EPA/600/6-89/001
May 1989
DEVELOPMENT OF RISK ASSESSMENT METHODOLOGY
FOR LAND APPLICATION AND DISTRIBUTION AND
MARKETING OF MUNICIPAL SLUDGE
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
r
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental
Protection Aaencv policy and approved for publication. Mention of trade
names or co^ercial products dSes not constitute endorsement or recommenda-
tion for use.
ii
-------
PREFACE
This 1s one of a series of reports that present methodologies for
assessing the potential risks to humans or other organisms from the disposal
or reuse of municipal sewage sludge. The sludge management practices
addressed by this series Include land application practices, distribution
and marketing programs, Iandf1ll1ng, Incineration and ocean disposal In
particular, these reports provide methods for evaluating potential health
and environmental risks from toxic chemicals that may be present in sludge.
This document addresses risks from chemicals associated with land
application and distribution and marketing of municipal sludge.
These proposed risk assessment procedures are designed as tools to
assist In the development of regulations for sludge management practices.
The procedures are structured to allow calculation of technical criteria for
sludge disposal/reuse options based on the potential for adverse health or
environmental Impacts. The criteria may address management practices (such
as site design or process control specifications), limits on sludge disposal
rates or limits on toxic chemical concentrations In the sludge.
u J{e "»tho
-------
DOCUMENT DEVELOPMENT
Authors and Contributors
Reviewers: Food Chain Impacts
R.J.F. Bruins, Document Manager
A. Jarabek, Co-Document Manager
A. Molak
L. Fradkin
W.B. Peirano
Environmental Criteria and Assess-
ment Office
Office of Health and Environmental
Assessment
U.S. Environmental Protection Agency
Cincinnati, OH 45268
J. Ryan, Co-Document Manager
Water Engineering Research Laboratory
Office of Environmental Engineering
Technology and Demonstration
U.S. Environmental Protection Agency
Cincinnati, OH 45268
J.D. Dean and P.A. Mangarella
Woodward-Clyde Consultants
Walnut Creek, CA 94596
G. Dawson
ICF Northwest
Richland, WA 99352
E.E. Niebla and B.A. Corcoran
Wastewater Solids Criteria Branch
Office of Water Regulations and
Standards
U.S. Environmental Protection Agency
Washington, DC 20460
Reviewers: Food Chain Impacts
Dr. Dale E. Baker
Department of Agronomy
Penn State University
University Park, PA 16802
Dr. Rufus Chaney
U.S. Department of Agriculture - ARS
Beltsville, MD 20705
Dr. Thomas Hinesly
Department of Agronomy
University of Illinois
Urbana, IL 61801
Dr. Lee W. Jacobs
Department of Crop and Soil Science
Michigan State University
East Lansing, MI 48824
Dr. Terry J. Logan
Department of Agronomy
Ohio State University
Columbus, OH 43210
Dr. Al Page
Department of Soil and Environmental
Science
University of California
Riverside, CA 92521
Dr. Bill Sopper
Institute for Research on Land and
Water
Penn State University
University Park, CA 16802
Reviewers: Surface Runoff Modeling
Dr. Carl Anderson
Davidson Hall
Department of Agricultural
Engineering
Iowa State University
Ames, IA 50011
Dr. Douglas A. Haith
Department of Agricultural
Engineering
Cornell University
Ithaca, NY 14853
1v
-------
DOCUMENT DEVELOPMENT (cont.)
Reviewers: Health Effects
Dr. Dale Johnson
Department of Environmental Health
University of Cincinnati Medical Center
Cincinnati, OH 45267
Dr. Martha Radike
Department of Environmental Health
University of Cincinnati Medical Center
Cincinnati, OH 45267
Document Preparation
Bohanon-
-------
r
TABLE OF CONTENTS
Page
1.
INTRODUCTION AND DESCRIPTION OF GENERAL METHODOLOGIC APPROACH 1-1
1.1 PURPOSE AND SCOPE } \
1 2 DEFINITION AND COMPONENTS OF RISK ASSESSMENT 1-2
1.3 RISK ASSESSMENT IN THE METHODOLOGY DEVELOPMENT PROCESS ... 1-3
131 EXPOSURE ASSESSMENT ^~3
l!s'.2 HAZARD IDENTIFICATION AND DOSE-RESPONSE ASSESSMENT. 1-7
1.3.3 RISK CHARACTERIZATION 1~8
1 4. POTENTIAL USES OF THE METHODOLOGY IN RISK MANAGEMENT . . . 1-10
1.5. LIMITATIONS OF THE METHODOLOGY I-'1
2. DEFINITION OF MANAGEMENT PRACTICES 2-1
2.1. INTRODUCTION 2~\
2.2. LAND APPLICATION PRACTICES 2~3
2.2.1. Agricultural Utilization 2-3
2.2.2. Forest Land Utilization z~3
2.2.3. Drastically Disturbed Land Utilization 2-6
2.2.4. Dedicated Land Disposal Site 2-7
2.2.5. Summary z 8
2.3. DISTRIBUTION AND MARKETING PRACTICES 2-9
2.3.1. Assumptions 2-11
2.3.2. Requirements or Potential Requirements
-------
TABLE OF CONTENTS (cont.)
4.3.
4.4.
4.5.
4.7,
4.8.
Page
4.1.2. Contaminant Uptake Relationships 4-10
4.1.3. Toxicity Thresholds for Nonhuman Organisms] . ' ' 4-21
Human Diet 4_23
Health Effects in Humans. '.'.'.'.'. 4-29
4.1.4.
4.1.5.
4.2. SLUDGE-SOIL-PLANT-HUMAN TOXICITY EXPOSURE PATHWAY 4-41
Assumptions . 4_41
Calculation Method '.'.'.'.'.'.'.'.'.'. 4-41
Input Parameter Requirements ..'.'.'.' 4-63
4.2.1.
4.2.2.
4.2.3.
SLUDGE — HUMAN TOXICITY (SOIL INGESTION) EXPOSURE
r MInWM!»••••,.»
4-72
4.3.1. Assumptions 4_72
4.3.2. Calculation Method ..'."!.' .' .' ,' .' .' 4-72
4.3.3. Input Parameter Requirements. .......... 4-75
EXPOSURE PATHWAYS FOR HERBIVOROUS ANIMALS FOR HUMAN
CONSUMPTION
4-78
4.4.1.
4.4.2.
4.4.3.
Assumptions 4_78
Calculation Method .......... 4-78
Input Parameter Requirements. .......... 4-33
EXPOSURE PATHWAYS FOR TOXICITY TO HERBIVOROUS ANIMALS 4-86
4.5.1. Assumptions ' 4_87
4.5.2. Calculation Method 4-87
4.5.3. Input Parameter Requirements. .......... 4-97
4.6. SLUDGE-SOIL-PLANT TOXICITY EXPOSURE PATHWAY 4-;
87
4.6.1
4.6.2
4.6.3.
Assumptions 4_87
Calculation Method .'....'.. . 4-88
Input Parameter Requirements 4-88
PATHWAYS FOR TOXICITY TO SOIL BIOTA AND THEIR
4-89
4.7.1,
4.7.2.
Sludge-Soil-Soil Biota Toxicity Exposure
Pathway 4_89
Sludge-Soil-Soil Biota-Predator Toxicity
Exposure Pathway 4-90
EXAMPLE CALCULATIONS 4_93
4.8.1,
4.8.2.
4.8.3.
Sludge-Soil-Plant-Human Toxicity Exposure
Pathway 4_94
Sludge-Human Toxicity (Soil Ingestion) Exposure
Pathway 4_105
Exposure Pathways for Herbivorous Animals
for Human Consumption 4-107
vii
-------
TABLE OF CONTENTS (cont.)
Page
4.8.4. Exposure Pathways for Toxicity to
Herbivorous Animals 4 MI
4.8.5. Sludge-Soil-Plant Toxicity Exposure Pathway . . . 4-111
4.8.6. Exposure Pathways for Toxicity to Soil
Biota and Their Predators 4-1 M
5. EXPOSURE AND ASSESSMENT OF HEALTH EFFECTS FROM INHALED
PARTICULATES
5-1
5.1. INHALATION OF PARTICULATES ................ 5-1
5.2. DUST EMISSION FACTOR
5.3. DUST EMISSION RATE AND AIR CONCENTRATION DETERMINATION .
5.4.
'
5-2
EXAMPLE CALCULATION 5~4
ATTfiM MFTHnns FOR SURFACE RUNOFF EXPOSURE
6-1
6. CRITERIA CALCULATION METHODS FOR SURFACE RUNOFF EXPOSURE
PATHWAY
6.1.
6.2.
6.3,
OVERVIEW OF THE METHOD
ASSUMPTIONS
6.2.1. Loading Algorithm Assumptions . . . . 6-7
coo Acciimn-Hnnc in Rereivina Water Analysis b-lb
.... 6-16
6.2.2. Assumptions in Receiving Water Analysi
CALCULATIONS
6.3.1.
632
Tier 1
Tier 2/3
Setting National Criteria, Surface Runoff
Algorithms
6.4. INPUT PARAMETER REQUIREMENTS
6.4.1. Loading Algorithms
6.4.2. Data Inputs for Receiving Water Analysis,
6-31
6-34
6-34
6-49
6.5. HEALTH AND ENVIRONMENTAL EFFECTS 6-50
6.5.1. Aquatic Life Protection . |>-50
6.5.2. Wildlife Protection &-5"
6.5.3. Human Health Effects 6-51
6.6. EXAMPLE CALCULATIONS 6~70
6.6.1. Site-Specific Application 6~70
CRITERIA CALCULATION METHODS FOR THE GROUNDWATER EXPOSURE
PATHWAY
7-1
7.1. OVERVIEW OF THE METHOD
7.2. ASSUMPTIONS
7.3. CALCULATIONS '~b
7-1
7-6
-------
TABLE OF CONTENTS (cont.)
Page
7.3.1. Source Term 7_6
7.3.2. Unsaturated Zone Transport '.'.'.'.'. 7-17
7.3.3. Saturated Zone Transport ' 7-17
7.3.4. Setting National Criteria •'.'.'.'. 7-17
7.4. INPUT PARAMETER REQUIREMENTS 7-2Q
7.4.1. Fate and Transport: Pathway Data 7-20
7.5. EXAMPLE CALCULATIONS 7_20
7.5.1. Site-Specific Application 7-20
7.5.2. National Criteria Site-Specific Application . .' .' 7-30
8. CRITERIA CALCULATION METHODS FOR THE VAPORIZATION EXPOSURE
PATHWAY . . . . g_1
8.1. OVERVIEW OF THE METHOD 8_i
8.2. ASSUMPTIONS .".*."!.'!!!.'.* .8-4
8.2.1. Vapor Pressure 8_4
8.2.2. Loss Rate .'.'*''* 8-4
8.2.3. Atmospheric Transport '.'.'.'.'.'.'.'.'. 8-6
8.3. CALCULATIONS 8_7
8.3.1. Tier 1 a_7
8.3.2. Tier 2 '.'.'.'.'. 8-9
8.3.3. Tier 3 ........ 8-13
8.3.4. Setting National Criteria ............ 8-13
8.4. INPUT PARAMETER REQUIREMENTS 8-17
8.4.1. Fate and Transport: Pathway Data 8-17
8.4.2. Fate and Transport: Chemical-Specific Data. '. '. ' 8-17
8.4.3. Health Effects Data 8_18
8.5. EXAMPLE CALCULATIONS . . . 8_30
8.5.1. Site-Specific Application 8_30
8.5.2. National Criteria \ \ \ 8_35
9. REFERENCES 9-1
APPENDICES
APPENDIX
APPENDIX
APPENDIX
APPENDIX
APPENDIX
Reanalysis of the FDA Revised Total Diet Food List
Parameter Guidance for USLE Parameters
Rainfall Depths for the 5-Year Recurrence Interval for
Storm Durations of 30 Minutes to 24 Hours
Procedure for Incorporating Site-Specific Considerations
in Receiving Water Analyses
5: Distribution Coefficients
4
ix
-------
LIST OF TABLES
No. Title Page
2-1 Distribution of Sludges Between Disposal/Reuse Alternatives . 2-2
3-1 Exposure Pathways Applicable to Current (C) or Future (F)
Land Use 3~3
4-1 Assumptions for Terrestrial Food Chain 4-2
4-2 Average Daily Dry-Weight Consumption of Food Groups, Based
on a Reanalysis of the FDA Revised Total Diet Food List . . . 4-26
4-3 Distribution of Daily Consumption of Six Food Groups
from a Survey Using 24-Hour Recall 4-28
4-4 Food Consumption of Lacto-Ovo-Vegetarians and Average
25- to 30-Year-Old Males 4-30
4-5 Illustrative Categorization of Carcinogenic Evidence
Based on Animal and Human Data 4-38
4-6 Assumptions for Sludge-Soil-Plant-Human Toxicity
Exposure Pathway 4-42
4-7 Summary of Criteria Derivation Procedure Based on
Curvilinear Uptake Response Model and Relative Uptake
Response Values 4-44
4-8 Relationship Between the Experimental Basis for Reference
Sludge Concentration (RSC) and Rules Governing Use of
Sludges Meeting RSC 4-51
4-9 Summary of Criteria Derivation Procedure Based on Linear
Uptake Response Model and Relative Uptake Response Values . . 4-54
4-10 Experimental Basis for the Reference Application Rate of
Pollutant (RP) and Situations Where RP Applies 4-58
4-11 Summary of Criteria Derivation Procedure Based on Linear
Uptake Response Model Without Using Relative Uptake
Response Values 4-60
4-12 Choice of Input Parameter Values for Sludge-Soil-Plant-
Human Toxicity Pathway, As Affected by Exposure Scenario. . . 4-65
4-13 Human Population, Sludge Production, Cropland and Cropland
Required Annually for the Application of Sewage Sludge in
Illinois, New Jersey and the United States in 1970, and
Projected Values for 1985 4-66
-------
LIST OF TABLES (cont.)
No.
4-14
4-15
4-16
4-17
4-18
4-19
5-1
6-1
6-2
6-3
6-4
6-5
6-6
6-7
6-8
6-9
7-1
7-2
7-3
Title
Annual Consumption Homegrown Foods. . .
Assumptions for Sludge-Human Toxicity (Soil Ingestion)
Exposure Pathway
Various Estimates of Daily Soil Ingestion in Children
of Ages 1-3
Assumptions for Pathways Dealing with Herbivorous Animals . .
Exposure Pathways for Herbivorous Animals for Human
Consumption: Effect of Management Practice on Various
Parameters
Assumptions for Sludge-Soil-Soil Biota-Predator Toxicity
Exposure Pathway
Particulate Exposure to Cadmium from Land Application of
Dewatered Sewage Sludge . .
Surface Runoff Methodology Assumptions
Input Parameters for the Runoff Pathway Methodology
Runoff Curve Numbers for Hydrologic Soil-Cover Complexes
(for Antecedent Rainfall Condition II)
Antecedent Rainfall Conditions and Curve Numbers
Daily Intakes of Drinking Water by Adults
Water Ingestion and Body Weight by Age-Sex Group in the
United States
U.S. Annual Per Capita Consumption of Commercial Fish
and Shellfish, 1960-1984
Fish Consumption by Demographic Variables . . .
Input Parameters for the Example Calculations ....
Assumptions for the Groundwater Pathway Methodology
Water Content of Sludges from Various Treatment Processes . .
Input Parameters for Example Calculations — Groundwater. . .
Page
4-69
4-73
4-76
4-79
4-80
4-91
5-5
6-8
6-35
6-40
6-43
6-54
6-55
6-64
6-67
6-72
7-7
7-16
7-21
xi
-------
LIST OF TABLES (cont.)
No.
8-1
8-2
8-3
8-4
8-5
Title
Page
Assumptions for the Vapor Pathway Methodology 8-5
Parameters Used To Calculate az 8~14
Dally Respiratory Volumes for "Reference" Individuals
(Normal Individuals at Typical Activity Levels) and for
Adults with Higher-than-Normal Respiratory Volume or
Higher-than-Normal Activity Levels 8-21
Illustrative Categorization of Carcinogenic Evidence
Based on Animal and Human Data 8~28
Input Parameters for Example Calculations: Vapor Loss
8-31
xll
-------
LIST OF FIGURES
No.
1-1
4-1
4-2
4-3
6-1
6-2a
6-2b
6-3
6-4
6-5
6-6
6-7
7-1
8-1
Title
Page
Relationship of Risk Assessment Methodology to Other
Components of Regulation Development for Sewage Sludge
Reuse/Disposal Options 1-4
Curvilinear Uptake of Sludge-borne Metal by Crops ...... 4-13
First-Year vs. Multi-Year Observations 4-15
Limitation by Phytotoxicity of Linear Uptake Response .... 4-19
Schematic of Surface Runoff and Erosion from a Sludge
Land Application Area As Addressed by the Methodology .... 6-2
Flow Chart for Estimating Long-Term Average Concentrations
As Addressed by the Methodology 6-3
Flow Chart for Estimating Event Mean Concentrations As
Addressed by the Methodology 6_4
Mineral Bulk Density of Soils of Varying Textures 6-38
Erosion Potential for Storms of Various Durations for
Soils With Selected Infiltration Properties 6-44
Field Capacity Water Content for Soils of Various Textures. . 6-47
Wilting Point Water Content for Soils of Various Textures . . 6-48
Soil Classification Chart Developed by Bureau of Public
Roads 6-49
Logic Flow for Groundwater Pathway Evaluation of Sludges
Applied to Land 7_3
Logic Flow for Vapor Loss Pathway Evaluation of
Land-Applied Sludges 8-3
-------
LIST OF ABBREVIATIONS
«2
Y*
At
At
e
0
ee
0fc
©wp
w
%OM
A
Aa
ac
ADI
AMC
AR
ARa
ARC
As
AWQC
b
B
Standard deviation of the vertical concentration distance (m)
Density of sludge liquid (kg/9.)
Elapsed time since the beginning of operation (cannot equal
zero)
Time period over which application is proposed (years)
Available water capacity of soil (dimensionless)
Available volumetric water capacity of the top cm of soil
Effective porosity
Field capacity of soil (dimensionless)
Wilting point of soil (dimensionless)
Average windspeed (m/sec)
Percent organic matter
Acres tilled/8-hour day
Sorbed contaminant mass in top centimeter (mg/ha-cm)
Acres
Acceptable daily intake (mg/kg bw/day)
Antecedent soil moisture condition
Sludge application rate (t DW/ha)
Annual sludge application rate (t DW/ha)
Cumulative sludge application rate (t DW/ha)
Sludge application rate (kg/ha-year)
Ambient water quality criteria
Slope of matric potential and moisture content plot
(dimensionless)
Bulk density of soil (g/cm*)
xlv
-------
LIST OF ABBREVIATIONS (cont.)
BB
BC .
BCF
BCFa
BCFU
BI
BI
Bm
BS
BS
BU
bw
C
c
CAG
cc
Cd
C1
ci test
C.E.C.
CN
Cus
Background concentration In soil biota (pg/g DW)
Background concentration in crop tissue (yg/g DW)
Bioconcentration factor (8,/kg)
Adjusted BCF (Si/kg)
Unadjusted BCF (a/kg)
Background intake of pollutant (rag/day)
Background intake of pollutant from a given exposure route
(mg/day)
Mineral bulk density (g/cm3)
Bulk density saturated zone
Background soil concentration of pollutant (yg/g DW)
Bulk density unsaturated zone
body weight (kg)
Concentration of contaminant in sludge/soil mixture (mg/kg)
"Cover management" factor (dimensionless)
Carcinogen Assessment Group
Compliance point concentration (mg/ma)
Cadmium
Concentration of contaminant in water (mg/S.)
Receiving water criteria determined from test case (mg/S.)
Cation exchange capacity
Contaminant concentration in the liquid (mg/S.)
Concentration of i in the solution (mole/ma)
SCS runoff curve number (dimensionless)
Contaminant concentration exiting the unsaturated zone (mg/9.)
xv
-------
LIST OF ABBREVIATIONS (cont.)
Cv-j
C(X)1
d
D
Da
DA
DC
D&M
DR
ds
Dv
DW
e
£
EDA
EH
FA
FC
Fi
FL
foe
N
FS
Concentration of i in air (mass/volume)
Atmospheric concentration (g/m3)
Incorporation depth (cm)
Dispersion coefficient
Dissolved contaminant mass in top centimeter (mg/ha-cm)
Daily dietary consumption of animal tissue food group
(g DW/day)
Daily dietary consumption of crop food group (g DW/day)
Distribution and marketing
Total storm runoff depth (cm)
Distance to property boundary (m)
Drainage volume (ma/ma-yr)
Dry weight (dried at 105°C until a constant weight is reached)
Base of natural logarithms, 2.71828 (unitless)
Emission factor (kg/ha)
Exposure duration adjustment (unitless)
Emission rate (g/sec)
Fraction of animal tissue food group (unitless)
Fraction of crop food group (unitless)
Contaminant loading rate to the SMA (mg/ha-yr)
Fraction of diet that is adhering soil (g soil DW/g diet DW)
Organic carbon content (of soil or sludge) (dimensionless)
Saturated soil moisture content (mVm2)
Fraction of animal diet that is sludge (g soil DW/g diet DW)
xv1
-------
LIST OF ABBREVIATIONS (cont.)
ha
HCB
Hi
hy
la
If
JP
IR
Is
!D
k2
Hectare
Hexachlorobenzene
Henry's Law constant for i (atm-mVmol)
Depth to groundwater (m)
Air inhalation rate (m'/day)
Human consumption of fish (kg/day)
Acceptable chronic pollutant intake rate (mg/day)
Irrigation (mVm2-yr)
Soil ingest ion rate (g DW/day)
Total water ingestion rate (8,/day)
Rate constant for contaminant: loss from soil (years-1)
Particle size multiplier (dimensionless)
"Erodibility factor" (metric tons /acre-year-unit 'R')
Lumped zero-order loss rate constant (mg/ha-year)
Lumped first-order loss rate constant (yr-i)
First-order loss rate coefficient for degradation (yr-i)
First-order loss rate coefficient for surface runoff losses
^ow
First-order loss rate coefficient for infiltration (yr-1)
FT - k0 (mg/ha-year)
Soil-water partition coefficient (cmVg)
Organic carbon distribution coefficient for the contaminant
(l/kg)
Octanol-water partition coefficient
xvli
-------
LIST OF ABBREVIATIONS (cont.)
ksat
L
LC(j
e
L0
LOV
LS
M
Ma
Mc
MEI
Mg
MGD
MI event
Mm
MO
MS
mt
Saturated soil hydraulic conductivity (m/yr)
Initial moisture content of sludge (kg/kg)
Length of buffer strip (m)
Soil concentration (yg/S. DW)
Lipid content of dietary seafood (kg/kg)
Lipid content of experimental organism (kg/kg)
Average annual erosional loss delivered to stream (kg/year)
Average annual erosional loss of contaminant from SMA (kg/year)
Lacto-ovo-vegetarian
"Topographic or slope/length" factor (dimensionless)
Total mass
Total available mass of contaminant
Long-term average contamination level in sludge management
area (mg/ha)
Most-exposed individual
Megagram = 1 metric ton (10s g)
Million gallons/day
Mass of contaminant lost to surface water (mg/year for
long-term case, mg for event case)
Total event loading (mg)
Maximum contaminant mass per area of soil in the SMA (mg/ha)
Original mass loading of contaminant (mg/ha)
2x103 t/ha = assumed mass of soil in upper 15 cm
metric tons
Total snowmelt depth (cm)
XV111
-------
LIST OF ABBREVIATIONS (cont.)
MUSLE
N
N
MI
NI criteria
Ni test
NREL
OHEA
OS
P
P
P
P.
Pa
PB-PK
PCB
PFRP
Pi
PI
POTW
Pqt
Modified USLE
Nitrogen
Dry weight concentration of contaminant in sludge
Contaminant concentration in sludge (mg/kg)
Sludge concentration criteria (mg/kg)
Sludge concentration in test case (mg/kg)
NIOSH-recommended exposure limit
Office of Health and Environmental Assessment
OSHA standard
Phosphorus
Plateau value (yg/g DW)
"Supporting practice" factor (dimensionless)
Precipitation (ma/ma-yr)
Total contaminant mass in top centimeter (mg/ha-cm)
Physiologically based pharmacokinetic
Polychlorinated biphenyl
Process to further reduce pathogens
Partial pressure of i above the solution (atm)
Plateau increment value (ng/g DW)
Publicly owned treatment works
Dissolved contaminant loss from SMA (mg/ha)
Dissolved contaminant loss from buffer strip (mg/ha)
Mass of dissolved contaminant from buffer strip (mg)
x1x
-------
LIST OF ABBREVIATIONS (cont.)
Pt
Pt
Pxt
Pxt'
Pxt"
Q
Qf
qp
Qv
r'
R
R
Ra
RAC
RC
RC
Rd
RE
RfD
RF
RFC
RIA
RL
Maximum contaminant level (mg/ha)
Total pressure in the system (atm)
Sorbed contaminant loss from SMA (mg/ha)
Sorbed contaminant loss from buffer strip (mg/ha)
Mass of sorbed contaminant loss from buffer strip (mg)
Volume of runoff (ma)
Volumetric flow rate (mVsec)
Human cancer potency [(mg/kg/day)"1]
Mass flux input to the unsaturated zone
Peak runoff rate (ma/sec)
Contaminant flux rate due to volatilization (g/ma-sec)
Distance from center of source to receptor (m)
"Erosivity" factor (year -i)
Recharge (m3/m2-yr)
Average annual precipitation (cm/yr)
Reference air concentration (mg/m3)
Recharge rate (m/yr)
Reference starting aquifer concentration (mg/S,)
Retardation factor (dimensionless)
Relative effectiveness of exposure (unitless)
Reference dose (mg/kg/day)
Acceptable mass flux of contaminant (mg/m^-year)
Reference feed concentration (pg/g DW)
Adjusted reference intake (yg/day)
Risk level (unitless)
xx
-------
LIST OF ABBREVIATIONS (cont.)
RLC
RMCL
RO
Rp
RP
RPa
RPC
RPM
RPS
RQ
RSC
Rt
RTI
RU
RWC
RX
s
S
S
S
SC
Reference soil concentration of pollutant (yg/g DW)
Recommended Maximum Contaminant Level
Runoff (mVmz-yr)
Annual application rate of pollutant in sludge (mg/m2-yr)
Reference application rate of pollutant (kg/ha)
Reference annual application rate of pollutant (kg/ha)
Reference cumulative application rate of pollutant (kg/ha)
Maximum pollutant application rate based on phytotoxicity
(kg/ha)
Reference single-application rate of pollutant, with no
waiting period (kg/ha)
Reference single-application rate of pollutant, followed by
waiting period or land-use conversion period (kg/ha)
Reference flux rate due to volatilization (g/m2-sec)
Reference sludge concentration (yg/g DW)
Total storm rainfall depth (cm)
Reference tissue concentration increment (yg/g DW)
Relative uptake response (unitless)
Reference water concentration (mg/fi.)
Reference starting leachate concentration (mg/8.)
Silt content of soil (%)
Water retention parameter (cm)
Storage capacity of sludge (kg/kg)
Solids content of the sludge (kg/kg)
Sludge concentration (ug/g DW)
Sediment delivery ratio (dimensionless)
xx1
-------
LIST OF ABBREVIATIONS (cont.)
SL
SMA
SRR
SRR
SW
t
t
T
TA
TBI
TC
TCLP
TKN
TL
TP
Tr
TT
Tu
UA
UB
UC
USLE
Site length
Sludge management area (ha)
Source receptor ratio
Source-receptor ratio (sec/m)
Site width
Metric ton
Time (years)
Waiting or land-use conversion period (years)
Threshold feed concentration (yg/g DW)
Total background intake (mg/day)
Tissue concentration (tig/g DW)
Toxicity Characteristic Leachate Procedure
Highest tissue concentration increment (yg/g DW)
Total Kjeldahl nitrogen
Plant tissue concentration limit (yg/g DW)
Threshold phytotoxic application rate of the pollutant (kg/ha)
Storm duration (hours)
Total travel time across all layers of unsaturated zone (years)
Steady-state travel time across an unsaturated zone soil layer
(years)
Uptake slope of pollutant in animal tissue [v»g/g(vg/g)-i]
Uptake response slope in soil biota [vg/g(kg/ha)-i]
Uptake response slope of pollutant in crops
[yg/g(kg/ha)-i] or [vg/gdig/g)-*]
Universal Soil Loss Equation
xxli
-------
V
vave
W
US
X
X
V1
LIST OF ABBREVIATIONS (cont.)
Volume of receiving water (mVyear for long-term case,
ma for event case)
Vertical term for transport (dimensionless)
Average velocity across the unsaturated zone (m/year)
Width of the buffer zone (cm)
Water content of sludge (kg/kg)
Leachate concentration (mg/cma)
Leachate concentration of contaminants (mg/S.)
Sediment loss rate for long-term case (mt/ha-year)
Length or width of source (m)
Sediment loss rate for event case (mt)
Lateral virtual distance (m)
Mole fraction of i in the gas phase (dimensionless)
Density of the sludge liquid (kg/a)
XX111;
-------
-------
1. INTRODUCTION AND DESCRIPTION OF GENERAL METHODOLOGIC APPROACH
1.1. PURPOSE AND SCOPE
This is one of a series of reports that present methodologies for
assessing the potential risks to humans or other organisms from management
practices for the disposal or reuse of municipal sewage sludge. The
management practices addressed by this series include land application
practices, distribution and marketing programs, landfilling, incineration
and ocean disposal. In particular, these reports deal with methods for
evaluating potential health and environmental risks from toxic chemicals
that may be present in sludge. This document addresses risks from chemicals
associated with land application and distribution and marketing of sludge.
These proposed risk assessment procedures are designed as tools to
assist in the development of regulations for sludge management practices.
