ID Number: RX000006380 Call Number: EPA 0765
Title: Compilation : 1) regulation of municipal sewage
sludge under the clean water act section 503, 2) heavey
metals and toxic organic pollutants in MSW-composts, 3)
treatment of sludge for land application
-------
0765
Ryan, J.A. and R.L. Chaney. 1992. In H.A.J. Hoitink et al. (eds.). Proc.
International Composting Research Symposium. In Press
Regulation of municipal sewage sludge under the Clean Water Act Section
503: A model for exposure and risk assessment for MSW-compost. 1
James A. Ryan
US Environmental Protection Agency
Risk Reduction Engineering Laboratory
5995 Center Hill Road, Cincinnati, OH 45224
and
Rufiis L. Chaney
USDA-Agricultural Research Service
Environmental Chemistry Laboratory
Bldg. 318, BARC-East, BeltsviUe, MD 20705
ABSTRACT
Efforts have not begun to regulate MSW-compost at the Federal level, nor In most states.
However, municipal sewage sludge regulations under the Clean Water Act Section 503 have
been proposed and are being finalized. The CWA-503 regulations will be applicable to
municipal sewage sludge products including MSW-sludge composts. MSW-compost, being
similar to municipal sewage sintige-compost and having similar uses, will ultimately be
required to meet similar standards or at least similar methodology will be used to develop
standards for MSW-compost. Therefore, an understanding of the methodology and its data
requirements will be beneficial in the development of appropriate data for MSW-composting.
Further, voluntary compliance with the CWA-503 regulations should enable the
MSW-composting industry to avoid waiting for state regulations. In particular, compliance
with the no observed adverse effect level (NOAEL) sludge quality limitations, including Class A
pathogen reduction, should be acceptable for MSW-compost products. Therefore, a discussion
of the proposed CWA-503 proposed regulation and its pathway analysis for agricultural
utilization, non-agricultural land application and distribution and marketing of sludge is
provided.
INTRODUCTION
Societal problems in siteing new landfills and the success of
composting and marketing of sewage sludge has renewed public interest in
the composting of the compostable portion of municipal solid waste (MSW).
Composting is simply a processing technology, and not a magical process for
disposal of MSW. It must be recognized that the MSW-compost product
must be safely placed in the environment. In this vein, beneficial use of
MSW-compost in agriculture or horticulture may add further value to the
effort of society to separate waste components, and recycle where possible.
As illustrated by past attempts at MSW composting however, one must be
cognizant of the potential failure of the process when it is touted as a money
making process.
Composting is a widely used method of organic waste stabilization
which relies on biological degradation of organic waste. It is carried out by
naturally occurring microorganisms which grow in mixed organic waste.
JThis document has been subjected to the United States Environmental Protection
Agency's administrative review. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
-------
The rate of degradation is affected by particle size, oxygen supply.
temperature, moisture content, and nutrients (C/N ratio). The end product,
commonly called compost, is a stable humus-like soil conditioner containing
various amounts of macro and micro plant nutrients. Relative to chemical
fertilizer it has low plant nutrient content. However, it is considered an
excellent medium for plant growth and increases the soil moisture holding
capacity; soil aeration; porosity; and permeability; provides natural biological
control of diseases caused by soil borne plant pathogens; as well as providing
a source of slow release organic nitrogen.
Society's concerns with location of landfills, awareness of groundwater
contamination from landfill leachate, and high cost of incineration as well as
its air pollution potential and associated ash disposal problems have made it
willing to reconsider MSW composting. If beneficial utilization of MSW-
compost is to be successful the idea of composting as a money making
venture must be abandoned and the quality of the product needs to be
appropriate for its end use. Then MSW composting might become the
alternative to landfills and incinerators.
A number of issues or questions about the environmental acceptability
of composting municipal solid waste (MSW), and use of MSW-compost
products on cropland or gardens must be addressed before MSW
composting becomes acceptable to society. This is not to say that
MSW-compost has problems with each question, but that attention must be
given to management and regulations to assure that these items do not cause
problems. These include 1) food-chain safety, 2) product/soil ingestion
safety, 3) potential for persistent phytotoxicity, 4) C:N ratio management,
and 5) acceptability of product. Each must be addressed if composting of
MSW and marketing of MSW-compost are to become the dominant method
used by US communities to handle the compostable fraction of MSW. This
paper considers the first principal issues which require evaluation for land
application of compost. Throughout this paper, it will be presumed that the
compost has been prepared to kill pathogens according to state and federal
rules — that it will be pasteurized such that there is no pathogenic risk
involved in compost utilization.
An approach used by some European countries has been to start with a
policy decision that soils will not be allowed to change with respect to their
"pollutant" contents. If this policy based decision is not clearly defined the
public is erroneously led to believe that science has dictated the conclusion
and unless the policy is followed, unreasonable risk will be encountered. In
contrast, the CWA-503 proposed sludge rule, which is designed to allow the
use of sludge products in soils and at the same time protect human health
and the environment, differs markedly from the approach of no-change in
soil-pollutant-content. We will critically examine the CWA-503 approach
and attempt to point out the science vs. policy decisions. We will attempt to
show the degree of protection that results in a reasonable-risk science-
based approach and how it might be used directly or modified for
developing guidance and regulations pertaining to MSW composting. We
-------
will argue that it is better to clearly define policy-based issues, science-
based issues, and mixtures of the two and allow for an informed public to
determine the acceptability of change.
SOIL QUALITY
The addition of foreign material to soil causes a change in the soil
which may alter its utility. To define alterations in the utility of the soil (i.e.
soil modification), requires a definition of soil. Several issues must be
considered in this definition. One is the pedologic versus engineering
versus edaphic concepts of soil. The pedologic concept of soil as the
unconsolidated mineral/organic matter on the surface of the earth that has
been subjected to and influenced by genetic and environmental factors of
parent material, climate, organisms, and topography, all acting over time
resulting in a product (soil) that differs from the material from which it is
derived in many physical, chemical, biological, and morphological properties
and characteristics. To the engineer soil is the unconsolidated material
between ground surface and consolidated bedrock upon which structures
are built. The edaphic concept of soil is restricted to soil as a medium for
plant growth. Once anthropogenic activity has occurred it is improbable
that the pedologic definition can be applied as a standard whereas the
edaphic definition, if expanded to include uses of soil other than just plant
growth may be applicable. From a practical standpoint, it is feasible to
define soil "conditions" as a function of land use. It must be recognized that
once a soil has been modified by man, the soil cannot be restored to its
original, pedologic condition. This change in utility can increase or
decrease the potential uses of the soil. Some judgement will be made as to
the acceptability of change in soil utility. This will depend on the intended
short- and long-term uses for the land and the societal attitude of changing
use of the soil. It must be remembered that change is inevitable and it is
society's responsibility to guide change. Changes in soil must consider not
only the effects of contamination on the soil itself, but also the off-site
impacts of mobilized soil (erosion), soil contaminants (runoff and leaching)
and soil vapor losses (volatilization) on the environment. Trace levels of
metals and xenobiotic organics could have little effect on soil processes and
soil ecology and yet require limitations because of the low threshold
concentrations established for drinking water quality, or the ability for
contaminants immobilized in surface soil to be mobilized from eroded
sediment in aquatic environments. At the very least, the soil changes must
be limited so as to provide the self-regulation of biological systems required
for sustaining soil processes such that they are not an unreasonable threat to
man or the environment.
The decision to allow anthropogenic changes in soil is a complex issue
which involves technical as well as policy decisions. Efforts toward rational
assessments of risk associated with changes in soils are frustrated by lack of
adequate understanding of the dynamics of the interaction of the
contaminant and soil, as well as the underlying social ramifications
associated with change. The basic concerns of soils, ecology, and toxicology
-------
assure that few individuals possess the education and experience to address
the complex nature of the contaminants environmental and toxicological
behavior from a technical prospective. Further, Vegter et al. (1988) argue
that soil quality standards are normative statements (ie. desirability of one
state of the environment vs another is a political not scientific choice) and
therefore science is unlikely to yield these values unless the results of
scientific investigations reflect the investigators' normative choices (ie. his
feelings about a good quality soil). In contrast, Simms and Beckett (1987)
argue that assessing the effects and consequences of hazards from
contaminated soil can be based on proper interpretation of sound scientific
data and this information can be utilized to establish safe concentrations.
The core of the debate may be the scientific method, which operates by
constantly questioning what is known and leads to new information, but
does not convey "good or bad" values, or it may be an unwillingness to accept
change. Either way the argument over the use of the scientific method in
the establishment of soil standards does not portray the spectrum of risk or
the limits of scientific understanding. Rather it becomes an argument over
semantics resulting in a divergence of opinion concerning the value of
science in establishment of regulatory limits and often leads to misuse of
science in establishment of regulations.
Misuse of science in formulation of public policy occurs when decision
makers without technical training fail to become informed on the technical
issues or are forced to rely on conflicting input from the technical
community. The obvious solution is to allow the technically trained person
to be the decision maker. However, his myopic view caused by a
preoccupation with the technical perspective results in an unwillingness to
recognize that the basis for policy may be scientific, but as the laws and
regulations are developed, the focus moves away from science to public will.
Further, their predilection for confidence in the answer is not balanced by
the importance of the question. When decisions have great economic or
social impacts, enormous pressure may be brought to bear on technical
advisors to "get the right answer", but this pressure for the right answer
dissipates when the economic or social impacts are not critical. In fact it
may be acceptable for the decision maker to know trends and not values
associated with a particular endpoint if the decision is not perceived as
critical.
From the above discussion it becomes apparent that the scientist must
communicate what is known along with what is not known and not allow
non-scientific factors - e.g., the regulatory outcome - to influence the
technical description. It is then the informed public through its policy
process that determine what is "good or bad", how to meet regulatory
objectives and/or serve political purposes. At the same time the technically
trained person as well as the policy maker must recognize that change is
inevitable and as members of society they have a responsibility to guide
change in the way that society dictates. If the scientist merges his "good or
bad" values in his evaluation of the technical issues the decision is
subverted, and endlessly confusing rhetoric that does not allow the public to
-------
have an informed opinion may result. At the very least the scientist should
identify where science and policy have been mixed rather than purport that
it is all science.
Mindful of the aforementioned factors, we will review the CWA-503
methodology for agricultural and non-agricultural land application and
distribution and marketing of sludge, particularly as they pertain to the
contaminant loading limits. We will attempt to define the size of the
potentially exposed population and their probability of reaching the
maximum exposure. Also, we will present a brief account of the review
comments submitted to the Agency. Although MSW-compost is different
than sewage sludge in several ways, they have enough similarities that the
risk assessment undertaken for sludge should be applicable to MSW-
compost or at least the methodology utilized in the CWA-503 rule will be
similar to those used to develop regulations for MSW-compost. Final
promulgation of US EPA's Standards for the Disposal of Sewage Sludge is
scheduled for later in 1992. Information contained in the present paper was
assembled in early 1992, and therefore must be considered as strictly
opinions of the authors and representative of technical limits without
consideration of policy issues.
PROPOSED RULE
The proposed CWA-503 rule represents EPA's first attempt to
establish comprehensive regulations for all sludge management options and
is applicable to sludge monofills, surface disposal, incineration,
non-agricultural land application, agricultural land application, and
distribution and marketing (EPA, 1989 a). Additionally it represents the
Agency's first comprehensive assessment including ecological as well as
human health effects. The contaminant loading limits are based on results of
a series of risk assessment exercises which consider the health and
environmental risk from each disposal option for 23 contaminants in sludges
(Table 1). The contaminants considered were screened from a larger list of
potentially harmful metals and organic compounds, including known or
suspected carcinogens (EPA, 1985). In establishing standards for the
disposal of sewage sludge, the Agency evaluated the risk a contaminant may
pose to either the most exposed individual (MEI) who represents the
exposed population of an exposure pathway, or the general population as a
whole (aggregate risk analysis). The computed contaminant loading of the
most restrictive exposure pathway becomes the official discharge limit for
each contaminant.
-------
Table 1. Contaminants regulated under the proposed rule
Organic Chemical Inorganic Element
Aldrin/Dieldrin Arsenic
Benzo(a)pyrene(BaP) Cadmium
Chlordane Chromium
DDT/DDE/DDD Copper
Dimethyl nltrosamine Lead
Heptachlor Mercury
Hexachlorobenzene (HCB) Molybdenum
Hexachlorobutadiene Nickel
Polychlorobiphenyls (PCBs) Zinc
Toxaphene
Trichloroethylene (TCE)
MEI: Most Exposed Individual
Many individuals may be exposed to the contaminants in the
sludge. However, it is assumed that the MEI is the individual with the
greatest exposure and therefore, if its exposure is protected, the rest
of the population is protected (EPA, 1989 b). The MEI may be a
human being, plant, animal, or any living organism. As the MEls
represent a certain segment of general populations, information or
assumptions regarding dietary habits, exposure duration, fraction of
diet derived from animals grazing on or food grown on lands on which
sludge has been applied, etc. need to be made. In the case of a human
MEI, the Agency assumed:
(a) a 70 year duration of exposure,
(b) water consumption of 2 liters per day,
(c) dietary intake equals the composite of the highest
consumption of each food group,
(d) 2.5% to 60% of the MEl's diet comes from foods grown on
sludge-treated soils,
(e) 34 to 48% of the MEl's dietary animal products was from
animals raised on feed produced from sludge-treated land
and/or grazed on sludge-treated land, and
(f) a respiration rate of 20 m3/day. In the case of the livestock
and avian species, 100% of their diet was assumed to come from
feed grown on or derived from sludged soils. In the case of
plants, the most sensitive plant species was used.
Pathways
In the proposed regulation for agricultural land application, 14
pathways were constructed to determine the contaminant loadings
(EPA, 1989 a,b). In the case of Distribution and Marketing only 6
pathways were considered. In the case of non-agricultural land, an
aggregate risk analysis was conducted based on pathways 11 (Sludge -
Soil - Surface water) and 12W ( Sludge - Soil - Ground water - Human). The
-------
aggregate risk analysis indicated the current non-agricultural land
application practices were environmentally safe and, possibly, no
regulation may be necessary. Recognizing that no regulatory limitation
would encourage utilization of highly contaminated sludges in
non-agricultural situations and at some future time non-agricultural
land may be converted to agricultural land, the Agency elected to base
limits on either the 98th percentile approach (98th percentile
concentrations of each contaminant found in a survey of sludges from
40 cities conducted in 1979 and 1980) or the agricultural land
application pathways (Table 2)— whichever resulted in the higher
number.
Table 2. Pathways models for land application of sewage sludge.1
Pathways
Description of the MEI
1: Sludge-Soil-Plant -Human
IF: Sludge - Soil - Plant - Human
2F: Sludge - Soil - Human
3: Sludge - Soil - Plant - Animal - Human
4:Sludge - Soil - Animal - Human
5: Sludge - Soil - Plant - Animal
6: Sludge - Soil - Animal
7: Sludge - Soil - Plant
8: Sludge - Soil - Soil -biota
9: Sludge - Soil - Soil-biota Predator
10: Sludge - Soil - Airborne dust - Human
11: Sludge - Soil - Surface water
12A; Sludge - Soil - Air - Human
12W: Sludge - Soil - Ground water - Human
Consumers in regions heavily affected by
landspreading of sludge.
Farmland converted to residential home garden use
5 years after reaching maximum sludge application.
Farmland converted to residential use 5 years after
reaching maximum sludge application with children
ingesting soil.
Farm households producing a major portion of their
dietary consumption of animal products on sludge-
amended soil.
Farm households consuming livestock that ingests
soil while grazing.
Livestock ingesting food or feed crops.
Grazing livestock Ingesting soil.
Crops grown on sludge-amended soil.
Soil biota living in sludge-amended soil.
Animals eating soil biota.
Tractor operator exposed to dust.
Water Quality Criteria for the receiving water.
Farm households breathing fumes from any volatile
contaminants in sludge.
Farm households drinking water from wells.
1 From: U. S. Environmental Protection Agency, 1989.
Proposed Numerical Standards
Based upon this methodology the Agency arrived at contaminant
loading limits for Agricultural Land Application, Non-Agricultural Land
Application, and Distribution and Marketing (EPA 1989 a). In keeping with
the concept of beneficial use, the limits were subjected to adherence to a
number of management practices specified in the rule. For agricultural land
application the annual sludge application rate was limited to amounts
required to supply adequate nitrogen for the crop grown or 50 metric tons
per hectare, whichever was greater; the sewage sludge had to meet
specified pathogen reduction requirements; and it could not be applied to
land that was 10 meters or less from a surface water source. For
nonagricultural land application a vegetative cover had to be established;
-------
food crops production was prohibited during periods when sewage sludge
was applied and for a period of five years after the final application of the
sewage sludge; and animals were prohibited from grazing during the period
when sewage sludge was applied and for a period of five years after the final
application of sewage sludge. For distribution and marketing, the Agency
required the following information accompany the product: (a) statement
that the product was derived from sewage sludge, (b) the name and address
of the distributor of the product, (c) list of nitrogen and contaminant
concentrations that were present in the product (at a minimum to include
the contaminants in Table 5), (d) warning to keep the product out of the
reach of children, (e) a statement prohibiting use except in accordance with
instructions, (f) instructions on the appropriate use of the product, (g) a
statement prohibiting the use of the product on frozen, snow covered or
flooded land, (h) statement prohibiting the use of the product 10 meters or
less from a surface water source, (i) rate at which the product could be
applied for stipulated uses, (j) a statement prohibiting the grazing of animals
intended for human consumption, on land where the product was applied,
and (k) statement prohibiting the use of crops grown on land where the
product was applied as feed for animals which were intended for human
consumption.
Comments on the Proposed Standards
In the preamble to the part 503 proposal, the Agency solicited public
comment on a wide range of issues including the fundamental principles of
the rule, the carcinogenic risk levels used, other human health and
environmental criteria that could be used in establishment of numerical
limits, changes that might occur because of other Agency actions (e.g.,
changes in MCL and air standards for lead), the models, the MEI and
aggregate risk analysis, and data deficiencies. In addition, the Agency
committed to facilitate and support scientific review of the technical bases
of the proposed rule during the public comment period. EPA's Science
Advisory Board conducted a review on the technical bases of the sludge
incineration regulation and the U.S. Department of Agriculture Cooperative
State Research Service Regional Technical Committee W-170, with
assistance from EPA, academia, environmental groups, and units of state and
local government agencies conducted a review of the technical bases for the
sludge regulations on land application, distribution and marketing, and
monofill and surface disposal. In addition to these two reports, the Agency
received in excess of 5500 pages of comments from 656 respondents
during the 180 day public comment period on the proposed rule.
The public and scientific peer review groups provided a
comprehensive range of opinions, comments and recommendations which
we will not attempt to summarize. Rather, we will consider the USDA-CSRS
W-170 Technical Committee report (1989). This Peer Review Committee's
report (PRC), although applauding the Agency's attempt at using the risk
assessment methodology in establishing pollution control regulations, was
critical of the assumptions and data selections made by the Agency. The
-------
primary criticism of the Agency's efforts of developing contaminant loading
limits for agricultural land may be outlined as follows:
(a) defined the MEI in an unrealistic manner,
(b) used a hypothetical and inappropriate diet scenario,
(c) used incorrect, incorrectly interpreted and inappropriate data
(d) used overly conservative relative effectiveness and dose coefficients for
the absorption of contaminant by humans and animals,
(e) used inappropriate and inadequate models to describe the transfer of
contaminant(s) from sludge source to surface and ground water,
(f) failed to take into consideration no-effect data, and
(g) arbitrarily limited sludge applications to no more than 50 metric tons
per hectare.
EVALUATION OF POTENTIAL REVISIONS TO THE PROPOSED STANDARD
In response to the comments, the Agency, in their November 9, 1990
announcement in the Federal Register (U. S. Environmental Protection
Agency, 1990) stated "...many of the assumptions and data used in the
exposure models used to generate numerical limitations for the proposed
rule will be changed to reflect more up-to-date information and more
realistic scenarios describing the expected conditions in which sewage
sludge will be land applied." At the same time EPA gave an indication of the
effect of these changes on the regulatory limits (EPA, 1990). We will
explore the nature and ramifications of these changes from a technical
perspective and evaluate how conservative the final limitations might be.
Pathway 1 & IF
Pathway 1 &1F (sludge--soil—plant—human toxicity) assumes that
sludge contaminants are taken up from the soil through plant roots. Direct
adherence of sludge or soil to crop surfaces is assumed to be minimal, and
the small amounts of contaminants on the plant's surface are presumably
washed off before consumption. The MEI for this pathway is a person
consuming food crops produced on sludge amended soil. In the case of
pathway 1, the consumer resides in a region heavily affected by
landspreading of sludge and is assumed to consume 2.5% of the plant food
groups (potatoes, leafy vegetables, root crops, garden fruits, dried legumes,
nondried legumes, grains and cereals, and peanuts) grown on
sludge-amended soils for his lifetime. In contrast, pathway IF assumes
sludge amended land is converted to a residential home garden and the
MEI produces a substantial fraction of their diet from the garden. Thus the
major difference between 1 and IF is the fraction of the food groups
assumed to be produced on sludge amended soil (FC) and the daily dietary
consumption of food groups (DC). In the case of the home gardener it is
assumed that the MEI will not produce grain, cereals, or peanuts.
However, it is assumed that the MEI produces for their consumption up to
60% of the garden food groups.. As a result of the higher consumption of
these more responsive crops pathway IF is more limiting than pathway 1
and thus will be the focus for our discussion (EPA, 1989 a,b; Page et al
-------
1987). For inorganics, a cumulative reference application rate (RP in
kg/ha) is calculated according to the following equation:
RfDxBW_TBI|xlo3
RP- L
x DCj x
where:
RP = reference (allowed cumulative) application rate of
contaminant (kg/ha)
RfD = reference dose (mg/kg day)
BW = human body weight (kg)
TB = total background intake rate of contaminant (mg/day)
RE = relative effectiveness of ingestion exposure (unitless)
103 = conversion factor (|ig/mg)
UCi = uptake response slope for the food group i (|Jg/g
DW) [kg/ha]-1
DQ = daily dietary consumption of the food group i (g DW/day)
PQ = fraction of food group i assumed to originate from
sludge-amended soil (unitless)
Although not expressed in this equation, duration of exposure (DA)
and exposure averaging time (AT) are implied. As the human health
endpoint (RfD) is for chronic lifetime exposure for the inorganics DA and
AT must also be considered as lifetime values. In fact the Agency has
assumed these to be 70 years.
