specifying a cone index of at least 150 psi;
(5) Compact waste material in order to reduce cap
settlement so that the final cap slopes would be at least 3%
by specifying at least 80 percent compaction requirements.
(6) The stabilized material was to be easily
excavatable after completion of the placement of the so-
lidified material.
After completion of this test pile, the solidified mate-
rial was removed and the liner material was inspected. No
appreciable damage was done to the geosynthetic layers.
5. Cap System Test Fills for Difficult Sites. Since uncon-
trolled landfill sites were usually located in the least de-
sirable locations, it follows that most sites offer unique
construction problems for cap construction. Site-specific
test fills can be used to determine overall constructability
of caps over difficult sites. Constructability problems in-
clude:
(1) Placement of soil layers and geosynthetics over
steeply sloping sites;
(2) Subgrade stabilization of soft subsoils to fa-
cilitate fill placement;
(3) Compaction over unconsolidated landfill materials
which result in large differential settlements;
(4) Construction at sites at which there is limited
space for staging material and equipment;
(5) Placement of fill in marshy areas with high ground
water levels.
1367
-------
(6) Placement of geosynthetics over landfills which
generate large amounts of gas.
Test fills offer the designer insight on how best to
solve those constructability problems rather than trying to
solve them during actual site construction when delays mean
slipped schedules and expensive contract modifications. The
small-scale test fill offers the designer an opportunity to
try new materials and methodologies in order to solve some of
these constructability problems.
6. Test Fill Specifications. It is recommended that a test
fill be constructed for all large scale cap/liner construc-
tion projects. In order to write an effective test fill
specification, the designer must have a clear idea of the ob-
jectives to be accomplished during and after construction of
the test fill. The designer should include the following in
the test fill specification:
(1) The size and location of the test fill including
the thickness and material type of each layer;
(2) The slope of the layers as well as compaction arid
density requirements;
(3) Horizontal and vertical survey requirements;
(4) Clearing and grubbing of site subgrade;
(5) Regrading and clay cap/liner soil requirements to
establish placement requirements such as compaction equip-
ment, moisture and density requirements, maximum clod size
and suitable fill material types for each layer;
(6) Placement requirements for each layer of
geosynthetic material;
(7) Establishment of a vegetative layer for long term
test fills;
(8) Development of a system to determine layer slippage
of geosynthetic materials during material placement on
critical slopes;
(9) Post construction testing such as infiltrometer or
other in-situ permeability testing;
(10) Special post construction testing; i.e., settle-
1368
-------
ment of test fill;
(11) Removal of test fill layers to verify lift thick-
nesses, bond between layers and condition of the underlying
geosynthetic layer.
The most important requirement of all to include in the
test fill specification is the QA/QC program. The designer
must specify construction quality control testing including
test frequencies, methods and pass/fail criteria. Third
party QA/QC is as important during the test fill as during
the full-scale cap/liner construction.
7. Conclusion. The design and construction of a large-scale
cap/liner system for hazardous waste site closure/remediation
projects can be a very complicated task. One way to reduce
and eliminate project problems due to site and construction
material unknowns is to specify and construct a test fill.
The objectives of a test fill must be clearly understood
so that a viable test fill method specification will result.
The goal is to establish a set of index properties which can
be used in the full-scale project which will result in a
functional cap/liner system. Test fills can be used to de-
fine placement parameters for waste pile construction and to
address constructability problems at problem sites. The
single most important factor for test fill construction is to
have an effective, well-organized QA/QC program. Along with
this, is the need to record, the test fill placement specifi-
cations, site-specific test results, and performance records
on a data base system so that engineers faced with designing
future cap/liner systems can consult them.
1369
-------
References
1. Construction Quality Assurance for Hazardous Waste Land
Disposal Facilities, EPA 530-SW-86-031, Oct 86, pgs 21-25
2. Covers for Uncontrolled Haz Waste Sites, EPA 540/1-85
/002, Sep 85, pgs 5-7
3. Earthen Liners for Land Disposal Facilities, David
Daniel, Proceedings, Geotechnical Practice for Waste Dis-
posal. ASCE, Ann Arbor, Michigan, pg 8
4. Geosynthetic Landfill Cap: Laboratory and Field Tests,
Design and Construction, Giroud and Swan
5. Lining of Waste Containment and Other Impoundment Fa-
cilities by Matrecon Inc., EPA 600/2-88/052, Sep 88, pg 8-8
For Further References:
1. Landfill Liners and Covers: Properties and Application
to Army Landfills (Final rept), Shafter, R.; Renta-Babb, A.;
Smith, E. ; Bandy, J. , Construction Engineering Research Lab,.
(Army), Champaign, IL
2. Test Fill for a Double Liner System, Forslund, B. L. ;
Smith, L. J. Shekter; Young, M.A. Corporate Source: Neyer,,
Tiseo & Hindo Ltd., Farmington Hills, MI, USA Conference
Title: Geotechnical Practice for Waste Disposal, 1987
3. In-Depth Look at Landfill Covers, Hatheway, Allen W.;
McAneny, Colin C. Corporate Source: Univ. of Missouri --
Rolla, MO Source: Waste Age v 18 n 8 Aug 1987 lOp between p
135 and 156
4. Design of Final Covers for Landfills, Dezfulian, Houshang
Corporate Source: Woodward-Clyde Consultants, Santa Ana, CA
Conference Title: Environmental Engineering, Proceedings of
the 1986 Specialty Conference
5. Design and Construction of Effective Soil Liners,
Anderson, David C. (K. W. Brown & Assoc, TX) and Anderson,
Myron C. (Univ. of Texas), EPA/et al Hazardous Wastes and
Hazardous Materials, 5th Natl Conf, Las Vegas, Apr 10-21, 88
1370
-------
P 202(3)
6. Technical Guidance Document: Construction Quality Assur-
ance for Hazardous Waste Land Disposal Facilities, EPA Report
530-SW-86-031, Oct 86 (100)
7. Construction Quality Assurance for Hazardous Waste Land
Disposal Facilities, EPA Report 530-SW-85-021, Oct 85 (112)
8. Field Measurement of Landfill Clay Liner Permeability,
Edwards, R.; Yacko, D.G.; Bell, J.M. (ed.), Peoria Disposal
Co., Peoria, IL 61615
9. Field Investigation of Clay Liner Performance, Miller,
C.J., Wayne State Univ., Civ. Eng., Detroit, MI 48202
1371
-------
M-l
CO
-------
t>0
cfl
c
•H
CO
§
C
o
•H
•U
CO
U
•l-l
cr
"O
V
Q.
O.
•r-l
o
-------
>-
C£
<
ID
CO
UJ
O
O
eg
LU £
CO t
00
Lu <
O 2
UJ
4t
03
COg
X
LU
Q
CM -T
- CO
- (O
O
UJ
a
6
UJ
a
CO
V)
a.
o
U)
UJ
CO
CO
a:
ui
m
3,
H
CO
tJ
a;
O
o
N-
O
a
(O
_r
o
s
o
o
I
O
o
o
(ISd) X3QN! 3NOO
1374
-------
00
I
00
LU o.
00 -
00 S
-------
00
I
00
CO
00
<
QL
*JL,
o
*
z:
LU
o:'
00
a:
x
<3
u_
_J
_J
5
2
ISORBED
cu
o
CO
C
1 ID
U (0 - 0)
0) CO /—> CO
a. a* 30)
O | •>-' H
in c e u
iw w n)
>-> co M c a.
ii 6 4-1 o c
U 0) O O 'H
« M
-------
LJ
00
n
£
1
L_
^
Q
UJ
CO
CO
_J
,C 4J
CJ 4-1 -H
0) 60 3
0) • W
-------
^^M«
^
f^S
^^J^
^fm
^^*^
^^•M
^^^^
^^^
^^^^
^Cfc.
jj^fr
•MM
ID
00
I
•
o
MMM £^
1 — esi
O w
<
o 6
0 1
NO Q
CO
00 g
^* CO
TT °°
h-
O
LU
h-
00
>J-a
cd o 1
Cfl CL, Q) G
1 T3 en o « o
6 x en -H
O f~! (13 d) 4J
O 4-i -H C CO
W) ,C 0) O r-(
TJ C 4-J XI -H QJ
(U CU -rH 4J 4-1 M
C M |S cfl M
•rH 4-1 XI 0 O
M-l CO Ol 4J O O
C M -H r-t
0 C CO > M
O O M-l • CO
C -rl M W O <-H (U •
3 CO 4J 4-1 rH T3
CO R -H C 4J 0 C
•H 0) -H 0 O >4H 3
iH M O -H -H -H O O
^-, ^-N
•-( CN;
W
H
0
f—J
1—
1
_J
d
LJ
Q.
V)
X
1
CO
,-j
CM
1
—
1
1 1 1
tr
~
o
U
CL
00
£=
2
CD
O
CJ
O
O
O
CO
CD
O
\-
o
<
OL_
O
O
CD
0)
r~
0)
3
M
•H
W
J3
M
,-J
H
M
pt,
£-<
CO
w
CM
O)
O
O)
O
LO
O
n
o
CM
1378
-------
i
»-
g
£
g
»-
S
UJ
O-
o
to
g
•~* oc
** w
W C
*- 1?
s'
0
u.
o. ui uuuSSS
1 •*
55 5 5 * 3 u 3
u. u. t u.
w
(9 UJ U. U O U 1 «B
$ S S S 2 „ 3 u,
s « o « o s s s
'i - s «
i *ftf ci w o v x SB
u.
•• ^- ta a.
^ «• - u, a J.
U.TJ -%i u. a. |f g ^
u.u. ua.tt.u'Y'V
U. Ik
U,UI U.O.B.W^^
o C"i • C! "— " 1- V
S," v» «« **Soo5£! o«i<5^ ?«
CCtl— >*— *-0*"~O v>_ _o<"">
•*"«*C T O t O m O O •? "C g v, *u lu
si 5a la j *; « la Ja sa
U
1
*
V
*
^_
* '
g
e
5
•;
?"
e S
Wl
! -1
.* «
o -o L.
f 1 3
• § «
^ »1
U £ S
5 1 •
• : 1 i
U. S 4*
•» vi i.
! s ° *
O 5 *• o
7 - c *s
"•sis
ff*
Q
= « s J
13 1. i I
H *P* O tt
x iS a
UJ <• A U
1379
-------
VI
c
o
VI
S
u
§
u
c
It
VI
§
4-1
<0
tl
VI
a
o
5 =
TJ ^
cf
CM MO 3
& fk
r-t °* S'n
& s> c "•
CO O£ g 'g
H is §
7* 3
ri« §
°S 3
U VI
•I- VI
•r- O.
S.
K •
J.2
iZ
tl
&t5
K
C
O
a
*L
u
V)
S
c
VI «— 0
•r- •»- VI 4-1
E E 01 U
•*• V* «O
*~ o S §"
•o co Q. o
«l — ^_ 0
C VI O 01
'r- 4-1 >
0 C » O
01 01
VI 01 O
1- VI
*J -f *-» Ol
Ol 3 C VI
oi cr 01 vi
e oi a n
«- c a.
VI 01
01 C CX KO
VI O 41 1
C
a. o r- -o
a u
0 6 -Q S
o o
•fc O O .4-1
•~- ** "o
— VI C
ro 4-» i- Ol Ol
01 (O "O 1-
S g £8
a. •—
V> VI fl QJ «O
*•»+•* fc. L.
Vt VI C (O QJ
41 o> o c
4J 4J ••- Vt 0)
XX U I/I
QJ QJ m 4/t 4->
•D -o a. >o 3
C C E Q. -C
•v* *p- O
00 *~
QJ QJ 2O
O O C O.
U U OJ U
U (U O *T
CM (O CM CM
r- r- CM CO
C VI
O 01
£ v* ai
•^ C* iQ lA
wi vi 10 a. v>
Q ai o. M n
vi e O H- a.
Svt O 00 O
«9 0 **-
^- Q. C 4M *^H O
»* -f O
E ^n 3C wi & -m
5 i > c
B VI VI C ID VI
•£ > > 0 C a >
C T- 0
v- X X 4-» — Jt *J
X ai a> u 4-1 be
•D -O go d S
•*- -r- E Q. *^ QJ
x o e >- r-
4) a> w o o as *^
-o c c u w *>
CO 0 4-» JE QJ
*-* O o C k4 UJ 1/1
3"" ^ £" fe""
^ CM rn
OJ
O)
(U C7» $-
VI CO)
VI T- >
•o i~ n
0. 3
0 C
H- ft) O
o e v-
•i- o o>
1_ L- irt jD
**- U =3 3 «*- U «)
O 3 O O 0 .C I.
O JZ -C *o
•M ^ *-» CO Ct
c ^r co c «t E
(U r-* CM ^ QJ O
XI -0 O U
C 4J 4-» +-» C 4-»
0. Q. L L-
Q> i/l i/l 1/1 QJ JZ rtj
-0 4J 4-» *J ^
•^ 1 1 tf
-O -O *J C
C C C **- OJ
O O O O 0 TO -0
•a -r* -^ -^ +J
4-» +-»•*-* QJ I-
Vi (TJ rt3 fO co . — rtj
t_ O O U 1 -O QJ
Q.« UOOD. >C
Q. CL O. U TJ
4Ta* v»v»[A£: XL:
cr» «3 4-* 4-> 4-* ai QJ
C tti aj CJ c. c: a> :-•>
t-CU EEE*-Ut>r-
C 4-» V» C -t~> Ul
•*-
a> *o 0*0 i<~ T- 'f o co >r~
Vi^- -^. Q r_ 4_» 4_> 4J »- ^ U QJ
vi QJ v>2a> zc:a:ia: °" "* ">
vi
c
o
*->
o
QJ
Q.
VI
1 XI X X X X X
OJ
>
0
t.
o
o
4->
"O
co CM ^o o "*• co t-j (L>
r— CM CM ro f*j ro ^a* £
o
^^ fSj ^^ ,-. , J^
O)
ro ^r ^ ^ ro ^r ^r tn «> v
V)
QJ
JC t/i
O c ro
C v» O >J cx
GJ ft) >r- 4-)
t- VI 4-» C C '»-«+-
4-> VI <_> C O 0 V»0
4O «) IO O •*- -r- C
U- Q. Q. •*- 4-> 4-» 4*1-
oc — • E-*-*O«J a QJ
O '«"• H— O O m 10 QJ -O
« -«"V»O 0«> CXCLE »- E
v>CL a. E £ - <9=}
Vt VI %* X E -^* O O h- OJ C
> dj LT» O C O U •—
L. CM V) V)OO V) O "3"
tV> ^"M'^**" > >J^
-r-£ .CiaviVI.CZ 4J
f— a. a* o oj a>i- -c -c QJ .a a* o
4-> E c -— i_ L x: >-, u «j c *-i
a)O'*-A4J *Jja^co4-» *->o \—
t/> u **- -*^to LO r- — - c\j *r oo o
o « x? ar
o • • ... o c ui
CM c r— CM n ^-trtko Xia £
O
^r in vo a:
1380
-------
VI
C
C
U
c
o
o
•o
c
Id
c
c
fl
L.
0
c
VI
g -M
_, -o
~ g V,
s t:
f"> Y\ OJ E i-
Ol 4-* CT» CT •"•"
3 3 U
V) r- W— C
C VI Cvt C C
O O O •*-
4_» 4J 4-» <_
1_ o *- O id ft
O *-» O4-> U C
CL CL r- £
O «— O- H- C
U •,- fc.-r- •— L
cx o 0,0 o
Vt VI OlX
X X Q.
•r- r— •«-! — I/I **-
EEC
jc jc E
t— D"> i— • O"t 3 */
r- 3 »— 3 E 4-» •*-
fl ,_ «^- E V
•O iO 4-i
V) vt X) •*-
vi c vt c ai -r-
Ol O Ol O U JC
+> 4-» 01 01
Ol (J O> O VI >
-O ••- XJ *- r— ••-
•»— •»- -•~^- -F-
• 01
C d v> c O.V1 4-* O
O VI 4-> O Vt+-< tO 4-1
4-J *3 4->*3 r~
t— Ol VI r— OJ Vt r— -r-
*< E t- «C E t- O VI
F- *rt .r- V* A
A vi 4-1 O a
ft «o o oo i
a B.
un »*- e •- 3
— o ox
U k
S «k |
r» v» vt O
r. 4-» 3
z: x x u -
O> Ol *O -*
i-o -o a j
c c E o
u *~ "*" o i"
~O Ol Ol 3
C C C *-• J
•-* o o c
0 «_> 0) *
Ol U
<->^— CM O4-
" CM
vt
4-»
JO
Jc
X
01
n
VI
0>
VI
VI
«
cx
>
CO
VI
O)
VI
IA
IO
0,
1
.
•o
iO
o
4-*
c
s
, *•* .
' 10 C
Ol t-
• C 4->
en c
0
3 4-»
VI JO O
Ol «O
V) 1~ O.
vt 0 E
o. cx u
«4- tO »-
•0 >
»k a o
X
CO
m
CM
1*1
i/)
IA CT)
01 C
IA at
IA C
S4-»
V)
•— H**- c
01 0 0
C -*h vi
« vt
OC vi O>
Jt D.
-1C 0
*— aj o
•MO) "0
C — Ol
OJ 4-» C
0) O) <*-
»— oO C
•M O
4-> U
CO r- 3
n «*•
VI C
4-> O
i ^
01 U
01 •*- •»- en
M JC 3 C •»- C
(A 4J CT O U -r-
10 en at •*- o> u
CX C 1- 4-» CX 3
0) « *A U
«4- C C U
O ** O f- E H-
«A >*- ««- 3 O
4k 4J «r- E r-
«*- 3 U 0) C t-
O E -r- CX -*- JC
t- H- Vt E
4J C -^- CO
C •»- O VI Ol O"
oi • E 01 4-» >
"O CX Ol Ol O
c a> vi o> ••" 4->
0> > E JC
CX o> vi O t-
Ol -•- 4-* >» « JC
•O JC 01 •—
C O O> *— 1— i—
•r- <0 E « O 4
c z u •
JQ *— o» en v> 4-»
4-1 -O U 01 t-
O C 3 10 O 10 1
4-* O» r- E "O
4-* v> •*-* a> i
IA vt oi en oi r- >
t- •/-•4-»cnccn jD
10 • vt - — • L- *a o *o *o *-
CX-Q O C CX r- 4-» ,— tO
CL IO OO V* VI Vt > |
10 E i- .— 10
O 4-» 4-» O -*-» U 1
JCOl 4->V1Ofc-OU Xf
en 10 •— o cx cx
c M- o — *- a* :
L. Ol **- O **> iO T
4-> (A C r— Vt O— ' O i
1/lVtOVI" 4-> 4-» . l/t (
ID ••" r— *** VI VI V* ^— i
0 °" «0 •»- 1- -^ C T- C •* "*
-<— O •— !_•>- 10 OO> OOI L- d
vi jt o> oivt cx VIE *rt e oi c
Vt4->l_WCX Ol OI4~*<
Ol t- «a ro *> t_ 4-» t- rtj t
I- i- o E O u -^ i-^- £
O- OJ O r— >> <0 =3 10 D •(
E +-> . — i — CL O" CX O < — C
O«*-O« — *»-C Ol Ol •— n
O«J Z <0 O •— I- •— »- < W
"H T. T.
1 W3 ^J ^J
CM CM r*)
CM
X 1 «0 X >
m ro CM
fl
\o o *f co CM 10
10 rx rx. r-» CO co
r- r— CM ro «r in
••a- -«r «T- -o- •* •* m*£
ur> ui ui mu> ui ui tr
c >
VI O 4-1
01 •*- **-
Vt 4-» C C V
VI U C O O C
IO 10 O **• •«- <
• — ex ex >*- 4-» 4J o
r- E 4~> U U
VIM- O O «OIO Ol t-
0- o U*E OO 4-> #-
CM 0 U U C
V Vt Vt U VI 3
> > >* *< > >i z;
•f >* *J
VI JZ JC vt Vt JT -r- VI
CX 4-* 4-» V» > > 4-» r- >
en en > en «*• CO *-* » LJ
tO (/» ^- CM ^ OO >
« -c
u c
X «
^- con ^r*n vo uj^
in *£
U Ol
71 4-» 4~»
« jc IA o
4-1 IO C
J ^ ^"^
Q a> o> oi vi
= ^ *J *J C
& JC V» Oj
* ° i
-vi Oi *"* L
O C 33
Ol O> IO CT
o e jc jc oi
1 Ol 4-» i-
r 1- 0
3 -r- 4J 4-> C
J 3 "O O
cr "- L. |-f-
1> 01 > Ol*-*
. L. O -r- 1 «O
O I- I- 1 O
c cx cxl.-
>> o **-
J -r- »— X) •*-
4-> r- CO
A
J L- d»
3 Ol V- O OJ
= ~O O C E
3 > t- OJ >^
: o a> r: r-
CL E «0
S U C J3
- 4-* I-
n "x vt CL
^ 4-* V>
U L. - M -—
CL VI -r-
\t CX r- jt
S CO O QJ
> O vt D>
4-» O TD
7 *-» O) 3
= V. 1- —
a 4-» T3 -r- V>
•* I-
CM o oi CL
.— JC
**- 4-> r-
O vt
•r- CO
X O 4-1
E •*-* T3 *—
o> ci •*-
ai c • E o
Qi Ol 4-* I- Vt
-o u i. o
3 4_> JQ »*_ 4-1
r- Vt CX (- i-
vt O) 10
C ^* CX D.