The procedures are structured to allow calculation of technical criteria for
sludge disposal/reuse options based on the potential for adverse health or
environmental impacts. The criteria may address management practices (such
as site design or process control specifications), limits on sludge disposal
rates or limits on toxic chemical concentrations in the sludge.
The methods for criteria derivation presented in this report are
intended to be used by the U.S. EPA Office of Water Regulations and Stan-
dards (OWRS) to develop technical criteria for toxic chemicals in sludge.
The present document focuses primarily on methods for the development of
nationally applicable criteria by OWRS. It is suggested that a user-
oriented manual based on these methods be developed for wider use in
deriving site-specific criteria for these sludge management practices.
Additional uses for the methodology may exist, such as developing guidance
for the selection of sludge management options by local authorities, but
these uses are not the focus of these documents and will not be discussed.
1-1
-------
These documents do not address health risks resulting from the presence
of pathogenic organisms in sludge. The U.S. EPA will examine pathogenic
risks in a separate risk assessment effort. These documents also do not
address potential risks associated with the treatment, handling or storage
of sludge; transportation to the point of reuse or disposal; or accidental
release.
1.2. DEFINITION AND COMPONENTS OF RISK ASSESSMENT
The National Research Council (NRC, 1983) defines risk assessment as
"the characterization of the potential adverse health effects of human
exposures to environmental hazards." In this document, the NRC's definition
is expanded to include effects of exposures of other organisms as well. By
contrast, risk management is defined as "the process of evaluating alterna-
tive regulatory actions and selecting among them" through consideration of
costs, available technology and other nonrisk factors.
The NRC further defines four components of risk assessment. Hazard
identification is defined as "the process of determining whether exposure to
an agent can cause an increase in the incidence of a health condition."
Dose-response assessment is "the process of characterizing the relation
between the dose of an agent ... and the incidence of (the) adverse health
effect " Exposure assessment is "the process of measuring or estimating
the intensity, frequency and duration of ... exposures to an agent currently
present or of estimating hypothetical exposures that might arise " Risk
characterization is "performed by combining the exposure and dose-response
assessments" to estimate the likelihood of an effect (NRC, 1983). The U.S.
EPA has broadened the definitions of hazard identification and dose-response
assessment to include the nature and severity of the toxic effect in
addition to the incidence.
1-2
-------
Figure 1-1 shows how these components are Included in the development of
these risk assessment methodologies for sludge management practices. The
figure further shows how each methodology may be used to develop technical
criteria, and how these criteria could be used or modified by the risk
manager to develop regulations and permits;
1.3. RISK ASSESSMENT IN THE METHODOLOGY DEVELOPMENT PROCESS
As illustrated in Figure 1-1, the methodology development process begins
by defining the management practice. Even within a given reuse/disposal
option, "real world" practices are highly variable, and so a tractable
definition must be given as a starting point. As a general rule, this
definition should include the types of practices most frequently used. That
is, the definition should not be limited to ideal engineering practice but
also need not include practices judged to be poor or substandard (unless the
latter are widespread). This definition, presented in Chapter 2 of this
document, helps to determine the limits of applicability of the methodology
and the exposure pathways that may be of concern. However, as also shown in
Figure 1-1 and as discussed in Section 1.4., this definition could be
modified as the methodology is applied, since the methodology itself will
help to define acceptable practice.
1.3.1. Exposure Assessment. The exposure assessment step begins with the
identification of pathways of potential exposure. Exposure pathways are
migration routes of chemicals from (or within) the disposal/reuse site to a
target organism. For those pathways where humans are the target of concern,
special consideration is given to individual attributes that influence
exposure potential. Individuals will differ widely in consumption and
contact patterns relative to contaminated media and, therefore, will also
vary widely in their degree of exposure.
1-3
-------
i
-------
HI
LU
(3
CO
£
e
u
UJ
2
1-5
-------
An ideal way to assess human exposure is to define the full spectrum of
potential levels of exposure and the number of individuals at each level,
thus quantifying the exposure distribution profile for a given exposure
pathway. The methodologies described in these reports will not attempt to
define exposure distributions in most cases, for the following reasons.
First, it is very difficult to estimate the total distribution of exposures,
since to do so requires knowledge of the distributions of each of the
numerous parameters involved in the exposure calculations and requires the
modeling of actual or hypothetical population distributions and habits in
the vicinity of disposal sites. Such a task exceeds the scope of the
present methodology development effort.
Second, while knowledge of the total exposure distribution may be useful
for certain types of decision-making, it is not necessarily required for
establishing criteria to protect human health and the environment. If
criteria are set so as to be reasonably protective of all individuals,
including those at greatest risk, then as long as the risk assessment proce-
dures can reasonably estimate the risk to these individuals, the quantifica-
tion of lesser risks experienced by other individuals is not required.
The drawback, however, of examining only a maximal-exposure situation is
that the true likelihood of such a situation occurring may be quite small.
The compounding of worst-case assumptions may lead to improbable results.
Therefore, the key to effective use of this methodology is a careful and
systematic examination of the effects of varying each of the input param-
eters, using estimates of central tendency and upper-limit values to gain an
appreciation for the variability of the result.
1-6
-------
Therefore, exposure will be determined for a most-exposed individual, or
MEI.* The definition of the MEI will vary with each human exposure
pathway. Chapter 3 of this document will enumerate the exposure pathways
and will define the MEI in qualitative terms; for example, for the "home
garden" scenario, the MEI is a person producing much of his or her own crops
on sludge-amended soil. The MEI will not be quantitatively defined in this
chapter, but relevant information that allows the user to do so (such as
available data on the ranges of crop consumption rates) will be provided in
later chapters. For exposure pathways concerning organisms other than
humans, the term "MEI" is not applied, but conservative assumptions are
still made regarding the degree of exposure. The remaining chapters
(Chapters 4-8 in this document) explain the calculation methods and data
requirements for conducting the risk assessments for each pathway.
1.3.2. Hazard Identification and Dose-Response Assessment. To determine
the allowable exposure level for a given contaminant, the hazard identifica-
tion and dose-response assessment steps must be carried out. For human
health effects, these procedures already are fairly well established in the
Agency (although they still require improvement, and specific assessments
for many chemicals remain problematic). Hazard identification in this case
consists first of all of determining whether or not a chemical should be
treated as a human carcinogen. Procedures for weighing evidence of carcino-
genicity have been published in the Federal Register (1986a) and are further
*The definition of the MEI does not include workers exposed in the produc-
tion, treatment, handling or transportation of sludge. This methodology is
geared toward protection of the general public and the environment It is
assumed that workers can be required to use special measures or equipment
to minimize their exposure to sludge-borne contaminants. Agricultural
workers, however, might best be considered members of the general public
since the use of sludge may not be integral to their occupation
1-7
-------
discussed in later sections of this document. If a chemical is treated as
carcinogenic, dose-response assessment would then include the use of
Agency-accepted potency values. If none are available, cancer risk
estimation procedures published by the Agency (Federal Register, 1986a)
would be used to determine potency.
If a chemical is not carcinogenic, hazard identification and
dose-response assessment normally consist of identifying the critical
systemic effect, which is the adverse effect occurring at the lowest dose,
and the reference dose (RfD), which is "the daily exposure ... that is
likely to be without appreciable risk of deleterious effects during a
lifetime" (U.S. EPA, 1987a). Further description and procedures for
deriving RfDs are published in U.S. EPA (1987a).
For certain disposal options, effects on other organisms are of concern.
In these cases, existing Agency methodologies have been used where avail-
able. For example, existing guidelines for deriving ambient water quality
criteria (AWQC) (U.S. EPA, 1984c) are used to determine levels for
protecting aquatic life. Where effects on terrestrial species are of
concern, there are no existing Agency guidelines, but suggested procedures
for identifying adverse effects (hazard identification) and threshold levels
(dose-response assessment) are provided.
1.3.3. Risk Characterization. Risk characterization consists of combin-
ing the exposure and dose-response assessment procedures to derive criteria.
Risk assessments ordinarily proceed from source to receptor. That is, the
source, or disposal/reuse practice, is first characterized and contaminant
movement away from the source is then modeled to estimate the degree of
exposure to the receptor, or MEI. Health effects for humans or other
organisms are then predicted based on the estimated exposure. The calcula-
tion of criteria, however, involves a reversal of this process. That is, an
1-8
-------
allowable exposure, or an exposure that is not necessarily allowable but
corresponds to a given level of risk, is defined based on health effects
data, as specified above. Based on this exposure level, the transport
calculations are either operated in reverse or performed iteratively to
determine the corresponding source definition. In this case, the resulting
source definition is a combination of management practices and sludge
characteristics, which together constitute the criteria. These steps are
carried out on a chemical-by-chemical basis, and criteria values are derived
for each chemical assessed and each exposure pathway. An example illustrat-
ing how these calculations may be carried out is provided in this document
for each pathway assessed. However, as indicated by Figure 1-1, the compi-
lation of data on specific chemicals to be used as inputs to the methodology
is a process separate from methodology development. Health effects data for
individual chemicals must be collected from the scientific literature. In
many cases, the U.S. EPA has already published approved values for cancer
potency or RfD. Data pertinent to a chemical's fate and transport charac-
teristics, such as solubility, partition coefficient, bioconcentration
factor or environmental half-life, must also be selected from the litera-
ture. In some cases, data for particular health or fate parameters were
gathered for a variety of chemicals in the process of developing the method-
ology. Where this was done, the information may appear as an appendix. In
most cases, however, such information does not appear in the methodology
document and must be gathered as a separate effort.
Once these data have been selected, even on a preliminary basis, it may
be useful to carry out a rough screening exercise, using these data plus
information on occurrence in sludges, to -set priorities for risk character-
ization. Screening could reveal that certain pollutants are unlikely to
1-9
-------
pose any risk, or that data gaps exist that preclude more detailed charac-
terization of risk. Methods for carrying out such a screening procedure
will not be discussed in this document.
Following chemical-specific data selection, risk characterization or
criteria derivation may be conducted. The values derived as limits on
sludge concentration or disposal rate, together with the management practice
definitions, will constitute the criteria. When calculating the numerical
limits, it is advisable to vary each of the input values used over its
typical or plausible range to determine the sensitivity of the result to the
value selected. Sensitivity analysis helps to give a more complete picture
of the potential variability surrounding the result.
1.4. POTENTIAL USES OF THE METHODOLOGY IN RISK MANAGEMENT
The results of the risk characterization step can then be used as inputs
for the risk management process, as shown in Part II of Figure 1-1.
Although this document does not specify how risk management should be con-
ducted, some potential uses of the methodology in the risk management
process are briefly described here. These optional steps are shown as
dashed lines in Figure 1-1.
As suggested by the National Research Council (NRC, 1983), a risk
manager may evalute the feasibility of a set of criteria values based on
consideration of costs, available technology and other nonrisk factors. If
it is felt that certain chemical concentrations specified by the
calculations would be too difficult or costly to achieve, the management
practice definition could be modified by imposing controls or restrictions.
For example, requirement of subsurface injection or waiting periods before
grazing is permitted on sludge-applied lands could result in higher
permissible concentrations for some pollutants. The same degree of
protection would still be achieved.
1-10
-------
Following promulgation of the criteria, it may also be possible to
evaluate sludge reuse or disposal practices on a site-specific basis, using
locally applicable data to rerun the criteria calculations. Criteria could
then be varied to reflect local conditions. Thus, the risk manager can use
the methodology as a tool to develop and fine-tune the criteria.
1.5. LIMITATIONS OF THE METHODOLOGY
Limitations of the calculation methods for each pathway are given in the
text and in tabular form in the chapters where calculation methods are
presented. However, certain limitations common to all of the methods are
stated here.
Municipal sludges are highly variable mixtures of residuals and
by-products of the wastewater treatment process. Chemical interactions
could affect the fate, transport and toxicity of individual components, and
risk from the whole mixture may be greater than that of any single compo-
nent. At present, these methodologies treat each chemical as though it acts
in isolation from all the others. It should be noted that EPA's mixture
risk assessment guidelines (Federal Register, 1986b) caution that a great
deal of dose-response information is required before a risk assessment may
be quantitatively modified to account for toxic interactions. Future
revisions to these documents to include consideration of interactions will
most likely be limited to qualitative discussion of such interactions.
Transformation of chemicals occurring during the disposal (including
combustion) practice or following release may result in exposure to
chemicals other than those originally found in the sludge. In many cases
these assessment procedures may not adequately characterize risks from these
transformation products.
1-11
-------
In addition, these methodologies compartmentalize risks according to
separate exposure pathways. The use of an MEI approach, which focuses on
the most highly exposed individuals for each pathway, reduces the likelihood
that any single individual would simultaneously receive such exposures by
more than one pathway, and therefore the addition of doses or risks across
pathways is not usually recommended. However, it is possible that risk
could be underestimated in a small number of instances.
Finally, the methodologies look at exposed organisms in isolation.
Population-level or ecosystem-level effects that could result from a reuse
or disposal practice might not be predictable by this approach.
1-12
-------
2. DEFINITION OF MANAGEMENT PRACTICES
2.1. INTRODUCTION
Municipal sludge production was estimated at 6.84 million metric tons
dry weight* (t DM) in 1982. By the year 2000, production is expected to
nearly double (12 million t DW/year). A survey of 6.5% of the U.S.
treatment plants in 1982 revealed the distribution between disposal/reuse
options presented in Table 2-1. Land application and distribution and
marketing (D&M) practices accounted for 2.87 t DW, or 42% of the total (Booz
Allen and Hamilton, 1982; U.S. EPA, 1983b). If the relevant distribution
among alternatives remains constant, it would be expected that 5.0 million t
DW of municipal sludge will be applied to land annually in the year 2000.
This chapter will briefly define the many practices included within the
land application and D&M management options. A detailed description of the
process design for each of these practices will not be given here; addi-
tional information may be found in U.S. EPA publications such as the Process
Design Manual for Land Application of Municipal Sludge (U.S. EPA, 1983b) and
Composting of Municipal Wastewater Sludges (U.S. EPA, 1985b). The defini-
tions given here will be limited to certain aspects that help to outline the
scope for the risk assessment methods that follow. These definitions
include stated assumptions or requirements regarding the forms that these
practices may take. Some of these assumptions or requirements allude to the
potential exposure pathways from each practice. The pathways themselves
will be more fully discussed in Chapter 3.
m^n 9HH.H »t M?™™*"?-', the tm dry Weight" should be Considered to
mean dried at 105'C until a constant weight is obtained. Analysis of a
K?JU h C1°"taminant maV be done on a wet weight or as is basis, but the
results should be expressed on a dry weight basis.
2-1
-------
TABLE 2-1
Distribution of Sludges Between Disposal/Reuse Alternatives^
Alternative
Million Dry Metric Tons
(10*t DW)
Percent
Incineration
Land application
Human food chain crops
Nonfood chain crops
Distribution and marketing
Landfilling
Ocean disposal
Other0
TOTAL
1.85
0.82
0.82
1.23
1.03
0.27
0.82
6.84
27
12
12
18
15
4
12
100
asource: Booz Allen and Hamilton, 1982; U.S. EPA, 1983b
bBased on a 1982 random survey of 1011 treatment plants (6.5% of total
number of plants)
cother includes lagoons and impoundments that constitute storage rather
than disposal. Ultimately, these facilities will be closed as landfills or
the sludge will be exhumed and sent for disposal by one of the alternative
means. There are no data at this time from which one can quantitatively
ascertain the distribution of sludge volumes among the ultimate disposal
alternatives.
2-2
-------
2.2. LAND APPLICATION PRACTICES
Land application refers to the distribution of sludge on or just below
the soil surface where it is employed as a fertilizer or soil conditioner to
grow human food-chain and non-food-chain crops or to utilize the land as a
sludge treatment system. Potentially adverse environmental or animal and
human health effects can be prevented by establishing acceptable pollutant
concentrations, application rates, good management practices (such as
physical barriers or record-keeping requirements) or, in certain cases, land
deed stipulations to manage conversion to uses that could have greater
potential for exposure. Four major designations for land application are
listed below:
Agricultural utilization
Forest land utilization
Drastically disturbed land utilization
Dedicated land disposal site
These options are not mutually exclusive; there is an overlap in several of
their characteristics. However, for the purpose of regulating the use of
sludge in each option, it is necessary to clearly define each practice and
differentiate between them.
2.2.1. Agricultural Utilization. Agricultural use of sludge includes
sludge application to land used for a wide range of crops including grains,
animal feeds and non-food-chain crops. The objective of this practice is to
improve the soil-conditioning properties and nutrient status, and to
increase crop production. This practice is probably of most concern because
it may involve incorporation of the pollutant in the human food chain. The
following assumptions or requirements will be made regarding this practice.
2-3
-------
2.2.1.1. ASSUMPTIONS —
1. Human food-chain and non-food-chain crops are expected to be
grown where this practice is utilized.
2. Work practices that reduce the possibility of offsite contami-
nation, such as washing of application equipment, will be
assumed.
3. Agricultural lands have a potential of being converted to
residential usage after sludge application to crops.
2.2.1.2. REQUIREMENTS OR POTENTIAL REQUIREMENTS —
1 Sludge will be applied at no greater than agronomic rates,
defined as annual rate at which nitrogen (N) or phosphorus (P)
supplied by the sludge does not exceed the annual N or P
requirement of the crop. The range varies from 2 to 35 t
DW/hectare (ha) for most sludges, but may be as high as 70 t
DW/ha for some composts. (It should be noted that P often is
more limiting than N.)
2 A waiting period after application may safely allow higher
application rates of sludge because of chemical immobilization
or degradation of constituents. Therefore, the need for a
waiting period before planting crops will be evaluated, as will
a waiting period before grazing of treated pasture land.
3. The criteria derivation procedures given herein will evaluate
the need for differences in criteria based on soil pH.
4 The criteria derivation procedures given herein will evaluate
the need for criteria based on varying certain physical charac-
teristics of potential sites. These characteristics include
slope, depth to water table, permeability, infiltration and
proximity to surface water.
All publicly owned treatment works (POTWs) will be required to
analyze the chemical composition of sludge using proper quality
assurance procedures.
Records on the locations and amounts of sludge application and a
copy of the POTWs analysis of sludge composition will be main-
tained by the POTW.
Incorporation of sludge into the soil (either by injection or
tilling) is required before crops are planted for human consump-
tion. The criteria derivation procedures given herein will
evaluate the need for requiring soil-incorporation before other
uses (such as use for pasture or other crops).
5.
6.
7.
2-4
-------
2.2.2. Forest Land Utilization. Application to forest lands decreases
the concern over direct entry of the pollutant into the human food chain and
thus has an advantage over agricultural utilization. However, forest lands
may be converted to agricultural or residential use, so protection of these
lands for their potential future use is generally accepted. The following
assumptions and requirements will be made regarding this practice.
2.2.2.1. ASSUMPTIONS —
1
3.
The majority of plants grown where this practice is utilized
are not in the human food chain. The only human food chain
crops, for example, might be mushrooms and wild berries con-
sumed by humans or plants consumed by wildlife, such as deer or
birds.
Sludge may be applied at levels higher than the agronomic
rates. It has been shown that forest surface litter layers
have a comparatively high storage capacity; therefore, higher
application rates than those allowed on agricultural soils may
be applied in many instances without changing the degree of
protection from nitrate leaching to sensitive aquifers. The
range may vary from 10 to 100 t DW/ha every 3-5 years.
Work practices that reduce the possibility of offsite contami-
nation, such as washing of application equipment, will be
assumed.
Forest application of sludge will occur at different stages of
plant growth. Therefore, application is not limited by season,
except that application on frozen land is not permitted.
Forest lands have a potential of being converted to agricul-
tural or residential use after sludge application. However,
such conversion would not be immediate; an elapsed time or
conversion period following sludge application may be assumed.
2.2.2.2. REQUIREMENTS OR POTENTIAL REQUIREMENTS —
Public access is restricted by signs adjacent to public roads.
The signs prohibit use of the area or its products for human
food consumption.
The public will be restricted to a given distance downwind
during spray application.
2-5
-------
4.
5.
The criteria derivation procedures given herein will evaluate
the need for criteria based on varying certain characteristics
of potential sites. These characteristics include slope, depth
to water table, permeability, infiltration and proximity to
surface water.
The POTW will be required to analyze the chemical composition
of sludge using proper quality assurance procedures.
Records on the locations and amounts of sludge application and
a copy of the POTW's analysis of sludge composition will be
maintained by the POTW.
6. Sludge application on frozen land is not permitted.
2.2.3. Drastically Disturbed Land Utilization. Sludge application to
drastically disturbed lands is often referred to as land reclamation. In
this practice, barren lands are treated to improve site aesthetics and
utility through regrowth of vegetation and landscaping. These barren lands
can be a result of mines, quarries or sand and gravel pits. They may have
problems such as acid runoff, high erosion rates, low nutrient levels and
high toxic levels of polutants. All these characteristics can be improved
by sludge application if the site and sludge use are properly managed. In
these practices, the amounts of sludge applied can be drastically different
than those applied for agricultural and forest use. The following is a list
of assumptions and requirements that will be made for this practice.
2.2.3.1. ASSUMPTIONS —
1. Because sludge application is expected to drastically enhance
or to be a substitute for topsoil, a much larger application of
sludge may be necessary to establish vegetation and improve the
physical properties of the surface material. It is often
applied to disturbed land in a one-time, large application, but
repeated applications may also occur.
2. Work practices that reduce the possibility of offsite contami-
nation, such as washing of application equipment, will be
assumed.
2-6
-------
Disturbed lands have a potential for being used as agricultural
or residential lands following reclamation with sludge. Unless
the same restrictions that protect these latter uses are
applied to reclamation practices, it is assumed that the land
will be tested and evaluated to determine its suitability
before it is used for agriculture or residences.
Land conversion to its intended use would not be immediate; an
elapsed time or conversion period following sludge application
may be assumed.
2.2.3.2. REQUIREMENTS OR POTENTIAL REQUIREMENTS —
2,
3.
The criteria derivation procedures given herein will evaluate
the need for criteria based on varying certain characteristics
of potential sites. These characteristics include slope, depth
to water table, permeability, infiltration and proximity to
surface water.
The POTW will be required to analyze the chemical composition
of sludge using proper quality assurance procedures.
Records on the locations and amounts of sludge application and
a copy of the POTW's analysis of sludge composition will be
maintained by the POTW.
Public access is restricted by signs adjacent to public roads.
The signs prohibit use of the area or its products for human
consumption.
2.2.4. Dedicated Land Disposal Site. The primary purpose of the site is
long-term disposal of sludge. The objective is to employ soil as a
treatment system by allowing soil to retain the metals and allowing
sunlight, microorganisms and chemical processes to degrade organic matter in
the sludge. The following is a list of assumptions and requirements for
this disposal practice.
2.2.4.1. ASSUMPTIONS —
1. It is assumed that any crops produced on "dedicated land" are
unfit for human consumption (or for consumption by animals that
will be consumed by humans). Crops produced on dedicated sites
may be utilized for human consumption or for consumption by
animals that will be consumed by humans only if their
compositon is found to be acceptable. Also, feed crops for
animals not consumed by humans may be grown if
analyzed and evaluated before use.
crops
they
are
2-7
-------
2. It is assumed that these sites are totally controlled by a
responsible governmental agency.
2.2.4.2. REQUIREMENTS OR POTENTIAL REQUIREMENTS —
1. Work practices that reduce the possibility of pffsite contami-
nation, such as washing of application equipment, will be
required.
2. Public access is restricted by signs and fences adjacent to
public roads. The signs prohibit use of the area or its
products.
3. The criteria derivation procedures given herein will evaluate
the need for criteria based on varying certain characteristics
of potential sites. These characteristics include slope, depth
to water table, permeability, infiltration and proximity to
surface water.
4. The POTW will be required to analyze the chemical composition
of sludge using proper quality assurance procedures.
5. Records on the locations and amounts of sludge application and
a copy of the POTW's measurement of sludge composition will be
maintained by the POTW.
6. Future property owners are notified by a stipulation in the
land record or property deed that states that the property has
received solid waste at high contaminant application rates and
that unless soil analyses show an absence of hazard, human
food-chain crops should not be grown and young children should
not be permitted access to the site because of possible health
hazards.
7. It is assumed that closure and hazard evaluation procedures
will be established and will be required before sale of a site
or use for another purpose.
2.2.5. Summary. There are differences among land uses and sludge appli-
cation practices. The potential changes in land use may blur some of these
distinctions where risk assessments are concerned. Forest, agricultural and
reclaimed lands can potentially become agricultural or residential. For
agricultural use, sludge application is limited to agronomic rates, whereas
for forest and reclamation uses it typically is higher. A second difference
is that for forest use, there will be restricted use of the forest and its
2-8
-------
products for human consumption, and a conversion period Is assumed before
agricultural use occurs. Third, an assumption for forest and reclamation
usages is that there will be a conversion period before agricultural use.
Conversion periods before residential use are also assumed for nonresiden-
tial lands. A critical difference between forest, agricultural and land
reclamation use vs. dedicated-land disposal is that in the latter, conver-
sion of land use is not assumed. Therefore, if a dedicated site is to be
converted, its suitability for the intended use needs to be carefully evalu-
ated. This evaluation is ensured by the requirement of a warning statement
in the land record or property deed.
2.3. DISTRIBUTION AND MARKETING PRACTICES
Distribution and marketing (D&M) refers to the give-away or sale of
sludge or sludge products either in bulk or bagged to the public, commercial
growers or local governments for use as fertilizers or soil conditioners for
food- and non-food-chain vegetation. Potentially adverse environmental or
animal and human health effects can be prevented by establishing acceptable
pollutant concentrations, application rates and labeling requirements.
Establishing and monitoring of good management practices or record-keeping
requirements by users are not feasible using this option.
Usually, in this practice the sludge has undergone some method of treat-
ment to either dewater or reduce the volume of sludge before distribution.
Sludge treatment before D&M varies among POTWs and may include, but is not
limited, to the following:
Aerobic digestion
Anaerobic digestion
Heat drying/treatment
Mechanical dewatering
Composting
Air drying
2-9
-------
In addition, humus material or nutrient additives may be blended with the
sludge to increase its fertilizer or soil-conditioning value.
Distribution and marketing programs currently practiced by POTWs range
from simple give-away programs, in which the local citizen picks up sludge
that has been stockpiled at the treatment plant, to detailed marketing
programs, such as the distribution of a bagged product to retail and whole-
sale outlets. The various end-uses associated with these programs can be
classified as shown below:
Residential
Gardens
Lawns
Landscaping
Commercial
Nurseries
Turf farms
Golf courses
Other horticul-
tural uses
Institutional
Parks and
recreation areas
Cemeteries
Roadsides
School grounds
and other public
lands
The end-use and demand for sludge-derived D&M products is based upon the
characteristics of the sludge, which include nutrient content (concentration
and availability of nitrogen and phosphorus) and physical nature (moisture
content and consistency).
These end-uses of D&M sludge products determine the populations and
environments initially exposed to contaminants that may be present in the
product. For instance, expected human exposure to sludge-borne contaminants
would be much higher for sludge applied to home gardens than for sludge
applied to cemeteries or roadsides. Sludge ingestion by children by either
hand-to-mouth play or pica would be a possibility for the residential
settings, but is less likely for commercial or institutional applications.
2-10
-------
In addition to end-use of the D&M sludge product, however, future use of the
land to which the product has been applied is also important in determining
potential risk. Therefore, the potential for land-use conversion must also
be considered.
The following assumptions are made for this option.
2.3.1. Assumptions.
1.
2.
3.
Both food-chain and non-food-chain crops are grown using D&M
sludge products.
Few, if any, site controls or record-keeping requirements are
employed.
Animals consumed by humans are not expected to forage on crops
grown on sludge-amended soil in the residential, commercial or
institutional D&M end-uses. Applications to parks or recrea-
tional areas where hunting is permitted would constitute forest
land utilization rather than D&M.
All lands to which D&M sludge products are applied, with the
exception of roadsides, are considered to have the potential
for conversion to residential use, including use for home
gardens. In many cases, however, such conversion would not be
immediate; an elapsed time or conversion period following
sludge application may be assumed.
2.3.2. Requirements or Potential Requirements.
1. A printed handout (in the case of bulk distribution) and a
label (for bagged products) will provide information on essen-
tial plant nutrient content, instructions for proper use on
different plant types and loading rates (such as number of
square feet per bag, ratio of sludge to soil in sludge-soil
mixture) that should not be exceeded.
2-11
-------
-------
3. EXPOSURE PATHWAYS AND MOST-EXPOSED INDIVIDUALS (MEIs)
Humans or other organisms may be exposed to the contaminants in land-
applied sludge by a variety of pathways. The relative importance of a given
pathway is influenced by the type of land application practice employed and
by potential future uses of the land receiving sludge. The definitions,
assumptions and requirements stated for these practices in Chapter 2 will be
used here to identify those pathways that may be of concern for a given
practice and those that can be eliminated a priori.
If a particular exposure pathway is stated here to be of concern for a
given management practice, this means that the assessment methods for that
pathway, described in the latter chapters of this document, should be used
to determine whether criteria will be required. It does not necessarily
mean that criteria will be needed.
For each exposure pathway, it is important to identify the most-exposed
individual, or MEI. Occupational exposures (other than of agricultural
workers) are not considered, as discussed in Chapter 1. While many individ-
uals of the general public may be exposed to a varying degree, the MEI is
that individual who would be expected to experience the greatest risk and,
therefore, requires the, greatest protection. The MEI is a hypothetical (not
actual) individual, but care should be taken that the definition be real-
istic. The definition depends on many of the assumptions and requirements
made concerning each management practice listed in Chapter 2. The pathways
and MEIs will be described qualitatively in this chapter and quantitatively
in the following chapters where criteria calculation methods are given.
3-1
-------
The upper portion of Table 3-1 summarizes the selection of pathways that
are relevant to each land application or distribution and marketing (D&M)
practice, according to either current or future land use. Some judgment was
required in making these selections, since practice definitions may vary or
overlap. The pathways selected are those judged to have a reasonable
probability of becoming important for each management practice.
