United States
Environmental Protection
Agency
     Office of Air Quality
     Planning and Standards
     Research Triangle Park, NC 27711
EPA-450/l-89-002a
August 1990
AIR/SUPERFUND
vvEPA
AIR / SUPERFUND
NATIONAL TECHNICAL
GUIDANCE STUDY SERIES
            VOLUME II - ESTIMATION OF
            BASELINE AIR EMISSIONS AT
            SUPERFUND SITES
  * This document revises earlier edition, EPA-450/1-89-002.

-------
       AIR/SUPERFUND NATIONAL TECHNICAL
              GUIDANCE STUDY SERIES
                    Report ASF-2a
VOLUME H - ESTIMATION OF BASELINE AIR EMISSIONS
           AT SUPERFUND SITES (REVISED)
                     Prepared for:

        Ms. Anne Pope, Work Assignment Manager
          U.S. Environmental Protection Agency
        Office of Air Quality Planning and Standards
      Research Triangle Park, North Carolina   27711
                 Revised: August 1990

-------
                  EPA-450/1-89-002a
                     August 1990
AIR/SUPERFUND NATIONAL TECHNICAL GUIDANCE
STUDY SERIES.  VOLUME II - ESTIMATION OF

BASELINE AIR EMISSIONS AT SUPERFUND SITES

* This document revises earlier edition, EPA-450/1-89-002.

                        By
                     Bart Eklund
                        and
                   Charles Schmidt
                 Radian Corporation
                    Austin, Texas
               Contract Number 68-02-4392
        U. S. ENVIRONMENTAL PROTECTION AGENCY
                   Office Of Air and Radiation
            Office Of Air Quality Planning And Standards
            Research Triangle Park, North Carolina 27711

-------
                             PREFACE


    This report revises and expands an earlier report, Procedures
For  Conducting Air  Pathway Analyses  For  Superfund  Activities.
Vtp^ugie II,  Estimation Of Baseline Air Emissions At Superfund Sites f
EPA 450/1-89-002.  It  is  one in  a series of reports that provide
guidance on conducting air pathway analysis  at Superfund hazardous
waste  sites.    It was developed  for the  Office  of  Air Quality
Planning and Standards in cooperation with the Office of Emergency
and Remedial Response  (Superfund).

    This  report  have  been reviewed by  the National  Technical
Guidance  Study  Technical  Advisory Committee,  State  agencies,
various groups within  the U.S. Environmental Protection Agency,
and the private sector.  It provides technical guidance for use by
a diverse  audience including  EPA Air and  Superfund Regional and
Headquarters staff, State Air Superfund program staff, Federal and
State remedial and removal contractors, and potentially responsible
parties in analyzing air pathways at hazardous waste sites.  This
report  is  written  to serve  the  needs of  individuals  having
different  levels  of  scientific  training  and   experience  in
designing, conducting and reviewing air pathway analyses.  Remedial
Project Managers, On  Scene Coordinators,   and  the Regional Air
program  staff, supported  by  the technical  expertise of  their
contractors, will use this volume when developing baseline emission
estimates for  undisturbed hazardous waste sites.

    Because assumptions  and judgments are  required in many parts
of an air pathway  analysis, an analysis requires a strong technical
background  in  air  emission   estimation  methods,  measurements,
modeling and monitoring.   Air  pathway analyses cannot be reduced
to  simple  "cookbook11  procedures.   Therefore,  this  volume  is
designed to be flexible, allowing the use  of  professional judgment.
The procedures presented in this report are  intended  solely for
technical guidance.  They are not  intended,  nor can they be relied
upon, to create rights substantive or procedural,  enforceable by
any party in litigation with the United States.

    This  edition  of Volume II will be  periodically  updated to
incorporate  new  data  and  information  on  air  pathway  analysis
procedures.  The Agency reserves the right to act at variance with
these  procedures  and  to  change  them  as new  information  and
technical tools become available  on air  pathway  analyses without
formal public  notice.  The Agency will,  however,  attempt to make
any revised or updated  manual available to those who currently have
a copy through the registration form included with the report.

    Copies of this report are available as supplies permit, through
the Library  Services  Office (MD-35),  U.S. EPA, MD-35,  Research
Triangle Park, NC   27711 or from the National Technical Informa-
tion Services, 5285 Port Royal Road, Springfield,  VA   22161.

-------
This report has been reviewed by the Office of Air Quality Planning
and Standards, U. S. Environmental Protection Agency, and has been
approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
                 EPA 450/l-89-002a
                           n

-------
                                   CONTENTS

                                                                           Page

Disclaimer       	  i
Tables           	vii
Figures          	ix
Glossary of  Frequently Used Terms and Acronyms	xi
Acknowledgements   	  xv


1.0   INTRODUCTION	1
      1.1  The  Problem	 1
      1.2  The  Objective	3
      1.3  Manual Organization 	 3
      1.4  Recommended Uses of this Manual 	 4
      1.5  Document Organization 	 7


2.0  AIR EMISSIONS FROM  HAZARDOUS WASTE SITES 	 9
     2.1  General Description 	  12
          2.1.1  Landfills	12
          2.1.2  Lagoons	14
          2.1.3  Equivalent Units 	  15

     2.2  Routes of Exposure  	  16
          2.2.1  Key Parameters and Critical Factors Affecting  Emissions
                 from Contaminated Soils (Landfills and Lagoons)   ....  19
          2.2.2  Key Parameters Affecting Emissions from Lagoons   ....  19

     2.3  Magnitude of Air Emissions	". -	21

     2.4  Emissions of Potential Interest at NPL Sites	21

     2.5  Summary of Potential  Receptors	23


3.0  PROTOCOL FOR BASELINE EMISSION ESTIMATES 	  27
     3.1  Protocol Steps for Developing BEEs  	  27
          3.1.1  Define the APA Objective	  27
          3.1.2  Site Scoping	34
          3.1.3  Evaluate Available Site Data	34
          3.1.4  Design and Conduct the Site Screening Study	37
          3.1.5  Design and Conduct the In-Depth Site Characterization   .  47

     3.2  Use of the BEEs in the Mitigation Process 	  50
                                     m

-------
                             CONTENTS (Continued)
                                                                           Page

4.0  AIR EMISSION MEASUREMENT TECHNIQUES	53
     4,1  General Considerations  .	53
          4.1.1  Comparison of Techniques 	  53
          4.1.2  Uses of Emission Estimates	59
          4.1.3  Determining Worst-Case Conditions  	  61

     4.2  Direct Emission Measurement Technologies	62
          4.2.1  Head Space Samplers  (Screening Techniques)	64
          4.2.2  Head Space Analysis  of Bottled Samples	66
          4.2.3  Emission Isolation Flux Chamber (In-Depth Technique)  .  .  69
          4.2.4  Portable Wind Tunnels (In-Depth Technique)	73
          4.2.5  Soil Vapor (Ground)  Probes (In-Depth Technique)   ....  78
          4.2.6  Soil Vapor Monitoring Wells (In-Depth Technique)  ....  82
          4.2.7  Downhole Emissions Flux Chamber (In-Depth Technique)  .  .  85
          4.2.8  Vent Sampling In-Depth Technique)   	  88

     4.3  Indirect Emission Measurement Techniques	89
          4.3.1  Upwind/Downwind (Screening Technique)	91
          4.3.2  Mass Balance (Screening Technique)  	  93
          4.3.3  Real-Time Instrument Survey (Screening Technique)   ...  94
          4.3.4  Concentration-Profile (In-Depth Technique) .	96
          4«305  Transect (In-Depth Technique)   .	99
          4.3.6  Boundary Layer Emission Monitoring	104

     4.4  Air Monitoring  Technqiues	105
          4.4.1  Ambient  Air Monitoring	106
          4.4.2  Indoor Air Monitoring	107

     4.5  Emissions  (Predictive) Modeling 	  .  	  Ill
          4o5.1  Summary  and Comparison  of Emission  Models  .......  113
          4.5.2  Emission Models for  Closed  Landfills  Without  Internal
                 Gas Generation	126
          4.5.3  Emission Models for  Closed  Landfills  With Internal Gas
                 Generation	*	137
     >    4.5.4  Emission Models for  Open Landfills  	  144
          4.5.5  Emission Models for  Landtreatsnent	151
          4.5.6  Leaks  and Spills on  Soil  	  154
          4.5.7  Emission Models for  Fugitive Dust	155
          4.5.8  Additional  Models  	  161
          4.5.9  Non-Aerated Lagoons	!  !  161
          4.5.10  Aerated  Lagoons	! .  !  !  167
          4.5.11  Sources  of  Model Input Data	! ....  168
                                     iv

-------
                             CONTENTS (Continued)
                                                                           Page

5.0  CASE STUDIES	171
     5.1  Case Study 1:  Petroleum Waste Landfill/Lagoon   	  171
          5.1.1  Site History	171
          5.1.2  Objectives	172
          5.1.3  Scoping	175
          5.1.4  Overview of Fieldwork for Site Characterization   ....  175
          5.1.5  Undisturbed Emissions Survey 	  177
          5.1.6  Disturbed Emissions Survey 	  182
          5.1.7  Development of BEEs	189
          5.1.8  Summary	192

     5.2  Case Study 2:  Bruin Lagoon	192
          5.2.1  Site History	192
          5.2.2  Objectives	196
          5.2.3  Scoping	196
          5.2.4  Overview of Fieldwork for Site Characterization   ....  196
          5.2.5  Undisturbed Emissions Survey 	  198
          5.2.6  Disturbed Emissions Survey 	  198
          5.2.7  Development of BEEs	202
          5.2.8  Summary	203

     5.3  Case Study 3:  Lowry Landfill	  203
          5.3.1  Site History	203
          5.3.2  Objectives	v	206
          5.3.3  Scoping	206
          5.3.4  Overview of Fieldwork for Site Characterization   ....  206
          5.3.5  Undisturbed Emissions Survey 	  207
          5.3.6  Disturbed Emissions Survey 	  207
          5.3.7  Development of BEEs	211
          5.3.8  Summary	211

     5.4  Case Study 4:  Western Processing Landfill   	  212
          5.4.1  Site History	212
          5.4.2  Objectives	213
          5.4.3  Overview of Fieldwork for Site Characterization   ....  213
          5.4.4  Scoping	216
          5.4.5  Undisturbed Emissions Survey 	  216
          5.4.6  Disturbed Emissions Survey 	  216
          5.4.7  Development of BEEs	218
          5.4.8  Summary	218

     5.5  Case Study 5:  Outboard Marine Corp.  Lagoon/Landfill  	  219
          5.5.1  Site History	219
          5.5.2  Objectives	221
          5.5.3  Scoping	221
          5.5.4  Overview of Fieldwork for Site Characterization   ....  222
          5.5.5  Undisturbed Emissions Survey 	  223
          5.5.6  Disturbed Emissions Survey 	  223
          5.5.7  Development of BEEs	224
          5.5.8  Summary	224

-------
                             CONTENTS (Continued)
                                                                           Page
6.0  REFERENCES	  191
     APPENDICES
          A -  Annotated Bibliography
          B -  Chemical  and Physical  Properties Affecting Baseline.Emission
               Estimates
          C -  Estimation of an  Overall  Source Emission Rate Using Emission
               Flux Measurements
          D -  Description of Sources of Model  Input Data
          E -  Description of Remote  Sensing  Techniques
          F -  Physical  and Chemical  Property Data
          G -  Physical  and Chemical  Property Data for 25 Compounds of
               Potential  Concern and  for Compound  Classes

-------
                                    TABLES
                                                                           Page
 1   Important Parameters in Determining Air Emissions and Their
     Qualitative Effects on Baseline Emission Estimates (BEEs)  	   20
 2   Summary-of Average Baseline Emissions for Various Emission Sources  .   22
 3   Most Frequently Reported Substances at 546 National Priority
     List Sites	24
 4   Toxic Pollutants Most Commonly Addressed by State and Local Agencies   25
 5   Examples of APA Objectives for BEEs  	   31
 6   Potential Air Contaminants by Generic Type of Contaminant	33
 7   Activities for Developing BEEs:  Evaluate Available Site Data   ...   35
 8   Factors to Consider in Selecting an Indicator Species for Study   .  .   40
 9   Examples of Broad-Band, Class, and Indicator Species 	   42
10   Screening Technologies Applicable to Site Screening APA for
     Landfills and Lagoons  	   43
11   Types of In-Depth Methods for Estimating Emissions 	   48
12   Comparison of Screening Techniques 	   54
13   Relative Ranking for Screening Techniques  	   55
14   Comparison of In-Depth Techniques  	   56
15   Relative Ranking for In-Depth Technqiues 	   57
16   Summary and Comparison of Emission Models  	  115
17   Data Sources for Selected Model Parameters 	  170
18   APA Activities Conducted at the Site	174
19   Summary of Screening Measurements of Undisturbed Waste 	  180
                                     vii

-------
                              TABLES (Continued)
                                                                           Page

20   Case Study 1:  Summary of Undisturbed Site Emissions Data	181
21   Downwind/Border Monitoring Results 	  „  .  187
22   Summary of Downhole Emissions Data 	  188
23   APA Activities Conducted at the Case Study 2 Site	193
24   APA Activities Conducted at the Case Study 3 Site  .........  204

25   Summary of Air Monitoring at Lowry Landfill   	  209
26   Maximum and Average Concentrations in Soil  for Selected Contaminants  217
                                    viii

-------
                                    FIGURES
                                                                           Page
 1   Superfund  Flow Chart, Noting Elements Where This Manual May Apply   .  .  5
 2   Conceptual  schematic of a landfill	10
 3   Conceptual  schematic of a lagoon  	   11
 4   Conceptual  schematic showing air contaminant pathways from a  landfill  17
 5   Conceptual  schematic showing air contaminant migration pathways
     from a lagoon	18
 6   Flowchart  of activities for developing screening and in-depth
     baseline emission estimates  	   28
 7   Checklist  of factors affecting air emissions per unit  	   39
 8   Use. of the  BEEs data in site mitigation	51
 9   Schematic  diagram of a soil core sample sleeve 	   68
10   A cutaway  diagram of the emission isolation flux chamber and
     support equipment   	   70
11   A cutaway diagram of the surface emission isolation flux chamber
     and support equipment for liquid surfaces  	   71
12   Illustration of MRI wind tunnel  	   75
13   Schematic of portable wind tunnel  	   77
14   Schematic diagram of a simple ground probe 	   79
15   Ground probe design with minimal internal volume 	   80
16   Vapor monitoring well constructions  	  83
17   Schematic diagram of the downhole emissions flux chamber 	  86
18   Real-time instrument survey  	   95
19   Mast sample collection system for C-P sampling 	   97
20   Example of transect technique sampling 	  100
21   Pick's correction factor,  Fv, plotted against dequivalent vapor
     pressure, Ce	147
                                      1x

-------
                              FIGURES  (Continued)
                                                                           Page
22   Location of suspected disposal  area  .	173
23   Location of waste soil coreholes  .....  	  185
24   Generalized flow regime of perched zone and bedrock aquifer   ....  195
25   Monitor well and soil boring locations  at the Bruin Lagoon Site   .  .  200
26   Western Processing site	214
27   Map of Case Study 5 site	220

-------
                 GLOSSARY  OF  FREQUENTLY USED TERMS AND ACRONYMS
 ACGIH


 Adsorption
 Air
 Monitoring
 APA


 BEEs



 Calibration
CERCLA
Co-disposal
Site

Detection
Limit
Direct
Emissions
Measurement
 American Conference  of  Governmental  Industrial  Hygienists,
 6500 Glenway Ave., Building D-5, Cincinnati,  OH  45211.

 A physical  process in which molecules of gas, dissolved
 substances, or liquids  adhere in an  extremely thin  layer to the
 surfaces of solid bodies with which  they are  in contact.

 A gas phase sampling technique where ambient  air is  sampled.
 It can be used to develop emission rate estimates and  is
 similar to  indirect emission measurement except measurements
 usually are taken at greater distances from the waste  site.

 Air Pathway Analyses.  APA are designed to assess the  potential
 for air emissions from a hazardous waste site.

 Baseline Emission Estimates.  These are estimates of baseline
 emission rates from a hazardous waste site in its undisturbed
 conditions.

 Establishment of a relationship between the response of  a
 measurement system obtained by introducing various calibration
 standards into the system.  The calibration levels should
 bracket the range of levels for which actual measurements  are
 to be made.

 Comprehensive Environmental Response, Compensation and
 Liability Act of 1980.  Modified by SARA in 1986.  The Acts
 created a special tax that goes into a trust fund, commonly
 known as Superfund,  to investigate and clean up  abandoned  or
 uncontrolled hazardous waste sites.

 A waste site that has received and mixed municipal and
 hazardous wastes.

 The minimum quantity of a compound which yields  a "measureable
 response."  Measurable response has many statistical
 definitions.  Be careful to differentiate "instrumental
 detection limit," which refers to the minimum quantity of
 material introducible into a measurement system that can be
 detected, from "method detection limit," which means the
 minimum concentration of a compound in a sample which, when put
 through the entire sampling and analysis process, can be
 detected.

A measurement made directly on or above the waste to determine
 the emission rate of volatile species from a liquid or solid
 surface.

-------
 Disturbed      Changes  in a hazardous waste site as remediation  takes place
 Condition      that  usually involve increasing the emission rate of volatile
                species  and particulate matter.

 Emissions      The total of substances discharged into the air from a discrete
                source.

 EPA            U.S.  Environmental Protection Agency.

 FS             Feasibility Study.  Analysis and selection of alternative
                remedial actions for hazardous waste sites.

 Fugitive  Dust   Atmospheric dust arising from disturbance of granular  matter
                exposed to the air; called "fugitive" because it  is  not
                released to the atmosphere in a confined flow stream.
 Hazardous
 In-depth
 Technologies

 Indicator
 Species

 Indirect
 Emissions
 Measurement
Lagoon
Landfill
mg/m3
NIOSH
NPL
 Those wastes that are regulated or "listed" under RCRA  (40
 CFR Part 261} or wastes that are ignitable, corrosive,
 reactive, or toxic.
 Very detailed methods for measuring emissions.
 technologies produce detailed,  reliable data.
These
 Species found in hazardous waste that can be used to represent
 a group of species in determining emissions from a site.

 A gas phase sampling technology that measures ambient air
 concentrations at short distances down-wind of a hazardous
 waste site.  Data are collected to satisfy specific needs of
 specialized models used to estimate air emissions.

 In this manual,  lagoon encompasses surface impoundments or
 impoundments designed to hold  liquid wastes or wastes
 containing free  liquids.

 For purposes of  this manual, a landfill  is a facility,
 usually an excavated pit,  into which wastes are placed for
 permanent disposal.

 Milligrams per cubic meter.  This is a measure of mass per
 unit volume.  The units mg/m3  are commonly used to describe
 concentrations of particulates,  dusts, fumes, and mists.

 National  Institute for Occupation Safety and Health, Centers
 for Disease Control,  Public Health Service, U.S Department of
Health and Human Services.

National  Priorities  List.  A list of waste sites for which EPA
has assessed the relative  threat of site contamination on soil,
air,  surface water,  ground water,  and the population at risk.
Site listing is  found under CERCLA (Section 105) and is updated
three times a year.
                                     xii

-------
OSHA
 Occupational Safety and Health Administration, U.S.  Department
 of  Labor.
OVA

Particulate
Matter

PEL
ppb

ppm

Probe
Quality
Assurance

Quality
Control

RI
RRM


Sampling




SARA
Screening
Technologies

Undisturbed
Condition
VOCs
Organic vapor analyzer.

Airborne solid or liquid matter.
OSHA  permissible exposure limit, expressed as ppm or mg/m3 of
substance in air.

Parts per billion.

Parts per million.

A tube used for gas phase concentration sampling or for
measuring pressures at a distance from the actual collection
or measuring apparatus.

A system of activities designed to assure that the quality
control system is performing adequately.

A system of specific efforts designed to test and control the
quality of data obtained.

Remedial Investigation.  Field investigations of hazardous
waste sites to determine pathways and nature and extent of
contamination.

Remedial Project Manager, equivalent to a site manager at
non-NPL sites.

The process of withdrawing or isolating a fractional part of
the whole.  In air or gas analysis,  it is the separation or a
portion of an ambient atmosphere with or without the
simultaneous isolation of selected components.

Superfund Amendments and Reauthorization Act.  Modifications
of CERCLA enacted on October 17, 1986.

Quick and simple methods for estimating baseline emissions.
The condition in which a hazardous waste site is discovered
or may be left if a no-action  remedial  alternative is
selected.

Volatile organic compounds.  An  organic compound (containing
carbon) that evaporates (volatilizes)  readily at room
temperature.
                                    xiii

-------
xlv

-------
                               ACKNOWLEDGEMENTS


     This manual was prepared for the U.S. Environmental Protection Agency  by

Radian Corporation.  Mr. Leigh Hayes (project manager) and Ms. Susan  Fernandes

(contract manager) managed the project.  Dr. Charles E. Schmidt served  as the
original project director and author of several sections.  Other contributors
included Mr. John Clark, Mr. Mark Galloway, Ms. Susan Penner, and Mr. Bart
Eklund.  Revisions to the document were made by Mr. Bart Eklund and Mr.  Doug
Orr.


     Mr. Joe Padgett and his staff at the Office of Air Quality Planning and

Standards (OAQPS), U.S. Environmental Protection Agency, provided overall
program direction.  Ms. Margaret McDonough and Mr. Tom D'Avanzo, of the  U.S.
EPA Air Management Division, Region I, along with Ms. Anne Pope of EPA-OAQPS
directed efforts on this particular manual.


     Other support was provided by the program's Technical Advisory Committee,
the members of which include:

          Mr. Joe Padgett, U.S. EPA OAQPS
          Mr. Jim Durham, U.S. EPA OAQPS
          Ms. Anne Pope, U.S. EPA OAQPS
          Ms. Donna Abrams, U.S. EPA Region III
          Ms. Grace Musumeci, U.S. EPA Region III
          Mr. Norm Huey, U.S. EPA Region VIII
          Mr. Mark Hansen, U.S. EPA Region VI
          Mr. Jim Southerland, U.S. EPA OAQPS
          Mr. Joe Touma, U.S. EPA OAQPS
          Mr. Wayne Kaiser, U.S. EPA Region VII
          Mr. Tom Pritchett, U.S. EPA Emergency Response Team
          Mr. David Dunbar, PEI Associates, Inc.
          Mr. Bart Eklund, Radian Corporation


     This program also received support from the Regional Air Superfund

Program and its participants.
                                      xv

-------
                                  SECTION 1
                                 INTRODUCTION
1.1  BACKGROUND
     The United States Environmental Protection Agency (EPA) is responsible
for the assessment and cleanup of the National Priority List (NPL) sites under
CERCLA and SARA.  EPA's Remedial Program Managers (RPMs) are required to
assess the potential for air emissions and air quality impacts caused by NPL
sites prior to and during cleanup.  To date, no standard approach for
assessing the air pathway at NPL or other hazardous waste sites has been
available.  As a result, performing air pathway analyses (APA) has been less
straightforward than evaluating other pathways such as the impacts on ground
water or surface water quality.  This manual assists RPMs in determining if an
uncontrolled site has the potential for significant air emissions and, if so,
how to characterize the baseline air emissions potential from the site.

     This volume is one in a series of manuals prepared for EPA to assist
its RPMs in the assessment of the air contaminant pathway and developing input
data for risk assessment.  Volume I (1) of the series provides generalized
guidance for addressing air issues throughout the overall  Superfund process.
This manual (Volume II) provides guidance on developing baseline emission
estimates for hazardous waste sites.  Baseline emission estimates (BEEs) are
defined as emission rates estimated for a site in its undisturbed state.
Volume III (2) provides guidance on estimating emissions from cleanup
activities, and Volume IV (3) provides guidance on ambient air monitoring and
on dispersion modeling.  Together these four manuals provide a complete
treatment of air issues for superfund applications.

-------
     CERCLA and SARA mandate the characterization of all  contaminant migration
pathways from the waste or hazardous  material  to the environment and
evaluation-of the resulting environmental  impacts.   However,  air pathway
analyses are often overlooked because many sites have little  or no perceptible
air emissions in their baseline or undisturbed state.  Even low level
emissions, however, may be significant if  toxic or carcinogenic compounds are
present.  Also, emissions during clean-up  may  be much higher  than baseline
emissions.  Emissions of potential  concern include volatile and semi-volatile
organics, acid gases, particulate matter,  and  toxics associated with windblown
particulate matter such as metals,  PCBs and dioxins.

     A remedial investigation is typically necessary to provide data on air
emissions from the site.  These emission can be measured  directly, or
estimated indirectly from chemical  and physical  data collected during the RI
and used as inputs to predictive models.  Remedial  investigations (RIs) often
include ambient air monitoring to assess baseline air quality impacts from the
site, but measurements of emission rates or soil-gas concentrations are less
widely employed.  An introduction to  these techniques is  a major emphasis of
this manual.  Emission rate or soil-gas data can be useful  for:  1) identifying
"hot spots" e.g. areas of higher than average  waste content or pockets of
subsurface gases, 2) serving as model  inputs (source terms) to estimate
ambient air concentrations under meteorological  conditions other than those
encountered during the RI, and 3) estimating emissions during remediation.
For this last use, the air emissions  investigation  during the RI stage would
include emission measurements of both the  undisturbed wastes  and the exposed
or disturbed wastes.

     While not strictly part of baseline emission estimates,  measurements of
emissions from exposed or disturbed wastes can generally  be performed during
the RI using the same techniques presented in  this  manual  for performing
baseline emission measurements.  These data along with the BEEs can be used in
the procedures outlined in Volume III of this  series to help  evaluate
remediation options, design an engineering approach to the site mitigation,
and determine whether air emission control technologies or an air monitoring
program may be necessary as part of the remedial  alternatives.

-------
1.2  OBJECTIVES

     The overall objective of this manual is to assist RPMs or site managers
in assessing the impacts on air quality from undisturbed sites.  Specifically
the manual  is  intended to:

           Present a protocol for selecting the appropriate level of effort  to
           characterize baseline air emissions.

           Assist site managers in designing an approach for estimating
           baseline emissions.

     •     Identify available methods for developing site-specific baseline
           emission estimates (BEEs).

1.3  APPROACH

     To meet the objectives of this program, three steps were undertaken to
compile and assess existing information:  1) Conduct a literature search, 2)
Perform a  survey of key researchers, and 3) Review and evaluate the collected
information.   This work served as the basis for developing the protocol for
estimating air emission factors for remediation presented herein.  Each step
of the approach is discussed below.

     A computer-assisted search of 15 databases was performed to identify
published  literature of potential  interest.  Keywords were formulated into a
search strategy to identify abstracts related to both baseline and remedial
emissions.  Approximately 1400 abstracts were reviewed,  and over one hundred
publications were identified as pertinent and obtained by staff librarians.

     The literature search was augmented by a telephone survey to locate and
access unpublished data or research in progress.   A list of contacts was
developed that included regional  EPA personnel, employees of EPA research
offices, EPA contractors,  university researchers,  and referrals from those
initially contacted.   A set of questionnaires was  used to put the responses
obtained during the phone survey in a standard format.

-------
     The collected information was reviewed and evaluated with respect to  its
applicability to estimating baseline air emissions from NPL and other
hazardous waste sites.

1.4  RECOMMENDED USES OF THIS MANUAL

     This manual is, to the extent possible,  a complete,  stand-alone document.
It is, however, intended to complement existing guidance  manuals for the
Superfund program.

     This manual has certain limitations:

          The manual is a decision making  tool  but it  is  not intended to
          relieve the site managers of their  decision  making responsibility.

          The protocol is not a "cookbook"  for designing  air pathway
          investigations or for determining BEEs.

          The determination of BEEs for a  site will  not by itself,  yield an
          assessment of actual  or  potential air impacts,  but it is  a useful
          part of that evaluation  process.

     The steps in the overall  NPL  site clean-up process are shown in Figure 1.
The primary intended use for this  manual  (Volume II) is for estimating air
impacts as part of the evaluation  of the undisturbed site.   Therefore the
manual's guidance is input to the  record of decision (ROD)  step,  as well  as,
the RI/FS step.

-------
                                Site  Discovery
                           Preliminary Assessments
                               Site  Inspections
                             National  Priorities
                                 List   (NPL)
               Removal
             Actions at
               Non-NPL
                Sites
                          Removal
                        Actions at
                         NPL Sites
            Air
Remedial Investigations/
  Feasibility Studies
                            Records of Decision
                                 Air
                              Remedial Designs
                                  Air
                              Remedial Action
                         Operation and Maintenance
Figure 1. Superfund Flow Chart, Noting Elements Where This Manual May Apply.

-------
     Furthermore, this manual  provides the important function of standardizing
the air pathway analysis (APA)  for baseline NPL sites,  thereby ensuring that  a
uniform and systematic approach is followed for the diverse universe of NPL
sites.  The manual  provides a  protocol for estimating air quality impacts
resulting from undisturbed sites.   For each step,  a three-tiered approach  is
presented.  The approaches in  order of preference  are:

     K   Use of site-specific  field data;

     2.   Use of predictive models using site-specific  inputs;

     3.   Use of tabulated default values when  site-specific information is
          unavailable.

Therefore, emissions can be estimated regardless of the state of knowledge
regarding a given site.  Of course, the confidence of the emissions estimates
depends on the associated confidence of the inputs to the estimation
procedure.

      Limitations of the emissions  estimation procedures should be borne in
mind.  The primary limitation  is that the data  quality  of any emissions
estimate  is dependent on the data  quality of the inputs and on the quality of
the assumptions that are made.   The use of site-specific data as input to the
estimation procedure is preferable to the use of predictive models, which in
turn are preferable to the use  of  tabulated generic emission factors.  Data of
known quality (confidence) should  be used whenever available.  In many cases,
the conceptual site model  will  be  developed from a limited database.  The
resulting estimates of volume  of contaminated material, the type of
contaminants present,  the  concentration of the  contaminants, etc. will have
large associated uncertainties.  Therefore any  emissions derived from such
data will have an even larger overall  uncertainty.

-------
1.5  DOCUMENT ORGANIZATION

     There are five remaining sections to Volume II.   General  information on
the potential for air contaminant emissions from hazardous waste sites is
presented infection 2.  Section 3 offers a protocol  for determining if BEEs
are required and how to develop site-specific BEEs.   Information on sampling
methods that can be used to obtain BEEs is provided  in Section 4.   Section 5
describes case studies in which BEEs were needed and/or determined for
hazardous waste sites.  References are given in  Section 6.

     An annotated bibliography of the information reviewed  for this project is
included as Appendix A.  Appendix B identifies chemical  and physical
properties of waste material  that may affect is  emissions potential.   A guide
to developing an overall emission rate from individual  emission rate
measurements is included as Appendix C.   Information  on databases  containing
potential input values for predictive models is  given in Appendix  D.
Descriptions of remote sensing systems are contained  in Appendix E.

-------
8

-------
                                   SECTION  2
                   AIR EMISSIONS FROM HAZARDOUS WASTE SITES

     This section presents information on landfills and lagoons, the two
general types of sites used in this manual  to demonstrate methods for
estimating the potential for air emissions.  For this manual, all types of
uncontrolled solid waste sites, land disposal sites in particular, will be
referred to as "landfills" and all types of uncontrolled liquid waste sites
will be referred to as "lagoons."  The estimation methods described for
application to landfills and lagoons may generally be applied to solid and
liquid hazardous waste, respectively.

     This section addresses potential emission sources and potential air
quality impacts.  Discussion of potential air quality impacts covers the
general types of air quality impacts by waste site category, and the basic
transport mechanisms involved with the movement of contamination from lagoons
and landfills.  Where not otherwise specified, the general term hazardous
waste site is used to refer to both landfills and lagoons that contain
hazardous wastes and/or substances.  Figures  2 and 3 depict these two types of
sites in generalized schematic drawings.

     The site and contaminant characteristics discussed below are general
background information for working with the protocol presented in Section 3.
The information provided will  assist the site manager in developing conceptual
models of landfills and lagoons.   Based on  this conceptual understanding, the
site manager can then develop strategies for  assessing the potential impacts
and for estimating potential  air  emissions  from these sites.  The references
cited in this section and those listed in the annotated bibliography contain
further background material.

-------
                               SOLID
                            WASTE BODY
                                                     STRATIFIED
                                                   '  SOLID WASTE
UNLINED
SOILS CONTAMINATED
    WITH WASTE
                                                         CONTAMINATED
                                                         GROUND WATER
         Figure 2.   Conceptual schematic of a landfill.
                                10

-------
                                                      LIQUID WASTE
            STRATIFIED
           LIQUID WASTE
SOILS CONTAMINATED
  BY LIQUID WASTE
    CONTAMINATED
    GROUND WATER


              Figure 3.   Conceptual  schematic of  a lagoon.
                                     11

-------
 2.1  GENERAL DESCRIPTION

 2.1.1  Landfills

      Landfills are facilities into which wastes are placed for permanent
 disposal, and often are simply excavated pits.   Landfills may vary in  size
 from a few tenths of an acre to several  hundred acres, and other landfill
 characteristics can also vary greatly from one  site to the next.  Most
 variations are attributable to the types of stored wastes, the operating
 practices and the age of the facility,  and hence,  its design.

      Commercial landfills can be categorized by design criteria such as liners
 and gas-venting systems.  Older landfills are usually unlined.  Newer landfill
 designs may specify liner systems to retard transport of leachate and  wastes
 into soils and ground water.  Some landfills have  built-in gas venting systems
 to prevent build up of landfill  gases.

      Landfills can be further differentiated by the types of waste they store.
 Commercial landfills are commonly classified as municipal  or hazardous,
 depending on the types of waste  accepted.   Municipal  landfills accept solid,
 semi-solid,  and liquid nonhazardous  wastes,  including garbage, glass,
 plastics,  paper,  plant matter, ashes, some  industrial  wastes,  and demolition
 and construction  wastes.

     Hazardous  waste  landfills accept hazardous  sludge,  liquids,  semi-solids,
 residues,  concentrates, or leachate  or ash  originating from a waste.   Much of
 the hazardous waste originates from  manufacturing,  petrochemical,  and chemical
 industries.  Federal, state, and  local regulations  establishing  minimum design
 standards and restricting types of acceptable landfill  wastes  have  evolved
 over the last 25 years.  In the past, mixtures  of liquid  and  solid  waste were
 common practice.  Today, landfills can no longer accept  liquid wastes or
solids that contain free liquids unless they've been treated with fixatives
and stabilizers to eliminate the free liquids prior to disposal.
                                      12

-------
     Co-disposal  landfills  are  sites that have received and mixed municipal
and hazardous  wastes.  Any  available disposal records may relate types  of
wastes  and  location/mixing  within the landfill.

     Most landfills  that  are  selected as Superfund sites have undergone some
form of closure.   In some cases, currently operating facilities may have
abandoned hazardous  landfills at the same site.  The appearance of a closed
landfill will  depend on when  it was closed.  Closure may mean that the  site is
covered with vegetation or  that no waste is exposed.  Telltale signals  of
covered waste  are seeping leachate and odors.  In the past, landfills were
often sited in unpopulated  areas close to the industry or industries
generating  the wastes, but  population growth and development may result  in
people  living  or  working  in close proximity to the site.

     Superfund sites may  differ in significant ways from the types of
commercial  landfills discussed above.  In some cases, the site history will be
one of  relatively indiscriminate disposal of hazardous substances.  The  sites
may contain buried drums  or pockets of dumped/spilled wastes that are not
uniformly distributed across  the disposal area.  Information on waste types,
disposal practices,  date  of disposal, etc. may be limited or non-existent.
However, the protocol and measurement techniques presented in this manual are
sufficiently generalized  that they can be applied to such uncontrolled sites.
Conversely, the emission  models presented here may be of only limited use for
sites where the chemical  and  physical parameters that serve as model  inputs
are poorly  defined.

Potential Site Conditions—
     The conditions  encountered when investigating landfill  sites will  vary
from site to site because of the differences in location,  design, use,  and
operation.  Figure 2 presents a conceptual schematic of a landfill site.  The
condition of the  site cover material  will vary greatly and is dependent, to a
large extent,  on  the landfill's operational  history.   In the best situation,
the landfill cover will extend over the entire land disposal area; the cover
will  have been  constructed to minimize rainfall  percolation into the waste
body and regulated to minimize erosion of the cover.   The degree to which the
air,  the surrounding native soils,  and the ground water are protected from
                                      13

-------
 contamination from the wastes stored in  the landfill  will  depend both on the
 landfill  design and on the construction  and operation of the waste storage
 facility.  If the wastes have been  disposed of in  an  unlined storage area, the
 likelihood of contamination of surrounding  soils and  eventual contamination of
 the ground-water beneath the landfill  increases.   If  the landfill has been
 constructed at or below the ground-water table, the rate of transport of
 pollutants into the ground water further increases.   Waste material  in the
 landfill  may be stratified by age of disposal  and/or  settling of the more
 dense waste.

      Emission of air pollutants from landfills is  dependent on the chemical
 and physical properties of the stored wastes and on the  landfill  design
 components which may have been implemented  to  reduce  air emissions.   Municipal
 landfills are sources of significant amounts of methane  and carbon dioxide,
 and variable amounts of other non-methane hydrocarbons.   Hazardous waste
 landfills often are sources of non-methane  hydrocarbons,  including volatile
 organic compounds (VOCs) and semi-volatiles, including pesticides.  Co-
 disposal  sites combine the emission potentials of  both municipal  and hazardous
 waste sites.  The methane gases generated often can increase the  migration
 potential of the high concentrations of  non-methane hydrocarbons  by  acting as
 a carrier medium during bulk flow transport of these  contaminants.

 2.1.2  Lagoons

      For  purposes  of this  discussion,  the term "lagoon"  refers to the class of
 facility  also known  as  surface  impoundment or  impoundment.   This  type of
 facility  generally includes  a natural  topographic  depression,  a man-made
 excavation,  or a diked  area  formed  primarily of earthen  materials.   Lagoons
 are designed  to hold  liquid wastes  or wastes containing  free liquids.   Lagoons
 include holding, storage,  settling,   and  aeration ponds.

     These waste sites may range  in  surface area from a few  tenths of an acre
to hundreds of acres.  Man-made lagoons typically range in depth anywhere from
2 to 30 or more feet below land surface.
                                      14

-------
      In  some cases,  for  certain wastes, lagoons may be lined to minimize any
fluid seepage.  Clay,  asphalt, soil sealant, and synthetic membranes are
typical  lining materials.

     To  prevent migration of pollutants into the native soils and ground water
beneath  the lagoon,  lagoons are usually built above the naturally occurring
water table and take advantage of any impermeable surface or subsurface soils.
In areas with high ground-water tables, lagoons may be constructed on the land
surface  to minimize  ground-water contact.  Of course, use of liners and
building above the water table will not prevent the release of contaminated
air emissions from the surface of the lagoon, but these practices limit the
possible routes of air emissions (see Section 2.2).  Equally important are
operating practices  and  the characteristics of wastes in the lagoon.  The
importance of these  factors tends to parallel the landfill discussion above,
as does  the importance of lagoon siting practices.

Potential Site Conditions--
     Actual site conditions that will be encountered when investigating lagoon
sites will vary because  of siting, lagoon design, lagoon usage,  and
differences in lagoon  operations.  Figure 3 presents a conceptual  schematic of
a lagoon site.  The  condition of the lagoon will depend,  in large part, on the
wastes stored there  and  the lagoon's operational history.  Mixed wastes
typically separate into  stratified layers with the lighter materials near the
surface  and the denser liquids, sludges, and sediments having settled to the
lagoon bottom.  Contaminated soils around and beneath the lagoon are likely,
as well  as possible  contamination of the underlying ground water.

2.1.3  Equivalent Units

     Sites containing  mixed wastes or having other types  of inherent
variability may require  separate remedial  options to be considered for each
equivalent area.  During the remedial  investigation, the  site should be
theoretically divided  into equivalent units for estimating baseline emissions
and evaluating potential  emissions during remediation.  For example, a site
containing an abandoned  landfill, sludge pits,  and buried drums  would have at
least three distinct units and maybe more.  If the type,  concentration, or
                                      15

-------
distribution of a given form of contamination varies, then further  subdivision
of the units should be considered.  Similarly, if the soil media or proximity
of receptors varies significantly across the site, then further subdivision of
the units may be warranted.
            •*
2.2  ROUTES OF EXPOSURE

     Waste site characterizations performed during a remedial  investigation
are intended to determine potential or existing contaminant migration  by the
direct contact, surface water, ground water, and air pathways.  Each pathway
represents a potential route of exposure to the public and the general
environment.  Figures 4 and 5 show the potential routes for contaminant
migration from landfills ind lagoons, respectively.  The focus of this manual
is the air pathway, and several routes exist for contaminant emissions within
this  one pathway.

      Organic and  inorganic emissions from surface wastes may occur  as  gases,
aerosols, and contaminated particulate matter.  Where gas migration controls,
(e.g., gas venting systems) have been installed, volatile emissions from the
controls are likely to be higher than emissions from the site  surface.   The
lateral migration of solid ind liquid wastes into the surrounding soils  and
beneath the containment area ean create large areas of contaminated subsurface
sells.  The contaminated soils also represent a source of potential  air
emissions via the transfer of contaminants into the air-filled spaces  in the
soil matrix.  The contaminated soil gas can then transfer contaminants into
the atmosphere at the surface soil/atmosphere interface.

     The generation of leachate from landfills and lagoons can accelerate the
pollutant migration into the ground water below and provide an additional
source of air emissions resulting from the volatilization of dissolved
contaminants in the ground water.  The contaminated groundwater also can
transfer contaminants into the soil gas and hence the atmosphere.
                                      16

-------
                              DIRECT AIR EMISSIONS OF
                          VOLATILES 4 PARTICULATE MATTER
      GAS VENTING
       FROM VENTS
         VOLATILIZATION
       OF DISSOLVED SPECIES
         IN GROUND WATER
                                                                 LATERAL MIGRATION
                                                                   OF VOLATILES
                                                                 FROM SOLID WASTE
                                                               LATERAL MIGRATION
                                                                  OF VOLATILES
                                                               FROM CONTAMINATED
                                                                SOILS t LE ACHATE
Figure  4.   Conceptual  Schematic  Showing Air Contaminant  Pathways From an
            Unllned  Landfill.
                                       17

-------
LATERAL MIGRATION
   OF VOLATILES
FROM CONTAMINATED
 SOILS & LEACHATE
     VOLATILIZATION
   OF DISSOLVED SPECIES
     IN GROUND WATER
                                     DIRECT AIR EMISSIONS
                                   OF VOLATILES & AEROSOLS

Figure  5.   Conceptual  Schematic Showing Air Contaminant Migration From an
            Unlined  Lagoon with No  Cover.          	
                                       18

-------
2.2.1  Kev  Parameters  and Critical Factors Affecting Emissions from  Landfills

     The generation  of landfill emissions depends on several key chemical  and
physical properties  of the waste materials stored at these sites'along with
site and meteorological factors.  Table 1 presents these key factors  along
with the qualitative effects these factors may have on baseline emissions.
These qualitative effects can be used to help estimate the potential  air
emissions from  the site, given critical factors imposed by the site.  A
discussion  of each parameter is outside the scope of this document, but in
general, for volatile  compounds the rate limiting step is the movement of
vapors through  the soil.  Volatilization into the soil pore spaces is usually
quite rapid as  is the  transfer from the soil-gas into the atmosphere  once  the
soil/air interface has been reached.

     Additional  information about the key physical and chemical properties of
the waste material are presented in Appendix B.  Users should consult the
references  cited below and those listed in the annotated bibliography
(Appendix A) for additional background material.  General reading materials
that may prove  helpful include:  Guidance Document for Cleanup of Surface
Impoundment Sites,(4)  Model Prediction of Volatile Emissions,(5)  Air
Pollution Assessment of Toxic Emissions from Hazardous Waste Lagoons  and
Landfills,(6)   Air Quality Assessment for Land Disposal  of Industrial
Wastes,(7)  Estimating Air Emissions from Disposal Sites,(8)  and Air
Pollution Problems of  Uncontrolled Hazardous Waste Sites.(9)

2.2.2  Kev  Parameters  Affecting Emissions from Lagoons

     Figure 5 presents a generalized schematic of the volatilization process
from lagoons.   In general, the process consists of two steps: vaporization
from the surface liquid layer into the boundary air layer and then mass
transfer from the boundary layer to the bulk atmosphere.   The rate of
vaporization is  dependent on factors such as the compound's concentration,
Henry's Law constant,  water solubility and the system temperature.  Mass
transfer into the bulk atmosphere is dependent on compound properties such as
molecular weight and diffusion coefficient,  and site-related factors  such  as
                                      19

-------
     TABLE 1.  IMPORTANT PARAMETERS  IN  DETERMINING AIR  EMISSIONS AND THEIR
               QUALITATIVE EFFECTS ON BASELINE  EMISSION ESTIMATES (BEEs)
 Parameter
                                         Qualitative  Effect  on  BEEsa
                                        Volatiles
                                                            Particulate Matter
Site Conditions

Size of Landfill  or Lagoon
Amount of Exposed Waste
Depth of Cover on Landfills
                               Effects overall
                               magnitude of emissions
                               but not rate per area.
                                         High
                                        Medium
Compaction of Cover on Landfills
Aeration of Ligoons
Ground Cover

Weather Conditions

Wind Speed
Temperature
Relative Humidity
Barometric Pressure
Precipitation
Solar Radiation

Soil/Waste Characteristics
                                       Medium
                                        High
                                       Medium
                                       Medium
                                       Medium
                                        Low
                                       Medium
                                        High
                                        Low
                                                            Effects overall
                                                            magnitude of
                                                            emissions, but not
                                                            rate per area.

                                                                 High
                                                                  Low
                                                                  High
                                                                  High
                                                                  High
                                                                  Low
                                                                  Low
                                                                  Low
                                                                  High
                                                                  Low
Physical Properties of Waste
Adsorption/Absorption
Properties of Soil
Soil Moisture Content
Volatile Fraction of Waste
Semi -Vol ati 1 e/Non-Vol ati 1 e
Fraction of Waste
Organic Content of Soil
and Microbial Activity
High

Medium
High
High

Low

High
High

Low
High
Low

High

Low
a High,  medium,  and low in this table refer to the qualitative effect that the
  listed parameter typically has on baseline emissions.
                                      20

-------
temperature, pressure, and wind speed.  Berms, wind breaks, and lagoon
geometry affect the wind speed at the liquid surface and can thereby control
the rate of mass transfer.  In general, volatility increases as the molecular
weight of the compounds present decreases.
             *
2.3  MAGNITUDE OF AIR EMISSIONS

     The magnitude of baseline air emissions from landfills and lagoons is
dependent on waste-specific chemical and physical factors and site-specific
environmental factors.  Limited data are available on measured air emission
fluxes (rate per area) from previously studied waste sites (Table 2).  These
data can be used to give the site manager some idea of typical  baseline
emission estimation (BEEs) and a limited comparison of BEEs for different
types of waste sites.  Emission flux data for disturbed or exposed wastes are
included to demonstrate the potential for increased emissions of volatiles
during waste remediation.

2.4  EMISSIONS OF POTENTIAL INTEREST AT NPL SITES

     The types of emissions at a hazardous waste site are dependent on the
types of waste present, and these in turn are dependent on the  types of
industries and manufacturers that produced the waste.   A listing of the
typical wastes generated by 30 various industries and manufacturers can be
found in the Handbook of Industrial  Waste in California (11).   The 25 most
frequently detected compounds at 546 hazardous waste sites are  summarized in
                                      21

-------
  TABLE 2.   SUMMARY OF AVERAGE  BASELINE  EMISSIONS  FOR VARIOUS EMISSION SOURCES
 Waste Type
                  Source Type
        line
Emission Estimate
   For TNMHC*
  (ug/m2-min)
   Disturbed
   or Exposed
Waste Emissions
   For TNMHC
  (ug/m2-min)
NPL/Hazardous
Waste Sites




Landfills
Site A
Site B
Site C
Lagoons
Site D
360 190,000
740 26,000
29 170,000

43 640,000
Industrial  Waste  TSDFC Facilities

                  Active  Landfills
                   Site E
                   Site F
                   Site 6

                  Inactive  Landfills
                   Site H (covered)
                   Site I (covered)

                  Land Treatments
                   Site J
                                                                    44-150
                                                                        47
                                                                         9
                                                                   610-9600
Lagoons
Site K
Site L
Site M
Site N

120
570
9-31b
630

_ _-

—

a  TNMHC = Total Non-Methane Hydrocarbons.
   Different assessment techniques were used.
c  Transfer, storage, and disposal facilities (RCRA)

Source:  Reference 10=
                                      22

-------
Table 3 according  to the type of media, i.e., groundwater, surface water,  or
air.  The table  shows the number of sites where each contaminant was detected
and the contaminant's relative rank for each type of media.  Another useful
listing for selecting contaminants with the potential for emissions of
concern, is the  list of toxic compounds most commonly addressed by state and
local regulatory agencies given as Table 4.

2.5  SUMMARY OF  POTENTIAL RECEPTORS

     Receptors can be divided into three broad categories:

     •    On-site  workers;
     t    Off-site populace; and
     •    Non-human receptors.

     The on-site workers are in the closest proximity to the hazardous waste
site and are potentially subject to the most acute exposure to hazardous
substances.  The use of personal protective equipment,  real-time field
instruments, personnel monitoring, site controls,  and designated work zones
are designed to  ensure that field personnel are properly protected against the
hazards present  at the work site.

     The off-site  (and any on-site) population in  close proximity to the
hazardous waste  site is another receptor of primary concern.   These people are
often acutely aware of the hazardous waste site and the potential  for
contaminant exposure.  Section 3 provides an approach for estimating emissions
which can subsequently be used to predict the airborne  contaminant
concentration for  downwind receptors to assist in  the establishment of
appropriate action  levels.  Use of this protocol,  coupled with an effective
monitoring and modeling program, will  provide useful  information for the
site's community relations program.
                                      23

-------
       TABLE 3.   MOST FREQUENTLY REPORTED  SUBSTANCES AT  546  NATIONAL  PRIORITY
                  LIST SITES
 Substance Identified at
    Hazardous Waste
    Disposal Sites
Sites*
                                            Air
 Number of
Sites (Rank)bc
Ground Water
 Number of
Sites (Rank}6
                                     Surface Water
                                       Number of
                                     Sites (Rank)b
 Host Frequently Occurring
1.
2,
3.
4.
5,
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
Trlchl oroethyl ene
Lead
To! uene
Benzene
Polychlorinated Biphenlys (PCBs)
Chloroform
Tetrachl oroethyl ene
Phenol
Arsenic
Cadmium
Chrotni um
1,1,1-Triehloroethane
Zinc and Compounds
Ethyl ebenzene
Xylene
Methyl ene Chloride
Trans-1 , 2-D1 chl oroethyl ene
Mercury
Copper and Compounds
Cyanides (Soluble Salts)
Vinyl Chloride
1.2-01 chl oroethane
Chlorobenzene
l,l-D1chloroethane
Carbon Tetrachl oride
179
162
153
143
121
111
90
84
84
82
80
79
74
73
71
63
59
54
47
46
44
44
42
42
40
8
7
1
1
6
1
3
3
2
31
1
3
2
7
9
2
1
4
6
2
4
2
0
0
2
(5)
(6)
(3)
(2)
(4)

(16)
(16)
(17)
(17)

(18)

(7)
(4)


(10)


(ID




127
77
81
84
29
70
57
43
45
28
34
58
28
36
32
36
42
27
17
16
28
25
23
28
25
(1)
(4)
(3)
(2)
(21)
(6)
(7)
(9)
(8)
(16)
(14)
(6)
(17)
(12)
(15)
(13)
(10)
(20)
(24)
(25)
(18)
(21)
(23)
(19)
(22)
49
84
40
36
54
24
17
28
35
28
33
20
27
14
8
17
17
20
16
16
10
17
9
8
12
(3)
(1)
(4)
(5)
(2)
(11)
(14)
(8)
(6)
(9)
(7)
(12}
(10)
(20)
(25)
(15)
(16)
(13)
(18)
(19)
(23)
(17)
(23)
(24)
(21)
  Number of sites  at  which substance is  present.   Substances may be present in
  one, two, or all  three environmental media  at all sites  at which it  is  known
  to be rpesent.   Therefore, the number  of  sites  at which  each substance  is
  detected in environmental  media may not equal the number in this column.
b

  Not all  ranks will  be  represented in all  media  because not all chemicals
  found in media are  among those found most frequently at  site.
c

  Volatile organics not  otherwise specified were  reported  as being detected
  most often (rank  1)  in the air medium.

Source: Reference  12
                                        24

-------
TABLE 4.  TOXIC POLLUTANTS MOST COMMONLY ADDRESSED BY STATE AND LOCAL AGENCIES
Acetaldehyde
Acrolein
Acrylonitrile
Allyl Chloride
Arsenic
Asbestos
Benzene
Benzidine
Benzo(a)pyrene
Benzyl Chloride
Beryl 1i urn
Bi s(chloromethyl)ether
1,3-Butadiene
Cadmium
Carbon Tetrachloride
Chlordane
Chlorobenzene
Chloroform
Chloroprene
Chromium
Cresol
1,4-Dichlorobenzene
3,3-Dichlorobenzidine
Dimethyl Sulfate
1,3-Dioxane
Dioxins
Epichlorohydrin
Ethylene Dibromide
Ethylene Dichloride
Ethylene Oxide
Ethylenimine (azridine)
Formaldehyde
Heptachlor
Hexachlorocylopentadi ene
Hydrazine
Hydrogen Sulfide
Lead
Lindane
Maleic Anhydride
Manganese
Mercury
Methyl Bromide
Methyl Chloride
Methyl Chloroform
Methylene Chloride
beta-Naphthylami ne
Nickel
n-Nitrosodimethylamine
Ni trosomorpholi ne
Parathion
Perchloroethylene
Phenol and Chlorinated Phenols
Phosgene
Polychorinated Biphenyls (PCBs)
Polycyclic Aromatic Hydrocarbons (PAH)
Propylene Oxide
Radionuclides
Styrene
1,1,2,2-Tetrachlorethane
Tetrahydrofuran
1,1,2-Trichloroethane (vinyl trichloride)
Toluene
Trichoroethylene
Vinyl Chloride
Vinylidene Chloride
Xylene
Source:  Reference 12.
                                      25

-------
     Non-human receptors also may be a concern at some hazardous waste sites.
Disturbance of the site may lead to exposure through inhalation of
contaminated air or exposure through ingestion of or direct contact with
contaminants deposited on plant and inert surface*   The possibility of
inhalation exposure may affect feral  or domesticated animals downwind of the
site.  This also impacts humans in that animal  exposure to pollutants can lead
to contamination accumulation in the food chain.

     Certain gaseous pollutants (e.g.,  ozone,  oxides of sulfur and nitrogen),
if present in high concentrations,  also can  affect plant and animal  growth.
Deposition of airborne contaminants may cause  stressed  vegetation,  release
pesticides and herbicides,  or impact  the value/usability of agricultural
crops,,   Also, deposition of metals  or other  pollutants  in surface  waters  may
impact  marine life.   Copper and some  other metals can cause fish kills  at very
low concentrations.
                                     26

-------
                                   SECTION 3
                   PROTOCOL FOR BASELINE EMISSION ESTIMATES

     This section presents a protocol for developing baseline emission
estimates (BEEs).  This protocol is a component of an air pathway analyses
(APA) program to assess potential air quality impacts from hazardous waste
sites.  While not all sites will require BEEs, the first three steps in the
protocol should be implemented to see if BEEs are necessary for a given site.
The protocol is a recommended guideline; the level of effort that is required
or the need to develop BEEs for individual sites must be determined on a
case-by-case basis.

3.1  PROTOCOL STEPS FOR DEVELOPING BEEs

     Figure 6 diagrams a protocol for developing BEEs.  The protocol was
developed to help the site manager to determine baseline emission rates and
absolute levels of emissions.  These values can be used as inputs to
dispersion models to assess the air impacts for receptor locations of
interest.  The activities identified in this flow chart are consistent with
the steps of the CERCLA remedial investigation process that involve the
assessment of the air contaminant migration pathway.   Although the protocol
was developed for NPL sites, it also applies to assessing air emissions from
other hazardous waste sites.  The flow chart is applicable to all sites,
regardless of the type of site (landfill, lagoon, waste pile, etc.), type of
waste, or the potential for the site to generate air emissions.   Each step of
the protocol is described below.

3.1.1  Define the APA Objective

     CERCLA and SARA legislation highlight the basic objectives  for all
remedial investigations.  Simply stated, these objectives are to provide data
that are "necessary and sufficient" to characterize the "nature  and extent" of
contamination on site.  In addition, they mandate that "all potential
                                      27

-------
     €«te* df Guidance -
                                      Define the APA Objectives
                                      Design and Conduct the
                                          Site Scoping
                                            Site Data
                                        Potential Exists for
                                     Air Pathway Contamination
   Design and Conduct a
Screening APA to Determine If
 In-Depth Baseline Emission
Estimate Data Are Necessary
                                     In-Depth BaseRne Emission
                                    Estimate Data Are Necessary
                                    Design and Conduct Detailed
                                     APA To Determine In-Depth
                                     Baseline Emission Estimate
  Report In-Depm Baseline
    Emission Estimate
                                                                                Document
                                                                              We Potenfla! (or
                                                                              Contamination
                                                                 In-Depth Baseline
                                                                Emission Estimate
                                                                    Data Are Not
                                                               	Necessary
                                                                       Document
                                                               Screening In-Depth
                                                                Baseline Emission
                                                                    Estimate Data
                                                               Sufficient
                                                               APA Data
                                                                                  Site
                                                                                Mitigation
Figure 6.    Flowchart  of  activities  for developing  screening and  in-depth
               baseline emission estimates.
                                              28

-------
migration pathways for contaminations" require characterization.  As the first
step of the protocol to assess baseline emissions, site-specific objectives
should be developed; this will generally occur simultaneously with the
performance of Steps 2 and 3 (data collection and review) of the protocol.
The site manager should consider the following issues when formulating
site-specific objectives:

     •    What information is already available?  As described in Section
          3.1.2, available information should first be reviewed before
          developing final site-specific objectives.  This preliminary review
          of information will provide necessary background information and aid
          in identifying data gaps.

     §    What pathways must be considered?  Except in rare cases,  all
          pathways, namely air, soil, and water,  must be considered.

     t    What is technically possible?  Site and situational  factors that may
          adversely affect the air pathway investigation should be  identified
          to avoid establishing unrealistic objectives.   These factors can
          range from complexity in site geology/hydrogeology and complex
          terrain to the feasibility of detecting the contaminants  of
          interest.

     t    What time deadlines exist?  Schedule constraints can affect the
          nature of the investigation and must be balanced with technical
          concerns.

     0    What data quality objectives are required?  Data must be  of a known
          accuracy and precision for use in evaluating the air pathway.

     •    What program is most cost effective?  The type,  level,  and  extent of
          contamination per migration pathway will  primarily determine how the
          available resources are apportioned.   It is also necessary  to
          collect,  integrate, and consider a variety of types  of data,
          including technical information,  institutional  issues,  political
                                      29

-------
          issues, public protection, community relations issues, and community
          concerns.

     •    What contaminants must be considered?  The investigation will
          typically involve identifying and characterizing the extent of the
          contamination.  Concentrations of individual  or summed indicator
          compounds are often selected to represent the extent of total
          contamination.  Site-specific objectives should neither identify
          specific indicator compounds nor require characterization of all
          compounds.

     Table 5 provides examples of objectives related to baseline emissions
estimates.  The first step in developing site-specific  APA objectives is to
collect and review readily available site historical  records.   The potential
for air emissions can be inferred from the review of preliminary site
information.  While baseline emissions may be low,  during remediation sites
have the potential for air emissions of particulate matter (semi-volatile
organics, metals, and other inorganic contaminants)  and enhanced volatile
organic emissions.

     Among the types of information  that can be reviewed and  used to develop
site-specific APA objectives are:  waste characteristics;  distribution  of the
waste;  orientation of the  general  public to  the emission source concerns;
technical feasibility;  and program resources.   The  first three items are
discussed below.

     •     Waste Characteristics.   Knowledge  of the  industrial  process or the
          waste source  involved  can  suggest  the types of chemicals  or agents
          that  may be  in the waste.   For example, volatile  emissions are
          likely  from waste  sites  with  organic solvents or  petroleum waste.
          Wastes  in  liquid  form  tend to have higher  baseline emissions  than
          wastes  in  solid  form.   In  addition,  identifying common  decomposition
          products of the  chemicals  identified in the waste may be  useful.
          It  is helpful  to categorize  potential  air contaminants  by  their
          generic  volatility:  volatiles,  semi-volatiles, and non-volatiles.
                                     30

-------
            TABLE 5.   EXAMPLES  OF  APA OBJECTIVES  FOR BEEs
t    Characterize the air emissions potential for volatile species and
     particulate matter from the undisturbed site.
•    Characterize the air emissions potential for volatile species and
     particulate matter from the disturbed site.
•    Identify contaminants of concern.
•    Provide baseline emission estimates that can be used to assess the
     health risk and the need to mitigate.
•    Provide baseline emission estimates that can be used to assess the
     need for on-site or fence!ine ambient monitoring.
                                 31

-------
           Examples of the  types of compounds  in each category  are listed in
           Table 6.  The  categories group together compounds with  similar
           physical behavior  in the atmosphere and thus are useful  for
           predicting emissions potential.

      •     Distribution of  the Waste.  The relative position of the waste can
           influence the  potential for air emissions and, thus, the APA
           objectives.  The amount and nature of overburden strongly influences
           the  baseline emissions at a site.  Semi-volatiles and non-volatiles
           present  near the surface can be emitted as windblown particulate
           matter.   Waste piles may have relatively high emissions  of  both
           volatile organics  and particulate matter due to their geometry and
           surface  area to  volume ratio.

      •     Location of the  Affected Population and Community Concerns.  The
           population potentially at risk from exposure to toxic air emissions
           from  the site must be identified and the exposure characterized.
           The close proximity of residential  areas may require the addition of
           measurement and monitoring activities  for health and safety  purposes
           that  are beyond what is necessary to develop BEEs for the site.
           Thus, the site-specific objectives  may include a component related
           to determining the potential  impact on the nearest  population  (e.g.,
           air monitoring at the site boundary).

     The APA objectives should be  documented  and circulated for peer review by
staff members that have knowledge  of the  site, APA,  and  the site mitigation
process.  The review process  may  also help  identify  site  characterization data
needs.  Developing site-specific  objectives  is an  iterative process; more than
one round of data gathering,  review,  and  discussion  may  be needed  to develop
satisfactory objectives.
                                     32

-------
      TABLE 6.   POTENTIAL AIR CONTAMINANTS  BY  GENERIC  TYPE  OF  CONTAMINANT
Volatiles  (>1 mm mercury vapor pressure at 25*C)
     a    All monochlorinated solvents; also trichloroethylene,
          trichloroethane, tetrachloroethane
     t    Most simple aromatic solvents: e.g.  benzene, xylene, toluene, and
          ethyl benzene
     •    Most alkanes up to decane (C10)
     t     Inorganic gases: e.g. hydrogen sulfide9 chlorine, and sulfur dioxide
Semivolatiles d-10'7  mm mercury  vapor pressure at 25*C)
     •    Most polychlorinated biphenyls, dichlorobenzenes, aniline,
          nitroaniline, and phthalates
     •    Most pesticides: e.g. dieldrin, toxaphene,  and parathion
     •    Most complex alkanes: dodecane and octadecane
     a    Most polynuclear aromatic's: e.g. napthalene, phenanthrene, and
          benz(a)anthrencene
     •    Mercury
Non Volatiles or Particulate Matter (
-------
 3.1.2  Site Scoping

      The second step in  the development of BEEs is collection^  available
 information about  the site.  This should be a quick, straightforward
 information search,  involving the collection of records, reports,  shipping
 manifests,  newspaper clippings, and information from interviews with  people
 living  close to or affiliated with the site.  For NPL sites, data  should be
 available from the preliminary assessment and site inspection conducted  prior
 to inclusion on the,NPL.  The type of information to be collected  parallels,
 for the most part,  the factors considered in creating the objective.  These
 include:

      t     Source of the waste (type of industry);

      •     Composition  of the waste (organic-volatile/semi-organic;
           inorganic=metals, others; biological;  radioactive);

      •     Distribution of the waste and cover material  limiting volatilization
           and  uptake of particulate matter from the waste;

      •    Distance from the waste to the property  fence!ine; and

      •    Representative meteorological  data.

 3.1.3   Evaluate Available Site Data

     The existing site information (including  the  site  inspection report)
 should be evaluated to determine  the potential  for release of air emissions.
 Examples of types of contaminants and situations to be  evaluated are provided
 in Table 7.  If it is determined  through this  assessment that the site poses
no significant potential  for air  pathway contaminant  emissions,  then no
further evaluation of the baseline emissions  is  required.   The  site manager
must record the basis for this  decision  and  include these  data  in  the site
investigation documentation.   In  most cases,  insufficient  information will be
available at this stage,  and further work
                                      34

-------
     TABLE 7.  EXAMPLES OF TYPES OF CONTAMINANTS AND SITUATIONS THAT MAY
               INDICATE A POTENTIAL FOR AIR PATHWAY  CONTAMINATION
Situation/Condition
           Particulate
Volatiles    Matter
          Comment
t Site Odors,           /
  Neighborhood
  Complaints

t Observation of Dust
  Clouds During Wind

t Evidence of Metal     J
  Corrosion

• Vent pipes            J
• Seeps of Waste        /

0 Weathered Waste       J
  Surface

t Aged and Layered      J
  Waste
• Aerated Lagoons       J

• Exposed Waste         J

• Industrial Wastes     J



t Petroleum Wastes      7
t Industrial Wastes/
  Paint Wastes          J
          J


          J
          7

          J
                      Indicates  moderate to high
                      levels  of  BEEs.
Check soil cover and look for
waste piles.

Look for corrosive agents.
Check records to determine if
the site is a codisposal
facility.

Probable buried wastes.

Emissions of disturbed waste
may be very high.

Likely that volatiles are
higher in underlying lay
chemicals possible.

Increased emissions.

Increased emissions.

Waste mixtures likely, check
for particular solvent types,
aromatic and halogenated
organic solvents.

Tar/wastes with volatile
emissions likely.

Organic volatiles likely.
                                                               (Continued)
                                      35

-------
                           TABLE 7.   (Continued)
                ion
           Particulate
Volatiles     Matter
                                                         Comment
a Recycling or
  Plating Wastes

• Municipal Wastes
• Hospital  Wastes
 J
 J
y
• Site Inspection
                          y
                                       y
Metal-containing particulate
matter likely.

Methane/carbon dioxide
volatiles likely; look for
industrial waste.

Solvent used likely;
biological hazards and
radioactive waste possible.
                     specific chemicals likely.

                     Gas detection results indicate
                     the presence of gas species
                     and the potential for
                     emissions.