-— o
•r~ -r~ O VI . —
O VI 4-1 -.-
VI • VI **-
c vi a> 4~> en o
«0 4-> U V, C
O OL «) f- X
C 10 £ CX NI -
O CX O -f- E
E 0^ c
•O -r- O) QJ
Ol "O •*- Co Cn
EQJ Ol O O TD
•— c e 3
03 -^ X 0 —
01 JC O "O • —
CX O O Ol C •»-
vi c en
*t- O» C 4-> . — V>
o c ai «a 10 «o
•«— -r— CL
L. i_ t- «3 U
J Olttl 01 •»* Ol T -— •
jo -a jt u 4-> -o
§•••" O Ol fO " O>
vi . — 4-» E
-------
NUCLEAR WASTE DENSIFICATION
BY
DYNAMIC COMPACTION
Cliff Schexnayder, P.E.
Chief Engineer
Nello L. Teer Company
P.O. Box 1131
Durham, N.C. 27702
(919) 682-6191
Robert G. Lukas, P.E.
Senior Principal Engineer
STS Consultants, Ltd.
Ill Pfingsten Road
Northbrook, Illinois 60062
(708) 272-6520
Introduction
Dynamic Compaction was used to achieve densification of
buried nuclear waste at the Department of Energy's, Savannah
River Plant. This procedure was the first step in the permanent
closure capping of fifty-eight acres of buried low-level waste
within the plant's Mixed Waste Management Facility. Before
constructing a RCRA standard clay cap, the waste was compacted
to reduce the potential for future subsistence which could
possibly crack the permanent cap.
During the operation of this part of the burial ground from
before 1976 until 1986, wastes had been deposited in a series of
parallel trenches. The trenches were 20 feet wide by 20 feet
deep with each trench separated by a 10 to 20 foot berm of
natural undisturbed soil. The lower 16 feet of the trenches
were filled with waste. In most cases, the waste had been
simply dumped into the trenches. However, some trenches had
been filled with waste which had first been placed in metal
boxes. These metal boxes, known on site as B25 boxes, are
similar to connex containers. Sometimes the boxes had been
stacked in an orderly matrix within the trench, but some B25's
had been randomly dumped into trenches. In all cases, loose
dumped waste or boxed waste, the trenches had been covered with
four feet of sandy silt.
The nuclear waste consisted of miscellaneous materials that
had been exposed to nuclear radiation, including clothing,
building materials, metal vessels, pipes, construction
1382
-------
equipment, and fluids, such as oil, that were mixed with
absorbent substances and placed in 55 gallon drums. The waste
is classified as ranging from low level to intermediate level
beta gamma.
It was observed that the initial soil cap which had been
shaped to shed surface water was settling and water was
beginning to pond in the low spots. This was considered
undesirable since there was the likelihood of surface water
seeping through and becoming contaminated from the nuclear
deposits. The contaminated water could then possibly percolate
downward to the groundwater table. To alleviate this problem,
it was decided to densify the nuclear waste within the trenches
to reduce future settlement and then to construct a new
impervious cap.
In-Situ Improvement by Dynamic Compaction
Dynamic compaction is the process of dropping heavy
tampers, typically in the 6 to 30 ton range, from heights
varying from 30 to 100 ft. The tamper is raised and dropped by
a single cable with a free spool which results in an energy lost
of about 12% due to drum and sheave friction. On some projects,
the tamper has been allowed to free-fall. In both situations,
the high impact energy imparted to the soil causes deep
densification. Dynamic compaction has been described in
numerous technical papers (Charles et al, 1981; Leonards et al,
1980; Lukas, 1980, 1985; Mayne et al, 1984; Menard and Broise,
1975). The advantages and disadvantages of dynamic compaction
are outlined by Lukas (1986) in a FHWA study. The process is
ideal for compaction of nuclear or hazardous waste for several
reasons.
1. Densification of the buried nuclear waste takes place
from the existing ground surface without exposure to
the waste. This is a critical requirement for the
safety of construction personnel.
2. The weight of the tamper and the drop height can be
adjusted to insure that compaction is obtained to the
depth and degree necessary. In general, the heavier
the tamper and the higher the drop height, the greater
the depth and degree of compaction.
3. Wherever resistant materials are encountered below
grade, additional energy can be applied to crush drums
or displace large objects, thereby collapsing potential
voids within the waste.
4. Dynamic compaction is generally the most economical
site improvement process and for deep densification, it
1383
-------
is one of the safest. Other methods such as stone
columns would have had to penetrate the nuclear waste
and thereby expose construction personnel to radiation.
Excavation followed by recompaction with conventional
compaction equipment would, also, have led to
unnecessary exposure.
Project Specification for Densification
The project specifications stated the following:
1. Prior to any waste trench being treated by dynamic
compaction, a 2 ft. thick (+3",-0") soil blanket would
be placed on top of the existing cover. This fill had
no compaction requirement. (Most burial trenches had 4
ft. of initial earth cover so there would now be
effectively 6 ft. of cover.)
2. Dynamic Compaction:
a. Tamper - 20 tons with a flat bottom, eight (8) ft.
diameter.
b. Drop Height - 42 ft.
c. Drop Pattern - The trench surface was subdivided
into 10 ft. x 10 ft. grids designated primary and
secondary grid drop locations. All primary
craters within a work area would be compacted and
backfilled prior to dynamic compaction of the
secondary grid drop locations, Fig. 1.
d. To facilitate tamper recovery, the tamper would
remain attached to the crane cable during all
drops.
e. Each crater would be driven using 20 drops, or
until a maximum crater depth of six (6) ft. was
achieved. This maximum depth was specified for
safety reasons, in order not to encounter the
radioactive waste material. It was expected that
on the average, 4.5 ft. deep craters would result
from 20 drops.
3. Crater Backfilling:
a. Place a uniform 4'-0" (+6") loose lift of fill
material into the driven crater.
b. Compact the loose fill by dropping the 20-ton
tamper, five (5) times from 42 ft.
1384
-------
TRENCH WIDTH
GRID PATTERN
PER SPECIFICATIONS
8 ft.(2.4 m)
50 ft.2 (4.6m2)
P = PRIMARY DROP POINTS
S = SECONDARY DROP POINTS
FIG. 1. Dynamic Compaction Tamper Drop Pattern, Mixed Waste
Management Facility, DOE, Savannah River Plant
138
-------
c. Continue backfilling and compacting as outlined
in a & b until a 2'-0" maximum crater depth,
measured from the surface of the soil blanket, is
obtained.
d. Backfill would be compacted to 95% of maximum dry
density (ASTM D698-79) at +2 percent of optimum
moisture.
Dynamic Compaction Equipment
The machine utilized to perform this dynamic compaction
work was specifically designed for the task, a Lampson LDC-350
"Thumper," Fig. 2. During compaction operations, the quick
release and sudden stop, when the tamper strikes the ground,
cause the boom and upper works of machines used for this work to
experience severe rocking. The severity of this motion places
unusual stresses on the undercarriage of full-revolving cranes.
Consequently, the LDC-350 has no turntable and the upper works
are fixed to the undercarriage, not pined. Each track of the
machine has an independent motor. Therefore, instead of
revolving, the machine is turned by counter direction travel of
the tracks.
The LDC-350 has a larger than usual diameter hoist drum and
a duel braking system. The braking system is a combination of
an air-operated caliper disc and a non-self energizing 60 inch
diameter band brake. The main brake applies sufficient
resistance so that the tamper can be stopped and held at a
desired height. It is rated to hold a 50-ton load. The second
brake is more like a drag on a fishing reel. When dropping the
tamper, the operator applies the drag brake just before the mass
strikes the ground. This prevents drop line backlash. If, as
on most other machines, the operator has to use the main brake
for this purpose, there can be severe damage to the machine when
the brake is applied prematurely. With the combination system,
the operator cannot inadvertently shock-load the machine by
attempting to stop the dropping tamper.
The LDC-350 boom is raised into its operating position with
an erection line and then tied off with two (2) rear boom
pendants. Additionally, there are two front kickback pendants.
Once these pendants are connected, the boom angle is fixed and
there is no stress on the boom hoist line drum. This is a
separate hoisting system, independent of the system used for
dropping the tamper.
Elevating scrapers, Fig. 3, were used to haul backfill
material to each crater location. A 335 HP track bulldozer,
Fig. 4, would then push the fill into the crater. One blade
load of a machine this size provided all the required backfill.
138G
-------
1 ft.= 0.305m
1 ft.2 = 0.093m 2
1 U.S. ton = 0.91 mt
120 ft. NO. 22 M.E.C.
BOOM W/ 4.5' OFFSET
TIP & BOOM POINT
SHEAVE ADAPTER
BOOM PENDENTS
ERECTION PENDANTS
20 U.S.TON
DROP WEIGHT in
iAREA 50 ft/ g
DROP LINE
BOOM RAISING
TIEOFF AFTER
ERECTION
FIG. 2. Lampson LDC-350 "Thumper," Dynamic Compactor, Mixed Waste
Management Facility, DOE, Savannah River Plant
1387
-------
FIG. 3. LDC-350, "Thumper" and Elevating Scraper Hauling Crater
Backfill, Mixed Waste Management Facility, DOE, Savannah River
Plant
1388
-------
FIG. 4. Bulldozer Pushing Crater Backfill, Mixed Waste Management
Facility, DOE, Savannah River Plant
1389
-------
This meant the dozer was idle a large portion of the time;
however, the effect on total cycle time per crater justified the
use of such a large machine. Employing such a large machine
enhanced safety since it could fill the crater without having to
maneuver directly under the hoisted tamper.
An analysis of hoist line wire rope performance was made
during the first 60,674 dynamic compaction tamper drops. This
represented about 25 percent of all project drops. A summary of
that data is presented in Table 1. From the analysis, it was
decided that new 1 1/2 inch 6x25 IWRC (Independent Wire Rope
Core) wire rope would be used on the hoist line. This decision
was based on safety. Whereas the new 1 1/2 inch 6x25 had a
better average number of drops than the 1 1/2 inch 6x37 surplus
rope, the cost difference was 255 percent greater while the
performance was only improved by 11 percent. Another point of
interest from the Table 1 data is the performance of the 1 1/2
inch 6x41 rope. This rope contains too many fine wires and is
not good for dynamic compaction type work.
In most cases, the line was replaced before complete
failure, because periodic inspection noted distress. The most
common distress observed was broken and crushed wires at the
point where the extended cable would break over the boom's point
sheave. There were, however, three sudden separation failures
during this analysis phase at the beginning of the project.
During the next 86,419 drops, there were 13 replacements of
new 1 1/2 inch 6x25 rope. The best rope life was 7,530 drops,
the least was 5,514 drops and the average was 6,648. There were
no further sudden separation failures during the remainder of
the project. This can be attributed to the prescribed
inspection and replacement procedures which resulted from the
analysis. At 5,000 drop cycles, close visual inspection of the
rope was performed on a weekly basis. The inspection included
climbing the boom in order to view the 42 feet of rope that was
continuously running over the point sheave. Additionally, once
a rope had experienced 7,000 drops, it was replaced during the
next machine maintenance period even if inspection did not find
evidence of excess stress.
Dynamic Compaction Test Program
Three test sections were proposed, each was the full width
of the trench and a minimum of 200 ft. long. Two sections were
situated in the low level alpha trenches and one section in an
intermediate level trench. The first drop at each location was
from a height of 42 feet to confirm that there was not a loose
layer below an upper crust. For the second drop the height was
varied as shown on the Table 2. If the difference in crater
depth between the first and second drop was less than 1 foot,
1390
-------
Table 1. Dynamic Compaction Hoist Line Wire Rope Study
for 20-Ton Tamper, Dropped by a Single Line from 42 Feet;
Boom Height - 130 Feet
AVERAGE
NUMBER
OF DROPS
(1)
1450
1753
2780
4499
4865
6058
6725
SIZE
inch
(2)
1 1/2
1 3/8
1 3/8
1 1/2
1 1/2
1 1/2
1 1/2
CLASS
IWRC
(3)
6x41
6x19
6x37
6x25
6x19
6x37
6x25
PURCHASED
NEW/ SALVAGE
(4)
New
Surplus
Surplus
Surplus
Surplus
Surplus
New
NUMBER
OF
REPLACEMENTS
(5)
1
1
2
1
2
4
2
1391
-------
Table 2. Contractor Dynamic Compaction Test
Program Drop Height Sequence, Mixed Waste
Management Facility, DOE, Savannah River Plant
NUMBER
OF DROP
POINTS*
(1)
4
4
4
4
4
4
Height of Drops, Ft.
1ST
DROP
(2)
42
42
42
42
42
42
4 1 42
2ND
DROP
(3)
42
42
42
42
REMAINING
DROPS
UNTIL CRATER
DEPTH OF
(4)
50
60
70
80
60
70
80
6 FT
All tests were with a 20-ton tamper.
*A minimum limit, repeat the most promising drop
height.
1392
-------
then additional drops were undertaken from the height specified
in the test program until such time as the crater depth reached
6 feet. The reason for using the higher drop heights was to
achieve the compression as quickly as possible with the least
number of drops. Because the 20-ton tamper had already been
constructed, no variation in tamper weight was attempted.
If, after any individual impact the crater depth was more
than 1 foot deeper than the previous depth, the drop height was
maintained at 42 feet until the incremental crater depth was
less than 1 foot per drop.
Safety was maintained during the program by:
1. Using the reduced drop height during the initial
tamping to confirm there was no weak spot directly
below an upper stiff layer.
2. Using the incremental crater depth measurement of 1
foot maximum per drop as an indicator for reducing the
drop height.
3. Limiting the crater depth to 6 feet.
4. Measuring for nuclear emissions at all times with air
monitors and wipe tests on the tamper.
Monitoring was undertaken during the test sections and
consisted of the following:
1. The depth of crater following each drop was measured.
2. The volume of the crater was determined by using a
depth measurement, a top of ground diameter measure-
ment and the known diameter of the tamper for the
bottom of the crater measurement. The volume of the
crater was determined for each drop.
3. Long spikes were driven into the ground adjacent to the
craters from which ground elevations were obtained to
determine if heave of the adjacent land mass was
occurring. Heave was compared with the volume
measurements obtained under Step 2 which was an
indicator of how effective each drop was in compacting
the mass.
4. The time taken to complete each test section was
monitored to determine the most efficient dynamic
compaction procedure.
5. During the dynamic compaction of the test sections,
1393
-------
measurements of peak particle velocity were taken at
the ground surface with a seismograph. Seismograph
readings were obtained at distances of 25, 50, 75, 100
and 125 feet from the drop point in both down-treach
and cross-trench directions.
Test Sections D-4, D-5 and E-10
Trench D-4, which was 390 feet long, contained low level
alpha waste. Trench E-10 was 239 feet long and contained
intermediate level waste. In both areas the miscellaneous
nuclear contaminated debris had been either dumped loosely into
the trenches or was in cardboard boxes placed within the
trenches. Test Section D-5, a 200 foot long portion of Trench
D-5, was filled with randomly dumped metal B25 boxes containing
low level alpha nuclear waste.
At all three test sections, the drop pattern was undertaken
as shown in Figure 1. For the initial drop points on both the
primary and secondary pass, the first two drops of the weight
were both from 42 feet after which the following drops were all
from a higher height. It was immediately apparent that the
advantage was very slight for the 50 foot height. Therefore, on
the second set of tests, the first drop within each test section
was undertaken from a height of 42 feet, and then the additional
drops were undertaken from heights varying from 60 to 80 feet.
The number of drops reguired to reach a crater depth of
approximately 5.5 feet at test sections D-4 and D-5 is
summarized in Figures 5 and 6.
For Test Section D-4, the most efficient method of applying
the energy was to use the highest drop height, in this case 80
feet. After the initial drop from 42 feet, it took only 5.8
additional drops from a height of 80 feet to reach the reguired
crater depth at the primary grid points, and approximately 7.4
drops at the second grid points. The number 5.8 and 7.4
represent an average for various locations, thereby resulting in
something other than a whole number of drops. The amount of
energy applied for the various drop heights is summarized in
Tables 3 and 4. It can be seen that approximately the same
amount of energy was applied for each grid point, even though
the drop height and number of drops varied. At the D4 primary
grid point locations, the average energy reguired to achieve the
densification was approximately 9,174 foot/tons. While at the
D4 secondary grid points, the average energy reguired was 13,387
foot/tons. More energy is reguired for the secondary grid
points, because some densification takes place in these areas
during the impacting at the primary grid point locations.
1394
-------
14—i
CO
812
CC
U.
O
cr
UJ 1A
CD
UJ
I
UJ
8^
INITIAL 1 OR 2 DROPS
FROM 42 FEET. REMAINING
DROPS AT DROP HEIGHT
SHOWN BELOW.
SECONDARY PASS
PRIMARY PASS
D
SFCTION D4
40
50 60 70
DROP HEIGHT - FEET
80
90
FIG. 5. Number of Drops from Varying Drop Heights to Induce a
Crater Depth of 5 1/2 Feet, with 20-Ton Tamper, Test Section D-4,
Mixed Waste Management Facility, DOE, Savannah River Plant
12-
oo
0-
o
o 10 —
a:
UJ
CD
UJ
o
UJ
SECONDARY PASS
INITIAL 1 OR 2 DROPS
FROM 42 FEET, REMAINING
DROPS AT DROP HEIGHT
SHOWN BELOW.
PRIMARY PASS
SECTION D5
55
60 65 70
DROP HEIGHT - FEET
75
80
FIG. 6. Number of Drops from Varying Drop Heights to Induce a
Crater Depth of 51/2 Feet, with 20-Ton Tamper, Test Section D5,
Mixed Waste Management Facility, DOE, Savannah River Plant
1395
-------
Table 3. Energy Required to Induce Crater Depth of 5.5 Feet at
Primary Drop Points, Mixed Waste Management Facility, DOE, Savannah
River Plant
Test
Section
(1)
D4
D5
E10
Energy
Energy
1
1
1
1
1 50
1 (2)
1
| 8,680
1
1
- Weight
Units in
Drop Height - Feet
60
(3)
9,276
11,040
9,792
1 1 1
1 1 1
65 | 70 | 75 |
(4) | (5) | (6) 1
1 1 1
1 9,380 | |
11,430 | 11,060 | 10,185|
1 1 1
of Tamper x Drop Height x Number of
Table Expressed in Foot-Tons
1
I Average
| Energy
|A11 Drop
80 (Heights
(7) | (8)
1
9, 360| 9,174
| 10,929
1 9,792
Drops
Table 4. Energy Required to Induce Crater Depth of 5.5 Feet at
Secondary Drop Points, Mixed Waste Management Facility, DOE, Savan-
nah River Plant
Test
Section
(1)
D4
D5
E10
50
(2)
Drop Height - Feet
1 1
1 1
1 1
60 | 65 | 70
(3) | (4) | (5)
1 1
1 1
14,040 | | 13,440
11,740 | 12,310 | 11,860
I | 12,684
75
(6)
11,010
80
(7)
12,680
Average
Energy
All Drop
Heights
(8)
13,387
11,730
12,684
Energy = Weight of Tamper x Drop Height x Number of Drops
Energy Units in Table Expressed in Foot-Tons
1396
-------
Test Program Heave Measurements
Heave measurements were taken during driving of 41 Trench
D-4 craters. Twenty-two of these were primary craters and 19
were secondary craters. At the primary craters, heave occurred
at 16 and a mixture of heave and settlement at three (3). At
the closest measuring point, which was seven (7) feet from the
center of the crater, heave was on the order of six (6) inches.
This is not considered significant. Volumetric calculations of
ground displacement indicated that heave ranged from 15 to 25
percent of the crater volume. On that basis, it was concluded
that most of the dynamic compaction energy was being transmitted
into the ground, causing compression.
During driving of Trench D-5 primary craters, heave was
generally less than six (6) inches when measured seven (7) feet
from the center of crater. At the secondary craters, the ground
adjacent to all craters exhibited heave. This is to be expected
because of the area wide densification effected during
compaction of the primary crates. At four locations, the heave
adjacent to the secondary crater was in the range of six to ten
inches; this was not considered major.
Ground heave was more noticeable at the Trench E-10
craters. At the primary craters, heave was not significant.
However, at twelve (12) of the 23 secondary craters, heave was
in the range of six (6) to twelve (12) inches adjacent to the
crater. This was still not considered an excessive amount of
heave and it was concluded that most of the energy was still
effective in causing densification.
Test Program Seismograph Readings
Seismograph readings were taken both parallel to the trench
and perpendicular. This was done because perpendicular to the
trench the ground vibrations were transmitted through both fill
and through natural soil, whereas vibrations parallel were
transmitted entirely through waste fill. In order to minimize
damage to adjacent facilities, it was recommended that the peak
particle velocity be kept to about 1 inch per second or less.
When using a 20-ton tamper from a height of 75 feet, the
Trench D-4 data for the parallel case translated into a reguired
distance of 79 feet from the point of impact. The data points
for peak particle velocity measurements taken perpendicular to
the trench exhibited wider scatter. This is attributd to the
ground vibrations traveling through both loose waste and dense
natural soil. For the 20-ton tamper and a height of 75 feet,
the safe perpendicular distance was 72 feet.