The resulting matrix of pathways and practices is complex, reflecting
the complexity of sludge utilization practices and of the environment. To
provide a more manageable framework for conducting risk assessments, a
reduced number of practice categories is suggested, as shown in the lower
portion of Table 3-1. For example, since all land application sites other
than those dedicated to sludge disposal are considered to have a potential
for eventual conversion to agricultural lands or residential gardens, all 13
pathways may apply either currently or in the future. Except for roadsides,
all lands to which D&M sludge products may be applied are also considered to
have the potential for conversion to residential use (see Section 2.3.1.).
Therefore, highway landscaping has been singled out from all other D&M uses
for special treatment. Any other D&M later shown to fit this criterion
could be handled in a similar fashion, however.
A future exposure may entail lower risk than a current one, since some
contaminants applied in sludge may be lost over the time period elapsed
before land-use conversion. For simplicity, the suggested categorization
ignores the conversion period. However, methods to account for the
conversion period in criteria calculations are provided in Chapter 4 and can
be used if desired.
3-2
-------
cu
VI
^
T3
IO
_J
CU
&—
3
3
Lu
S.
O
. t
c
£
i-
3
O
4-1
0}
r—
•fi
ie
_o
O.
D.
l/l
i
.5
ro
£
cu
~
VI
O
Q.
X
i . t
ro
|
+J
OL
£
5
O
O
X
UJ
J
t
en
CM
r—
O
Ol
CD
"-
itn
in
•*
«
CNJ
r—
a
VI
cu
u
u
2
a.
C
CU
DJ
C
i
CU
OJ
o
/5
XXXX Illtlll 1 XXII
XXXX XXXXXXX X XXXX
XXXX XXXXXXX X XXXX
XXXX IIXIXI1 X XXXX
XXXX XXXXXXX X XXXX
XXXI XXXXXXX X XIXX
XXXI XXXXXXX X XIXX
XXXI XXXXXXX 1 XIXI
XXXI XXXXXXX 1 XIXI
XXXI Illllll 1 XIII
XXXI Illllll 1 XIII
XXXI XXXXXXX 1 XIXI
u
XXXI XXXXXXX 1 XIXI
1—
ro
t/1
O OJ
-^ CL C
OJ 1/1 •!—
C «r- O.
••—1/1 t/1 -o to
ro I- 4-» c 4-* *o vi
U >
4-> O O 1/1 4-> IO
•— * • CCO V1IO3
•EC ••" O *-^ .C *^ (J <—
O0 3 *—*'^ <_> «r— OJ
O CO T— 4-> Vll/l Jii-O T-
*-^i — r- CO U ro t/1 CU -C
ro 4-> 3 eu • -i- -a
OJ -4-> c -a L. vi vi ae. 4->
•— CCCCUO COCUT3 +J I— 0.
ro »t— cu cu E= &~ «f-c i-cxrocu
VI 4^'OEraa. r— S- fl3 OCUV1U
O CUUraC roCUr— «- u O X
^. o.jxracs-o. C4-> OJ XCLCUOJ
vi ^o cu -o rau.aia i-cu oro
O 4->eZ C34->CU4->DJL.»CLI/I OJ'i- C 4-> VI
•*— r- rora ro o •(— O c u vi -a CU4-*rou'O
4-> ID E 1— I— Q.W- Q.-^- CUCUl-C 4-> O I— rO C
ro t- cu ro 1 c: ro 1 1 *— * E&-VJCUIO rorol^-ro
•'-•SS'u^ •?4^eSeueuroT335'~ "o.'S0"'"
1— I — OJ4-» 4-1C»f-»i— VJU-COO>> "O 4->E>»
CL34->l-rO 3CUS-i-3 rOUrO «_ioQ3
ro S-
+•" J= 1.
•I- 03
•" eu U
X tu !_
O *~* VI CU
sr >, o 4J1 ^
C 4-> C -i- .r- CU -t^
o *t— ro 4-» u u cu
4-> T- 3 CU X 4-1 O
vi x :r OJ o Ti Jj
cu 0 1 > c 1— ro
>> OJ 1— .— . 4-> I-H I. 4J
4-> C C i- >S_ Q.C
•r-l-0 Jo
O rd -r- •(- O *r- 4-> 4J >
•>- 1— E 4-> x eu oro c cu
ooareuh— -r->ixcu E 'cu
1— CO 1 OJ a 4-> O L. m Si
= r- C r- •— T- 1— Q. C OJ
c — to I-H ro u 10 rovi
*~* jo E E>»'r-rarO'r- c >>
01 E >-r- 4->-r-4->X4->+Jvl IO fO
!34->COC-r-OOOC £ 3
•— 3= -r- OOt/)(/)00(/)(/50000000-(/)CS> It- S
V) g} .^
s-i— cvjco^-inior-cocnor-CMco s_ .^
i u- x
-Q u
3-3
-------
3.1. TERRESTRIAL FOOD-CHAIN PATHWAYS AND MEIs
These pathways refer not only to the human food chain but also to the
ecological food chain. Terrestrial trophic relationships are numerous and
complex. A few pathways have been selected as perhaps most important for
assessing sludge applications. These ecological pathways involve plants,
soil biota and their consumers. Pathways leading to human exposure are also
given special consideration.
3.1.1. Crops for Human Consumption. This pathway (sludge-soil-plant-
human toxicity) is important wherever crops for human consumption are grown
or may be grown subsequent to sludge application. Uptake of sludge contami-
nants is assumed to occur through the plant roots. Direct adherence of
sludge or soil to crop surfaces is minimal; crops are washed before consump-
tion.
The relevant practices for this pathway include agricultural use and use
of D&M products for home gardens or in commercial enterprises where crops
for human consumption are raised, whether in pots (e.g., hothouse
production) or in the field (e.g., truck farming).
For agricultural use and field use of D&M products, the MEI could be
defined as an individual residing in a region where a relatively high
percentage of the available cropland receives sewage sludge applications.
All crops in the diet could be presumed to be affected owing to crops from
sludged and nonsludged lands being completely mixed within the region. The
percentage of the diet potentially affected would be much higher, however,
where D&M products are used for home gardening. A scenario in which the
Individual grows a large proportion of his or her own food would result in
higher risk.
3-4
-------
The future uses of agricultural lands may include residential develop-
ment. Risk is therefore assessed on the basis of eventual suitability for
home gardening.*
It is assumed that few plants grown on forest lands are ordinarily con-
sumed by humans and that foods present where sludge has been applied will
not be harvested, since required warning signs restrict access (see Section
2.2.2.). This pathway is, therefore, not considered for forest land under
its current use. However, the future use of forest land may include
agricultural use or residential development. Since criteria are intended to
protect future use, risk is assessed on the basis of eventual suitability
for agriculture or home gardening.* Similarly, reclaimed land may
eventually be used for agriculture or residences.
Lands dedicated to sludge disposal do not require assessment by this
pathway, however, since crops for human consumption are not grown and future
property owners are notified that such crops should not be grown.
D&M products not suited for home garden use may be applied only where
there is considered to be virtually no likelihood of land conversion to
residential use (i.e., roadsides). All other D&M uses require an assessment
by this pathway.
3.1.2. Soil Ingestion by Children. Human adults may ingest some soil,
but the amounts consumed by young (i.e., preschool) children are much
greater. These children constitute the MEI for this pathway (sludge-human
*If physical terrain or some other factor(s) completely preclude the possi-
bility of any such future use, it may be desirable to conduct a site-
specific risk assessment based on continued use as an agricultural or
forest site.
3-5
-------
toxicity). Since D&M sludge is not necessarily soil-incorporated, it is
assumed that children may ingest sludge directly. Preschool children are
assumed to be exposed in residential areas where D&M sludge has been applied
to gardens, lawns, landscaped areas or turf farms providing turf for
residential use. Park and recreational areas could also be sites of
exposure. Sludge application to agricultural, forest and reclaimed lands is
assumed not to result in exposure of preschoolers. Following land-use
conversion, however, such exposures are considered to be possible.
Dedicated land is not assessed according to future use because of warning
statements in the deed and required hazard evaluation procedures if land is
converted (see Section 2.2.4.2.).
3.1.3. Herbivorous Animals for Human Consumption. Two separate pathways
are considered whereby animal products may become contaminated: 1) sludge-
son-plant-animal-human toxicity, and 2) sludge-animal (direct ingestion)-
human toxicity. By the first pathway, row crops (e.g., corn) or other
forage crops (e.g., grasses) are grown on sludge-amended soils and take up
contaminants through the roots. The crops are harvested for animal consump-
tion. By the second, sludge is applied over growing forage crops and
adheres to crop surfaces or remains in the thatch layer on the soil surface.
The crop is then harvested or grazed shortly after sludge application,
resulting in direct ingestion of sludge particles. Alternatively, sludge is
incorporated into the soil. The land is then grazed, and heavy grazing
pressure is assumed to maximize soil ingestion. The MEI is the human
consumer of these animal products.
The first of these two pathways (uptake) is clearly important when
sludge is applied to agricultural lands, since animal feeds may be grown.
3-6
-------
The importance of the second (direct ingestion) is partially dependent on
whether or not surface applications without soil-incorporation are permit-
ted. The methodology assumes soil incorporation is not required for pasture
crops; the criteria calculation procedures can be used to determine whether
or not this practice would be acceptable (see Section 2,2.1.2.).
When sludge is applied to forests, forage plants may be contaminated by
uptake or by direct adherence. Herbivores, such as deer that are wide-
ranging, may forage in sludged areas and may be taken by hunters in other
areas. An individual consuming large amounts of wild game would be the MEI.
Conversion of forest or reclaimed lands to agricultural use would also
result in human exposure through meat consumption, as discussed above.
Reclaimed lands are to be protected according to potential future use,
including forest or agricultural use, as stated earlier. Dedicated lands
for sludge disposal need not be regulated for these pathways, since human
food-chain crops are not produced, and conversion of use is only permitted
following hazard evaluation (see Section 2.2.4.). Grazing or production of
feed for animals consumed by humans is assumed not to occur where D&M sludge
products are used (see Section 2.3.)
3.1.4. Toxicity to Herbivorous Animals. The exposure pathways for
herbivorous animals are as described in the previous section, but the end-
point of concern is toxicity to the animals themselves, which constitute the
MEI. The pathways are 1) sludge-soil-plant-animal toxicity, and 2) sludge-
animal toxicity (direct ingestion). For these pathways, it does not matter
whether the animals are subsequently consumed by humans. The management
practices to which these pathways apply are the same as described above,
except that it is assumed that wildlife may forage on lawns, gardens, etc.,
where D&M sludge products are used.
3-7
-------
3.1.5. Phytotoxlcity. This pathway is described as sludge-soil-plant
toxicity. Toxic effects in plants are of concern for most sludge applica-
tion practices, since the purpose of most types of utilization is to promote
plant growth. On land dedicated to sludge disposal, plant growth may be
important in stabilizing the soil to curb erosion and in protecting ground-
water by removing available nitrogen. However, not all dedicated-land
disposal programs encourage vegetation, and therefore evaluation of this
pathway is not required. This pathway should be evaluated for all other
management practices, however. The MEI, or vegetation type to be protected,
may be varied to match the land use if sufficient data are available to
adequately assess effects for different vegetation types; otherwise, the
most sensitive plant for which data are available will be assumed to
represent all plants.
3.1.6. Toxicity to Soil Biota or Their Predators. Two pathways are con-
sidered here: 1) sludge-soil-soil biota toxicity, and 2) sludge-soil-soil
biota-predator toxicity. The term "soil biota" is intended to be interpret-
ed broadly. The first pathway examines effects on a broad range of organ-
isms including microorganisms, soil invertebrates such as earthworms, or
various arthropods living in or near the soil, as long as potential effects
in these organisms can be related to soil concentrations. The second
pathway examines effects on predators of these organisms, especially small
mammals and birds. These predators could include insectivores, for example,
as long as available data permit the contaminant concentrations in their
prey to be related to soil concentration.
3.2. PARTICULATE RESUSPENSION PATHWAY
Particulate resuspension leading to human inhalation exposure may be of
concern in certain situations. This concern probably can be limited to
3-8
-------
practices in which sludge is applied to large areas (>1 ha) where estab-
lished vegetation is lacking or where mechanical resuspension, such as by
tilling, may occur. These situations include agricultural use, land recla-
mation use and dedicated land disposal, and also outdoor nursery operations,
truck farms, turf farms and highway landscaping projects using D&M sludge
products. A tractor driver performing tilling operations is examined as the
MEI.
3.3. SURFACE RUNOFF PATHWAY
Contaminants in sludge applied to soil may be transported in surface
runoff to receiving waters including streams, lakes or estuaries. Harmful
effects could, therefore, occur in aquatic organisms residing in the water
or in humans and animals drinking the water or consuming aquatic organisms
living in the water.. The runoff pathway should be considered to pose a
potential problem except where the sludge is completely contained, as in
indoor nursery operations. Site size tends to be more limited for most D&M
applications than for the other land application practices, and therefore.
the likelihood of runoff impacts is usually minimal. However, since appli-
cation to large sites is not ruled out, D&M uses are not excluded.
3.4. GROUNDWATER PATHWAY
Subsurface transport of contaminants to groundwater, and subsequent
ingestion by humans, may be of concern where sludge is applied to land.
Like surface runoff, this pathway is considered potentially important except
where sludge is completely contained. The MEI is an individual obtaining
drinking water from a nearby well.
3.5. VAPORIZATION PATHWAY
Volatile contaminants may vaporize from land-applied sludge, and subse-
quently may be transported downwind to cause human exposure. Vaporization
3-9
-------
1s not a potential problem where composted, heat-dried or air-dried sludge
1s used; in these cases volatile contaminants are considered to have had
adequate time to escape. Therefore, D&M uses are not evaluated for this
pathway, but agricultural, forest and land-reclamation uses and dedicated-
land disposal require evaluation.
3-10
-------
4. CRITERIA CALCULATION METHODS FOR TERRESTRIAL
FOOD-CHAIN EXPOSURE PATHWAYS
4.1. TERRESTRIAL FOOD CHAIN — GENERAL CONSIDERATIONS
Several factors affect many of the exposure pathways associated with the
terrestrial food chain. These factors are addressed below and will then be
referred to in later chapters dealing with specific pathways. The assump-
tions involved are presented in Table 4-1 and also are discussed in the text.
4.1.1. Calculation of Application Rates To Achieve Specified Soil Concen-
trations. The criteria calculation procedures given in the following pages
frequently calculate a reference soil concentration, RLC (in yg/g DW),
that will protect human health or the environment. It is then necessary to
translate the RLC into a corresponding reference application rate of the
pollutant, RP (in kg/ha). Calculation of the RP must reflect soil
incorporation practices and also loss rate of the contaminant from the soil.
4.1.1.1. SOIL INCORPORATION OF SLUDGE -- In many, but not all, land
application practices, sludge is incorporated into the upper layer of soil
before crops or other vegetation are grown. Incorporation is usually
accomplished by disking or chisel plowing of surface-applied sludge or by
direct injection into the soil. An assumption typically used is that sludge
is mixed into the soil to a depth of 15 cm (6 in), and that the soil has
a bulk density of 1.33 g/cm3; therefore, the dry mass of this upper-
layer of soil is 2xl03 t/ha (Naylor and Loehr, 1982; Donahue et al.,
1983). This assumption results in the following relationship between
contaminant application rate and initial contaminant concentration in soil:
where:
RP = RLC x MS x 10 3
RP = reference application ratie of pollutant (kg/ha)
RLC = reference soil concentration of pollutant (yg/g DW)
MS = 2x103 t/ha = assumed mass of soil in upper 15 cm
10-3 = conversion factor (kg/g)
4-1
(4-1)
-------
IA
c
o
4-*
ro
4-»
Rani1f1cations/L1m
c
•*~
rt)
JC
CJ
E
£
r~
e
< 5
in
VI
re
£
r—
S
O
•r*
U
•j
LL.
c m
tO L_ *r~
Ul ^->* (U
•O >, 0)
c j= 10 *- c:
O 4J r— O O
CL 0. t*- M
I/I
en o CL o x:
*^ IO L* *r~ 4-1
en o u-
SLU- 0 M- U-
>-• e E o •
t— >r- re m
C£ VI Ql
u. • 01 tn tn
o ra x: oi re
JS +J • r— Ol
o oi c re oi o
•i— ^ H— m u oi
4J 1- CTJ
re CM re oi f-
L. 4J > tn x:
4-1 U- O 4J
c o *»— o> • CL
0) X3 O !• Ol
o tt> i- ro -a
e 4J i— > oi
o re •«- r- c
U I- O T3 U O
r- CZ re VI 4->
•<- o ez — ' in re
Of- o oi t-
VI 4-> T3 1— O
re m 01 o.
re u 4J 4J a i-
M- O O J- O
• r- re T- re o
c o. cx-o c
o o. E 01 in •!-
M- (0 -^ i. 4->
4-> CL C in
t4-> - i- ro re
C VI 01 l~
3 re in -a CLXI
in 4J o> cz oi
in 3 i — 3 c 4-1
re r— o re
r— in oi c
in o f- xi tn T-
T- a. 4-> E
x: ^-»-o o re
4J re in r- Ol 4J
tn 3 u- e
>> o re o u- o
ca 4-» E o ai o
•
5 0
a) re co
E CO
VI E r—
Vt U
re M
in o
V) 4->
ezi- T-
o m
4J +J Ol
re CL-O
S- 01
o -a *c
CL I—
S. C 3
o o xi
0 t-
c 4-> re
i- ro
s- in
f™ O (Q
•se*
(A O r-
U- C O
h-l «r- VI
^^
•
'[ , ^_
O •
c •
O *f
re o
0 *J
CL (J
i. a>
o to
o ^^
•r- Ol
o>
r~* ^3
O r—
LO VI
m c at
•i- 3 tn
o o
c s- x;
0 014-1
4-> 0 t.
re re o
L. _a u-
c o -a
Ol I- Ol
o oi tn
C N 3
0 C
u o at
C XI
TZI
C 01 T3
3 4-> r—
o re 3
t- 0 O
oiv- x:
££ "O in
0 C
d be underpredicted if ba
here site-specific data i
ns, actual concentrations
"~ 3.?
o 4-> tn
o • re oi
0 U.r-
m i- 4-> 4->
4-> Ol C •!-
0 M Ol I—
oi o re
U- 4-> C 0
u- o o o
UJ C O r—
1
•o o
t- •!-
0 4->
Ol
C
f— Z*l
o tn
•»—
re o
4->
C O
01 S-
O Ol
r~ hj
Soil background coi
narily considered ;
organic compounds.
1 JZ>
oi >» re
S- 0 X>
3 CO
in O>
E o S- c
•i- 0 O
• TJ in •»-
in o> •!- 4->
I— t- -O CL
•r- CL E
O t- VI 3
tn 01 •!— tn
> .e tn
o o 1— re
cz m
•CJ 01 • •«—
C 4-> C J=
3 U- O 4J
OO-i-
jQ 4-> -
c re 01
>1 O 0 1-
4-> 4-> r— «-
x: CL CL o>
Ol E 0. S-
•r- 3 re 0)
4-> in -C
in OI 4J
a> re 01
|_ 13 .. .
re vi 3 in m
•r- i — in -O
in x: m o s-
(— 4-> r— re
re en N
4-> 4-> C 01 re
§ro H- E J=
x: 3 o
4-> o m e
>) r— •<-
> 3 i — in re
re o o CD 4->
a> .c u- 4-> *-
x: tn re a>
tn o o
4-> 0 C •!-
in 4-> O TZI tn
§•1- e 4->
T3 4-> ••- 0
c re ••—
.C OI I- >T3
O)4-> 4-> i — Ol
3 e xi s-
o m 01 to CL
.C 4-> 0 XI S-
4J c c o ai
r- O) O !- >
«x E o CL o
O S- 4-> 4-> >>
4-> O) in re xi
CL O .C
TD CL E 4-> in
a> 3 in
E i — 4->
re o •• c
c c c re
a> -i- •<- r-J o •
S- i- Ol
re >i 01 •«- c
vi 01 CD c_9 c ^
4-> 4J X: Ol 0
c -i- 1— - -i- re
re c xi m 01
C T- CL. i —
•r- It- • O
E 0) r- -4-> i-
ra -a T- T3 o
4-> C 0 CJ 4->
c i- in o c
o - ai o
ai x: oixi 4->
t — > 4^ . 3 re
U 0> tn M
>, o> u- -^ •<-
U VI O 4J r—
O c tn o "—
Ol O i- i— C 4->
01 o a> re re
4-1 >*4J 0> I —
re ai re ai s_ o
o xi r- E re >
in
lr_
o
in
E
o
s-
l*- *~*
VI OJ
tn
O r—
"~ ,J
c •*
re
c c
•r- O
E •*"-
4-> 0
C Ol
O CO
^»
1 u
flj i/i OJ«r- »
X tn 3 U- »r- C -O *-
0) C in O -C O 0) -0 CD
"a, •«- «r- -O O. 3 (D
o •*- CD c o a> u o
(j "O +•*+-* 3 *r- *O X;
•O C tfl O "O C t/I
m (o ro c u- CD 3 *- ro
JT r— E 4-> r— O fO IO I- •
3 tn -r-tO3 -M>rt3C
•* xi-o>>ro 0.0
0) OJ C EC E 0> I/* 4J
tO -M *r— t/1(JCD fO (O 4-1 *DC
l_ i-+J tnCr— +J O) O CD •«-
C U-i- i- U fO -o M~ » CDt/l
+J TDCU 0)01(13 f03> E
to O)0 -O > »3CU'«-
to uo 4->r— t. ico .QX:
t_* (OU (/itt>4JO^ 4->
Dl U O) CU J= on "O
ajdj L.-"O r~ T3
•OT3 3 JDOO .i-OJ. 3C
l_ l/»trt *r- U- (/I (/> O 3
+JQ *^ (/» t/>C» EO^- .CO
U -r- LOOt. C JH 0) • tA U-
.,- +J 34-* O -r- C C
T3 t/» (O Q. •*-» > VI O* O V» O
flii. ua> "acr-T-co
S_-r- 4->C QJC3 O>O4^O
CL<4- " fO(O JD3O t/)«r-C(J»r-X3
U r— r— ti-Z (O-POC4-Jr-
I*2" 2"^ 1 m • ml'5'-vi 5 1
3^ Otn tnOl COC-r-'r-
re ere 4-ire+J o i— -i- x) m m
<_m oai -i— re •»- re E 01
om i- cs-'+Jcjremaii—
10 IDC id ••- c i- i— cxiaire
>c 4-1 t. +J c r— in -a re
O4J tno tn4->reo re «- r- c c>
o oic oicO'*-ut-re3«r-
T3 C •<- "- 0) ™- 4-> ••- O S-
"3-0 "clr^ 'o.c'oi.r— inl-'<-C VIX:
OC E-r- EOD.01-r-OICO'i-4J
Ore >-l3 worel-QS4Ji-OQ£0
o in
Ol TJ 1 '
2i£t! § . o ojc'S
3oreoo s-re 4-" S-TZIO
tno-4-ii— rei— ^>c id e •!- m
EOIL.VI 013 ere o>re4-> < —
01OEO1OO CET3OI OCS- 3OI tO 3t-
t/jO X) •»— T-3C4-* 'r-.f-ttl re"O»O rt)O
o >> ec i— o re o +JEx: oisin-t- oiin
^:o>>>ore -i— 4-* rere4-> 4-»« — t-i — 4-1
Jj -r- re re T- 01 > u- - a. t-4->o reinreo. rec
COIE4JS- S-OCE 4->C i— 0( O. r— •(-
o>rex: reo 3 o>> COTZI cut- >> re o.
S-OI4->N OC-r-in O10C O fc.
re &- 01 re «t- u- > O4^re o re T3 t-i — xio)
tn 4-^4-1 *r- Ol 'i— 4-1 O f— OOOI* COre4-*3 reOI*
4-io> 4->ct3 o*»-re o oiin oirec -r-o
CT3T3--~reOl. 01 C i— 4-1 CT3O Ol >>i— C OIX:-
reScm^-r-o in3Q.e 0103-1- n. re o. 10
c^-re cu-o-o 3 ••- 1— 4-> 04->c ooiAl
,^re4-* o reN m4-»tntn i — in«r--o i — t-
cczT3» -otn O-in •(- tno •«- ins- 01 vireoi
r5 T- ro i- mre+J4- -i-cci- -1-C4-" m
4-1 -o xiS-'r- 01 co 4-* 3 •<- oi oiM-rere oivio
c >> re • oiu- t-tsrox: <*- 4-1 in j=o> in i— x:
OOlt-OIr— Ol OIC • CO COI4-"O- Cra4-"
oxzoit/^'r-Qtn 01 in ^— re • «reoro ox: 01 04-*
CM Tzt-in reoireins .m4-*re in o inEre
01 >in 4->i-4-'o»3 -r-T-rox: oic-i-oi cu x:
s_tns-x:cc 3Xooo> 34J re -r- in
omre-4->T-re CD o ID +-> 4->recoi o>t-o3 oioo
oio • x:c 4-> ore xroioioi j«:oi-i-o J£E-
Oil — 4-»OlEO'i— CCOIi-r— O14->O'C3 (OX:i — I — W lO
4-> re-oreE reoi> O-_CL •<- re c 3 4J _oi 0.1— 4^ t ^
m
o
c
re
Ol
o
•i- T
!—•
i^ *
O CM
Ol r—
re ^*
0- C
3 O
•r-
C 0
re oi
i— co
Ou
4-2
-------
C
O
o
CO
1
in
C
o
+J
IO
'i
•1—
1
in
O
•^
ro
o
•r-
u_
10
06
c
0
+J
i-
3
in
in
cn+j
o. o o
3 > ro
«-
O in
> ro ro
'•£ >>
ro +J XI
"o "Jo >
S— L. S_
•o ro
u- >
o o
-c >>
••- E E
i—OO
•r— S— v^
XI «- ro
ro +J
•r- in o.
S- C 3
ro O
> -r- 0
O ro ••-
x: o +J
+J -i- ro
u- c o
O-i- t-
c >+J
0 s- ro
•i- ro x:
ro -i-
3 E 0
10 i — ro
> 0
o S- +J
Q. S-
i— O
3 Q.
«- • o
0 -0 S-
i_ o
ro -0 O.
0 O O
0) J=
•a: c in
+j
c
o
+J
in
O> VI
c c
§0
o
IO
o i— in
in S- o
C -r- O>
010*0
O.U- 3
in i —
O in in
•&
O in c
.££ O. (O
ro O
O. 0 r—
+J 0
O C in
> o
+J o in
ro «- 0
i— U- S-
o -i- o
an -a ro
in <~s
o •
c c
ro o
01 o
O ^
c ^~.
•r- •
r—
U- •
O CM
O f—
ro «S-
0. C
3 O
+j +J
ro o
i— <<0
Q. —
u-
"~
O
j <
Q
O.
ro
C
ro
S
ro
+J
O
O
o •
o o
o <*—
O.*r~
in ^
O S-
O
0
3 >>
in i —
in 3
+J 4->
+J in
O 1-
O '^
j_ (—
o. in
J_ ^
> *r-
ro
ro O
+J
3 o o
S— X O
to o ro
Q£ -4-J OJ
r— f— O
in O) ro
x: o
O r~ in
+J *^~ in
+J ^ c
^ »fHf
V> +J
>r" 2 c
^ c .3
(O o 1 *
•4-J «r~ (Q
=3 (O +J
4J ." S
(0*0. C
<~~ O. O
o.
Q.
IO
E
10
.
oo
me
Olw. .i- Q.
•r- 3
O
O
CCJ.Q
°
o
o
u
03
o
O O)
C C -i-
O E
3 >>
in i—
o-^
>"~
0. "t-1
ro
re-
0 S»
VI'1-r—
e s-s
^ -So
°
fe
O.U-
in o u-
o Q. o
D>
C S- +J
ro O C
oi--i-
£ = >,
JCIOIO
C30
O Q.
•I- S-
f— O
>1
re 3
jg i
wo
o
3 >
"Si
-o o
o +J
>
g,s
-
O>XI
•o
r— ro
in E
in •
IO r-
O S-
'irS
O 3
O (O
+J O +J
O Lu O
IOX1C:
"II
>
i-
0
I/I
ro •
c
T3 O
O -i—
+J +J
ro ro
(0 H- o = -r- 01 •
r-TS-i- >OU1+J
-^ T3 0 -r- i— +J
|0>.p ET30W> 0
>«4- OCOO J=
o
0.
0
3>,
1.10
(0-^4-
.— -00.
XI
ca
>r--rO
O
o
o -^
°§
ro
O)
ro co
+J •
C CM
o •
O r—
ro +J
+J o
Q. O
S.
O
in
•o
m c: co
o ro •
iff-:
+J O ^1-
O 3 +J
••- J= O
X C O
o
-------
From this equation, a soil concentration of 1 yg/g DW corresponds to a
pollutant application rate of 2 kg/ha.*
Equation 4-1 is simplified in that is does not take into account back-
ground concentrations of the pollutant that may already be present, whether
natural or from other pollution sources. If it is assumed the total mass of
the soil-sludge mixture is equal to MS, then the amount of soil is MS-AR,
where AR is the sludge application rate (in t DW/ha). The reference appli-
cation rate, RP, must be reduced by an amount corresponding to the contami-
nant mass contributed by the soil:
RP = [(RLC x MS) - BS(MS-AR)] x 10~3 (4-2)
where BS = background concentration of pollutant in soil (yg/g DW). If AR
is very small compared with MS, as is often the case, AR may be dropped from
Equation 4-3. The equation then reduces to the following:
RP = (RLC-BS) x MS x 10 3 (4-3)
where:
RP = reference application rate of pollutant (kg/ha)
RLC = reference soil concentration of pollutant (yg/g DW)
BS = background concentration of pollutant in soil (yg/g DW)
MS = 2x103 t/ha = assumed mass of soil in upper 15 cm
10-a = conversion factor (kg/g)
For organic pollutants the background concentration in soil (BS) ordi-
narily is assumed to be zero, so that criteria for sludge application are
calculated independently of other contamination episodes. Equations 4-2 and
4-3 thus ordinarily reduce to Equation 4-1 for organics. However, Equation
4-2 or 4-3 may still be appropriate for organics on a site-specific basis,
if warranted by existing contamination.