                     Visual inspection and particle
                     counting/detection results
                     indicate the potential for
                     particulate matter emissions.
                                     36

-------
will be warranted.  If air emissions are a potential concern, the next step of
the protocol (site screening study) should be implemented to provide
additional information to make a judgement regarding the potential for air
emissions from the site.  At this point the site-specific APA objectives
should be reviewed by the site RPM to ensure they are still realistic,
attainable, and applicable.

3.1.4  Design and Conduct the Site Screening Study

     Designing a site screening study to assess the air emissions potential
involves the selection of an air emissions measurement/assessment method.  The
four broad categories of measurement/assessment approaches include:

     «    Direct emissions measurement;
     •    Indirect emissions measurement;
     •    Air monitoring/modeling; and
     •    Emissions (predictive) modeling.

     Each approach includes a range of possible methods that can.be
categorized according to their level of complexity as screening (quick and
simple) methods or in-depth (very detailed) methods.

     The activities necessary to design and conduct the site screening study
are:

     •    Determine the feasibility of obtaining the screening data.
          (Identify any site factors that may limit this activity.)

     •    Select appropriate tracer species,  screening methods, and applicable
          equi pment/i nstrumentati on.

     •    Design the site inspection technical  approach and test plan,
          including the Quality Assurance/Quality Control  Program.  Make sure
          that  all  units of a combined site are studied.
                                      37

-------
     0    Circulate the site screening approach for review and ensure the
          screening addresses the site-specific objective(s).

     c    Modify the site screening program,  as necessary.

     •    Conduct the site screening study and document the findings.

     •    Determine if the site screening study was adequate to characterize
          the air contamination migration pathway and if detailed BEE data are
          necessary.  If detailed BEEs are necessary, initiate the in-depth
          site characterization study.  If not, document the site inspection
          survey results and the basis for discontinuing the APA.

One preliminary step is to evaluate those key factors that affect the air
emissions of volatiles and particulate matter.   The factors were presented in
Section 2 and are summarized in the check-list presented in Figure 7.  This
figure can be used to summarize site information and facilitate the decisions
regarding selecting and implementing screening methods.   If the site contains
a waste type that has the potential  to create air emissions,  the most
important factors which determine the baseline emissions are typically site
conditions and weather conditions.   Once  the  checklist (Figure 7) has been
completed and some knowledge of the factors affecting air emission processes
is gained, the site manager must select appropriate indicator species and
select air emission screening methods, equipment,  and instrumentation.

     Indicator species are species  found  in the waste that can be used to
represent a group of species in determining emissions.  If little is known
about which specific species are present  at the site, select an indicator that
represents a family or class of species so that gross data on emissions can be
obtained and then later refined. Table 8 provides information that will aid
in this selection process.   All  highly toxic  compounds likely to be present at
the site should be on the target compound list, along with the indicator
species.   The ideal  indicator species or  class  of species is:
                                      38

-------
                                                                 -Effect on Emissions-
                          General Effect      Site            Volatilea	     Partieulate Matter
  Parameter                Volatile   PM    Information    Increase   Decrease     Increase   Decrees*
SITE CONDITIONS


Amount of Exposed Haste      High    High
Depth of Soil              Medium   High
Presence of Oil Layer        High    High
Compaction of Cover         Medium   Low
Aeration of Lagoons          High    High
Ground Cover               Medium   High
HEATHER CONDITIONS


Hind Speed                 Medium   High
Temperature                Medium   Low
Relative Humidity            Low     Low
Barometric Pressure         Medium   Low
Precipitatioa               High    High
Solar Radiation             Low     Low
SOIL/HASTE CHARACTERISTICS


Physical Properties          High    High
Sorption of Soil            Med     Low
Soil Moisture               High    High
Volatile Fraction            High    Low
Semi/Non-Volatile  Fraction   Low     High
Organic Content of Soil      High    Low
  Figure  7.   Checklist of factors  affecting air emissions  per  unit.   The  site
                manager  should use this  to summarize  site  data on critical
                factors  to determine  how these factors may affect the air
                emissions potential.
                                               39

-------
TABLE 8.  FACTORS TO CONSIDER IN SELECTING AN INDICATOR SPECIES FOR STUDY
     1)   Homogeneity of waste and representativeness  of proposed
     2)   Variety of types of air contaminants  (organic,  inorganic,
          biohazard, radioactive);
     3)   Physical state of air contaminants  (gas,  liquid,  solid);
     4)   Level  of air contaminant  emission;
     5)   APA objectives;
     6)   Feasibility of air monitoring  for proposed  indicator species;
     7)   Availability of standard  sampling/analytical/monitoring
          techniques;
     8)   Potential  interferences for the proposed  indicator  species;
          and
     9)   Health effects.
                                     40

-------
     t    Present  in the air emissions in a fixed ratio;

     •    A non-reactive or stable species;

     •    Present  at levels above analytical detection limits;

     t    Unique to the site (not in background air samples);

     t    Representative of the "worst case" toxicity for compounds at the
          site; and

     t    Applicable for existing measurement and monitoring technologies; and

     «    Of known toxicity and exposure criteria.

     A list of candidate indicator species can be developed from those species
previously identified in analysis of the waste or by identifying broad-band
type indicators that represent the type of waste identified in the scoping.
Candidate species  should match as closely as possible the characteristics of
an ideal indicator species.  The two main required characteristics are
presence in the air emissions from the site and the ability to measure/monitor
the species using commercially available methods and instruments.  Examples of
broad-band, class and indicator species are given in Table 9.

     In addition to selecting indicator species, the site manager must select
the most suitable air emissions sampling methods for screening.   Screening
methods are summarized in Table 10.   Air emissions measurement/assessment
methods are described in detail in Section 4.   The four general  approaches for
screening emissions are described below.
                                      41

-------
        TABLE 9.  EXAMPLES  OF  BROAD-BAND, CLASS, AND  INDICATOR  SPECIES
  BROAD BAND
Volatile Organics
                           CLASSES OF COMPOUNDS
                           Aliphatics

                           Aromatics
                           INDICATOR.SPECIES
                           Alkanes, Total
                           Hydrocarbons as Pentane
                           Benzene, Xylene, Toluene
Volatile Inorganics
                          Halogenated Species
                          Oxygenated Species
Nitrogen Containing
Species
Acid Gases
                          Sulfur Containing
Semi-Volatile Organics     Polynuclear Aromatics
Non-Volatiles
                          Polychlorinated
                                  is
Metals
                           Trichloroethene,
                           Trichloroethane, Vinyl
                           Ethanol,  Formaldehyde
                                                     Benzonitrile
                                                     Sulfur Dioxide, Hydrogen
                                                     Chloride
                                                     Hydrogen Sulfide
                                                     Napathalene, Benzo-
                                                     (a)Pyrene
                                                     PCBs As Aroclor 1254
                                                     Lead, Chromium, Zinc
                                     42

-------
        TABLE 10.   TYPES OF SCREENING METHODS FOR ESTIMATING EMISSIONS*
          Direct Emissions Measurement

          -  -Head Space Sampler
          -  Head Space Analysis of Bottled Sample
          Indirect Emissions Measurement

          -  Upwind/Downwind
          -  Mass Balance
          -  Real-time Instrument Survey
          Air Monitoring/Modeling
          Emissions (Predictive) Modeling

          -  Superfind Exposure Assessment Manual (SEAM) models
          -  Any model using literature values and assumed concentrations
*See Section 4 for more detail
                                      43

-------
 Direct Emissions Measurement.  Concentration measurements can be
 made of the air directly  above the waste, either  the  "head space" or
 air space of a sampler  placed on the waste, or  the  head space of a
 sample bottle half-filled with waste material.  The screening
 provides a relative measure of the emission potential  of various
 wastes/locations.  High concentrations of volatile  species in the
 head space can indicate high potential for air  emissions from the
 site.   Likewise,  low concentrations can indicate  low potential  for
 air emissions.   The advantage of these screening  methods is  that
 they are relatively quick, easy and inexpensive to  perform.   They
 also can have the highest sensitivity (i.e., detect the  contaminants
 even at low concentrations), since they measure concentrations  at
 the source.   The techniques are only applicable to  volatile  gas
 species.   These data can be used to estimate emission  factors to
 indicate the potential for air emissions.

 Indirect Emissions Measurement.   These types of sampling methods can
 be  used for  any type of contaminant such as volatiles  and/or
 particulate  matter.  It is probably the most common screening
 approach  used, though it is usually not used to develop  emission
 estimates.   Screening measurements  can be  made upwind  and downwind
 and  directly  above waste material  using real-time instruments to
 estimate  potential for air emissions.   Air concentration
 measurements made at short distances  (<40  meters)  downwind of the
 waste can be used to indicate  the  potential  for air emissions from
 the  site.  Downwind measurements  should be corrected for any
 instrument bias and upwind interferences.   Meteorological factors
 can  influence the air concentration of volatiles and particulate
matter so field notes must include  on-site observations and
meteorological conditions  during testing.
                            44

-------
     •    Air Monitoring/Modeling.  Air monitoring and modeling methods  are
          equivalent  to  the  indirect emissions measurement methods  except  that
          tht samples are collected at greater distances from the waste,
          typically at the fenceline or property line, and the sampling
          locations are  not  as clustered.  If meteorological data are
          collected concurrently, these data can be used to validate
          dispersion  model estimates and estimate the potential for
          contaminants to reach off-site receptors.  However, they  may be
          limited  by  the sensitivity of the instrumental measurement used  and
          the dispersion and dilution of the air emission.  Unless  estimates
          of off-site impact potential are needed, fenceline monitoring  is not
          recommended for the screening study.

     •    Emissions (Predictive) Modeling.  Emissions modeling can  be used if
          the preliminary assessment has provided fairly detailed information
          that can be input  into a model.  The model  estimates can  usually be
          improved by using  air emissions data from the site.  The  Superfund
          Exposure Assessment Manual (SEAMS) models are appropriate as a
          screening technology for landfills, and the Mackay model  is
          appropriate for lagoons.  Modeling has the obvious advantage of
          being an off-site  activity.

     Once an appropriate screening method and associated equipment/
instruments are selected, a technical approach to applying the method should
be developed.  The Data  Quality Ob.iective For Remedial Response Activities
(13) can provide assistance  in designing the site screening.  Again, each unit
of a combined site must  be studied independently.  This may involve the use of
different screening methods for the various site units.

     Quality control  should be an integral part of any screening study, and
direction in developing  and documenting a quality assurance project plan can
be obtained from the  interim Guidelines and Specifications For Preparing
Quality Assurance Pro.iect Plans.(14)  One very useful  exercise in the design
of the screening study is a "dry-run" survey, including a mock exercise
involving data use.   This will  help in determining if the technical  approach
                                      45

-------
will satisfy the intended objective of obtaining an estimate of the potential
for air emissions from the site.

     After design is completed, the slti screening approach should be
circulated for review and then modified, as necessary.

     Once the site screening has been completed, the screening data should be
evaluated to determine if further data (in-depth measurements) are required
for the site characterization.  Screening data may consist of a concentration
number representing volatile content (headspace sampler or analysis of
headspace above the waste in a bottled sample) or a preliminary BEE (indirect
emission measurement/model estimate, air monitoring/model  estimate, predictive
model estimate).  The BEE should be an addition of all of the site units if
the units are studied individually.  Any emissions data for disturbed waste
should not be included in the BEE and do not have to be summed separately
since the site mitigation planning will  probably use the disturbed site
emission estimates on a unit basis in developing mitigations for "operable
units."

     Absolute criteria against which to  compare these BEE  data are not
available.  Estimated ambient levels (determined from the  BEEs) can be
compared to state ambient air toxic levels  and to EPA cancer risk values if
available.

     If high levels of air emissions from the site are possible and more
detailed information is required to meet the APA objectives, then site
characterization should be conducted using  in-depth technologies to develop
representative BEEs.   If, however,  little or no potential  for air emissions
surfaces from the data or if screening BEEs  provide enough information for a
site manager to evaluate air emissions,  in-depth BEEs development may not be
necessary in the site characterization.
                                      46

-------
3.1.5  Design  and  Conduct the  In-Deoth Site Characterization

     The activities  to design  and conduct an in-depth study are  similar  to
those described  in Section 3.1.4 for screening except that in-depth  assessment
methods rather than  screening  methods are used.  The steps are:

     •    Determine  the feasibility of obtaining the detailed BEEs.

     •    Select appropriate in-depth methods, indicator species, and
          applicab!e equipment/instrumentation.

     t    Design the site characterization technical approach and test plan,
          including  the QA/QC  program.

     •    Circulate  the detailed technical approach and test plan for review
          and  ensure it addresses the site-specific objective(s).

     •    Conduct  the site characterization program and document the findings.

     •    Determine  if the BEE data are sufficient and adequate for site
          mitigation decisions.  If adequate,  document them;  if inadequate,
          evaluate the data needs and reiterate,  as necessary.  Re-evaluate
          and  document.

     In-depth  techniques for developing BEEs are  summarized in Table 11 and
presented in detail  in Section 4.  In general,  direct emissions measurement
methods offer  several advantages over the other approaches and are considered
to be the preferred  methods for most sites.   With the exception of the wind
tunnel, the direct measurement methods applicable to landfills and lagoons are
limited to volatiles.  These technologies generate BEEs as a  function of the
site conditions and  can be used with a variety of analytical  techniques.
Measurements at the  source (waste)  will  be the highest in concentration
compared to the other measurement techniques.   The direct emission measurement
methods are not suited to sites that are heterogeneous.
                                      47

-------
        TABLE  11.  TYPES OF  IN-DEPTH METHODS  FOR  ESTIMATING EMISSIONS*
     §    Direct  Emissions Measurement

          -  -Surface  Emission  Isolation Flux Chamber
          -  Portable Wind Tunnels
          -  Soil Vapor  Probe
          -  Soil Vapor  Monitoring Well
          -  Downhole Emission Flux Chamber
          -  Vent Sampling
           Indirect  Emissions Measurement

           -  Concentration Profile
           -  Transect
          Air Monitoring/Modeling

          -  Concentration Profile
          -  Transect
          -  Exposure Profile
          Emissions (Predictive) Modeling

          -  Any model using site-specific values
          -  RTI and Scholl Canyon (Landfills)
          -  Thibodeaux, Parker, and Heck (Lagoons)
* See Section 4 for more detailed information.
                                      48

-------
     For those applications where the waste is not homogeneous, is
inaccessible, or consists of multiple sources which need not be individually
studied, the indirect methods are preferable to direct methods.  Total site
air emissions can be obtained and used to estimate BEEs and disturbed waste
emission data.  Indirect methods are susceptible to meteorological influences
and require analytical techniques with greater sensitivity than those used for
direct measurements.  Also, upwind interferences can create problems.

     Air monitoring/modeling methods are similar to indirect methods, but used
farther downwind.  They have the same applicability and limitations, except
that even  lower air concentrations can be expected farther downwind.  This
approach is not generally recommended unless fenceline data are required for
other needs.  Air monitoring/modeling methods are discussed in detail in
       IV  of this series of guidance manuals.
     Emissions  (predictive) modeling can be used; however, site-specific data,
particularly diffusion coefficient data, are often necessary to obtain
representative  BEEs.  These data may be difficult and expensive to gather.  In
addition, demonstrating the validity of the models for specific applications
may be difficult.   It is emphasized that site-specific data are required for
site characterization; the use of literature data should be limited, if
possible.

     Once the in-depth site characterization has been designed and conducted,
the BEEs should be evaluated and documented.  Data should be organized like
the screening assessment data, but these data are generally better estimates
of the emissions potential since more sophisticated technologies are used.

     If the baseline emission estimates are considered necessary and
sufficient to satisfy the project needs, then the field studies for the APA
are complete and these data are documented.  If the BEEs are considered
inadequate or additional  APA data are required, then follow-on site
characterization should be designed and conducted.
                                      49

-------
3.2  USE OF THE BEEs IN THE MITIGATION PROCESS

     Figure 8 illustrates how the BEEs fit  in  the CERCLA mitigation process.
Data uses for non-NPL sites will  probably be very similar to those identified
in Figure 8. 'BEEs generated in  the  remedial investigation can be used in two
ways.  First, BEE data provide information  regarding  the potential air quality
impact and health risk(s) posed  by the site should the  no-action remedial
alternative be selected.   Essentially,  this is the air  impact that can be
expected from the site over time  if  the site is  left  alone.   Thus, the BEEs
used in this way may help justify the  decision to mitigate the site.   These
data may also indicate the need  for  an immediate removal  action to protect the
    lie from possible air  contaminants.
                                     50

-------
Uo&tfkn* Emission
EstlmalM
• REMEDIAL INVESTIGATION •
1 J


UndisiuftMd
SM
Conduoru

1
i
i
!
OislurtXKi
Si.. .
Condaions'*


    Data indieal* in* Nc«d lor
  Imrmdial* Removal Adwn lo
     Protect Public tteaMt
      DM Oaia to Control
       Removal Action
        Au Emissions
   Data Input lor ihe
No Aaten AI«malM« in Iho
    FeasibMy Study
OMa Input to Owatopmg Emoaan Eitatuiut
Owing Ctem-4^an4 EvatuMng tw Ranunng
     R*m«dMt TcchnatogMt and AMugr
       • FEASttUTY STUDY •
                                                                    One* HoiMdul AIWnMv« Is Sattkxl
                                    AlkMiaiv* a< Choicu
                                  (HECCMOOf OEClStON)
                    Data Input lor Mitigation Thai
                    Oo«s Not Invotv* OisturtMd
                        SMCondMions
                                                                   N Huniimy. Duvutop an A« Momomi^
                                                                 Mudukng Pragiam lo 0 Poll Cleanup to Oc-lcrrruru
                                 Am^ranuni ol Chun up GojU
                                (OPERATION i MAIN It NANCE)
*As previously noted,  measurements of emissions  from exposed  or disturbed
 wastes can generally be performed  during  the RI  using the same  techniques
 used  for  performing  baseline emission  measurements.
                  Figure  8.   Use of  the  BEEs data  in  site  mitigation,
                                              51

-------
     Second, data for the disturbed  site  are  necessary  to  estimate short-term
impacts to air quality and risks  to  on-site workers  and neighboring residents
during cleanup.  These data assist in the evaluation of remedial  alternatives
considered in the feasibility study.  Remember that  this protocol  is designed
specifically *to develop BEEs (and disturbed emission estimates).   These data,
along with dispersion modeling, can  be used to assess off-site  impacts to air
quality.   Assessing off-site impacts is one of the topics  addressed in Volume
IV.
                                    52

-------
                                   SECTION 4
                      AIR EMISSION MEASUREMENT TECHNIQUES

     Section 4 describes recommended air emission assessment techniques
 (methods) for landfills and lagoons.  The recommended techniques are organized
 into four types of approaches: direct emission measurement techniques,
 indirect emission measurement techniques, air monitoring/modeling techniques,
 and emissions (predictive) modeling techniques.  Each type of approach is
 further divided into two classes:  screening methods, and in-depth assessment
 methods.  The screening methods provide some level of air emission assessment
 but may not accurately represent the site's potential for air emissions.  The
 in-depth assessment methods are much more rigorous and generally provide a
 more accurate estimate of the potential for air emissions from the site.
 Screening techniques are typically used in the site inspection stage of the
 RI, whereas in-depth assessment techniques are typically used during site
 characterization.

 4.1  GENERAL CONSIDERATIONS

     General considerations for the various air emission measurement
 techniques are given below before individual air monitoring and modeling
 techniques are described.  The methods are first compared,  followed by a
 discussion of the uses of the emission estimates.   A discussion of how to
 determine worst-case conditions for estimating emissions is then given.

 4.1.1  Comparison of Techniques

     A comparison and relative ranking of the air emission  assessment
 techniques are provided here to assist the user in selecting an appropriate
 assessment technique.  Comparisons and relative rankings for screening
 techniques are summarized in Tables 12 and 13.   A more detailed comparison of
 various predictive models is given later in Section 4.5.  Similar information
 for the in-depth techniques is presented in Tables 14 and 15.   The comparisons
of available techniques indicate the applicability of each  technique and the
                                      53

-------
                                   TABLE 12.  COMPARISON OF SCREENING TECHNIQUES
"Preferred"
Screening Techniques Technology
Head Space Samplers
Head Space Analysis
of Bottled Samples X
Upwind/Downwind X
Mass Balance
Real-Time Instrument Survey X
Predictive Models X
Typical Date Outputs
Applicable Species Indieiitm
Participate Potential Soil Vapor Ambient Air
Volatile Matter Emissions Flux Kate Concentration Concentration
X XX
X XX
X X
XXI K
XX X
Ln

-------
                                 TABLE 13.   RELATIVE RANKING FOR SCREENING  TECHNIQUES
in
Ui
Cost Parameters"
Screening Techniques
Head Space Samplers
Head Space Analysis
of Bottled Samples
Upwind/Downwind
Mass Balance
Real -Time
Instrument Survey
Predictive Models
Equipment
4
3
5
1
3
3
Manpower
1
1
1
1
1
1
Time
3
3
3
3
3
5
Analytical
5
5
8
10
1
1
Quality Parameters6
Accuracy
2
2
3
2
3
5
Complexity
*
8
8
6
10
10
5
Variability
2
1
3
2
4
5
        8  Ranking  is  on  a  1  to  10  scale  (1  being  lowest  cost,  10  being highest cost); Manpower  ranking
          is the number of people  required.
        b  Scale:   Accuracy
                  Complexity
                  Variability
1 = least accurate,  10 = most accurate
1 = most complex, 10 = least complex
1 = most variable, 10 - least variable

-------
TABLE 14.  COMPARISON OF IN-DEPTH TECHNIQUES
"Preferred"
Screening Technique Techniques
Emission Isolation
Flux Chanter X
Portable Wind Tunnel
Soil Vapor Probes X
Soil Vapor Monitoring Well
Downhole Flux Chamber X
Vent Sampling
Concentration Profile
Transect X
i_n Boundry Layer Emission
O\ Monitoring
Predictive Models X
Applicable Species Typical Data Outputs
Particulate Baseline Disturbed Soil Vapor Ambient Air Point Source
Volatiles Matter Flux Rate flux Rate Concentration Concentration Emission Rate
X X M
XX X X
X X
X MX
X M
X X
X X X X X
X X X K X
X X X X X
XX X X

-------
                                    TABLE 15.   RELATIVE RANKING FOR IN-DEPTH TECHNIQUES
On
Cost Parameter"
In-Oepth Techniques
Emission Isolation
Flux Chamber
Portable Wind Tunnel
Soil Vapor Probes
Soil Vapor Monitoring Well
Downhole Chamber
Vent Sampling
Concentration Profile
Transect
Boundry Layer Emission
Monitoring
Predictive Models
8 Ranking 1s on a 1 to 10
of people required.
b Scale: Accuracy 1
Complexity 1
Variability 1
Equipment Manpower Time Analytical
5
7
5
10
6
5
10
8
8
3
scale (1
= least
« most
= most
2
2
2
2
2
2
3
3
2
1
= lowest
accurate
complex,
variable,
5
5
3
7
7
2
10
10
10
5
cost, 10 =
, 10 = most
10 = least
10 = least
6
6
6
6
6
6
10
10
6
1
highest cost)
accurate
complex
variable
Quality Parameters'*
Accuracy
9
9
5
8
8
10
7
5
6
5
; manpower

Complexity
4
5
6
4
4
4
10
1
3
3
5
ranking 1s

Variability
5
5
3
4
3
10
3
3
3
5
the number


-------
type of output data that each technique will  provide.   Where possible
"preferred" techniques are identified and recommended  for use.  The preferred
techniques are not necessarily those with the highest  numerical rankings in
the tables, since not all  the ranking factors are equally important; rather
they are the*techniques that have proved to be the most cost-effective in past
field applications.  The preferred technique, however,  will  not always be the
best choice of assessment technique for every application.   The intent of
identifing preferred techniques is to assist  the RPM or site manager by
identifying those techniques that are thought to be preferrable for the
majority of sites.

     The type of output data that each technique provides varies
significantly.,   For example, some screening  techniques are  designed to
measure emission flux rates and ambient air concentrations,  while others will
simply indicate if there is a potential  for emissions  (i.e.  a qualtative
output).  Data output from the in-depth techniques can  also  vary.

     The techniques are also ranked according to their  relative cost (i.e.
equipment, manpower, time, and analytical)  and data quality  (i.e. accuracy,
complexity, and variability).  Among the screening techniques, upwind/downwind
monitoring has the highest relative cost while mass balances and predictive
models provide the lowest cost alternative.   However, as one would expect,
there is a trade-off between cost and data  quality.  The same is true of the
in-depth techniques.  The selection of an appropriate technique must be based
on the data requirements of the site inspection and site characterization
stages of the RI, as well  as on the cost of the available techniques.

     The use of screening versus in-depth techniques will  depend on the
ultimate uses of the data.  Screening techniques are generally employed at all
sites to assess the potential for air emissions or the  approximate levels of
contaminants to which on-site workers are exposed.  In-depth techniques,
because of their cost and comlexity,  are generally only employed at sites
where air emissions are considered likely to  be a potential  problem.
                                      58

-------
     One area of emerging technology is the use of remote sensing systems for
measuring background levels of emissions.  There has been only limited
evaluation of these techniques for Superfund or hazardous waste applications.
These techniques are still undergoing rapid development, but are generally not
sufficiently-sensitive yet to measure large numbers of volatile and semi-
volatile compounds at concentration levels of interest from a risk assessment
standpoint.  Therefore, these techniques are not included in Section 4.  An
overview of the currently available types of remote sensing systems, however,
is included as Appendix E of this manual.  Remote sensing has the potential to
become an important tool for estimating baseline emissions.

4.1.2  Uses of Emission Estimates

     The various monitoring and modeling techniques presented in this manual
provide either a relative indication of emissions potential (e.g. soil vapor
concentration) or an emission rate estimate.  The relative indications of
emissions potential are primarily useful as a screening tool.  The data can
yield useful information about the types of emissions at a given site, the
general magnitude of these emissions, and the spatial variability of these
emissions, i.e. the presence of "hot spots".  This information,  in turn,  is
input to the process of evaluating whether more detailed emissions data are
necessary.

     the emission rates generated from monitoring or modeling activities
generally provide data for a portion of the site or equivalent area (see
2.1.3) for a given time frame.  The primary use of such data is  for risk
assessment purposes to evaluate the potential effect of baseline emissions on
the health and safety of on-site workers and the surrounding populace.  For
on-site workers, the emissions from a given equivalent area are  usually of
greatest concern.  For off-site receptors, the summed emissions  from all  the
equivalent areas at the site need to be considered.  A simple arithmetic  sum
should be sufficient, but the time dependence of the emission processes should
be considered.  The maximum emissions for different equivalent units may  not
take place in all units during a given time frame.
                                      59

-------
     Emission rates are given in terms of mass of pollutants per unit  area per
given time frame.  Emissions over several different time frames may  of
interests including:

     1)  Maximum emissions during any 15-minute period;
     2)  Maximum emissions during any 60-minute period;
     3)  Average emissions during an 8-hour work period;
     4)  Average emissions during a 24-hour period;
     5)  Average annual emissions; and
     6)  Average exposure over a 70 year period.

The time frames of interest will depend on the objectives of the air pathway
analysis.  In general, the emissions will be compared to either health-based
standards as part of a risk assessment or to regulatory emisson standards  to
assess compliance.  The six time frames listed above include peak,  short-term,
and long-term periods as commonly defined in health-based standards and
emission standards.

     It is usually necessary to convert emission  rates to concentrations at
the receptors of interest, in order to assess health and safety concerns
arising from air quality impacts.   This conversion  must be made for conducting
quantitative risk assessments.   The most valid means to accomplish  this is to
use the emission rate as a source (input) term to an atmospheric dispersion
model.   The models identified by the EPA for volatile compounds and for
particulate matter are discussed in Volume IV of  this series (3).   The
downwind concentrations determined from the dispersion modeling should be
compared to any existing downwind ambient air data  as a check of the modeling
results.  Dispersion models are of limited use for  estimating dispersion over
very short distances,  such as may be the case for on-site personnel working in
close proximity to emission sources.   The results of the emission rate
estimates coupled with dispersion modeling are typically compared to
established action levels  to determine if the site  emissions pose a potential
health  or safety risk or used as input to a risk  assessment.
                                      60

-------
     The  advantage of using emission rate estimates rather than performing  air
monitoring at the locations of interest (e.g. fenceline) is the additional
information and flexibility the emission rate data provide.  Air monitoring
results are valid for the locations used for the specific metorological
conditions encountered.  Emission rates determined from monitoring or modeling
can be used to estimate ambient air concentrations under a wide variety of
meteorological conditions and for almost any receptor locations of interest.
The emission rates can also generally be modified to account for changes  in
the emitting area or the levels of emissions without the need for additional
field work.  As new information on the nature of the site becomes available,
the emission rates can be adjusted to reflect this additional information.
For example, if removal actions are instigated or a soil cover is added,  the
emitting  area or absolute rate can be changed to reflect these changes.
Additional data developed during the RI/FS process can also be used to modify
the inputs to predictive models for estimating emissions or their dispersion.
For example, if new receptors are identified, it would be simpler to run  new
model cases than to perform additional  ambient air monitoring to estimate the
concentration levels at the new receptors.

4.1.3  Determining Worst-Case Conditions

     The  desciptions of the monitoring and modeling techniques in the
following subsections include information on the limitations of the
techniques.  All of the techniques are only useful  for certain ranges of
conditions and will not provide data of acceptable uncertainty if applied to
situations that exceed the worst-case conditions.   The acceptable worst-case
conditions vary from technique to technique and must be determined from the
available literature and field experience.

     The monitoring techniques generally have specific validation criteria.
For the indirect monitoring techniques,  this involves  some set of
metorological  conditions that must be met.  For the direct monitoring
techniques, this usually involves some  evaluation  of the spatial  and temporal
variability in emissions to determine if the series of direct emission
measurements are adequate to determine  the average emissions for the entire
site or some portion of the site.
                                      61

-------
     The modeling techniques generally do not have specific  validation
criteria, but the validity of the modeling results are dependent  on  the
uncertainty associated with the model inputs.  Use of the models  requires
knowledge of the types of compounds present ind their concentrations in the
soil or water, along with information on the physical nature of the  site and
the applicable meteorological conditions.  The exact inputs  required vary
greatly from model to model.  The evaluation of whether or not the inputs are
adequate for modeling purposes is rarely clear-cut; the worst-case conditions
for modeling are generally not well defined.  Therefore, it  is generally
necessary that field data be collected to assess the validity of  the results
from predictive models.

4.2  DIRECT EMISSION MEASUREMENT TECHNIQUES

     A general discussion of direct measurement techniques is followed  by
descriptions of the individual  techniques.   The direct emission measurement
techniques presented in this section are:

     Screening Techniques--
     4..2.1     Head Space Sampler
     4.2.2     Head Space Analysis of Bottled Sample

     In-Depth Techniques--
     4.2.3     Emission Isolation Flux Chamber
     4.2.4    ^ Portable Wind Tunnels
     4.2.5     Soil  Vapor Probes
     4.2.6     Soil  Vapor Monitoring Well
     4.2.7     Downhole Chamber
     4.2.8     Vent Sampling

     Direct emission measurement techniques  are often the best techniques for
investigating the air pathway.   The emission rates  that  are  typically gener-
ated can be plugged into dispersion models  to predict ambient concentrations
at various  locations under varying meteorological conditions.  The techniques
generally  consist of isolating or covering  a small  section of the site  surface
or subsurface using  a chamber or enclosure.   The concentration of emissions
                                      62

-------
produced by the  isolated surface is measured within the chamber or from  an
outlet line.  These concentration measurements, along with other technique-
specific parameters, are then used to calculate an emission flux or relative
concentration value.  The emission flux (rate per area) can generally be
related to an emission rate for the entire source.

     The types of volatile or particulate species that can be measured by the
techniques are essentially unrestricted; their measurement depends on the
sampling media.selected and analysis technique rather than the emission
measurement technique.  However, few of the techniques are applicable to both
volatile and particulate emission rate measurement.  Selection of sampling
media and analysis techniques are addressed in Volume IV of this series.
Direct emission measurement techniques can be used to determine the emission
rate variability of a site by performing multiple measurements at selected
locations across the site.  In addition, these techniques allow for the
evaluation of individual waste areas at the site so that the investigation can
focus on those areas with the greatest potential emissions.

     The cost of the direct emission measurement techniques varies
considerably.  Most of the direct techniques, however, are cost-effective
relative to other approaches since several  measurements can usually be made in
a given day.  Real-time instruments can be used with all  the direct techniques
to provide immediate data for decision-making during the sampling program, and
for the relative ranking of the emission rate at locations across the site.
This procedure can be used to reduce the number of samples requiring
laboratory analysis by screening for those samples with significant
concentrations.

      Direct emission measurement techniques, as a class of assessment
techniques,  are generally preferable to other classes of techniques because
they have been proven to be a relatively cost-effective approach for obtaining
emission rate and concentration data and they avoid the necessity of modeling
to develop BEEs.   Direct emission measurement techniques and equipment are
generally relatively simple and straightforward.
                                      63

-------
4.2.1  Head Space Samplers (Screening Technique)
          space samplers can be used as a chamber to isolate part of the
emission source surface.(15,16,17)   The quantity or concentration of vapors
and/or gas emitted from the surface that build up in the chamber over a period
of time is measured, rather than measuring a rate.   The head space sampler
technique was a predecessor to the  emission isolation flux chamber described
in 4.2.3.

     Head space samplers may be operated in one of  two modes, referred to as
static and dynamic modes.  In the static mode,  the  sampling enclosure is
placed over the emitting surface for a given period of time.(15)  The
enclosure may be purged initially with clean air or nitrogen.  Surface
emissions then enter the chamber from the exposed surface and are allowed to
concentrate in the chamber before sample collection.

     A time-integrated emission flux for the static mode is calculated as:(15)

                E, - (C,  VE)/(t  A)                        (Eq.  1)

     where     E,  -  emission  flux for  component  i (ug/m2-sec);
               C1  -  concentration of component  i  (ug/m3);
               VE  -  volume  of the enclosure  (m3);
               t  - length  of time  enclosure is in  place (sec);  and
              'A  - surface area enclosed by chamber  (m2).

     The build-up of gas species within  the chamber improves the sensitivity
of the method relative to ambient air  sampling.   However,  the accuracy of the
calculated emission rate is dependent  on the duration of sampling relative to
the time required to reach  steady-state  concentrations  within the chamber.
For long sampling periods,  the  concentration gradient of the soil/air
interface is reduced and the emission  flux is underestimated.   Also,
instantaneous changes in the flux cannot be measured.
                                      64

-------
     In the dynamic mode, the sampling enclosure also is placed over the
emitting surface for a given time period and the chamber may be initially
purged with clean air or nitrogen.  However, collected emissions are
continuously withdrawn from the enclosure.(15)  The chamber can be operated
with a second port allowing ambient air to enter the chamber to prevent a
negative pressure within the chamber.  The emitted species is concentrated on
sampling media or may be continuously monitored.  When the sample is
concentrated on the sampling media (i.e., sorbent), the emission flux is
calculated as:(15)

                E, -  (C, V,)/(A t)                       -(Eg.  2)

     where     Ei - emission flux for component i  (ug/m2-sec);
               Ci * concentration of component i (ug/m3);
               Vs * total  volume of sample withdrawn  (m3);
               t  = length of sampling interval (sec); and
               A  = surface area enclosed by chamber (m2).

     An advantage of the dynamic mode is that the sampling duration and air
sampling rate can be varied to adjust the volume of air sampled in order to
achieve the required analytical sensitivities.  The disadvantage of operating
in the dynamic mode is that as the atmosphere within the enclosure is
withdrawn, the emission flux value may be affected.  This can occur by the
addition of bulk flow of the soil gas into the chamber or,  alternately,  air
entrainment occurring within the enclosure because of leakage at the
enclosure's bottom edge, or by air moving through the soil  at the enclosure's
bottom edge.  If a second port is used to allow atmospheric air to enter,  a
means of removing volatiles from the atmospheric air must be used.

     The major advantage of the technique is that the method is simple.   The
enclosures are typically 55 gallon drums sawed in half lengthwise or a
similarly simple enclosure.   The method is independent of meteorological
conditions (except for changes in barometric pressure).   The analysis can
involve sampling the headspace with a portable analyzer after specified time
intervals.   The field crew can consist of a single technician.
                                      65

-------
Applicability--
     The emission isolation flux chamber (see Section 4,2.3) is preferred  to
head space samplers except where the emission flux is expected to be extremely
low.  Under these conditions, use of head space samplers in static mode  for
long intervals (possibly days) may allow emission fluxes to be measured  by the
head space sampling technique where other techniques may fail to produce a
measurable emission flux.

     Head space samplers are applicable to emission flux measurement from  all
forms of solid area sources including landfills, open dumps, and waste piles,
and for homogenous, quiescent lagoons.  The technique can be used at open  and
closed landfills, with or without internal  gas generation.   The technique  can
be used to assess emission rites from cracks in the surface cover and from
vents that have minimal or no volumetric flow.  The technique is applicable
both for undisturbed and disturbed site conditions.
   ritations--
     As previously mentioned, head space samplers may enhance (dynamic mode)
or suppress (static mode) the emission flux.  The dynamic mode also may be
subject to air entrainment resulting in deceptively low emission flux values.
The technique does not assess the effects of wind speed on the emission flux.
For lagoons, removal of air from the chamber will induce a pressure change
unless make-up air is able to enter the chamber.  The chamber will  require
some type of support of flotation system, and this may be affected by waves or
agitation.

     Head space samplers are not applicable to the measurement of particulate
emission fluxes.  Also, no validation of the technique has not been reported.

4.2.2  Headsoace Analysis of Bottled Samples (Screening Technique)

     Headspace analysis of bottled sample is a preferred technique for field
screening of wastes and soils to determine their relative emissions potential.
In this method, liquid, soil, or waste from the site surface or subsurface is
collected.  The material is immediately placed in a sampling container,
typically 1-liter or a 40 ml volatile organic analysis (VOA) vial  with septa;
                                      66

-------
and the container sealed.  Transferring the soil or waste immediately after
collection into the sample container to prevent loss of volatiles is the key
to successful use of the technique.  The container is allowed to stand for a
given period of time, typically 5 to 30 minutes.  The container lid is then
cracked open-and the probe of a field instrument is inserted to determine if
soil-vapors are present.  Syringes can be used to withdraw gas samples via
parts of the VGA vial septa for more sophisticated analysis techniques.
Typical field instruments used include portable flame ionization detectors,
photoionization detectors, combustion meters, and colorimetric tubes.

     A second type of headspace analysis involves analyzing the headspace gas
or extracted solids of a soil waste core.  To obtain a sample to measure the
headspace, an undisturbed core is collected using an auger or by driving a
tube into the ground.  The sample is then sealed in a sample container with
minimal headspace.

     The core sampler shown in Figure 9 consists of a brass core sleeve which
is pressed into the soil to a sufficient depth to fill  the sampler but not
compress the sample.  After excess soil is removed, the sleeve is sealed with
a Teflon-lined cap.  The samples are stored at room temperature.  Headspace
gas is removed from the core by a syringe and analyzed by GC.

Applicability--
     The headspace analysis is a useful screening technique for all  types of
soil or waste which have a volatile component.  Some experimentation may be
required to determine the optimal  time to allow for volatilization to  occur
before measurement.  The technique can be used to identify surface contamin-
ation boundaries,  select sampling locations for detailed techniques, and
identify health and safety concerns.

     The technique is best suited for measuring adsorbed organics, although it
can also be effective with free organics in the pore space if the material  is
rapidly transferred to the sample container,  or.if tube or core samplers are
used.
                                      67

-------
.. WINS NUT
                       OtOCAf
                                         t BRASS cone SLEEVE i
                             TEFLON
                     TEFLON CA^ UNEH
    Figure  9.  Schematic diagram of  a soil  core sample sleeve,
                                    68

-------
Limitations--
     The technique does not provide for calculation of an emission rate, but
rather identifies soils and wastes which are potential source of air
emissions.  The technique generally only provides qualitative data on species
type, however, species specific data can be obtained if suitable analytical
techniques are selected.  The technique is only applicable to undisturbed
conditions.  The technique is not applicable to particulate emissions.

Preferred Technique--
     Headspace analysis of bottled samples is one of the two preferred
techniques for the emission screening study.  This direct approach is
preferred because it is simple to implement and effective at identifying
volatile content, which represents volatile emissions potential.  Chemical
analysis of thi wiste will identify the potential for contaminated particulate
matter emissions.  Thus, the headspace analysis and the analysis of the waste
can provide effective screening of emissions potential.  The other preferred
screening technique is an indirect technique, simple upwind/downwind sampling.

4.2.3  Emission Isolation Flux Chamber (In-Depth Technique)

     The emission isolation flux chamber is one of the preferred in-depth
techniques for the direct measurement of volatile species emission rates.(18-
23)  The technique uses a surface enclosure (flux chamber)  to isolate a known
surface area for emission flux (rate per area)  measurement.   The emission
isolation flux chamber for solid surfaces is illustrated in  Figure 10 and for
liquid surfaces in Figure 11.

     Emissions enter the open bottom of chamber from the exposed surface.
Clean, dry sweep air is added to the chamber at a metered rate.   Within the
chamber,  the sweep air is mixed with emitted vapors and gases by the physical
design of the sweep air inlet.  The sweep air creates a slight wind velocity
at the emitting surface, preventing a build up  of the emission concentration
in the boundary layer directly above the surface.  The exit  port is used for
measurement of the concentration of the air within the chamber or for sampling
and subsequent-*analysis.  A pressure relief port in the enclosure prevents
pressure  build up within the chamber that might otherwise occur when sampling
                                      69

-------
                                            THEflMOCOUPt£
                                                                    SYRINGE/ SANlSTEfl
                                                                     SAMSUNG PONT
CABBIES
  GAS
STAINLESS STEEL
 OR PLEXIdLAS
                                    CUT AWAY TO SHOW
                                   SWEEP AIM INLET UNE
                                   ANO TMf OUTLET UN£
   Figure 10s  A cutaway  diagram of the emission isolation flux  chamber
   	and support  equipment for solid  surfaces.
                                        70

-------
                                                SAMPiE CQiUCTION
                                                 ANOiOA ANALYSIS
                                       THiAMOCOUfHJ
Figure 11.
A cutaway diagram of the surface emission  isolation flux
chamber and  support  equipment for liquid surfaces.
                                  71

-------
liquid surfaces.  A positive or negative pressure in the enclosure  could
affect the emission event and, thus, the assessment of the emission flux.   For
lagoons, a support or flotation system is necessary.

     The technique directly measures essentially an instantaneous emission
flow (flux) from that surface.  The emission flux is calculated from the
surface area isolated, the sweep air flow rate, and the emission
concentration.  Statistical methods are used to determine the number of
measurement locations required to characterize the emission from an  area
source.  These methods are based on the surface area of the source  and the
variability (precision) of the measured emission rate at randomly selected
locations across the site.  Use of the emission isolation flux chamber is
described in the draft "Measurement of Gaseous Emission Rates from  Land
Surfaces .Using an Emission Isolation Flux Chamber User's Guide (23)."  The
emission isolation flux chamber was validated for EPA for measuring  volatile
emissions from landfills.(22,23)

     The emission flux is calculated as (23):

                               E,  - (C, Q)/A                          (Eq.  3)

     where     E1  - emission  flux  of component  i  (ug/m2-min);
               Cf  = concentration  of component  i  at  chamber  outlet  (ug/m3);
               Q  - sweep air flow rate into chamber (m3/min);  and
               A  - surface area enclosed by chamber (m2).

Applicability--
     The emission isolation flux chamber is applicable to emission flux
measurement from all types of area sources including lagoons,  landfills, open
dumps,  and waste piles.   The  technique can be used at open and closed
landfills,  with or without internal gas generation.   The technique can be used
to assess emission rates from cracks in the surface  cover and  from vents that
have minimal  or no volumetric flow.
                                      72

-------
     The technique  is  applicable both for undisturbed and disturbed site
conditions, and for the testing of emissions control techniques.  The
technique can be used  to satisfy data needs for all phases of the RI/FS
process, as well as post-remediation monitoring and, therefore, can provide
directly comparable data throughout the process.

Limitations--
     The emission fluxes of volatile species may be enhanced or suppressed
since the flux chamber alters the environmental conditions (e.g., wind speed)
at the sampling locations.  The technique does not assess the effects of wind
speed on the emission  rate.

     The technique  is  not particularly well suited to large emission sources
with a high degree  of  heterogeniety.  Comparison of sample data and
variability estimates  can determine the number of sampling locations needed to
determine representative area emissions.  Also, the technique is not
applicable to the measurement of particulate emission fluxes.

Preferred Technique--
     The emission isolation flux chamber is a preferred technique for
developing in-depth BEEs.  As discussed, the advantages generally out-weigh
the limitations of  the technique.  The technique is well documented in the
user's guide and well characterized.

4.2.4  Portable Wind Tunnels (In-Depth Technique)

     Wind tunnels are  in-depth techniques used to  directly measure the
emission rate of erodible material.  They also can be used to measure the
emission flux of volatile compounds.  In either application,  measurements can
be made under varying wind conditions to examine the effect of wind speed on
emissions.   The required equipment consists of portable, open-bottomed
enclosures  used to  isolate a known surface area, a blower used to simulate
wind conditions,  and sampling devices.
                                      73

-------
     The Cowherd wind tunnel, shown in Figure 12, is a portable wind tunnel
developed for "in situ measurement of emissions from representative test
surfaces under predetermined wind conditions"(24).  The tunnel was developed
to measure particulate matter emissions from open waste piles.  "The
open-floored"test section of the portable wind tunnel is placed directly over
the surface to be tested.  Air is drawn through the tunnel at controlled
velocities.  The exit air stream from the test section passes through a
circular duct fitted with a sampling probe at the down-stream end.  Air is
drawn through the probe isokinetically by a high-volume sampling train"(24).
The sampling train consists of a trapper probe,  cyclone precollector,
parallel-slot cascade impactor,  backup filter,  and high volume motor.   Air
flow is provided by a blower located downstream of the sampling train.

     The authors state that, "although a portable wind tunnel does not
generate the larger scales of turbulent motion found in the atmosphere,  the
turbulent boundary layer formed  within the tunnel  simulates the smaller scales
of atmospheric turbulence.  It is the smaller scale turbulence which
penetrates the wind flow in direct contact with  the pile surface and
contributes to the particle entrainment mechanisms"(24).

     The emission flux is calculated from the isolated surface area, emission
concentration, simulated wind speed, and time period during which particulates
are collected. Because the emission concentration is collected over time,  the
technique measures the overall (time integrated)  emission rate rather than the
emission flux.  Varying the simulated wind speed  between measurements  allows
for development of a weighted average emission rate.   This is preferred to
using an average wind speed because the total  erosion may be greatly
influenced by infrequent periods of high wind speed.

     The loss of erodible material is calculated  as (24):

                 E, - (C,  Q t)/A                      (Eq.  4)
                                      74

-------
Figure 12.  Illustration of MRI wind tunnel
                      75

-------
      where     E, = emission rate of component i (g/m2);
                C, - average participate concentration of component  i  in tunnel
                    exit  stream  (g/m3);
                Q  » tunnel  flow  rate  (sn3/sec);
             *   t  » duration of  sampling  (sec); and
                A  = exposed test area  (m2).

      Elimination  of the time factor in the calculation will  provide an
 emission  flux  on  a unit area per time basis  (g/m2-sec).  The average
 particulate  concentration  (C4)  may be reduced to account for background dust
 levels  by sampling under  light wind conditions and subtracting  the  resulting
 average concentration  from C,  values generated during simulations of  higher
     The Astle wind tunnel, shown in Figure 13, is a form of surface  enclosure
developed for "measurement of odor source strength"(25); but it  also  may be
applicable to volatile emissions measurement.  This portable wind  tunnel
consists of an open-bottom enclosure that is placed over the emitting surface.
Ambient air is blown through the chamber at typical wind speed rates  (e.g.,  1
to 15 mph) and collected near the enclosure outlet.  Test results  for volatile
emission rate measurement were not identified in the literature.

     The emission flux is calculated from the simulated wind speed, emission
concentration, and surface area isolated.  Varying the wind speed  between
measurements allows for development of a weighted average emission rate.   The
measurement is essentially instantaneous.

     The emission flux calculation is (25):

                  E, - (C,  Q)/A                        (Eq.  5)

     where     E,  = emission  flux of component  i  (ug/m2-sec);
               C,  « concentration of component  i  (ug/m3);
               Q  - air flow rate through tunnel  (m3/sec);  and
               A  - surface area enclosed by the tunnel (m2).

                                      76

-------
                                                         Activated
                                                         Carbon Filter
          Variae to Control
             Blower Speed
                                                         Air Inlet
Sampling Probe for
 Trench Samplers
                                          .Lead Weight to Aid Seal
                    '3-inch thick Acid-Assistant
                        Foam Cushion for Seal
          Figure 13.   Schematic of portable  wind  tunnel
                                77

-------
Applicability--
     Wind tunnels are applicable to emission measurements from all forms of
solid area sources, including quiescent lagoons, landfills, open dumps, and
waste piles.  The technique can be used at open and closed landfills, with or
without internal gas generation.  The technique is applicable both for
undisturbed and disturbed site conditions.

Limitations--
     The portable wind tunnels do not account for macro-atmospheric effects on
the emission flux.  Repeated measurement at a given location may deplete the
reservoir of erodible material.

4,2.§  Soil Vapor (Ground) Probes (In-Depth Technique)

     Soil vapor or ground probes are a preferred in-depth technique for
mapping the horizontal extent of soil gas plumes in near-surface soils
(26,27,28).  The technique directly measures the soil  gas concentration at a
given depth.  As such, it can provide a relative estimate of the emissions
potential of the subsurface soils.  One variation of a  ground probe is shown
in Figure 14.  Probes with small chambers or openings  at the probe end have
also been used (see Figure 15).  Reference 27 includes  descriptions of a wide
variety of ground probe designs.

     The probe assembly is driven into the ground to the desired sampling
depth.  Emissions enter the probe from the exposed subsurface soil, depending
on the design via existing openings or by raising the  pipe away from the drive
tip.  Generally, either a small aliquot of soil  gas is  collected to avoid
disturbing the soil gas equilibrium or a known amount  of soil-gas is pumped
from the probe.  The concentration measured is used as  a relative indicator of
contamination and potential  emissions.  Clean, dry sweep air can be added to
the probe at a metered rate if a relative emission "flux" is desired.  The
flux cannot be measured with any degree of confidence  since the exact surface
area of exposed waste is not known.
                                      78

-------

                       • 4
                             Ortvta«:  ria 4Ua«c«r

                                         0.3.
                       '•\/•':
                        •\i«  • 3
                              •«
                             • i


                             .4
                                kl
  Mil «BCT7 ea pipe.


         «

•4 UMfUISC;  Tu4« lj ratj«4 2 la.

           ea ie^<«« toil
                              X  *  !
                              ••^  !
Figure 14.   Schematic diagram of a simple ground probe.
                               79

-------
                               syringe

                               3 way valve
                               Tenax GC  trap
                               fitting
                                      pounding plate
                              coupling

                              air holes

                              point
Figure  15.  Ground  probe design with minimal  internal vol
                                                     ume.
                           80

-------
     The major  advantage  of the technique is that it allows for rapid mapping
of the horizontal extent  of soil gas plumes in near-surface soils.  Generally,
mapping is performed  by measuring the soil gas concentration without use of
sweep air.  The mapping can be used to determine the approximate subsurface
boundaries of buried  waste or immiscible liquids floating on the water table.
Ground probes are very useful for investigating the migration of soil vapors
as part of the  air  pathway analyses.

     It is not  necessary, although desirable, to know the waste composition to
use the technique.  Knowing the waste composition will help in the selection
of appropriate  instrumentation or sampling apparatus.
     The technique can be used to map soil gas plumes to indicate approximate
waste boundaries or  soil gas migration routes.  Ground probes are also
applicable to the measurement of potential volatile emissions from disturbed
subsurface soil and  waste.  The technique measures the emissions that would
occur if the subsurface soil or waste were disturbed by excavation.  The
technique also can be used to measure the soil gas concentration by operating
the probe without sweep air.  Soil gas concentration measurements can be used
as an input to predictive models, as discussed in Section 4.4.

     The technique is applicable for estimating emissions from solid waste
sites, including open and closed landfills, with or without internal gas
generation, open dumps, and waste piles.  The technique also can be used to
estimate emissions below the edges of tanks,.buildings, ponds, etc., by
driving the probe into the soil at an angle.  In addition,  ground probes can
be hand- or machine-driven.  Hand-driven probes allow measurements to be
performed in areas where surface access by machines is not  available.

Limitations--
     Ground probes do not measure the undisturbed emission  rate or flux from a
waste site.  Rather,  the technique identifies the relative  concentration of
vapors in the sub-surface.  This information can be used to estimate locations
that may have*significant air emisisons during any soil excavation or other
site disturbance.
                                      81

-------
     Use of ground probes is generally limited to near-surface soils,
typically less than 12 feet deep.  However, the operational depth will depend
on site and equipment characteristics which limit the ability to insert  the
probe.  Ground probes do not measure particulate emission rates.
            V
4.2.6  Soil Vaoor Monitoring Wells fln-Deoth Technique)

     Soil vapor monitoring wells (29) are used to measure the emission flux
from subsurface soil and waste and to monitor soil  vapor concentrations  and
the effects of soil vapor remedial actions.  The in-depth technique uses a
monitoring well consisting of a screened chamber installed during drilling
activities.  As such, the soil vapor monitoring well  is a permanent or
semi-permanent structure.  Figure 16 illustrates a typical soil  vapor
monitoring well.

     Soil vapor monitoring wells are installed through the annulus of a
hollow-stem auger.  The well consists of an enclosed,  screened chamber with
sampling lines leading to the surface.   The chamber is enclosed  in a sand
pack.  The well casing is used to provide support during well  construction and
protection for the sampling lines, but soil vapors  and gases do  not contact
the well casing.  Soil vapors and gases enter the soil vapor monitoring well
through the chamber screen.  Within the chamber,  the  vapor and gases are mixed
with clean, dry sweep air.  The resulting mixture is  withdrawn through the
exit port to measure the emission concentration and/or for sampling and
subsequent analysis.  The sweep air flow rate is selected to approximate the
instrumentation or sampling apparatus flow rate.

     Because the measured emission concentration is directly related to an
emissions event from an isolated surface over an essentially instantaneous
time period, the technique can be used to estimate  the emission  flux from the
isolated soil  surface.  The estimated emission flux is calculated from the
assumed surface area isolated, the sweep air flow rate,  and the  emission
concentration.
                                      82

-------
                   mO, WMUT
                 «««- 8T««H»« 9T8RL
                                    i i
                                 1
                                    Is
                                             «»fl» TIQHf
                                                   STIM,
                                           • ITAM4.KS* STCIL
                                           r* i 3«" SCMCM
                                           WCkMO BOTTOM m,Aft
Figure 16.   Vapor  monitoring well  constructions.   (Not  to scale)
                                   83

-------
     The emission flux is calculated as (26):

                             E, = ^

     whirs  .  Ei - emission flux for component i  (ug/m2-min);
               C, * concentration of component i  (ug/m3);
               Q  - sweep air flow rate (m3/m1n);  and
               A  - exposed surface area (m2) » ffdh
              -d. - screen diameter (m); and
               h « height of sand pack (m).

     Typically, soil vapor monitoring wells are used to measure the soil vapor
concentration, rather than the emission flux.  The soil vapor concentration is
measured by operating the well without sweep air.

     The technique is not dependent on knowledge  of the emissions process or
the waste composition, although knowledge of the  waste composition will help
in selecting appropriate instrumentation and/or sampling apparatus.
    lieability--
     Soil vapor monitoring wells are applicable to the monitoring of soil
vapor concentrations and viper migration, and are the preferred method for
monitoring the effects of soil vapor extraction systems.  Soil vapor
monitoring wells may also be used to estimate the potential emission flux from
subsurface soil and waste in the same manner as the soil vapor probes
discussed in the previous subsection.

     The technique is applicable to the estimation of emission fluxes and
concentrations from open and closed landfills, with and without internal gas
generation, and open dumps.  The technique also is applicable to measurement
of emission concentration from immiscible liquids floating on the water table.
                                      84

-------
Limitations--
     Soil vapor monitoring wells do not measure the undisturbed emission flux,
but rather the flux that would occur during site disturbance.  The technique
is not applicable to the measurement of particulate emission fluxes.  The
actual exposed surface area is an assumed or estimated value.

4.2.7  Downhole Emissions Flux Chamber f!n°Depth Technique)

     The downhole emissions flux chamber is one of the preferred in-depth
techniques for direct measurement of potential volatile emissions from
subsurface soils (30).  The technique uses a flux chamber to measure emission
fluxes from subsurface soils exposed by drilling operations.  The chamber
isolates a known area of soil at a desired depth within the annul us of a
hollow-stem auger.  Figure 17 depicts the downhole emissions flux chamber.

     Emissions enter the chamber from the exposed surface.  Clean,  dry sweep
air is added to the chamber at a metered rate.  Within the chamber, the sweep
air is mixed with emitted vapors and gases by the physical design of the air
inlet.  The mixture of sweep air and emitted vapors and gases is withdrawn
through the exit port for measurement of the emission concentration or for
sampling and subsequent analysis.  The sweep air flow rate must match the flow
rate of the analytical instrumentation or sampling apparatus that is used to
withdraw sample gas.

     Because the measured emission concentration is directly related to an
emissions event from an isolated surface over an essentially instantaneous
time period, the technique directly measures the emission flow (flux) from the
surface.  The emission flux is estimated from the assumed surface area
exposed, the sweep air flow rate, and the emission concentration.

     The emission flux is calculated as (30):

                                                                      (Eq. 7)
                                      85

-------
            ,23 i». in-
0.25 la. ouesue
 uwt.Tiw.j3N
       7  isu  lesge,h  of
                                //  /
                                    //
                                1
Figure 17.  Schematic diagram of the downhole  emissions flux chamber.

-------
     where     E1 - emission flux of component i (ug/m2-min);
               C, = concentration of component i (ug/m3);
               Q  - sweep air flow rate (m3/min); and
               A  » exposed surface area (m2).

     The major advantage of the technique is that it allows the investigation
of subsurface areas without excavation.

     Although desirable, knowledge of the waste composition is not necessary
to use the technique to assess the air pathway.  However, knowledge of the
waste composition will help in the selection of appropriate emission
concentration measurement instruments.

Applicability--
     The downhole flux chamber is applicable to measurement of the potential
emissions from subsurface soil and waste.  The technique estimates the
emission flux that could occur if the subsurface soil or waste were exposed by
excavation.  Therefore, the technique is most suitable for determining
emissions potential for remedial alternatives evaluation and/or site
disturbances.  The technique is useful for characterizing the volatile species
present in subsurface waste or soil layers.  The technique also can be used to
measure the soil gas concentration by operating the chamber without sweep air.
However, for shallow depths, ground probes are preferred for gas concentration
measurements (see Section 4.2.5).

     The technique is applicable for estimating emission fluxes from all
materials that can be investigated using hollow-stem auguring techniques.  The
technique can be used at open and closed landfills, with or without internal
gas generation, and at open dumps.

Limitations—
     The downhole emissions flux chamber does not measure the baseline
emission flux from a waste site.  Rather, the technique measures the potential
emission rate that could occur during site disturbance.
                                      87

-------
     The emission flux of volatile species may be enhanced or suppressed if
the sweep air flow rate does not closely match the flow rate of the  sample gas
extraction system.  The downhole flux chamber cannot be used to measure
particulate emission rates.

Preferred Technique--
     The downhole emissions flux chamber is a preferred technique for
characterizing potential emissions from subsurface disturbed waste.  Most
investigationsvinvQlve subsurface sampling using a hollow-stem auger and drill
rig and downhole flux chamber work can be incorporated in the investigation.

4.2c8  Vent Sampling (In°Depth Technique)

     Vent sampling for emissions measurement has been well  documented else-
where (32), and will not be discussed in detail  here.  Vent sampling is
performed when the waste site contains vents with measurable flow rates  (i.e.,
active venting, not passive venting), such as vent systems  at some landfills.
Tht in-depth technique requires measuring the emission concentration and the
volumetric flow rate, typically as the exhaust velocity and cross-sectional
area of the vent.  Volumetric flow rates for vents (or ducts) can be obtained
using the procedures given in the Code of Federal  Regulations (CFR) Title 40
Pirt 60.   Those procedures indicate how to determine  the exhaust velocity and
appropriate sampling location.   The emission rate for a vent is  calculated as
(31):

                  E1 - C,  U  A                        (Eq. 8)

     where     E1  »  emission rate  of  component i  (ug/sec);
               Cf  «  concentration  of  component i  (ug/m3);
               U  -  gas velocity through vent (m/sec);  and
               A  -  cross-sectional  area of vent (m2).

-------
Applicability--
     Vent sampling is applicable to any waste site that contains active or
passive venting systems.  However, active venting systems are not typical of
uncontrolled hazardous waste sites.  Vent sampling can be applied to landfill
vents, tanks; building vents, machinery, and equipment.  Where present, vents
may be a major source of air emissions from the site.

limitations—
     Vent sampling using standard stack sampling technique is not applicable
when the vent has minimal or no flow.  For these situations,  the emissions
isolation flux chamber technique is preferred,  provided that  the chamber's
cross-sectional area is larger than the vent's.   When the vent is too large
for use of the flux chamber, emission rate estimates may be based on head
space concentration measurements at the vent outlet and diffusion in air
4.3  INDIRECT EMISSION MEASUREMENT TECHNIQUES

     A general discussion of indirect measurement techniques is followed by a
description of specific techniques.  The indirect emission measurement
techniques presented in this section are:

Screening Techniques--
     4.3.1  Upwind/Downwind
     4.3.2  Mass Balance
     4<,3.,3  Real-Time Instrument Survey

In-Depth Techniques—
     4.3.4  Concentration-Profile (C-P)
     4.3.5  Transect
     4.3.6  Boundary Layer Emission Monitoring
                                      89

-------
      Indirect emission estimation techniques generally consist of measuring
the ambient concentration of the emitted species and then applying  these  data
to an equation  (air dispersion model) to determine the emission rate.   Many  of
the equations were developed to determine downwind concentrations resulting
from  stack emissions.  For area emission sources, the source is treated as a
virtual point source or line source.

      The in-depth techniques are very similar as all involve clusters of
ambient air samplers..  The C-P technique involves a vertical array of samplers
directly over the source.  The transect technique involves vertical and
horizontal arrays of samplers within the downwind plume.   The boundary  layer
technique is a  simplified version of the transect technique and involves
several downwind samplers each at a different height.  Because of their cost
and complexity, the Indirect, in-depth techniques are usually only used for
measuing emissions of very toxic compounds,  and then only if direct
measurement techniques prove to be unsuitable.

     A disadvantage of indirect emission measurement techniques is that the
results are highly dependent on meteorological  conditions.  The indirect
techniques require meteorological  monitoring to properly  align the sampling
systems and to analyze the data following sample analysis.  Changing
meteorological conditions significantly affect  the liklihood of collecting
useful data.  Unacceptable meteorological conditions may  invalidate much of
the data collected, requiring an additional  sampling effort.  The techniques
also may product false negative results if the  emitted species are present in
low concentrations which are below the sampling and analysis detection limits,
or if upwind sources cannot be fully accounted  for.  The  techniques also may
not be feasible at some sites where the source  area is excessively large,  or
where insufficient space exists downwind of the source to set up the sampling
array without disturbance of tha air flow pattern by obstructions (e.g.,
buildings,  tanks).

     The types of volatile and particulate species that can be measured by the
techniques  are essentially unrestricted and  are dependent on the sampling
media selected and analysis technique rather than the emission measurement
technique.   Sampling media and analysis techniques are not discussed here.
                                      90

-------
        Indirect emission measurement techniques generally do not provide
   significant data on the emission rate variability for different locations
   across a site.  This is because the emission concentration is measured
   downwind of the site after some atmospheric mixing.   The techniques generally
   do not allow- for the evaluation of individual  contaminated areas at the site
   unless the treas are separated from one another and  are not located upwind of
   one anothero
lt
        The costs j»f the indirect emission measurement  techniques vary
   considerably.  The screening techniques are relatively simple and straight-
   forward to implement, and require minimal  labor and  analytical  costs.   The in-
   depth techniques are complex and require considerable equipment,  labor,  and
   analysis costs.   All of the techniques are subject to data loss or sampling
^  delay due to inappropriate meteorological  conditions.

   4.3.1  Upwind/Downwind (Screening Technique)

        The upwind/downwind technique is an indirect screening technique  (33).
   As the name implies, in this approach one  monitor (or set  of monitors}  is
   located upwind of the area source and a second monitor (or set of monitors) is
   located downwind.  The monitoring stations include detectors or samplers for
   thi species of interest as well  as devices for measuring wind speed and
   direction.  The upwind monitor serves as a blank or  background sampling
   location.  Concentration is measured primarily along the downwind axis  only.
   The average surface emission flux for a particular trajectory is  equal  to the
   increase in column concentration (downwind minus upwind) divided  by the
   transit time across the source.   Transit time  is a function of the distance
   between the source and downwind location,  and  the average  wind velocity.

                              E.R.I - (C0-Cu) ™y °z °                     (E<1- 9)

        where     E.R.i » emission flux of species (ug/m2-sec);
                  C0 -  downwind concentration of  species  1  (ug/m3);
                  Cy -  upwind  concentration of species  i  (ug/m3);
                   * - 3.141 ...
                                         91

-------
               ay - lateral extent of Gaussian plume;
               az = vertical extent of Gaussian plume; and
                 0 - mean wind speed (m/sec).

Applicability--
     The  upwind/downwind technique is applicable to emission flux measurement
from all  forms of area  sources, including lagoons, landfills, open dump,  and
waste  piles.  The technique can be used at open and closed landfills, with  or
without internal gas generation.  The technique is applicable both for
undisturbed and  disturbed site conditions, and for testing emission control
techniques.  The technique can be used for both volatile and particulate
emission  flux assessment.

     The .method  is most frequently used when an approximate emission rate
needs  to  be determined, and the only existing air data are ambient
concentrations from a few upwind and downwind locations.  Thus the emission
rite can  be modeled without additional field measurements being performed.

Limitations--
     Upwind/downwind requires that meteorological  conditions during sampling,
particularly wind speed and variability, match the predetermined conditions
used to select the sampling locations.  The sampling locations must be on the
approximate plume centerllne.

     The technique also assumes that the site is fairly homogeneous and
that the plume is well mixed at the downwind sampling location.   Therefore,
the technique may not be applicable to heterogeneous sites.  The technique
also may not adequately collect emissions from point sources within an area
source, such as cracks in landfill covers or vents,  unless the plume is well
mixed at the downwind sampling location.  The technique is not applicable
during quiescent or unstable wind conditions, and  may produce false negative
results during these conditions.
                                      92

-------
Preferred Technique--
     The upwind/downwind screening technique is a preferred indirect technique
for indirectly screening the air impacts for a site.  The technique is similar
to a simple real-time inspection survey, however, it is superior in that it
specifies data collection consistent with a dispersion model (e.g. ISC model)
so that emissions can be estimated from a variety of area sources.