By assuming that the sludge is mixed into the plow layer of the soil,
RP can be converted to a soil concentration (RLC) by the following equation:
MS x 10 9
where:
RLC = reference (allowed cumulative) soil concentration of
contaminant (ng/g DW)
RP = reference application rate of contaminant (kg/ha)
MS = 2x109 g/ha = assumed mass of soil in upper 15 cm
lO-9 = conversion factor (kg/jog)
The variables in the equation utilized to calculate RP can be classified
as dose (RfD) or exposure variables (all others). The risk reference dose
(RfD) utilized in pathways 1 & IF require a dietary intake of a contaminant
as a measure of the potential for adverse effects. Therefore, a brief
-------
description of this endpoint is worthwhile. The Food and Agricultural
Organization and the World Health Organization have defined ADI
(acceptable dally intake) 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). It is apparent that this value
is developed to protect the more susceptible members of the population
and .thus allows greater protection for the majority of the population (Ryan
et.al.,1982; EPA, 1988; Barnes and Dourson, 1988). The Agency prefers the
term reference dose (RfD) to avoid the connotation of acceptability. The
Agency's Integrated Risk Information System (IRIS) has a list of RfDs for
most noncarcinogenic chemicals. For chemicals not listed, RfD values
should be derived according to established Agency procedures (EPA, 1988).
Doses less than the RfD are not likely to be associated with adverse health
risks, and are therefore, less likely to be a regulatory concern. As frequency
and/or magnitude of the exposures exceeding the RfD increases, the
probability of adverse effects in the exposed population increases and
therefore becomes of regulatory concern (Hallenbeck and Cunningham,
1986; EPA., 1988; Barnes and Dourson, 1988). Thusrthe calculated RP
represent the maximum allowable application of contaminants in sludge to
land before exposure to the MEI has reached a level of regulatory concern.
MEI must be Real
In defining exposure, the MEI is of critical importance. A MEI can be
human, plant or animal that is supposed to represent a living organism that.
because of individual circumstances, has the maximum exposure to a given
contaminant for a particular disposal practice. While this concept seems
simple, it presents severe methodological problems to a risk assessment.
Risk assessment is fundamentally a probabilistic analysis dealing with a
random variable. Traditionally, risk assessment has dealt with two extreme
ends of the risk scale. One is the low probability-high consequence risk
(e.g., nuclear reactor meltdown). The other is the high probability-low
consequence risk (e.g., car accidents). The MEI approach which is utilized
by the Agency represents another extreme, namely a low probability-low
consequence risk. That is, the probability that an MEI as defined actually
exists is certainly very small, and it may approach zero. The health
consequence based on Agency policy, if this hypothetical person does exist,
is 10-4, or less for carcinogenic chemicals or no greater than the RfD for
noncarcinogenic chemicals. It is possible to discuss the upper 99th
percentile (or 90th or 95th), but an improperly defined MEI (the individual
with the greatest exposure) is a concept without statistical relevance and
represents a bounding estimate whose exposure is irrelevant. When worst
case assumptions about the MEI are made, do they lead to the 95th
percentile, the 99th percentile, the 99.99999th percentile? At a certain
point, which is a function of the size of the exposed population, there is a
percentile which is not defined because there are no individuals in the
group. Thus, exposure to this undefinable group is irrelevant as no one is at
risk. Therefore, the MEI must be defined and corresponds to a very small,
-------
but statistically meaningful, percentage of the population before it is
appropriate to create algorithms to attempt to quantify its exposure.
MEI and Exposure must be Linked
The purpose of an exposure assessment is to estimate exposure and
combine it with chemical specific dose response data to estimate risk. It is
important that assessments for specific chemical source demonstrate a link
between source and the exposed or potentially exposed population (EPA,
1991). Further, when the exposure assessment is predictive in nature, a
modeling and scenario development approach is recommended and the link
between individuals and source is emphasized (EPA, 1991). Thus,
information on chemical concentration and time of contact data (duration of
exposure) as well as information on the exposed population become critical.
It is apparent that not only is the definition of the MEI important, but also
its exposure and the two must be linked if the scenario approach is utilized.
Point Estimates of Exposure of the MEI
In theory, statistical tools can be used to enter the values as frequency
distributions and calculate the results in a frequency distribution. This
requires that the frequency distribution of the variables be known and that
they are independent. Unfortunately, it is only in rare cases that such
information is known. Thus, the alternative approach of selection of
discrete values from the ranges of each variable is utilized to make the
predictions. This approach results in a less precise estimate that is
described with ill-defined terms (e.g. worst case, maximally exposed
individual, etc.). As historically illustrated, use of these exposure scenarios
has been a source of controversy regarding how conservative they are. In
part conservatism can occur because of attempts to account for data
uncertainty by becoming more conservative in expression of the data;
without specifying what was done. A clear distinction between the
variability of exposures received by individuals in the population and the
uncertainty of the data would help resolve the controversy. In many cases
where estimates are termed "worst case", both a focus on the high end of
the exposed population and a selection of high end value from the data set
(for uncertainty) are used, leading to values that are quite conservative. By
using both the high end individuals (variability) and upper confidence
bounds on data (uncertainty), the estimates might be interpreted as
approaching upper bound exposures received by high end individuals.
Descriptions of these point estimates: During the time the Agency has been
working on the CWA-503 exposure assessment, others have been
attempting to communicate where on the distribution these loosely defined
terms such as "high end exposure", "reasonable worst case", "worst case"
and "maximally exposed individual" might fall (Figure 1). The question
now being ask is how does the 503 Proposed Rule fit in with these
definitions? As the mathematical product of several conservative
assumptions is more conservative than any single assumption alone,
-------
information about the distribution of each variable and its interactions must
be known if the final distribution is to be known. This being the case an
understanding of the distribution of each variable utilized in the 503
regulation and its impact on the size of the most exposed population (MEI)
as well as where on the exposure distribution the (MEI) falls needs further
clarification. Therefore, we will examine the exposure variables and
evaluate their impact on the exposed population and its potential exposure.
DIDIVTDUAI- WTTH
9O HIGH END 98
WORST CASE WORST OVSE
MAXIMTJQMI EXPOSURE
ESTIMATE
Figure 1. Terms used to identify exposure (EPA 1991). Individual with
highest exposure represents 100% of the population.
Subsistence Home Gardeners
As previously discussed, pathway IF ( home gardening) represents the
most limiting of the pathways for human consumption of agricultural crops.
It is necessary to attempt to quantify the number of home gardeners and
their production. If the term "households" is equated to population, then
46% of the population produces some of the food it consumes (Kaitz, 1978).
The data on percent of gardens vs garden size would suggest that the
median garden size was 800 ft2 and that less than 8% of the gardeners had a
garden of greater than 21,000 ft2 (Figure 2). If one equates garden size to
the amount of food production and assumes that a 21,000 ft2 garden is
required to produce all garden foods consumed each year (Ryan et al,
1982), less than 12% of gardeners can produce half their yearly
consumption (Figure 2). Thus most of the population of gardeners do not
have a large enough area to produce a large part of their annual
consumption, and only (46 x .12) = 5.5% of the population would be in the
defined subsistence home gardener category. Further, demographics
suggest that less than 2% of the population live in the same county for a
lifetime and thus a change in garden location would alter their exposure
(EPA, 1977). It would seem that the subsistence home gardener is no more
than 2% of the population (assuming all persons who do not move are
subsistence gardeners) nor less than 0.1% (5.5 x .02), but most likely less
than 1%. This is a conservative estimate and the actual number is much
less because the data on gardeners is short term information and only a
-------
small part of them will continue to garden for a lifetime. It is evident that
the subsistence lifetime gardener population would be classified as
maximum exposure and may be worst case or bounding estimate. While
one might argue that this population exists, the size of the population
diminishes drastically as some of the other assumptions are placed on them.
For example how many subsistence lifetime gardeners use sludge or have
gardens that are located on previously sludge soils? How many of these
subsistence lifetime gardeners who use sludge are at the maximum
application rate (RP)? How many of these subsistence lifetime gardeners
who use sludge and are at the maximum application rate are unaware of
good agronomic management practices? Answers to these questions not
only impact the potential MEI population but have significant impacts on its
exposure.
TOO
1000 -10000
GARDEN SIZE (sq. ft)
1OOOOO
Figure 2. Garden size distribution (Kaitz, 1978b)
Lifetime Subsistence Gardeners Who Use Sludge
No direct measurements of the number of gardeners who use sludge
as a soil amendment/fertilizer is available, nor is information on the
conversion of past land application sites to home gardens available. It would
seem that only a small percentage of the gardeners would use sludge as
there are many sources of organic material available and the subsistence
gardener would already have their favored material which they were
satisfied with and thus unlikely to change to an unproven product. An
estimate on conversion (land application sites becoming gardens) would be
the percent of lands receiving sludge. The 5% value utilized by the Agency
as a conservative estimate of the agricultural land in a region which is
heavily impacted by land application of sludge could be used. It would be a
conservative estimate as not all lands converted to residential use would
have been agricultural land, thus the actual percent would be less. It is thus
apparent that only a small percentage of the lifetime subsistence gardeners
will be utilizing soil which received sludge. The estimate of lifetime
subsistence gardener on soil which received sludge is less than 0.005% of
-------
the population (. 1 x .05). This does not consider how many are at the
maximum application rate and/or the number of gardeners that utilize
agronomic management practices.
What is the Potential Exposure
It is important to recognize that in addition to Cd, other materials are
being added with sludge which make the soil concentration dependent on
the sludge concentration. In fact the ultimate soil concentration will be a
function of the sludge concentration and percent sludge in the 15 cm. zone
of incorporation. If no decomposition of the organic fraction in sludge
occurs, then as the mixed 15 cm. incorporation zone approaches 100%
sludge its concentration will approach that of the sludge. As sludge is
approximately 50% organic material when the added organic fraction of the
sludge is decomposed and all that would be left is the inorganic material,
the mixed 15 cm. incorporation zone could be 50% sludge residue. If
sufficient sludge is applied such that the 15 cm. incorporation zone is 100%
sludge residue, its concentration would be twice that of the sludge applied
(McCalla et al., 1977). Thus, the projected soil composition distribution
when it is 100% sludge can be obtained using sewage sludge composition
information (EPA, 1990). The frequency distribution illustrates that 98% of
the sludges are below 58 mg/kg, 90% are below 15 mg/kg and 50% are
below 6 mg/kg Cd (Figure 3). Assuming the soil is 100% sludge residue,
then the concentration would be equal to at most twice the sludge
concentration.
1.0
Sludge Concentration
UJ
O
Q
(f)
LJL
O
O
h-
o
cc
LI-
0.8-
0.6-
Sludge Concentration x 2
0 20 40 60 80 100
SLUDGE Cd CONCENTRATION (mg/kg)
Figure 3. Sludge Cd concentration distribution (EPA, 1990)
-------
The other side of the question is how long it takes for the 15 cm.
incorporation zone to become 100 % sludge. For example if a sludge is
continuously mixed into the upper 15 cm. of soil at a annual rate of
application of 50 MT/yr it will be 30 yrs. before the mixed 15 cm. zone of
incorporation becomes 90 % sludge or 60 yrs. before it approaches 100%
sludge (Figure 2). At a more reasonable agronomic rate, it will take in
excess of 60 to 100 yrs. of continuous application before the mixed 15 cm.
zone of incorporation is 50% sludge and on the order of 300 to 600 yrs of
continuous application before it approaches 100% sludge. It will be a long
time before soil concentration approaches that of the material being
applied. The same time frame will be required before the exposure can be at
the projected RLC. With a DA of 70 yrs. required at the RLC before the RfD
is reached it is virtually impossible for exposure to be as great as predicted
in the next several hundred years.
63
o
ffi 1OO~
8 801
I
g 601
5"£
II ^
§ 20-1
O
I
6.7 MT/ha
**
3.3 MT/ha
1.7 MT/ha
— i
SO
1OO ISO 2OO 25O
YEARS OF APPLICATION
3OO
350
Figure 4. Effect of time and rate of continuous application on the
composition of the soil
The implication is that even if there exists a population of subsistence
lifetime gardeners who use sludge, it will take several lifetimes of
continuous application at agronomic rates before the soil reaches a
concentration equal to the sludge. It is unlikely that continuous yearly
applications will occur for these time frames; therefore soil concentrations
are not likely to reach the levels in sludge.
As the foregoing discussion illustrates the MEI for Land Application
(Agricultural) pathways 1 & IF, is an example of the problem of piling
conservative assumption on top of conservative assumption. It may be
-------
impossible to define the size of the population who are the lifetime
subsistence gardeners who use sewage sludge and have reached the
maximum sewage sludge application rate (RP), but to have such a person
have the highest consumption of all food groups which have the highest
observed plant response for 70 yrs exposure certainly makes for an
infinitely small population and one which cannot be calculated. These
requirement make the exposed population an upper bound estimate. At any
rate it becomes apparent that the layering on of conservative assumptions
about the exposed population makes it infinitely small and potentially
undefinable.
Diet
The use of this dietary exposure information for chronic exposure
situations requires an integration of exposure over time. The Proposed Rule
(EPA, 1989) used the highest food consumption group to represent the diet
of individuals from birth to age 70 (Mega Eater). This means that the diet of
the teenage male (14-16 yrs) was used for the food groups: grain, potatoes,
root vegetables, dairy, and dairy fat. The diet of the adult female (25-30 yrs)
was used to represent the food groups: lamb and lamb fat. The diet of the
adult male (25-30 yrs) was used to represent the food groups: legume
vegetables, garden fruit, beef, pork, poultry, beef fat, poultry fat, and pork
fat. The diet of the adult female (60-65 yrs) was used to represent the food
group leafy vegetables. The diet of the adult male (60-65 yrs) was used to
represent the food groups: beef liver, eggs, and beef liver fat. This results in
an over estimate of dietary consumption (DC) during a 70 year life time. As
illustrated by the comments received this assumption was viewed as a
bounding estimate for exposure making it impossible to define the exposed
population; thus leading to the conclusion that the exposed population did
not exist (EPA, 1990).
To develop a more reasonable exposure we started with the
Pennington (1983) revision of the total diet study as modified by EPA
(1989), averaged the dietary consumption rates across sex in the 14-16,
25-30, and 60-65 age categories and calculated the Estimated Lifetime
Average Daily Food Intake (Chaney, 1990 a,b). The use of the estimated
lifetime vs the mega eater diet results in a difference in daily consumption
rate of 52-75% depending on food group. The use of the estimated lifetime
diet as DC, results in an RP which is 1.4 times that when the mega eater is
used as DC (Table 3). The use of the lifetime DC should be encouraged as it
is logical from the duration of exposure perspective and from a population
perspective. The calculated lifetime DC value is based on short term dietary
data and thus must be considered a over estimation (conservative estimate)
of the true value. One must be aware that extrapolation of short term
exposure data to estimate long term exposure results in an overestimate of
the true exposure (EPA, 1991). However, at this time no data on lifetime
consumption of individuals or the population exist; thus, this data is the best
available.
-------
TABLE 3 Effect of changing daily dietary consumption (DC) and plant
response slope (UO on reference application rate (RP) for Cd.
Dietary
Consumption
Neutral
All
UC1
Acid
Highest
KP(ke/ha)
Mega Eater
Lifetime
164
234
108
152
48
67
4.2
6.0
l)Highest = the highest UC observed for each food group
Neutral = the geometric mean of UC where pH > 6.0.
All = the geometric mean of all observed values of UC
Acid = the geometric mean of UC where pH < 6.0
Plant Response
One of the strongest comments the Agency received was that their
data sets were flawed in that they did not utilize field data. The errors
associateoTwith using salt rather than sludge or greenhouse pot data to
predict field response are well documented (Logan and Chancy, 1983; Page
et al., 1987). Therefore an extensive effort to evaluate the existing data sets
and add all relevant sludge field data including the no observed adverse
effect data from valid field studies was undertaken. The data were classified
as on the basis of field sludge studies (A), sludge pot studies (B), and all
others (C). The geometric mean of type A was used to represent UC in the
calculation. On the surface this would appear to represent an average
exposure value but as will be illustrated this isn't true.
Lower bounding estimate: In the case of plant uptake (UC), if the data
showed no significant increase in plant concentration with application or if
the slope was negative, it was arbitrarily given a value of 0.001 mg/kg
increase in plant tissue [kg/ha rate of application]-!. This lower cutoff value
results in an over estimation of the actual values of UC's and thus any value
of UC utilized to represent the distribution will be higher than if the real
values were used (conservative estimate of the distribution). The degree of
overestimation will be in part a function of the number of points which have
values at the lower bounding estimate. As might be expected this varies
with contaminant. For example, of the 52 data points for Pb cited in the
revised data set 38 (73%) had plant uptake values of 0.001, whereas of the
196 data points for Cd cited in the revised data set 28 (14%) had plant
uptake slopes of 0.001. At this time it is not possible to determine how far
this assumption shifted the distribution, but one can understand that
because of data uncertainties, the use of this lower boundary on the
distribution results in higher exposure calculations than would occur if the
actual data points were utilized. The amount of this overestimation of UC is
unknown.
Linear extrapolation: It has been observed in field studies with sewage
sludge (and some pot studies) that plant uptake is curvilinear (asymptotic to
-------
a maximum) rather than linear ( Chancy et al. 1982; Mahler et al.,1987;
Corey et al., 1987; Hinesly and Hansen, 1984; Hinsely et al., 1984; and EPA,
1989) This phenomenon has been argued to be due to competition
between soil and sludge solids for metal binding. As sludge application rate
increases, the binding capacity of the sludge solids becomes the controlling
factor in metal chemistry of the system. The model indicates that metal
adsorption (and occasionally co-precipitation) with sludge constituents
(specific or selective adsorption in the presence of 3mM Ca2+ common to
root zone soil) is the controlling factor in metal availability in sludge
amended soil. Because of the importance of sludge capacity for specific
adsorption of metals, the concentration of metals in sludge is important and
the ratio of metals :metal adsorption capacity controls metal availability and
thus plant uptake reaches a maximum as sludge application increases (Corey
et al., 1987; Logan, 1989). An examination of the plant uptake data from
field applied sludge shows the response curves tend to be curvilinear,
however most of the individual observations of UC are based on experiments
which do not have sufficient rates of application to test their lack of
linearity. Therefore, the revised data on plant response (UC) utilized a
linear response and must be recognized as an overestimation of UC
(conservative estimate and may in fact be a bounding estimate). It becomes
obvious that linear regression and extrapolation of plant concentration
results in an overestimation of plant concentration as the extrapolation
exceeds the bounds of the data. The degree of overestimation will be a
function of where the maximum occurs and how far past it a linear
extrapolation is used. The degree of conservatism in the estimated RP
could be anywhere from 1, if the response were linear, to (RP/point at
which curvilinearity occurs). The Cd application rate on which the UC data
set is based, range from 0.08 to 20 kg Cd/ha with an average of 7 kg Cd/ha
for the food groups utilized in the equation for pathway IF. Therefore, the
calculations of RP outside this range must be considered conservative and if
we assume that plant concentration reached the maximum within the data
set, the overestimation of exposure will be at least RP/20.
Sludge quality: It has been observed that sludges with low metal
concentrations have lower plant uptake at the same metal loading than do
sludges with higher metal concentrations (Corey et al., 1987 and Logan,
1989). In a recent pot study a linear relationship between the total Cd
concentration of 17 anaerobically digested sludges and Cd concentration of
Sudax was observed when the sludges were applied at a constant Cd rate
(Jung and Logan, 1992). In the present data set for UC, sludge composition
has not been considered and it is important to note that most of the
available studies with sewage sludge were conducted with sludges
containing metals at levels higher than the median concentrations in
current U.S. sludges (EPA, 1990). The cited studies were either
deliberately conducted with high metal sludges (e.g., the Chicago sludge of
the 1970's which had a Cd concentration of approximately 200 mg/kg) or
simply reflect the higher metal concentrations that were present prior to
pretreatment As not all studies give the metal concentrations of the
sludges used, it is not possible to determine sludge metal concentration for
-------
all studies. It is apparent that the data set will overestimate the UC values
that would be observed from current lower metal sludges, but the amount of
overestimation is not known.
Sludge equilibrium: It has been observed from long term field sludge
studies that plant availability of sludge-borne metals is highest during the
first year after sludge is applied (Hinesly and Hansen, 1984; Bidwell and
Dowdy, 1987; and Chang et al.,1987). This is contrary to the long-held
popular belief that once the sludge applied organic matter is oxidized
complexed metals will be released and plant uptake will increase (Beckett
and Davis, 1979). Additionally, this "Sludge Time Bomb" mentality is not
supported by studies of sludge chemistry which indicate that digested
sludge in addition to being 50% organic matter is 50% inert inorganic
mineral forms (including Fe and Al oxides, silicates, phosphates, and
carbonates) that are reactive with the metals and environmentally stable
(McCalla et al., 1977; Essington and Mattigod, 1991). Using early-year
response curves to develop UC will overestimate UC derived from long term
well stabilized sludge/soil systems. Nevertheless, most of the field studies
used in the UC data set are from the early years of the experiment (less than
5 years after establishment) and are being utilised to develop long term (70
yrs) exposure assessments. Based on the observations of Bidwell and Dowdy
(1987) and Chang et al. (1987) the overestimation could be as large as a
factor of 5.
It is apparent that these two assumptions [lower bounding estimate
(plant response slope of 0.001 for all non responsive data) and simple linear
regression of the response curve)] and two variables (sludge composition
and sludge equilibrium) are included in the data set of UC and in all cases
imply that the true long term data set for UC has been overestimated. It is
not possible to determine how large an overestimation these conservative
approaches cause, but it could cause UC to be bounding estimates. Just
considering linearity and long term sludge equilibrium could result in an
overestimation of exposure by RP/4 [(RP/20) x 5]. It is thus apparent that
even though we have eliminated the error caused by utilization of salt pot
studies for the development of the UC data set, we have allowed data
uncertainties to cause the revised UC data set to overestimate the true long
term data set and in fact the new set may represent upper bound exposures.
With this conservatism built into the data set, it is apparent that any
representation of the data set will also yield a conservative estimate of the
true long term data set it is representing. Therefore, the use of a
conservative estimate of the distribution would only layer on another
conservative factor which doesn't appear to be justified. As the UC data set
for any food group appears to represent a log normal distribution, it is our
contention that the geometric mean which best represents the distribution
should be used. This may appear to say that the UC value in the exposure
assessment is a mid range value but as discussed above could be considered
a bounding estimate.
-------
Soil variables: Of all the soil variables which have been reported to affect
plant uptake of sludge applied metals (organic matter content, cation
exchange capacity, soil texture, pH, etc.) only pH has been shown to have a
consistent significant effect (Page et al., 1987). Therefore, it is necessary to
consider this variable in the selection of UC from the data set. It is
important to recognize that in natural soil systems as the pH decreases
below 5.5 a rapid (exponential) increase in soluble Al and Mn occurs. This
increase in soluble Al and Mn plays havoc with plant growth and
development in all but the hardiest species (Pearson and Adams, 1967).
Therefore, consideration of plant uptake in these strongly acid soils
becomes questionable as even without the increased level of metals
associated with high accumulative application of sludge, yield will suffer and
little or no edible product will be available for consumption. It would also
seem that even before this reduction in yield associated with extremely acid
soils could occur the visual symptoms of Zn, Cu or Ni phytotoxicity would
likely occur and the subsistence gardener would learn about soil pH and
lime the soil. Thus the required duration of exposure (70 yrs) would not
occur and the MEI would not exist. Further, if the MEI is defined as the
subsistence gardener it-would seem unreasonable to assume that they were
unaware of agronomic practices (i.e. pH management) which would imply
they would manage soil pH in the more desirable agronomic range of >6.0.