*If soil incorporation depth were greater, i.e., 20 cm, an area/mass of
2.6x103 t/ha would be calculated. In this case, if RLC = 1 yg/g DW,
then RP = 2.6 kg/ha.
4-4
-------
If the contaminant persists indefinitely in the upper soil layer once
applied, then RP in the above equations represents a cumulative pollutant
application rate, which will be denoted as RP . if the pollutant is lost
over time, then the calculated RP gives the amount that could be applied in
a single sludge application, where no waiting or conversion period is
assumed. This rate will be denoted as RP^ To calculate a permissible
single-application rate that is to be followed by a waiting period or
land-use conversion period (RP^) or to calculate an annual application
rate (RPa), the loss rate from soils must also be considered.
4.1.1.2. CONTAMINANT LOSS FROM SOILS - Contaminants may be lost from
soils as a result of numerous processes, including leaching, volatilization,
and chemical and biological degradation. These processes may be occurring
simultaneously and at different rates.
Many of the inorganic contaminants of concern are not subject to vola-
tilization or degradation, and leaching is minimal because they are tightly
bound to soil. Particulates may be lost to runoff, but if sludge is soil-
incorporated, loss of some soil and sludge particles will not affect contam-
inant concentration in the remaining soil. Therefore, this methodology
usually assumes that inorganic contaminants are conserved indefinitely in
the upper layer. [Inorganics may be treated in the same manner as organics,
however, if a loss rate constant can be estimated.]
Organic contaminants, on the other hand, may be subject to all of these
loss processes that are extremely difficult to model in order to predict the
rate of loss. A simpler means to estimate loss is based on empirical data
from soil systems where soil concentrations have been followed over time.
These data may be used to estimate a first-order loss rate constant for the
pollutant. The use of such a rate constant is recognized to be an
4-5
-------
over-simplification, since the processes involved are complex and not
necessarily first order. Rate constants should be derived from field data
wherever possible, or may be estimated by analogy to other closely related
chemicals. Where no basis for an estimate is available, no loss should be
assumed.
First-order loss is represented by the following equation:
-kt
coe
(4-4)
where:
Ct = concentration at time t (yg/g)
C0 s concentration at time zero (yg/g)
e - base of natural logarithms, 2.718 (unitless)
k = loss rate constant (years-*)
t - time (years)
If BS is assumed to be zero, the following is true:
-kT
where:
RP<
RPS = RPsT
reference single-application rate, with no waiting
period (kg/ha)
(4-5)
RPsT s reference single-application rate followed by waiting
period (kg/ha)
k » rate constant for contaminant loss from soil (years-*)
T = waiting (or land-use conversion) period (years)
e « base of natural logarithms, 2.718 (unitless)
Thus, from Equations 4-1 and 4-5, an expression can be derived for calculat-
ing the permissible application rate of a pollutant subject to loss over
time when a one-time application will be followed by a waiting period:
RPsT = RLC x MS X 10
x e
kT
(4-6)
where:
RPsT ~ reference single-application rate followed by waiting
period (kg/ha)
L-K
-------
RLC = reference soil concentration of pollutant (yg/g DW)
MS = 2xl03 t/ha = assumed mass of soil in upper 15 cm
k = rate constant for contaminant loss from soil (years-1)
T = waiting (or land-use conversion) period (years)
e = base of natural logarithms, 2.718 (unitless)
Ordinarily, however, it is desired to determine an amount that can be
applied annually, RPa (1n kg/ha). To do so, it is first necessary to
solve Equation 4-2 for the reference soil concentration, RLC:
RLC =
RP x 103 + BS
MS
(4-7)
where:
RLC = reference soil concentration of pollutant (yg/q DW)
RP = reference application rate of pollutant (kg/ha)
IDs = conversion factor (g/kg)
BS = background concentration of pollutant in soil (vg/g DW)
MS = 2x103 t/ha - assumed mass of soil in upper 15 cm
AR = sludge application rate (t DW/ha)
When the first annual application, RPa, is applied, the background BS is
assumed to be zero. Therefore, the initial soil concentration (LC ) is as
follows:
LC0 =
RPa x IP3
MS
(4-8)
After 1 year's time, but before a second application, the soil concentration
(LC1,) is as follows:
. = LC
|RPa x 1Q3
MS
(4-9)
\ /
When the second application is made at year 1, the soil concentration
(L^) is determined from Equation 4-7, using the residual concentration
from the first application, LC],, as the value of BS:
= Rpa x 1Q3 + LCr (MS-ARa)
MS
4-7
-------
RPa x 103 + [(RPa x 103)/MS]e-k(MS-ARa)
MS
RPa x IP3
MS
(4-10)
Similarly, when applications are repeated annually for n years, the concen-
tration Immediately following the nth application (LCn) is as follows:
,c =
Rpa x 1Q3
MS
De-k + D2e-2k
where:
D - (MS-ARa)/MS
When the reference soil concentration, RLC, from the criteria calculation
procedures given later is used for LCn, then RPfl is the corresponding
reference application rate that can be applied annually. Solving for RPa,
RPa = RLC x MS x 10-3 C1
D2e-2k+
Dn-le(l-n)k]-l (4-12)
This equation can be further generalized to include the case where a waiting
or conversion period, T, follows the nth application:
RPa = RLC x MS x ID- ekT n + De-k + D2e-2k + . . . + on-led-")*] (4-13)
RPa = reference annual application rate of pollutant (kg/ha)
RLC = reference soil concentration of pollutant (yg/g DW)
MS = 2xlQa t/ha = assumed mass of soil in upper 15 cm
10-3 = conversion factor (kg/g)
e = base of natural logarithms, 2.718 (unitless)
k = loss rate constant (years-1)
T = waiting (or land-use conversion) period (years)
D = (MS-ARa)/MS
ARa = annual application rate (t DW/ha)
Thus, annual application rates for dissipating compounds may be calculated
as. a function of the limiting soil concentration, pollutant loss rate
constant, waiting period since last application and sludge application rate.
4-8
-------
As noted previously, in most cases ARa is very small compared with MS, and
0 reduces to 1 and may be dropped from Equation 4-13.
4.1.1.3. INPUT PARAMETER REQUIREMENTS FOR CALCULATING APPLICATION
RATES — Several input parameters may be required for calculating pollutant
application rates from the reference soil concentration. The soil mass, MS,
and loss rate constant, k, have been discussed previously. The other
parameters are discussed in the following text.
4.1.1.3.1. Waiting Period or Land-Use Conversion Period (T) — As
mentioned in Section 2.2.1.2., waiting periods following sludge application
may be evaluated as a useful means of reducing risk from certain activities,
such as grazing of cattle or planting of crops. Suggested values for T in
these instances will be given as part of the discussion of those specific
exposure pathways rather than here. As also mentioned in Chapters 2 and 3
(see Table 3-1), certain exposure pathways may not be relevant until
land-use conversion occurs. It may be assumed that some period of time, T,
is required for such conversion to occur. This report will not attempt to
specify the time periods that could be associated with various types of
conversion, however.
4.1.1.3.2. Number of Annual Sludge Applications (n) — Since there is
no established limit to the number of sludge applications that may occur in
most utilization practices, an infinite number should ordinarily be assumed.
In practical terms, if n = 5.6/k then the final term, e(1~n)/k, in
Equation 4-13 will be <0.01, and the result of further increasing n will be
negligible. This approach will be used here. In certain utilization
practices, however, especially where site-specific information can be used,
it may be appropriate to assume a more finite value of n. For example, land
reclamation may involve only 1 or a few years of sludge application.
4-9
-------
4.1.1.3.3. Sludge Application Rate (AR) — A cumulative rate, ARC
(in t DW/ha), is used in Equation 4-2, and an annual rate, ARg (in t
DW/ha), is used in Equations 4-12 and 4-13. Since large values of AR
slightly increase RP, low values should ordinarily be assumed for a protec-
tive approach. AR typically is ~5 t DW/ha for many agricultural uses.
9
AR can be much higher, since it is assumed that sludge applications may
be repeated indefinitely, but if only a single application is made,
AR = AR . Therefore, AR should ordinarily be considered zero for
c a
purposes of criteria calculation.
4.1.1.3.4. Background Soil Concentration of Pollutant (BS) —As
stated previously, BS (in yg/g DW) may be the natural background
concentration or may result from other pollution sources. In national
criteria derivation, it is suggested that mean values, such as those arising
from natural background or very widespread pollution (such as lead), be
used. Locally elevated concentrations, either from unusual natural
background levels or localized pollution sources, should be dealt with on a
site-specific basis. Values of BS should be derived from national or
regional surveys of soil concentration. For metals, pollutant additions in
sludge are considered on the basis of total metal, rather than extractable
metal. Therefore, soil analyses should also be of total metal. Extractable
metal is that which can be extracted by dilute acid (such as 1 N HNOg) or
a chelating agent (such as EDTA), whereas total metal is approximated
following a complete digestion of the sample with concentrated acid (such as
nitric-perchloric acid digestion).
4.1.2. Contaminant Uptake Relationships.
4.1.2.1. PLANT UPTAKE OF CATEGORY 1 CONTAMINANTS — Rates of uptake
of inorganic chemicals, especially metals, by plants grown in sludge-amended
4-10
-------
soils have been intensively studied. Recent reviews of plant-metal uptake
relationships include those by CAST (1980), Ryan et al. (1982), Logan and
Chaney (1983) and U.S. EPA (1987a). In an effort to relate the response of
total dietary cadmium (Cd) to sludge-applied Cd for the purpose of assessing
risk, Ryan et al. (1982) used linear regression (of plant tissue Cd
concentration against applied Cd) to derive uptake response slopes for
various crops.* The slope of the crop response curve to added soil Cd was
different for annual and cumulative Cd additions. The annual response curve
represented the more responsive situation; therefore, they felt it was
judicious to use this curve and thus err on the conservative side (i.e.,
overestimate human exposure). Presently from our understanding of long-term
metal availability from land application of sewage sludge (U.S. EPA, 1987a),
it is known that the first-year response curve [annual response curve from
Ryan et al. (1982)] generated by a large, single sludge addition will
overestimate long-term metal accumulation in vegetative tissue and that
response curves generated 4 or more years following sludge application are
more appropriate models. At any rate, the underlying assumption is that for
each incremental increase in soil metal there is a constant linear increase
in plant tissue concentration. By this approach a cumulative metal
application limit can be reached above which no further additions of sludge
containing metal would be allowed.
docume.nt, the term "uptake response" (or simply "response") is used
V, tissue concentration in response to' exposure tla
P sure
4-11
-------
More recently, however, others have argued that Cd response is curvi-
linear and approaches a plateau, the level of which is not dependent on
cumulative application rate but rather on the metal activity in the
soil/sludge solution. This argument, as summarized by Logan and Chaney
(1983) and U.S. EPA (1987a), is based on the premise that as the total
amount of sludge added to soil increases, Cd availability becomes controlled
by the adsorptive characteristics of the sludge rather than of the soil.
Since with increasing sludge addition the soil Cd increases, crop response
increases with application rate at low application rates where the soil's
adsorptive capacity controls metal activity. However, if the metal
adsorptive capacity of the sludge is high compared with the soil (usually
assocated with high sludge rates), the soil adsorption sites that can be
filled at the activity supported by the sludge will result in only a small
decrease in solution activity, and the sludge adsorption properties will
control solution activity and thus plant uptake. At some point, where the
sludge adsorption capacity is controlling metal solution activity, plant
concentration will reach a maximum (plateau) and further addition of sludge
will not change plant tissue concentration. The sludge properties that are
thought to control its adsorption capacity include Cd concentration, Fe
concentration, Al concentration, P concentration and pH (U.S. EPA, 1987a).
A thorough critical review and acceptance of this hypothesis by the
scientific community and development of the data base will be necessary
before it can become a basis for regulatory criteria. However, in an
attempt to reflect the current state of knowledge, the present methodology
will tentatively accept this hypothesis.
A comparison of the response curves generated by the linear and
curvilinear models are illustrated in Figure 4-1. In the case of the
4-12
-------
Crop Metal
Concentration
(ug/g)
A - linear response
B - curvilinear response
Cumulative Sludge Application Rate (t/ha)
FIGURE 4-1
Curvilinear Uptake of Sludge-borne Metal by Crops
4-13
-------
curvilinear model the plant tissue concentration reaches a maximum value P
(plateau), whereas the linear model assumes that no maximum value is
reached. At low loading rates both models predict the same plant tissue
concentration, whereas at higher loading rates the linear model predicts
higher plant tissue concentration. If the acceptable plant tissue
concentration were greater than P, the curvilinear model would allow for an
infinite amount of sludge to be applied, whereas the linear model would
allow for a finite amount of sludge to be applied.
An additional complicating factor is that crop response during the first
year of sludge application is higher than that observed several years after
the first sludge application (U.S. EPA, 1987a). This is illustrated in
Figure 4-2 for both the linear and curvilinear model. The implication of
this is that for long-term consideration, response curves generated several
years after an initial sludge addition, or after multiple years of sludge
addition, would be more applicable.
Both the linear model and the plateau model can be useful for assessing
effects of sludge applications. Linear response slopes are easily derived
and usually available for various crops, thereby providing a means for
comparing relative responsiveness among crops. Plateau data, on the other
hand, may be most appropriate for estimating long-term impacts of cumulative
sludge addition. However, where insufficient data exist to determine
plateau levels in sensitive crops, linear response slopes provide a highly
conservative means of assessing risks from crop uptake of metals. Guidance
for determining linear response slopes and plateaus is given below.
Specific use of these values to derive criteria for various exposure
pathways is discussed in later chapters.
4-14
-------
Crop Metal
Concentration
(WJ/9)
A - linear response
B - curvilinear response
Cumulative Sludge Application Rate (t/ha)
FIGURE 4-2
First-Year Vs. Multi-Year Observations
a = Crops grown in the first year of sludge application
b = Crops grown 1 or more years following sludge application, or where appli
cations have been repeated annually
4-15
-------
The report (U.S. EPA, 1987a) also advances the hypothesis that relative
response among crops is fairly consistent across soils and sludges, at least
for similar soil pH. In other words, if the linear response slope of a
metal in crop A is 5 times higher than that in crop B in a well-conducted
experiment using a given soil and sludge, then it will also be 5 times as
high when a different sludge or soil (of a similar pH) is used, even though
absolute response may differ substantially between the two studies.
Therefore, the response of any crop theoretically can be expressed in terms
of any other crop, as long as the two have been studied in the same valid
experiment. The response of all crops could thus be expressed in terms of a
single, frequently studied (and relatively responsive) index crop such as
lettuce.
The hypotheses regarding the uptake response plateau and relative uptake
response relationships remain to be rigorously validated for each chemical
that may require criteria derivation. Therefore, criteria calculations for
specific chemicals should include discussions of pertinent uptake data
verifying that these hypotheses hold in each case.
4.1.2.1.1. Determination of Linear Uptake Response Slopes — Linear
response slopes can be calculated from any data set where tissue analyses
and cumulative metal application rates have been recorded. Tissue analyses
from various application rates in a single year (whether or not previous
sludge additions have been made) are generally more easily interpreted
because year-to-year variability of conditions is eliminated. Data from the
first year of sludge application will generally result in higher slopes than
those from later years, as mentioned above. Multi-year slopes may be more
representative of typical sludge application practices.
4-16
-------
Data derived from sludge applications in the field are most appropriate
for use in risk assessments. Greenhouse studies where plants are grown in
pots are known to often overpredict uptake under field conditions (Logan and
Chaney, 1983). However, in the absence of field data, data from pot studies
may be useful, especially where large pots are used to minimize restriction
of root growth. Studies using metal salts, added either to soil or a
sludge-soil mixture, also tend to overpredict field response. Therefore,
metallic salt uptake data should not be used for risk assessment if sludge
data are available.
Studies where plants are grown in solution culture also should not be
used, as there is no reliable way to relate concentration in solution to
total soil concentration in the field or application rate. Studies where
sludge has been applied over growing plants demonstrate physical adherence
rather than physiological uptake and therefore should not be used. Their
use in assessing effects from contaminated animal forage is discussed later
(see Sections 4.4. and 4.5.).
Uptake response slopes are calculated by regressing plant tissue contam-
inant concentration (in pg/g OW) against cumulative contaminant applica-
tion rate (in kg/ha) for the various treatment levels, including the
control. Note that where the control application rate is zero, the tissue
concentration is greater than zero because of background occurrence of these
elements. Where data from a pot study are used, soil concentrations must
first be converted to application rates using the relationship discussed
previously (see Section 4.1.1.1.).
Some additional considerations may be important when linear response
slopes are calculated. If a data set shows a curvilinear tendency, linear
regression may underestimate the slope at low application rates (and
overestimate the slope at high rates). Since many studies use very high
4-17
-------
application rates to demonstrate an effect, and rates of concern may be In
the lower part of the curve, it may be appropriate to recalculate the slope
after dropping some of the higher points.
Conversely, a linear slope is sometimes used to estimate tissue concen-
trations at application rates well beyond the range of the available data.
If the contaminant is phytotoxic, this extrapolation could result in
predicting very high tissue concentrations that in actuality would have
resulted in death of the plant (and therefore cause zero risk to higher
trophic levels). A straightforward means of addressing this problem is to
determine a maximum tissue concentration for a given crop, based on avail-
able phytotoxicity data, and to assume this value as an upper limit to
uptake. The plant tissue concentration limit (TL) is not that which occurs
at a phytotoxic threshold (that is, the level at or above which symptoms of
phytotoxicity, such as a modest yield reduction or foliar discoloration,
first appear), since threshold effects do not necessarily preclude contami-
nant passage up the food chain. Ratherp maximum concentrations are those
associated with severe yield reduction (>75%) or death of the plant. When a
maximum tissue concentration (TL) is imposed as a ceiling, the resulting
assumed uptake relationship is as shown in Figure 4-3. Plant tissue
concentration at which phytotoxicity occurs is dependent upon a number of
soil and sludge factors and can be expected to vary over a wide range.
Therefore, the value used in criteria calculations for specific chemicals
should include a discussion of pertinent data.
4.1.2.1.2. Determination of the Uptake Response Plateau— If uptake
response is assessed on the basis of a plateau, the tissue concentration at
which a plateau occurs must be estimated for a given study. The shape or
4-18
-------
TL
Crop
Metal
Concentration
Cumulative Contaminant Application Rate (kg/ha)
FIGURE 4-3
Limitation by Phytotoxicity of Linear Uptake Response*
*For explanation, see text.
4-19
-------
slope of the lower portion of the curve is not important. Nonlinear regres-
sion techniques for fitting a curve to the data and determining and placing
confidence limits around the asymptote should be used. If an upper confi-
dence limit of the asymptote is used as the plateau, the number of points
and goodness-of-fit of the data will help determine how low a plateau can be
assumed. This would provide a built-in measure of safety when using less-
than-ideal data and would be consistent with procedures currently used in
the Agency for estimating cancer potency, that is, the use of an upper bound
value rather than the maximum-likelihood estimate of the parameter sought.
However, the maximum-likelihood estimate would also be useful, especially in
illustrating the variability associated with this parameter.
Nonlinear regression may be carried out with the same type of data as
used for linear regression, that is, tissue contaminant concentrations (in
pg/g DW) vs. cumulative contaminant applications (in kg/ha DW).
4.1.2.2. PLANT UPTAKE OF OR6ANICS — Linear uptake response is
assumed for organic chemicals and is calculated as described for inorganics,
but with some important differences. Because organic compounds tend to
degrade in soil, plant tissue concentration is usually expressed as a
function of a measured soil concentration, rather than application rate, in
most available studies. Therefore, the soil concentration (in yg/g dry
weight) is used as the x axis to determine the slope.
In addition, because soil concentration rather than application rate is
used, and because most of the compounds of concern are xenobiotics, tissue
concentration can be assumed to be zero when soil concentration is zero.
Therefore, the slope reduces to a bioconcentration factor (BCF) that can be
derived from a single data pair, as is commonly done to derive BCFs for
aquatic organisms. Few studies have quantified uptake of organic compounds
from land-applied sludge. Estimations of response slope may need to rely on
4-20
-------
other data, such as from soil-applied pesticide studies where plant uptake
through roots occurred.
4.1.2.3. CONTAMINANT UPTAKE BY ANIMAL TISSUES — Linear response
slopes are derived for uptake of inorganics or organics by animal tissues
consumed by humans. Tissue concentration is regressed against concentration
in feed. Tissue concentrations in the literature may be expressed in dry or
wet weight, but dry weight is preferred. For uniformity in applying this
methodology, all slopes should be derived based on dry-weight (moisture-
free, but including fat) concentrations in tissue and feed. Conversion from
wet to dry weight for various tissues should be made according to percent-
moisture values given in USDA (1975).
For lipophilic organics, tissue concentration is often expressed on a
fat basis (ng/g fat). If so, the slope should also be expressed on a fat
basis rather than converted to dry-weight basis. Also, the slope for
organics may be the same as a BCF derived from a single data point (that is,
animal tissue concentration * feed concentration), as described previously
for plant uptake of organics (Section 4.1.2.2.).
The best studies for deriving uptake information are those in which
sludge or sludge-grown crops are used as part or all of the diet for animals
(such as cattle, sheep, swine and poultry) typically consumed by humans.
Studies in which the diet has been amended with pure chemicals should be
used only in the absence of these data, since the added forms often are more
bioavailable. Tissues analyzed may include muscle, kidney, liver and
various other organs less frequently consumed by humans.
4.1.3. Toxicity Thresholds for Nonhuman Organisms. Detailed methods have
been developed by the U.S. EPA for determining toxicant exposure levels that
should not be exceeded in humans (Federal Register, 1980). When sludge is
applied to land, other organisms, such as soil biota and their predators,
4-21
-------
plants and grazing animals should be protected as well as humans. However,
specific methods for selecting threshold levels for protecting these diverse
groups have not been articulated. It may be difficult to determine what
studies are most appropriate, what effects are of concern, and how to select
protective values based on the available information.
The U.S. EPA Guidelines for Deriving Numerical National Water Quality
Criteria for the Protection of Aquatic Organisms and Their Uses (U.S. EPA,
1984c) comprehensively address these issues with regard to aquatic organ-
isms. These Guidelines list acceptable types of toxicity tests and estab-
lish a minimum data base for the number of species required to have been
tested. The threshold, or "chronic value," is defined as the geometric mean
of the lowest exposure level causing, and the highest level not causing, a
statistically significant, adverse effect in a given species. The final
value is then calculated as the level that protects 95% of the tested genera
of aquatic species. If a commercially or recreationally important species
requires greater protection, the final value is lowered accordingly.
It would be desirable to use similar procedures for protecting terres-
trial organisms; however, sufficient information may not be available for a
comparably systematic approach. The following general guidelines are
suggested for determining toxicity thresholds in plants, invertebrates and
vertebrates:
1. Long-term inhibitory effects should be considered adverse
unless evidence to the contrary is available. These effects
usually include reductions in growth, fecundity, lifespan or
performance, as well as symptomatic manifestations of
toxicity. However, temporary reductions in soil microbial
activity or diversity should not be considered adverse unless
there are demonstrated long-term effects on ecosystem
parameters. Where effects cannot be attributed to one
chemical, such as studies where exposure is to sludge,
thresholds ordinarily cannot be determined.
4-22
-------
2. The geometric mean of exposure levels bracketing an adverse
effect should be used as the threshold. For example, if expo-
sure levels are 1, 10 and 100, and effects are significant at
100, a value of -30 (~[10xlOO]l/2) should be used. Where
effects occur at the lowest exposure level, and other studies
better defining the threshold value are unavailable, no thresh-
old value can be determined. Where results were not statisti-
cally tested, careful judgment should be used to determine the
meaningfulness of a given change.
3. The form of a contaminant used in a study should not be con-
sidered equivalent in bioavailability to the form present in
sludge or migrating from sludge unless no better data are
available. Studies with sludge or sludge-grown crops usually
cannot be used to establish a toxicity threshold because of the
presence of other pollutants. Therefore, it may be necessary
to rely on results of studies using pure forms of the contami-
nant. If, however, a study using sludge, sludge-grown crops or
some other appropriate vector demonstrates lower bioavailabil-
ity in a realistic exposure situation, such information should
be taken into account when establishing the threshold. How-
ever, a study showing little response to a sludge-borne chemi-
cal in species A cannot be taken to indicate that a more sensi-
tive response to the pure form in species B should be dis-
regarded.
4. Where studies with few species are available for a given chemi-
cal, the results with the most sensitive species should be used
to determine the threshold. Where eight or more diverse
species have been studied, procedures for estimating the fifth --—
percentile of response may be used, as described in the Aquatic
Life Guidelines (U.S. EPA, 1984c), as long as commercially
important species are protected.
5. Where several tests have been conducted for a given species,
results appearing to be outliers should be disregarded.
These general guidelines should be applied together with careful scien-
tific judgment to derive toxicity threshold values for various types of
organisms.
4.1.4. Human Diet. Humans may be exposed to sludge-borne contaminants in
crops or animal products that have taken up the contaminants by the soil or
diet, respectively. To quantify potential dietary exposures, it is
necessary to estimate the amounts of various types of foods in the human
diet. The most up-to-date and detailed source of information regarding food
4-23
-------
consumption habits of the U.S. population is the FDA Revised Total Diet
Study food list (Pennington, 1983). This list is based on combined results
of the USDA Nationwide Food Consumption Survey (1977-1978) and the Second
National Health and Nutrition Examination Survey (1976-1980). The list
provides average, fresh-weight consumption data for over 200 foods (201
adult foods; 33 infant/junior foods) for eight age/sex groups ranging from
Infancy to 60-65 years of age.
While the Pennington (1983) food list provides a very detailed picture
of the human diet, it cannot be used in its published form for risk assess-
ments of the present type. Many of the food items listed are complex pre-
pared foods (such as soup, pizza), rather then the raw commodities (such as
crops, meats) for which contaminant uptake data are available. Therefore,
to predict the impact of sludge application using uptake data, it is
necessary to reorganize the diet to determine the respective consumed
amounts of these raw commodities.
Two previous efforts have been made to reorganize the Pennington (1983)
diet. In 1981, the U.S. EPA Office of Solid Waste proposed an approach that
grouped the 201 adult foods into the 12 dietary categories used in the
previous FDA Total Diet Study food list, for the purpose of estimating the
amount of cadmium in the typical U.S. diet (Flynn, 1981). This was
necessary because metal analyses for foods on the revised food list were not
then available, and still are not. These twelve categories include several
to which uptake data may be applied (such as grains'and cereals, potatoes,
leafy vegetables) and therefore can be used to estimate the impacts of
sludge application on contaminant amounts in the diet. However, the indi-
vidual foods were not broken down according to their contents; for example,
beef and vegetable stew was listed in the "meat, fish and poultry" group.
4-24
-------
In addition, some of the listed items consist largely of added water, such
as canned, reconstituted bouillon (also listed under "meat, fish and
poultry"). Therefore, the resulting consumption values for each still did
not reflect the raw commodities.
A second approach was presented in the draft Air Quality Criteria
Document for Lead (U.S. EPA, 1984b). Here many of the individual foods from
the Pennington (1983) diet were fractionally apportioned into different food
groups. For example, the food item "pancakes" was apportioned as follows:
60% food crops, 10% dairy and 30% meat, representing the contribution from
grains, milk and eggs, respectively. However, the number of food groups
employed was too few for use with the present methodology; that is, all
crops were lumped into a single category. In addition, the apportionments
were made not on the basis of weight of each ingredient as desired for this
analysis, but on the basis of the amount of lead in each ingredient.
Therefore, a new analysis of the Pennington (1983) diet was required for
this methodology. Each item in the Pennington diet (including the infant/
junior foods) was broken down into its components based on information
available in FDA (1982) and USDA (1975). The percentages of dry matter and
fat for each component were also listed. These components were then aggre-
gated into the specific commodity groups required for this methodology. A
summary of consumption for each category by each age/sex group is presented
in Table 4-2. Data for the entire analysis are compiled in Appendix 1.
The consumption values from the Pennington (1983) diet, and therefore
from Table 4-2 as well, are average values based on 24-hour recall by sur-
veyed individuals. Additional statistics indicating the range of consump-
tion values for the populations studied would also be useful for indicating
ranges of exposure; this information was not given, however. Yost and Miles
(1980) reanalyzed data from a 1965-1966 USDA survey of food consumption and
4-25
-------
t
M
•_!
1
u.
•M
fl)
0
ro
2
in
•r-
^
0)
cc
«s
u_
01
JC
u.
o
tn
in
rO
"—
ro
CM
1 C
•a- o
ui -o
_j a,
CO V)
1— CO
in
3
s
CO
^
O
tl.
If
o
i
»r-
-M
VI
C
o
+J
JS
at
01
^^
?>
o
>)
r—
•*~
n>
^^
OI
CD
ro
i_
o>
<«
ro
•o
•£
o
•r—
>
I.
•o
D
C
1
O
X
a
t
ai
o
>
.c
c
c
1
V
c
o
u
in
1-
ra
a>
in
IO
I
o
to
in
u
ra
0)
5-
O
CO
1
in
CM
in
ra
at
«3-
O.*»-
3 C
O =
t- C
0 L.
O
*a i*"
o =
0 I/
Uu «*•
I/I
(U
i
I/I
0
2
u.
in
0)
i
in
0)
i
ai
u.
in
a>
r—
a>
u.