4.3o2  Mass Balance (Screening Technique)

     A mass balance technique can be used to indirectly determine overall,
long-term emission rates by accounting for material  in and out of a system and
assuming the difference is lost as air emissions (31),,  The use of mass
balance for baseline emission estimates at uncontrolled landfill or lagoon
sites was not identified, but it might be applicable to waste lagoons that
have minimal leaching losses.  To apply the technique at an uncontrolled site,
the concentration of the species contained in the lagoon (or landfill)  would
be measured infrequently over time along with the flowrate and concentration
for any influent and effluent streams, and the emission rate would be
estimated as the loss of species over time.

Applicability--
     Thi technique would be best suited for homogenous sources containing
highly volatile wastes.  Application of the technique to "fresh waste", when
emission rates are typically highest, is more feasible than application to
"weathered" waste.- The technique does not appear applicable to particulate
matter emission assessment.

Limitations--
     The technique is generally suited only to process applications.   It is
unsuited to uncontrolled waste sites because of the  source types present and
since the losses of material  are difficult to identify due to the imprecision
of the sampling and analytical  methods.  In addition, the mass balance
technique does not distinguish between material lost to other pathways.
                                      93

-------
4.3.3   Real-Time  Instrument Survey (Screening Technique)

     Real-time  instrument surveys are a preferred screening technique  for
identifying potential emission "hot-spots'6 or near surface waste  bodies  at
hazardous waste sites.  The survey consists of inspecting the site with
real-time instruments to determine the average species concentration in  the
air layer directly above the site, and to identify "hot spots" with above
average emission  concentration.  Real-time instrument surveys can also include
taking  headspace  concentration measurements in cracks and vents.  They can be
used for both volatile species and particulate matter, but it is typically
used for volatile species emission.

     The site is  inspected by placing the inlet of the real-time instrument at
a specified height above tht surface, typically 2-3 inches to several feet
ibove the site surface.  The site is walked 0n a 25-foot grid as shown in
Figure  18, although the grid may be adjusted to accommodate site size.  Upwind
measurements are mide before and after inspecting the site by measuring the
ambient air at 5-feet above the site at an upwind location.  Sampling for
volatiles should be performed during quiescent wind conditions (i.e., average
wind speed less than 5 miles per hour).

     Personal-size airborne particulate monitors can be used to survey ambient
particulatt matter concentrations.  These monitors are used in accordance with
standard NIOSH industrial  hygiene methods (34).   Personal-size monitors whose
operating principle is based on the  detection of scattered electro-magnetic
radiation in the near infared spectrum have been used successfully as field
monitoring devices at hazardous waste sites.   These monitors can also be used
to collect particulate (respirable and non-respirable) samples for subsequent
chemical analyses.  Sampling for worst-case particulate emissions should be
conducted during periods with high wind speed and dry soil conditions.
    icability--
     The real-time instrument survey is applicable to all  types of hazardous
waste sites, assuming sampling personnel  can reach all  areas of the site.
                                      94

-------
                         100'
                                   25'
   500'
Figure 18.  Real-time instrument survey.
                    95

-------
Limitations--
     The technique does not provide for calculation of an emission rate,  but
rather determines the number of particles per volume of air at a given
receptor.  This information can be used to estimate the risk from baseline  -
participate matter emissions at the site.  The quantity of particulate matter
sample that can be collected using the personal-sized monitors is small,  and
this may limit the type of analytical  analyses that can be conducted.

4o3.4  Concentration-Profile (In-Deoth Technique)

     The concentration-profile (C-P)  technique measures the concentration of
the emitted species at logarithmically spaced heights at a downwind location
on the anticipated plume center!ine (21,35).   This technique has been tested
under a variety of waste site conditions and  has  been shown to produce valid
results.  Figure 19 illustrates the C-P sampling  approach.

     The C-P technique was developed  by L.J.  Thibodeaux and co-workers at the
University of Arkansas under an EPA contract-   The technique is based on
measurements of wind velocity,  volatile species concentration,  and temperature
profiles in the boundary layer above  the waste body.   These measurements are
used to estimate the vertical  flux of  the volatile species as  (31):
                    E
                     i
                                  H
W
     where     E1  *  emission rate (flux) of organic species i  (g/cm2-sec);
               D1  -  molecular diffusivity of organic species i  in  air
                    (cm2/sec);
               H20 « molecular diffusivity of water vapor in air (cm2/sec);
                 n » exponent for diffusivity ratio;
                Sv - logarithmic slope of the air velocity profile  (cm/sec);
               .Si - logarithmic slope of the concentration-profile  for
                     organic species  i  (g/cm3);
                                     96

-------
157"-
SENSOR
 ARMS
                  \
               J
   52*
     29 TT^ —
      '   13  ««
                            WIND BIBtCTlGN
                               SENSOR
                                          SAMI»UN@
                                            MAST
                                     MAST SI3E
                                       PANEL
                                               PONO
                                                            COMPUTE*
                                                            DATA SYSTEM
     Figure  19.   Mast sample collection  system for C-P  sampling.
                                   97

-------
                  K  »  von  Karman's constant;
                #m  -  Businger wind shear parameter; and
                Sc - turbulent Schmidt number.

     The term  (*Js  )-1  represents an atmospheric stability correction  factor
and  is expressed  as a function of the Richardson number.  The function  is an
empirical correlation which corrects the estimated emission rate for water
vapor to measured values  under various atmospheric stabilities.  For this
reason, the correction  factor is valid only under specific meteorological
conditions.

     As seen in Equation  10, calculation of an emission rate using C-P data
requires the use ©f i complex equation that includes several estimated
aprameters.  Contrast this to the relatively simple equations (e.g. flux
chamber) presented for  previous methods, where all  parameters could be
measured in the field.  Users should see the cited literature (21, 35) for
more discussion of the  C-P data reduction procedures and for sources of the
estimated constants and parameters.

     The sampling equipment consists of a 4-meter mast with a wind direction
indicator,  wind speed sensors, temperature sensors, and air collection probes
at six logarithmically  spaced heights above the area source; a continuous
real-time data collection system; and a thermocouple for measuring water
temperature.  Prior to  sample collection, meteorological conditions must be
monitored to determine  if sampling conditions meet the necessary
meteorological  criteria.  Once acceptable meteorological conditions are
documented, the sample collection period is initiated.  During the sample
collection period, wind speed, air temperature, and relative humidity are
measured.

Applicability--
     The C-P technique  is applicaole to emission rate measurement from many
types of large area sources including landfills, lagoons,  and areas of
contaminated soils.  The technique is applicable for both  volatile and
                                      98

-------
participate emission rate measurement.  It is applicable both for undisturbed
and disturbed site conditions, and for testing emission control techniques.

limitations--
     The technique requires that meteorological conditions during sampling,
particularly wind speed and direction, match the predetermined conditions used
to select the sampling location.  The equipment required to monitor and
process the meteorological conditions is much more complex than that for any
of the other direct or indirect techniques.  The sampling location must be on
the approximate plume center line.

     The technique also assumes that the site is fairly homogeneous;
therefore, the technique may not be applicable to heterogeneous sites.  The
technique is not applicable to sites where there is insufficient distance
downwind to set up the sampling equipment.  The technique also may not
adequately collect emissions from point sources within an area source, such as
cracks in landfill covers or vents.

     The technique requires upwind sampling to account for other potential
sources.  Finally, the technique is not applicable during quiescent or
unstable wind conditions, and may produce false negative results during these
4o3,§  Transect (In-Depth Technique)

     The transect technique, also referred to as plume mapping,  measures the
concentration of the emitted species at several  downwind locations aligned
perpendicular to the anticipated plume centerline (21,36).   The  in-depth
transect technique is an indirect emission measurement approach  that has been
used to measure fugitive particulate and gaseous emissions  from  area and line
sources.  This technique has been successfully tested at a  variety of waste
sites, including landfills.  Figure 20 illustrates the transect  sampling
array.
                                     99

-------
  Vlilual
Point Soufcc
                            Wind Ol««cfik»ft
             Figure 20.  Example of  transect  technique sampling.
                                        100

-------
     Tht transect technique uses horizontal and vertical  arrays  of samplers  to
measure concentrations of species within the effective cross-section  of the
emission plume.  The volatile species emission flux  is then obtained  by
spatial integration of the measured concentrations over the assumed plume
area.  For volatile species, the emission flux is calculated  as  (21):
                      Ei *"   f  FA   Ci(M) dhdw
                            s   J J  p
               E1 « emission flux of component 1 (ug/m -sec);
               y - wind speed (m/sec);
               As - surface area of emitting source (m2);
               Ap * effective cross-sectional area of the plume (m2);
               C1 « concentration of component i at point (h,w), corrected for
                    upwind background (ug/m3);
                h » vertical distance coordinate (m); and
                w • horizontal distance coordinate (m).
     For particulates, the emission flux is calculated as (36):
     where     E » emission flux (ug/m2-sec) ;
               t » sampling time (sec);
               As - surface area of emitting source (n»2);
               Ap * effective cross-sectional  area of plume (m2);
               m - mass of particulates collected after correction for back-
                   ground concentration (ug);
               h - vertical distance coordinate (m);
               w - horizontal distance coordinate (m); and
               a - intake area of sampler (m2).
                                      101

-------
     The cross-sectional area of the source (As)  term can be eliminated from
both equations, if only the total site emission per time is required.  An
alternative equation for volatile species, based on diffusion theory  and
measurements,  is  (37):
                   E.R., - I a X.K. o a. C 0                          (Eq.  13)
                        i   j    i  i  y L  „

     where     E.R., - emission rate of species  i  (gm/sec);
                  Xt - peak concentration of species  i  (Gaussian Fit Curve);
                  Ki = conversion  factor gm/ppm  for species  i;
                  oy - lateral  extent of Gaussian  plume;
                  GZ = vertical extent of Gaussian plume;
                  C  = instrument response factor;
                  ft  = 3.141; and
                  0  = mean wind speed.

All parameters are obtained from field measurements.   (In some instances, oz
is estimated from ay).   As  for  the C-P technique,  users  should  see the cited
literature for further guidance on the data reduction procedures.  The key
point to note is that the equations are relatively complex.   Furthermore,
ttrms such as ay and @z will  require  curve-fitting  of the data  and will
typically have a large associated uncertainty.

     The sampling equipment consists of a central  3.5-meter mast supporting
three equally spaced air sampling probes, and single  wind direction, wind
speed, and temperature sensors  at the top; and five 1.5 meter masts with
single air sampling probes.  The central mast is aligned downwind along the
expected plume center-line.  Two masts are placed on  each side of the central
mast perpendicular to the plume center!ine at equal spacings; and one mast  is
used to collect air samples at  an upwind location.  The spacing of the
associated masts is selected to cover the expected horizontal plume
cross-section, as defined by observation and/or profiling with real-time
analyzers.  Additional sampling locations, both vertically and horizontally,
can be added as required to provide sufficient coverage of the plume

                                      102

-------
cross-section.  Prior to sample collection, meteorological parameters must be
monitored to determine  if.sampling conditions meet the predetermined
meteorological criteria.

     The transect technique is somewhat less susceptible to changing
meteorological conditions than the concentration profile technique, but it
does not account for the vertical dispersion of the emitted species due to
their varying molecular weights.  A more complex array of samplers can be
employed to overcomt this shortcoming, if necessary.  Tha transect is often
the preferred technique because the technique is applicable to a variety of
some types and the resulting data can be more useful since the data are
collected across the plume area.

Applicability--
     The transect technique is applicable to emission rate measurement
from all forms of area sources, including lagoons,  landfills,  open dumps,  and
waste piles.

     The technique can be used for both volatile and particulate matter
emission rate assessment.   The technique is applicable both for undisturbed
and disturbed site conditions, and for testing of emission control techniques.
The technique is applicable to emissions measurement during
all phases of the RI/FS process and can, therefore, provide directly
comparable data throughout the process, including post remediation monitoring.

     While the method assumes a relatively homogeneous site and a well mixed
plume, these conditions are not necessarily required to use the method.  The
placement of sufficient sampling stations across the plume can allow the
technique to be used at a heterogeneous site,  or where the distance downwind
for equipment set up is limited.  However, data collected under these
conditions should be carefully evaluated before use.
                                     103

-------
Limitations--
     The technique requires that meteorological conditions during sampling,
particularly wind speed and direction, match the predetermined conditions  used
to select the sampling locations.  The center mast should be on the
approximate plume center!ine.  The technique may not adequately collect
emissions from point sources within an area source, such as cracks in landfill
covers or vents, unless the plume is fairly well mixed at the sampling
locations.  The technique provides only limited vertical profiling of the
plume.  The technique is not applicable during quiescent or unstable wind
conditions; it may produce false negative results during these conditions.

Preferred Technique--
     Thi transect technique is a preferred indirect emission assessment
technique.  The technique has been used for several different types of area
sources and is documented in the literature.  The applicability of the
technique, the conditions required for sampling, and the moderate level  of
equipment and man-power needs suggest this technique as a preferred technique
relative to other indirect approaches.

4.3.6  Boundary Layer Emission Monitoring

     Boundary layer emission monitoring can be used to determine the emission
rate of pollutants from large hiterogenous area sources (38).  The technique
1s similar to the transect technique in that samples are collected along an
array that is perpendicular to the emission plume.   However,  rather than fixed
sampling locations, the boundary layer technique involves a cart with three or
more samplers traversing the plume with each sampler at a different height.
During the plume traversing the sampling rate is adjusted to be proportional
to the sine of the angle between the wind vector and the direction of the
traverse path.  Each run takes about an hour.  The  average concentrations are
adjusted for any upwind concentration and then used to calculate an average
vertical concentration profile.  This profile is numerically integrated (with
the wind velocity profile) over the contaminant boundary layer to derive an
emission rate for the source.
                                      104

-------
     The emission  rate  is  (38):
D
                     C
WVn x 10 w      A    (z/loni-z/Z. )u dz      (Eq. 14)
  0        •     n    \fl\i  n'           \ -i    i
     where      E  - emission  rate  (g/sec);
                C0 - ground level concentration (ug/m3)
               W  - cross-wind distance.
              ,vVe « average wind speed at 10m (m/sec);
               Zb - boundary layer thickness (m);
               p  - exponent of wind-velocity profile; and
               b  » exponent of concentration profile.

    I ications--
     The technique can be used for measuring an emission rate from any  type  01
source with constant emissions.

limitations--
     The method assumes that both the contaminant emission rate and the wind
speed and direction are reasonably steady while sampling is conducted.  The
major sources of error are in the average measured wind velocity and  in
accounting for upwind contamination.  The method should be limited to area
sources that do not have significant plume buoyancy, to periods of 1-10 m/sec
winds, and to sampling heights less than the depth of the mixing layer.  The
technique has not been validated by the EPA.

4.4  AIR MONITORING TECHNIQUES

     Techniques for ambient air monitoring and indoor air monitoring  are
discussed below.
                                      105

-------
4.4.1 Ambient Air Monitoring

     Air monitoring techniques that measure the ambient air concentration
resulting from area emission sources are combined with air dispersion  modeling
t© calculate'the irei source emission rate.  The primary difference  between
indirect emission measurement techniques and air monitoring techniques  is  the
distance at which measurements are made downwind from the source.  Indirect
measurements are made near the source or units of a combined site  (usually on
site) and may be able to distinguish between multiple units within a site,
depending on the spacing between units.  Air monitoring is generally performed
at considerable distance downwind from the source and usually cannot
distinguish between multiple units within a combined site.  Air monitoring
typically measures lower concentrations because the contaminant plume  is
subject to additional air dispersion.

     The first step to use the ambient air sampling data to develop emission
rate estimates is to select an air dispersion model which accurately reflects
the site-specific conditions, including regional  and local terrain, typical
wind stability, etc.  Guidance for selecting an appropriate model is given in
the EPA's Guideline on Air Quality Models (39).  Preferred models given in the
guidance document include the Climatological Dispersion Model  (COM 2.0),
Gaussian Plume Multiple Source Air Quality Algorithm (RAM), Industrial Source
Complex Model (ISC), and Single Source (CRSTER) Model.   A number of other
potentially applicable models are included in the guidance document.   Models
not included irrEPA's manual  also may be applicable at  uncontrolled hazardous
waste sites, including the Point Plume Model (PTPLU),  and the Gifford and
Hanna Simple Box Model (40).  The models art used with  air monitoring and
meteorological  monitoring data to estimate emission rates.

     Air monitoring and air dispersion models are used  to determine the
emission rate through an iterative process.  An emission rate is first
estimated for the area source.   This estimated emission rate,  along with
meteorological  data collected during air monitoring, is used to calculate  a
predicted downwind concentration.  The predicted concentration is then
compared to the measured downwind concentration^   Based on this comparison,
the estimated emission rate is adjusted appropriately,  and the process is
                                      106

-------
repeated until acceptable agreement is reached between the measured and
predicted downwind air concentrations.

     The air monitoring techniques that can be used to develop BEEs are listed
below.  These techniques were described in Section 4.3 and can be used at
greater distances downwind of the emission source as air monitoring
techniques.
     4.3.1

In-Depth Techniques--
     4.3.4  Concentration-Profile
     4.3.5  Transect
     4.3.6  Boundary Layer Emission Monitoring

4.4.2  Indoor Air Monitoring

     This section describes techniques for assessing the impacts on the indoor
air quality of homes or other structures from adjacent hazardous waste sites,
The primary routes for contamination of indoor air are by migration of
contaminated water into or near subsurface structures and subsequent
vaporization of the contaminants, and by contaminated ambient air entering
into the structure.  Airborne particulate matter from hazardous waste sites
may settle on surfaces near the structure and may then be brought into the
structure via shoes, clothing or other items.  Contaminants entering the
structure from this route may subsequently become airborne when they are dis-
turbed by foot traffic, vacuuming, or other means. Assessing these impacts may
involve a variety of techniques, including soil  and ground water testing, and
indoor and outdoor air testing.

     In the absence of site-specific indoor air data, soil vapor studies are
usually the most useful environmental data for assessing the potential for
indoor air impacts.  Soil  vapor migration is the most common route of
subsurface contaminant entry Into homes from hazardous waste sites.
                                      107

-------
Performing soil vapor studies may also aid in identifying specific  structures
requiring detailed indoor air testing.

     Indoor air testing programs may be burdensome on occupants and generate
considerable'concern, since individuals are not likely to be fully  aware  of
the relative hazards associated with certain chemical exposures and other
site-specific factors that may affect exposure patterns.  These considerations
dictate that indoor air programs be focused on those structures most likely
affected, and that the testing methods be designed to accurately reflect
airborne concentrations without being overly inconvenient or burdensome to the
occupants.  Occupant cooperation and agreement to participate in the testing
program is an important consideration in performing these studies.

     Techniques for Confirming Indoor A1r Impacts - Indoor air testing
programs are performed to determine the nature and extent of indoor air
impacts from waste site contaminants.  These testing programs may be screening
studies designed to identify target analytes and specific areas or structures
for additional study, or comprehensive testing and detailed analyses.

     Screening studies typically involve limited sampling and analytical
efforts in structures identified as high priorities, i.e. those structures
roost likely to be impacted.   Tasting may be performed by using direct reading
rill time analyzers that provide instantaneous measurement and feedback of the
presence and concentration of selected contaminants.  The most common devices
used are portable flame ionization detectors (FID) and photoionization
detectors (PID).  These instruments are capable of detecting a wide variety of
organic gases and vapors,  but are not applicable to detecting non-volatile
contaminants or particulate  matter.

     The main limitation with portable, real-time analyzers are their
detection limits.   These limits ire based on chemical and physical properties
of the specific chemical in  question, and are usually in the one to five  part
per million range.   Also,  these devices are not "capable of determining
chemical identities.   If a greater degree of sensitivity or chemical
identification is  required,  more sensitive sample collection and laboratory
analytical  methods  would be  required.
                                      108

-------
     The real time analyzers are very useful  in investigating point source
locations where volatile contaminants may be  entering the structure, such as
floor/wall junctions in grade or below grade  level  rooms (eg. basements) and
voids-or penetrations in slabs and below grade walls.  Calibration of the
analyzers to.standard gases of analytes of interest is an essential quality
control feature for these devices.

     Detailed studies to identify indoor air  impacts may be designed to
thoroughly identify structures that may be impacted, and to examine variations
in contaminant concentrations that may occur  over time.   Possible features of
the detailed studies are summarized below:

     •    Sampling outdoor air in conjunction with  indoor air sampling to
          determine the contribution of outdoor sources  to indoor
          contamination (ie. to differentiate them  from  subsurface contaminant
          migration);

     •    Testing multiple rooms to determine distribution and exposure pat-
          terns within a structure;

     §    Inventorying products and other potential  sources of target com-
          pounds (eg. cleaning, hobby and maintenance products containing
          volatile chemicals) to determine impacts  of occupant activities on
          air contaminant concentrations; and

     «    Conducting similar testing programs in non-impacted "comparison
          structures" to compare the results  from study  area structures.
          Numerous sources of many indoor air contaminants have been iden-
          tified inside residential and other structures,  requiring testing in
          similar structures to determine if  impacts are indeed due to the
          adjacent waste site.   Comparison structures must be selected with
          construction type, building materials,  indoor  activities, age,
          heating system,  and outdoor air impacts carefully considered to
          assure proper comparisons are made.
                                     109

-------
Detailed studies may be performed over extended  periods of time to examine the
influences of seasonal  weather changes.   This  is advisable since structure
ventilation and infiltration of soil  vapors  are  affected by weather condi-
tions.  Repeat or follow-up testing may be necessary if initial testing
suggests in adverse impact is present,  new environmental data indicates
changes in contaminant  migration,  or  mitigation  techniques are initiated.

     Study design considerations - Before initiating an indoor air quality
study, it is essential  that proper study design  features be carefully
considered to assure the appropriateness of  the  data that are produced and the
accuracy, precision and completeness  of the  data.  Some of these
considerations are:

     t    Collecting an adequate number  of samples;
     •    Duration of sample collection;
     *    Location of sample collection  in the structure;  and
     c    Quality assurance and quality  control  samples.

     Worst-Case Conditions - Factors  influencing  infiltration of contaminants
into structures from adjacent hazardous  waste sites  can  significantly affect
indoor concentrations.   Examples of these factors are:

     «    Structure ventilation (eg.  open vs. closed windows,  operation of
          air-to-air heat  exchangers, mechanical ventilation  systems,  etc.);

     t    Rate of soil  vapor infiltration (eg. indoor stack effect  from
          indoor/outdoor temperature  differentials, mechanical  devices
          creating negative pressure  inside  the  structure  enhancing soil  vapor
          infiltration); and

     §    Rate of outdoor  ambient  air infiltration if a  significant route of
          contaminant entry is  from outdoor  air.
                                     110

-------
     By recognizing  these factors, it is possible to conduct indoor testing
under conditions that would result in worst case indoor concentrations.   For
example, if volatile organic compounds had migrated under a home and there
were-no significant  outdoor air impacts, worst case concentrations would  be
expected with all outdoor air ventilation routes closed (eg. windows and  doors
closed, clothes dryer not operating, air-to-air heat exchanger not operating,
etc.), and a large temperature differential between indoor and outdoor air
creating a significant stack effect drawing in soil vapor.

     Interpretation  of Results - There are several  ways that indoor air
quality data may be  examined and interpreted. There is no universally
recognized protocol.  These methods include comparison to established health
criteria values such as Occupational Safety and Health Administration (OSHA)
Permissible Exposure Limits (PELs), and American Conference of Governmental
Industrial Hygienists (ACGIH) Threshold Limit Values (TLVs).  Some fraction of
the health-based criteria may be used to add an extra safety factor.

4.5  EMISSIONS (PREDICTIVE) MODELING

     Emissions models have been developed to predict emission rates for a
variety of waste site types including landfills without internal  gas
generation, landfills with Internal gas generation  (typically co-disposal
sites), open dumps,  waste piles, spills, land treatment operations, aerated
lagoons, non-aerated lagoons, and lagoons with an oil  film.   These models are
almost exclusively theoretical, and each model is generally applicable to only
one type of waste site.

     The predictive models can be used as screening or in-depth techniques.
Emissions models, used as screening techniques, employ data that can be
obtained or calculated from information available in the literature, or can be
assumed with some level  of confidence.  Emissions models,  used as in-depth
techniques, require site-specific site and waste characterization data.  The
selection of model  input sources (site-specific,  literature value,  or assumed)
should be based on the requirements of the decision-making process and the
level  of resources available.  Site-specific data should be used whenever
possible to Increase the accuracy of emission rate  estimates.
                                      Ill

-------
     Several predictive models are presented below to acquaint the  reader with
the types of available models for emission rate estimation.  The EPA's
Suoerfund Exposure Assessment Manual (SEAMS) (41) presents simplified
predictive models for many of the waste site types.  The SEAMS models,
simplified versions of the snore complex models, are included in this section.
In general, the SEAMS models are recommended for estimating emissions from the
various types of uncontrolled waste sites.  However, the complex models can be
used if more precise estimates are required and/or more detailed site
information i s avai1able.

     Specific methods for calculating the model input variables,  such as
diffusion coefficients, have been presented by the authors of the models,  but
art not included here for sake of brevity.  Data bases containing various
input parameters are described in Appendix D.   Commonly needed input data  for
selected compounds are listed in Appendices F  and G.  Each model  requires
estimating the emission rate of the individual  components of the waste; and
then summing the emission rates to determine the overall  emission rate.  For
complex waste, application of the models is best performed on a computer to
speed the calculation.  An emission flux can be calculated by dividing the
emission rate by the emitting area.

     A wide variety of variables are associated with each of the predictive
models; however, a number of key inputs art required by many of the models.
These key inputs for landfills include:  the vapor diffusion coefficient
through the soil or mass transfer coefficient  across the air/soil  boundary for
waste constituents; the physical  size of the source expressed as  area,  length,
and/or width,  depending on the model  used; physical  parameters of the landfill
cover,  such as depth of cover, permeability, and soil  porosity (total,
air-filled, and/or effective porosity); physical/chemical parameters of the
waste,  including chemical  composition,  weight  or mole fraction for
constituents,  vapor concentration of constituents at the waste surface or
within  the soil  cover, and partial  pressures of constituents; atmospheric
conditions, such as temperature,  wind speed and direction,  and barometric
pressure;  and  estimates of the soil gas velocity through the soil  cover.   The
key inputs for lagoons include:   mass transfer coefficients;  physical/chemical
parameters of  the waste,  including chemical composition,  weight or mole
                                     112

-------
fractions, partial pressures, and Henry's Law Constants; physical dimensions
of the lagoon surface; atmospheric conditions, such as temperature, and wind
speed and direction; layering of waste within the lagoon; and physical/
chemical parameters of a surface crust.

     While all of these parameters can be estimated with varying levels of
confidence, it is best, when possible, to collect site-specific data.
Physical/chemical measurements of waste constituents can be obtained from
sampling and analysis programs, although i records review is advisable to
identify key constituents and ensure representative sampling.  Likewise a
sampling and analysis program combined with a records search is desirable to
determine the physical size and shape of the source and the porosity and
permeability of any soil cover.  Atmospheric conditions are easily obtained
from various weather services which can provide regional data; however,
collecting some site-specific meteorologic data to ensure representativeness
is desirable.  Diffusion and mass transfer coefficients are typically
calculated based on the wastes' chemical composition and their known chemical
properties, such as Henry's Law Constants, although tabulated diffusion
coefficients are now available.  The referenced literature includes suggested
methods for calculating the diffusion coefficients as well  as some tabulated
data.  A list of references is given at the end of this section that contain
tabulated values for selected input model  parameters.   Diffusion and mass
transfer coefficients can also be determined experimentally in the lab;
however, disturbance of the waste and landfill cover to obtain site-specific
materials would probably introduce uncertainty.

4.Sol     Summary and Comparison of Emission Models

     The landfill models presented here can be categorized into six types:

     «    Closed landfills without internal  gas generation »- Fanner Model,
          Shen Model, SEAMS Model, Thibodeaux a Model,  Logarithm Gradient
          Model,  and RTI Closed Landfill Model;
                                     113

-------
     «    Closed landfills  with  internal  gas  generation  --  Scholl  Canyon
          Model, (Thibodeaux)  Convective  "Add-On"  Model,  SEAMS Model,
          Thibodeaux b Model,  and  (Thibodeaux)  Exact  Model;

     «    Open landfills -- Scholl  Cinyon Model, Arnold's Open Landfill Model,
          Shin Open Landfill  Model,  and RTI Open Landfill Model;

     «    Lindtreatment --  RTI Landtreatment  Model  and SEAMS Model;

     •    Leaks and spills  on soils  -- SEAMS  Model  (Fresh Spills)  and SEAMS
               Spills); and
     «    Fugitive Dust -- AP-42 Vehicular  Traffic,  SCS  Model,  Cowherd Model,
          tnd Gillette Model.

     The lagoon models presented in this  document  can  be categorized into two
            i-aerated lagoons -- Mackay-Lienonen;  SEAMS  Model;  Thibodeaux,
          Parker, and Heck; Smith,  Bomberger,  and  Haynes;  Shen; and RTI; and

     ®    Atrited lagoons -- Thibodeaux,  Parker, and  Heck; and  RTI.

     The applicability and chancteristies of  etch of these models is
summarized in Table 16.  The complexity of the models varies depending on the
number of emission mechanisms included* in the  model.   In general, the SEAMS
models are the most simplified models  and have been recommended by the U.S.
EPA for estimating emissions from Suptrfund sites  (41).   The SEAMS models
apply to a majority of Superfund sites; however, other emission models may be
more appropriate when more accurate emission estimates are required or the
basts of the SEAMS models are not appropriate  for  t specific site.  The user
should consider the amount of site specific data  available, the compatability
of the site characteristics with the model assumptions,  and the level of
effort required when selecting an appropriate  model.
                                      114

-------
                                  TABLE  16.   SUMMARY AND COMPARISON OF EMISSION MODELS
                            Soil                     High Con-8              Barometric      Time  Complexity6
                            Phase  Air/Waste  Liquid  tration of     Stogas    Pressure        Fune-     of          Comments/     Reference
Waste Site/Model    Species  Cover  interface  Phase   Contaminant  generation   Pimping  Mind  tion  Input Data    Applicability    Number
Closed Landfill Without
Internal Gas Generation
Fanner
Shen
VOC
VOC
SEAMS
Thibodeaux8
                      VOC
VOC
SS        1     Hazardous Waste      42  • 45
               Sites

SS        1     Hazardous waste      46-48
               sites;  applies
               to wastes with
               high
               concentration of
               constituent
               (i.e.,  Raoult's
               Law applies).

SS        1     Hazardous waste      41
               sites;  EPA
               recommended.

SS        I     Hazardous waste      45,  49
               sites.
Logarithim Gradient   VOC
RTI Closed
VOC
                                                                        SS
                                                                                              TD
               Hazardous waste
               sites! waste
               concentrations
               >5% volume.

               Hazardous waste
               sites; predicts
               long-term
               emissions over
               time.
45, 48
                                                                                                                              (Continued)

-------
                                                       High Con-88
                             Phase  Air/Waste  liquid  tration of     Slogas
Waste Site/Model    Species  Cover  Interface  Phase   Contaminant  Generation
                                                                                Barometric       Time  Complexity
                                                                                Pressure        Fune-     of          Comments/     Reference
                                                                                         Mind  tSon  Input Data    Applicability    Nuiber
Closed landfill with Internal Gas fieneration

Schotl Canyon         VOC


Convectfve "Add-on"   VOC      X
                                                                                                 SS/TD      1     Co
                                                                                                                sites

                                                                                                 SS         2     C
                                                                                                                Chazardous and
                                                                                                                municipal
                                                                                                                wastes) sites.
                     n
SEAMS
                      VOC
                                                                                                 SS        1     Co-disposal
                                                                                                                sites; iPA
                                                                                                                recommended.
Thibodeaux
          ti-
                                                                                                 SS
                                                                                                                Co-disposal
                                                                                                                sites.
                                                                                                                                         49
Thibodeaux Exact      VOC      X
                                                                                                 SS
Co-disposal
sites.

-------
                                                      TABLE  16.   (Continued)
                            Soft                      High Con-a               Barometric      Time  Complexity6
                            Phase  Air/Waste  Liquid  tratfon of     Biogas    Pressure        Func-     of          Comments/     Reference
Waste Site/Model     Species  Cover  interface  Phase   Contaminant  Generation   Pumping  Mind  tion  Input Data    Applicability    Number
Open Landfills
Scholl Canyon VOC
*
X SS/TD 1 Hazardous waste
sites with
Methane
generation.

72
Arnold's Open
VOC
                                                                         TD
Hazardous waste
sites; diffusion
from a liquid
surface.
45
Shen Open
VOC
                                                                    X    TD
Hazardous waste
sites; accounts
for wind speed.
45. 46,
47, 50
RTI Open
VOC
                                                                    X    TD
Hazardous waste
sites; more than
one volatile
constituent in
42, 45
                                                                                                                               (Continued)

-------
                             Soil                       High Con'*
                             Phase  Air/Waste  Liquid  tratSon of
Waste Site/Model    Species  Cover  Interface  Phase   Contaminant   Generation
                                                                          Time  Compte«itye
                                                                          Func-     of
                                                                    Wind  ««on  Input Data
     Comments/
                                                                                                                                   Reference
                                                                                                                                    Number
Land Treatment

RT1
VOC
Liquid waste
applied fto soU
surface; long-
term and short-
term emissions
over time;
bfodegradation*
                                                                                                                                    44, 45
SEAMS
                      VOC
       and leaks on Soil
SEAMS (Fresh Spills)  VOC
                                                                                                 SS
SEAMS (Old Spills)    VOC
                                                                                                 TO
                                                                                          Liquid waste
                                                                                          applied to soil
                                                                                          surface;
                                                                                          predicts average
                                                                                          emission rate
                                                                                          over time, EPA
                                                                                          recommended.
                                                                                          Predicts
                                                                                          emissions from
                                                                                          visible licguid
                                                                                          pools on soil
                                                                                          surface; EPA
                                                                                          recommended.

                                                                                          Predicts
                                                                                          emissions of
                                                                                          liquids
                                                                                          Incorporated
                                                                                          into soil;  EPA
                                                                                          recommended.
                     41
                     41
                                                                                                                                  (Continued)

-------
                                                                  TABLE  16.    (Continued)
                                        son
                                        Phase Air/Waste
           Waste Site/Model    Species  Cover  Interface
                                          Con-"
                                     tration of     Biogas
                                     Contanfnant  Generation
Baronet He      Time  Complexity6
Pressure        Func-     of          Cements/    Reference
 Puipine  Uind  tion  Input Data    Applicability    Number
            Fugitive Dust Emissions

            AP-42             Partieulate
            SCS
            Cowherd
 Partieulate
 Particulate
                                                                                                          SS
                                                                                                          SS
                                                                                                          SS
\O
            Gillette
Particulate
                                                                              SS
                                Emissions caused     54
                                by vehicular
                                traffic on
                                unpaved road.

                                Annual average       41
                                emissions of
                                wind-blown
                                surface soils.

                                Annual average       55
                                emissions of
                                wind-blown dust
                                from surface
                                with a "limited"
                                amount of
                                erodible
                                material.

                                Emissions of         55
                                wind-blown dust
                                from surface
                                with an
                                "unlimited"
                                amount of
                                erodible
                                material; annual
                                and 24-hour
                                average
                                emissions.
                                                                                                                                          (Continued)

-------
TABLE  16.   (Continued)
Soii
Phase Air/Meste liquid
Ueste Site/Model Species Cover Interface Phase
Hon-aerated lagoons8
Mackey-Lienonen VOC X


SEAMS VOC M



Thibodeaux. Parker, VOC JC
and Heck

Smith, Bamberger, VOC %
and Waynes
Shen VOC X

RTI VOC X



Hfigh Core* Barometric Time Complexity*1
tratfon of Biogas Pressure Func- of Ccsimcfitfi/
Contaminant SemratSon Pumping Uind tion Input Data AppikabiHty
«
TD 2 Quiescent lagoon
no influx of
contaminants.
2 Quiescent lagoon
constant liquid
phase concen-
tration.
SS 3 Constant Uquld
phase coneen*
tration.
SS 3 Highly volatile
compounds only.
SS 2 Empherical
screening model.
SS I Constant liquid
phase
concentration,,
biodegradation.
Reference
Number

56, 57


41



5S


37

48, 58

U



                                                           (Continued)

-------
                                                       TABLE  16.    (Continued)
                             Soil                      High Con-*              Barometric      Tine  Complexi
                             Phase  Air/Waste  Liquid  t rat ion of     B togas    Pressure        Fune-     of         Conments/     Reference
Waste Site/Model     Species  Cover  Interface  Phase   Contaminant  Generation   Pumping  Wind  tiorc  Input Data   Applicability    Number
            Aerated Lagoons

            Thibodeaux, Parker,   VOC
            and Heck
                                                                                                                                   53
RT!
                                  VOC
                                                                                               SS
N)
* Greater than five percent by volume.

b SS « steady state; TD * time dependent.

c Scale: 1 « low complexity
         2 » medium complexity
         3 * high complexity
 ,
° Nodel can be used to predict emissions from landfills with internal gas generation  if a multiplication factor of six is used.

e Can also be used to estimate emissions from contaminated water in basements  and other subsurface structures.

-------
     Each general category of emission models is discussed in the  following
paragraphs.  A detailed discussion of each model is presented in sections
4.5.2 through 4.5.10.

     Closed Landfills

     Most of the closed landfill models assume that the emission of volatile
species is controlled by vapor diffusion through the soil cover.  Additional
closed landfill models have been developed to account for convective transport
mechanisms which can also influence emission rates.  The most significant of
these convective mechanisms is biogas generation.  Biogas is typically
generated within co-disposal site where a mixture of municipal and solid
wastes have been dumped.  The upward movement (convective sweep) of the biogas
within the landfill can greatly increase the upward migration of volatile
species resulting in increased emissions compared to simple vapor diffusion
through the soil pores.  When biogas is present, it usually becomes the
significant controlling factor and therefore, it is the most important
consideration when selecting an appropriate landfill emission model.  Other
less significant convective transport mechanisms include waste concentration
and atmospheric effects (i.e.  barometric pressure and wind).

     Waste concentrations greater thin approximately 5 percent by volume can
cause a convective sweep of vapor through thi soil  which also increases
emissions.  The logarithm gradient model and the Thibaodeaux exact model were
developed to account for this  mechanism.

     Changes in barometric pressure have also been  shown to have an effect on
landfill  emissions rates.  This mechanism, referred to as barometric pressure
pumping,  has been incorporated into two of the landfill  models (RTI closed and
Thibodeaux b).
                                      122

-------
     Open Landfills

     The Scholl Canyon Model is a single stage, first order kinetic model that
can be applied to calculate methane generation and emissions of volatile
constituents from open landfills.  The Scholl Canyon Model is based on
empirical factors.  Thi model is applicable to estimating emissions over long
time periods.

     The other models presented for estimating emissions from the surface of
open landfills are very similar to one another since the Shen Open and the RTI
Open are simply modifications of the original Arnold Open Landfill model.
These models are based on Pick's Second Law of unsteady-state diffusion (i.e.
time dependent emission rates) and they assume that the emission rate is
controlled by molecular diffusion of the volatile species through the air
above the liquid surface.  The models are designed for hazardous waste sites
and do not apply to co-disposal sites since they do not account for increased
emission rates caused by biogas generation.

     The RTI model 1s the most complete of the models; accounting for both
convective transport by wind at the waste surface and multiple species in the
volatilizing liquid.  One limitation of the Shen, Arnold, and RTI models is
that they do not account for depletion of the volatilizing species from the
surface.  Because of this limitation, reviewers have judged these models to be
inadequate for estimating open landfill emissions over long time periods
(months or longer) (44).

     Landtreatment

     The 1andtreatment models are based of the premise that emissions from the
surface are limited by vapor diffusion through the air-filled soil pores.
They are applicable to sites where liquid or semi-liquid has been applied to
the soil surface and become incorporated into the soil pores.  The short-term
version of the RTI model also accounts for convective wind losses of volatile
species that can occur while liquid pools are present of the soil surface
(immediately after waste application).  The RTI models (short and long term)
can be modified to account for biodegradation losses.
                                      123

-------
     The models available for estimating emissions from landtreatment areas
have been shown to produce similar results for cases where biodegradation is
not important (i.e. short times and/or no soil biological activity).  The
SEAMS model has been recommended by the U.S.  EPA for estimating landtreatment
eraisslons; h6wev@r9 1f b1©degradation losses  are significant the SEAMS model
   1 tend to over predict emissions, particularly for longer times.  Since tht
    model accounts for losses due to biodegradation it is more useful for
estimating long term emissions.  For short times, both models will produce
similar results.

     Spills and Leaks on Soil

     The U.S. EPA's Superfund Exposure Assessment Manual  (SEAMS) recommends
the two models listed 1n Tible AA for estimating emissions from fresh and old
spills on soil.  The model for fresh spills 1s applicable when a contaminant
pool 1s visible on the soil surface or when the soil is saturated from the
surface down.  The model does not consider diffusion through the soil phase
and therefore does not apply to cases where contaminants  have seeped into
surface soils.  In this case, the SEAMS model  for old spills (equivalent to
the SEAMS model for landtreatment) is recommended for estimating emissions of
volatile species.

     Lagoons

     Models are available for predicting emission from two types of lagoons;
non-aerated (quiescent) lagoons and aerated (turbulent) ligoons.  Since most
abandoned hazardous waste lagoons will probably be inactive, the model
squations for non-aerated lagoons will apply  in most cases.  The models for
aerated lagoons are simply modified versions  of the non-aerated models and are
presented here for completeness*

     The SEAMS model for non-aerated lagoons,  recommended by the U.S. EPA, is
a simplified steady-state version of the of the Mackay-Leinonen model and is
based on the following assumptions: (1) a constant concentration of
             v. •
contaminant in the liquid phase (i.e. no input of contaminant to the lagoon);
and (2) negligible atmospheric background levels of the volatile contaminant.
                                      124

-------
The SEAMS model is a valid estimation technique for lagoons which meet these
conditions.  However, the liquid phase concentration will decrease over time
if there is not a constant influx of contaminant to the lagoon to maintain the
concentration.  Under these conditions, the SEAMS model will tend to over-
predict emissions; the unsteady-state Mackay-leinonen model is more
appropriate since 1t 1s designed to account for a depletion of the contaminant
in the liquid phase over time.

     The models developed by Thibodeaux, Parker, and Heck , and RTI are very
similar to the SEAMS model for the case of non-aerated lagoons.  These models
can also be modified to estimate emission from aerated lagoons.  The RTI model
has also been designed to account for biodegradation losses.

     Other models are less useful because of uncertainties associated with
Input data or model applicability.  The Smith, Bomberger, and Haynes model is
only applicable for emissions prediction for highly volatile compounds.  In
addition, representative values for some input parameters are difficult to
obtain.  The model developed by Shen is an empirical relationship that is
applicable as a screening technique to determine if lagoon emissions may be
significant.

     Fugitive Dust

     The models presented here are useful for predicting fugitive dust
(particulate) emissions from contaminated soil surfaces.  Two types of models
are available: (I) those that predict emissions caused by vehicular traffic on
contaminated unpaved roads (AP-42 model); and (2) those that predict wind
blown dust emissions (SCS, Cowherd and Gillette models).  All- of the models
are designed or can be modified to estimate respirable (less than 10 urn
diameter) and non-respirable (greater than 10 urn diameter) fractions.

     The models for predicting wind-blown dust emissions are similar; however,
there are some minor differences.  The SCS model and the Cowherd model are
designed to predict annual average emissions, whereas the Gillette models can
be used to predict annual and 24-hour worst case emissions.  In addition, the
models developed by Cowherd and Gillette are applicable to surfaces with
                                      125

-------
limited erodibility and unlimited credibility, respectively.   The wind erosion
potential of the surface, which  is a function of vegetation  cover and number
of nonerodible elements such as  stones, must be characterized  before an
appropriate model can be chosen.

4D§,,2  Emission Models for Closed Landfills without  Internal Gas  Generation

Farmer Model--
     The Fanner Model (42-45) was one of the first models developed  and
generally accepted to predict emission rates from covered landfills.   The  rate
at which a compound is lost to the atmosphere from the land  surface  is
controlled by tht compound's molecular diffusion through the soil  covering the
waste.  Farmer et al, developed  this model to determine hexachlorobenzene
vapor diffusion through a soil cover.  They found that the two prime  factors
controlling/determining vapor movement through the soil were soil  depth and
soil air-filled porosity.  The model should be applicable to other compounds
as well.

     The Farmer Model combined Pick's First Laws for steady-state  diffusion
with the Millington and Quirk evaluation of the diffusion coefficient.   The
latter included a porosity term  accounting for the soil's geometric effects on
     The Farmer equation is (43):

                            J - D1(C,-Cs)(P.lfl/3/PT2)/L                  (Eq.  15)

     where      J - volatilization vapor flux through the soil cover
                    (ug/cm2-day);
               D^  • vipor diffysion coefficient in air (cmz/day);
               P8 • air-fill id soil porosity (cm3/cm3);
               PT - total soil porosity (cm3/cm3);
               Cs » concentration of volatilizing material  at the surface  of
                    soil layer (ug/cm3);
                                      126

-------
               C1 - concentration of the volatilizing material at the bottom
                    of the soil layer (ug/cm3); and
               L  - depth of the soil layer (cm).

     Fanner et al. simplified the equation somewhat by assuming a worst-case
scenario, where the soil is completely dry (Pa equals PT)  and  where  the
concentration at the surface (C2) equals 0,  meaning any increase in C2 would
effectively reduce the driving force behind the vapor flux and, thus, reduce
the vapor flux from the soil surface.  Farmer et al. called this equation the
Assessment Application (43):

                                J - D,PT4/3Cs/L                         (Eq. 16)

Appllcablllty--
     The Farmer Model provides an estimate of individual compound emissions.
The intended applicability of the Farmer Model is quantification of
steady-state volatile chemical fluxes from hazardous waste landfills.  This
model does not account for convective transport due to biogas generation and
is not applicable to municipal landfills or other landfills containing gas
generating wastes.  Use of the Farmer Model  assumes the transport of  a
volatile compound through the soil cover layer is controlled by a molecular
diffusion process.

Limitations--
     The Fanner Model assumes that the soil/waste below the soil cover layer
is saturated with constituent i.  This assumption tends to overestimate the
emissions by not accounting for the true concentration gradient below the soil
cover.  Additionally, the Farmer Model does not account for the emissions
associated with atmospheric/meteorological effects such as barometric pressure
fluctuations.  Also, at high concentrations (greater than 5 percent by
volume), the diffusion process creates its own convective sweep or apparent
velocity within the landfill.  This convective mechanism is not accounted for
by the Farmer Model (59).  These latter two limitations are likely to result
in underestimates of landfill emission rates.
                                      127

-------
Shen Model--
     Shen modified the Farmer Model (46-48) to determine a vapor emission
rate, a$ opposed to the vapor flux rate, and to enable calculation of  the
volatilization of specific components of the complete waste mixture.   This
modification'assumes Raoult's Law applies.  Shen multiplied the Farmer
equation by the exposed contaminated surface area and by the weight fraction
of tht component in the mixture.  The modified equation is
                            E, - D^AfP/'3) _W,_                      (Eq. 17)

     where     E, - emission rate of the component  i  (g/sec);
               Dj - diffusion coefficient of component in air (cm2/sec);
               Ci « saturation vapor concentration  of component i (g/cm3);
               A  - exposed area (cm2);
               Pt - total soil porosity  (dimensionless);
               L  « effective depth of the soil cover (cm); and
             wyw » weight fiction of component i  in  the  waste (g/g).

     The model assumes completely dry soil and zero concentration of
volatilizing material at the soil surface, a worst-case scenario considered
appropriate for cap design and most analyses for volatilization release from
landfill wastes,,  This assumption should be used in all cases except where
cover soils can be shown to have significant soil  moisture.  In these
instances, the soil lir-filled porosity should be substituted into the
equation for the total porosity by replacing Pt4/3  with Pa10/3/PT2.

Applicability--
     The Shen Model differs from the Farmer Model  in that it relates  emissions
to the waste composition with a weight factor (wi) and multiplication of the
flyx by the landfill area.  Like the Farmer Model, the intended applicability
is quantification of sttady-stite volatile chemical emission rates from
hazardous waste landfills.  This model does not account for convective
transport due to biogas generation and is not applicable to municipal
landfills or other landfills containing gas generating wastes.   In the use  of
the Shen Model, it is assumed that the relatively toxic properties of organic

                                      128

-------
waste placed  in  hazardous  and  industrial  waste  landfills minimize  gas
production  due to  biodegradation.

Limitations--
     The  ShiD Model  does not account for  the  landfill  gas  losses in  leachate
systems,  run  off,  or soils.  But  here again,  due  to  the inert  properties  of
the  volatile  constituents,  this accountability  is considered by Dr»  Shen  to be
minute.   The  Shen  Model also assumes that the soil is  completely dry with no
internal  gas  generation.   However,  the Shen Model  cm  be modified  to account
for  biogas  generation with  a multiplicative factor of  6.   This assumption
would tend  to overestimate  emissions by not accounting for actual  wet  soil
conditions  below the soil  cover layer.  As with the  Farmer Model,  the  Shen
Model does  not account for  emissions due  to meteorological  fluctuations (e.g.,
barometric  pressure  pumping).

     Another  limitation of the Shen Model  is  the  incorporation of  Raoult's  Law
to relate the waste  composition to  emission rate.  Raoult's Law is applicable
only to waste saturated with constituent  i and  ideal solutions.  Application
of the Shen Model  to wastes containing dilute concentrations of the
constituent i is likely to  result in an overestimate of emission rate.

SEAMS Model--
     The  model recommended  in  the U.S.  Environmental Protection Agency's  SEAMS
jnanual is a slightly modified  version of  the  Shen Model (41).  This  modified
model was proposed by Farino et.al. (49)  who  found that a  more accurate
approach  would be  to multiply  by  the mole fraction of  the  volatile component
in the buried mixture.
                          E1  - D.C.A (P)    i                       (Eq.  18)

     where:     E, •   emission rate of the component  i  (g/sec);
                Di -   diffusion coefficient  of component  in  air  (cm2/sec);
                C, -   saturated vapor concentration of  component i  (g/cm3);
                A -   exposed area (cm2);
                Pt •   total  soil  porosity (dimensionless);

                                      129

-------
               L  »  effective depth of soil  cover (cm); and
               M,  -  mole fraction of component i in the waste (gmole/gmole).

     The SEAMS manual provides guidance on methods for estimating or calculate
values for the model input parameters (41).
     The SEAMS model applies to the same situations described for the Shen
model.

timitations--
     The SEAMS model and Shen model have similar'limitations; however, the
SEAMS model relates the waste composition to the emission rate more
Thibodeaux a Model--
     Tht Thibodeaux a Model (45,49) was developed by Thibodeaux to estimate
the emissions ef volatile constituents due to interphase vapor transport from
landfills with no internal gas generation.  The model is derived from Pick's
Law of steady state diffusion.  Molecular diffusion is the controlling and
only transport mechanism addressed by the Thibodeaux a Model for the movement
ef volatile constituents toward the soil/air interface and then to the
overlying air.  To describe this mechanism, the two-resistance theory is used
to describe the two-film resistance in which the movement of chemical
constituents Is limited by their ability to diffuse through the soil and after
migration from the surface, through the air.

     The model assumes that a pure component i exerts its pure component vapor
pressure under the earth, subject to normal geophysical and meteorological
factors.  Thibodeaux defines an overall mass transfer coefficient to describe
vapor movement which is hindered-by both the resistance due to soil
characteristics and diffusion resistances at the air interface.

                             E, - 'K,  (CrCn) A                       (£q.  19)
                                         Ea1r/so11                      (Eq.  20)
                                      130

-------
where     1K1  - overall soil  phase mass-transfer coefficient (cm/sec);
          Ctl  - concentration of i above the soil/air interface (g/cm3);
           Ct - concentration of i in the sand-filled chamber  pore  spaces
                 (9/cm3);
        •  Ei *  rate of vapor movement within  the  soil phase (g/sec); and
           A •  landfill  surface area  (cm2).
                                     (C,-C,)A
where     1D% - effective diffusivity of constituent i within the pore
                 spaces  (cm/see);
             L  «  depth of the  soil  fill  cover  (cm);  and
           Cg * concentration of i at the air/soil  interface  (g/cm3).

          E.ir/»n  " 3°i (C.-Cn) A                    (Eq. 22)

where     3D, -  gas phase mass-transfer coefficient using  the equation
                 developed by  MacKay  and Matsugu  (m/hr).

           U| m  UoO&9E  Vx    LX    §c                          '^.q. tJ)

where:    Vx »   wind  speed measured  at  10 m  (m/hr);
          Lx ^   length  of the ground emission  source in  the direction (
                 the wind (m);  and
          Se -   Schmidt  number for the  gas.

Overall mass-transfer coefficient:
                                 1    +
where     1D, - D,|/rh
           rh « tortuosity, taken to be 3
             £ - porosity  of the  cover  material
                                 131

-------
    icability--
     L1ke the Farmer and Shen models, the intended application of the
Thibodeaux a Model is a hazardous waste landfill.  This model does not account
for convective transport due to internal gas generation typically present in
municipal landfills.
Limitations--
     The Thibodeaux a Model does not account for the possible emissions due to
barometric pressure fluctuations or internal gas generation.  In addition, the
Thibodeaux a Model does not account for the convective sweep of a volatile
constituent caused by high concentrations greater than 5 percent by volume.  A
number of factors, such as waste composition, multicomponent systems, and
biological or chemical reactions, greatly increase the uncertainty in the use
of the two-resistance theory.
     Tht logarithm Gradient Model (45) is the modern day interpretation of the
Farmer Model developed to address volatile constituent emission rates for
landfill concentrations greater than 5 percent by volume.  In general, this
model takes into account both the diffusive mechanism as described by Pick's
Law and the convective mechanism due to the sweep or apparent velocity which
diffusion can create with high concentrations.

     The Logarithm Gradient Model is (45):
                                                                     (Eq. 25)
Ei -DE
P JL In
PT RTL Ln
PT-pi
PT - PI*
                                      132

-------
     where      E, * emission rate (g/sec);
                DE - effective diffusion coefficient (cm2/sec);
                PT - total pressure below the cover layer (mm Hg);
                 M  - mole weight of constituent (g/mol);
                 R  - molar gas constant (cm3 mm Hg/mol °K);
                 T  - absolute temperature (°K);
                 L  - length of soil cover (cm);
                P, - partial  pressure at the air-soil  surface (mm Hg);
               P,* - partial  pressure of the volatilizing material in soil gas
                     at the bottom of the soil depth (mm Hg); and
                 A - landfill surface area (cm2),

Applicability--
     This model can generally be applied to landfill situations where
molecular diffusion is the controlling vapor transport mechanism.  The model
accounts for the apparent velocities associated with high volatile
concentrations, but does not account for the convective sweep movement
associated with co-disposal (biogenic gas production) practices.

Limitations--
     The limitations indicated for both the Fanner Model and the Shen model
apply to this model.  However, the Logarithm Gradient Model accounts for
corrective transport resulting from the diffusion of highly volatile
constituents (greater than 5 percent by volume).

RTI Closed Landfill Model--
     The Research Triangle Institute (RTI) closed landfill model  (45848)
accounts for emissions via two mechanisms: diffusion through the soil cap and
convective loss from barometric pumping through passive landfill vents.  The
model is based on the Fanner Model (above) which was modified to account for
convective losses due to barometric pumping and the decline in emission rate
over time.
                                      133

-------
     The total instantaneous emission rate is a function of the total  initial
emission rate at the time of landfill closure which is the sum of the
instantaneous emissions associated with diffusion through the cap and
barometric pumping.
                                 Ij.
     where     E,   - total  initial  emission rate at the time of closure
                     (9/sec);
               En  - emissions associated with diffusion through the cap
                     (g/sec); and
               E21  * emissions associated with barometric pumping (g/sec)

     The equations for estimating E2i and E21 are present below.
                           L    l_"      U 
-------
     where
flow rate of gas through the vent (cm3/cm2-sec);
concentration of constituent in the gas within the
landfill (g/cm3);  and
landfill surface area (cm2).
hAEfw
Pr
.V

^+273
Tr+273
-1
                                                            (At  A)
                                                  (Eq. 29)
     where       h - thickness  of waste bed within landfill  (cm);
               Efw - air porosity fraction of fixed wastes;
                Pr - reference  barometric pressure (mm Hg);
                Px - final barometric pressure (mm ug);
                Tr - reference  landfill temperature (°C);
                T! « final landfill temperature (°C);  and
                At - time  interval over which change  in  pressure and/or
                     temperature occurred  (sec).
The total instantaneous  emission  rate at any time  is then computed  via  an
exponential decay function:
                      31.56  E1 exp(-At)
                                   (Eq.  30)
     where    E,(t)« total  time-dependent emission rate  (mg/yr);
                 *
                Et - initial emission rate, at time of landfill closure
                     (g/sec);
                 t - time since landfill closure  (mo); and
                 A » decay  constant  (mo"1)  - 2.63 x 106 E^/M^; and
               M,
                'ol
total mass of the constituent in the landfill (g)
The average emission rate from a closed, vented landfill  over  the  time since
landfill closure is given by the following expression:
 EA (t) - ^P i [1-exp (-At)]
                                      135
                                                                     (Eq.  31)

-------
     where  EA1(t) -  average emission rate over the time since landfill
                     closure (mg/yr);
                 t - time since landfill  closure (mo);
                 A - decay constant (mo"1); and
                 *
                E1 - initial emission rate at time of landfill closure
                     (9/sec).

     The RTI closed landfill model assumes that n© biodegradation occurs and
that the landfill is passively vented to  the  atmosphere.  Transport of the
constituent in moving water is assumed not to occur.

Applieability--
     The RTI Model estimates the time-dependent behavior of emissions from
landfills.  RTI modified the Farmer Model to  account  for the convective loses
from barometric pumping through vents.  In addition,  the RTI Model was
designed to account for the decline in the emission rate from closed landfills
ovtr an extended period of time.  Th§ time dependency was incorporated simply
by adding an exponential decay function.   The decay constant was taken as the
ratio of the emission rate at the time of landfill closure to the total mass
of the constituent in the landfill.
     The RTI Model does not include convective or purgi-ng action associated
with biogas production commonly found at municipal solid waste landfills.  It
assumes, as did the Shen and Fanner models, that the toxic properties of the
waste will inhibit biological processes and, thus, prevent biogas generation.
Furthermore, the liquid waste which contains the volatile constituent i is
assumed to be bound in the fixed waste within the landfill cell.  No
experimental or field verification has taken place.
                                      136

-------
4.5.3  Emission Models for Closed Landfills with Internal Gas Generation

The Scholl Canyon Model--
     The Scholl Canyon Model was developed by Emcon Associates and Jacobs
Engineering Company as part of a feasibility study of the recovery of methane
from the Scholl Canyon Sanitary Landfill in alifornia.  The model is a single-
stage, first-order kinetic model which assumes that after a lag time of
negligible duration, during which anaerobic conditions are established and the
microbial biomass is built up and stabilized, the production rate of meathane
is it its peak.  The gas production is then assumed to decrease as the organic
fraction of the landfill decreases.  The Scholl Canyon Model is:

                         QCH4 - k LQ 2 Mj .^ exp(-kti)              (Eq. 32)

where     QCH4  » methane flowrate m3/yr.
          k    • methane generation rate constant 1/yr.
          L0  - potential  methane generation of refuse,  m3/Mg of refuse.
          Mj  - mass of submass i,  Mg.
          t,  • year since the initial  placement for submass i,  yr.
          n    - number of submasses in the landfill.

The total methane generation from the entire landfill is at its peak upon
landfill closure.  If a constant annual acceptance rate is assumed, the form
of the model can be simplified to:

                  QCH4 - L0 R [exp (-Ice) - exp (k-t)]                 (Eq. 33)

where     R «  average annual acceptance rate of refuse, Mg/yr
          c -  the time since landfill closure.

A lag time during which anaerobic conditions are established can be
incorporated into the model by substituting (c + lag time) for  c and (t +  lag
time) for t.   Typical lag times range from 200 days to several  years depending
on landfill conditions.
                                      137

-------
The emission of volatile constituents can be calculated as a function  of
methane generation rate.

                              Qe - (QcH4) 2 (Cc)                        (Eq. 34)

where     Qc - Volatile constituent emission rate,  mVyr.
          2  - multiplication factor (assuming landfill gas consists of 50%
               methane and 50% carbon dioxide).
          Ce - volatile constituent initial  concentration within the landfill
               10°6 ppmv.
          QCH4 - methane flowrate, m3/yr.

Applicability--
     The Scholl Canyon Model can be used to estimate emissions from closed
landfills with internal gas generation assuming that all the gas that  is
generated is emitted.  The mass emission rate can be computed by using the
ideal gas law to convert from volumetric flowrate.   This model was selected
for use for the Clean Air Act Regulations for municipal solid waste landfills
which are to be proposed in the Fall of 1990.

limitations--
     The model assumes no lateral migration of emitted gases takes place*  No
verification of a  kinetic model using field data from landfills to describe
the time dependency of gas production has been performed.  This is the case
for most, or all,  of the models discussed in this document.  EPA's Air and
Engineering Research Laboratory is conducting work to field validate gas
production models  for the global climate change program.

(Thibodeaux) Convective "Add On" Model--
     The Convective "Add On" Model was developed by Thibodeaux to account for
both the diffusion and convective mechanisms present in landfills at which
co-disposal of municipal solid waste and hazardous organic waste has been
practiced.  The model describes the migration of a chemical constituent i due
to the convective  gas sweep of biological gas production within the soil cover
                                      138

-------
layer.  As the apparent velocity, Vy, of the constituent, approaches zero, the
model reduces to the diffusion controlled Farmer Model described above.
     The Corrective "Add Onfi Model is  (45):

                                 (crcs)
[exp (LyDE)-l]  + V y
                                             V  C  . A                (Eq. 35)
     where      Et - rate of vapor movement within the soil phase (g/sec);
                Vy • mean gas velocity in pore spaces (cm/sec);
                C, - concentration of  i  in sand chamber filled pore spaces
                     (9/cm3);
                Cs - concentration of  i at the air-soil  interface (g/cm3);
                  L - depth of fill cover  (cm);
                DE • effective diffusivity of  i within the soil pore space
                     (cn»2/sec); and
                  A » landfill surface  area (cm2).

Applicability--
     The Convective "Add On" Model can be used to estimate volatile emissions
from landfills with internal gas generation.  This model accounts for both
diffusion and convective transfer.  However, the transfer due to net upward
gas flow greatly  overshadows the diffusion transfer mechanism  (59).  This
deduction is based, in  part, on laboratory experiments of simulated gas flow
through a soil cover.

Limitations--
     A major limitation of this model  is  the required input of mean gas
velocity through  the soil cover.  The  limitations applied to the Farmer and
Shen Models apply to this model as well.  However, the "Add On" Model does
account for biogas convection.
                                      139

-------
SEAMS Model--
     The model recommended in the SEAMS manual for landfills with  internal gas
generation is a simplified version of the Convective "Add-On" model.  This
model assumes that the effect of biogas generation are so great that  soil and
gts phase diffusion becomes insignificant.  The simplified model is (41):

                           E. - C.V A                                 (Eq. 36)
                            i    i y
     where     E, -  emission rate of component 1 (g/sec);
               C1 -  concentration of component in the soil pore spaces
                     (9/cm3);
               Vy -  mean landfill gas velocity in the soil pores  (cm/sec);
               A  -  landfill surface area (cm2).

     This model was shown to produce emission rates within one percent of the
values obtained using the Conveetive "Add-On" model.  The SEAMS manual
provides guidance on estimating values for model parameters.
     The model is recommended by EPA-Superfund for estimating volatile
emissions from landfills with internal gas generation.

limitations--
     This model is limited by the accuracy of the values used for the mean gas
velocity^  A representative value for this parameter may be difficult to
obtainf or a specific site; however, Thibodeaux provides an average value of
1,63 x 10"3  cm/sec.   The accuracy of this  model  for  estimating  emissions from
landfills with moist or wet soils is unknown.

Thibodeaux b Model--
     The Thibodeaux b Model (45,49)  includes emissions due to barometric
pressure pumping.  The model accounts for the air emissions resulting from
concentration gradients  (diffusion), biogas generation sweeps  (convection),
and barometric pressure pumping  (convection) (51).  The barometric pressure
                                      140

-------
fluctuations create a pressure gradient within the landfill cell, pumping
vapors to the atmosphere.

     The Thibodeaux b Model incorporates Darcy's Law to characterize the
laminar flow, of gases flowing through porous media due to pressure gradients.
The 
-------
               k  -  permeability of soil cover layer material  (cm2 cp/sec
                     a tin);
               P  -  landfill cell pressure (atm);
               u  -  landfill cell gas viscosity (cp);
            "  *  -  atmospharic pressure (atm);
               L  -  depth of soil cover layer (cm);
               DE -  effective diffusivity of i within the air-filled soil
                     pore space (cm2/sec);  and
               A  - landfill surface area (cm2).

Applicability--
     The Thibodeaux b Model can be applied to situations where molecular
diffusion, convection, and barometric pressure fluctuations are to be
expected.  Efforts by Springer indicate that the annual barometric pressure
fluctuation does not average out to one and, therefore, should be considered.
Furthermore, Springer observed through computer simulations of flux with
benzene that barometric pressure fluctuation only slightly influenced the flux
ratt at co-disposal facilities, and greatly influenced the flux rate at
hazardous waste landfills where no internal gas generation is expected.

Limitations--
     The limitations observed in Thibodeaux's Convective "Add On" Model
applied to this modi! with the exception of barometric pressure fluctuations.
Springer's observations indicate that the Thibodeaux b Model is applicable to
landfills with no internal gas generation (51).

Exact Model--
     The Exact Model, (45) developed by Thibodeaux, is a steady state model
which accounts for diffusion due to concentration gradients, the convective
velocity created by highly volatile compounds, and convection due to biogas
generation (50).  Application of the model requires an iterative procedure,
since the flux term, Jis appears on both sides of the equation.  The Logarithm
Gradient Model discussed above is used to calculate the initial estimation of
the flux to start the iterative calculation.  Since it is a steady-state
                                      142

-------
model, the flux is not given as a function of time and the concentration
constituent i is assumed to be constant within the landfill cell.

     The Exact Model is (45):
WA"A
  RT
                              P.
                         1 -   -
               exp
+
                                                      RT
LRT/PTD
Eq. 41)
     where
               I
               0
               EN
volatile chemical flux (g/cm2-sec);
apparent biogas velocity (cm3/cm2-sec) ;
vapor pressure of chemical  A (atm);
molecular weight of chemical A (g/mole);
atmospheric pressure (atm);
soil layer thickness (cm);
effective diffusion coefficient (cmz/sec);
(EN, DA1
enhancement factor from experimental  data;
diffusion coefficient of chemical  A in air;
porosity of soil layer;
air-filled porosity of soil  layer;
molar gas constant; and
absolute temperature (*K).
Applicability—
     The intended applicability of this model is municipal landfills with
internal gas generation.  This model can be used to estimate the flux of a
volatile constituent i from closed landfills due to both diffusion and
convective transport mechanisms.  However, this model can only be used for
landfills with an internal gas velocity greater than zero.
                                      143

-------
Limitations--
     When the model predictions were compared to the experimental data
developed by Thibodeaux, large discrepancies were observed.  The experimental
emission rates were higher thin the model predictions, and Thibodeiux
attributes this deviation to surface diffusion which occurs in parallel with
pore diffusion and, in general, enhances the total diffusion rate.