The magnitude of the ground vibration produced during
1397
-------
dynamic compaction of Trench D-4 was representative of the
Trench D-5 and E-10 data. Therefore, in order to limit the peak
particle velocity to 1 inch per second or less, it was
recommended that for dynamic compaction utilizing a 20-ton
tamper and a 75 foot drop height, a clear distance of 79 feet in
line with critical objects or 72 feet for objects at right
angles to the trench be maintained. If compaction had to be
performed at distances less than the above values from critical
facilities, the drop height should be reduced and more blows
applied to achieve desired crater depth.
Field Operations
A Spectra-Physics El-1 electronic level laser system was
used to check the final depth of every crater. Average
incremental crater depth for primary craters was 0.51 feet and
0.40 feet for secondary craters. This difference was to be
expected because the construction sequence created a stiffer
matrix around the secondary craters. Some voids were
encountered with resulting incremental depths as high as 1.73
feet per blow. The result of achieving an average incremental
depth of 0.45 feet per blow was that on the average, only 13
drops were required to drive the craters. The resulting average
crater depth for the project was 5.63 feet.
The average ground compression was 12.85 percent, as
computed by the following expression:
D x Am
AG = T
G.S. x D^
r
where: AG = average percentage ground compression
D^ = depth of crater: 5.63 ft
O
AT = area of tamper: 50.2 ft2*
DF = depth of fill: 22 ft**
G.S. = grid spacing: 10 ft x 10 ft
*For 8 ft diameter tamper
**Two ft blanket plus 20 ft trench
The craters could easily have been driven deeper but as a safety
measure, driving was stopped when the depth reached or passed
the 5.5 foot mark. This policy was instituted when it was
realized that the incremental depths being experienced were
close to one-half foot.
1398
-------
The effect of the B25 boxes which must have had large void
spaces both within and between boxes was apparent. In the D
area, the average number of drops for all craters, primary and
secondary, in the B25 trenches was 11.7. The average number of
drops for trenches having random mixed waste was 14.1. The
point should be made that this is not an equal comparison; the
11.7 drops in the B25 trenches produced an average crater depth
of 5.69 ft. The 14.1 drops in the mixed trenches produced only
a 5.41 ft. deep crater. Table 5 further illustrates the
differences between miscellaneous mixed waste trenches and
trenches with B25 boxes.
The difference between expected average crater depth, 4.5
ft., and required blows to achieve as stated in the project
specifications, and what was actually realized during the
project, was the result of the differences in the equipment used
during the original design test program and the machines
selected for use by the contractor. The result was a final
product very close to the high end of the expected 11 to 13
percent waste matrix compression versus 12.85 percent actual.
A depth of greater than 5.5 ft. was achieved with less than
20 blows for 85.6 percent of the crates. The average depth for
those 1,875 craters which did receive the specified maximum 20
blows was 4.9 ft. The average number of backfill 42 foot drops,
was 6.78.
Initially, there were problems with backfill compaction.
There was one trench in the D area which required a total of 517
drops to drive the craters and 521 drops to compact the
backfill. The specifications called for a backfill density of
95% of standard proctor. Standard Proctor Energy is 12,375 ft.-
Ibs./cubic foot. Additionally, the specifications limited the
backfill lift to a maximum of 4.5 feet and called for five drops
of the 20-ton tamper to compact. A 20-ton weight free falling
from 42 feet imparts 1,680,000 ft.-Ibs./drop. Mechanical losses
for the LDC-350 using a single line are 11.1 percent.
Therefore, for five drops, the resulting energy is 7,467,600
ft.-Ibs. A 4.5 foot lift 8 feet in diameter is 226.2 cubic feet
or 33,014 ft.-Ibs./cubic foot. This is 2.67 standard proctor
energy. Heave or rebound was experienced in the top part of the
compacted backfill. Additionally, it was found that with so
much energy, the backfill operation was actually a secondary
driving operation.
To correct this situation, a change was instituted in the
backfill operations. The number of blows required was adjusted
to the depth of backfill lift and the lift thickness increased.
When the thicker backfill lift was tried, difficulties were
experienced with keeping the weight level. The final procedure
adopted was to fill the crater completely including about one
1399
-------
Table 5. Dynamic Compaction Results, "D" Area, Mixed
Waste Management Facility, DOE, Savannah River Plant
TYPE
OF
TRENCH
(1)
Loose Mixed
Waste Trenches
B25 Trenches
All Trenches
Primary Ci
AVERAGE
CRATER DEPTH
ft.
(2)
5.50
5.79
5.59
raters
AVERAGE
NUMBER
OF DROPS
(3)
12.79
9.75
11.85
Secondary C]
AVERAGE
CRATER DEPTH
ft.
(4)
5.31
5.60
5.40
raters
AVERAGE
NUMBER
OF DROPS
(5)
15.34
13.62
14.81
1400
-------
foot of overfill; drop the weight one time from 15 feet to take
the fluff out of the loose fill; then push in additional soil so
as to again completely fill the crater; and finally, at this
point, to apply five drops from 42 feet. This resulted in an
energy application of about 2.2 times standard proctor. Density
tests were taken at different levels in the compacted backfill
to verify that the 95 percent compaction specification was being
achieved for the full depth, Table 6. Once the procedure was
proven, it became the standard for the project.
Backfilling normally required two fillings. The first was
as described above. The second filling, after the initial full
depth and 5 blows, usually had a depth of less than three feet.
For this second filling, only 2 blows from 42 feet were applied
to achieve compaction.
It should be noted that the first backfill step which
imparted roughly 2.2 times standard proctor energy, is about the
limit that can be applied effectively. When too great an amount
of energy is applied, the material is found to rebound or
experience tension in the uppermost portion. Even at 2.2 times
standard proctor energy, half the tests performed in developing
the revised procedure, Table 6, had greater densities at the
five (5) foot depth than at the three (3) foot depth, test 1, 3,
9-13.
When a second backfilling was required because the
compression during the first filling left a crater having a
depth greater than two (2) feet, the energy level applied was
about 2.4 times standard proctor. In this shallow crater
situation, there was insufficient lateral restraint provided by
the two ft. of uncompacted soil blanket and only one foot of
original trench fill. The backfill material would be driven
laterally in many cases. In some cases, heave of as much as two
feet resulted at the edge of the crater during backfill
compaction. The heave would taper out over a distance of about
six feet. This was a heave situation during backfilling and
should not be confused with the nominal heave observed during
initial driving of the craters.
The majority of the craters were completed by:
--Thirteen initial driving blows from 42 feet.
—A first backfill with one 15 foot fluff compaction
drop and five 42 foot compaction drops.
—A second backfill with two 42 foot compaction drops.
There were 13,002 drop point locations in the project area.
These required a total of 161,096 initial driving drops and
1401
-------
Table 6. Crated Backfill Density as Percent of ASTM D698-79
Maximum Dry Density, Backfill Compacted According to Revised
Procedure, Mixed Waste Management Facility, DOE, Savannah
River Plant
TEST
NUMBER
(I)
I
2
3
4
5
6
7
8
9
10
11
12
13
14
TRENCH
(2)
D4
D5
D5
D5
D5
D5
D5
D5
D5
D5
D5
D5
D5
D5
HOLE
(3)
PI 5
S5
S6
S7
S8
S9
S10
Sll
S12
S15
S16
S18
S19
S20
COMPARATIVE
DENSITY
%
(4)
100.7
102.3
99.0
91.8*
97.4
98.0
101.9
98.9
98.4
97.7
99.0
98.5
103.6
101.0
103.4
| 102.2
I
98.4
| 103.1
| 100.7
102.3
| 100.3
104.0
95.0
| 99.3
| 98.8
103.2
102.6
97.6
DEPTH OF
TEST
ft from
surface
(5)
3.0
5.0
3.5
5.0
3.0
5.0
3.0
5.0
3.0
5.0
3.0
5.0
3.0
5.0
3.0
5.0
3.0
5.0
3.0
5.0
3.0
5.0
3.0
5.0
3.0
5.0
3.0
5.0
*Moisture was too high, greater than 2% above optimum.
1402
-------
85,879 backfill drops. The first machine mobilized, worked for
five days before the second machine was ready to begin. These
two machines worked from March thru July 1989. In August, a
third machine was mobilized and three machines were utilized
from 11 August 1989 until 21 February 1990. The total dynamic
compaction duration was 357 calendar days. That period included
205 workdays, 39 days lost due to weather, and 113 non-workdays
(Saturday, Sunday & Holidays).
Of the total work time available, 86.4 percent was
productive. The remaining 13.6 percent was lost to machine
availability. Considering production time only, 3.35 craters
were driven and backfilled per hour or conversely, it required
17.9 minutes to complete all work at a crater location. Twelve
and one-half minutes were required to complete the 20 driving
and backfill drops. The remainder of the time was for pushing
backfill and positioning the machine. Weather was not a major
hinderance. The machines could not work in the rain or lighting
and sometimes the backfill material became too damp, but
weather accounted for only 11 percent of all non-work time. The
dynamic compaction phase of the project was completed three
months ahead of schedule.
Dynamic Cone Penetrometer Measurements
To check on the degree and depth of improvement throughout
the full depth of the nuclear waste facility, dynamic cone
penetrometer readings were undertaken. These tests were
performed after the 2 ft. soil blanket had been removed.
Therefore, both the initial and final tests were from the same
elevation, top of the original cap.
The dynamic cone penetrometer consists of a 2 inch diameter
conical cone with a 60 degree cone angle which is connected to a
drill rod that is driven by a 140 pound hammer falling 30
inches. This cone is driven into the nuclear waste and the
number of blows per foot that are required to advance to the
penetrometer are recorded. Readings were taken in the nuclear
waste before and after dynamic compaction. In all cases, there
was a significant increase in the penetration resistance
following dynamic compaction. Fig. 7 is the test results for a
specific location, B25 Trench D-13 at secondary crater S-21,
while Fig. 8 is at primary crater P-53 of miscellaneous mixed
waste Trench A-4. An illustration of the range in cone
penetrometer values before and after dynamic compaction covering
two different areas of the burial grouns is presented in Fig. 9.
Typically, the cone penetrometer values before dynamic
compaction were found to range from about 5 to 20 blows per foot
with occasional higher values. It was assumed the high values
were the result of encountering large objects within the waste.
1403
-------
Feet Below Grade
0
Backfill
Estimated Base
of Crater
10
Soil / Waste Matrix
15
Estimated Base
of Trench
In Situ Soil
25
30
x
0.1
10
100
Blows/Foot
Pre-Densification
Post-Densification
FIG. 7. Cone Penetroraeter Data for Trench D-13, Containing Metal
"B25" Boxes, Secondary Crater Location S-21, Mixed Waste Management
Facility, DOE, Savannah River Plant (from Chas. T. Main, Inc.
Letter Report May 31, 1989)
1404
-------
Feet Below Grade
Backfill
Estimated Base
of Crater
Soil / Waste Matrix
Estimated Base
of Trench
In Situ Soil
I ! I I I I I
100
Pre-Densification
Post-Densification
FIG. 8. Cone Penetroraeter Data for Trench A-4, Containing Miscel-
laneous Mixed Waste, Primary Crater Location P-53, Mixed Waste
Management Facility, DOE, Savannah River Plant (from Chas. T. Main,
Inc. Letter Report December 20, 1989)
1405
-------
UJ
Q
o
0
LU
m
UJ
l_
U.
U
-10-
~z(J
jU
/ !##/>
R^GE OF xx RANGE OF ^X\
K™ACI10N N. POST COMPACTION N
; CPT TESTS N - ^ CPT TESTS
X
N
s
*
II I II I II I I II I I
BACKFILL
SOIL / WASTE
MATRIX
X
\
\
\
IN SITU SOIL
I I I I
0 10 20 30 40 50 60 70 80 90 100
BLOWS/FOOT ift.=.304m
FIG. 9. Comparison of Cone Penetrometer Test Results, "A" and "D"
Areas, Mixed Waste Management Facility, DOE, Savannah River Plant
140G
-------
After dynamic compaction, the typical cone penetrometer values
showed a relatively uniform penetration record throughout the
entire depth of fill, a desirable result.
Conclusions
Based upon the data from the actual dynamic compaction of
this 58 acre site, it is concluded that:
1. Densification of the nuclear waste was accomplished by
the dynamic compaction procedures.
2. The craters formed by the impact of the drop weight
averaged 5.63 feet which resulted in an average ground
compression of the nuclear waste of approximately 12.85
percent.
3. Cone penetrometer tests taken before and after
dynamic compaction indicated a significant increase in
the penetration resistance, thereby confirming the high
degree of densification within the nuclear waste
deposit.
4. Safety was maintained at all times by limiting the
crater depth to 6.0 feet. Measurements were taken with
air monitors around each compactor, monitoring of
craters before backfill, and wipe tests of the weights
during the entire dynamic compaction construction
period. These confirmed that radioactive debris was
not discharged.
Acknowledgements
The authors wish to thank Mr. Jay D. Rasmussen, Civil
Engineering student, Iowa State University and Miss. Ann M.
Schexnayder, Nello L. Teer Company, for their efforts in
assembling and sorting the raw field data from which this paper
was developed. Mr. Patrick E. Geluso, P.E., and Mr. William N.
Lampson of Neil F. Lampson, Inc. gave valuable counsel and
shared their knowledge concerning machine mechanics. A special
word of appreciation goes to Mr. Walter D. Munn, P.E., Senior
Editor, Highway & Heavy Construction, for his advice and
guidance concerning equipment and methods both before and during
the project.
1407
-------
Appendix I. References
Charles, J.A., Burford, D. and Watts, K.S., "Field Studies of
the Effectiveness of Dynamic Consolidation," Tenth Internation-
al Conference on Soil Mechanics and Foundation Engineering,
Stockholm, 1981, Volume 3, pp. 617-622.
Leonards, G.A., Cutter, W.A. and Holtz, R.D., "Dynamic Compac-
ion of Granular Soils," Journal of the Geotechnical Engineering
Division, ASCE, Vol. 106, GT4, January, 1980, pp. 35-44.
Lukas, R.G., "Densification of Loose Deposits by Pounding,"
Journal of the Geotechnical Engineering Division, Vol. 106,
American Society of Civil Engineering, GT4, April, 1980, pp.
435-446.
Lukas, R.G., "Densification of a Decomposed Landfill Deposit,"
Eleventh International Conference on Soil Mechanics and
Foundation Engineering, San Francisco, CA, 1985, Vol. 3, pp.
1725-1728.
Lukas, R.G., "Dynamic Compaction for Highway Construction, Vol.
1, Design and Construction Guidelines," Federal Highway
Administration, Report No. FHWA/RD-86/133, July, 1986.
Mayne, P.W., Jones, J.S. and Dumas, J.C., "Ground Response to
Dynamic Compaction," Journal of the Geotechnical Engineering
Division, American Society of Civil Engineer, Vol. 110, No. 6,
pp. 757-774, June 1984.
Menard, L. and Broise, Y., "Theoretical and Practical Aspects of
Dynamic Compaction," Geotechnigue, March 1975.
Appendix II. Metric Conversion Factors
1 foot = 0.305 meters
1 foot2 = 0.093 meters2
1 U.S. ton =0.91 metric ton
1408
-------
U.S. EPA Region II Treatability Trailer for
Onsite Testing of Soils and Sludges
William Smith
Camp Dresser & McKee Inc.
Raritan Plaza I
Raritan Center
Edison, New Jersey 08818
(908) 225-7000
Keith Kollar
U.S. Environmental Protection Agency
26 Federal Plaza
New York, New York 10278
(212) 264-1576
INTRODUCTION
Over the history of the Superfund program, treatability studies
have generally been postponed until the remedial design phase,
following selection of the remedy. However, remedial planning
guidance and directives from the U.S. Environmental Protection
Agency (EPA), and recent administrative reviews of the Superfund
program, have emphasized the importance of conducting these
studies during the remedial planning process. The emphasis on
supporting remedy selection with treatment data will increase the
need for timely and cost-effective performance of treatability
studies.
In anticipation of this need, EPA Region II has developed a
treatability trailer for onsite testing of treatment technologies
for soils and sludges. The treatability trailer has been
designed to provide a generic working platform capable of
supporting a number of different bench-scale test programs with a
minimum of modifications. Provisions have been included to
support testing of treatment technologies for soil, sludges and
water, and to support screening-level chemical analyses of these
matrices. Although the optimum application of the trailer is
within a field operations center with separate facilities to
support field office, sampling activities and analytical
services, the trailer is capable of operation at remote locations
with minimal support. The first use of the trailer is scheduled
for August 1991 as part of a Superfund RI/FS.
Among the potential benefits to be realized from the use of the
treatability trailer are an increased availability of equipment
and facilities for onsite testing, increased experience with
these technologies at the contractor level, reduced need to ship
waste offsite or to obtain permits for testing, and avoidance of
hidden costs associated with offsite studies. Realization of
these benefits will make it easier to implement studies of
innovative and conventional treatment technologies, and will help
1409
-------
to promote performance of treatability studies in the remedial
planning phase.
CANDIDATE TREATMENT TECHNOLOGIES
The number of technologies available for treatment of soil and
water has increased over the history of the Superfund program.
Application of these technologies generally requires that testing
be performed to evaluate the feasibility of using a treatment
technology to attain remedial criteria or to develop design data
for implementation of a selected remedy.
As an initial design activity, existing data on treatability
studies that were planned or performed under the Superfund
program were reviewed to identify the technologies most suitable
for onsite testing.
The review focused on the application of bench-scale treatability
tests as the most appropriate studies to support the remedial
planning process. As expressed in the Guide for Conducting
Treatability Studies Under CERCLA (EPA 1989a), "bench-scale
testing can verify that the technology can meet the expected
cleanup goals and can provide information in support of remedy
selection." These tests usually employ standard laboratory
equipment or other simple test apparatuses, and are performed
over short periods of time using relatively small amounts of
material. Data quality objectives for bench-scale screening are
generally quantitative in nature and require fairly rigorous
quality assurance and quality control measures (QA/QC). Less
sophisticated methods with limited QA/QC are also used to direct
the studies or evaluate the applicability of treatments.
The bench-scale tests were also reviewed for how easily they
could be performed in the field. Tests that required specialized
equipment or posed unique hazards, such as some solvent
extraction processes or incineration, or tests of proprietary
technologies were not considered to be suitable for inclusion in
the field trailer. Identification of appropriate test equipment
was oriented toward supporting the tests which were most
frequently required during remedial planning or for which testing
services were not readily available in the marketplace.
Requirements for sample preparation and phase separation were
also considered.
Based on this review, a number of technologies were identified as
being potentially feasible for testing in the treatability
1410
-------
trailer. Selected technologies for soil and sludge treatment
were:
o Low-temperature thermal desorption
o Soil washing
o Solidification/stabilization
o Alkaline dechlorination (KPEG)
o Solvent extraction
The technologies selected for water treatment were:
o Air stripping
o Carbon adsorption
o Ion exchange
o Metals removal
Lists of recommended physical tests, chemical analyses and
bench-scale test equipment were produced for each technology that
was reviewed. These served as the basis for the recommended
design criteria and equipment inventory for the treatability
trailer (FPC 1990). Tables 1 through 3 present the lists
generated for bench-scale testing of stabilization/solidification
as examples of the information that was developed.
Other technologies that were not included in the review may also
be suitable for testing in a mobile field laboratory. The
requirements for field-testing any technology should be evaluated
on a project-specific basis.
TRAILER DESIGN AND OPERATION
The principal objective of the treatability trailer design was to
provide a facility that would support a broad range of
treatability study programs with a minimum of modifications.
This made it necessary to design the trailer as a generic working
platform.
Design criteria were developed to define the key features of a
functional laboratory environment. These were grouped into eight
areas, which are discussed herein:
o Trailer construction
o Transportation requirements
o Space utilization
o Laboratory furniture and appliances
o Utility requirements
o Heating, ventilation and air conditioning
o Laboratory safety requirements
o Analytical support requirements
The following sections present the design criteria for the
trailer and discuss the specific features that are provided to
1411
-------
support performance of the treatability studies. The elevation
and plan view sketches presented in figures 1 through 8
illustrate the manner in which the various features of the
trailer have been integrated to create a functional laboratory
environment.
Although the trailer is capable of operating with a minimum of
support, its optimum application will be as part of a field
operations center where its functions would be supported by a
separate facility for field office and sampling activities and a
close support analytical laboratory. Some of these additional
support activities could be accommodated within a larger
treatability trailer with little additional design effort, but at
a significantly greater cost. Others, particularly close support
analytical activities, can be performed most effectively in a
separate onsite facility.
Trailer Construction
The principle criteria for construction of the trailer addressed
its durability, maintainability and conformance to applicable
codes, standards and practices.
Exterior elevations of the treatability trailer are shown in
figure 1. The trailer body is of the standard semi-trailer, or
"box trailer", configuration. Another common trailer style is
the tow-trailer, which sits low to the ground, allowing for
easier entry and exit. However, the semi-trailer allows the
towing hitch to be placed above the rear wheel axles of the
towing vehicle, providing greater stability and leverage than can
be obtained when the hitch is placed behind the rear wheel axles.
