OSWER DIRECTIVE 9502.00-6D
INTERIM FINAL
RCRA FACILITY INVESTIGATION (RFI) GUIDANCE
VOLUME IV OF IV
CASE STUDY EXAMPLES
EPA 530/SW-89-031
FEBRUARY 1989
WASTE MANAGEMENT DIVISION
OFFICE OF SOLID WASTE
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
ABSTRACT
On November 8, 1984, Congress enacted, the Hazardous and Solid Waste
Amendments (HSWA) to RCRA. Among the most significant provisions of HSWA are
§;3004(u), which requires corrective action for releases of hazardous waste or
constituents from solid waste management units at hazardous waste treatment,
storage and disposal facilities seeking final RCRA permits; and ;§^3004(v), which
compels corrective action for releases that have migrated beyond the facility
property boundary. EPA will be promulgating rules to implement the corrective
action provisions of HSWA, including requirements for release investigations and
corrective measures.
This document, which is presented in four volumes, provides guidance to
regulatory agency personnel on overseeing owners or operators of hazardous waste
management facilities in the conduct of the second phase of the RCRA Corrective
Action Program, the RCRA Facility investigation (RFI). Guidance is provided for the
development and performance of an investigation by the facility owner or operator
based on determinations made by the regulatory agency as expressed in the
schedule of a permit or in an enforcement order issued under §i 3008(h), §' 7003,
and/or 53013. The purpose of the RFI is to obtain information to fully characterize
the nature, extent and rate of migration of releases of hazardous waste or
constituents and to interpret this information to determine whether interim
corrective measures and/or a Corrective Measures study may be necessary.
-------
DISCLAIMER
This document is intended to assist Regional and State personnel in exercising
the discretion conferred by regulation in developing requirements for the conduct
of RCRA Facility Investigations (RFIs) pursuant to 40 CFR 264. Conformance with this
guidance is expected to result in the development of RFIs that meet the regulatory
standard of adequately detecting and characterizing the nature and extent of
releases. However, EPA will not necessarily limit acceptable RFIs to those that
comport with the guidance set forth herein. This document is not a regulation (i.e.,
it does not establish a standard of conduct which has the force of law) and should
not be used as such. Regional and State personnel must exercise, their discretion in
using this guidance document as well as other relevant information in determining
whether an RFI meets the regulatory standard.
Mention of company or product names in this document should not be
considered as an endorsement by the U.S. Environmental Protection Agency.
-------
RCRA FACILITY INVESTIGATION (RFI) GUIDANCE
VOLUME IV
CASE STUDY EXAMPLES
TABLE OF CONTENTS
SECTION PAGE
ABSTRACT
DISCLAIMER
TABLE OF CONTENTS '"
TABLES V"
FIGURES IX
LIST OF ACRONYMS X"
in
-------
VOLUME IV CONTENTS (Continued)
SECTION
14.0
14.1
14.2
15.0
CASE
INTRODUCTION
USE OF CASE STUDIES
ORGANIZATION OF VOLUME IV
CASE STUDIES
STUDY 1.
CASE STUDY 2.
CASE STUDY 3.
CASE STUDY 4.
CASE STUDY 5.
CASE
CASE
CASE
CASE
STUDY 6.
STUDY 7.
STUDY 8.
STUDY 9.
CASE STUDY 10.
CASE STUDY 11.
USE OF THE 40 CFR 261 LISTING BACKGROUND
DOCUMENTS FOR SELECTING MONITORING
CONSTITUENTS
ESTIMATION OF DEGRADATION POTENTIAL OF
CONTAMINANTS IN SOIL
SELECTION AND EVALUATION OF A SOIL
SAMPLING SCHEME
SAMPLING OF LEACHATE FROM A DRUM
DISPOSAL AREA WHEN EXCAVATION AND
SAMPLING OF DRUMS IS NOT PRACTICAL
USE OF QUALITY ASSURANCE/QUALITY
CONTROL (QA/QC) AND DATA VALIDATION
PROCEDURES
PRESENTATION OF DATA COLLECTED DURING
FACILITY INVESTIGATIONS
CORRELATION OF CONTAMINANT RELEASES
WITH A SPECIFIC WASTE MANAGEMENT UNIT
USING GROUND-WATER DATA
WASTE SOURCE CHARACTERIZATION FROM
TOPOGRAPHIC INFORMATION
SELECTION OF GROUND-WATER MONITORING
CONSTITUENTS AND INDICATOR PARAMETERS
BASED ON FACILITY WASTE STREAM
INFORMATION
USING WASTE REACTION PRODUCTS TO
DETERMINE AN APPROPRIATE MONITORING
SCHEME
CORRECTIVE MEASURES STUDY AND THE
IMPLEMENTATION OF INTERIM MEASURES
PAGE
14-1
14-1
14-1
15-1
15-1
15-6
15-10
15-14
15-19
15-29
15-43
15-47
15-50
15-54
15-58
IV
-------
VOLUME IV CONTENTS (continued)
SECTION
CASE STUDY 12.
CASE STUDY 13.
CASE STUDY 14.
CASE STUDY 15.
CASE STUDY 16.
CASE STUDY 17.
CASE STUDY 18.
CASE STUDY 19.
CASE STUDY 20.
CASE STUDY 21.
CASE STUDY 22.
CASE STUDY 23.
CASE STUDY 24.
PAGE
USE OF AERIAL PHOTOGRAPHY TO IDENTIFY 15-63
CHANGES IN TOPOGRAPHY
WASTE MIGRATION ROUTES.
IDENTIFICATION OF A GROUND-WATER 15-68
CONTAMINANT PLUME USING INFRARED
AERIAL PHOTOGRAPHY
USE OF HISTORICAL AERIAL PHOTOGRAPHS 15-74
AND FACILITY MAPS TO IDENTIFY OLD WASTE
DISPOSAL AREAS AND GROUND-WATER FLOW
PATHS
USING SOIL CHARACTERISTICS TO ESTIMATE 15-78
MOBILITY OF CONTAMINANTS
USE OF LEACHING TESTS TO PREDICT 15-87
POTENTIAL IMPACTS OF CONTAMINATED SOIL
ON GROUDWATER
USE OF SPLIT-SPOON SAMPLING AND ON-SITE 15-97
VAPOR ANALYSIS TO SELECT SOIL SAMPLES
AND SCREENED INTERVALS FOR MONITORING
WELLS
CONDUCTING A PHASED SITE INVESTIGATION 15-105
MONITORING BASEMENT SEEPAGE 15-110
USE OF PREDICTIVE MODELS TO SELECT 15-114
LOCATIONS FOR GROUND-WATER
MONITORING WELLS
MONITORING AND CHARACTERIZING 15-119
GROUND-WATER CONTAMINATION WHEN
TWO LIQUID PHASES ARE PRESENT
METHODOLOGY FOR CONSTRUCTION OF 15-124
VERTICAL FLOW NETS
PERFORMING A SUBSURFACE GAS 15-137
INVESTIGATION
USE OF A SUBSURFACE GAS MODEL IN 15-144
ESTIMATING GAS MIGRATION AND
DEVELOPING MONITORING PROGRAMS
-------
VOLUME IV CONTENTS (Continued)
SECTION
CASE STUDY 25.
CASE STUDY 26.
CASE STUDY 27.
CASE STUDY 28.
CASE STUDY 29.
CASE STUDY 30.
CASE STUDY 31.
USE OF METEOROLOGICAL/EMISSION
MONITORING DATA AND DISPERSION
MODELING TO DETERMINE CONTAMINANT
CONCENTRATIONS DOWNWIND OF A LAND
DISPOSAL FACILITY
USE OF METEOROLOGICAL DATA TO DESIGN
AN AIR MONITORING NETWORK
DESIGN OF A SURFACE. WATER MONITORING
PROGRAM
USE OF BIOASSAYS AND BIOACCUMULATION
TO ASSESS POTENTIAL BIOLOGICAL EFFECTS
OF HAZARDOUS WASTE ON AQUATIC
ECOSYSTEMS
SAMPLING OF SEDIMENTS ASSOCIATED WITH
SURFACE RUNOFF
SAMPLING PROGRAM DESIGN FOR
CHARACTERIZATION OF A WASTEWATER
HOLDING IMPOUNDMENT
USE OF DISPERSION ZONE CONCEPTS IN THE
DESIGN OF A SURFACE WATER MONITORING
PROGRAM
15-158
15-165
15-1-74
15-185
15-188
15-194
-VI
-------
TABLES (Volume IV)
NUMBER PAGE
14-1 Summary of Points Illustrated 14-2
15-1 Uses and Limitations of the Listing Background Documents 15-2
15-2 Results of Original Surface Soil and Tap Water Analyses 15-23
15-3 Laboratory QC Results 15-25
15-4 Field QC Results 15-26
15-5 Summary of Data Collected 15-33
15-6 Typical Methods for Graphically Presenting Data Collected 15-.42
During Facility Investigations
15-7 Indicator Parameters 15-52
15-8 Results of Monitoring Well Sampling 15-55
15-9 Average Values of Parameters in Ground Water and Stream 15-73
Samples
15-10 Relative Mobility of Solutes 15-82
15-11 HEA Criteria, Constituent Concentrations and Relevant 15-91
Physical/Chemical Property Data For Constituents
Observed At Site
15-.12 Leaching Test Results (mg/l) 15-93
15-13 Ground-Water Elevation Summary Table Phase II 15-127
15-14 Model Results 15-149
15-15 Summary of Onsite Meteorological Survey Results 15-156
15-16 Relationship of Dissolved and Sorbed Phase Pollutant 15-169
Concentrations to Partition Coefficient and Sediment
Concentration
15-17 Parameters Selected for Surface Water Monitoring Program 15-170
15-18 Selected Surface Water Monitoring Stations and Rationale 15-171
VII
-------
TABLES (Volume IV - Continued)
NUMBER PAGE
15-19 Mean Concentrations (ug/l) of Organic Substances and 15-178
Trace Metals in Leachate and Surface Waters
15-20 Mean Sediment Concentrations (ug/kg Dry Wt) of Organic 15-179
Substances and Trace Metals
15-21 Mean Liver Tissue Concentrations (ug/kg Wet Wt) of 15-180
Organic Substances and Trace Metals
15-22 Mean LC50and EC50Values (Percent Dilution) for 15-181
Surface-Water Bioassays
15-23 Relative Toxicity of Surface-Water Samples 15-182
15-24 Arsenic and Lead Concentrations (ppm) in Runoff 15.-1-87
Sediment Samples
15-25 Summary of Sampling and Analysis Program for 15-191
Wastewater Impoundment
15-26 Relationship of Dissolved and Sorbed Phase 15-198
Contaminant Concentrations to Partition
Coefficient and Sediment Concentration
15-27 Parameters Selected For Surface Water 15-199
Monitoring Program
15-28 Selected Surface Water Monitoring Stations and 15-200
Selection Rationale
Viii
-------
FIGURES (Volume IV)
NUMBER PAGE
15-1 Results of Laboratory Bench Tests for Pesticide 15-8
Degradation
15-2
Isoconcentration Map of the Lead Concentrations in ppm 15-13
15-3 Around the Smelter
15-3 Schematic Diagram of Gas Control System Utilized at Pit 15-16
15-4
Schematic Drawing of Wireline Drill Bit and Reaming Shoe 15-17
15-5 Map of the Smelter Site and Associated Tailings Ponds 15-30
15-6 Locations of Copper Leach Plant and Waste Storage Ponds 15-31
15-7 Schematic of Surface Water System 15-35
15-8 Ground-Water Flowlines Based on Measured Water Levels
15-9 Selected Surface Water Quality Parameters at Key Stations 15-37
15-10 Changes in Sulfate Over Time at Selected Wells Located 15-38
Within the Site
15-11 Field Sketch Tailings Trench T-3 15-40
15-12 Depth vs. Concentration Profiles for Selected Variables 15-41
for Borehole 88A
15-13 Location of Ground-Water Monitoring Wells 15-44
15-14 Site Map Showing Waste Disposal Areas 15-48
15-15 Site Map and Monitoring Well Locations 15-56
15-16 Ground-Water Elevations and Flow Directions in 15-60
Upper Limestone Aquifer
15-17 October 1983 Aerial Photograph of Land Disposal Facility 15-65
15-18 Aerial Photograph Interpretation Code 15-66
15-19 February 1984 Aerial Photograph of Land Disposal Facility 15-67
15-20 Facility Plan View 15-69
IX
-------
FIGURES (Volume IV- Continued)
NUMBER PAGE
15-21 Generalized Geologic Cross-Section 15-71
15-22 Infrared Aerial photograph of the Site 15-72
15-23 Site Layout: LWDA-2, SDWA-2 and Stream Channel 15-76
Identified Through Use of Aerial Photograph Analysis
15-24 Schematic Cross-Section of Waste Disposal Site 1.5-80
15-25 Hypothetical Adsorption Curves for a) Cations and 15-83
b) Anions Showing Effect of pH and Organic Matter
15-26 Schematic Diagram Showing Plumes of Total Dissolved 15-86
Solids (TDS), Total Organic Halogens (TOX) and Heavy
Metals Downgradient of Waste Disposal Site
15-27 Facility Map Showing Soil Boring and Well Installation 15-88
15-28 Facility Map Showing Ground-Water Contours 15-90
15-29 Site Plan Showing Disposal Areas and Phase I Well 15-98
Locations
15-30 Geologic Cross-Section Beneath Portion of Site 15-100
15-31 Ground-Water Elevations in November 1984 15-101
15-32 Example of Borehole Data Including HNU and 15-102
OVA/GC Screening
15-33 Proposed Phase II Soil Borings 15-107
15-34 Proposed Phase II Monitoring Wells 15-108
15-35 Geologic Cross-Section Beneath Site 15-111
15-36 Estimated Areal Extent of Hypothetical Plumes 15-116
from Four Wells
15-37 Consideration of Solute Migration Rates in Siting 15-118
Sampling Wells
15-38 Well Locations and Plant Configuration 15-121
15-39 Behavior of Immiscible Liquids of Different Densities 15-123
in a Complex Ground-Water Flow Regime
-------
FIGURES (Volume IV - Continued)
NUMBER PAGE
15-40 Top of Lowest Till Contour Map-and Location of 15-125
Vertical Flow Net
15-41 Recharge/Discharge Areas and Flow Directions 15-130
15-42 Vertical Flow Net T-T 15-132
15-43 Site Plan 15-138
15-44 Gas Monitoring Well 15-140
15-45 Facility Map 15-145
15-46 Unconnected Mignation Distances fon 5 and 1.25% 15-147
Methane Concentnations
15-47 Connection Factons fon Landfill Depth Below-Gnade 15-148
15-48 Impenvious Connection, Factons (ICF) fon Soil Sunface 15-150
Venting Condition Anound Landfill
15-49 Landfill Penimeten Gas Collection System Wells 15-152
15-50 Site Map Showing Location of Meteonological Sites A and B 15-154
-ic -\ c q
15-51 Site Plan and Locations of Meteonological Stations
15-52 Sampling Station Locations fon Sunface Waten Monitoning 15-167
15-53 Site Plan and Waten Sampling Locations 15-176
15-54 Bioassay Responses to Sunface Waten Samples 15-183
15-55 Sunface Waten and Sediment Sample Locations 15-186
15-56 Schematic of Wastewaten Holding Impoundment 15-190
Showing Sampling Locations
15-57 Sampling Station Locations fon Sunface Waten 15-196
Monitoning
XI
-------
LIST OF ACRONYMS
AA
Al
ASCS
ASTM
BCF
BOD
CAG
CPF
CBI
CEC
CERCLA
CFR
CIR
CMI
CMS
COD
COLIWASA
DNPH
DO
DOT
ECD
EP
EPA
FEMA
FID
Foe
FWS
GC
GC/MS
GPR
HEA
HEEP
HPLC
HSWA
HWM
ICP
ID
Kd
Koc
Kow
LEL
MCL
MM5
MS/MS
NFIP
NIOSH
Atomic Absorption
Soil Adsorption Isotherm Test
Agricultural Stabilization and Conservation Service
American Society for Testing and Materials
Bioconcentration Factor
Biological Oxygen Demand
EPA Carcinogen Assessment Group
Carcinogen Potency Factor
Confidential Business Information
Cation Exchange Capacity
Comprehensive Environmental Response, Compensation, and
Lability Act
Code of Federal Regulations
Color Infrared
Corrective Measures
Corrective Measures Implementation
Corrective Measures Study
Chemical Oxygen Demand
Composite Liquid Waste Sampler
Dinitrophenyl Hydrazine
Dissolved Oxygen
Department of Transportation
Electron Capture Detector
Electromagnetic
Extraction Procedure
Environmental Protection Agency
Federal Emergency Management Agency
Flame lonization Detector
Fraction organic carbon in soil
U.S. Fish and Wildlife Service
Gas Chromatography
Gas Chromatography/Mass Spectroscopy
Ground Penetrating Radar
Health and Environmental Assessment
Health and Environmental Effects Profile
High Pressure Liquid Chromatograhy
Hazardous and Solid Waste Amendments (to RCRA)
Hazardous Waste Management
Inductively Coupled (Argon) Plasma
Infrared Detector
Soil/Water Partition Coefficient
Organic Carbon Absorption Coefficient
Octanol/VVater Partition Coefficient
Lower Explosive Limit
Maximum Contaminant Level
Modified Method 5
Mass Spectroscopy/Mass Spectroscopy
National Flood Insurance Program
National Institute for occupational Safety and Health
XII
-------
LIST OF ACRONYMS (Continued)
NPDES National Pollutant Discharge Elimination System
OSHA Occupational Safety and Health Administration
OVA Organic Vapor Analyzer
PID Photo lonization Detector
pKa Acid Dissociation Constant
ppb parts per billion
ppm parts per million
PDF Polyurethane Foam
PVC Polyvinyl Chloride
QA/QC Quality Assurance/Quality Control
RCRA Resource Conservation and Recovery Act
RFA RCRA Facility Assessment
RfD Reference Dose
RFI RCRA Facility Investigation
RMCL Recommended Maximum Contaminant Level
RSD Risk Specific Dose
SASS Source Assessment Sampling System
SCBA Self Contained Breathing Apparatus
SCS Soil Conservation Service
SOP Standard Operating Procedure
SWMU Solid Waste Management Unit
TCLP Toxicity Characteristic Leaching Procedure
TEGD Technical Enforcement Guidance Document (EPA, 1986)
TOC Total Organic Carbon
TOT Time of travel
TOX Total Organic Halogen
USGS United States Geologic Survey
USLE Universal Soil Loss Equation
Ultraviolet
VOST Volatile Organic Sampling Train
VSP Verticle Seismic Profiling
WQC Water Quality Criteria
XIII
-------
14.0 INTRODUCTION
14.1 Use of case studies
This document Volume IV of the RCRA - Facility Investigation of (RFI) Guidance,
contains case studies selected to illustrate various concepts and procedures
presented in Volumes I, II, and III. These case studies are provided to explain,
through example, how various tasks can be conducted during RFIs. The case studies
also identify some of the potential problems that can occur if the RFI sampling and
analytical programs are not carefully designed and executed. The case studies,
however, should not be used as the primary source of guidance for RFI program
design and conduct. Instead, Volumes I, II and III should be consulted. The studies
do not necessarily address details specific to individual facilities, and omission of
certain RFI tasks should not be interpreted as an indication that such tasks are
unnecessary or of less significance. Most of the case studies are based on actual
sites. In some cases, existing data have been supplemented with hypothetical data
to illustrate a particular point.
14.2 Organization of Volume IV
The case studies are organized primarily by the order in which the subject
matter was presented in Volumes I, II and III. In some cases, individual case studies
present materials relevant to more than one topic or media. Table 14-1 lists the
points illustrated and identifies the case studies-which provide information relevant
to these points.
The following general format was used as appropriate for each case study
• Title
• Identification of points illustrated
• introduction/Background
• Facility description
• Program design/Data collection
• Program results/Data analysis
• Case discussion.
14-1
-------
TABLE 14-1
SUMMARY OF POINTS ILLUSTRATED
POINTS ILLUSTRATED
CASE STUDY
NUMBER
SELECTION OF MONITORING CONSTITUENTS
• Use of 40 CFR Part 261 Listing Background Documents in selecting
monitoring constituents
I • Consideration of degradation as a factor in identifying monitoring
constituents
1
2
SAMPLING SCHEMES
• Selection of a sampling scheme that appropriately characterizes soil
contamination
• Evaluation of the effectiveness of a sampling scheme using
statistical analyses
• Use of release monitoring leachate collection to characterize wastes
when the actual waste stream is inaccessible, as in the case of buried
drums
3
3
4
QUALITY ASSURANCE AND CONTROL
• Use of quality assurance and control and data validation procedures
DATA PRESENTATION
• Techniques for presenting data for facility investigations involving
multimedia contamination
WASTE CHARACTERIZATION
• Correlation of a contaminant release with a specific waste
management unit using ground-water data
• Use of site topographic information to select test boring and
monitoring well locations at facilities where large volumes of waste
have been disposed.
• Use of waste stream information to select indicator parameters and
monitoring constituents in a ground-water monitoring program to
minimize the number of constituents that must be monitored
• Use of information on possible waste reaction products in designing
a ground-water monitoring program
7
8
10
CORRECTIVE MEASURES INCLUDING INTERIM MEASURES
• Use Of biodegradation and removal for interim corrective measures
Corrective action and the implementation of interim corrective
measures
11
AERIAL PHOTOGRAPHY
• Use of aerial photographs to identify actual and potential waste
migration routes and areas requiring corrective action
• Identification of a ground-water contaminant plume using infrared
aerial photography
• Use of historical aerial photographs and facility maps to identify old
waste disposal areas and ground-water flow paths
12
13
14
14-2
-------
TABLE 14-1
SUMMARY OF POINTS ILLUSTRATED (continued)
POINTS ILLUSTRATED
CASE STUDY
NUMBER
SOIL
• Use of soil characteristics to estimate mobility of contaminants in
soil
• Effects of degradation in determining the fate of a contaminant in
soil
• Use of leaching tests to predict potential impacts of contaminated
soils on ground water
15
2
16
GROUND WATER
• Use of split-spoon sampling and organic vapor monitoring to select
screened intervals for ground-water monitoring
• Development of a two-phase boring program to investigate
ground-water contamination
• Use of basement monitoring to estimate contaminant migration
• Use of mathematical models to determine locations of ground-
water monitoring wells
• Monitoring and characterization of ground-water contamination
when two liquid phases are present
• Methodology for construction of vertical flow nets
17
18
19
20
21
22
SUBSURFACE GAS
• Design of a phased monitoring program to adequately characterize
subsurface gas migration
• Use of predictive models to estimate extent of subsurface gas
migration
23
24
AIR
• Use of dispersion modeling and meteorological/emissions
monitoring data to estimate downwind contaminant
concentrations
• Design of an upwind/downwind monitoring program when
multiple sources are involved
25
26
SURFACE WATER
• Use of existing site-specific data to design a surface water
monitoring program
• Use of bioassays and bioaccumulation studies to assess potential
biological effects of off-site contaminant migration
• Use of sediment sampling to indicate off-site contaminant
migration via surface runoff
• Design of a sampling program to account for three-dimensional
variations in contaminant distribution
• Use of dispersion zone concepts in the design of a surface water
monitoring program
27
28
29
30
31
14-3
-------
15.0 CASE STUDIES
CASE STUDY 1: USE OF THE 40 CFR 261 LISTING BACKGROUND DOCUMENTS FOR
SELECTING MONITORING CONSTITUENTS
Point I 11 u s trated
• . The 40 CFR 261 Listing Background Documents can be of direct help in
selecting monitoring constituents.
I n t r o d u c t i o n
The RCRA Hazardous Waste Listing Background Documents developed for the
identification and listing of hazardous wastes under 40 CFR Part 261 represent one
source of potential information on waste-specific constituents and their physical
and chemical characteristics the documents contain information on the
generation, composition, and management of listed waste streams from generic
and industry-specific sources. In addition to identifying hazardous constituents that
are present in the wastes, the documents may also provide data on potential
decomposition products. In some background documents, migratory potential are
discussed and exposure pathways are identified.
Appendix B of the Listing Document provides more detailed information on
the fate and transport of hazardous constituents. Major physical and chemical
properties of selected constituents are listed, including molecular weights, vapor,
pressures and solubilities, octanol-water partition coefficients, hydrolysis rates,
biodegradation rates, and volatilization rates. Another section of the appendix
estimates the migratory potential and environmental persistence of selected
constituents-based on a conceptual model of disposal in an unconfined landfill or
lagoon.
The appropriate uses and limitations of the Listing Documents are outlined in
Table 15-1. A case study on how the Documents may be used in investigating a
release follows.
15-1
-------
TABLE 15-1
USES AND LIMITATIONS OF THE LISTING BACKGROUND DOCUMENTS
Uses
Limitations
• Identifies the hazardous constituents for
which a waste was listed
• Applicable only for listed hazardous wastes
• In some cases, provides information on
additional hazardous constituents which
may be present in a listed waste
• Industry coverage maybe limited in scope
(e.g., the wood preserving industry). Listing
Documents only cover organic
preservatives, not inorganics(15 percent of
the industry), such as inorganic arsenic salts
• In some cases, identifies decomposition
products of hazardous constituents
• Data may not be comprehensive (i.e., not all
potentially hazardous constituents may be
identified). Generally, limited to the most.
toxic constituents common to the industry
as a whole
• Provides overview of industry; gives
perspective on range of waste generated
(both quantity and general characteristics)
• Data may not be specific. Constituents and
waste characteristic data often represent an
industry average which encompass many
different types of production processes and
waste treatment operations
• May provide waste-specific characteristics
data such as density, pH, and leachability
• Listing Documents were developed from
data/reports available to EPA at the time,
resulting in varying levels of detail for
different documents
• May provide useful information the
migratory potential, mobility, and
environmental persistence of certain
hazardous constituents
• Hazardous waste listings are periodically
updated and revised, yet this may not be
reflected in the Listing Documents
• May list physical and chemical properties of
selected constituents
• Listing Documents for certain industries
(e.g., the pesticides industry) may be subject
to CBI censorship due to the presence of
confidential business information. In such
cases, constituent data may be unavailable
(i.e., expurgated from the document)
15-2
-------
Facility Description
The facility is a wood preserving plant located in the southeast. The facility
uses a steaming process to treat southern pine and timber. Contaminated vapors
from the wood, treating process are condensed and transported to an oil/water
separator to reclaim free oils and preserving chemicals. The bottom sediment
sludge from this and subsequent waste water treatment units is a RCRA listed
hazardous waste: K001.
Use of Listing Background Documents
Due to the presence of small, but detectable, levels, of phenolic compounds in
the ground water of an adjacent, property, a RCRA Facility Assessment (RFA)was
conducted and it was determined that a release from the facility, had occurred. The
owner was instructed to conduct a RCRA Facility Investigation (RFI). Before
embarking on an extensive waste sampling and analysis program, the owner
decided to explore existing sources of information order to better focus analytical
efforts.
The owner obtained a copy of the Wood Preserving Industry Listing
Background Document from the RCRA Docket at EPA Headquarters. He also had
available a copy of 40 CFR Part 261, Appendix VII, which identifies the hazardous
constituents for which his waste was listed. For K001, he found the following
hazardous constituents listed: pentachlorophenol, phenol, 2-chlorophenol, p-
chloro-m-cresol, 2,4 dimethylphenyl, 2,4-dinitrophenol, trichlorophenols,
fluoranthene, benz(b)fluoranthene, benz(a)pyrene, ideno(1,2,3-cd) pyrene,
benz(a)anthracen, dibenz(a)anthracene, and acenaphthalene.
From the Summary of Basis for Listing section in the, Listing Document, the
owner found that phenolic, compounds are associated with waste generated from
the use of pentachlorophenol-based wood preservatives, and that polynuclear
aromatic hydrocarbons (PAHs) (i.e., chrysene through acenaphthalene in Appendix,
VII) are associated with wastes from the use of creosote-based preservatives.
Examining the facility records, he determined that pentachlorophenol had been the
15-3
-------
sole preservative used; moreover, it had come from a single manufacturer. Based
on a demonstrable absence of creosote use, the owner felt confident in excluding
creosote and PAHs.
To help focus on which phenolics might be present in his waste, the owner
turned to the Composition section of the Listing Document. In Table 4, he found
typical compositions og commercial grade pentachlorophenol. The sample from his
manufacturer contained 84.6 percent pentachlorophenol, 3 percent
tetrachlorophenol, and ppm levels of polychlorinated dibenzo-p-dioxins and
dibenzo-furans. The owner was surprised by the absence of the other phenoiics
mentioned in Appendix VII, and he was concerned by the presence of dioxins and
furans. Reading the text carefully, he discovered that the majority of the phenolic
compounds listed as hazardous constituents of the waste are actually
decomposition products of penta- and tetrachlorophenol. He also learned-that
while the Agency had ruled out the presence of tetrachlorodibenzo( p)dioxin (TCDD)
in the listed waste (except where incinerated), it had not ruled out the possibility
that other chlorinated dioxins might be present: "... chlorinated dioxins have been
found uncommercial pentachlorophenol and could therefore be expected to be
present in very small amounts in some wastes. " Due to their extreme toxicity and
because his facility had historically used the commercial pentachlorophenol, with
the highest concentration of dioxins and furans, the owner thought it prudent to
include a scan for dioxins in his waste analysis plan.
The owner found no further data in the Composition section to help him
narrow the list of phenolics; however, Table 6 gave a breakdown of organic
compounds found in different wood preserving plants (i.e., steam process vs.