     Thibodeaux incorporates the enhancement factor, EN, in the Exact Model to
account for the discrepancies observed between the effective diffusivity
(obtained by laboratory experiments) and the Millington-Quirk diffusivity.
Therefore, one major limitation of Thibodeaux's Exact Model is the
availability and accuracy of the enhancement factor.  Another limitation of
Thibodeaux's Exact Model is that the value of the apparent gas velocity, V,
needs to be known.  In order to apply Thibodeaux's Exact Model, the
enhancement factor needs to be experimentally obtained for the specific
constituent, soil type, soil cover depth, gas type,  gas velocity, and gas
     Thibodeaux's Exact Model may be useful in estimating the order of
magnitude value of the emission flux if the soil .cover depth is less than 10
cm.  The enhancement factor for a shallow soil cover (i.e., 7.62 cm) ranges
from L58 to 4,93 compared to the range of 5,54 to 17o2 for a deep soil cover
layer (i.e., 38,1 cm),

4o5.4  Emission Models for Open Landfills

Emission Models for Open Landfills The Scholl Canyon Model--
     The Scholl Canyon Model, a single stage, first order kinetic model, can
be applied to calculate methane generation within an open landfill.  For open
landfills it has the following form:
                    QCH4  - k L0
                                      144

-------
where     QCH4 - methane  flowrate  m3/yr
             k - methane generatioh rate constant,  1/yr
            L0 - potential  methane generation of refuse, m3/Mg of refuse
            Mj - mass of submass i, Mg
            tt - year since the initial placement for submass i, yr
            n  * number of submasses  in the landfill

A simplified form of the equation  can be used if a  constant annual acceptance
rate is assume.  This is the form  of  the simplified model:

                   QCH4 - L0 R [1-exp  (kt)]                            (Eq.  43)

where     R - average annual acceptance rate of refuse, Mg/yr

A lag time during which anaerobic  conditions are established can be
incorporated into the model by substituting (c + lag time) for c and  (t +  lag
time) for t.  Typical lag time range  from 200 days  to several years depending
on landfill conditions.

     The emissions of volatile constituents can be  calculated as a function  of
the methane generation rate.

                          Qc ' (Qow) 2 (Ce)                            (Eq.  44)

where     Qc * Volatile constituent emission rate,  m3/yr
          2  - multiplication factor  (assuming landfill gas consists  of 50%
               methane and 50% carbon dioxide).
          Cc • Volatile constituent initial concentration within the  landfill
               10"6,  ppmv.
        QCH4 »  methane fTowrate m3/yr.
                                      145

-------
Applicability--
     The Scholl Canyon Model can be used to estimate emissions from open
landfills assuming that all the gas that is generated is emitted.  The mass
emission rate can be computed by using the ideal gas to convert from
   umetric flow rate.
     There are no available field data from landfills which will allow
verfication of
-------
                     Equilibrium Vapor Pressure, %
R«f«rtne«: Sh«n, TT^ "estimating Hazardous Air Emissions from Disposal Slt««." Pollution
                     August 1981.
 Figure  21.   Pick's  correction factor,  Fv,  plotted against equilibrium mole
              fraction,  y .
                                      147

-------
where     V° - volume of vapor released at ambient pressure and temperature
               (cm3);
          y* - equilibrium mole fraction of the volatilizing component  in  the
               gis phase at the liquid-gas interface;
          A »  area of the liquid surface (cm);
          D, <* diffusivity of volatilizing component in air (cm2/sec);
           t - time  (sec);
          Fv - Pick's Law correction factor;  and
           * - 3.1416

Applicability--
     The Arnold Open Landfill Model is applicable to open landfills and dumps
where emissions are due to volatilization of liquid at the landfill surface.
         model assumes that emissions do not occur due to biogas production or
barometric pumping.  The model also does not account for the effects of wind
speed which would tend to increase thi emission rate.  Finally, the model
assumes a constant waste source, whereas, a surface crust may form as volatile
constituents are lost from the surface soil.  This crust may then act similar
to a landfill cover.  This model does not account for transport of chemicals
by diffusion through the soil pores.
     '$
     Shin's Open Landfill Model  (44,45) provides the average volumetric
emission rate of the volatile constituent from open landfill surfaces.   Shen
modified the Arnold Open Landfill Model to account for the convection due  to
wind speed.  Shin took the time  derivative of the Arnold Model and changed the
time function, t, in the model to a position function.  This position function
is related to the length of the  open dump and the wind speed.
                                      148

-------
     The Shen Open Landfill Model is
                                   2 * W
                                   *"  WL       -
                                              * FV
                               _           ^

     where
M                   (Eq. 46)
              dt
                                                   »
                     average emission rate (cm /sec;;
               y" »  equilibrium mole fraction;
               WL  « width of open landfill  (cm);
               D,  « diffusion coefficient (cm2/sec);
                U  - wind speed (cm/sec);
               Fv  - Pick's law correction factor;
               LL  - length of open landfill  (cm);  and
                w  - 3.1416

     The model has also been presented as Ziegler's modification of Arnold
as (46,50):

                         dV * 2 CeW (DLvAFY)1/2Wi                    (Eq. 47)
                         dt

when     dV  - emission rate;
          dt
          W  « width of landfill;
          L  * longest dimension of the landfill;
          v  - wind speed;
          Wi  - weight fraction of a specific compound in the waste;
          Ce  * equilibrium vapor concentration;
          D   « diffusion coefficient; and
          FY  - correction factor.

     The emission rate can increase with increasing wind speed; however, the
dilution fraction also increases.  The net effect of wind speed on ambient
concentration, therefore, becomes compensative and depends on receptor
location.
                                      149

-------
Applicability--
     The Shen Open Landfill Model is applicable to open  landfills  where
emissions are due to volatilization at the landfill surface.   It appears that
this model is useful as a screening process to examine whether volatilization
will be significant for a given contaminant (47).

Limitations--
     The model does not provide for biogas generation or barometric  pumping.
The model also does not provide for the formation of a weathered surface which
would tend to reduce emissions similar to a closed landfill cover.

RTI Open Landfill Model--
     Research Triangle Institute (RTI) modified the Shen Open  Landfill  Model
by introducing the mole fraction of the constituent to account  for more than
one volatile constituent in the liquid.  The RTI Open Landfill  Model,  shown
bilow, uses the  ideal gas law to convert the volumetric emission of  the Shen
Open Landfill Model and provide an average mass emission rate.

     The RTI Open Landfill Model is (42,43):

                              2PM.Y.* U,       D.L  U
                        E  .     11   L      JJ.                    (Eq,  43,
                          B        s\ I          " r»|

     where     Ef - average mass emission rate of component i  (g/sec);
               P  - ambient pressure (mm Hg);
               M, - molecular weight of component i;
                *
               Y, - equilibrium mole fraction of component i in the gas  phase;
               WL - width of open landfill;
               D, • diffusivity of component i in air (cm2/sec);
               LL * length of open landfill  (cm);
               U  * wind speed (cm/sec); and
               Fv - Pick's Law correction factor.
                                      150

-------
Applicability--
     The RTI Open Landfill Model is applicable to open landfills where
emissions are due to volatilization at the landfill surface.

Limitations-.-
     The model does not account for biogas production, barometric pumping, or
formation of a surface crust.  The model does not account for depletion of the
volatiizing chemical from the waste surface; thus, it is not applicable for
estimating emissions over long time periods (i.e., months).

4o5.5  Emission Models for Landtreatment

RTI Land Treatment Model--
     Research Triangle Institute (RTI) has developed a model (44,45) for
estimating emissions from land treatment areas.  The model is comprised of two
equations; one for short time (immediately after application or tilling) and
one for longer times.  The RTI land treatment equations are based on the
premise that emissions are limited by vapor diffusion through the soil.  The
model accounts for the removal of organic material from the land treatment
area by both biological degradation and air emissions.

     The expression for the  instantaneous emission rate for short time periods
immediately following initial waste deposition is given by:
     where
               M,
                o
                1 -

1
e
IT t
Keq°e
1/2
                                  gt/tb
                                                                     (Eq. 49)
emission rate of constituent (g/cm2-sec);
area loading of constituent (g/cm2);
depth of waste in open landfill (cm);
volume fraction of air-filled voids in the soil
(dimensionless);
                                      151

-------
               Keq -  ratio of gas-phase constituent to total constituent in
                     solid waste (dimensionless);
               Kg -   gas-phase mass transfer coefficient (cm/s);
                t  -  time after waste application  to the landfill site (sec);
            •   De *   effictivt diffusion coefficient of constituent in the
                     solid waste (cm2/sgc);  and
               tb -   time constant for biological  decay (i.e., time required
                     for 63.2% of constituent to be degraded.

     For longer times after application or tilling, when most  of the
constituent is not present in the soil, the short-term equation will over
estimate air emissions.  Under these conditions, the following equation is
applicable:
     where
                         Keq

                                            4  I
                                                          e
                                                                     (Eq. 50)
emission rate a long time after application or tilling
(g/cm2-sec);  (all  other  parameters  are  the  same as
presented above).
     The RTI Land Treatment Model is applicable to sites where liquid or
semi-liquid waste is applied to the soil  surface.   The model  assumes that
emissions from the surface are limited by diffusion of vapors through the pore
space of the soil/waste mixture.  The model  accounts for removal of organic
material from the soil/waste mixture by both biological degradation and air
emission.

Limitations—
     The model is not applicable to wastes which are easily biodegraded, or to
sites with highly porous soils which allow easy vertical migration of the
liquid waste.  The model does not account for losses to these or other
pathways.
                                      152

-------
SEAMS Model--
     The SEAMS 1andtreatment model is a simplified version of the Thibodeaux-
Hwang model.  The Thibodeaux-Hwang model was simplified by assuming that the
oil layer diffusion length value is low (i.e., that the spilled contaminant
has become incorporated into surface soils and is not present as a discrete
film on the soil particles).  The SEAMS model is designed to calculate an
average emission rate over time.
                                                                     (Eq. 51)
                                  d +
                                   2 OC$t
                                   ~
where
          0
          R
          T
          A
          d
          t
                     average emission rate of component i over time (g/sec);

                     Di  
-------
4.5.6     Leaks and Spills on Soil

SEAMS Model (Fresh Spills)--
     The SEAMS manual presents the following model for estimating volatile
releases from spills or leaks where a contaminant pool is visible on the  soil
surface (41).
                      E. - Ki6 C* A                                  (Eq. 52)

     where    ^Et  -  emission rate of component i (g/sec);
               Kig-  gas phase mass transfer coefficient of component  i
                     (cm/sec);
               C.  -  vapor concentraiton of chemical i (g/cm3); and
               A   -  contaminated surface area (cm2).
Guidance on methods for estimating K.g and C. is presented in the SEAMS
    iicability--
     This model can be used for estimating air emissions of volatile species
from contaminant pools on soil surfaces.
     The model does not consider the soil phase mass transfer resistance,
therefore it is not appropriate for use when spilled contaminants have seeped
into surface soils.  Also, since the model does not consider the liquid phase
resistance, it is only useful for estimating releases of pure compounds.

SEAMS Model (Old Spills)--
     The SEAMS manual recommends the use of the SEAMS land treatment model  in
cases where past spills, leaks or intentional disposal directly onto surface
soils have resulted in contaminated surface soils with liquids in the pore
spaces.  This model is described in Section 4.5.5.
                                      154

-------
4.5.7     Emission Models for Fugitive Dust

     Fugitive dust at hazardous waste sites (airborne wastes or contaminated
soils) most commonly results from wind erosion of the wastes or vehicular
travel over unpaved contaminated roads.  The U.S. EPA has developed equations
to estimate fugitive dust emissions arising from vehicle travel on unpaved
roads (AP-42).  The U.S. Soil Conservation Service (SCS), and Cowherd and
Gillette have developed models for predicting fugitive dust emissions
resulting from wind erosion from soil surfaces.

SCS Model--
     The SCS model (71) takes into account such factors as surface soil
moisture content, roughness, and cloddiness, type and amount of vegetative
cover, wind velocity and the amount of soil surface exposed to the eroding
wind force.  The SCS equation can be expressed as:

                           E - f(I-,CMC',L',V)                   -   (Eq. 53)
where     E  - potential annual average wind erosion soil loss;
          I* « soil credibility index;
          C* • climatic factor;
          K* - soil ridge roughness factor;
          I" - field length along the prevailing wind direction; and
          V  - vegetative cover factor.

     For the sake of brevity, the details of the calculation method are not
presented in this document.  The reader is directed to the Skidmore and
Woodruff (59) source document and the SEAMS Manual (41) for further guidance.

Vehicular Traffic-
     The US EPA has developed the following equations which can be used to
estimate fugitive dust emissions resulting from vehicular travel on
contaminated unpaved roads (54);
                                      155

-------
             EVJ - k(5.9) (fg)

or In metric form:
                                           -} Q 5 f    "  p)          (Eo.   55)
                                            '     *        '          l  q'   '
                                                     365
where EVT  -    emission factor for vehicular traffic (Ib/vehicle mile
               traveled, kg/vehicle kilometer traveled);

     k    -    0.45 * particle size multiplier for particles <10 urn  (i.e.,
               particles that may remain suspended once they become  airborne
               and which can be inhaled into the respiratory system);

     s    -    silt content of raod surface material (percent);

     Sp   -    mean vehicle speed (mph, kph);

     W    -    mean vehicle weight (tons, Mg);

     w    -    mean number of wheels; and

     D    »    number of days with at least 0.254 mm (0.01 inch) of
               precipitation per year.

     The SCS wind equation computes total fugitive dust emissions due to wind
erosion which result from the combination of surface creep, saltation,  and
suspension.  If only the fraction of soil loss that is suspendable and  trans-
portable over significant distances by wind is desired, the wind equation must
be adjusted (reduced) to reflect emissions from only this phenomenon.

Applicability --
     The SCS model is applicable to wind-blown dust.
                                      156

-------
Limitations --
     The SCS model is designed to estimate annual erosion losses and is not
reliable when altered to estimate short term emissions (e.g., 24-hour
emissions).

Cowherd Model --
     Cowherd (55) has developed a predictive equation for estimated respirable
particulate emissions from soils of "limited erodibility."  The annual average
nti of respirable particulate emissions is a function of surface and climatic
factors represented by the following equation:

                     E10 -  0.83  f (1-V)  P(UJ  (50/PE)2               (Eq. 56)

where     E10   -    Annual average PM10  emission  factor  (mg/m2 - hr).

          f    »    Frequency of disturbance per month.

          U+   -    Fastest mile of wind for the period between disturbances
                    (m/sec).

       P(lT)   -    Erosion potential,  i.e., quantity on the surface prior to
                    the onset of wind erosion (g/mz)

          V    *    Fraction of contaminated surface area covered by
                    continuous vegetative cover.

          PE   -    Thornthwaite's Precipitation Evaporation Index.

     Soils of limited erodibility are nonhomogeneous surfaces impregnated with
nonerodible elements (stones, clumps of vegetation, etc) and contain a finite
reservoir of erodible material.  In contrast, bare surfaces of finely divided
material are characterized by an "unlimited reservoir" of erodible particles.
Guidelines for characterizing the wind erosion potential of a surface and for
evaluating the terms in the Cowherd model are presented in the EPA manual
                                      157

-------
entitled Rapid Assessment of Exposure to Participate Emissions from  Surface
Contamination Sites (55).

Applicability --
     The Cowherd model is applicable for estimating the emission rate of
respirable (less than 10 microns in diameter) wind-blown dust from surfaces of
limited erodibility.

Limitations --
     The model is designed for estimated an annual average emission  rate.  It
is not designed to estimate short-term emissions.

Gillette Model --
     Gillette (55) has developed the following model for estimating  annual
respirable particulate emissions from wind erosion of surfaces with  an
"unlimited reservoir" of erodible particles:
               E10 - 0.036 (1-V)
                                  JL
U
                                    t
       F(x)                       (Eq. 57)
where     E10   *    Annual average PM10  emission rate  (g/m2 -hr).
          V    *    Fraction of contaminated surface vegetative cover.
          U    *    Mean annual wind speed (m/sec).
          Ut   *    Threshold value of wind speed at 7m (m/sec).
          x    -    0.886 Ut/U (dimension less  ratio).
          F(x) *    Empirical function.

     Soil moisture is not taken into account in this equation because highly
arodible soils do not readily retain moisture.   Gillette modified equation  55
slightly to obtain the following expression for estimating worst-case 24-hour
average emission rates:

                   E10  -  0.036 (1-V)  (U.  -2)3                          (Eq. 58)
                                      158

-------
where     E1Q  -     24-hour average PM10 emission rate (g/m2  -hr).

          V    *     Fraction of contaminated surface  area covered  by
                     vegetation.
            V
          U+   *     Mean  annual fastest mile of wind  (m/sec).

     It should be noted,  however, that the 24-hour average emission rate
predicted by this equation is based on annual average parameter values  for  U+.
Guidelines for evaluating the terms in the Gillette equations are  presented in
the EPA manual entitled Rapid Assessment of Exposure  to Particulate Emissions
from Surface Contamination Sites  (55).  The reader is referred to  this
document for a more  detailed discussion of the models.

Applicability --
     The Gillette models  are applicable for wind-blown dust emissions from
highly erodible surfaces. Both annual and short-term emissions can be
estimated.
Limitations  --
     The short-term model  is  based on  annual average parameter values  and may
overestimate the  24-hour average  emission rate.

Vehicular Traffic-
     The U.S. EPA has  developed the  following equations which can  be used   to
estimate fugitive dust emissions  resulting  from vehicular travel on
contaminated unpaved roads (54):


              r      t/c qi ,_ii ,Sj>>  ,W, 0.7 ,w. 0.5  r365-DD~|       (Eq.  59)
              tyy - Kp.sj 112> ^30J  13;     ^4^      [__ 355  _J

or in metric form:
                                      159

-------
                                                               0=3
                                                         365-Dp
                                                          365  J

where     Eyj*  emission factor for vehicular traffic (Tib/vehicle mile
               traveled, kg/vehicle kilometer traveled);

          k -  0.45 « particle size multiplier for particles <10um  (i.e*,
               particles that may remain suspended once they become airborne
               and which can be inhaled into the respiratory system);

          s -  silt content of road surface material (percent);

          Sp - mean vehicle speed (mph,kph);

          W •  mean vehicle weight (tonssMg);

          w -  mean number of wheels; and

          Dp - number of days with at least 0.254mm (0.01 inch) of
               precipitation per year.

     To estimate fugitive emissions due to vehicle travel for a given time
period, the emission factor, EVT defined above, is multiplied by the  vehicle
miles traveled during that time period.  Maximum release conditions  may  be
estimated by using a small value of Dp in the model  to reflect assumed
drought conditions.  Average emissions can be estimated by using annual
average value for Dp.  Whenever possible, values for climatic and soil
parameters should be obtained for the particular site in question.  Cowherd
et.al. (55) provides default values for model parameters that can be  used when
site-specific data are not available.
                                      160

-------
Applicabillty--
     The AP-42 dust model  is only  applicable to dust  resulting  from  vehicular
traffic.

4.5.8  Additional Models

     Additional models  identified  but not  included here are: Hwang's
modification of Farmer,  (52,60) RTI Open Dump Model,  (44) Hartley  Method,
(37,61) Hamaker Method,  (61) and Dow Method (61).  The latter three  equations
were developed for volatilization  of pesticides applied to soil.

4.5.9  Non-Aerated Lagoons

Mackay and Leinonen Dynamic Two-Film Model —
     This dynamic model  (56,57) best serves those instances involving  isolated
disposal of a given quantity of waste, as  opposed to  the steady-state  scenario
offered by the other models.   Laboratory validation of this model  was  reported
(57).

     This model assumes  that nearly stagnant films of well mixed bulk  air  and
water systems occur on  both sides  of the liquid/air interface.

                        NI " KiL (Ct - PiA)                           (Eq.  61)
                    C,  - P,/H,  + (C10  -  P/H,) exp (-K1L t/L)
where     N,  - mass flux rate (mol/m2->
          K1L - overall  mass transfer coefficient (m/hr);
          C1  * concentration of i at time t (mol/m3);
          P,  » equilibrium partial pressure of i in the vapor (atm);
          H,  - Henry's Law constant for i (atm-m3/mol);
          Cio • initial  concentration of i  at t - 0 (mol/m3);
          t    - time  (hr);  and
          L    - depth of lagoon  (m).
                                      161

-------
    icability--
     The model offers an alternative to the steady-state scenario.  However,
Wetherold (37) reports that despite its theoretical validity, the model  is
difficult to apply to the "real world."  The input parameters are difficult to
determine or" find in available literature.  And, this modal best applies to
the emissions of single compounds.

Limitations--
     The model fails to provide for white caps, thermoclines, eddy diffusion
and other similar phenomena, tending to under-predict emissions when these
conditions occur at the lagoon.

Thibodeaux, Parker and Heck Model--
     This model (53)s which can be applied to both non-aerated and aerated
lagoons, evolved from basic accepted theories of mass transport.  It is used
to determine emission rates of individual compounds,  assuming that the
concentration of each compound remains constant in the aqueous phase (it does
not interact with the other compounds present}.,  It also assumes that the
influx of the compound is steady, that its bi©degradation rate is steady, and
that the lagoon surface can be clearly separated into either quiescent (non-
aerated) or turbulent (aerated) zones.

     To use the equation, four mass transfer coefficients must be determined
from impartial relationships.

                            q,  - M,^, (X,  -  X,*)                      (Eq. 62)


For each volatile component i:

                                      162

-------
                                                                     (Eq-65)
where     q^  » flux of component i from the lagoon surface (g/cm2-sec);

          Mt  » molecular weight of component i (g/g-mol);

          K^* overall liquid-phase mass transfer coefficient for component i
               (mol/cm2-s);

          X1  * mole fraction of component i in the aqueous phase (this must be
               measured); and

          X,*- mole fraction of component i in equilibrium with the mole
               fraction of i in air, Y1  (if Yt  is  assumed to be negligible,
               X,* can equal  0);

     Kp kj *  overall liquid-phase mass transfer coefficient for
               aerated non-aerated zones of a lagoon,  respectively (mol/cm2-
      A^.,  A,, »  surface areas of aerated and unierated zones,  respectively
               (cm);

     1C, kj -  individual liquid phase mass transfer coefficients for
               the aerated and unaerated zones, respectively (mol/cm-s);
     K*, K" -  individual gas phase mass transfer coefficients for the
               aerated and unaerated zones, respectively (mol/cm-s); and

           H • Henry's Law constant in mole fraction form (y * Hx).
                                      163

-------
Applicability--
     The Thibodeaux, Parker, and Heck model is applicable to aerated,  non-
aerated, and combined lagoons.  The model assumes steady-state conditions and,
therefore, is applicable to undisturbed lagoons.  The model is probably not
applicable to disturbed (disturbed sludge) conditions.
     The accuracy of this model has not been verified as of 1982.  Wetherold
(37) expressed some skepticism regarding the accuracy or availability of some
parameters necessary for the calculation of the mass transfer coefficients.
Also, the model requires additional development to satisfy the need for a
predictive model capable of predicting total VOC emissions from a lagoon
containing a complex mixture of compounds (37)»
         way to deal with this, as suggested by DeWolf (53), is to sum up the
emissions estimated for several classes of compounds by selecting a
representative compound from each class.  Acknowledging this selection as
"arbitrary", DeWolf provides some suggestions, noting that some compounds are
more likely to be encountered and those in mid-molecular weight range of 4 to
8 carbons are "likely to dominate in frequency of occurrence".  He suggests:

                                         Compound
          Olefins
          Aromatics                       Toluene
          Haloganated hydrocarbons        Methylene chloride
          Oxygenated hydrocarbons         Acetone

Smith, Bomberger, and Haynes Model--
     The Smith et al. model (37) is applicable to emissions prediction for
highly volatile compounds in a lagoon setting.  The model is not applicable to
low and intermediate volatility compounds.  Also, liquid phase resistance
should be the controlling resistance.

     The volatilization rate is expressed as a first-order kinetic equation.
                                      164

-------
                                 E -  (Ka)env (c)                       (Eq. 66)

                                      Ka                               (Eq. 67)
                           env
where      E • mass emission rate per unit volume (Ibs/gal-day);

     (Ka)env *  volatilization rate constant for compound a in the environment
               (day'1);

           c » concentration of compound a (Ibs/gal);

      Ka     » ratio of volatilization constants of compounds a and oxygen as
     K*
       L*b      K lab measured in laboratory (dimensionless); and

     (K')env - oxygen reaeration rate in the environment (day"1).

Applicability-
     The model is applicable to volatilization of high volatility compounds
from non-aerated waste disposal lagoons.
     The model is limited in that it is designed to predict emission rates of
highly volatile individual compounds and it may be difficult to apply to
complex multicomponent wastes.  The model is not appropriate for estimating
emissions of low or intermediate volatility compounds.  Also, Wetherold (1982)
notes that determining the ratio of volatilization constants of a compound
(K*/K*)  is expensive in the laboratory; attempts to estimate this ratio simply
using diffusion coefficient values increase the model's overall uncertainty.

Shen Model--
     The Shen Model (48,58) presents an empirical equation for determining
volatile emissions from lagoons.  The Shen Model is (48):

                        ERP1  -  18 x  10'6  KL1 AC,                      (Eq. 68)

                                      165

-------
where   ERpi  - emission rate potential of  compound  i  (g/sec);
           A » lagoon surface area  (cm2);
          Cf  » concentration of compound i in lagoon  (mg/1); and
         KL1  - liquid-phase mass transfer  coefficient of compound i  (g-mol/
          KL1  - 4.45 x 10'3 (M^T0-5 (1.024)t'20 (U)  °'67 (H)'0'85          (fq.  69)

where     M, * molecular weight of compound i (g/mole);
           t •  lagoon  surface temperature  (oc);
           U »  surface velocity - 0.035 wind speed  (cm/sec);  and
           H »  average liquid depth of the lagoon (meter).

Applicability--
     The Shen Model  is applicable as a screening technology to  estimate
volatile emission rates from lagoons.  It  appears,  from Shen's  discussion,
that the model  assumes the lagoon is a dilute water solution, although  this  is
not explicitly  stated.  Shen indicates that the model  should  only  be used  when
"emission rates and  risks are clearly acceptable or unacceptable.'
     The model should be limited to use as  a screening technology.
RTI Model--
     The RTI Model is a simple volatile constituent mass  transfer model  (44):

                               E, * K,A C,                             (Eq.  70)

where     E, - air emissions for component  i  from  the  liquid surface (g/sec);
          K, - overall mass transfer coefficient for component i  (m/sec);
           A - liquid surface area  (m2); and
          C, - concentration of component  i  in  the liquid phase (g/m3).

                                      166

-------
     The calculation of the mass transfer coefficient (K,)  will  depend on
whether the lagoon is quiescent, turbulent, a combination of quiescent and
turbulent, or has an oil film.  In addition, the equation can be adjusted to
account for losses due to biodegradation.  Several methods for calculation of
KL are given«in the listed reference as well  as  examples  for applying the
model to specific site types.

Applicability—
     The RTI Model is applicable to assessing volatile emissions from aerated
and non-aerated lagoons.  The model is applicable to quiescent and turbulent
lagoons and can be adjusted to include biodegradation, although the toxic
nature of most waste lagoons will limit biological activity.  The model is
applicable to both undisturbed and disturbed site conditions.

Limitations--
     The model is not applicable to lagoons with a surface crust.  While the
calculation of the mass transfer coefficient includes wind speed for quiescent
lagoons, the turbulent lagoon calculations appear to consider wind speed to
have negligible iffect on the emission rate.

4o5.10  Aerated Lagoons

Thibodeaux, Parker, and Heck Model--
     This model is described in 4.5.9.

RTI Model--
     This model is described in 4,5.9, and can be adjusted for aeration.

Chemdat 6--
     EPA has published a number of models for RCRA sites, including models for
aerated lagoons (75).  This same model is also available in a more user-
friendly version known as the Surface Impoundment Modeling System (SIMS).
                                      167

-------
4.5.11  Sources of Model Input Data

     The models presented in the previous sections require a wide variety of
input parameters.  Table 17 provides a list of references which contain
tabulated data for some af the common chemical and soil property parameters;
hwoever, many of the models require additional input data such is mass
transfer coefficient, soil vapor concentration,  etc.  In these cases, the
reader is referred to the source document for each model.  The authors
generally provide guidance on methods for estimating speicific model
parameters.

     In addition, a number of existing data bases may be useful for supplying
the data requirements of the emission model.   Data bases that contain landfill
facility data, chemical property data, geophysical data, and meteorological
data are described below.  Hore detailed information regarding the data bases
is contained in Appendix D.  Physical and chemical properties for compounds
frequently encountered at Superfund sites are given in Appendices F and G.

     Landfill Data Bases

     Five data bases have been identified that may provide the landfill
facility data required to estimate emissions  using available techniques.
     «    Solid waste landfill survey;

     «    National Survey of Hazardous Waste Treatment,  Storage, Disposal, and
          Recycling Facilities;

     «    National Survey of Hazardous Waste Generators;

     *    Industrial Subtitle D Facility Study - Mail  Questionnaire;

     •    Industrial Subtitle D Facility Study - Telephone Survey,
                                      168

-------
Chemical Property Data Bases

*    EPA Chemical Properties Data Base; and
©    National Library of Medicine Online Service (HSDB)
«    EPA GEMS Database

Geophysical Data Bases

*    GEMS Geoecology Data Base; and
          Soil Temperature Data Base.
Meteorological Data Bases

«    STAR Data Base; and
«  '  NCDC Data Bases.
                                 169

-------
             TABLE 17,   DATA SOURCES FOR SELECTED MODEL PARAMETERS
Model Parameters                                 Reference  Number

Henry's Law Constant (H)                         44,  51, §6,  62
Diffusion Coefficient  (D)                        41,  44, 53,  63
Soil Porosity (Pt,P8)                             41,  64
Vapor Pressure (P*)                              44,  51, 53,  56,  63
Methane Generation Rate Constant (k)             73,  74
Potential Methane Generation of Refuse  (tT0)      73,  74
                                      170

-------
                                   SECTION 5
                                 CASE STUDIES

      Section 5  is a collection of five case studies that demonstrate  the
protocol described in this manual for developing BEEs.  The purpose of this
section  is to document different experiences regarding site investigation  and
characterization and to demonstrate the protocol for developing BEEs as
applied  to these case studies.  The sites selected represent different  regions
of the country, different types and distributions of waste, varying levels of
air emissions potential, and varying levels of historical air pathway  analyses
(APA) performed  in support of the Remedial  Investigation/Feasibility Study
(RI/FS)  process.

     Only the first case study demonstrates application of the complete
protocol.  The protocol was implemented only partially at the other sites.
The assessment techniques used (or not used) in these case studies do not
necessarily represent the best or most technically suitable assessment
techniques.  Many factors influenced the decision-making process concerning
the development of BEEs leading to air pathway evaluations.  Also, the work at
these sites was conducted without the benefit of a formalized protocol  for
designing APA programs and developing BEEs.

5.1  CASE STUDY 1:  PETROLEUM WASTE LANDFILL/LAGOON

     Case Study 1 is a disposal area for wastes from a defunct refinery.

5.1.1     Site History

     The petroleum waste site resulted from years of dumping bottom sludge
from refinery vessels and tanks at a disposal area located adjacent to the
refinery.  The on-site disposal activity was performed as general  refinery
upkeep and was typical  of the oil  industry at that time.
                                      171

-------
     This small refinery was located in Southern California between  the
foothills of a mountain range and a small community.  The refinery dumped  its
wastes on site from about 1930 to 1950.  In 1952, the refinery was dismantled
except for an old garage and several tanks.  Since 1953, the site has  been a
crude oil pumping station.  No known dumping has resulted from the pumping
station operation; therefore, all waste dates back over 35 years.  At  present,
the property is separated from an elementary school and a number of  residences
by a fence and a drainage channel.  The site is shown in Figure 22,

     The refinery dumped most of its sludge in a landfill on the western edge
of the property.  The landfill  covered approximately one acre of surface area
and was bermed at the middli at some unknown time,  thereby separating  the
landfill at the north end from a lagoon at the south end.  The landfill is
bilieved to have resulted from dumping soil into the landfill  to solidify the
liquid waste.  Investigations of the site disclosed that the entire
landfill/lagoon contained roughly 11,100 cubic yards of waste to a depth of
about 6 feet.  The waste was an oily sludge, with an odor and appearance
typical of refinery wastes.

     The site is subject to hot summers and mild winters.  Precipitation is
approximately 20 inches per year, occurring predominantly during the winter
months.  During site work, winds generally were light and easterly or
northeasterly during cooler periods.  During warmer periods, onshore sea
bfiizts yielded moderate breezes from the west and  southwest.   The residential
neighborhood was downwind of the lagoon and landfill  most of the time.

5.L2     Objectives

     The objectives of the site work,  from an air pathway perspective, were
several-fold: provide estimates of the undisturbed  and disturbed site
emissions; to use BEEs to develop a litigation plan;  and to conduct ambient
monitoring during the remedial  investigation and mitigation to ensure  worker
and community protection through the setting of appropriate action levels.
Table 18 summarizes the activities conducted at the site to address the air
pathway.   These activities are  described below.
                                     172

-------
N
    30-foot M«t«orologieai Tower
    1 0-foot Meteorological Toww
    Suspcatd Disposal Areas. 1-8
  0  '00 200  300 <00 JOB
               Figure  22.  Location of suspected disposal  areas.
                                     173

-------
                TABLE 18.   APA ACTIVITIES CONDUCTED AT THE SITE
APA Objectives
Determine the baseline and disturbed emissions for the site using direct
emission measurement techniques.  Protect on-site workers and the public from
air emissions during the investigation by using air monitoring and action
levels to stop site work.

Scoping
The lagoon and landfill were determined to contain petroleum waste that had a
moderately high volatile organic air emissions potential.  Particulate
emission potential from the sludge/tar-like solid waste was considered to be
low.
Screening Measurements
The site was surveyed using real-time instruments (indirect technique) for
indicator compounds on a grid system.  Soil samples were collected for head
space analyses to assess air emissions potential (direct technique}.  These
data were used to design the in-depth measurement strategy.
In°Depth Measurements
The in-depth emission measurements included:
     t    Undisturbed baseline emission measurements using the surface
     •    Disturbed baseline emission measurements using the downhole emission
          chamber; and
     •    Air monitoring for worker and public protection.
These data were used to develop undisturbed and disturbed site BEEs.
Mitigation
Undisturbed and disturbed BEEs were used to develop a remedial alternative
that included excavation and removal of the waste and air emissions control
                                      174

-------
 5.1o3      Scoping

      In  late  1980,  the owner sought  regulatory  agency  approval  to  remove  the
 waste  and  apply  it  to site roads.  California's Regional Water  Quality  Control
 Board  denied,the request  after  two sets of  samples, sent to  two independent
 laboratories,  disclosed high lead concentrations  in the waste.   The  owner then
 sought a more  thorough environmental  evaluation of the site  and
 recommendations  for remediation.

     The initial  task in  completing  the scoping phase of the site
 characterization was a review of existing data.   Information was collected
 from the owner's site files,  from files at  a similarly operated refinery,  from
 public agencies,  and from available  references.   The collected  data  were
 reviewed to provide a working knowledge of  the  site history, conditions,  and
 environmental  setting.  Essentially  no activities had been conducted to
 determine  waste  volume, environmental impacts,  emissions characteristics, or
 waste  existence  and type  in  other suspected disposal areas.

     The site  investigation  was initiated by a  site inspection.  The
 inspection served to familiarize the  crew with  the site, to locate special
 features,  and  to assist in the development of appropriate sampling methods.
 This undertaking identified  two types of wastes:  a tar-like waste and a
 granular waste that gave  off fumes and white vapors when it came in contact
 with water. The  granular  waste caused eye irritation and hindered breathing.
 Based  on this  site  scoping,  the potential for volatile air emissions during
 site mitigation  was  deemed to be high and further site characterization
 activities were  initiated.

 5.1.4  Overview of  Fieldwork  for Site Characterization

     An undisturbed  emission  survey  (see 5.1.5) was completed to assess the
 atmospheric impacts  of volatilized compounds under prevailing site conditions.
The emission survey  used  indirect real-time instrument measurement techniques.
Knowledge of the type of waste indicated that total  hydrocarbons and benzene,
as representative of aromatic species, were good indicator compounds for
volatile emissions  from the petroleum waste.  It was also possible that sulfur
                                      175

-------
dioxide (S02) could be an air contaminant from the waste.  Thus, total
hydrocarbons, benzene, and S02 were selected as the indicator compounds.
Other compounds may have been equally well suited for use as  indicators,  but
the compounds selected proved to be adequate.  Testing for these compounds
showed low emissions  (less than three times background levels) of total
hydrocarbons, benzene, and S0r   This result was obtained by comparing on-site
emission rate data to background or off-site data.  It was concluded that no
significant  atmospheric impacts existed on or off site for undisturbed site
conditions.

     To assess the potential impacts of air emissions of the combined site
during remediation and to provide adequate monitoring for on-site personnel
and nearby residences, a monitoring program was conducted that included:
constant meteorological monitoring at two stations; monitoring for emissions
from waste disturbance activities at the property boundary between the waste
and the nearby residences; and surveying of corehole borings.

     Two meteorological towers (see Figure 22) were used to collect continuous
data.  A 33-foot tower equipped with meteorological instrumentation collected
the primary  meteorological site data during the field activity.  In addition,
a 10-foot tower provided micro-meteorological wind speed and wind direction
data.  The towers were positioned upwind and downwind of the landfill and
lagoon.  Meteorological data were used in receptor modeling, in conjunction
with the measured emission rate data, for planning remedial  options.

     Additional waste characterization efforts were performed.  Five core
holes were drilled through the waste and into the soils below the waste to
permit sampling of the waste and soils and examination of waste and soil
stratification and physical properties.   They also provided a means for
measuring volatile species emissions as  a function of depth in the waste.
Wastes were  found to be only 5 or 6 feet deep over the §0,000 square-foot pit
area, for a total of approximately 11,100 cubic yards of wastes in the
landfill  and ligoon.  Wastes were generally soft and semi-fluid in the lagoon
and hard or  soil/waste mixtures in the landfill.  The waste had a pH below -
2.0,  contained varying levels of trace metals (including high concentrations
                                      176

-------
of lead  in some  samples),  and  had a very high percentage of organic material.
Soils below the  wastes were  predominantly alluvial gravels mixed with  sand  and
silt.  The soils rapidly buffered acidic waste leachate.  Trace metals were
found in varying concentrations and no apparent trend existed with depth.   The
underlying soils had been  impacted by low levels of hydrocarbons originating
in the waste pit.

     In  addition to the air-related work described above, extensive work was
undertaken to assess the impact of the site on local ground water.  Also,
eight small areas suspected  of being former waste disposal areas were
investigated.

     The site inspection data  were reviewed and a subsequent site
investigation plan was developed.  This Included an undisturbed emission
survey,  air monitoring on-site and at the fenceline during drilling and
sampling of wastes in the  landfill and lagoon, and an emissions survey under
disturbed conditions to estimate the potential for emissions during possible
future disturbance of the  wastes during an excavation activity.

S.I.5  Undisturbed Emissions Survey

     Both screening and in-depth measurements were performed.

Screening Measurements--
     A survey of the undisturbed surface emissions was conducted.   First, the
main waste pit disposal area was surveyed and a map was prepared with a grid
system overlying the waste area to provide location reference.  An emissions
survey was performed which consisted of a real-time instrument survey (an
indirect measurement screening technique) for indicator species at randomly
selected grid points mapped over the landfill  and lagoon.  This survey
provided general data on the level of gas concentrations from the landfill and
lagoon,  identified potential areas of higher concentrations, documented
background conditions (i.e., gas species concentration background levels), and
provided input into the safety program,  ensuring adequate worker/operator
protection.  This screening technique was selected because it was a quick and
                                      177

-------
inexpensive way to survey the site for areas of high air emissions  potential.
These data could then be used to design the in-depth measurement approach.

     During the emission screening, portable, ret!-time monitors were used to
determine sulfur dioxidi (S02), total hydrocarbon compounds (THC),  and
aromatic compounds (benzene).  THC measurements were made with an organic
vapor analyzer (OVA).  S02 measurements were made with  an electrochemical cell
instrument.  The HNU analyzer was used to detect aromatic species reported as
benzene.  S02,  THC,  and benzene,  as well  as surface  and air temperature, were
measured at 27 grid node points under quiescent conditions.  The portable
analyzers provided rapid feedback, but could not differentiate between various
hydrocarbon or sulfur species.

     For.this site, benzene, S02,. and THC were  used  as  indicators of air
emissions.  Benzene is a carcinogenic contaminant representative of aromatic
compounds.  Sulfur dioxide was a possible inorganic  air contaminant on site.
Total hydrocarbons were monitored for an indication  of total organic air
emissions.

In-Depth Measurements--
     After completion of the screening measurements, in-depth measurements
were conducted te quantitite th@ gas emissions  from  the undisturbed site for
risk assessment purposes and to aid in siting drilling locations.  The need
for this type of sampling was determined before any  screening measurements
were performed, based on the waste type and the proximity of receptors (i.e.,
the suspected large volume of highly volatile wastes was considered likely to
cause significant air impacts during any remedial waste removal or disturbance
activities).  Direct emission measurements were performed using an emission
isolation flux chamber.  This in-depth technique was selected because it is
ideally suited to obtain emission rates from homogenous area sources.  The
flux chamber is relatively easy to use and multiple  measurements (i.e., 8 to
10) can be obtained in one day.  Gas samples collected from the flux chamber
outlet line were analyzed using real-time analyzers.  Samples also were
collected in gas canisters for detailed hydrocarbon  speciation in an off-site
laboratory.
                                      178

-------
     The  site was  divided  into zones of high and low emissions potential  based
on plotted results from  the  real-time  instrument survey.  Locations for  in-
depth measurements were  randomly selected from grid cells in these different
zones.  Based on the results of the real-time instrument survey., nine flux
chamber measurements were  performed to assess the undisturbed emission from
the main  landfill/lagoon.  The flux chamber was constructed and operated  as
described in Section 4.  Measurements at a given grid point were typically
made over a 40-minute time period.

     Chemical measurements performed on the air leaving the emission chamber
included:

     t    Continuous determination of S02 (InterScan);
     •    Continuous determination of THC (OVA);
     •    Continuous determination of benzene (HNU); and
     «    Grab sampling  for organic speciation (Photovac 1010).

Undisturbed Emission Survey Results--
     A total of 27 grid  nodes were sampled, including sampling at five
background locations (upwind of each block), five duplicate sample points, and
sampling  at one control  point location at three different times of the day
(morning, noon, and afternoon).  The control point was  one of the grid points
that wis  regularly sampled to establish an estimate of  the temporal
variability in emissions at the site.

     Very low gas  concentrations were observed over the exposed waste.   Most
of the screening measurements showed background levels  at the locations
sampled for S02,  THC,  and benzene  as  shown  in  Table  19.   Moderately  low  levels
of undisturbed emissions were observed over exposed waste in the lagoon and
the landfill.   Table 20 provides undisturbed site emissions data.
                                      179

-------
       TABLE 19.   SUMMARY OF SCREENING MEASUREMENTS OF UNDISTURBED WASTE
                                        Range of Values
 Number of          SO* (pptn)           THC fppm)              Benzene (ppm)
                         Average     Peak   Average        Peak
    41            OoOOS   0.005      2-4a     2-4fi       0.01-0.80b  0.01-0.70b
a  Differences probably due to instrument drift.
b  Background levels only detected at  30  of 41  points.
                                     180

-------
      TABLE 20.   CASE STUDY 1: SUMMARY OF UNDISTURBED SITE  EMISSIONS DATA


 Lagoon                   S02                  THC*                 Benzeneb
Location
(Surface)            (mg/m2, min"1)        (ug/nr,  min"1)         (ug/m2, min'1)
            »

   #1                     0.14                 1.8                   4.7
   n                     0=14               120                    470
   #3                     0.14                 7.3                  43
   14                     0.14                44                      1.8
   IS                     5.6                   7.3                  11	
8 As determined by portable FID (OVA)

b As determined by portable PID (HNu),
Average                   1.2                  36                     98
Landfill
Location
#1
n
Average
0.14
UL.
0.77
7.3
££_
18
3.6
3^6
3.6
                                      181

-------
     The results of the screening and in-depth emissions testing  showed that:

     •    The emissions were highest in the landfill and lagoon where wastes
          were exposed (especially where natural disturbances  occurred; i.e.,
          cracking of surface, waste seeps, etc.);

     •    Control point sampling at various times of the day (same  location)
          indicated a large temporal variation in emissions due primarily to
          solar surface heating;

     f    Areas surrounding the combined site or in overburden on top  of the
          waste material showed background levels of emission; and

     s    Volatile emission rates from the combined site (landfill  and lagoon)
          were low for S02 and benzene  under undisturbed conditions.   For
          steady-state conditions:  S02 amission  rates  ranged from background
          to 5.6 ug/m2 minute"1; THC emission rates ranged from background to
          120 ug/m2s  minute'1; and benzene from background to 470  ug/m2,
          minute'1.   Emissions did show  a high dependence on diurnal
          temperature fluctuations with more emissions observed during the
          hottest periods, as expected.  The field photovac analytical
          capability provided limited hydrocarbon speciation data that helped
          direct more detailed hydrocarbon sampling and analysis.   The
          photovac data were not used to determine emission estimates.

5ol.6  Disturbed Emissions Survey

     Both screening and in-depth measurements were performed.

Screening Measurements--
     Samples of soil/waste were collected during drilling at each sampling
point as part of the screening survey.   Screening for volatile organic
compounds (VOCs) in these samples was performed at the field site with a
Photovac 1010 portable photoionization  gas chromatograph (GC).  This GC has
part-per-billion (ppb) level sensitivity for milliliter volumes of  air.  Soil
                                      182

-------
and waste  samples  were collected  (2  to  3  grams)  and  stored  In  40  ml  VOA vials
with Teflon\  SEPA.   The vials were equilibrated  In a 30'C water bath for 30
minutes prior to the head  space analysis  for VOCs.   This sampling procedure Is
considered  a  direct  measurement screening-technique.
            »
     During disturbance activities,  fenceline monitoring for S02  and  benzene
was conducted using  portable real-time  instruments.   Also meteorological
conditions  were monitored  during  all work disturbing  the site.  Action  levels
were established to  require cessation of  site activities if exceedances were
noted  (none occurred)  to protect  the local community.

     Downwind and  border monitoring  consisted of three activities:

     •    S02 and THC were monitored immediately downwind of the disturbance
          activity;

     •    S02 and THC were monitored at a mobile unit approximately 40  feet
          downwind of the  initial disturbance activities; and

     •    S02 and benzene were monitored at a mobile unit at the downwind
          fenceline  between the drilling  operation and the nearby residences.

Standard monitoring  procedures (see  Volume IV of this series of manuals) were
used in operating the  border and  downwind stations.  Air analyzers were
operated according to  written quality control protocols and continuous  data
printouts were collected using strip chart recorders.  The monitoring stations
were positioned each day based on wind direction data from the meteorological
stations.

In-Depth Measurements--
     Downhole emission measurements  at various depths in the waste were
conducted during drilling activities to determine an emissions "profile" in
the waste.  These data were used to characterize the waste properties and to
predict potential  gas emission from the wastes if they were excavated.
                                      183

-------
     Downhole emissions measurements were performed using the direct  emissions
measurement technique (i.e. downhole flux chamber).  This technique  is
considered an in-depth measurement technique and was applicable for the
landfill and the lagoon.  The plexiglas chamber had an exposed surface area of
0.00318 m2. 'The chamber input and output lines were 40 feet long,
facilitating flux measurement to 30 feet below land surface.

     Five locations for drilling were selected as part of the solid waste
investigation.  They are shown in Figure 23.  They were representative of
waste bodies and were equally spaced across the waste areas.  Hydrocarbon
samples were collected in 2.8-liter stainless steel canisters.  After
collection, the canisters were shipped to an off-site laboratory for analysis.

     A total of 18 downhole emission measurements were performed and realtime
data for S02,  THC,  and benzene were  collected at  each  point  using  the real-
time analyzers.   Canister samples were collected at six locations (at various
depths within the five core holes) for speciation analyses.   Canister samples
were not collected at all  sampling locations in an effort to conserve project
resources.   The indicator compounds  were used to represent air emissions
potential  in the absence of the canister samples.
                                     184

-------
RADIAN
   §
   I
   •Q
   Q
   €
   14.

                                                      CONCaETE
                                                      PAD
                                                                     N
                                                                          'CO
               Figure 23.  Location of waste  soil coreholes.
                                      185

-------
Disturbed Emission Survey Results--
     The results of air monitoring conducted during drilling  are  summarized  in
Table 21.  The results of border and immediate downwind monitoring  for
disturbed site activities indicated no significant impact  from  fugitive
emissions downwind from the site (northeast) boring activities.   Even though
the disturbed waste had a high emissions potential, small  amounts of waste
were exposed in the drilling operations following the conservative  operating
procedure.  The border station was positioned on the west  border to assess
neighborhood safety.  This location also provided valuable onsite safety
information for site operations personnel.

     The results of the downhole emissions survey included peak and steady-
state emissions for S02 and  THC at  the  given depth.   Benzene data were not
collected due to the lack of an extractive pump in that analyzer that could
pull the sample through the long sampling lines.   The steady-state values are
more representative of the level of emissions expected involving a disturbance
of the waste (i.e., removal).  Data results are presented  in Table 22 to
illustrate the S02 and  THC emissions  observed as  a function of the type  of
waste/soil.  This format illustrates the relative emissions characteristics of
S02 and THC per depth  in  the cores  as well  as the emission  tendencies  of the
waste/soil.  A comparison of peak to steady-state emission values is used to
identify emission sources in the waste  pit (i.e., limit of vertical
contamination).

     Volatile emission rates from the site under disturbed conditions were
higher than the undisturbed site and demonstrated potential for volatile
emissions during waste disturbance  activities.  The S02 emission rates under
disturbed conditions ranged from background (1.6 EE4 ug/m2, minute"1) to 1.1.
EE6 ug/m2,  minute"1, and the THC emission rates ranged from 3.8  EE3  ug/m2,
minute"1  to 3.8 EE6 ug/m2, minute"1.  The  results  of the hydrocarbon  speciation
analyses indicated the following hydrocarbon composition:
                                      186

-------
                 TABLE 21.  DOWNWIND/BORDER MONITORING RESULTS
Location
»
West Border
(Fence)
Core S-l
Downwind
40 m
Core S-2
West Border
(Fence)
Core S-2
West Border
(Fence)
Core S-3
West Border
(Fence)
Core N-l
West Border
(Fence)
Core N-2
Date
08/02/83
08/02/83
08/02/83
08/03/83
08/03/83
08/03/83
08/05/83
08/05/83
08/08/83
08/08/83
08/09/83
08/09/83
Species
Benzene*
S02b
Benzene
S02
Benzene
S02
Benzene
S02
Benzene
S02
High Leveld
(ppmv)
0.12C
Background
Background
Background
Background
Background
0.35
0.12
0.12C
Background
0.16C
Background
Duration
Elevated
Background


<5 minutes


a  Benzene instrument (HNU) background typically 0.1 ppmv.
b  S02  instrument  (InterScan Analyzer)  background  typically  0.05  ppmv.
c  Reading attributed to instrument drift.
d  Background refers to instrument reading  in "clean" air.
                                      187

-------
                  TABLE 22.  SUMMARY OF DOWNHOLE EMISSIONS DATA
SO. (ua/m2 min"1)
Core
S-l
Average
S-2
Average
S-3
Average
N-l
Average
N-2
Average
Depth
(ft)
5
7-1/2
15
2-1/2
4-1/2
15
20a
20
5
10
15
20
30
5
10
15
25
10
20
Peak
<1.9E4
>1.1E4
2.2E3
1.1E4
6.0E3
2.5E3
>1.1E4
4.0E3
1.6E2
4.6E2
7.6E2
3.0E2
6.0E2
LIE3
1.7E3
9.7E2
LIE3
1.6E3
6.7E2
Steady"
State
7.2E3
3.8E3
<8.3E2
3.9E3
5.4E3
1.9E3
1.7E3
>1.1E4
2.9E3
4.6E3
1.6E2
4.6E2
4.6E2
1.6E2
1.6E2
2.3E2
1.6E2
1.7E3
7.6E2
7.6E2
8.5E2
1.3E3
6.8E2
9.9E2
THC (ua/m2min~M
Peak
1.6E2
1.3E4
>3.8E4
1.9E3
8.3E3
>3.8E4
>3.8E4
>3.8E4
4.6E2
>3.8E4
1.6E4
>3.8E4
>3.8E4
>3.8E4
LIE4
2.3E4
>3.8E4
>3.8E4
>3.8E4
Steadyb
State
3.8E1
6.8E3
6.8E1
2.3E3
3.831
3.8E1
9.5E1
>3.8E4
3.8E2
9.3E5
7.6E1
>3.8E4
7.6E3
2.3E2
6.0E2
9.3E5
3.8E2
8.8E2
2.0E2
2.9E2
4.4E2
5.6E3
2.6E2
2.9E2
'Range of emissions,  two measurements were conducted.
 Steady-state values  were averaged by core by operable unit to determine BEEs

             E - Exponential Notation  (7.2E3 = 7.2 x 103 = 7200).
                                      188

-------
          Hydrocarbon Class            Average %          _Ranqe %a	
          Alkanes                        79.0               68-87
          Alkenes                        15.0             0.87 - 21
          Aromatics                       4.2             2.4  - 8.1
          Oxygenates                      0.62            0.15-1.5
          Halogenated                     1.2             0.091-3.3
          Sulfonated                      ND                  ND
          Unidentified                    2.0             0.42 - 6.6
a5 cores, 6 canister samples


These data show that most of the air emissions were alkane species and of no
significant concern regarding toxicity.  The aromatic fraction was, as
expected, around 5 percent, and was composed of numerous compounds.

5.1.7  Development of BEEs

     Calculation of baseline emission estimates (BEEs) can be developed from
either ambient concentration data (indirect techniques) or from emission rate
measurement data (direct techniques).  BEEs can be calculated for each
contaminant species detected or for a group of contaminant species.  BEEs
obtained from direct measurement techniques which provide rate data (i.e.,
mass per unit time per given surface area) are preferable.  The BEE is
normalized for the area of the source and has units of mass of contaminant or
group of contaminants per time.

     BEEs can be calculated from individual emission sources and summed for
sites containing multiple emission sources (operable units), such as a lagoon
and a landfill with each source characterized by different air emission rates
and contaminant species.

     The calculation of an undisturbed emission estimate for this site
included the following considerations.  The site consisted of a waste area
containing a landfill in the northern portion and a lagoon in the southern
portion, separated by a berm.  Each portion of the site was evaluated
separately.  Undisturbed emission factors for each of the two operable units
were calculated separately and then averaged to provide an overall site

                                      189

-------
emission estimate.  The undisturbed emission  rates  for  the  lagoon and the
landfill, and the combined  site emission estimate calculations  are presented
below.

     The undisturbed BEE for the lagoon (average) was calculated  for S02  and
THC.  The surface emission  isolation flux chamber survey  results  from five
single measurements were averaged by species  and multiplied  by  the lagoon
surface area to determine the unit BEE.

                            2            2              3
Lagoon BEEcn  = (1.2ug/m-min)(6860 m) =  8.2  x  10  ug/min of SO,
          MJ2
          THC
    Lagoon BEETHC »  (36  ug/m2-min)(6860 m2) = 2.5 x 105 ug/min of THC
Similarly, the BEE for the landfill operable unit  (average) was calculated  for
S02 and THC.   The emission isolation flux chamber survey results were averaged
for each species and multiplied by the surface area of the landfill to
determine the unit's BEE.

   Landfill BEE$()  - (0.77 ug/m2-min)(7240 m2)  =   5.6  x  103 ug/min  of  SO
   Landfill BEETHC -  (18 ug/m2-min)(7240 m2)  = 1.3 x 10s  ug/min  of THC

     The overall site BEE (for S02 and THC)  can be obtained by summing the
respective unit BEEs by species.

     Calculation of emission estimates for disturbed  site conditions  can  be
performed from either air monitoring data (concentration measurements) during
waste disturbances or from direct emission rate measurement data.
Concentration values can be expressed as a concentration (ppm-v) for  each
species, as a ratio of the species concentration to the total  concentration
from all species, or as a percentage value for the species of  interest over
the total concentration contributed from all other species.  The disturbed
site emission flux data has the units of mass per time per unit area.
                                      190

-------
       The emission estimate is calculated by multiplying the average measured
  emission rate by the total surface area of disturbed material.  This results
  in a single value of mass per unit time which provides a relative estimate of
  the rate of air emissions from the source.  Emission estimates can be
  calculated for a site with either individual sources or with multiple sources.
,  These operable units often are investigated and remediated independently.
  Emission estimates for a combined site can be calculated by using the highest
  disturbed emission estimates for each source and summing the emission per time
  for each source.

       Summaries of the average disturbed emissions flux for both units of the
  Case Study 1 site are presented below:

            Landfill
                 S02  -  9.2 x  102  (ug/m2-min)
                 THC  -  1.7 x 103  (ug/m2-min)

            Lagoon
                 S02  -  2.9 x  103  (ug/m2-min)
                 THC  -  6.4 x 103 (ug/m2-min)

       These data can be used in conjunction with estimates of exposed disturbed
  waste to predict air impacts from various waste disturbance and treatment
  techniques.

       Example

       Excavation of the landfill  would typically expose 50 m2  of waste  at  a
       time.   The estimate of THC air emissions from the site activity would be:

               THC (landfill) = 1.7 x 103  (ug/m2, minute'1)  x 50m2
                              - 8.5 x 10s  (ug/min)

  Depth-specific information could  be used to provide area and depth-specific
  emission estimates as needed.
                                        191

-------
5.1.8  Summary

     The Case Study 1  investigation was generally a thorough,  well  documented
study that fully addressed the air pathway for volatile contaminants.   The
study followed closely the steps outlined in this manual's  protocol  and all
objectives were met.   Furthermore, site personnel report that  the  knowledge of
the potential for emissions, ultimately resulted in a safer and more cost-
effective remediation  of the site.  BEEs were used in risk  assessment  and in
designing removal plans.  Air emission control techniques were selected based
on the BEEs.

5.2  CASE STUDY 2:  BRUIN LAGOON

     Case Study 2 is a disposal lagoon that received various wastes  from a
mineral oil refinery.  This site was under remediation in 1984 when  subsurface
gases were unexpectedly released during cleanup.  Work was  halted and  the
remedial design was reassessed.  A second Remedial  Investigation/Feasibility
Study (RI/FS) was then performed.  This case study focuses  on the APA
conducted during this second RI/FS (see Table 23).

5.2.1  Site History

     Bruin Lagoon is located about 45 miles north of Pittsburgh, in  Bruin
Borough of Butler County, Pennsylvania.  The 4-acre site is situated along the
western bank of Bear Creek's South Branch, approximately 7 miles upstream of
the creek's confluence with the Allegheny River.  Part of the site lies  within
the creek's 100-year floodplain.  To the west, the site is  bordered  by  private
homes and State Route 268.  Bruin Borough's main residential and commercial
areas are within five blocks of the site and more than 30 residences are
within 500 feet of the lagoon.   To the south is an abandoned refinery,  which
is the source of the wastes deposited in the lagoon.   Also, adjacent to the
site are two ponds and a small  stream that drain into Bear Creek.
                                      192

-------
         TABLE 23.  APA ACTIVITIES CONDUCTED AT THE CASE STUDY #2 SITE


APA Objectives

Assess the extent and composition of  subsurface gas pockets.  Monitor the
ambient air for health and safety reasons.


Scoping

Initially, the lagoon was determined  to contain wastes that had minimal
potential for volatile or particulate matter emissions (first RI/FS).  The
second RI/FS did  assume that volatile emissions from subsurface gas pockets
were likely.


Screening Measurements

A variety of portable, real-time analyzers and detector tubes were used to
monitor the ambient air during drilling activities.  Monitoring took place at
the site perimeter and in the ambient breathing zone near the drill rig.  Soil
samples were collected and the emissions from the samples were scanned.  The
screening data were used to design the in-depth measurement strategy.


In-Depth Measurements

The in-depth emission measurements involved collecting and analyzing grab
samples of gas from boring/wells whenever the ambient breathing zone
monitoring showed elevated concentrations significantly over background
levels. However, these data were not used to develop undisturbed and disturbed
site BEEs.
Mitigation

The disturbed waste emissions data were used in the development of the
remedial action plan.  Although disturbed emission rates were not calculated,
knowledge of areas considered to be "hot spots" were used to conduct site
operations in a way that prevented major releases of air toxics to the air.
The remedial alternative included gas monitoring during on-site stabilization
and neutralization of the unstabilized sludge and collecting, venting, and
treating as necessary.
                                      193

-------
     Currently in remediation, Bruin Lagoon  is an unlined  earthen  diked lagoon
that has been partially covered with 9300 cubic yards of stabilized soil/
sludge mixture treated during the first attempt at remediation.  Beneath this
material is approximately 17,200 cubic yards of unstabilized  sludge/tar,  with
up-welling of the waste in a number of areas.  The sludge/tar contains
sulfuric acid and heavy metals, along with other contaminants.  The lagoon
area of the site is generally level and lacks vegetation.  A  cross-sectional
view of the site is shown in Figure 24.

     Bruin Oil Company, producer of white (mineral) oil, began disposing  of
its wastes at the lagoon in the 1930s.  This continued for more than 40 years.
Materials discarded there included:

     •    Residues scraped from crude oil  storage tanks;
     •    Used bauxite, charcoal  filtering agents, and bone powder;
     •    Oils not meeting specification;
     t    Coal fines;
     •    Lime;
     •    Spent alkali; and
     •    Boiler house coal  and ashes.

     The lagoon attracted national  attention in 1968 when about 3,000 gallons
of acidic sludge spilled into the South Branch of Bear Creek through a breach
in the dike.  In the Allegheny River,  roughly 4 million fish died and many
downstream communities temporarily lost their water supplies.  The  spill was
addressed,  but the remedial  investigation  of the site didn't begin  until  1981.

     The abandoned refinery and the lagoon were owned by AH & RS Coal Company,
which underwent bankruptcy proceeding in 1986.
                                      194

-------
                                                                Existing
                                                                Lagoon
Clean
SoKF*
                                                    <-v'
                                        ' < v' '• *'?  V
                                             •          -'
Top ol Bediock
 Walat Table
   Aquilur
                                                                                                                    0
                                                                                                                    5
   LEGEND:
          Indicalus Flow OiiocUon
   NOT TO SCALE
       Figure 24.   Generalized  flow  regime  of perched zone  and  bedrock aquifer.

-------
5.2.2  Objectives

     The objectives of the site work, from an air pathway perspective, were
two-fold:  to monitor the ambient air for health and safety reasons;  and to
assess the extent and composition of subsurface gas pockets.  To meet the
latter objective, sampling and analysis of vapors contained in the  shallow
wells were performed to:  identify the composition and extent of gases trapped
under the site; determine their regeneration rates; and assess the  potential
for their release into the atmosphere during remedial construction  work.

5.2.3  Scoping

     The existing data were reviewed to provide a working knowledge of the
site history, conditions, and environmental  setting.   Based on this review, no
specific potential emission characterization was called for in the  first
RI/FS.  After unexpected emissions were encountered when drilling through the
bottom of the lagoon during remediation, the second RI/FS did address the air
pathway to a greater extent.

5.2.4  Overview of Fieldwork For Site Characterization

     EPA contractors began what would become the first RI/FS at Bruin Lagoon
in July 1981.  Air monitoring during this remedial  investigation failed to
find detectable levels of organics,  sulfur dioxide (S02),  hydrogen  sulfide
(H2S),  hydrogen chloride (HC1),  or hydrogen  cyanide  (HCN)  in  ambient air at
the site.  Although one well boring showed organic vapors during drilling
operations, the levels were not detectable at the ambient breathing zone.
Significantly, no borings through the open lagoon were performed in this
initial effort.

     With the RI/FS completion in January 1982,  EPA and the Pennsylvania
Department of Environmental  Resources (PADER) selected a remedial alternative
that included sludge stabilization,  dike reinforcement, debris removal, and
construction of a multi-layer cap to cover the lagoon.   Design kicked off  in
September 1982 and cleanup actually started in August 1983.
                                      196

-------
     The project  proceeded  until May 4,  1984, when hazardous gas and acid mist
escaped from an unanticipated crust, thought to be the bottom of the lagoon,
that was broken during mitigation.  These organic vapors and sulfur dioxide or
hydrogen sulfide  reached the ambient breathing zone but were not detectable at
the site perimeter.  Gas sampling was performed for worker and public
protection.  Gas  samples from beneath the crust revealed high concentrations
of carbon dioxide, hydrogen sulfide, and sulfuric acid mist.  Consequently,
EPA suspended cleanup activities and immediately launched in emergency
response, which included some removal, covering the lagoon with stabilized
sludge, installing 13 shallow gas monitoring wells, and collecting and
analyzing additional sludge and soil samples.

     The site remained in the emergency mode until September 1984.  The second
RI/FS was initiated the following January.  Air monitoring conducted
throughout this RI included:

     •    Health  and safety;
     •    Site perimeter;
     t    Ambient breathing zone;
     •    Downhole concentration sampling and analysis; and
     •    Sample  screening.

     Conclusions drawn from a review of all of these activities (the two RI/FS
endeavors and the emergency action) included identification of a "hot spot"  in
the unstabilized portion of the lagoon that contained carbon dioxide, hydrogen
sulfide, sulfur dioxide and methane at levels deserving attention during
remediation.  The RI/FS concluded, however, that subsurface gases were not
present throughout the site.  To address the "hot spot" the remedial
alternative selected in September 1986 included gas monitoring,  venting, and
treating during excavation, followed by post-closure monitoring.
                                      197

-------
5.2.5  Undisturbed  Emissions  Survey

     Based on the available documentation, no screening  or in-depth mea-
surements were made to  assess the undisturbed emission at  the  site.  Since the
site contains heavy metals and  is unvegetated, an evaluation of the entrained
particulate matter  from the site would have been advisable, and some screening
measurements for particulates and inorganic gases may have been warranted.
Some air monitoring was performed immediately prior to site disturbances.
These limited data do address undisturbed emissions.

5.2.6  Disturbed Emissions Survey

     Both screening (air monitoring and sample headspace)  and  in-depth  (soil
vapor well) measurement techniques were used to assess the emissions  during
site disturbances such  as drilling.  The following discussions  are  largely
taken from the second RI/FS prepared for the Bruin Lagoon  site  (65).