The trailer hitch would be of the fifth-wheel configuration to
provide the necessary stability and versatility for transporting
a trailer of this size.
Construction of the trailer body conforms to standard industry
practices. This should provide sufficient durability for the
trailer to perform its expected services through the ten-year
life of the ARCS II contract, with a minimum of maintenance or
repair. The trailer also complies with the requirements and
standards of the U.S. Department of Transportation (DOT) and the
Interstate Commerce Commission and Society of Automotive
Engineers (ICC/SAE) for over-the-road vehicles:
o Reinforcement for loading in excess of 6500 pounds
o Rating labeled on the exterior of the trailer
o Wheel assembly sized to accommodate loading
o Furnished with break away braking system
o Furnished with emergency disconnect brakes
o Double axle for trailers over 25 feet in length
o Finished exterior with running lights at the rear and
sides
1412
-------
The trailer frame has been constructed with extra reinforcement
to accommodate the stresses of travel over unfinished roads and
off-road conditions. The structural members and walls are also
reinforced to support overhead cabinets and wall-mounted
apparatus.
The running gear consist of a heavy-duty double axle assembly
with air-glide suspension and shock absorbers. The braking
system is air-operated on both axles to provide 110 percent
braking of the trailer and its controls can be interconnected
with the tow vehicle. An emergency braking system is provided to
cause total braking of the trailer in case of on-the-road
breakaway.
The exterior is finished with riveted aluminum sheeting and
coated with polyacrylic paint. The interior sidewalls and
ceiling are continuous sheeting of nontextured panel with the
number of seams minimized to facilitate decontamination. All
corners, seams and penetrations are sealed to be leak free for
the life of the trailer.
The trailer is provided with two doors along the curb side of the
trailer. The rear door is of the standard 3-foot width. The
front door is extra-wide to allow loading and unloading of large
equipment. Aluminum stairs and platforms are provided as part of
the trailer's equipment. An all-weather canopy, supported by an
aluminum frame, has been provided for outdoor work such as sample
preparation. These items fold for storage in compartments below
the trailer.
Transportation Requirements
The treatability trailer has been designed to withstand stresses
associated with transportation over paved and non-paved roadways.
The trailer will be sturdy enough for occasional off-road
activities, however, frequent travel over rough terrains would
shorten its useful life.
A commercial trailer transporter will be contracted to tow the
trailer between sites. The semi-trailer configuration with
fifth-wheel towing hitch is commonly used for long-distance
hauling, and transport services are widely available. A
medium-duty tractor or larger vehicle will be used for towing.
As discussed above, the trailer design is in conformance with the
pertinent federal transportation regulations. Local limitations
relative to permissible height or weight may be exceeded along
some routes. No special permits will be required for over-the-
road travel, as the trailer width does not exceed 8 feet.
1413
-------
Space Utilization
The design of the treatability trailer must provide a safe,
controlled and open environment for the workers. The permanent
furniture and fixtures have been arranged to provide adequate
open area for operations and allow unimpeded movement of
personnel and equipment. Sufficient open area is provided to
allow large or floor-standing equipment to be mounted in the
trailer or removed as needed. The arrangement also allows the
work areas to be segregated into "clean" and "dirty" zones to
facilitate control of contamination.
Adequate bench space and storage space is provided to allow safe,
uncluttered performance of the tests and analyses, and maintain
the necessary equipment and supplies within the trailer. The
arrangement of the benches and cabinets uses the available space
as efficiently as is practical.
Cabinet and bench space arrangements for the laboratory interior
are illustrated in figures 2 and 3, Plan View and Interior
Elevations. This arrangement provides about 34 linear feet oE
bench space over about 256 cubic feet of cabinet, cupboard and
drawer space. About one-third of this space is occupied by
utilities and fixed equipment. Overhead storage provides about
72 cubic feet of storage space in large cabinets with sliding
panel fronts. There is about 100 square feet of open floor
space, 30 of which is available for free-standing laboratory
equipment.
Laboratory Furniture and Appliances
The principle criteria for selection of fixtures and furniture
for the treatability trailer were their durability and
maintenance requirements, and their suitability for use in a
generic laboratory environment. Standard items have been
selected to the maximum extent practical so that replacements can
be readily obtained. The materials and construction techniques
used are expected to last the duration of the ARCS program
(10 years). Assemblies and finishes are designed to facilitate
cleaning and minimize decontamination efforts.
A plan view and elevation of the laboratory interior have been
presented in figures 2 and 3.
Laboratory furniture. Materials of construction that were
considered for the base units and overhead cabinets include wood,
plastic laminate and enameled steel. Steel was chosen because of
its superior durability. Wood and plastic laminate base units
are estimated to cost less than steel by 8 and 20 percent,
respectively, but such units would be more likely to require
repair or replacement over the life of the trailer.
1414
-------
Stainless steel and epoxy were considered as materials for the
worktop surfaces. Epoxy surface with a plywood base was
identified as a suitable material because of its strength and
chemical resistance. Epoxy worktops are less expensive than
stainless steel and replacement parts are readily available in
standard sizes.
The worktops are provided with a lipped front and an integral
backsplash to control spills and facilitate cleaniong. The base
cabinets are installed about 8 inches away from the wall to allow
access to service lines through service shelving or the back
panels of the base units. The epoxy worktops are designed to be
tilted into place so that the backsplash fits under the service
shelving. Furniture modules can be removed to provide additional
floor space when needed.
Laboratory appliances. The fume hood selected for the
treatability trailer is large enough to accommodate test
equipment, such as furnaces or distillation apparatus, that may
be used during the treatability studies. It is equipped with an
integral exhaust blower, capable of drawing 1120 cubic-feet-per-
minute of air.
Other appliances that have been provided include a deionized
water system and a standard laboratory refrigerator. The water
purifier unit has four modular purifiers with a built-in pump and
a resistivity/temperature monitor. Purification modules include
a prefilter, an activated carbon cartridge, ion exchange
cartridges and an adsorbent resin cartridge for removal of trace
organic contaminants.
Other features. In many applications, the treatability trailer
would be located at a field operations center, for which
communications and security will have been arranged. For
independent operation, the trailer has been provided with
telephone jacks and the necessary wiring to allow connection of
telephone service. A standard telephone is included as part of
the trailer inventory. A mobile phone or two-way radio would be
needed in those situations where telephone service was
unavailable.
A security system has been installed for the exterior doors and
windows, to provide a 90 decibel audible exterior alarm upon
unauthorized entry. This system would be turned on or off with a
key from the exterior of the trailer. The trailer wheels have
been equipped with anti-theft bars and a protective cover has
been installed over the rear window.
Utility Requirements
The treatability trailer provides fully-developed utility systems
to support the execution of bench-scale studies in the field.
1415
-------
The general design criteria for these systems were developed to
enable the trailer to operate with a minimum of outside support.
These systems are capable of supporting an array of different
appliances and equipment without rewiring or replumbing. Utility
connections are designed for quick connection to minimize mobili-
zation and demobilization times. The installation of the utility
should meet the requirements of the local codes and regulations
in the areas that the trailer will be used.
Provisions have also been made to accomodate electrical and weiter
supply in remote areas/ and for proper segregation and handling
of wastes generated during the studies.
Electrical systems. The electrical systems for the trailer
provide a high voltage power supply for appliances and test
equipment and a battery-powered low voltage system for the
security system and exterior lighting.
The electrical plan for the high voltage system is shown in
figure 4. Power supply and electrical distribution are provided
through a 200 amp service panel mounted inside the trailer.
Electrical power at 220 VAC and 120 VAC electrical power outlets
and services to appliances are provided throughout the trailer.
Outlets for 220 VAC power are also provided at the exterior of
the trailer for operations performed out-of-doors. Exterior
outlets and outlets in the equipment area are provided with
tight-fitting caps for protection from water and dirt. Voltage
surge dampening is provided on selected 120 VAC circuits to
protect analytical and data management equipment.
The low voltage system provides power to operate the security
system and security lights. Cable and connectors are provided
for connection to the tow vehicle's power system. The standby
power system includes a 12-volt deep-cycle battery and automatic
battery charger.
The maximum high voltage requirement for the trailer is estimated
to be 32 kilowatts (KW), including power surges at startup. This
power can readily be supplied through a high voltage line and
transformer.
The potential need for a generator should be assessed on a
project-by-project basis. For many projects, electrical power
will be available at the site, or the electrical demands may be
low enough to allow use of a smaller generator. At remote
locations, it may be necessary to provide a fuel-powered
generator to power the trailer and equipment. The costs of
leasing a generator for occasional use under these conditions
should be much less expensive than purchasing a large system.
Lighting. Interior lighting is provided by three ceiling-mounted
fluorescent tube fixtures with wrap-around lenses. An additional
1416
-------
single-tube fluorescent light fixture is incuded as part of the
fume hood equipment. Emergency lights with self-contained power
supplies have also been installed inside the trailer.
Exterior lighting is provided by low-pressure sodium-vapor
security lights, mounted on the sides of the trailer. A
low-power courtesy light would be provided over each door.
The trailer is equipped with highway running lights, tail lights
and brake lights as required under DOT and ICC/SAE rules and
guidelines.
Area lighting is not included as part of the trailer equipment.
If such lighting was needed, equipment could be leased for the
specific project, and powered through the trailer's electrical
distribution system.
Pressurized water system. The pressurized water system plan is
shown in figure 6. This system includes an onboard storage tank,
pressure pump, water heater, distribution system, hot and cold
service fixtures and provisions for connection to an external
domestic water system. Freeze protection has been provided near
hatches and through the underbelly of the trailer.
Pressure control is provided by small pressure tanks of the type
used with household well systems. This system has been designed
to operate over the pressure range of 20 to 40 psig and prevent
excessive pump cycling during episodes of high water usage.
Two pressure tanks, stored in a cabinet inside the trailer,
provide pressure to the distribution system. Two additional
tanks provide pressure to the safety shower. A check valve
installed before the shower tanks prevents loss of pressure from
the safety shower to the distribution system.
The pump has been sized to deliver a minimum of 10 gpm to the
safety shower at 30 psig. This will allow the system to operate
near the high end of the pressure range under heavy water demand.
Pressurized water can also be pumped directly into the
distribution system from outside, bypassing the trailer's pump
and pressure tanks.
Compressed air system. The treatability trailer has been
provided with an onboard air compressor and distribution system.
Four deck-mounted air nozzles have been provided: one near each
sink, and two at the fume hood. Media for preparing the air to
meet experimental needs (e.g., filters or desiccant tubes) will
be provided by the specific studies.
Waste drain system. Schematic drawings of the waste drain system
are presented in figure 7. Drains have been provided from the
sinks, hood and safety shower, and from two locations in the
1417
-------
trailer floor. The drain outlets are about 4 feet above ground
level, which will allow drainage by gravity flow to external
containers. In situations where gravity flow is not sufficient,
the waste may drain to a sump, and then be pumped to the disposal
point. The drains at each end of trailer have been installed as
separate systems, allowing segregation of the wastewater streams.
Collected wastewater will be removed for offsite disposal,
whenever required during the study.
Heating, Ventilation and Air Conditioning (HVAC)
The HVAC system for the trailer will regulate the temperature and
humidity within the trailer against the full range of external
atmospheric conditions that may be encountered during field
operations at sites in Region II. It will maintain adequate
indoor ventilation rates, as required by the Occupational Health
and Safety Administration (OSHA), and be capable of a rapid
turnover of the room air when necessary.
A wall-mounted air conditioning and ventilation unit has been
installed at the front of the trailer. Heating is provided by a
heat pump, with an electric heating element as a backup system.
The system has been sized to maintain an interior temperature of
75°F and about 60 percent relative humidity against air exterior
temperatures of minus 10 to plus 110 F and humidities up to 100
percent. Air temperature is thermostatically controlled.
The ventilation system can maintain up to four air changes per
hour through air diffusers distributed throughout the trailer
interior. More rapid air changes can be provided by operating
the fume hood with its below-counter air intakes closed.
The components of the HVAC system are indicated on figure 8.
Laboratory Safety Requirements
The safety features incorporated in the laboratory design meet
generally accepted safe laboratory practices and the applicable
requirements of OSHA (for example: 29 CFR Part 1910 Subpart H).
Specific features include exhaust vents to provide adequate
indoor ventilation, two entrances and exits, and provisions to
segregate corrosives and flammables in separate cabinets. The
layout of furniture and fixtures also facilitate segregating the
work areas into "clean" and "dirty" zones.
A fume hood has been provided for operations that pose a fume or
splattering hazard. The hood fan is capable of maintaining a
flow of up to 100 linear feet per minute. A charcoal filter pack
has been installed in the exhaust line to capture particulates
and organic contaminants.
1418
-------
Specific safety equipment that has been provided includes a
full-deluge safety shower, a portable face-and-eyewash station,
fire extinguishers and fire blankets, smoke detectors, and
emergency lights. Signs will be placed in the interior of the
trailer as warning or precautionary devices. A laboratory
first-aid kit is also provided, and test equipment is provided
for monitoring performance of the ventilation system. Other
safety equipment and information, such as material safety data
sheets (MSDS), personnel protective equipment or air monitoring
equipment, will be supplied by the individual projects.
The acid and solvent storage cabinets incorporate the required
safety features. Solvent storage cabinets are specially designed
to prevent spread of flames within the cabinets, and to shield
their contents from exterior fires. The cabinet floor is
recessed below the door sill to contain spills. Acid storage
cabinets are made with acid-resistant material and have vents in
the doors. Both types of storage cabinets have been modified to
vent directly into the fume hood. Safety containers and labels
for temporary storage of spent solvents and chemical solutions,
will also provided by the specific projects.
ANALYTICAL SUPPORT REQUIREMENTS
There are several advantages to supporting an onsite treatability
study with onsite analytical capabilities. These include shorter
turnaround times, increased flexibility in the experimental
program, and better characterization of unstable chemical species
generated by the treatment process. However, it would be
necessary to segregate the more sophisticated analytical
equipment in a separate trailer to minimize cross-contamination
and prevent power fluctuations and vibrations that would be
associated with operation of the treatability test equipment.
The costs of maintaining onsite analytical capabilities are much
greater than the costs of performing the analyses off site, and
maintenance of data quality is more difficult.
The relative benefits and disadvantages of performing chemical
analyses in a trailer on site or at a fixed facility off site
must be evaluated during the design of each treatability study.
The quality of treatability testing data should be appropriate
to the potential impacts of the decisions that will be based on
those data. The EPA has published guidance that defines the
framework and processes for developing appropriate data quality
objectives (DQOs) (EPA 1987).
The treatability trailer provides working areas and basic
laboratory equipment to perform screening tests to direct the
progress of the study or determine whether a treatment process is
potentially applicable (DQO Levels I or II). Selected samples
can be analyzed at a fixed facility to verify the screening data,
or to attain the more stringent data quality objectives required
1419
-------
to evaluate treatment performance or develop design data (DQO
Levels III or IV).
The pre-design review of bench-scale studies identified
screening-level chemical analyses that could provide reliable
data at the lower DQO levels in the treatability trailer
environment. These tests and the equipment required to perform
them are listed in tables 4 and 5, respectively.
A summary list of the test and analytical equipment recommended
for eventual inclusion in the trailer inventory is presented in
table 6. This list was compiled from the technology-specific
lists developed from the pre-design review.
CONCLUSION
The treatability trailer will be deployed for its first
assignment in August 1991, when it will serve as a field
laboratory for treatment of contaminated soils for a Superfund
RI/FS. Test plans for these studies are presently under
development.
In this application, the trailer will be operated as part of a
field operations center where commercial power and city water
will be provided. Its operations will be supported by a separate
field office and a close support analytical laboratory. While
this situation will not test the trailer's full capabilities, it
will provide a demonstration of the applicability of an onsite
facility for treatability testing.
The particular benefits expected from the use of the treatability
trailer for the upcoming study are an improved flexibility in
executing the test program, a reduced need to ship waste offsite
or to obtain permits for testing, and an increase in experience
and understanding of the the technologies that will carry through
the later phases of the site remediation and that can be applied
to remedial planning activities at other sites.
1420
-------
REFERENCES
American Public Health Association (APHA), American Water Works
Association (AWWA), Water Pollution Control Federation (WPCF),
1985. Standard Methods for the Examination of Water and
Wastewater. 16th Edition. Washington: American Public
Health Association Publication Office.
Bates, Edward and Paul DePercin. 1989. Field Treatability
Studies and Processes. Draft discussion paper from the RREL,
EPA-Cincinnati 16 November 1989.
CDM Federal Programs Corporation (FPC). 1990. Draft Preliminary
Design Report for the Treatability Trailer Design and
Construction Technical Support. CDM Federal Programs
Corporation, New York, New York. Report prepared for U.S.
EPA, Region II, 26 Federal Plaza, New York, New York.
Office of the Federal Register. 1989. Code of Federal
Regulations. Title 29-Labor. Parts 1900 to 1910. Washington,
D.C.: U.S. Government Printing Office. (see esp. Parts
1910.24, 1910.106, 1910.120, 1910.1200, 1910.1450)
U.S. Environmental Protection Agency. 1983. Methods for
Chemical Analysis of Water and Wastes. Office of Research and
Development. EPA-600 4-79-020, Rev. March 1983.
U.S. Environmental Protection Agency. 1986a. Contract
Laboratory Program Statement of Work for Inorganic Analysis.
SOW No. 786. Rev. October 1986.
U.S. Environmental Protection Agency. 1986b. Contract
Laboratory Program Statement of Work for Volatile Organic
Analysis. SOW No. 10/86.
U.S. Environmental Protection Agency. 1986c. Contract
Laboratory Program Statement of Work for Base/Neutrals and
Acid Extractable Organics Analysis, SOW No. 2/88.
U.S. Environmental Protection Agency. 1987. Data Quality
Objectives for Remedial Response Activities. Development
Process. Office of Emergency and Remedial Response.
EPA/5 4 0/G-87/003.
U.S. Environmental Protection Agency. 1988a. Guidance for
Conducting Remedial Investigations and Feasibility Studies
Under CERCLA. Office of Emergency and Remedial Response.
EPA/540/G-89/004. October 1988.
U.S. Environmental Protection Agency. 1988b. Methods for the
Determination of Organic Compounds in Drinking Water. EPA-600
4-88-039. December 1988.
1421
-------
U.S. Environmental Protection Agency. 1989a. Guide for
Conducting Treatability Studies Under CERCLA - Interim Final.
Office of Research and Development and Office of Emergency and
Remedial Response.
U.S. Environmental Protection Agency. 1989b. A Management
Review of the Superfund Program. U.S. Government Printing
Office: 1989-623-682/10263.
U.S. Environmental Protection Agency. 1989c. Stabilization/-
Solidification of CERCLA and RCRA Wastes, Physical Tests,
Chemical Testing Procedures, Technology Screening and Field
Activities. Center for Environmental Research Information and
Risk Reduction Engineering Laboratory and Office of Research
and Development,. EPA/625/6-9/022, May 1989.
1422
-------
o-O
«°=:^> oo
S§8£ tz~
^rzo;
o!8£
1423
-------
t
t
/^
v:
£
(a>
$
©
(
(
m
0
?\
^
D
©
-,
t/
)
— 1
00) I
---
^-/
a
N-X
vy
©
©
-.
©
©
/Fh
O
©
Ov
P)
VS/
r*
U> V) ,x
oc cc V
1 1 !
-» •». ^
8 S? £
.x .x o
GO CD O
M f i, or
I IS
a a 5
©-1
" " /tjO^ T* CC ^
^ § n-'
•L* £Q «\i
' ^^ ^
ff ^* O co
HI x O
r» C3 ** Ml *™"
U-^ • = c/5
li 8 s|
C3" O XO
cc£ ^ gj
S^ o o^
© (D©
T=^~ 1 1
V 3 S
a & ^
J S UJ
Q z U.
0 E OH
II 1 I
§1 i 22 *
CC
£T
H
H
_l
fQ
g §
| 5 §
.21 2
Q. 0
O
LU
CC
UJ
C/3
1424
-------
Figure 3
OR ELEVATIONS
TREATABILITY TRAILER
ON
cc -s
LU
CO
13
1425
-------
Ofc
UJ
O
UJ
• u i
UJ i
> a- o z
§ K Z 0.