Boneton conditioning, but only two phenolics were listed. A note in the text
highlights one of the limitations of using the Listing Document: "The absence in
this table (Table 6) of certain components . . . probably indicates that an analysis for
their presence was not performed rather than an actual absence of the
component. It should be kept in mind that the waste analyses in the Listing
Background Documents are not comprehensive and that they are based, as the
Agency acknowledges, on data available at the time. In the absence of more
detailed waste-specific data, the owner decided to include pentachlorophenol,
tetrachlorophenol, unsubstitued phenol, and the six listed decomposition-product
phenolic compounds in his waste analysis plan.
15-4
-------
In reading the Listing Documents, the owner found useful information for
other phases of the RFI. In the Migratory Potential Exposure Pathways section, he
learned that pentachlorophenol is highly bioaccumulative, with an octanol/water
partition coefficient of 102,000. Tetrachlorophenol, tri-chlorophenol, and
dichlorophenol are likewise bioaccumulative, with octanol/water coefficients of
12,589, 4,169, and 1,380, respectively. He also learned that the biodegradability of
pentachlorophenol is concentration limited.
In Appendix B of the Listing Background Documents; Fate and Transport of
Hazardous Constituents, the owner found data sheets for six out of nine phenolic
compounds, also some for dioxins and furans. Information on water chemistry, soil
attenuation, environmental persistence, and bioaccumulation potential were listed
along with chemical and physical properties such as solubility and density.
Case Discussion
Although the Listing Background Document did not provide the owner with
enough specific data to fully characterize his waste, it did help him refine the list of
monitoring constituents, alert him to the potential presence of dioxins, and gave
him physical and chemical waste characteristic data which could be useful in
predicting contaminant mobility.
15-5
-------
CASE STUDY 2: ESTIMATION OF DEGRADATION POTENTIAL OF CONTAMINANTS
IN SOIL
Point III ustr a t e d
• Degradation, either chemical or biological, can be an important factor in
determiningcthe fate of a contaminant soil, and can also be a factor in
identifying constituents to monitor. The. degradation rate can also be
accelerated as a means of conducting interim or definitive corrective
measures.
Introduction
Degradation of contaminants in the environment can occur through several
mechanisms, and can be a factor in identifying monitoring constituents. Under
natural conditions, these processes are often very slow, but studies have shown that
chemical and biological degradation can be accelerated in the soil by modifying soil
conditions. Parameters such as soil moisture content and redox condition can be
altered to encourage contaminant degradation in soils.
Site Description
The site is situated in an arid region that was used during the 1970s by aerial
applicators of organochlorine and organophosphate pesticides. The applicators
abandoned the site in 1980 and homes were built in the vicinity. The site can be
divided into three areas based on past use. The most contaminated area, the "hot
zone", is a 125 feet by 50 feet area at the north end of the site that was used for
mixing, loading, and unloading the pesticides. Soil samples from this area
contained toxaphene, ethyl parathion, and methyl parathion at concentrations up
to 15,000 mg/kg. The present residential area was used as a taxiway and an area to
rinse tanks and clean planes. Soils from this zone were low in parathion but
toxaphene concentrations ranging from 20 to 700 mg/kg were found. This area is
approximately 1.7 acres in size and located immediately south and west of the hot
zone. The runway itself was approximately 10 acres in size and south of the
residential zone. Soil sample results from the runway area had low concentrations
of all three pesticides.
15-6
-------
A number of factors influence degradation of organic compounds in soils.
These include:
• chemical nature of the compound
• organic matter content of the soil
• soil pH
• oxidation/reduction environment of the soil
• concentrations of the compound.
At the subject site, the soils were low in moisture content, were oxidizing, and
exhibited soil pH values of 6.8 to 8.0. Under such conditions, parathion can be
degraded slowly by alkali catalyzed hydrolysis reactions. The rate of these reactions
increases with increasing soil pH. Parathion can also be biodegraded to 0,0-Diethyl
phosphoric acid. At a nearby site, it was shown that toxaphene will degrade
anaerobically if reducing conditions can be achieved in the soil. It has also been
observed that the loss of toxaphene by volatilization is enhanced by high soil
moisture content. Other data indicated that toxaphene will degrade in the
presence of strong alkali, by dechlorination reactions. This information can be used
in identifying monitoring constituents and in performing interim and definitive
corrective measures.
To test the feasibility of chemically degrading the contaminated soil, in_ silu,
laboratory bench-scale tests were performed. Two treatments were evaluated,
application of calcium oxide (quicklime) and sodium hydroxide (lye). Figure 15-1
shows that the pesticides were degraded by both of these strong alkalis.
Those responsible for the remedial measures felt that the hot zone was too
contaminated for in_ situ_ treatment to be effective over" reasonable time periods.
The upper 2 feet of soil from this area was excavated and transported to an
approved landfill for disposal. However, the 1.7-acre residential area was treated in.
situ. To promote degradation, approximately 200 g/ft2of sodium hydroxide was
applied using a tractor with a fertilizer-spreading attachment. A plow and disc
were used to mix the sodium hydroxide into the soil to a depth of 1.5 feet. At 70
days after the application; concentrations of ethyl parathion had decreased by 76
percent, methyl parathion by 98 percent, and toxaphene by 45 percent.
15-7
-------
1000
800 -
600
400 -
200 -
25
20
15
I
10
• NaOH
o CaO
Laboratory Bench Test, Ethyl Parathion Degradation
2 4 68
DAYS
.Laboratory Bench Test, Methyl Parathion Degradation
17.500
15,000
12,500
i
10,000
7,500
5.000
2,500
2 46 8
DAYS
Laboratory Bench Test, Toxaphene Degradation
Figure 15-1. Results of Laboratory Bench Test for Pesticide Degradation
Source: (from King M al., 1985).
15-8
-------
Case Discussion
Knowledge of the properties of a contaminant as well as its environment are
important in assessing the potential for degradation, and this information can be
used to identify monitoring constituents and conduct interim or definitive
corrective measures. it may be possible to alter the site's physical or chemical
characteristics to enhance degradation of contaminants, under appropriate
conditions, in. situ treatment of contaminated so\\s can be an effective corrective
measures method.
Reference
King, J., T. Tinto, and M. Ridosh. 1985. In Situ Treatment of Pesticide Contaminated
Soils. Proceedings of the National Conference of Management of Uncontrolled
Hazardous Waste Sites. Washington, D.C.
15-9
-------
CASE STUDY 3: SELECTION AND EVALUATION OF A SOIL SAMPLING SCHEME
Points Illustrated
• Sampling methodologies must be properly selected to most
appropriately characterize soil contamination.
• Statistical analyses can be used to evaluate the effectiveness of a chosen
sampling scheme.
Introduction
Selection of a sampling scheme appropriate for a soil contamination problem
is dependent on the objectives of the sampling program. A grab sampling scheme
may be employed; however, grab sampling can produce a biased representation of
contaminant concentrations because areas of gross contamination are most often
chosen for sampling. Random sampling can provide an estimate of average
contaminant concentrations across a site, but does not take into account differences
due to the proximity to waste sources and soil or subsurface heterogeneities. A
stratified random sampling scheme allows these factors to be considered and, thus,
can be appropriate for sampling. Depending on the site, additional sampling using
a grid system may be needed to further define the areas of contamination.
Facility Description
The example facility operated as a secondary lead smelter from World War n
until 1984. Principal operations at the smelter involved recovery of lead from scrap
batteries. Air emissions were not controlled until 1968, resulting in gross
contamination of local soils by lead particulates.
Land use around the smelter is primarily residential mixed with
commercial/industrial. A major housing development is located to the northeast
and a 400-acre complex of single family homes is located to the northwest. Elevated
blood lead levels have been documented in children living in the area.
15-10
-------
Program Design/Data Collection
Initial soil sampling was conducted at the lead smelter and in the surrounding
area to document suspected contamination. Sample locations were selected based
on suspected areas of deposition of airborne lead and in areas where waste
dumping was known to have occurred. High lead concentration were documented
in samples collected from these sources. Because data obtained in the exploratory
sampling program (grab sampling) were not adequate to delineate the areal extent
of contamination, a stratified random sampling scheme was developed.
Based on wind rose data and the behavior of airborne particulate matter, a
sampling area was selected encompassing a 2-mile radius from the smelter. Specific
sampling sites were selected using a stratified random sampling scheme. The study
area was divided into sectors each 22.5 degrees wide and aligned so that prevailing
winds bisected the sectors. Each sector was further divided into approximately one-
tenth mile sections. A random number generator was used to select first the
direction and then the section. Random numbers generated were subject to the
following restrictions: two-thirds of the sites selected had to fall in the major
downwind direction; both residential and non-residential sites had to exist in the
sector; sampling sections were eligible for repeat selection only if they were
geographically within 1/2 mile of the smelter or if the section contained both
residential and non-residential sites. Sites that were biased towards lead
contamination from other than the lead smelter were not sampled (e.g., gas
stations and next to roads). A total of 20 soil sampling locations were selected, 10 at
residences and 10 at non-residential sites such as schools, parks, playgrounds and
daycare centers.
Sample cores were collected using a 3/4-inch inner diameter stainless steel
corer. Total sample depth was 3 inches. A minimum of four and maximum of six
samples were collected at each sampling location within a 2 ft radius. Cores were
divided into 1 inch increments and the corresponding increments were composite
from each depth to make up one sample. This approach provided data on lead
stratification in the top 3 inches of soil. All samples were analyzed for total lead.
15-11
-------
The results of the stratified random sampling indicated several acres with over
2,000 ppm lead in the soil. To further define the extent of these areas, a grid
sampling plan was designed. Seven hundred and fifty foot increments were used.
The grid was oriented along the axis of the release. Both residential and non-
residential areas were sampled. At each grid point, four 3-inch cores were collected
30 m from the grid point in each major compass direction. The cores were
composited by depth as discussed above.
Program Results/Data Analysis
Analytical results from the soil sampling program indicated significant lead
contamination within the study area. Maximum concentrations observed were
2,000 ppm lead with a background level of 300 ppm. Krieging of the data from the
grid sampling plan was used to develop a contour map as shown in Figure 15-2.
Lead concentrations were highest northwest and southwest of the smelter.
Case Discussion
Because of the large area potentially affected by lead emissions, development
of a sequential sampling plan was necessary to determine the maximum soil lead
concentrations surrounding the smelter and the areas having elevated
concentrations. A grab sampling scheme was first used to confirm that soil
contamination existed. A stratified random sampling scheme was developed to
provide representative data throughout the study area. This type of sampling
allowed consideration of prevailing wind directions and the need to sample both
residential and non-residential areas. To further define areas of contamination, a
grid sampling plan was developed. From these data, lead isoconcentrations maps
were prepared delineating areas with elevated concentrations.
15-12
-------
-i r
Lead Isovalves
• SMELTER '
—1 , I ... I.——1 II 1- 1 ~r—-—l —r
EST1MATED LEAD CONCENTRATIONS-(719/9), IN SOIL .
..... . - , . . • SCALE
0 800
Feet
Figure 15-2. Isoconcentration Map of the Lead Concentrations in ppm
Around the Smelter
15-13
-------
CASE STUDY 4: SAMPLING OF LEACHATE FROM A DRUM DISPOSAL AREA WHEN
EXCAVATION AND SAMPLING OF DRUMS IS NOT PRACTICAL
Points Illustrated
• It is not always possible to perform waste characterization prior to
establishing the RFI monitoring scheme because the waste may not be
directly accessible, as in the case of buried drums.
• When direct waste characterization is not practical, release monitoring
should be performed for the constituents listed in Appendix B of Volume
I of the RFI Guidance.
Introduction
Insufficient waste characterization data existed for a former drum disposal
facility that was suspected of releasing contaminants into the subsurface
environment. Leachate within the disposal pit was sampled and analyzed for all
constituents listed in Appendix B of Volume I of the RFI Guidance. The resulting
information was used to determine the major waste constituents to be monitored
during the RFI.
Facility Description
The unit of concern was a pit containing an estimated 15,000 drums. Due to
poor recordkeeping by the facility operator, adequate information regarding the
contents of the drums was not available. It was also not known if the drums were
leaking and releasing contaminants to the environent. Because insufficient data
existed regarding the drum contents, it was not known what constituents should be
monitored in nearby ground and surface waters. Due to the risk to workers and the
potential for causing a multi-media environmental release, excavation and
sampling of the drums to determine their contents-was not considered practical.
Instead, it was decided that leachate around the perimeter of the drum disposal pit
would be sampled to identify constituents which may be of concern.
15-14
-------
Program Design/Data Collection
To determine the physical extent of the buried drums, a geophysical survey
was conducted using a magnetometer. Borings were located at positions having
lower magnetometer readings than surrounding areas in order to minimize the
potential for drilling into drums.
Soil borings were drilled around the perimeter of the drum disposal pit, as
defined by the magnetometer survey. Drilling was accomplished using a hydraulic
rotary drill rig with a continuous cavity pump. Water was used as the drilling fluid.
To prevent surface runoff from entering the borehole and to control gaseous
releases from the borehole, primary and secondary surface collars were installed.
These consisted of 5-foot sectitons of 4-inch steel pipe set in concrete. A device to
control liquid and gaseous releases from the borehole was threaded onto the collars
to form a closed system (Figure 15-3)
Drilling was performed using a wireline operated tri-cone roller bit with a
diamond tipped casing advancer (Figure 15-4). Water was pumped down inside the
casing and out the drill bit, returning up the borehole or entering the formation.
The use of water to aid in drilling also helped reduce the escape of gases from the
borehole. Air monitoring showed no releases. Split-spoon samples were collected
at 5-foot intervals during the drilling and a leachate monitoring well was installed
at each boring location.
The soil and leachate samples were analyzed for the compounds contained in
Appendix B of Volume I of the RFI Guidance.
Program Results/Data Analysis
The leachate samples were found to contain high levels of volatile organic
compounds including 2-butanone, 4-methyl-2-pentanone, and toluene.
Concentrations were higher on the downgradient side of the pit.
15-15
-------
KELLY ROD
KELLY S1MVEL
KELLY HOSE
BALL-VALVE OPERATED
SAMPLING PORT
POTABLE
WATER
THREADED STEEL ADAPTER
RUBBER CASKET
ENCLOSED RETURN
TANK (200 GAL.)
/ SURFACE
4 FLEXIBLE PIPE
THREADED BALL-VALVE
(EMERGENCY SHUT-)N)
THREADED Vl.O. STEEL
SURFACE CASING
Figure 15-3. Schematic Diagram of Gas Control System Utilized at Pit
15-16
-------
WIRELINE CABLE
OVERSHOT LATCHING
DEVICE
CASING
RETRACTABLE 2 15/16"
TRI-CONE ROLLER BIT
W/ LOCKING INNER SUB
DIAMOND TIPPED CASING
ADVANCER (REAMING SHOE)
Figure 15-4. Schematic Drawing of Wireline Drill Bit and Reaming Shoe
15-17
-------
Case Discussion
Leachate sampling can be useful in determining whether buried drums are
leaking and to identify materials that are being released. This methodology can be
safer than excavation and sampling of individual drums. It can also identify the
more soil-mobile constituents of the leachate.
The data gathered in this case study were used in designing a monitoring
program, and the contaminants found were used as indicator compounds to link
downgradient ground-water contamination to this waste disposal unit.
15-18
-------
CASE STUDY 5: USE OF QUALITY ASSURANCE/QUALITY CONTROL (QA/QC) AND
DATA VALIDATION PROCEDURES:
Points Illustrated
•. A comprehepsive field and laboratory QA/QC program is necessary for
assessing the quality of data collected during an RFI.
• Timely validation of laboratory data can uncover problems correctable by
reanalysis or by resampling, thus preventing data gaps.
Introduction
A company in the mining and smelting industry sampled domestic wells and
surface soils in the vicinity of a tailings pile to monitor possible leaching of metals
into the aquifer and possible soil contamination due to wind-blown dust. Because
the data would be used to assess corrective measures alternatives and to conduct a
health and environmental assessment, the company chose to conduct both its
sampling and analysis efforts under a formal QA/QC Project Plan and to subject all
laboratory data to rigorous data validation procedures.
Facility Description
At this facility, a tailings pond had received smelter waste for many years.
Local water supply wells were potentially at risk due to percolation of water
through the pile and possible leaching of heavy metals. Local surface soils in nearby
residential areas (e. g., yards, public playgrounds) were also subject to
contamination from wind-blown dust originating from the pile during dry windy
weather.
Sampling Program
Before sampling began, a set of documents were drafted following U.S. EPA
guidelines (U.S. EPA 1978, 1980a, 1980b, 1981, 1982, 1985a, 1985b) that specified in
detail sampling sites and parameters to be measured, field and laboratory
15-19
-------
procedures, analytical laboratory protocols, and all field and laboratory QC checks
including frequencies, and corrective actions. The important elements of each
document are described below.
Standard Operating Procedures (SOPs)-
This document contained step-by-step procedures for the following items:
• Calibration, operation, and maintenance of all instruments used in the
field and laboratory.
• Equipment decontamination.
• Ground water and soil sampling, including compositing.
-,;•" Use of field notebooks and document control.
• Sample packaging, shipping and chain-of-custody.
Field Operations Plan (FOP)~
This document included the following:
• Rationale for choice of sampling locations, sampling frequency, and
analytes to be measured
• List of sampling equipment and SOPs to be used for each sampling event.
• List of field QC checks to be used and their frequency for each sampling
event.
• Health and safety issues and protective measures for field personnel.
• Sampling schedule.
15-20
-------
Laboratory Analytical Protocol (LAP)~
This document included the following:
• Sample size, preservation, and analysis protocol for each analyte.
• List of laboratory QC checks, QC statistics to be calculated and their
. -,. control limits, and corrective actions for QC checks outside control limits.
• Detailed list of deliverable documents and their formats.
• Procedures for sample custody, independent audits, and general
laboratory practices.
QA/QC Project Plan (QAPP)-
This document gathered into one place the overall data quality objectives for
the sampling and detailed QC procedures needed to attain those objectives.
Included were:
• Quality assurance objectives in terms of precision, accuracy,
completeness, comparability, and representativeness.
,• Procedures for the screening of existing data.
• •, Data management, reduction, validation, and reporting.
• Overview of both field and laboratory QC checks and their frequencies,
control limits, and corrective actions.
.,•. Data assessment procedures.
Results
Five surface soil samples were taken in high traffic areas of two playgrounds
and three residential yards. Five tap water samples were collected at two public
15-21
-------
drinking fountains at the playgrounds and at the three private residences. The
analysis results, as received from the laboratory, are shown in Table 15-2. The data
indicated that a soil hot spot existed for cadmium, that elevated lead occurred at all
five soil stations, and that all of the domestic wells showed elevated levels of
mercury.
The laboratory data package was subjected to a thorough data validation, as
detailed in the QA Project Plan. The following information and QC results were
checked by examination of original documents or photocopies of the documents.
Sampling, Sample Shipping, Chain-of-Custody-
Copies of field and field laboratory notebook pages were examined to insure
that all SOPS were correctly followed, that there were no notations of anomalous
circumstances (such as sample spillage) that may have affected analysis results, and
that the samples were correctly preserved, packaged, and shipped. Copies of all
chain-of-custody forms, bills-of-lading, and sample analysis request forms were
examined to insure that chain-of-custody was not broken and that samples arrived
intact at the laboratory.
Laboratory Raw Data-
The QAPP had specified that one of the deliverables from the laboratory was
copies of all instrument readouts and laboratory notebook pages. The digestion
raw data were checked to insure that no holding time violations had occurred. This
is important for mercury because the holding time is only 28 days for aqueous
samples.
All raw calibration data were recalculated and tested against instrument-
calculated sample results. Recoveries of calibration verification standards and
continuing calibration standards were checked to-insure that all instruments were
correctly calibrated, were not drifting out of calibration, and were correctly
calculating raw analysis results.
15-22
-------
TABLE 15-2
RESULTS OF ORIGINAL SURFACE SOIL AND TAP WATER ANALYSES
Sample3
SOIL-1
SOIL-2
SOIL-3
SOIL-4
SOIL-5
WATER-1
WATER-2
WATER-3
WATER-4
WATER-5
Cd
14
7
<20C
19
1200
<50
<50
<50
<50
<50
Cu
5200
2400
720
680
1080
NA
NA
NA
NA
NA
Pb
800
400
4530
350
460
<30
<30
<30
<30
<30
Hg
N Ab
NA
NA
NA
NA
1.5
1.3
1.0
1.4
1.2
Zn
1200
190
70
350
420
NA
NA
NA
NA
NA
a Soils in units of mg/kg, water in ug/L.
b Not analyzed.
c Undetected at detection limit shown.
15-23
-------
Final analysis results were recalculated from raw data using dilution and
digestion factors, as summarized in the lab notebooks, and compared to the data
summary sheets. No transcription errors were found. However, the cadmium result
for SOIL-5 contained a calculation error, and the correct final result was 12 mg/kg
instead of the 1200 mg/kg reported.
Laboratory QC Checks-
The QAPP had specified that the laboratory had to analyze pre-digestion
duplicates and spikes, U.S. EPA laboratory control samples, and reagent blanks. The
laboratory QC" results are summarized in Table 15-3 and indicated accuracy and
precision well within U.S. EPA guidelines. The mercury preparation blank also
indicated that the tap water results were not due to laboratory digestion reagents
or procedures.
Field QC Checks-
As specified in the QAPP and FOP, the follgwing field QC samples were
included with each of the soils and tap water samplings: bottle blank, field blank,
standard reference material (SRM), triplicate, and an interlaboratory split to a
"reference" lab. The results are summarized in Table 15-4.
Although no U.S. EPA control limits or corrective actions exist for field-
generated QC checks, the results of their analysis can aid in the overall assessment
of data quality. The triplicate, SRM, and interlaboratory split analyses indicated
good overall analysis and sampling precision and accuracy. The field blanks
indicated the possibility of mercury contamination from one of the four possible
sources: the pre-cleaned bottles, the preservation reagent, the distilled water used
in the field, or an external contamination source such as dust. The high positive
mercury result in the water bottle blank eliminated all of these sources except the
first because the bottle blanks remained sealed throughout the sampling effort.
The laboratory was immediately called and, upon personal inspection, the
laboratory manager discussed the remnants of a broken thermometer bulb in the
plastic tub used to acid-soak the bottles. An unused bottle from the same lot and
15-24
-------
TABLE 15-3
LABORATORY QC RESULTS
Analyte ,
, , Cd
. Cu
Pb
Hg
Zn
Duplicate RPDa(%)
SOIL-2
18 :,
5
14
NA
7
WATER-4
NCf
NAh
NC
NC
NA
Spike Recovery b(%)
SOIL-2
]
, TOO
93
1:10
NA
85
WATER-4
98
NA
92
-103
NA
LCSC
.(%)
- 101
97
106
NA
99
Soil
Preparation
Blank<*
<5Q9
<100
<200
NA
<150
Water
Preparation-
Blanks
".;• <50;:'
NA
<30
<0.20
NA ; :
a RPD = relative percent difference = (difference/mean) xl 00. Control limits = ± 35% for
solids and ± 20% for aqueous samples.
b Spike Recovery = (spike + samle result) - (sample result) x100. Control limit = 75-125%
(spike added)
LCS = laboratory control sample. Control limit- 90-1, 10%. '"
mg/kg.
ug/l.
NC = not calculated-due-to one or both concentrations below detection limit.
Undetected at detection limit shown.
h NA = not analyzed.
15-25
-------
TABLE 15-4
FIELD QC RESULTS
Analyte
Cd
Cu
Pb
Hg
Z n
Triplicate
Cva(%)
SOIL-1
22
3
7
NA
1
WATER-1
NCh
N A1
NC
18
NA
SRM
Recovery b(%)
BCSS-f
8 3
94
9 7
NA
110
U.S. EPAd
105
NA
101
1 03
N A
Interlab.
RPD"(%)
SOIL-1
-12
0
14
NA
24
WATER-1
NC
NA
NC
19
NA
Field
Blanks'
SOIL-1
<50j
-------
still at the laboratory as well as two bottles washed in previous lots were analyzed.
The bottles previously washed contained no detectable mercury, and the bottle
from the same lot as used in the sampling effort contained 0.75 ug. The water
mercury data were rejected, and a second sampling effort using new bottles was
conducted. All of the new samples contained no detectable mercury.
Discussion
This case study demonstrates the need for the establishment of a formal
QA/QC program that not only specifies field QC protocols but also incorporates
thorough data package validation. In this instance, a potential hot spot was found
to be due to a calculation error, and potential mercury contamination of domestic
well water was found to be a result of using contaminated sample containers. In
the latter case, timely QA/QC review allowed for a speedy resampling effort which
could be done at this site. In situations where resampling is not possible, adequate
QA is crucial.
References
U.S. EPA. 1978 (revised 1983). NEIC Policies and Procedures. EPA-330/9-78-001-R.
U.S. EPA, National Enforcement Investigations Center, Denver, CO.
1978 (revised 1983). NEIC Policies and Procedures. EPA-330/9-78-001-R. U.S. EPA,
National Enforcement Investigations Center, Denver, CO.
U.S. EPA. 1980a. Interim guidelines and specifications for preparing quality
assurance project plans. QAMS-005/80. U.S. EPA, Office of Monitoring Systems an~
Quality Assurance, Washington, DC. 18 pp.
U.S. EPA. 1980b. Samplers and sampling procedures for hazardous waste streams.
EPA-600/2-80-018. U.S. EPA, Municipal Environmental Research Laboratory,
Cincinnati, OH.
U.S. EPA. 1981. Manual of qroundwater quality sampling procedures.
EPA-600/2-81-160. Robert S. Kerr Environmental Research Laboratory, Ada, OK. 105
PP.
U.S. EPA. 1986. Test methods for evaluating solid waste. SW-846. 3rd ed. U.S. EPA,
Office of Solid Waste and Emergency Response, Washington, DC.
15-27.
-------
U.S. EPA. 1985a. Contract laboratory program .statement of work. Inorganic
analysis, multi-media, multi-concentration S.OW. NO. 765. July, 1965. U.S. EPa,
Environ-mental Monitoring Support Laboratory, Las Vegas, NV.
U.S. EPA. 1985b. Laboratorydata validation. Functional guidelines for evaluating
inorganic analysis. October, 1985. U.S. EPA, Office of Emergency and Remedial
Response, Washington, D.C.
15-28
-------
CASE STUDY 6: PRESENTATION OF DATA COLLECTED DURING FACILITY
INVESTIGATIONS
Point Illustrated
• Techniques for presentation of data for facility investigations involving
multimedia contamination.
Introduction
Data acquisition and interpretation are integral parts of facility investigations.
Depending on the size, complexity, and hazards posed at a particular site,
significant quantities of meteorologic, hydrologic, and chemical data can be
collected. To make the best use of these data, they should be presented in an easily
understood and meaningful fashion. This case study focuses on widely used and
easily implemented graphical techniques for data presentation.
Site Description
The site is a former copper smelter that ceased operation in the early 1980s.
During the operation of the smelter, large quantities of mine tailings were slurried
to tailings ponds that remain today (Figure 15-5). The tailings contain high solid
phase concentrations of inorganic contaminants such as copper, zinc, lead,
cadmium, and arsenic. In the Smelter Hill area, flue dust and stack emission
deposition have contaminated surficial soils. Numerous other units were operated
at the complex including an experimental plant designed to leach copper using
ammonia. The copper leach plant is shown in Figure 15-6. Three disposal ponds (I,
II, and III), received wastes slurried from the plant.
As a result of smelting and waste disposal practices, multimedia contamination
of ground water, surface water, and soils has occurred. Also, episodes of air
contamination have been documented due to entrainment of tailings during windy
periods.
15-29
-------
MILE / <(.
KILOMETER /
Figure 15-5. Map of the Smelter Site and Associated Tailings Ponds
-------
in
i
u>
KEY:
[23 BUILDINGS :;
| | WASTE STORAGE
0 235 650
Figure 15-6. Locations of Copper Leach Plant and Waste Storage Ponds
-------
Field Sampling and Data Collection
Data collection activities at this site were comprehensive. Over 100,000 pieces
of data were collected in the categories shown in Table 15-5.
Data Presen t a t i o n
This section illustrates a number of graphical techniques that can be used to
present data from facility investigations. Graphical presentations are useful for the
following general purposes:
Site feature identification, source identification, and mapping;
• Hydrologic characterization; and
i Water quality characterization.
For large sites, aerial photography is often very useful for defining the locations and
boundaries of waste deposits, and for establishing time variability of site
characteristics. Figure 15-6, for example, was developed from aerial photographs at
a 1:7800 scale. Types of information obtained by comparing this photograph to one
taken 10 years earlier include:
I Pond III was originally constructed earlier than Ponds I and II, and was not
lined. Ponds I and II were lined.
i The red sands (a slag deposit) shown in Figure 15-6 are present only north
of the railroad tracks. Earlier photographs showed that the red sands
extended to the highway, but were leveled and covered with alluvium
during construction of the copper leach plant.
This type of photographic information is valuable for locating waste deposits,
estimating quantities of wastes, and determining waste proximity to sensitive areas.
15-32
-------
TABLE 15-5
SUMMARY OF DATA COLLECTED
Category
Ground Water
Surface Water and
Sediment
Alluvium9
Soil3
Tailings3
Slag and Flue Dust*
Miscellaneous
Parameters
Water level elevations, potentiometric heads
Concentration of Al, Sb, As, Ba, Be, Bo, Cd, Ca, Cr,
Co, Cu, Fe, Pb, Mg, Mn, Hg, Mo, Ni, K, Se, Ag, Na,
Sn, V, Zn, P, CI, F, SO4, pH, 02, EC, Eh, Alkalinity,
TDS
Flow rates, bed particle size distributions,
suspended solids concentrations, dissolved
concentrations of same inorganic parameters as
ground water
Moisture content, soil, pH, EC, Sb, As, Cd, Cu, Fe,
Pb, Mn, Se, Ag, Zn, particle-size distribution
Cd, Cu, Fe, Pb, Mn, Ni, Zn, Sfa, As, Cd, Cr, Hg, Se,
Ag, Zn, particle-size distribution, Eh, S, TOC
Sb, Ar, Be, Cd, Cu, Fe, Pb, Mn, Ag, Se, Zn, particle
size, moisture, pH, EC, sulfur, carbonate
Sb, As, Cd, Cu, Fe, Pb, Mn, Se, Ag, Zn, SO4, EC, pH,
alkalinity
Meteorology, aerial photographs and other
photographic documentation, well log data,
surface topography, volumetric surveys of waste
piles
. Element data are sol id phase.