Screening Measurements--
     The following air monitoring equipment was available on site during all
drilling activities:

     •    HMD photoionization detector (PID)  with 11.7 and 10.2 eV  probes;
     •    OVA (organic vapor analyzer) flame ionization detector (FID);
     •    H2S portable gas monitor;
     •    S02 portable gas monitor;
     •    H2S monitor  alarms;
     •    Explosimeter/oxygen monitor; and
     •    Detector tubes S02,  H2S, H2S04,  02, C02, natural gas.

     Based on gases detected in past site work,  portable direct reading  real-
time instrumentation was primarily used for gas characterization and  health
and safety purposes.  Detector tubes were used for screening of possible
instrumentation interferences, confirmation of direct reading concentrations,
and analysis of gases  not detected on available instrumentation.
                                     198

-------
     The direct  reading  monitoring  instruments determined to be most  effective
for monitoring drilling  operations  were the HNU PID  (11.7 eV), H2S portable
gas monitor, H2S monitor alarms, and S02 portable  gas monitor.   The HNU PID
was selected over  the  OVA  FID  due to the sensitivity of this instrument  to
hydrogen sulfide gas.  The HNU PID  could detect both hydrogen  sulfide gas and
organic vapors.  This  selection was made because of past historical data
demonstrating possible H2S gas  release.

     Periodic monitoring of ambient air at the site perimeter  was  routinely
performed during drilling  operations.  Also, if ambient breathing  zone
concentrations indicated possible gas  releases, perimeter monitoring  was
immediately initiated.   Seventeen monitoring locations were established  along
the site fence line  at intervals of approximately 150 feet and marked with
stakes.  Figure  25 shows the perimeter monitoring locations.   Perimeter
monitoring was conducted with  the H2S and S02  portable gas monitors and the
HNU PID.

     During all  drilling operations, the H2S and S02  portable gas  monitor, H2S
monitor alarms,  explosimeter/oxygen monitors, and HNU PID were used for
characterization of  the  ambient breathing zone.  Background levels  were
determined prior to  starting the drilling.

     Portable instrument readings provided continuous, real-time monitoring of
each split spoon and drilling  depth to determine at what depth, if any, gas
releases occurred.   Each split  spoon and core sample was scanned with Figure
25 all  direct reading  instrumentation  immediately after collection.   Samples
showing positive readings were  usually selected for chemical analysis.

In-Depth Measurements--
     If ambient breathing zone  monitoring showed elevated concentrations
significantly above background  levels, a grab sample of the gas present in the
boring/well  was collected for  analysis.  These grab samples were analyzed to
characterize the emitted gases.  Grab  samples were collected from  a point
approximately 3 feet below the  ground  surface by inserting tubing  into the
                                      199

-------
                                    Btluntu* t Ohu HR
                                                                                  LEGEND:
                                                                                  _4.	Sueam
                                                                                  —*r—  Fonce
                                                                                  	Piopoi ty Lino
                                                                                     •   Bodiock WuH
                                                                                     A   Shallow Wtil
                                                                                     ©   SodBonog
                                                                                     •   Pufhnolof An
                                                                                          Momloiing Location
                                                                                  NOTE: AW4 was not in&tuNud
                                                                                      050   100
                                                                                       Seal* In F»«i
:o
;5
:z
Figure 25.   Monitor  well  and soil boring locations at  the Bruin  Lagoon  site.

-------
 boring/well  and pumping the gas into an  air bag  collector.   The bag sample was
 then  sealed  and analyzed on-site using available instrumentation and detector
 tubes.

 Disturbed  Emission Survey Results--
      The following results are taken from the  second  RI/FS  prepared for the
 Bruin Lagoon site (65).

      Air monitoring conducted  during the 1981  remedial  investigation of Bruin
 Lagoon  revealed no detectable  levels of  organics,  S02, H2S,  HC1, or HCN in the
 ambient air  at  the site.   Organic vapors were  detected  within  one  well  boring
 during  drilling operations;  however,  no  detectable levels were found in the
 ambient breathing zone  at this location.   It should be  noted that  no borings
 were  constructed through  the open lagoon during  the initial RI  and,  as  a
 result, the  gases trapped below the  crust were not encountered.

      Background air monitoring performed during  field work  in  June  1984 showed
 no detectable levels of H2S  or methane.  Air samples collected from the soil
 boring  indicated the presence  of H2S, C02, methane, and  aromatic hydrocarbons.
 S02 was not detected  in the  downhole samples.  H2S was present  in the soil  gas
 on the  average  at about  300  to 400 ppm;  initial  concentrations were  greater by
 an order of  magnitude or  more.

      Low levels  of organic vapors, sulfur  dioxide  and hydrogen sulfide  were
 released into the ambient breathing  zone when  the  subsurface of the  site was
 disturbed by drilling operations.  S02 concentrations  were observed as high as
 50 ppmv but typically were found  at  0.5  to  18  ppmv in the breathing  zone
 during drilling  operations.  H2S concentrations were lower with high
 concentrations observed at 14  ppmv with  typical  concentrations of  1  to  10
 ppmv.   However,  concentrations of these gases were not detectable at the site
 perimeter.

     The analytical  results  for the subsurface gas samples collected from  the
 13 shallow wells  installed during the 1984 emergency action showed various
concentrations of volatile organics, S02, H2S04,  and methane in the wells.

                                      201

-------
Methane was observed in all the wells ranging from 2.6 to 4,400 ppm.   These
concentrations showed good correlation with the organic CH4 levels measured in
the field by the OVA.  Analysis of carbon tube samples collected  for  each well
showed no detectable volatile organic compounds.  Sulfur dioxide  was  found to
be present iri 11 of the 13 wells, with four wells having S02 concentrations
greater than 100 ppm.

     The results of the 1984 sampling of subsurface gases at the  Bruin Lagoon
site showed elevated levels of S02.   The presence of H2S04 mist was limited to
three wells (A-8, A-10, and A-13).  Additionally, reactions and gas releases
occurred during the installation of each of these wells.  Elevated levels of
S02 also were detected in well  A-2.   These wells  are all  located within 50
feet of one another, and, as a result, the data suggest that this area of the
site is a "hot spot" with respect to potentially harmful trapped  subsurface
gases.  Additional sampling one year later confirmed the presence of a "hot
spot" area located in the central part of the site.

5.2.7  Development of BEEs

     No baseline emission estimates or disturbed emission estimates were
generated.  It was noted, however, that remediation may result in the release
of pockets of hazardous gases trapped below a crust at the bottom of the
lagoon.  BEEs would allow performance of risk assessments for various release
and meteorological scenarios at receptor points of interest.

     Given the available data,  the best method for determining BEEs for this
site would be to take the existing ambient air monitoring data and back-
calculate an emission rate using an air dispersion model.  This is done by
setting up the model's run conditions to match those at the site  as closely as
possible, and then varying the source term to find an emission rate that
produces downwind concentrations equal to those actually measured.  Using the
respective air monitoring data sets,  this procedure could be used for both
undisturbed and disturbed conditions.  However, the accuracy of this procedure
is limited by the amount and representativeness of the available  air
monitoring data.
                                      202

-------
 5.2.8   Summary

     The  Case  Study  2  investigation  points  up  the  need  to  consider  potential
 emissions  from the disturbed  site  before  any remedial actions  are undertaken.
 Here, no  soil  borings  through the  open  lagoon  were performed during  the  first
 RI/FS.  The  initial  failure to do  so, or  to even consider  potential  air
 emissions  at this site,  resulted in  the need for a second  RI/FS and  associated
 schedule  delays and  extra  expenses.  The  study did not  closely conform to  the
 steps outlined in the  protocol  of  this  manual.  No undisturbed emission
 measurements were performed (beyond  some  background air monitoring)  and  no
 emission  rate  data were  collected.

     The  best  technique  for screening particulate matter (PM)  emissions  at
 this site  would have been  to  collect upwind/downwind samples on filters  using
 high-volume  sampling pumps (hi-vols).   The  total particulate matter  present in
 the air would  be determined by dividing the filter weight  gain by the volume
 of air  sampled.  Analysis  of  the filter catch  for selected metal species would
 assist  in  assessing  the  health impacts  from undisturbed emissions.   A less
 acceptable alternative would  have  been  to measure the ambient  particulate
 matter  loadings using  a  portable particulate matter analyzer.  The activities
 conducted, however,  provided  valuable data  that were very  useful in  the
 design, selection, and implementation of the remedial alternative.

 5.3  CASE  STUDY 3:   LOWRY  LANDFILL

     Case  Study 3 is an  active  municipal landfill that  formerly also accepted
 liquid  and solid industrial wastes and  domestic sewage  sludge.  This case
 study focuses  on APA activities conducted at the site (see Table 24).

5.3.1  Site History

     The Lowry  Landfill  is co-owned  by  the  City and County of  Denver,
Colorado.   It  opened for business as a  municipal  landfill   in 1965.   The  site
is located about 20 miles  southeast  of  Denver  and two miles east of  the  City
                                      203

-------
         TABLE 24.   APA ACTIVITIES CONDUCTED AT THE CASE STUDY #3 SITE
APA Objectives

A soil gis study was conducted to locate waste pits, determine the extent of
off-site subsurface gas migration, and determine the waste pit contribution to
such migration.  An air monitoring program was conducted to measure  ambient
air pollutants during installation of monitoring wells in the waste  pits.


Scoping

The landfill was determined to contain both hazardous and municipal  wastes.
The generation of off-gases from the wastes was considered to be a hgih
probability.


Screening Measurements

The ambient air was monitored upwind and downwind of the site during
monitoring well installation.  Samples were collected using a variety of
adsorption media.  Soil samples were collected and the emissions from the
samples were scanned.


In-Depth Measurements

The in-depth emission measurements involved collection of soil gas samples at
a large number of points using vapor monitoring wells for deep samplign and
ground probes for shallow sampling.   However, these data were not used to
develop undisturbed and disturbed site BEEs.


Mitigation

Migration plans have not yet been prepared.  The site is currently under study
to further understand the air emissions potential  from the site.   These study
activites include air monitoring at  nearby receptors of concern (e.g.,
community school).
                                      204

-------
 of Aurora,  in Arapahoe County.   The site  covers  approximately  480  acres.   The
 surrounding area was mostly undeveloped when  the landfill  was  established,  but
 is growing  rapidly today.   (Proximity of  the  closest  residence was not given
 in the  Remedial  Investigation  (RI)  report.
             *
      From 1965 until  the  advent  of  the Resource  Conservation and Recovery Act
 (RCRA)  in 1980,  the facility accepted municipal  refuse,  liquid and solid
 industrial  waste (some of which  was hazardous),  and domestic sewage sludge.
 The  landfill  handled these wastes by excavating  pits, filling  them three-
 quarters with liquids and then covering the waste with refuse  until  a  mound
 several feet above the land surface was created.  Landfills were dug
 repeatedly,  sometimes into old,  filled landfills.  The landfills at  the south
 end  of  the  facility were  covered with as  much as 30 to 60  feet of  refuse.

      In 1975,  Continental  Oil Company contracted with site owners  to set  up
 and  run an  oil  sludge disposal operation  in the  southeastern portion of the
 site.  This  operation and  the acceptance  of industrial waste stopped with RCRA
 in 1980.  At  that  time, the City and County of Denver hired a  private  firm to
 manage the  site  as a municipal waste facility only.  This contractor,  Waste
 Management,  Inc.,  formed  a subsidiary which opened a hazardous  waste disposal
 facility just  north of Lowry.  This facility was closed  in 1982.

      In the  early  1980s,  Lowry Landfill began to be closely scrutinized by a
 number of public agencies  due to odor problems and other concerns.   These
 first cursory  looks focused primarily on  the groundwater contamination
 pathway, studying  only shallow groundwater.  Not all landfills  were  located or
 confirmed.   Initial  investigations  disclosed that records of types and
 locations of waste  were incomplete  and inaccurate.  Also, no measures  had been
 taken to prevent leachate  or seepage  from these pits.

     Among the 60  pits  identified through aerial photographs,  it was estimated
that roughly 100 million gallons of liquid wastes were disposed on-site over
15 years.   Wastes  identified include: acid and alkaline sludges; caustics and
solids;  brines,  including  plating wastes and other water-based  sludges;
organics,  both natural and  synthetic, such as petroleum-based  oils,  grease,
and chlorinated  solvents and sludges; watersoluble oils; municipal sewage
                                      205

-------
sludge; low-level radioactive wastes; pesticide wastes; asbestos;  and  metallic
wastes.

     EPA did not become intimately involved at the site until about  1981 when
Lowry was first considered as a candidate for the NPL.  In 1984, Lowry was
placed on the NPL and formal investigations were initiated.

5.3.2  Objectives

     At the Lowry site, a soil gas study and an air quality investigation were
performed as part of the RI.  The soil gas study was conducted to  help locate
the waste pits, to determine the extent of any off-site subsurface migration,
and to determine the waste pit contribution to such migration.  The  air
quality investigation was conducted to measure contaminants in the ambient air
during installation of monitoring wells in the waste pits.

5.3.3  Scoping

     Existing data were collected and reviewed to provide a working  knowledge
of the site history, conditions, and environmental  setting.  A topographic map
of the site and its environs was developed from aerial photos.  Surveying was
performed to map sampling locations and determine the relative coordinates and
elevation of each location.

5.3.4  Overview of Fieldwork For Site Characterization

     The purpose of the first phase, of a two-phase remedial  investigation,
was to characterize the site geology and climate; identify the location and
contents of all landfills; characterize the extent of contamination, including
air and soil gas; and identify data gaps to be filled in during rememdial
investigation Phase II.

     Investigatory work into some of the landfills was thwarted by three piles
of vehicle tires, two piles of roughly 2 million tires each and one  pile of  up
to 8 million tires.  Findings disclosed that soil gas, air, groundwater,
surface water, and soil all were contaminated and that some migration  was
                                      206

-------
 occurring.   To detect air contamination,  contractors sampled upwind and
 downwind  of the site, and.50 feet downwind from the five waste pit wells
 shortly after installation  (to demonstrate a worst-case emissions scenario).
 The  results showed that air quality was degraded as it crossed the site from
 south  to  north (the direction of the prevailing wind).  Volatile organic
 compounds (VOCs)  were detected at 0.020 to 16 parts per billion (ppb)  higher
 in the downwind samples relative to the upwind samples.

     Separately,  samples for soil  gas emissions were taken  from the landfills,
 from gas  well  points and probes,  and from 10 gas sampling wells installed
 around the  site perimeter by the contracted  facility operator  in 1981.   The
 results disclosed 19 VOCs,  found in ranges of 37 to 160,000 ppb,  and verified
 that contaminated soil  gas  was migrating  off-site in the vicinity of one of
 the  perimeter wells.

     A complete meteorological  monitoring station has  been  operating at  Lowry
 since  April  21,  1985.   It measures  wind speed and direction, temperature,
 relative  humidity,  barometric pressure,  and  precipitation on a  10-meter  tower.
 Measurements  are  taken  by a Climatronics  Electronic  Weather Station  (EWS)
 connected to  a  Campbell  Scientific  CR21 data  logger.   The data  are  read
 periodically  into a  mainframe DEC-10 computer.

 5.3.5  Undisturbed  Emissions  Survey

     No screening or  in-depth measurements were made to  assess  the undisturbed
 emissions at the  site.   However, the  Phase I  RI  report (66) does  recommend
 that ambient air  and meteorological  monitoring  be  performed in  the planned
 Phase  II work to  collect  background  data.  Based  on  the  types of  waste present
 and the presence  of contaminated soil gas, screening measurements  (at a
minimum)  would have been warranted  for  this site  during  Phase I activities.

5.3.6  Disturbed  Emissions  Survey

     Both  screening (headspace sampling and upwind/downwind air monitoring)
and in-depth (soil vapor wells) measurement techniques were used  to  assess the
air pathway.
                                      207

-------
Screening Measurements--
     Split spoon samples collected during drilling activities were  retrieved
and opened, then the air space above the samples was scanned using  an  HNU
portable organic vapor analyzer.
             *
     The ambient air was monitored over a 12-day period during  installation of
monitoring wells in the waste pits.  The monitoring took place  in November and
December of 1985 during which there was some snow cover; therefore, the
results do not equal a worst-case scenario.  Samples were collected upwind and
downwind of the site and 50 feet downwind of the waste pit well installations.
A controlled release of waste pit well gas was permitted to help predict
ambient air impacts associated with remediation of the waste pits.  Samples
were collected by concentrating air on carbon molecular sieves  (CMS),
polyurethane foam, Tenax, and glass fiber filters.  The sampling methods are
listed in Table 25.

In-Depth Measurements--
     Soil gas samples were collected from a number of locations, including 10
existing soil vapor wells around the perimeter of the facility  and wells
installed at four locations in suspected waste pits.   The wells at the waste
pits were drilled to within 2 feet of the water table.   Ground  probes also
were driven into the waste pits at the same four locations to measure the gas
emanating from the waste pits and municipal refuse and reaching the near-
surface.  Three more ground probes were installed at areas without underlying
waste pits.

     Samples were collected from all locations using a pump to  transfer gas to
a tedlar bag within a rigid-wall container.  Sample gas was extracted at
roughly 1 L/min for 4 to 5 minutes.  The perimeter soil vapor wells required
purging (three well volumes) before sample collection;  the other sampling
points had free-flowing gas.

     Samples were analyzed for priority pollutants by GC/MS using EPA Method
624 at an off-site location.
                                      208

-------
                      TABLE  25.  SUMMARY OF AIR MONITORING AT LOWRY LANDFILL
Collection
Media
Tenax
Carbon
Molecular
Sieve
Glass Fiber
Filters
Polyurethane
Foam (PUF)
N/A
Equipment Number of
Description Samples
Gillian HFS 52
personal pumps
Gillian HFS 52
personal pumps
Sierra Accu-Vol 20
high-volume
samplers
GMW Model PS-1 26
high-volume
samplers
Climatronics Continued
Wind Mark III
Analytes
Highly Volatile
Organic Compounds
VOCs
TSPa, Metals
Sem1-volatiles and
Pest1c1des/PCBs
Met Data
Duration of
Sampling
(Hours)
8-12
8-12
8-12
8-12
N/A
Method of Analysis
4
GC/MS
(Method TO-1)
GC/MS
(Method TO-2)
EPA Reference
Methods
GC/MS (Method 625
and Method TO-4)
N/A
Total suspended particulate matter.

-------
Disturbed Emission Survey Results--
     A large data base was developed during this program and  is  summarized
here.

     The upwind/downwind sampling indicated that the site  is  a VOC  emissions
source.  Total VOCs on-site were 54 ppb higher than upwind values,  and
downwind concentrations were 25 ppb higher than upwind values.   Downwind
concentrations of acetone, carbon disulfide, and toluene were 3  to  10 times
the upwind values.  During the controlled release, these compounds  were found
at 8 to 100 times the upwind values.  Other compounds, such as 1,1,1-
trichloroethane, benzene, and TCE also were found at elevated (4 to 20 times)
levels downwind.  These emissions could be expected to be greatly higher
during non-winter weather conditions.

     The upwind/downwind sampling showed particulate matter emissions to be a
problem at the site.  The upwind values averaged 187 ug/m3  and the downwind
samples averaged 325 ug/m3 with a range of 112 to 643  ug/m3.   The average
downwind total solid particulates (TSP) exceeded the Primary TSP National
Ambient Air Quality Standard of 260 ug/m3.

     Nineteen hazardous volatile organic compounds were detected in soil gas
samples emanating from waste pit liquid.  These compounds were similar to
those found in the liquid samples.  Concentrations ranged from 460 to 291,000
ppb.   Nineteen volatile organic compounds were found in the refuse gas
samples,  in ranges of 37 to 160,000 ppb.  Compounds were nearly  identical to
those found in gas samples above waste pit liquids except that the
concentration of compounds above these liquids was two to five times greater
than  in the overlying refuse and six times greater than in refuse with no
underlying pits.  It is reasonable to conclude that the liquids and refuse are
contributing to gas contamination.
                                      210

-------
      The  results of perimeter well  gas  sampling  indicate that  subsurface
 contaminant  migration has occurred  at Well  GPM-3  and  possibly  at  GPM-7.
 Thirteen  hazardous  substances were  detected in gases  at  Well GMP-3,  near waste
 pits  and  refuse  disposal  areas.   Substances included  volatile  organics  at
 ranges  of 9-to  1,200 ppb.

 5.3.7  Development  of BEEs

      No baseline emission estimates for either the  undisturbed or disturbed
 wastes  were  generated,  though sufficient data exist to estimate a disturbed
 BEE.  Given  the  available data,  the best method for determining BEEs  for this
 site  would be to take the existing  ambient  air monitoring data and back-
 calculate an emission rate using an air dispersion  model.  This is done  by
 setting up the model's  run conditions to match those  at  the site  as closely  as
 possible,  and then  varying the source term  to find  an emission rate that
 produces  downwind concentrations equal  to those actually measured.  Using the
 respective air monitoring data sets, this procedure could be applied  to  both
 undisturbed  and  disturbed conditions.

 5.3.8  Summary

      The  Case Study 3 investigation did not closely conform to the protocol
 steps outlined in this  manual.   No  undisturbed emission  measurements  were
 performed  and no emission rate data were collected.  While substantial data
 were  collected,  scheduling the air  monitoring during cold temperatures and
 snow  cover conditions limited the data's applicability.

      The best technique for screening undisturbed emissions at  this site would
 be to:  1) perform  ambient air monitoring around the perimeter of  the facility
 to determine the  magnitude of baseline  emissions from the site  and to verify
 if any  adverse health or  safety  risks are present;  and 2) survey the  site
 using a portable  analyzer and windscreen  to delineate any localized emission
 "hot  spots."  The results  of  these  screening studies would need to be
 interpreted to determine  if any  further  undisturbed emission measurements were
warranted.
                                      211

-------
5.4  CASE STUDY 4:  WESTERN  PROCESSING LANDFILL

     Case Study 4 is a former industrial waste processing and  recycling
facility.

5.4.1  Site History

     The 13-acre Western Processing site is situated in the Green River Valley
between Seattle and Tacoma,  five miles inland from Puget Sound.  The site was
used for agricultural purposes until 1951 when it was leased to the Department
of Defense.  An anti-aircraft artillery base operated there until the lease
expired in 1960.  The owner  opted for a cash settlement and left in place the
installation's buildings and on-site drainage system that linked its
facilities to a septic tank, a tile subsurface drain field, a  500-gallon
chlorination tank and a ditch leading to Mill Creek which runs along the
site's western border.

     Western Processing, a waste recycling operation, purchased the site in
1960 and claimed to have reclaimed or recycled millions of gallons of liquid
waste and thousands of tons  of solid waste before it was shut  down in 1982.
The wastes handled included:  animal blood, brewer's yeast, chrome baths,
corrosive liquids, crank case oil, flue dust, lead, pickle liquor, plating
bath solutions, solvents and paints, and zinc skimmings.  These wastes were
handled, stored, or disposed of in storage lagoons (acid/caustic/cyanide
wastes), a fertilizer plant, a solvent recovery plant,  bulk storage tanks,
cooling water lagoons, a chlorine gas tank storage house, a laboratory,
naphtha storage tanks, a 55-gallon drum storage area, and piles of flue dust.
By the late 1970s, the below-ground surface impoundments had been filled and
were being used to store waste material.

     The site is located in an industrial area.  A barbed wire fence separates
the site from a bicycle and jogging trail,  which follows a railroad right-of-
way.  This is the nearest community exposure to the site.  Trail users
reported seeing hoses draped over the fence discharging into a ditch along the
railroad tracks that feeds into Mill Creek, prior to the closure of the
recycling plant (67).  The site is shown in Figure 26.
                                      212

-------
5.4.2  Ob.iectives

     The Remedial  Investigation/Feasibility Study  (RI/FS) process did  not
address the  air  pathway  for  contaminant transport  to any meaningful extent.
Therefore, no objectives were  set or met.

5.4.3  Overview  of Fieldwork for Site Characterization

     State and local  inspections of Western Processing or its vicinity date
back to 1977.  These  were  initially concerned about the quality of water in
Mill Creek.  In  1982,  EPA  determined that the company's management practices
were resulting in  the release  of priority pollutants and other contaminants to
the environment.

     A remedial  investigation  initiated in the fall of 1982 led to emergency
and interim  remedial  site  activities in April 1983.  These included removal of
some liquids, solids,  and  drums, and reorganization of concrete blocks from
five surface impoundments  to create a large diked area where excavated
materials were then placed.  The excavated materials contained solvents, paint
sludge, and  some heavy metals.  Also, buried storage tanks and drums were
encountered during this  activity.  Later in the fall, the state of Washington
led an effort to prevent storm water infiltration and runoff, which included
further excavation and berming as well as paving of the reaction pond.

     A Phase I^surface cleanup, funded by the potentially responsible parties,
started with release  of  a  remedial action plan (68) in July 1984.   In that
plan,  it was clear that  air  emissions work had been limited.  The report noted
the northerly direction  of winds would primarily deposit any contaminants
stirred up or blown off  waste  piles in or near Puget Sound.   The report cited
a 1982 air analysis that showed only trace amounts of trichloroethene,
toluene,  xylene,  and tetrachloroethene.  The low levels and high volatility of
these  compounds,  coupled with  wide atmospheric dispersal,  were thought to
create only limited effects  on receptors during .past operations.  "Although
past releases into the atmosphere may have been greater than observed in the
1982 samples and undoubtedly included particulate matter and contaminants.
                                      213

-------
RADIAN
                SANITARY
                CHAIN FIEUO
                DISCHARGE LINE
                                          WESTERN PROCESSING
                     Figure 26.   Western processing site.

-------
Other  than  volatile organics,  this  pathway  is  not  believed  to  have  been
significant."

     That the  air  pathway  was  overlooked  to  some extent during the  RI/FS  stage
is apparent..  However,  Phase  II  remediation  work began in 1987 to remove
shallow wastes at  the  site, and  a comprehensive air monitoring program was
initiated.  This program  is outlined  below  (as described in  Lepic and Foster
(69).)

   An  air monitoring program was implemented to ensure adequate protection of
both the field team and the surrounding community.  Work area monitoring  was
conducted to identify  action levels where personnel protection levels must be
upgraded.   Continuous  upwind and downwind perimeter monitoring was  conducted.

   Direct reading,  real-time instruments were used to determine total gases
and vapors, cyanide, gamma radiation  and combustible gas; particulate
concentrations also were measured.  The field instrumentation used  at Western
Processing  included:

   t    OVA 128: total  organic vapors;
   •    HNU PI  101:  total  organic vapors;
   •    Hand-held  aerosol  monitor (HAM): total particulates;
   •    Gastech CGI: combustible gases;
   t    Ludlum 19:  gamma radiation;
   t    Monitox Compur  4100:  cyanide;
   •    Draeger pump and colorimetric detector tubes: cyanide and methane;
   •    Hi-volume  air samplers:  suspended particulates; and
   •    Recording meteorological station: wind direction and speed.

   Air was monitored regularly at 16 fixed locations around the site perimeter
to detect any  possible  off-site  migration of airborne contaminants.   In
addition,  monitoring was conducted at each sample location to determine
adequate protection levels and to ensure worker safety.  These monitoring
procedures are described below:
                                      215

-------
   •    Borehole and excavation site monitoring: an OVA, HNU and HAM were used
        to monitor the breathing zone at the drill rig and backhoe during the
        subsurface exploration and sampling activities.  A cyanide detector
        and combustible gas indicator were used regularly; and
            •»
   •    Drum, tank and utility monitoring: an OVA and HNU, radiation detector,
        combustible gas indicator and cyanide monitor were used to test the
        atmosphere within containers for flammable vapors.

5.4.4  Scoping

   Collection and review of the existing data were performed.  Based on this
data review, no need for an air pathway analysis was perceived.

5.4.5  Undisturbed Emissions Survey

   No screening or in-depth measurements were made to assess the undisturbed
emissions at the site.  Table 26 lists some of the contamination found at the
site.  Based on the very high concentrations of heavy metals (e.g., lead at
31,000 ppm) found in the surface soil, an evaluation of the entrained
particulate matter from the site would have been advisable, and some screening
measurements were warranted.

5.4.6  Disturbed Emissions Survey

   No screening or in-depth measurements were made to assess the disturbed
emissions at the site.  Based on the very high concentrations of organic
compounds in the subsoil, an evaluation of the emission potential would have
been advisable.
                                      216

-------
TABLE 26. MAXIMUM AND AVERAGE CONCENTRATIONS IN SOIL FOR SELECTED CONTAMINANTS
Contaminant
*
Chromi urn
Zinc
Arsenic
Antimony
Lead
Cyanide
Phenol
Aldrin
Dieldrin
PCB-1248
Hexachloroethane
Phenanthrene
Pyrene
1,1,1 -Tri chl oroethane
Methyl ene Chloride
Toluene
Trichloroethene
Maximum
Surface
5,300
81,000
38
98
31,000
15
19.0
0
0.145
3.30
5,090
20,000
16,000
0
0.130
0
0.037
Cone, (com)
Subsurface
7,600
40,500
102
130
141,000
179
65.0
2.86
3.34
19.6
1,80
62.4
11.0
174
49
394
580
Average3
Cone, (ppm)
594
2,580
3.28
8.59
5,450
11.2
1.65
0.006
0.007
0.341
0.0192
720
184
2.87
1.49
6.44
19.3
  Based on geometric  averaging  approach.
                                     217

-------
5.4.7  Development of BEEs

     No baseline emission estimates for either the undisturbed  or  disturbed
wastes were generated.  Given the available data, the best method  for
determining BEEs for this site would be to take the existing ambient air
monitoring data (discussed below) and back-calculate an emission rate  using an
air dispersion model.  This is done by setting up the model's run  conditions
to match those at the site as closely as possible, and then varying the source
term to find an emission rate that produces downwind concentrations equal to
those actually measured.  Using the respective air monitoring data sets, this
procedure could be applied to both undisturbed and disturbed conditions.

5.4.8  Summary

     The best technique for screening undisturbed particulate matter emissions
at this site would have been to collect upwind/downwind samples on filters
using high-volume sampling pumps (hi-vols).  The total  particulate matter
present in the air would be determined by dividing the filter weight gain by
the volume of air sampled.  Analysis of the filter catch for selected metal
species would assist in assessing the health impacts from undisturbed
emissions.  A less acceptable alternative would have been to measure the
ambient particulate matter loadings using a portable particulate matter
analyzer.

     Screening VOC emissions also would have been advisable, based on the
waste composition data.  The best technique for screening undisturbed VOC
emissions at this site would have been to:  1) perform ambient air monitoring
around the perimeter of the facility to determine the magnitude of baseline
emissions from the site and to verify if any adverse health or safety risks
were present; and 2) survey the site using a portable analyzer and windscreen
to delineate any localized emission "hot spots."  The results of these
screening studies would need to be interpreted to determine if further
undisturbed emission measurements were warranted.
                                      218

-------
     The best  technique  for  assessing  disturbed  VOC  emissions would  have  been
 to expose  representative areas  of waste  using  a  backhoe  (or  drill  rig), and  to
 measure emission  rates using the  flux  chamber  technique.  The best technique
 for  assessing  particulate matter  emissions would have  been to use  hi-vol
 samplers arrayed  downwind (i.e.,  transect technique) to  capture  emissions
 during site disturbances.

 5.5  CASE  STUDY 5:  OUTBOARD MARINE  CORP. LAGOON/LANDFILL

     Case  Study 5 is  a manufacturing site where  harbor sediments and nearby
 land are contaminated with PCBs.

 5.5.1  Site History

     Outboard  Marine  Corp. (OMC)  sits  on the west shore of Lake Michigan, 37
 miles north of Chicago and 10 miles  south of the Wisconsin border.  This
 hazardous  waste site  evolved from an outboard motor manufacturer that used
 PCBs in die cast  machines from  the early 1950s to the early  1970s.

     Over  the  years,  the facility discharge created three areas of contami-
 nation (see Figure 27).   The first is  Waukegan Harbor, a 37-acre irregularly
 shaped harbor  feeding into Lake Michigan.  The operation also led to
 contamination  of  "North  Ditch," a small tributary that drains surface water
 runoff into Lake  Michigan.   A nine-acre parking lot north of the plant was
 identified as  another area, of significant PCB contamination.

     Concerns  about possible receptors of site contamination included the
 harbor's biological community and fish in Lake Michigan.   The City of
 Waukegan, population  67,653  in  1980, is nearby, but the harbor area is zoned
 industrial.  The  15 businesses  in the  immediate harbor area that employ about
3,500 people were the immediate concern.   Also, the local Port Harbor received
heavy recreational use and long-term plans included development of the Upper
Harbor.   Potentially,  people in a variety of locations could be exposed to the
contamination via direct contact,  fish consumption or possible drinking water
contamination.
                                      219

-------
RADIAN
                      EAST-WEST PORTION
                      OF NORTH DITCH
NORTH SHORE
SANITARY
DISTRICT
             OUTBOARD MARINE CORP.
             JOHNSON OUTBOAROS OIV.
             PLANT NO. 2
         NEW
         DIE
         CAST
         COMPLEX
                            OMC CORPORATE
                            HEADQUARTERS
                            OFFICES
                                                                  OMC OUTFALL
                             JOHNSON OUTBOAROS
                             PLANT NO. 1
                   Figure 27.   Map of Case  Study 5  site.

-------
     OMC purchased  roughly  9 million pounds of PCBs from Monsanto Co.  over  a
20-year period beginning  in the early  1950s.  OMC used the PCBs as hydraulic
fluids in die casting machines and related equipment.  This equipment  leaked
routinely and the fluids  ran from the  plant floor into floor drains that
discharged into Waukegan  Harbor and North Ditch.  EPA estimated that as much
as 20 percent of the PCBs purchased could have been discharged.

     It was not until 1975  that the Illinois Environmental Protection  Agency
(IEPA) discovered the high  levels of PCBs in soils and harbor sediments near
CMC's plant.  This  discovery was triggered by a 1971 EPA study that showed  PCB
concentrations in Lake Michigan fish.  In 1976, the EPA began to regulate PCB
disposal.  At that  time,  OMC began to  sample its outfalls and then sealed two
outfalls leading to North Ditch, pursuant to a joint Administrative
Enforcement Order by EPA  and IEPA (70).  OMC later declined to immediately
remove sediments contaminated with PCBs, as demanded by EPA.  When clean-up
negotiations among  EPA, IEPA, and OMC  failed, legal  actions were filed.  These
legal actions were  still  pending in early 1988.  Superfund money for this site
became available in 1983.

5.5.2  Ob.iectives

     The objective  of the APA for this site was to model  the exposure of
downwind receptors  to PCBs  during baseline conditions.
5.5.3  Scoping

     Existing data were collected and reviewed to provide a working knowledge
of the site history, conditions, and environmental setting.  No information
was uncovered that indicated a need to modify the air pathway analysis
objectives.
                                      221

-------
5.5.4  Overview of Fieldwork for Site Characterization

     Discovery steps through site characterization led to the conclusion that
PCB concentrations were significant and that PCB release to the surrounding
environment could follow a number of pathways.

Waukegan Harbor--
     EPA contractors estimated that in Slip Number 3 in the harbor about 7,200
cubic yards of-muck, varying in thickness from 2 to 5 feet, was contaminated
by about 167,200 pounds of PCBs.  Concentrations typically exceeded 500 ppm.
Another 3,700 cubic yards of sand and silt (about 7 feet thick) were
contaminated by about 138,000 pounds of PCBs.  In one localized area near a
former OMC outfall, concentrations exceeded 10,000 ppm.

     In the upper harbor about 35,700 cubic yards of muck, 1 to 5 feet thick,
were contaminated with approximately 5,000 pounds of PCBs.  Concentrations
here typically were 50 to 500 ppm.

North Ditch--
     Contractors broke the North Ditch into three areas for study.  In the
"crescent ditch," about 28,900 cubic yards of soil, roughly 25 feet thick,
were contaminated by about 403,700 pounds of PCBs, creating concentrations
ranging from 5,000 to 38,000 ppm.  Another 2,300 cubic yards of soil 3 feet
thick north of the die storage area were contaminated by an estimated 200
pounds of PCBs.   Concentrations here typically were about 200 ppm.

     The "oval lagoon," about 27 feet deep, contained about 14,600 cubic yards
of soil contaminated by about 85,500 pounds of PCBs in the top 5 feet.
Concentrations within those 5 feet were about 26,000 ppm; no data were
available for below 5 feet.

     In the main part of North Ditch, about 25,000 cubic yards of soil about
25 feet thick were contaminated by at least 4,300 pounds of PCBs.
Concentrations in 200 feet of the ditch's western portion were typically above
5,000 ppm PCBs and another 1,000 feet of the central/western portion of the
ditch showed concentrations of 500 to 5,000-ppm.
                                      222

-------
Parking Lot--
     The parking lot also showed significant contamination  in contractor
studies.  Approximately 278,000 pounds of PCBs were found in 105,000 cubic
yards of soil.  Volatilization was thought to be slight because of soil cover
and partial pavement.  Air emission estimates were not made.

5.5.5  Undisturbed Emissions Survey

     No air monitoring for the presence of PCBs was conducted at the site.
Dispersion modeling was used to estimate rates of PCB volatilization.  The PCB
concentration expected in solution at the sediment/water interface was
estimated by mixing contaminated sediment with water, decanting the mixture,
and measuring the PCB concentration in the water.  This concentration number
was plugged into transport rate equations.  Contractors assumed a
volatilization rate of 3.8 mg/m2/hour from a saturated solution,  based on data
provided by General Electric Corporation.  Assuming volatilization to be
proportional to the PCB concentration in the solution, calculations showed
that roughly 3.3 pounds of PCB were leaving the harbor portion of the OMC site
through the atmosphere per month.  The rate would vary positively with
temperature.  EPA estimated that 12 to 40 pounds of PCBs were volatilizing
from the harbor each year.  In addition, the North Ditch was estimated to be
contributing another 15 pounds of PCBs to the atmosphere per year.

5.5.6  Disturbed Emissions Survey

     No field measurements or modeling estimates were made to assess the
emissions during site disturbances.  It would be advisable to conduct a
laboratory or field study to determine the degree to which volatilization will
increase during dredging or other site remediation work.  The emission
estimates could then be used as inputs to dispersion models to assess the
impact on downwind receptors.
                                      223

-------
5.5.7  Development of BEEs

     As discussed above, undisturbed (baseline) emission estimates were
developed for two of the three operable units at tht site.  These estimates
were 12 to 40 Ib. PCB/year and 15 Ib. PCB/year.  No disturbed emission
estimates were developed.  If emission rate data were available for waste in
the disturbed state (e.g., flux chamber test data), then disturbed emission
estimates should have been developed using the same modeling approach used to
develop the BEEs.

5.5.8  Summary

     The air pathway for contaminant transport was assessed at this site for
undisturbed conditions using modeling techniques.  This was a valid, cost-
effective option, given the logistical  problems of making direct field
measurements at this site and the low probability of detecting PCBs in the
ambient air downwind of the site.

     The best technique for assessing the disturbed PCB emissions would have
been to dredge up representative contaminated material  and directly measure
emissions with a flux chamber.  As an alternative, this approach could be
modified to perform the work in a laboratory setting.
                                     224

-------
                                   SECTION 6
                                   REFERENCES
             •
 1.   US  EPA.   Procedures  for  Conducting Air Pathway Analyses  for  Superfund
     Applications  -  Volume  I,  Application  of Air  Pathway Analyses for
     Superfund Activities.  EPA-450/1-89-001.   Intermin Final  Manual.  July,
     1989.

 2.   US  EPA.   Procedures  for  Conducting Air Pathway Analyses  for  Superfund
     Applications  -  Volume  III,  Estimation of Air Emissions From  Clean-up
     Activities  at Superfund  Sites.   EPA -450/1-89-001.  Interim  Final Manual
     January 1989.

 3.   US  EPA.   Procedures  for  Conducting Air Pathway Analyses  for  Superfund
     Applications  -  Volume  IV, Procedures  for Dispersion Modeling  and  Air
     Monitoring  for  Superfund  Air  Pathway Analysis.  EPA-450/1-89-004.
     Interim Final Manual.  July,  1989.

 4.   Camp, Dresser and McKee,  Inc.  Guidance Document for Cleanup  of Surface
     Impoundment Sites.   Prepared  for U.S. Environmental Protection Agency,
     June 1985.

 5.   Hwang, S.T.   Model Prediction of Volatile Emissions.  Environmental
     Progress, Vol.  4, No.  2,  May  1985.

 6.   Shen, T.T.  Air Pollution Assessment of Toxic Emissions  from  Hazardous
     Waste Lagoons and Landfills.  Environmental International, Vol.  II, pp.
     71-76, 1985.

 7.   Shen, T.T.  Air Quality Assessment for Land Disposal of  Industrial
     Wastes.    Environmental Management, Vol. 6, No. 4, pp. 297-305, 1982.

8.   Shen, T.T.  Estimating Air Emissions from Disposal Sites.  Pollution
     Engineering,  13(8), pp. 31-34, 1981.
                                      225

-------
9.   Shen, T.T. and G.H. Sewell.  Air Pollution Problems of Uncontrolled
     Hazardous Waste Sites.  Civil Engineering for Practicing and Design
     Engineers, 3(3):241-252, 1984.

10.  Balfour; W.D. and C.E. Schmidt.  Sampling Approaches for Measuring
     Emission Rates from Hazardous Waste Disposal Facilities.  Presented at
     the 77th Annual APCA Meeting, June 1984.

11.  Storm, D.L. Hazardous Materials Common to Specific Industries.  In:
     Handbook of Industrial Waste Compositions in California.  Cal. DHS 1978.

12.  Radian Corporation.  Air Quality Engineering Manual for Hazardous Waste
     Site Mitigation Activities, Revision 2.  Air Quality Engineering and
     Technology, Department of Environmental Protection, New Jersey, Division
     of Environmental Quality CN027, 1987.

13.  COM Federal Programs Corp.  Data Quality Objectives for Remedial Response
     ctiviteis.  EPA 540/G-87-004, U.S.  Environmental Protection Agency,
     Washington, DC, 1987.

14.  U.S. Environmental Protection Agency.  Interim Guideliens and
     Specifications for Preparing Quality Assurance Project Plans.  QAMS-
     005/08.  Office of Monitoring Systems and Quality Assurance, Office of
     Research and Development,  Washington, DC, 1980.

15.  Kaplin,  E.J., A.J. Kurtz,  and M. Rahimi.  VOC Sampling for Emission Rate
     Determination and Ambient Air Quality on an Inactive Landfill.  Presented
     at New England Section, Air Pollution Control Association,  Fall 1986
     Conference, Worcester, MA, October 6-7, 1986. 27 pp.

16.  South Coast Air Quality Managmente District.  Landfill  Gas Emissions:
     Report of the Task Force,  El Monte, CA, 1982.
                                     226

-------
17.  Wood, J.A.,  and  M.L.  Porter.  Hazardous Pollutants  in Class  II  Landfills.
     South Coast  Air  Quality  Management District, El Monte, CA, 1986.

18.  Eklund, B.M., W.D.  Balfour, and C.E. Schmidt.  Measurement of Fugitive
     Volatile Organic Emission  Rates.  Environmental Progress, 4(3):199-202,
     1985.

19.  Balfour, W.O., B.M. Eklund, and S.J. Williamson.  Measurement of Volatile
     Organic Emissions from Subsurface Contaminants.  Radian Corporation,
     Austin, TX,  1985. 20  pp.

20.  Dupont, R.R.  Measurement  of Volatile Hazardous Organic Emissions.
     Journal of the Air  Pollution Control Association, 37(3):168-176, 1987.

21.  Balfour, W.D., R.G. Wetherold, and D.L. Lewis.  Evaluation of Air
     Emissions from Hazardous Waste Treatment,  Storage, and Disposal
     Facilities.  EPA  600/2-85/057,  U.S. Environmental Protection Agency,
     Cincinnati,  OH,  1984. 2 vol.

22.  Eklund, B.M., M.R.  Kienbusch, D. Ranum, and T.  Harrison.   Development of
     a Sampling Method for Measuring VOC Emissions from Surface Impoundments.
     Radian Corporation, Austin, TX, no date. 7 pp.

23.  Keinbusch, M.R.  Measurement of Gaseous Emission Rates from Land Surfaces
     using an Emission Flux Chamber:  Users Guide, EPA Contract 68-02-3889,
     Radian Corporation, Austin, TX, 1986.

24.  Cowherd, C. Measurement of Particulate Emissions from Hazardous Waste
     Disposal Sites.  For presentation at: 78th Annual  Meeting, Air Pollution
     Control  Association, Detroit, MI,  June 16-21, 1985.
                                      227

-------
25.  Astle, A.D., R.A. Duffee, and A.R. Stankunas.   Estimating  Vapor and Odor
     Emission Rates  from Hazardous Waste Sites.   In:  National  Conference on
     Management of Uncontrolled Hazardous Waste Sites, U.S.  Environmental
     Protection Agency, et al., Washington, D.C., 1982.  pp. 326-330.

26.  Eklund, B. Detection of Hydrocarbons in Groundwater by  Analysis of
     Shallow Soil Gas/Vapor.  Radian Corporation, Austin, TX, 1985.   78 pp.

27.  Devltt, D.A., R.B. Evans, W.A. Jury, T.H. Starks, B. Eklund,  and A.
     Gholson.  Soil  Gas Sensing for Detection and Mapping of Volatile
     Organics.  National Water Well Association.  1987.

28.  Kerfoot, H.B. Soil-Gas Measurement for Detection of Groundwater
     Contamination by Volatile Organic Compounds.  Environmental Science  and
     Technology, 21(10):1022-1024, 1987.

29.  Schmidt, C.E.,  R. Vandervort, and W.D. Balfour.  Technical Approach  and
     Sampling Techniques Used to Detect and Map Subsurface Hydrocarbon
     Contamination.  For presentation at the 79th Annual  Meeting,  Air
     Pollution Control Association, Minneapolis, MN, 1986.   31 pp.

30.  Schmidt, C.E.,  and W.D. Balfour.  Direct Gas Emission Measurement
     Techniques and  the Utilization of Emissions Data from Hazardous Waste
     Sites.  Reprinted from:  National Conference on Environmental Engineering
     Proceedings, Environmental Engineering Division, ASCE,  1983.  8 pp.

31.  Balfour, W.D.,  C.E. Schmidt, and B.M. Eklund.  Sampling Approaches  for
     the Measurement of Volatile Compounds at Hazardous Waste Sites.  Journal
     of Hazardous Material 14 (1987) 135-148.

32.  EPA Reference Methods, Environment Reporter, December 5, 1980.

33.  Radian Corporation.  Review of Soil Gas Sampling Techniques.  Austin, TX,
     1983.  26 pp.
                                      228

-------
34.  National Institute for Occupational Safety and Health  (NIOSH).  NIOSH
     Manual of Analytical Methods.  1985.

35.  L.J. Thibodeaux, D.G. Parker, and H.H. Heck.  Measurement of Volatile
     Chemical Emissions from Uastewater Basins.  U.S. EPA,  Hazardous Waste
     EngineerinResearch Laboratory, EPA/600/5-2-82/095.  Cincinnati, OH   1982

36.  C.Cowherd, K.Axetell, C.M. Guenther, and G.A. Jutze.   Development of
     Emission Factors for Fugitive Dust Sources.  EPA  450/3-74-037.  U.S.
     EPA, Cincinnati, OH  1974.

37.  Wetherold, R.G., and D.A. Dubose.  A Review of Selected Theoretical
     Models for Estimating and Describing Atmospheric Emissions from Waste
     Disposal Operations.  EPA Contract 68-03-3038, U.S. Environmental
     Protection Agency, Office of Research and Development, Industrial
     Environmental Research Laboratory, Cincinnati, OH, 1982.  73 pp.

38.  Esplin, G.J.  Boundary Layer Emission Monitoring.  JAPCA Vol. 38, No. 9,
     1158-1161 September 1988.

39.  U.S. Environmental Protection Agency.  Guideline on Air Quality Models
     (Revised).  Office of Air Quality Planning and Standards, PB86-245248.
     Research Triangle Park, NC  July 1986.

40.  Baker, L.W., and K.P. MacKay. Screening Models for Estimating Toxic Air
     Pollution Near a Hazardous Waste Landfill.  Journal of the Air Pollution
     Control Association, 35(11):1190-1195, 1985.

41.  U.S. Environmental Protection Agency.  Superfund Exposure Assessment
     Manual.  EPA/540/1-88/001, Washington, DC, April 1988.
                                      229

-------
42.  Farmer, W.J., M.S.  Yang, J.  Letey, W.F.  Spencer,  and M.H.  Roulier.  Land
     Disposal of Hexachlorobenzene Wastes:  Controlling Vapor Movement in
     Soils.  In: Land Disposal of  Hazardous  Wastes.  Proceedings  of the Fourth
     Annual  Research Symposium, U.S.  Environmental  Protection Agency,
     Municipal Environmental Research  Laboratory, San  Antonio,  TX,  March 6, 7,
     and 8,  1978. pp.  182-190.

43.  Farmer, W.J., M.S.  Yang, J.  Letey, and W.F. Spencer.   Land Disposal  of
     Hexachlorobenzene Wastes:  Controlling Vapor Movement  in Soil.
     EPA-600/2-80-119, U.S. Environmental Protection Agency,  Office  of
     Research and Development, Municipal Environmental  Research Laboratory,
     Cincinnati, OH, 1980.  69 pp.

44.  U.S. Environmental  Protection Agency.  Hazardous  Waste,  Treatment,
     Storage and Disposal Facilities  (TSDF) -- Air  Emission Models.   Draft
     Report, Office of Air Quality Planning and Standards,  Research  Triangle
     Park, NC, 1987.  pp. 6-1 to  6-50.

45.  Radian Corporation.  Survey  and Assessment of  Air  Emission Modeling
     Techniques for Landfills.  Draft  Final Report.  EPA  Contract 68-01-7287,
     U.S. Environmental  Protection Agency, Washington,  D.C.,  1988.   115  pp.

46.  Shen, T.T.  Estimating Hazardous Air Emissions from  Disposal Sites.
     Pollution Engineering, 13(8):31-34, 1981.

47.  Shen, T.T.  Air Quality Assessment for Land Disposal of  Industrial
     Wastes.  Environmental Management, 6(4):297-305,  1982.South Coast  Air
     Quality Management District.   Landfill Gas Emissions: Report of the  Task
     Force,  El  Monte, CA, 1982.

48.  Shen, T.T.,  and G.H. Sewell.   Air Pollution Problems of  Uncontrolled
     Hazardous Waste Sites. Civil  Engineering for Practicing  and Design
     Engineers, 3(3):241-252,  1984.
                                      230

-------
 49.   Farino,  W.,  P.  Spawn,  M.  Jasinski,  and B.  Murphy.   Review of Landfill
      AERR Models.   In:  Evaluation and Selection of Models for Estimating Air
      Emissions  from  Hazardous  Waste  Treatment,  Storage,  and Disposal
      Facilities.   Revised Draft  Final  Report.  Contract No.  68-02-3168,  U.S.
      Environmental Agency,  Office of Solid Waste,  Land Disposal  Branch,  1983.
      pp.  5-1  -  5-13.

 50.   Shen, T.T.  Air Pollution Assessment  of Toxic Emissions  from Hazardous
      Waste Lagoons and  Landfills.  Environment  International,  ll(l):71-76,
      1985.

 51.   Thibodeaux, L.J.,  and  S.T.  Hwang.   Landfarming of Petroleum  Wastes  -
      Modeling the Air Emission Problem.  Environmental Progress,  l(l):42-46,
      1982.

 52.   Hwang, S.T. Toxic  Emissions  from  Land  Disposal Facilities. Environmental
      Progress,  l(l):46-52,  1982.

 53.   DeWolf, G.B., and  R.G. Wetherold.   Protocols  for Calculating VOC
      Emissions  from  Land Applications  Using  Emission Models.  Radian
      Corporation, Austin, TX,  EPA  Contract  No. 68-02-3850, U.S. Environmental
      Protection Agency, Research Triangle Park, NC, 1984.  28 pp.

 54.   U.S. EPA AP-42:  Compilation of Air  Pollutant  Emission Factors, Fourth
      Edition.  USEPA/OAQPS  RTP, NC  September 1985.

 55.  Cowherd, C., 6.E. Muleski, P.J.  Englehart, and D.A.  Gillette.  Rapid
     Assessment of Exposure to Particulate  Emissions From Surface
     Contamination Sites.   EPA/600/8-85/002.  Prepared for U.S. Environmental
     Protection Agency,  Office of Research and Development.  Washington, D.C.,
     February 1985.

56.  Mackay,  D., and P.J. Leinonen.  Rate of Evaporation  of Low-Solubility
     Contaminants from Water Bodies to Atmosphere.  Environmental Science and
     Technology, 9(13):  1178-1180, 1975.
                                     231

-------
57.  Mackay, D., and A.T.K. Yeun.  Mass Transfer Coefficient  Correlations  for
     Volatilization of Organic Solutes from Water.  Environmental  Science  and
     Technology, 17(4):211-217, 1983.

58.  Shen, TIT. Hazardous Air Emissions from Industrial Waste Treatment
     Facilities.  In:  Industrial Waste:  Proceedings of the  Fourteenth
     Mid-Atlantic Conference, June 27-29, 1982, J.E. Alleman  and J.T.
     Kavanagh, eds., Ann Arbor Science, Ann Arbor, MI, 1982.  pp.  361-372.

59.  Springer, C., K.T. Valsaraj, and L.J. Thibodeaux.  In Site Methods to
     Control Emissions from Surface Impoundments and Landfills.  Journal of
     the Air Pollution COntrol Association, 36(12): 1371-1374, 1986.

60.  Hwang, S.T.  Estimating and Field-Validating Hazardous Air Emissions from
     Land Disposal Facilities. In: Third Pacific Chemical Engineering
     Conference, Seoul, Korea, 1983. pp. 338-343.

61.  Thomas, R.G. Volatilization from Soil.  In: Handbook of  Chemical Property
     Estimation Methods, W.J. Lyman, W.F. Reehl, and O.H. Rosenblatt, eds.
     McGraw-Hill, New York, NY, 1982.  pp. 16.1 - 16.50.

62.  Mackay, D. and W.Y. Shiu.  A Review of Henry's Law Constants  for
     Chemicals of Environmental Interest.  Journal of Physical Chemistry
     Reference Data, 10(4):  1175-1199, 1981.

63.  Perry, R.H. and C.H. Chilton.  Chemical Engineer's Handbook.  Sixth
     Edition,  McGraw-Hill Book Company, New York, NY  1973.

64.  Brady, N.C.  The Nature and Properties of Soils.  Eighth  Edition,
     MacMillian Publishing Company,  Inc., New York.

65.  Record of Decision,  Remedial  Alternative Selection, Bruin Lagoon Site,
     Bruin Borough,  PA.  September 29, 1986.
                                      232

-------
66.  CH2M Hill.  Phase  I Remedial  Investigation, Lowry Landfill, Vols.  I  and
     II. EPA  No  38.8L08.3  Milwaukee,  WI.   September  2,  1986.

67.  CH2M Hill.  Final  Remedial Investigation Data Report:  Western
     Processing. RA-WA-37-OL16-1,  Kent, WA.  December  17,  1984.

68.  Dames and Moore  and Landau Associates Western Processing  Technical  Basis
     for Remedial Action Plan, Phase  II.  October 3, 1984.

69.  Lepic, K.A. and  A.R.  Foster.  Superfund 1987: Proceedings of  the  Eighth
     National Conference.  Washington, DC. November  16-18,  1987.

70.  U.S. Environmental Protection Agency. Superfund Record of Decision:
     Outboard Marine  Corp. Site,  IL.  EPA/ROD/R05-84/007, Washington,  DC,
     1984.

71.  Skidmore, E.L. and N.P. Woodruff.  Wind Erosion Forces in the United
     States and Their Use  in Predicting Soil Loss.  Agriculture Handbook No.
     346. Washington, D.C., U.S.  Department of Agriculture, Agricultural
     Research Service,  1968.

72.  Emcon Associates.  Methane Generation and Recovery from Landfills.  Ann
     Arbor Science Publishers, Inc.   1982.

73.  U.S. EPA.  Air Emissions from Municipal Solid Waste Landfills.
     Background Information for Proposed Standards and Guidelines.
     Preliminary Draft.  March 1988.

74.  McGuinn,  Y.C., Radian Corporation.  Use of landfill Gas Generation Model
     to Estimate VOC  Emissions from Landfills.   Memorandum to Susan A.
     Thorneloe.  (PB/10AQPS).  June 21, 1988.

75.  U.S. EPA.  Hazardous Waste Treatment, Storage,  and Disposal Facilities
     (TSDF)  Air Emission Models.   EPA-450/3-87-026 (NTIS PB88-198619). 1987.
                                      233

-------
      APPENDIX A
ANNOTATED BILIOGRAPHY

-------
                                  APPENDIX A

                            ANNOTATED BIBLIOGRAPHY
             *

  I.       Adams,  D.F.   Sulfur Gas Emissions from Flue Gas Desulfurization
          Sludge  Ponds.  Journal  of the A1r Pollution Control  Association,
          29(9):963963, 1979.

               This article describes an enclosure used to measure natural
               sulfur species emissions.  This article was used to assist in
               the development of the Radian surface flux chamber.


 II.       Aim,  R.R., C.P. Hanauska, K.A. Olson,  and M.T.  P1ke. The Use of
          Stabilized Aqueous Foams to Suppress Hazardous  Vapors -- A Study of
          Factors Influencing Performance. In: Superfund  '87:  Proceedings of
          the 8th National Conference, The Hazardous Materials Control
          Research Institute, Washington, D.C.,  November  16-18, 1987. pp. 480-
          484.

               This article presents results from laboratory and field testing
               of 3M vapor suppression foams.   Field testing was performed on
               excavated material using a direct emissions approach:  surface
               emissions Isolation flux chamber.


III.       Aim,  R.R., K.A. Olson,  and R.C. Peterson. Using Foam to Maintain Air
          Quality During Remediation of Hazardous Waste Sites. Presented at
          the Air Pollution Control Association's 80th Annual  Meeting and
          Exhibition,  New York, NY, June 21-26,  1987. 17  pp.

               This paper is not  useful for determining emission measurements
               or estimating techniques.  It also is not  useful as a case
               study.   It provides some data on  the effectiveness of foams
               from laboratory tests and Radian  testing of 3M foam, using a
               surface isolation  flux chamber.


 IV.       Aim,  R.R., K.A. Olson,  and E.A. Reiner. Stabilized Foam: A New
          Technology for Vapor Suppression of Hazardous Materials. Presented
          at the  International Congress on Hazardous Materials Management,
          Chattanooga, TN, June 8-12, 1987.  13  pp.

               This paper is not  useful for determining methods of emission
               measurement or estimating, or as  a case study.   It provides
               data on the effectiveness of vapor suppression foam from
               laboratory tests using a "Radian-style" flux chamber and GC.
               The paper Includes few details and references Radian field
               testing of 3M foams.

-------
  V.       Asolan, M.F., and M.J. Barboza.  A Practical Methodology for
          Designing and Conducting Ambient Air Monitoring at Hazardous Waste
          Facilities.  For Presentation at the 79th Annual Meeting of the Air
          Pollution Control Association, Minneapolis, MN, 1986,  16 pp.

             •  This paper presents an approach for designing ambient tir
               monitoring programs, including a decision tree.  The emphasis
               1s placed on establishing program objectives, including why
               sampling 1s performed, for who, and what 1s to be sampled.  The
               approach is intended for use for project planning rather than
               project execution.


 VI.       Astle, A.O., R.A. Ouffee, and A.R. Stankunas.  Estimating Vapor and
          Odor Emission Rates from Hazardous Waste Sites.  In:  National
          Conference on Management of Uncontrolled Hazardous Waste Sites, U.S.
          Environmental Protection Agency, et al., Washington, D»C., 1982.
          pp. 326-330.

               This article discusses sampling and evaluation of emissions for
               odor.  The sampling tunnel may have some usefulness for
               investigating volatile emission rates versus wind speed;
               however, the technique would require development and testing.


VII.       Baker, L.W., and K.P. Mackay. Screening Models for Estimating Toxic
          A1r Pollution Near a Hazardous Waste Landfill.  Journal of the Air
          Pollution Control Association, 35(11):1190-1195, 1985.

               Baker and Mackay evaluate performance of four air dispersion
               models to calculate the ambient air concentration of vinyl
               chloride versus measured concentration downwind of a landfill.
               The vinyl chloride emission rate was calculated using Shen's
               modification of Fanner's gas migration equation.  The air
               dispersion models used are a ground level point source model,
               two virtual point source mdoels, and a simple box model.

VIII.     Balfour, W.D., R.G. Wetherold, and D.L. Lewis. Evaluation of Air
          Emissions from Hazardous Waste Treatment, Storage, and Disposal
          Facilities. EPA 600/2-85/057,  U.S. Environmental Protection Agency.
          Cincinnati, OH, 1984. 2 vol.

               Emission rates based on direct and indirect emission
               measurements were compared to emission rates calculated  from
               predictive models.  Emission rate measurement techniques were
               also compared against each other.  Emission measurement
               techniques include surface isolation flux chamber, vent
               sampling, concentration profile, transect, and mass balance.
               Predictive models include the Thibodeaux, Parker, and Heck
              ^Models (non-aerated surface  impound), the Thibodeaux-Hwang
              'Model  (land treatment), and  an API model  (tanks).  Details for
               each measurement technique and model are provided.

-------
 IX.      Balfour, U.O., C.E. Schmidt, and 8.M. Eklund.  Sampling Approaches
          for the Measurement of Volatile Compounds at Hazardous Waste Sites.
          Radian Corporation, Austin, TX, no date. 29 pp.

               Sampling techniques for measuring volatile emission rates  and
               for measuring soil gas concentrations are discussed.  Emission
               rate techniques Included are emission Isolation flux chamber,
               vent sampling, concentration profile, transect, and mass
               balance.  Soil gas concentration techniques are headspace
               analysis of soil cores, soil gas probes (ground probe), and
               passive samplers.

               Each approach 1s described, and the applicable equations are
               presented.


  X.      Balfour, W.D., and C.E. Schmidt.  Sampling Approaches for Measuring
          Emission Rates from Hazardous Waste Disposal Facilities.  Radian
          Corporation, Austin, TX, 1984.  13 pp.

               Balfour and Schmidt present five sampling approaches for
               measuring volatile emissions:  surface emission isolation  flux
               chamber, vent sampling, concentration-profile, transect
               technique, and mass balance.  A comparison of the applicability
               of each technique to various TSOF sites or units is given.


 XI.      Balfour, W.O., B.M. Eklund, and S.J. Williamson.  Measurement of
          Volatile Organic Emissions from Subsurface Contaminants.  Radian
          Corporation, Austin, IX, 1985. 20 pp.

               This paper presents results of field measurement performed with
               the surface emission isolation flux chamber.  It also evaluates
               the effect of six operating variables on measured emission rate
               and appropriateness of statistical sampling procedure for  area
               .sources, and provides an analysis of variability in the method.

XII.      Banerjee, P., and O.H. Homer. The Impacts of Using Assumed Versus
          Site-Specific Values in Determining Fate and Transport. In:
          Superfund'87: Proceedings of the 8th National Conference, The
          Hazardous Control Research Institute, Washington, O.C., November 16-
          18, 1987. pp. 126-128.

               This article does not discuss emission rate measurement;
               rather, 1t emphasizes the Importance of the need to obtain
               site-specific data rather then assumed or literature-derived
               values as inputs to risk assessment.


XIII.      Berrafato, L.R., and R.A. Wadden.  Predicted vs. Measured Air
          Emissions of Volatile Organics from a Simulated Hazardous Waste
          Lagoon.  In:  Toxic Hazardous Wastes, Proceedings of the 18th Mid-
          Atlantic Hazardous Waste Conference, Chera. Ind. Inst. Toxicol.,
          Research Triangle Park, NC, 1986.  pp. 515-525.

-------
               Evaporation of toluene and chlorobenzene from a simulated
               lagoon was measured based on the liquid concentration of these
               chemicals 1n the lagoon.   The evaporation rate was compared to
               a predictive node!  similar to the Mackay-leinonen Model.  The
               rtsults showed the model  may be useful for order of magnitude
            .  estimates.


XIV       Bilslcy, I.L.  Air Pollution Aspects of Hazardous Waste Disposal in
          Texas. Environmental Progress, 5(2):123-129, 1986.

               This article examines the Texas administrative review process
              -for proposed hazardous waste disposal facilities.  A case study
               of a waste disposal facility application is reviewed.


XV.       Blasko, M.J., B.F. Cockroft, W.C. Smith, and P.P. O'Hara.  Design of
          Remedial Measures and Waste Removal Program, Laekawanna Refuse
          Superfund Site. In: Superfund '87: Proceedings  of the 8th National
          Conference, The Hazardous Control Research Institute, Washington,
          D.C., November 16-18, 1987.  pp. 367-370.

               This article discusses the development of the design and con-
               struction bid package for the remedial measures and removal
               program.  No information on emissions data is given.  The RI
               report would have to be reviewed directly to see if a case
               study exists.


XVI.      Breton, M., T. Nunno, P. Spawn, W. Farino, and R. Mclnnes.
          Evaluation and Selection of Models for Estimating Air Emissions from
          Hazardous Waste Treatment, Storage, and Disposal Facilities.  EPA-
          450/8-34-020, U.S. Environmental Protection Agency, Office of Air
          Quality Planning and Standards, Emission Standards and Engineering
          Division, Research Triangle Park, NC, 1984.  157 pp.

               Mathematical models describing the release rate of volatile air
               emissions from hazardous waste treatment, storage, and disposal
               facilities were compiled and reviewed.  Mathematical modeling
               techniques which predict volatile air emission release rates
               from landfills, landfarms, surface impoundments, storage tanks,
               wastewater treatment processes, and drum handling and storage
               facilities were assessed.  Existing field test validation
               efforts were also reviewed.  Models reviewed include:  landfill
               -- Fanner Model, Shen modification of Fanner Model, Thibodeaux
               a Model, Thibodeaux Convective "Add On" Model, Thibodeaux  b
               Model, and Shen's Open Dump Model; land treatment --
               Thibodeaux-Hwang Model and Hartley Model; lagoons -- Mackay  and
               Leinonen Model, Mackay and Wolkoff Model, Thibodeaux, Parker
               and Heck model, Shen Model, Smith Model, and McCord Model.

-------
XVII.     Caputo, Jr., K., and R.L. Blttle.  Case History:  A Superfund
          Cleanup In Central Pennsylvania.  In:  Hazardous and Toxic Wastes:
          Technology, Management and Health Effects, S.K. Majumdar and E.W.
          Miller, eds.  Pennsylvania Academy of Science, Easton, PA, 1984.
          pp. 228-241.

               This chapter 1s not relevant to the current project; no
               emission measurements are reported.  Caputo and EHttle describe
               the general details of an emergency cleanup at an Industrial
               site.


XVIII.    Caravanos, J., and T.T. Shen. The Effect of Wind Speed on the
          Emission Rates of Volatile Chemicals from Open Hazardous Waste Dump
          Sites.  Source unknown.