This would suggest that only those studies in well managed near neutral pH
systems should be considered. Concerns for deviations of soil pH during the
chronic lifetime exposure (70 yrs) coupled with the known effect of pH,
resulted in utilizing plant response curves (UC) from all available sludge
field data including both the acid and neutral soil conditions. Within the
UC data set the observations on systems with a pH < 6.0 comprised 40% of
the total data set and varied from 15 to 55 % of the observations within a
food group. Thus it is apparent that the acid soil system is well represented
within the data set.
The choices for UC (highest observed, geometric mean of acid soil
conditions, geometric mean of neutral soil conditions, geometric mean of all
possible conditions) has been debated and continue to be debated. As
illustrated, the use of these different UC's in the calculation of RP for Cd
result in a range of values from 6 to 230 kg/ha (Table 3). The use of the
highest observed UC (RP = 6 kg/ha from Table 3) would assure safety, but
would be a difficult position to defend for the reasons discussed above.
Additionally, assuming that the UC values for each of the food groups utilized
is independent the probability of each having the highest UC at the same
time is, 2 x 10-8, and for this to occur 70 times in a row (to account for DA)
is even less likely. Thus not only may the MEI be nonexistent, the exposure
is an upper bound estimate, of a conservative data set. If one then went to
the assumption of the acid garden scenario and utilized only the UC's from
experiments with pH < 6.0 (RP = 67 kg/ha from Table 3) it is still difficult
to imagine that such an MEI exists. At this time there is no information to
allow for a probability of occurrence of continuous production on acid soils,
but as discussed above it is hard to believe that anyone who depends on a
garden for subsistence will not somewhere along the way learn about soil pH
-------
management and break the insidious chain of events. It is apparent that the
other two conditions (occasional acid and neutral) could occur and thus a
potential MEI may exist and therefore become more tenable exposure
events. But, the necessity of adding the extra layer of conservatism
(inclusion of both acid and neutral observations in the data set) on an
already conservative representation of UC is questionable.
Fraction of Food Produced
As presently defined, the subsistence gardener is assumed to produce
37% of his lifetime consumption of potatoes and 59% of his lifetime
consumption of all other food groups. As discussed in the MEI section,
these are conservative assumptions and make for a small number of people
within the defined MEI population. Changes in the fraction of food
originating from sludge-amended soil (FC) would alter the size of the
exposed population (MEI). Additionally, changes in this FC have a
significant effect on RP (Table 4). As indicated, the 60% FC would
represent a high end value with further increases in FC having little impact
on RP, whereas reductions in FC lead to large changes in RP. An issue
which is not apparent is that the requirement of a 60% FC makes the MEI
an upper bond estimate of both the population and exposure. Leafy
vegetables are consumed in a fresh state and it is not possible to produce
them throughout the year, except in extreme situations. Most gardeners
are lucky to harvest a few weeks production, which means that they might
have garden produced leafy vegetables for one month each year. They
certainly would not harvest 60% of their yearly consumption during this
time period. Assuming that the gardener may obtain 10% of his leafy
vegetables from his home garden, and allowing 60% for the other food
groups, the calculated RP becomes 250 kg/ha rather than 145 kg/ha. Thus,
this failure to consider the actual fraction of leafy vegetables a gardener can
produce allows significant overestimation of both the size of the exposed
population and their exposure [approximately a factor of 2 (250/145)],
causing an overestimation of risk.
TABLE 4. Effect of changing fraction of food group originating from sludge
amended soil (FC) on the reference application rate (RP) of Cd where UC is
the geometric mean of all data and DA is the lifetime diet.
Reference %FC
Application 1 10 60 100
Rate
RP (kg/ha) 8630 870 145 87
It is apparent that the way the data set for UC was constructed [using
all data regardless of pH (approximately 30 out of the 70 yrs of exposure the
garden would be assumed to be strongly acidic, based on the distribution of
pH among studies in the data set)] and the 60% value of FC, results in an
overestimation of exposure. The exact amount of the overestimation is not
known, but could be RP/2 [(RP/4) x 2].
-------
In order to evaluate the percentages of sewage sludges that could
cause a garden soil that was 100% fully decomposed sludge (total
disappearance of sludge applied organic matter) to exceed the RP, the
calculated RP (as a function of UC and FC) was converted to RLC and
displayed with the sludge Cd concentration distribution (Figure 5).
Assuming that all sludges are utilized, allows an evaluation of the percent of
the MEIs that could reach the estimated exposure when their gardens are
100% sludge residue. As discussed this is unrealistic as it requires several
hundred years of continuous application at agronomic rates. It is apparent
that even with the overestimation of exposure that has occurred, the only
data that would suggest that greater than 10% of the sludges could exceed
the exposure limit (RLC) were when the extreme values of UC are used.
Even the assumption of a lifetime subsistence gardener who used sludge and
never learned common agronomic practices (always had a strongly acidic
garden), indicates that less than 10% of the sludges would exceed the RLC.
If you considered the consumption of leafy vegetables to be 10% rather than
60%, less than 2% of the sludges would exceed the RLC. Thus, this is at
least high end and most likely maximum exposure for a questionable MEI.
As the potential number of MEFs increases with less absurd assumptions
(i.e. the subsistence gardener learns about pH and limes his soil) the
potential for exceeding the calculated conservative soil concentration (RLC)
moves from high end/reasonable worst case to maximum exposure to worst
case to upper bound.
o
co
o
LU
o
1.O
Slud
•* O.8-
LU
O
Q
co
o
LU
O
O
o
cc
CL
Q
O
O
O
I—
O
•<
cc
O.6-
O.4-
O.2 -
O.O
SI udga^ Concentration
pH > 6.0
UC = Highest value
1.0
- 0.8
O.6
CO
LU
O
ra
i
co
u_
O
- 0.4 ^
O.2
CD
o
cc
O.O
O 2O 4O 6O 8O 1OO
SLUDGE/SOIL Cd CONCENTRATION, RLC (mg/kg)
Figure 5. Effects of variation of UC and FC on the calculated soil
concentration and the fraction of sludges which could exceed the soil
concentration.
-------
In other efforts we have illustrated that the methodology and data
utilized appear to result in high end/reasonable worst case exposure
assessments for the organics and for direct ingestion by children (Chaney et
al. 1991; Chaney and Ryan, 1991). Thus, even the drastic changes in the
limitations from the proposed 503 rule (EPA 1989a) to those contained in
the Nov. 9, 1990 Federal Register which were equal to those of the PRC
(1990) appear to be a great deal of conservatism built in. The limits are
protective of the high end MEI (Table 5). In their development of the
NOAEL (no observed adverse affect level) sludge limits Chaney and Ryan
(1991) illustrated that some of the limitations derived on a technical basis
could be adjusted to include value judgements, but that these changes need
to be identified.
Table 5. Comparison of the EPA 503 proposed regulations (1989a),the
W-170 Peer Review Committee limits (1989), the limits as suggested by
EPA (1990), and the NOAEL sludge limit from (Chaney and Ryan, 1991)
POLLUTANT
Cd
Cr
As
Pb
Hg
Zn
Cu
Ni
Mo
Se
503
kg/ha
18.4
530
14
125
15
172
46
78
5
32
PRC
kg/ha
>20
NA
1600
300*
NA
2600
1200
500
NA
NA
EPA 1990
kg/ha
>20
NA
1600
580
NA
2600
1200
500
NA
NA
NOAEL
Sludge
Limit
mg/kg DWCd
251
>3000L2
601
300
15L3
2700
1500
500
354
32
1 Adjusted downward for pretreatment considerations.
2 No adverse effects reported for any Cr3+ level in municipal sludge.
3 Valid for all sludge uses except mushroom production.
4 Mo limit raised because Mo slowly leaches from "worst-case" alkaline soil.
It is apparent from a comparison of the NQAEL sludge limit and MSW-
compost quality that MSW-compost should not have a problem if it behaves
environmentally similar to sludge.
RESEARCH NEEDS FOR MSW-COMPOST:
In order for MSW-composting and D&M of MSW-compost to become
acceptable to the public and marketability as desired by the industry,
research and demonstrations will be required. Research on fate and effects
of nutrients, metals, and organics in sewage sludge were critical for public
acceptance, and are providing the data needed to prepare appropriate
-------
regulations. Thus, the most important research needs or questions
remaining for MSW-composting and MSW-compost marketing are
confirmation of the assumption that its environmental behavior is like
sludge. As these issues are discussed in detail in our other paper in the
conference they will not be repeated. Research needs which are related to
the risk assessment may not be specific for MSW-compost, but will help its
regulation development and our understanding of the risk. These include:
1) Plant uptake assumptions
A.) The assumption that plant response is relative (all crops can be
represented by the response of one crop) needs to be validated.
B.) Description of the plant response curve (linear or curvilinear)
needs to be confirmed.
C.) Does MSW-compost influence plant uptake of contaminants like
sewage sludge?
2) Dietary Data
A) What is the long term variability in dietary consumption for the
U.S. population?
B) What are the gardening habits of the U.S. population and what
percentage of their consumption of various products do they
grow.
C) What are the agronomic management practices (mulching, pH
control, water management, etc.) of the gardening population
and how do they impact plant uptake.
D) What effect does source of a chemical in the diet have on its
bioavailability.
3) Collect information on statistical distribution of parameters utilized
in risk assessment in order to develop probability distribution for
exposure rather than rely on point estimates and the MEI approach.
4) A true ecological risk assessment including system level impacts,
[e.g. species diversity and population impacts]) needs to be made
rather than rely on specific points of information or geometric
means. Additional data for many species need to be collected.
5) Soil transport models need to consider transport in non uniform
media; better information on contaminant desorption is needed.
SUMMARY:
The use of the proposed CWA-503 methodologies for development of
soil loading limits represents a valid technical approach for calculating
maximum loading limits for the contaminants of concern. It must be
remembered that the limits are conservative both by design as well as
because of conservative expressions of the data caused by its uncertainties.
These data uncertainties need to be expressed and quantified where
possible so that the conservatism becomes apparent rather than hidden.
-------
After these limits are calculated it is possible to make policy decisions
which modify the limits, but it is necessary that the public understand what
is being done and why.
In conclusion, emphasis must be on production of composts which are
1) within the NOAEL standards, 2) pasteurized to the PFRP standard, 3)
stabilized to allow use as nitrogen fertilizer, and 4} stored to prevent
production of phytotoxic anaerobic biodegradation by-products. This will
allow the composting technology to become a widely accepted national
program for handling of MSW in the United States.
LITERATURE CITED
Barnes, D.G. and M. Dourson. 1988. Reference dose (RfD): Description and
use in health risk assessments. Regulatory Toxicology and Pharmacology
8:471-486.
Beckett, P. H. T., R. D. Davis, and P. Brindley. 1979. The disposal of sewage
sludge onto farmland: The scope of the problems of toxic elements. Water
PoUut. Contr. 78:419-445.
Bidwell, A.M. and R.H. Dowdy. 1987. Cadmium and Zinc availability to corn
following termination of sewage sludge application. J. Enviro. Qual. 16:438-
442.
Chaney, R.L., S.B. Sterrett, M.C. MoreUa, and C.A. Lloyd. 1982. Effect of
sludge quality and rate, soil pH, and time on heavy metal residues in leafy
vegetables, pp. 444-458. In Proc. Fifth Annual Madison Conf. Appl. Res.
Pract. Munic. Ind. Waste. Univ. Wisconsin - Extension, Madison, WI.
Chaney, R.L. 1990 a. Twenty years of land application research Regulating
beneficial use. BioCycle 31(9):54-59.
Chaney, R.L. 1990 b. Public health and sludge utilization Food chain impact.
BioCycle 31(10):68-73.
Chaney, R.L. and J.A. Ryan. 1991. The future of residuals management after
1991. In Proc. AWWA/WPCF Joint Residuals Management Conf. Aug. 1991.
Chaney, R.L., J.A. Ryan, and G.A. O'Connor. 1991. Risk assessment for
organic micropollutants: U.S. point of view, pp In P. L'Hermite et al.
(eds.) Proc. EEC Symp. Treatment and Use of Sewage Sludge and Liquid
Agricultural Wastes (Athens, Sept. 1990).
Chang, AC., T.D. Hinesly, T.E. Bates, H.E. Doner, R.H. Dowdy, and JA. Ryan.
1987. Effects of long-term sludge application on accumulation of trace
elements by crops, pp. 53-66. In Page, A. L., T. J. Logan, and J. A. Ryan
(eds). Land application of sludge: Food chain implications. Lewis Publishers
Chelsea MI
-------
Corey, R. B., L. D. King, C. Lue-Hing, D. S. Fanning, J. J. Street, and J. M.
Walker. 1987- Effect of sludge properties on accumulation of trace elements
by crops. In Page, A. L., T. J. Logan, and J. A. Ryan (eds). Land application of
sludge: Food chain implications. Lewis Publishers, Chelsea ML
Essington, M.E. and S.V. Mattigod. 1991. Trace element solid-phase
associations in sewage sludge and sludge-amended soil. J. Enviro. Qual.
55:350-356.
Hallenbeck, W.H. and K.M. Cunningham 1986. Quantitative Risk Assessment
for Environmental and Occupational Health. Lewis Publishers, Chelsea, MI.
199pp.
Hinesly, T. D, and L. G. Hansenl984. Effects of using sewage sludge on
agricultural and disturbed lands. U. S. Environmental Protection Agency.
600/S2-83-113. PB# 84-117142, National Technical Information Service,
Springfield, VA
Hinesly, T. D., L. G. Hansen, and D. J. Bray. 1984. Use of Sewage Sludge on
Agricultural and Disturbed Lands. U. S. Environmental Protection
Agency.600/S2-84-127. PB84-224419, National Technical Information
Service, Springfield, VA
Jung, J. and T.J. Logan. 1992. Effects of sewage sludge cadmium
concentration on chemical extractability and plant uptake. J. Environ. Qual.
21:78-81.
Kaitz, E.F. 1978. Home gardening national report. 1975-77. Presented at the
American Seed Trade Association, Inc. 95th Annual Convention, Kansas City,
MO., June, 1978.
Kaitz, E.F. 1978 b. Personnel communication.
Logan, T.J. 1989. Sludge metal bioavailability. In Muralidhara H.S. (ed) Proc.
Int. Symp. on Solid/Liquid Separations. Battelle Press, Columbus, OH.
Logan, T.J. and R. L. Chaney. 1983. Utilization of municipal wastewater and
sludge on land - Metals. In Page, A. L., T. L. Gleason, J. E. Smith, jr., I. K.
Iskandar, and L. E. Sommers (eds) Utilization of municipal wastewater and
sludge on land. Univ. Calififornia, Riverside, CA.
Lu, F.C. 1983. Toxicological evaluations of carcinogenic and noncarcinogens:
Pros and cons of different approaches. Reg. Toxicol. Pharmacol. 3:121-132.
Mahler, R.J., JA. Ryan, and T. Reed. 1987- Cadmium sulfate application to
sludge-amended soils I. Effect on yield and cadmium availability to plants.
The Sci. Total Envir. 67:117-131.
-------
McCalla, T.M., J.R. Peterson, and C. Lue-Hing. 1977. Properties of
agricultural and municipal wastes. In Elliott, L.F. and F.J. Stevenson (eds)
Soils for Management of Organic Wastes and Waste Waters. Soil Science
Society of America, Madison, WI.
Page, A. L., T. J. Logan, and J. A. Ryan (eds.) 1987. Land Application of
Sludge—Food Chain Implications. Lewis Publishers, Chelsea, ML 168pp.
Pearson, R.W. and F. Adams (eds). 1967. Soil Acidity and Liming. Agronomy
Monograph # 12. Am. Soc. Agron., Madison, WI. 274pp.
Pennington, J.A.T. 1983. Revision of the total diet study food lists and diets.
J. Am. Diet. Assoc. 82:166-173.
Ryan, J.A., H.R. Pahren, and J.B. Lucas. 1982. Controlling cadmium in the
human food chain: A review and rationale based on health effects. Environ.
Res. 28:251-302.
Simms, D.L. and M.J. Beckett. 1987. Contaminated land: Setting trigger
concentrations. The Sci. Total Envir. 65:121-134.
U. S. Environmental Protection Agency. 1977 Hazardous waste: A risk-
benefit framework applied to cadmium and asbestos. EPA-600/5-77/002.
U. S. Environmental Protection Agency. 1985. Summary of environmental
profiles and hazard indices for constituents of municipal sludge. U.S. EPA,
Office of Water Regulations and Standards, Wasterwater Critera Branch,
200pp.
U. S. Environmental Protection Agency. 1986. Guidelines for Estimating
Exposures. Federal Register 51:34042-34054.
U. S. Environmental Protection Agency. 1988 Reference Dose (RfD):
Description and use in health risk assessments. Integrated Risk Information
Systems (IRIS). Online. Intra Agency Reference Dose (RfD) Work Group,
Office of Health and Environmental Assessment, Environmental Criteria and
Assessment Office, Cincinnati OH.
U. S. Environmental Protection Agency. 1989.a. Standards for the Disposal of
Sewage Sludge; Proposed Rule 40 CFR Parts 257 & 503. Federal Register
54:5746-5902.
U. S. Environmental Protection Agency. 1989.b. Development of risk
assessment methodology for land application and distribution and marketing
of municipal sludge. EPA/600/6-89/001
-------
U. S. Environmental Protection Agency. 1990. National Sewage Sludge
Survey: Availability of Information and Data and Anticipated Impacts on
Proposed Regulations; Proposed Rule 40 CFR Part 503.Federal Register
55:47210-47283.
U. S. Environmental Protection Agency. 1991. Guidelines for exposure
assessment, Draft final. Risk Assessment Forum. Washington DC.
Veger, J.J., J.M. Roels and H.F. Bavinck. 1988. Soil quality standards:
Science or science fiction. In K. Wolf, J. van der Brink and F.J. Colon (eds.)
Contaminated Soils, Klumer Academic Publishers.
W-170 Peer Review Committee. 1989. Peer Review of Standards for the
Disposal of Sewage Sludge (U.S. EPA Proposed Rule 40 CFR Parts 257 &
503) USDA-CSRS W-170 Regional Research Committee. 122pp.
-------
Chaney, R.L. and J.A. Ryan. 1992. ID H.A.J. Hoitink et al. (eds.). Proc.
International Composting Research Symposium. In press.
HEAVY METALS AND TOXIC ORGANIC POLLUTANTS IN MSW-COMPOSTS:
RESEARCH RESULTS ON PHYTOAVAILABILITY, BIO AVAIL ABILITY, FATE, ETC.
Rufus L. Chaney, Environmental Chemistry Lab, USDA-Agricultural Research
Service, Bldg 318, BARC-East, Beltsville, MD 20705 and James A. Ryan, Risk
Reduction Engineering Lab, US-Environmental Protection Agency, 5995 Center Hill
Road, Cincinnati, OH 45224.
KEYWORDS: Heavy metals; PCBs; PAHs; food-chain; cadmium; lead; boron; soil
pH; phytotoxicity; Mn-deficiency.
ABSTRACT:
This paper is a review and interpretation of research which has been
conducted to determine the fate, transport, and potential effects of heavy metals
and toxic organic compounds in MSW-composts and sewage sludges. Evaluation
of research findings identified a number of Pathways by which these contaminants
can be transferred from MSW-compost or compost-amended soils to humans,
livestock, or wildlife. The Pathways consider direct ingestion of compost or
compost-amended soil by livestock and children, plant uptake by food or feed
crops, and exposure to dust, vapor, and water to which metals and organics have
migrated.
In research on these questions, the chemical properties of sludges and
composts were found to be very important in binding the metals and toxic
organics. Amorphous oxides of Fe, Al, and Mn provide persistent specific metal
adsorption capacity for the heavy metals of concern in MSW-compost and sludges.
When properly cured modern MSW-composts containing low levels of metals and
organics were land applied, there was no evidence of adverse effects to humans,
livestock, or wildlife except temporary B phytotoxicity. Adverse effects have only
been found when highly metal contaminated sludges or MSW +sludge-composts
with highly metal contaminated sludges were used at high cumulative application
rates, at very strongly acidic soil pH. Based on the quantitative estimates of
sludge constituent cumulative loadings or concentrations which cause No
Observed Adverse Effect (NOAEL sludges) according to the Pathway Approach for
risk analysis, and strong evidence that this quality sludge and MSW-compost may
be regularly used as part of sustainable agriculture, EPA has proposed using sludge
composition limits (APL = Alternative Pollutant Limits) to regulate low contaminant
sludges. High contaminant concentration sludges would continue to be regulated
by cumulative contaminant application limits.
The "bioavailability" of contaminants in MSW-composts describes the
potential for accumulation in animals of metals or organics from ingested sludges
or composts, or from food/feed materials grown on sludge or compost amended
soils. Risk assessment for direct ingestion is very important since this allows the
greatest potential for transfer for many constituents. Limited feeding studies have
been reported for sludges, while research on ingestion of properly composted MSW
has only recently begun. The presence of high levels of humic materials and
hydrous Fe oxides in sludges, and the presence of other elements with the element
being evaluated, cause the bioavailability of Pb, Cd, and other elements and
-------
organics in sludges to be quite low. Because Cd is ordinarily about 0.5% of Zn in
MSW-composts, it is not possible for compost Cd to cause injury to the most
exposed home gardeners who grow a large fraction of their garden foods on
compost amended soils for a lifetime.
Presently, it appears that the most limiting heavy metal in MSW-composts
may be Pb. A large body of data from feeding studies, and risk evaluation using
the EPA Pb Uptake Biokinetic Model, indicate that composts with up to 300 mg
Pb/kg will not comprise a significant risk to children who inadvertently ingest
compost products. Thus, MSW-compost may provide fertilizer and soil conditioner
benefit in agriculture and horticulture if compost manufacturers carefully reject Pb
rich wastes.
INTRODUCTION
During the preparation, review and revision of the Clean Water Act-503
Proposed Regulation (US-EPA, 1989b), a Pathway Approach to risk assessment
was developed (US-EPA, 1989a). This Pathway Approach is a comprehensive
evaluation of potential worst-case risk to humans, livestock, soil fertility, and
wildlife. It considers all receptors and pathways identified by researchers. As a
result of the 503 process, important lessons have been learned about risk
assessment for land application of sewage sludge, a residual with properties
somewhat similar to those of MSW-Compost. This paper reviews the limited
research on the potential environmental problems which might result from land
application of MSW-compost, and relevant research on sludges and sludge
composts which we believe should provide the basis for development of limitations
for utilization of MSW-composts.