15-33
-------
For sites with complex hydrologic interaction, it is often helpful to graphically
represent the flow system. Figure 15-7 illustrates the surface water system at the
site. The diagram is useful because it shows the-hydrologic interconnections of the
drainage system.
For the ground-water system, flow direction and velocities provide
information needed for contaminant transport predictions. This information is
generated by plotting water levels on a site map, and then drawing contours
through points of equal elevation. An example is shown in Figure 15-8. Because the
contours form a relatively simple pattern in this case, they were drawn by hand.
However, computer-based contour packages exist that could be used to plot more
complicated contour patterns.
Inferred flow directions are also shown in Figure 15-8. From a knowledge of
the hydraulic gradient, hydraulic conductivity and effective porosity, the average
linear velocity can be calculated as shown in the upper left hand corner of the
figure. A velocity of 79 m/yr is calculated, for example, which means that
approximately 126 years would be required lot conservative solutes to move across
the site (approximately 10,000 meters).
Water quality data can be presented as shown in Figure 15-9. This figure
shows the spatial distribution of calcium, sulfate, and TDS at key surface water
stations. This data preservation method provides a broad areal view of these
parameters
Time series plots are useful for showing temporal variations in water quality.
For example, time trends of S04at three ground-water monitoring locations are
shown in Figure 15-10. Well 19 is slightly downgradient from the source, and the
high S04 levels reflect that the well is receiving solutes generated within the source.
Wells 26 and 24 are further upgradient, and reflect better water quality conditions.
The plot indicates that variability between stations generally is more significant
than time variability at a given location. One exception is at well 24 where a
temporary increase in sulfate levels was noted in 1975-76.
15-34
-------
UNCAGED
DIVERSIONS
RANGUS
CREEK
HILL
CREEK
GOLDEN
CREEK
NEW
LIME
DITCH
SOUTH
DITCH
OLD
LIME
DITCH
SEWAGE
DITCHES
NORTH
DRAIN
DITCH
'DECANT '
! DITCHES/*"
BYPASS
COLD
CREEK
GARDINER
DITCH
UNCAGED
DIVERSIONS
GREEN
RIVER
PONDS
Figure 15-7. Schematic of Surface Water System
15-35
-------
ui
i
u>
8 =
AH
H = water level
x = distance inflow direction
K - hydraulic conductivity
'Example:
AH = 300' for Ax = 5 mi
K =• 10
-2
GROUHDUMER fltVAIION
IN lltt ABOVt SU UVtl
Figure 15-8. Ground-Water Flowtines Based on Measured Water Levels
-------
r"
1
m
'%
P
I
WS-6
(Ji
WC-11
MC-9
Figure 15-9. Selected Surface Water Quality Parameters at Key Stations
-------
2400-1
2000-
1600-
O
CO
1200
800
400
• Well 19
o Well. 24
• Well 26
i i
1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985
Year
Figure 15-10. Changes in Sulfate Over Time at Selected Wells
Located Within the Site
15-38
-------
To identify leachate and soil interactions beneath a waste site, trenches may
be dug. The trench walls are then logged and photographed. Detailed sampling
may be done at closely spaced intervals to confirm that reactions such as
precipitation have occurred. Figure 15-11 shows a cross-section of a tailings deposit
that was developed based on a trench excavated through the tailings into the
underlying alluvium. The plot shows the demarcation between wastes and natural
alluvium.
Figure 15-12 shows the detail of the chemical composition of one borehole
through the tailings and into the underlying alluvium. The chemical composition is
shown to vary significantly with depth. These types of plots contain a wealth of
chemical information that can help to explain the geochemical processes operative
in the tailings. Figure 15-12 also shows the marked contrast between the
composition of the tailings (in the top 16 feet)and the underlying galluvium.
Summary
The graphical presentations illustrated in this case study are a few of the many
techniques available. With the proliferation of graphical packages available on
microcomputers, scientists and engineers have a wide range of tools available for
data presentation. Some of these tools are summarized inTable 15-6.
15-39
-------
GROUND
cn
*»
O
WITH PLANT MATTER
'///mil I n rLANt MAI I tH
miutmMVMfL
DARK ORANGE
• MIXED BLACK SLAG
AND CRUSHED ROCK
• Fe OXIDE AT
INTERFACE
• .••rSsNJ
CWQ
'r^j-iom c VJ
£»£*&
'*<>*aV__
9X&&&SF
DARK GRAY SOIL WITH
PLANT MATTER
%^^&&?&%W3S' PLANT MATTER
S S-W^^ b^^Cr^ tNTERBEDDED PALE GRAY J
^^.*O-2'^»^toofe «V*o AND ORANGE SAND
"•^.-^* ^%. ^»~ :^^*_ .'F.*^^. l^\' ' •^'_-* " *fc."\ '• - ' ' -
^03
^ o
•°.'0Vs>
.-O*
FERRICRETE
HARDPAN
.
r?.
0
I
FEET
Notw: A BLUE-GREEN PRECIPITATE FORMED ON THE EXPOSED SURFACE OF TOP 6 INCHES OF ALLUVIUM.
THIS PRECIPITATE WAS NOT THERE WHEN TRENCH FIRST DUG.
Figure 15-11. Field Sketch of Tailings Trench T-3
-------
BORE HOLE BSA
Cl Pond
uit*
Sutlil. SuNut (V.)
Iron (gA(g|
Ul
(gA(g|
)W m •>0> m 1000 IMO
Ar»«nic
» B « »
Cadmium
unooiUDonoaanooonocniiaxi
Copp«r
joo m too KO ion »» i4a> iuo
?o«i «ooo tan »aa loom inxn
l» » »J W »
Zinc |mg4(g}
Cta,
l~l Ultingi
[ID Aluvkim
Figure 15-12.
Depth vs. Concentration Profiles for Selected Variables
for Borehole 88A ; \ i .
-------
TABLE 15-6
TYPICAL METHODS FOR GRAPHICALLY PRESENTING DATA COLLECTED
DURING FACILITY INVESTIGATIONS
Data
METEOROLOGICDATA
Wind speed and direction
Air temperature
Precipitation
Evaporation
SURFACE WATER DATA
Flow rates
Water quality
it
GEOHYDROLOGIC DATA ,
~
GROUND-WATER DATA
MISCELLANEOUS
Graphical Presentation Methods
-"•'; . „ , -;•• . ."-;--".
• Wind rose showing speed, direction and percent of
observations for each 10° increment
• Bar chart, by month
• Bar chart, by month
• Bar chart, by month
• Hydrographs; distance profiles, cumulative frequency
distributions, flood frequency plots
Hydrologic network depiction and water budgets
Trilinear diagram
Stiff diagrams
Contour showing vertical concentration or temperature
variability in two deep water bodies
Time history plots showing daily/annual variability
Bar charts of major cations/anions or contaminants at
multiple locations shown on a single map
• Geologic map of site and vicinity
• Stratigraphic cross-sections of site in direction of and
perpendicular to ground water flow
• Well logs
• Cross-sections near waste deposits
* Solid phase chemical analyses by depth at borings near
.waste deposits and into alluvium
Water level contours
Flow directions and velocities
Time history of water table at important locations
Stiff diagrams
Trilinear diagrams
Contaminant plumes, showing isopleths
• Figures with irnportant site features, including waste
sources, storage ponds, disposal areas, buildings,
,. sampling locations, well locations
• Operational aspects for special sampling equipment
(e.g., lysimeters)
15-42
-------
CASE STUDY 7: CORRELATION OF CONTAMINANT RELEASES WITH A SPECIFIC
WASTE MANAGEMENT UNIT USING GROUND-WATER DATA
Point Illustrated
• Development of an effective ground-water monitoring program can tie
releases of contaminants to specific waste mangement units.
Introduction
Documentation of a release from a specific waste management unit may
require the development of a comprehensive ground-water monitoring program
coupled with an extensive hydrogeologic investigation. Determination of ground-
water flow direction and horizontal and vertical gradients are necessary to assess
the direction of potential contaminant migration. Historical data on wastes
disposed in specific units can provide information on contaminants likely to be
detected downgradient.
Facility Description
Chemicals were manufactured at a 1000-acre facility for over 30 years.The
facility produced plastics including cellulose nitrate, polyvinyl acetate, polyvinyl
chloride and polystyrene, and other chemicals such as phenols and formaldhyde.
Wastes produced in the manufacturing process were disposed on site in an
unlined liquid waste impoundment and in two solid waste disposal areas. Readily
combustible materials were incinerated in four burning pits. Ground-water
contamination has been documented at the site. Figure 15-13 shows, the facility
plan and locations of ground-water monitoring wells.
The site is located in a glacial valley and is adjacent to a major river. A minor
tributary runs through the southwestern portion of the facility and drains into the
river. Approximately 200 dwellings are located downgradient of the site.
15-43
-------
BURNING PIT C ®
A;!-
ui
t
LIQUID WASTE DISPOSAL AREA
' '
(SOLID WASTE
> DISPOSAL
AREA
N
Kev
Groundwater Monitoring Well
Scale
I ' '
0 200 400
Feet
SOLID; WASTE DISPOSAL AREA 1
Figure 15-13. Location of Ground-Water Monitoring Wells
-------
Data Collection
Initial studies to assess the extent of ground-water contamination began in
1981. Studies focused on ground water in the vicinity of various wasted disposal
units. A limited numberof monitoring wells were installed in 1983. These wells
provided general data on the direction of ground-water flow and chemical
constituents that had entered the ground water. In 1984, a two-phased approach
was developed to define the areal and vertical extent of contamination and to
identify, contaminant, releases from specific waste management units. The first
phase involved the characterization of facility geologic and hydrogeologic
conditions using historical data, determination of the chemical nature of
contaminants in the ground water using existing monitoring wells and
development of a contaminant contour map delineating horizontal boudaries
of contamination. Based on this data, 33 soil borings were drilled in Phase 2. The
goals of the second phase were to 1) detail subsurface geologic characteristics,
vertical and horizontal water flow patterns, contaminant migration, and site-
specific chemical contaminants; and 2) install wells that would be used to monitor
contaminants being released from ail units of concern at the facility.
Continuous split spoon samples were collected in each boring and headspace
analyses for volatile organic compounds (VOC) were conducted on-each sample.
Chemical constituents were identified using a field gas chromatograph.
Confirmational analysis by GC/MS were conducted on selected samples.
Geotechnical analyses were also conducted on the split spoon samples.
Chemical and hydrogeologic data (direction of flow, gradients) obtained from
the borings were used to select appropriate ground-water monitoring well
locations and screen depths. Fifty-two (52) nested monitoring wells were if installed
at 25, locations upgradient and downgradient of each waste management unit, and
near the river and its, tributary. Screen depths were determined by the depth of
maximum VOC contamination observed in the borings and the permeability of soil
layers.
15-45
-------
Data Analysis
Ground-water contamination data from new wells, coupled with historical
waste disposal data allowed releases from three specific waste management areas
to be defined. Sample analyses showed organic solvents in nearly all locations.
However, more unusual constituents associated with specific manufacturing
progesses were detected in some samples, allowing them to be correlated with
releases from specific waste management units The two situations below illustrate
how these correlations were accomplished:
1) PCBs detected in some samples were correlated with Solid Waste Disposal Area
#1. This area received construction debris, resins, plastics, metals, drums, and
PCB containing transformers. Records indicated that this unit was the only
location where transformers were disposed onsite. PCBs could not be
associated with any of the other waste management units.
2) The solvent dimethylformamide (DMF) detected in some samples was
correlated with Burning Pit B. It was discovered that the building that housed
this unit had been used to tint windshields and that DMF is a component of
the dye used in this process. DMF could not be tied to any of the other waste
management units. A leachfield in which waste dyes had been disposed was
discovered under the building and the contamination was traced back to that
source.
Case Analysis
An extensive hydrogeologic investigation of the facility was completed and, in
conjunction with historical, data, was used to develop a comprehensive ground-
water monitoring program. Placement of the monitoring wells of the screens was
essential in providing data that unequivocally linked contaminant releases to
specific waste management units and manufacturing processes.
15-46
-------
CASE STUDY 8: WASTE SOURCE CHARACTERIZATION FROM TOPOGRAPHIC
INFORMATION
Points Illustrated
• Mapping of changes in site topography can support the selection of
locations for test borings and monitoring wells.
• This technique is especially useful at sites where large volumes of waste
have been disposed of over several years.
Introduction
Topographic surveys conducted prior to and at different times during the
operation of a waste management facility can be used to help characterize the
vertical and horizontal extent of waste disposal areas. Because the resolution of
this technique is limited, it is most-useful when large volumesof waste are involved.
Facility Description
This facility is the same as discussed in Case Study 7 above.
Topographic Survey
In 1984, a topographic survey measuring elevations in feet relative to mean
sea level was conducted for the areas shown in Figure 15-14. These elevations were
plotted on a map of appropriate horizontal scale and contoured in 2-foot intervals.
This topography was transferred to an existing site plan (horizontal scale 1" to
200'). Topographic maps from 1935 (showing the natural topography before waste
deposition) to 1960 (showing the topography in the earlier stages of the facility
operation) were compared to the 1984 map. By examining the changes in
elevations which occurred over time, contours were developed showing' the
estimated changes in vertical and horizontal units of the liquid waste and solid
waste disposal areas.
1 5-47
-------
BURNING PIT C
LIQUID WASTE DISPOSE
I SOLID WASTE
V DISPOSAL
AREA 2
SOLID WASTE DISPOSAL AREA 1
N
$ BURNING PIT
WASTE DISPOSAL
LANDFILL
Scale
0 200 400
Feet
Figure 15-14. Site Map Showing Waste Disposal Areas
-------
Results
From the analysis, it was apparent that the deepest portion of Solid Waste
Disposal Area (SWDA) No. 1 (Figure 15-14) was approximately 48 feet, and the
Liquid Waste Disposal Area (LWDA) was approximately 30 feet deep. The
horizontal limits of the disposal areas were also defined in part by this review, but
other field surveys provided more accurate information on the horizontal
boundaries of the waste disposal areas.
Case Discu s s i o n
Topographic surveys can provide useful information for characterizing
disposal areas. The results of these studies can facilitate the selection of
appropriate test boring locations, and may reduce the number of borings necessary
to describe the subsurface extent of contamination. It should be noted that
techniques such as infrared aerial photography and topographic surveying are
approximate in their findings. They are useful methods in the early phases of an
investigation, but do not replace the comprehensive characterization of the
environmental setting needed for the full investigation.
15-49:
-------
CASE STUDY 9: SELECTION OF GROUND-WATER MONITORING CONSTITUENTS AND
INDICATOR PARAMETERS BASED ON FACILITY WASTE STREAM
INFORMATION
Points I 11 u s t r a t e d
• Waste stream information can be used to identify potential
contaminants and thus to select appropriate ground-water monitoring
constituents and indicator parameters.
• The number of initial monitoring constituents analyzed may be
significantly reduced from the 40 CFR Part 261 Appendix VIII list when
detailed waste stream information is available.
Introduction
Hazardous waste treatment, storage, and disposal facilitity subject to RCRA
are required to identify all waste streams managed the facility, waste volumes,
concentrations of waste constituents, and the waste management unit in which
each waste type is disposed. Ground-water monitoring programs should be
developed to adequately monitor contaminant migration from each unit.
Constituents to be analyzed in the ground-water monitoring program should be
established prior to sample collection. When waste stream data are not available,
the full set of Appendix VIII monitoring constituents may be required to
characterize ground-water contamination. Knowledge of the waste streams
managed by a facility simplifies the selection of monitoring constituents and
indicator parameters because potential contaminants and their likely reaction and
degradation products can be more easily identified.
Facility Description
The 600-acre facility is a permitted waste disposal site operated since 1980.
Solid waste management units occupy 20 acres of the site and include four surface
impoundments and one container storage area subject to RCRA. Until 1985, three
units (two surface impoundments and one solids disposal unit) not subject to RCRA
were used for geothermal waste disposal. However, the two surface impoundments
15-50
-------
were replaced by a RCRA regulated landfill. RCRA wastes managed by the facility
include metals, petroleum refining-wastes, spent non-halogenated solvents,
electroplating wastewater treatments sludge, spent pickle liquor from steel finishing
operations, and ignitable, corrosive, and reactive wastes. Ground-water monitoring
wells have been installed downgradient of each waste mangement unit.
Program Design
Prior to disposal, each load of waste received is analyzed in an on-site
laboratory to provide a complete characterization of waste constituents. Periodic
sampling of the waste management units is, also conducted to identify waste
reaion products and hazardous mixtures. Even though the incoming wastes have
been characterized, the facility owner also analyzed initial ground-water samples
from each monitoring well far all Appendix, VIII constituents. The resulting data
were used to establish existing concentrations for each constituent and to select a
set of monitoring constituents and indicators parameters to identify migration of
waste to the ground-water system. Table 15-7 includes a list of the indicator
parameters analyzed at the facility. Rationale for indicator parameter selection are
included in this table. A separate list of hazardous constituents to be monitored
was also developed based on the waste analysis.
Because the facility accepts only a limited number of 40 CFR Part 261 Appendix
VIII constituents and initial monitoring verified the absence of many constituents,
the facility owner or operator was able to minimize the total number of
constituents monitored in ground water. The process of constituent elimination is
dependent on the actual wastes received by the facility and the physical and
chemical properties of these constituents that influence their migration potential
(e.g., octanol/water partition coefficients, volubility, absorptivity, susceptibility to
biodegradation).
Non-halogenated solvents have relatively low partition coefficients
(Kow: benzene = 100; toluene = 500) and are not readily retained by soils.
Conversely, polycyclic aromatic hydrocarbons, constituents of petrochemical wastes,
have very high partition coefficients (e.g., chrysene = 4x105) and are generally
immobile in soils. Migration rates of metals are also influenced by the exchange
15-51
-------
TABLE 15-7
INDICATOR PARAMETERS
Parameter
Total Organic Carbon (TOC)
Total Petroleum Hydrocarbons
Total Organic Halogen (TOX)
Nitrates
Chloride
Sulfides
pi-i
Total phenols
Citeria for Selection
Collective measure of organic substances present
Indication of petroleum waste products
Halogenated organic compound are generally
toxic, refractory, and mobile
Mobile contaminant, degradation product of
nitrogen compounds
Plating solution constituent, highly mobile in
ground water, Early indicator of plume arrival
Toxic biodegradation by product, strong
reducing agent, may immobilize heavy metals
Good indicator of strongly acidic or alkaline waste
leachates close to sources.
Collective measure of combounds likely to be in
waste. Even small concentrations can cause
olfactory problems following water treatment by
chlorination.
15-52
-------
capacity of the soil. Different metal species are sorbed to different extents.
Following an assessment of the migration potential of each waste constituent, the
need for analysis of that constituent can be prioritized.
Case Discussion
Waste stream information was used to determine appropriate monitoring
constituents and indicator parameters. The use of the existing initial ground-water
quality data and the incoming waste analyses allowed for prediction of
contaminants of concern in ground water and reduced the number of constituents
requiring analysis.
15-53
-------
CASE STUDY 10: USING WASTE REACTION PRODUCTS TO DETERMINE AN
APPROPRIATE MONITORING SCHEME
Point Illustrated
• It is important to consider possible waste reaction products when
developing monitoring procedures.
Introduction
Volatile organic priority pollutants have been detected in ground water at
various areas across the country. These compounds, widely used as solvents, are
generally considered environmentally mobile and persistent. Increasing evidence,
however, indicates that chlorinated solvents can be degraded under anaerobic
conditions by reductive dehydrochlorination. The sequential removal of chlorine
atoms from halogenated 1 and 2 carbon aliphatic compounds results in formation
of other volatile priority pollutants which can be detected during investigations of
solvent contamination.
Facility Description
The facility is a small municipal landfill sited on a former sand and gravel
quarry. In addition to municipal wastes, the landfill accepted trichloroethane and
tetrachloroethene contaminated sludge from a local fabrication plant until 1975. In
1983, a municipal well located downgradient of the landfill tested positive for
dichloroethane, dichloroethene isomers, and vinyl chloride. This prompted the city
to investigate the cause and extent of the problem.
Site Investigation
According to records kept at the facility, some of the compounds found in the
municipal well were not managed at the facility. This prompted the city to request
that a monitoring program be developed to determine whether another source was
causing well contamination. A careful search of the city records, however, failed to
indicate a credible alternative source of the compounds. Suspecting that the
landfill was the source of the well contaminants, five monitoring wells were
15-54
-------
installed (Figure 15-15) and water samples were analyzed for halogenated
compounds using EPA Method 601. The results, given in Table 15-8, show an
increase in degradation products of trichlorciethane and tetrchloroethene with
increasing distance from the landfill. Using these data, supported by hydrogeologic
data from the monitoring wells, the munitipal landfill was shown to be the source
of the observed contamination.
TABLE 15-8
RESULTS OF MONITORING WELL SAMPLING
v .
Chlorinated Ethanes
(l)Trichloroethanes
(2) 1,1-Dichloroethane
1,2-Dichloroethane
Chloroethane
Chlorinated Ethenes
(1) Tetrachloroethene
Trichloroethene
(2) 1,2-Dichloroethenes
1,1-Dichloroethene
Vinyl Chloride
WELL NUMBER (See Figure 1 5-15 for
well locations)
1
10(3)
71
NO
ND
80
12
ND
ND
ND
2
68
240
12
21
13
100
990
ND
120
3
-•
ND(4)
130
21
18
ND
62
950
ND
59
4
ND
11
ND
160
ND
ND
150
ND
100
5
ND
13
ND
ND
ND
ND
ND
ND
ND
(1) Parent Compounds
(2) Degradation Products
(3) All Concentrations In Micrograms/L
(4) ND means < 10 Micrograms/L
Case Discussion
Based on the compounds found in the municipal well, the city believed that
the municipal landfill could not be the source of the contamination. If this
reasoning had been followed, then a system of monitoring wells might have been
needlessly installed elsewhere in the attempt to find an alternate source of the
15-55
-------
.Property. Line
.*
//General
Approximate Scale L'>50b'
// Direction
V Of
" Ground Water
• 'Flow
.Municipal
MOTE: .Locations of nearby industrial
facilities not shown.
Figure 15-15. Site Map and Monitoring Well Locations
15-56
-------
contamination. Instead, after carefully researching local industries] it was
determined that the landfill was the most reasonable source of the contamination
and that the observed well contaminants were probably degradation products of
the landfilled solvents. The progressive dehalogenation of chlorinated.ettiaries-a—
ethenes, as shown in Table 15-8, is commonly encountered in situations where
chlorinated solvents are subjected to anaerobic conditions (Wood, 1981). Different
degradation reactions may occur when pesticides are subjected to acidic or alkaline
conditions or biological degradation. Therefore, it is important to keep reaction
products in mind when designing any monitoring scheme or interpreting
contamination data.
Reference
Wood, P. R., R.F. Lang, I.L. Payan, and J. DeMarco. 1981. Anaerobic Transformation.
Transport and Removal of Volatile Chlorinated Organics in Ground Water.'" First
International Conference on Ground Water Quality Research, October 7-. 1:0,': 1981",
Houston, Texas.
15-57
-------
CASE STUDY 11: CORRECTIVE: MEASURES STUDY AND THE IMPLEMENTATION
OF INTERIM MEASURES
Points Illustrated
• Interim corrective measures may be necessary to protect human health or
the environment.
• The evaluation of the need for definitive corrective measures.
Introduction
The development and implementation of a comprehensive Corrective
Measure Study can be a time-consuming process. Between the time of the
identification of a contaminant release and the completion of definitive corrective
measures, existing conditions or contaminant migration can endanger human
health or the environment. Under such conditions interim measures may be
necessary. The case study presented below illustrates the implementation of
interim measures to reduce contaminant migration and to remove the imminent
threat to the nearby population from exposure to contaminants in drinking water,
and also illustrates the decision making process as to whether definitive corrective
measures may be necessary.
Facility Description
The facility in this case study is an underground tank farm located at a
pharmaceutical manufacturing plant. The tank farm encompasses an area
approximately 140 feet by 260 feet and contains 30 tanks ranging in size from
12,000 to 20,000 gallons. The tanks are used to store both wastes and raw materials
for the various batch manufacturing processes performed at the plant. Typical
wastes include carbon tetrachloride, acetonitrile and chloroform. At the time of the
release, the tank farm had no cap to prevent the infiltration of rainfall or runoff. It
also did not have berms to provide containment for surface spills. No leak detection
or leachate collection systems were present.
15-58
-------
Geological and Hydrological Setting
The site is underlain by silty soil overlying limestone. The weathered limestone
beneath the site is very: permeable (up to 210 ft/day) due to the solution of rock
along joints and bedding planes in the limestone. Depth to the limestone varies
from 3 to 80 feet beneath the tanks and from 15 to 190 feet downgradient of the
site.
The ground-water system beneath the site consists of two aquifers. The uper
one, an unconfined limestone aquifer, is about 300 feet below the surface. The
deep aquifer is an artesian aquifer, in another limestone formation about 200 feet
below the land surface. Ground-water flow in the upperaquife is controlled by
both the regional flow system and local channelized flow through solution
conduits. The upper aquifer discharges to a canal 3 miles north of the site. Figure
15-16 shows the ground-water elevation contours in the vicinity of the site.
Regional average ground-water flow velocity was estimated at 4ft/day, but ground-
water velocities on the order of 50 ft/day have been measured in some channelized
areas. Channlized flow is also is responisible for local deviatiorrs in flow direction.
Release Characerization
A contaminant release from the tank farm was discovered when one of the
tanks used for waste storage was found to be empty. The waste sored in the tank
was predominately carbon tetrachloride (CCU) (a carcinogen with an MCL of 0:005
mg/l, with some acetonitrile (a systemic toxicant with a water-based health, criterion
of 200 ug/l) and chloroform (a systemic toxicant with a water-based health criterion
of 400 ug/l) reference dose (RfD) is 0.4 mg/l). Approximately 15,000 gllons of waste
liquids had been routed to the tank before the leak was discovered. Excavation of
the tank revealed ruptures in at least three locations. Initial ground water
monitoring after the tank rupture was discovered identified CCU in a well 2500 feet
downgradient of the site, at concentrations above the MCL for CCU of 0.005 mg/l.
Contaminants from the leaking tank were found to have dispersed laterally
within a two-foot-thick sand bed which underlies the tanks. The contaminated area
15-59
-------
®MW22, f.
3.2V3-4
4:0-
®3.19
Elevation above
Mean Sea Level
Groundwater ,
Flow Lines
Grouhdwater
Monitoring Well
.(and-Elevin Ft.) .
Contours based on Water Levels
taken.on 5/2/84
(Contour Interval 0.2 Ft )
®MW1
Figure 15-16. Ground-Water Elevations and Flow Directions in Upper Limestone
Aquifer
15-60
-------
was approximately 5600 ft2. High levels of CCI4were found throughout the sand
layer. Concentrations of CCIJn the natural soil ranged between" undetected and
2200 mg/kg. Observed consentrations were well above the soil RSD for CCI4(2.7
mg/kg). Concentrations generally decreased with depth due to adsotptioh onto the
clay particles in the soil. Carbon tetrachloride apparently moved downward with
little lateral dispersion until reaching the soil-limestone interface. Upon reaching
the unsaturated limestone, the contaminants then appeared to have rapidly
dispersed over an area of about 12 acres before entering the aquifer.
Interim Corrective Measures
Immediate action to contain the release in the aquifer was taken. This
involved pumping the well where CCI4had been found continuously at its full
capacity of 450 gpm.
All drinking water in the vicinity of the release was obtained from wells
installed in either the shallow or artesian aquifer. Immediately after the detection
of the release, all domestic and industrial wells north of the facility were tested for
CCI4contamination. Test results showed contamination of several shallow water
supply wells. Based on this information and the inferred ground-water flow
direction to the north-northeast, wells serving two small communities and a nearby
motel were closed. The facility operator hired all mobile water tanks available and
supplied water for immediate needs until a temporary water supply could be
implemented. Water from an unaffected artesian well was then used to supply
water to these communities.
The design and operation of the tank farm was altered in an attempt to avoid
similar problems in the future. A fiber-reinforced concrete cap was installed over
the tank farm to prevent the infiltration of rainfall and runoff, thus minimizing
further contaminant migration in the soil. The ruptures were repaired, and a tank
monitoring system was also developed and implemented at the site.
Definitive Corrective Measures: Saturated and Unsaturated Zones
A comparison of CCI4concentrations within the ground water to the MCL for
CCI4(0.005 mg/l) indicated that definitive corrective measures may be necessary.
15-61
-------
Due to the high mobility of CCI4within the unsaturated zone, and the potential for
continued inter-media transfer, from this zone to the ground water, definitive
corrective measures for both the saturated (ground water) and unsaturated zones
should be evaluated in a Corrective Measures Study (CMS).
Case Discussion
The development and implementation of definitive corrective measures at a
site may take a substantial length of time. Depending on the nature and severity of
the release and the proximity of receptors, interim measures, such as alternative
water supplies, were required to minimize the effects on human health and the
environment. Comparison of constituent concentrations with health and
environmental criteria indicated that definitive corrective measures may be
necessary and that a Corrective Measures Study (CMS) should be initiated.