               This article presents a modified diffusion equation (Shen
               Model), which includes wind speed as a variable.  Experimental
               data are presented for benzene, carbon tetrachloride, and
               trichloroethylene applied to soil, which approximate a spill.
               The equation also could be applied to waste exposed at surface.


XIX.      Cassis, J.A., E.P. Kunce, and T.A. Pederson.  Remedial Action at
          Uncontrolled Hazardous Waste Sites:  Problems and Solutions.  In:
          Hazardous Waste Management for the 1980s, T.L. Sweeney, H.G. Bhatt,
          R.M. Sykes, and O.J. Sproul, eds.  Ann Arbor Science, Ann Arbor, MI,
          1982.  pp.241-264.

               This chapter is not relevant to this program.  The authors
               describe a remedial action plan for the Pollution Abatement
               Services Oswego Site in Oswego, New York.


XX.       Clraorelli, A.J.  Palraerton Zinc National Priorities List Site:
          Atmospheric Deposition Analysis of Cadmium, Zinc, Lead and Copper  in
          the Vicinity of the New Jersey Zinc Palmerton Facility. U.S.
          Environmental Protection Agency Region III Air Management Division,
          1986.  93 pp.

               Cimorelli discusses heavy metal deposition from stack
               emissions.  Meteorological data are used to identify areas of
               high deposition, to design a soil sampling program.


XXI.      Countess, R.J., R. Brewer, and R.J. Gordon.  Sampling Airborne Toxic
          Organics by Remote Control.  Presented at the 78th Annual Meeting  of
          the A1r Pollution Control Association, Detroit, MI, 1985.  16 pp.

               This paper describes a radio-controlled air sampler, which
             > would be useful as a sampling method.

-------
XXII.     Cowherd, C. Measurement of  Part1culate  Emissions  from Hazardous
          Waste Disposal Sites.  For  presentation at:  78th  Annual  Meeting,  Air
          Pollution Control Association, Detroit,  MI,  June  16-21,  1985.

               Article describes the  MRI wind tunnel  and  exposure  profiling
             • techniques for partieulate emissions  rite  measurement.   The  MRI
               wind tunnel is a portable wind tunnel  which  can be  used for
               direct emissions measurement.  The exposure  profiling technique
               is used for indirect emissions measurement and is similar to
               the transect technique.


XXIII.    Cox,'R.D., K.J. Baughman, and R.F. Earp.   A Generalized  Screening
          and Analysis Procedure for  Organic Emissions from Hazardous  Waste
          Disposal Sites.  In:  Proceedings of the 3rd National  Conference  and
          Exhibition on Management of Uncontrolled Waste  Sites,  Washington,
          D.C., 1982.

               The authors describe a technique developed by Radian
               Corporation for analysis of gas, liquid, and solid
               environmental samples.  The technique uses gas chromatography
               with flame ionization, photoionization and Hall  electrolytic
               conductivity detectors, as well as mass spectrometry.


XXIV.     DeWolf, G.B., and R.G. Wetherold.  Protocols for  Calculating VOC
          Emissions from Surface Impoundments Using  Emission Models:
          Technical Note.  Radian Corporation, Austin,  TX,  EPA Contract No.
          68-02-3850, U.S. Environmental Protection  Agency,  Research  Triangle
          Park, NC, 1984.  34 pp.

               OeWolf and Wttherold present the Thibodeaux  models  for aerated
               and rsonaeratad steady-state impounds  and the Mackay and
               Liinonen Model for unsteady-state  impounds.   Input  variables
               are defined, sources of these variables are  suggested,  and
               approximate precision  levels for the  variables are  given.
               Physical property inputs are discussed and methods  for their
               estimation are given,  along with selected  values for some
               materials.


XXV.      OeWolf, G.B., and R.G. Wetherold.  Protocols for  Calculating VOC
          Emissions fro«> Land Applications Using  Emission Models.   Radian
          Corporation, Austin, TX, EPA Contract No,  68-02-3850,  U.S.
          Environmental Protection Agency, Research  Triangle Park, NC, 1984.
          28 pp.

               The authors present the Thibodeaux-Hwang Model for land
               treatment and the Fanner Model for landfills.  Input variables
               are defined, sources of the variables are  suggested, and
              -approximate precision  levels for the  variables are  given.
               Physical property inputs are discussed and methods  for their
               estimation are given,  along with selected  values for some
               materials.

-------
XXVI.      Dupont, R.R. A Flux Chamber/Solid Sorbent Monitoring System  for  Use
          In Hazardous Organic Emission Measurements from Land Treatment
          Facilities.  Presented at the 79th Annual Meeting of the Air
          Pollution Control Association, Minneapolis, MN, 1986.  15 pp.

               This paper reports on testing of the emission Isolation flux
               chamber with Tenax\ tubes as the sampling media.


XXVII.    Dupont, R.R. A Flux Chamber/Sorbent Tube Monitoring System for
          Hazardous Organic Emission Measurements from Land Treatment
          Facilities.  In:  192nd National Meeting, American Chemical  Society,
          Division of Environmental Chemistry, 26(2):394-397, 1986.

               This 1s not useful for a case study; the technology 1s  already
               known.  The article reports that testing of the emission isola-
               tion flux chamber with Tenax\ tubes for sample collection was
               effective for measuring specific volatile species under both
               laboratory and field conditions.


XXVIII.   Dupont, R.R.  Measurement of Volatile Hazardous Organic Emissions.
          Journal of the A1r Pollution Control Association, 37(3):168-176,
          1987.

               An emissions isolation flux chamber was laboratory tested In
               combination with Tenax\ and charcoal tube sampling to determine
               the recovery efficiencies for selected organics.  The testing
               validates the use of a "Radian-style" flux chamber.  Testing
               included the use of flow rates significantly below the  standard
               protocol.


XXIX.      Elclund, B.M., W.D. Balfour, and C.E. Schmidt.  Measurement of
          Fugitive Volatile Organic Emission Rates.  Environmental Progress,
          4(3):199202, 1985.

               This article describes the design and operation of a "Radian-
               style" emission isolation flux chamber.  It also provides
               limited data from several projects, which can be used as
               emission measurement case studies.

XXX.      Elclund, B. Detection of Hydrocarbons in Groundwater by Analysis  of
          Shallow Soil Gas/Vapor.  Radian Corporation, Austin, TX, 1985.   78
          pp.

               This report describes five method.s of measuring soil vapor
               concentrations:  surface flux chamber, soil probe, downhole
               flux chamber, accumulator, and soil coring.  All five methods
             , s/aiou1d be useful for data collection for direct measurement
               and/or predictive modeling.

-------
XXXI.     EJclund, B.M., W.O. Balfour, and C.E. Schmidt.  Measurement  of
          Fugitive Volatile Organic Compound Emission  Rates with  an  Emission
          Isolation Flux Chamber.  For presentation at:  AIChE  Summer National
          Meeting, Philadelphia, PA, 1984.  8 pp.

             .  The authors present the procedures for  using the surface
               emission Isolation flux chamber for direct emission rate
               measurement.  Also presented are the results of  measurements at
               two spill sites, three landfills, several surface  impounds, a
               landfarn, and a remedial action.


XXXII.    EJclund, B.M., M.R. Klenbusch, D. Ranum, and  T. Harrison.
          Development of a Sampling Method for Measuring VOC  Emissions from
          Surface Impoundments.  Radian Corporation, Austin,  TX,  no date. 7
          pp.
               This paper describes the development program for evaluating and
               modifying the design and operation of the surface  isolation
               flux chamber for use on surface impounds.


XXXIII.   Enfield, C.G., R.F- Carsel, S.Z. Cohen, T. Phan, and  D.M. Walters.
          Approximating Pollutant Transport to Ground  Water.  Ground  Water,
          20(6): 711-722, 1982.

               This article does not provide information on emission  rate
               determination.  It present three transport models  for
               evaluating the movement of organic chemicals through  the soil
               to the groundwater.  The models include losses due to
               degradation and sorption.  Field data are compared to  the
               models.

XXXIV.    Engineering Science, Inc. Determination of Air Toxic  Emissions from
          Non-Traditional Sources in the Puget Sound Region.  EPA 910/9-86-
          148, U.S. Environmental Protection Agency, Region X and Puget Sound
          A1r Pollution Control Agency, Seattle, WA, 1986.  108 pp.

               This report develops emission estimates for several selected
               sources in the Puget Sound Region including POTWs, industrial
               wastewater treatment plants, Superfund  sites,  municipal land-
               fills, and TSDFs.  Emissions are based  on theoretical
               equations.  The report contains some useful examples  of
               theoretical equation applications.  The report may also provide
               case study site information.


XXXV.     Farino, W., P. Spawn, M. Jasinski, and B. Murphy.   Review  of
          Landfill AERR Models.  In: Evaluation and Selection of  Models for
          Estimating Air Emissions from Hazardous Waste Treatment, Storage,
          and Disposal Facilities. Revised Draft Final Report.  Contract No.
          68-02-3168, U.S. Environmental Agency, Office of Solid  Waste, Land
          Disposal Branch, 1983. pp. 5-1 - 5-13.

-------
               This section describes six predictive models that estimate
               volatile emissions:  Fanner, Shen's modification of Farmer,
               TMbodeaux (three variations), and Shen's Open Dump.  The first
               five models are based primarily on gas diffusion through the
               landfill cover.
XXXVI.    Fanner, U.J., M.S. Yang, J. Letey, W.F. Spencer, and H.H. Roulier.
          Land Disposal of Hexachlorobenzene Wastes: Controlling Vapor
          Movement In Soils. In: Land Disposal of Hazardous Wastes.
          Proceedings of the Fourth Annual Research Symposium, U.S.
          Environmental Protection Agency, Municipal Environmental Research
          Laboratory, San Antonio, TX, March 6, 7, and 8, 1978. pp. 182-190,

               Fanner presents a predictive equation for determining
               hexachlorobenzene vapor diffusion through a soil cover.
               Volatilization through soil, water, and polyethylene film was
               studied in laboratory simulations.  The predictive equation
               should be applicable to other waste types.


XXXVII.   Fanner, W.J., M.S. Yang, J. Letey, and W.F. Spencer.  Land Disposal
          of Hexachlorobenzene Wastes:  Controlling Vapor Movement in Soil.
          EPA-600/280-119, U.S. Environmental Protection Agency, Office of
          Research and Development, Municipal Environmental Research
          Laboratory, Cincinnati, OH, 1980.  69 pp.

               The volatilization fluxes of hexachlorobenzene through a
               covering of soil, water, and polyethylene film were simulated
               1n the laboratory.  Volatilization through soil was directly
               related to soil porosity.  Fanner develops a diffusion equation
               for determining flux rates through the soil covering.


XXXVIII.  Glllespie, D.P., F.J. Schauf, and J.J. Walsh.  Remedial Actions at
          Uncontrolled Hazardous Waste Sites, Survey and Case Study Investiga-
          tion.  In:  Proceedings of the Second National Symposium on Aquifer
          Restoration and Ground Water Monitoring, National Water Well
          Association, Worthington, OH, 1982.  pp. 369-374.

               This paper 1s not useful.  It discusses how nearly half of all
               remedial actions completed by 1980 were ineffective at cleaning
               up the sites.
XXXIX.     Gravitz, N. Derivation and Implementation of Air Criteria During
          Hazardous Waste Site Cleanups. Journal of the Air Pollution Control
          Association, 35(7):753-758, 1985.

               This article presents an approach for developing fenceline air
             ^monitoring criteria to protect community health.  The approach
               is dependent on developing acceptable community exposure
               levels, and back calculating the fenceline concentration by
               assuming Gaussian wind dispersion.  The method does not require
               emissions measurement.

-------
XL        Greenberg, M.  A Review of: 1) A Technical Approach  for Sampling and
          Analysis of VOCs at Hazardous Waste Sites and 2)  Some  Case Studies
          in New Hampshire. Presented before the Fall Meeting, Air Pollution
          Centre! Association New England Section, Conference  on Air Toxics,
          Worcester, MA, October 6-7, 1986.  30 pp.

               This document contains overhead slides for a conference
               presentation on sampling VOCs in ambient air at landfills.
               Case studies are not worked up as emission estimates.


XLI.      Hani sen, R.C., and M.A. McOevltt.  Protocols for  Sampling  and
          Analysis of Surface Impoundments and Land Treatment/Disposal Sites
          for VOCs.  EPA Contract No. 68-02-3850, U.S. Environmental
          Protection Agency, Office of Air Quality Planning and  Standards,
          Research Triangle Park, NC, 1984.  88 pp.

               The report gives detailed description for analysis of volatile
               organics by Radian gas chromatography with multiple detectors
               (GC/MD) system and gas chromatography with mass spectrometry
               (OS/MS) system.  The report gives a brief discussion  of predic-
               tive emission models and lists inputs to models,  but  does not
               present models.  A description of a statistical approach to
               cellecting sufficient samples for representativeness  is
               included.


XLIL     Helslng, L.O., M.P. Morningstar, J.B. Berkowitz,  and T.T.  Shen. Risk
          Analysis of Pollutants at Hazardous Waste Sites:  Integration Across
          Media is the Key. Ins Superfund '87: Proceedings  of the 8th
          National Conference, The Hazardous Control Research  Institute,
          Washington, D.C., November 16-18, 1987.  pp. 471-475.

               This paper describes the types of pollutants frequently found
               at hazardous waste sites and how they can be transferred from
               one media to another.  This paper describes  how all media need
               to be taken into account when performing risk analysis.  No
               information on emission rate determination is provided.


XLIII.    Hwang, S.T. Measuring Rates of Volatile Emissions from Non-Point
          Source Hazardous Waste Facilities.  Presented at  the 75th  Annual
          Meeting of the Air Pollution Control Association, New  Orleans,  LA,
          1982.  22 pp.

               This paper presents Concentration Profile, Plume  Mapping  (tran-
               sect), and Upwind/Downwind models for indirectly  measuring
               emission rates.


XLIV.     Hwang, S.T. Toxic Emissions from Land Disposal Facilities. Environ-
          mental Progress, l(l):46-52, 1982.

-------
               Hwang gives theoretical equations for emission rate estimates
               from surface  Impoundments  (natural surface and aerated),  land-
               fills (based  on soil diffusion), and land treatment (oily
               wastes).  The equations require knowledge of waste and  site
               characteristics to determine variables and coefficients.   All
               equations would be difficult to apply for a complex waste,
               requiring computer calculation.  The equations can be applied
               more easily to a single component waste, or Its primary compo-
               nents.


XLV.      Hwang, S.T. Comparison of Model Predicted Volatile Emission  Rates
          Versus Results of  Field Measurements at Hazardous Waste Sites.
          Presented at:  American Institute of Chemical Engineers, National
          Meeting, Summer 1984.  American Institute of Chemical Engineers, New
          York, NY.  18 pp.

               Paper presents a comparison of measured and estimated emission
               rates based on field sampling results and theoretical models.
               Measurements  were performed at landfills, surface Impounds, and
               land treatment facilities.  Field sampling techniques used were
               concentration profile, transect, and surface emission Isolation
               flux chamber.  The predicted emission rates were generally
               within the confidence  Intervals of the measured emission  rates,
               although the  author indicates that more study is required to
               validate the  models.   The specific models used are referenced
               but not discussed.


XLVI.     Hwang, S.T.  Estimating and Field-Validating Hazardous Air Emissions
          from Land Disposal Facilities.  In: Third Pacific Chemical
          Engineering Conference, Seoul, Korea, 1983. pp. 338-343.

               Models for estimating  volatile emissions are reviewed,
               Including Shen's modification of Farmer's equation (landfill),
               the  Thlbodeaux-Hwang  equation (land treatment) and surface
               'impound equation.  Predicted versus measured emissions
               (concentration profile and upwind/downwind techniques)  are
               compared, but limited  data are given.


XLVII.    Hwang, S.T.  Model Prediction of Volatile Emissions. Environmental
          Progress 4(2):141-144, 1985.

               Hwang presents a comparison of measured and predicted emission
               rates.  Measurement techniques Include transect and
               concentration profile  techniques.  Isolation flux chambers and
               upwind/downwind are also discussed.  The article contains
               limited data  from potential case studies.


XLVIII.   ICF, Incorporated. The RCRA Risk-Cost Analysis Model Phase III
          Report Appendices. Submitted to the Office of Solid Waste, Economic
          Analysis Branch, U.S. Environmental Protection Agency, 1982.
          Appendix E, 31 pp.

-------
               Appendix E describes natural chemical, physical, and biological
               processes that reduce the concentration of chemicals in the
               environment.  These processes are the basis for deriving
               surfaci and groundwater decay rates for the chemicals  included
              . 1n the RCRA Risk-Cost Analysis Model.  Important for baseline
               emission rate estimates 1s the discussion on volatilization
               from water.


XLIX.     Jubach, R.W., R.R. Stoner, T.F- laccarino, and D.R. Smiley.  An
          Atmospheric Field Program Conducted at a Hazardous Waste Site.
          Presented at the 78th Annual Meeting of the Air Pollution Control
          Association, Detroit, MI, 1985.  14 pp.

               This study characterizes the atmospheric dispersion at a
               hazardous waste site from ground level release.  The study
               consists of releasing a tracer gas and measuring the concentra-
               tion with two sampling arrays (i.e., transect techniques).
               Measuring the of actual waste releases are not included.
               Downwind data could be useful as a theoretical case study.


L.        Kaplin, E.J., A.J. Kurtz, and M. Rahirai.  VOC Sampling for  Emission
          Rate Determination and Ambient Air Quality on an Inactive Landfill.
          Presented at New England Section, Air Pollution Control Association,
          Fall 1986 Conference, Worcester, MA, October 6-7, 1986. 27  pp.

               This paper describes sampling performed at an inactive landfill
               in New York (a municipal waste landfill containing industrial
               waste). Sampling included ambient air, emission flux chamber
               (crude), crevices, and vent, sampled with Tenax, PUF/GFF, mylar
               bags, impingers, and high volume samplers with GFF.  Flux
               measurements included both covered and uncovered areas.  Data
               are somewhat limited, but this study may be useful as  a case
               study.


LI.       Karably, L.S., and K.B. Babcock. Effects of Environmental Variables
          on Soil Gas Surveys. In: Superfund '87: Proceedings  of the 8th
          National Conference, The Hazardous Control Research Institute,
          Washington, D.C., November 16-18, 1987.  pp. 97-100.

               This may be useful as a case study if sufficient soil  data  are
               available.  Soil gas was measured at 40 locations and  over  time
               at a fire fighting training area.  This paper reports  how
               environmental conditions (principally weather) affected
               results.  Limited analytical data are presented.


LII.      Karirai, A.A., W.J. Fanner, and M.M. Cliath.  Vapor-phase Diffusion
          of Benzene in Soil.  Journal of Environmental Quality, 16(l):38-43,
          1987.

-------
               The authors use Farmer's diffusion model and laboratory testing
               method to determine benzene emission rates through soil (I.e.,
               landfill emissions).  The effects of soil bulk density and
               moisture content are Investigated.


 LIU.     Kelnbusch, N.R.  Measurement of Gaseous Emission Rates from Land
          Surfaces using an Emission Flux Chamber:  Users Guide, EPA Contract
          68-02-3889, Radian Corporation, Austin, TX, 1986.

               This document Includes a detailed description for using the
               emission Isolation flux chamber, Including design of sampling
              sprogram, QA/QC, data reduction, and examples.


LIV.       Kerfoot, H.B. Soil-Gas Measurement for Detection of Groundwater
          Contamination by Volatile Organic Compounds.  Environmental Science
          and Technology, 21(10):1022-1024, 1987.

               This article describes sampling using a ground probe system for
               soil gas.  Soil gas data are correlated with groundwater data
               showing good correlation.  Soil gas data are correlated with
               soil probe depth below ground surface, also showing good
               correlation.  Ground water was at shallow depth in this study
               (3-4m).  Data could be used as a case study If additional  soil
               characteristics data are available.


LV.        Kerfoot, H.B., and P.B. Durgin. Soil-Gas Surveying for Subsurface
          Organic Contamination: Active and Passive Techniques. In: Superfund
          '87: Proceedings  of the 8th National Conference, The Hazardous
          Control Research Institute, Washington, O.C., November 16-18, 1987.
          pp. 523-524.

               This article provides an overview of considerations in
               designing a soil-gas survey.  It does not provide information
               on sampling techniques for air pathway assessment.


LVI.       Leplc, K.A., and A.R. Foster.  Remediation at a Major Superfund
          Site: Western Processing -- Kent, Washington.  In: Superfund '87:
          Proceedings  of the 8th National Conference, The Hazardous Control
          Research Institute, Washington, O.C., November 16-18, 1987.  pp. 78-
          84.

               This article summarizes a remedial Investigation conducted at a
               Western Processing site.  Continuous upwind and downwind air
               monitoring was conducted, as well as some location-specific
               sampling.  It may be useful as a case study.  No data on air
               monitoring are presented; the author would have to be
               contacted.

-------
LVII.     Liwis, R.G., B.E. Martin, D.L. Sgontz, and J.E. Howes.  Measurement
          of Fugitive Atmospheric Emissions of Polychlorinated Blphenyls  from
          Hazardous Waste Landfills.  Environmental Science and Technology,
          19(10):986-991, 1985.

             .  This article describes tir sampling for PCBs at three
               uncontrolled landfills and one controlled landfill.  Emission
               rates were not calculated.  The data set may be usable for
               calculating emission rates.  Sampling was performed using  low-
               volume and high-volume samplers and PDF cartridges.


LVIII.    Mackay,- D.M., P.V. Roberts, and J.A. Cherry.  Transport of Organic
          Contaminants in Groundwater.  Environmental Science and Technology,
          19(5): 384-392, 1986.

               This article describes the mechanisms involved in the transport
               of organic chemical contaminants in ground water, including
               advection, dispersion, sorption and transformation.


LIX.      Mackay, 0., and P.J. Leinonen.  Rate of Evaporation of Low-
          Solubility Contaminants from Water Bodies to Atmosphere.
          Environmental Science and Technology, 9(13): 1178-1180, 1975..

               The authors present predict models (Mackay and Leinonen) for
               emission rates from aqueous systems.  Equations are presented
               for both steady and unsteady state systems.


LXo       Mackay, D., and A.T.K. Yeun.  Mass Transfer Coefficient Correlations
          for Volatilization of Organic Solutes from Water.  Environmental
          Science and Technology, 17(4):211-217, 1983.

               Volatilization rates of organic compounds in water were
               measured in a wind-wave tank for compounds of varying Henry's
              'Law Constants.  The data yielded correlations for liquid and
               vapro-phase transfer coefficients as a function of windspeed,
               and showed that interactions did not occur when mixtures of
               compounds volatilized simultaneously.


LXI.      Marquardt, G.O.  Toxic Air Quality Investigation at a Hazardous
          Waste Site.  In: Superfund '87: Proceedings  of the 8th National
          Conference, The Hazardous Control Research Institute, Washington,
          O.C., November 16-18, 1987.  pp. 284-295.

               This article does not provide data for baseline conditions.
               However, it may be useful as a case study.  Sampling was  per-
               formed upwind, on site, and downwind during field investigation
                (drilling) and during a controlled release from a gas well.
               Sampling techniques could be used for indirect measurement of
               baseline emissions; techniques include Tenax\, carbon molecular
               sieve, high-volume particulate sampler, and PDF.

-------
LXII.      Meegoda, N.J., and P. Ratnaweera. A New Method to Characterize
          Contaminated Soils. In: Superfund '37: Proceedings  of the 8th
          National Conference, The Hazardous Control Research Institute,
          Washington, D.C., November 16-18, 1987.  pp. 385-389.

               This paper 1s not related to emission rate determination;  it
               presents an electrical method to determine engineering
               characteristics of soil.


LXIII.    Miller, G.C., V.R. Herbert, and R.G. Zepp.  Chemistry and
          Photochemistry of Low-Volatility Organic Chemicals on Environmental
          Surfaces.  Environmental Science and Technology, 21(12):1164-1167,
          1987.

               This article discusses factors affecting the transformation,
               mobility, and fate of xenobiotic chemicals (specifically
               dioxins and PAHs).  Emphasis is placed on discussing effects on
               and near the soil surface and factors that affect the rate of
               photolysis.


LXIV.      Panaro, J.M. Air Monitoring and Data Interpretation During Remedial
          Action at a Hazardous Waste Site.  In:  Hazardous Wastes and
          Environmental Emergencies: Management, Prevention, Cleanup, Control,
          Hazardous Materials Control Research Institute, Houston, TX, 1984.
          pp.160-164.

               Panaro describes the air monitoring program employed during
               Initial remedial measures at a chemical recycling plant.  He
               does not address emission estimates.


LXV.      Polcyn, A.J., and H.E. Hesketh.  A Review of Current Sampling and
          Analytical Methods for Assessing Toxic and Hazardous Organic
          Emissions from Stationary Sources.  Journal of the Air Pollution
          Control Association, 35(1):54-60, 1985.

               This article is a brief review of sampling methods (i.e.,
               sample collection media) and analytical methods.  The methods
               are summarized in tables, and give references for detailed
               descriptions.  The article does not discuss measurement
               techniques (I.e., sampling design) or modeling.


UCVI.      Qulmby, J.M., R.W. Cibulskis, and M. Gruenfeld.  Evaluation and Use
          of a Portable Gas Chromatograph for Monitoring Hazardous Waste
          Sites.  Source unknown.

-------
               This paper evaluates the use of the Century Systems Organic
               Vapor Analyzer Model OVA-128 which 1s a portable gas
               chrofflttograph with a fume ionization detector  (GC-FID).  The
               piper addresses instrument operating performance, QA/QC
               consideration, operational difficulties and recommended  field
               uses.  Instruments can be used to perform field screening of
               ambient air, as well as other uses.


LXVII.    Radian Corporation.  Review of Soil Gas Sampling Techniques.
          Austin, TX, 1983.  26 pp.

              ^Report presents the results of a literature review to  identify
               techniques applicable to soil gas sampling and measurement.


LXVIII.   Radian Corporation.  3M Foam Evaluation for Vapor Mitigation:
          Technical Memorandum.  Sacramento, CA, 1986.  9S pp.

               The effectiveness of temporary and stabilized foam for control-
               ling VOC emissions from petroleum refinery waste was tested.
               Testing was performed using the surface emission isolation flux
               chamber.


LXIX.     Radian Corporation.  A1r Quality Engineering Manual for Hazardous
          Waste Site Mitigation Activities, Revision No. 2.  Sacramento, CA,
          1987.  291 pp.

               This guidance document provides general information for
               designing and reviewing air monitoring programs for hazardous
               waste site remedial programs.  An overview of New Jersey's
               agency involvement in the remedial process is given.   Sampling
               and analysis methods are discussed in detail.


LXX.      Radian Corporation.  Ambient Air Monitoring at Hazardous Waste and
          Superfund Sites, Revision No. 2.  Sacramento, CA, 1987.  389  pp.

               This guidance document provides information for reviewing air
               quality engineering activities for hazardous waste site
               remediation programs.  It also provides an overview of the
               phases of the hrzardous site remediation process and the role
               of New Jersey's agencies 1n the process.  Descriptions are
               given for types of waste sites, potential air contaminants,  and
               basic remedial processes.


IXXI.     Radian Corporation.  Survey and Assessment of Air Emission  Modeling
          Techniques for Landfills.  Draft Final Report.  EPA Contract  68-01-
          7287, U.S. Environmental Protection Agency, Washington, D.C., 1988.
          115 pp.

-------
               Article reviews theoretical models for landfill air emissions
               Including:  Farmer Model* Shen Model, Thibodeaux a Model,
               Thibodeaux Logrithnric Gradient Model, RTI Closed Landfill
               Model, Thibodeaux Convective "Add On* Model, Thibodeaux b
               Model, Thibodeaux Exact Model, Arnold's Open Landfill Model,
             .  Shen's Open Landfill Model, and RTI Open Landfill Model.


LXXII.     Saeger, M.L., and E.E. Rlckman, Jr. Final Report: Ambient Air
          Monitoring at Hazardous Waste Treatment, Storage, and Disposal
          Facilities, Phase I. EPA Project Number 68-02-4326, U.S.
          Environmental Protection Agency, Emissions Standards and Engineering
          Division, Office of Air Quality Planning and Standards, Research
          Triangle Park, NC, 1968.

               The authors contacted several state and federal regulatory
               agencies and treatment, storage and disposal facilities (TSOFs)
               to determine the air monitoring requirements for TSOFs.  Few of
               the states contacted had established requirements, and there
               was no consistency among states.  The program was of a limited
               scope and was Intended to determine if a study should be
               conducted to assess the feasibility of conducting air
               monitoring programs at TSOFs.


LXXIII.    Schmidt, C.E., and D.L. Gordy.  Designing Air Monitoring Programs
          for Remediation at Hazardous Waste Sites.  Presented at the Annual
          Conference of the Air Pollution Control Association, San Francisco,
          CA, 1985.  14 pp.

               This paper presents a methodology for developing site-specific
               air monitoring programs for hazardous waste site remediation.
               The Information presented focuses on fugitive gas phase air
               contaminants.


LXXIY.     Schmidt, C.E., and J.K. Meyer-Schmidt. Assessment, Monitoring, and
          Modeling From a Superfund Site Remedial Action.  Presented at the
          Air Pollution Control Association Annual Conference, San Francisco,
          CA, 1985.  20 pp.

               This paper gives a brief overview of RI/FS activities performed
               at the McColl waste site in Fullerton, California.
               Measurements included surfaces screening survey, surface flux,
               ground probes, and downhole flux.  Indirect measurements were
               performed by ambient air sampling.  Modeling was used to
               predict downwind concentrations from direct measurement.  Data
               can be used directly as a case study for both direct and
               Indirect emission estimates.  Data can also be used with
               predictive models.


LXXV.      Schmidt, C.E., R. Stephens, and G.A. Turl.  Case Study:  Control  and
          Monitoring of A1r Contaminants During Site Mitigation.  Radian
          Corporation, Sacramento, CA, 1987.  8 pp.

-------
               This article presents an RI/FS/RM case study where  an  air
               pathways assessment was performed at a hazardous waste site.
               Both baseline and disturbed site air emissions were measured
               and reported.


UCXVI.    Schmidt, C.E., and M.W. Eltgroth.  Off-Site Assessment of A1r
          Emissions from Hazardous Waste Disposal Facilities.  Reprinted from:
          Management of Uncontrolled Hazardous Waste Sites, Hazardous
          Materials Control Research Institute, Silver Spring, MD, 1983.14 pp.

             »F1eld data and modeling (tugrangian model) were used to
               estimate downwind air concentrations of contaminants using
               measured disturbed site emissions data.  Modeled data  were
               compared to measured data.


UCCVIL   Schmidt, C.E., and W.D. Balfour.  Direct Sas Emission Measurement
          Techniques and the Utilization of Emissions Data from Hazardous
          Waste Sites.  Reprinted from:  National Conference on Environmental
          Engineering Proceedings, Environmental Engineering Division, ASCE,
          1983.  8 pp.

               This article -describes direct emissions measurement techniques
               and discusses various applications of these techniques to waste
               management.  Techniques included are surface and downhole
               Isolation flux chambers, ground probes, and soil vapor
               monitoring wells.


IXXVIIL  Schmidt, C.E., Re Vandervort, and W.D. Balfour.  Technical  Approach
          and Sampling Techniques Used to Detect and Map Subsurface
          Hydrocarbon Contamination.  For presentation at the 79th Annual
          Meeting, Air Pollution Control Association, Minneapolis, MN, 1986.
          31 pp.

               This paper presents a case study where several direct  air emis-
               sions measurement techniques were used to detect emissions  from
               a gasoline plume on groundwater about 90 feet below the land
               surface.


LXXIX.    Shen, T.T., and T, J. Tofflemlre.  Air Pollution Aspects of Land
          Disposal of Toxic Waste.  Ini National Conference on Hazardous
          Material Risk Assessment, Disposal and Management, Miami Beach,  FL,
          April 25-27, 1979. pp. 153-159.

               This paper provides a general discussion of the air pollution
               dangers inherent in landfill ing industrial waste, especially
               co-disposal with municipal waste.  Volatilization processes are
               discussed.  Also discussed are methods for reducing the rate of
               volatile loss.

-------
LXXX.     Shen, T.T.  Estimating Hazardous Air Emissions from Disposal Sites.
          Pollution Engineering, 13(8):31-34, 1981.

               This article presents Shen's modification of Fanner's equation
               for the diffusion of volatile* from a landfill, and Ziegler's
               modification of Arnold's equation for open dumps  (also  referred
             '  to as Shen's open dump equation).  A table of diffusion
               coefficients for selected compounds 1s given.  Two example
               calculations for PCS emission are also given.


LXXXI.    Shen, T.T., and G.H. Sewell.  Air Pollution Problems of Uncontrolled
          Hazardous Waste Sites. Civil Engineering for Practicing and Design
          Engineers, 3(3):241-252, 1984.

               Shen and Sewell provide three theoretical equations for
               predicting emission rates (Shen landfill model, Shen landfarm
               model, Shen lagoon model) for volatlles and one for dust.  The
               equations require some field data about soil, waste
               characteristics, site size, etc., but no direct or indirect
               measurement of emissions.  The equations should be used with
               caution, and some spedesspecifie coefficients may be difficult
               to determine or infer.

LXXXII.   Shen, T.T.  Air Pollution Assessment of Toxic Emissions from
          Hazardous Waste Lagoons and Landfills.  Environment International,
          ll(l):71-76, 1985.

               Shen briefly discusses the available methods for determining
               emission rates and their drawbacks.  He does not provide
               technical detail for the use of methods, but includes isolation
               flux chamber, transect, concentration profile, and predictive
               models.


LXXXIII.  Shen, T.T.  A1r Quality Assessment for Land Disposal of Industrial
          Wastes.  Environmental Management, 6(4):297-305, 1982.

               Shen presents a models for predicting emission rates from land-
               fills (Shen modification of Farmer), dumps (Arnold Model) and
               lagoons based on the diffusion theory.  Shen also presents data
               on the comparison of predicted versus measured ambient concen-
               tration of PCBs at a Niw York landfill.  This article may be
               useful as a case study.


LXXXIV.   Shen, T.T. Hazardous Air Emissions from Industrial Waste Treatment
          Facilities.  In:  Industrial Waste:  Proceedings of the Fourteenth
          Mid-Atlantic Conference, June 27-29, 1982, J.E. Alleman and J.T.
          Kavanagh, eds., Ann Arbor Science, Ann Arbor, MI, 1982.  pp. 361-
          372,,

               Shen presents predictive models for dust and volatile organics
               emissions from lagoons (Shen Model), and discusses the  fate  of
               volatile emissions in the environment in general terms.

-------
LXXXV.    Sherman, S., W. Dickens, and H. Cele. Analysis Methods and Quality
          Assurance Documentation of Certain Volatile Organic Compounds  at
          Lower Detection Limits, In: Superfund '87: Proceedings  of the 8th
          National Conference, The Hazardous Control Research Institute,
          Washington, D.C., November 16-18, 1987.  pp. 280-283.

               This paper 1s not useful.  It discusses modification of EPA
               Method 624 to allow lower detection limits for groundwater
               analysis.


LXXXVI.   Skipa, K.J., D.F. Ellas, and J.O. Gram.  Monitoring and Evaluating
          Multiple Source Emissions at Hazardous Waste Sites.  Presented at
          the 78th Annual Meeting of the Air Pollution Control Association,
          Detroit, MI, 1985.  11 pp.

               This paper discusses, in a generic form, the issues to be con-
               sidered 1n selecting monitoring and modeling approaches to
               assess the air impacts from a hazardous waste site.


LXXXVII.  Smith, P.G., and S.L. Jensen. Assessing the Validity of Field
          Screening of Soil Samples for Preliminary Determination of
          Hydrocarbon Contamination. In: Superfund '87: Proceedings  of  the
          8th National Conference, The Hazardous Control Research Institute,
          Washington, O.C., November 16-18, 1987.  pp. 101-103.

               This paper compares results from vapor screening of samples in
               tht field using portable analyzers" (FID and PID) to results
               frera laboratory TPH analyses.  Comparison of the results  shewed
               peer comlation indicating the field screening of vapors
               should not be used as the sole criteria for selecting samples
               for analysis.


LXXXVIII. South Coast Air Quality Management District.  Landfill Gas
          Emissions: Report of the Task Force, El Monte, CA, 1982.

               This report describes task force activities, including sampling
               of emissions at several landfills in the South Coast Air
               Quality Management District.  Data presented are very limited;
               sampling methods were headspace over landfill, shallow ground
               probe, and vent sampling.  Site-specific sampling results could
               possibly be worked up as a case study.
LXXXIX.   Springer, C., K.T. Valsaraj, and L.J. Thibodeaux.   In Situ  Methods
          to Control Emissions from Surface  Impoundments and  Landfills.
          Journal of the Air Pollution Control Association, 36(12):1371-1374,
          1986.

-------
               This article discusses various control measures to  reduce  air
               emissions  from  landfills and lagoons.  Included are  air-
               supported  structures, floating objects, shape modification,
               aerodynamic redesign, of! and surfactant covers,  and synthetic
               membrane covers.  The effectiveness and other considerations
               for use of each method are discussed in general.
              •*


XC.       St. Clalr, A.E., and K.T. Ajraera. Remedial Action at Uncontrolled
;,          Hazardous Waste Sites. Environmental Progress, 3(3): 188-193,  1984.

               This paper describes the feasibility study approach  used to
               develop and select  a cost-effective remedial action  alternative
               for the McColl  Superfund site 1n Fullerton, California.  Data
               from the McColl remedial Investigation, not presented  in this
               paper, can be used  as a case study of baseline emission rate
               determination.


XCI.      Thibodeaux, L.J., and S.T. Hwang.  Landfarming of Petroleum Wastes -
          - Modeling the  A1r Emission Problem.  Environmental Progress,
          l(l):42-46, 1982.

               This article reviews volatilization from land fanning of
               petroleum  wastes, discusses distribution of oil waste  in the
               soil, and  presents  a gradient!ess diffusion model for
               estimating emissions (Thibodeaux-Hwang Model).  The  article
               also gives predicted versus measured emission rates  for
               Oeldrin.


XCII.     Thomas, R.G.  Volatilization from Soil.  In: Handbook  of  Chemical
          Property Estimation  Methods, W.J. Lyraan, W.F. Reehl, and  D.H. Rosen-
          blatt, eds. McGraw-Hill, New York, NY, 1982. pp. 16.1  -  16.50.

               This chapter presents a discussion of the theory  of  volatiliza-
               tion of organics from soil.  Thomas presents five models for
               calculating volatile loss from the soil.  The models  are:
               Hartley Model;  Hamaker Model; Meyer, Letey, and Fanner Model;
               Jury, Grover, Spencer, and Fanner Model; and Dow  Methods.  This
               work 1s based on pesticides applied to soiT.  A decision tree
               Indicating which models apply for varying conditions  is
               included.  The  models could be applied to sites with  known
               contaminants.


XCIII.    Thomas, R.G.  Volatilization from Water.  In:  Handbook  of Chemical
          Property Estimation  Methods:  Environmental Behavior of  Organic Com-
          pounds, W.J. Lyman,  W.F. Reehl, and O.H. Rosenblatt, eds., McGraw-
          Hill,  New York, NY,  1982.  pp. 15-1 to 15-34.

              "Chapter describes the volatilization process from water.
               Various methods for estimating volatilization rates  are
               discussed.  The Mackay and Leinonen model and others  are
               presented.

-------
XCIV.     Thorntloe, S.  Summary of Reports Prepared for the  Development  of
          Air Emission Stindirds fer Hazardous Wasti Treatment,  Storage,  and
          Disposal Facilities* U-.S. Environmental Protection  Agency,  Office of
          A1r Quality Planning and Standards, Emission Standards and
          Engineering Division, Chemicals and Petroleum Branch,  Research
          Triangle Park, NC, 1987.

               This report contains annotated bibliographies  and contacts for
               several studies and reports.  Most are directly applicable to
               treatment, storage, and disposal facilities, but  some  may
              •provide sampling techniques applicable to uncontrolled sites.


XCV.      Tucker, W.A., and L.H. Nelken.  Diffusion Coefficients  in Air and
          Hater.  In?  Handbook of Chemical Property Estimation  Methods:
          Environmental Behavior of Organic Compounds, W.J. Lyman, W.F. Reehl,
          and D.H. Rosenblatt, eds. McGraw-Hill, NY, 1982.  pp.  17-1  to 17-25.

               Chapter discusses molecular diffusion in air and  water.
               Methods are not useful for estimating dispersion  in air.
               Methods may be relevant to uncontrolled sites  (i.e., within
               lagoon or landfill), however, factors such as  wind mixing
               (lagoons) or soil gas flow (landfills) may outweigh molecular
               diffusion.


 XCVI.    U.S. Environmental Protection Agency. Letter from J.R.  Farmer,
          Director, Emission Standards and Engineering Division,  to D. Kolar,
          Environmental Counsel9 Browning Ferris Industries,  dated October 30,
          1987, regarding EPA's investigation of techniques   for controlling
          air emissions from municipal landfill facilities.

               The U.S. EPA sent a questionnaire to Browning  Ferris  Industries
               requesting Information on their landfills.  The results, des-
               cribed in this memorandum, do not provide information  for  base-
               line emissions.


XCVII.    U.S. Environmental Protection Agency.  Material on  RCRA Facility
          Investigation Guidance (RFI) provided by the U.S. Environmental
          Protection Agency, Office of Air Quality Planning and  Standards,  to
          Radian Corporation, Sacramento, CA, December 1987=

               The material provided includes a Section 12 on Air, Case Study
               14: Use of Air Monitoring Data and Dispersion  Modeling to
               Determine Contaminant Concentrations Down-Wind of a Land
               Disposal Facility, and Case Study 15: Use of Meteorological
              (Data to Design an Air Monitoring Network.  This material
              "provides a recommended strategy for characterizing releases  to
               the air.  Air monitoring and modeling are discussed in general
               form.  The Field Methods section provides considerable
               information on available sampling media and appropriate soecies
               applications.

-------
XCVIII.   U.S. Environmental Protection Agency.  RCRA Facility Investigation
          (RFI) Guidance, Volume III, Section 12, A1r and Surface Water
          Releases.  OSWER Directive 9502.00-6C, Office of Solid Waste,  1987.

             '  Provide recommended strategy for characterizing releases  to
               air, which Includes characterizing the source and the
               environmental setting for removal actions at RCRA facilities.


XCIX.     U.S. Environmental Protection Agency.  Superfund Exposure Assessment
          Manual.  OSWER Directive 9285.5-1, Office of Solid Waste and
          Emergency Response, Washington, O.C.

               Section 2.3 of this manual presents eight equations for
               predicting short-term and long-term emission rates for
               landfills with and without internal gas generation, lagoons,
               landfarms, spills and leaks.  Both particulates and volatile
               emissions are discussed.  Equations and discussions are given
               for determining input variables.


C.        U.S. Environmental Protection Agency.  Hazardous Waste Treatment,
          Storage and Disposal Facilities (TSDF) -- Air Emission Models.  EPA-
          450/3-87-026, Office of Air and Radiation, Office of Air Quality
          Planning and Standards, Research Triangle Park, NC, 1987.

               Analytical models are presented for estimating air emissions
               from hazardous waste treatment, storage, and disposal
               facilities (TSDF).  Air emission models have been developed for
               aerated and nonaerated surface impoundments, land treatment
               facilities, landfills, and wastepiles (RTI models).  Emission
               model predictions are compared to available field data.   This
               report also Includes emission factors for transfer, storage,
               and handling operations at TSDFs.  The models have been
               assembled into a spreadsheet that is included in this report as
               floppy diskette for use on a microcomputer.  Appendices include
               a list of  physical-chemical properties for approximately 700
               compounds and a comprehensive source list of pertinent
               literature in addition to that cited in the report.


CI.       Vandervort, R., C.E. Schmidt, and W.D. Balfour.  Surface and Subsur-
          face Gas/Vapor Monitoring Techniques Applied to Environmental
          Contamination Caused by Petroleum Products and Processing Wastes.
          Radian Corporation, Sacramento, CA.  12 pp.

               This paper discusses the application of surface and subsurface
               emission rates and vapor concentration measurements for
              ; investigating petroleum leaks and hazardous waste site
               investigations.

-------
CII.      Vaught, C.C. A Basic Programming Technique for the  Estimation  of  VOC
          Emissions from Hazardous Waste Treatment, Storage,  and  Disposal
          Facilities.  Presented at the 78th Annual Meeting of  the  Air
          Pollution Control Association, Ditreit, MI, June 16-21,  1985.   9  p.

             . This paptr describes a BASIC program for use on  microcomputer.
               The program uses unspecified predictive models to  estimate
               emission rates for surface impounds, landfills,  land treatment,
               and storage tanks.  The program includes a chemical  library  of
               required model variables which are accessed by use of CAS num-
               bers, as well as estimation programs for filling data gaps,  and
               adjustments for site- specific temperature conditions.  The
               ^program was developed in 1985 and may have been  updated.


CIII.     Vogel, G.A.  Air Emission Control at Hazardous Waste  Management
          Facilities. Journal of the Air Pollution Control Association,  35(5):
          558-566, 1985.

               This article is not related to emissions measurement.   It
               identifies a method to control toxic air emissions from tanks,
               lagoons, landfills, land treatment facilities, and waste  piles.
               Control cost information is also included.


CIV.      Walker, K.A. Air Emissions from Hazardous Waste Treatment,  Storage
          and  Disposal.  Presented at the 77th Annual Meeting of  the Air
          Pollution Control Association, San Francisco,. CA, 1984.   14 pp.

               Summarizes OSW's work which is presented in detail by Balfour,
               Wttherold, and Lewis (1984) included elsewhere is  this biblio-
               gnphy.  Volatile air emissions it TSDFs were  compared for
               measured versus predictive models.


CY.       Weston, R.F. Performance of Remedial Response Activities  at Uncon-
          trolled Hazardous Waste Sites (REMII):  Draft Remedial
          Investigation/ Feasibility Study Report for the Bruin Lagoon Site,
          Bruin  Borough, Pennsylvania.  IKS. EPA Contract No. 68-01-6939, U.S.
          Environmental Protection Agency, 1986.

               This report presents the results of the remedial investigation
               performed at a refinery waste lagoon locatei 45  miles north  of
               Pittsburg in Bruin Borough, Butler County, Pennsylvania.   Air
               pathway analysts were performed for determining  the  health and
               safety requirements for workers and nearby residences during
               excavation of the lagoon.  Also, sampling and  analysis of sub-
               surface soil gas from wells located in the lagoon  were
               performed to determine the soil gas composition, regeneration
               rates, extent of trapped soil gas within the lagoon  and to
               assist in the assessment of the potential for  the  release of
               soil gas into the atmosphere during future excavation.

-------
CVI.       Wetherold, R.G., and O.A. Oubose.  A Review of Selected Theoretical
          Models for Estimating and Describing Atmospheric Emissions from
          Waste Disposal Operations.  EPA Contract 68-03-3038, U.S.
          Environmental Protection Agency, Office of Research and Development,
          Industrial Environmental Research Laboratory, Cincinnati, OH,  1982.
          73 pp.

               This report presents models for determining emissions from
               landfills, land treatment, lagoons, waste piles, and tanks.
               Models Include both predictive models and Indirect measurement
               techniques.  Estimates of model precision and accuracy, and
               potential sources for model variables are included*  Models
               Included are:  Hartley Model; Th1 bodeaux-Hwang Model; Farmer
               Model; Smith, Bomberger, Haynes Model, Mackay and Leinonen
               Model; Thlbodeaux Concentration Profile; and Thibodeaux,
               Parker, and Heck Model.


CVII.     Wetherold, R.G., 8.M. Eklund, and T.P. Nelson. A Case Study of
          Direct Control of Emissions from a Surface Impoundment. In:
          Proceedings of the llth Annual Research Symposium on Incineration
          and Treatment of Hazardous Waste, Annual Solid Waste Research
          Symposium, U.S. Environmental Protection Agency, Cincinnati, OH,
          1985. pp. 85-92.

               Testing was performed to determine the effectiveness of an
               Inflated flexible dome enclosure in controlling VOC emissions
               from an aerated wastewater lagoon.  Effectiveness was investi-
               gated by performing a mass balance of VOCs around the system.
               The article is not directly applicable to baseline emission
               estimates.


          Wood, J.A., and M.L. Porter.  Hazardous Pollutants in Class II Land-
          fills. South Coast Air Quality Management District, El Monte, CA,
          1986.

               This report describes sampling for air toxics at several Class
               II landfills.  Landfill gas sampling was performed at 20 sites
               from vents (when present) and headspace over the site.  Ambient
               air sampling was performed at five sites'.  Air toxics were
               detected even though the species cannot be legally disposed of
               at Class II landfills.

-------
             APPENDIX  B

   CHEMICAL  AND  PHYSICAL  PROPERTIES
AFFECTING BASELINE EMISSION ESTIMATES

-------
  CHEMICAL AND PHYSICAL PROPERTIES OF THE WASTE MATERIAL AFFECTING EMISSIONS
       Property
                     Effect
Saturation Concentration
Diffusion Coefficient
Molecular Weight
Partial Pressure of
Constituents
Weight Fraction
Combination of Constituents
Concentration of Waste
Henry's Law Constant
The waste will  tend to reach equilibrium with
the soil vapor.  If sufficient waste is
present, the equilibrium concentration within
the air-filled  voids of the soil matrix will
reach saturation.   Because the rate of
emission to the atmosphere is directly
proportional to the soil vapor concentration,
the emission rate will increase as saturation
concentration increases.

Compounds with  high overall diffusion
coefficients will  be emitted at higher rates
than those with lower diffusion coefficients
via increased transport, on a relative basis.
The overall diffusion coefficient may be
comprised of diffusion through the soil-water
interface, soil-air interface, soil, water,
air, and soil vapor.

Lower molecular weight compounds typically
have higher volatilization and diffusion
coefficients.  Other compound characteristics
may predominate.  Molecular weight is used to
determine diffusion rates in some predictive
models.

High partial pressure increases the emission
rate of a species by increasing its soil
vapor concentration.

An effect similar to partial pressure, it is
used as an input to some predictive models.
Not as  important as Henry's Law constant.

This increases the complexity of the
emissions process and determines the emission
rate.   It may change over time as more
volatile species are lost.

Increasing waste concentration increases the
emission rate for dilute wastes by  increasing
the vapor pressure and, therefore, vapor
concentration.

This is used to determine diffusion
coefficients.  A high Henry's Law constant
produces a higher diffusion rate.
                                                                   (Continued)

-------
       Property
                     Effect
Porosity
Adsorption/Absorption
Properties of Soil
Soil Moisture
Wick Effect
Particle Size Distribution
Organic Content of  Soil
Microbial Activity
One of the controlling factors for diffusion
through the soil.  Emission rates typically
increase with increasing soil porosity.
Total porosity, i.e., dry soil, may represent
worst-case conditions for predictive models.
Air-filled porosity may be more a realistic
parameter for many sites.

Soil with high sorption properties will
reduce the vapor density of the sorped
compounds and, therefore, the emission rate.
The effect may be minimal where high waste
concentrations saturate the sorption sites.
The effect may be reversed causing increased
emissions.

Its effect varies.  High moisture will reduce
the air-filled porosity, with pores being
filled under worst-case conditions and,
therefore, should reduce the emission rate.
Moisture may be preferentially adsorped by
the soil, releasing volatiles and increasing
the emission rate.  Drying of soil may
increase available sorption sites.  Moisture
is required for the wick effect.

Sell moisture may draw waste constituents to
the surface through the soil pores.  This
process can increase the concentration of the
constituents at the surface and, therefore,
increase the emission rate.

This affects the total soil porosity and soil
pore continuity.  Increased soil pore
continuity increases the emission rate.  A
higher percent of fines will typically
increase particulate emissions.

High organic content will  increase the
sorptive characteristics of the  soil  and
reduce the emission rate.  High  organic
content also will increase microbial  action.

Its effect varies.   It may reduce the
emission rate  by biological reduction  of  the
waste present.   It also may  increase  the
emission rate  due to gas formation which
carries volatile species to surface.
                                                                   (Continual)

-------
       Property
                     Effect
Depth of Landfill Cover
Compaction of Landfill Cover
Ground Cover
Size of Landfill/Lagoon
Amount of Exposed Waste
Water Depth in Lagoon
Aeration of Lagoons
Temperature
Wind
Emission rates decrease with increasing depth
(thickness) of cover as the diffusion path
increases.  For an open dump or landfill, the
cover thickness is zero.

Increasing compaction reduces the soil
porosity and disrupts continuity of the soil
pores, thereby, reducing the emission rate.

Soil cover, typically vegetation, will reduce
particulate emissions by reducing the
erodability of the soil.  It also will help
hold soil moisture, which reduces the air-
filled porosity and reduces volatile
emissions.

The emission rate is directly proportional to
the size of the landfill or lagoon.

Emission will increase when waste is exposed
at the surface, both due to volatilization
and wind erosion.

Water overlying waste will act as a cover.
Diffusion through water may control the
emission rate.

Aeration increases emission of volatile and
particulates with increasing volume of air
used and/or agitation.  The effect is due to
air stripping of volatiles and bulk transport
of liquid particles.

Increasing temperature  increases the
volatilization rate for organic.species and,
therefore, the emission rate.  Increasing
temperature reduces soil moisture, increasing
air-filled porosity and the emission  rate.

Wind removes the volatilized compound
concentration  in the boundary layer over  the
site, maintaining the driving force for
volatilization.  Increasing wind speed
reduces  the boundary layer over  the site.
Wind causes turbulence  within the  boundary
layer, providing the driving force for
surface  soil/waste erosion and increasing
particulate emission rate.
                                                                   (Continued)

-------
       Property                                       Effect


Cloud Cover   •                   Increased cloud cover reduces solar heating
                                 of the surface and, therefore, the
                                 volatilization rate from surface.  It also
                                 affects wind stability.

Precipitation                    Emissions are reduced by reducing the air-
                                 filled soil porosity.  It may increase
                                 landfill emission by displacing soil vapor
                                 from soil voids.  It may increase surface
                                 water and air emissions by floating waste
                                 constituents to the surface.  Precipitation
                                 increases agitation of the lagoon surface,
                                 potentially increasing emissions, but it also
                                 increases water depth over waste in the
                                 lagoon.

Humidity                         Increasing partial pressure of water vapor in
                                 air reduces the capacity for some types of
                                 volatilized material.  It may reduce air-
                                 filled soil porosity.

Barometric Pressure              Changing barometric pressures cause bulk flow
                                 of soil vapor into/out of soil.  The overall
                                 net effect is to  increase the emission rate.
                                 The effect increases with frequency and scale
                                 of barometric changes.

-------
                 APPENDIX C

ESTIMATION OF AN OVERALL SOURCE EMISSION RATE
      USING  EMISSION  FLUX MEASUREMENTS

-------
C.O       ESTIMATION OF AN OVERALL SOURCE EMISSION RATE USING EMISSION  FLUX
          MEASUREMENTS

          This appendix provides guidance on how measured emission fluxes can
be used to estimate an overall emission rate for an area source.  The
            •*
information applies to any of the direct emission measurement technologies
which are designed to measure an emission flux at a specific point on the
emitting surface.  A statistical sampling strategy and computation techniques
are discussed, and an example calculation is provided.

C.I       Statistical Sampling Strategy

          To obtain an overall emission rate for a source, a statistically
valid sampling strategy must be developed before samples are collected.
Development of this strategy involves the following:

          0    Division of the total area into zones;

          •    Subdivision of the zones into smaller units of equivalent area;

          t    Determination of the number and location of units to be sampled
               per zone;

          t    Measurement of emission fluxes from the predetermined grid
               points; and

          •    Calculation of an overall emission rate based on the average
               flux for each zone.

          Sampling strategies can be designed to obtain data that will satisfy
different statistical  criteria.   The strategy described below provides an
estimated average emission rate within 20 percent of the true mean with 95
percent confidence.
                                      C-l

-------
C.I.I     Zones

          Based on  area  source  records  and/or  preliminary screening data (e.g.
real-time instrument survey of  surface  emissions),  subdivide the total  area
source  into zones where  heterogeneous chemical distribution  is exhibited,  i.e.
areas expected to exhibit comparable emission  rates.   The zones should  be
arranged to maximize the between-zone variability and  minimize within-zone
variability.

C.I.2     Grids

          Divide each zone into an imaginary grid with unit  areas  that  depend
on zone area size (Z) as follows:

          i    Z <500 m2, divide the zone into units with areas equal to five
               percent of the total zone area;

          •    500 m2 < Z <4,000 m2, divide  the zone area into units of 25  m2;

          •    4,000 m2 < Z <32,000 m2,  divide  the zone area  into 160 units;
          •    Z > 32,000 m2,  divide the zone area into units with area equal
               to 200 m2.

Assign a series of consecutive numbers to the units in each zone.

C.I.3     Sample Size

          The number of units to be sampled for the Kth zone (nK) is
calculated based on the area of the zone as follows:
               nK = 6 + 0-15Leaof zoneK(m2
                                      C-2

-------
C.I.4
Sample Localions
          The nK units to be sampled in each zone are selected using a table
of random numbers.  A unit is selected for measurement only once.   Emission
flux measurements are then obtained for each randomly selected unit using one
of the direct emission measurement techniques.
C.I.5
Zone Emission Rates
          After all the unit emission rates are determined, a preliminary mean
emission rate for each zone is calculated as follows:
                                                                     (Eq. C-2)
where:    EK - mean emission rate for zone K;
          n« - number of units sampled in zone K;
               measured emission rate for the ith unit  in zone K
In addition, the variance ($£) and coefficient of variation (CVK) are
calculated for each zone using the following relationships:
                                                                     (Eq. C-3)
                  CVL
                (100)(SK)
                    L
(Eq.  C-4)
                                     C-3

-------
          Before calculating an overall emission  rate  that  represents all  the
zones measured, the data for each zone must be tested  for  level  of confidence.
That is, for the calculated coefficient of variance  (CVK) of zone  K,  the zone
sample size (nK) must be equal  to or greater than the sample size  required
(n«) to estimate the overall emission rate within 20 percent of the overall
emission rate with 95 percent confidence.  The total number of samples  (nK) to
be collected for different confidence levels can  be  calculated as  follows:
                           t2  CV2
                     n,/ >   <*    K                                    (Eq.  C-5)
                              t
where a study requires 100 (l=2a) percent confidence that the  emission  rate
estimates will be within p percent of the true mean.  The parameter  ta  is the
(1-a) percentage point of a student's t-distribution with NK degrees of
freedom.  Tabulated t-values can be found in any book on standard  statistical
techniques.  Recommended values for ta are listed in Table C-l.  The sample
sizes required for 95 percent confidence and a 20 percent confidence interval
(a - 0.025, p - 20) are listed in Table C-2.

          For each zone, the calculated coefficient of variation (CVK) is used
to determine the required sample size (nK).   If NK  > nK, then  (NK - nK)
additional units must be sampled in zone K.  Additional units  are  identified
using a random numbers table.  If additional samples are required  to meet  the
specified confidence limits,  then the mean zone emission rate  (£„),  variance
  2
(S^), and of coefficient of variation (CV.,) be recalculated using  equations
C-l, C-2, and C-3.  NK samples  are  used  instead of nK in the  recalculations.

          If NK is significantly larger  than nK (i.e.,  emission rates from
units within the zone are very different), it may be most effective  to  rezone
the entire area using the preliminary measured emission rates  as a guide.  The
new zones will need to be gridded as discussed previously.
                                      C-4

-------
                 TABLE C-l.   TABULATED VALUES  OF  STUDENT'S  "t"
Degrees of
Freedom*
*
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Tabulated
"t" Value**
12.706
4.303
3.182
2.776
2.571
2.477
2.365
2.306
2.262
2.228
2.201
2.179
2.160
2.145
2.131
2.120
2.110
2.101
2.093
2.086
Degrees of
Freedom*
21
22
23
24
25
26
27
28
29
30
40
60
120
CO






Tabulated
"t" Value**
2.080
2.074
2.069
2.064
2.060
2.056
2.052
2.048
2.045
2.042
2.021
2.000
1.980







 * Degrees of freedom (df) are equal to the number of samples collected less
   one.

** Tabulated "t" values are for a two-tailed confidence interval  and a
   probability of 0.05 (the same values are applicable to a one-tailed
   confidence interval and a probability of 0.025).
                                     C-5

-------
    TABLE C-2.  TOTAL SAMPLE SIZE REQUIRED BASED ON THE PRELIMINARY SAMPLE
                COEFFICIENT OF VARIATION ESTIMATE*
           Coefficient of                         Number of Samples
        Variation - CV (%)**                   Required (NK)  per Zone K
0 -
19.2 -
21.7 -
24.1 -
26.1 -
28.1 -
29.8 -
31.6 -
33.2 -
34.7 -
36,3 -
37.7 -
1 39.0 -
40.3 -
41.6 -
42.9 -
44.0 -
45.2 -
46.3 -
47.4 -
48.5 -
49.6 -
50. 8 -
51.7 -
52.4 -
19.1
21.6
24.0
26.0
28.0
29.7
31.5
33.1
34.6
36.2
37.6
38.9
40.2
41.5
42.8
43.9
45.1
46.2
47.3
48.4
49.5
50.7
51.6
52.3
53.4
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
 * Value given is the sample size required to estimate the average emission
   rate with 95 percent confidence that the estimate will  be within 20 percent
   of the true mean.

** For CVs greater than 53.4, the same size required is greater or equal to
   CV2/100.
                                     C-6

-------
C.I.6     Overall Area Emission Rate
            i
          Once the zone emission rates have been determined within the
specified confidence limits, the overall area source mean emission rate  (E)  is
calculated as follows:
                      E -  2  WK •  EK                              (Eq. C-6)


where:    EK - mean emission rate for zone K;

          WK - fraction of the site area covered by zone K (zone area/site
               area); and

          7  - total number of zones sampled.

C.2       Example Calculations

          The following example calculations are presented for a hypothetical
site contaminated by a spill of JP-4 aviation fuel.  The total contaminated
area is 1,000 m2.  Other pertinent data and calculations are summarized in
Table C-3.

          Results from a preliminary emission survey performed with a portable
real-time analyzer  (organic vapor analyzer) were used to divide the area
source into emission zones.  The survey indicated that two zones of
contamination were  present; a small zone with a high concentration of
contaminant (250 m2) and a larger zone with low level  contamination (750 m2).
                                      C-7

-------
         TABLE C-3.  HYPOTHETICAL SITE DATA AND EXAMPLE CALCULATIONS
Site Data                             ====================

          Total  Site Area  =  1,000 nr
          Number of Zones  *  2
               of Zone  I   -  250 m2
          Area of Zone  2   «  750 m2

Zone 1
          Total  number  of  units - 20  (12.5 nr)
          Number of units  to be sampled:
                           nK = 6 + 0.15 ^area(m2)
                            J3.. = 6  + 0.15 V250 m2
 Measured Emission Flux  from  Randomly
            Selected Units	EK1 /*g/iii2«min
Unit 4
Unit 6
Unit 8
Unit 13
Unit 14
Unit IK
unit is
Unit 19
Unit 23
Unit 25
14.4
12.6
19,6
14.9
10.0
1 S. £
10 = 0
10.7
11.5
11.4
                                               -     K
                   Mean Emission Rate - EK - —  y^
                                              121.7
                                       I - 13
                                     C=8

-------
                           TABLE C-3.  (Continued)
Variance =
                                =  —i-j   f  (*JJ - nK E*
                            * = -^- [1724.5 - (9) (13. 5) 2]
                                         = 10.5
                                         = 3.2
                                                  (100) (Sr)
                Coefficient of Variation = CVK =  	-——
                                            (100) (3.2)
                                          CVi = 23.7
NK (from Table C-2) = 8
NK <  nK (8 < 9)
Zone  2
          Total  number of units - 30 (25 m2)
          Number of units to be sampled:
                            nK - 6 + 0.15 /750 m2
                                   nr = 10
                                     C-9

-------
                     TABLE C-3.   (Continued)
Measured Emission
Flux from Randomly
Selected Units
Unit 2
Unit 4
Unit 7
Unit 12
Unit 13
Unit 17
Unit 22
Unit 23
Unit 25
Unit 29

EK<
/id/in «min
0.10
0.25
0.30
0.17
0.18
0,25
0.20
0.21
0.15
0.30
Mean Emission Rate » E2 = 0.211 /tg/m2»min



Variance « S^ - j^y  |o.4829 - (10)(0.211)M
              - 4.19 x 10"3
                0.0647
Coefficient of Variation - CV, - (100)(0-0647)
                             *       0.211
                           CV., = 30.7
 NK (from Table C-2) = 12

 N  > n   (12 > 10)
                                                            (Continued)
                              C-10

-------
                            TABLE C-3.   (Continued)
        Emission Rates for 2 additional units:
        Additional Units         K., ng/mz • min
                   Unit 3       OJ
                   Unit 10      0.15
                                          —           2
        Recalculated Mean Emission Rate = E2 - 0.205 m
        Recalculated Variance «  S
2 __L_
2 ~ 12-1
                                                                  71
0.5454 - (12)(0.205)2
                                    - 3.74 x 10"3
                                 S2 = 0.0611
        Recalculated Coefficient of Variation = CV£ =
                                                CV2 = 29.8
        NK (from Table C-2) - 12

Overall Source Mean Emission Rate
             1
             '   WK'EK
            K-l
        7- 2
        _             2
Zone 1: E, - 13.5 /Kj/m  • min
        u     250 m
         1   1,000 nT
                                     C-ll
                                                                  —I
                                                                   (Continued)

-------
                           TABLE C-3.   (Continued)
Zone 2:  E^ - 0.205 /ig/ffl   • min


        «. - -^
             1,000
        E - (0.25)(13.5) + (0.75)(0.205)
        -          2
        E =  3.5 /jg/m  • min
                                   O12

-------
          To obtain an overall emission rate for the total area, the zones
were divided into units according to the criteria specified in Section C.I.2
and assigned consecutive numbers.  The number of units to be sampled in each
zone was calculated using Equation C-l.  This calculation and hypothetical
emission rates for each randomly selected unit are summarized in Table C-3.
The mean, variance and coefficient of variation were calculated for each zone
using Equations C-2, C-3, and C-4.

          For zone 1 the level of confidence check indicates that a sufficient
number of samples were collected since NK < nK  (8 < 9).   No additional  samples
were required from this zone.  The coefficient of variation (CVK)  for Zone 2
indicates that two additional samples were required (CV2 - 30.7,  NK  «  12  from
Table C-2).  The two new zones were sampled and a new mean, variance, and
coefficient of variation were calculated using the new sample size (nK =  12).
The level of confidence check then indicated that a sufficient sample size (nk
= 12) had been collected.