I/I
i-
ro
0)
CM
in
.c
I
r;;
T
to
1
^
o o o ?— co incMininr*como cooir— co«*o«*iflr— CM in in «* yD en
r^cnuaoita CMenmOOCT>OlO CMP^CM^TCOi— OOi— r— OOr- OiO ^ GO
pgJggJS S-cv^-r-cdr^piua u3cn«3-CT>r-»Kt*r-->C3 r»r— CNJr— C*3WO«1 cnr-COCOf— CO^OlOimd 0000 m CM
ro rn CM S iri cocooai— mcomo cymio^Of— i— intnr»p-CM ma» 01 ui
oacooooco or*r-inmcoor-> ^-r».oicNjrMCMinaieMCMio CQCJ r- ua
tncy^r— O CNJCVlOOi— ^-r— OCO ^-OOOOOP-dtfir— CO OlCNJ P- GO
^j. ^_ p_ CNJ^-^- ^-r—r*
t3 r*-tOMcomc\j oito^coocor— ^•encMP- r^io p*> CO
S J^ iG ^^ ^Q S- J\j c\I Si i/i ift ID CM co oo i~ I"* ^3 cyi r^ c\j n 10 o^ ^f ID ^J* ^t*
CD CO 00 tn CO ^-inUlCXJOr— COO Otr-CNJWSlOiO^'UICOeOiO «tOO CO CO
SS™So cor-r-o«3-cOf-r- r-oia»cMCMCM^-cvr-cx3UJ m^ o m
m r- 10 •— ^t r-cxji— CMincoo^f e7it£)O*C3OOcoeyicni— «*• CMCM o in
}iicv]WCMCM^ ^ CMCM r- r- COCMr-O
Sr=S§S gSSKSSSr? gSSSrtSSSKr?J2 ^^ S S
r^ptlScniO Spip-r--OCOCOO p-CMCOCOCO«*OOOin«*O 0«J in 00
030OP-OOO CMr— OOini— ini— P- COOOCMCOCOCOOP-r— CMin mi— IO 00
CMCO^-f— CM COCMP-r— <3-r— OCn r-COr-OOOP-CniOr-CO ^jJfZ ^ §
a, cs o inOiococMOOOen cMiococooiini — CM«d-o^ coco m en
r-coco§S o P- n- en co CM CM i- p. 10 10 en co in o «a- r- p- to CM CD in 10
r^S-cop-to coair-coiOo oo o * to i— r- in •— o •» CM
CMeniccncM cM«s-OCM«3-CMinr- r-entoinr-«a-cnoop-coen i— to p- to
SS-coCTio p-coenr-p-CDini— tocor— p-toioor— CMCMCO i— o en r—
S5r-0° en r- CO CO 0 r- 0 CM 10 in CM 0 0 0 •>* CM CO CM in OP- O r-
r- co in r- o mi— tor- coi— oo tocvjoooor— oioi— CM «*oo us in
V0r- r- COr-r- r— COr- 00
_*.. inrui in«d-r-ooocMr— P- i — *d"i — toooinp-ococnin to^d- in 00
S 5 ?)• O r-- coinp-p-encDCMtO cMcnoinioenoOP-i— o in«3- co CO
coinooinr— cooomtotoOr-oo *or~^1"*OP;Cai<:r'l?2o. 22 ~ i£
CMCoinioo o^fmtotOCMOiO UB^-CMOOOCMI—P-COCM en * en p-
CMtn^-CMO OO^Oi — CMOP- cntOOOOO"*OOCOOi — CMtO tO P-
•*r- r- r- <~
e^tnrnooto coo^l-oocoi— on oooiotncococMcncop-r— OP- en CM
CMCoS-eno p-eo en o in to 0 0 en «3- to in en «*• oo in en in oo P- to o to
ocncMCMr— tocoo^tocoocM oO'S-td-s-co-a-coOtOcnco "J?00 IT: 2
tOOlCMP-O lOOOOOOtOCOOlO O>«S-r—Or-l— COOCMOCO tOOn CM 00
r~ co CM co o m O co co O O O r- co CM o o O O i— CM CM r- o O oo co oo
CM CM T n I—
', «'
u-
£ VI CO
r— in ro *r"
™ 7;-° " « "~ •!->?
a> c jaror- in •— ^^ ior^
U *1B +-»0)ro'r- O O)0) t*-O 4-»
^3 t_ Q) en +^ 3 ^™* ^ ^ ^* ^* ^ ™
c: en oia>O)S- via) ro-r-'i- re a >> >> -r- «_
TO lflO)>Dltt-^er^ u_r-r-ll-«-j_i..^-. _^
viaca>via> o a>>c+Joro ""'J'S'aj'g'Eif'Jf'a'a'S >,.r^ i-
^ ^ o »r- ro -4-* ro *i 3 +•* *o c r- aj +ja)ajoja)rorooooo "r" s- ro in o>
gsuuoo •gSS'SroSSS' S^^-a-a--13-0-0-0-"- '5^ S |
5 eu _i _J ee co a. z > z a ui 0
CM
oo
en
r—
„£
a
u-
•C3
C
ro
CO
00
o>
'
•.
o
+J
D)
•r^
C
a>
a.
c
o
Q)
ro
00
a>
Sf
3
o
00
4-26
-------
presented cumulative-percent distributions of consumption for six food
groups. These data, shown in Table 4-3, were also from a survey of 24-hour
consumption recall. The data show that for each of six food groups, con-
sumption at the 95th percentile exceeds the mean by a factor ranging from
3.6 (potatoes) to 7.5 (root vegetables).
These factors indicate interindividual variability on a given sampling
day. To estimate the interindividual variability in long-term average
intake, intraindividual variability over several sampling days must be
known. Sempos et al. (1985) have shown that the ratio of intraindividual to
interindividual variability (as expressed by the ratio of the variances) is
>1 for a variety of dietary parameters. For example, the ratios for four
elements (Ca, Fe, Mg, Zn) were 1.2, 2.6, 1.3 and 2.4, respectively. This
would seem to indicate that the variability associated with 1-day consump-
tion is greater than that for long-term consumption. Therefore, the 95th
and 99th percentile values given in Table 4-3 overestimate those values for
long-term consumption. Another limitation of these higher percentile values
is that they cannot simply be summed to predict total dietary response,
since no single individual would be likely to consume all affected food
items in 95th- or 99th-percentile amounts. Matrix methods, or stochastic
methods such as Monte Carlo analysis, would be required for combining these
values to estimate total dietary intakes representative of the 95th or 99th
percentile (Yost et al., 1980).
Vegetarians have been viewed as a group possibly at higher risk from
land application practices. Ryan et al. (1982) showed that under certain
assessment scenarios, cadmium intake of lacto-ovo-vegetarians could be about
50% higher than that of the population as a whole. This was determined by
comparing the Loma Linda lacto-ovo-vegetarian (LOV) diet (Loma Linda Univer-
sity, 1978) with the 1974 FOA Total Diet (after adjustment of the latter to
4-27
-------
re
|MM
|» >
re
o
CO
^
s_
3
o
3=
^*
CM
cn
•i™
in
n
>>
0)
j_
3
CO
re
lj
£
<*-
in
ex
3
~f_
VpJ
•o
0
u.
X
«p«
VJ
U*
0
C
Q
•P
Q.
€
—y
o
o
>>
j —
re
a
<4-
o
c
0
J_l
•—I
•r™"
s_
^
o
^««
ifj
x:
cr
«
2~
o
%••»
C
O
•r-
C
|
u
c
c
o
•P
re
a
o
a.
£
•i™ •
•p
c
UJ
in
re
•o
Q
•
s_
re
CO
>
o
CM
C
^^
Id
j-J
^^
en
en
XJ
in
t>
^
^0
O
in
c
re
0)
X
XI
en
X)
en
u
X
o
in
c
re
CO
a.
3
o
CD
•o
o
o
u.
o in in in o m
4- CM CM co o en
**t* ^$* m if* 10 10
r~
O m o o o o
«s- «3- in o o >d-
CM CM CM * * «*
o o o m o o
eo r~-
V
M CO i— CM UO vO
CM * r~ i— 00 r—
in «* co O O CM
i— r— r—
O m o in o in
^- O CM CO O CO
«f in co o en en
r— r—
O O O O O O
CM CM CM in in in
o o o m o i—
co en
r- ^ O CO iO •*
in 10 CM r— •* O
in in co m co CO
^— J3 re r— m co
re -H x» •)-» o
CO CD -H 3 T3
CD co a> i_ e
co > cn in <4- re
> CO CO
eo > o c oo
> E 4-> CO C
»*- 3 4J re -o •>-
re CD o 4-> v re
co co o o re L.
_j -j oe cu, ca cfl
E
3
I/I
O
u
o
c
•o
•n*
T3
2 in
cn ro
T 3
in 1^
en .^1
"" -o
_ c
C ,(—
•r~
•° "S
s >>
o "
3 ^
^3 *"
1 3
0 O "*
>> -p _ •
r— I/I C i/i
re T- aj s-
c x: x: 3
re 3 o
i- 0 O
re co co CM
3 N
c cr CD
o co a> <=
t3 <*- CO "O
co •* co
«n E o
re p „ co
x> i- a> h
.. r_
co CD 5 x:
en -p -P
i— re _^s
in '-p 'Z "~
co m I, O.
r- eo •£. 3
•E . <" S-
"o co
C 3 g -P
re r— °- re
re x:
^™* > XI 4^
O CO O E
*" ^ ^ °
.. .p u-
co c c
o co re m
s- o -i- E
3 i- -a a>
O CO CO -P
CO CU SE «r-
re x> u
4-28
-------
a comparable caloric basis). Consumption data from the LOV diet (as report-
ed by Ryan et al., 1982) are presented in Table 4-4. Comparison of the LOV
data with fresh-weight consumption data for 25- to. 30-year-old males from
Appendix 1 shows that LOV consumption is substantially higher for several
vegetable crop categories and lower for potatoes.
The Pennington (1983) data base is more adequate for most features of
this analysis. Therefore, the Pennington data, as reanalyzed here (Table
4-2), will be employed in the following criteria derivation procedures. If
desired, the results may be adjusted to reflect the LOV diet by adjusting
the dry-weight consumption values for these categories by the ratio of the
fresh-weight values shown in Table 4-4.
4.1.5. Health Effects in Humans. An adjusted reference intake (RIA, in
yg/day) will be defined as the increase in dietary intake of a contaminant
that is used to evaluate the potential for adverse effects on human health
as a result of land application of sewage sludge. That is, given the
practice definitions and assumptions stated previously in this methodology,
the criterion for a given sludge contaminant is that concentration in the
sludge, or that application rate to land, which is calculated to result in
dietary intakes not exceeding the RIA in exposed individuals. To exceed the
RIA would be a basis for concern that adverse health effects may occur in
those individuals.
The RIA is termed "adjusted" because it is a health-based reference
intake value that has been adjusted from a per-weight basis to a particular
human body weight and also to account for contaminant intake from other
sources.
The procedure for determining RIA varies according to whether the
pollutant acts by a threshold or nonthreshold mechanism of toxicity.
4-29
-------
TABLE 4-4
Food Consumption of Lacto-Ovo-Vegetariansa
and Average 25- to 30-Year-Old Males (selected va1ues)b
Food Group
Consumption (g/dav fresh weight)
Lacto-Ovo-Vegetarians
25- to 30-Year-Old Males
Dairy products
Grains and cereals
Potatoes
Leafy vegetables,
root vegetables and
garden fruits
Legume vegetables
Fruits
Oily fats, shortening
Sugars and adjuncts
Beverages
Meat, fish and poultry
584
203
43
252
166
284
107
110
600
0
193
68
107
50
aSource: Loma Linda University, 1978
^Source: Pennington, 1983, as reanalyzed in Appendix 1 (see Table Al-2)
4-30
-------
4.1.5.1. THRESHOLD-ACTING TOXICANTS — Threshold effects are those
for which a safe (subthreshold) level of toxicant exposure can be estimated.
For these toxicants, RIA is derived as follows:
RIA = [(RfD x bw/RE) - TBI] x 103 (4-14)
where:
RIA = adjusted reference intake (yg/day)
RfD = reference dose (mg/kg/day)
bw = human body weight (kg)
TBI = total background intake rate of pollutant (mg/day) from
all other sources of exposure
RE = relative effectiveness of ingestion exposure (unitless)
103 = conversion factor (pg/mg)
The definition and derivation of each of the parameters used to estimate RIA
for threshold-acting toxicants are further discussed in the following
sections.
4.1.5.1.1. Reference Dose (RfD) -- When toxicant exposure is by
ingestion, the threshold assumption has traditionally been used to establish
an "acceptable daily intake," or ADI. The Food and Agricultural Organiza-
tion and the World Health Organization have defined ADI as "the daily intake
of a chemical which, during an entire lifetime, appears to be without
appreciable risk on the basis of all the known facts at the time. It is
expressed in milligrams of the chemical per kilogram of body weight (mg/kg)"
(Lu, 1983). Procedures for estimating the ADI from various types of toxico-
logical data were outlined by the U.S. EPA in 1980 (Federal Register, 1980).
More recently the Agency has preferred the use of a new term, the "reference
dose," or RfD, to avoid the connotation of acceptability, which is often
controversial.
4-31
-------
Values of RfD for noncarcinogenic or systemic toxicity have been derived
by several groups within the Agency. An effort is currently under way to
corroborate these values and to produce a master list of RfDs for use by the
various Agency programs. Most of the noncarcinogenic chemicals that
currently are candidates for sludge criteria for the land application food
chain pathways are included on the Agency's RfD list, and thus no new effort
will be required to establish RfDs for deriving sludge criteria. For any
chemicals not so listed, RfD values should be derived according to
established Agency procedures (U.S. EPA, 1987b).
4.1.5.1.2. Human Body Weight (bw) — The choice of body weight for
use in risk assessment depends on the definition of the individual at risk,
that in turn depends on exposure and susceptibility to adverse effects. The
RfD (or ADI) was defined before as the dose on a body-weight basis that
could be safely tolerated over a lifetime. Food consumption on a body-
weight basis is substantially higher for infants and toddlers than for
teenagers or adults. Certain behaviors, such as mouthing of dirty objects
or direct ingestion of soil, which could also contribute to exposure, are
also much more prevalent in children than adults. Therefore, infants and
toddlers would be at greater risk of exceeding an RfD when exposure is by
food or soil ingestion. However, the effects on which the RfD is based may
occur after a cumulative exposure period, in some instances approaching the
human lifespan. In these cases, it may be reasonable to base the derivation
of criteria upon adult values of bw. In cases where effects have a shorter
latency (<10 years) and where children are known to be at special risk, it
may be more appropriate to use values for toddlers or infants.
4.1.5.1.3. Total Background Ingestion Rate of Pollutant (TBI) — It
is important to recognize that sources of exposure other than the sludge
4-32
-------
reuse or disposal practice may exist, and that the total exposure should be
maintained below the RfD. Other sources' of exposure include background
levels (whether natural or anthropogenic) in drinking water, food or air.
Other types of exposure that are due to occupation or habits such as smoking
might also be included depending on data availability and regulatory
policy. These exposures are summed to estimate TBI.
Data for estimating background exposure usually are derived from
analytical surveys of surface, ground or tap water, from FDA market basket
surveys and from air-monitoring surveys. These surveys may report means,
medians, percentiles or ranges, as well as detection limits. Estimates of
TBI may be based on values representing central tendency or on upper-bound
exposure situations, depending on regulatory policy. Data chosen to
estimate TBI should be consistent with the value of bw. Where background
data are reported in terms of a concentration in air or water, ingestion or
inhalation rates applicable to adults or children can be used to estimate
the proper daily background intake value. Where data are reported as total
daily dietary intake for adults and similar values for children are unavail-
able, conversion to an intake for children may be required. Such a conver-
sion could be estimated on the basis of relative total food intake or
relative total caloric intake between adults and children.
As stated in the beginning of this subsection, the TBI is the summed
estimate of all possible background exposures, except exposures resulting
from a sludge disposal practice. To be more exact, the TBI should be a
summed total of all lexicologically effective intakes from all nonsludge
exposures. To determine the effective TBI, background intake values (BI)
for each exposure route must be divided by that route's particular relative
effectiveness (RE) factor. Thus, the TBI can be derived after all the
background exposures have been determined, using the following equation:
4-33
-------
Tnr / ,,, x
TBI (mg/day) =
BI (food) BI (water) BI (air)
RE (food) + RE
BI (n)
RE
(4-15)
where:
TBI = total background intake rate of pollutant from all other
sources of exposure (mg/day)
BI = background intake of pollutant from a given exposure route,
indicated by subscript (mg/day)
RE = relative effectiveness, with respect to dietary exposure,
of the exposure route indicated by subscript (unitless)
When TBI is subtracted from the weight-adjusted RfD, the remainder
(after adjusting for RE) defines the increment that can result from sludge
disposal without exceeding the threshold. If upper-bound data (such as 95th
percentiles) were used to estimate TBI, then an increase in exposure
corresponding to this increment, if realized, would cause the RfD to be
approached or exceeded in a relatively small percentage (5%) of the exposed
population. If central-tendency data (the median) were used to estimate
TBI, such an increase would cause the RfD to be approached or exceeded in
about half of the exposed population. If TBI were set at zero for lack of
exposure data, the allowed increase resulting from sludge disposal would
result in an unknown degree of exceeding the RfD, depending on whether other
sources of exposure exist.
4.1.5.1.4. Relative Effectiveness of Exposure (RE) — RE is a unit-
less factor that shows the relative toxicological effectiveness of an expo-
sure by a given route when compared with another route. The value of RE may
reflect observed or estimated differences in absorption between the
inhalation and ingestion routes that can then significantly influence the
quantity of a chemical that reaches a particular target tissue, the length
of time it takes to get there, and the degree and duration of the effect.
The RE factor may also reflect differences in the occurrence of critical
4-34
-------
toxicological effects at the portal of entry. For example, carbon
tetrachloride and chloroform were estimated to be 40 and 65% as effective,
respectively, by inhalation as by ingestion based on absorption differences
(U.S. EPA, 1984e,f). In addition to route differences, RE can also reflect
differences in bioavailability due to the exposure matrix. For example,
absorption of nickel ingested in water has been estimated to be 5 times that
of nickel ingested in diet (U.S. EPA, 1985d). The presence of food in the
gastrointestinal tract may delay absorption and reduce the availability of
orally administered compounds, as demonstrated for halocarbons (NRC, 1986).
Physiologically based pharmacokinetic (PB-PK) models have evolved into
particularly useful tools for predicting disposition differences due to
exposure route differences. Their use is predicated on the premise that an
effective (target-tissue) dose achieved by one route in a particular species
is expected to be equally effective when achieved by another exposure route
or in some other species. For example, the proper measure of target-tissue
dose for a chemical with pharmacologic activity would be the tissue concen-
tration divided by some measure of the receptor binding constant for that
chemical. Such models account for fundamental physiologic and biochemical
parameters such as blood flows, ventilatory parameters, metabolic capacities
and renal clearance, tailored by the physicochemical and biochemical prop-
erties of the agent in question. The behavior of a substance administered
by a different exposure route can be determined by adding equations that
describe the nature of the new input function. Similarly, since known
physiologic parameters are used, different species (e.g., humans vs. test
species) can be modeled by replacing the appropriate constants. It should
be emphasized that PB-PK models must be used in conjunction with toxicity
and mechanistic studies to relate the effective dose associated with a
certain level of risk for the test species and conditions to other
4-35
-------
scenarios. A detailed approach for the application of PB-PK models for
derivation of the RE factor is beyond the scope of this document, but the
reader is referred to the comprehensive discussion in NRC (1986). Other
useful discussions on considerations necessary when extrapolating route to
route are found in Pepelko and Withey (1985) and Clewell and Andersen (1985).
Since most exposures in this group of pathways occur by food consumption,
the RE factors applied are all with respect to ingestion in food.*
Therefore, the value of RE in Equation 4-14 gives the relative effectiveness
of the exposure route and circumstances on which the RfD was based when
compared with food. Similarly, the RE factors in Equation 4-15 show the
relative effectiveness, with respect to exposure in food, of each background
exposure route and matrix.
An RE factor should be applied only where well documented/referenced
information is available on the contaminant's observed relative
effectiveness or its pharmacokinetics. When such information is not
available, RE is equal to 1.
4.1.5.2. CARCINOGENS — For carcinogenic chemicals, the Agency
considers the excess risk of cancer to be linearly related to dose (except
at high dose levels) (Federal Register, 1986a). The threshold assumption,
therefore, does not hold, as risk diminishes with dose but does not become
zero or background until dose becomes zero.
The decision whether to treat a chemical as a threshold- or nonthresh-
old-acting (carcinogenic) agent depends on the weight of the evidence that
it may be carcinogenic to humans. Methods for classifying chemicals as to
their weight of evidence have been described by the U.S. EPA (Federal
*The only exception is exposure from soil ingestion. In this case RE values
should take into account the soil matrix if supporting data are available.
4-36
-------
Register, 1986a), and most of the chemicals that presently are candidates
for sludge criteria have recently been classified in Health Assessment
Documents or other reports prepared by the U.S. EPA's Office of Health and
Environmental Assessment (OHEA), or in connection with the development of
Recommended Maximum Contaminant Levels (RMCLs) for drinking water
contaminants (Federal Register, 1985). To derive values of RIA, a decision
must be made as to which classifications constitute sufficient evidence for
basing a quantitative risk assessment on a presumption of carcinogenicity.
Chemicals in classifications A and B, "human carcinogen" and "probable human
carcinogen," respectively, have usually been assessed as carcinogens,
whereas those in classifications D and E, "not classifiable as to human
carcinogenicity because of inadequate human and animal data" and "evidence
of noncarcinogenicity for humans," respectively, have usually been assessed
according to threshold effects. Chemicals classified as C, "possible human
carcinogen," have received varying treatment. For example, lindane,
classified by the Carcinogen Assessment Group (CAG) of the U.S. EPA as
"B2-C," or between the lower range of the B category and category C, has
been assessed both by using the linear model for tumorigenic effects (U.S.
EPA, 1980b) and based on threshold effects (Federal Register, 1985). Table
4-5 gives an illustration of these EPA classifications based on the
available weight of evidence.
The use of the weight-of-evidence classification, without noting the
explanatory material for a specific chemical, may lead to a flawed conclu-
sion since some of the classifications are exposure route dependent.
Certain compounds or elements, such as nickel, have been shown to be-
carcinogenic by the inhalation route but not by ingestion (U.S. EPA,
1985d). Similarly, arsenic has been shown to cause carcinogenic effects
when certain inorganic forms are ingested in water, but no carcinogenic
4-37
-------
TABLE 4-5
Illustrative Categorization of Carcinogenic Evidence
Based on Animal and Human Data*
Animal Evidence
Human
Evidence
Sufficient
Limited
Inadequate
No data
No evidence
Sufficient
A
Bl
B2
B2
B2
Limited
A
Bl
C
C
C
Inadequate
A
Bl
D
D
D
No Data
A
Bl
D
D
D
No
Evidence
A
Bl
D
E
E
*The above assignments are presented for illustrative purposes. There may
be nuances in the classification of both animal and human data indicating
that different categorizations than those given in the table should be
assigned. Furthermore, these assignments are tentative and may be modified
by ancillary evidence. In this regard all relevant information should be
evaluated to determine if the designation of the overall weight of evidence
needs to be modified. Relevant factors to be included along with the tumor
data from human and animal studies include structure-activity relationships,
short-term test findings, results of appropriate physiological, biochemical
and toxicologies! observations, and comparative metabolism and pharmaco-
kinetic studies. The nature of these findings may cause an adjustment of
the overall categorization of the weight of evidence.
4-38
-------
potential has been demonstrated for the organic forms normally present in
many foods (U.S. EPA, 1984a). The issue of whether or not to treat an agent
as carcinogenic by ingestion remains controversial for several chemicals.
If a pollutant is to be assessed according to nonthreshold, carcinogenic
effects, the adjusted reference intake, RIA (in yg/day), is derived as
follows:
RIA =
where:
x BW
x RE
- TBI
X 103
(4-16)
RIA
Ql*
RL
BW
RE
TBI
adjusted reference intake (pg/day)
human cancer potency [(mg/kg/day)-i]
risk level (unitless) (e.g., 10-s, 10-*, etc.)
human body weight (kg)
relative effectiveness of ingestion exposure (unitless)
total background intake rate of pollutant (mg/day); from
all other sources of exposure
103 = conversion factor (yg/mg)
The RIA, in the case of carcinogens, is thought to be protective recognizing
that the estimate of carcinogenicity is an upper limit value. The defini-
tion and derivation of each of the parameters used to estimate RIA for
carcinogens are further discussed in the following sections.
4.1.5.2.1. Human Cancer Potency (q *) — For most carcinogenic
chemicals, the linearized multistage model is recommended for estimating
human cancer potency from animal data (Federal Register, 1986a). When
epidemiological data are available, potency is estimated based on the
observed relative risk in exposed vs. nonexposed individuals, and on the
magnitude of exposure. Guidelines for use of these procedures have been
presented in the Federal Register (1980, 1985) and in each of a series of
4-39
-------
Health Assessment Documents prepared by OHEA (such as U.S. EPA, 1985c). The
true potency value is considered unlikely to be above the upper-bound
estimate of the slope of the dose-response curve in the low-dose range, and
it is expressed in terms of risk-per-dose, where dose is in units of
mg/kg/day. Thus, q^ has units of (mg/kg/day)"1. OHEA has derived
potency estimates for each of the potentially carcinogenic chemicals that
currently are candidates for sludge criteria. Therefore, no new effort will
be required to develop potency estimates to derive sludge criteria.
4.1.5.2.2. Risk Level (RL) — Since by definition no "safe" level
exists for exposure to nonthreshold agents, values of RIA are calculated to
reflect various levels of cancer risk. If RL is set at zero, then RIA will
be zero. If RL is set at 10"6, RIA will be the concentration that, for
lifetime exposure, is calculated to have an upper-bound cancer risk of one
case in one million individuals exposed. This risk level refers to excess
cancer risk, that is, over and above the background cancer risk in unexposed
individuals. By varying RL, RIA may be calculated for any level of risk in
the low-dose region, that is, RL <10~2. Specification of a given risk
level on which to base regulations is a matter of policy. Therefore, it is
common practice to derive criteria representing several levels of risk
without specifying any level as "acceptable."
4.1.5.2.3. Human Body Weight (bw) -- Considerations for defining bw
are similar to those stated in Section 4.1.5.1.2. The CAG assumes a value
of 70 kg to derive unit risk estimates for air or water. As discussed
previously, ingestion exposures may be higher in children than in adults
when expressed on a body-weight basis. However, if exposure is lifelong,
values of bw are usually chosen to be representative of adults.
4-40
-------
4.1.5.2.4. Total Background Intake Rate of Pollutant (TBI) — As
discussed in Section 4.1.5.1.3., it is important to recognize that sources
of exposure other than the sludge reuse disposal practice may exist. The
total exposure to a given pollutant should be maintained below the deter-
mined cancer risk-specific exposure level (RL).
4.1.5.2.5. Relative Effectiveness of Ingestion Exposure (RE) — In
some cases potency estimates have been derived on the basis of a different
type of exposure than may occur from food-chain contamination. In these
cases, the use of RE for carcinogens is similar to that described earlier
for threshold-acting toxicants (see Section 4.1.5.1.4.). As stated in that
section, an RE factor should only be applied where well documented/refer-
enced information is available on the contaminant's observed relative
effectiveness or its pharmacokinetics. When such information is not
available, RE is equal to 1.
4.2. SLUDGE-SOIL-PLANT-HUMAN TOXICITY EXPOSURE PATHWAY
4.2.1. Assumptions. In addition to many of the assumptions listed in
Table 4-1, some additional assumptions relating to relative uptake response
of crop food groups and percent of diet affected by sludge application are
made for this pathway. These assumptions and their potential ramifications
are summarized in Table 4-6 and further discussed in the following text.
4.2.2. Calculation Method.
4.2.2.1. CATEGORY 1 CONTAMINANTS —
4.2.2.1.1. Procedure Based on Curvilinear ("Plateau") Uptake Response
Model and Relative Uptake Response Values — Criteria for inorganic pollu-
tant concentrations in sludge applied to land where crops for human consump-
tion are grown may be calculated using a curvilinear response model and
relative response values, if sufficient data are available. The procedure
4-41
-------
10
I
UJ
_)
CD
re
re
a.
a>
s_
3
I/I
O
a.
x
u
X
o
•»->
C
re
a>
en
•o
s_
o
in
o
in
in
ions/Limitations
Ramificatl
Assumptions
re
CD
^£
r—
re
o
•r-*
•*->
g_
3
LU
>,
*-•* r—
QJ "
i/i re
CD O O CD
of crop respons
jre dietary resp
restimated, espe
ata are availabl
>,<+_ o> -o
+J O) T3
•r- 4- C 3
l — CD 3 (U
•r- x: **-
XJ •»-> CD
re JQ CD
.1- -0 4-
4- C > CD
re re re xz
^^ >-^ E 3
"O C
o re 4- re
o o o +•»
u— *^-**o re
^-*. CD "o
c m CD -i->
CD (U 3 U CD
•r- xi 're r- c
en re > CD o
4-> i/l Q.
CO 03 CD ^^
Cnr— x-x CD
c a> en in 4-
•r- > C CD
•i- 3 CD O.
t/1 CD I/I r— r— 3
a. E re xi o
o 3 re > re 4_
4- en r— en
O CD >} O i-
r- xs 3 re CD
it- 4-> > x:
o m -o re +•»
(O CD ^^
CD .(-> CD CD C
i/i x: c
O 3 t/i o in
CL in CD CO.
l/l i- CD O O
CD O. E 4-
4_ CL CD O T3 0
3 S- on CD
CD O i/l t-
x: 4- CD c re o
t— en xj T- xj M-
CD
in
c
O
CL
in i/>
CD O.
i. 3
o
CD S-
^ Cn
re
••-> -a
a. o
3 O
0)
> Q.