15-62
-------
CASE STUDY 12: USE OF AERIAL PHOTOGRAPHY TO IDENTIFY CHANGES IN
TOPOGRAPHY INDICATING WASTE MIGRATION ROUTES
Points Illustrated
• Aerial photographs can be used to obtain valuable data on facility
related topographic features, including type of waste management
activity, distance to residences and surface waters adjacent land use and
drainage characteristics.
• Detailed interpretation of aerial photographs can identify acual and
potential waste migration routes and areas requiring corrective action.
Introduction
Stereoscopic pairs of historical and current aerial photographs were used to
assist in the analysis of waste management practices at a land disposal facility.
Stereo viewing enhances the interpretation of aerial photographs because vertical
as well as horizontal spatial relationships can be observed, and because the
increased vertical resolution aids in distinguishing various shapes, tones, textures,
and colors within the study area. Typical items that should be noted include pools
of unknown liquid that may have been released from buried materials which could.
migrate off site through drainage channels. Soil discoloration, vegetation damge
or enhanced vegetation growth can also be indicative of contaminant migration.
Facility Description
The site contains an active land disposal facility which receives bulk hazardous
waste, including sludges and contaminated soil for burial, and liquid waste for
disposal into solar evaporation, surface impoundments. Operations at the facility
began in 1969. Historical and current aerial photographs were reviewed to asses
waste management practices and to identify potential contaminant migration
pathways requiring further investigation and corrective action.
15-63
-------
Data Collection and Anaysis
Low altitude color aerial photographs of the facility (scale =" 1:8400) were
obtained in October 1983 and February 1984. The photos were interpreted by an
aerial photo analyst at the U.S. EPA Environmental Monitoring and Support
Laboratory at Las Vegas, Nevada. Figure 15-17 shows the analyzed photograph. The
interpretation code is givin in Figure 15-18. Analysis of the photograph indicates
several areas of seepage at the base of the surface impoundments. This seepage
indicates that either the impoundments are not lined or the liners have failed.
Drainage from the western portion of the facility which contains most of the
impoundments flows into a drainage reservoir formed by a dam across the main
drainage. Drainage from the northeast portion of the facility where seepage was
also observed appears to bypass this reservoir and enter the main drainage which
flows offsite. Besides possible surface contamination, this seepage also indicates
potential subsurface contamination.
The aerial photograph obtained in February 1984 (Figure 15-19) indicates the
continued existence of see page from the surface impoundments. There is evidence
of possible discharge from the drainage reservoir to a stream channel, as a pump
and piping were observed. Additional material in the solid waste disposal area has
altered the dranage pattern. At the south end of this area, seepage is evident in
association with damaged vegetation. Drainage from this area enters a drainage
system and appears to be diverted offsite.
Case Discussion
Analysis of aerial photographs of the land disposal facility enabled
investigators to identify potential contaminant sources and migration, pathways.
This information was used by investigators to identify areas for surface water,
sediment, soil, and subsurface sampling.
15-64
-------
< •* -/... •
^ ^ i" «*PU-j;^H»mi -*n
UNDERGROUND^
PIPELINE
WASTE MATERIAL
BEING REMOVED
AERIAL
SURVEY
MARKER
RESERVOIR
Figure 15-17. October 1983 Aerial-Photograph of Land Disposal Facility
15-65
-------
INTERPRETATION CODE
BOUNDARIES AND LIMITS '
i.i—i— FENCED STE BOUNDARY
^_^» UNFENCEO are BOUNDARY
.» i'» * *,. PENCE ''''','
— —— PROPERTY LINE
—,/— GATE/ACCESS POINT
. ^- SECTION CORNER
"I .1 -
DRAINAGE
»——- DRAINAGE •;.,".
* ' • FLOW DIRECTION
«—** INDETERMINATE DRAINAGE
TRAMSPORTATION/UTIUTY ,
= === = VEHICLE ACCESS
». ti. )- I- RAILWAY
........ PIPELINE ] "
•——— POWERLINE ; ;
STT6 F€ATURES ; .;
IIMIIHIIIII DIKE -
STANDING LIOU1O
q, STANDING UOUIO
(SMALL)
en-r^ EXCAVATION. PIT -
•~_J, (EXTENSIVE!
MOUNOKJ MATtBtAL
IEXTEMSIVE)
MOUNDED MATERIAL
(SMALL)
CR
0«-
HT
PT
VT
CA
:OG
FL
IM
LC
-------
TANK TRUCK
UNLOADING
AERIAL
SURVEY
MARKER
TANK
TRUCK
AERIAL
SURVEY
MARKER
DAMAGED
VEGETATION
tt
RESERVOIRS
AERML
SURVEY
^-MARKER
DISCHARGE
INTO STREAM
Figure 15-19. February 1984 Aerial Photograph of Land Disposal Facility
15-67
-------
CASE STUDY 13: IDENTIFICATION OF A GROUND-WATER CONTAMINANT
PLME USING INFRARED AERIAL PHOTOGRAPHY
Point Illustrated
Infrared photography can assist in identifying contaminant plumes and
in locating monitoring wells by showing areas of stressed vegetation and
contaminated surface water.
Introduction
Infrared aerial photography can assist in identifying contaminant plumes at
sites where little or no monitoring has been conducted. By identifying areas of
stressed vegetation or contaminated surface water, it may be possible to focus on
contaminant discharge points and roughly define the extent of a release.
Hydrogeologic investigations and surface water sampling can then be performed to
further characterize the release. Infrared photography offers the potential to
increase the efficiency of a sampling program.
Facility Description
The facility is a municipal solid waste landfill which has served a population of
22,000 for 30 years. The facility covers an area of 11 acres, holding an estimated
300,000 tons of refuse. The majority of waste in the landfill was generated by the
textile industry. Until July 1978, the facility was operated as an open dump with
sporadic management. City officials indicated that original disposal occurred in
open trenches with little soil cover. After July 1978 the facility was converted to a
well-operated sanitary landfill. Figure 15-20 shows the facility.
Geologic Setting--
The landfill is located on a sandy to silty till varying in thickness from 23 feet at
the hill crest to 10 feet on the side slope. A swamp is present at the base of the hill
at about 255 feet above sea level. There is a dam at the southern drainage outlet
15-68
-------
O BACKGROUND WELL B
METAL
BACKGROUND WELL
LIMIT OF REFUSE 7/78
N
SCALE: (APPROXIMATE)'
0 333' .564'
SWAMP
TREE LINE
• WATER
O WELL LOCATION
Q VEGETATION SAMPLING
. STREAM
» STREAM SAMPLING POINT
• HOUSING
Figure 15-20. Facility Plan View
15-69
-------
of the swamp, a distance of 2,500 feet from the landfill. Ground water is
approximately 20 feet below the surface at the crest of the hill, while on the slope it
is approximately 6 feet below the surface. The swamp at the foot of the hill is the
surface expression of the ground water (Figure 15-21).
Aerial Photograhv and Sampling Program
Figure 15-22 shows the infrared aerial image of the site. The landfill
corresponds to the light area in the northwest portion of the photograph (Figure
15-21). The dark area to the south of the site is stressed vegetation, and the-light
area within it is contaminated swamp water. The 33-acre area of tree kill and stress
is clearly visible in the original photograph. Plants, under stress may be detected by
infrared photography because of changes in infrared reflectance.
Ground-water monitoring wells and vegetation sampling points are shown in
Figure 15-20. Data collected from the wells indicated elevated levels of chromium,
manganese, iron, and total organic carbon (TOC). Table 15-9 lists the average
concentrations of the parameters tested. The vegetation study indicated an
accumulation of heavy metals.
Case Discussion
The vegetative stress apparent in the infrared photography was confirmed by
the data from the ground water and vegetation sampling. However, the site
requires further characterization to determine the vertical extent of contamination
and to assess the potential for impact beyond the present area of stressed
vegetation.
It should be emphasized that infrared photography is not a substitute for
hydrogeologic characterization. However, it is a useful tool for identifying areas of
stressed vegetation that may be associated with releases from waste disposal sites.
15-70
-------
Ol
SHAPE OF LANDFILL AFTER
MODIFICATION MID 1978
SCALE:
HORIZONTAL
VERTICAL
GROUND WATER MOUND
LEACHATE BREAKOUT
so1 100!
10 20
GROUND WATER FLOW
Figure 15-21. Generalized Geologic Cross-Section
-------
Area of
Tree kill
and Stress
Ul
Figure 15-22. Infrared Aerial Photograph of the Site
-------
TABLE 15-9
AVERAGE VALUES OF PARAMETERS INGROUND WATER AND STREAM SAMPLES1
Background1
Well #2
Well #3
Well #4
Well #5
Well #6
Stream
BOD5
<10
<20
<20
/
20
19
<20
TOC
/•'-
119
56
300
45.5
45.5
72.5
TKN
' V
1.7
5.5
2.5
0.6
3.4
43.7
NH3
/: .
0.54
3.9
<0.01
1.22
2.47
49.7
N03 =
/ .;
0.48
0.10
<6.01
0.15
<0.1
0.05
Tot.R
1.73
0,11
0.15
0.13
0.18
1.5
0.10
Fe
2.7
108=8
39.6
27
71.6
177
18.4
Mn
0.4
4,2
1,6.3
9.4
3.8
7.65
1.73
Hardness
18.2
5t25.6
414.2
1
353
659.9
230
Cl
5.2
67.3
103.5
980
74.7
120
76.7
CIBO
102
10QO
450
240
1300
1300
Cr
k. ;.
0.03 :
0.02
0.62 ;
0.03
0.71
0.017
IDS
98
828.5
230
/
863.0
780
817
Cd
0.05
1
—
/
1
/
/
pH
/' '
2.9
5.5
5:4
5.8
5.7
68
vl
1 All values as mg/l, except conductivity (umhos) and pH (standard units)
2 Average uf background weiis A and B
/ = N6t Analyzed
-------
CASE STUDY 14: USE OF HISTORICAL AERIAL PHOTOGRAPHS AND FACILITY
MAPS TO IDENTIFY OLD WASTE DISPOSAL AREAS AND
GROUND-WATER FLOW PATHS
Points Illustrated
• Aerial photographs taken over many years in the life of a facility can be
used to locate old solid waste management units (SWMUs).
• Historical aerial photographs can be used to identify
geologic/topographic features that may affect ground-water flow paths.
Introduction
In gathering information pertaining to investigation of a release, historical
aerial photographs and facility maps can be examined and compared to current
aerial photographs and facility maps. Aerial photographs can be viewed as stereo
pairs or individually. Stereo viewing, however, enhances the interpretation because
vertical as well as horizontal spatial relationships can be observed. The vertical
perspective aids in distinguishing various shapes, tones, textures, and colors within
the study area.
Aerial photographs and facility maps can be used for the following:
• Providing evidence of possible buried drums. Historical photographs can
show drums disposed of in certain areas where later photographs show
no indications of such drums, but may show that the ground has been
covered with fill material.
• Showing previous areal extent of landfill or waste management area.
Earlier photographs might show a much larger waste management area
than later photographs.
• Showing areas that were dry but now are wet, or vice versa, indicating a
possible release from an old waste management area.
15-74
-------
• Showing changes in land use patterns (e.g., a landfill in an early
photograph could now be a park or be covered by buildings).
• Soil discoloration, vegetation damage, or enhanced vegetative growth
can sometimes be detected, indicating possible contaminant migration.
• Geologic/hydrologic information, such as faults, fracture or joint systems,
old stream courses (channels), and the contact between moraines and
outwash plains.
Facility description
This facility is the same as previously described in Case Studies 7 and 8.
Data collection and analysis
Over the past 50 years aerial photographs were taken of the facility area.
Interpretation of the photographs produced important information that is shown
diagramatically in Figure 15-23. Solid Waste Disposal Area 2 (SWDA-2)was lower in
elevation in 1940 than it is now. In fact, the area appears to have been leveled and
is now covered by vegetation, making it difficult to identify as a SWMU at ground
level. Another area was identified as a possible waste disposal area from a historical
review of photos. Further study of photographs, facility maps and facility files
revealed this to be a former Liquid Waste Disposal Area (LWDA), designated as
LWDA-2 on Figure 15-23.
The use of these historical photographs also revealed geologic features that
could affect the ground-water flow system under the facility. In this case,
monitoring well data indicated a general northwesterly ground-water flow
direction, in addition to a complex flow pattern near LWDA-1 and SWDA-1 (Figure
15-23). Recent photographs were analyzed, but because of construction and other
nearby activities (e.g., cut and fill, sand and gravel mining) conclusions could not be
drawn. A review and analysis of old photographs revealed the existence of a buried
stream channel of the river (Figure 15-23). This buried stream channel was
identified as a preferential path for ground water and consequently contaminant
15-75
-------
BURNING PIT C,
en
BURNING PIT D
WASTE DISPOSAL AREA
BURNING PIT B
BURIED STREAM
CHANNEL
i SOLID WASTE
.>. DISPOSAL
AREA
> ^BURNING PITA
N
BURNING PIT
& WASTE DISPOSAL
LANDFILL
GROUND-WATER FLOW
Scale
0 200 400
Feel
SOLID WASTE DISPOSAL AREA 1
Figure 15-23. Site Layout. LWDA-2, SWDA-2 and Stream Channel Identified
Through Use of Aerial Photograph Analysis
-------
migration. Additional monitoring data and further analysis of subsurface geologic
data is needed to determine the full impact of the buried stream channel on the
ground-water flow regime.
Case Discussion
Analysis and interpretation of a series of historical aerial photographs and
facility maps spanning a period of over 50 years enabled facility investigators to
identify the following:
(1) Location of waste disposal areas (e.g., old SWMUs);
(2) Changes in topography (related to earlier disposal activities); and
(3) Possible preferential pathways (e.g., old stream channel) for migration of
ground water and contaminants.
This information was used to identify areas for more detailed sampling and
analysis.
Analysis of historical facility maps and historical aerial photographic
interpretation can be a very powerfull tool in a RCRA Facility Investigation, but
should be used in combination with other investigative techniques to result in a
thorough characerization of the nature, extent, and rate of contaminant
migration.
15-77
-------
CASE STUDY 15: USING SOIL CHARACTERISTICS TO ESTIMATE MOBILITY OF
CONTAMINANTS
Point Illustrated
• Information on soil characteristics can be used to estimate the relative
mobility of contaminants in the surface environment.
Introduction
The relative mobility of contaminants can be estimated using soil
characteristics and aquifer hydraulic characteristics. Although metals do precipitate
at higher concentration, at the levels encountered in most subsurface
environments, sorption is the dominant attenuation process. The degree to which a
metal sorbs onto soil particles depends on the soil pH, the percent clay, the percent
soil organic matter, the presence of particular coatings (e.g., iron, manganese, and
aluminum oxide/hydroxides) and, to a lesser extent, the type of clay present. For
organic contaminants, there are several processes which may be important in
predicting their fate in soils. These include sorption, biodegradation, hydrolysis
and, to a lesser extent, volatilization. The sorption of a given organic compound
can be predicted based on its octanol-water partition coefficient, the percent
organic carbon in the soil, and the grain-size distribution of the soil.
Determining the relative mobility of contaminants can be helpful in selecting
appropriate sampling locations. For example, if wastes containing metals were
present in an impoundment, samples to determine the extent of any downgradient
metal contamination would normally be collected within a short distance of the
impoundment. On the other hand, for fairly mobile waste constituents such as
trichloroethylene (TCE), samples could be taken over a much larger downgradient
distance. The case study presented below illustrates how contaminant mobility can
be estimated.
Facility Description
A 17-acre toxic waste dump was operated in a mountain canyon for 16 years.
The facility received over 32 million gallons of spent acids and caustics in liquid
15-78
-------
form. These wastes were placed in evaporation ponds. Other wastes sent to the
facility included solvents and wastes from electroplating operations containing
chromium, lead, mercury and zinc. Pesticides including DDT had been disposed of in
one corner of the site.
Site Description
The site was underlain by alluvium and granitic bedrock (Figure 15-24). The
bedrock, as it was later discovered, was fractured to depths of between 50 and 100
feet. Ground water occurred in the alluvial deposits at depths of 10 to 30 feet.
Several springs existed in the upgradient portion of the site. A barrier dam was
built across part of the canyon at the downgradient edge of the site in an effort to
control leakage. Because of the extensive fracture system, this barrier was not
effective. Instead, it appears to have brought the ground-water table up into the
wastes and, at the same time, pressurized the underlying fracture system, thereby
creating seepage of contaminated water under the dam.
Estimation of Contaminant Mobility
Because of the variety of constituents accepted at this site, an estimate of their
relative mobility was needed prior to designing the remedial investigation. The first
step was to estimate the average linear velocity using the following equation:
«. 5
He
where
v = horizontal seepage velocity, ft/day
K = hydraulic conductivity, ft/day
i = ground-water gradient
Tie = effective porosity, decimal fraction.
The hydrogeologic data needed were obtained from existing site assessment
reports. The alluvium underlying the site had an average hydraulic conductivity of
0.8 ft./day and an estimated effective porosity of 11 percent. The average ground-
15-79
-------
Barrier Dam
00
o
GHOUNO_ WATtH TABtE
FRACTURED BEDROCK
Figure 15-24. Schematic Cross-Section of Waste Disposal Site
-------
water gradient below the site was 0.06. Using the above equation, the average
linear velocity was estimated to be 160 ft/yr. This represents the average velocity at
which a conservative constituent would migrate downgradient along the centerline
of the plume. Examples of such constituents include chloride and bromide. As
shown in Table 15-10, nitrate and sulfate also behave conservatively in many cases.
Due to the absence of highly weathered, sesquioxide soils, sulfate behaved
conservatively at this site. Using the above average linear velocity, an estimate was
made of the distance a conservative solute would travel in a given time (T) using
d =v'T. Limited water quality data were available for 1980. Wastes were first
disposed at this site in 1956. The average extent of plume migration along the
centerline was thus estimated to be 3800 feet.
With respect to metals, additional data were needed to estimate their fate
including soil pH, presence of carbonates, organic ligands, and percent soil organic
matter and clay. At this site, the soil PH varied from less, than 3.0 within 400 feet of
the acid ponds to 7.2 at a distance 2000 feet downgradient. As shown in Figure
15-25, the partition coefficients for metals are dependent on pH and organic matter
content. For example below a pH of 5.6, for the types of soil encountered at the
site, the partition coefficient (Kp) for cadmium is about 10 ml/g. At a pH of 7.2, Kp is
about 6500 ml/g, (Rai and Zachara, 1985). The relative mobility of attentuated
constituents can be estimated as follows (Mills et al.. 1985):
= v/Rd
where
VA = average velocity of attenuated consitutent along centerline
of plume, ft/day
v * average linear velocity as defined above, ft/day
Rd = retardation factor (unitless)
and
Rd , 1 +
15-81
-------
TABLE 15-10
RELATIVE MOBILITY OF SOLUTES
Group
Conservative
Slightly Attenuated
Moderately Attenuated
MoreStrongly
Attenuated
Examples
TDS
;BR'
NO,'
S042
B
TCE
Se
As
Benzene
Pb
Hg.
Penta-
chlorophenol
Exceptions
- •• ; -. '-•
_;' ' •
Reducing conditions
Reducing conditions
and in highly
weathered soils coated
withsesquioxides
Strongly acidic systems
Anaerobic conditions
- - .- I- - i .
Master Variables*
V
v, pH, organic matter
v , organic matter
v , pH, Fe hydroxides
v , pH, Fe hydroxides
v, organic matter
V , pH, SO*2'
V,pH,C1
v , organic matter
Variables which strongly influence the fate of the indicated solute groups.
Based on data from Mills et al., 1985 and Roi and Zachara, 1984.
15-82
-------
100
Percent
Adsorbtion
by Soil
50
Shift due
to presence
of soil organic
matter
Typical
adsorbtion
curve for.
heavy metal
x, on a clean
silica or
aluminum
silicate
surface
Typical adsorbtion
curve for heavy
metal x, on silica
or aluminum silicate
surface coated with
soil organic matter
pH of the Soil Solution
a) Generalized Heavy Metal Adsorbtion Curve for Cationic Species (e.g., CuOH+)
100
Percent
Adsorbtion
by Soil
50
»
Typical adsorbtion \
curve for heavy \
metal species, x,
on iron hydroxide
-
\
\
•
\
1\
\
A
Shift \
due to \
presence t
of soil \
organic \
matter
pH of the Soil Solution
b) Generalized Heavy Metal Adsorbtion Curve for Aniotic Species (e.g., CrO|')
SOUFM: (MilUrt»!.. 1985).
Figure 15-25. Hypothetical Adsorption Curves for a) Cations and b) Anions
Showing Effect of pH and Organic Matter
15-83
-------
where
Kp = soil-water partition coefficient for solute of concern, ml/g
PB = soil-bulk density, g/ml_
Tie = effective soil porosity (decimal fraction).
For example, the relative mobility of cadmium at a pH of 7.2 was estimated for this
site as shown below:
Rd = 1 + 6500(1.7) = 100,456
0.11
VA= 160/100,000 = 0.002 ft/yr
This estimate was consistent with the field data which indicated that the metals
migrated only until the pH of the contaminant plume was neutralized, a distance of
about 2000 feet. Cadmium concentrations decreased from 1.3 mg/l at a distance of
1400 feet from the ponds to below detection (<0.1 ug/l) at a distance of 2000 feet.
Estimates of mobility for organic contaminants which sorb onto soil particles
can be made in an analogous manner. The partition coefficient for organic
constituents can be calculated using the following equation (Mills et al.. 1985):
Kp = Koc[0.2(1-f)Xsoc + fXfocl
where
Kp = soil-water partition coefficient, ml/g
KOC - organic carbon partition coefficient, ml/g
and
K0<; = 0.63 Kow
KQW - octanol-water partition coefficient
f - mass of silt and clay (0
-------
For example, the solvent trichloroethylene (TCE) has a Kowvalueof200. Using the
above equation and site data (f = 0.1, Xsoc= 0.001, Xf0c= 0.01), the partition
coefficient Kpwas estimated to be 0.2 ml/g. The relative mobility of TCE at the site
was then estimated to be approximately 40 ft/yr (Rd = 4 and VA = 40 ft/yr).
Methods for considering-additional processes influencing-the fate of organics (e.g.,
hydrolysis and biodegradation) are presented in the manual entitled Water Quality
Assessment: A Screening Procedure for Toxic and Conventional Pollutants in
Surface and Ground Water (Mills et al.. 1985).
Case Discussion
As shown in Figure 15-26, contaminants downgradient of a waste disposal site
may migrate at different speeds. Using the methods illustrated above, estimates of
the relative mobility of constituents can be made. Such estimates can then be used
to locate downgradient monitoring wells and to assist in the interpretation of field
data.
References
Mills, W.B., D.B. Porcella, M.J. Ungs, S.A. Gherini, K.V. Summers, L. Mok, G.L. Rupp,
and G.L. Bowie, 1985 Water Quality Assessment: A Screening Procedure' for Toxic
and Conventional Pollutants in Surface and Ground Water. EPA/6QO/6-85/Q02a. Vol.
I, II and III.
Rai, D. and J.M. Zachara. 1984. Chemical Attenuation Studies: Data Development
and Use. Presented at Second Technology Transfer Seminar: Solute Migration in
Ground Water at Utility Waste disposal Sites. Held in Denver, Colorado, October 24-
25,1985.63 pp.
15-85
-------
FEET
I
800
Figure 15-26.
Schematic Diagram Showing Plumes of Total Dissolved Solids (TDS),
Total Organic Halogens (TOX) and Heavy Metals Downgradient of
Waste Disposal Site
15-86
-------
CASE STUDY 16: USE OF LEACHING TESTS TO PREDICT POTENTIAL IMPACTS OF
CONTAMINATED SOIL ON GROUND WATER
Point illustrated
o Soil leaching tests can be used in conjunction with waste and site-specific
factors to predict potential impacts on ground water.
Introduction
Contaminated soil, whether deep, or surficial in nature, has the potential to
impact ground water, primarily through leaching. In many cases, soil
contamination has already lead to contamination of the ground water and
decisions can be made regarding clean-up of the contaminated soil and ground
water based on the constituent concentrations observed in these media. However,
in cases where contaminated soil has not yet resulted in contaminated ground
water, but has some potential to do so, decisions need to be made regarding the
contaminated soil and whether it should be removed or some other action-should
be taken because of the soil's potential to contaminate ground water above levels
of concern. This evaluation may be especially critical in those cases where only deep
soils are contaminated, or where constituent concentrations within surficial soils do
not exceed soil ingestion criteria. Both theoretical (mathematical) and physical
(leaching test) models can be used in this evaluation, as well as or in conjuration
with a qualitative evaluation of release and site-specific factors. This case illustrates
the, use of leaching tests and consideration of release and site-specific factors to
determine whether contaminated soil has the potential to contaminate ground
water above levels of concern.
Facility Description
The facility is an industrial chemical and solvent facility located on a leased 2.5
acre site within the corporate limits of a major city in the north-central United
States (see Figure 15-27). Periodic overtopping of the surface impoundment, which
is now empty, and suspected contamination of the soil with organic solvents
from the surface impoundment, resulted in an RFI in which the facility was directed
to characterize the nature, extent and rate of release migration.
15-87
-------
RIVER
LEGEND
• MONITORING WELL
O SOIL BORING
SCALE. FEET
Figure 15-127. Facility Map Showing Soil Boring And Well Installation
-------
Release characterization revealed that the soil surrounding the surface
impoundment, which was mostly fine sand and silt with some clay, was
contaminated with tetrachloroethylene and 1,1,1 -trichloroethane at concentrations
ranging from 0.05 to 0.10 and 2 to 20 mg/kg, respectively. This contamination was
observed at depths of up to 5 feet, which was approximately 20 feet above the
water table (i.e., the water table was approximately 25 feet, below the land surface).
The soil beneath the site was relatively permeable, with a hydraulic conductivity of
approximately 9x10"4cm/sec.
Ground-water monitoring conducted during the RFI showed no current
contamination of the ground water, which flows in a northerly direction and
eventually intersects the river (Figure 15-28). The river is used for irrigation and
drinking at downstream locations. Grab, samples taken from the river and river
sediments showed no contamination.
The soil in the immediate vicinity of the railroad spur also showed isolated
pockets of mercury contamination ranging in-concentration from 1 to 2 mg/kg, and
to a depth of 1 foot below the land surface. The source of the mercury
contamination could not be determined.
Contamination Evaluation
The relevant health and environmental (HEA) criteria, the constituent
concentrations observed at the site, and selected physical/chemical properties for
the three constituents are shown in Table 15-11. Although comparison of the HEA
criteria for ingestion with the consituent concentrations observed at the site
showed no exceedances, the regulatory agency overseeing the RFI was concerned
that leaching of the contaminated soil could lead to eventual contamination of the
underlying ground water. This concern was based on the relatively high
permeability of the soils beneath the site and the relatively high mobility of the two
organic constituents detected. The facility obtained the regulatory agency's
approval to conduct a leaching evaluation using EPA's Method 1312 (Synthetic Acid
Precipitation Leach Test for Soils).
15-89
-------
807.2
ui
UD
O
807.4
807.8
Figure 15-28. Facility Map Showing Ground-Water Contours
-------
TABLE 15-11
HEA CRITERIA, CONSTITUENT CONCENTRATIONS AND RELEVANT
PHYSICAL/CHEMICAL PROPERTY DATA FOR CONSTITUENTS, OBSERVED AT SITE
Chemical
Tetrachloroethylene
1,1,1-Trichloroethane
Mercury
CAS No.
127 -18-4
71-55-6
7439-97-6
HEA
Criteria
(Ingestion)
(mg/kg)
69
2,000
--
H20
S o I
(mg/l)
150
1500
HEA
Criteria
(Water)
(mg/l)
0.0069
0.2
0.002
Constit.
Cone.
(mg/kg)
0.10
20
2
Koc
(mg/l)
364
152
Low
Log
Kow
2.6
25
Det.
Limit*
(mg/l)
0.01
0.01
0.0004
Detection limits presented are those for water. Detection limits for soil vary greatly, but may
be assumed to be approximately 1 mg/kg.
15-91
-------
Leaching Evaluation
Prior to collecting samples and applying the leaching test, the facility first
decided to determine if the contaminated soils could possibly result in leaching test
(extract) concentrations that exceed the relevant HEA criteria (See Table 15-11 ). To
do this, the facility calculated the maximum theoretical extract concentration by
assuming that 100 percent of the constituents would leach from the soil. The
Maximum Theoretical Concentration of Toxicant
Extract Concentration(mg/l) in Soil (ma/ka)
20
where 20 refers to the liquid to solid ratio applied in EPA Method 1312.
Using this simple equation, the facility determined that the maximum leachate
concentration for tetrachloroethylene was, in fact, below the HEA criteria for water
(see Table 15-11), and that the level could not possibly be exceeded even if 100
percent of the contaminant leached from the waste. For 1,1 ,1-trichloroethane and
mercury, however, it was determined that the HEA criteria level could be reached if
only a portion of the contaminant present leached from the soil, and that
application of the leaching test would be necessary. Using this screening-type
evaluation, the facility was able to reduce the number of constituents that would
need to be analyzed when applying the leaching test, from three to two.
Samples of the contaminated soil were then collected at selected locations
(i.e., those expected to produce the more heavily contaminated samples) and
Method 1312 applied. Total constituent analyses were also conducted in order to
ensure that the samples represented the more heavily contaminated areas of the
site. Analyses of the soils and leaching test extract were conducted for 1,1,1-
trichloroethane and mercury. The results are shown in Table 15-12.