Table 1 shows the Pathways which may allow transfer of compost-applied
contaminants to most exposed individuals (humans, livestock, plants, microbes, or
wildlife) (see Ryan and Chaney (1992) for detailed review of the risk analysis
protocols). As summarized in Chaney (1990a, 1990b, 1992), Chaney, Ryan, and
O'Connor (1991), and other papers, several pathways predominate in risk for
metals or organics because of the chemical properties of the contaminants, soils,
etc. The importance of these pathways was identified during the last 20 years of
sludge risk analysis research (Logan and Chaney, 1983; Chaney et al., 1987;
Chaney and Giordano, 1977). Phytotoxicity from compost-applied Zn, Cu, Ni, and
B is the principle limitation for these elements. Direct ingestion of composts or
sludges by children, livestock or wildlife is the principle limitation on potentially
toxic organics such as PCBs, DDT, etc, and from Pb, Fe, and F. Plant uptake and
transfer to the human food chain is the principal limitation on Cd application, while
transfer to the feed chain for ruminant livestock is the principle limitation for Mo
and Se.
Although these summaries are based on a large body of sludge research in
the field, it is necessary to consider the data from studies of MSW-compost
application to see if results are sufficiently similar to allow development of
limitations for MSW-compost to be based on the more complete sludge database.
-------
Table 1. Pathways for risk assessment of potential transfer of sludge-applied trace contaminants to humans, livestock, or the
environment, and the Most Exposed Individual to be protected by regulation to be based on the Pathway Analysis (US-EPA, 1989a).
Pathway
Most exposed individual
1 Sludge—Soil—Plant—Human
1-Future Sludge—Soil—Plant—Human
1-D&M Sludge-Soil-Plant-Human
2-Future Sludge-*Soil-*Human child
2-D&M Sludge-Human child
3 Sludge—Soil—Plant—Animal—Human
4-Surface Sludge—Animal—Human
4-Mixed Sludge—Soil—Animal—Human
5 Sludge—Soil—Plant—Animal
6-Surface Sludge—Animal
6-Mixed Sludge—Soil—Animal
7 Sludge-Soil-Plant
8 Sludge—Soil—Soil biota
9 Sludge—Soil—Soil biota—Predator
9-Direct Sludge—Soil—(Soil biota)—Predator
10 Sludge—Soil—Airborne dust—Human
11 Sludge—Soil—Surface water—Human
12 Sludge—Soil—Air—Human
12-Water Sludge—Soil—Groundwater—Human
General food chain; 2.5% of all plant-derived foods for lifetime.
Home garden 5 yr after last sludge application; 50% of garden foods for lifetime.
Home garden with annual sludge application; 50% of garden foods for lifetime.
Residential soil, 5 years after last sludge incorporation; 200 mg soil/d.
Sludge product; 200 mg sludge/d for 5 years or 500 mg sludge/d for 2 years.
Rural farm families; 40% of meat produced on sludge amended soil, for lifetime.
Rural farm families; 40% of meat produced on sludge sprayed pastures, for lifetime.
Rural farm families; 40% of meat produced on sludge amended soils, for lifetime.
Livestock fed feed, forages, and grains, 100% of which are grown on
sludge amended land.
Grazing livestock on sludge sprayed pastures; 1.5% sludge in diet.
Grazing Livestock; 2.5% sludge-soil mixture in diet.
Crops; vegetables in strongly acidic sludge amended soil.
Earthworms, slugs, bacteria, fungi in sludge amended soil.
Shrews or birds; 33% of diet is earthworms from sludge amended soil.
Shrews or birds; habitat is sludge amended soil.
Tractor operator.
Water-quality criteria; fish bioaccumulation, lifetime.
Farm households.
Farm wells supply 100% of water used for lifetime.
-------
Unfortunately, potential risks from utilization of MSW-Compost research has not
had the intensity of research using modern scientific technology that sludge
application has received. When sludge research began in the early 1970's, some
research on MSW +sludge composts was included, but little new or detailed work
was conducted on MSW-compost in the U.S. until the 1990s.
Perhaps the most important perspective on the potential for persistent risks
from utilization of composts from separated MSW (ignoring the short term
problems from N-immobilization, inadequately curing, salts, etc.) is the simple
statement that no adverse effects from contaminants in MSW-compost have been
reported other than B toxicity to plants (a temporary problem). Neither Zn, Cu, or
Ni phytotoxicity has been observed, nor have Cd, Pb, or xenobiotic organic
compounds been observed to cause injury to humans, livestock or wildlife.
One adverse effect of compost has been lime-induced Mn-deficiency in low
Mn light textured soils (Haan, 1981). Where other problems from metals or
organics have been identified, they have resulted from composting MSW with
highly contaminated sewage sludge. Although increases in metals or organics in
compost-amended soils have been found as expected, demonstrations of potential
risk from the increases in soil metals have not been reported. Some have
expressed concern that soil metals or organics have exceeded background levels
for agricultural soils. We conclude that the basis for regulating land application of
MSW-composts and sewage sludge should be the potential for compost utilization
to cause adverse effects on agriculture or on the environment due to the metals or
organics in these resources, not the simple soil enrichment with known potentially
toxic metals and organics.
In general, we believe that soil enrichment without demonstrable risk is a
different perspective that agronomists and ecologists must learn how to deal with.
We conclude that utilization of MSW-composts and sewage sludge can provide
significant benefit to sustainable agriculture; compost utilization can safely
continue for an indefinite period without risk to agriculture or the environment.
Thus, this paper is a review of the limited data on the potential adverse effects of
land-applied MSW-compost, and perspectives on risk analysis from our work on
municipal sewage sludge and sludge composts. We believe that an appropriate
risk analysis methodology for potentially toxic contaminants in land-applied organic
residuals has been developed, and that there is little evidence that compost
prepared from MSW will be found to comprise risk to highly exposed individuals
even at very high cumulative applications.
Others have reviewed these subjects, and readers should consider this paper
an extension of the information summarized by these previous workers. The
MSW-compost research in the 1960s and 1970s is important in increasing the
efficiency of our research in the 1990s. We need not "reinvent the wheel" about
many of the questions about MSW-compost, considering that new plans to pre-
separate the compostable fraction of MSW before it becomes contaminated by
other materials will substantially decrease the concentration of many potentially
toxic constituents. Some important reviews include those of Haan, 1981;
Andersson, 1983; Herms and Sauerbeck, 1983; Sauerbeck, 1991; Petruzzelli,
1989; Terman and Mays, 1973; Gallardo-Lara and Nogales, 1987).
-------
COMPOSITION OF MSW-COMPOST:
MSW-compost contains higher levels of many trace elements than do US
background soils, but lower levels than do sewage sludges (Table 2). Modern
sludges contain far lower mean concentrations of metals than found in earlier large
surveys, but many sludges still exceed levels attainable by industrial pretreatment
and treating the drinking water to reduce corrosiveness of the tap water (a
significant source of Pb now that gasoline is Pb-free). The so called "green-
wastes" composts prepared by separate collection of only the compostable fraction
of MSW allow production of composts with lower metal residues than can be
attained by general pre-separation, or by central-separation of MSW into different
fractions. However, just because lower concentrations can be reached in MSW-
composts doesn't mean that they have to be attained to make utilization of MSW-
compost on cropland a valuable practice of sustainable agriculture. Comparison of
US soil metal levels with sludge and MSW metal levels indicates that modern
MSW-composts are only somewhat enriched in metals compared to soils (although
Pb is now higher in MSW-compost than in sewage sludges). The "non-volatile"
fraction of MSW-compost (30-60% depending on the nature of the wastes and
methods of separation utilized [Lisk et al., 1992a]) indicates the maximal
concentration which would be in soils if the soil were comprised of biodegraded
MSW-compost. As noted in Ryan and Chaney (1992), if a compost contains 50%
inorganic matter, the maximum concentration of contaminants in undiluted oxidized
compost would be double the original compost. Analytical results of Lisk et al.
(1992a) are in agreement with the above discussion; further, they showed that
PCBs were quite low in yard waste-, sludge-, and MSW-composts. Lisk et al.
(1992b) noted small variance in metals, etc., in yard waste compost and sludge
compost.
TABLE 2. Geometric mean heavy metal content of composts from mixed MSW
from the United States and separated organic wastes from Europe (dry matter
basis) (MSW-composts from on Epstein et al., 1992) (US sludge data [lognormal
means with multi-censoring] from US-EPA, 1990); "Green" MSW-composts from
Fricke, Pertl, and Vogtmann (1989); NOAEL sludge limits from Chaney (1992); US
soil metals data from Holmgren et al. (1992) (Cd, Cu, Pb, Ni, Zn) or Shacklette and
Boerngen (1984) (Cr).
Element
As,
Cd,
Cr,
Cu,
Pb,
Hg,
/jg/g
^g/g
/^g/g
^g/g
x^g/g
x/g/g
jug/g
MSW-comoosts
No.
Samples
8
72
66
73
73
31
66
Geometric
Mean
2.6
2.0
32.6
107
169
1.09
22.7
"Green"
MSW
Compost
40 '.
86.
0.17
17.
NOAEL
Sludge
Limits
100
25
>3000
1200
300
20
500
US Sludges
Geo. Mean
NSSS
9
6
118
741
134
5
42
.9
.9
»
.2
.7
US Soils
0.
53.
18.
10.
16.
18
0
6
5
72 418 255. 2700 1200. 42.9
Cd/Zn,"/7g/^g 71 0.0055 0.0020 0.015 0.0058 0.0041
-------
IDENTIFIED PERSISTENT PROBLEMS FROM LAND-APPLIED MSW-COMPOSTS
A number of short-duration problems have occurred when high rates of
MSW-composts were applied to cropland (phytotoxicity from biodegradation by-
products in inadequately cured compost; excess soluble salts; N-immobilization).
Fortunately, good management of MSW composting or utilization can avoid these
serious limitations to beneficial use of MSW-compost.
However, two significant persistent agricultural problems have occasionally
been observed in fields amended with MSW-compost: Boron phytotoxicity and
Mn-deficiency. Each has occurred under unusual conditions, and the potential for
yield reductions were very site specific. Further, high rates of compost application
used in research were required to cause the B phytotoxicity or Mn-deficiency, and
these rates are much higher than commonly applied in normal agricultural
practices.
Boron Phytotoxicity: In contrast with municipal sewage sludge, MSW-
compost contains substantial levels of soluble boron (B). B toxicity from sewage
sludge application was reported only for an unusual case of a sensitive tree species
growing in soils amended with a sludge containing lots of glass fibers
(Vimmerstedt and Glover, 1984; see also Neary et al., 1975, regarding high B
levels in phosphate-free detergents). The glass fibers contained borosilicate and
release of B caused phytotoxicity. Research has shown that much of the soluble B
in MSW-compost comes from glues (Volk, 1976). It has long been known that
plant samples placed in paper bags can become contaminated from B from glue
used to hold the bag together. El Bassam and Thorman (1979) and Gray and
Biddlestone (1980) noted that the B level in MSW-composts was quite variable as
might be expected if composts are not well mixed.
In general, B phytotoxicity has occurred when high application rates were
used, and B-sensitive crops were grown. However, when MSW-compost is used
at fertilizer rates in normal fields, the B might be important as a fertilizer rather
than as a potential phytotoxicity problem.
Boric acid and most borates are quite water soluble, although B can be
adsorbed on clays and by organic matter. Low soil pH facilitates B uptake by
plants because the H3B03 molecule (predominant form at lower soil pH) is absorbed
by roots rather than anionic borates (Oertli and Grgurevic, 1975). Although most
B toxicity has been reported on alkaline soils, this is due to the lack of leaching for
most of these soils. Excess applications of soluble B are much more phytotoxic in
acidic soils, and liming can correct B phytotoxicity. The usual liming action of
compost should help prevent this problem.
There are large differences among crop species in tolerance of excessive soil
B. Some crops are very sensitive, and these are the species which have suffered
phytotoxicity from compost-applied B (bean, wheat, and mum). Francois has
summarized the significant differences among several groups of crops (Francois
and Clark, 1979; Gupta, 1979; Francois, 1986). Ornamental horticultural species
have been examined to some extent (information on individual species can be
found by literature searching), but many horticultural crops have not been studied.
This is one research need related to practical microelement phytotoxicity from
-------
compost.
Perhaps the first report on B toxicity from MSW-compost is that of Purves
(1972) who noted B phytotoxicity to beans on field plots which received high rates
of MSW-compost. The full description of the compost experiment is reported in
Purves and Mackenzie (1973), and a careful examination to prove B phytotoxicity
was reported by Purves and Mackenzie (1974). Bean (but not potato or other
species examined) suffered severe yield reduction at high compost rates; this yield
reduction was proportional to rate of compost application. Bean is known to be
especially sensitive to B phytotoxicity. Gray and Biddlestone (1980) also found B
phytotoxicity in sensitive species grown in field plots with high rates of MSW-
compost.
Gogue and Sanderson (1975) reported B phytotoxicity to chrysanthemums
in potting media containing MSW-compost. Foliar analysis clearly supported the
conclusion that B was toxic and that Mn, Cu, Zn, and other elements were not at
toxic levels. They conducted a calibration experiment to determine the sensitivity
of chrysanthemums (Gogue and Sanderson, 1973), and the levels found in the
mums grown on the test media were in the phytotoxic range. In their research,
they adjusted the pH of the media to 6 using sulfur, rather than allowing the MSW-
compost to raise the pH of the media. This probably contributed to the severity of
B phytotoxicity observed. Some other horticultural species also suffered B
phytotoxicity in compost-containing media (Gilliam and Watson, 1981). Sanderson
(1980) reviewed B toxicity in compost amended potting media. In contrast to
MSW-compost, sewage sludge composts with wood chips have not been found to
cause B phytotoxicity (Chaney, Munns, and Cathey, 1980). Only a few acid-loving
species require acidification of media to do well on neutral compost-amended
media.
Interestingly, because the B which causes phytotoxicity is water soluble, the
B phytotoxicity problem from MSW-compost is short-lived. Purves and Mackenzie
(1973) noted that pre-leaching MSW-compost prevented B phytotoxicity. Other
studies noted that the B-phytotoxicity occurred only during the year of application,
and that soluble B was leached out of the root zone over winter (Volk, 1976) or by
leaching potting media with normal horticultural watering practices. Sanderson
(1980) noted that perlite also adds B to potting media, and that use of both may
cause B toxicity when either perlite or MSW-compost alone might not have done
so. Lumis and Johnson (1982) studied leaching of B in relation to toxicity of salts
and B to Forsythia and Thuja. They reported that a simple leaching treatment
removed excess soluble salts, but was unable to remove enough B to prevent
phytotoxicity (the compost they studied contained 225 mg B/kg, higher than most
reports). Nogales et al. (1987) also found compost-applied B leached quickly such
that crop B was reduced in each successive ryegrass crop.
B phytotoxicity is significantly more severe when plants are N-deficient
(Gogue and Sanderson, 1973; Nogales et al., 1987; Gupta et al., 1973). This
makes the B in MSW-compost which is not properly cured (to avoid N
immobilization) potentially more phytotoxic than in well cured composts. Further,
B flows with the transpiration stream and accumulates in older leaves. In
environments with low humidity, more transpiration occurs (e.g., greenhouses),
and B toxicity is more severe, B and salt toxicity are easily confused; both are first
-------
observed in leaf tips or margins of older leaves. Diagnosis of B phytotoxicity
requires a knowledge of relative plant tolerance of B, or analysis of the leaves
bearing symptoms.
Thus, in general use, compost application at a reasonable fertilizer rate
would simply add enough B to serve as a fertilizer for B-deficiency susceptible
crops such as alfalfa or cole crops. However, use of MSW-compost at high rates
in soils or potting media could cause phytotoxicity if high soluble B were present.
The B phytotoxicity would not be persistent because soluble B would leach from
the root zone with normal rainfall or irrigation. Compost-applied B would be more
phytotoxic in N-deficient soils, which might result from application of improperly
cured compost. Water soluble B should be one chemical which is regularly
monitored in MSW-composts so that the need for warning about rates of
application and use with sensitive crops can be identified. Deliberate use of MSW-
compost as a B fertilizer for high B-requiring crops such as the cole crops (cabbage
family) might become a regular agronomic practice. Sources of soluble B in
modern MSW-compost should be evaluated, and alternative to B use identified.
Compost-induced Mn deficiency. In contrast with most sewage sludges,
application of MSW-compost usually raises the pH of the soil-compost mixture.
Sludges usually contain more reduced N and S, and oxidation of these after mixing
sludge with soil generates acidity. Some sludges from areas with hard water do
contain enough lime equivalent to correct the acidity they add to the soil, but all
MSW-composts have been reported to contain lime equivalent. This could come
from use of CaCO3 and other materials as fillers in paper, or from stabilization of
crop residues.
When MSW-compost was added to naturally low Mn acidic soils, the
resultant high pH was been found to cause Mn-deficiency in some cases. Haan
(1981) noted that Mn deficiency occurred in several cases in the Netherlands, and
Andersson (1983) noted this effect in some Swedish soils.
One way to assure that MSW-compost does not cause Mn deficiency is to
add Mn to the MSW during composting (inclusion of identified industrial Mn
wastes or Mn ore). Composts usually contain fairly low Mn levels. Most alkaline
soils do not cause Mn deficiency if they contain high enough total Mn, and
composts with added Mn should prevent this problem. Crops differ substantially in
susceptibility to lime-induced Mn deficiency. Soybean and wheat are well known
to suffer severe Mn deficiency when other crops (e.g. corn) grown on the same
soil have no Mn deficiency.
Besides the pH of the soil and the susceptibility of the crop, the native Mn
level of soils are important in whether Mn deficiency will be induced by lime rich
sludges or composts. In general, Mn concentration in soils increases with
increasing clay content. Besides coarse texture, a very important factor in
affecting loss of Mn from soils is height of the water table. Soils which were
submerged during soil formation have had Mn02 reduced to Mn2+ and leached from
the soil. Thus, coarse-textured, Coastal plain soils are often very susceptible to
Mn deficiency. In a long term field experiment with a single 1976 application of
high rates of lime-treated anaerobically digested sewage sludge applied to
Galestown loamy sand at Beltsville, severe Mn deficiency was noted in wheat and
8
-------
soybean grown in 1991 and 1992 (R.L. Chaney and B.R. James, unpublished). In
previous years, corn had been grown and no apparent deficiency occurred.
Lime induced Mn-deficiency has also become a problem in some cases when
high metal sludges were used at such high cumulative rates that the soils had to
be limed to prevent metal phytotoxicity. Spotswood and Raymer (1973) noted
that crops on a sewage farm which also received a high metal concentration sludge
suffered Mn deficiency when lime was applied to prevent Zn toxicity. In that case,
the repeated heavy irrigation with sewage caused depletion of soil Mn, increasing
the potential for liming to induce deficiency (as was observed at sewage irrigated
light textured soils on sewage farms at Paris, France, and Berlin, Germany; Doring,
1960; Rinno, 1964; Rohde, 1962; Trocme et al., 1950).
It seems clear that MSW-compost manufacturers need to consider the
potential of MSW-compost to induce Mn deficiency if the soils in their marketing
region are susceptible to Mn deficiency, and the crops commonly grown include
susceptible species. The manufacturer could warn users of this potential problem,
or could choose to add Mn during composting to assure that Mn deficiency would
not occur. Research has not yet clarified the amount of compost-Mn required to
avoid Mn deficiency on susceptible soils.
HEAVY METAL CONCERNS IN USING MSW-COMPOST ON CROPLAND
Because metals in MSW-compost are conserved in the soil-compost mixture,
application of MSW-compost to cropland causes an increase in the concentration
of potentially phytotoxic heavy metals (Zn, Cu, Ni) in soils. Many scientists have
expressed concern about this simple increase in soil metals, and have implied that
this is a problem. As noted above, we believe the potential for adverse effects of
heavy metals should be the basis for concern, not the simple presence of metals in
soils. It is important that we understand that metals in sludges and composts with
low concentrations of metals have not been shown to cause adverse effects, and
that an improved understanding of the chemistry of sludges and composts appears
to explain the low potential for phytotoxicity and phytoavailability of metals in low
metal concentration sludge and compost materials.
Proper approach to evaluate potential compost heavy metal questions: Over
25 years of research have been conducted to better understand the potential for
risk from heavy metals in sewage sludge applied to agricultural land. During this
period, a number of principals of "heavy metal agronomy" have been identified.
Foremost among these is the recommendation from the W-170 Peer Review
Committee Report (Page et al., 1989): In development of regulations, use results
from "field studies with municipal sludge instead of non-field studies with metal
salts or pure organic compounds." This recommendation was made because
research showed that pot studies in greenhouses, metal sources other than sludge,
or even studies on high contaminant concentration sludges were not valid for
evaluation of risks from sludges with low concentrations of these contaminants.
Many studies were conducted to determine the relationship between plant
uptake and tolerance of metals in pots vs. the field, and from metal salts, metal
-------
salt amended sludges, and sludges of different quality (see Logan and Chaney,
1983; Page et al., 1987). Some studies included comparison of plants grown in
pots inside and pots outside the greenhouse compared to plants grown with equal
sludge applications in the field (deVries and Tiller, 1978; Davis, 1981). When
sludge was applied in the field, much lower [(plant metal concentration):(soil metal
concentration)] slopes were obtained than when outdoor pots were used with the
same soil; indoor pots had even higher slope, about 3-10 fold higher than in the
field. This is now understood in terms of the differences between salts and
sludge, and between pots vs. the field (see also deVries, 1980). Pot studies
overestimate metal phytoavailability because: 1) The indoor and outside
environments differ in soil temperature and water use patterns (the humidity
microenvironment in a greenhouse is quite unlike the field; in the greenhouse,
transpiration is increased which increases metal flow to the root by convection and
transfer to leaves in the transpiration stream); 2) In pots, the whole amount of
fertilizer nutrients required to support the growth of the test plants must be applied
to a limited soil volume; this soil volume has much higher soluble salt concentration
which increases the concentration of metals and diffusion of metals from the soil
particles to the roots; 3) When fertilizers contain NH4-N, rhizosphere acidification in
the small volume of soil in a pot can increase metal uptake; and 4) In pots, the soil-
sludge mixture comprises the whole rooting medium, while in the field the sludge is
only mixed into the tillage depth (usually < 20 cm deep) and much of the plant
root system is below this depth.
Perhaps the biggest source of difference among these incorrect methods to
evaluate sludge metals is the difference in uptake and toxicity from metal salts vs.
sludge-metals. In many studies, sludge or metals equivalent to the sludge were
added to the same soil, and crops grown. In many studies the salts caused severe
phytotoxicity, while the sludge caused yield increase. Although many of these
studies suffered from errors due to difference in pH between the salts and sludge
(added metal salts displace protons from the soil and lower pH), some had equal
pH. For example, in the greenhouse pot study of Korcak and Fanning (1986),
equivalent metal salts or 224 Mg/ha of sludge were added to a number of soils
with widely different properties; salts caused phytotoxicity to corn on all soils, but
sludge caused no phytotoxicity. Soil properties strongly affected metal uptake on
the metal-salt-amended soils, but had little effect on the sludge-amended soils.