°- 5 z =1
§2§
3 111 UJ
CO * Z
diSg
2 CC o
CC
CD
K
1
O
o
§IL
^ w << z
_i CC
ai a.
z <
O
u
i=
uj
ceo r~ "
Q.H- T l-
T^ CVJ CO •* «
J
QC
UJ
CC
1-
Z t
a. ^
r^ LJJ
^T ^~
O K
o o
y 5
UJ j£
UJ
1426
-------
Q
^^
UJ
O
LU
_J
_l
g
i-
o
UJ
_J
LU
Z
O
t
cc
o
CO
UJ
a
5
o
cc
o
Z
0
1—
0-
cr
0
CO
LU
Q
b
D
a:
o
LU 111 111 111 ID
CC CC CC CC CC
r\ Q Q^ j^ Q^
CO CO CO CO CO
O O O O O
CM CM CM CM CM
O O O O O
CM CM CM CM CM
»— T— ^ ^ T-
T- CM co ^ m
CM CM CM CM CM
111 UJ UJ LU 111
K- ,
2 LU
c I o Q
S z£ S Q
DODQQOODT £5,<^ LUX
LULULUUJLULULUUJi N ^ r? CCin
^* H~ ^ 1" H* H* H* H" CJ yj t-v •»•) n •••
<<<<<
Z b
UJ -J
O CD
LU g
in — ' <
§) **• *~
"• E 5
s §
LU =
CL
LU
CO
Z)
1427
-------
CO
3
2
:5
CL
O
2
E5
cr
LU
cr
i—
>
CD
LU
cr
i-
0
o
LU
CC
UJ
CO
1428
-------
0
ID
CO
<
tr
Q
cc
LJJ
-J
<
DC
"• Z
LLJ Q
O O
rf UJ
01
CO
in
i
-------
UJ <
m cc
o1-
1~£
O °
Z CC
PQ
OE
3 01
CC
Q
CC
UJ
X
UJ
LL
o
UJ
V)
<
CD
UJ uj
x£
%*
O
-------
TABLE 1
PHYSICAL TEST METHODS - STABILIZATION/SOLIDIFICATION
Parameter
Test method
Applicability
Particle Size
Analysis
ASTM D422-63
Moisture content ASTM D2216-80
Permeability
(falling head and
and constant head)
USEPA method 9100
(SW-846)
Unconfined com- ASTM D2166-85
pressive strength
of cohesive soils
Unconfined com- ASTM D1633-84
pressive strength
of concrete specimens
Flexural strength ASTM D1635-87
Cone index
Reference:
ASTM D3441-79
Particle size distribution
influences the effective-
ness of treatment
To determine water content
To measure the rate which
water will pass through a
soil-like material.
To evaluate how cohesive
soil-like materials behave
under mechanical stress
To evaluate how cement-
like materials behave
under mechanical stress
To evaluate the treated
material's ability to
withstand loads over a
large area
To evaluate the treated
material's stability and
bearing capacity
Stabilization/Solidification of CERCLA and RCRA
Wastes, EPA.625/6-89/022
1431
-------
TABLE 2
CHEMICAL TEST METHODS - STABILIZATION/SOLIDIFICATION
Parameter
Test method
Applicability
EPA method SW-9045
Major Oxides
ASTM C114
Total Organic
Carbon
Oil and Grease
Combustion method
EPA method 413.2
Elemental Analysis EPA method SW-846
Target Compound
List Parameters
Alkalinity
EPA methods 624,
625 or current CLP
methodology
Titrometry
Leachability of hazardous
constituents (e.g.,
metals) may be governed by
the pH of the solid
Mineralogy of the
stabilized/solidified
waste may aid in
interpretation of leach
test results
Used to approximate the
nonpurgeable organic
carbon in wastes and
treated solids
Presence of oil and grease
in the untreated wastes
will influence the
effectiveness of the
treatment
Used to determine the
fraction of metals leached
to the total metals
content of the untreated
and stabilized/solidified
wastes
To
of
quantify
concern.
contaminants
Alkalinity changes in
leachates may be used to
evaluate changes in
stabilized/solidified
waste form
Stabilization/Solidification of CERCLA and RCRA
Wastes, EPA/625/6-89/022
Reference:
1432
-------
TABLE 3
EQUIPMENT LIST FOR STABILIZATION/SOLIDIFICATION
Item Quantity
1) Single-use cardboard cylinder molds 48
2) Single-use cylinder lids 250
3) Stripping tool 1
4) Beam molds 30
5) Cube molds 30
6) Laboratory mixer, 12-qt. capacity 1
7) Jar mill - 2 tier 1
8) Compaction vibrator 1
9) Cylinder carrier 1
10) Sample cart 1
11) Autogenous concrete curing container 1
12) Wet sieve tester 1
13) Round test sieves, U.S., standard sizes
2-inch, 1-inch, 3/8-inch-inch, No.4, No.10, No.20 6
14) Scalping apparatus, with sample trays 1
15) Scalping screens, U.S. standard sizes
2-inch, 1-inch, 3/8-inch-inch, No.4, No.10 5
16) Mechanical soil compactor (with molds) 1
17) Unconfined compression tester (with molds) 1
1433
-------
TABLE 4
RECOMMENDED ANALYSES FOR CONTAMINANT SCREENING
Analysis
Volatile Organics
Semi-Volatile Organics
Selected Metals
PH
Turbidity
Total Dissolved Solids (TDS)
Total Suspended Solids (TSS)
Alkalinity
Biological Oxygen Demand (BOD)
Chemical Oxygen Demand (COD)
Methlyene Blue Active Substances
(NBAS)
Hardness
Specific Ion Analyses
Method
Headspace analysis by
gas chromatograph or
extraction followed by
gas chromatograph anal-
ysis
Extraction followed by
gas chromatograph anal-
ysis
Colorimetric
pH electrode
Light Scattering
Gravimetric
Gravimetric
Titration
5-day incubation
Digestion
Colorimetric
Titration
Specific ion electrodes
1434
-------
TABLE 5
EQUIPMENT FOR SCREENING-LEVEL ANALYSES
Portable Gas Chromatograph
COD Reactor
Spectrophotometer
Turbidity Meter
Zero Headspace Extractor
Moisture Determination Balance
Soil pH Kit
Benchtop pH/mV Meter with:
Chloride Electrode
Cyanide Electrode
Oxygen Electrode
Redox Electrode
BOD Incubator
1435
-------
TABLE 6
RECOMMENDED EQUIPMENT INVENTORY
General Laboratory Equipment
Item Quantity
Gravity Convection Oven1
Microscope with 35 mm Camera 1
Analytical Balance 1
Pressure Filter Apparatus 1
Heavy Duty Balance 1
Dual Memory Electronic Timer 1
Peristaltic Pump Kits 2
Submersible Pump with Level-Activated Switch 2
Heavy Duty Mixer 1
Flow Meters 4
Handheld Thermocouple with Type K Probe 1
Additional Type K probes 2
Temperature Recorder 1
Refrigerated Circulation Bath 1
Rotating Shaker Unit with Shaker Carrier 1
Vacuum Filter Pump 2
Tool Kit 1
Conductivity Meter with Probes 1
Stop Watch/Timer 2
Dessicator 2
Hot Plate 2
Magnetic Stir Plate 2
1436
-------
TABLE 6
RECOMMENDED EQUIPMENT INVENTORY
(continued)
General Laboratory Equipment
Item
Quantity
Compressed Gas Cylinder Holder
Pressure Regulators
Nitrogen
Air
Jar Mill
55-Gallon Polyethylene Containers
Laboratory Racks
Micro Dispensers
1
1
1
2
3
2
Miscellaneous Laboratory Supplies
Beakers
Large glass containers
Test tubes
Graduated cylinders
Volumetric flasks
Steel bowls
Buchner funnels
Filter paper
Membrane filter paper
Measuring spoons
Tweezers
Tubing
Stirring rods
Pipet micro tips
Magnetic stirrers
Scrub brushes
Erlenmeyer flasks
Watch glasses
Pipets
Thermometers
Steel pitcher
Filter flasks
Microfiltration filter
holders
Spatulas
Tongs
Bungi cord
pH paper
Metal weighing dishes
Squirt bottles
1437
-------
TABLE 6
RECOMMENDED EQUIPMENT INVENTORY
(continued)
Analytical Equipment
Item Quantity
Portable Gas Chromatograph with Adapter 1
COD Reactor 1
Spectrophotometer 1
Turbidity Meter 1
Zero Headspace Extractor 1
Moisture Determination Balance 1
Soil pH Kit 1
Benchtop pH Meter with: 1
Chloride Electrode 1
Cyanide Electrode 1
Oxygen Electrode 1
Redox Electrode 1
BOD Incubator 1
1438
-------
TABLE 6
RECOMMENDED EQUIPMENT INVENTORY
(continued)
Equipment for Soil Testing and Sample Preparation
Item Quantity
Wet Sieve Tester 1
Sieve Mesh Protector
Rubber Gaskets & Filter Paper
Round U.S. Standard Size Test Sieves
Sizes: 3-inch, 2-inch, 1 1/2-inch,
1-inch, 3/4-inch, 3/8-inch, No. 4
8-inch diameter 7
12-inch diameter 7
Sizes: No. 10, No. 20, No. 40 No. 60
8-inch diameter 4
12-inch diameter 4
No. 140
8-inch diameter 1
12-inch diameter 1
No. 200
8-inch diameter 1
12-inch diameter 1
Hydrometers 2
Portable Concrete Mixer 1
Scalping Apparatus, with Sample Trays 1
Scalping Screens, U.S. Standard Sizes
2-inch, 1-inch, 3/8-inch, No.4, No.10 5
Unconfined Compression Apparatus 1
Mechanical Soil Compactor 1
Compactor Molds:
4-inch ID Split 1
6-inch ID Split 1
1439
-------
TABLE 6
RECOMMENDED EQUIPMENT INVENTORY
(continued)
Filtration Apparatus
Item Quantity
Centrifuge 1
Filter Leaf Apparatus 1
Plate and Frame Filter press 1
Air-Driven Diaphragm Pump 1
Additional Filters and Screens
Test Equipment For Soils Treatment Technologies
Item Quantity
Muffle Furnace 1
Tube Furnace 1
Combustion Tubes 6
Distillation Glassware Kits 6
Single Use Cardboard Cylinder Molds with Lids 48
Stripping Tools 1
Beam Molds 30
Cube Molds 30
Laboratory Mixer 1
Compaction Vibrator 1
Cylinder Carrier 1
Autogenous Concrete Curing Container 1
1440
-------
TABLE 6
RECOMMENDED EQUIPMENT INVENTORY
(continued)
Test Equipment For Water Treatment Technologies
Item Quantity
Bubble column - 6-inch diameter,
9-ft height; fine bubble diffuser,
flange in center of column, bottom
and middle sample port, drain (custom-made) 1
Column test assembly - 4-inch diameter,
6-ft height, connected in series, connector
piping, valves, sample ports, pressure
gauges, flow meter 1
Variable-speed centrifugal pumps 2
Jar test apparatus - six paddle stirrer 1
DAF bench-scale test kit - includes 1
pressurized tank, flotation receiver, air
release valve, and other fittings as needed
1441
-------
Summary of Issues Affecting
Remedial/Removal Incineration Projects
(Author(s) and Address(ea) at and of papar)
INTRODUCTION
Incineration is a very popular method of remediating superfund sites.
This is because it is a proven technology that is capable of decontaminating a
wide variety of waste. The residues from incineration can often be disposed of
without further treatment whereas residues from many other treatment technologies
require incineration prior to final disposal.
As with other remedies, implementing incineration is not always
straightforward. Identifying Applicable or Relevant and Appropriate Requirements
(ARARs) and complying with them is difficult. Incineration is also a costly
remediation method. Further, there is often opposition to the use of
incineration because of the belief that incineration emissions are harmful to
the environment and the health of the surrounding community. Because of the cost
and controversy surrounding incineration, a decision to use it at a site is
usually subject to challenge.
To assist the Remedial Project Manager (RPM) and On-Scene Coordinator (OSC)
in responding to these challenges and in directing the progress of remedial and
removal incineration projects, the Engineering Forum and EPA's Risk Reduction
Engineering Laboratory (RREL), have prepared a summary report entitled Issues
Affecting the Applicability and Success of Remedial/Removal Incineration
Projects.
The purpose of this summary is not to provide an encyclopedic account of
all relevant incineration information in one volume. That would be difficult,
if not impossible, and would soon become obsolete as the state-of-the-art
advances. Rather, this summary is intended to alert the RPM/OSC to issues
affecting the successful implementation of incineration projects, and to alert
them to both written and human resources that can help to address these issue',.
The remainder of this paper summarizes the content and key points of the summary
report.
BACKGROUND
Incineration has been chosen as the remedial method of choice in 32% of
the Records of Decision through FY89.(1) Most of these sites contain soil
contaminated with both organics and metals. Incineration has been used for a
number of years to treat a variety of waste including contaminated soils. EPA
sponsored tests over the last ten years indicate that properly operated
incinerators can successfully decontaminate waste without producing highly
contaminated residual streams. (2) A large body of knowledge exists regarding
how to successfully design and operate an incinerator.
To successfully implement an incineration remedy, it is important to access
this body of knowledge. To help provide this access, the Engineering Forum and
EPA's RREL developed a Summary for the RPM/OSC concerning issues affecting tie
successful implementation of remedial/removal incineration projects. This
document provides a summary of incineration hardware, design and maintenance
practice, ARARs and other compliance issues, vendors, and lists of state,
regional, headquarters and other technical experts capable of providing
assistance in a number of relevant areas. This paper summarizes the information
1442
-------
contained in the summary report.
TYPICAL INCINERATION CONFIGURATION/OPERATION
A typical incineration system includes not just the combustion device, but
also the processes necessary to deliver feed, fuel and air to the incinerator,
remove ash from the kiln and remove residual hydrocarbons, particulate and acid
gasses from the exhaust. Multiple chemical processes are used in an incineration
facility. The operation of each of these systems needs to be integrated and
controlled so that the entire system runs most efficiently. For this to happen,
the entire incineration system must be controlled by a series of process
controllers, thus making the entire system sophisticated and expensive.
The figure on the next page is a block diagram of a typical incineration
facility. Waste is first processed to remove any large-scale debris such as tree
limbs and animal carcasses and to blend it with wastes having different heating
values, if appropriate. Feed systems introduce the blended waste into the
incinerator and usually consist of combinations of conveyors, weigh hoppers, ram
feeders and nozzles (for liquid waste).
The typical incinerator used at Superfund sites is an 8 to 100 Million
BTU/hr rotary kiln operating at 75-200% excess air.(3) Since the waste treated
at Superfund sites typically has little heating value (<200 BTU/lb), auxiliary
fuel is used to provide the heat needed to volatilize and incinerate the organic
contaminants of the waste. Typically, incinerators operate at a gas temperature
of at least 2000 °F. If this temperature is uniformly maintained throughout the
rotary kiln and afterburner, emissions of unburned hydrocarbons should be
minimized.
Exhaust gas from the incinerator is treated using a similar sized
afterburner to complete the combustion of the organic contaminants volatilized
from the waste. A venturi scrubber is commonly used to control particulate
emissions. Caustic scrubbing in a packed tower is used to control the emissions
of acid gasses. Slowdown from these scrubber operations and ash from the
incinerator are two waste streams generated by incineration which must be
disposed of as hazardous waste and may require additional treatment.
Tables 1 and 2 summarize design and operating characteristics of hazardous waste
incinerators.(3)
Regulations call for the incineration process to be monitored closely and
for feed to the incinerator to be automatically shut off when conditions
indicative of a process upset are observed. Much of the process monitoring
occurs at the exhaust stack since the composition of the exhaust gasses is
indicative of combustion conditions in the incinerator. In addition, the levels
of some of the exhaust gas constituents are limited by some of the ARARs. Table
3 lists some of the continuous emission monitors routinely used and the expected
range of concentrations of exhaust gas constituents.(3)
Some of the indicators typically used to trigger a cessation of feed
include carbon monoxide in the exhaust gas, low temperature in either the rotary
kiln or afterburner, low burner pressure, flame loss and many others. A
comprehensive list of parameters which should trigger an automatic cessation of
feed is listed in Table 4.(3)
1443
-------
Incineration System Concept Flow Diagram
Auxiliary
Fuel
Combustion
Air
Waste
Processing
Waste
Feeding
Combustion
Unit
Ash
Removal
Exhaust to
Atmosphere
Acid Gas
Control
t
Particulate
Removal
Gas
Conditioning
Residue
Treatment
Wastewater To disposal
Source: U.S. Environmental Protection Agency I988b.
1444
-------
TABLE 1. DESIGN AND OPERATING CHARACTERISTICS OF A TYPICAL
INCINERATION SYSTEM
Parameter
Typical values
Rotary kiln
Operating temperature, *F
Ashing kiln
Slagging kiln
Types of waste
Ashing kiln
Slagging kiln
Solids residence time, min
Ashing kiln
Slagging kiln
Gas residence time, s
Gas velocity through kiln, ft/s
Heat release levels, Btu/ft3 per h
Small kiln, million Btu/h
Large kiln, million Btu/h
Kiln loading, X kiln volume
Ashing kiln
Slagging kiln
Kiln operating pressure, in.HjO
Excess air, X
Liquid injection unit
Operating temperature, *F
Residence time, s
Excess air, X
Waste heating value, BTU/lb
1200 to 1800
2200 to 2600
•Low Btu waste (e.g., contami-
nated soils) < 5000 Btu/lb
•High Btu waste >5000 Btu/lb
•High Btu waste >5000 Btu/lb
•Moderate moisture and halogen content
•Both drums and drummed wastes
30 to 60
60 to 100
1 to 2
15 to 20
25,000 to 40.000
8 to 35
35 to 100
7.5 to 15
4 to 6
-0.5 to -2.0
75 to 200
1800 to 3100
Milliseconds to 2.5
10 to 60
< 4500
Secondary coabustor (afterburner)
Residence time, s
Operating temperature, *F
TSCA wastes
RCRA wastes
Excess air, X
2200 typical
>2250
1600 to 2800
10 to 60
Sources:
Tillman, Rossi, and Vick 1990; Schaefer and Albert 1989.
1445
-------
TABLE 2. TYPICAL DESIGN PARAMETERS FOR AIR POLLUTION
CONTROL EQUIPMENT ON HAZARDOUS HASTE INCINERATORS
Air pollution control equipment
Typical design parameters
Participate
Electrostatic precipitators
Fabric filters
Venturi scrubbers
Acid gases
Packed towers
Spray dryers
SCA « 400-500 ft /1000 acfm
Gas velocity = 0.2 ft/s
Pulse jet A/C » 3-4:1
Reverse air A/C = 1.5-2:1
AP « 40-70 in. W.C.
L/G = 8-15 gal/1000 acfm
Superficial velocity « 6-10 ft/s
Packing depth = 6-10 ft
L/G * 20-40 gal/100 acfm
Caustic scrubbing medium,
maintaining pH = 6.5
Stoichiometric ratio « 1.05
Lou temperature:
Retention time 15-20 s
Outlet temperature 250-450*F
Stoichiometric ratio (lime)
= 2-4
SCA > specific collection area
A/C * air-to-cloth ratio in units of ft/min
L/G * liquid-to-gas ratio
Source: Buonicore 1990.
1446
-------
TABLE 3. SUMMARY OF CONTINUOUS EMISSION MONITORS
Pollutant
K
CO
NO,
so:;
Organic
compounds
(THC)
Monitor type
Paramagnetic
NDIRC
NDIR
Chemi luminescent
Flame photometry
FID
Expected
concentration
range
3-UX
2-14X
0-100 ppm
0-4000 ppm
0-4000 ppm
0-20 ppm
Available
range
0-2SX
0-21X
0-5000 ppm
0-10000 ppm
0-5000 ppm
0-1000 ppm
Typical
value
8%
8X
40 ppm
200 ppm
Varies by waste
<20 ppm
Source: Oppelt 1987.
For available instruments only. Higher ranges are possible through dilution.
Nondispersion infrared.
1447
-------
TABLE *. TYPICAL AUTOMATIC UASTE FEED SHUT OFF (AWFSO) PARAMETERS"
Purpose of AWFSO
Parameter (example value)
High CO in stack (100 ppm)*
Low chamber temperature* (KOO'F for rotary
kiln. 1700' F for SCO
High combustion gas flow (Varies by size)
Low pH of scrubber water (4) (e.g. not less
than 6.5)
Low scrubber water flow (Varies by size)
Low scrubber pressure drop (20 inches U.G. for
venturi)
High scrubber temperature (220* F)
Low sump levels (variable)
High chamber pressure (positive)
High chamber temperature (2000*F for rotary
kiln, 2600*F for SCO
Excessive fan vibration
Low burner air pressure (1 psig)
Low burner fuel pressure (3.0 psig for natural
gas)
Burner flame loss
Low oxygen in stack (3 percent)*
Loss of atomizing media
High stack SO,*
High waste feed flow
High Opacity >5X
Excess Worker
emissions safety
X
X
X
X
X
X
X X
X X
X X
X
X
X
X
X
X
X
X
Equipment
protection
X
X
X
X
X
X
X
* Rolling averages of these parameters can sometimes be used. (Leonard, Paul comments 10/23/90)
Source: Oppelt 1987.
1448
-------
ARARs
The requirement to achieve 99.99% Destruction and Removal Efficiency (ORE)
is the performance standard which is most often associated with incineration.
Although this is an important regulation, it is certainly not the only one which
must be complied with. Table 5 is a summary of typical incineration ARARs.(3)
This list is not exhaustive. Other ARARs may apply depending upon the process
being used and the location of the site. A process that has a water discharge
stream, for example, will have to comply with the Clean Water Act and other ARARs
pertaining to water discharge. A process which exhausts a gas stream heavily
laden with NOX may have a very difficult time complying with air ARARs in urban
areas in Southern California, for example, but may have no trouble at all in more
rural areas of the country.
Compliance with the Clean Air Act and the Toxic Substances Control Act
typically require permits. On-site remedial/removal actions undertaken at
Superfund sites do not require permits. However, all such activities must comply
with the substantive requirements of those permits. Determining what those
requirements are necessitates close coordination with the appropriate regional,
state and local authorities.
In addition to Federal laws, State and Local laws may also constitute ARARs
and must be complied with. The CERCLA Compliance With Other Laws Manual (EPA
540/G-89-009) provides a summary of other ARARs which may apply to incineration
projects.