The leaching test results for 1,1,1 -trichloroethane and mercury showed extract
concentrations above the respective HEA criteria (action levels) for these
15-92
-------
TABLE 15-12
LEACHING TEST RESULTS
Constituent
1,1,1-Trichloroethane
Mercury
D-C
0.3
0.002
C-C
0-2
0.002
C-B'
0,5
0.003
* Resampled at locations close to original sampling point. Samples analyzed
are result of composite of three grab-samples. All samples were taken from
the top 0-1 ft of the soil surface.
15-93
-------
constituents, indicating that there might be a basis to require some sort of
corrective action. The facility, however, presented arguments to show that mercury
would be attenuated in the soil column as the leachate percolates towards the
water table, and that 1,1,1-trichloroethane would be degraded to below the level
of concern in the ground water. Below is a synopsis of the two arguments.
Mercury: The facility first examined theoretical Eh-pH fields of stability for the
various aqueous mercury species, determined that the predominant mercury species
would be elemental mercury, and further predicted (using Eh-pH diagrams) that the
maximum equilibrium concentration of elemental mercury in water would be 0.025
mg/l. The facility interpreted the substantially lower leaching test concentration to
indicate that attenuation processes such as sorption play a major role in restricting
the mobility of elemental mercury. The facility cited high soil/water partition
coefficients (i.e., Kd values), and several scientific studies to further support their
contention that mercury would strongly sorb to both organic and inorganic
components of the soil before any leachate reached the ground water.
1.1.1-Trichloroethane: The facility recognized that due to its high solubility
(1500 mg/l) and low Kd (0.011 ml/g), 1,1,1-trichloroethane would not be attenuated
appreciably as the leachate percolates towards the water table. The facility argued,
however, that abiotic hydrolysis would significantly degrade 1,1,1-trichloroethane
during leaching. Several environmental half-life studies were cited which indicated
that the half life for 1,1,1-trichloroethane ranged between 0.5 and 2.5 years. Based
on these studies, the facility predicted that 1,1,1-trichloroethane would be
degraded to below levels of concern within one to three years. Using additional site
information and simple time of travel calculations, the facility predicted that
concentration levels for 1,1,1-trichloroethane would be decreased to below the
level of concern well before reaching any potential receptors.
The regulatory agency's evaluation of the facility's arguments is presented
below:
Mercury: The facility's argument with respect to mercury is essentially correct
if it can be assumed or proven that the mercury originally present at the site is
inorganic in nature. If, however, the mercury present is organic in nature (e.g.,
15-94
-------
methyl mercury), the potential for migration of the mercury is, substantially
increased. Organic mercury compounds generally have higher volatility, higher
solubility, and much lower Kd values, than inorganic mercury compound. It should
also be noted that even if the original release was of inorganic mercury,
biotransformation (i.e., biomethylation) of elemental mercury may occur. The
facility should be required to determine the actual form(s) of mercury present at the
site.
1.1.1-Trichloroethane: The facility's argument with respect to 1,1,1-
trichloromethane raises many technical questions. For example, the facility uses
published data, on the half life of 1,1,1-trichloromethane, which may not be
applicable to the facility's soil and ground-water environment. As another example,
the half-life degradation, rate argument may only be applicable for ground-water
transport. The facility does not address degradation in soil or effects on surface
water (assuming that contaminated ground-water will eventually migrate to the
river). Most important, however, is the fact that the facility did not address the
degradation products of 1,1,1-trichloroethane, one of which is 1,1-
dichloroethylene, which is also a hazardous constituent. 1,1,1-trichloroethane
should be assumed to pose a threat to ground water.
Conclusions
The next step in the RFI process would be to determine if interim corrective
measures or a Corrective Measure Study was warranted for the release. Although
none of the soil ingestion HEA criteria were exceeded at the site, application of the
leaching evaluation indicated that 1,1,1-trichloroethane could leach to ground
water and result in exceedance of the HEA criterion for water. On, this basis, the
facility should be directed to perform a Corrective Measures Study.
To prevent further contaminant migration, the application of interim
corrective measures may also be considered. Construction of a temporary cap over
the contaminated area is one option Perhaps a more appropriate measure would
be to remove the contaminated soil. Such an action, taken as an interim corrective
measure, may negate the need for a formal Corrective Measures Study.
15-95
-------
Case Discussion
Leaching tests and similar evaluations (e.g., application of validated
mathematical leaching models) can be used to identify potential problems due to
leaching of contaminated soils. In this case; application of a leaching evaluation
was instrumental in identifying a potential threat to ground water as a result of
leaching of contaminated soil. This finding was particularly significant as HEA
ingestion criteria were not exceeded.
It should be noted, however, that in some cases leaching tests may provide
result that are difficult to interpret. For example, consider what would have
happened if the soil underlying the facility was predominantly clay with a
permeability on the order of IO"8 cm/sec. In this case, demonstrating" that leaching
will most likely occur within the foreseeable future may be difficult. As another
example, if the soil underlying the facility were predominantly sand, leaching would
be probable. In both these cases, application a leaching test may not provide any
more useful information than is already available. Careful consideration of release
and site-specific information is always warranted prior to application of leaching
tests.
15-96
-------
CASE STUDY 17: USE OF SPLIT-SPOON SAMPLING AND ON-SITE VAPOR
ANALYSIS TO SELECT SOIL SAMPLES AND SCREENED
INTERVALS FOR MONITORING WELLS
Point Illustrated
HNU and OVA/GC screening can provide a relative measure of
contamination by volatile organics. It can be used to select soil sample
locations and can assist in the selection of screened intervals for
monitoring wells.
Introduction
On-site vapor screening of soil samples during drilling can provide indications
of organic contamination. This information can then-be used to identify apparent
hot spots and to select soil samples for detailed chemical analyses. In this manner,
the use of higher powered laboratory methods can be focused in an effective way
on the analysis of samples from critical locations and depths. The vapor analyses on
site can also be helpful in selecting screened intervals for monitoring wells.
Facility Description and History
Manufacturing of plastics and numerous other chemicals has occurred at the
site over the past 30 years. Some of the major products included cellulose nitrate,
polyvinyl acetate, phenol, formaldehyde, and polyvinyl chloride. The entire site
covers 1,000 acres. The location of the buildings and waste disposal areas are shown
in Figure 15-29. This is the same facility as used in Case Studies 7,8 and 14.
Three disposal methods are known to have been employed at the site. Readily
combustible materials were incinerated in four burning pits, while non-
combustibles were either disposed of in landfills or in a liquid disposal area. All on-
site disposal operations were terminated in 1970, and monitoring programs have
been implemented to identify contaminants, define and monitor ground-water
contaminant plumes, and assess the resulting environmental impacts.
15-97
-------
BURNING PIT
ui
i
l£>
00
LIQUID WASTE DISPOSAL AREA
N
Scale
0 200 400
Feet
SOLID WASTE DISPOSAL AREA 1
Figure 15-29.
LEGEND
D PHASE | WELLS
|p BURNING PIT
€SB> WASTE DISPOSAL AREA
'-- ^SS> LANDFILL
| 1 SEISMIC REFRACTION LINE
Site Plan Showing Disposal Areas and Phase I Well Locations
-------
Geologic and Hvdrologic Setting
The site is located in a well-defined glacial valley, adjacent to a river. Three
major units underlie the site, consisting primarily of sand and gravel outwash
deposits; fine-grained lacustrine sands; and till. The middle sand unit contains
lenses of silt, clay and till. Only the deep till formation appears to be continuous
across the site. A geologic cross-section beneath two of the disposal areas is shown
in Figure 15-30.
The ground-water flow direction at the site is to the northwest. However,
there appears to be a buried stream channel running across the site which strongly
influences the local ground-water flow regime (see Figure 15-31). Ground water
from the site is thought to discharge to the river. The depth to ground water varies
from 10 to 40 feet.
Sampling Program
As part of the remedial investigation at this site, 33 borings were drilled using
a hollow-stem auger rig. Continuous soil samples were collected using split-spoon
samplers. Samples for laboratory chemical analysis were selected based on the
volatile organic concentrations detected by initial vapor screening of the soil
samples in the field.
This field screening was achieved by placing a portion of each sample core in a
40 ml glass headspace vial. An aliquot of gas was extracted from the vial and
injected directly into a portable OVA gas chromatography (OVA/GC). The
chromatography was equipped with a flame, ionization detector to identify
hydrocarbons. Each sample was also screened using an HNU photoionization
detector because of its sensitivity to aromatic hydrocarbons, particularly benzene,
toluene and the xylenes. Following completion of drilling, gamma logs were run on
ail boreholes!
An example of the vapor screening results (HNU and OVA/GC) and geological
and gamma logs for one of the boreholes are shown in Figure 15-32. The data
shown demonstrate the differential sensitivity of the HNU and OVA/GC detectors.
Because the OVA/GC is more sensitive to the organics of interest (aliphatics),
15-99
-------
in
_i
in
01
Ul
UJ
u.
O
brown fine SAND
brown medium
SAND
brown fine-in HI um
SAND
gray fin* SAND
brown medium-coarse SAND
brown fine-medium
gray-fin* SAND with
partings SILT and CLAY
gray varved SILT and CLAY
gray fine SAND with partings
SILT and CLAY
LEGEND:
*:•:•: WASTE DISPOSAL
:$:$ AREAS
*v.V- TILL
VERTICAL SCALE
SAND
gray-brown fine SAND FEET
jwith SILT partings
20
HORIZONTAL SCALE
0 80
FEET
Figure 15-30. Geologic Cross-Section Beneath Portion of Site
-------
BURNING PIT
ui
o
•~~- P~A>BURNING\PITA
N
Scale
0 200 400
Feet
CONTOUR INTERVAL, 5 Feel
LEGEND
BURNING PIT
WASTE DISPOSAL AREA
LANDFILL
" !• .; :
:i5d GROUND WATER
ELEVATION. Feet
Figure 15-31: Ground-Water Elevations in November 1984
-------
en
X
S
s
10
' IS
10
.«
JO
JS
10
.*s
JUU
B
i
i
5
C
>
1
_9_
10
II
1
1
11
£
I
IT
li
19
Id
1
11
*
*
Lil
0M H l"l 1- 1 H H H M PKOVEHYI
W lol l-l IJ LI LI LI LI 1 i»t»Ti 1
t.»
1.)
,|
,n
.9
-3.
.i
,4
.&
.1
.*
,}
.1
•1
.S
Lil
ttt
ii
1
H
1
11
i
10
ii
lo
13
ii
ii
1C
i*
ii
_U_
11
Ml
])
ff
11
19
1)
It
14
Jli
fOlL CLAfllFICMIOM
Light brown coarse land with some
(in* - medium q ravel, (try. loose -
firm.
1
Light brown line - medium sand, ory.
lirn.
Grades lo light brown fine sand, with
some silt, molsl, firm.
19
Red fine - coarse sand and silt with
some gravel, Moisl, compact.
ICiacial tllil ,
Layer grey silt. ^ ^
Brown silt with partings red sill,
1 rading to clay.
1>«
lot lied brown line - coarse sand mna
ine - medium gravel, wet. firm.
1 GEOLOGIC I
I COLUMN I
^
O^Q.
>^«
i]5Sft'
pi
.^«° •
'^;?
5??°.9,
^•^n
• .
#
#
*
•
to
-<;
-
C*H
C
S
\\
~"
4
u*
OUHI
IOO
(
- "7
. 4
1
<>
LOG
S
~\
•*.
~t
-<
v
I
S
_J
^
>
H
\
HHU
SCMCNING
ffM
o
«
2
1
0
— s —
0
~9
D
— 0 — r
o
~ii
0
0
0
— o
— Ti
-?
0
0
1
A
ffH
9
to
9
9
9
9
— i —
9
i
•
— 1 —
— 1 —
••
i
7
1
7
—»
— n —
~7I
i>
»
MS
190
S
o
V* !
-------
L/l
I
o
UJ
X
i
•" ss
iS
•
ts
•
70
-»
*IO
•S
90
m
,'^
.
loo
K
27
!•
19
~
Jl
31
iT
5
(
37
1
1
lit
i
<1
II
..
4
i
It
7
1
—
9
So
§:
hi —
S
1.3
1.7
ill
1.1
1.7
.7
1.3
u
,6
.4
.6
. 1
.3
.7
. 1
.4
)
.*
6
0
3
*
Ii
II
IB
Ii
3D
111
11
V)
ii
11
1*
31
19
20
11
NO
ii
3D
61
7S
6l
•6
MI
MIL CLMSIFICMID*
net. firm.
Grades to brown fine - medium sand
trace coarse sand, set, firm.
grades grey - brown. to
firm.
Grades In grey-green fine - medium
sand, little coarse sand. wet. firm.
Grading to trace coarse sand.
86
Red silt with some clay, some sand.
some gravel, wet. firm.
(Glacial Till) 91
Dark brown fine - medium sand with
varves sill.
Dark brown varved silt and (me sand.
wel, hard.
99
NBP-103 1
31001030 I
*
V* f c
i ;o*4^t
SS.SK
S^A ^
s
-
o
[J
^
f
n
(
/
\
i
i
y
C4
s
Lk
)-
V
f
^
CM
1C
r
^
*
J
1 I OS
a/rs
6
-
~
»^_
1
•^
-=•
V
s
^
V,
^
U H
*v
^
V
J
L
HNU
SCIKEHUM
n
- IL
0
J
_>_lJ
> -1
0
0
Tr
Tr
.S
Tr
1
a
8
0
a
B
0
0
17t
— Ml .
1*
lid 7
381!
NO
31
13
IN
S6
56
93
96
0
j
2 :
a
OV» -SCHEMING
IO 10O I.OOO K>,<
jj ml i nit i nil i ii
r--
S
I
j.
/
.>
i.
-
V,
^
\
**
^ •
s
^ '
fl
L
U
F
n
-
-
-•
I4JJ
-
a
MM M
1
roc
ttiffi
B ~
tfH
•
u
!U
•
=
' .
irfClflt
CCHDUC1
SJ
ft?
NOTES
Figure 15-32. Example of Borehole Data Including HNU and OVA/GC Screening
(continued)
-------
these results were used to select samples for detailed chemical analysis in the
laboratory. As shown in Figure 15-32, samples in zones with OVA/GC readings of
365 ppm (45 feet deep), 407 ppm (65 feet deep), and 96 ppm (85 feet deep) were
selected. In the laboratory, samples were first analyzed for total organic carbon
(TOC). The ten samples with the highest TOC levels were then analyzed for
purgeable organics using EPA Method 50-30 and extractable organics using EPA
Method 82-,50 (U.S. EPA, 1982 - Test Methods for Evaluating Solid Waste, SW 846).
The OVA/GC results were also used to select well screen intervals. Examination
of the data in Figure 15-32 show that the highest levels of volatile organics (by
OVA/GC) were found a depth of 65 feet. In addition, the gamma and geologic
logs indicated that the permeable medium at that depth was coarse sand which
would be a suitable location for the placement of a well screen. Thus, a 5-foot
stainless steel screen was set over the depth interval of 62 to 67 feet. -
Case Discussion
This sampling program incorporated field techniques that detect the presence
of volatile organics and allow on-site, rapid identification of likely contaminant
"hot spots" for detailed laboratory analysis and to select depths for monitoring well
screens.
15-104.
-------
CASESTUDY 18: CONDUCTING A PHASED SITE INVESTIGATION
Points Illustrated
When ground-water contamination is known or suspected at a site, a set
of initial borings is typically made to determine site hydrogeologic
characteristics and to identify areas of soil, and ground-water
contamination (Phase I).
• These findings are then used to select well locations to fully delineate the
extent of contamination (Phase II.
I n t r o d u c t i o n
To identify the extent ground-water contamination in an efficient manner,
information is needed on the ground-water flow regime. Phase I investigations
typically focus on determining site geologic characteristics and ground-water flow
directions and velocities: Waste sources are also identified. The Phase I results are
then used in planning the Phase II investigation to determine the extent of
contamination and to refine estimated rates of contaminant migration.
Facility and Site Description
Descriptions of the facility and site geologic characteristics were included in
Case Studies 7, 8, 14 and 17.
Sampling Program
The Phase I sampling program included geophysical surveys, water level
monitoring, soil sampling, and ground-water quality sampling. Three seismic
refraction lines were run to estimate the depth to the top of the deep till. The top
Available historical data indicated that the general ground-water flow
direction was to the northwest across the site. The ground water was thought to
discharge to the river. This information and historical drawings and maps of known
15-105
-------
disposal areas were used to locate the Phase I borings (see Figure 15-29 in Case
Study 17). One well (MW4) was located on the suspected upgradient side of the
site. The other wells were located near waste sources to determine which sources
appeared to be contributing contaminants to the ground water. For example, two
wells (MW6 and 7) were immediately downgradient of solid waste disposal area #2.
To determine the presence, of vertical gradients, three two-well clusters were
drilled-each with one well screened just below the water table and a second well
screened considerably below that at the base of the till.
The results of the Phase I investigation indicated that all the wells contained
solvents. Thus, investigations of the waste sources and contaminant plumes were
continued in Phase II. The highest solvent concentrations were found in wells
located near the liquid waste disposal area where downward vertical gradients
were present. The contaminants had migrated down to depths of 75 feet in this
portion of the site. The Phase I data confirmed the general northwest ground-
water flow direction but showed a complex flow pattern near the buried stream
channel. A second concern was whether observed lenses of fine-grained till under
the site were producing localized saturated zones which could be contaminated.
Based on the Phase I results, a Phase II monitoring program was designed to
determine the extent of contamination around the major disposal sites. Typically,
two soil borings were made - one up- and one downgradient of the waste source.
Because of the high solvent concentrations observed in the wells downgradient of
the liquid disposal area, a more intensive field investigation of this area was
included in Phase II. instead of two borings per waste source at the liquid disposal
area, 11 soil borings and five new monitoring wells were drilled. This represented
one-third of the total effort for the entire 1,000 acre site. The total number of
Phase II soil borings was 33 (Figure 15-33) and the total number of Phase II wells was
15 ,( Figure 15-34). The phase II data indicated, that most of the solvent
contamination originated, from the liquid disposal area and not from solid waste
disposal area #1 which is upgradient of the liquid disposal area. The Phase II data
did identify PCBs from solid waste disposal area # 1 but not from any of the other
sources. This was consistent with site records indicating that transformers had been
disposed at this area.
15-106
-------
LSI
o
vl
BURNING PIT C
MW-2n
MMM2 Q MW-1
MW-3
D •
MW-5
D
LIQUID WASTE: DISPOSAL AREA
*$%&&
• " &r&*)fjfSsSA:*
•„.., .^#^%j
*\ °
BURNING PIT A
SOLID WASTE DISPOSAL AREA 1
N
Scale
i ' I
0 200 400
Feet
LEGEND
BURNING PIT
WASTE DISPOSAL SITE
LANDFILL
PROPOSED SOIL BORINGS
Figure 15-33. Proposed Phase II Soil Borings
-------
BURNING
Y1
o
00
MW-2D
MW.,2 MW-1 ci
D
1OLID WASTE
DISPOSAL
AREA
BURNING PIT A
N
Scale
I 'I
0 200 400
Feet
SOLID WASTE DISPOSAL AREA 1
LEGEND
D TEST BORING W/GROUNDWATER MONITORING WELL
BURNING PIT
WASTE DISPOSAL AREA
LANDFILL
O PROPOSED MONITORING WELL
Figure 15-34. Proposed Phase II Monitoring Wells
-------
Case Discussion
Investigation of a large complex site is commonly conducted sequentially.
Basic information is needed on site geologic characteristics and ground-water
velocities and directions to appropriately locate wells for determining the extent of
contamination. Thus, the initial installation of a limited number of exploratory
borings and wells can provide the data needed to design a complete and effective
investigation. Results from the latter investigation can then be used to determine
the need for remedial action and to evaluate alternative remediation methods.
15-109
-------
CASE STUDY 19: MONITORING BASEMENT SEEPAGE
Point Illustrated
• Basement monitoring can be used to estimate the extent of contaminant
migration.
Introduction
Leachate produced in a landfill can be transported downgradient in ground
water by advection and dispersion. Shallow ground water may surface and seep
into basements.
Site Description
A channel, originally constructed as part of a hydroelectric power generation
system, was used as a disposal site for a variety of chemical wastes from the 1920s
through the 1950s. More than 21,000 tons of waste were dumped in and around
the site before its closure in 1952. After closure, homes and a school were
construed on and around the site. In the 1960s, residents began complaining of
odors and residues. During the 1970s, the local water table rose, and contaminated
ground water seeped into nearby basements.
Geologic and Hydrologic Setting
Figure 15-35 shows a cross-section of the site. The site has both a shallow and
a deep aquifer. The shallow aquifer consists of approximately 5 feet of interbedded
layers of silt and fine sands overlying beds of clay and glacial till. The deeper aquifer
is a fractured dolomite bedrock overlying a relatively impermeable shale. Travel
times from the shallow to the deeper aquifer are relatively long. Contamination has
occurred in the shallow aquifer because of the "bathtub effect". The impermeable
channel filled because of infiltration, and leachate spilled over the channel sides.
The leachate contaminated the shallow ground water and was transported laterally
in this system.
15-110
-------
BASEMENT
'&:l::!1::!!::!i:'::::;WKn::^!!:;jijji.^V:Kftn^
ftiiaat^"'-"iiiiii riiffliiiiiiiiii-iii-ii-iij;!:
LEGEND: BURIED UTILITIES ARE
S -STORM SEWER ' :.
A - SANITARY SEWER .
W- WATER MAIN
*• ACTUAL FLOW PATH FOR LEACHATE
SEASONAL HIGH WATER TABLE
BASEMENT ;
- GROUND
- 1-8-2.5 ft.
- $£6"-
^ 12:0 ft
- 23.0ft.
- 38.0ft.
LEGEND
4#= SILT FILLIK>10-^cm/s
p:- SILTY SAND.K>10-6cm/s
HARD CLAY
TRANSITION CLAY >K=10-7 to10-8cm/s
SOFT CLAY
GLACIAL TILL
DOLOMITE
figure 15-35. Geologic Cross-Section Beneath Site
-------
Sampling Program
The houses surrounding the channel were grouped into three sets
(upgradient, downgradient, and on-site) based on preliminary data on the
underlying strata and ground-water flow directions. Four houses from each group
were selected for sampling for a total of 12 houses. Samples of water and
sediments were collected, from the sump pump wells in each basement. Water
samples were collected when the sump pumps were running and 24 hours after
pumping had ceased. Water and sediment samples were analyzed for purgeable
and extractable organics. Benzene, carbon tetrachloride, chloroform, and
trichloroethylene (TCE) were found in the water samples. Water samples taken
while the sump pumps were running had higher concentrations of volatile organics.
Sediment samples contained PCBs and dioxin, possibly due to cosolvation.
Relatively immobile organics can become dissolved in another more mobile solvent.
The mobile solvent containing traces of other organics can be advected along with
the water. This process (cosolvation) is one facet of enhanced transport which has
recently been proposed as a possible mechanism for the observed mobility of
otherwise immobile organics. Samples of water and sediment from storm drains
were also collected and analyzed to determine if discharge from the sumps to the
storm drains were a significant source of organics in the storm runoff.
In addition to determining water quality, indoor and outdoor air quality was
measured in the basements at each house. Tenax and polyurethane foam tubes
were placed in air monitoring systems in each basement to measure 12-hour
average concentrations of volatile organics (e.g., carbon tetrachloride, benzene,
and TCE) and semi-volatile-organics (e.g., pesticides). Volatile organics were present
in the indoor air samples but semi-volatile organics were not detected. The highest
volatile organic concentrations were observed when the sump, pumps were
operating.
Case Discussion
At sites where hydrogeologic factors favor shallow lateral ground-water flow,
initial site characterization may involve sampling of basements. Results from such
an initial site characterization can provide information on contaminant migration
15-112
-------
which can be used in the design and implementation of detailed soil and ground
water monitoring programs.
The results of the sampling program described above led to the evacuation
and destruction of a number of homes. A system of monitoring wells has been
installed to replace the basement sump sampling sites. The shallow aquifer is being
pumped and treated to arrest contaminant migration.
15-113
-------
CASE STUDY 20: USE OF PREDICTIVE MODELS TO SELECT LOCATIONS FOR
GROUND-WATER MONITORING WELLS
Point Illustrated
• Simple mathematical models can be used to estimate the longitudinal
and transverse spread of a contaminant plume. Wells can then be
located in areas expected to have elevated contaminant concentrations
and in areas thought to be both up-and downgradient of the plume.
Introduction
The use of mathematical models to estimate the migration of contaminants
can be helpful for several reasons, including: 1) fewer wells may be needed to
delineate a contaminant plume, and 2) wells can be rationally located in an attempt
to determine the maximum concentrations in a plume, its furthest extent, and
locations where concentrations should beat background levels.
Facility Description
The site was an electronics manufacturing plant that had been in operation for
20 years. Four large diameter, rock-filled "dry wells" had been used to dispose of
solvents and process wastes. These disposal units were between 35 and 60 feet
deep. Depth to ground water was over 460 feet. Disposal Units 1 and 2 had
received paint wastes and solvents, including trichloroethylene (TCE) and
tetrachloroethylene, between 1964 and 1979. Disposal Units 3 and 4 had been used
to dispose of plating solutions and spent acids between 1971 and 1977. These
solutions contained copper, chromium, nickel, lead and tin. All the disposal units
were closed in 1982. Exact quantities of wastes disposed are not known.
Geologic and Hvdrologic Setting
The site is located in a large alluvial basin in an arid region. The basin alluvium
is over 1,000 feet thick and consists of an upper sand and gravel unit, a middle silty-
clay unit, and a lower sand and gravel unit. Granitic bedrock underlies the
unconsolidated formations. Prior to large withdrawals of ground water, the upper
15-114
-------
unit had been saturated. At present, the silty-clay unit acts as an aquitard so that
water beneath it is under confined conditions. The potentiometric surface is now
350 feet below the land surface. In addition to a drop in water level elevations, the
ground-water flow direction has changed over the years from east to north in
response to changing pumping regimes. Estimated horizontal flow velocities have
varied from 10 to 40 feet/year.
Site Investigation
In 1982, city water officials discovered TCE in water samples from wells within
3 miles of the site. On its own initiative, the site owner began a pre-remedial
investigation, and then later a remedial investigation, to determine whether his site
could be a source of the TCE. The pre-remedial investigation provides an example
of how simple models can be used to determine well locations. The pre-remedial
investigation included sampling nearby wells and drilling a single deep sampling
well (over 500 feet deep).
Original plans called far locating the deep monitoring well between the waste
disposal units in an attempt to determine whether solutes had contaminated the
underlying ground water. However, site constraints, including an overhead power
transmission line, underground power lines and major manufacturing buildings,
necessitated that the monitoring well site be moved. The next step was to
determine an appropriate location for this well. Because of the changing ground-
water flow direction at this site, it was decided to use a simple mathematical model
to predict the areal extent of contamination from the disposal units. The results
would then be used in selecting a new location for the deep monitoring, well. Data
were collected to determine historical hydraulic gradients, pumping histories, and
aquifer hydraulic characteristics (e.g., conductivity, porosity). Following data
collection, a vector analysis model "the method of Mido" (1981) was used to predict
plume evolution. The results showed that the major plume migration was to the
north (Figure 15-36). Thus, the well was located north of the disposal units at a
distance of 60 feet from Unit 4.
15-115
-------
I
0
Final Site of Deep
Monitoring Well
DISPOSAL UNIT #4
DISPOSAL UNIT i
Original Planned
Deep Monitoring
Well Location
DISPOSAL UNIT #2-
DISPOSALUNITrT
Scale
100
Feet
BUILDING 2000
BUILDING 1000
Figure 15-36. Estimated Areal Extent of Hypothetical Plumes from Four Wells
15-116
-------
Case Discussion
Use of a model to predict potential plume migration at this site provided a
means of evaluating the long-term consequences of changing ground-water flow
directions and velocities. Thus, the pre-remedial investigation deep monitoring
well could be sited in the direction of net plume displacement, rather than at a
location which might have had a low probability of intercepting contaminated
ground water. A concentration below the detection limits from a well located
beyond the expected plume boundaries would have been inconclusive (for example,
see Figure 15-37). However, the deep monitoring well was located close to the
disposal units and in the direction of plume migration. Additional wells are now
being planned for the full-scale remedial investigation.
Reference
Mido, K.W. 1981. A economical approach to determining extent of ground water
contamination and formulating a contaminant removal plan. Ground Water,
Vol. 19, No. 1, pp. 41-47.
15-117
-------
WASTE SOURCE
SAMPLING WELL
•. ' • SATURATED ZONE • •--,",
YEARLY INCREMENTS OF WATER AND
CONSERVATIVE SOLUTE MOVEMENT
Figure 15-37. Consideration of Solute Migration Rates in Siting Sampling Wells.
If a monitoring well is sited farther downgradient than solutes could
have traveled in the time since disposal, low concentrations in the well
would certainly not prove that ground-water contamination had not or
was not occurring. Prior to locating a well, average linear velocities
should be estimated( v := Ki/ne where v = average linear velocity for
conservative solutes, K = hydraulic conductivity, i = ground-water
gradient, and ne = effective porosity). Using these estimates, and the
age of the disposal unit, T, an approximate migration distance, D, can be
computed (D = T/v) for conservative solutes associated with the waste.
For soil interactive solutes, migration distances will be less. Methods for
estimating these distances are given by Mills et at. (1985).
15-118
-------
CASE STUDY 21: MONITORING AND CHARACTERIZING GROUND-WATER
CONTAMINATION WHEN TWO LIQUID PHASES ARE PRESENT
Point Illustrated
• Monitoring and characterizing ground-water contamination when two
or more liquid phases are present requires knowledge of the physical and
chemical properties of each phase.