•r- O
+J S-
re u
CD <4-
rv O
x:
o ^~
•r- O CD
(/* m i «
3 X3 C T3 **- 4-
C O 3 t- O
t_ r— re "— r— x:
CD i — i — •)-> 00 i/l i/»
> CD U C
0 3 T3 =3 >> 0
CD "a • re '•—
C •)-» TD O m E 4->
o o c i- o. re
•r- c CD a. O m i —
en E i- CD 3
CD in re CD o o en
S_ -r- 1 > T- CD
CD re c +J 4-
CD T3 CO CD T- O
x: CD "o i — re re o •
4J X 3 X: 4_ +J to
•r- i — 4- O CL 4-
M- E 00 O 1 CD O
0 T3 •«->
CD 4. O C C U
CD 4. * CD O CD O re
M re -a •!-» n- E cm-
•r- CD C CD I/I
inmccDcencD4.
-o T- re re 4. CD
CD o «- >» E c x:
xiooresrec-M
H-<4-T3 EX: E'r- O
«- 1
T3 O •<- CD C
The percentage of commercially dis-
tributed foods in an individual's
diet that derives from sludge-amende-
soil can be estimated as a function
regional sludge production, regional
cropland area, agronomic sludge appl
cation rates and percentage of sludg
typically applied to human food-chai
land.
•o
CD
•H
U
CD C
14— O
11— *f—
re 4->
re
•»-> u
CD -r-
•5 "Q.
CL.
«- re
o
CD
c en
o -o
«r- 3
4-> r—
O Wl
re
4. >>
U_ JO
diet exposure.
May overpre
T3
All of an individual's homegrown foo
could come from sludge-amended soil.
•!-»
re c CD
o re i/i
•r- 3
14- I/I 0
fl •(••» ^J ^^^
•r- C •!-
enx: CD
• •r- O. I/l
C m re CD
o 4- x:
.r- 3 cn-t->
•4-> O O
re x: E <+-
E m x: re o
£ re o cnx:
CD
5 E
The percentage of homegrown food in
diet of the MEI can be estimated fro
USDA (1966) survey data on rural
farm households, which constituted
6% of all households.
4-42
-------
for deriving criteria for a given contaminant is summarized in Table 4-7 and
described in detail as follows:
steP A- Determine Relative Uptake Response Values for Each Crop
Relative response values for the metal at hand should be determined for
as many crops as permitted by the available data. As discussed in Section
4.1.2.1., relative response is determined within and not across studies and
is determined with respect to an index crop that is both sensitive (in terms
of response) and frequently studied. (Lettuce is an example of a crop often
meeting these criteria.) The index crop should also be one for which data
showing a plateau are most widely available (see Step D). A relative
response value for crop A would be determined as follows:
EA
UCi
(4-17)
where:
RUAI = uptake response for crop A relative to index crop
(unitless)
UCA = linear uptake response slope of crop A (yg/g [kg/ha]-*)
UCi = Linear uptake response slope of index crop in the same
experiment in which UCA was measured
Some studies (Davis and Carl ton-Smith, 1980; Carlton-Smith and Davis,
1983) showing the response of multiple crops to different sludge-amended
soils do not report contaminant application rates and, therefore, values of
the linear response slope UC cannot be calculated. However, if tissue
levels for crops A and I were both measured in two different soils, then
RUA]. can be estimated as follows:
RUAI =
A2 ~ AT
12 - II
(4-18)
4-43
-------
TABLE 4-7
Summary of Criteria Derivation Procedure Based on Curvilinear Uptake
Response Model and Relative Uptake Response Values
Step
A
B
C
0
E
F
Description
Determine relative uptake response values for
each crop
Determine relative uptake response values for
each food group
Determine the reference tissue concentration
increment (RTI) for the index crop
Sort available uptake response data for the
index crop
Determine plateau increment values (PI) for the
index crop
Determine reference sludge concentration (RSC)
Text Page
4-43
4-45
4-46
4-47
4-48
4-48
G
not causing reference tissue concentration
increment (RTI) to be exceeded
Check the reference sludge concentration (RSC)
by substituting other crops for the index crop
4-49
4-44
-------
where A? denotes tissue concentration of crop A in soil 2, and so forth.
As a practical matter, this procedure should only be used when it. is clear
from the data that the contaminant concentration is substantially higher in
soil 2 than in soil 1, so that the difference between I and I is
meaningful. A control soil, if used, is the best choice for soil 1. This
procedure was used in U.S. EPA (1987a).
Uptake response relative to the index crop should be determined for as
many crops as possible. Where more than one relative response value can be
determined for one crop, a value determined under high-response conditions
(such as at low pH) should ordinarily be selected as a conservative measure.
If a given crop (A) has not been studied in the same experiment as the index
crop (I), but has been co-studied with another crop (B) for which response
relative to the index (RUgI) has been determined, relative response for
crop A (RUAI) can be estimated as follows:
RUAI - RUBI X RUAB
(4-19)
where RUAg is uptake response of crop A relative to crop B.
steP B-—Determine Relative Uptake Response Values for Each Food Group
Once relative response has been determined for as many crops as
possible, these crops should be divided into the crop food groups shown in
Table 4-2. A single relative response value should be assigned to each food
group in most instances. This value may be determined as a weighted mean of
all the available response values where weighting is according to the dry-
weight consumption of each crop. Where there are no response values for a
particular crop, care should be taken in assigning values so that the
response value is from a closely related crop. If the data do not permit
determination of a weighted mean, an unweighted mean may be taken, or to be
conservative the highest single value may be chosen to represent that food
group.
4-45
-------
If the data indicate that a food group consists of some crops that are
relatively high accumulators and some that are relatively low (such as
lettuce, spinach and chard vs. cabbage and the other coles in the leafy
vegetable food group), then it may be worthwhile to subdivide the food group
on this basis. If no relative response value can be determined for a
particular food group, the highest value for any of the other comparably
responsive food groups should be assigned to that group.
Step C. Determine the Reference Tissue Concentration Increment (RTI) for
the Index Crop
Once a relative response value has been determined for each food group
potentially affected by sludge application, the response values are coupled
with the dry-weight consumption values for each group. In this manner,
total dietary response to sludge application can be related to the response
of the index crop. A reference tissue concentration increment (RTI, in
yg/g DW) based on human health effects and fraction of diet affected can
then be determined for the index crop:
RIA
RTI =
(RUi x DCi x FCi)
(4-20)
where:
RIA = adjusted reference intake (yg/day)
RUi = relative uptake response for ith food group (unitless)
DCi = daily dietary consumption of ith food group (g DW/day)
FCi = fraction of food group assumed to originate from sludge-
amended soil (unitless)
Reference intake, RIA, is derived based on health effects data as discussed
1n Section 4.1.5. RTI, as derived from Equation 4-20, is the tissue concen-
tration increase for the index crop that, if exceeded, indicates that the
health-based intake level used to derive RIA will be exceeded.
4-46
-------
Step D. Sort Available Uptake Response Data for the Index Crop
The available response data for the indicator crop should be grouped
according to whether soil pH was <6.0 or >6.0, and according to whether the
plateau was observed in the first year of sludge application or over several
years. If criteria are to be derived based on the assumption (or the
requirement) that pH will be maintained >6.0 in sludge-amended soils where
food crops are grown, then studies with pH <6.0 should not be used for
derivation of criteria. If soil pH will not necessarily be maintained >6.0,
then studies with pH <6.0 should be used as a conservative measure, since
these studies normally show higher metal uptake. This procedure differs
from the existing approach (40 CFR 257.3-5) of using pH 6.5 as the cutoff
between different permissible application rates. The Las Vegas workshop
report (U.S. EPA, 1987) concludes that plant response to metals is not
significantly increased when the soil pH (as determined using a 1:1
soil-water suspension) decreases from 6.5 to 6.0 (although it may increase
at pH <6.0). The report further concludes that the use of soil cation
exchange capacity (C.E.C.) as a basis for determining metal application
limits is not supported by current evidence. Therefore, no difference in
approach based on C.E.C. is recommended here. The use of this methodologic
approach should be conditioned upon acceptance of the final workshop report
by the workshop participants and other scientists knowledgeable in this
field.
That report also indicates that data based on the first year of sludge
application will tend to show higher response rates and higher plateau
values than data from annually repeated sludge additions. It is unlikely
that any individual would be exposed, year after year, to crops grown imme-
diately following a first application of sludge, and therefore basing risk
assessments solely on first-year data may be unreasonable. On the other
4-47
-------
hand, some exposure to such crops would presumably occur. In the case of
chronic exposure this would not be a problem. However, for acute exposure,
separate criteria calculations should be carried out based on first-year and
multi-year application to determine the potential range of results.
Step E. Determine Plateau Increment Values (PI) for the Index Crop
All available response data in the index crop for the metal at hand for
the groups identified in Step D should be examined to determine which
studies show a plateau for tissue concentration. As suggested earlier (see
Section 4.1.2.1.2.), the plateau value (P, in vg/g DW) may be determined
using nonlinear regression and estimation of confidence limits around the
asymptote. A value derived from P will be used in subsequent calculations,
e.g., the plateau increment value, PI (in yg/g DW), which is the plateau
value (P) minus the background tissue concentration (BC) for the experiment
in which P was determined.
Step F. Determine Reference Sludge Concentration (RSC) Not Causing
Reference Tissue Concentration Increment (RTI) to be Exceeded
For each grouping of plateau data described in Step D above, it should
be determined whether a reference sludge concentration (RSC) can be identi-
fied that gives a plateau increment (PI) not exceeding the RTI. To make
such a determination, several steps are necessary. First, it must be
demonstrated that all sludges with metal concentration at or below the RSC
always result in tissue concentration increments below the RTI in all
available sludge-field studies within that grouping, regardless of the
sludge (or contaminant) application rate. This finding must hold true in
studies that do not show a plateau, as well as those that do. Second, an
adequate number of studies must be available to clearly demonstrate the
relationship between sludge concentration and plateau increment (PI), so
that there can be a high degree of confidence that the use of a different
4-48
-------
sludge or soil will not result in exceeding the RTI. This methodology will
not attempt to define the number or quality of studies necessary to make
this determination; the validity of a derived value of RSC should be
determined by expert peer review.
Studies at soil pH <6.0 may be used to help validate an RSC for soil pH
>6.0, but the reverse does not apply. Similarly, first-year studies may be
used to validate an RSC for multi-year applications, but the reverse should
not be done.
steP 6- Check the Reference Sludge Concentration (RSC) bv Substituting
Other Crops for the Index Crop
According to the relative response hypothesis, it should be possible to
use any crop as the index crop and arrive at about the same value for RSC.
Thus, RSC can be checked against other crops, even if plateau data are not
available for those crops. To do so, RTI$ for the substitute crop is
first determined as follows:
RTI - RTI x RU
^ 1
(4-21)
where:
RTIS = reference tissue concentration increment for substitute crop
(yg/g DU)
RTIi = reference tissue concentration increment for index crop
(vg/g DM)
RU$i = uptake response in substitute crop relative to index crop
(unitless)
Steps D-F are then repeated using data for the substitute crop. If data
showing a plateau are not found, it can still be determined from the avail-
able data whether application of sludge at or below RSC ever results in
tissue concentration increments exceeding RTI . If so, RSC should be
O
revised downward so that RTI$ is not exceeded.
4-49
-------
RSC should be tested in this manner using all crops for which RU has
been determined. If RU for a given food group was derived as a weighted
mean (see Step B, above), then RTI may be exceeded for certain crops in
O
that food group. However, the following condition should hold:
TIs
RTIS ~ RUG
(4-22)
where:
highest tissue concentration increment (i.e., increase
above background) observed in substitute crop when
sludge having concentration
-------
TABLE 4-8
Relationship Between the Experimental Basis for Reference Sludge
Concentration (RSC) and Rules Governing Use of Sludges Meeting RSC*
Data Used To
Generate Limit
Soil pH
>6.0 and/or <6.0
<6.0 only
Multi-year and/or
first-year
First-year only
A.
B.
C.
D.
Applies to cumulative application if soil pH will remain >6.0 without
liming. Separate, annual application limit applies.
Applies to cumulative application. The pH requirement is dependent upon
data. Separate, annual application limits apply.
Applies to cumulative application if soil pH will remain >6.0 without
liming. Separate, annual application limit not required.
Applies to cumulative application. The pH requirement is dependent upon
data. Separate, annual application limit not required.
*For explanation, see text.
4-51
-------
studies in which RSC was determined. That is, even though annual appli-
cations of sludge with a concentration of metal X of 10 pg/g never caused
RTI to be exceeded when 10 t OW/ha of sludge was applied over several years,
it cannot be assumed that a single application of 50 t OW/ha will not cause
it to be exceeded. Therefore, an annual contaminant application limit of
(10 t OW/ha x 10 yg/g) 0.1 kg/ha should apply in this case, although no
cumulative limit is needed. Alternatively, the annual application limit may
be based on the procedure using the linear model (see Section 4.2.2.1.2.).
When crops for human consumption are grown on agricultural lands that
have received sludge, liming is often practiced to maintain soil pH >6.0 so
as to lessen uptake of metals. However, it cannot be assumed that liming
will continue after the land is sold to another farmer or developed for
residential use, since there is no requirement that future owners be
informed of sludge applications. Therefore, it is recommended that criteria
based in whole or in part on data from soils of pH >6.0 be employed only
where soils are expected to remain at pH >6.0 even when liming is not
practiced. Similarly, if RSC is derived using data from soils at pH >6.0,
sludges meeting this RSC value could be applied without cumulative applica-
tion limits as long as soil pH was expected to remain >6.0. If based only
on soils with pH <6.0, the pH requirement would be that of the data used.
For example, if the pH of the study used was 5.5 there would be a pH 5.5
requirement.
4.2.2.1.2. Procedure Based on Linear Uptake Response Model and
Relative Uptake Response Values — Criteria based on the assumption of
linear uptake (as opposed to curvilinear uptake) will also be required for
use in the following situations: 1) if the available data are insufficient
to support derivation of RSC values; 2) if RSC for acid soils cannot be
4-52
-------
derived and pH maintenance is not viable; or 3) to govern application of
sludges that do not meet RSC. The criteria derivation procedure based on
linear uptake is summarized in Table 4-9 and described in detail in later
sections.
Step A. Determine Relative Uptake Response Values for Each Crop
This step is the same as described for the curvilinear model procedure
in Section 4.2.2.1.1.
Step B. Determine Relative Uptake Response Values for Each Food Group
This step is the same as described for the curvilinear model procedure
in Section 4.2.2.1.1.
steP C. Determine the Reference Tissue Concentration Increment (RTI) for the
Index Crop
This step is the same as described for the curvilinear model procedure
in Section 4.2.2.1.1.
Step D. Sort Available Uptake Response Data for the Index Crop
This step is the same as described for the curvilinear model procedure
in Section 4.2.2.1.1.
Step E. Determine Appropriate Uptake Response Slopes for the Index Crop
The highest valid uptake slope for each of the categories established in
Step D should be identified. That is, field studies where sludge was
applied should be considered to provide the most valid information; pot
studies with sludge should be used where the latter are not available. Pot
studies with added metal compounds are considered less useful, even if
sludge was also applied. It should be realized from previous discussion
(Section 4.1.2.1.1.) that pot studies will overestimate plant response in
field situations.
4-53
-------
r
TABLE 4-9
Summary of Criteria Derivation Procedure Based on Linear Uptake
Response Model and Relative Uptake Response Values
Step
E
F
G
Description
index crop
Determine appropriate uptake response slopes
for the index crop
Determine reference" application rate of the
pollutant (RP)
Check the reference application rate of the
pollutant (RP) by substituting other crops for
the index crop
Adjust the reference application rate of the
pollutant (RP) for phytotoxic effects
Text Page
A
B
C
D
Determine relative uptake response values for
each crop
Determine relative uptake response values for
each food group
Determine the reference tissue concentration
increment (RTI) for the index crop
Sort available uptake response data for the
4-53
4-53
4-53
4-53
4-53
4-55
4-55
4-55
4-54
-------
Step F. Determine Reference Application Rate of the Pollutant (RP)
For each category established in Step 0, the reference application rate
of the pollutant (RP, in kg/ha) is calculated from the reference tissue
concentration increment (RTI, in v>9/9 PW) and the index crop response
slope (UCr, in yg/g DW [kg/ha]'1), as follows:
RP • RTI/UCj (4-23)
This is analogous to the determination of a reference sludge concentration
(RSC) in Step F of the curvilinear model method (see Section 4.2.2.1.1.)
except that instead of a plateau, that relates tissue concentration incre-
ment to a sludge concentration, a slope is used that relates the tissue
concentration increment to a pollutant application rate.
Step 6. Check the Reference Application Rate of the Pollutant (RP) bv
Substituting Other Crops for the Index Crop
This step is analogous to Step G of the previous section (see Section
4.2.2.1.1.). A reference tissue concentration increment for a substitute
crop is determined using Equation 4-21. Using response slopes for the
substitute crop from the appropriate category established in Step D, RP is
recalculated from Equation 4-23. Studies at soil pH <6.0 may be used to
help validate an RP value for soil pH >6.0, but the reverse does not apply.
Similarly, first-year studies may be used to validate RP for multi-year
applications, but the reverse should not be done. If the recalculated RP is
less than RP based on the index crop, the recalculated value should be used,
except when Equation 4-22 holds true.
Step H. Ad.iust the Reference Application Rate of the Pollutant (RP) for
Phvtotoxic Effects
Since the linear uptake model projects continually increasing tissue
concentrations with increasing pollutant application, it is possible using
4-55
-------
this model to predict increases in tissue concentration that in actuality
would be prevented by toxic effects in the plant. If so, the application
rate calculated in the soil-plant pathway (see Section 4.6.) will be lower
than that calculated in the present pathway, and therefore adjustments to
the rate calculated here may be unnecessary. However, an adjustment method
will be presented so that each pathway may stand alone and to avoid
overestimating the potential for human health effects from phytotoxic
pollutants.
A maximum pollutant application rate (RPM.) based on phytotoxicity
should be determined for each crop group that was used in Steps B and C to
help determine RTI:
r) (4-24)
RPM1 = (TL. - BC.)/(RU. x
where:
TL.j = maximum tissue concentration for crop category i, above
which crop production is virtually eliminated by phyto-
toxicity (yg/g DW)
BCi = background concentration for crop category i (yg/g DW)
RUi = relative uptake response value for crop category i
(unitless)
UCj = uptake response slope for index crop, used to determine
the RP value being adjusted (yg/g DW [kg/ha]-i)
The calculated RPM. represents the pollutant application rate at
which, according to the linear response model, the tissue concentration in
crop category i reaches its maximum level; higher concentrations would
effectively terminate crop production. Values of RPM. are compared with
RP. If RPM,
-------
indicate effective removal of crop i from the diet. Since RP is a
permitted, but not required, application rate, is is assumed that lower
rates would be used on more sensitive crops.]
For all crops for which RPM. RP. Thus, the summation in
the numerator has units of pg/day and denotes the constant amount of
contaminant in the diet contributed by those crops assumed to be at their
maximum concentration. This amount is subtracted from the adjusted
reference intake, RIA, to determine the remaining intake amount that can be
partitioned into those crops still uptake-dependent in the denominator.
If Equation 4-25 is used, RTI will increase, since uptake response has
been limited in some crop categories. Therefore, RP will also increase
(from Equation 4-23), and may exceed RPM. for additional crops, requiring
additional recalculation of RTI using Equation 4-25. This iterative process
is repeated until there is no further change in RP.
Once the above steps have been followed, RP represents a cumulative
application rate for inorganic pollutants that should not be exceeded where
crops for human consumption are grown. As was the case when the curvilinear
response model was used (see Section 4.2.2.1.1.), the interpretation of RP
depends on the data used in its derivation (see Step 0 above). If only
first-year data were used in Steps E-G, then RP constitutes a cumulative
4-57
-------
limit. If multi-year data were used (with or without support from first-
year data, see Step G), RP is a cumulative rate limit, but may not be fully
protective in the first year of application. RP based on first-year data
could be used as an annual limit.
If derived using data from soils at pH >6.0 (with or without support
from data derived at pH <6.0), RP is a cumulative application limit for
soils expected to remain at or above that pH value. If based only on
studies with pH <6.0, then RP is a cumulative application limit that can be
applied to all soils, regardless of pH. The relationship between the
experimental basis for RP and the resulting rules is illustrated in
Table 4-10.
4.2.2.1.3. Procedure Based on Linear Uptake Response Model Without
Using Relative Uptake Response Values — Criteria are calculated using this
procedure if relative response values cannot be determined for an adequate
number of crops to characterize the various affected food groups. Instead
of relative uptake response values, which are unitless, absolute uptake
response slopes, in units of pg/g DW (kg/ha)"1, are used to determine
dietary response to the sludge-borne contaminant. The main disadvantage of
using slopes is that the slopes for each crop used will likely originate
from different experimental conditions. To examine total dietary response
to a change in conditions (such as soil pH), all of the response slopes need
to be changed, instead of individually examining the change in response of
an index crop and several substitute crops as can be done using relative
response values. The use of the curvilinear model is also precluded in
practical terms. The steps for this procedure are summarized in Table 4-11
and are described in detail below.
4-58
-------
TABLE 4-10
Experimental Basis for the Reference Application Rate of Pollutant (RP)
and Situations Where RP Applies*
Data Used To
Generate Limit
Soil pH
>6.0 and/or <6.0
<6.0 only
Multi-year and/or
first-year
First-year only
A. Applies to cumulative application if soil pH will remain >6.0 without
liming. Separate, annual application limit applies.
B. Applies to cumulative application. The pH requirement is dependent upon
data. Separate, annual application limits apply.
C. Applies to cumulative application if soil pH will remain >6.0 without
liming. Separate, annual application limit not required.
D. Applies to cumulative application. The pH requirement is dependent upon
data. Separate, annual application limit not required.
*For explanation, see text.
4-59
-------
TABLE 4-11
Summary of Criteria Derivation Procedure Based on Linear Uptake
Response Model Without Using Relative Uptake Response Values
Step
Description
Text Page
A
B
C
Sort available uptake response data for all
food crops
Determine uptake response slopes for each food
group
Determine reference application rate of the
pollutant (RP)
Adjust the reference application rate of the
pollutant (RP) for phytotoxic effects
4-61
4-61
4-61
4-62
4-60
-------
Step A. Sort Available Uptake Response Data for All Food Crops
The data for all crops consumed by humans are sorted as described in
Step D of the previous procedures (see Sections 4.2.2.1.1. and 4.2.2.1.2.).
Step B. Determine Uptake Response Slopes for Each Food Group
This step is analogous to that described in Step B of the previous
procedures except that response slopes, with units of yg/g DW
(kg/ha)"1, are substituted for unitless response values. For each food
group, a separate response slope should be derived for.. each of the
categories established by sorting in Step A. However, as described in Step
F of the curvilinear procedure (see Section 4.2.2.1.1.) and Step G of the
linear procedure using relative response values (see Section 4.2.2.1.2.),
studies at pH <6.0 may be used to augment the data for pH >6.0, and
first-year studies may be used to augment the data for multi-year studies,
but the reverse should not be done.
Step C. Determine Reference Application Rate of the Pollutant (RP)
Since there is no index crop, a reference tissue concentration is not
calculated. Instead, the reference application rate of the pollutant (RP,
in kg/ha) is calculated directly from the response data by the following
equation:
RIA
RP =
.
1=1
i x DCi x
(4-26)
where:
RIA = adjusted reference intake (vg/day)
UC-j = uptake response slope for ith food group (yg/g DW [kg/ha]"1)
DC-j = daily dietary consumption of ith food group (g DW/day)
FCj = fraction of food group assumed to originate from sludge-
amended soil (unitless)
4-61
-------
Step D. Adjust the Reference Application Rate of the Pollutant (RP) for
Phvtotoxlc Effects
This step is identical to Step H of the linear procedure using relative
response values (see Section 4.2.2.1.2.) except that, instead of first
recalculating RTI (see Equation 4-25) before recalculating RP, RP is calcu-
lated directly from the following equation:
RP
RIA - I [(TLj-BCj) x OCj x FCjJ
k=l
(UCk x DCk x FCk)
(4-27)
where subscript j indicates those food groups for which RPM. RP. RPM.. values are
calculated as in Equation 4-24, except that UCj must be substituted for
the quantity RLL x UCj. The interpretation of values of RP derived by
this procedure is as illustrated in Table 4-10.
4.2.2.2. CATEGORY 2 CONTAMINANTS — The criteria derivation procedure
for organic contaminants by the sludge-soil-plant-human toxicity pathway is
largely the same as the procedure for inorganics based on the linear
response model and not using relative response values (see Section
4.2.2.1.3. and Table 4-11). One difference is that uptake data are not
segregated on the basis of either soil pH or years of sludge application,
since no reason for doing so has been demonstrated. In addition, the
procedure for calculating response slopes for organics differs somewhat from
that for inorganics, in that tissue concentration is treated as a linear
function of soil concentration rather than application rate, as explained in
Section 4.1.2.2. Thus, Equations 4-26 and 4-27 yield a reference
4-62
-------
concentration in soil (RLC) in yg/g DW, rather than the reference
application rate of pollutant, RP, in kg/ha. This assumes that the soil
background concentration, BS, is zero. If BS^O, then Equations 4-26 and
4-27 yield RLC-BS. That is, BS must be added to the result to derive RLC.
RLC is the soil concentration that should not be exceeded. The value of
RLC may be used in Equation 4-13, along with values for degradation rate
constant in soil (k, in years"1), waiting period before planting (T, in
years), and assumed annual sludge application rate (AR , in t DW/ha), to
3
find RPg, the annual application rate of the pollutant (in kg/ha). RP
a
is the application rate that can be repeated indefinitely without exceeding
RLC. If only a single sludge application is to be made, the single applica-
tion rate of the pollutant, RP (in kg/ha), is derived from Equation 4-6.
The preplant waiting period, T, may be employed if desired.
4.2.3. Input Parameter Requirements. The following sections discuss in
more detail the individual parameters required to calculate criteria for the
sludge-soil-plant-human toxicity pathway. In many cases the assignment of
parameter values depends on the specific type of land application practice
assumed. As explained in Chapters 2 and 3, it is assumed that all lands
other than dedicated land and highway roadsides can eventually become agri-
cultural land or home gardens. Therefore, this methodology will not make
distinctions between the risks associated with various sludge utilization
practices; only the practices with highest risk need be examined. However,
it is difficult to state a priori whether application to agricultural lands
or home gardens entails the highest risk. A home gardener may grow a high
percentage of certain types of foods, but a wider number of food types would
probably be affected by agricultural use. Therefore, both of these exposure
4-63
-------
scenarios should be examined to determine for each contaminant of concern
that entails the higher risk. The resulting restrictions should be applied
to all lands except dedicated lands and highway roadsides.
No special consideration will be given to hothouse or other indoor crop
production except to examine potential risk from any crops grown only in
that setting. The only such crop not already covered under the home garden
or agricultural use scenarios is mushrooms. An analysis of contaminant
uptake data for mushrooms should be conducted to determine whether pre-
cautions based on other crops will be sufficient to prevent any excessive
accumulations. For example, Logan and Chaney (1983) presented a review of
data showing particularly high uptake of mercury by mushrooms. They
suggested that sludge composts might not be suitable for use in mushroom
production. Parameter choices are discussed below and also are summarized
in Table 4-12 for agricultural and home gardening practices where crops for
human consumption are grown.
4.2.3.1. FRACTION OF FOOD GROUP ASSUMED TO ORIGINATE FROM SLUDGE-
AMENDED SOIL (FC) — This exposure pathway deals with crops for human
consumption and thus potentially includes grains and cereals, potatoes,
leafy vegetables, legume vegetables, root vegetables, garden fruits, peanuts
and mushrooms (see Table 4-2). Some food groups might be assumed to be
unaffected by certain sludge application practices; in these cases, FC would
be set at zero. For food groups that may be affected, FC is set between
zero and one, based on the percentage likely to have originated from sludge-
amended soils.
4.2.3.1.1. Agricultural Utilization—Sludge production in the
United States is not sufficient to enable application to all U.S. cropland
(Table 4-13). Therefore, not all of the U.S. diet will be affected, even if
4-64
-------
TABLE 4-12
Choice of Input Parameter Values for Sludge-Soil-Plant-Human Toxicity
Pathway, As Affected by Exposure Scenario*
Input Parameter
Exposure Scenario
Agricultural Use
Home Garden
Crops Included
Fraction of crop (FC)
all except
mushrooms
0.025
all except grains
and cereals,
peanuts, mushrooms
0.60 vegetables
0.45 potatoes
0.17 dried legumes
*For explanation, see text.
4-65
-------
fjl
S,
"f
GO
«•
— in
^!
© >
•p "D
IB
o o
t- a>
fl
c "§
11
C- O»
i-c
J -
M
•o ID
(0 *0
e. oo
o
"O *""
!:§
11
<0 T"
P =
t. "5
•o c
oo —
§ -
t
a.
=
o
.2
oo
•o
ndicate
^™
U
*•
•a
£
o
&
•o
c
1C
c
o
o
z
_j
„
<0
w^
^
cr
c
'c
'5
o
ID
|
C/)
03 -V
£ 1
Q.
S- O
0 i-
fe^
tA
•o £
C ID
(A O
1— M-
O
^^
I/I I/I
C ID ID
.2 S t
a. o o>
o j= x:
o^^
•o
ID M- /-.
O O I/I
T3 I/) O
CL S o
I/I •—
ID 3 i.