Some comparisons of metal-salts and sludge were conducted in the field. For
instance, Ham and Dowdy (1978) compared metal uptake by soybean when
equivalent metals and sludge were applied in the field, and found much higher
metal uptake from the salts. Although metals added as salts may approach the
phytoavailability of sludge-applied metals over time, the lack of other sludge
constituents makes results from study of additions of single metals of little value
(Bell, James and Chaney, 1991).
Another pattern related to the effect of sludge metal binding properties
became apparent in the early 1980s. R.B. Corey (University of Wisconsin,
Madison) had predicted (at the 1980 annual meeting of the W-170 Regional
Research Committee) that sludge adsorption chemistry should control the activity
of free metal ions in the soil solution of sludge-amended soils after reaching the
sludge application rate which saturated the soil metal binding sites (see also Corey
10
-------
et al., 1981). Based on this model, Chaney et al. (1982) used orthogonal contrast
analysis of variance to analyze data from a long-term study of lettuce uptake of Cd
from sludge-amended field plots and found that the rate-squared term was highly
significant. This indicated that use of simple linear regression to evaluate data
from sludge studies was in error. Subsequently, Logan and Chaney (1987) used
plateau regression to evaluate these data. Figure 1 shows several approaches to
evaluate the effect of application rate of a low Cd sludge on the uptake of Cd by
lettuce (averaged over 1976 to 1983). The plateau regression predictions, and
their 95 percent confidence intervals are shown for each soil pH, as are the simple
linear regressions. These data clearly demonstrate the over-estimation of Cd
uptake when simple linear regression is used to evaluate plateau response data.
With time, other studies were evaluated and found to fit this curvilinear response
pattern (Corey et al., 1987; Chang et al., 1987).
Based on these understandings, researchers attempted to characterize the
chemical aspects of sludge which made metals so much less available to plants
(phytoavailable) than were metal-salts. A review and interpretation of this
information was published by the Corey et al. (1987) workgroup. In short, the
specific metal adsorption capacity (ability to selectively adsorb heavy metals in the
presence of 3-10 mM Ca2+ present in the soil solution of most fertile soils) of
sludge persistently increases the ability of the soil-sludge mixture to adsorb metals,
thereby reducing the phytoavailability of sludge-borne metals. As noted below,
because the sludge chemistry controls the phytoavailability of sludge-applied
metals, plant uptake approaches a plateau with increasing sludge application rate
rather than showing the usual linear increase with increasing applications of metal-
salts.
Another aspect of these data showing that sludge chemical factors reduce
the phytoavailability of sludge metals is that it takes time for the reactions of
metals to reach their lowest "free energy" condition; by this we mean that by the
time sludge metals are applied to soils, the metals have reached strong adsorption
sites in the sludge, greatly reducing their phytoavailability compared to fresh
additions of metal salts to soils. Soils and sludges contain metal binding with a
wide range of specificity for metal adsorption; freshly added metals are bound to
the population of all binding sites, then slowly equilibrate to the strongest specific
adsorption sites. Several scientists evaluated the extractability and
phytoavailability of sludge metals when the metals were added to the sludge
before anaerobic digestion, or after digestion (Bloomfield and McGrath, 1982;
Cunningham et al., 1975a, 1975b, 1975c; Davis and Carlton-Smith, 1981, 1984).
In each case, adding the metals after digestion (immediately before application to
soil) caused the metals to be much more phytoavailability than metals added during
sewage treatment or before sludge stabilization. However, metals added to sludge
were less phytoavailable than metal salts added to the soil without the sludge.
This could result from the presence of high levels of many metals competing for
the strong adsorption sites. The weaker sites are filled and equilibration is more
rapid during sludge stabilization when concentrations are higher (compared to
dilution with soil). Reaching the strongest binding sites should take a long time
when metals salts are added directly to soils.
11
-------
~ 3
O
CD
O
=5
CD
0
Geometric mean of field
data for lettuce and spinach /
on acidic soils (pH < 6).
Hayden Farm
Mean for 1976-1983.
0 250 500 750 1000 Mg/ha
0 3.4 6.7 10.0 13.4 kg/ha
SLUDGE and CADMIUM APPLICATION RATES
Figure 1. Linear vs. plateau regression analysis of lettuce uptake of Cd from Christiana fine sandy loam amended with 0, 56, 112, or 224
Mg dry heat-treated sludge/ha, and pH adjusted to 6.2-6.5 with limestone (Hi pH) or uncontrolled (£5.5 in 1983) (Lo pH). Predicted
responses extrapolated to 1000 Mg/ha to show implications of the data. Results are average for 1976 to 1983. Data points shown are
arithmetic means ± one stnd. error; plateau regressions show predicted (dashed lines) with ±95% confidence interval (dotted lines).
Equations for linear regressions (solid lines) are: Lettuce Cd= 1.22 + 0.291 -Rate (low pH); Lettuce Cd = 0.774 + 0.0900*Rate (High
pH). Sludge applied in 1976 contained 13.4 //g Cd, 1330 fjg Zn, and 83 mg Fe/g dry weight (data originally reported in Chaney et al.
[1982]). Small dashed line shows the geometric mean simple linear regression slope for increased lettuce Cd on all strongly acidic
(pH<6) field soils used in the CWA-503 final rule, over-estimating effect of low Cd sludges.
-------
The importance of metal adsorption by sludges was also seen when
researchers examined the relationship of pH to solubility of metals in sludges or
soil-sludge mixtures. In all heavy metal cation (Zn, Cd, Cu, Ni, Pb, Hg) studies,
solubility increases with decreasing pH. Sanders and co-workers found for each
sludge and metal, that as pH was decreased, some threshold pH was reached
below which metal solubility was sharply increased. They then studied the effect
of metal concentration in the sludge on this threshold pH. Adams and Sanders
(1984) found that the higher the sludge metal concentration, the higher the
threshold pH point of increasing metal solubility (see also Sanders and Adams,
1987; Sanders et al., 1986). This can readily be interpreted in terms of filling the
specific metal adsorption sites vs. sludge metal concentration.
These bodies of data on specific metal adsorption by sludge constituents is
very important in understanding sludge metal research. In studies of phytotoxicity
of sludge-applied metals, it is now clear that phytotoxicity to sensitive crop species
has only resulted when high metal concentration sludges were used, or extremely
low pH was reached: 1) When high cumulative applications of low metal sludges
(NOAEL quality) were applied, and soil pH allowed to drop to near 4.5,
phytotoxicity to soybean (Lutrick et al., 1982) and rye (King and Morris, 1972)
were observed; simple correction of soil pH to near 6 completely corrected yield
reduction; normally, good agricultural practice requires that soil pH be ^ 5.5 for
nearly all crops to avoid natural Al and Mn phytotoxicity of more strongly acidic
soils; Mn and Al contributed to or caused the yield reductions noted by Lutrick et
al. and King and Morris; and 2) High metal sludges at lower cumulative applications
caused metal phytotoxicity which was not simply corrected by liming the soil
(Marks et al., 1980; Webber et al., 1981; Minnich et al, 1987). When sludge Zn,
sludge+ MSW compost Zn, and ZnS04 were applied at equal Zn rates, only the Zn
salt caused phytotoxicity even when soil pH levels were made equal by addition of
sulfur to acidify the sludge and MSW +sludge compost plots (Giordano et al.,
1975).
Specific metal adsorption is involved in the effect of sludge metal
concentration on the phytoavailability and bioavailability of sludge metals. It had
been apparent from many studies that sludges with higher metal concentrations
could cause higher metal uptake by plants when equal amounts of metals were
applied (i.e., different amounts of sludge dry matter and hence adsorption capacity
were applied). This was part of the plateau response data set. Recently, Jing and
Logan (1992) reported on the phytoavailability of sludge applied Cd from many
different sludges, where equal amounts of Cd were applied in each pot. Crop
uptake of Cd increased with increasing sludge Cd concentration. This is explained
in terms of the filling of specific Cd binding sites in the sludge; the population of
Cd binding sites vary widely in strength of specific Cd adsorption; as sludge Cd
concentration increases, the least strongly bound Cd is more phytoavailable.
Similarly, when amounts of metals required to reduce yields of barley or vegetables
were determined with salts in greenhouse pots, with mixtures of high metal
sludges in pots in the greenhouse, or with normal quality sludges in the field, the
salts and high metals sludges caused phytotoxicity (Davis and Carlton-Smith,
1984), but the normal quality sludges caused only yield increase (Johnson,
Beckett, and Waters, 1983).
13
-------
One question of importance for use of sludge and MSW-compost in
sustainable agriculture is: "How long does the reduced metal phytoavailability of
sludge-applied metals due to sludge specific metal adsorption capacity last?".
Some field plots have been studied up to 20 years after the last sludge application.
Other soils from long-term sludge or sewage farms have been examined by basic
studies in the greenhouse. The demonstration of persistence of the "sludge effect"
on metal sorption was well illustrated by the data of Mahler et al. (1987, 1988a,
1988b) in which Cd rich sludge or Cd salts were added to soils from long-term
sludge plots, and a high Cd accumulating crop grown. The slope of the crop
response to the added salt-Cd or fresh sludge-Cd was lower for soils with historic
sludge application due to the increase in metal adsorption capacity of the sludge-
amended soils (pH was not different between treatments). All evidence available
indicates that the specific metal adsorption capacity added with sludge will persist
as long as the heavy metals of concern persist in the soil. Although this effect
strongly confounds estimating the phytoavailability of Cd in different soils which
received different amounts of different sludges, it is clear that the specific metal
adsorption capacity added by sludge plays a very significant role in controlling the
phytoavailability of metals of concern regarding phytotoxicity or food-chain
contamination.
The inorganic part of the sludge contributes much of the sludge-applied
specific metal adsorption capacity. As summarized by Corey et al. (1987), Fe, Al,
and Mn oxides in soil and sludge exhibit specific metal adsorption properties. As
noted above, even though sludge organic matter is oxidized over time, if soil pH
does not fall, the ability of crops to accumulate soil metals is only decreased over
time. This indicates that the non-organic matter adsorption sites are adequate to
protect against metals added in sludges. Part of the sludge-applied specific metal
adsorption capacity is due to humic acids formed from sludge organic matter;
interestingly, metals stabilize soil humic acids against biodegradation. Further, in
the long term, part of the added metals become occluded in Fe oxides (Bruemmer
et al., 1986).
All these data from research on sludge vs. metal salts, and the effect of
sludge metal concentration on phytoavailability of sludge-applied metals (including
the plateau response finding of Chaney et al., 1982) led the Corey et al. (1987)
workgroup to conclude that specific adsorption of metals by sludge surfaces would
normally be the controlling factor in metal phytoavailability in soil-sludge mixtures.
They concluded that a plateau response would be the expected pattern of
response, and that some sludges could be so low in metals, and so high in metal
specific adsorption capacity that addition of sludge could actually reduce metal
uptake by plants. This response has been observed for Cd with several studies in
pots and field. This model integrates data from many studies which initially
appeared to offer conflicting results. Sludge-applied Cd is additive, but along a
plateau response curve rather than a linear response curve.
Very similar conclusions about sludge constituents binding metals could
have been drawn from the animal literature with sludge or compost amended diets.
In numerous studies to assess the risk from sludge contamination of diets of
grazing livestock, different livestock species were fed sludges or composts for
prolonged periods. Sheep and cattle are notoriously sensitive to excess dietary Cu,
14
-------
and the sludges added as much as 5-10 times higher Cu than required to kill cattle
or sheep if Cu salts are mixed into practical diets. Surface application of high Cu
swine manure to pastures for sheep did not cause Cu toxicity (e.g. Poole et al.,
1983; Bremner, 1981). Moreover, depletion of liver Cu reserves or even frank Cu
deficiency was the common result unless sludge Cu concentration was above
1000 mg/kg (Baxter et al., 1982, 1983; Bertrand et al., 1981; Decker et al.,
1980a; Sanson et al., 1984). Thus, the bioavailability of sludge metals was very
low compared to metal salts (based on both toxicity and on liver metal
concentrations). Similar results were seen for bioavailability of Pb and Cd in
ruminants fed sludge, and for Cu and Cd in non-ruminants fed sludge (Logan and
Chaney, 1983).
Another source of over-estimation of sludge metal phytoavailability has
resulted from high rates of application of sludge in field research studies. Often,
high rates are applied at one time to apply high cumulative rates of sludge in a
short time rather than applying N-fertilizer rate sludge application rates for 20-50
years. In numerous studies, crop uptake of Cd and other metals has been followed
for a number of years after application ceased. Crop uptake fell by as much as 80-
90% compared to the last year sludge was applied (e.g., Bidwell and Dowdy,
1987; Chang et al., 1982; Hinesly et al., 1979). One significant cause of this
pattern is the biodegradation of sludge organic matter. When high rates of sludge
application are used, the biodegradation rate can be so high that anaerobic
biodegradation by-products are formed in the soil, and these increase metal
diffusion from soil particles to plant roots. This is well illustrated by the study of
Sheaffer et al. (1981) reported in Logan and Chaney (1983). On plots treated with
112 Mg/ha of a higher metal concentration sludge, soil temperatures were varied.
Immediately after mixing sludge and soil and imposing soil temperature, radishes
were sown. At high soil temperature (which hastens biodegradation) severe
phytotoxicity resulted in stunted radishes, no edible globes, and high enough Zn
and Cu in leaves to indicate phytotoxicity (>1000 mg Zn/kg and >60 mg Cu/kg);
in the second year and 6th year after the sludge application, radishes were again
grown but no phytotoxicity resulted. In year 2, soil pH on the sludge treated plots
had dropped, and pH was corrected to the pH of the control soil before cropping in
the 4th year. Not only was no phytotoxicity seen in either year 2 or year 6, but
also foliar Zn and Cu were appropriate for healthy radish; and normal radish globes
resulted. Thus, rapid biodegradation of higher rates of sludge can cause temporary
increase in sludge metal phytoavailability and over-estimate the risk of sludge metal
phytotoxicity. This error would be expected to be greater for higher metal
concentration sludges.
These conclusions should have been apparent to the scientific community
earlier than 1980, but concern about metal enrichment of soils caused great
caution by researchers. Not only are NOAEL sludges and composts able to be
used as fertilizer and soil conditioner with very low risk of phytotoxicity or
excessive food-chain transfer of metals, but these sludges have also been found to
be able to correct (remediate) soils which were already metal toxic (Gadgil, 1969;
Bergholm and Steen, 1989), although sludges are clearly more effective than
MSW-composts in correcting severe phytotoxicity from soil metals. Metal
phytotoxicity from mine or smelter wastes or corrosion residues were corrected in
15
-------
a number of studies (increase in soil pH was not the basis for correction of
toxicity). This too shows the specific metal adsorption capacity of sludges can
control phytoavailability in the soil-sludge mixture.
Thus, only data from field studies of low contaminant concentration sludges
or composts are appropriate for development of regulations for these materials.
The lack of adverse effects from use of NOAEL sludges, and even lower
concentrations of metals in MSW-composts, should be considered a valid basis for
development of risk-based quality standards for MSW-compost products which
could be marketed for general use.
Can Cd in MSW-compost cause risk to the human food-chain: Since 1969
when the itai-itai disease of Japanese farm families was attributed to consumption
of rice containing high levels of Cd, scientists have expressed high concern about
food Cd and about Cd contamination of soils. However, we now know that this
concern was based on ignorance of the factors which control risk to humans from
soil containing increased levels of total Cd (Ryan et al., 1982). McKenna and
Chaney (1991) McKenna et al. (1992), and Chaney (1990b; 1992) recently
summarized new concepts of the food-chain risk from Cd in land-applied MSW-
compost and sewage sludges. Excessive dietary Cd can accumulate over one's
lifetime in the kidney cortex and cause renal tubular dysfunction (Fanconi
syndrome), a disease in which low molecular weight proteins are excreted in urine.
Although farm families in Japan experienced this disease after prolonged
consumption of rice grown on highly Zn + Cd contaminated paddies, the properties
of rice and flooded soils, and malnutrition in Japan before, during, and after World
War II, played very important roles in allowing high transfer of soil Cd to kidneys.
The rice grain was greatly increased in Cd but its Zn concentration was not
increased because ZnS was formed in flooded soils; crops grown in aerobic soils
usually have a greater increase in Zn than Cd in edible crop tissues.
In another case, New Zealand oyster fishers and their families consumed
high amounts of Cd-rich oysters, ingesting nearly as much Cd as the Japanese
who suffered Cd disease. However, because oysters or the New Zealand diet are
not deficient in Ca, Zn, or Fe, these persons did not suffer tubular proteinuria
(Sharma et al., 1983), and did not accumulate high amounts of Cd in their kidneys
(McKenzie-Parnell and Eynon, 1987; McKenzie et al., 1988). Thus, the
bioavailability of Cd in different foods or diets can be quite different. In two
locations (Shipham, UK [Strehlow and Barltrop, 1988] and Stolberg, FRG [Konig et
al. 1991]) vegetable garden soils were highly contaminated with Zn and Cd from
mining wastes, which caused garden crops to be Cd enriched, yet no tubular
proteinuria resulted in long-term residents who consumed high amounts of garden
crops.
This difference between effect of Cd in rice and Cd in other foods is
evidence that Cd has different bioavailability depending on the presence of
different nutrients in the same food, and perhaps depending on the chemical
speciation of Cd in the foods. In studies of the bioavailability of Cd in sludge-
grown food, Chaney et al. (1978a; 1978b) fed lettuce and Swiss chard (grown on
both control and sludge amended soils) to mice or guinea pigs, respectively. Chard
had up to 5-fold higher Cd when grown on strongly acidic sludge amended soils,
16
-------
but caused no change in kidney or liver Cd concentration. When grown on
digested sludge-compost amended soil, lettuce had 2-times the Cd of the control
crop, yet caused significant reduction in kidney Cd compared to the control. Thus,
Cd concentration in crops is not related to the risk of Cd from those crops because
the bioavailability of the crop Cd can be affected by other elements in the sludge or
compost.
In order to estimate the maximum allowable increase in Cd in garden crops,
Chaney et al. (1987) extended the dietary models of Ryan et al. (1982) relating Cd
in lettuce vs. Cd in the garden foods part of the diet grown on a Cd enriched soil
(Table 3). In strongly acidic soils which cause increased Cd levels in foods, the
relative uptake of Cd was fairly consistent (Chaney et al., 1987). By multiplying
the dry weight of each food group by its relative increased Cd uptake on acidic
sludge amended soils, one can estimate that diet Cd will be increased 1.67 jjg/day
when lettuce is increased by 1 //g/g dry weight (100% of garden foods grown on
the amended soil). [As discussed in Ryan and Chaney (1992), it is extremely
unlikely than individuals will grow a substantial fraction of their garden vegetables
for a lifetime, always using strongly acidic soils, and always having a poor quality
diet which favors Cd absorption.]
Cd in MSW-composts appears to be even less likely to cause food-chain Cd
problems than can the Cd in sewage sludges because the Cd:Zn ratio of
MSW-composts is about 0.005 compared to the 0.010 of domestic sludges
(Table 2) (Chaney, 1992). For many years, Chaney has noted that the Zn which
accompanies Cd in sludge and compost provides further protection against
excessive dietary Cd (see Logan and Chaney, 1983; Chaney, 1990b; McKenna and
Chaney, 1991). The worst-case scenario for food-chain Cd risk (Pathway 1F) has
always involved acidification of the amended soil to very acidic pH which promotes
Cd and Zn uptake by plants. Besides interactions of Zn and Cd which reduce plant
Cd bioavailability, another basis for the protection from Cd risk due to Zn in sludge
and compost is that Zn phytotoxicity occurs in crops if leaf Zn exceeds about 500
mg/kg, thus limiting yield of Cd-rich foods. Poor yields and visual symptoms of
Table 3. 1991 Home garden dietary Cd risk assessment, using lifetime diet
model, and relative Cd uptake among garden crops (Chaney, 1990b; Chaney et
al., 1987).
FOOD GROUP Food Relative Increased Diet Cd
Intake Cd Uptake if lettuce Cd increased
by 1 fjg/g DW
g DW/d Lettuce=l
Leafy Vegetables 1.97 0.536 1.056
Potato 15.60 0.020 0.312
Root Vegetables 1.60 0.096 0.154
Legume Vegetables 8.75 0.010 0.088
Garden Fruits 4.15 0.014 0.058
All Garden Foods 1.67
17
-------
problems such as chlorosis (in more sensitive crops such as bet and lettuce) alert
the gardener to the need to identify the reason for the toxicity. Thus, in sludge- or
compost-amended soil, Zn becomes a "natural" factor which limits Cd risk to
gardeners who consume a substantial portion of their diet grown on amended soils.
Either they maintain reasonable soil pH for vegetable crop production (which
protects them from increased crop and diet Cd), or eventually, when the soil pH
drops enough to allow high Cd uptake and potential Cd risk, Zn phytotoxicity
reduces yield and hence reduces potential for consumption of Cd-enriched garden
crops.
Clear evidence of this protection is found in Chaney's (1992) analysis of
data published by Baker and Bowers (1988). They grew lettuce in gardens
contaminated by Zn-smelter emissions over the last century. Garden soil Cd
reached as high as 100 mg/kg, and Zn, 10,000 mg/kg. Gardeners added limestone
and livestock manure to their soils to reduce the effect of soil Zn and many grow a
wide variety of garden crops. As part of an effort to assess need for remediation
under a Superfund "Remedial Investigation", Baker and Bowers grew Romaine
lettuce in many gardens. Chancy (1992) calculated the Cd:Zn ratio for each
garden, and designated each point on Figure 2 as belonging to one of three classes
of Cd:Zn, < 0.010, 0.010-0.020, and > 0.020. It is clear that all gardens with
Cd:Zn < 0.010 produced lettuce which would increase diet Cd no more than
about 10 /JQ Cd/day. This was for 100% of garden crops grown on the acidic
garden rich in Cd+Zn for 50 years, an extremely unlikely event. MSW-compost
has Cd:Zn about 0.005, which indicates that the likely worst case for gardens with
high rates of MSW-compost would be about 5 fjg Cd/day. Present U.S. daily
intake of Cd is about 12 /jg/day (lifetime diet model) (based on Adams, 1991).
The Risk Reference Dose (RfD) for Cd is 70 /yg/day, with a difference between RfD
and normal intake of 58 jwg/day (if lettuce were increased about 35 /jg Cd/g DW
[(58 j/g/day) •*- (1.67 //g/day if lettuce increase by 1 fjg/g) = 34.7 j/g/g allowable
increase in lettuce Cd], 100% of garden vegetables would be increased by 58
//g/day). The RfD is designed to protect the highly exposed persons with sensitive
kidneys from lifetime consumption of excessive Cd. Of course, the protections
from Zn reducing Cd bioavailability in sludge-grown crops discussed above would
also occur, making this small increase in crop Cd of even lower significance to
humans.
Researchers have worried about Cd in sludges and composts since the
1970's, and conducted much research on this subject. Although we still conduct
research on crop Cd bioavailability to settle other specific questions about risk from
Cd in foods, we now conclude that uncontaminated sludges and MSW-composts
comprise no Cd risk even in extremely worst-case risk analysis scenarios. The
improvement in our understanding of soil Cd risk during the last few years, ending
with the more valid soil Cd:diet Cd model summarized here, strongly supports this
conclusion. The low bioavailability of crop Cd noted above supports this
conclusion. And the new evidence on natural limitation of increased diet Cd due to
Zn which accompanies Cd shown in Figure 2, supports this conclusion.