Introduction
Ground-water supplies are susceptible to contamination by immiscible organic
liquids. Organic liquids such as PCB-contaminated transformer oils, petrochemical
solvents, and motor fuels, because of their nature, often form a second liquid phase.
This separate liquid, in either the vadose or saturated zone, represents a problem in
multiphase flow. It is necessary to understand how these separate phases behave
where designing monitoring and sampling programs for sites contaminated with
such liquids. Techniques commonly used for single-phase flow systems may not be
appropriate.
Site D e scri p t i o n
The facility is a transformer manufacturing plant which experienced a major
discharge of polychlorinated biphenyls (PCBs) and trichlorinated benzenes (TCBs).
The discharge resulted from a break in a buried pipeline, but surface spillage may
have also occurred during production. The volume and duration of the subsurface
discharge is not known; neither is the quantity released by above ground spillage.
Geological and Hydroloqic Setting
The site is comprised of 10 feet of fill over lacustrine clay which varies in
thickness from 20 to 30 feet. Fractures with openings of approximately 0.1 cm have
been observed in the clay. Below the clay lies a thin silt layer. Below that is a 40- to
60-foot-thick layer of glacial till composed of fine sand near the top, and gravel,
sand, and silt below.
15-119
-------
Perched water about 3 feet deep flows laterally in the fill. The permanent
water table, located in the till, is partially confined. Potentiometric levels in this
latter system are between 25 and 30 feet below the land surface.
Sampling Program
Over 1000 soil samples were taken as part of the site investigation. A mobile
atmospheric pressure chemical ionization mass spectrophotometer (APCI/MS) was
employed for rapid, on-site characterization of soil samples. This instrument can
detect PCBs down to a minimum concentration of 100 mg/kg. About 20 percent of
the PCB analyses were replicated by conventional gas chromatography.
Granular dry materials were sampled from an auger with care taken in
cleaning sampling equipment to avoid cross-contamination. In taking samples from
the clay, special effort was made to sample the surfaces of obvious fractures. This
was done to maximize the changes of detection of PCBs in largely uncontaminated
soil. Due to dilution, large bulk samples can prevent the detection of contaminant
migration through fractures in low permeability soils.
Vertically, the soil sampling program showed PCBs to be distributed in a non-
homogeneous pattern within the clay zone. Concentrations of PCBs greater than
500 mg/kg PCBs were detected. The lateral spreading of PCBs throughout the fill
was much more extensive than the vertical movement. This could be due to the
nature of the discharge/spillage, pressure from the broken pipe, or the fact that the
fill is more permeable than the clay. The PCBs appear to have formed a layer along
the fill/clay interface. Movement of PCBs more than 300 feet laterally from the
original spill site has been confirmed.
Based on the soil sampling results, 12 well locations.(Figure 15-38) were chosen
to further characterize the site. Four boreholes were drilled into the till aquifer.
One well, 686-B, was placed upgradient of the spill site with a screened interval
between depths of 45 and 50 feet. The three downgradient wells in the till aquifer
were screened over different, intervals to increase the possibility of detecting a
separate organic liquid layer. The screened intervals used were at depths 45 to 50
feet (well 686-A), 50 to 55 feet (well 686-C), and 55 to 60 feet (well 686-D). Eight
15-120
-------
08
07
04
. 06
PARKING LOT
MANUFACTURING
PLANT
•636-B
direction of
ground water
flow
t
N
• deep well locations
o shallow well locations
Figure 15-38. Well Locations and Plant Configuration
15-121
-------
shallow wells were also placed in the fill to monitor the perched water. The fill is
approximately 10 feet deep and a layer of PCBs was suspected at the fill/clay
interface. The depth of the perched water fluctuates between 7 and 8 feet. Six of
the eight wells in the fill, 1, 3, 4, 6, 7, and 8, are screened from 7 to 10 feet. Samples
from wells 1, 6, 4, and 7 showed PCB levels much higher than the volubility limits.
The sampling results suggest that two separate liquid layers exist at these locations
and that the liquids are being mixed during sampling. Wells 2 and 5 were screened
from 5 to 8 feet to determine if a floating liquid layer was present. Again, samples
having concentrations far in excess of volubility limits indicated the existence of a
layer of organic liquid.
Ground-water systems contaminated with immiscible liquids require special
attention. Well screen intervals should be placed to intercept flow along
boundaries between soil layers of differing hydraulic conductivities and at water
table surfaces. Sampling results must also be interpreted properly. Samples
showing contaminant concentrations far in excess of volubility limits may indicate
that two layers of different liquids are being pumped and mixed.
Finally, Figure 15-39 is offered as an illustration of the types of complexity
which can be encountered with immiscible liquids having densities both greater
than and less than water.
15-122
-------
i'1-..,'.1-'.,1';,',:".'
Figure 15-39. Behavior of Immiscible Liquids of Different Densities in a Complex
Ground-Water Flow Regime
15-123
-------
CASE STUDY 22: METHODOLOGY FOR CONSTRUCTION OF VERTICAL FLOW
NETS
Point Illustrated
• Construction of a vertical ground-water flow net can be a valuable tool
for evaluating ground-water (and contaminant) pathways and for
determining additional actions that may be necessary to accurately
delineate the ground-water flow regime at a facility.
Introduction
Constructing a vertical flow net at a facility provides a systematic process for
analyzing the accuracy of ground-water elevation and flow data, and can therefore
foster a better understanding of the ground-water flow regime at the site.
Facility Description and History
The site contains a large chemical manufacturing facility of approximatley 300
acres located beside a major river in the northeastern United States. The site has
been used for chemical manufacturing by different companies since 1904 and has a
Icing history of on-site waste management. Several solid waste management units
have been identified at the facility. This is the same facility as discussed in Case
Studies 7,8,14,17 and 18.
Geologic and Hydrologic Setting: At depths of 150 to 200 feet the site is
underlain by bedrock identified as arkosic sandstone. Above this bedrock are glacial
deposits consisting of a thick bed of hard till, overlain by lacustrine sediments and
deltaic and outwash deposits. Discontinuous lenses of till were identified within the
deltaic deposits. A trough cut into the thick-bedded till and trending approximately
southeast to northwest has been identified. See Figure 15-40.
The river beside the facility flows westward and discharges into the main stem
of a larger river approximately 4 miles west of the facility. A small tributary (brook)
borders the facility to the southwest and west. Swamp-like areas are present near
15-124
-------
—— Top 6f Lower Till Contour Line
• Monitoring Wells
Figure 15-40. Top of Lowest Till Contour Map and Location of Vertical Flow Net
-------
the tributary. It is suspected that the arkosic sandstone outcrops in the river
adjacent to the facility. Whether this visible rock is a large glacial erratic or an
outcrop of the arkosic sandstone bedrock is an issue identified during previous
investigations and may be important in characterizing the ground-water flow
regime at the facility.
Program Design
The site was investigated in two phases. Phase I (1981-1984) included the
installation and monitoring of wells MW-1 through MW-12, while Phase II (1984-
1985) consisted of 34 soil borings, installation of wells MW-13 through MW-57, and
monitoring and sampling of all wells. This two-phased approach allowed the use of
the initial monitoring well data and soil boring data to determine the placement of
the Phase II monitoring wells. Further discussion of this two-phased approach is
provided in Case Studies 7 and 18.
Data Analysis
Evaluation of the data was conducted based on information provided by the
owner or operator, including the water-level elevation data presented in Table
15-13. Well locations and water-level elevations in the wells were mapped and
compared to elevations of the midpoint of the well screens to show relative
hydraulic head differences from well to well. Vertical gradients are a reflection of
different head values at different elevations. For each well, the head can be
determined at the elevation of the midpoint of the well screen by measuring the
water-level elevation in the well. Different head values corresponding to different
screen elevations were used to evaluate vertical gradients. During the plotting of
this map, anomalous data were identified and marked for further investigation.
The geology of the site and the depositional processes forming the aquifer
were studied to determine what sorts of hydrogeologic phenomena might be
expected. Glacial outwash deposits exhibit trends in sediment size and sorting.
Sediment size decreases and sorting increases from the marginal to the distal
portions of the deltaic/lacustrine deposits.1 It is expected that this tendency will be
'Mary P. Anderson, "Geologic Fades Models: What Can They Tell Us About Heterogeneity, presented to the American
Geophysical Union, Baltimore, May 18,1987.
15-126
-------
TABLE 15-13
GROUND-WATER ELEVATION SUM MARY TABLE PHASE II
Well
Number
MW-1
MW-2
MW-3
MW-4
MW-5
MW-6
MW-7
MW-8
MW-9
MW-10
MW-11
MW-1 2
MW-1 3
MW-1 4
MW-1 5
MW-1 6
MW-1 7
MW-1 8
MW-1 9
MW-20
MW-21
MW-22
MW-23
MW-24
MW-25
MW-26
MW-27*
MW-28
MW-29
MW-30
MW-31
MW-32
MW-33
Ground
Elevation
M
162.80
162.50
174.20
201.90
186.30
144.30
144.60
155.10
160.50
160.40
154.70
159.50
162.20
162.10
162.00
162.00
162.00
161.90
137.10
137.20
141.40
141.60
204.30
143.90
143.80
143.80
142.70
142.80
172.00
172.20
203.10
174.20
Well ,
Depth
(*)
76.50
22.50
31.00
54:00
47,50
39.50
19.50
24.00
61.00
30.00
27.00
26.50
29.00
29.00
29.00
29.00
71.00
72.00
24.00
17.00
26.50
15.10
225.50
70.00
39.00
24.00
46.00
23.00
85.50
24.85
61.00
94.00
Midpoint of
Well Screen
Elevation1
145.7
1504
141.3
107.3
127.6
133.6
135.0
132.9
130.2
135.5
139.2
139.1
139.1
135.5
104.5
103.4
116.6
123.7
118.4
13.0
-10.2
76.4
107.3
123.2
100.2
123.3
90.0
150.8
145.6
83.7
Screen
Length
(ft)
3,
3
3
3
3
3
3
3
3
3
3
3
10
10
10
3
25
25
5
5
5
5
20
5
5
5
5
5
5
5
5
5
Water
Level Elevation
9/1/82
150.54
1 56.85
149.95
135.78
135.94
149.04
141.53
144.62
140.57
141.05
141.22
140.66
140.67
140.87
140.52
140.53
127.83
127.82
135.39
135.35
184.98
136.47
130.20
130.17
127.86
127.88
152.70
151.68
154.78
150.49
*Not installed.
''Assumes screens are installed one foot above the bottom of the well.
15-127
-------
TABLE 15-13 (continued)
Well
Number
MW-34
MW-35
MW-36
MW-37
MW-38
MW-39
MW-40
MW-41
MW-42
MW-43
MW-44
MW-45
MW-46
MW-47
MW-48
MW-49
MW-50
MW-51
VIW-52
MW-53
VIW-54
MW-55
MW-56
MW-57
Screen
Reference
Points
SRP-1
SRP-2
JRP-3
SRP-4
SRP-S
SRP-6
SRP-7
SRP-8
Ground
Elevation
(ft)
186.20
203.20
189.40
189.50
189.30
154:90
173.80
173.70
134.20
133.50
139.50
144.32
144.15
141.50
141.60
143.00
143.00
157.00
157.00
159.30
145.80
145.90
133.60
141.30
Well
Depth
(ft)
75.80
106.25
101.20
48.00
135.30
68.00
47.50
75.30
64.00
32.10
28.00
35.00
25.00
34.00
17.00
72.20
30.20
70.30
34.00
77.90
52.00
35.00
20.30
Midpoint of
Well Screen
Elevation
113.9
100.4
91.7
145.0
57.5
90.5
129.8
101.9
73.7
80.9
115.0
112.8
122.6
111.0
128.1
74.3
116.3
90.2
126.5
84.9
97.3
114.4
116.8
Screen
Length
(ft)
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
114.41
114.92
116.05
115.86
NA
128.81
137.28
134.11
Water
Level Elevation
9/1/82
149.72
144.31
143.22
150.51
145.04
142.45
146.59
141.95
117.62
117.24
119.62
128.97
126.48
131.91
131.74
123.22
123.85
149.58
139.48
141.09
120.18
121.63
119.84
*Not installed.
'Assume screens are installed one foot above the bottom of the well.
15-128
-------
reflected in hydraulicconductivities throughout the outwash deposits at the facility.
There is some suggestion of such a trend in the head data from the site.
The map of hydraulic head values and screen midpoint elevations were
evaluated considering both the possible hydrogeologic phenomena expected for
the geology of the area and the depositional processes creating the aquifer. Several
working hypotheses were developed to explain the apparent ground-water flow
patterns and the identified vertical gradients.
• Hypothesis 1: Vertical gradients can be explained by classifying areas
where the vertical gradients were reflective of discharge and recharge
areas. (See Figure 15-41.)
• Hypothesis 2: The top surface of the till forms a trough with a saddle.
(See Figure 15-40.) The vertical gradients showing higher head with
depth reflect the movement of water as it flows upward over the saddle.
• Hypothesis 3: The vertical gradient may correlate with locations of
buildings and parking lots at the site. Recharge occurs primarily where
the ground is not paved. The downward gradient near the river may be
caused by runoff flowing downhill and recharging the ground water at
the edge of the pavement.
• Hypothesis 4: Most of the ground-water flow is horizontal. The vertical
gradients reflect, phenomena whose scale is. smaller than the resolution
of available data, and an accurate interpretation cannot be made.
Geologic systems exhibit heterogeneity on different scales, causing
fluctuations in head on different scales. The small-scale fluctuations
detected at the site are due to undefined causes and may represent:
1. details of stratigraphy (such as till beds in parts of the outwash
deposit),
2. artificial recharge and discharge (such as leaky sewer pipes), or
3. errors in the data.
15-129
-------
01
;
Mil MIIIIVI HUHHIII WPWtTI UFMM MtntHH
WUIral MIHHin IHWtll »•»»[> CH1MNT1
J
/
Legend
Wa«te Dlcppaal Area
Direction of Ground-Hater Flow
J
Figure 15-41. Recharge/Discharge Areas and Flow Directions
-------
To characterize flow at the site and to support the design of corrective
measures (if needed), a working (conceptual) model of flow at the site should be
developed. This model, in this case a vetical flow net, can be used to identify data
gaps and to prioritize gathering of the necessary additional information.
Considering the hypotheses developed, an area for characterizing the vertical flow
regime was selected. Determination of this area, where a geologic cross-section
and flow net will be constructed, was based on:
• Assumptions and requirements necessary to construct flow nets, as
identified in the Criteria for Identifying Areas of Vulnerable
Hydrogeology, Appendix B: Ground-Water Flow Net/Flow Line
Construction and Analysis (Vulnerable Hydrogeology, Appendix B). For
example, ground-water flow should be roughly parallel to the direction
of the cross-section and vertical flow net.
• Flow being representative of the hydrogeologyof the facility.
• Flow representing the major paths of ground-water movement. For
example, the aquifer is shaped like a trough and a major portion of the
ground-water flow occurs in the middleof this trough; therefore, a cross-
section and flow net should be constructed along the axis of the trough.
A geologic cross-section was constructed for the area of interest and is
identified as T-T in Figure 15-40. A flow net was then constructed following the
methodology described in Vulnerable Hvdrogeoloav, Appendix B; see Figure 15-42.
Construction of a vertical flow net requires a graphical solution of Darcy's Law.
Data that do not fit the solution become evident in Figure 15-42 as shown, for
example, by the head value for MW 52.
Construction of a vertical flow net allowed for a systematic evaluation of the
various hypotheses. Hypothesis 1, where vertical gradients are-labeled recharge and
discharge, is rejected because the gradients vary significantly in a very irregular
pattern (compare well clusters MW 14-18 and MW 12, and 53); there is no apparent
reason that natural recharge would vary so irregularly. Hypothesis 2 seemed
reasonable initially but, after closer inspection, is rejected because upward
15-131
-------
Ul
i
U)
Figure 15-42. Vertical Flow Net T-T'
-------
gradients are not consistently found near the saddle. Hypothesis 3 is feasible and
deserves further study. Aerial photographs were examined to identify paved and
unpaved areas, but the available ground-water data are insufficient for detailed
correlation to these distinct areas. Additional data are needed to construct a more-
detailed flow net to further evaluate this hypothesis. Hypothesis 4, which asserts
that most of the flow is horizontal addresses the area of the site where the major
portion of ground-water flow occurs. Although it relies on undefined causes to
explain fluctuations, it reflects the most logical explanation of the data.
Results
During construction of the flow net and testing of the hypotheses several
issues were identified. One of the most important gaps in the study to date is how
localized flow at the site fits into the regional ground-water flow regime. Regional
flow issues would need to be resolved prior to determining the extent and type of
corrective measures, if necessary. The following regional flow issues were
identified:
• Geologic information beyond the facility property boundary is necessary
to explain the suspected bedrock in the middle of the river directly beside
the site to characterize the regional ground-water flow (i.e., to
determine the possibility for contamination of regional ground water).
The difference in elevation of the top of the bedrock in the river and the
top of the bedrock throughout the facility is approximately 120 feet.
How can this be explained? Is the bedrock surface irregular or is this rock
a glacially-transported boulder exposed in the river? How does this
affect regional ground-water flow?
• , Data consistently show a downward gradient (i.e., recharge conditions)
near the river. This is difficult to explain because rivers in this region are
not expected to be losing streams (Heath, 1984). The expected flow
direction near a ground-water discharge area, in this case a gaining
stream, is upward. Data points showing downward flow near the river
are not included in flow net T-T. (Further investigation of vertical
gradients near the river is recommended). If this downward gradient
near the river is confirmed, near-water-table contamination could move
15-133
-------
downward and contaminate deeper ground water. If deeper, regional
contamination must be addressed, corrective measures maybe
significantly more difficult and extensive.
Other issues deal with localized flow patterns that may affect design of corrective
measures. Resolution of these issues will probably not change the overall scope of
corrective measures, but would need to be considered in the detailed design.
These localized flow pattern issues are as follows:
• The hydraulic head in the brook is higher than the head in the closest
wells in the aquifer, but the water slopes toward the stream. This is
inconsistent. If ground water from the site is not discharging into this
stream, fewer interceptor wells may be needed.
• Anisotropy must be taken into account in determining the region of flow
captured by interceptor wells, drains, etc.
• Till identified as lenses in outwash deposits may actually be continuous
with upgradient till, causing the aquifer to flow-under confined
conditions. Are the till beds isolated lenses or are they continuous? If the
till beds in the outwash aquifer are continuous and isolate adjacent
zones within the aquifer, they will have the potential of blocking flow to
interceptor wells that may be included in the corrective measures plan.
• Vertical gradients of 0.25 and 0.002 in the same geologic unit are
presented. Are these gradients accurate and how can they be explained?
There could be artificial discharge (pumping) or recharge (possibly from a
leaking sewer) near the wells showing a high vertical gradient. The areas
labeled discharge areas show no signs at surface water or other surficial
evidence of discharge. Artificial recharge and discharge may create areas
of relatively constant head, such as where ground water contacts leaky
sewers; these areas could limit the growth of cones of influence of any
interceptor wells or drains. Also, any contaminated water that may be
discharging from pipes should be identified and corrected.
15-134
-------
Case Discussion
Further investigation is necessary to resolve the above issues. Regional flow
issues should be resolved first. This information would be used to better
understand localized flow patterns which would affect the design of corrective
measures. The following options for further investigation are suggested:
i. Study the regional geology and hydrogeology. Techniques that could be
employed using existing data include review of geologic maps, analysis
of well logs, and interpretation of existing surface geophysical data (e.g.,
gravity and magnetic surveys). Measurement of water level elevations in
wells outside the site would also be useful.
2. Conduct a detailed study of the depositional, environment of the glacial
deposits on the site. This should provide a better understanding of flow
patterns.
3. Collect a full-year series of head data at existing wells to differentiate
transient from steady-state (e.g., artificial from natural) effects in the
measured heads.
4. Conduct multiple well pumping tests to determine the degree of
connectivity of geologic formations using wells at different depth and
locations. [Note: this should be done with careful attention to details of
well construction so that it is understood exactly what is being
measured.]
5. Collect detailed chemical data (including major ions and contaminants)
at the existing wells and interpret them to aid in characterizing the flow
regime.
6. Drill one or more wells into the bedrock near the river to determine the
vertical component of ground-water flow at this location.
Options 1 through 5 above are recommended prior to drilling additional wells
in the outwash deposits, unless more wells are needed to delineate the release.
15-135
-------
Further single-well hydraulic conductivity tests in the glacial deposits are not
recommended at this time. The large-scale flow in the outwash aquifer should be
determined by the location and relative degree of continuity of the till versus the
sand because the permeability contracts between the till and sand is so much
greater than the variability among the different sands. (See paper by Graham Fogg
in Water Resources Research, 22, 679.) Single-well tests would be useful for
determing localized hydraulic conductivities of the sand bodies, not their
connectivity.
Gathering existing data and constructing an initial vertical flow net proved
useful in identifying data gaps in defining ground-water flow, and identified
problems due to differing interpretations of the existing data. Determining options
for gathering additional data necessary to resolve these issues was based on a
qualitative understanding of the ground-water flow regime gleaned from
construction of the vertical flow net.
References
Fogg, Graham. Water Resources Research. 22. 679.
Heath. 1984. Ground Water Regions of the U.S. USGS Water Supply Paper No.
2242.
U.S. EPA. 1986. Criteria for Identifying Areas of Vulnerable Hvdroaeoloav.
Appendix B: Ground-Water Flow Net/Flow Line Construction and Analysis. Office
of Solid Waste. Washington, D.C. 20460.
15-136
-------
CASE STUDY 23: PERFORMING A SUBSURFACE GAS INVESTIGATION
Points Illustrated
• Design of a phased monitoring program to adequately characterize the
extent and nature of a subsurface gas release.
• The use of ambient air and basement monitoring to supplement
monitoring wel I data.
• The importance of subsurface characterization prior to design of a
monitoring network.
)
Introduction
Gases produced in a landfill will migrate via the path of least resistance.
Subsurface, lateral migration of landfill gas can occur due to natural and man-made
barriers to vertical gas migration, such as impermeable overlying soil layers, frozen
soil, or surface water. Installation of a gas-monitoring well network, in conjunction
with sampling in buildings in the area, can be used to determine the need for
corrective measures.
Facility Description
The unit in question is a landfill covering approximately 140 acres and
bordered by a river on one side and a floodwall on the other. Beyond the floodwall
lies a residential area (Figure 15-43 ). Several factors contribute to the subsurface
gas migration problem at this landfill. The site reportedly received large quantities
of organic wastes which, when decomposed in the absence of air, produce methane
and carbon dioxide gases. The presence of "tight", low permeability soils at the
ground surface (12 feet of clayey silt at the surface grading to coarse sand and
gravel at a depth of 55 feet) in the residential area, combined with a rapidly rising
water table below the landfill due to increased infiltration, restrict the vertical area
available for gas migration and encourage lateral movement.
15-137
-------
ui
U)
00
$$&$^^
FLOOOWALL D'6*
m-5 • • n-it
• n-7
n-i«
n-2»
n-4»
• nr-i
• nr-2
•n-3
• n-s
Figure 15-43. Site Plan
-------
Investigation of the gas migration began when foul odors and explosive levels
of methane (5 to 15 percent by volume in air) were discovered in the basement of a
home approximately 200 feet from the landfill. Residents in the area were
evacuated a sampling network was installed, and monitoring was conducted.
Sampling Program
The sampling was conducted in four phases, an initial screening phase and a
more detailed three-phase sampling program. The monitoring network for the
initial screening phase consisted of four wells (W1 through W4) aligned
perpendicular to the long axis of the landfill, in the direction of (and extending
beyond) the house where the gas was initially detected (Figure 15-43). The wells
were drilled to an approximate depth of 30 feet below the land surface with the
farthest well located about 1000 feet from the landfill boundary. These wells were
sampled twice a day for a month. Samples were analyzed for methane and
combustible hydrocarbons. The results of this initial monitoring showed average
methane levels to be highest at the monitoring well closest to the landfill (30
percent by volume), and roughly grading to below the detection limit at the well
farthest from the landfill.
Grab and composite ambient air samples were also taken at the landfill and
around houses in the neighborhood where gas was detected during the initial
monitoring phase. These samples were analyzed for methane and other
combustible, hydrocarbons. No gases were detected above normal background
levels in any of these above ground samples.
The next phase of monitoring (Phase I of the detailed sampling) involved the
installation of 14 new gas monitoring wells (1-1 through 1-14 in Figure 15-43). Most
of these were placed in a line 250 feet from and parallel to the longitudinal axis of
the landfill. Seven of these wells were drilled to an average depth of 55 feet, at
least 5 feet below the water table so that ground-water levels could be monitored.
The other seven wells averaged 30 feet and did not intercept ground water. As
shown in Figure 15-44, each well consists of three separate gas monitoring probes at
evenly spaced depth intervals. Each probe was packed in gravel to allow gas to
collect in its vicinity. Clay plugs were installed between each probe interval and
15-139
-------
CAST IRON COVER SET IN CONCRETE
GROUND SURFACE
VALVE
PROBE A
1/4" DIAMETER
POLYETHYLENE
TUBING
PROBE B
LEGEND
NATIVE SOIL
BENTONITE PLUG
PEA GRAVEL
VALVE
PROBE A
PROBE B
2" DIAMETER
PVC PIPE
PROBE C
2" DIAMETER PVC
WELL SCREEN
PROBE C
Figure 15-44. Gas Monitoring Well
15-140
-------
between the top probe and the surface to minimize vertical movement of gas in the
well. After two months of monitoring the well headspace twice monthly, concern
over the high levels of methane that were being measured prompted an expansion
of the monitoring well system.
The Phase II monitoring network involved another 14 wells (11-1 through 11-14)
installed to a depth of 6 feet along, three radial lines from the landfill. These wells
were monitored twice monthly with the. Phase I wells. Methane was not detected at
these wells because they were not deep enough to penetrate the clayey silt layer
which in this area extended to a depth of 12 feet. Had adequate boring logs been
compiled prior to the placement of these wells, the time and money involved in
their installation and sampling could have been saved.
Detailed soil boring logs were compiled during the installation of the Phase in
wells (111-1 through III-8 in. Figure 15-43). These wells were drilled to ground water,
averaging 55 feet in depth, were located in the vicinity of the Phase II wells, and
were constructed in the same manner as the Phase I wells, with three gas probes
placed in each well. The Phase III, wells were located from 510 to 900 feet from the
landfill. These wells were monitored twice a month for two months concurrently
with the Phase I wells. Methane levels at all but two Phase III wells (which are
located along the same radial line) exhibited explosive concentrations, ranging up
to 67 percent by volume in air. These high concentrations of gas prompted another
round of sampling of homes in the vicinity of wells exhibiting high methane
concentrations.
Methane and combustible hydrocarbons were measured in basements, crawl
spaces, and living areas of 28 homes adjacent to the landfill. All proved to be well
below the lower explosive limit of methane.
Wells were then selected based upon proximity to houses exhibiting the
highest levels of combustible gases, and sampled to determine gas composition and
concentration. The proportion of constituents in the collected gas was similar in, all
samples analyzed, and concentrations decreased with increasing distance from the
landfill.
15-141
-------
Ambient air sampling for organic gases at the landfill and in the residential
area was also performed at this time and showed low levels of several organic
compounds. Air samples collected in houses near the landfill showed the presence
of two of the gas components measured in the test wells (methane and ethane).
The gas migration hazard had been sufficiently characterized so that a plan for
corrective measures could be developed. This involved the installation of 31 gas
extraction wells which were located along a line between the landfill and the
residential areas, and a blower system to "pump" the gas out of these extraction
wells.
Results
The monitoring program implemented for this case was, for the most part,
effective in characterizing the extent and concentrations of subsurface gas
contamination. The four initial monitoring wells verified that the landfill was the
source of contamination. Phase I monitoring confirmed that the high levels of
methane were present at all depths monitored and along the entire length of the
landfill. The horizontal location of the Phase II wells, in lines radiating from the
landfill, was appropriate, although the lack of subsurface characterization rendered
them useless. Phase III sampling established the vertical and lateral extent of
subsurface contamination into the residential area.
Throughout the study, ambient air sampling as well as monitoring of homes in
the area of concern provided adequate safety control, as well as an additional
indication of potential migration of landfill-generated gases.
Case Discussion
Subsurface gas migration can occur when atmospheric ventilation of gases
generated in a landfill is insufficient. The gas produced migrates along the paths of
least resistance. Conditions restricting release to the atmosphere, such as saturated
or tight surficial soils, may force the gas to move laterally over considerable
distances.
15-142
-------
This case was selected as an illustration of a phased approach to monitoring a
subsurface gas release. The results of one phase of monitoring were incorporated
into the design of the next phase throughout the study. Monitoring was performed
at discrete vertical levels below the surface and at distances from the landfill that
were adequate to confirm the extent of the contaminant plume.
The study also illustrates the importance of characterizing subsurface
conditions prior to installing monitoring wells. Fourteen unusable wells were
installed and then monitored for two months because of insufficient preliminary
soil (stratigraphic) characterization.
The use of ambient and basement monitoring for gas to supplement
monitoring well data is also noted in this case study. The location of new wells can
be based in part on readings from these sources.
15-143
-------
CASE STUDY 24: USE OF A SUBSURFACE GAS MODEL IN ESTIMATING GAS
MIGRATION AND DEVELOPING MONITORING PROGRAMS
Point Illustrated
• Predictive models can be used to estimate the extent of gas migration
from a suspected subsurface source. This information can be used to
estimate human exposure and to determine appropriate locations for
monitoring wells and gas collection systems.
Introduction
Methane is a common landfill gas and is often used as an indicator of landfill
gas migration. The subsurface methane predictive model, described in Volume II,
Appendix D of this document, will yield a methane concentration contour map and
predict the distance that methane will migrate. The model consists of a series of
charts developed by imposing a set of simplifying assumptions on a general
methane migration computer model.