O> 0 +•
"o j= gj
13 H- e
00
O 1/i
£ o
10 •—
li
I
vO — 00 — ""d" 03
l^s [^, O\ in *"i Ov
6 vg d - [n -
in "a- f- i^ o o
S ^ * S
* ^ ^ " » ^
O vO O O "^ O
•tt K\ o oo in o
— — CM 1^ CM m
l<^ vO
° CO
z. —
CM i3 00 vO vO VO
CM — in in — in
* s * S
f**k I*x CN O "^ GO
* £i S ^ S £i
^
«• O O VO O\ O
— CM ON in ^l- in
— ^. IA CM 00 *
_ O _ ro
CM CM
^
in S in vt
O ID O ID
c -> ^c ->
E|| =||
§
1Z
^
'i
S
5\
T3
C
<0
O
CM
CM
VO
„
•
~ ^
1 II
s- H
S T, *".
0) I/I
•f- «)
ID Mi-
y s-s
O <0 0)
%> 3= ^
ID ^
4-66
-------
all sludge produced were applied to agricultural soils. However, the degree
to which any individual's diet will be affected will depend on the degree of
mixing of food products before consumption. If mixing were complete on a
nationwide basis, FC for the individual's overall diet could not exceed the
fraction of cropland available in the United States that would be required
to receive all sludge produced (assuming that fertilization/irrigation
achieves equivalent yields where sludge is not used). Table 4-13, taken
from CAST (1976), illustrates the percentage of total available cropland
that would be required, based on nitrogen content, to dispose of the total
sludge production for the entire United States. Estimates for 1985 (based
on 1975 information) ranged from 0.49 to 1.98%. If mixing were complete
only on a more geographically limited basis, such as statewide, the fraction
could be much higher, especially in areas where available cropland is small
compared with population size (such as 55% for New Jersey). It is quite
unlikely that all crops consumed in any state, especially a highly urbanized
one, would originate within that state. However, it may be insufficiently
conservative to assume complete mixing nationwide. If an average of the
U.S. and New Jersey cases based on 4% available nitrogen is taken, a
somewhat arbitrary value of 29% results.
Not all of the sludge produced is presently land applied, and the
percent applied to human food-chain land is lower (16%) for large treatment
plants [>10 million gallons/day (MGD); characteristic of population centers]
than for small plants (31% for plants <1 MGD; characteristic of rural
areas). A weighted average of 17% applies to all sludge produced (Pierce
and Bailey, 1982).
It may be desirable to base values of FC on a more recent and thorough
regional analysis of sludge application, land use and food distribution
4-67
-------
practices, as well as sludge nitrogen content. No such analysis will be
attempted here. For purposes of example only, a value of 5% or 0.05
(~0.29x0.17) will be employed here as a reasonable worst-case estimate.
The use of this value of FC for all food groups will tend to overestimate
exposure, since not all sludge-grown crops are for human consumption. For
example, livestock feed, export and seed uses of grain produced in the
United States exceed the amounts used directly for human food products
(CAST, 1976). Thus, a value of 2.5% could be employed as a worst case
estimate of FC for crops grown for human consumption.
4.2.3.1.2. Home Gardens — It will be assumed that home gardeners
produce and consume potatoes, leafy vegetables, legume vegetables, root
vegetables and garden fruits, but not grains and cereals, peanuts or
mushrooms. The USDA (1966) survey of U.S. food consumption in 1965-1966
includes data on the percentages of foods consumed that were homegrown, for
urban, rural nonfarm and rural farm households. The highest percentages of
homegrown foods were for rural farm households, which consitituted c.6% of
all U.S. households (Table 4-14). The rural farm dweller will be taken as
the MEI for the home gardening scenario in this pathway. Values of FC
(after rounding) are 0.60 for all vegetables (except dried legumes), 0.45
for potatoes and 0.17 for dried legumes.
4.2.3.2. UPTAKE RESPONSE DATA: RELATIVE UPTAKE RESPONSE VALUE (RU),
PLATEAU INCREMENT VALUE (PI) AND UPTAKE RESPONSE SLOPE (UC) -- Uptake
response data are required for as many crops as possible in the food groups
for which FC#), as determined in the previous section. Instructions for
deriving the needed uptake parameters are given previously: for RU see
Section 4.1.2.1. and Step A of 4.2.2.1.1.; for PI see Section 4.1.2.1.2. and
Step E of 4.2.2.1.1.; and for UC see Section 4.1.2.1.1.
4-68
-------
TABLE 4-14
Annual Consumption Homegrown Foods3»D
Food Group
Milk, cream, cheese
Fats, oil
Flour, cereal
Meat
Poultry, fish
Eggs
Sugar, sweets
Potatoes, sweet potatoes
Vegetables (fresh, canned, frozen)
Fruit (fresh, canned, frozed)
Juice (vegetable, fruit)
Dried vegetables, fruits
Percent Homegrown bv llrhaniyatinn
Rural Farm
39.9
15.2
1.6
44.2
34.3
47.9
9.0
44.8
59.6
28.6
n.o
16.7
Rural Nonfarm Urban
4.0
1.6
0.8
6.1
11.9
9.1
5.0
14.6
26.7
11.5
4.1
6.9
0.04
0
0
0.8
2.9
0.6
1.6
1.2
5.1
2.8
0.7
2.6
aSource: Calculated from data presented in USDA (1966)
bfhe percentage of the U.S. population, represented by households from each
urbanization at the time of this survey were 6%, 24% and 70% for rural
farm, rural nonfarm and urban, respectively.
4-69
-------
Response slope (UC) or plateau increment value (PI) for many contami-
nants varies inversely with soil pH and number of years of previous sludge
applications (or years since most recent application). The following
assumptions should be applied when selecting response data to represent
different types of land application. Soil pH may be controlled (>6.0) or
uncontrolled (possibly <6.0, as determined using a 1:1 soil-water paste) in
agricultural utilization. Therefore, both situations should be evaluated to
determine whether pH control is warranted. However, pH control should not
be assumed in home gardens; studies with soil pH <6.0 should be used to
determine UC or PI. In agriculture and home gardening, crops may be grown
in the first year of sludge application. Therefore, data of this type
should be used to determine annual application limits, but data based on
prior sludge applications (or intervening years without sludge application)
may be used to determine cumulative limits. For site-specific calculations,
if future use of land may include raising food crops but present use does
not, then first-year response data should be avoided.
4.2.3.3. DAILY DIETARY CONSUMPTION OF FOOD GROUP (DC) — Values for
daily dietary consumption (DC, in g DW/day) are needed for each food group
for which FCj^O. The values chosen should be appropriate for the MEI, an
individual with higher than average intake of the affected food groups.
Consumption data presented in Pennington (1983) and reanalyzed in Table 4-2
are mean values for each of eight age/sex groups. One could define the MEI
in terms of the age/sex group having the highest consumption for all of the
food groups combined, or for each individual food group. Alternatively, one
could estimate a 95th-percentile consumption level based on variability of
1-day consumption; however, this procedure risks overestimation of long-term
4-70
-------
consumption. The appropriate use of dietary data to characterize the MEI is
more fully discussed in Section 4.1.4.
4.2.3.4. BACKGROUND CONCENTRATION IN CROPS (BC) — Background concen-
tration data are required for index crops (BCj) and substitute crops
(BCS) used to determine the plateau increment, PI (see Section 4.2.2.1.1.,
Step E). Whenever possible, values for BCj should be taken from the study
used to determine the plateau value, P, and reference sludge concentration,
RSC. Values for BCS should be taken from each study used to check RSC.
Background data are also required whenever a maximum pollutant appli-
cation rate based on phytotoxicity (RPM) is calculated (see Section
4.2.2.1.2., Step H). If possible, BC should be taken from the same study
from which TL was derived (see Section 4.2.3.5.).
4.2.3.5. MAXIMUM TISSUE CONCENTRATION (TL) — TL is the tissue
concentration limit, above which crop production is virtually eliminated by
phytotoxicity. TL is used as a limit to linear response, as explained in
Section 4.1.2.1.1. The maximum pollutant application rate based on phyto-
toxicity (RPM) must be calculated for each food crop. If a given crop was
used to determine the linear response slope, UC, for a food group, then TL
for the same crop should be used, if possible, to calculate RPM. If the
same crop cannot be used, or if UC was based on a weighted mean, then TL
should be based on that crop for which the highest tissue TL value is found.
4.2.3.6. WAITING PERIOD (T) - If a waiting period (T, in years)
occurs following the last sludge application before crops are grown, the
value of RPa for organic compounds will increase. (Refer to Section
4.1.1. for methods of calculating RPa, from RLC.) Ordinarily, no waiting
period is assumed. However, current regulations [40 CFR 257.3-6(b)(3)] call
for a waiting period of 1.5 years if the sludge has not been treated by a
4-71
-------
"process to further reduce pathogens" (PFRP) and if crops that contact the
sludge are grown. It is suggested that values of T = 1.5 years and T = 0
years be used for comparison when deriving criteria for land application
practices for this exposure pathway. For D&M practices, T = 0 should be
assumed.
4.2.3.7. ADJUSTED REFERENCE INTAKE" (RIA) — Values for adjusted
reference intake (RIA, in vg/day) are derived based on health effects
data, as described in Section 4.1.5.
4.3. SLUDGE — HUMAN TOXICITY (SOIL INGESTION) EXPOSURE PATHWAY
4.3.1. Assumptions. In addition to many of the assumptions listed in
Table 4-1, some additional assumptions are made for this pathway relating to
the degree of sludge ingestion that could occur and the method of assessing
potential effects. These assumptions and their potential ramifications are
summarized in Table 4-15 and are further discussed in the following sections.
4.3.2. Calculation Method. A reference soil concentration (RLC, in
yg/g DW) may be determined based on human health effects data, as follows:
RLC = RIA/(Is X DA) (4-28)
where:
RIA = adjusted reference intake (yg/day)
Is = soil ingestion rate (g DW/day)
DA = exposure duration adjustment (unitless)
The procedure for ensuring that RLC will not be exceeded depends on whether
or not the sludge is soil-incorporated. If it is not, then it will be
assumed that the uppermost soil layer, which children may ingest, is 100%
sludge. Therefore, sludge concentration should be controlled. If soil
incorporation is practiced, then pollutant application rate is the basis for
criteria calculation. For organic compounds, soil degradation must also be
4-72
-------
0)
=3
00
o
Q.
X
Ul
c
o
•I—
•P
Ul
co
Cn
c
£ £
00
c
o
•r—
•P
(O
•P
00
c
o
CO
O
CO
OS
oo
O
CO 3
=J -0
•P O
•r- S_
00 Q.
CO
i- y_
3 O
00
o c
o. o
3^ •("*
CO -P
(D
01 i.
CO O
•P 0.
(O t» •
E O "O
•r- O CO
•P «= O
00 »r- •>-
O) 4->
i-r- O
CO «r- to
> O i.
O 00 Q.
c
o
00
oo
O
00
> eu
+J
O
3
CO
CO
CO
C
o
U
c
3
_
s- >> a.
(O CO
oo £ TO
CyO CU
a> c +->
O
CU
a> +•> o
en co •(->
TO i-
3 o -o
c— Q.
3 Q. CO CO
O CX I. O
> co 3 T-
<4- CO oo u
•r— CD CO CO
TJ CX i-
> 3 CX
r— i— &_
C 00 O -P
O O
O 00 C
CO -P 4->
i- oo in
3 oo CO ••!-
oo oo i.
O CO O C
Q. <_I <4- O
X O T-
CO CO oo +J
CO CO
•Pr- x: O
CO T— O CX
^E CO ^3 &••
"- "O oo O
•P O
oo CU * C
CO > oo -r-
S- CO to
CU JC CO r—
•O i- •!-
C C CO O
3 CO oo
I— I 'I- S_
3 -r- +J CO
o x: co x:
o o o 3
cu +->
i- TJ CO O
3 C S- 3
oo cb co "O
a^cg
x^|^-
O i— O>
•P S- -I- +J
o x: 3
S- <*- O r—
O -t-
S- CU
Q. E
3
C 00
O oo
00 C
a> 3
u
co oo
S_ »r-
O
Q. C
i- CU
O i-
O XJ
r- o
o u- a. o
co o
(O
v. o
0.+-)
C T3
O CU
•i— oo
•4-> O
ca a.
o x
Id i-
JZ O)
C oo
a> ro
•4-> 01
X 5-
QJ to
a> c a.
i_
a. &- a>
X <4- X»
oo i — o
<1> •<- C
•M x:
O
oo •<- a*
O) Q. CD
S- T5
3
> JC i—
O •(-» oo
00 r—
>> eu i-
co +-> o
E co oo
S-
• i.
r- CO O
•I- jC <4-
o +->
oo (o
•P O
T3 CO «i-
cu a.
TJ C
c a>
cu i-
E -a
co r— in
I -r- CU
«U J= .1-
O3 O T3
•O 3
3 >+->
r^ XJ OO
OO
•M I
3 O)
O -O T3
•P (0 C
3
09 CD
r— .e c
-a +-> -i-
•P C
CX CO
CO -C
O -P
00
3 oo
oo T-
oo
CO CO
!w c
§CO
en
o
oo a
u
-o &-
I— CO
•i- o
CO
u
O (O
O M
(O
•(-> x:
c
oo C
3 O
CO CO
EO-r-
O) r- 4->
^= \ oo
U IT) CD
CO
3
CU co oo
o
O X
o o
u c
O CO
O T-
E • co co
CO i. ,
u ca o I
+J O
i. CO -r-
O -P
in xs
• a> a,
co en >
en c s_
•o -I- co
3 oo
r— cu .a
(O
x: i- oo JZ
•P CO 3 -P
CO T-S
T3 >>T3 i-
eo co co
E 10 jc
3
m o t— to
4-> 4_
CO >>
i— O S-
oo c CO
•i- E co co
O -P >
•P i- O I
•P
c
CO
oo
oo
CO
oo
oo
CO
U
CO
u-
LU
4-73
-------
taken into account. Equations relating soil concentrations to sludge appli-
cation rates, with and without degradation, are presented in Section 4.1.1.
4.3.2.1. SURFACE APPLICATION — It is assumed that surface applica-
tion, without soil incorporation, may occur for many D&M sludge uses. These
practices could result in exposure of children, either immediately or fol-
lowing land-use conversion. The calculations for determining the reference
sludge concentration, RSC (in yg/g DW) are as follows.
4.3.2.1.1. Inorganics — Since degradation ordinarily is not assumed,
the reference sludge concentration, RSC, is equal to RLC.
4.3.2.1.2. Organics — If immediate exposure potential exists follow-
ing application, the procedure for organics is the same as for inorganics.
This assumption is followed for the purposes of deriving criteria, since
conversion periods are neglected (see Chapter 3). In some instances,
however, it may be desirable to assume a land-use conversion period (T, in
years) before exposure of children is likely. Since degradation may occur,
the initial soil concentration prior to conversion, RLCQ, is calculated
from RLC by the following equation, which is derived from Section 4.1.1.2.,
Equation 4-4:
RLC = RLC x e
o
kT
(4-29)
The reference sludge concentration, RSC, is then set equal to RLCQ.
4.3.2.2. SOIL INCORPORATION — Soil incorporation is required in
agricultural utilization when crops for human consumption are grown, and
such a requirement will be evaluated for pasture lands as well (see Section
2.2.1.2.). If agricultural, forest or reclaimed lands are converted to
residential use, it is assumed that soil incorporation will occur during the
process of land conversion, if it has not already occurred. The reference
application rate of the pollutant (RP, in kg/ha) is determined from RLC as
4-74
-------
described in Section 4.1.1., using Equation 4-3 for inorganics, and Equation
4-6 or 4-13 for organics, for single or multiple applications, respectively.
4.3.3. Input Parameter Requirements.
4.3.3.1. SOIL INGESTION RATE (y-Soil ingestion has been recog-
nized as an important source of exposure to pollutants such as lead. For
adults, a value of 0.02 g/day has been used to estimate dust ingestion (U.S.
EPA, 1984b). Children may ingest soil by either inadvertent hand-to-mouth
transfer or by intentional direct eating. When such behavior is frequent,
it is called pica. Lepow et al. (1975) estimated that children frequently
mouthing their hands may inadvertently ingest at least 100 mg of soil per
day. Studies aimed at more accurately determining the range of ingestion
rates have yielded some data but are as yet inconclusive. Binder et al.
(1986) conducted a pilot study to establish methods for determining soil
ingestion rates in children living near a lead smelter. Three tracer
elements, aluminum, silicon and titanium, were measured in soil and dust
samples and in stool samples of 70 children. Results for the three elements
were often in disagreement, indicating either a metabolic loss or
unrecognized sources of one or more of the elements.
Assuming that the lowest estimate for a given child was accurate, and
that the higher estimates for the other two elements represented sources
other than soil, the following results were obtained (in g DW/day): for 59
children with soil and stool measurements, mean ingestion was 0.108; the
median was 0.088; the geometric mean was 0.065; the range was 0.004-0.708
and the 95th percentile was 0.386 (Table 4-16).
Estimates based on aluminum and silicon were possibly the most accurate,
since these were in relatively close agreement, and the titanium results
should probably be dismissed as anomalous (since these averaged higher by a
4-75
-------
j£
i
*
uu
CO
*
n
i/i
01
en
u_
e
e
0)
t_
VJ
c
e
o
5
o
w
r—
5
,>»
03
O
e
1
5
3
O
k.
-
us
•o
2
o
1/1
e
*"*
"o
>
I
r-
«-» e
en W
CU
Q.
01
e
«a
ae.
a
*
u
»
J
e
CO
0;
e
*
o e
at
o ^*
^3 ffll
^3 fis
O *"
O. U1
^^
e •*••
^j bt
^E fi|
V jB
r- C3
^U ^""
00
CP
CM
.
^B
I
LO
CM
O
O
CM
C3
CO
CM
*
o
5
o
,_
00
o
in
vO
e
c
«^B
e
^»
<
CO
Ln
*
0
en
i^b
I ™
C3
t
rt
Q
C3
hO
r—
0
O
eo
O
sO
CO
O
^
00
C3
in
*°
e
o
u
^«
«^>
CO
o
en
en
0
P^
f^
i
0
Q
C3
i
*
,_
o
e>
CO
i—
vO
*
O
«*•
en
00
^
in
vO
s
JJJ
(^
e
*«
«^
h—
CO
en
e— •
«
b.
•o
(B
oa
«•
a?
o
^^
3
O
^O
*
4-76
-------
factor of 10). This assumption produces slightly higher estimates, as shown
in Table 4-16. Based on these preliminary data, a value of 0.5 g/day is
suggested as a reasonably protective value for I . Further studies are
being undertaken by the U.S. EPA to improve these estimates.
4.3.3.2. ADJUSTED REFERENCE INTAKE (RIA) — Values for adjusted
reference intake (RIA, in yg/day) are derived based on health effects
data, as described in Section 4.1.5. Some special provisions may apply,
since this exposure occurs only in children (1-6 years of age).
4.3.3.2.1. Human Body Weight (bw) — A body weight of 10 kg, the
approximate median body weight at 12 months of age, should be used.
4.3.3.2.2. Total Background Ingestion Rate of Pollutant (TBI) — The
value of TBI should be based on a body weight of 10 kg. FDA data for
infants/toddlers may be available; otherwise it is suggested that ingestion
amounts be adjusted downward. A factor of 3 is suggested for such an
adjustment. TBI should not include pollutant ingestion in soil, since the
background concentration in soil is handled separately in Equation 4-3.
4.3.3.3. EXPOSURE DURATION ADJUSTMENT (DA) — An adjustment to the
RIA may be required based on the brief duration of this exposure. Values of
RfD and q^ are usually calculated to be representative of a lifetime
exposure. Adjustment of RfD values for exposure duration is not recommended
because higher exposures may lead to more severe toxic effects. However,
time adjustment of cancer risk estimates is consistent with the method in
which potency estimates (q^) are derived, and has been used previously.
Therefore, a DA value is suggested for use with carcinogenic chemicals. The
value is derived on the basis of exposure duration divided by assumed
lifetime, or 5 years/70 years = 0.07. This adjustment should be carefully
4-77
-------
evaluated on a case-by-case basis to ensure that an exposure of RIA/DA will
not lead to toxic effects other than carcinogenesis.
4.4. EXPOSURE PATHWAYS FOR HERBIVOROUS ANIMALS FOR HUMAN CONSUMPTION
4.4.1. Assumptions. Animal forage may be contaminated by uptake of
sludge pollutants through the plant roots (sludge-soil-plant-animal-human
toxicity) or by adherence of sludge to plant surfaces or roots [sludge-
animal (direct ingestion) human toxicity]. As explained in Section 3.1.3.,
soil incorporation is not assumed when sludge is applied to pasture, and
direct ingestion may occur. Therefore, the criteria calculation procedure
should be used to evaluate whether soil incorporation should be required.
In forest use, soil incorporation often is not feasible, and direct inges-
tion is possible by animals such as deer that will be taken by hunters in
unsludged areas. Deer and other game may also feed in agricultural areas
and be taken and consumed by hunters. However, the amount of game consumed
is assumed not to exceed the consumption of home-produced meat by farm
dwellers. Therefore, the farm dweller is taken as the MEI. Criteria that
protect the MEI are assumed to be protective of hunters as well. It is
assumed that sludge application to this farmland may have occurred as a
current or previous agricultural application practice, or as a previous
forest or disturbed-land application practice. Additional assumptions are
listed in Table 4-17. Various aspects of the assessments for these pathways
are shown in Table 4-18.
4.4.2. Calculation Method.
4.4.2.1. UPTAKE PATHWAY: SLUDGE-SOIL-PLANT-ANIMAL-HUMAN TOXICITY -- A
reference concentration of contaminant in animal feed (RFC, in yg/g DW)
4-78
-------
I/I
e •
<
vi
53
o
fc.
O
^^
n
fc.
35
^j
(^.
3
cn
^_
— _
ft
VI
^^
(Q
3
^j
fM
&
&_
o
U-
VI
c
o
^J
f*
s
VI
*
VI
e
0
«^"
^rf
t?
^«*
«^»
5
•^
_i
V,
VI
e
o
•^"
^J
flg
U
UM
•(••»
S
rg
as
0
•**
VI
VI
^£
gq
4>
fc.
^
^~
rg
e •
o
•*~
•*•>
u
c
3
U_
*
e
o
«^B
4^
S
S3
w.
O
«*••
•o
e
t*m
e
4>
^3
O
+*
^3
i
3
VI
VI
rg
VI
•^
i—<
^
4>
.C
r-
•o
4)
U
4f
•o
u-
O
C
o
*•<
u
rg
^
u.
3
O
£ e
VI •<—
•«•» VI
JC 41
cn en
•Z c
S rg
• u
4)
r**> *»
j2 C
rg rg
r- 0
«•— tf—
>
LB
*^»
rg
•o -o
4)
^ W
e s
rg T3
O
VI fc.
O> CL
cn i
S
« o
>»JC
fc.
•** y-
f— O
o o
^^ c^
rg
» ^^
*/* C
•M 4)
rg u
4) L-
E 4>
CL
4)
O JC
)••
VI
«^
jC •
VI
*o **
3
J= T3
u-o
3 fc.
S CL
e
0
*J
u
i^»
^"
CL
CL
C^
•<->
ffl
^
*J
i/i
4>
VI
•^
LU
4)
JC
^j
u_
O
t-t
41
•^
^5
^>
id
^^
e
•^
VI
^3
O
O
<4-
>l ,
o
IA
+J
•^
j2i
Vi •
3 vi
O ^0
jC r-
_ 0
S J=
rQ vt
a. 3
O
f— jC
fc. f—
3 (—
fc. rg
u- u-
0 O
^> ^^
4)
i- •O
3 4>
VI -«rf
3
— »**
tiO i^
VO H-.
O^ VI
r- C
-_- O
u
0 JC
«/> u
^3 «^™
jC
rg 3
1
e
»g
«*J
(/I
•fi
3
VI
O
e
o
•^
ts
E
fc.
O
u-
e
e
4>
en
rg
O
u_
•<->
rg
*d
4)
E
*^
cn
^^
•*••
3
CT
C
•»*•
E
3
VI
C
o
u
VI
r"»
rg
3
"O
*>
iff"
•o
c
l—t
4)
VI r—
"~ jCi
rg
C i—
o *•
**"* rg
^^ ^
O« rg
^^
VI (—
vi 4>
rg "i*4
^5
^ *^**
^M ^J
JC 4)
f» £
g
4) «-
*J
•2 "o
*. e
u
4)
*J
S
fc. •
en 4>
0 0
e ja
4)
> -o
rg 4)
UM
O •»-
c
*O 4^
41 -a
§t^
V> r— *
V) UJ
t3 3C
^^ ^^
fc. JC
rg *-
VI C
rg rg
4> JC
t. H-»
rg
41
^ *"
Cn i/i
•o o
3 CL
r— X
vi 4)
r- 4>
r— fc.
^^ M
U r—
^M E
*rg "5
^^
*5 O*
e e
V) N
^^ ^9
O fc.
L. en
o
^^ •
^^^B ^
J3 C U
o e
4) 4)
Jtt ** fc.
rg 3 4)
•*< J3 JC
CL -a
3 » rg
VI
** -o >»
e 4> ja
rg 4>
C U- TJ
*- 4»
rg re u
"c — a.
O C u-
(_> rg rg
o -o
VI V VI
•— e CL
S 4> O
E E fc.
C 1
rg 4) 4)
cn cn
J3 3 l_
r- O
e vi «-
o
-^ 4T O
w CP
4> -0 i—
O» 3 —
C r- O
1— 1 VI VI
e
4)
^^
e
o
VI
u
3
U
U
O
4)
cn
•o
3
^—
VI
y,,.
O
e
o
^
VI
4>
Cn
C
u
4)
L.
•f~
O
41
fc.
•o CL
e o
u
«
•O 4)
4) fc.
U 3
•<-> VI
U rg
rg CL
a. cn
c
o "s
C 3
l/l
> •
U r— -O
C C 4>
1 U
i— C 4)
w- 4) «fc-
O J= U_
VI -M rg
^j
U
•O 4>
4) vi
^ ^j
CL O
fc. CL
4) VI
> 4)
O U
O CL
*•* O
"O U
41 4)
-M en
P"* ^
i-" a>
3B rg
4>
.*
VI
e
O 4)
CL en
i/i rg
4) fc.
fc. O
IU
VI l—
O r-
S rg
4) 4->
•M 4)
VI
u_ 4)
O fc.
CL
4) 4)
CL fc. t->
• o 4)
r— O •«-
VI *•* ^
4> "O r—
vi 4) rg
C VI E
O 3 —
a. c
vi vi rg
41 *-
fc. 4)
CLJC
fc. O i->
4) 0 C
C »-
f— cn vi
4) fc. O
JC O fc.
r— «- U
4-79
-------
• *
c
o
g-iB
3 U
«£ o>
o 2
c *°
S «B
•1 Cha
= V.
• 2
o
i^ **"*
L*
t/i -^^
. !•
T £ »
^- °
LL. 3 O
S o *-
»— >
!S ***
. c
0* i=
«K ^
el
""I
VI*
>•<*.
1°
5.£
°" 0>
o> £
3 ^^
VI
O
Q.
X
e
o —
S CU
t- -I-*
u_ L.
0
to
•^
_l
u.
u.
•o
0>
u
u
^g
VI
^J
(B
0)
^
0>
a
c
"*""
e
0
u.
a>
o
^^
u
e
x
<^_
fQ
3
•^
IB
0.
Q.
ce.
=
•*• o ^- oo
o o o o
_^
0
a
»
1
r- >»
(_
• >>•*>«
U- t. I— VI
u — 3 en
0^ ^S ^3 ^7^
^3 ^3 O» a^
VI
•o
o>
.•*•
VI fm
vi 52
o)"c
e u
en's
VI
a.
o
L.
U
9}
3
VI
g^
L.
O
^
o
9>
^B
^^1
a.
=3
a.
o
r- •
C3
II
^
^ O
0 0
r-
^
O
1
rO
r~
* >^
^^ ^B
IS
0> -M
•o o
e o
rt3 ^3
1 C
9)
cu
j=
•^3
^
CO
at
A
o oo
PO O
o o
II
u- O
•r o
o o
g»
o
1
^
1^
• >*»
U- i.
O) »-
9> IB
A -o
S,
2
C vi
o a» j=
u u
d> IB ^*
O>O- (B
3 3 **
r- vi
vi e
9) C
t.
W VI iB
^
a> ••-» o
v. 3 a.
3 O t-
-M ^ O
VI <4-> U
/B •*" C
a. 3 —
•
VI
O)
*
•B
O
o>
c
IB
0.
_
I/I
•o
o
««~
^B
9>
a.
en
•*-»
3
>,
•o
1
o
CO
IB
O
u.
o
t_
3
• c
*•» o
X
9t I/I
*•• ^
e
4> 0)
o a.
vi a>
•o
•»
C 0)
0 3
•^ ^"
% >
r- O
a.
X 9>
fl^ O
U *O
O £
U*> ^J
IB ^
4-80
-------
may be calculated based on human health effects data, uptake by animal
tissues and consumption of those tissues by humans:
RIA
RFC =
i x DA-i x FA-j)
(4-30)
where:
RIA = adjusted reference intake of pollutant in humans
= uptake response slope of pollutant in the ith animal
tissue [yg/g tissue DW (yg/g feed DW)-i]
= daily dietary consumption of ith animal tissue food group
(g OW/day)
FAi = fraction of food group assumed to be derived from sludge-
amended soil or feedstuffs
RFC is the increase in feed concentration of the animal's total diet that
will cause contaminant intake in exposed individuals to equal the RIA. The
actual feed concentration includes background concentration as well and
therefore may be higher.
Following the calculation of RFC, the reference application rate of the
pollutant (RP, in kg/ha) is calculated. The procedure differs for category
1 and category 2 contaminants, since the units of crop uptake differ.