Henceforth we should no longer consider that Cd in uncontaminated (NOAEL
sludge) sludges or MSW-composts comprise any food-chain Cd risk to humans
consuming Western diets under any conditions.
18
-------
^50
~D
CD
3.40
•N
~o
o
D
CD
D
CD
O
c
^20
10
0
; Baker & Bowers, 1988
Presumes 100% of Garden Foods
Corrected for Yield Potential
0 Cd/Zn»<0.01
D Cd/Zn=0.0 1-0.02
A Cd/Zn»=>0.02
A
A
a
0
25 50 75 100 125 150 175
Lettuce Cd, mg/kg DW
Figure 2. Predicted Yield Corrected Increased Dietary Cd for consumption of 100% of garden vegetables from Zn + Cd contaminated
gardens near Zn smelters in Pennsylvania (corrected for yield reduction due to Zn toxicity). Romaine lettuce was grown in 48 gardens at
varied distances from the smelters and hence varied soil Zn + Cd. Predicted increased Cd in garden crops obtained by: Lettuce Cd
concentration (-0.5 mg/kg for control crop) times 1.67 (to convert fjg Cd/g dry lettuce to increased fjg Cd/day) and then multiplied by
[actual yield/34.25 ( = control yield)] to calculate the predicted "Yield-Corrected Increased Dietary Cd". Based on the data of Baker and
Bowers (1988).
-------
Evaluation of the potential for Pb risk to children who ingest MSW-compost:
The risk from Pb in compost-amended soil or MSW-compost products ingested by
children provides the basis for limiting compost Pb concentration. This limit will
require management and planning in the MSW-compost community. Research on
lead poisoning of children has shown that children live in a dusty environment. We
and our pets and environmental processes like mud on shoes and dust blowing in a
window, bring dust (soil-derived environmental dust) into our homes. We bring soil
into our homes on our shoes, it dries, is crushed, and becomes part of the
housedust pool. When automotive exhaust was high in lead, children got about
four times more lead from the dust they ingested than from the inhalation of Pb in
air directly. The dust was always more important, it just took us decades to
understand the role of Pb in dust (US-DHEW, 1991).
We have all heard about Pb poisoning of children from paint chips.
However, other sources, such as soil ingested by children, can be a significant
source if the soils are highly contaminated. House side soil can contain up to 5%
lead (50,000 jt/g Pb/g) around old painted houses (Chaney and Mielke, 1986;
Chaney, Mielke, and Sterrett, 1989), whereas there are background levels (10-20
fjg Pb/g) in soils around newer houses.
All children ingest some soil by normal hand-to-mouth play (Calabrese et al.,
1989; Calabrese and Stanek, 1991; Davis, 1990; Binder et al., 1986; Clausing et
al., 1987; Van Wijnen et al., 1990; Stanek and Calabrese, 1991). But when we
consider Pb risk to children eating soil, we must consider pica children because
these children consume the most soil (pica is the consumption of non-food items).
Many of these studies of middle class children found children with pica for soil.
Thus, we need to protect pica children from sources of Pb which could provide
excessive bioavailable Pb if ingested. As long as we protect that child, everybody
else is protected.
An example of the clearest demonstration of the risk to children from Pb in
dust was found at a Pb-battery recycling factory in Memphis, TN (Baker et al.,
1977). At this factory, the workers did not change their clothing and shower
before going home (as since has been required by OSHA). They carried highly
contaminated smelter dust into their homes. The children of the smelter workers
were shown to have lead poisoning but their neighbor's children did not. Blood Pb
concentration in the workers' children was related to the concentration of lead in
the house dust of their home. Most of these children had very high blood Pb
levels, due to the industrial dust exposure, and required medical treatment to
remove Pb from their bodies. This result illustrates the principal that if you bring a
high-lead dust or product into the home, it could be a risk to children who have
high ingestion of dust. Other research, summarized in Chaney and Mielke (1986)
and Chaney et al. (1989) has shown that for children exposed to Pb-rich dusts, the
blood Pb concentration rises quickly after the children start crawling, reaches a
peak at about 18 months with the peak of hand-to-mouth play, and declines until
lower blood Pb levels are reached by about 4-6 years of age when mouthing
generally reaches low levels.
Another reason we have such concern about Pb in children is the
demonstration over the last decade that blood Pb levels above 10-15 //g/dL (dL =
deciliter = 100 mL) can significantly reduce IQ and learning ability in children.
20
-------
This phenomenon has been labeled "neuro-behavioral impairment", and appears to
result from Pb interference with nerve growth during brain development in children.
Adults are much less sensitive to blood Pb because they are not undergoing brain
development. Previous limits for acceptable blood Pb were 25 //g/dl_, but the
Center for Disease Control has now lowered the recommended maximum blood Pb
concentration to 10 //g/dL (US-DHEW, 1991). If blood Pb is above this level,
parents and public officials are advised to identify the source(s) of Pb, reduce the
Pb exposure, and to improve nutrition to prevent Pb absorption, etc.
Because of the reduction of Pb in automotive emissions, and reduction of Pb
in food due to change in canning technology (both food and automotive emission
Pb levels have decreased nearly 10 fold in the last 15 years) (Bolger et al., 1991),
median blood Pb levels in suburban children have fallen from about 20-25 //g/dL in
1970 to about 3-4 jwg/dL in 1990. With the normal variance (and varied amounts
from Pb in plumbing systems, etc.), some suburban children exceed the 15 //g/dL.
But over 50% of children in the center city exceeded 15 //g Pb/dL limit (ATSDR,
1988). Children exposed to high levels of soil and dust Pb have been found to
have high blood Pb in numerous cases (reviewed in Chaney and Mielke, 1986;
Chaney, Mielke, and Sterrett, 1989). In other cases, social factors or soil chemical
factors altered the exposure or bioavailability of the soil Pb and little or no increase
in blood Pb was observed even with soils containing 5000 mg Pb/kg (Cotter-
Howells and Thornton, 1991).
In order to better understand the risk from Pb in soil, feeding studies were
conducted with rats. Previous work summarized in Chaney et al. (1989) showed
that less Pb was absorbed from soil than from soluble Pb salts or paint chip
powder. Thus, rat feeding studies were conducted to determine the bioavailability
of lead in garden soils. They found 1) compared to Pb acetate (a soluble Pb salt
considered to be 100% bioavailable in diets) added to purified diets, bone Pb was
increased only 53% as much in a diet containing 5% control low Pb soil as in the
diet without soil (the added diet Pb and soil were equivalent to adding a soil with
1,000 tig Pb/g dry soil); and 2) If the rats were fed urban garden soils with about
1000 fjg Pb/g, bone Pb was only about 20% as high as when equivalent Pb
acetate was added to the control diet, while one soil with 10,200 //g Pb/g caused
bone Pb to be 70% as high as with Pb acetate. So we have to consider lead in
soil as being partially bioavailable. We now interpret these findings as indicating
that soil Pb bioavailability increases with increasing soil Pb concentration because
of weaker Pb adsorption by the soil at higher Pb concentration (Chaney et al.,
1989).
Besides the effect of compost chemical properties on the bioavailability of
Pb in compost, new findings on the effect of "soil dose" (g soil ingested per day)
on absorption of soil Pb are also very important in assessing this risk. Based on
our model of sludge/compost chemistry controlling the activity of free metal ions in
the equilibrium solution, we would expect that as the amount of soil ingested
increases, blood Pb concentration should approach a plateau because soil is
present to adsorb soil-Pb in the intestine. Because adsorption controls Pb
solubility, the solution concentration of Pb should be nearly independent of the g
soil/ml of intestine contents. In short, this is the linear versus plateau response
concept we found with plant uptake of Cd from salts vs. from sludge. This
21
-------
concept had not been tested until November 1990, when results were reported by
Freeman et al. (1991). They found that lead acetate in purified diet caused a huge
smooth increase in bone and blood Pb, but two different soils with up to 3,000 //g
Pb/g caused tissue Pb to increase up to a plateau, far below that from equal Pb
from the Pb acetate (Figure 3). This work says soil Pb has both a low
bioavailability and a non-linear or plateauing dose-response. Thus, the pica child is
protected much more than we previously believed because soil continues to adsorb
Pb in the intestine and reduces absorption of Pb into blood.
The effect of exposure to high soil Pb on blood Pb of individual children is
highly variable, and appears to be related to social, nutritional, behavioral, and soil
chemical factors. Pb in mining soils appears to have lower bioavailability than Pb
in urban dusts (Steele et al., 1990; Freeman et al., 1991; Davis et al., 1992; Ruby
et al., 1992). In particular, the least soluble known compound of Pb in soils is
pyromorphite [(Pb5(PO4)3CI], and this compound has been found in the weathering
products of galena (PbS) in mine waste contaminated soil. Cotter-Howells and
Thornton (1991) report low blood Pb levels in children living in an area with soils
(about 5000 mg Pb/kg) derived from Pb mining wastes. High levels of soil
phosphate may be required to facilitate formation of pyromorphite from other
forms of soil Pb.
Bioavailability of Pb in sludge and compost: Because Pathway 2F is most
limiting for compost-Pb, and because human feeding studies with Pb-rich soil or
compost have not been reported, we must consider the available information on
bioavailability of Pb in sludges and compost. Studies have been conducted to
assess the bioavailability of metals in many different sewage sludges ingested by
livestock. In many of these studies, no increase was found in bone Pb under
conditions relevant to pica children (Decker et al., 1980). Their studies involved
sludge compost that had 215 jjg Pb/g dry weight, at 0, 3.3 and 10% of diet for
180 days. With this sludge compost, there was no significant change in the
indicator tissue lead levels even though the fecal analyses show that the animals
ingested greatly increased amounts of Pb (Table 4). However, in comparable
studies by Kienholz et al. (1979), tissue Pb was significantly increased by ingesting
12% of a sludge containing 780 //g Pb/g (Table 4).
One laboratory conducted cattle feeding experiments with a material similar
to MSW-compost, in the early 1970's. Utley et al. (1972) fed 20% "digested
garbage" and found that Pb accumulated in kidney and liver. Johnson et al.
(1975) fed 17.5% compost (prepared from pre-separated MSW using the Fairfield
digester (containing about 152 mg Pb/kg DW). This experiment found a small
increase in Pb in liver and kidney, but the significance was not evaluated; bone Pb
is a much better indicator of absorbed Pb during chronic feeding studies (this
compost contained low Fe and P, which may have allowed Pb absorption to be
higher than found with typical sewage sludge materials). These studies were
conducted at a time when Pb analyses were less reliable than those of today, and
the mixed diet Pb concentration does not agree with the compost Pb level and
amount of compost in the diets. Unfortunately, bones (the best indicator of
chronic Pb absorption) were not analyzed. Interestingly, there has never been
evidence of Pb accumulation in animal fat even when diets are high in Pb.
However, Johnson et al. (1975) reported a significantly higher level of Pb in fat
22
-------
_Q
CD
C
o
m
80
70
60
40
30
20
10
0
10 L
o oPbOAc in feed
* A 810 ppm Pb Soil
a °3908 ppm Pb Soil
Female rats fed 30 days
Freeman et al., 1991
Mine Waste Soils
AIN-76 Pufified Diet
0 25 50 75 100 125 150 175 200 225
Measured Feed Pb, jug/g
Figure 3. Effect of increasing soil dose (g soil ingested per day) on response of Pb in bone of rats fed two soils and
Pb acetate for 30 days (Freeman et al., 1991). Statistical analysis indicated that bone Pb in the soil fed animals
approached a plateau with increasing soil dose.
-------
(<2.0 in control vs. 3.58 /yg/g FW in the garbage diet). This indicates they
suffered Pb analysis problems common to that era. Thus, these studies are not
reliable evidence that Pb in MSW-composts comprises a risk to children. We
believe that Pb in the colored waste paper fed by Heffron et al. (1977) is
comparable to the "digested garbage" studies of Utley and Johnson. Heffron et al.
fed 23% colored paper (538 ppm Pb) to sheep for a long period, and found
increased Pb in tissues which accumulate Pb (liver, kidney, bone), but not in other
tissues.
Based on all the available research involving ingestion of sludge or compost
(grazing or controlled feeding studies), our best judgement now is that limiting
sludge and compost products to 300 JJQ Pb/g dry weight will allow adsorption of
lead by the compost material to be strong enough so that it does not significantly
contribute to blood lead increase, even for the pica child (see Table SHChaney and
Ryan, 1991). We believe that the Pb adsorption capacity (due to Fe oxides and
organic matter), Ca, and phosphate of sludges can strongly bind Pb and reduce Pb
absorption by animals which ingest sludges or composts. Increasing Fe (and
possibly increasing P) in MSW-composts may further reduce bioavailability of
compost Pb, although Pb levels in most MSW-composts are low enough that
compost Pb comprises insignificant risk.
Because present MSW-compost prepared from MSW separated at a central
facility often contains about 200-500 mg Pb/kg dry weight, there will need to be
an improvement in Pb diversion from the compost stream. Old painted wood, Pb-
batteries, Pb caps from wine bottles, bullet residues, etc., must be diverted to
household hazardous waste collections rather than be put in the MSW. This will
require extensive education efforts. Cessation of using Pb-soldered cans for foods
has reduced one source of Pb compared to 1980, but discarded electronic
equipment has lots of solder which can be leached during hauling and separation.
Some central separation facilities produce MSW-compost with < 300 mg Pb/kg.
Thus, it may not be necessary to require pre-separation of the compostable or
"green wastes" at the home in order to allow production of acceptable quality
MSW-compost for marketing.
Evaluation of potential food-chain risks through mushrooms produced on
media containing MSW-compost: Commercial mushrooms are usually produced on
special "mushroom composts" and these have been considered one possible
market for MSW-compost. Some research has been conducted to determine if
mushrooms grown on media which include MSW-compost or sludge cause high
transfer of metals to edible mushrooms. Some mushroom species accumulate Hg
or Cd to concentrations higher than the media on which they are grown. Uptake
of Hg by vascular plants and transport to edible plant tissues is so small that diets
are not enriched in Hg when soils would provide appreciable bioavailable Hg to
animals which ingest soil. Thus, the unusual Hg food-chain of
compost-*media-»mushrooms-*humans requires consideration.
24
-------
Table 4. Effect of ingesting sewage sludges, composts, or similar materials with
different properties on the concentration of Pb in bones of livestock.
Study Sludge Pb
Source In
Concn. Sludge
sludge in diet
mg/kg %
1.
2.
3.
3.
4.
4.
5.
6.
Ft. Collins
Ft. Collins
Denver
Denver
Washington, DC
Washington, DC
Las Cruces
Chicago
Added after W-170
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
WSSC, MD
WSSC, MD
WSSC, MD
WSSC, MD
WSSC, MD
WSSC, MD
Pensacola
Pensacola
Chicago
466
387
780
780
215
215
150
-
11
12
4
12
3
10
7
.5
.0
.0
.0
.3
.0
.0
-
Dietary Pb Duration Bone/Liver Pb
Cont.
Test
---mg/kg DW---
0.86
1.8
0.6
0.6
6.0
6.0
-
-
56.6
50.0
26.
77.
11.2
19.9
+10.5
-
Fed
days
106
270
94
94
180
180
1440
Control
+Sludge
mg/kg DW
B: 5.
B: 1.
B: 1.
B: 1.
B: 3.
B: 3.
L: -
L: -
0
6
7
7
7.
4.
4.
11.
4.
3.
-
-
2 *
3 *
*
*
7 NS
4 NS
NS
NS
process:
190
185
380
250
257
215
397
397
774
Melbourne 56-241
Ohio
Fairfield
Fairfield
Chicago Dig.
Las Cruces
Chicago
Netherlands
Netherlands
Colored Paper
Glen field
557
163
169
937
150
260
?
165
514
254
3
9
3
3
3
1
2
5
6
(soil)
<1
22
17
3
50
10.
23.
•
.5
.3
.8
.2
.0
.0
.7
.2
.0
•
^
.0
.5
*
.5
*
?
6
0
4.3
4.3
5.6
5.6
6.0
6.0
0.8
0.8
1.4
3.4
4.5
4.8
3.6
4.
•
1.5
2.5
1.1
1.1
9.07
12.0
22.4
19.2
11.2
13.6
7.4
11.0
20.5
40.0
12.
3.8
39.2
35.3
8.
+5.2
130.
8.0
13.0
138.
8.74
150
150
200
200
200
200
168
168
141
365
700
140
91
>1000
730
63
840
90
124
1800
B: 5.
B: 5.
B: 4.
B: 4.
B:12.
B:12.
L: 0.
L: 0.
L: 0.
L: 0.
L: 0.
K: 0.
L: 0.
L:<0.
B: 1.
B:21.
K: 0.
K 0.
K: 0.
B: 2.
K: 0.
7
7
1
1
1
1
32
32
10
93
40
42
62
50
8
00
66
26
6
99
3.
7.
4.
4.
14.
12.
0.
0.
0.
DW 1.
WW 0.
WW 0.
WW 3.
WW 1.
DW 0.
DW 18.
WW 0.
WW 0.
FW 0.
DW 19.
DW 1.
9 NS
4 NS
4 NS
4 NS
8 NS
6 NS
31 NS
49 NS
26 *
12 NS
52 NS
72 *
96 *
60 NS
6 NS
NS
00 NS
42 *
31 NS
0 *
25 *
* Bone or liver Pb concentration significantly increased by sludge ingestion.
1. Johnson et al. (1981). Hereford steers. Selected samples analyzed also by Boyer et at.,
1981).
2. Baxter et al. (1982). Cows and steers.
3. Kienholz et al. (1979). Feedlot steers.
4. Decker et al. (1980). Cows, calves, and steers. Composted sludge,
high in Fe and CaC03.
5. Smith et al. (1985) Sheep. No significant change of Pb in liver.
6. Hansen et al. (1981). Foraging sows. Soil Pb in 504 mt/ha plot =
131 mg/kg, while control plot soil was 37.7 mg Pb/kg. Feces were 7.9 and 41. 7 mg Pb/kg
FW in March. Bone not analyzed, but liver and kidney showed no significant change in
tissue Pb.
7,8 Decker et al. (1979). Cows, calves, and steers grazed on pastures with spray applied sludge
every 4 weeks; 7 is for sludge applied 21-days before grazing; 8 is for sludge applied 1 day
before grazing. In the 1-day treatment, high sludge Fe (11 %) caused induced Cu deficiency
and severe toxicity and weight loss, with higher liver Pb. Dietary sludge and Pb estimated at
25
-------
50% of fecal concentrations.
9-12 Decker et al. (1979, 1980a, 1980b). Cows, calves, and steers grazed on pastures with
spray applied sludge every 4 weeks (Nos. 9, 11) with cattle entering the paddocks 21 days
after sludge application. Alternatively, sludge compost was topdressed on the pastures
intermittently to provide adequate N (Nos. 10, 12). In the second year of the study,
compost was applied only once because of residual N release.
13-14. Bertrand et al. (1980). Bahaigrass pastures spray applied repeatedly during grazing
season, 9 (No. 13) or 16 (No. 14) times. Blood, liver and kidney Pb not increased by sludge
application.
15. Bertrand et al. (1981). Chicago heat dried activated sludge mixed into practical diet for
steers.
16. Evans et al. (1979). Cattle grazed pastures which had received sewage from Melbourne,
Australia, for about 60 years. Soils had accumulated high levels of metals in surface 2-5
cm. Cattle continuously grazed on the pastures.
17 Reddy et al. (1985). Dewatered sludge surface applied in pastures, and cattle grazed about
30 days later. Much less sludge ingestion than from spray-applied sludge in other studies.
Blood Pb was significantly higher on sludged farms, 0.43 vs. 1.21 //g/dL in cows, but not
calves; kidney Pb but not liver or blood Pb was sig. higher in calves; bone, kidney and liver
Pb unchanged in cows.
18. Utley et al. (1972). Fairfield "garbage digest" for 5 days, then dried and palletized. Fed to
beef steers and cows. Poor analysis. Report significant increase in Pb in kidney and liver;
no bone analysis. Milk analyzed, but no Pb detected.
19. Johnson et al. (1975). Fairfield "garbage digest" fed to beef cattle for 91 days. Poor
agreement between direct analysis of garbage (140 ppm Pb) and garbage as part of feed
(198 ppm Pb). Fat was reported to be sign, increased in Pb (<2.0 vs. 3.58) in contrast with
any other study of Pb at chronic doses. Kidney reported to also be sig. increased.
20. Fitzgerald et al., 1985. Cows grazed up to 8 yr on pastures with spray applied or
incorporated Chicago fluid digested sludge. Liver, kidney, and bone not increased in Pb.
21. Sanson et al., 1984. Breeding ewes fed complete ration ± 3.5% irradiated sludge for 2 yr.
No changes found in tissue Pb levels. No adverse effects of sludge in diet.
22. Osuna et al., 1981. Fed 50% Chicago dried activated sludge to weanling swine for 63 days,
compared to control and 79 ppm Cd as salt.
23. Vreman et al., 1986. Cows fed salts vs. sludge in indoor management for 2-3 yrs.
Concentration of metals in sludge not reported, but level in concentrate feedstuff was 50 vs.
168 mg Pb/day and whole diet was 2.5 vs. 8.0 ppm Pb. Few replications. Sludge caused
smaller increase in kidney Pb than did PbOAc [1.19 //g Pb/gFW (salt), 0.66 //g Pb/g (sludge.)]
24. Veen and Vreman, 1986. Lambs fed about 1100 g concentrate and 225 g hay DW for 90
days in an enclosed environment. Sludge included in concentrate at 10% for 42 days, and
then reduced to 5% for the duration. Gain not reduced by sludge addition.
25. Heffron et al., 1977. Colored paper from newspapers and magazines fed at 23% of
practical diet to sheep for 124 days. This could be considered "uncomposted" MSW, similar
to the poorly composted Fairfield compost used by Utley et al. (1972) and Johnson et al.
(1975). All Pb accumulating tissues increased (control diet/colored paper diet): Blood,
0.2/0.7 //g/gDW; kidney, 0.85/7.6 j/g/gDW; liver, 0.45/5.0 //g/gDW.
26. Ross and Short, 1990. Managed to produce fat lambs for 3 yr with 37.5 Mg/ha sludge DW
applied each year. Details of waiting period not reported. No adverse effects on ewes or
lambs. Lamb kidney also significantly lower on sludge amended paddocks.
Data not used due to internal disagreement:
27. Beaudouin et al. (1980). Tissue Pb results varied in no pattern, with control 2 mg Pb/kg and
sludge fed swine have <0.01 mg Pb/kg in some tissues, and reversed in others.
28. Cibulka et al. (1983). Tissue Pb levels increased without clear relationship with increasing
sludge level in diet, 0-4.5%. Muscle increased as much as kidney, while other research did
not observe increases in muscle with such low diet Pb levels.