FACTORS AFFECTING THE PERFORMANCE OF INCINERATORS
When properly designed, incineration should be able to decontaminate waste
to appropriate levels while complying with all ARARs. Since it is impossible
to determine with certainty whether ARARs will be met under all possible sets
of circumstances, it is necessary to use certain guidelines pertaining to
characteristics of the system which most affect the performance of incineration
systems.
Of the incineration parameters which most affect performance, compatibility
of the feed with the feed system is probably the most important. The feed system
must be reliable and capable of continuous operation even when the feed varies
widely in size, density, moisture content, and other properties. The feed system
must also be capable of reducing the size of the incoming feed if necessary and
it must be reliable. A feed system which constantly breaks down will adversely
affect the economics of the entire system and will significantly lengthen the
time required to complete the remedial/removal action. Other operating
parameters which affect performance are listed in Table 6.(3)
Of the waste feed properties which most affect performance, the H:C1 ratio
of the key waste components most affects the tendency of the waste stream to form
undesirable byproducts during the incineration process. While these compounds
are not always formed, other stable byproducts can be formed and emitted even
as the original compounds are being oxidized in the incinerator. As key waste
components become more chlorinated (i.e as the H:C1 ratio decreases), the
byproducts formed become more stable. This means that they are less likely to
1449
-------
c
o
4-«
5
u
5. POTENTIAL INCINERATION ARARS8
Requirement
w
s
^
X
!S
CO
u
1
u.
o
«4-
0>
*-*
'«
'1
i-
u
O-
V—
HI
\s
f\)
oe
u
O
and chemical
e the waste feed to determine physical
ition limits.
N CO
sl
< U
41
V)
§
ai
|
CO
N
a
*| <
ul 5
acl a;
*!
3
CM
a:
u
0
t_
4>
2
including ash, scru
licable requirements
e of all hazardous waste and residues,
and scrubber sludge, according to app
CO ••
U. 01
Ul 4-1
0 HI
Ki ro
>s >s
ru f\j
c* tx.
u o
0 0
c
CO W O* ••"
•* v ' c> «•> o
&j 2 "J 41 ^
CA Q 4-* ^"« H— ^ ** — '
CD CA — ^ «^- ttl O *-* L-
C.C.O ^ ^ v £
— M_.i.2c!x^ >-Si-
4> O <-• W 01— ^ R X
L. ux«-> c S150-
0 4) CD ~^ «~ Ai g) 4f w
*> O «. S •- "a> .2 S S
5 . * S w H-coS
No further requirements for wastes tl
ous solely because they exhibit one or
teristics of ignitability, corrosivity
ail the TCLP leaching test and a waste
endix VIII constituent is present that
ed to be present.) Such wastes may all
ix VIII constituents are not present a
mance standards:
Achieve a destruction and removal ef
percent for each principal organic h
designated in the waste feed and 99.
and PCB contaminated liquids.
A) i- ID u U C O
O N CD 4) U. CL C.
o 1.8 kg/hr or to 1
before entering any
Reduce hydrogen chloride emissions t
percent of the HCI in the stack gas
pollution control device.
scm (0.08 gr/dscf)
No release of particulates >180 mg/d
corrected to 7% Oxygen.
tA
'c "~
K o
° JC
^ , 4^
< ^
CJ *
Of
4^
t_>
o
to
•sj-
(A
C.
Disposal Restrictiol
All residues must meet the RCRA Land
t fugitive emissions by:
o
t_
4-*
i
u
'*-
41
i
4-t
CO
C
CD
41
2
O
Keeping combustion zone sealed; or
Maintaining combust i on- zone pressure
pressure.
""!
V
r\j
a:
u.
5
1
C
(U
.c
2 .
W
'a *^
tomatic cutoff system to stop waste fee
ions deviate or exceed established limi
CD .—
41 C
:3 o
Tl
Oi
3
C
1450
-------
UD
CO
tt
cc
u.
o
o
a:
c •<-
o o
•— o
•D
Ol
8.
o
i - o
2-0
V? Ol
10 .C
X W
C t-
o o
c
-c oi
in x
to o
01 ••-
o 01
4->
TD
E <1J
£ 01
+-> o
to x
>. Ol
C-
it- O
1-
O Ol
T3 Ol ,*
+> O
to 10 (o
m 01 C
g 01 •--
I- Ol
10 Ol T3
O. -U •—
to oi
c o
O (-
•^ 01
t/1 Q.
to
U Ol
IO t- C
> 3 O
(- 10 t-
o i- - o
•o
CD
C
O
CO
§1
» QJ QJ
*•> 'S 3
tfl (0 i—
10 3E O
X >
o>
•D O. Ol
^ >, 4->
O I 10
w '— x
•o
Ol Q. Ol •
C — C U
•r- O "~ Ol
C — O 4->
t- C 3 4-1
3 3 "O
-------
g
U
O
a:
u.
U
o
-*
oc
u.
U
O
T
ID
?
ID
L.
01
a.
I
L.
O
§
u
UJ
_J
at
u
4^
a
O)
u
a. •
T3 4!
•f- *^
3 ID
& a>
1
c
1453
-------
if
g
4_»
g
o
»»
ai
_/
CD
<
>—
c
o
4"^
a
4^
o
4-*
L.
1
ex.
X
'I.
£
a
u
*^
|
c_
o
*4-
*erequisite
a.
e
*
s
at
Ik
U
o
>^
PS
n- at
O -
4J — »
•?¥
Monitoring must occur:
When the incinerator is first used or modified; moi
must measure for 02, CO, CO,, oxides of nitrogen, 1
PCBs, total particulate matter.
^
1
6
I
w
31
£1
CA
1 X
— —
Ml
Whenever PCBs are being incinerated, the 02 and CO
must be continuously checked; CO, must be periodic)
checked.
e s s
• * •
s s s
at at at
Ik U. Ik
O ec
C -D u.
1 fe o Z
%
x
c .
D "^
o ^
t- V
1?
U E
c o
o) a!
t_ L.
O v
a"
& C
U 41
X CA
41 C.
CO
O>T3
Si c
•Q CO
O •>
••Z E
4J 5
CA ••-
^1
U U
l-o
o> eg
fc^X
CO
in
o
at
u
o
^
CA
4J
.c c
4J 4£
3 v
4i .i:
Dispose of liquids, sludges, or solid residues in accordanc
applicable Federal, State, and local pollution control requ
CO
m
•0
at
u
o
^^
> ..
CA
CO
4^
i
Chemically or physically treat pesticides to recover heavy
incinerate in same manner as organic pesticides.
x
C
3
y
U
(A
4J ^3
8-S
U O
x u-
* S
- o
CA
*|
0 4>
— CO
4-* C-
$ «
B-fe
u
tallo-organ
ad, cadmium
ecommended)
41 41 i-
z — ^
c>
in
<0
ae
u
o
1
u
4^
CA
2L
u
'E
CO
O)
!_
O
CA
CO
c_
CA
C
4)
«•>
CO
ii
|
4->
8" E
cj 3
•D X •-
— - cu f=
41 -B
^ - CO
CO O
X 01
t."2?
CU u CO
C. 4-» i.
O CA U
H- 41 —
2L C
4^ 4)
CO U CO
f '^ L,
4J C CO
CO
CO O) •
c. 1- TJ
4) 0 CO
.S i-
CO — •
§CD X
1 . ^ .
O 41 3 **
E o p
4) l- 4f
— L. CU •§
£ 0 E C
4J U O E
3 E C 5
n eg a u <|
e TO o> 4> xl
B L. t. c. cnl
cj o o ^ Ol
o -o
Sni
o o*
•— w—
ae at
u. u.
u u
aa
•D
All remediation activities must comply with the policies an
programs established for worker safety.
+j
'>
4-1
CO
§
4^
CO
at
CD
1
1
(D
O*
.X
8
CU
D)
<
C
o
k*
0
CU
4^
o
L.
Q.
a
4^
C
1
o
L.
>
C
UJ
c/>
Source: U.
CD
CU
U
C
CO
4-f
E
u
u
CD
C
c
o
4J
a
^
u.
orovisions or variances applicable to the specific site under
• any decisions are formulated.
— o>
. L.
.2-2
S-8
«"0
0
C 4-1
io 3
S c
0 O
0 0
|.S
c -o
•5|
fc-g
£
•D f-
5 c
s °
"to1
c^
.£?
4J U
(0 _,
3 3
ffS
2 S
Is
4^
xi
1453
-------
TABLE 6. EXAMPLE OPERATING PARAMETERS AND HOU THEY AFFECT PERFORMANCE
Operating parameter
Effect
Temperature
Combustion gas flow rate
Waste feed rate and heat content
Moisture Content of the Waste
Air input rate
Waste atomization
Feed System
Mixing/Turbulence
Combustion reactions rates of burning are faster at
high temperatures until the rate is limited by
mixing. High temperatures can also elevate NO*
emissions.
For a fixed chamber volume, the waste constituents
remain in the chamber for a shorter time (have a
lower residence time) as the flow rate increases.
As the combustion gas flow rate increases, gas
velocity through the chamber increases. This can
result in increased entrainment of solid material
(fly ash) and emission of particulates.
As waste feed rate decreases, the heat release in
the combustion chamber will decrease and
temperature may drop. Waste heat content can
affect combustion temperature. Insufficient heat
content can result in the need for auxiliary fuel
which will adversely affect the economics of the
process. Wide variations in heating value of the
waste can cause puffing (positive pressure surges)
in rotary kiIns.
Moisture decreases the heat content of the waste
and, as a result, reduces the combustion
temperature and efficiency when high moisture waste
is burned.
Air supplies oxygen for the combustion reactions.
A minimum is needed to achieve complete combustion;
however, too much air will lower the temperature
(because the air must be heated) and quench
combustion reactions due to excessive cooling. The
additional air will increase combustion gas flow
rate, which then lowers the residence times.
Increased air input can increase combustion
efficiency by increasing the amount of oxygen
available to oxidize organic contamination.
Atomizing liquid waste into smaller droplets will
increase the effectiveness of fuel/air mixing and
the burning rate. Waste feed and atomizing fluid
(air or steam) flow rates and pressures affect
atomization. Suboptimal waste feed and atomizing
fluid flows will result in less efficient
atomization resulting in the production of larger
fuel/waste droplets.
Consistent, reliable delivery of waste feed into
the incinerator is critical to the efficient
operation of an incinerator. The design of
appropriate feed systems can be difficult for
inconsistent or difficult feed streams.
A burner must be selected which induces adequate
turbulence into the combustion air/fuel/waste
mixture. This promotes good mixing of air and fuel
which leads to efficient combustion.
Source: ASME 1988.
1454
-------
be destroyed in the incinerator or afterburner and are more likely to be emitted.
Under oxygen starved conditions, the tendency to form byproducts increases. It
should be noted that even though combustion byproducts are routinely formed,
dioxins and furans have rarely been observed in emissions from hazardous waste
incinerators. EPA has sampled a number of incinerators to measure performance
relative to the RCRA incineration regulations. As part of that effort samples
were taken to determine whether dioxins and furans were formed as byproducts of
the incineration process. None were found despite extensive sampling.(2)
Despite the fact that dioxins and furans are seldom formed as byproducts
of the hazardous waste incineration process, other byproducts can be formed if
incineration performance is suboptimal. Certain failure modes can lead to the
incomplete combustion of organic contaminants and, as a result, exacerbate the
formation of these combustion byproducts. The byproducts which are formed under
these conditions depend largely on the chemicals which are being incinerated.
Chloroform, for example, has been shown to form nine different regularly
occurring byproducts. Eight of these are short chain (C-l and C-2) chlorinated
hydrocarbons. The other one is hexachlorobenzene. A listing of compounds and
the byproducts usually observed from their combustion is included in Table 7.
(3)
INCINERATION EXPERTS, VENDORS, AND RODS
The OSC/RPM responsible for directing an incineration project needs access
to a wide variety of expertise. Each State and Regional office has incineration
experts who are available to advise OSCs/RPMs on incineration issues. These
technical specialists are located in each Regional office and are often involved
in RCRA incineration permit review. They should be consulted on every
incineration project since they can be of great help in providing needed
technical support. A list of these individuals is provided in the summary
report.
A survey of Superfund related incineration activity around the nation
reveals some interesting facts. As of 1989, incineration RODs had been written
for sites throughout the nation. Region V had the most incineration RODs
(eighteen) while Region IX had the fewest (one). In general, the western United
States had fewer incineration RODs than the Eastern and Midwestern parts of the
country. The agricultural sections of the country (roughly comprised of Regions
7 and 8) also had relatively few incineration RODs.
If the volume of material requiring incineration is too small to justify
the expense of bringing an incinerator on-site, it may be practical to ship the
waste to an off-site commercial facility for disposal. The summary report lists
eight off-site commercial facilities that may be used for disposal of wastes.
Any use of these facilities must comply with the "off-site" policy (OSWER
Directive 9330.2-1). Although Region 5 had the most incineration RODs, the
largest number of off-site commercial incineration facilities (6) are located
in the Southern U.S., specifically in Region 6. Region 5 does, however, have
three off-site commercial incineration facilities.
Vendors of mobile/transportable incinerators are located in the regions
with the most incineration RODs, with the exception of Region 9. In California,
there are 2 mobile incinerator vendors while Region 9 only has 1 incineration
ROD. Of the eleven mobile/transportable incinerators identified in the report,
1455
-------
TABLE 7. REACTION PRODUCTS OBSERVED FROM THERMAL DECOMPOSITION STUDIES8
Parent (POHC)
Product (PIC)
Condition
Carbon Tetrachloride
Pentachlorobenzene
Chloroform
Chloroform
Mixture of
CCl, 53% (mole)
CHCT, 33%
CHpCT, 7X
CHjCl 7X
Tetrachloroethene
Hexachloroethane
Hexach I orobutadi ene
Hexach I orobenzene
1,2-C,H?Ct?
*
Air atmosphere. tp* = 2.0 s
Air atmosphere, tp = 2.0 s
o = 0.67, tp = 2.0 s
C,HCl,
CC1
C2C12
C,H,tl,
* 4
Carbon Tetrachloride
Trichloroethene
PentachIoroethane
Dichloroethyne
Tetrachloroethene
Tetrachloropropyne
1,1,2,4-Tetrachloro-1-buten-3-yne
Hexach I orobutad i ene
o=0.76 and Nitrogen
atmospheres
CCl,
CHCC,
CH2CT2
CH,Cl
c28i2
171-E2H2CI2
C,HCl,
CCl
24
Pyrolytic, t = 2.0 s
etc*
a This table was excerpted from a table appearing in a DDR I report on PIC minimization entitled
Minimization and Control of Hazardous Combustion Byproducts Final Report and Project Summary prepared for
U.S.E.P.A under cooperative agreement CR_813938-01-0 summarizing the results of flow reactor studies
conducted at the University of Dayton Research Institute. The complete table can be found in the above
listed reference.
1458
-------
seven are rotary kilns, two are infrared incinerators, one is a circulating
fluidized bed and one is a conventional fluidized bed. The average size is 30
Million BTU/hr with an average processing cost of $350/ton.(3)
The distribution of on-site public and private sector thermal remediation
projects is slightly different than that for the distribution of RODs with most
of the activity being located in the Southeastern U.S. (Region 4) rather than
the Midwest (Region 5) and the Northeast (Region 2). There is significant on-
site thermal remediation activity in California (six sites) despite the fact that
there is only one incineration ROD in Region 9. (3)
Of the fifty-one on-site thermal remediation projects identified in the
summary report, 53% are finished, 39% are contracted and only 8% are currently
ongoing. The average site has 27,000 tons of contaminated material and is being
cleaned up with a 34 Million BTU/hr incinerator. The incinerators used at these
sites are provided by twenty different vendors with no vendor providing
incinerators for more than 12 % of the projects listed. Most of the incinerators
(43%) were rotary kilns. The second most frequently used technology was low
temperature direct desorption. This was used at 29% of the sites listed. Other
technologies used were infrared incineration, high temperature direct and
indirect desorption and circulating fluidized bed incineration.
CONCLUSIONS
Because incineration is a controversial and expensive remedial method,
completing an incineration project is difficult unless the RPM/OSC has access
to the most up-to-date information available. Fortunately, incineration has been
used widely and a large body of knowledge about the proper implementation of this
technology exists. Access to this information is easiest through consultation
with Regional and State incineration experts and with other RPMs/OSCs who have
recently or are currently implementing incineration projects. In addition,
current literature on incineration, especially ORD publications and OSWER
guidance documents can provide in-depth information on selected topics. The
summary report summarized in this paper and prepared by the Engineering Forum
and the Risk Reduction Engineering Laboratory will help OSCs and RPMs to make
effective use of this large body of incineration experience.
REFERENCES
1. OSWER Directive 9835.13 A Comparative Analysis of Remedies Selected in
the Suoerfund Program During FY 87. FY 88 and FY 89. June 1990
2. Oppelt, E. T. Incineration of Hazardous Wastes, A Critical Review.
Journal of the Air Pollution Control Association, Vol. 27 No. 5 May 1987.
3. Superfund Engineering Issue: Issues Affecting the Applicability and
Success of Remedial/Removal Incineration Pro.iects EPA/540/2-91/004
February 1991.
Author(s) and Address(es)
Laurel J. Staley
U.S. Environmental Protection Agency
26 W. Martin Luther King Dr.
Cincinnati, Ohio 45268
(513) 569-7863
1457
-------
REMEDIATING TCE-CONTAMINATED SOILS: A CASE STUDY
OP A FOCUSED RI/FS AND VACUUM EXTRACTION TREATABILITY STUDY
W. Winslow Westervelt, P.E.
Thomas R. Hundt, Ph.D.1
Gannett Fleming, Inc.
Suite 200, East Quadrangle
Village of Cross Keys
Baltimore, Maryland 21210
(301) 433-8832
Michael c. Marley
Vapex Environmental Technologies, Inc.
480 Neponset Street
Canton, Massachusetts 02021
(617) 821-5560
ABSTRACT
A focused remedial investigation/feasibility study (RI/FS) was
conducted for EPA Region III to determine the extent of
trichloroethene (TCE) contamination in soils at a former
sanitary landfill site and to evaluate alternatives for soil
remediation. The investigation revealed high concentrations of
TCE (up to 330,000 /ig/kg) trapped in a 50-foot-deep vadose zone,
and high concentrations of TCE and acetone (up to 840,000 ng/kg)
in the saturated soils above bedrock. The overburden soils in
the vicinity of the spill areas are between 40 to more than
100 feet deep and were classified as predominately silt. Due to
the depth of contamination and potential problems of controlling
volatile organic compound (VOC) emissions, a combination of
capping and in-situ vacuum extraction was considered to be the
most promising alternative for this site.
To evaluate the effectiveness, implementability, and cost of
vacuum extraction, a pilot-scale treatability study was
performed at the site. Physical and chemical data were
collected over a two-week period that allowed for determination
Currently with EA Engineering Science and Technology, Sparks, Maryland
1458
-------
of the radius of influence of vacuum pressure in various soil
units, an evaluation of the effects of key operating parameters
and system designs on performance, and an estimation of the time
required to remediate the contaminated soils. Subsurface air
flow and contaminant removal models were calibrated to the
pilot-scale data and used to predict the performance of various
full-scale system configurations that included nested vacuum
extraction wells, surface capping, and air injection wells.
Preliminary costs and designs for full-scale remediation systems
were developed.
INTRODUCTION
In 1967, the Heleva Landfill Site began operations as a sanitary
landfill, accepting between 250 and 350 tons per day of general
mixed refuse from the Allentown, Pennsylvania, area. In
addition to the municipal wastes, industrial wastes consisting
of chlorinated organic solvents were sent to the site and
improperly disposed by dumping the liquids onto the ground in
one or more "spill areas." The organic solvents appeared in a
neighboring town's water supply wells, alerting citizens and
regulatory agencies to a potential public health threat. The
landfill was closed in 1981 under consent order because of
operational deficiencies, and in 1982 was placed on the National
Priorities List (NPL) for hazardous waste sites in accordance
with the Comprehensive Environmental Response, Compensation and
Liability Act (CERCLA).
A focused remedial investigation/feasibility study (RI/FS) was
initiated in 1988 to determine the location of the spill area(s)
and to evaluate remedial alternatives for the soil that
continues to be a source of contamination to the bedrock
aquifer. A total of 42 soil borings were drilled during the
subsurface investigation to classify soils and obtain soil and
water samples for laboratory analyses. The suspected locations
of the spill areas are depicted in Figure 1. Quick-turnaround
time chemical analysis of target compounds allowed the field
team to focus the placement of the soil borings in the potential
spill areas, limiting the total number of borings required to
define the extent of contamination. The investigation revealed
high concentrations of trichloroethene (TCE) trapped in a
50-foot-deep vadose zone, and high concentrations of TCE and
acetone in the saturated soils above bedrock. The soils were
classified as predominately silt interspersed with sandy silt
and lean clay. The depth of overburden soils in the vicinity of
the spill areas range between 40 to more than 100 feet. It is
estimated that approximately 392,000 cubic yards of soil are
contaminated above a TCE remediation goal of 30 jig/kg.