          The overall area mean emission rate were then calculated using
Equation C-6, as shown in Table C-3.  The final result was an overall  emission
rate of 3.5 /jg/m2«min.
                                     C-13

-------
                APPENDIX D
DESCRIPTION OF SOURCES OF MODEL INPUT DATA
                   D-l

-------
     Solid Waste Landfill Survey

     The Solid Waste Landfill Survey was initiated by the U.S. Environmental
Protection Agency in 1986.  Detailed questionnaires were sent to a total of
1,200 municipal landfills; 98 percent responded.  The information provided in
these responses was complied into a SAS data file by Development Planning and
Research Associates, Inc. (DPRA).  The information included in this
computerized data base includes:

     Operations Information
     t    Owner and operator
     •    Location
     §    Number of other municipal landfills in the vicinity
     •    Shortest distance from the property line/landfill to a residence
     •    Total area
     t    Total design capacity
     t    Total remaining capacity
     •    Landfill method
     t    Waste processing techniques
     •    Quantity of waste from transfer stations
     t    Quantity of waste from resource recovery stations
     •    Ratio of waste to cover
     •    Operational hours

     Hvdroqeological Information
     •    Terrain

     •    Soil  type beneath landfill

     t    Landfill relation to the water table

     •    Distance from bottom of landfill  to bedrock
                                     D-2

-------
•    Average permeability, porosity, and hydraulic gradient
     of uppermost aquifer

§    Average groundwater flowrati

Waste Composition Information
•    Average annual quantity of waste

•    Percent of wastes by category (including industrial waste
     characterization)

•    Percent accepting liquids

§    Acceptance of liquid solvents

•    Segregation of waste types

Design Information—for Closed. Active and Planned Units
•    Date opened
t    Date last received waste
§    Total volume
•    Total area
•    Maximum height above original grade
•    Maximum height below original grade
•    Average number of lifts
•    Average lift height
•    Liner type, thickness,  and permeability
•    Final cover material,  thickness,  and permeability
•    Type of leachate collecting system
•    Volume of leachate collected
•    Method of leachate disposal
•    Run-on/run-off systems

-------
     Monitoring  Information
     §    Gas monitoring and recovery systems
     •    A1r emissions monitoring

     Nationa> Survey of Hazardous Waste Treatment. Storage. Disposal, and
     Recycling Facilities -- The National Survey of Hazardous Waste Treatment,
Storage, Disposal, and Recycling Facilities was initiated in 1987.  The
database includes the following type of information:

     •    Permit data
     •    Commercial status of landfill
     •    Operating status
     t    Expected closure date
     •    Total capacity
     •    Surface area
     •    Liner data
     •    Quantity of hazardous waste received and generated
     •    Quantity of non-hazardous waste
     •    Remaining capacity
     •    Lab pack waste data
     •    Expected physical changes to be made
     •    Waste segregation
     t    Air pollution controls
     •    Air emission monitoring systems
     •    Cover type planned
     •    Waste restrictions
     •    Waste composition

     National Survey Of Hazardous Waste Generators --  The U.S.  Environmental
Protection Agency sent this questionaire to a number of hazardous waste
generators 1n 1987.   Information requested in the questionnaire included the
following.
                                      D-4

-------
Operational  Information
«     Ownership
 §     Total  land  area
 •     Lofigitude and  latitude of  facility

 Waste Composition Information
 t     NPDES  Permit
 §     Waste  water quantity
 §     Destination of hazardous waste
 •     Quantity of hazardous waste
 •     Management  and quantity of gaseous hazardous waste
t    Quantity and management of reactive materials
•    Quantity and management of radioactive waste
t    Waste minimization programs

Hazardous Waste Management Unit Information
•    Any present onsite

•    Terrain

•    Soil type

•    Relation to water table

•    Average permeability, porosity, and hydraulic gradient of the
     uppermost aquifer

t    Groundwater flowrate

•    Shortest distance of waste management unit to property boundary
                                D-5

-------
     Solid Waste Management Unit Information
     •    Presence of SWMU's
     «    Number receiving hazardous waste
     t    Number receiving non-hazardous waste
     •    Type of unit
     •    Annual quantity of waste received
     •    Active or closed
     •    Start-up date and completion date
     •    Total quantity waste
     •    Waste types
     t    Groundwater monitoring
     •    Releases of hazardous waste from SWMU
     •    Waste piles
     t    Drummed hazardous waste in temporary accumulation areas

     Hazardous Waste Information
     t    Types of waste
     •    Sources and processes producing waste
     •    Waste quantity generated per year
     •    Physical form of waste
     • -   Liquids
     •    Characteristics

     Industrial Subtitle D Survey

     The Industrial Subtitle D Survey was initiated by the U.S.  EPA and sent
to facilities in January 1988.  The results of the survey will  be used to aid
the U.S. Environmental Protection Agency in developing strategies for the
management of Subtitle D waste.  The information included in this data base is
listed below.
                                      D-6

-------
Facility Information
•    Location
•    Total area
•    Waste minimization
•    Wastes generated and managed
t    Physical and chemical characteristics of waste
§    Waste constituents
•    Small quantity generator hazardous waste

Active and Closed Landfill Information -- (Also impoundments, land
application units, waste piles)

•    Total area
•    Total design capacity
•    Maximum height above original grade
•    Maximum height below original grade
•    Average number of lifts
t    Average height of lifts
e    Liner type and thickness
e    Year opened
•    Years till closure
•    Amount of waste disposed of in landfill
•    Wastes present in landfill
•    Time spent adding, covering, and compacting waste
•    Waste processing
•    Landfill method
•    Ratio of waste-to-cover
t    Percent capacity used
f    Leachate collection systems
•    Volume of leachate collected
•    Management of leachate
t    Run-on/run-off systems
                                 D-7

-------
Other Waste Management Practices Information
§    Incineration:  design parameters, control devices, feed compositions
     and residual management

•    Energy Recovery:  design parameters, control devices, feed
     composition, and residual management

•    Storage/Treatment tanks:  design, process, amount treated, feed
     composition, and residual management

•    Wastewater treatment system:  design, process, waste quantity, and
     composition;

•    Recycling refuse:  types, quantity, and residual management;

•    Offsite management:  quantity;

•    Underground injection: waste quantity and composition.

Monitoring Information
•    Gas systems
•    Air emissions
t    Soils
§    Groundwater
•    Surface Water

Hvdroqeological Information
•    Terrain
•    Soil type
•    Relation to water table
•    Permeability, porosity, and hydraulic gradient of uppermost aquifer
•    Average groundwater flowrate
                                 D-8

-------
     Industrial Subtitle D Telephone Survey -- The Industrial Subtitle D
Telephone Survey was also designed to gather information which would aid the
U.S. Environmental Protection Agency in developing strategies for the
management of Subtitle D waste.  The survey began in 1986 and includes the
following information:

     •    Number of active landfills at a facility

     •    Total area

     t    Total waste quantity in 1985
     •    Remaining design capacity

     •    Number receiving offsite waste

     •    Is offsite waste household waste?

     •    Also, similar set of questions for surface impoundments, land
              ication units, and waste piles
     t    Is the facility a small quantity hazardous waste generator (SQG)?

     •    Does the facility dispose of SQG waste?

     •    Does the waste management unit receive solvents or metals?

     CHEMICAL PROPERTY DATA BASES

     EPA Chemical Properties Data Base -- This chemical  properties data base
is presented in Appendix D of a Office of Air Quality Planning and Standards
(OAQPS) report (44).  This data base was developed for the use of calculating
emission rates from hazardous waste, treatment, storage,  and disposal
operations with theoretical and empirical models.  This data base is available
                                      D-9

-------
 in LOTUS  format  on  a  floppy  disk  and  contains data on about 760 chemical
 compounds.  The  data  provided  in  this data base are:

     e    Chemical  name
            •*
     t    CAS number

     t    Molecular weight

     «    Vapor  pressure at  2S*C

     t    Solubility

     •    Henry's law constant

     •    Diffusion coefficient in water

     •    Diffusion Coefficient in air

     •    Boiling point

     •    Coefficients for the Antoine equation for estimating vapor pressure
          at temperatures other than  25*C

     •    Cancer unit risk value

     §    Allowable daily intake  in air

     •    Ratio of  biochemical  oxygen demand to chemical  oxygen demand

     The major limitations of this data base are:   (1)  the physical  property
data is not complete for all  compounds and (2)  the properties  data provided in
the data base  represent a mixture of cited values  and estimated values.
                                     D-10

-------
     National Library of Medicine Online Service =- The National Library  of
Medicine maintains i Hazardous Substances Data Bank (HSDB) that contains  data
on chemical properties.  The HSDB data bank is derived from a core set of
standard texts and augmented with Information from government documents,
technical reports, and journal literature.  The data file is maintained,
reviewed, and updated on the National Library of Medicine's (NLM) Toxicology
Data Network (TOXNET), where it is also searchable.  The data bank is
organized by chemical, with over 4100 chemical substance records contained in
the file.  Types of data available include:

     •    Substance name
     •    CAS number
     •    Boiling point
     t    Freezing point
     •    Specific gravity
     •    Critical temperature and pressure
     •    Dissociation constants
     •    Heat of combustion
     •    Heat of vaporization
     •    Octanol/water partition coefficient
     •    Solubilities
     •    Vapor density
     •    Vapor pressure
     t    Relative evaporation rate
     •    Viscosity

     The HSDB also contains data in an environmental  fate/exposure potential
file on the following:

     t    Biodegradation
     t    Abiotic degradation
     •    Bioconcentration
     t    Soil  adsorption/mobility
     •    Volatilization from water/soil
                                     D-ll

-------
     Geophysical Data Bases

     The GEMS Geoecology Data Base and the soil temperature data base,
available through the National Climatic Data Center (NCDC), can be used to
obtain geophysical data for a specific site.

     GEMS Geoecology Data Base--
     This data base contains county-level data on a variety of environmental
parameters including climatic and terrain data.  The terrain data includes
county soil types from national atlantes, together with some physical and
chemical properties of general soil classes.  The climatic parameters archived
include annual average and monthly maximum and minimum temperatures and
monthly precipitation dati.

     NCDC Soil Temperature Data Base--
     This data base is available on computer tape and provides data collected
over the period 1967-1982.  The primary source of the archived data is daily
measurements taken by the Federal Government, State governemnt, and University
sponsored Agriculture Research and Experimentation station network.  The data
are stored by State number, Station Index number, and Division number by daily
data, monthly extremes, and monthly average.  Each record contains the daily
measurements for a particular depth.  The major parameters that make up the
data file are:

     t    Depth of soil temperature measurement;
     •    Soil type; and
     •    Daily temperature.

     Meteorological Data Bases

     Meteorological Statistical Array (STAR) data bases and National Climatic
Data Center (NCDC) data bases may be useful sources of meteorological
information.
                                     D-12

-------
     STAR Data Base--
     Two STAR data sets were identified, the data set created for the
Industrial Source Characterization - Long-term (ISCLT) model and the data set
created for the Human Exposure Modal (HEM).  The STAR data set generated for
ISCLT is available through the GEMS data management system.  This data set was
derived from NCDC data and contains data for 394 weather stations in the
continental United States.  The ISCLT STAR data set provides the following
information for each STAR site:

     •    Frequency of occurrence for different combinations of wind speed,
          stability class, and wind direction;

     •    Years of record;

     •    Annual  average mixing height;

     •    Annual  average precipitation;

     t    Annual  average temperature;  maximum and minimum temperatures;

     t    Average morning mixing height;

     •    Average afternoon mixing height;

     •    Stability classes and frequency of occurrence;

     •    Average weighted wind speed  for each  stability  class;

     •    Average weighted wind direction for each wind  speed;

     •    Average wind speed;  and

     •    Representative wind  speeds and frequency of occurrence.
                                     D-13

-------
     NCDC Data Bases--
     The NCDC archives meteorological data collected from cooperative weather
stations.  Daily and monthly average are available for:

     t    Maximum and minimum temperature;
     •    Precipitation;
     •    Snowfall and snow depth;
     •    Evaporation;
     t    Soil temperature;
     t    Relative humidity;
     •    Mixing height;
     •    Wind Speed;
     §    Wind Direction; and
     •    Pressure tendency and pressure change.
                                     D-14

-------
               APPENDIX E




DESCRIPTION OF REMOTE SENSING TECHNIQUES

-------
        REMOTE SENSING TECHNIQUES
        FOR MEASURING TRACE GASES
           IN THE AMBIENT AIR

           DRAFT TECHNICAL NOTE
              Prepared For:

              Ms.  Anne  Pope
The U. S. Environmental Protection Agency
 Non-Criteria Pollutants  Programs  Branch
                   MD-15
     Research Triangle Park, NC   27711
               Prepared By:

            Radian Corporation
             November 7,  1989
                     E-l

-------
                               TABLE OF CONTENTS
1.0       INTRODUCTION  ......................... 1=1

2.0       REMOTE SENSING TECHNIQUES	2-1
          2.1  Active Remote Sensing	 2-4
          2.2  Passive Remote Sensing	  2-16

3.0       APPLICATION OF REMOTE SENSING TECHNIQUES  ........... 3-1
          3.1  Point Source Monitoring  ...... 	 ....3-2
          3.2  Area Source Monitoring	 3-8
          3.3  Permiter Monitoring Applications 	 	  3-11
          3.4  Advantages/Disadvantages of Remote Sensing Techniques   .  3-12
                                LIST OF TABLES

2=1       Summary of Various Remote Sensing Techniques  .........2-2
                               LIST OF FIGURES

2-1       Optical Configuration for Long Path FTIR  ...........2-6

2-2       Differential bsorption Lidar Operated in A Range Resolved or
          Topographical Refectance Mode ................  2=11

2-3       Basic DIAL Lidar System Configuration ............  2-12

2-4       Differential Absorption Lidar Using a Non-Laser Source  ...  2-14

3-1       EPA Rose Infrared Spectrometer System ............. 3-4
                                     E-2

-------
1.0       INTRODUCTION

          The environmental  impacts from the release of airborne toxic
chemicals is a topic of great interest among air pollution scientists.  This
interest extends to all types of emission sources including stationary point
sources (e.g. incinerator stacks, ground water strippers) as well as area
source emissions (i.e., landfills and lagoons).  It is important that
measurement methods be developed to accurately assess the impact of airborne
chemical emissions on the environment.  Until now, traditional air sampling/
analytical techniques have been used to characterize emission impacts of
airborne toxic chemicals in  the environment.  While these techniques can
provide useful information,  very often these methods are unable to provide
adequate temporal and spatial resolution due to inherent limitations (i.e.,
physical, manpower, capital  cost, operating cost, etc.).  One major limitation
is the significant time delay typically encountered between the start of
sample collection and the reporting of analytical results.

          Significant advances have been made in recent years to develop
practical remote sensing methods for measuring trace gases in the ambient
environment.  The development of cost-effective and very powerful
microcomputer technology and reliable medium to low powered gas and semi-
conductor lasers has made a  possible the development of remote sensing
techniques which can be used to measure the emission impacts of toxic airborne
chemicals in the ambient environment.  As a result, several promising remote
sensing technologies have emerged, and ultimately may provide scientists with
useful tools to conduct environmental assessment studies.  Remote sensing
could, potentially, allow for rapid screening or in-depth studies of ambient
concentrations of airborne toxic compounds.  This has obvious implications for
the Superfund Program, where there is a need for real-time concentration data
for specific analytes for both baseline conditions and during remediation.
Such data are essential for  making on-site decisions that may affect the
health and safety of on-site workers and the surrounding community.
                                      E-3

-------
          This document briefly summarizes some of the more recent
developments in remote sensing technology as it applies to the
characterization of emission impacts both area and stationary point emission
sources.  This document discusses some of the advantages and practical
limitations of selected remote sensing technologies.   Finally, some general
comments and recommendations are provided regarding the future technological
development required to advance the development of remote sensing for
environmental applications.
                                     E-4

-------
2.0       REMOTE SENSING TECHNIQUES
                 /
          Remote sensing methods can be divided into two general categories.
These categories include active remote sensing and passive remote sensing
techniques. * Active remote sensing typically entails directing a focused beam
of energy (usually light energy from a controlled source) through the
atmosphere and then sensing the interaction of the beam of electromagnetic
energy with the constituent(s) of interest (1).  This process usually involves
sensing the amount of reflected energy returned to the sensor, through the
process of atmospheric backscatter, or by measuring the amount of molecular
absorption or neutral absorption (attenuation) that occurs when the reflected
energy beam interacts with those constituents of interest.

          Passive remote sensing involves two modes of detection (e.g.,
emission or absorbance).  In the emission mode, the target compound(s) emit
electromagnetic energy which is sensed directly.  In the absorbance mode the
target compound(s) react in a manner whereas to change the transmissivity of
natural light (e.g., direct sunlight, scenery light skylight, or thermally
emitted light).  Because uncontrolled light sources are used in conjunction
with most passive remote sensing techniques, operational periods are generally
limited to periods of sufficient natural lighting.

          There are a number of remote sensing techniques which deserve
serious consideration for environmental measurement applications.   An
extensive review of remote sensing methods has been recently compiled by
Saeger et.al. (1) for the U. S. Environmental protection agency.  A summary
listing of those remote sensing techniques which are in common use for
environmental measurement applications are given in Table 2-1.  The table
includes both active and passive remote sensing technologies, which appear
suited for measuring trace concentrations of gas or vapors from a variety of
emission sources.  A brief summary of these methods is provided.
                                      E-5

-------
                                         TABLE  2-1.    SUMMARY OF VARIOUS  REMOTE SENSIMG  TECHNIQUES
                    Technique
                                 Spectral
                                  Region
                                                           Measurement
                                         Heasurment
                                         Application
                           Effective Range
                               (oatars)
          Active Remote Sensing,

          Long Path Interferometry
            Fourier Transform IR (FTIR)
                                 IR
                                             Gases  and Vapors
            Fabry-Perot Intarferometry     UV and       Some  gasae
                                          VIS
                                                                               Area,  perimeter
                                                                     Stack
                                                             8,3  to X kro
                                                                                             Undetsrmiuad
                                                Most «pplie«tions
                                                use * telescopic
                                                traziacBittmr/
                                                receives.

                                                High resolution in
                                                UV and visible
                                                regions.  Technique
                                                has nofc been field
                                                tested.
T
0\
Long Path Lidar
  Atmospheric  Backscatter
  Lida
  Differential Absorption
  Lida (DIAL)
VIS and     Opacity, aerosol
Hear IR     cone,, aersol depth.
UV and      Atmospheric gases
IR          and organic vapors
                                                                               Area and Stack
Area, stack,  and
perimeter.
                                                                                                       O.S te 15 km
                                                                                             0.1 to 8 km.
              used
to measure Hie
aeatteeing Ceerosols
and fine dust
partieles).

DIAL sen be used  in
ranging as
coneentsatiaii modes.
Technique is useful
in eliminating
atmoapheEie
interfesenees.
Raman Scattering Lidc UV and
VIS




Gas Imaging
Backscatter/Absorptlon Gas IR
Imaging (BAGI)




Atmospheric gases
and organic vapors





Plume ranging and
detection.




Area, stack, 0.1 to 1.0 km.
particle mattar.
smoke, temperature
profiles, and
tracking atmospheric
dyes and tracers.

Stacks, area, 0.1 to O.S km
perimeter.
'



Technique is
generally limited to
cleat sky. nighttime
applieati'ins . The
technique t>££#s»
limited sensitivity.

Employee DIAL
measurement
approach, Cursently
method eannot
measure
concentration.
                                                                                                                                     (Continued)

-------
                                                TABLE 2-1.  (Continued)
T
Technique
Paaijv* Remote Sensing
Radiometers
Gaa Correlation Spectroscopy
Non-dispersive DCS
Mask correlation
Spectroscope
Inter ferometry
Spectral
Region

UV and
VIS

IR
UV and
IR
IR
Measurement Measurement
Parameter(s) Application

Opacity, velocity. Stack
and atiDoapheric
pollutants

Atmospheric Stack perimeter
pollutant*
Atmospheric Stack perimeter
Pollutants
gases and vapors Stack
Effective Range
(meters) Commits

0.1 to 1.0 km Technique* usas
wavelengths to
eliminate
atmospheric
intee£*;enc«a .



two

0.1 to 1.0 km Typically limited to
one compound
0.1 to 1.0 km Use* skylight ac
source>. Limited
daylight hours.
0.1 to 2.0 km Utilises Fourier
Transform IR
to


-------
2.1       Active Remote Sensing

          Several promising aetlve remote sensing measurement techniques have
been developed for environmental measurement applications.  Techniques which
appear promising for measuring the presence of trace levels of toxic gases and
vapors in the atmosphere include:

          •    Long Path Interferometry;
          •    Long Path Lidar; and
          •    Gas Imaging.

          Lone; Path Interferometers

          Long path (open path) interferometry can be used in both the active
and passive remote sensing modes.  As such these devices are useful in
assessing the impacts of airborne chemical emissions from a variety of
emission sources.  There are several types of interferometers used in
environmental remote sensing applications (1,2,3,4,5,6,7,8).  The most common
types are the Michelson interferometer and the Fabry-Perot interferometer.

          The Michelson interferometer has replaced most wavelength dispersive
spectrometers for applications in the infrared region.   This device uses a
sophisticated optical arrangement to split incoming radiation into two paths.
The two beams are then recombined after introducing a path difference with a
moveable mirror.  This results in the encoding of a multiplexed spectral
inteferogram.  A microcomputer is used to perform digitized encoding of the
interferogram, and a Fourier transformation of the interferogram is then
performed to produce a spectrally resolved output.  This technique results in
a substantial improvement in spectral resolution and throughput over
conventional grating spectrometers.  The Fabry-Perot interferometer is used to
obtain high resolution and throughput in the ultraviolet and visible ranges.
Circular apetures are used instead of slits to improve throughput of these
devices.
                                      E-8

-------
          Long  path  (open path)  interferometery can be used in either  the
active or passive modes.  In active remote sensing applications, the system
incorporates either  a single ended or double ended telescopic
transmitter/receiver configured  to measure the absorbance of features  of those
constituents of interest.  A coaxial telescope arrangement is used to  direct a
line of site beam of infrared  light across an area of concern.  In a double
ended configuration, the light source is mounted on a tripod or some other
mounting device at some fixed  path distance from the receiver
optics/interferometer as shown in Figure 2-1.  This arrangement can be
modified by the addition of collimator optics such that the transmitter and
receiver are housed  in the same  unit, and corner cube retroflectors are then
used to reflect the  transmitted  IR beam back to the receiving telescope.  The
latter design offers several advantages over the double ended design in that
the retrorefleetors  are much more portable and easier to align.  Both
arrangements can be  configured to permit multiple reflectance targets to be
used to extend  path  coverage over a wider area.

          Typically, long path Fourier Transform IR (FTIR) is used in the
absorbance mode to make horizontal measurements over distances of 0.01 to 2.0
km.  These devices are used to gather path averaged concentration measurements
for a variety of gases and vapors that have moderate to strong absorbance
features in the infrared region.  Because these systems employ an
interferometer  and sophisticated signal processing computer hardware and
software, they  can be programmed in either a spectral search mode or used to
selectively scan selected regions of the infrared spectrum if the absorbance
frequencies of  the compounds of  interest are known.

           Long path FTIR is capable of measuring path integrated
concentrations  in the low ppm-meter range.  The ultimate sensitivity will
depend on the width  the pollutant plume and distribution of the gas or vapor
molecules along the  directed path.  In most instances it is convenient to
relate the path integrated concentration measurements in terms of one meter
closed path cell.  System calibrations can be performed in the field by
inserting closed gas cells in  a  specially designed mount and directing the
light source through the cell.

                                      E-9

-------
                                          VAN HALL
      satinet, TELESCOPE
USHT
MWKf
               Figure  2-1.   Optical Configuration for Long Path FTIR.
                                           E-10

-------
          Long path FTIR. is quite versatile because it can be used to obtain
both qualitative and quantitative information regarding the presence of
compounds which exhibit strong absorbance features in the infrared spectrum.
Spectral libraries exist  for a large number of gases and vapors, and software
programs designed to perform spectral matching for complex matricies are
available.  This latter feature is very important if the system is being used
near emission sources which have not been previously characterized.  However,
the technique is subject  to atmospheric and chemical interferences (i.e.,
carbon dioxide, carbon monoxide, water vapor, etc.), and may not possess the
sensitivity or selectivity to accurately characterize extremely low
concentrations of gases and vapors.  Spectral subtraction and data
manipulation techniques are used to minimize the effects of atmospheric and
chemical interferences (4).  However, due to inherent limitations in the
sensor design, it may not be possible to completely eliminate atmospheric or
chemical interferencies using reference path and spectral subtraction
techniques.

          Long path FTIR  has been used to effectively characterize fugitive
emission sources from surface impoundments and landfarming operations.
Additionally, there are commercially available, open path FTIRs used in
perimeter monitoring applications.  In the spectral search mode, long path
FTIR can be used to screen for the presence of infrared active species.   Once
compounds of interest are detected, the system can then be placed in a scan
mode which will simultaneously scan for several compounds of choice.

          Long Path Lidar

          The development of laser technology has made possible the
development of long path  Light Detection and Ranging (Lidar) techniques for
environmental and atmospheric measurement applications.  Bordonali et.al. (2)
provide an excellent theoretical overview of Lidar based sensing techniques. A
number of Lidar methods have evolved.  These include atmospheric (Mie)
backscatter, Differential Absorption Lidar (DIAL), fluorescence Lidar, Raman
backscatter Lidar, wedge  Lidar, and Doppler Lidar.  These methods have been
used in the following applications:

                                      E-ll

-------
          •    Area and perimeter monitoring of airborne chemical emissions
               from area sources;

          •*   Measurement of pipeline leaks and other stationary point
               sources, and fugitive emissions;

          •    Monitoring of trace concentrations of vapors and gases which
               -participate in photochemical reactions;

          •    Measurement of atmospheric parameters (i.e., temperature and
               wind velocity);  and

          «    Validation of pollutant transport models.

          Of the Lidar methods cited,  atmospheric backscatter and DIAL have
been demonstrated to be effective in assessing impacts of airborne chemical
emissions from area and stationary point sources.  Depending on the particular
application, these types of measurement techniques offer the ability to
characterize the temporal and spatial impacts of air emission sources.

          Atmospheric Backscatter Lidar

          The most common and effective backscattering Lidar technique
involves the measurement of Mie scattering.  In this application the
transmitter/receiver are collocated.  The use of lasers permits the use of
intense monochromatic sources of light energy which are tunable and can be
emitted as continuous waves (CW) or in short duration pulses.  The latter
permits useful range information to be obtained by measuring the time interval
over which backscattered light energy is returned to the receiver (2).  When a
light beam is transmitted through the atmosphere a portion of the transmitted
light is scattered in many directions when it interacts with atmospheric
constitutents.  A portion of the transmitted light energy is returned along
the axis of directed beam by a process known as backscatter.
                                     E-12

-------
           There  are  three  scattering processes which occur.  These are Mie,
Rayleigh, and Raman scattering.  Hie scattering occurs when the parameters of
interest  (i.e., aerosols, dust, water droplets, etc.) are the same size
(diameters) as the wavelengths of the incident light.  Mie scattering is not
wavelength dependent, and untuned lasers can be used to measure aerosol
concentration, particulate  emissions, and opacity.  Sensors, which operate
based on  the measurement of Mie scattering, employ lasers which operate in the
infrared  and visible  range  (1).  Pulsed Mie backscattering sensors are able to
determine both concentration, plume opacity and ranging information.  For the
most part these types of devices have been used to characterize emissions from
power plant plumes.   In these applications, the plume measurements are
referenced to clean air Lidar returns.  These types of sensors are capable of
remotely measuring particle and aerosol concentrations at ranges of 0.1 to 15
km.  However, these devices are not as sensitive as other remote sensors which
use optical reflectors to increase throughput of light energy from the
controlled source.

          Differential Absorption Lidar (DIAU

          Differential Absorption Lidar (DIAL) relies on the wavelength
dependence of those gases of interest to obtain quantitative concentration
data for  specific analytes  of interest.  DIAL remote sensing systems typically
employ a  laser(s) tuned to  two different wavelengths.  DIAL devices employ
either a  single or double ended transmitter and receiver configurations.  In
most cases, the single ended system is preferred due to the increased
sensitivity due to path doubling when reflectance targets are employed.

          The DIAL technique involves tuning one laser to a wavelength for
which the compound of interest contains a strong spectral absorbance feature
(wavelength "on"), and the  second laser tuned to a wavelength in a region of
low absorption (wavelength  "off").  The technique of simultaneous one "on" and
one "off" sensing permits DIAL systems to be operated under a variety of
conditions and virtually free of other atmospheric interferences (aerosols,
water vapor, dust, fog, snow and water droplets).  In some system designs, it
is possible to accomplish the same effect by tuning a single laser to two

                                     E-13

-------
different wavelengths.  Thus, by comparing the attenuation in the intensity of
the two signals (e.g. differential absorption) it is possible to detect the
presence of a particular compound ©f interest.  This technique has been
particularly useful in measuring those gases and vapors which have strong
absorption features in the infrared and ultraviolet ranges.

          The DIAL technique can be configured to measure the differential
absorbance of light returned to the receiver by the distributed reflectance of
aerosols and molecules.  A modification of this technique involves using
topographical reflectors (hills, walls of buildings, etc.) to reflect the
light beam to the receiver.  If the system is configured with pulsed lasers in
the UV visible range, it is possible t© not only obtain concentration data,
bue als© obtain ranging information t© determine the location and path width
of the plume.  Typically, lasers in the infrared do not have sufficient power
to obtain effective ranging information.  Figure 2-2 illustrates the use of
DIAL sensors in both the distributed reflectance and topographical reflectance
modes.

          Figure 2-3 illustrates a DIAL system developed by TECAN REMOTE which
employs cornercube retroflectors in a topographical reflectance mode (4).  In
this configuration the corner cube reflector acts as a man-made topagraphical
reflector with extremely high reflectance efficiency.  This approach results
in greater signal reflectance than other natural or man made topographical
reflectors (i.e., walls, buildings, etc.).  It is important to note that when
topographical"reflectors are used, the resulting measurement represents a path
averaged concentration measurement over the entire path, and ranging
measurements cannot be effectively obtained.  The use of cornercube reflectors
permits lower powered lasers to be used with minimal degradation in sensor
performance.  This is extremely important from the standpoint of eye safety.
Lasers operated in the IR range above 1.4 urn. and in the UV range below 0.4 urn
can present eye safety dangers, and may restrict their use in certain
                                      E-14

-------
     A . Ring* Resolved Measurmerrt
     B • Integrated Mea*urment
                                                              B
                                                                    Topographic
                                                                     Reflector
Figure 2-2.
Differential  Absorption Lidar Operated in A Range Resolved
 or Topographical  Refectance Mode
                                      E-15

-------
                  R
     =!=~~^
     I  LmerjJ
        rv
          BS
               as
      Receiver
                                                        Cube comer
No gas
present
	^— Sign** 1
           2
Low
quantity
of gas
present
—*J	SfeuJ 1
High
quantity
of gas
present
Signal 1

Signal 2
                       BS « 8«am«plftt>r
                        M - Mirror
                      .Rj * UMT power mtartne* signal
              Figure 2-3.   Basic DIAL Lidar System Configuration.
                                     E-16

-------
instances  (i.e., populated areas, workplace area, etc.)-  The  effective  path
range over which measurements can be made varies depending  on  system
configuration and application.  Pulsed  lasers can provide effective
measurements over a range of 5 km and continuous lasers can be used  at
effective  ranges up to  1 km.

           Various types of lasers can be used to emit emission lines ranging
from 0.27  to 337 fan.  There are four classes of lasers which are in widespread
use.  They are distinguished based on the lasing medium, and include:  gas
lasers, liquid lasers,  semiconducting   lasers, and dielectric  crystals and
glasses (10).  Those  lasers which emit  laser lines in the ultraviolet and
infrared spectrum are the most popular  for use in remote sensing applications
involving  the trace measurement of gases and vapors.  One of the most
effective  lasers used in commercial DIAL applications is the C02 laser.   This
laser emits 60 emission lines between 9 and 12 /im.

           It is important to note that  differential absorption techniques can
also be used with a combination of blackbody radiation sources and narrow
bandpass filters (9)  to obtain results which are functionally  comparable to
conventional DIAL techniques which use  laser sources.  Figure  2-4 illustrates
an example of an infrared differential  absorption system developed by TECAN
REMOTE.  This double  ended system, is limited in the effective range over
which measurements can measurements can be made (-200 meters).  However, it
offers the ability to sequentially measure more than one compound using a
filter wheel and chopper arrangement to alternate between absorbing and
non-absorbing frequencies.  This represents a practical advantage over
conventional DIAL systems which are typically designed to measure only one
compound at a time.

           DIAL techniques are effective in characterizing chemical emissions
from stationary point sources, fugitive emissions from area sources.  These
types of sensors offer reasonable effective ranges over which  measurements can
be made, and are designed to minimize atmospheric interferences.  These
systems can be configured for use in mobile labs or stationary shelters.
                                     E-17

-------
              THANSMTTTER
                                               RECEIVER
                         • DIAL MONITORING PATH •
                                                                 REFLECTING
                                                                  CHOPTER
Figure 2=4.   Differential Absorption Lidar Using A  Non-Laser Source
                                    E-18

-------
This permits both  transient or semi-permanent/permanent monitoring to be
performed for both stationary and area monitoring source characterization
applications.  While  the systems are sensitive in the low to sub-part-per-
million meter range,  the usefulness of these techniques as screening tools to
assess potential health hazards depends, to a large part, on the source
strength and other emission source characteristics.  As such, these methods
may not be as sensitive as integrated point monitors in some applications.
The same also holds true for long path interferometery techniques.
Furthermore, it is important to note that it is possible to select wavelengths
where more than one compound will absorb, thus increasing the potential for
chemical interferences.  In some cases the absorption characteristics of other
atmospheric or chemical interferents are such that either positive or negative
interferences can  occur, thereby making data interpretation difficult.
Therefore, there may  be times when the use of this technique may result in the
measurement of erroneous data.  Physical obstructions (i.e., animals, man,
man-made objects,  etc.) are easily detected and the measurement output can be
flagged accordingly.

          Gas Imaging

          A sensing technique which involves the use of Backscatter/
Absorption Gas Imaging produces a visible image of invisible gas plumes on
a video monitor (1).  This technique measures the differential absorption of
backscattered radiation at the absorption wavelength of interest.
Sophisticated signal  processing techniques are used to display a real time
image of the plume on a video monitor.  A laser beam is directed to points
within the plume and  outside of the plume to measure the amount of radiation
which is differentially absorbed by the gas species of interest.  This system
has been used primarily in imaging gas plumes from stationary emission
sources, but can be used in other applications.  Carbon dioxide gas lasers are
used as a laser source to produce emission lines in the infrared range.  This
particular technique  when coupled with other sophisticated data processing
techniques can be  used to estimate plume volume, plume height, and plume
velocity.  This particular technique has not been used to measure plume
concentration.
                                     E-19

-------
2.2       Passive Remote Sensing

          Several passive remote sensing techniques have been developed te
measure atmospherie parameters (i.e., temperature, water vapor, and
atmospheric 'gases), and chemical measurement parameters.  In general these
devices have been used from satellite, aircraft and mobile ground based
platforms and for stationary stack measurement applications.  These devices
include radiometers, gas correlation spectrometers, and interferometers.  A
brief description of these types of remote sensing techniques follow.

          Radiometers

          Radiometers are n@n=dispersive measurement devices which have been
used over a wide spectral range including UV,  visible and IR regions.  These
devices are designed to measure electromagnetic radiation over a specified
spectral region (10).  A series of interference filters can be use to isolate
the spectral regions of interest.  Radiometers are low resolution devices.
The most common use of radiometers has been to measure temperature, opacity,
plume velocity, and gas concentration (in limited applications).  As such,
these devices are rather limited in the ability to selectively monitor trace
gas concentration of airborne chemicals for environmental measurement
applications.  However, they do have some application in obtaining infomration
(I.e, temperature, opacity) to track the transport of pollutants form specific
sources.

          Gas Correlation Speetroscopy

          Gas correlation spectroscopy simply measures the degree of
correlation between the spectrum of radiation emanating from the target gas
emissions compared to a known spectrum (1).  This technique is extremely
sensitive and selective, and can be used to sense passive emissions from  a
target gas over a large area.  There are two basic types of gas correlation
spectrometers: non-dispersive and dispersive gas correlation techniques.  Most
correlation spectrometers operate in the infrared spectrum.
                                     E-20

-------
          Non-dispersive  gas correlation spectrometers use a gas  cell  filled
with the component  of  interest  to mask the incoming radiation  (1).   In most
designs, the spectrometer consists of two cells  (e.g. sample and  reference).
The sample cell is  filled with  the gas constituent of interest.   The reference
cell is either evacuated  or filled with an inert gas.  The incoming  radiation
is alternated between  the two cells  (e.g., reference and sample cells).  This
is accomplished typically by chopping the incoming radiation which is  directed
through each cell.  In the absorbent mode, the incoming radiation is
selectively absorbed by the target gas of interest.  Due to differences in the
intensity of the incoming radiation and the absorption spectra of the  gas in
the sample cell, there is an attenuation in the radiant energy passing through
the sample cell.    Thus,  the difference in the modulated signal is inversely
proportional to the gas concentration ©f the target constituent in the
atmosphere.  In the emissions mode, the gas concentration is proportional to
the concentration of target gas concentration in the atmosphere.

          In the dispersive mode, the incoming radiation is di-spersed  into its
spectrum (typically using an optical grating).  The dispersed radiation is
then superimposed on a machined mask positioned at the entrance to the
detector.  The mask is machined such that the slits in the mask correspond to
the wavelength minima.  The mask is specific to the spectrum of the  target gas
of interest.  The part of the incoming spectrum which matches the mask causes
an oscillation in the  detector.  The portion of spectrum of the incoming
radiation that does not match the mask causes a low amplitude variation in the
detector signal.  The  concentration of the target compound of interest is
proportional to the difference in the detector output (max vs. min)  referenced
to the average.

          Gas filter correlation techniques and mask correlation  spectrometers
are typically designed to measure only one compound of interest.  This limits
the overall utility of the method, unless the component of interest  is an
atmospheric tracer  which  can be used to estimate the concentration of  other
compounds of interest.
                                     E-21

-------
          Interferometrv

          As previously mentioned, Fourier Transform interferometers can be
operated either as active of passive remote sensors.  In the passive mode, the
infrared emissions can be detected using long path FTIR techniques.  The
sensitivity in the emissions mode is typically on the order of a factor of lOx
less than when used in the absorbance mode.  However, in most applications
involving the measurement of gas concentrations from stationary point source
plumes at elevated temperatures,  it is necessary to determine the temperature
from the emission spectrum (7).
                                     E-22

-------
3.0       APPLICATION OF REMOTE SENSING TECHNIQUES

          Traditionally, direct source sampling techniques have been used  to
estimate the emissions from stationary point sources (i.e., boiler stacks,
flares, tail gas treatment systems, incinerators, process vents, etc.).  These
direct sampling techniques include a variety of integrated sampling and
continuous emissions monitoring devices (i.e. extractive and in-situ
monitors).  Similarly, fugitive emissions from industrial sources, chemical
processes, and various types of treatment, storage, and disposal facilities
(i.e., landfills, surface impoundments, landfarms, aeration lagoons, etc.)
have been characterized using direct integrated sampling methods as well as
arrays of continuous measurement devices.  While these devices are generally
effective in assessing air emission impacts, they can be quite costly in terms
of man-power, analytical costs, and/or routine operational costs.
Furthermore, the number of samples required to adequately characterize air
emission impacts may in some instances be cost prohibitive.

          The use of remote sensing techniques may offer a more attractive
alternative for conducting both screening and routine surveillance monitoring.
This need is particularly critical around Superfund sites which are either in
the remedial investigation or remediation stages.  Emissions of airborne toxic
chemicals from these sites pose a potential threat to both the workers on
site, and to local citizens who live and work in close proximity to these
sites.  Additionally, the chemical manufacturing and petroleum refining
industries have shown interest in the applicability of remote sensing
technology to detect routine releases of airborne chemical emissions which are
reportable under SARA Title III.  There is also interest in developing
perimeter and in-plant monitoring systems which can provide an early detection
of process malfunctions for emergency response purposes.   This section
examines some practical applications of remote sensing technology in assessing
the emission impacts for a variety of trace gas constituents.
                                     E-23

-------
          For the purpose of this discussion, the examination of remote
sensing teehniques will be limited to those which are capable of measuring
trace levels of airborne ehemicals which might be emitted from a variety of
stationary point sources and area sources.  A discussion of the practical
applications' of these technologies as well as the current limitations is
                                                      *
presented.  These applications include the following:

          «    Point source monitoring (i.e., boiler stacks, flares,
               incinerators, process vents, tailgas treatment systems, etc.);

          «    Screening of fugitive emissions from stationary area sources
               (Io«., landfarms, surface impoundments, wastewater treatment
               systems, landfills, etc.);  and

          •    Routine Perimeter Surveillance Monitoring (i.e., chemical
               plants, refineries, TSDF's, etc.).

3,1       Point Source Monitoring

          A number of remote sensing techniques have been developed which are
designed to measure the concentration of various gases emitted from stationary
point sources (6,7) . To date these methods have been evaluated from sources
including: utility boilers, flares, cement plant kilns, etc.  Both active and
passive remote sensing techniques have been used to characterize plume
temperature, opacity, particulate loading, and the concentration of combustion
gas products and other pollutants (i.e., N02,  NO,  CO,  C02, HCL, NH3, HF, S02,
and H2CO.   This  technique is amenable to measuring a variety of organic and
inorganic compounds which exhibit strong absorbance features in the infrared
spectrum.

          The most common remote measurement systems which have been used to
evaluate airborne chemical emissions from stationary point sources include:

          •    Long path FTIR operated in both the emissions, and absorbance
               modes;

                                      E-24

-------
          «    Gas filter and mask correlation spectroscopy; and

          •    Differential absorption Lidar,
            «
          The U. S. Environmental Protection Agency has conducted extensive
experiments to evaluate long path FTIR.  The remote optical sensing emissions
(ROSE) system developed by EPA  (3-8), is a good example of a mobile remote
sensing system based on the use of a commercially available Fourier Transform
infrared system.  The ROSE system consisted of a Nicolet model 7199 FTIR which
was van mounted.  The system was configured with a Dall Kirkham f/5 telescope.
A diagram of the system is shown in Figure 3-1.  This system is capable of
measuring both in an emissions measurement mode (passive), or in an absorbance
mode (active) using cross stack sampling techniques.

          In the cross stack mode, ports are installed in the stack, and a
collimated beam is aimed through the stack and reflected  back to the receiver
located in the mobile van at the base of the stack.  In the cross stack mode,
the system is capable of obtaining sensitivities on the order of 1 ppb.  In
the emissions mode the receiver telescope is use to sense the IR emissions
signatures of pollutants of interest which exit the stack.  The sensitivities
for most pollutants of interest is on the order of 1 ppm.  In the emissions
mode, the FTIR system is particularly sensitive to plume temperature.   Thus,
estimates of plume temperature are needed to correct the concentration
measurement data.  Temperature measurements are estimated based on measuring
the spectral radiance of two different C02 lines.   Of the two  methods,  the
cross stack measurement method is more sensitive and accurate than the
emission measurement mode.  However, the logistics associated with the
emission measurement mode is more convenient, and yields acceptable accuracies
(+- 20X) when compared to extractive sampling techniques.  Systems such as
this cost on the order of $200,000 to $300,000.
                                      E-25

-------
                         VANWftil
                 INFRARED
                 S16NAI.
                  FROM
                 PtUME
                                                  DETECTOR   . ,
Figure  3-1.  EPA Rose Infrared Spectrometer  System.
                           E-26

-------
          Another useful  tool which can be used to measure gas concentrations
from stationary point sources involves the use of gas filter correlation  and
mask correlation spectrometers.  These techniques have been previously
described in Section 2.   The Barringer COSPEC is an example of a mask
correlation spectrometer  (which can be used in either a passive or active or
passive mode.  The COSPEC has been commercially developed to measure either
N02 or S02.  In the active mode the system is reported to have sensitivities
of approximately 2 ppb over a range of 300 meters.

          In the passive  mode, the COSPEC system has been used in a variety of
plume tracking modes.  In the look up mode, it is possible to track plumes
from stationary sources for long distances.  The system is easy to use, it is
rugged, and can be operated by semi-skilled technicians.  In the emissions
mode, the performance of  the device is subject to drift based on the changes
in the source spectra (1). thus is limited to daylight use only.  Furthermore,
these devices are also are sensitive to changes in the plume temperature.  The
sensitivity in the passive mode is between 2-5 ppm-meters.  The cost of the
COSPEC ranges from $30,000 to $50,000 depending on options.   Gas filter
devices (which operate in a similar fashion) are also commercially available.

          Differential Absorption Lidar can be used for measuring a variety of
gas species and can be operated in an emission or absorbance measurement mode.
These devices can be used to measure UV, VIS, and IR active species.   Since
most DIAL systems employ  the use of lasers, the specific compounds of interest
which can be detected will be dependent primarily on the emission lines of the
laser.  These devices can be employed in a similar manner to the long path
FTIR, and can be used to  perform range resolved, cross stack measurements, and
or used with topographical reflectors.  Because the DIAL systems are active
remote sensing devices, these systems can be used both during the day and at
night, and have the ability to minimize atmospheric and chemical interferences
by use of differential absorption.
                                      E-27

-------
          While DIAL systems are very effective measurement tools, they do
have certain limitations.  It must be noted that the current commercially
available systems are capable of measuring only one compound at a time, and
must be manually tuned for each compound of Interest.  Furthermore, other
compounds wHich have absorption features at the designated absorption minima
of a compound of interest, can interfere (false negative)- and result in
erroneous measurement results.

          Most commercially available DIAL systems have sensitivities and
selectivities which are comparable or better than long path FTIR's.  The cost
of these systems are also generally comparable with that of FTIR's and range
from $100,000 to $200,000.  These costs are dependent on the application
(stationary vs. mobile), and types ©f features dasired with regard effective
range, data processing capabilities, and data display.

          Most remote sensing devices which measure the concentration of
                                ^
pollutants emitted from stationary sources- are housed in mobile vans or on
lightweight tripods, and can easily be move to various vantage points to
conduct measurements.  Some systems like the COSFEC device and other gas
correlation spectrometers can also be mounted in mobile platforms (i.e., vans,
airplanes, helicopters, etc.) and measurement results can be obtained while
the platform is in motion.  This is particularly desirable for tracking plumes
from a point source, identifying "hot spots", or screening for gas leaks from
pipelines, etc.).

          The applicability of using remote sensing techniques for measuring
emissions from stationary point sources will depend on a number of factors,
such as the characteristics of the source,  site characteristics, meteorology,
and the limitations associated with the design and operation of the remote
sensor.  Factors include:
                                     E-28

-------
Source Characteristics

«    Pollutant concentrations;
«    Plume width;
« ,   Plume temperature;
•    Plume exit velocity;
Site Characteristics

*    Physical Obstructions which limit field of view;

«    Objects which might interfere with plume dispersion (i.e.,
     building wave effects);

•    Accessibility of convenient vantage points to allow the set-up
     and operation of remote sensing systems;
Meteorology

•    Meteorological conditions that effect plume dispersion;


Limitations in Sensor Design

•    Variations in measurement results for those devices which
     require natural lighting;

•    Field of view of the sensor;

•    Selectivity and sensitivity;

•    Number of compounds which can be measured;
                            E-29

-------
          •    Interferences (atmospheric or chemical);

          «    Logistics required to support measurements (available power,
               need for eryogens to cool some detectors, physical space
               requirements, etc.);

          •    The need for environmental enclosures required to filter out
               vibration, and maintain constant temperature and humidity; and

          «    Overall complexity.

The methods described above have been shown to yield results comparable (+20%
t© 30%) to extractive point sampling techniques (i.e. long path FTIR,, gas
correlation, spectroseopy, and DIAL).

3.2       Area Source Monitoring

          The use of remote sensing devices to assess impacts of emissions
from stationary and other area sources (i.e., chemical process units,
wastewater treatment systems, landfarms, landfills,  waste handling facilities,
etc.) is an area of great interest.  These devices offer the following
          •    Rapid assessment of toxic airborne chemicals over a wide area;

          •    Real-time assessment temporal and spacial variation of
               emissions on ambient air quality; and

          •    Cost effective alternative to integrated point source
               monitoring.

          There are several remote sensing techniques which offer the ability"
to assess the ambient emission impacts from a variety of stationary and area
sources.  Both active and passive sensing techniques have been used.  They
include: long path interferometry, Lidar, DIAL and gas correlation

                                      E-30

-------
spectrometers.  These operation of these devices have previously been
described.

          EPA has recently evaluated several types of remote sensing
techniques fer conducting pre-remedial air investigations near waste sites
(4).   The purpose of these investigations was to assess the feasibility of
using remote sensing techniques to screen for air emissions from these sites.
In this preliminary investigation, three types of remote sensors were used.
They were long path FTIR, long path UV, and differential absorption Lidar.

          The long path FTIR system consisted of Nicolet model 730 equipped
with a double ended Cassgrain receiver and transmitter.  The energy source
consisted of a 100 watt halogen lamp and infrared globar source.  A mercury-
cadmium-telluride liquid-nitrogen cooled detector was used to operate over a
range from 700 to 1400 cm-1.  The system was equipped with a microcomputer
data system to process and display the measurement results from the sensor.
The receive.r optics and interferometer were located in a mobile van, and the
transmitter scope was tripod mounted at fixed distances from the
receiver/interferometer.  The system was operated in the absorbance mode.

          The UV system consisted of long path UV system designed and built by
the University of Denver.  The system consisted on a single ended
transmitter/receiver.  The system contained UV source (xenon arc lamp),
focusing mirror, retroreflectors, beam splitter, and photodiode array
monochromator.  The system was used in both a qualitative and
semi-quantitative mode to measure a variety of aromatic and other compounds
with absorption features in the UV range.

          The differential absorption Lidar system was manufactured by TECAN
REMOTE, and consisted of a dual C02 laser each capable of generating emission
lines over the range of 9 to 11 mircrometers.  The LASERSAFE system contained
a photoconductive, mercury-cadmium-telluride detector, and lock-in amplifier
for signal processing, and optics were housed in a single ended transmitter/
receiver design.  Retroreflectors positioned at a fixed distances from the
transmitter/receiver unit were used to reflect the infrared beams from the

                                     E-31

-------
analytical and reference lasers tuned in a one "on" and one "off"
differential absorption mode.

          Ix* EPA's limited evaluations studies,  the IR and W systems have
been used to' monitor air emissions upwind, downwind, and cross wind of waste
sites.  The purpose of these investigations was  to assess air emission impacts
of toxic airborne chemicals emitted from hazardous waste sites.  The Hazard
Ranking System (HRS) model was used to rank sites for possible inclusion on
the National Priorities List (NFL).   The use of  remote sensing offers the
potential for reducing the time and effort required to complete investigations
of this type.  A total of 21 contaminants were designated for evaluation based
on previous investigations of the test sites.

          The long path FTIR and UV systems were used in a qualitative mode to
first identify various volatile organic compounds emitted from each site, and
then the DIAL system was used to conduct quantitative assessments of those
contaminants in the DIAL library.  The long path FTIR and UV systems were also
used in a semi-quantitative mode to determine path averaged concentrations of
toxic airborne chemicals at locations positioned both upwind, downwind, and
cross wind of the designated site areas.  Because of shifting winds at the
test site, it was not always possible to position the equipment simultaneously
upwind and downwind of the designated facility.

          The results of the limited evaluation  suggest that remote sensing
can be used as an effective screening tool (at least in limited applications)
to screen for the presence of fugitive airborne  toxic emissions from area
sources.  However, it must be recognized that no comparative data was obtained
using alternative point sampling methods, and therefore no statement can be
made regarding precision and accuracy.  Other techniques which are limited to
the number and quantity of contaminants which can be monitored (gas filter
correlation or mask filter correlation devices)  may have limited applicability
in similar applications.  However, Ohese types of devices are generally more
selective.  Backscatter absorption gas imaging (BAGI) may have limited
application in identifying the presence of airborne toxic emission sources.
However, this method currently cannot yield quantitative data and lacks
sufficient sensitivity for certain types of fugitive sources.
                                      E-32

-------
           In investigations where the source of contamination is unknown,  it
 is  desirable to use techniques such as the long path FTIR and UV systems  to
 identify possible contaminants of interest.   Depending on site specific
 factors  (i.e.,  source type, source strength, the presence of background
 contamination,  meteorology, terrain considerations,  etc.),  these devices  may
 be  subject to both chemical and atmospheric  interferences which could mask the
 measurement results.   Therefore care must be taken to operate the systems  in a
 manner so as to minimize the effects of atmospheric  inferences.

 3.3        Perimeter Monitoring Applications

           Remote sensing techniques offer a  unique means  of  providing routine
 continuous surveillance along the perimeter  of  waste sites under investigation
 or along the perimeter of petroleum refining and petrochemical manufacturing
 facilities.   These measurements may be useful in detecting emissions  of toxic
 airborne chemicals emanating from stationary point or area sources.    The
 routine  surveillance of these chemicals can  provide  data  to  assess both
 community and worker exposure,  and in estimating the magnitude of  the  release
 of such  materials.   This type of information is useful  in assessing the
 effectiveness of control devices or the need to assess  the effectiveness of
 routine  maintenance procedures  aimed at minimizing the  fugitive  emissions from
 various  types of chemical processes,  waste handling  facilities,  etc.
 Furthermore,  this type of monitoring can be  used to  warn  of  the  presence of
 dangerous  releases  of acutely toxic or flammable gases  or vapors.
              ,"
           Perimeter surveillance monitoring  is  usually  performed for a limited
 number of  compounds where the sources are known.   For applications such as
 this, it  is vital that the remote sensing device be  capable  of operating under
 a variety  of conditions  (i.e.,  fog,  rain,  dust  storms,  etc.).  Furthermore, it
 is necessary that the device  be capable of measuring those constituents of
 interest both during  the day  and at night.   Thus,  active  remote  sensing
 techniques  are  far  more  suited  for this application  than  passive devices,
since these device  employ controlled light sources.   Differential  absorption
Lidar (DIAL). long  path  FTIR, and UV systems  offer the  greatest  potential for
conducting  these  types of measurements.
                                     E-33

-------
          These types of systems  (e.g. DIAL, FTIR, and UV), can be easily
adapted for perimeter monitoring  applications when operated in the absorption
mode.  These devices can be used  in single ended transmitter/ receiver
configuration in combination with strategically placed eornereube
retroreflectors.  Some systems are configured with a separate transmitter and
receiver positioned apart  (e.g.,  double ended design). In either case, these
devices can be equipped with rotating optical heads to direct a light beam to
a series of reflectance targets (e.g., multiple targeting configuration).
This reduces the number of remote sensing detectors required to conduct
continuous surveillance along the entire perimeter of a facility or site.  In
addition, vertical reflector arrays can also be mounted to provide information
about plume dispersion characteristics.  Micro-computers are used to control
the sequencing of the sensor head, and collect, process, and store path
averaged concentration measurements for one or more compounds of interest.
The remote sensing system can be wired to alarm systems, and the alarms can be
triggered when the processed signals from the sensor exceed pre-established
path-averaged concentration thresholds.

3.4       Advantages/Disadvantages of Remote Sensing Techniques

          There are several obvious advantages to using remote sensing methods
over conventional point sampling and measurement techniques to conduct area
and perimeter surveillance.  Remote sensing can yield a great deal of
information over the entire path of interest that could not be obtained
otherwise at a reasonable cost using conventional point source sampling and
measurement techniques.   Furthermore,  most remote sensing techniques can
provide virtual real-time measurements.  This provides rapid feedback of
measurements data to allow the investigator to react to changing conditions in
the field,  while avoiding the lengthy delays encountered by conventional
integrated sampling methods that depend on subsequent sophisticated analytical
steps.   By measuring over longer path distances, it is possible to provide
surveillance over larger areas.   This feature permits greater temporal and
spatial resolution to be obtained.
                                     E-34

-------
           Path averaged concentration measurements  infer that the  concen-
 tration of the constituents of interest are  uniformly distributed  over the
 entire  path.   Depending on the width of the  plume and pathlength over which
 measurements  are made,  the concentration of  pollutants along a path may not be
 uniform.   Thus,  the actual ambient concentrations measured at sensitive recep-
 tors  downwind of a source may be different.   Thus remote sensing results which
 yield path averaged measurements may not be  ideal for estimating dose exposure
 at  discreet locations.   Puff releases near the  fenceline or other  emission
 "hot  spots" may not be  identified using remote  sensing systems.

           Remote sensing methods are generally  effective for measuring a
 variety gases and volatile vapors which have  absorption features in the  infra-
 red or  ultra-violet spectrum.   However,  depending on  the type of source  and
 emission characteristics,,  current remote sensing techniques  may not possess
 sufficent  senstivity or resolution to identify  trace  concentrations  of consti-
 tuents  of  interest in extremely complex matrices.  Thus,  it  may be  necessary
 to  augment remote sensing methods with other  sampling and analytical
 techniques to confirm the identity of some gases and  vapors.

           The support systems  for some remote sensing detectors may not be
 amenable to routine continuous surveillance.  Thus,  the  operational costs may
 be  prohibitive.   This is particularly true of those systems  which may  require
 liquid  cryogens  to cool the detector in order to achieve  optimum system
 performance.

           Finally,  it must be  recognized that remote  sensing  systems are
 currently  not in widespread use.   Thus,  the capabilities  and  long term
 operational effectiveness  have yet to be determined.  While  these systems
 offer tremendous  potential for measuring trace gases  from  a variety of
 sources, more effort  is required at  this  time to assess  the-most suitable
 application of this technology.   As  with any application  of  measurement
 technology, one must  be carefully match  the capability with  overall program
 goals.  While  remote  sensing may offer the only practical  alternative  for some
 environmental  measurement  applications,  it may not be well suited for  others.
Therefore,   as  with  any  well  designed measurement program,  the  limitations of
any measurement methodology  must be  thoroughly understood  and applied
correctly  to  achieve  optimum results.
                                      E-35

-------
             APPENDIX F
PHYSICAL AND CHEMICAL PROPERTY DATA
   (From  ChemDat 7  Documentation)

-------
     Tht  following  properties  are given for each chemical (listed by name
and Chemical Abstract  Source  [CAS] number):
     •    Density
     •    Vapor pressure at 25 *C
     •    Solubility
     •    Henry's  law constant
     •    Diffusion coefficient in water
     •    Diffusion coefficient in air
     •    Boiling  point
     •    Coefficients for the Antoine equation for estimating vapor
          pressure at temperatures other than 25 'C.
To estimate vapor  pressures at temperatures other than 25 *C, the
Antoine equation coefficients are used with the following equation:
               log/jQj Vapor Pressure (mm Hg) - A -
where
     A, B, and C • the Antoine equation coefficient
               T » temperature in *C.
     Two approaches may be used to introduce a new compound and its
properties into CHEWJAT7.  First,  the data for one compound in
CHEMOAT7 may be replaced with data for the compound of interest in the
                                 F-l

-------
          Estimate the biodegradation rate constants using the following
          methodology:
               Approximate Kfnax fro* available data for %ax for eowpeynds
               of siillif structure ind/or functional groups; and
               Approximate Kj either by using the following correlation:
                         KjU/h/g) • 0.135  KQW 0.38
               where K^ « octanol -water partitioning coefficient
               or fay using the default (average)  value for KI,  which is
               * 1 L/h/g,  and then calculate Ks as:   Ks
     The following properties are given for each chemical  (listed by n&m
and Chemical Abstract Source [CAS]  number)?
     •    Density
     •    Vapor pressure at 25 *C
          Solubility
     «    Henry's law constant
     «    Diffusion coefficient in water
     «    Diffusion coefficient in air
     •    Boiling point
     •    Coefficients for the Antoine equation for estimating vapor
          pressure at temperatures other than 25 *C.
To estimate vapor pressures at temperatures other than 25 *C,  the
Antoine equation coefficients are used with the following equation:
               log HO) Vapor Pressure (mm Hg) « A - y ^ •
where
     A, B, and C • the Antoine equation coefficient
               T • temperature in *C.
     Two approaches may be used to introduce a new compound and  its
properties into CHEMOAT7.  First, the data for one compound in
CHEMOAT7 may be replaced with data for the compound of interest  in the
                                    F-2

-------
columns specified above.  With this approach, the  list of compounds  in
CHEMOAT7 remains constant at 62.  The second approach involves append"
nig the new compound and its properties to the existing list of chemi-
cals in CHEMOAT7.  All the equation!/calculations mist then be copied
from one of the existing rows via Lotus 1,2,3 into the appropriate
cells in the new row of the spreadsheet.  As mentioned above, the
inclusion in CHEMOAT7 of all or a large part of the chemicals listed
in this appendix could result in increasing the time required to exer-
cise CHEMOAT7 and could prevent its use on some microcomputers.
     The properties of interest listed above, with the exception of
the CAS number, ninic those in columns 8,  0-M,  and Q of the CHEMOAT7
spreadsheet.
                                    F-3

-------
                                                                                                                                 V
                                             oe«nv
                                                                         N urn oMSf&mm-m
 OBfQUOMMC
                           CASf
                                                                                                     Wlt-WO   VArat WfSSUm COSTKIBttS
                                                                                                     POINT    	—	—	
 M.V.
 1,1-ouMneftifnwML
 M-4UE1HAJUEA
 i,i-0inem£itM»c
                           Mtt-UM    Itt.U
 l.t-mUNIHMCBC
l-IMOUDElMtte
                          4T4
a,a,c-nuicnM.-
t.».7.i
M.»-l

              I.».
2,4-OIAt-*-*UKML
a,4-MMEWA.-l-
t,« MOUMOSTMBC
                          ion-«t-«
a MflM> 4 OUMD-fr-OMNOrHUDliM
                   ALCOHOL 110
                          ion-M-«
                                21
                               112
HMOMD-4-Or/HMEIHM. BBCOA1E      S»
2-OAjOMDMi.n. ALCOHOL        S97A-47-4
2-OUMeUTADIBC-l.a        120-8»-«
2-OUMOCOUMiAN            2Q61-i»-4
                                      IM.OO
                                      1M.U
aoi.it
iw.to
 aaa.c
1C7.M
 •7.00
m.oo
IM.St
m.oo
aot.oo

1*4.41
   1M
in.u
m.oo
1IO.W
 1M.C


10.00
1R.OO
ie.i7


1W.OO
IM.OO
240.00
 K.M
 M.iO
itO.M
                                  a
                               aooo
                                 ao   rioE-oi
                              toooo   1.1
40.«



   •


   •
 1OOO
 O.OB   l.ME«Oi
    a   a.aoE-oi

  o.a   t.HE-oa
    a
    a   7.ME-OI
aOOOO   4.HE-OT
    a   s.HC-oi
 O.H   1.0K«OI
    a   •.<
    a   i.i
   10
  o.a   •.aac-ot
                                                                           7.
                                                             a
                                                             a
                                                           i.S
            a
            a
           ao
          too

         10000
           n
         ano
            a
           to
            2
         UXD
          O.2
         10000
        •.llC-Oi
        1.10E-04
      O.OOOQ04U


        a.ME-Ot
        •.OK-OI
        7.«l£-0i
                                                                           •.4K-OI
                                                                           1.1
                                                                                      .OD£«QO O.OQEtOO   148 .0
                                                                           3.36E-OI
                                                                                                                                    Q

-------
                                          oofinv
                         CAS f
                                    M.V.
                         miUUTV H Uft OMBr.D9F.HHBI DIV.A
                         *•/»    <*4»«M»l)
                                                                                                          vAfoi ncsaME OBFicrans
trOtanaOjOHEMNQL         IMl-tt-O
                         107-07-4
                         Ut-Ot-l
KCttan-MOiu-t-ABfturaacxMC
                UMMN-*-OMOmUC AC
a-CHJtt 4 AtflNDCaii
t-OUHM-OMOBmL ALOML
                              lit
                               tl
traiMM-iaiNMNISU
t-QUM < flOmnMDBC
t-oumi Aypo-«-nMDBCa
t-CVOJMEXn.-t,M>mBM£IIM.
t-ianxuiMUN
t-mnucEimniaLi
t-tCMDUOUMNAN
                         tOV-n-«
                         IU-II-T
t-mMLvnumnauMN
                         107-1*-?
                               M
t.4-MOUM01E11UIM«DfUtMI
                         nn-n-t
t-tMMO-4-CY/MBCIHA. 0OOA1E


»-OUIHnE1RMflW0UWN
                              no
                              KM
                              42t
1M.M
 to.a
llt.0
 %^^k K
 tlt.t
ut.oo
10
 1
 t
 t
                                     MV.t
                                    uo.ao
tf7.00
ia.i4

 •.It
M0.lt
 •.II
 •.00
 tr.oo
 •4.00
M7.M
 •.ot
 •.ot
Mt.OO
Mt.M
141.00
1M.OO
ia.it
   to
yyHXB>
  100

  tooo
  MOO
   to
    t
  tooo
   to
  100
   to
    t

    t
  1000
    t
t.ME-04
LOK-flt

4.UE-OI

t.UE-OB
4.00E-O*

t.tOE-Ot
t.«E-04
t.CIC-OI
t.Ott-04
                                                                   .t
                                                                 tooo
                                                        11 .t
 ID. II
1M.M
241.00
22*.00
121.00
lU.fO
2O4.M
211.00
                                            l.tIS
                                                          •
                                                          t
      MXUVTE
       tooo
          t
      100000
      100000
        100
       20000
       20QOO
       10000
          1
        0.2
      100000
      20000O
          2
       10000
S.74E-OB
7.00E«00
t.lTE-Of
1.0K«OI
«.»4E-Ot
t.ME-04


t.UE-Ot
l.ME-Ot
1.41E-OT
l.ME-OP
B.OK-4M


•.2BE-OB


2.41E-OI
l.UE-01
I.OBC-04
•.07C-OB
i.ME-01
2.11C-00

-------
                                         oeem   v*.mesa
                                                                                                     wo*
                                                                   H Um ODNCT.OWF MOOt DW.Att HUNT
 OBMUOMUE
                        CAS|
MM.
 a-€MJMO-4-4ElHA.-4MC1tMJBa
                             4tO
                        UO-C4-4)    KM.il
                             ia?    aa.io
 a-tfiona-4-MEiHiLiEiMMMMonj     an
                                                                 10
                                                                 •0   2.HC-OI
                                                                100   2.40E-OI
                                                              HOOD   4.0MNM
                                                             100000   i.MC-0*
                                                        S
                                                        4

4-atancwtacxMO.
4-.••
                                     114
                                  no.00
                                                               100
i umnn t tnmu>ncmi(L na
4-4CMN. iBSM. AXML
4-ionuHianea.
                             m
                             at
4-*amisoB8aw. M.OMOL
4-nemijCvaacxMoc
                        100-40-»
  1M
1M.OO
  1M
la.oo
1H.OO
m.oo
m.ao
IM.iO
                                                        a

                                                        a
                                                               O.t
                                                                 t
                                                                 t
                                                                 t
                                       O.t
                                       o.a
                                      woo

                                      atoo
                                        K>
                                       100
                                      1000
                                         2
                                         2
                                       0.2
                                   1.
                                    .•IK-OS
                                    .74E-«a
                                    .74E-OX
                                    .HEr04

                                    .40E-A
                                    .1
                                   l.OK-01
, 406-0?
.?aE-oa

OK-04
.MC-0§
7K-OI
ME«OO
7JC-M

-------
                                             OB«ITV    VAP.
                                                               •OUAILirr N LAI CONST.OKRF.IMTBI  OVF.Att POINT
                          CAS|
                                                                                                                VAFOR ncssuc oooficiaas
                                                                                                                            ft
t4UMftcnm.-a-
                                IS*
t-mMJUffuuL
ACCMODE
ACCIKAOD
ACETIC AMMNUDC
                          TMV-0
ACCIOC
AQEKMTMLE
Accra, atone
Accra, i
                           Kff-fift^
AOMJCACB
                           MV-U-1
totmnttf. oicMBomuc ACB
AMMMMC OiatOUDE
ANPK AOD
     171
124-O4-4)
AlUL ALCOHOL
AU.1L OajOMDE
Aim mm. sroec
AJMA mm. STMBC ones
                           10I-M-*
MfHMMMMYACETMJienDC
                  (-4)
                                42*
                           M-M-a
                                       114.1
                                       •.00
                                      110.11
                                      ut.ai
                                       44 X>
                                       M.OV
                                       M.OB
                                        M.O
lao.M
 7C.IO
1M.OO
 M.O4
  M.I
 n.o»
 78.1O
  M.I

aor.io
14S.14
             M.I
            M.M
           1W.OO
           2M.OO
           1M.M
               •O
           in.oo

            M.14
                                 O.I
                                200
                                o.a
                                  a
                     0.7*

                     1.01
                     1.O4
                     0.7*
                     O.7*
                     0.7*
                     1.O1
                     1.11
                                               O.M
                                               O.OO
                                               0.00
                                                1.S7
          O.M
          O.M
          O.OO
                                                          O.OOS
                        1
                    lf.41
                                           o.a
                                             a
                                             a
                      7.B1C-01
                      s.ooE-oa
                      4.0K-OB
                      a.aoc-o?
                      4.0U-01
                       .TIC-OB
                       .14E-M
    M7.«
              1000
      40      IOM
    a44.a
    o.ota
     •.a   C.lTi-07
     114
       4         a
               o.a
O.OOOOUf
              •000
              O.01
                                                 ®.ooc«oo ox>oc«ao

                                                 S.41C-O1 l.a4C-01
                                                                                                         27B.O   7.72*
                                                            •.OOf  MQO.OI7
                       ,aic-oa
                       .tlC-OB
                       .nc-oi
                                                                             .41C-01
                                                                             .OOE«00
                                                 (.ME-«B
                                                 •.aoc-io
                                                 I.OOC-O7
                                                 a.
                   l.UC-01
           0.ME-OB a.ME-01
           1.14E-OI l.a4C-Ot
           a.t4E-Ol 1.24C-OI

           O.ODE«00 OXX&OO
           l.UC-Oi •.•OC-Ot
           a.aac-oi 14K-01
           S.OBE-OB t.TOE-Oa
           i.OBE-01 •.•OC-Ot
           i.B«E-oi i.aac-oi
                   iia.o

                    M.a
                    M.2
                    •!.•
                   aoa.a
                    ta.o
         .M7
         .140   1444.7U IM.BU
         .117   I210.MC a29.M4
         7.117  laiO.MC 229.0A4
                    .4   aao
                    i        273
               tm.0S4 223.664
                    M.O 2.M7SM
                    •7.0        O        &
                   i4i .a  s.aca   «•.•»   u«.
                    77.4    7.oat  laaa.st   322.47
                                                          O.07*
               100
                 a
                 a
                 2

              1000
                 2
             20000
I.04E-OB
f.OBE-11
 .17C-OB
 .ME-oa
 .aoc-oi
 .71E-01
 .•IE-OB
 .iac-ot
 .aa£-oB
 .BQE-OK
 .aae-t»
 .aiE-oa
 .MC-0?
 .ME-OB a.MEHM    MS.O
                                                                                                                               0
                                                                                                                                        O
S.14E-01 1.14C-M
®.COE«OD O.OOE«00
1.14C-01 a.«4E-01
 •7.0 1.M740B
 «-0        O
-M.«  a.823   14M.M
                                                                                                                                 2OB.4O

-------
                                              Deem   wr .ness souaurv H urn OWCT.MRF
                                                                                                       •OILMQ
                                                                                               OVF.A» ramr
                                                                                                                 VAHM
 COMMONMC
                            CAS|
            M.V.
            ACB
4MMCVQUHEXMC
          L<-*>
                            ItO-U-O
            U7.lt
             tt.lt
            Mt.ia
            Wt.12
 MfHETMOC
    . ACEUTE(-«)
JMMMOBe
 AUUDBC
 MMU1E
 •MM. OijOMDE
                           111-4*4
                           ltl-tt-4
       ftlfOOC AC»
•MOC AOD
MO-tt-7
     1*0

     107
     4-0
     Itl
     -O
KB-47-O
                           lOO-tl-t
                           100-44-7
mm. ICIHn. P»O
UU-nUPIOLACTOe
tiorao(2.2.i)-2,
tWOM.
                           C7-«7-t
                                172
            ut.n
            uo.it
             tt.io

            ITt.tt
   4t.M»
woooo.oo
   Ml .00
   10t.lt
   IttXXt
    Tt.l
   Ut.17
   44t.lt
   SM.IO
   Itt.U
   lOt.O?