26
-------
The potential of mushrooms to bioaccumulate Hg has been demonstrated
both in compost- or sludge-amended media and in natural environments (Brunnert
and Zadrazil, 1983; Enke, et al., 1979; Frank, Rainforth, and Sangster, 1974;
Zabowski et al., 1990). Some mushroom species, especially cellulolytic species,
accumulate very high levels of Hg compared to other vegetable foods, even when
grown on media which are not contaminated. However, research has shown that
only a small fraction of the total Hg in mushrooms is in the form of methyl-Hg
(D'Arrigo et al., 1984; Bargagli and Baldi, 1984; Minagawa et al., 1980; Quinch,
Bolay and Dvorak, 1976; Stegnar et al., 1973; Stivje and Roschnik, 1974; Stivje
and Besson, 1976). Methyl-Hg is much more toxic than inorganic Hg2*.
Methyl-Hg is lipophilic, and is efficiently absorbed and can cross the blood-brain
barrier to cause neurologic effects in animals. Because inorganic Hg is far less
toxic than methyl-Hg, WHO and US-FDA recommendations and/or regulations
about Hg are based on methyl-Hg. Thus, the finding that mushroom species which
bioaccumulate Hg contain <3% of their total-Hg as methyl-Hg is very important.
Some other mushroom species which do not bioaccumulate Hg to a high extent
can have a greater fraction of their total Hg as methyl-Hg (up to 36% for one
sample), but the methyl-Hg concentration in these species is lower than that in the
species which accumulate high total Hg levels.
Domsch et al. (1976) evaluated the absorption of Hg by commercial
mushrooms (Agaricus bisporus) which were grown on a mushroom compost which
included MSW + sludge compost at 0, 25, 50, and 75% of the medium (the
compost contained 2.4 mg total Hg/kg DW). The mushrooms grown on media
containing 50 or 75% compost contained slightly over 0.5 mg Hg/kg FW, the US
Food and Drug Administration numerical limit for Hg in fish. The concern about Hg
in MSW-compost and sludge-compost used in production of mushrooms has not
adequately taken into account the finding that methyl-Hg was only a small fraction
of total Hg in mushrooms. Further, the response of increased Hg in mushrooms
vs. fraction of MSW + sludge in the mixed mushroom compost shown in the
Domsch et al. study is clearly plateauing (same mushroom-Hg concentration for 50
and 75% MSW + sludge in the mushroom compost). Thus, modern low Hg MSW
compost materials appear to comprise little risk to persons who ingest unusual high
quantities of mushrooms. In forest ecosystems, the combination of low methyl-Hg
in the mushrooms, coupled with low annual ingestion, indicates that Hg should not
be a practical limit on forest utilization of compost (Zabowski et al., 1990).
Unfortunately, the bioavailability of Hg in mushrooms has not been reported, and
the effect of modern MSW-composts on mushroom Hg levels or bioavailability
have not been evaluated, so limits for Hg in MSW-compost products can not be
accurately estimated. Compost programs clearly need to promote separate
collection of Hg rich wastes (e.g. batteries).
Cd in mushrooms is only important in those species which bioaccumulate
high levels of Cd, which does not include the commercial mushroom, Agaricus
bisporus. Some Cd-accumulating mushroom species contain over 50 ppm Cd DW
on uncontaminated substrates. Rat feeding studies have been conducted by Diehl
and Schlemmer (1984) to test the retention in animals of mushroom Cd; about 1%
of diet Cd reached the kidney and liver by the end of 6 weeks of feeding 15%
mushroom diet with 3.9 ppm total Cd. Human feeding studies have also been
27
-------
conducted, and over 90% of ingested Cd was excreted within a few days
(Schellmann et al.f 1980, 1984); if normal retention of Cd in the intestine for a
prolonged period is considered, the human studies support very low bioavailability
of mushroom Cd. Several hypotheses have been suggested to explain the low
bioavailability of mushroom Cd: 1) the presence of chitin in mushrooms may
adsorb Cd in the intestine and reduce absorption; 2) Cd in mushrooms may be in
the form of Cd-phytochelatins or metallothioneins which have lower bioavailability;
and/or 3) presence of other nutrients in the mushrooms may inhibit retention of Cd.
In any case, there is no evidence that mushroom Cd would be a significant source
of transfer of soil-compost mixture Cd to humans (forest worst case scenario)
compared to the worst case acidic garden scenario.
Evaluation of potential risks to wildlife. Although the evidence summarized in
this review paper has shown that modern sludges and MSW-composts can be
safely utilized in agriculture and protect humans, plants, and livestock, there has
been less research on the potential effects of sludge or compost-applied heavy
metals and toxic organics on soil organisms and wildlife than on agricultural
ecosystems. It is now clear that soil fauna are particularly important in risk
analysis because earthworms have been found to bio-concentrate Cd and PCBs
from soils (e.g., Ireland, 1983). Although most wildlife animals consume seeds or
forage materials, a few mammals or birds ingest substantial amounts of
earthworms and other soil fauna which might serve to accumulate and transfer the
toxic constituents from soils when other food webs do not. Although some crops
absorb Cd to high concentrations, there is no evidence that herbivorous wildlife are
at higher risk from eating crops growing on Cd-rich sludge-amended soils than are
omnivorous wildlife eating earthworms living in the soils. Beyer (1986) noted that
there is little evidence for biomagnification of heavy metals (other than methyl-Hg)
in food webs except for the earthworm pathway. Little dietary Cd is retained over
a lifetime, so the body contains little Cd when consumed by the next higher trophic
level. This situation makes the earthworm pathway much more significant that
plant based foods.
Studies of Cd in ecosystems has consistently shown that shrews are "close
to soil" regarding Cd, PCBs, and Pb risk. Ecologists studying metal transfer and
risk in smelter-contaminated soils, or in mine soils, repeatedly showed that animals
which consumed earthworms comprised the most exposed receptors for these
contaminants. Comparison of other mammal species to shrews or other
earthworm consuming mammals has shown that Cd, Pb, or PCB transfer from soil
is perhaps 10-fold higher for the shrew than for mice, voles, or other non-
earthworm consumers (Cooke et al., 1990; Hegstrom and West, 1989; Hunter et
al., 1983; Ma, 1989; Scanlon, 1987). Studies of other wildlife collected on sludge
amended soils, or fed sludge-grown crops (e.g. rabbits, deer, deer-mice, voles,
pheasants etc.) (Alberici, et al. 1989; Anderson et al., 1982; Beardsley et al.,
1978; Dressier et al., 1986; Hinesly et al., 1982) failed to find appreciable
contaminant transfer to wildlife. Often the increase in plant biomass production on
disturbed sites caused significant increases in population density. The increased
exposure to soil Pb by earthworm consumers results from the high fraction of soil
in this food source, which causes higher soil ingestion than other behaviors.
28
-------
Shrews, moles, badgers, and red fox consume an appreciable amount of
earthworms (Macdonald, 1983; Ma, 1987), and might thus be at higher risk than
other mammals. Although birds may be exposed to soil PCBs and Cd due to
earthworm ingestion, few species are known to inhabit a small territory for their
lifetime which might provide them unusual exposure to high amounts of
earthworm-transferred sludge or compost contaminants from an amended site
similar to shrews. In one bird study (with redwinged blackbirds, a species not
known to consume earthworms), little or no Cd accumulated in liver or kidney in
birds nesting on mine spoil amended with a very high rate of a high Cd sludge
compared to birds nesting on non-sludged areas (Gaffney and Ellerston, 1979).
Considering the lifetime exposure of wildlife species to sludge contaminants, it is
clear that the earthworm-consuming small mammals with limited territory must
comprise the most exposed individuals rather than birds which have much more
limited exposure over time.
Earthworm accumulation of Cd from soils: Because earthworms can
bioaccumulate at least Cd and PCBs, and some animals ingest earthworms with
the worm digestive system full of soil, ingestion of earthworms comprises a
significant exposure route to metals in soils amended with sludges or composts.
Research has characterized the ability of earthworms to accumulate different
metals. Many researchers attempted to purge the worms of soil by allowing them
to live in a moist environment on filter paper. However, Helmke et al. (1979)
found that traces of soil remaining in the digestive system can explain nearly all of
the residues of many metals. They used neutron activation to measure many
elements, and then used non-absorbed elements to estimate soil contamination of
the worm samples. Soil explained most of the residue of most elements, but Cd,
Au, and a few other elements were bioaccumulated. Because soil normally
comprises 45% of the dry weight of an intact non-purged worm (Beyer and
Stafford, 1992), the soil can provide much more exposure to many elements than
can the worm tissues (except for Cd). Also, soil in the gut of an animal which
consumes an earthworm should provide ability to adsorb the metal in the intestine
and reduce bioavailability. Beyer et al. (1982; 1987) and Beyer and Cromartie
(1987) have shown many characteristics of earthworms on metal salt or sludge
amended soils. Ma (1982) examined earthworm accumulation of elements from
the long term MSW-compost plots described by Haan (1981). He found only Cd
and Zn were increased in purged worms from these plots. The pattern found for
worms from MSW-compost amended soils was quite similar to that for the sludge
amended soils described by Helmke et al. (1979) and Beyer et al. (1982).
Estimating allowed soil or compost Cd which protects wildlife mammals
which consume earthworms: Two separate approaches for estimating the
maximum allowed soil-sludge mixture Cd concentration protective of the most
sensitive wildlife (predator) species from lifetime excess Cd including
bioaccumulation of Cd by soil biota, were identified by Chaney and Ryan (1991):
1) The first approach follows that of the original US-EPA (1989a) Technical
Support Document for the Clean Water Act-503 Regulation: A tolerable Cd level in
wildlife is divided by the slope for the (soil biota-Cd):(soil-Cd) ratio [the increment
29
-------
in diet Cd due to sludge utilization]; fraction of earthworms in the total diet must
be taken into consideration, as well as bioavailability of Cd in the biota (or biota
with ingested soil) to the predator; or 2) The second approach avoids the
uncertainties of the Cd-bioaccumulation ratio for earthworms, the fraction of
chronic wildlife diet comprised by earthworms, and the bioavailability of Cd in
earthworms to wildlife, by computing a direct ratio between soil-Cd and tolerable
Cd in the kidney cortex of earthworm-predator wildlife.
Approach 1: This analysis follows the suggestion of Beyer and Stafford
(1992), with adjustments for factors used in calculations for other elements in the
revision of the CWA-503 risk analysis. They noted that a number of studies have
found earthworms with high Cd levels due to use of sewage sludge. On soils
amended with high Cd sludges, earthworms may contain as high as 100 JJQ Cd/g
earthworm DW for soil-purged worms. Their work has shown that the worm:soil
bioaccumulation ratio for Cd is about 10 for soil-purged worms, or about 5-6 for
non-purged worms (Beyer and Stafford, 1992). For 10-fold enrichment in purged
worms:soil (dry matter basis), (non-purged worms contained 45% soil (DW basis)),
the worm tissue provides about 92.4% of the Cd and soil only 7.6%; for 10-fold
increase in purged worms, the increase in non-purged worms is only 6.0-fold.
Bioavailability must be taken into account, and the soil or soil-sludge mixture in the
earthworm gut should lower Cd bioavailability to animals which ingest earthworms
(in nearly all cases, worms are ingested intact with internal soil). Readers should
recall the errors in assessing risk of dietary metal salts compared to sludge-borne
metals. In study of the relative toxicity of sludge-Cd and Cd-salt to pigs, Osuna et
al. (1981) fed diets containing 50% high Cd sludge (147 mg Cd/kg DW) or Cd-salt
(79 mg Cd/kg DW). Cd-salts caused severe anemia and toxicity, while the pigs fed
50% sludge had no anemia (the low energy of the sludge containing diet reduced
gain rates, but caused no other adverse effects). Kidney Cd was increased 21.4%
as much by sludge-Cd as salt-Cd per unit diet Cd. Further, in chronic feeding
studies of the effect of feeding earthworms to Japanese quail, Stoewsand et al.
(1986) and Pimentel et al. (1984) found no adverse effects of feeding 60% control
or 50% Cd-enriched earthworms (dry matter basis); the accumulation of Cd in
kidneys showed that worm Cd had low bioavailability.
Based on a number of studies, taking into consideration the short biological
half life of Cd in rodents and birds (e.g. Freeman et al., 1983), 100 //g Cd/g diet
DW can be tolerated by sensitive individuals [the 0.5 mg Cd/kg diet recommended
by the US National Research Council (1980) was set to protect use of liver and
kidney as human food, not the health of the livestock]. Using this as the lowest
chronic toxic concentration, and 6 as the bioaccumulation factor for whole
earthworms (and assuming 50% bioavailability [higher than the 21.4% from Osuna
et al. 1981] or even lower bioavailability based on Decker et al., [1980]), and
assuming the diet contains at maximum 1/3 earthworms (non-purged) over a
lifetime chronic exposure period, the allowed soil Cd would be:
30
-------
100 mg bioavailable-Cd 1 kg diet DW _300 mq bioavailable earthworm-Cd
kg diet DW *0.33 kg earthworm-DW kg earthworm-DW
1 mq total earthworm-Cd 1.0 mq soil-Cd
0.5 mg bioavailable worm-Cd * 6 mg earthworm total-Cd
= 100 mg Cd/kg soil DW = 200 kg Cd/ha.
Approach 2: This is the direct approach in which the kidney Cd relationship
with sludge-applied Cd is estimated for sludge-amended soils. There are a few
valid sludge field data to allow calculation using the second approach. A study by
Hegstrom and West (38) looked at tissue metals in several species of small
mammals from forest sites which received sludge applications. They collected
insectivorous Towbridge's shrews (Sorex towbridgii) and shrew-moles
(Neurotrichusgibbsi), and granivorous deer mice (Peromyscus maniculatus) from
sludge-treated and control sites at Pack Forest, where Seattle sludge was surface
applied at 51 Mg/ha several years before the wildlife collections. Heavy metals
were higher in tissues of Towbridge's shrews from the sludge-treated areas than in
control, and much more accumulated in the shrews than in the other species.
A second collection of shrews was made from forested sites with much
higher cumulative applications in order to identify any kidney or liver lesions which
may result from sludge use. Despite the high levels of heavy metals found in
kidney of Towbridge's shrews (mean = 126 mg Cd/kg DW), no lesions were found
in their organs. Of course, this concentration is far below the level expected to
cause the first health effect in mammals (696 mg Cd /kg whole kidney DW, shown
below).
To estimate transfer from soil to kidney as a basis for limiting sludge Cd
applications, the following information was used: 51 Mg dry sludge was applied to
forest sites where shrews were sampled. The sludge applied in the studies
contained 50 ppm Cd, 2000 ppm Zn, 900 ppm Cu, and 1200 ppm Pb. The
application of 51 Mg DW sludge/ha containing 50 mg Cd/kg DW applied 2.55 kg
Cd/ha. The shrew whole kidney Cd concentrations were 33 (25-43, N = 66) on
sludged plots, and 9 (8-10, N = 50) on equivalent control forested sites. The
increment in kidney Cd due to sludge utilization was 24 mg Cd/kg whole kidney-
DW.
The concentration of Cd in the whole kidney has to be related to the potential
toxic level (200 //g Cd/g kidney cortex-FW; this value is considered a measure of
the lowest Cd concentration which can cause tubular dysfunction in sensitive
individuals for many animal species):
200 ua Cd f 1.00 UQ Cd f q kidnev cortex 1 _ 160 LIQ Cd
g kidney cortex-FW " [ g whole kidney * 1.25/Aj Cd J g whole kidney-FW
(based on using the conversion factor 1.25 for (kidney cortex-Cd concentration) :
(whole kidney-Cd concentration) for humans from Svartengren et al. [1986]).
To complete this calculation, one also needs to convert from kidney-FW to
kidney-DW. In the absence of data specifically for shrews, the mean dry matter
content of fresh beef, calf, hog, and lamb kidney from USDA Handbook 8 (Adams,
1975) was used, = 23% solids. Thus, (160 mg Cd/kg FW)-(1.00 g FW/0.23 g
31
-------
DW) = 696 mg Cd/kg whole kidney DW.
Then the slope for (shrew kidney-Cd):(soil-Cd) is divided into the tolerable
whole kidney Cd concentration on a dry weight basis: 696 fjg Cd/g DW (maximum
permissible Cd concentration in whole kidney) •+• [(24 mg Cd increase/kg whole
kidney DW)/(2.55 kg Cd/ha)] = 74 kg Cd/ha when sensitive shrews would be
expected to reach their first health effect on kidney function due to dietary Cd
exposure. The disagreement between the estimates for Approach 1 and Approach
2 might be due to the surface application of sludge in the forest compared to the
slopes obtained for earthworms which habitated soils with sludge mixed about 20
cm deep into the soil. Different earthworm species feed at different depths in the
soil; the earthworm species and feeding habit in the forest was not reported
(Hegstrom and West, 1989).
The relationship of kidney Cd, or bone or kidney Pb, to survival of shrew
populations has been studied somewhat. In the study by Ma (1989) of the Pb
transfer to mammals at a shooting range, shrews with excessive organ Pb had high
population density. In the study by Hunter et al. (1983), shrews had evidence of
excessive organ Cd, but the population was well established. In studies by Beyer
et al. (1985), many animals were collected in areas where vegetation persisted in
the vicinity of a Zn smelter; no evidence of excessive kidney Cd or Cd health
effects were seen, but high Pb caused depressed ALAD activity in some animals
with high tissue Pb. Deer in the area suffered Zn-induced Cu deficiency with loss
of cartilage in joints of long bones, but kidney Cd levels were not sufficient to
indicate tubular dysfunction due to excessive Cd (Sileo and Beyer, 1985). Based
on these evaluations, it seems clear that the acidic garden scenario for Cd, and
children ingesting compost scenario for Pb are much more restrictive on sludge and
compost metal concentrations than are wildlife scenarios.
Evaluation of potential risks to soil microbes: Starting in the 1980s, studies
by McGrath, Brookes, Ciller, and their associates identified apparent adverse
effects of sludge-applied heavy metals on the soil microbial biomass and on the
Rhizobium strain which forms nodules in white clover and related species (Brookes
and McGrath, 1984; Brookes et al., 1986; Ciller et al., 1989; McGrath, Brookes
and Ciller 1988; McGrath, Hirsch and Ciller, 1988). In a long-term experiment (the
Woburn Market Garden Experiment), about 766 Mg/ha of moderately high metal
concentration sewage sludge (average metals were about 3000 mg Zn/kg, 1300
mg Cu, 200 mg Ni, 100 mg Cd, 900 mg Pb, and 1000 mg Cr/kg DW; McGrath,
1984) was applied to field plots of vegetable crops on a sandy soil from 1942 to
1961, and the soil microbe populations were examined more than 20 years after
the last sludge application. No legume had been grown since 1942. Their
research found that the historic sludge applications had caused selection in these
soils of a strain of Rhizobium leguminosarum biovar trifolii which formed nodules
on white clover, but the nodules were ineffective in fixing N. Although the sludge
utilization practice caused selection of this ineffective Rhizobium strain, no
phytotoxicity occurred to white clover if N-fertilizer was added to the pots.
Further, inoculation of the plots with an effective strain allowed normal nodulation
of white clover, although the population of effective strains in the soil declined
after inoculation. Further, Rhizobia for other legume species (other biovars) have
32
-------
not been found to be inhibited by soil metals levels below those which cause
significant phytotoxicity (soybean: Heckman et al., 1986; 1987a; 1987b; Kinkle et
al., 1987; alfalfa: Angle and Chaney, 1991; Angle et al., 1988; El-Aziz et al.,
1991).
Besides the inhibition of N fixation by this strain of Rhizobium, N fixation by
blue-green algae was also inhibited on these plots and some other high metals soils
(Brookes, McGrath and Heijnen, 1986) and N fixation by free living bacteria was
also inhibited on high metal mine soils (Rother, Millbank and Thornton, 1982a).
Many other studies have shown that soil microbial activities were not
inhibited on sludge-amended soils, including ammonification of organic-N,
nitrification of NH4-N, mineralization of C and N, etc. (e.g., Minnich and McBride,
1986; Rother, Millbank and Thornton, 1982b). These studies on white clover
Rhizobium vs. other soil microbes appear to be replicated well, but to disagree
regarding the toxicity of soil metals to soil microbes compared to the toxicity of
soil metals to higher plants. Angle and co-workers have conducted some work to
evaluate metal tolerance of US strains of white clover Rhizobium and found these
strains were less sensitive to metals than the UK strains described by McGrath et
al. (Angle et al., unpublished). We have found effective strains in nodules of white
and red clover growing in farmers fields in the vicinity of the Palmerton, PA, Zn
smelter, in soils with higher Zn and Cd levels than in the Woburn study.
In attempting to explain the adverse effects of sludge application on the
Worburn plots, some workers have hypothesized that the finding that the
Rhizobium strain was more sensitive to soil metals than was the host plant, may
have resulted from the very light texture of the soil studied, the somewhat high
level of metals in the sludge applied, or from the long period of exposure without
reinoculation of the soil. It is clear that simple inoculation of seeds when sowing
white clover can allow normal nodulation. The causal agent for selection of
ineffective strains has not yet been identified. Few long-term sludge plots with
very high cumulative sludge applications have been examined for this phenomenon,
while some high metal mine spoils have been found to cause rapid decline in white
clover Rhizobium populations (S.P. McGrath and K.C. Jones, personal
communication; Rother et al., 1983). It is clear that further research is needed to
establish whether the first adverse effect of very high cumulative applications of
NOAEL quality sludges will be phytotoxicity to highly sensitive vegetable crop
species when the soil is acidified, or decline in population of white clover
Rhizobium strains. Clovers are not as sensitive to excessive Zn and other metals
as lettuce, beet, chard, and some other well known highly sensitive species (e.g.,
Hewitt, 1954).
EVALUATION OF RISKS FROM POTENTIALLY TOXIC ORGANIC COMPOUNDS IN
MSW-COMPOST
Reviews by Harms and Sauerbeck (1983), Chaney (1985), Jacobs et al.
(1987), O'Connor et al. (1991), Chaney et al. (1991), and Chaney and Ryan
(1991) cover the concepts and research data on the potential for risk of sludge
PCBs and other organics to humans, livestock, crops, or wildlife. When the W-170
Peer Review Committee used the Pathway Approach to make quantitative
33
-------
estimates of cumulative applications of PCBs, PAHs, and other compounds, none
were found to occur in sewage sludge at high enough levels to be a risk to the
most exposed individuals (Page et al.( 1989; Chaney et al., 1991; Chancy and
Ryan, 1991). Table 5 shows the limits required for PCBs to avoid risk under each
Pathway for which quantitative estimates were completed. Thus, Pathway 2
{ingestion by children), Pathway 4-Surface (surface applied compost ingested by
grazing livestock), and Pathway 9 (accumulation by earthworms which are
ingested by wildlife as one-third of the dry matter of their diet) are most limiting to
application to persistent potentially toxic organic compounds.