4.4.2.1.1. Category 1 Contaminants — For inorganics that are
conserved in the soil, a cumulative application rate (RP , in kg/ha) is
calculated:
RP = RFC/UC
(4-31)
where:
RFC = reference feed concentration of pollutant (yg/g OW)
UC = linear response slope of forage crop [yg/g crop DW (kg/ha)"1]
4.4.2.1.2. Category 2 Contaminants — For organic compounds and
inorganics that are not conserved in the soil, before determining pollutant
application rate, a reference soil concentration (RLC, in yg/g DW) is
calculated:
RLC = (RFC/UC) + BS (4-32)
4-81
-------
where:
BS = background soil concentration of pollutant (yg/g DW)
UC = linear response slope of forage crop [yg/g crop DW (yg/g
soil
The procedure for estimating application rate of pollutant, RP, from RLC is
given in Equations 4-6 and 4-13 for single and multiple applications,
respectively.
4.4.2.2. ADHERENCE PATHWAYS: SLUDGE-ANIMAL (DIRECT INGESTION) — HUMAN
TOXICITY — The reference feed concentration (RFC, in yg/g DW) is calcu-
lated as for the uptake pathway (see Equation 4-30) but the values will
differ because fewer human food categories will be affected. If soil incor-
poration is assumed, the upper parts of crops are assumed to be uncontami-
nated in agricultural use, and thus only grazing animals (e.g., cattle,
sheep) are affected. If soil incorporation is not assumed, the upper parts
of pasture crops may be contaminated and may be harvested as feed for cattle
or sheep, or they may be grazed directly by these animals. Sludge can also
accumulate in the thatch layer of pasture and can be directly consumed by
grazing animals. Domestic animals consuming grains or other nonpasture
crops (such as pork, poultry) are assumed not to be affected.
4.4.2.2.1. Soil Incorporation of Sludge -- Following the calculation
of RFC, RFC is adjusted by the amount of soil in the diet to determine the
reference soil concentration (RLC, in yg/g DW) :
RLC = (RFC/FL) + BS (4-33)
where:
RFC = reference feed concentration of pollutant (yg/g
FL = fraction of diet that is adhering soil (g soil DW/g diet DW)
BS = background soil concentration of pollutant (vg/g DW)
4-82
-------
A pollutant application rate, RP, is calculated from RLC as described in
Section 4.1.1. Equation 4-2 or 4-3 is used for inorganics; Equation 4-6 or
4-13 is used for organics.
4.4.2.2.2. Surface Application Without Soil Incorporation — When
sludge is consumed directly from plant or soil surfaces, contaminant intake
by the grazing animal is more closely related to sludge concentration than
contaminant application rate. Therefore, criteria are derived in terms of a
reference sludge concentration (RSC, in yg DW/g) rather than RP. For
inorganics and organics, RSC is derived as follows:
RSC =
where: FS
FS = fraction of animal diet which is sludge (g sludge DW/g
diet DW).
(4-34)
If a waiting period (T, in years) is assumed and occurs between the most
recent sludge application and grazing, RSC may be multiplied by the expres-
kT
sion e , where k is the loss rate constant (in years'1).
4.4.3. Input Parameter Requirements.
4.4.3.1. FRACTION OF FOOD GROUP ASSUMED TO BE DERIVED FROM SLUDGE-
AMENDED SOIL OR FEEDSTUFFS (FA) — As was the case with FC in the crops-
for-human-consumption pathway (see Section 4.2.3.1.), this parameter deter-
mines which food groups are included in the analysis.
For the first of these two pathways (sludge-soil-plant-animal-human
toxicity), all meat groups except fish will be assumed to be affected.
These include beef, lamb, pork, poultry, dairy products and eggs. The
second pathway [sludge-animal(direct ingestion)-human toxicity] is assumed
to affect only grazing animals; beef, lamb and dairy food groups are
included.
4-83
-------
As for the pathway dealing with crops for human consumption, the MEI for
these two pathways is chosen as a farm family raising a substantial percent-
age of their own meat and other animal products. The choice of FA values is
based on the percentage of homegrown foods consumed by rural farm households
(see Table 4-14). These values have been reiterated in Table 4-18. As
stated previously, it is assumed that these FA values are sufficiently high
that an individual consuming wild game from sludged areas will also be
protected.
4.4.3.2. DAILY DIETARY CONSUMPTION OF FOOD GROUP (DA) — Values for
daily dietary consumption (DA, in g DW/day) are needed for each food group
for which FA#). As was described for DC (see Section 4.2.3.3.),
consumption data are taken from Table 4-2 or other information presented in
Section 4.1.4. The food item "beef liver" in Table 4-2 includes various
other organ meats consumed by humans in smaller amounts, such as kidneys and
hearts. Individuals with a preference for those organs are expected to
consume them at the rates given for beef liver. It is also assumed that
consumption of wild game does not exceed the values of DA for other meats
(beef, lamb) and that criteria derivation procedures based on consumption of
these meats will also protect hunters.
4.4.3.3. FRACTION OF ANIMAL DIET THAT IS SOIL (FL) OR SLUDGE (FS) —
Studies of grazing animals indicate that soil ingestion ordinarily ranges
from 1 to 10% of dry weight of diet (but may range as high as 20%) for
cattle and may be 30% or higher for sheep during winter months when forage
is reduced (Thornton and Abrams, 1983). Since lamb contributes relatively
little to the U.S. diet, a value of FL = 10% or 0.10, based largely on
cattle, will be used to represent a reasonably high exposure situation.
4-84
-------
Studies of sludge adhesion to growing forage following applications of
liquid or filter-cake sludge show that when 3-6 t/ha of sludge solids is
applied, clipped forage initially consists of up to 30% sludge on a dry-
weight basis (Chaney and Lloyd, 1979; Boswell, 1975). However, this contam-
ination diminishes gradually with time and growth, and generally is not
detected in the following year's growth. Where pastures amended at 16 and
32 t/ha were grazed throughout a growing 'season (168 days), average sludge
content of forage was only 2.17 and 5.17%, respectively (Bertrand et al.,
1981).
In the draft Las Vegas report, Chaney et al. (1987) review these and
other data on sludge adherence to forage crops, and show that by 21-28 days
following sludge application, the amount of sludge in or on forage was
0.5-5.4% (DW/DW), depending on the sludge moisture content and forage
species. Sludge in the feces of cattle rotationally grazed on these
pastures at least 7-21 days following sludge application was slightly
higher, as high as 7.7%, probably indicating some ingestion of sludge from
the soil surface. Therefore, a value of FS = 30% or 0.30 is recommended if
no waiting period is employed before grazing or harvest of sludge-amended
pasture. A value of 0.08 is recommended if a waiting period of T = 0.082
years (30 days) is employed.
4.4.3.4. UPTAKE SLOPE OF POLLUTANT IN ANIMAL TISSUE (UA) -- Uptake
slopes for affected tissues consumed by humans should be calculated as
described in Section 4.1.2.3. The form and availability of uptake response
data will affect the types of food groups included in the calculation of RFC
(see Equation 4-30) and therefore the choice of values for FA and DA as
well. If values of UA can be calculated for several different types of
meats (beef muscle, pork muscle, beef liver, etc.), then each of these meats
4-85
-------
can be separately included in Equation 4-30. If data for only a single type
of meat are available, it may be necessary to as'sume that similar types of
meat from different species have similar UA values. These tissues would
then be grouped in Equation 4-30. It should not be assumed, however, that
organ meats and muscle have similar UA values, since organ meats often
accumulate more highly.
This analysis also assumes that UA values for wild game species are no
greater than those for the domestic animals to which most available data
apply. If data contradicting this assumption are available, these values
may be used in Equation 4-30 to calculate RFC.
4.4.3.5. LINEAR UPTAKE RESPONSE SLOPE OF FORAGE CROP (UC) -- The crop
chosen for UC should be the one showing the highest response slope appro-
priate for the exposure scenario involved. Grains, legumes (such as soy-
beans), silage or grasses could be affected in agricultural use.
4.4.3.6. ADJUSTED REFERENCE INTAKE (RIA) — Values for RIA (in
yg/day) are derived based on health effects data as described in Section
4.1.5.
4.4.3.7. WAITING PERIOD (T) — If a 30-day waiting period is assumed
as discussed previously (see Section 4.4.3.3.), then a value of T = 0.082
years is used in calculations dealing with the nonincorporation scenario
(see Section 4.4.2.2.2.).
4.5. EXPOSURE PATHWAYS FOR TOXICITY TO HERBIVOROUS ANIMALS
These two pathways [sludge-soil-plant-animal toxicity; sludge-animal
toxicity (direct ingestion)] are similar to the two just discussed.
However, since toxicity to the animal itself is now the endpoint of concern,
the list of animals to be considered is broadened to include all herbivores
found in the agricultural or forest environment. For example, herbivorous
4-86
-------
rodents or birds should be considered, as well as large herbivores and other
domestic animals. However, the pathway for adherence of agricultural soils
to plant roots still is limited to grazing animals that ingest significant
amounts of soil (cattle, sheep).
4.5.1. Assumptions. The assumptions pertaining to this pathway have been
stated in Tables 4-1 and 4-17.
4.5.2. Calculation Method. The calculations for these pathways are the
same as given in the previous section, except that a feed concentration
corresponding to a toxicity threshold in a sensitive herbivore should be
substituted for the reference feed concentration, RFC, in all calculations.
Since RFC was defined as the increase in feed concentration resulting in an
adverse condition in humans, RFC is redefined for these pathways as follows:
RFC = TA - BC (4-35)
where:
RFC = reference feed concentration (pg/g OW)
TA = threshold feed concentration (pg/g DW)
BC = background concentration in feed crop (pg/g DW)
Equations 4-31 to 4-34 are then used to determine criteria, in the form of
either RP or RSC.
4.5.3. Input Parameter Requirements. The focus of the methodology for
these pathways is the identification of the appropriate threshold feed con-
centration, TA. Procedures for selecting TA are generally discussed in
Section 4.1.3. An important source of relevant data for inorganic chemicals
is found in NRC (1980), "Mineral Tolerances of Domestic Animals." Values of
TA are needed for various types of herbivorous animals, including grazing
animals, many birds and many small mammals.
4.6. SLUDGE-SOIL-PLANT TOXICITY EXPOSURE PATHWAY
4.6.1. Assumptions.
4-87
-------
The assumptions pertinent to this pathway have been stated in Table 4-1.
4.6.2. Calculation Method.
Determination of criteria for this pathway is straightforward, involving
few or no calculations. A threshold phytotoxic application rate of the pol-
lutant, TP (in kg/ha), can in some cases be derived directly from
phytotoxicity studies. If the pollutant is not subject to loss, then a
reference cumulative application rate, RP (in kg/ha), is simply
RPc = TP (4-36)
If the pollutant is subject to loss from soils, then RP is derived by
substituting TP for the expression RLC x MS x 10 either in Equation
4-6, if a single application is to be used, or Equation 4-13, if multiple
applications are to be used.
If instead of TP, a value of the threshold phytotoxic soil
concentration, RLC (in yg/g DW), is derived, then RP is calculated using
either Equation 4-3 (for nondegrading pollutants) or 4-6 or 4-13, as
appropriate (for degrading pollutants).
4.6.3. Input Parameter Requirements. The focus of the methodology for
this pathway is the selection of the appropriate application rate (TP) or
soil concentration (RLC) that corresponds to the threshold level for adverse
effects in plants. The threshold is generally defined in Section 4.1.3. as
the geometric mean of the lowest exposure level causing, and the highest
level not causing, a significant adverse effect. The terms "significant"
and "adverse" are further defined in that section.
As for uptake data, it would be preferable to derive these values from
sludge application studies conducted in the field. When such data are not
available, less appropriate data, such as from greenhouse studies or pesti-
cide application studies, may be used.
4-88
-------
Values of TP or RLC may be available for numerous plant species for a
given chemical. Ordinarily the recommended procedure is to identify the
most sensitive species and ensure that it will be protected by the value
selected, unless that species does not grow in the United States (or in a
given region, if site-specific criteria are derived).
Phytotoxicity of metals may be altered by soil pH. If TP or RLC values
can be segregated based on pH (i.e., >6.0 or <6.0), then separate criteria
can be established on this basis as well. However, criteria based on pH
>6.0 should be applied only where soil pH is expected to remain >6.0 without
addition of lime.
4.7. EXPOSURE PATHWAYS FOR TOXICITY TO SOIL BIOTA AND THEIR PREDATORS
This section deals with two pathways: toxicity to soil biota (sludge-
soil-soil biota toxicity) and toxicity to predators of soil biota (sludge-
soil-soil biota-predator toxicity). As explained in Section 3.1.6., the
term "soil biota" is intended to refer to a broad range of organisms includ-
ing microorganisms and various invertebrates living in or on the soil.
Their predators similarly include a variety of organisms. The availability
of data determines what species are considered.
4.7.1. Sludge-Soil-Soil Biota Toxicity Exposure Pathway. Procedures for
this pathway are identical to those described above for phytotoxicity
(Section 4.6.), except that the threshold levels, TP and RLC, deal with
effect thresholds in soil biota rather than in plants. As was touched on in
Section 4.1.3., "adverse" effects can be particularly difficult to define
where microorganisms are concerned. It is assumed that long-term reductions
in soil microbial activity or diversity that can be attributed to the
chemical should be considered adverse in lieu of information to the contrary.
4-89
-------
4.7.2. Sludge-Soil-Soil Biota-Predator Toxicity Exposure Pathway.
4.7.2.1. ASSUMPTIONS — In addition to many of the assumptions listed
in Table 4-1, some additional assumptions pertinent to this pathway are
stated in Table 4-19.
4.7.2.2. CALCULATION METHOD — Calculations of criteria for this
pathway may take either of two forms, depending on the type of data
available concerning contaminant uptake by soil biota. If uptake response
(that is, the increase of concentration) in soil biota, UB, can be expressed
in terms of a pollutant application rate,, then criteria are calculated as
follows:
TA - BB
RP =
UB
(4-37)
where:
RP = reference application rate of pollutant (kg/ha)
TA = threshold feed concentration for the predator (yg/g DW)
BB = background concentration in soil biota (yg/g DW)
UB = uptake response slope in soil biota [yg/g (kg/ha)-1]
If the chemical is not subject to degradation or loss in soil, RP is a
cumulative rate (RP ). Otherwise, the value of RP from Equation 4-37 can
I*
be substituted for the expression RLC x MS x 10~3 in Equation 4-6 to
derive a single-application rate, RP , or in Equation 4-13 to derive an
annual rate, RP .
3
If soil biota response is measured in terms of a soil concentration, as
is more often the case, the following equation is used:
TA - BB
RLC =
UB
+ BS
(4-38)
4-90
-------
TABLE 4-19
Assumptions for Sludge-Soil-Soil Biota-Predator
Toxicity Exposure Pathway
Functional Area
Assumption
Ramifications/Limitations
Contaminant uptake by
soil biota
Use of available data
to protect a variety
of species
Response in tissues
of soil biota can be
represented by a
linear function.
It is assumed that use
of the highest avail-
able response slope in
soil biota and the low-
est available dietary
threshold in predators
will result in protec-
tion of untested species,
Probably oversimplifies
a more complex relation-
ship.
This conservative assump-
tion could be overprotec-
tive of some species; the
extent to which it under-
protects others is
unknown. A "match" of
data for a consumed organ-
ism and its predator
usually is not possible.
4-91
-------
where:
RLC = reference soil concentration of pollutant (yg/g DW)
UB = uptake response slope in soil biota [y/g (yg/g)~l]
BS = background soil concentration of pollutant (yg/g DW)
and other parameters are as defined above. RLC is then used in Equation 4-3
(for nondegrading chemicals) or Equations 4-6 or 4-13 (for degrading chemi-
cals), as appropriate.
4.7.2.3. INPUT PARAMETER REQUIREMENTS —
4.7.2.3.1. Threshold Feed Concentrations for the Predator (TA) — A
wide variety of organisms may prey on soil invertebrates, including birds
and insectivorous mammals. For a given chemical, information on subchronic
or chronic toxicity for oral administration may be available for only a few
species, and thus the value chosen for TA may be for a species not actually
preying on soil biota. In general, the lowest dietary adverse effect level
(and the accompanying no-adverse-effect level) found for birds or small mam-
mals are used to determine the threshold. The threshold is calculated as
discussed in Section 4.1.3.
4.7.2.3.2. Uptake Response Slope in Soil Biota (UB) — UB may take
any of several forms, depending on the characteristics of the chemical and
the data available. UB is derived by linear regression of tissue concentra-
tion on either pollutant application rate or soil concentrations, depending
on which is reported. Use of a pollutant application rate is preferable.
As for plant and animal tissues (see Section 4.1.2.), two or more points are
needed to derive the slope for inorganics; for synthetic organic compounds a
single data pair can be used to derive a bioconcentration factor.
The highest available uptake response slope will be used to estimate the
level of dietary contamination to which predators of soil biota would be
subject. This fact is important to bear in mind when evaluating data for
4-92
-------
earthworms. Some studies go to lengths to distinguish between contaminant
physiologically absorbed and that which is due to the gut contents (the
"cast") or soil contamination of the sample. While one may "suspect that the
absorbed contaminant is rendered more available to the higher trophic
levels, this is probably not proven or quantified. Since a predator would
ingest the gut contents as well as the rest of the organism, there is no
need to distinguish between absorbed and unabsorbed contaminant. Wherever
possible, analyses used should be based on the whole organism.
4.7.2.3.3. Background Concentration in Soil Biota "(BB) -- This value
should be obtained from the same study from which UB was derived.
4.8. EXAMPLE CALCULATIONS -
In this section, examples will be provided for each of the criteria cal-
culation procedures for the different terrestrial food-chain pathways des-
cribed in Sections 4.2-4.8.
One of the most important steps in criteria calculations is the selec-
tion of values for each of the parameters involved. Value selection must be
based on careful literature searches and evaluations of available data. For
many parts of this methodology, value-selection will be a very data-inten-
sive exercise. Since the calculations presented in this section are only
for the purpose of examples, the values employed will not be based on liter-
ature searches, but will rely on a few'sources of readily available informa-
tion. Therefore, the results do not constitute actual recommendations for
criteria. The toxic metal cadmium and the organic compound hexachloro-
benzene (HCB) are used for the following examples.
4-93
-------
4.8.1. Sludge-Son-Plant-Human Toxicity Exposure Pathway.
4.8.1.1. PROCEDURE BASED ON CURVILINEAR ("PLATEAU") UPTAKE RESPONSE
MODEL AND RELATIVE UPTAKE RESPONSE VALUES —
Step A. Determine Relative Uptake Response Values for Each Crop
Equation 4-17
UCA
RUAI -
UCJ
where:
= uptake response for crop A relative to index crop
(unitless)
- linear response slope of crop A [yg/g (kg/ha)-1]
UCj = linear response slope of index crop [yg/g (kg/ha)-1] in
the same experiment in which UCA was measured
Equation 4-18 (an alternative procedure to Equation 4-17)
A2 - A-|
~ 12 - 11
where A_ denotes tissue concentration of crop A in soil 2, etc.
Data Examples for Cadmium (Source: Dowdy and Larson, 1975)
Crop type
Linear response slope, UC [yg/g (kg/ha)-*]
Tissue concentration (mean) (yg/g)
Control tissue concentration (yg/g)
Example Calculation of Equation 4-17
RU = 0.20/0.605 = 0.33
Example Calculation of Equation 4-18
Crop
A
Carrot
0.20
0.93
0.48
RUAI =
0.93 - 0.48
1.89 - 0.61
= 0.35
Index Crop
I
Lettuce
0.605
1.89
0.61
4-94
-------
In this last example, the mean of tissue concentration from three treat-
ments was used for A2 and l^ and the control concentration tissue con-
centration was used for A and I .
steP B. Determine Relative Uptake Response Values for Each Food Group
Data Examples for Cadmium
For the "root vegetable" food group, two crops have values of RU (Dowdy
and Larson, 1975). Consumption data from Appendix 1 are used to derive a
weighted average RU. Consumption data for the 25- to 30-year-old male are
used.
Crop type Carrot
Relative uptake response value, RU (unitless) 0.33
Dry-weight consumption (g DW/day) 0.55
Radish
0.093
0.021
Example Calculation for Weighted Mean (Equation not presented in text)
RUroot veg. - <0-33x0.55) * (0.093 x 0.021) =
0.55 -i- 0.021
5i£e—C. Determine the Reference Tissue Concentration Increment for the
Index Crop
Equation 4-20
RTI =
RIA
I (RU-j x OC-j x FC-j)
where:
RTI
RIA
RUj
FCn
= reference tissue concentration increment in index croo
(yg/g DW)
= adjusted reference intake (jag/day)
= relative uptake response for ith food group (unitless)
= daily dietary consumption of ith food group (g DW/day)
= fraction of food group assumed to originate from
sludge-amended soil (unitless)
4-95
-------
Data Examples for Cadmium
Values of RU in this example are based on very few crops (data of Dowdy
and Larson, 1975). Values of DC are obtained from Table 4-2 (and Appendix 1
for legumes). DC values chosen are those for the age/sex group with highest
consumption of that food group. Values of FC are those for the home garden
scenario, from Table 4-12. The only legume data available in this example
data set are for pea pods and pea fruit. The mean RU value will be used for
dried legumes as well. The RIA value for cadmium is not based on Equation
4-14, as would normally be the case, since an Agency-approved RfD value is
still pending. Instead, the FAO/WHO (1972) provisionally tolerable daily
intake of 57-71 pg/day is used to derive a mean of 64 yg/day. When the
Agency's RfD value for cadmium is determined it should be used to derive the
RIA based on Equation 4-14. The value of BCj in this example is 0.61 jjg
Cd/g and is that for lettuce from Dowdy and Larson (1975). In actuality,
BC should be taken from the study from which RSC is derived.
Food Group
Potatoes
Leafy vegetables
Legume vegetables, nondried
Legume vegetables, dried
Root vegetables
Garden fruits
Grains and cereals
Peanuts
Mushrooms
TOTAL
RU
DC
FC
RU X DC x FC
0.063
1.0
0.008
0.008
0.32
0.12
31.85
2.78
3.38
8.51
2.28
5.94
0.45
0.60
0.60
0.17
0.60
0.60
0
0
0
0.903
1.668
0.016
0.012
0.438
0.428
0
0
0
3.465
Example Calculation
RTI =
64 ug Cd/dav
3.465 g/day
= 18.5
Cd/g
4-96
-------
Steps D and E. Sort Available Uptake Response Data for the Index Crop, and
Determine Plateau Increment Values for the Index Crop
Data Examples for Cadmium
Examples are not readily available for these very data-intensive steps.
Therefore a hypothetical data set for response in lettuce will be created to
illustrate these steps. For each hypothetical study, it is stated whether
soil pH (when the plateau was achieved) was <6.0 or >6.0, and whether the
plateau was observed in the first year of sludge application or after >1
year. The sludge concentration (SC, in pg/g DW) of cadmium is listed, and
the plateau increment value (PI, in vg/g DW), which is assumed to have
been confirmed by appropriate statistical techniques and represented by a
95% confidence limit, is also listed.
Stud\
1
2
3
4
5
6
7
8
<6.0
>6.0
Year
first
multi
first
multi
SC
DM)
200
10
200
15
300
300
200
15
PI
(ug/g DW)
50
5
30
3
20
10
12
2
Step F. Determine Reference Sludge Concentration (RSC) Not Causing Reference
Tissue Concentration Increment (RTI) to be Exceeded
Four groupings of studies 1-8 are possible based on soil pH and on
whether the data are from a first-year or multi-year study, as shown in
Table 4-8. ;..''...
Group A: Soil pH >6.0 or <6.0; multi-year or first-year data
All eight hypothetical studies can be included in this category, but
studies with pH <6.0 and first-year response data are included only if
necessary to round out the available data. Studies 6, 7 and 8 yield three P
4-97
-------
values (10, 12 and 2 yg Cd/g, respectively), all of which are well below
the RTI value of 18.5 yg Cd/g. Assuming that these sludges represent a
variety of sludge types and study conditions, the highest SC value, 300 yg
Cd/g, can be selected as the RSC. Examination of all other uptake data
(i.e., including studies not showing a plateau) must also show that no
sludge with SC <300 yg Cd/g will cause RTI to be exceeded in a multi-year
study using soil pH >6.0, regardless of the annual or cumulative application
rate. If this were an actual value, based on expert evaluation of all the
available data, this would indicate that all sludges having SC of 300 yg
Cd/g or lower could be applied at unlimited cumulative rates to soils
expected to remain >6.0 without pH control measures. Although no cumulative
rate would apply, an annual Cd rate would apply since crops in the first
year of sludge application could have a higher cadmium content (see Table
4-8). The annual Cd rate would have to be less than the highest rates
employed in study 5, for example, since RTI was exceeded at the plateau.
The annual rate is determined based on further evaluation of rates used in
studies 5-8, or based on the procedure using the linear model.
Group B: Soil pH <6.0 only; multi-year or first-year data
Studies 1-4 are potentially included in this group; studies 1 and 2 are
included only if needed to augment the available data. Studies 3 and 4
Indicate that sludges with SC of 200 yg Cd/g may cause RTI to be exceeded,
but those with SC of <15 yg Cd/g will not. Studies 1 and 2 strengthen
this conclusion; therefore, RSC is set at 15 yg Cd/g. If an actual value,
this would mean that sludges not exceeding this cadmium concentration could
be applied in unlimited cumulative amounts to all soils regardless of pH.
An annual application rate would still apply, however.
4-98
-------
Group C: Soil pH >6.0 or <6.0; first-year only
Studies 1, 2 and 5 potentially fit this group. Study 5 shows only that
one sludge with SC of 300 yg Cd/g will just cause RTI (18.5 yg Cd/g) to
be exceeded. Therefore studies 1 and 2 must be included as well, and they
indicate that an SC of 200 yg Cd/g caused RTI to be exceeded in the first
year, but that an SC of 10 yg Cd/g did not. Based on these observations
an RSC of 10 yg Cd/g is determined. If an actual value, sludges not
exceeding this concentration could be applied with no annual or cumulative
rate limit to soils expected to remain at pH >6.0 without liming.
Group D: Soil pH <6.0 only; first-year data only
Only studies 1 and 2 may be considered. These studies show only that an
SC of 200 yg Cd/g caused RTI to be substantially exceeded and that an SC
of 10 yg Cd/g stayed well under the RTI. RSC would be set at 10 yg Cd/g
if the data were judged to be reliable. If an actual value, sludges not
exceeding RSC = 10 yg Cd/g could be applied to all soils regardless of pH
and without annual or cumulative rate limits.
steP
-------
Example Calculation
RTI = 18.5 pg Cd/g x 0.33 = 6.1 pg Cd/g
O
Steps D-F are then repeated using plateau data and other response data
for the substitute crop. RSC is adjusted downward if necessary.
4.8.1.2. PROCEDURE BASED ON LINEAR UPTAKE RESPONSE MODEL AND RELATIVE
UPTAKE RESPONSE VALUES —
Steps A-C
Steps A-C are identical to Steps A~C using the curvilinear uptake
response model. For cadmium, an RTI of 18.5 pg Cd/g is derived.
Steps D-E
Steps D-E utilize a sorting procedure identical to that described for
the curvilinear approach, except that values of the uptake response slope
(UC) for the index crop are sorted, rather than plateau values (P).
Step F. Determine Reference Application Rate of the Pollutant (RP)
Equation 4-23
RP = RTI/UC,
where:
RP
RTI
'I
= reference application rate of pollutant (kg/ha)
= reference tissue concentration increment of index crop
(yg/g
UCj = uptake response slope of index crop [pg/g (kg/ha)-i]
Data Examples for Cadmium
Index crop RTI
Lettuce TsTs
Example Calculation
RP = (18.5 pg Cd/g)/[0.605
= 30.6 kg Cd/ha
UCi
0.605
"1
Cd/g (kg Cd/ha)"]
4-100
-------
Step G. Check the Reference Application Rate of the Pollutant (RP) by
Substituting Other Crops for the Index Crop
Data Examples for Cadmium
RTI is derived using Equation 4-21, as shown previously.
UCs
Substitute Crop
Carrot
RTIs
6.1
0.20
Example Calculation (see Equation 4-23. above)
RPS = (6.1 pg Cd/g)/[0.20 pg Cd/g (kg Cd/ha)-i]
= 30.5 kg Cd/ha
RPS is in virtual agreement with RP (except for rounding error)
because both of the UC values used were also used to derive values of RU and
RTC. This will not always be the case, however, since values of UC from
different sorting groups established in Steps D and E will be used to calcu-
late various values of RP, as shown in Table 4-10.
Step H. Ad.iust the Reference Application Rate of the Pollutant (RP) for
Phytotoxic Effects
A maximum pollutant application rate RPM. (kg/ha) based on phyto-
toxicity is determined for each crop group used to calculate RTC.
Equation 4-24
where:
RPMi = (TU - BCi)/(RUi x UCj)
TLi = maximum tissue concentration for crop category i, above
which crop production is virtually eliminated by phyto-
toxicity (pg/g DW)
BCi = background concentration for crop category i (pg/g DW)
RU-j = relative uptake response value for crop category i
(unitless)
UCj = uptake response slope for index crop, used to determine
the RP value being adjusted [pg/g DW (kg/ha)-i]
4-101
-------
Data Examples for Cadmium
Data for RU, DC and FC (which will be needed below) are those used above
to determine RTI. Data for TL and BC are matched values (i.e., from the
same study) and are selected or estimated from U.S. EPA (1985e).
Croc
RU
TL
NA
668
28.2
28.2
29.8
15
BC
NA
12.2
5.7
5.7
2.4
0.1
DC
31.85
2.78
3.38
3.38
2.28
5.94
FC
0.45
0.60
0.60
0.17
0.60
0.60
RPMi
NC
1084
4649
4649
142
205
t. (subscrip
0.903 (k)
1.668 (k)
0.016 (k)
0.012 (k)
37.5 (j)
53.1 (j)
Potatoes 0.063
Leafy vegetables 1.0
Legume vegetables,
nondried 0.008
Legume vegetables,
dried 0.008
Root vegetables 0.32
Garden fruits 0.12
RPM, for each crop group is compared with RP. If RPM. |