   Ut.lt
   Ut.47
   140.S7
   22t.tO
   20.tO
   2S2.t2
   10t.lt
   Ut.t
   122.lt
   72.10
   1M.OO
   iM.ao
                                                                      MOO
            O.OO
            O.OO
            O.OD

            O.tt
            l.Ot
                                                S.4t
                                                l.
                                                o.t?
                                                 i, a
                                                l.M

                                                1.11
                                                l.lt
                                                0.00
                                                           O.tlt
                                                           O.Mt
                                                         747t.Mf
      1
     10
I.IOE-Ot
a.ooE-ot

    MO

   QJOI
      I
            2OOQO
             MOO

              100
  0.2
    s
 MOO
t.MC-04
l.ME-0*
4.t4£-04
t.toc-ot
7.11C-OI

t.aoE-ot
t.ME-Ot
4.t4E-OI
t.OOE-01
7.41£-0t
4.1
1.1
S.tOE-flt
                                                                                               7.1
                                                                                               t.Mf-4M
 O.OOMM
      I
                                                             0.2
                                                             214
                                                           .K-O7
                                                           ME-04
80000
 torn
 tooo

WOOD
   t
  100
t.MC-Ot    ?,0IE-M

t.iie-ot
I.ME-4M
t.UE-Ot    &.&£<*» O4X£iOO
O.OOE«00
                                                 1.10
                                               O.M7
                                                l.lt
                                 O.lt
                                 1.21
                                    t
                                  S.4
            MOOO

               M>
           tfiOOQO
              0.2
               2
        l.ME-Ot   9.&m-m 4.10E-02
        2.OU-4X   @.90E«flO O.OQ£«00
        t.iOE-Of
        4.ME-04   ?.tCE-Ot 7.HE-02
        t.2aE-or

        1.01E-OI
                                      174.0        i
                                      M4.0  -S.«§7
                                      -M.4  7.K4M  VXM,m 947.1
                                      14t,4
                           7^IOE-0§    It4.0
                                                                                                           MO.O
                                                                                                           MO.O
                                                                               207.0
                                                                                                           W.I
                                                                                                                                         0
                                                                                                                                     .o<»
                                                                                                                                        O
                                                                                                                   t.tos
t4t.2  t.OM    «W* *
                                                                               221.0

                                                                                M.t
                                                                               •12.0
                                                                                 4.0
                                                              O        Q
                                                              O        ®
                                                              o        ©
                                                                   o

                                                                   o
                             178.4  O.OB27M
                                                                    O.OQE<00    264.0   7.24C
                                                                                                                                 308.733

-------
COMNUONMC
                          CASf
                                     M.V.
oacnv
(0/MC)
                                                        VAr.
                                                                                                        uiLBa   VAIW messuRE cm nemos
                                                                •auuLnv N LAV ca4sr.oi?.miB) OVF.AIM romr	
                                                1.11
                                                o.oo
                                                l.tt
                                                          •74.7
                                                            1.4
                                                           O.M
                                                         t.OE-07
                                                             •0
                                                                        to
                                                                        10
                                                                                         .SX-01 •.ME-02
ftXJMLp)
MHNDL-1
am. ACCTATE(-«)
•jnLAonLAiE
•m.*BffVL
UnLCAMITOL
•Jltt.CBlJDSa.VE
                          141-M-t
                          104-C1-*

                          llt-M-4
UnLBC
MUMUM
unuc ACID
                           va-m-o
JMBAZOLE
ANBOM BISULFIDE
                          107-W-4
                          70-*2-i
                          M-74-4
                                       IM.tt
                                       llt.M
                                       itt.n
                                       II*.20
                                        00.1*
                                        •0.14
                                        72.11
                                        M.10
                                       1M.2*
                                       191.90
                                         U.I
                                                                       4K>
                                                                      •000
                                                                      •000
                                                                      •000
                                                                      2*00
                                                                        1O
                                                                                                             tl.l
                                                                                                            17* .O
                                                                                HX-O4    i.4l£-Oi •^•hOt    1M.O
                                                                               .OQE-Or    9.ME-01 a.n£-00    M4.0
                                                                               .10C-O4    §.*K-O1 •.Mf«0t    104.0

                                                                               .74E-01
                                                                               .74C-0»
                                                l.*7



                                                O.OO

                                                0.00
                                                 O.M
                                                 O.*0
   O.M
   O.*0
                                                O.M

                                                i.aa
   M.1C



 1C01.M

tDM.O4f

     10
    *.•
     U
    •.•
      1

4.MC-0*


   0.0*

   O.M

   1
    •M
                                                                       »ao
                           t
                         •1*0
                         •00
                         o.t

                         400
                        77000

                        14000
  2.*
  O.t
4COOO

 2QQO
xxno
taaoo
   10
    2
                                                                              .20E-OI

                                                                              nc-or
                   i.a*i-o§ t.
          .nc-oi
          .tic-oi
          .KC-Ot    t.lOE-C* •^OE-01
          >v4|pi^0t    ^^ QQC^OO OvOOC^OO
          .11C-04    O.£CE«OD OjOa&OD
                                                                          o
                                                                          o
                                                                          o
                                                                          o
                                                                         o
                                                                         o
                                                                         o
                                                                                        ®.ODE«CO O4K&00    0O.O
                                                                               ,1*1-04
                                                                               MEiOl
                                                                               .Mf-04   ®.£OE«00 O.OOE«CO   140.S
                                                                               nf-OI   <».«OE«00 OXOEiOO     4.*
                                                                                                                       O
                                                                                                                       O
                                                                                                                                         0
                                                                    O
                                                                    o
                                                                                                            -4.4   *.M*  S*O.S4C   tM.M4
                            117.7    7.47*   IMa.a*  17».77
                            12i.O  7.127   14*0.41* tlO.74t
                            148.0        O        o       O
.IIC-OI   O.OOE«a> O.OOE«00   2M.O        O
.ME-ar   O.OOE«OO O.OOE«OO   m.o  *.*E*   it».
.2K-4M
.Mf-Ot
.ME-04
.7*E-01
.ME-OI   o.aae«oo o.ooe«oo   IM.O        o
                                                                                                                                o       o
                                                                                                                             •OS  172.1M
                                                                             i.a*£-at
                                           (.OQE-OS 1.04E-01     46.
                                                                                                                   «-M2
                                                  O       O

                                                 II   9«» fee

-------
 COMNUOHMC
                                                                                                      fiOILINQ   VAKM HKS9U8E COBRFKIBOS
                                             oeenv    wf .ness aouuuunv H urn COMKI.
                                                                                                  ,Att
                                      M.V.
CHUMMC
CMUXamiH) TAW
OUMMCErALDBMC
OUMOMZnC ACB>
OUMOAMU»C(2)
OLflKMNB P€(>)
             ACM) (-«)
QUnCVdJOHEXMC
QUMDUOETW.
ALCOHOL -
auMEiMM*-) VMM. fna
Oj&aaVLUMOUHHMC
           Accrnac
           EIHM.ICIBC
           newticioc
               •QOO-ll-7
               aoa-17-o
               •7-74-4)
                           •-•1-t
                           •n-M-t
                                MO
                          •Tt-aa-c
                              KM
                           74-«7-»
                          U2-X7-4

                          01-M-7
                                410
OHNMHWBB4ZBC(-«)
OUMOPteCL
                          96-97-t
 IM.t
•M.t?
147.40
410.00
aao.oo
 M.n

10 .M
10.17
tM.TO
 ua.«
iTa.tt
1M.K7

ua.c*
14K.M
14t.M
ua.oo
U7.CO
1U.U
 uo.s
 •4.C*
                            74.»
                          lOa.H
                         iM.ao
                            7*.»
                          itt.ai
                          2*0.00
                          1K7.M
                         2000.OO

                          ta.ao
l.C*

i.n
!„!!

1.11
                                                1.11
                                                1.
                                                1.24
    na
     o
    80
l.OE-Ot

    •0
     1
     1
     1
                                                        t.WE-Ot
                                                             i
                                                            00
                                                          t4.7
                                                                       10
                                                                    10000
                   •0000
                       ao
                       ao
                       s

                     aooo
                     tioo

                     •000
                     8000
                     8000
                       a
                       «
                       t

                     8740
                      100
                       ao
                                                                      ao
                                                             a
                                                         0.017
                                                             a
                                                           OS
                              •ooo
                              8000
                               o.x
                              1000
                                 2
                             2OQOO
                             20600
                             26000
.ooE-«a   tcUE-oa 7.«E-oa
>fc^t~OS   @j.€ffl£^P OjOO^fOO
WE-O4   O.ODi«aO O.OOE«OO
,a7E-ai   O.@OE«OO onocc O.OQE«OO
                             26E-00   O.aaEflOS O.OOE«OO
»?.•
 o.o

K.O
                                                                                                        sao.o
                                                                                             10».O

                                                                                  i^Mi-oi     «i.a
                             Ott-OI

                             aac-ot   ®.ooi«oo O.OOE«OO    2w.o
                                                176.a
                                                2H.O
                                                                                                                     O
230


  O
  O
                                                                                                                       1471 .

-------
                          CAS
                   oecm   VAT,
            M.V.
                                                                         H uw co>«T.evp.MUBi
                                                                 BOn-Ma   VAIW
                                                         Mff.ABI POMT     	
                          MO-M-t
amc AOD
      rtOHMDCUNDC
n»-ot-*
77-4tt-*
147-14-*
OCMJC ACD
CMKAMNIDDC
CMiTMUDDC (-4)
CYMUIIC ACB
C7-1S-4
4*M»-«
4C1-M-C
                           iio-n-7
                           10*-M-O
                           11O-M-C
1M.*0     I.U
  M.t    O.M*
 7*.M     O.M
 7*.M     0.*7
 7*.M
                                                14*
                                               O.O>
                                                1.11
Mt.OO
 M.tt
140.il
Ui.v>
1M.1S
•M.IO
14«.U
             m.t
             m.i
             m.o
^raacxn. ACETATE
 M.O*
 uo.t
nt.»
 n.oo
 •.04
 M.01
104.11
i».at
 M.90
loo.ao
  M.t
 M.lf
142.22
                                                           O.U
                                                            m
                                          tno>
          1. 01
          I.Oi
          1.01
          l.Oi
O.M
O.OO
o.oo
O.M

O.OO

O.I*
o.«
O.M
                                  t.*
                                  Ml
                             S.SK-0*
             •o
        S.MC-1O
           O.ODt
                                                            o.i
                                                           OJOB
                                                           O.M
                                                           0.11
                                  a
                                MO
                                MO
                                                                       •0
                                                                   100000
                                                                      o.t
                                                                             .OOE-Ot
                                                            4.*
                                                        o.oooou
                                                              o
                                  t
                             100000

                              10000
                                MO
                               •000
                                  100
                                 l.tt
                                  4.*
                                                                      200
                                        .UE-OB
           117.0
1.O4E-M    M.4
           47.t
           M.f
                                                          1M.O
                              ,Mf-O7   ®.flOE«00 O40E100    3JK.O
                              ,*4E-O4
                                                                      O
                                                                   •.1*1
                                                                                                             O
                                                                                                          17*. 7
                                        1*E-0*    ®.OOE«00 O.OO&OD   4M.O
                                        •K-10
                                                                                                                               O
                                                          .40I-O8    30B.O    7.fO*
                                                                   Iff}.*    *.M1
                                                                   aO*.O    7.OM
                                                                   tM.O -O.C22**
                                                             i-0*     M.O       O
                                                                               §   1M.07
                                                                           I4IS.C   1M.1*
                                                                          IS11.01   MI.M
*.M*  14*0.?*t
    O        Q
    O        0>
                                    XV.7*
                                        O
                                        O
         f .108-O> •.M-Of   1O.O
         O.OOE«00 O4XC«00   MC.O
J0t«*   O.OQE«00 OjOOEiOO     O.O
                           -21. t
            O.O
                                                f .ME-Oi *.ME^tt     M.O   AMI   MDi.W    m.M
                                       .47E-OS   8.S1E-01 1.14E-01    Mt.O   «.2H    *1S.*7   iW.iJ
                                       .IK-OS   *.«E-Oi 7.ME-02    167.0  O.M1MA
                                                   •OE-01    *.77f-OI 7

-------
                                                oaenv
                                                                    «UMLHV H
                                                                                aoiLMQ   VAPOR WKSSJB
                                                                        09P.AIN fona     ——  •	
                             CA1|
                                         y.v.
COT
MACCTN.
•0-70-0
124-40-1
MunLcitet
              1,2
              : (-*)
              CM)
                            M-74-2
         NZB*<1,4) C-f|
                (TA0A)
0IOUM0UTMC (1.4)

                 .»>
OIOiaU1U»*C(1.2)
MOUMrwrac(i.i)
DIOHA. AIDC
oienm. EDO
                                   201     00.10
                                   000 100003.00
                             0040-0     0S4.40
                                         tM.M
                                         W.M
                                         110.0
                                         •70.00
                                          147.0
                                         147.01
                                  -7      147.O
                                         220.10
                                         127.01
                                         120.02
                             2001-00-1    207.00
                             Tt-04-0       00.00
                             107-00-2       00.O
                                          07.00
             04.04
            102.02
            100.01
            100.00
            221.00

            111.00
            110.07
            120.00
            120.00
            12S.OO
             09.04
            300.0*
             71.14
             74.14
                                                   O.02
                                                2
                                              0.2
                                           0.0001
                            07-
                           ,404-70-7
                            M2-7C-*
                            704-41-O
                            04O-00-O
                            109-W-7
                                                   t.at
                      O.07
                      1.47
                      1.01
                      1.20
                      1.20
                      O.OO

                      o.ao

                      1.17
                      1.20
                      i.n
                      1.20
O.OO
  .10
  .20

  ,0S
  .0S
  ,10
                                                   o.oo
                                                           S.20QE-11
        f
  o.aoooi
      1.0
     t.20
      i.i
        o

     0000

      204
       00
    000.1
      200
      400
     1200
      0.1
    O.004
    200.C
       4O
       40

     O.27
        7
     2.07
      400
O.OaaOO12
                        0.2
                       4110
                      11000
                        200
         .OOE-00
                       O.02
                         10

                          i
                       MOO
                                                                           000
                                                                         20000
                                                                           0.2
2700
2100
                        .00f«oi    d.OQE«ao o.oo£«ao     40.0
                        ,006-01
                        .14E-01
                        .01C-OS
                        .01C-00
                        .00E«O0    @.@Q1«00 O.OOfiOD    122.0
                                ,01£-00
         .00E-02
         .10E«00
         .ou-a
         ,101-01
         .041-00
         4*001-02    140.O  0.010   1744.1&    110.00
        I0.0QE-48    170.0 0.170007
f .00E-40 0.02&Ot    172.O        ft         &        O
        10.0aE-ar    ira.4 o.ai0i0«
                      o.o        ®         ®        o
                                                                                             I.OK«OO O4XX«00    -20.0
                                                                                            ®.3QE«ao o.oai«ao
                                                                                                    14MC-0)
                                ,001-01
                                ,101-01
                                .04C-00
                                ,2iE«ai
                                                                                            ®.OE£«00 OXEfeOO
                     S7.0
                     01.4
                    -00.0
                     00.0  0.90S
                                                7.0»    ISTi.S    222.0

                                                           9    201.0
                                                                                   .00E-00
                                                                                   tlC-01
                                                                                   ,20E-00
                        000
                         to
                      aaoao
                      08800
                                                                                   286-08
                                                                                   00E-04
                                                                                 2.04E-01
                                                                                 2.MC-O4
                                                                                 0.0QE-QB
                                                                                 0.ME-00
©.SOI«00 O.901«00      0.0
®.@af«oo O.OD&OO    210.0
@.SD£«CO O9     VaHK
d.OQE«00 O.OQE«00    112.0        O

®.@QE«OO o.aoi«ao     02.0        o
®.9ffi«00 O.OO£«OO    IK.O        ©
S. 126-01 7.2K-03    160.0        O
   O
   O
22.0
   O

   O
   O
   O
                       2.6CE-04    0.016-01 7.02E-tie    246.O   / «3«    1939.4    W2_,

-------
                                            MMsnv    VAF .ness OUMLHY
                                                                                                     •OILBM
                                                                        H LAN CflN5T.01FF.miBI OFF .AM POINT
COMUONMC
                          CAS|
                                     M.V.
MEMN. EDO *CB> OUOMDE
MEM*. nmwuiE
MEWL MOEIIO
OKML fl,»Q 4NBJNE
MEMUAC O.VODL
MMOtt i MATAUHt
MICMME
DHCMKV IC1HMC
P1»M. AUDC
OHOM. DISLLF BC
ODCML FOMMIIDE
OHCMN. HIWUIHEd.l)
OHOM. NnrnSMONE
MUEMN. PHIH«iA«
OHCMN. ILLFA1E
OHEMIL AinDE
OnCMAJBmL MMOPBOOXC
                          •24-tt-O
                          M-lt-S
                          •7-14-7

                          l*l-ll-«
                          «-!»-•
OBCMACIHILNOC
OBEMAftCHDLo2
                          101-C4-*
                          II
                          90-M-l
                          1M-4I-*
                                      IM.i
Mt.n
m.M
   m
!•§.»
tM.tf
 W.IO
I44.tt
 4i.O»
 M.SO
 n.o>
                                      ut.at
1M.10
1M.OO
1M.M
1M.OO
M4.40
 •.2O
                          101-Ct-C
                                      iio.ai
                                      aoB.oo
                                      210.24
                                      1*4.23
                                      1M.2I
                                               l.U
                                                                     100
                                                                      t
                                                                      •0
                                                                   14400
         O.«0
         LOOS
         l.W
                                               O.OO
                                                .11
                                                .M
                                                .M
 1.01
 l.M
1.07K

 l.ld
               1

               1

             400

            BOO


             W7

        O^OOU7
             O.1
                                                                    •000
                                                                      t
                                                                   anoo
                                                                     10

                                                                    •too
     0.24
0*00000000
       to
   o.om
     OM

   O.OOS1
    O.OM
   .on
       •7
  1.C2E-OB
     O.OX

   0.0037C
                                                      0. 0000623
                               too
                                 t
                          •.OOOE-11
                                 t
                                 2

                                 2
                                 2
                       .llC-Ot
                        «£-«
                       .74E-OB
                       ,40£-0»

                       .07E-OI
                       .171-07
                       .tlE-M
                       .44E-01
                       .•4E-01
                                                                                     S.07E-0* i.iac-02    aw.o
                       .S4C-4M

                        ICE-OS
                              .nc-o«
                              .7K-IO
                                                                                      . ax-os i.oc-01
                                                                                         -O1 441E-O2
.21C-O4
.tOE-OS
.CK-O7   O.OOI«00 OXHE«00
.O7E-OS   7.0BE-O1 f .OK-OI
.40E-OS   O.OOE«00 O.OOE«00
.«C«OS
.•IE-OS   LOBE-OS 2.a«e~oi
.UC-OS   S.flOE-01 1.04E-O1
.24C-4V
.•!£-«
.7K-OB   i.SSE-Oi i.KE-oa
.OU-QB
.oK-oa
.OGE-IO   O.O£«O> O.OOE«00
                            1C2.«   «.«2t   1400.07  1M.41
                             M.O   7.401   laOfott   22C.&3
                              1BI
                            2M.«   4.C22    700.S1   SI .42
                                                                                                       477.0
                                                                                                                    O
 211.S        A        O       O
 •00.0   4.U7   X2».f    -ill
  •>••        O        O       O
 •00.0   C.7M   illt      •!.«
 2*4.7
-1M.7
 101.0   7.4M   1ES4.M   840.34
 421.4     12. W    «4aS.i     273
                                                                                                       3CC.O
                                                                                                       220. 0

-------
                                                                                                                    VAPOR «ESSUW° gHEflFIClBflS
                                               oa«nv
                                                                             H LAM c»«f jBs*
 COMUOMMC
                            CAtf
                                        M.V.
 OMMfLldlME
                                   14
                            I17-«4-O
                            •1-41-4
                            •4-17-4
                            14I-4I4
HMmETHttCL-t
mm. ACETATE
CTML ACETATE KMMK
CTMLAOniATC
CTMLIGMtBC
CHM.OUMD£
crmLCMMDC
                            141-71HI

                            140-W-t
                            MO-41-4
                            7B-00-*
                            M7-12-O
                                 -?
EVMACETATE
CHMM OUMHNMM
                            74-W-l
                            107-07-4
                            1O7-11HI
                            KV-Ot-S
flMUtt OiaCOUDC
CMABCOLVCOL
CMftBC O.VOL OHCBM. E1HB1110-71-4
CMftBC OJ.VCOL laOCTHtt, fTHK»-4»-«
CTHAfiC O.VOOL MMOCTtM. CWlKMfr*
                            104-74K7
ElMAfTtO
CDM.(t) MEXANQL
cnm.-(t)-fMrn.-(i)
                            20»-44-
                       111
                    100000
                    100000
                    100000
                    100000
                    100000
                    100000
                                                                        MOO
                                                                          2
                                                                       20000
                                                                                .OE-04
                                                                                .OK-OI
                                                                                                  •.•OE-Ot    117.0  1.2XMM
                               7».4   «.121
                               172.0   7.4M
                               I1S.O
                               .ne-04
                               .occ-ot
                               .tOE-0«
                               ,44C-Ot
                               .7R-4M
                               ,nc-oi
                               ,nc-o4
         i.
i.l4E-Oi I4MC-OI
®.OQE«00 &j(JE£*CQ
           S.nE-Ot f.»Df,-OB
           S,,90t-Ot ?.nE-OI
           g „!«-«• 1.71C-OI

           S.10I-4» 7.4QE-01
                                                                                 44E-«t   f .nE-01 7.HC-OI
                                                                                 nc-oi
  •ic-or
  llE-Ot
  42E-04
  •OE-04
  17E-Oi
  iHC^OB
  TJE-Oi
  7IC-01
  OCC-Ot
  TOC-Oi
7.00E-07
                                                                                           .«E-0§ SME-M
S.4K-01 I.O4E-OI
s.aoi-ai i .40E-at

§.iQE-oi i.nf-o2
O.OOE«OO O^OE«OO
O.OQE«00 O.OQE«00

I.9K-OS 1.7K-01
i.3?e-aB 7.80E-ae
                                                                                                                                        M7.C2

                                                                                                                                       tt4.2
                    1OD.O        O
                    ltt.2  «.97S   1424.
                     12.4  «.9»«   10M.OI  tM.«l
                                                                                                              M.i   «.MO   lOM.Of
                                                                                                               77.0
                                                                                                              110.2
                                                                      7.101
                                                                      •.»7S
                                                         217. M
                                                         211.21
                              !»!.•   «.721
                               tt.S   7.0gf   ISTi.i
                              «0.0   t.080
 10.7
 14.•

117.0
2GO.O
293.0
                                                                                                                       7.121   S054.S4   217.70
                                                                                                                          O
                                                                                                             -H.O     ?.S9S
                                                                                                             1O0.7   7 SflS
                                                                                                                                      lit
                                                                                                                                       241.2
                                                                                                                                91D.A    344.1

-------
                                              oacnv    VAF,
                                                                toujULm N u»
                                                                             MILMa
                                                                     ovrAnroim
                                                                                                                               OBFICIENIS
OMIUONMC
                          CAS
            M.V.
fUlMOC ACID
fUUN
lUnSCAOD
OlffMK ACID
aWMMODE
OMNDME. Nnrnso
    l-l
    4-2
1IO-C4-1

M-tt-C

•74-M-7
                           14»-«t-f
      jCNOCttUMEXMC (QAMM
  KMMJMOCVCUVBO/IOUMI
  KACHJMQETHMC
teaUMJOUKHtM*
M3IMC(-«4
M3UMUCACH)
M3UMX-1
MM
Ml*
                                41C
                           110-C4-4
                           111-S7-*
                           ao»-ot-t
MMDCWMC ACID
MDMMENSmDE
MNMOBJfDOC
MMOKV oncnm. EHO
MMOWAOmC ACB>
tMMKncnm. ACEDLBC
MfMDmCltftt ISOJ1UPM. IC1DC
                 EIHER
      211

      277
                                       UO.M    1.4M
                                       1M.O7
                                       •.Of    O.04
                                       M.0»    l.M
                                       lit .01
                                       1M.11
                                       tt.Ot    1.2i
                                       •.10
                                       74.10
             74.01
            m.w
            100.01
            2M.IO
             M0.«
            280.M
             272.«
27C.OO

1M.O1
m.u
  M.4
 144.0
   117
 •2.01
 27.00
 20.0O
 •4.10
11O.12
 7» .00
 7* .OK
    M
 n.oo
140.00
130.00
                                                 1.11
                                                 I.K7
                                                 O.M
                                                 2.0*
                                                 l.«7

                                                 1.71
                   iM.207
                        t
                        a

                   O.OOOU
                   O.OOOM
                    o.ooo*
                       4*
                        1
                     0.1S
                 O.OOOOOM
                                                   4.0UE-OI
                                           7000    l.ME-Ok
                                                   •••4E-01
                              1. ODE-OS l^MC-Ol

                              I.22E-OI 14ME-4M
                              1.0C-OS •.72E-02
           MOOO
          •40000   a.OK-OB
                              -29. •  ».M»lOa

                               •1.4   •.•7S  1090.07   227.74
                              Ml.7   «.K7S  11M.7    M2.C
                                                                                                          281.0
                               MOOO
                              100000
                   1.1
                   1.4
                   •.UE-Of
                   •.OBE-Of
                   2.aOE-4»
                   a.02E*00
                   •.tOE-04
           O.OQE«OO O.QOE«OO    Itt.O
           O.OOE«00 OUQOEtOO     10.2
           O.ODE«OO O.OQE«QO
                                                                                                           M.4  «.MC   1204.
                                                                                                          •22.O        O
                                                                                                          21S.O  -
                                        O       O
                                          2U.C4
                                        O       O
                      O.M

                      O.M
              17


             • a

           11000
                      O.M
                      O.OO
                      O.OO
o.nts
     i
     i
     i
  14.4
   7M
   •00
 1K200
                                                                    100000

                                                                     IOOOO
                                                                      1000
                                                                    100000
                                                                        10
                                                                     10000
                                                                         2
                                                                      1000
7.1
I.

l.ME-OI

l.OK-01

•.ME-01
•.ooE-oa
1.1
•.
4.1
a.«E-04
•.aoE-ot
1.I7E-07
1.ME-O2
1.SOE-4M

2.2K-OI
7.00E-01
•.2EE-Ot
                                                             •.ME-01 f^lC-Ot    214.0  -1.001H
                                                                                1M.O        O
                                                                                        7.77E-OI a.OOE-01
                                                                                         .4QE-01
                                                                                         .t4E-0§
,IDE-OS
,ME-OS
                                       O       O

                  M.O   •.•?•   9171.87   224.41

                 1M.O   7.MO   IMl.M   gM.M
                                                                       7K-Ot
                                                                      .40E-01
.07E-01
.ME-Ot
.74C-O1
 28.7   7.82*   g«2».f    M0.4
 1».7   7.217   12M.17   271.97
-«0.2   7.014   MS.itft   280.26

-------
                                                                                                              VAPOR ncssue
 CBMUONMC
CAS I
M.V.
                nUMLXTtf H LAV COMCT.ftlRF HIHI OBP.Afll HUNT
                                          o)   (•*/•
        ALCOHOL (ISOmftAMft.)
         ACETA1E
         CIHSt
     -I
•7-tt-O
IOB-tl-4
                           W-M-O
UA9ACEUTE
IfAft ftMCIEAlE
1MB
MHIKACD
MtOC 4MMMUDC
110-M-f
mttOMJC AOD

       aUOOCAOD
                          7*41-4
IdMBCMCCTIC
ICMfL 1-fpUnUMMO) CAJWMT      M
mm. ACETA1E              7»-tO-«
tcnm. AonuiE             M-M-*
IC1HN. OUMBE             74-4T7-*
kETMN. CVOOCXA»C
MEimtEIICIl
lenra. EIMVL KETOC
                          10KC7-2
                          11S-10-4
                          7S-M-*
 n.o»
tM.M
  74 U>
 71.1ft
KD.tt
IM.tl
 •0.0*
m.u
Mt.M
m.to
  ff.t
  7».4
iM.or
  M.I
!».*>
•OO.M
 M.lt
 M.M>
 M.O4
 •.II
 4*.ll
            0.00
            71.M
           2».00
             74.1
            •.10
            •>.»
            M.iO
           24J.M
            46 M
             7X.1
                     0.7*
      ^ f^yyyy]K^
             M>
            I7O
O.OO       M.07
O.tt       0.4M
0.7»       4t.i
             7S
            UO
                                                                  110000
                                                                       t
                     O.M
              I

         o.oaoi
             •o
          o.oot

            0.1
                                                                            .4
                                                                            .00*01    1.11E-OS •.HE-Ot    900.0  -4.00014
                                                                     100
                                          MOO
                                           M
                                         moo
                                               O.7»
                                 114
         O.tt

         O.M




         O.K
             41
           MK
            too
                                                                    noo
                                                                  tooooo
                                                                       s
                                                                    aooo
                                                                   00000
                                                                    •QQO
                                                                     100
                                                                      14
                                                                   74000
                                                   47C-04
                                                  .MfeOO   7.SK-Of 1
                                                                             KC-O«
                                                                                                        SM.O
                                                                                                        1O7.0  l.OOOOM

                                                                                                         O.O        O        &
                                                                                                        flC.O        O        O
                                                                                                         •.4   «.ii7  ine.it  tit.«i
                                                                             ilE-Off
                              nc-or
                              UE-OI
                             ,iow»
                             .7K-O4
                             .aK-ot
                             .oee-o«    i,ate-a§ I^ME-OI     M.O     7.00*
                             .44E-4V
                             .14E-01    »„&£•<* ioME-Ot    -24.0
ue-os
                                                                                             8.0K-02     70. •   «. 07421

-------
                                              oaenv    wr.
CAS •
                                       M.V.
                                                                  KULINQ   VAfO* nCSSUNE COHFICIBflS
                                    N MM COCT.OV7.W1BI  OSV.AOI TOJHT	
                                                          (««/«K)(*«.C)       A    .  *        C
      MMAZBC
      IODIDE
      UOMAL IBDC
      isoount ICIBC
      DOOMNMI
74
IMMt-O
MO-1O-1
024-4»-o
HEM*. ICDMOMUV
      miAiuc
      auuac ACB
                           •o-u-o
                           M-C7-4
                           tf^L»A£_Y
WMUMMMCCnC ACS
ICMMMrniMMBC (1)
MEMNJM lUOtOUDC
               01
044-00-0
      S
       04
                           IOO-M-0
                                 -»
                           1IO-M-4
NHrNMOIIJBOC-1.4
         O.YCOL
 120-BO-T
MACM
MACMMODC
 M-47-4
 M-98-O
 M7-t»-?
14I.M
114.00
 too.t
100.10
14t.tO
141.10
lit. 10
 llt.l
 07.10
 tt.Ot
 M.OO
10B.4K
  01.0
tao.oo
 14t.S
S4».O1
toi.to
 74.00
 71.11
  ot.o
 llt.0
   •7
 W.IO
 07.11
H0.»
 Ui.t
1M.OO
KM.17
12*. IX
121.14
11D.7S
0.00
0.00
                      O.
                                  000
                                  40.0

                                  4.10
                                  1C.7
                                   0.1
                                0.0301
                                                              OS
                                                             uao
                                                 1.S4         4St
 14000
  S400
 10000

 47000




 10000
                                        .OQE«00
                                           11SM
                                             100
                                              s
                                             MO
                                              1
                                            1000
1.00

1.14
1.42

i.a
                                  O.7
                                    1
                                    1
                                    1
                                  7401
                                   X>

                                  o.n
                               0.11111

                                 o.oot
     1.
 80000
    10

  1000
flHOOD
     2
 20000
 10000
     2
                                                                               JQE-OI
           144.0        O        O       O
t.OOE-00    11S.O    0.078   1101.4    101.8
                                        g.«E-08 7.90E-OI    101.0  0.400   SOBO.g     174.4
                                        d.OC«00 0.00&OO    141 .S
                               .OOE-CS   O.OOE«00 O.OOE<00    141.0
                                                     UC4S
                                                                                                            w.o
                                                                    7.400   10SS.O
                                                                                                                                      2M.«
                                                                               .OQtOS   i.ME-OS O.OK-O1
                                                                               12&«S   8.«i-08
                                                                               OBE-CB   S.ME-OS
                            -40.0
7K-4K    0.001-08 O.IOE-«t    120.0
41E-08

•lE-Oi    O.OOE«00 O4)OE«00    IOO.O
                                                                                        o.aaE«ao O.OQE«OO     (0.4
                    7.01
                       O

                       O
                                                                                                          XSt.M
                                                                                                               o
                               Mt-Of
                               22E-OI

-------
 OOPIUONUC
                            CAS f
                                               oecnv
                                                                                                           BOILDM   VAfOR
                                                         VAP.nBS SOLUBILITY N LAM CONST.OWJMBI DIFF.AIM fOWf	
                                                                                                                                  COBRFICIBrtS
                                                                                                                                •
 MVttfMLJ>C<-«)
 ICMMLVCEMM
                              -74-4     1M.14
                                         ltt.1
                            0004-70-O 10000.00
           «)
            U
            <-»>
                            100-00-7
            IC1MA.1KIOCAMUIM1C
            NnmiSMonLAi*     0is
•-MXAfiGCAlC                 044-70-0
tHMMKftCTHAftem. CAABMMT     471
                                 011
IHCMILOUMOACETAMBE            401
N-tOflUWM. CANBAUME           472
IMCWN., N-AeMC1HILAMDC  000-00-1
                                 47*
                            111-
OCltt. ALOOHQL                111-47-0
WALK AOD                  144-4t-7
WMOC ACID                  471-47-4
0,0-OIEIHrt. 0-2HETIM.1lliaE1MN.ia0-70-O
O-OUHHTIROKNIBC         M-70-0
o-€THN. s.s-Dmem. nasnaoi7o0-40-0
o-MLvaraic KETO                 sio
0-XN.VL CMjOfUDE             SS2-4C-4
PARA1MION
                           66-3«-2
                                        u0xn
                                        110.14
    110
 10B.OO

 1M.OO
  M.OO
  100.0
 lor.so
 140.00
 U0.t*
 140.00
 M0.lt
 M0.IO
 900.00
 1M.01
 114.00
 100.00
  00.04
  M.00
 2C0.M
 1S7.M
 MO.M
1000.00
 140.00
1000.00
 isa.ao
 291.SO
 101.4ft
           1.44
           1.90

           1.00

           1.40
 0.000
   O.0

O.OOM
  tf.0
 O.0H
         0.00E-OV
         1,•!£-<•
    M>   l.OOE-00
         0.00E-10
    00   t.ME-Ot
    10   0.04E-00
                                                                                          S.CnE-00 7.WE-Qt   204.1
                                                                                          •.00E-00 7.00E-00   210.0

                                                                                          O.QQE«00 O4BE«00   98O.O
                             270.0
                               220
                                            S330.H&
                                       7.11S    9740.A
    O
201.0
                                                                                                   .HC-Ot
                                                 O.7D
                                                    .I
    17
 O.124
                                                         l.»OE-l»
                                                 0.80
                                                 1.36
  as. a
o.ooa
   448
    10
  1000
    00
O.0000
  1000
    00
    00
  1000
  1000
    10
  MOO
     i
     •
  2000
     i
  O.M
   000
 90000
   100
     t
     2
     2
     2
    10
   100
120000

   0.2
                                                                               4.001-04
                                                                               1.1
                                                                                .09E«OI
                                                                                .ME-Oi
                                                                                .4K-OI

                                                                                .00E-00
                                                                                ,40E-0§
                                                                                .OOE-Oi
.07E«00
.B4E-C0
                                                                                                                     0.010
                                                                                                                                       100. is
                    .00E-OB
                    .21E-14
                    .ME-Oi
                    .SSE-04
                    .OCE-OB
                    .41E-00
                    .OOE-00
                                                                                          O.OQE<00 O,OOE«OO   M».i
                                                                                ICE-OS
                                                                                           .ace«oo o.ooe«oo
                             125.O
                             376.0

-------
(DrOJONMC
                          CA»|
                                     y.v.
                                                                  BOILMQ  VAKtt HESSUC COEFFICIB4I5
          deem    VA? .ness HUJULHV M urn atm^vf.wam DVF.ADI ramr	
          aiND~«-(t-4EitftM-4*-7
neOIKIMftC
ne«L tsocYMftiE
net*. IBKIIIIC ACEMI
neMAccnc ACB
iwBMJtcEnc raoanc
neMBC oiMac(-^
neMBC DUMDC(-«)
MMABC OIMOC(-»)
                          M-«4-t
                                ra
   i«o.oo
   144.M
   1M.14
   140.M
1  nt.10
   •O.M

   •M.4O
   1M.1C
   SM.17
   17».M
   ITt.tt
     M.I
   MO.M
   1M.M
   1W.IS
   ••.TO
   1M.14
   Ut.OO
                          100-M-O
                              1-1
                          TCOt-U-t
nmMUc ACID
nUHMJC AHMMUDC
                          K-41-4
PICOJHE(2-)
ra.voiiMimiB> unents
                          110-M-O
                               602

                               607
                               780
    IM.14
    101.14
    MB.M
    •0.40

       M
    1M.14
     14*.1
    147.10
     M.ll
    1M.20
     • .14
   1000.OO
      290
   1600.00
   1600.00
                                                                    0.1
                                                                    O.t
                                                                          T.
                                               t.ll
                                               i.n
                                               i.w
                                               o.oo
                                               l.U
                                               1.07
O.00001
 O.OOM
    4.4
o.aoow
o.ooon
o.ooon
  O.M1
   10
 0.01
1.41C-01
l.MC-Ot
T.KC-OI
t.ME-Of
O.OOIiOO O.OOE«OO
•.IOE-01 f.TaE-OI
f ,IOE-Oi •.•OE-01
•.ME-01 •.•Of-02
   10
  7«0
                                               1.14
                                               O.OO
                                               O.OO
                                               O.OO

                                               l.tt
                                               I.M

                                               O.M
                                               O.M
                                               O.OO

                                               1.46
  O.OJ»
0.007*7
 O.OOM

   M.T
   1«M
   •no
  uo.«
 O.OOli
  117.S
   10.4

   3.72
                                                                   1000
                                                                    MO
  MO
 •no

 •no
IOOQO
   i
10000
  0.2

  0.2
  0.2
D.ME-Of
i.tx-at
•.06C-OD
4.ME-07
1.20E-OI

1.1K-01
1.20E-0§
•.20E-4V
1.C3E-OK
1.13E-O1
l.Uf-03
1.13E-01
l.OK-04
4.nc-or
1.71E-01
                                                                                    O.ODE«00 O.OD£«OO
                                                                                    •.ME-01
                                                                                    O.ODE«OB O.OO£«00
                                                                                    O.OOC«OO O.OOE«00
                                                                                    O.OQE«00 OU)OE«OO
                                                                                    l.UE-Oi lUMC-Ot
i.3ac-oi
•.ooE-or
1.14E-OI
1.27E-04
4.4K-OI
4.22C-OS
t.OOE-0*
2.94E-O4
7.60E-C3
        •.40E-02
i.WE-03 7.ME-02
O.OOE«00 O.OOE«OO
».aOE-« 7.HE-O2
O.OOE«00 O.OOE«OO
O.OOEiOO *.1OE-O2

l.OOE-OK 1.O4E-O1
  0.0
177.0
102.0
MO.O
                   260.0
                   340.0
                   Itt.O
                                                                                                                   ©
                                                                                                                            O
                                                                                                                            O
                                                                                                                                   O
                                                                                                                                   o
                                                                                                             •.740    iS7»      107
                                                                                                                   G
                                                                                                               7.1M  liM.Tt  174.8K
2M.O
2M.O
M7.O

110.6
  •.1

 72 .•
2M.O
 09.6
120.0
166.0
140.0
                                                         O
                                                         O
                                                                                                                  o
                                                o
                                                o
                                                o
                                                                                                            ••Ml
                                                                                                                               130
                                                                                                                  o
                                                                                                               •.O2S
    O
3M».S
O
O
                                                                                                             7.032   I41C.71  811 .A3
                                                                                                                           0

-------
                                             MNSirv   we.raea MLUUUIY H LAV oHa&m,
 OftMUOMUC
                          CAS |
            y.v.
                                                                  BDILMQ
                                                          »•*. AM ramr
                                                          {0*/MC)(*0.C)
                                                                                                              VAPOR rfESMf COBFFlCIBflS
 NLttCB COHTAJMNB W1MHBI
                               •TO
                          •OOMV-O
           •000.00
          •9000.00
POLWMN. ACETA1E
ra.WMC ALCOHOL
roLtvwN. CHLOMK
fO-WBULlJOC CHLOMDC
                                no
                          «K»-X>-7
                               •74
    nac AOD
ttt-M-t
7*-0»-«
M7-U-0
11IHV-1
•7-
 toooo.oo
1MOOO.OO
110000.00
 vooo^n
 10000.00
   44.0*
   M.OI    o.n
   74.01
         ACETAK
                          m-oo-o
                          MO^M-1
                               MS
                          101-77-*
                              -i
                               IM
^-OCOHtBOOIMCMLOnDi
r-QUMDmem. mncmai
            AUOHDL
                          WM-M-l
                          Xffi-71-l
                          llM-4*rl
                                •t
                               47*
                          1M7-11-4
                               IM
                          2CMO-17-4
                          lfO-7*-S
            ALCOHU.
                        m
                    CHJQRIDM-74-«
                          019-7>-«
   4t.lt
   M.ll
    M.I
  Mt.U
  UO.I*
  tot.to
    7».l
  MO.OO
  1M.M
  Ml.11
  tit.oo

  1M.C7
  9M.fi

  I ft.CO
  1M.M

  1W.OO
  1*4.22
  124.1*
  20.10
  221.M
  1M.«
1.M
O.M
O.M
O.M
I.t7
O.M
 no
  10
  40
MOO
 O.I
                                                           t.c
                      O.t
                      O.t
                      MO
                      O.t
                      O.t
                    •0000
                      o.t
                      o.t
                     tOOD
                    JQOO(}
                    •0000
                    coooo
                      SCO
                                   •
                           O.OOOOQM7
                                         fQIJOft
                                         MOOO
                                            •0

                                         yyipQ
                      MO
                      O,
                                                            10
                                          O
                                            •0
                                          2000
                                            t
                                          •000
                                            2
                                         40000
                                            2
                                            60
                                         10000
                  .•OE-01
                  .OOE-OI
                  .OOE-OI
                  .OQE-OI
.OQC-Ot
.OOE-OI
.OOE-OI
,«-01
.1K-OI   S
•7C-CC
,7K-«I
                                                                                       S4E-OI IXBf-OI
                                                                             •OE-Oi
                                                                             .•4E-OI
                                                                             ME-04
                                                                                       .3K-OI t.nE-Ot
                                                                                       .SQE-Oi I.04E-OI
                                                                             oot-oi
                            i.90E«00 O.OQE
-------
                                           roam
                                                                                       •ULfttt   VArat
                                                                                       roiMr    	
                                                                                                                 MES9UC COBFIClBflS
CBtXUOMMC
                         CAS
                         M.V.
                         iao-n-t
                         m MO
•ODBJUACETA1E
aODUM HMftlE              141-M-7
ram)M&>B4-io-a>c.t,t^BCiHDa«T-f7-«
sraac                    ioo-4t-«
UCCMC ACB>               iio-u-4)
•XCMDODE                 ra-«*-4)
UMOL1C AOD             lSl-47-1
auuoc
fr4-auMcraj(ieM.-«.(M»ikcr     MS
t-auacm. •.•^IICIMN. nmnaatimi
O.t.t.4)

                         M».»
                          •.Oft
                         M».M
                         U4.M
                         •4S.OO
                           •40
                                     «M.4f
                                     400.00
                         ns.n
                         s».n
                         tlC.4D
                         Mi.OO
                          1M.O
                         14K.M
                         tn.to
                         m.to
                                     MJ.M
                                     1M.M
                                      72.12
                                              O.OB
                                              O.OO
                                              o.oo
                                              o.oo
                                              o.oo
                                              l.M
                                              1.40
                                              I.W
                                              1.434
                                              O.OO
                                              O.OO
                                                            1
                                                       O.OQtt
                                                       O.OOOM
                no
               1000
               1400
                 •
               tan
                 t
                                                                  tmo
                                                          7.1
                                                         O.ODt
                                                            4
                                                            4
                                                        o.ou
                                                         OjQi
                                                         OM
     0.4*
    .01

     O.K
O.OOOOQOM
        1
     72.1
                                                                   O.
                                                                            •c-ai
                                                                    O.OOE«O> S.VE-Ot
                                                                    «.WC-0» 7.«E-Ot
                                                                                                         ,O
                                                                                                         .O
O
O
                             O
                             O
                                                                           OK-1D
                                                                           JOE-OB   S.OOE-Ot 7.KC-OI   14B.O   7.14O
                                                                    .it    124.00
                                s.
                                                                                       0.0
                                                                           S4E-01
                                                                           nc-04
                                                                           TK-M
                                                             1.0E-4M
                                                             t.OOE-01
                                                     t
                                                     t

                                                     20
                                                  aaoao
                                                           .OQE-Oi
                                                           .•OE-04
                                                           4Dt-Ot
                                                           ,OE-Oi
                                                                                   d.OQE«OB OAC«00
                                                                                   9.0QE«00 OXttXD
                                                                                   7.SOC-OS 7.MC-4V
                                                                                   7.40E-0* 7.UE-OI
                                                                                   s.ace-o* r.20E-4K
                                                                                   O.OCE«OO O-ODEiOO
                                                                                   O.OOE«00 O.ODE«00
»4.0
244.0
•4».0
M4.»
149.2
121.4
1«4.0
                                                             Q
                                                                                                            •.Mt  St«S.M   to*
                                                          4. AM   IMt.
        O
        O
        O
    .74
    170.0
   217.M
        O
        O
                                                              .ME-Oi
                                                              .08E-4X
                                                             4.00E-OS
                                                                                           0.KE-4K
                                                                                                            o.me
                                                                                                                             aa« 26

-------
                                                                                                                      VAPOR emssjm. coBFiciBfts
                                               oeenv    vtf .r*esa nuuuLnv M
                                                                                                     ovr AW ramr
     OMKUOMME
                            CASf
                                       M.V.
     VIBUEA
       ; 0IAMBC(2,4)
       ! DIBOCYAMUEOM)
muaic-a.4-oiA20-ua-«cu-i
vuoc AOO it**-}
          <-o>
          untf
     XNOBB4ZBCa.a.4)
      ua
     -%
iaa-70-i

0aoi-«s-a
      •44
iao-ta-i
                    ,M>
    •MIO)UMEIHttC(l,l.a)
i   VOOiOUEimLBC
                            71-00-0
                            70-00-4
                            70-O1-0
                    .!.»
                    ,t.»
                               •77-4)
                               >-aa-
                              10-4
        W,l,2)WIH10UE1)MN70-ia-l
                           Ul-44-4
WEMUBC o-vco.           iia-a7-«
vmaunc AMMMUDE        wa-ao-7
                           7s-fo-a
                O.a.s)     M»47-a
                   )       110-00-7
                           •ot-as-o
ntinemmKnu>c raoo. CAW      a»7
IAEA                       K7-19-0
UETlWC                    K1-70-A
 01.14
 M.U
  00.4
iaa.oo
174.10
4IB«00
1M.10
iao.i4
Mf.17
414.00
            101.00
            101.40
             1M.4
             iaa.4
             iai.4
             197.4
            107.40
            147.40
            147.48
            147.0B
            147.40
            140.40
            107,10
            Mt.aa
            1B3.20
             04.00
            102.10
             00.11
            120.20
            227.10
            202.00
            •77.00
             00.00
             00.00
                      1.41
                      O.07
                      1.11
                      i.ao
1.00

1.11

o.oo
0.00
o.oo
i.aa
 i.a
1.40
1.40
0.00
o.oo
          o.oo

          l.M
            0.01
             14f
              ao
           o.ow
O.00001
   0.10
  o.a«a
    o.a

   0.10
   o.aa
   4.00
    iaa
    at
    7i
    700
 O.OOV0
    a.i
^QQQfr


 aooo

 MOO

  MO
                                                                        o.a
                                                 1.34
                                                 o.oo
            1.07
               a

             100
             400
               1
            0841
          o.oon
               1
            1.00
           O.O40
                     0.00
                       10
            •0000
            aoooo
               a

           MOOOO
               a
             1000
               2
              10
             2000
            aoooo
                                                                                .tOE-4M
                                                                                          .OOE«00
                     ,00E-0§
                     ,ME-«0
                                                                                             •.aoc-oa «.ME-OB
                                                                                    SX-Ot
                                                                                             O.OOEiOD O.OOE«00
                                        .4IC-01
                                        >43£^0ft
                                        00E-01
                                        .00£«00
                                        .72E-0§
                                        .40E-OI
                                                                               .77E-fli
                                                                                ,0QE-Oe
                                                                                0o£-oa
                                                                                0DE-01
                                                                                                 7.00E-00
                                                                                         t.ME-00 7.§OE-OB
                                                                                         i.70E-00 •.70E-00
                                                                                         O.OOE«00 O.OOE«OO
                                                                                                 7.MEHQI
                               7otOE-00 7.ME-OB
         4QE«at
         .ME-00

         .47E-OE
         .•7E-OI
         .•1E-0«
         .77E-0»
        2.04E-O4
        C.00E-Oi
                                                                                            O.OOE«00 O.OOE«00
                                                 .22£~01
                                         .CME-Oi •.•K-01
102.0
110,1
  0.0
2C1.O

aTS.o
204.0

  0.0

aia.o
200.K
  0.0
 7S.O
   74
 07.0
 aa.0
244, K
M7.O
14O.O
124.0
Ii0.0

 40.0
                                                                                                           -47. •
                                                                                                           242.C
                                                                                                               133.0
                                                                                                                            a
                                                                                                                         ©
O
O
                                                                                          o
                                                                                          o
                                                                                                                                  o       o
                                                  o        o       o
                                                  ©        ®       o
                                               0.«4i   tia0.«    aoe.0
                                               0.0S£   S014.4     209
                                               0.110   Mtt0.«    102.7
                                               0.004 1040.004   a«0.00
                                                  ©        &       O
                                                                                                                         •oti    iflet.t
                                                                                                                                      227. S

-------
OftNUOWtC
                                                                                                    •HUM  VAMi
                                            Deem    vAf.net «U*IUTV M u» coN^,»iFF.miBi  OWF.AB HUNT    -—
WLOOC AHD
WML ACETATE
VMA. OUNCE
MNNJDBC OUKBC
XftBC
                          It-OI-4
                          tt-4f-«
«ae<-»)
EMC ACETATE
10i.lt
 M.Ot
  M.t
  •7.0
IOt.20
10t.lt
 aot.t
10t.lt
O.M
0.«
i.n

O.M
O.M
O.M
  1
lit
                                                                    aooo
t.t
  t
  I
t.l
                                                                   SOOOD
                                                                            ne-of
                                                                            tot-oi
                                                                            27E-OI
                                                                            S7E-OI
                                                                            .I7C-O»
t.VE-Ot t.tCC-OI    71.0
l.SK-OI 14MC-O1   -ll.t
1.0C-OI t^UE-Ot    M.t

7.tOE-Ol 7.00E-Qt   1M.O
l.OOe-OI t.TOE-Ot   144.4
O.OX«00 OXHE«00   1M.4
                                                                                                              7.21O  12M.lt
                                                                                                                t.tn    10M.4    217.2
7.0W
  t.M
t.MO
                                                                                                                     14M.
                                                                                                                     14tt.4»
tlt.ll
  tU.M
ttt.U

-------
             Tablt 2. Default Values for Compound Properties
•n
k
  l 1A A A AC
o 3V A a V ^*^^V9 01 9 e ^^a^P^^^rW ^ e B^V 9, • V o 97 9 6 ^^P
1.41 i»i.0 10.90 0.009000 0.093 ,0.t« |0I0
0.7« 699.0 a.f 0.000009 0.190 If. a 1.0
1.00 0 0 0.000909 0.090 0 0
8 AC •• • • ••!• A AA^AAA ^ ••• A a
o ^^^ ^^^r • ^^ ^F • ^^^F P ^ 3T » ^^^^^^^^^F^ ^F • ^^^P^V ^p E^p
a. is 199.0 0.isaa 0.000011 0.110 g,? $.02
0.97 142.0 0.0072 0.000010 0.110 17. SC 4.48
1.19 10.0 4.420.0. 0.000009 0.090 If. I f.6a
1.26 192.0 0.0290 0.000010 0.100 if.i f.63
.

-------
            APPENDIX G

PHYSICAL AND CHEMICAL PROPERTY DATA
   FOR 25 COMPOUNDS OF POTENTIAL
 CONCERN AND FOR  COMPOUND CLASSES

-------
                            APPENDIX G.
CHEMICAL AND PHYSICAL PROPERTIES FOR  25 COMPOUNDS OF POTENTIAL CONCERN
9MMM.
No.
1
2
3
4
5
6
7


8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Henry's Henry's Law log Kow
Law Henry's Law Constant (octanol -water
Constant Constant (atm) partition coeff
Compound (atm-m/mol) (atm-m3/gmol ) @20 C (9 25 C)
Acetone
Acrylonitrile
Benzene 0.00550
Carbon TetrachloMde 0.03020
Chlorobenzene 0.0039
Chloroform 0.0039
Diehlorobenzene,, e-
Di chlorobenzene. w-
01 chlorobenzene. p-
1.1 Dlchloroethane 0.0056
1.2 Dlchloroethane 0.0014
Ethyl benzene
Isopropy'l Alcohol
Methyl ene Chloride 0.0032
Methyl Ethyl Ketone
Methyl t-Butyl Ether
Napthalene
Phenol
Tetrachl oroethyl ene
Tol uene
1,1,2 Trichloroethane 0.0049
1.1.1 Trichloroethane
Tr1 chl oroethyl ene
Vinyl Chloride 0.19000
o-Xyl ene
m-Xyl ene
p-Xyl ene
2.
8.
5.
3.
3.
3.
1.
3.
1.
5.
1.
6.
1.
2.
4.
7.
1.
4.
2.
6.
7.
1.
9.
8.
5.
5.
5.
50E-05
80E-05
50E-03
OOE-02
93E-03
39E-03
94E-03
61E-03
60E-03
54E-03
20E-03
44E-03
50E-04
24E-03
35E-05
80E-04
18E-03
54E-07
90E-02
68E-03
40E-04
72E-02
10E-03
60E-02
27E-03
20E-03
27E-03
-0.
-0.
24
92
2.40E+02 2.13 8 20 C
1.29E+03 2.
83
2.84 8 20 C
1.70E-H32 1.97 9 20 C
3.
3.
1.90E+02 3.
1.
6.10E+01 1.
3.
38
38
39
79
48
15
-0.16/0.28 calc
1.
0.
1.74-t-O.
3.01/3,
1.
25
26
12
45
46
1.10E+03 2.60 8 20 C
3.40E+02 2.
4.30E+01 2.
4.00E+02 2
5.50E+02 2.
3.55E+05 1.
2.
3
3.
73
47
.5
29
38
77
.2
15
Flash
Point
(C)
-18
0
-11
non-flani
23
non-flam
66
63
65.5
-5
13
15
12
non-flamm@<
-3
-28
88
79
non-flam
4.4
non-flam
non-flam
non-flam
-78.9
17.1
29.4
27.2
Maximum V-L
Water Equil.
Vapor Solubility Constant
Density mg/L K

1
2

3
4
5
5
5
3

3
2
2
2

4
3
5
3
4

4
2
3
3
3
2
.83
.77
5.3
.88
.12
.05
.08
.08
.44
3.4
.66
.07
.93
.42

.42
.24
.83
.14
.63
4.6
.53
.15
.66
.66
.66
1000000
79000
1750
757.
466
8200
100
123
790
5500
8520
152

20000
268000


93000
150
535
4500
1500
1100
2670
175
130
198


6L9
313.1
40.8
49



71.3
17.1


49.8






48.7


2760.3



                                                G-l

-------
                        APPENDIX G.
CHEMICAL AND PHYSICAL PROPERTIES FOR 25 COMPOUNDS OF POTENTIAL CONCERN
No.
1
2
3
4
5
6
7


8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Molecular
Compound We1 ght
Acetone
Acrylonitrile
Benzene
Carbon Tetrachlorid*
Chlorobenzene
Chloroform
Diehlorobenzene, o-
Dichlorobenzene. m-
Oi chlorobenzene, p-
1,1 Dlchloroethane
1,2 Oichloroethane
Ethyl benzene
Isopropyl Alcohol
Methyl ene Chloride
Methyl Ethyl Ketone
Methyl t-Butyl Ether
Napthalene
Phenol
Tetrachl oroethyl ene
To! uene
1.1.2 Triehloroethane
1,1.1 Trlchloroethane
Trichl oroethyl ene
Vinyl Chloride
o-Xylene
m-Xylene
p-Xyl ene
58
,,53
78
153
112
119
147
147
147
98
98
106
60
84
72
88
118
94
165
92
133
133
131
62
106
106
106
.08
.06
.12
.82
.56
.38
.00
.00
.00
.96
.96
.16
.09
.93
.11
.15
.19
.10
.83
.14
.41
.41
.40
.50
.20
.20
.20
Diffuslvity
Boiling Vapor in Water
Point Pressure (cm2/sec)
(C) (mti Hg) xlOE-05
56.2
77.4
80.1
76.7
131. S
61.2
17i.O
172.0
173.4
57.0
83.5
136.2
82.4
40.1
79.6
55.0
218.0
182.0
121.0
110.7
113.7
75.0
86.7
-13.9
144.0
138.8
138.5
266
114
95.2
113
11.8
208
1.5
2.28
1.2
234
80
10
42.8
400
100
245
0.023
0.0341
19
30
25
123
75
2660
7
8
9.5
1
1
0
0
0
1
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
1
1
0

.1400
.3400
.9940
.8840
.9000
.0600
.7900
.7860
.7900
.9880
.9880
.7800
.0400
.6600
.9800
.8026
.7500
.9000
.8200
.8600
.8800
.8800
.9100
.0400
.0000
.7800

roKSmaKSSKSS=333«X=3S3SSSS3SS3
Liquid Odor Oiffusivity
Density Threshold in Air
(g/cm3) (ppm) (cm2/sec)
0
0
0
1
1
1
1
1
1
1
1
0
0
1
0
0
1
1
1
0
1
1
1
0
0
0
0
.790
.810
.879
.595
.170
.489
.310
.290
.290
.256
.256
.870 0,
.790
.327
.820
.758
.140
.070
.624
.840
.320
.330
.400
.908
.880
.860
.860
100
21.4
0.84
21.4
0.21
675
2-4
0.02
15-30
120
3-100
,46-0.60
7.5
25-307
2

0.003
0.016
50
0.17

100
21.4
260
0.17
-1
-0.3
0.1240
0.1220
0.0932
0.0632
0.0730
0.0888
0.0690
0.0692
0.0690
0.0919
0.0907
0.0750
0.0980
0.1000
0.0808
0.0806
0.0590

0.0720
0.0870
0.0792
0.0780
0.0790
0.0900
0.0870
0.0700

                                               G-2

-------
                              TECHNICAL REPORT DATA
                       (Please read Instructions on the reverse before completing)
    EPA-450/l-89-002a
                         2.
                                                  3. RECIPIENT'S ACCESSION NO.
    Air/Superfund  National Technical Guidance
    Study Series.   Volume II - Estimation Of
    Baseline Air Emissions. At Superfund Sites,
            5. REPORT DATE
               August 1990
            ». PERFORMING ORGANIZATION CODE
    Bart Eklund and Charles Schmidt
                                                  8. PERFORMING ORGANIZATION REPORT NO.
   FORMING ORGANIZATION NAME AND ADDRESS
    Radian Corporation
    8501 Mo-Pac Boulevard
    Austin, Texas    78759
            10. PROGRAM ELEMENT NO.
            11. CONTRACT/GRANT NO.
               68-02-4392
12. SPONSORING AGENCY NAME AND ADDRESS
    U.S.  Environmental Protection Agency
    OAR,  OAQPS
    Research Triangle Park,  NC   27711
            13. TYPE OF REPORT AND PERIOD COVERED
               Interim Final
            14. SPONSORING AGENCY CODE
IS. SUPPLEMENTARY NOTES
    EPA Project Officers   Anne A. Pope
S. ABSTRACT
    This report presents available methods  for estimating air
    emissions at Superfund hazardous waste  sites prior to any
    remedial action.   Methods described  include direct emission
    measurement techniques, indirect measurements and predictive
    emissions modeling.   Information is  provided on selecting from
    among the range of methods available given the associated range
    of costs and uncertainties.  This report revises and expands  an
    earlier report, Procedures For Conducting Air Pathway Analyses
    For Superfund Activities. Volume II. Estimation Of -Baseline Air
    Emissions At Superfund Sites. EPA^-450/1-89-002.  It is one in a
    series of reports  that provide guidance on conducting air pathway
    analysis at Superfund hazardous waste sites.

    The purpose of this  report is to assist EPA Air and Superfund
    staff> State Air Superfund;program.staff,  Federal and State
    remedial and removal contractors, potentially responsible parties
    and others in designing, conducting, and reviewing air pathway.
    analyses at undisturbed hazardous waste sites.
                          KEY WORDS AND DOCUMENT ANALYSIS
              DESCRIPTORS
                                      b.lOENTIFIERS/OPEN ENOEO TERMS  C.  COSATI Field/Group
    Superfund
    Hazardous Waste Sites
    Air Pathway Analysis
    Emissions
8. DISTRIBUTION STATEMENT

   Unlimited
19. SECURITY CLASS (TttteReportl
     Unclassified
21. NO. OF PAGES

      390
2O. SECURITY CLASS (Tliis pagel
     Unclassified
                       22. PRICE
  F«o« 2220-1 (R«». 4-77).  i»«cviou» EDITION i* o«soucre

-------