A methane migration distance prediction chart is used to find a preliminary
migration distance based on the age of the site and the soil type. The remaining
chart are used to find correction factors which are in turn used to adjust the
migration distance. These factors are based upon site characteristics (e.g., depth of
the waste).
Facility Description
The unit is located on a 583-acre site in a suburb of a major metropolitan area.
Figure 15-45 shows the site layout. The landfill itself occupies 290 acres. 140 acres of
the landfill were used for the disposal of hazardous wastes. Both hazardous and
nonhazardous wastes were disposed at the site from 1968 to 1984. Hazardous waste
disposal ended in 1984. The disposal of sewage treatment sludges and municipal
refuse continues. As seen in Figure 15-45, residential development has taken place
with houses now bordering the facility to the south. A population of 30,000 to
40,000 people reside within a mile radius of the landfill center.
15-144
-------
Central
DlncEloa
of
Ground Utter
Flow
LANDFILL PLAN
WELL LOCATION, MAP
Approximate Seal* 1":1130'
Figure 15-45. Facility Map
15-145
-------
The unit is a V-shaped fill overlying sediment and bedrock. The rock type is a
poorly consolidated, fractured sandy silt offering no lithologic barrier to gas
migration. The shape of the water table has not been established. Also unknown
are the possible effects of local, permeable formations such as sand lenses, faults,
etc.
The warm climate at the site encourages rapid degradation of organic wastes
and therefore rapid gas production. Site characteristics suggest that vertical gas
migration is not hindered. However, the compaction of the fill cover by truck traffic
combined with the rapid production of gas has forced lateral migration through the
fractured sandy silt.
Applying the Subsurface Methane Predictive Model
The subsurface methane predictive model-allows the development of a
subsurface methane concentration contour map. The model predicts, the distance
methane will migrate from a unit based on its age, depth, soil type, and
environmental factors. A contour map for two different methane concentrations, 5
and 1.25 percent, is predicted. The likelihood of human exposure can be estimated
from the location of the contours with respect to on-site and off-site structures.
Application of the model involves three steps. The first step is the prediction
of gas migration distances, based on the age of the landfill and the local soil type.
The unit of interest is 18 years old and has sandy soils. Figure 15-46 shows the
uncorrected methane migration distances for various soils over time. From
Figure 15-46, the uncorrected migration distances for the subject site are 165 feet
and 255 feet for 5 and 1.25 percent methane concentrations, respectively.
The second step in applying the model involves the application of a correction
factor to the migration distances based on waste depth. The deeper the waste, the
greater the opportunity for subsurface migration. Figure 15-47 is used to find the
correction factors for depth. For the subject waste unit the depth is 25 feet, which
corresponds to a correction factor of 1.0 for both concentrations.
15-146
-------
cn
5
300
255
UJ 200
165
u
o
o
cc
o 100
i
•4-
-r-CLAY-1.25%
!SAND-1.25%!
'SAND-5%'
L=CLAY-5%:
8 10 12 14 16 18 20
SITE AGE-YEARS
Figure 15-46. Uncorrected Migration Distances for 5 and 1.25% Methane
Concentrations
-------
00
25
9fl
ON FACTOR
-• i
ui i
i
U
UJ
oc
oc
O
5
^^M
" " ^
X
^'
•*••*"
^^
*"^
^^
• OK* ricco •
• ZO Ubtr •
^^^"
^^-•^
KG' DEEP -
^- — — "
ABOVE GH
. 100' DEEP
i
=r=^=
i
i
24 6 8 10 12 14 16 18 20
SITE AGE- YEARS
Figure 15-47. Correction Factors for Landfill Depth Below Grade
-------
The final step in applying the model is the correction of migration distances
based on surface venting conditions. The following equation is used to calculate
the adjusted correction factor, ACF:
ACF = [(ICF-1)(fraction of site which is impermeable)] + 1
The impervious correction factor, ICF, is obtained from Figure 15-48. In the above
equation, ICF is adjusted to account for the fraction of time the solid is saturated or
frozen and the fraction of the land area that is impermeable due to natural or man-
made barriers. If corrections for both time and area are required, the fractions are
additive. From Figure 15-48, the ICF for a unit 18 years old arid 25 feet deep is 2.4.
Site characteristics together with weather conditions indicate a value of 0.4 for the
fraction of impermeable area. Substituting these values into the above equation
yields an adjusted correction factor of:
ACF = [(2.4-1 )(0.4)] + 1 = 1.56.
Results
Table 15-14 summarizes the results from steps one through three of the model
application. The predicted migration distances for methane are found by
multiplying the uncorrected distance from step one by the correction factors from
steps two and three. The predicted distances of travel for methane are 255 feet and
395 feet for 5 and 1.25 percent concentrations, respectively.
TABLE 15-14
MODEL RESULTS
Methane
Concentration
(percent)
5
1.25
Uncorrected
Distance
(ft)
165
255
Correction
for Depth
1.0
1.0
Correction
for Venting
1.56
1.56
Corrected
Distance
Oil
255
395
15-149
-------
LFl
Ul
O
tc.
o
<
li.
z
o
5
UJ
DC
DC
8
3.0
2.0
1.0-
2.0
3.0
•^
SAND
CLAY
•^-£":
^-<^^
:~ — -
*^^-
-^-- -""'
>*. ^^
•*"
•
WIOUS - 10
^ptpv*°°
-IMPEBVIC
-1MPERVK
"IMPERVK
^ 'MPEflvic
5-26
-------
Case Discussion
Figure 15-49 is a methane concentration contour map developed from the
predicted travel distances. The map indicates that the possibility of human
exposure to landfill gas is high. Landfill gas is known to be present and well drilling
operations at the landfill have caused minor explosions. The monitoring wells along
the facility perimeter and testing in nearby homes indicate that gas has migrated
off site. Both the 5 percent and 1.25 percent methane contours enclose homes
evacuated because of gas accumulation. Measures have been taken to mitigate the
immediate problems and the landfill operators have installed additional gas
collection wells and extended the monitoring system.
15-151
-------
. I
UI
LEGEND:
HH PRIORITY t HOMES EVACUATED
PRIORITY II HOMES EVACUATED
\mtm PRIORITY II HOMES NOT EVACUATED
• PERIMETER GAS COLLECTION
SYSTEM WELLS
PREDICTED METHANE CONCENTRATION
APPROXIMATE SCALE: :
1" = 230'
Figure 15-49. Landfill Perimeter Gas Collection System Wells
-------
CASE STUDY 25: USE OF METEROROLOGICAL/EMISSION MONITORING DATA
AND DISPERSION MODELING TO DETERMINE CONTAMINANT
CONCENTRATIONS DOWNWIND OF A LAND DISPOSAL
FACILITY
Point Illustrated
• How to use meteorological/emission monitoring data and dispersion
modeling to estimate contaminant concentrations.
Introduction
Concern over possible vinyl chloride transport into residential areas adjacent
to a land disposal facility prompted initiation of this study. As a followup to a
screening assessment (involving emission modeling) a survey and emission
monitoring program with the application of an air dispersion model were used to
assess potential health hazards.
Facility Description
The facility is a landfill which has been in operation since 1963. The facility
occupies an area of 583. acres, of which 228 acres contain hazardous and municipal
waste. The facility and surrounding terrain is hilly with elevations ranging from 600
to 1150 feet above mean sea level. Residential areas are located immediately
adjacent to the south and southeast facility boundaries, as shown in Figure 15-50.
The facility previously received waste solutions from the synthesis of polyvinyl
chloride which included the vinyl chloride monomer. Gas is generated by municipal
waste decomposition and chemical waste volatilization. The primary air release
from the particular unit is vinyl chloride. A gas collection system has not been
installed for this unit.
15-153
-------
I/I
l/l
Meterological
^Monitoring Station A Meterological
EMISSION AREA
ESIDENTIALAREAA
RESIDENTIAL AREAS
Figure 15-50. Site Map Showing Location of Meteorological Sites A and B
(Adapted from Baker and Mackav.
-------
Program Design/Data Collection
A screening assessment (based on emission/dispersion modeling) was
conducted to evaluate vinyl chloride emissions from the landfill. Evaluation of
these results indicated that emission monitoring should be conducted to more
accurately quantify the release. An isolation flux chamber was used to measure
vinyl chloride emissions during a three-day period in August. This sampling period
was selected based on the screening assessment results to represent worst case
emission and dispersion conditions.
An on-site meteorological survey program was also conducted to characterize
wind flows at this complex terrain site. Two meteorological stations were deployed
to evaluate wind flows, as influenced by complex terrain, which may impact the two
adjacent residential areas (see Figure 15-50.) A one-month data collection period
during August was conducted to characterize on-site wind and stability patterns
during worst-case, long-term emission/dispersion conditions. Although the facilty is
located in complex terrain, the diurnal wind pattern during the meteorological
survey was very consistent from day to day. Therefore, the one-month
meteorological monitoring period was adequate for this RFI application.
Program Results/Data Analysis
The emission monitoring and meteorological monitoring data were used as
input for dispersion modeling. The wind patterns were different for each of the on-
site meteorological stations (see Table 15-15). Therefore, two sets of modeling runs
were conducted (meteorological station A data were used to estimate
concentrations at residential area A and meteorological station B data were used to
estimate concentrations at residential area B).
The dispersion modeling results indicated that estimated concentrations at
both residential areas were significantly below the RFI health criteria. Therefore,
followup air release characterizations were not necessary and information was
sufficient for RFI decision making.
15-155
-------
TABLE 15-15
SUMMARY OF ON-SITE METEOROLOGICAL SURVEY RESULTS
Prevailing daytime
wind direction.
Prevailing nighttime
wind direction
Station A
S
NNE
Station B
SW
ENE
15-156
-------
Case Discussion
Emission sampling was appropriate for this application because of the
uncertainties associated with emission rate modeling for landfills (including
uncertainties in emission modeling inputs such as the waste composition and spatial
distribution). The isolation flux chamber technique provided a basis for direct
measurement of vinyl chloride emission rates for dispersion modeling input.
The conduct of an on-site meteorological monitoring survey provided the
required wind and stability input for dispersion modeling. The use of multiple
meteorological towers for this application was necessary to characterize wind flow
patterns in complex terrain and to account for off-site exposure at two residential
areas subject to different wind conditions. The combination of emission
monitoring, meteorological monitoring and dispersion modeling provided an
effective air release characterization strategy for this RFI application.
References
Baker, L.W. and K.P. MacKay. 1985. Screening Models for Estimating Toxic Air
Pollution Near a Hazardous Waste Landfill. Journal of Air Pollution Control
Association, 35:11.
15-157
-------
CASE STUDY 26: USE OF, METEOROLOGICAL DATA TO DESIGN AN AIR
MONITORING NETWORK
Points Illustrated
• How to design an air monitoring program
• How to conduct an upwind/downwind monitoring program when
multiple sources are involved.
Introduction
A screening assessment (based on emission/dispersion modeling)
commensurate with RFI guidance was concluded to characterize hazardous air
constituents being released from a wood treatment facility. Evaluation of these
screening results indicated that it was necessary to conduct a monitoring program
to more accurately quantify air emissions from units at the facility. Meteorological
data were first collected to determine the wind patterns in the area. The wind
direction data with the locations of the potential emission sources were then used
to select upwind/downwind air sampling locations.
Facility Description
The site is a 12-acre wood treatment facility located in a flat inland area of the
southeast. Creosote and pentachlorophenol are used as wood preservatives; heavy
metal salts have been used in the past. Creosote and pentachlorophenol are
currently disposed in an aerated surface impoundment. Past waste disposal
practices included treatment and disposal of the metal salts in a surface
impoundment, and disposal of contaminated wood shavings in waste piles. The
constituents of concern in the facility's waste stream include phenols, cresols, and
polycyclic aromatic hydrocarbons (PAH) in the creosote; dibenzodioxins and
dibenzofurans as contaminants in pentchlorophenol; and particulate heavy metals.
The potential emission sources (Figure 15-51) include the container storage facility
for creosote and pentachlorophenol, the wood treatment and product storage
areas, the aerated surface impoundment for the creosote and pentachlorophenol
wastes, and the contaminated soil area which previously contained both the surface
15-158
-------
PREVAILING
WIND
DIRECTION .
INACTIVE SURFACE
IMPOUNDMENT AND
CONTAMINATED
WOOD SHAVINGS
STORAGE AREA
AERATED,
SURFACE
IMPOUNDMENT
• STATION 2 (V)
OFFICE
STATION 4 (V)®
TREATMENT
AND- PRODUCT
STORAGE AREAS
I
-------
impoundment for treating the metal salts and the wood shavings storage area.
Seepage from these waste management units has resulted in documented ground-
water and surface water contamination.
The area surrounding the facility has experienced substantial development
over the years. A shopping center is now adjacent to the. eastern site perimeter. This
development has significantly increased the number of potential receptors of air
releases of hazardous constituents.
Program Design/Data Collection
Preliminary Screening Survey-
A limited-on-site air screening survey was first conducted to document air
releases of potentially hazardous constituents, to prioritize air emission sources, and
to verify screening assessment modeling results and the need to conduct a
monitoring program. Total hydrocarbon (THC) levels were measured with a
portable THC analyzer downwind of the aerated surface impoundment, wood
treatment area, and product storage area. Measurements were also made upwind
of all units to provide background concentrations. Because THC levels detected
downwind were significantly higher than background levels, a comprehensive
monitoring program to characterize releases to the air was designed and
implemented.
Waste Characterization-
To develop an adequate monitoring program, the composition of wastes
handled in each waste management unit was first determined to identify which
constituents were likely to be present in the air releases. Existing water quality data
indicated contamination of ground water with cresols, phenol, and PAHs and of
surface water with phenols, benzene, chlorobenzene, and ethyl benzene. A field
sampling program was developed to characterize further the facility's waste stream.
Wastewater samples were collected from the aerated surface impoundment and
soil samples were collected from the heavy metal salt waste treatment/disposal
area. Analytical data from this sampling effort confirmed the presence of the
constituents previously identified. Additional, constituents detected included
toluene and xylenes in surface impoundment wastes, and arsenic, copper,
chromium, and zinc in the treatment/disposal area.
15-160
-------
Based on their individual emission potentials and potentials for presenting
health and environmental hazards, the following constituents were selected for
measurement in the air monitoring program:
Volatile/semi-volatile constituents: toluene, benzene, total phenols, penta-
chlorophenol, PAHs, cresols
Particulateconitituents: arsenic, copper, chromium, zinc.
Meteorological Data Collection--
Meteorological information is critical for designing an air monitoring program
because stations must be located both upwind and downwind of the contaminant
sources. Therefore, a one-month meteorological monitoring survey was conducted
at this flat terrain site. The survey was conducted under conditions considered to be
representative of the summer months during which air samples would be collected.
Summer represented worst-case conditions of light steady winds and warm
temperatures. The collected meteorological data showed that the local wind
direction was from the southeast. No well-defined secondary wind flows were
identified.
Initial Monitoring-
Alternative methods were considered for monitoring emissions from the
aerated surface impoundment and contaminated storage area. Direct emission
measurements (such as use of isolation flux chambers) would not be practical for
aerated ponds or for monitoring particulate emissions from area sources.
Therefore, an air monitoring program with samplers located in proximity to the
other units of concern was selected for this application.
The on-site meteorological survey data were used with the EPA atmospheric
dispersion model, ISC (Industrial Source Complex Model), to estimate-worst-case air
emission concentrations and to help determine the locations for the air sampling
stations. The ISC model was used because it is capable of simulating conditions of
point and non-point source air emissions. Using the established southeast wind
direction, maximum downwind concentration were predicted for different
15-161
-------
meteorological conditions (e.g., wind speed). Upwind background stations and
downwind monitoring stations were selected based on the predicted dispersion
pathways. Because the releases from the individual waste management areas
overlapped, the model also provided a means for separating the incremental
contamination due to each source.
Figure 15-51 shows the locations of the selected sampling stations. Station 1 is
the upwind background station. Here background volatile concentrations,
particulate concentrations, and meteorological conditions were monitored.
Stations 2 and 4 were located to identify volatile emissions from the aerated surface
impoundment and wood treatment/product storage areas, respectively. Station 3
was located downwind of the inactive surface impoundment/wood shavings
disposal area. This station was sited to document releases from these waste
management units and to document worst-case concentrations of volatiles and
particulate at the facility property boundary. For this application the locations of
Stations 2, 3 and 4 were adequate to characterize, air concentrations at both the unit
boundary as well as the facility property boundary (due to the proximity of these
two boundaries in the area downwind, based on the prevailing wind direction, of
the units of concern). A trailer-mounted air monitoring station was used, to
supplement the permanent stations and to account for any variability in wind
direction.
Sample Collection-
The air quality monitoring was conducted over a three month period during
the summer. Meteorological variables were measured continuously on site
throughout the study. Air samples were taken over a 24-hour period approximately
even six days. The sampling dates were flexible to insure that worst-case conditions
were documented.
Volatile and semi-volatile constituents were sampled by drawing ambient air
through a sampling cartridge containing sorbent media. A modified, high, volume
sampler consisting of a glass fiber filter with a polyurethane foam backup sorbent
(EPA Method T04) was used to sample for total phenols, pentachlorophenol, and
PAHs. Benzene and toluene were collected on Tenax sampling cartridges (EPA
15-162
-------
Method T01) and cresol was collected on silicagel cartridges (NIOSH Method Z001).
Participate were collected on filter cassettes using high-volume samplers.
In addition to the constituents previously discussed, Appendix VIII metals were
analyzed on the first few sets of samples. These analyses were conducted to identify
air releases of constituents other than those known to be present. The results
indicated that no additional constituents were present in significant concentrations,
so the additional analyses were dropped for the remainder of the study.
Program Results/Data Analysis
Standard sampling/analytical methods were available for all the target
monitoring constituents. Analytical detection limits were below specific health
and environmental criteria for all constituents except cresol. The high analytical
detection limit for cresol which exceeded reference health criteria complicated data
analysis. This difficulty was handled by the routine collection and analysis of waste
water samples during the air monitoring program. These data were used to
estimate cresol levels in the air by comparing its emission potential to the other air
monitoring constituents which have relatively low detection levels.
Analytical results obtained during this sampling program established that
fugitive air emissions significantly exceeded reference health criteria. Source
control measures were implemented to reduce emission concentrations below
health criteria levels. Subsequent air monitoring was conducted at the same stations
used previously on a weekly basis immediately after implementation of the
remedial measures, and on a quarterly basis thereafter.
Case Discussion
This case illustrates a sequence of tasks which were taken to design an air
monitoring program at a site with multiple air emission sources. An initial field
survey was conducted to identify decal prevailing wind patterns and to identify
potential downwind receptors of fugitive air emissions. The meteorological survey
results were used to design an effective monitoring network. Monitoring station
locations were selected to obtain background conditions and to document air
releases downwind of each emission source. Also, the monitoring strategy included
15-163
-------
use of a portable sampling station to provide flexibility in sampling locations to
account for variation in wind direction. Spatial variability in air concentration was
assessed with the aid of an air dispersion model to assist in data interpretation.
Air, emissions data showed an air release of hazardous constituents
significantly above health criteria levels. Remedial measures were implemented,
and periodic subsequent monitoring was conducted to insure compliance with the
health criteria.
References
Methods T01 and T04. Compendium of Methods for Determination of Toxic Organic
Compounds in Ambient Air. 1984. EPA-600/4-84-041.
Method Z001, NISOH Manual of Analytical 1984, National Institute of
Occupational Safety and Health.
15-164
-------
CASE STUDY 27: DESIGN OF A SURFACE WATER MONITORING PROGRAM
Point Illustrated
• When designing a surface water monitoring program, site-specific
sediment and suspended solids information should be considered.
Introduction
Designing a surface water monitoring program to determine the extent of
contamination involves identifying the potential waste sources, the contaminants
likely to be present in each waste stream, and the flow paths by which the
contaminants could reach surface waters. The fate of the contaminants once they
reach the surface water must also be considered when selecting sampling stations
and parameters to be measured. The example described here illustrates-the design
of a monitoring program for a river system.
Facility Description
A facility which processed zinc, copper and precious metals from ores operated
along a river for five years. The plant was closed after being cited for repeated fish
kills which were reportedly due to failures of a tailings pond dike. At present, the
site is covered with tailings containing high concentrations of copper, zinc,
cadmium, arsenic and lead. There is no longer a tailings pond.
Site Setting
The site is located on coarse colluvium (hill-slope deposits of weathered
bedrock) and fine-grained alluvium. These deposits, are, typically 50 feet thick.
Metamorphic rock (phyllite) underlies the unconsolidated materials. Ground water
moves laterally in the gravel formations from the steep valley walls towards the
river.
The site is about 400 feet from the river. Two drainage ditches cross the lower
portion of the site and merge prior to leaving the site. The ditch carries the
15-165
-------
combined flow and discharges directly into the river (Figure 15-52). No other
tributaries enter the river within two miles of the location.
Sampling Program
A surface water monitoring program was designed as part of the Phase I
remedial investigation to determine the extent of contamination in the river.
Existing data from a reconnaissance visit had shown high concentrations of metals
in the drainage ditch sediments (e.g., 5,170 mg/kg Cu and 11,500 mg/kg Zn). Ground
water data from the plant's well showed concentrations of Cu (7 ug/l) and Zn (54
ug/l). The contribution of metals to the river by ground-water discharge at the site
was considered to be relatively small.
Based on a review of the plant history and the available water quality and
sediment data, a monitoring program, was designed. The potential pathways by
which metals could reach the river appeared to be direct discharge from the
drainage ditch, seepage of contaminated ground water, and storm water runoff.
Plant records indicated that typical flows in the drainage ditch at its confluence with
the river varied from 1 to 3 cubic feet per second (cfs) in the spring. During extreme
flood conditions, the flow in the ditch exceeded 20 cfs. In the summer, flows in the
drainage ditches at all locations were less than 0.5 cfs. Resuspension of
contaminated sediments in the ditches during storm runoff appeared to be the
most likely pathway for metals to reach the river. The specific metals of concern
were identified as As, Cd, Cu, Pb and Zn based on the processes used at the plant
and the composition of the ores which contained some arsenopyrites (As, Cu),
galena (Pb), and sphalerite (Zn, Cd).
The available soil and water quality data, from the reconnaissance visit were
reviewed to determine the likely fate of the metals. Soils in the area were
circumneutral (pH = 6.5) and contained about 0.5 percent organic matter by
weight. Thus the metals, particularly Pb, would, be, expected to adsorb onto the soil
particles. In the on-site tailings piles, the pH of core samples ranged between 3.3
and 4.9. Low soil pH values had been measured in sediments in the drainage ditch
just downgradient of the tailings pile. The pH of the river during the
reconnaissance was 6.9. The suspended solids concentration was 10 mg/l.
15-166
-------
River
—^
• 39 (High flow only)
To S8-
Area of
Former
Tailings
Pond
35
Site Operations
Drainage Ditch
Sampling Station
N
t
Scale
i I
0 160 feet
Figure 15-52. Sampling Station Locations for Surface Water Monitoring
15-167
-------
Estimates of the distribution of metals between the dissolved and adsorbed
phases for a range of partition coefficients (Kp) are shown in Table 15-16. For
example, if Kp= io4the suspended solids concentration was 10 mg/l, 90
percent of the metal present would be in the dissolved phase. This information
indicated that even though a metal (e.g., lead), was known to sorb strongly, a
significant amount could be transported in the dissolved phase. Thus, both water
and suspended solids should be analyzed for metals. The complete list of
parameters selected for measurement in the Phase I investigation and the rationale
for their selection are outlined in Table 15-17.
The sampling stations were selected to determine river quality up- and
downstream of the site and to determine whether particulates with sorbed metals
were deposited on the river banks or streambed. The sampling stations and the
rationale for their selection are listed in Table 15-18. The station locations are
shown in Figure 15-52. Because floods were considered to be one cause of
contamination incidents, samples were to be collected under both high and low
flow conditions.
Selected results of the surface water quality sampling program for spring
conditions are given below:
Station
S5 (mouth of ditch)
S7 (upstream)
SB (downstream)
Dissolved Copper
Concentration, ug/8.
1110
2.7
4.0
15-168
-------
TABLE 15-16
RELATIONSHIP OF DISSOLVED AND SORBED PHASE POLLUTANT
CONCENTRATIONS TO PARTITION COEFFICIENT AND SEDIMENT
CONCENTRATION
K P
10°
1 0'
1 O2
1 0"
1LT
ss
(ppm)
1
10
100
1000
10,000
1
10
100
1000
10,000
1
10
100
1000
10000
1
10
100
1000
10000
1
10
100
1000
10,000
Cw/C/
1.0
1.0
1 .0
1.0
1.0
1.U
1 . 0
1 .0
1 .0
0.9
1.0
1.0
1.0
0.9
0.5
1.0
1 .0
0.9
0.5
0.1
1.0
0.9
0.5
0.1
0.0
If CT = 100 p p b
cw=
100.
100.
100.
100.
100.
100.
99.9
99.0
90.9
100.
99.9
99.0
90.9
50.
99.9
99.0
90.9
50.
9.1
99.0
90.9
50.
9.1
1.0
x =
100.
100.
100.
100.
99.
1x103
1x103
999.
990.
909 .
1 x 104
1x104
9.9x10"
9.1x103
5x 1 O3
1x105
9 . 9 x 1 O4
9.1x104
5x1 O4
9x1 O3
9.9x10b
9.1x105
5x105
9.1x104
9.9x103
Cs =
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0 . 1
1.0
9.1
0.0
0.1
1.0
9.1
50.
0.1
1.0
9.1
50.
90.9
1.0
9.1
50.
90.9
99.0
After Mills et al., 1985.
The fraction dissolved (CJC,) is calculated as follows:
CT
KDx5x10'£
here Kp = partition coefficient,I/kg
SS = suspended solids concentration, mg/l
15-169
-------
TABLE 15-17
PARAMETERS SELECTED FOR SURFACE WATER MONITORING
PROGRAM
Parameters
Metals- As, Cd,Cu,Pb,Zn
PH
Dissolved Oxygen, Sulfide,
•Fe(1l), Fed1,1,)
Alkalinity
Total Dissolved Solids
Major Cations (Ca? * , Mg2 + ,
Na+,K+,NH+J
Major Anions (C1-, SOA2 ,NO,-)
Suspended Solids
Streamflow
Rationale
Determine extent of contamination
Predict sorption behavior, metal
solubility, and speciation
Determine redox conditions which
'Influence behavior of metals,
particularly the Jeaching of tailings
A measure of how well buffered a
water is; allows consideration of the
likelihood of pH change
Used as a water quality indicator and
for QA/QC checks
May identify other waste sources;
can influence fate of trace metals
Predict the fraction of metal in water
which issorbed
Compute mass balances and assist in
identifying sources of observed
contamination
15-170
-------
TABLE 15-18
SELECTED SURFACE WATER MONITORING STATIONS AND RATIONALE
Station
Drainage ditch west of site
(51)
Drainage ditches on site (S2
and S3)"
Downstream of confluence of
2 ditches (S4)
Mouth of drainage ditch (S5)
River (S6.S7, and S9)
River (58)
Media
Water and sediments
Water and sediments
Water and sediments
Water, suspended
sediment, bedload
Water, suspended
sediment, bedload
Water, suspended
sediment, bedload
Rationale , ,
Determine whether off-site drainage is
significant source of contamination
Identify on-site sources
Provide information for checking mass
balances from the two drainage ditches
Determine upstream water quality
Determine upstream water quality
/
Determine quality downstream of site
and provide data for mass balance
15-171
-------
A mass balance was computed to determine how much of the apparent decrease
from the ditch (S5) to the downstream river sampling point (S8) was due to dilution
and how much could be attributed to other processes (e.g., sorption, precipitation).
The concentration in the river considering dilution alone was predicted using the
following mass balance equation:
CyQu+ CWQW
CR=
Q u+ Qw
where
CR = downstream concentration of pollutant in river following mixing with
ditch waters (S8), ug/l
Cw= concentration in ditch water (S5), ug/l
Cu = concentration in river above ditch (S7) ug/l
Qw= discharge rate of ditch, ftVsec
QU = flow rate of river above ditch, ftVsec.
At the time of sampling, the flow in the ditch at station S5 was 1 cfs and the river
flow at station S7 was 155 cfs. Using the above equation, the predicted river
concentration for Cu was approximately 10 ug/l. (The observed concentration was 4
ug/l.) The observed decrease in concentration was primarily due to dilution,
although other attenuation processes (e.g., sorption) were probably occurring. The
expected sorbed concentration was estimated as follows:
X = KPC
where
x = sorbed concentration, ug/kg
K p = partition coefficient, I/kg
c = concentration of dissolved phase, ug/l.
Hererthe sorbed concentration of Cu was estimated as 8 x 105 ug/kg (800 mg/kg).
15-172
-------
Case Discussion
This case illustrates the use of site-specific data and the use of information on
the environmental fate of contaminants in the design of a surface water monitoring
program. Site data are needed to locate waste sources and to determine the likely
flow paths by which contaminants reach rivers. An understanding of the general
behavior of the contaminants of interest, and of the factors which influence their
fate is helpful in determining where samples should be collected and what
parameters, particularly master variables, should be measured. Collecting data on
such parameters (e.g., pH, suspended solids) ensures that the necessary information
is available to interpret the data.
15-173
-------
CASE STUDY 28: USE OF BIOASSAY5 AND BIOACCUMULATION TO ASSESS
POTENTIAL BIOLOGICAL EFFECTS OF HAZARDOUS WASTE ON
AQUATIC ECOSYSTEMS
Point Illustrated
• Measurements of toxicity (i.e., bioassays) and bioaccumulation can be used to
assess the nature and extent of potential biological impacts in off-site areas.