A number of technologies were considered during the development
of remedial alternatives, including capping; excavation with
either thermal, fluid extraction, or biological treatment; and
in-situ vapor recovery processes such as vacuum extraction and
1459
-------
Figure 1. Map of study area showing surface features and suspected boundaries of solvent spill areas.
steam stripping. Due to the depth of contaminated soil and
potential problems of controlling volatile organic compound
(VOC) emissions during excavation and treatment, capping and in-
situ vacuum extraction were considered to be the most promising
technologies for the spill areas. In response to a previous
Record of Decision (ROD) for the site, a synthetic membrane cap
conforming to Resource Conservation and Recovery Act (RCRA)
standards was constructed over the landfill area and a portion
of the spill areas shortly after the completion of the RI field
investigation.
To further evaluate the potential effectiveness, implementa-
bility, and cost of vacuum extraction as a remediation
technology, a pilot-scale treatability study was performed at
the site. Physical and chemical data were collected over a two-
week study period that allowed for a determination of the radius
of influence of vacuum pressure in various soil units, an
evaluation of the effects of key operating parameters and system
designs on performance, and an estimation of the time required
to remediate the contaminated soils to specified cleanup
criteria. Preliminary costs and conceptual designs for full-
scale remediation systems were developed.
1460
-------
BACKGROUND
Physical Characteristics of the Site
The overburden soils encountered during the subsurface
investigation were predominately silt, sandy silt, and lean
clay. Isolated and discontinuous lenses of silty sand with
gravel, fat clays, and elastic silts were also encountered.
Since the study area was once an open-pit iron ore mine, it is
possible that much of the natural stratigraphy has been altered
due to reworking of the soils.
A closer look at the engineering characteristics of these soils
revealed that the silts and clays are very similar. The
permeability of the silt/clay soil at the Heleva Landfill Site
was determined in laboratory analyses to have a range of 10~6 to
10~7 cm/sec. Permeabilities of this nature indicate a "tight"
soil matrix. The permeability of the silty sand with gravel
soil which occurs in isolated and discontinuous lenses was
estimated to be 10~3 cm/sec.
Water levels observed during the subsurface analysis generally
occurred between 400 and 410 feet above mean sea level,
approximately 50 to 60 feet below grade. In some instances, a
water level was observed in the sandy lenses above the static
water table, indicating localized areas of perched water.
Nature and Extent of Contamination
During the field investigation, soil samples were collected at
each sampling location at regular 10-foot intervals and analyzed
for VOCs within 24 hours for quick-turnaround time analysis and
within 14 days for standard Contract Laboratory Program (CLP)
analysis (EPA Method 601/602). A summary of the concentration
ranges and average concentrations for all VOCs detected is
presented in Table 1.
TCE was the most widespread and prominent soil contaminant
detected, at concentrations up to 330,000 /zg/kg. By using
kriging techniques to statistically correlate data between
borings, areas requiring further sampling were identified and
eventually two distinct spill areas were delineated. The
locations of the spill areas in plan view and kriged estimates
of log TCE isoconcentration contour lines are shown in Figure 2.
A cross section of the spill areas with kriged isoconcentration
lines of TCE contamination is presented in Figure 3. It is
noted that the isoconcentration lines do not extend into the
landfill or the bedrock since the scope of this focused RI/FS
was limited to contaminated natural soils and samples from the
landfill and bedrock were not collected.
Biological degradation of compounds such as TCE may have created
several "daughter" compounds in the soil where only one compound
1461
-------
may have been present initially. The breakdown of TCE to
1,1-dichloroethene (DCE), cis and trans 1,2-DCE, both 1,1 and
1,2-dichloroethane (DCA), vinyl chloride, and chloroethane leads
to the production of six additional chlorinated hydrocarbons.
DCE was present at more than one-half of the sampling locations
along with TCE, but at somewhat lower concentrations (up to
35,000 pig/kg) . Vinyl chloride was detected in soil gas samples
but was undetectable in nearly all soil samples, although its
absence is most likely related to its extreme volatility.
Tetrachloroethene (PCE) and 1,1,1-trichloroethane (TCA), which
are "parent" compounds of TCE, DCE, and DCA, were also present
at about half of the sampling locations, indicating that these
compounds were disposed at the site along with TCE.
Acetone was detected at moderate to high levels (up to
840,000 /ig/kg) in samples taken from the saturated soil zone.
Since acetone is completely miscible in water, it is possible
that acetone solutions disposed in the spill areas migrated
quickly through the vadose zone and were concentrated in the
saturated soil layer. Moderate concentrations of chloroform (up
to 3,700 /xg/kg) , another widely used industrial solvent, were
also detected at the site. Fuel-related compounds (benzene,
ethylbenzene, toluene, and xylenes) were detected at various
locations throughout the site.
Semivolatile organic compounds (SVOCs), pesticides,
polychlorinated biphenyls (PCBs) and inorganic elements were
TABLE i
SUMMARY OF VOC CONCENTRATIONS IN SUBSURFACE SOILS
Compound
Acetone
Benzene
2-Butanone
Carbon Oisulfide
Chlorobenzene
Chloroform
1 , 1 -Dichloroethane
1,1-Dichloroethene
Total 1,2-Dichloroethena
Ethylbenzene
Methylane Chloride
4- Methyl- 2-Pentanone
1,1.2. 2-TetracMoro«than«
Tetrachloroethene
Toluene
Total Xylanes
1,1,1 -Tnchloroethane
Trichloroethene
Vinyl Chloride
Concentration Range
(pg/kg)
1 5-840.000
2-56
68-9,000
11-11
3-58
4-3.700
4-19
10-49
2-35,000
1-460
3-1 1 ,000
10-150
4-4
1-1.700
0.2-91
1-1,600
3-3.400
1-330,000
10-540
Average Concentration
(//g/kg)
20.155
3
291
3
4
30
2
3
583
10
153
16
2
32
2
37
57
8.648
11
CRQL
(//g/kg)
10
5
10
5
5
5
5
5
5
5
5
10
5
5
5
5
5
5
10
No. of Detections/
Total No. of Samples
32/147
2/37
5/37
1/110
5/37
15/147
4/110
3/147
139/282
14/147
11/147
4/37
1/37
65/282
8/147
23/147
70/282
181/282
4/147
Note: CRQL = Contract Required Ousntitation Limit
1462
-------
analyzed for in seven samples taken from the spill areas. In an
isolated occurrence, phenol was detected at 4,100 jug/kg-
Moderate concentrations of dichlorobenzenes (170 to 8,700 jug/kg)
and phthalates (100 to 550 /ig/kg) were also detected. There was
no pesticide or PCB contamination found. Inorganic elements
were typical of average background concentrations for Eastern
U.S. soils (Shacklette and Boerngen, 1984), except for raised
concentrations of cobalt, iron, and manganese that are likely
related to iron ore deposits in this former mine area.
Groundwater samples were obtained using screened stainless steel
wellpoints whenever saturated conditions were encountered. The
relative distribution of contaminants observed for the soil
samples was observed in water samples as well; however, the
concentrations in groundwater tended to be much higher. Acetone
and TCE were encountered at concentrations up to 1,900,000 and
930,000 Mg/L, respectively. Vinyl chloride was also detected at
concentrations up to 19,000 nq/L. Several factors may have been
responsible for the higher concentrations of VOCs in the water
as compared to the soil: 1) the VOCs may have been concentrated
at the air/water interface at the top of the groundwater table,
2) the VOCs may have been partially flushed from the vadose zone
by percolating rainwater, and/or 3) the measurement of VOCs in
the water may have been more accurate than in soil because of
the zero headspace in the water sample vials.
Development of Remedial Action Goals
The primary concern at the site is contaminated soils acting as
a continuing source of contamination to the bedrock aquifer.
Since enforceable federal or state standards have not yet been
promulgated for soil contamination, the remedial action goals
were based on meeting contaminant-specific Applicable or
Relevant and Appropriate Requirements (ARARs) for the
groundwater beneath the site. Primary drinking water standards
Figure 2. Plan view of site showing locations of
soil borings and kriged isoconcentration lines
of TCE contamination in soil at 430' MSL,
approximately 30 feet below ground surface.
Figure 3. Vertical cross section of spill areas
showing kriged isoconcentration lines of TCE
contamination.
1463
-------
known as the Maximum Contaminant Levels (MCLs) and Maximum
Contaminant Level Goals (MCLGs), developed by EPA in response to
the Federal Safe Drinking Water Act, were determined to be
relevant and appropriate requirements since the groundwater may
be used for drinking water after remediation of the aquifer is
complete. The Final National Contingency Plan (NCP) promulgated
in 1990 specified that non-zero MCLGs shall be attained by
remedial actions for groundwater or surface waters that are
current or potential sources of drinking water. When there is
no MCL or MCLG set for a contaminant, a calculated level based
on the health risk from water consumption can be used. The
resulting groundwater protection standards for site contaminants
are summarized in Table 2.
A soil cleanup goal was defined as the concentration of a
contaminant in the vadose zone soil that remedial alternatives
need to achieve to prevent contamination of groundwater above
the groundwater protection standards. Cleanup goals for the
vadose zone soils were calculated using a combination of the
Hydrologic Evaluation of Landfill Performance (HELP) Model (EPA,
1984), the Summers Model (Summers, et al., 1980), and a
calculation to determine the effective aquifer mixing depth
(Woodward-Clyde, 1988). The HELP Model was used to estimate the
rate of rainwater infiltration through natural soils or through
a synthetic membrane cap system. The Summers Model assumes that
infiltration will desorb contaminants from the soil following
equilibrium soil/water partitioning theory. It is further
TABLE 2
GROUNDWATER PROTECTION STANDARDS FOR SITE CONTAMINANTS
Contaminant
Acetone
Benzene
2-Butanone
Chlorobenzene
Chloroform
1 , 1 -Dichloroethane
1 , 1 -Dichloroethene
cis 1 ,2-Dichloroethene
trans 1,2-Dichloroethene
Ethylbenzene
Methylene Chloride
Tetrachloroethene
Toluene
Trichloroethene
1 , 1 , 1 -Trichloroethane
Vinyl Chloride
Total Xylenes
MCL
U/g/L)
NP
5
NP
100
NP
NP
7
70
100
700
5
5
2,000
5
200
2
10,000
MCLG
U/Q/U
NP
0
NP
100
NP
NP
7
70
100
700
0
0
2,000
0
200
0
10,000
Calculated Values
Based on Risk
U/g/L)
3,500
1,890
100
440
Note: NP = Not Promulgated
1464
-------
assumed that contaminated infiltration upon reaching the
groundwater will mix completely within the mixing zone
calculated by the Woodward-Clyde model, resulting in an
equilibrium between groundwater and soil contaminant
concentrations. The soil contaminants with maximum
concentrations greater than or nearly equal to the calculated
soil cleanup goals are summarized in Table 3. The cleanup goals
would differ significantly if the infiltration rate was reduced
by capping the contaminated area. If a synthetic membrane cap
is installed (assuming a leakage factor of 0.1 percent), the
Summers Model predicts that only TCE, methylene chloride and
acetone soil concentrations would cause the groundwater
protection standards to be exceeded.
Development of Remedial Alternatives
Potential remedial technologies and process options were
identified and screened according to their overall applicability
to the conditions and contaminants at the Heleva Landfill Site.
The general categories of technologies initially considered were
containment, thermal treatment, soil vapor recovery, fluid
extraction, biological treatment, and soil dewatering. Process
options which required excavation of soil prior to treatment
(i.e., incineration, solid- and slurry-phase bioremediation, and
soil washing) would be difficult to implement because of the
depth of contaminated soils and the public health risks of VOC
exposure to site workers and the community during excavation.
In-situ process options that require subsurface injection of
fluids (i.e., in-situ biotreatment and soil flushing) would not
work well with the low permeability soil which would limit the
ability to contact contaminated soil particles and recover the
contaminated solutions. Process options considered to be the
most applicable to the site conditions were capping, vacuum
extraction, in-situ steam stripping, and soil dewatering.
Capping the contaminated areas with a composite soil and
synthetic membrane liner cap system would be expected to
eliminate most of the contaminated infiltration reaching the
groundwater. However, capping alone does not comply with the
statutory preference for treatment-based alternatives as
directed by the NCP. In-situ vacuum extraction was judged to be
the best treatment-based technology for contaminated vadose zone
soils. In-situ steam stripping or a combination of soil
dewatering and vacuum extraction were considered potentially
applicable for the treatment of contaminated saturated soils.
A combination of capping and treatment-based technologies would
greatly reduce the volume of soil requiring treatment since less
stringent soil cleanup goals would apply.
Treatabilitv Study Objectives
To facilitate a detailed evaluation of the application of vacuum
extraction at the Heleva Landfill Site, an onsite pilot-scale
1465
-------
TABLE 3
SOIL CLEANUP GOALS
Contaminant
Acetone
Benzene
2-Butanone
Chloroform
1,1-Oichloroethene
Total 1,2-Oichloroethene
Methylene Chloride
Tetrachloroethene
Trichloroethene
1,1,1-Trichloroethane
Vinyl Chloride
Concentration Range
in Soil
U/8/kg|
15-840,000
2-56
68-9,000
4-3,700
10-49
2-35,000
3-11,000
1-1,700
1-330,000
3-3,400
1O-540
Soil Cleanup Goal
With No Cap
(M/ka>
410
20
460
170
20
1 80 (as)
320 (trans)
2
100
30
1,600
6
Soil Cleanup Goal With
Synthetic Membrane Cap
to/kg)
715,000
NG
NG
NG
NG
NG
NG
4,100
NG
59,000
NG
NG
Note: NG « No Goal, calculated cleanup goal i* greater than highest observed concentration
treatability study was performed. The treatability study was
designed to satisfy several objectives:
• Radius of Influence—to determine the radius of influence
of vacuum pressure in various soil units at the site.
• Operating Parameters—to evaluate the effects of key
operating parameters on system performance, including vapor
extraction rate and vacuum pressure.
• System Configuration—to evaluate the effects of various
system components and configurations on system performance,
including capping and air injection wells.
• Remediation Time—to estimate the length of time required
to remediate the contaminated soils to the soil cleanup
goals.
• Cost—to evaluate the major cost items associated with a
full-scale system.
Pilot Test System Configuration and Installation
An area approximately 20 by 50 feet in size was selected within
the limits of a spill area for performing the treatability
study. Soil borings in this area revealed a soft silt to a
depth of 20 feet, a relatively coarse layer of slightly silty,
coarse to fine sand between the 20- to 25-foot depth, and stiff
silt with varying amounts of sand and gravel below 25 feet.
Soil moisture was visually classified as "moist to wet" from a
depth of 10 to 25 feet, and a noticeable decrease in soil
1466
-------
moisture was observed below 25 feet. The water table was
encountered at a depth of approximately 50 feet.
Based on this geological stratification, the recommended
installation for the vacuum well/vapor probe network consisted
of two vertical vacuum extraction wells nested in a single
borehole and 13 vapor probe monitoring points nested in four
additional boreholes. The nested vacuum well configuration was
a shallow well screened between 5 and 18 feet below grade, and
a deep well screened between 30 and 45 feet below grade. The
presence of the coarse to fine sand layer was a primary factor
in determining the screen intervals for the vacuum wells. Due
to the potential for short circuiting of air through the more
permeable sand layer during vacuum extraction operations, a
bentonite seal was installed between the two well screens over
the entire depth of the sand layer to isolate the two wells.
Vapor probes were installed in four boreholes located 4.75, 8,
15, and 47 feet from the vacuum wells to measure vacuum pressure
and soil gas contaminant concentrations at discrete depths and
at a range of distances from the extraction well. A cross
section depicting the placement of the vacuum extraction wells
and vapor probes is shown in Figure 4.
The pilot test system was assembled and installed adjacent to
the vacuum extraction wells. The system included a 15-cfm
liquid ring vacuum pump, a 10-cfm rotary vane oil-less vacuum
pump, two air/water separator drums, six 200-pound canisters of
activated carbon, and associated meters, gauges, valves,
fittings, and piping. A schematic diagram of the pilot-scale
extraction test system is presented in Figure 5.
Treatabilitv Study Procedures
Following assembly and installation, the pilot-test system was
activated and operated over a 14-day period. The shallow well
was tested over the first 10 days and the deep well was tested
- CONCKfJf
- OENfOHlIE
- SAND
- '.ROUND WV
Figure 4. Schematic diagram of vacuum
extraction well (VW) and vapor probe monitoring
well (VP) construction.
Figure 5. Schematic diagram of the pilot-scale
vacuum extraction test system.
1467
-------
over the final four days of the study period. Three air flow
rates were utilized during each test to produce data for
calibration and verification of subsurface air flow models. The
time-weighted average air flow rates for the shallow well tests
were 7.0, 11.0, and 12.9 cfm, and for the deep well tests, 4.6,
5.1, and 7.2 cfm.
Vacuum extraction system operating parameters were recorded on
a daily basis. The operating parameters included wellhead
vacuum, wellhead flow rate, flow meter temperature, wellhead
temperature, and vacuum at the pump. Vacuum readings were taken
at each vapor probe location at least once per day.
Samples were collected for VOC analysis at regular intervals
during the course of the treatability study from the vapor probe
soil gas, wellhead soil vapor discharge, carbon canister vapor
discharge, and air/water separator drain water. Onsite analysis
of vapor samples was performed with an HNU Model 321 Gas
Chromatograph equipped with an 11.7 eV photoionization lamp
(GC/PID). Sample screening was performed with a hand-held
Thermo Environmental Instruments Model 58OA OVA total organic
vapor analyzer equipped with an 11.8 eV lamp (TECO 580A) . Water
samples and vapor samples for confirmational analysis were
analyzed by an offsite laboratory by EPA Methods 601/602 and
T01/T02, respectively.
All field sampling and analyses were performed in accordance
with strict quality assurance/quality control (QA/QC)
procedures. QA/QC procedures for the GC/PID consisted of
routine analysis of field blanks, standards, and duplicate
samples in order to monitor the instrument's performance.
Calibration of the TECO 580A was checked on a daily basis
against a known standard of PCE.
DISCUSSION
Treatability Study Results
The relationship between vacuum levels and flow rates observed
at the wellhead is depicted in Figures 6 and 7. As expected,
best-fit lines plotted through the data points show a slightly
curvilinear relationship of diminishing flow rates at higher
operating vacuum pressures.
Vacuum pressure was observed in at least one vapor probe at each
borehole location over the test period at levels ranging from
0.005 to 3.1 inches of water during the shallow well test, and
from 0.005 to 0.45 inches of water during the deep well test.
Vacuum pressure was not detected at several probes over the
first five days of the shallow well test; it is likely that
condensation in the Teflon tubing connected to the vapor probes
may have been blocking the lines and interfering with vacuum
reading. Corrective measures were taken by injecting 150 ml of
1468
-------
air into the tubing 30 minutes prior to measuring vacuum at each
of the probes. Readings taken after clearing the tubing were
generally more stable and consistent than those observed prior
to clearing the lines. Vacuum pressure was consistently not
detectable at several probes in boring VP1 which was closest to
the extraction well. Since these probes were expected to have
the highest vacuum pressure, it was concluded that some of the
vapor probes in this borehole may have been sealed off from the
surrounding soil during installation from smearing of the
borehole walls with wet, clayey soil as the augers were
withdrawn.
Contaminant discharge concentrations for the shallow and deep
wells are shown in Figures 8 and 9, respectively. The total
target VOC concentration in the shallow well ranged from a
maximum of 11,787 ppm (v/v) on the fifth day of the test to a
minimum of 3,082 ppm on the ninth day of the test. For the deep
well, total target VOC concentration ranged from a maximum of
9,072 ppm at the start of the test to a minimum of 4,073 ppm at
the completion of the test. As expected, the primary
constituent in each wellhead discharge vapor sample was TCE.
The other prominent target VOCs detected were cis-DCE, total
xylenes, TCA, chloroform, ethylbenzene, PCE, and toluene.
Soil gas sampling of the vapor probes was performed before and
after the treatability study to verify that vacuum influence had
been achieved and to determine the effects of vacuum influence
on local soil vapor composition and concentration. TCE was
again the most prominent VOC detected in all probes. The
percent reduction of TCE at the vapor probes ranged from
72 percent at VP1-2 to 55 percent at VP3-1. Similar
concentration decreases were observed for other target VOCs with
the exception of chloroform which remained relatively unchanged.
Air Flow and Contaminant Removal Modeling
Proprietary computer models were used to evaluate air flow and
contaminant removal characteristics of the soil units in the
vadose zone at the Heleva Landfill Site. A description of the
theoretical development of the models has been presented by
Baehr, Hoag, and Marley, 1989. The soil units identified at the
test site—an upper soil unit of soft, sandy silt between the
surface and approximately 20 feet deep, a discontinuous five-
foot-thick sand unit at a depth of between 20 and 25 feet, and
a lower soil unit of stiff silt extending from a depth of
25 feet to below the water table (approximately 50 feet)—were
modeled as a two-layer system with surface and water table
boundaries and an intermediate boundary layer or lens.
Air flow modeling was used to determine the relative intrinsic
permeability tensors of the soil units through which air flow
occurs and to simulate system performance. Calibration of the
2-D, radially symmetric form of the air flow equations with the
1469
-------
16
14
12
10
r
5 10 15 20
WELLHEAD VACUUM (in. HP)
25
5 10 15
WELLHEAD VACUUM (in. Hg)
Figure 6. Shallow vacuum extraction well flow
rate as a function of vacuum pressure in inches
of water.
Figure 7. Deep vacuum extraction well flow
rate as a function of vacuum pressure in inches
of mercury.
10
24 6 8 10
RUN TIME (days)
• TOTAL VOC DISCHARGE
Figure 8. Results of GC/PID chromatographic
analyses of the shallow vacuum extraction well
vapor discharge over the 10-day test period.
0123
RUN TIME (days)
• TOTAL VOC DISCHARGE
Figure 9. GC/PID results for the deep vacuum
extraction well vapor discharge over the four-
day test period.
steady-state physical data obtained during the pilot test
allowed determination of the horizontal (Kr) and vertical (Kv)
intrinsic permeabilities of the
-8
upper soil unit; the calculated
i 1.0 x 10~8 cm2, respectively.
values were 2.29 x 10~a cnr and
Soils displaying an intrinsic air permeability value in this
range are considered to be moderately permeable. In addition,
the model provided an evaluation of the equivalent vertical
intrinsic permeability of the boundary at the soil surface. The
calculated value was 1.0 x 10~8 cm2. The surface boundary
condition is an important parameter that can significantly
influence the achievable radius of vacuum influence, the air
flow pathways, and the vacuum developed at the well. The value
of the permeability of the surface boundary condition calculated
for this test area indicates that the surface is relatively
1470
-------
permeable and that significant air flow to the well from the
atmosphere occurs within the near field of the well. The Kr and
Kv values for the lower soil unit were calculated to be 3.9 x
10~10 cm2 and 1.0 x 10~10 cm2, respectively. Soils displaying an
intrinsic air permeability value in this range are considered to
have a low permeability approaching the limits considered effec-
tive for the application of vapor extraction technology, where
significant secondary porosities do not exist. The K^ value of
the intermediate boundary lens was calculated to be 4.5 x
10~8 cm2. Since the boundary lens appeared discontinuous, it is
important not to lend too great an emphasis on its significance
with respect to projected full-scale system performance.
The calibrated air flow model was verified by utilizing the
model to project system performance under a secondary air flow
rate and comparing the projections with the observed field data.
The model calibration and verification results for the upper
soil unit are shown in Figures 10 and 11, respectively. The
calibrated and verified air flow model was used in the
simulation mode to predict the effective radius of vacuum
influence, the vacuum distribution in the subsurface and the air
flow pathways that would be observed under a variety of system
conditions. The predicted soil vacuum pressure distribution in
the upper soil unit over the range of flows from 7 to 120 cfm is
shown in Figure 12. The operating vacuum of the well for
different flow rates is read from where the curves intersect the
y-axis (at a radial distance of zero feet). It may be observed
that at the maximum air flow rate of 120 cfm, the operating
vacuum is in excess of 0.6 atmospheres, or 18 inches of mercury.
By reducing the design flow to 100 cfm, a more readily operable
vacuum of less than 15 inches of mercury is predicted. Due to
the significant mass of contaminants considered to be
distributed within the upper soil unit, the most cost-effective
and highest practical flow rate is desired. A 100-cfm design
flow rate per well is recommended. The effective radius of
vacuum influence is site-specific and was defined as the limit
of vacuum levels approaching atmospheric conditions. At soil
vacuum pressures approaching atmospheric pressure, it can be
observed from Figure 13 that the radius of influence of the
vacuum extraction well in the upper soil unit is in excess of
50 feet for the simulated air flow rates. An effective radius
of influence of 50 feet at a design flow rate of 100 cfm was
used in the full-scale conceptual design. Similar analysis of
the lower soil unit yielded an effective radius of vacuum
influence of 8 to 10 feet at a recommended flow rate of 7 cfm.
As previously stated, the surface boundary condition can have a
significant influence on the achievable air flow rates, air flow
pathways and on the effective radii of vacuum influence of an
extraction well. A decrease in the permeability of the surface
boundary (i.e., capping) may increase the radius of influence;
however, the increased radius of influence is generally
accompanied by a significant decrease in the air flow rate from
1471
-------
i.oo-r
1.00
£ 0.99
-2.
UJ
§ 0.98-
in
in
go.97
a.
| 0.96
1 0.95 H
o
20 40 60
RADIAL DISTANCE FROM WELL (feet)
— 13cfm • Field Data Point
Figure 10. Air flow model calibration for the
upper soil unit.
*
094
0 20 40 60
RADIAL DISTANCE FROM WELL (feet)
— 7.2 cfm • Field Data Point
Figure 11. Air flow model verification for the
upper soil unit.
the well under the same operating vacuum. Figure 14 presents
plots of the predicted operating vacuum and pressure
distribution for an extraction well in the upper soil unit under
an operating air flow rate of 50 cfm, where the surface boundary
is simulated as being capped. The upper and lower curves
represent the operating conditions for caps having equivalent
vertical intrinsic air permeabilities of 1.0 x 10~ cm2 and
1.0 x 10~12 cm2, respectively. As expected, the plots
demonstrate the significant increase in the operating vacuum
from 0.8 atm (uncapped) to 0.53 atm (1.0 x 10~12 cm2 cap) and the
significant increase in the effective radius of influence from
50 feet (uncapped) to greater than 100 feet (capped) . In
general, spacing extraction wells in excess of 200 feet on
center has the potential to introduce significant reductions in
remediation efficiency due to potential significant variations
in soil properties at this scale and due to potential extended
remediation time periods from lower air flow rates. Based on
the model and cost benefit analysis at this site, capping the
surface is not expected to improve the overall efficiency of the
full-scale conceptual design.
Air injection was also considered as part of the full-scale
design due to the predicted, limited achievable radius of vacuum
influence and the significant levels of contaminants observed in
the lower soil unit. Simulations were performed to predict the
operating pressures and pressure distribution in the lower soil
unit under a range of air injection rates. The predicted
pressure distribution in the lower soil unit over a range of air
injection rates from 20 to 70 cfm is presented in Figure 15.
From the plot, it may be observed that an operating pressure of
up to 2.9 atm is predicted at the well. The plot shows that, in
the region of one atmosphere, an effective radius of vacuum
influence of 12 to 13 feet is achieved. Although the radius of
influence is not substantially increased over the vacuum.
extraction case, the achievable air flow rate and contaminant
removal potential are enhanced. A configuration of wells in the
147;]
-------
LEGEND
A = 7.2 cfm
B «= 13 cfm
C = 20 cfm
0 = 50 cfm
E = 10Ocfm
F = 120dm
20 40 60
RADIAL DISTANCE FROM WELL (feet)
LEGEND
A » 72 cfm
B= 13cfm
C o 20 cfm
0 = 50 cfm
E= 100 cfm
F= 120 cfm
35 45 55 65
RADIAL DISTANCE FROM WELL (feet)
Figure 12. Predicted vacuum levels that would be
observed at the wellhead over the range of
achievable air flow rates in the upper soil unit.
Figure 13. Predicted radii of influence for the
range of achievable air flow rates in the upper
soil unit.
1.00T
0.98-
•P- 0.96
| 0.94
«0°:H
g 0.88
co 0.86
^ 0.84-
a_ 0.82
S °-80
W 0.78
3 0.76
5 0.74
g 072
§ 0.70
0.68
LEGEND
A = 1.0x10''° cm*
1.0x10'" cm7
20
40 60
DISTANCE (feet)
80
too
1.0009
"§• 1.0008
-1.0007
f 1.0006
CO
«u 1.0005
Q 1.0004
UJ
3 1.0003
| 1.0002
o
Z 1.0001
1.0000.
Figure 14. Predicted operating vacuum and pressure
distribution for the upper soil unit with a cap
installed over the surface.
8 10 12 14 16 18 20
DISTANCE (feet)
• 20 cfrn -f 40 cfrn « 70cfm
Figure 15. Predicted pressure distribution and
radii of influence of air pressure in the lower
soil unit over a range of air injection rates.
deep soil unit and flexibility in the design of the manifolding
system will allow reversal in well operation (i.e., extraction
wells may be used as injection wells and vice-versa) , and an
effective radius of influence of 12 to 13 feet at air flow rates
of 7 cfm (extraction) and up to 70 cfm (injection) is
achievable.
A semi-empirical contaminant transport model was used to
evaluate the vadose zone soil units with regard to contaminant
removal characteristics. The contaminant discharge data as
displayed in Figure 8 present a curve which is atypical of the
standard vapor extraction system discharge plot. This type of
curve is generally associated with the misalignment of the vapor
extraction well with respect to the center of mass of the
contaminants within the well's zone of influence. The existence
of a second peak at approximately five days into the test run
1473
-------
most likely represents the lag time for transport of the vapors
from the center of contaminant mass to the extraction well. In
predicting the removal of the contaminants from the upper zone,
the initial four days of data were not utilized since the data
from the second peak forward would be more representative of the
behavior of the full-scale system, and the initial four-day time
frame would represent an insignificant time period in the
prediction of the total time to achieve the soil cleanup goals.
Contaminant discharge from the lower soil unit, shown in
Figure 9, presents a more typical vapor extraction system
discharge plot when the extraction well is placed near the
center of mass of the contaminants within the zone of vacuum
influence of the well.
The contaminant transport model was used to extrapolate a
discharge curve from the field data to estimate the time
required to achieve the soil cleanup goals for specific site
contaminants. Figures 16 and 17 present theoretical graphs of
contaminant removal for the shallow soil unit at a design air
flow rate of 100 cfm, utilizing an initial mass of contaminants
within the radius of influence of the extraction well
corresponding to the highest concentration of soil contaminants
observed at the site. The model predicted that the time to
achieve the cleanup criteria at an extraction well for TCE, DCE,
and methylene chloride would be approximately 120, 40, and
30 days, respectively. Due to its lower volatility and mole
fraction, PCE is predicted to be removed more slowly and take
approximately 160 days to achieve the cleanup goal. In the
lower soil unit, the projected remediation times for the maximum
contaminant concentrations detected during the RI investigation
are 60 days for DCE, 40 days for methylene chloride, and up to
five years for TCE and PCE. Vacuum extraction is generally not
as effective for extracting highly water soluble VOCs such as
acetone and 2-butanone. It is expected that unless acetone and
2-butanone are present as a free phase, additional measures such
as groundwater extraction and treatment techniques may be
required to remove these contaminants from the soil.
Preliminary Conceptual Design
The preliminary conceptual design parameters for a full-scale
vacuum extraction system at the Heleva Landfill Site are
summarized in Table 4. Based on the information developed for
the field investigation, the preliminary design is presented
under the assumption that the soil properties and contaminant
composition and distribution are relatively consistent through-
out the areas of the Heleva Landfill Site designated for
remediation. It is more realistic, however, to assume that
within the designated remediation areas, localized high and low
levels of contamination and varying soil conditions will exist.
Where these conditions are observed in the field, it is impor-
tant to be flexible and to consider diverging from the concep-
tual design with particular respect to the spacing of the wells,
147/1
-------
a
<
cc
en
D
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
20 40 60 80 100 120 140 160 180 200
TIME (days)
Figure 16. Theoretical graph of time vs. removal
of TCE, DCE, PCE, and methylene chloride at a
design air flow rate of 100 cfm.
20 40 60 80 100 120 140 160 180 200
TIME (days)
Figure 17. Theoretical graph of time vs. total
VOC removal at a design air flow rate of
100 cfm.
the use of air injection points, and the prediction for time to
achieve the specified cleanup goals for these localized areas.
From the air flow analysis, utilization of air injection wells
within the deeper soil units at the Heleva Landfill Site would
tend to increase the effective radius of influence of the
wellpoints and enhance VOC removal through the higher air flow
rates achievable within the soil system. However, preliminary
estimates indicate that the relative costs associated with the
widespread utilization of air injection could be significant.
Further, the application of air injection would also transfer
the deep soil unit contaminants into the capture zone of the
shallow soil unit vapor extraction wells and therefore may
prolong the period of operation of the shallow wells. Assuming
field observations made during the full-scale installation would
demonstrate localized variations in soil properties and
contaminant composition and distribution, the utilization of air
injection points would only be recommended for the "hot spots"
of the deeper soil unit.
Based on the results of the pilot study, air control equipment
would be required for treatment of the vapor discharge from the
vacuum extraction system. During the treatability study, vapor
phase carbon was found to be effective in providing air emission
controls for all of the VOCs identified during the test. The
amount of carbon required for the full-scale systems would be
directly related to the amount of VOCs to be removed by the
system. A rough estimate of the amount of carbon required can
be based on a carbon adsorption capacity of 10 percent by
weight. The potential magnitude of contamination at the Heleva
Landfill Site warranted the consideration of onsite regeneration
techniques as opposed to offsite regeneration and/or disposal.
The estimated costs to install and operate a full-scale vacuum
extraction system for the shallow and deep soils is summarized
in Table 5. This estimate was prepared assuming an intermediate
1475
-------
TABLE 4
PRELIMINARY DESIGN PARAMETERS
Parameter Shallow Soil Deep Soil
Radius of Influence 50 feet 8 to 10 feet
Air Flow Rate 100 cfm 7 cfm
Vacuum Pressure 1 5 in. Hg 15 in. Hg
Remediation Time 1 year 5 years
Cost* $17/cubic yard $88/cubic yard
* Estimated cost for remediating soils to 1 mg/kg TCE
range soil cleanup goal of 1,000 /xg/kg for TCE with a
corresponding volume of 57,870 cubic yards of shallow soil and
48,520 cubic yards of deep soil requiring remediation. The
shallow system would include a total of 11 wells on 100-foot
centers with a 100-cfm pump at each well. The deep system would
need a total of 156 wells on 20-foot centers manifolded to
11 vacuum pumps with 100-cfm capacity. The unit costs for
treating shallow and deep soils are $17 and $88 per cubic yard,
respectively.
Analysis of Remedial Alternatives
Vacuum extraction and other appropriate technologies were
developed into a series of remedial alternatives for the site
that ranged from no action to complete treatment of all
contaminated soils. A major factor that had to be considered
was how the remedial options would work along with a recently
completed RCRA cap located over the landfill area and
approximately 50 percent of the spill areas. Due to the reduced
contaminant migration potential under the cap, the higher
cleanup goals presented in Table 3 could be applied to
contaminated soils under the cap, requiring less treatment to be
performed. Another consideration was saturated soils above
bedrock that retain approximately 40 percent of the VOC
contamination and essentially all of the acetone and 2-butanone
detected at the site. Vacuum extraction cannot draw air through
saturated soils and, therefore without dewatering, would appear
to be ineffective for remediating this contaminated area.
From the range of remedial alternatives, a remedy that includes
extending the existing landfill cap over the contamination
source areas, dewatering the saturated soils above bedrock, and
using vacuum extraction to remediate the "hot spots" of
contaminated soil that exceed the soil cleanup goals when a
synthetic membrane cap is in place was recommended. Dewatering
the saturated soils would be evaluated through pilot testing
during the Remedial Design phase before determining the
conceptual design. The present worth cost of this alternative
is estimated to be two million dollars, much lower than similar
alternatives without a cap extension that ranged from 22.6 to
1476
-------
TABLE 5
COST ESTIMATE FOR VACUUM EXTRACTION SYSTEM
Item
Shallow Soil
Deep Soil
Capital and Equipment Costs
Vacuum Extraction System
Vacuum Well Installation
Well Manifolding
Vacuum Equipment
Equipment Staging Areas
Subtotal Capital Costs
Air Control Equipment
Carbon with Offsite Regeneration
Canisters
Subtotal Air Controls
Subtotal Capital and Equipment
Contingency at 20%
Total Capital and Equipment
Operation and Maintenance
Monthly Costs
Electric
Operator/Maintenance
Analytical
Reporting/Oversight
Subtotal Annual O&M
Contingency at 20%
Total Annual O&M
Demobilization
Allowance
Total Demobilization
NET PRESENT VALUE
assuming 5% discount rate, 1 year of O&M for
shallow and 2 years of O&M for deep soil
Estimated Cost Per Cubic Yard
$94,230
$60,029
$132,480
$100.000
$386,739
$192,500
$40.000
$232,500
$619,239
$123.848
$743,086
$5,569
$7,900
$3,000
$1.300
$17,769
$213,225
$41.645
$255,870
$50.000
$50,000
$991,613
$17
$1,836,207
$1,001,490
$255,280
$200,000
$3,292,977
$179,900
$40.000
$219,900
$3,512,877
$702.575
$4,215,453
$5,528
$7,900
$3,000
$1,300
$17,728
$212,736
$42.547
$255,283
$100,000
$100,000
$4,254,091
$88
1477
-------
39 million dollars. The recommended alternative is expected to
be completed within two years, as compared to five years for an
alternative based on dewatering and vacuum extraction treatment
without a cap extension.
A phased approach was recommended for implementing a combination
vacuum extraction and dewatering system. During the initial
phase, the shallow vacuum extraction system and several
dewatering wells would be installed. In a later phase, the
remainder of the dewatering wells, the deep vacuum extraction
system, and the cap extension would be installed. The reasons
for a phased approach are several. First, the installation and
operation of the shallow system would allow for identification
of the more highly contaminated areas and for any necessary
debugging of the full-scale system operating parameters.
Second, the shallow soils are projected to achieve the cleanup
criteria within one year, whereas the deep soils and soils above
bedrock may require up to five years, hence the overall project
may be extended by only one year while valuable operating
knowledge is gained. Third, the operating equipment used for
both the shallow and the deep systems are similar and savings in
capital costs could be achieved by utilizing the same equipment
for the shallow and the deep systems. Fourth, the dewatering
system would require a more detailed subsurface investigation
and pilot-scale testing before the full-scale design is
performed. Once the dewatering system is functioning properly,
vacuum extraction of the saturated soil zone could be initiated.
SUMMARY
A field investigation of the Heleva Landfill Site delineated two
distinct solvent spill areas contaminated with chlorinated
hydrocarbons and ketones. The use of quick-turnaround analyses
and statistical correlation of data between borings (kriging)
allowed the field team to focus the placement of borings in the
potential spill areas, reducing the total number of borings
required to define the extent of contamination. Soil cleanup
goals were developed based on a combination of modeling
techniques to predict the concentration of contaminants in soil
that would correspond to acceptable groundwater quality beneath
the site. Remedial technologies capable of achieving the soil
cleanup goals were evaluated. Due to the depth of contamination
and problems associated with controlling exposure to VOCs,
in-situ vacuum extraction and surface capping were considered to
be the most applicable remedial technologies for this site.
A systematic evaluation of the parameters involved in operating
a vacuum extraction system was conducted by performing a pilot-
scale field study and utilizing air flow and contaminant
transport models to evaluate the results. It was determined
that the vacuum extraction process could successfully remove
VOCs from the sandy silt soil matrix in the shallow soil (from
ground surface to 25 feet) but VOCs were more difficult to
1478
-------
remove from the deep stiff silt soil strata (25 feet to 50 feet
below ground surface). The vacuum extraction wells in the
shallow strata would be capable of achieving an effective radius
of influence of approximately 50 feet at an optimal vapor
extraction rate of 100 cfm and a corresponding wellhead vacuum
pressure of 15 inches of mercury. Vacuum extraction wells in
the deep strata would be capable of achieving an effective
radius of influence of about 10 feet at an optimal vapor
extraction rate of 7 cfm and a corresponding wellhead vacuum of
15 inches of mercury. It is expected that if the saturated soil
above bedrock (50 to 70 feet below ground surface) were
dewatered, the air flow and chemical removal characteristics
would be similar to the lower vadose zone soils.
ACKNOWLEDGEMENT
The authors wish to sincerely thank Ms. Carol A. Royal for her
enthusiasm and effort in preparing and reviewing the manuscript.
REFERENCES
l. Baehr, A.L., G.E. Hoag and M.C. Marley, "Removing Volatile
Contaminants from the Unsaturated Zone by Inducing
Advective Air-Phase Transport," Journal of Contaminant
Hydrology, 4, pp 1-26, 1989.
2. Shacklette, H.T. and J.G. Boerngen, Element Concentrations
in Soils and Other Surficial Materials of the Conterminous
United States, U.S. Geological Survey Professional Paper
1270, U.S. Government Printing Office, 1984.
3. Summers, K.S., Gherini and C. Chen, Methodology to Evaluate
the Potential for Groundwater Contamination from Geothermal
Fluid Release, EPA-600/7-80-117, Tetra Tech, 1980.
4. U.S. Environmental Protection Agency, The Hydrologic
Evaluation of Landfill Performance (HELP) Model, Volume I.
User's Guide for Version I. EPA 530-SW-84-009, Office of
Solid Waste and Emergency Response, 1984.
5. U.S. Environmental Protection Agency, Determining Soil
Response Action Levels Based on Potential Contaminant
Migration to Groundwater: A Compendium of Examples,
EPA 540/2-89/057, Office of Emergency and Remedial
Response, 1989.
6. Woodward-Clyde Consultants, Multimedia Exposure Assessment
Model for Evaluating the Land Disposal of Hazardous Wastes,
Volume I, Environmental Research Laboratory, Office of
Research and Development, U.S. Environmental Protection
Agency, 1988.
OU.S. GOVERNMENT PRINTING OFFICE;! 991 .5 n i . 1 8 7/2 5 6 1 n
1479
------- |