Introduction
A study was conducted to determine whether leachate discharged into surface
waters had adversely affected biota in a stream adjacent to a waste site and in a
nearby lake. The components of the study included chemical analyses of the
leachate, surface waters, sediments, and tissue samples; toxicity testing of the
surface waters; and surveys of the structure and composition of the biological
communities. Tissue analyses are important for determining contaminant bio-
accumulation and assessing potential human exposure through consumption of
aquatic organisms. Toxicity testing is important for determining potential lethal
and sublethal effects of contaminant exposure on aquatic biota. Although
ecological analysis of community structure and composition is also an important
component of biomonitoring, it will not be discussed here since the focus is on the
relationships between the leachate source, the distributions of contaminants near
the waste site, and the toxic effects and bioaccumulation of the contaminants in the
tissues of local fauna.
Site Description
The 5-acre facility is an industrial waste processing site which accepts wastes
from nearby plastic manufacturing and electroplating industries. Liquid wastes are
dewatered on site prior to removal to an off-site disposal area. The principal wastes
processed at the faclity include several organic compounds and metals.
The site contains a wastewater impoundment with numerous seeps and
drainage channels that transport leachate into an adjacent river (Figure 15-53). The
river flows from northeast to southwest, and is joined by a tributary stream before
15-174
-------
entering a nearby lake. The RFA indicated an oily sheen associated with a strong
chemical odor on the surface of the stream below the treatment pond, and further
reported numerous violations of the NPDES permit. Subsequent analyses of samples
taken from the drainage channels and seeps flowing into the river showed high
concentrations of organic and trace metal contaminants, principally bis(2-
ethylhexyl). phthalate, ethylbenzene, phenol, copper, cadmium, and zinc.
Sampling Program
Six stations were sampled to assess possible toxicity and bioaccumulation of
released substances (Figure 15-53). Station 6, located upstream of the release, was
selected as a reference location for the stream. Station 17 was selected as a
reference location for the lake because it is distant from the river mouth and
because prevailing winds from the northwest direct the river discharge along the
southeast shore of the lake away from the station. Stations 7, 15, and 18 were
selected to determine the extent of toxic impacts on river and lake biota.
Water, sediments, and tissues of bottom-dwelling fishes (brown bullhead
catfish, Ictalurus nebulosus) were collected at each station. Concentrations of bis(2-
ethylhexyl) phthalate, ethylbenzene, phenol, copper, cadmium, and zinc were
measured in each matrix. Analyses were conducted according to U.S. EPA guidelines
for sediments, water, and tissues. Water quality variables (dissolved oxygen,
temperature profiles, and alkalinity), total organic carbon in sediments, and lipid
content of tissues were also measured.
Three independent bioassays were conducted on each water sample. The test
species and endpoints used in the bioassays were those recommended in the U.S.
EPA protocol for bioassessment of hazardous waste sites (Tetra Tech, 1983). Growth
inhibition in the alga Selanastrum capricornutum. and mortality in the crustacean
Daphnia maqna were determined using U.S. EPA (1985) short-term methods
for chronic toxicity testing. Inhibition of enzyme-mediated" luminescence in the
bacterium Photobacterium phosphoreum (i.e., the Microtox procedure) was
measured according to the methods established by Bulich eLaL (1981).
15-175
-------
en
Figure 15-53. Site Plan and Water Sampling Locations
-------
Results
Results of the survey indicated that concentrations of organic contaminants in
the surface waters were generally less than U.S. EPA water quality criteria, but that
concentrations of inorganic contaminants generally exceeded water quality criteria
at Stations 7, 15, and 18 (Table 15-19). In comparison with the reference stations,
significant sediment contamination was evident at Stations 7, 15, and 18 for the
three trace metals (Table 15-20). Tissue concentrations of organic substances
exceeded detection limits for bis(2-ethylhexyl) phthalate at Stations 7 and 15, and
for ethylbenzene at Station 7(Table 15-21 ). However, trace metal concentration in
tissues were highly elevated at Stations 7 and 15, but only slightly elevated at
The bioassay data showed a considerable range in sensitivity, with the algal
bioassay being the most sensitive (Table 15-22). Consequently, the bioassay results
were normalized to the least toxic of the reference stations (i.e., Station 6) to
compensate for the wide range of sensitivity among, the test species (Table 15-23).
Overall, the bioassay results showed a high degree of agreement with contaminant
concentrations in water and sediments (Figure 15-54, Table 15-19 and 15-20).
Stations 7 and 15 showed highly toxic results, and Station 18 indicated moderate
toxicity. Only the algal bioassay indicated significant, but low, toxicity at Station 17
(the lake reference station).
In summary, the results indicated that the organic contaminants were less of a
problem than the trace metals in terms of bioaccumulation and potential toxicity.
Most of the observed toxicity was attributed to trace metal contamination, which is
consistent with the elevated concentrations of trace metals measured in the water,
sediments, and tissues.
Case Discussion
This case study provides an example of a biomonitoring program designed to
characterize the relationship between a contaminant source, contaminant
concentrations in sediments and water, bioaccumulation in tissues, and receiving-
water toxicity. It should be recognized that in many instances, the relationship
15-177
-------
TABLE 15-19
MEAN CONCENTRATIONS (yq/l) OF ORGANIC SUBSTANCES AND TRACE METALS
IN LEACH ATE AND SURFACE WATERS3
Chemical Class
Base Neutral
Volatile
Acid Extractable
Metals
Chemical
• . '. .,
Bis (2-ethylhexyl)
phthalate
Ethyl benzene
Phenol
Copper
Zinc
Cadmium ,
-
Station
Seep
L1
600
100
1500
4300
35,000
4800
River
6
2
1
<1
<1
17
<1
River
7
11
1
18.37
489
4290
146
" Cake
15
10
<1
<1
56
1100
49
Lake
18
1
* 1
<1
26
37
<1
Lake
17
2
2
<1
2
35
<1
Water Quality.
Criteria"
Acute-
940
32,000
10,200
18
320
3.9
Chronic
3
NAC
2560
12
47
1,1
aRiver and lake alkalinity = lOOmgCaCOs/L
bTrace metal criteria adjusted for,.alkalinity
"Not available for this substance
15-178
-------
TABLE 15-20
MEAN SEDIMENT CONCENTRATIONS (pg/kg DRY WT) OF ORGANIC
SUBSTANCES AND TRACE METALS
Chemical Class
Base Neutral
Volatile
Acid Extractable
Metals
Chemical
;
Bis (2-ethylhexyl)
phthalate
Ethyl benzene
Phenol
Copper
Zinc
Cadmium
Station
S@€p
1 L1
NA*
NA
NA
NA '
NA
NA
River
6
216
10
<30
3
11
<0.1
.River
7
1188
34
<30
1663
28,314
19
Lake
15
1080
20
<30
190
7260
6
Lake
18
108
14
<30
ss
24
<0.1
Lake
17
216
8
<30
7
23
<0.1
aNot applicable (NA).
15-179
-------
TAB LEI 5-21
MEAN LIVER TISSUE CONCENTRATIONS (ng/kg WET WT) OF ORGANIC
SUBSTANCES AND TRACE METALS
Chemical Class
Base Neutral
Volatile
Acid Extractable
Metals
Chemical
Bis(2-ethylhexyl)
phthalate .
Ethyl benzene
Phenol :
Copper
Zinc
Cadmium :
Station
Seep
LI-
NAa
NA
NA
NA
NA
• NA-
River :
6
<25
<5
<30
1t8
983
ITS
River
; 7
95
9
<30
1600
28,400
,1500
Lake
15
86
<5
<30
750
8500
639
Lake
18
<25
<5
<30
237
2139
190,
Lake
17
<25
<5
<30
180
1420
.125
aNot applicable (NA),
15-180
-------
TABLE 15-22
MEAN LC5o AND ECSo VALUES (PERCENT DILUTION) FOR SURFACE-WATER
BIOASSAYS3
Bioassay
f
Algae
Daphnia
Microtox
End point
Growth inhibition
(EC50%)a
Mortal itytLGso^)11
Decreased
luminescence
(EC50%)a
Station
Seep
LI
NAb
NA
NA
River
6
>100<
>100
> too
River
7
0-4
3.3
5.6
Lake
IS
10.0
18.5
15.0
Lake
18
24.9
100.0
43.4
Lake •
17
75:0
9Q.O
>TOO
aPercent dilution required corresponding to a 50 percent response
bNot applicable (NA) because leachate toxicity was not tested
'Response of > 100 indicates that samples were not toxic at all dilutions tested
dPercent dilution corresponding to 50 percent mortality
15-181
-------
TABLE 15-23
RELATIVE TOXICITY OF SURFACE-WATER SAMPLES9
Bioassay
Algae
,. ...
Daphnia
Microtox
-.. - -
Endpoint
i
Growth inhibition
{EC50,%)
Mortal ity(LC50%|
Decreased
luminescence
(EC50%)a
Station
Seep
L1
NAb
NA ,
NA
River
6
0.0
, 0.0
0.0
River
7
99.6
36,7
94.4
Lake
15
90.0
S1.S
85.0
Lake
18
75.1
,..
0.0
56.6
Lake
17
25.0
10.0
0.0
"Relative toxicity = 100 x [(Reference Station - Impacted Station)/Reference Station]
bNot applicable (NA) because leachatetoxicity was not tested
15-182
-------
1
o
.a
*_/
O
K>
s
1_
0
O
IO
2
100
90 -
80
70 i
60
50 -
40 -
30 -
20 -
10 -
0
m
River 6
Late 15
Station
Lake 18
Laka 17
100
o
I
90 -
80 -
70 -
60 -
50 -
40 -
30 -
20 -
10 -
River 6
Rfv«r 7
Lake 15
Algae
Station
LNJSl Daphnia
Lake 18
Micretex
Lake 17
Figure 15-54. Bioassay Responses to Surface Water Samples
15-183
-------
between contaminant concentrations in the water and toxicity till not be as clear-
cut as described in this example. Consideration of the chemical composition in
leachate samples, mass balance calculations, and transport and fate mechanisms
may indicate that sediments are the primary repository of contaminants. In such
instances; sediment bioassays rather than receiving-water bioassays may be better
suited for characterization of potential toxic effects on local fauna.
R e fe re nce s
Bulich, A.A., M.W. Greene, and D.L. Isenberg. 1981. Reliabiltv of the bacterial
luminescence assay for determination of the toxicity of pure compounds and
complex effluent, pp. 338-347. In: Aquatic toxicology and hazard assessment:
Proceedings of the fourth annual symposium. ASTM STP 737. D.R. Branson and K.L.
Dickson (eds)., American Society for Testing and Materials, Philadelphia, PA.
Tetra Tech. 1983. Protocol for bioassessment of hazardous waste sites. EPA- 600/2-
83-054. Lafayette, CA. 42 pp. + appendices.
U.S. Environmental Protection Agency. 1985. Short-term methods for estimating
the chronic toxicity of effluents and receiving waters to freshwater organisms.
EPA/600/4-85/014. U.S. EPA, Environmental Monitoring and Support Laboratory,
Cincinnati, OH. 162 pp.
15-184
-------
CASE STUDY 29: SAMPLING OF SEDIMENTS ASSOCIATED WITH SURFACE
RUNOFF
Point Illustrated
• Contaminated sediments associated with surface runoff pathways
(rivulets or channels) are indicative of the migration of chemicals via
overland flow.
I n t r o d u c t i o n
This facility is a secondary lead smelting plant which began operation in 1976.
The plant reclaims lead from materials such as waste automotive batteries,
byproducts of lead weight manufacture, and wastewater sludges. Lead grid plates
from alvaged batteries are temporarily stored on site in an open pile prior to being
re-melted. It is therefore appropriate to conduct some form of-runoff sampling to
monitor migration of contaminants from the site via this route.
Facility Description
The facility covers approximately 2,000 ft2and is situated in an area primarily
used for farming. A creek flows adjacent to the plant and drains into a major river 6
miles west of the site. Population is sparse with the nearest town 4 miles to the
south. In the past, there have been four on-site impoundments in operation and
two landfills. In addition, blast furnace slag, lead grid plates, and rubber chips from
the recycled batteries have been stored in two on-site waste piles.
Sediment Sampling
Four sediment samples (020, 022, 025, and 027) were collected from surface
runoff pathways and a creek which receives runoff from the site. Figure 15-55
shows the locations of the runoff pathways relative to the facility and the four
sampling points. Additional sediment samples were collected from the creek at
various points upstream and downstream of known overland leachate seeps and
surface water runoff routes. The program design enabled comparison between
15-185
-------
_ CLOSED
IMPOUNDMENTS
CLOSED SLAG
STORAGE AREA
OW4
CLOSED
IMPOUNDMENTS 1 & 2
CLOSED RUBBER CHIP
STORAGE AREA
CLOSED SLAG
STORAGE AREA
IMPOUNDMENT4
DRILLHOLES
WASTE AREA
WELLS
\
OW13
I
0
FEET
200
Figure 15-55. Surface Water and Sediment Sample Locations
15-186
-------
concentrations at different sections of the creek and background locations in
relation to the runoff pathways.
Results
Table 15-24 presents,the concentrations of lead and arsenic measured on the
four surface runoff pathways and at location 029, which represents an upstream
background concentration (Figure 15-55). It is clear that highly elevated levels of
lead, were detected in all four of the runoff pathway samples. The highest
concentration of lead, 1,900 ppm, was detected in the western-most portion of the
site. Runoff pathway sediment at the northern end of the facility, adjacent to the
slag storage area, recorded 1,600 ppm of lead. Concentrations of this order
represent a substantial source of sediment contamination.
TABLE 15-24
ARSENIC AND LEAD CONCENTRATIONS (PPM) IN RUNOFF
SEDIMENT SAMPLES
Sampling Location
Contaminant
Arsenic
Lead
Case Discuss i o n
This case illustrates the importance of monitoring surface runoff pathways,
because they can represent a major route of contaminant migration from a site,
particularly for contaminants likely to be sorbed on or exist as fine particles. This
type of monitoring is especially useful for units capable of generating overland
flows. Such monitoring can establish the need for corrective measures (e.g., surface
runon/runoff controls and/or some form of waste leachate collection system).
# 020
11.0
1300
# 022
9.6
1 900
# 025
2.0
1600
# 027
8.9
1700
Backgroi
# 029
<0.1
11.0
15-187
-------
CASE STUDY 30: SAMPLING PROGRAM DESIGN FOR CHARACTERIZATION OF A
WASTEWATER HOLDING IMPOUNDMENT
Points Illustrated
• Sampling programs should consider three-dimensional variation in
contaminant distribution in an impoundment.
• Sampling programs should encompass active areas near inflows and
outflows, arid potentially stagnant areas in the corner of an
impoundment.
introduction
This study was conducted to assess whether an act
impoundment could be assumed to be of homogeneous composition for the purpose
of determining air emissions. This case shows the design of an appropriate
sampling grid to establish the three-d imensional-composition of the impoundment.
Facility Description
The unit being investigated in this study is a wastewater impoundment at a
chemical manufacturing plant. The plant primarily produces nitrated aromatics and
aromatic amines. Raw materials include benzene, toluene, nitric acid, and sulphuric
acid. Wastewater from the chemical processing is discharged into the
impoundment prior to being treated for release into a nearby water body. The
impoundment has an approximate surface area of 3,750 nfand a depth of 3 m.
Sampling Program
For the most part sampling involved the collection of grab samples using an
extended reach man-lift-vehicle. The program was designed to collect samples at
different locations and depths in the impoundment.
15-188
-------
Sampling Locations and Procedures
Sampling Grid - The wastewater impoundment was divided into 15 segments
of equal area. Within this grid, eight sampling locations were selected which
included all pertinent areas of the impoundment, such as active portions near the
inflows and outflows, potential stagnant areas in the corners, and offshore points
near the center line of the impoundment.
It was decided to take samples from four depths in the liquid layer and one
from the bottom sediments at each of the eight locations. Figure 15-56 shows the
impoundment schematic and sampling locations.
Liquid Sampling - A total of 32 liquid grab samples were taken. These were
analyzed for the following parameters: all identifiable volatile organic compounds
(VOCs) and semivolatile organic compounds (SVOCs) using gas chromatograph/mass
spectroscopy; and selected VOCs and SVOCs by gas chromatography using a flame
ionization detector.
Sediment/Sludge Sampling -The bottom layer was sampled using a Ponar grab
sampler. The same analyses were performed on the eight sediment/sludge samples
as on the liquid samples.
Meteorological Monitoring - The ambient meteorological conditions were
monitored throughout the sampling period, including wind speed, wind direction,
and air temperature. A video camera was also used to record the movement of
surface scum on the impoundment.
Table 15-25 summarizes the sampling locations and analyses, including
locations where QC data were collected.
Results
From the sampling program, it was discovered that approximately 99 percent
of the organic compounds (by weight) were contained in the bottom sludge layer.
15-189
-------
3<
..
.
I
f
.
„;
'.'
i
\
/
. D0
- -
1 ...
G H
1
•F
©
' ' •_
1-1
©D E©
1 u
"•: — ^7
C
©
A B
© ®
• - ' 1- .
Ill
ftDrtCCn CAUDI titf* i n/*>ATiA&
/
\
e
a
O f
c
•Q-
0.
(9
• K
w
tu
l"
' -
•' '
' "•
,^^_— _
N
PLANT SUMP EFFLUENT
INFLUENT/LA&OON
EFFLUENT
BOILER BACK WASH EFFLUENT
PRIMARY PLANT EFFLUENT
Figure 15-56. . Schematic of Wastewater Holding Impoundment Showing
Sampling Locations
15-190
-------
TAB LEI 5-25
SUMMARY OF SAMPLING AND ANALYSIS PROGRAM FOR
WASTEWATER IMPOUNDMENT
Location
A-1
A-2
A-3
A-4
A=5
B-1
B-2
B-3
B-4
8-5
C-1
C-2
C-3
C-4
D-1
D-2
D-3
D-4
D-5
Depth
(Feet)
0-1
2
4
6
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
0-1
2
4
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
Sample Analyses
GG'FiD
VGA
X
X
X
X
X
X
X
X
X
X
X
x
X
x
X
X
X
X
X
GO'PvIS
VOA,
.,;._,. X.
X
X
X
x
X
X
X
TOC
X
X
X
X
X
X
X
X
POC
X
1
. X
X
X
Gnsite
Parameters3
X
X
X
X .
X
X
X
X
X
X
X .
X
X
X
X
X
X
X
X
GO'FiD
svoc
X
X
X
X
X
X
, X
x
X
X
X
X
X
, X
X
X
X
X
X
GO'MS
svoc :
X
1 i
X
X
-
X
A
.,,. , -'
..x
X
X
a Includes pH, turbidity, specific conductance, and dissolved oxygen measurements.
X Indicates locations where QC samples were collected.
15-191
-------
TABLE 15-25 (continued)
Location
E-1
.£-2"
E-3
E-4
E-5
F-1
F-2
F-3
F-4
F-5
G-1
G-2
G-3
G-4
G-5
H-1
H-2
H-3
H-4
H-5
Depth :
(Feet)
0-1
2
4
6
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
0-1
2
4
6
Bottom
Sediment
Sample Analyses
GC/FID
VGA
X
X
X
X
X
"!
X
X
K ,
X
X
X
X
X
X
X
X
x
X
X
X
GC/MS
VOA
X
X
X
. ,., .
X
X
X
X
X
TOC
X
X
X
X
X
X
X
X
POC
X
. •
' •- :
, , ,.„ .
X
,1 1 1-1
X
X
--,-
Onsite
•'Parametersa
X
X
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
GC/FID
svoc
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
GGMS
SVOC
X
X
X
X
X
•
x
X
X
a Includes pH, turbidity, specific conductance, and dissolved oxygen measurements.
X Indicates locations where QC samples were collected.
15-192
-------
Vertical and horizontal variation in the composition of the lagoon was apparent.
The degree of horizontal variation was relatively small, but sample point "A"
showed considerably higher concentrations of 2,4-dinitrophenol than the other
locations. This could have resulted from a recent discharge from the outflow at the
southern end of the impoundment. Vertical variation in composition showed a
general trend of increasing concentration with depth, but certain chemicals tended
to have higher concentrations at mid-depth in the impoundment.
Case Discussion
This case provides an example of a sampling program at an areal source
designed to yield accurate information for characterizing air emissions from the
unit. The study illustrated the importance of characterizing the organic
composition of the lagoon in three dimensions and considering variations resulting
from inflow and outflow areas.
It should be mentioned that this study did not consider variation in the
chemical composition of the impoundment with time. T obtain this information, it
would be necessary to conduct subsequent sampling programs at different times.
From this study, it is apparent that chemical composition varies both horizontally
and vertically, and is likely to change depending on inflows and outflows of wastes.
This sampling program is therefore limited to effectively characterizing competition
at a single point in time.
15-193
-------
CASE STUDY 31: USE OF DISPERSION ZONE CONCEPTS IN THE DESIGN OF A
SURFACE WATER MONITORING PROGRAM
Point Illustrated
• Estimation of the dispersion zone of contaminants downstream of a
release point can be used to help design a surface water monitoring
program.
Introduction
When a contaminant is initially released to a body of water, the concentration
of the contaminant will vary spatially until fully dispersed. In streams, the
contaminant will disperse with the surrounding ambient water as the water moves
downstream and will eventually become fully dispersed within the stream.
Downstream of this point, the contaminant concentration will remain constant
throughout the stream cross-section, assuming that streamflow is constant and that
the contaminant is conservative (e.g., nondegradable). The area in which a
contaminant's concentration will vary until fully dispersed, referred to here as the
dispersion zone, should reconsidered when determining the number and location
of sampling stations downstream of the release point.
Facility Description
A facility that processed zinc, copper and precious metals from ores operated
along a stream for five years. The plant was closed after being cited for repeated
fish kills, reportedly due to failures of a tailings pond dike. At present, the site is
covered with tailings containing high concentrations of copper, zinc, cadmium,
arsenic, and lead. There is no longer a tailings pond. This is the same facility
described in Case Study 27.
Site Setting
The site is located on coarse colluvium (hill-slope deposits of weathered
bedrock) and fine-grained alluvium. These deposits are typically 50 feet thick.
Metamorphic rock (phyllite) underlies the unconsolidated materials. Ground water
15-194
-------
moves laterally in the gravel formations from the steep valley walls toward the
stream.
The site is located about 400 feet from the stream. Two drainage ditches cross
the lower portion of the site and merge prior to leaving the site. The ditch carries
the combined flow and discharges directly into the stream (Figure 15-57). No other
tributaries enter the stream within 2 miles of this location. Downstream of the
release point, streak width and depth remain fairly constant at 45 and 3 feet,
respectively. Mean stream velocity is 0.5 feet per second and channel slope is 0.0005
feet per foot.
Sampling Program
A surface water monitoring program was designed as part of a Phase I
investigation to determine the extent of contamination in the stream. Existing data
from previous sampling had shown high concentrations of metals in the drainage
ditch sediments (e.g., 5,170 mg/kg Cu and 11,500 mg/kg Zn). Ground-water data
from the plant's well showed {concentrations of Cu (7 ug/l) and Zn (54 ug/l). The
contribution of metals to the stream by ground-water discharge was considered to
be relatively minor.
Based on a review of the plant history and the available water quality and
sediment data, a monitoring program was designed. The potential pathways by
which metals could reach the stream appeared to be direct discharge from the
drainage ditch, discharge of contaminated ground water, and storm water runoff
over the general facility area. Plant records indicated that typical flows in the
drainage ditch at its confluence with the stream varied from 1 to 3 cubic feet per
second (cfs) in the spring. During extreme flood conditions, the flow in the ditch
exceeded 20 cfs. In the summer, flows in the drainage ditches at all locations were
less than 0.5 cfs. Resuspension of contaminated sediments in the ditches during
storm runoff appeared to be the most likely pathway for metals to reach the
stream. The specific metals of concern were identified as As, Cd, Cu, Pb and Zn,
based on the processes used at the plant and the composition of the ores which
contained some arsenopyrites (with As, Cu), galena (Pb), and sphalerite (with Zn,
Cd).
15-195
-------
i 36
River
——.,
• 39 (High flow only)
• 37
*
To S8-
Area of
Former
Tailings
Pond
N
f
Site Operations
Drainage Ditch
Sampling Station
Scale
I
160 feet
Figure 15-57. Sampling Station Locations for Surface Water Monitoring
* Located approximately 1030 feet downstream of the confluence of the ditch
with the stream.
15-196
-------
The available soil and water quality data from previous sampling were
reviewed to help determine the likely fate of the metals. The pH of soils in the area
is about 6.5 and they contain about 0.5 percent organic matter by weight. Under
such conditions,the metals, particularly Pb, would be expected to adsorb onto the
soil particles. In the on-site tailings piles, the pH of core samples ranged between
3.3 and 4.9. Low soil pH values had been measured in sediments in the drainage
ditch just downgradient of the tailings pile. The pH of the stream during the
previous sampling was 6.9. The suspended solids concentration was 10 mg/l.
Estimates of the distribution of metals between the dissolved and adsorbed
phases for a range of partition coefficients (Kp) are shown in Table 15-26. For
example, if Kp= 10"and the suspended solids concentration was 10 mg/l, 90
percent (0.9) of the metal present would be in the dissolved phase. This information
indicated that even though a metal (e.g., lead) was known to strongly sorb, a
significant amount could still be transported in the dissolved phase. Thus, both
water and suspended solids should be analyzed for metals. The complete list of
parameters selected for measurement in the Phase I investigation and the rationale
for their selection are outlined in Table 15-27.
The sampling stations were, selected to determine stream water quality up-
and downstream of the site and to determine whether particulate with sorbed
metals were deposited on the stream banks or streambed. The sampling stations
and the rationale for their selection are listed in Table 15-28. The station locations
are shown in Figure 15-57. Because floods were considered a cause of
contamination incidents, samples were to be collected under both high and low
flow conditions.
The location of the downstream station (S8) was determined after estimating
the stream length that may be required for complete dispersion of the
contaminants. The following equation was used for this estimation:
0.4 w2u
DZ =
15-197
-------
TABLE 15-26
RELATIONSHIP OF DISSOLVED AND SORBED PHASE CONTAMINANT
CONCENTRATIONS TO PARTITION COEFFICIENT AND SEDIMENT
CONCENTRATION
K p s P C w / C ra
10o
•;•
101
102
103
- • ,
104
1
10
100
1000
10,000
:• .'. 1.
10
100
1000
10,000
1
10
100
1000 ,
10,000
1
10
100
1000
10,000
r • '
10
100
1000
10,000
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.9
1.0
1.0
1.0
. 0.9
0-5
1.0
1.0
0.9
0.5
0.1
1.0
0.9
0.5
0.1
0.0
After Mills et al 1985.
"The fraction dissolved (CJCT) is calculated as follows:
Qri, 1
CT 1+KpX 5x10-6
where Kp = partition coefficient, 4/kg
SS = suspended solids concentration, mg/f
Cw = Dissolved concentration
CT = Total concentration
15-198
-------
TABLE 15-27
PARAMETERS SELECTED FOR SURFACE WATER MONITORING PROGRAM
Parameters
Rationale
Metals - As,Cd, Cu, Pb, Zn
PH
Dissolved Oxygen Sulfide, Fe(ll),
Fe(lll)
Alkalinity
Total Dissolved Solids
Major Cations (Ca+2, Mg+2, Na+, K+,
NH4+) and
Major Anions (CI-, S042", N0.3)
Suspended Solids:
Streamflow
Determine extent of contamination
Predict sorption behavior, metal
volubility, and speciation
Determine redox conditions which
influence behavior of metals,
particularly the leaching of tailings
A measure of how well buffered a water
is, allows consideration of the likelihood
of pH change
Used as a water quality indicator and for
QA/Q C checks
May identify other waste sources, can
influence fate of trace metals
Predict the fraction of metal in water
which is sorbed
Compute mass balances and assist in
identifying sources of observed
contamination
15-199
-------
TABLE 15-28
SELECTED SURFACE WATER MONITORING STATIONS AND SELECTION RATIONALE
Station
Media
Rationale
Drainage ditch west of
site(S1)
Drainage ditches on site
(S2 and S3)
Downstream of
confluence of two
ditches (S4)
Mouth of drainage
ditch (S5)
Stream (S6, S7 and S9)
Stream (S8)
Water and sediments
Determine whether off-site
drainage is significant source of
contamination
Water and sediments Identify on-site sources
Water and sediments
Water, suspended
sed i ment, bed I oad
Water, suspended
sediment, bedload
Water, suspended
sediment, bedload
Provide information for
checking mass balances from the
two drainage ditches
Determine quality of djrect
discharge to stream
Determine upstream water
quality
Determine quality downstream
of site fallowing complete
dispersion and provide data for
mass balance
15-200
-------
where:
DZ = dispersion zone length, ft
w = width of the water body, ft (45 ft)
u = stream velocity, ft/sec (0.5 ft/sec)
d = stream depth, ft (3 ft)
s = slope (gradient) of stream channel, ft/ft (0.0005)
g = acceleration due to gravity (32 ft/sec2).
Using the above equation, the estimated stream length required for complete
contaminant dispersion is 1030 feet. This can serve as an approximate distance
downstream of the release point at which a sampling station should be located.
Case Discussion
This case illustrates the use of contaminant dispersion zones in the design of a
surface water monitoring program. In this example, the data indicate that
approximately 1030 feet of flow within the described stream channel is required
before a contaminant will become fully dispersed. A downstream station should
therefore be located at or below this dispersion zone to fully characterize the
extent of the release. An adequate number of sampling stations should also be
located upstream of this point.
15-201
------- |