Interestingly, the amount of MSW-compost ingested by grazing livestock
would be expected to be significantly lower than found in the case of surface
applied fluid sludges. Although cattle grazing pastures to which fluid sludge was
applied 21-days before initiation of grazing consume about 2.5% sludge in their
diet (dry matter basis), when dewatered sludge or sludge composts were applied,
sludge comprised only about 1 % of diet dry matter or less (Reddy et al., 1985;
Decker et al., 1980a, 1980b).
Besides PCBs, phthalates, and many other chemicals, the family of
compounds called polycyclic aromatic hydrocarbons (PAHs) is known to occur in
sludges and MSW-composts. Many of the PAHs are carcinogenic, and research
has been conducted to clarify the potential risk from representative carcinogenic
PAH compounds (e.g., benzo(a)pyrene). PAHs are generated by combustion
processes, and are strongly adsorbed by humic acids. PAHs are biodegradable,
although the rate of biodegradation of sludge-borne PAHs is now known to be
somewhat slower than previously estimated using addition of pure compounds to
soils (Wild et al., 1990, 1991). Plant uptake of PAHs is significant only in the case
of carrots, and nearly all the PAH in carrot roots is in the peel (Wild et al., 1992)
Feeding studies with PCBs indicated that sludge organic matter could adsorb
PCBs strongly enough to reduce absorption by cattle by about 50% compared to
pure PCBs in corn oil (see Chaney et al., 1991). Using the q,* for benzo(a)pyrene
= 11.5 (mg/kg/day)'1, the allowed compost concentration would be 6.1 mg/kg DW
in order to protect 1-6 year old children who consume 0.2 g dry compost per day
for 5 years (or 0.5 g/day for 2 years) and assuming 100% bioavailability. If
bioavailability is lower due to adsorption by compost organic matter, the allowed
sludge concentration would be proportionally higher.
As noted by Chaney et al. (1991), the garden foods pathway (Pathway 1F)
for PCBs comprises much lower risk than does Pathway 2. In the garden food
pathway, the potential for transfer of potentially toxic organics in compost to
humans is very dependent upon the ability of carrots to accumulate the
compounds from amended soils since carrot is nearly the only crop with
appreciable accumulation of PAH or PCB from sludge-amended soils. Much like the
case for metals being bound by the specific metal adsorption capacity of sludges,
organics are bound strongly by sludge and compost organic matter. This reduces
"uptake" (transfer to edible plant parts) by plants growing on compost amended
soils compared to soils which receive applications of pure compounds without the
adsorption capacity of compost, and makes the response pattern a plateau in crop
PAH or PCB with increasing sludge application rate for a sludge. This was shown
for PCBs by O'Connor et al. (1990). For PAHs, Ellwardt (1977) and Wild and
34
-------
Jones (1992) have made similar observations for plateau response to compost or
sludge-borne PAH, again with focus on carrot which accumulates lipophilic organic
compounds in the peel layer.
Besides these considerations, there is the possibility that PAH compounds will
be biodegraded during composting of MSW. Little degradation was observed by
Muller and Korte (1976) in a model system. This question has not been
unequivocally settled for sludge or MSW at this time. The decreased microbial
diversity in composting organic materials compared to mesophilic populations may
prevent enhanced destruction of some organics during composting, and
management at lower temperatures may favor biodegradation of some persistent
organics.
Table 5. Comparison of PCB application limits for each pathway from the 503
Proposed Rule (US-EPA, 1989b) with the corrected versions based on US-EPA
(1989a) and Chaney, Ryan, and O'Connor (1991). Units are changed in some
corrected versions.
Proposed 503 Rule Corrected Approach
Pathway Limit Units Limit value Limit Units Limit value
1
IF
kg/ha-yr
kg/ha-yr
4.14
0.264
mg/kg soil max.
17.2
kg/ha-yr 2.31
2F kg/ha-yr 7.26 mg/kg soil max. 9.09
2D&M kg/ha-yr 7.26 mg/kg sludge DW 9.09
3 kg/ha-yr 0.0056 mg/kg soil max. 18.3
kg/ha-yr 2.46
4-Surface Application
4-Mixed With Soil
9
kg/ha-yr
kg/ha-yr
kg/ha-yr
0.0192 mg/kg sludge DW
0.0192 mg/kg soil max.
kg/ha soil max.
kg/ha- yr
mg/kg soil max.
kg/ha soil max.
kg/ha-yr
2.23
2.23
4.46
0.299
4.06
8.12
0.545
Annual applications based on 10 year half-life for PCBs in soil. Fraction of dietary
meat and milk products from sludge/compost amended soil presumed to be 45%
(Chaney, Ryan, and O'Connor, 1991); reassessment of this fraction indicates that
only 15% may now come from "homegrown" livestock based on more recent
dietary surveys. This would increase the allowed PCBs in sludge, or kg/ha-yr by
about 3-fold for Pathways 3, 4, 5, and 6.
35
-------
RESEARCH NEEDS FOR MSW-COMPOST:
In order for MSW-composting and distribution and marketing of MSW-
compost to win the degree of public acceptance and marketability desired by the
industry, research and demonstrations will be required. Research on fate and
effects of nutrients, metals, and organics in sewage sludge were critical in winning
public acceptance, and provided the data needed to prepare appropriate
regulations. Demonstration projects were required in many locations to convince
citizens that local agencies could utilize sludges within the regulations. We
conclude that the most important research needs or questions remaining for MSW-
composting and MSW-compost marketing include:
1) Will higher Fe concentration in MSW-compost persistently increase the
specific metal adsorption capacity of compost and thereby reduce the
potential for risk from compost metals, particularly focusing on:
A) Bioavailability of compost Pb to monogastric animals which ingest
compost;
B) Phyto-availability of compost Cd at pH ^ 5.5 as indicated by reduced
height of the plateau above the control, or slope for plant:soil
relationship;
C) Phytotoxicity of sludge-applied Zn, Cu, and Ni at pH 2: 5.5.
D) Effects on white clover Rhizobium.
2) Does addition of MSW-compost to Pb-rich urban soils reduce the
bioavailability of soil Pb to monogastric animals?
3) Can compost be efficiently converted to organic-N fertilizer instead of a
low N soil conditioner? Can methods be established to determine the first
year mineralization of organic-N in MSW-compost (see Gilmour and Clark,
1988)?
4} Can homogeneity of MSW-compost be improved by planned mixing
during processing?
5) Does aeration during storage prevent formation of phytotoxic
biodegradation by-products in MSW-compost as required for compost to
be utilized as a large fraction of potting media?
6) Can any MSW-compost cause metal phytotoxicity at pH 5.5 or above to
sensitive vegetable crops. We have no evidence that phytotoxicity could
result from MSW-compost use in short or long term. However, high
cumulative applications, studied after considerable time to allow
decomposition of organic matter in the soil-compost mixture, and
adjusted to very low pH to comprise the "worst case", should be
examined.
36
-------
7) How important is the potential for lime-induced Mn deficiency from land
application of MSW-compost compared to lime-treated sludges? Can
susceptible crops and soils be identified so that agronomic advice to avoid
Mn deficiency can be provided to MSW-compost users.
8) Do particular sources of compostable organics carry undesirable levels of
boron, Pb, Cd, or Zn, and what can be done to keep materials rich in
potentially toxic constituents out of the compost stream.
9) Do present levels of Hg in MSW-compost or MSW +sludge compost still
prevent their use in mushroom production? Do mushrooms produced on
media including MSW-compost have increased Hg or methyl-Hg? New
studies are needed to clarify Hg limitations for this use of MSW-compost.
37
-------
REFERENCES CITED
Adams, C.F. 1975. Nutritive value of American foods in common units. Agricultural Handbook
No. 456, US Gov. Printing Office 0100-03114.
Adams, M.A. 1991. FDA total diet study: Dietary intakes of lead and other chemicals. Chem.
Spec. Bioavail. 3:37-42.
Adams, T.M. and J.R. Sanders. 1984. The effects of pH on release to solution of zinc, copper
and nickel from metal-loaded sewage sludges. Environ. Pollut. 88:85-99.
Agency for Toxic Substances and Disease Registry (ATSDR). 1988. The Nature and Extent of
Lead Poisoning in Children in the United States: A Report to Congress. DHHS Doc. No. 99-2966.
US Dept. Health Human Services, Public Health Service. Atlanta, GA.
Alberici, T.M., W.E. Sopper, G.L. Storm, and R.H. Yahner. 1989. Trace metals in soil, vegetation,
and voles from mine land treated with sewage sludge. J. Environ. Qua). 18:115-120.
Anderson, T.J., G.W. Barrett, C.S. Clark, V.J. Elia, and V.A. Majeti. 1982. Metal concentrations
in tissues of meadow voles from sewage sludge-treated fields. J. Environ. Qual. 11:272-277.
Andersson, A. 1983. Composted municipal refuse as fertilizer and soil conditioner. Effects on the
contents of heavy metals in soil and plant, as compared to sewage sludge, manure and commercial
fertilizers, pp. 146-156. jn S. Berglund, R.D. Davis and P. L'Hermite (eds.) Utilization of Sewage
Sludge on Land: Rates of Application and Long-Term Effects of Metals. D. Reidel Publ., Dordrecht.
Angle, J.S. and R.L. Chaney. 1991. Heavy metal effects on soil populations and heavy metal
tolerance of Rhizobium meliloti, nodulation, and growth of alfalfa. Water, Air, Soil Pollut. 57-
58:597-604.
Angle, J.S., M.A. Spiro, A.M. Heggo, M. EI-Kherbawy, and R.L. Chaney. 1988. Soil microbial-
legume interactions in heavy metal contaminated soils of Palmerton, PA. Trace Subst. Environ.
Health 22:321-336.
Baker, D.E. and M.E.Bowers. 1988. Health effects of cadmium predicted from growth and
composition of lettuce grown in gardens contaminated by emissions from zinc smelters. Trace
Subst. Environ. Health 22:281-295.
Baker, E.L., Jr., D.S. Folland, T.A. Taylor, M. Frank, W. Peterson, G. Lovejoy, D. Cox, J.
Housworth, and P.J. Landrigan. 1977. Lead poisoning in children of lead workers. Home
contamination with industrial dust. New Engl. J. Med. 296:260-261.
Bargagli, R. and F. Baldi. 1984. Mercury and methyl mercury in higher fungi and their relation
with the substrata in a cinnabar mining area. Chemosphere 13:1059-1071.
Baxter, J.C., B. Barry, D.E. Johnson, and E.W. Kienholz. I982. Heavy metal retention in cattle
tissues from ingestion of sewage sludge. J. Environ. Qual. II:6I6-620.
Baxter, J.C., D.E. Johnson and E.W. Kienholz. 1983. Heavy metals and persistent organics
content in cattle exposed to sewage sludge. J. Environ. Qual. 12:316-319.
Beardsley, A., M.J. Vagg, P.H.T. Beckett, and B.F. Sansom. 1978. Use of the field vole for
monitoring harmful elements in the environment. Environ. Pollut. A16:65-71.
38
-------
Beaudouin, J., R.L Shirley, and D.L. Hammell. 1980. Effect of sewage sludge diets fed swine on
nutrient digestibility, reproduction, growth and minerals in tissues. J. Anim. Sci. 50:572-580.
Bell, P.P., B.R. James and R.L. Chaney. 1991. Heavy metal extractability in long-term sewage
sludge and metal salt-amended soils. J. Environ. Qua). 20:481-486.
Bergholm, J. and E. Steen. 1989. Vegetation establishment on a deposit of zinc mine wastes.
Environ. Pollut. A56:127-144.
Bertrand, J.E., M.C. Lutrick, G.T. Edds, and R.L. West. 1980. Effects of dried digested sludge and
corn grown on soil treated with liquid digested sludge on performance, carcass quality and tissue
residues in beef steers. J. Anim. Sci. 50:35-40.
Bertrand, J.E., M.C. Lutrick, G.T. Edds, and R.L. West. 1981. Metal residues in tissues, animal
performance and carcass quality with beef steers grazing Pensacola bahiagrass pastures treated
with liquid digested sludge. J. Anim. Sci. 53:146-153.
Beyer, W.N. 1986. A reexamination of biomagnification of metals in terrestrial food chains.
Environ. Toxicol. Chem. 5:863-864.
Beyer, W.N., R.L. Chaney, and B. Mulhern. 1982. Heavy metal concentrations in earthworms
from soil amended with sewage sludge. J. Environ. Qua). 11:381-385.
Beyer, W.N. and E.J. Cromartie. 1987. A survey of Pb, Cu, Zn, Cd, Cr, As, and Se in earthworms
and soil from diverse sites. Environ. Monitor. Assess. 8:27-36.
Beyer, W.N., G. Hensler, and J. Moore. 1987. Relation of pH and other soil variables to
concentrations of Pb, Cu, Zn, Cd, and Se in earthworms. Pedobiologia 30:167-172.
Beyer, W.N., O.H. Pattee, L. Sileo, D.J. Hoffman, and B.M. Mulhern. 1985. Metal contamination
in wildlife living near two zinc smelters. Environ. Pollut. A38:63-86.
Beyer, W.N. and C. Stafford. 1992. Survey and evaluation of contaminants in earthworms and in
soils derived from dredged material at confined disposal facilities in the Great Lakes region.
Environ. Monitor. Assess. In press.
Bidwell, A.M. and R.H. Dowdy. 1987. Cadmium and zinc availability to corn following termination
of sewage sludge applications. J. Environ. Qua). 16:438-442.
Binder, S., D.Sokal, and D. Maughan. 1986. Estimating soil ingestion: The use of tracer elements
in estimating the amount of soil ingested by young children. Arch. Environ. Health 41:341-345.
Bloomfield, C. and S.P. McGrath. 1982. A comparison of the extractabilities of Zn, Cu, Ni, and Cr
from sewage sludges prepared by treating raw sewage with the metal salts before or after
anaerobic digestion. Environ. Pollut. 63:193-198.
Bolger, P.M., C.D. Carrington, S.G. Capar and M.A. Adams. 1991. Reductions in dietary lead
exposure in the United States. Chem. Spec. Bioavail. 3:31-36.
Boyer, K.W., J.W. Jones, D. Linscott, S.K. Wright, W. Stroube and W. Cunningham. 1981. Trace
element levels in tissues from cattle fed a sewage sludge-amended diet. J. Toxicol. Environ. Health
8:281-295.
Bremner, I. 1981. Effects of the disposal of copper-rich slurry on the health of grazing animals.
pp. 245-260. jn P. L'Hermite and J. Dehandtschutter (eds.) Copper in Animal Wastes and Sewage
39
-------
Sludge. Reidel Publ., Boston.
Brookes, P.C., C.E. Heijnen, S.P. McGrath and E.D. Vance. 1986. Soil microbial biomass
estimates in soils contaminated with metals. Soil Biol. Biochem. 18:383-388.
Brookes, P.C. and S.P. McGrath. 1984. Effects of metal toxicity on the size of the soil microbial
biomass. J. Soil Sci. 35:341-346.
Brookes, P.C., S.P. McGrath and C. Heijnen. 1986. Metal residues in soils previously treated with
sewage-sludge and their effects on growth and nitrogen fixation by blue-green algae. Soil Biol.
Biochem. 18:345-353.
Bruemmer, G.W., J. Gerth, and U. Herms. 1986. Heavy metal species, mobility and availability in
soils. Z. Pflanzenernahr. Bodenk. 149:382-398.
Brunnert, H. and F. Zadrazil. 1983. The translocation of mercury and cadmium into the fruiting
bodies of six higher fungi. A comparative study on species specificity in five lignocellulolytic fungi
and the cultivated mushroom Agaricus bisporus. Eur. J. Appl. Microbiol. Biotechnol. 17:358-364.
Calabrese, E.J., R. Barnes, E.J. Stanek, H Pastides, C.E. Gilbert, P. Veneman, X. Wang, A. Lasztity
and P.T. Kostecki. 1989. How much soil do young children ingest: An epidemiologic study. Reg.
Toxicol. Pharmacol. 10:73-82.
Calabrese, E.J. and E.J. Stanek, III. 1991. A guide to interpreting soil ingestion studies. 2.
Qualitative and quantitative evidence of soil ingestion. Chem. Spec. Bioavail. 3:55-63.
Chaney, R.L. 1985. Potential effects of sludge-borne heavy metals and toxic organics on soils,
plants, and animals, and related regulatory guidelines. Annex 3, Workshop Paper 9, pp. 1-56. In
Final Report of the Workshop on the International Transportation, Utilization or Disposal of Sewage
Sludge Including Recommendations. PNSP/85-01. Pan American Health Organization, Washington,
DC.
Chaney, R.L. 1990a. Twenty years of land application research - Regulating beneficial use.
BioCycle31(9):54-59.
Chaney, R.L. 1990b. Public health and sludge utilization - Food chain impact. BioCycle 31(10):68-
73.
Chaney, R.L. 1992. Land application of composted municipal solid waste: Public health, safety,
and environmental issues, pp. 61-83. jn Proc. 1991 Solid Waste Composting Conference. Solid
Waste Composting Council, Washington, DC.
Chaney, R.L., R.J.F. Bruins, D.E. Baker, R.F. Korcak, J.E. Smith, Jr., and D.W. Cole. 1987.
Transfer of sludge-applied trace elements to the food-chain, pp. 67-99. In A.L. Page, T.J. Logan
and J.A. Ryan (eds.) Land Application of Sludge—Food Chain Implications. Lewis Publishers Inc.,
Chelsea, Ml.
Chaney, R.L. and P.M. Giordano. 1977. Microelements as related to plant deficiencies and
toxicities. pp. 234-279. in L.F. Elliott and F.J. Stevenson (eds.). Soils for Management of Organic
Wastes and Waste Waters. American Society of Agronomy, Madison, Wl.
Chaney, R.L., and H.W. Mielke. 1986. Standards for soil lead limitations in the United States.
Trace Subst. Environ. Health 20:355-377.
Chaney, R.L., H.W. Mielke, and S.B. Sterrett. 1989. Speciation, mobility, and bioavailability of
40
-------
soil lead. [Proc. Intern. Conf. Lead in Soils: Issues and Guidelines. B.E. Davies & B.C. Wixson
(eds.)l. Environ. Geochem. Health 11(Supplement):105-129.
Chaney, R.L., J.B. Munns, and H.M. Cathey. 1980. Effectiveness of digested sludge compost in
supplying nutrients for soilless potting media. J. Am. Soc. Hort. Sci. 105:485-492.
Chaney, R.L. and J.A. Ryan. 1991. The future of residuals management after 1991. pp. 13D-1
to 13D-15. in AWWA/WPCF Joint Residuals Management Conference (Research Triangle Park,
NC. Aug. 11-14, 1991). Water Pollution Control Federation, Arlington, VA.
Chaney, R.L, J.A. Ryan, and G.A. O'Connor. 1991. Risk assessment for organic micropollutants:
U.S. point of view. pp. 141-158. in P. L'Hermite (ed.) Treatment and Use of Sewage Sludge and
Liquid Agricultural Wastes. Elsevier Applied Science, New York.
Chaney, R.L., S.B. Sterrett, M.C. Morella, and C.A. Lloyd. 1982. Effect of sludge quality and rate,
soil pH, and time on heavy metal residues in leafy vegetables, pp. 444-458. jn Proc. Fifth Annual
Madison Conf. Appl. Res. Pract. Munic. Ind. Waste. Univ. Wisconsin - Extension, Madison, Wl.
Chaney, R.L., G.S. Stoewsand, C.A. Bache, and D.J. Lisk. 1978a. Cadmium deposition and
hepatic microsomal induction in mice fed lettuce grown on municipal sludge-amended soil. J.
Agric. Food Chem. 26:992-994.
Chaney, R.L., G.S. Stoewsand, A.K. Furr, C.A. Bache, and D.J. Lisk. 1978b. Elemental content of
tissues of guinea pigs fed Swiss chard grown on municipal sewage sludge-amended soil. J. Agr.
Food Chem. 26:994-997.
Chang, A.C., T.D. Hinesly, T.E. Bates, H.E. Doner, R.H. Dowdy, and J.A. Ryan. 1987. Effects of
long-term sludge application on accumulation of trace elements by crops, pp. 53-66. in A.L. Page,
T.J. Logan and J.A. Ryan (eds.) Land Application of Sludge — Food Chain Implications. Lewis
Publishers Inc., Chelsea, Ml.
Chang, A.C., A.L. Page, and F.T. Bingham. 1982. Heavy metal absorption by winter wheat
following termination of cropland sludge applications. J. Environ. Qual. 11:705-708.
Cibulka, J., Z. Sova and V. Muzikar. 1983. Lead and cadmium in the tissues of broilers fed a diet
with added dried activated sewage sludge. Environ. Technol. Lett. 4:123-128.
Clausing, P., B. Brunekreef, and J.H. van Wijnen. 1987. A method for estimating soil ingestion by
children. Int. Arch. Occup. Environ. Health 59:341-345.
Cooke, J.A., S.M. Andrews, and M.S. Johnson. 1990. Lead, zinc, cadmium and fluoride in small
mammals from contaminated grassland established on fluorspar tailings. Water, Air, Soil Pollut.
51:43-54.
Corey, R.B., R. Fujii, and L.L. Hendrickson. 1981. Bioavailability of heavy metals in soil-sludge
systems, pp. 449-465. in Proc. Fourth Annual Madison Conf. Appl. Res. Pract. Munic. Ind.
Waste, Univ. Wisconsin-Extension, Madison, Wl.
Corey, R.B., L.D. King, C. Lue-Hing, D.S. Fanning, J.J. Street, and J.M. Walker. 1987. Effects of
sludge properties on accumulation of trace elements by crops, pp. 25-51. in A.L. Page, T.J.
Logan and J.A. Ryan (eds.) Land Application of Sludge — Food Chain Implications. Lewis
Publishers Inc., Chelsea, Ml.
Cotter-Howells, J. and I. Thornton. 1991. Sources and pathways of environmental lead to
41
-------
children in a Derbyshire mining village. Environ. Geochem. Health 13:127-135.
Cunningham, J.D., D.R. Keeney and J.A. Ryan. 1975a. Phytotoxicity in and metal uptake from
soil treated with metal-amended sewage sludge. J. Environ. Qual. 4:455-460.
Cunningham, J.D., D.R. Keeney, and J.A. Ryan. 1975b. Yield and metal composition of corn and
rye grown on sewage sludge-amended soil. J. Environ. Qual. 4(4):448-454.
Cunningham, J.D., D.R. Keeney, and J.A. Ryan. 1975c. Phytotoxicity and uptake of metals added
to soils as inorganic salts or in sewage sludges. J. Environ. Qual. 4:460-461.
D'Arrigo, V., G. Santoprete, and G. Innocent!. 1984. Contenuto in mercurio totals ed inorganico
in funghi coltivati, non coltivati ed in alcuni alimenti (Total organic and inorganic mercury content in
cultivated and wild fungi [Italian foods]). Micol. Ital. 13:69-75.
Davis, A., M.V. Ruby and P.D. Bergstrom. 1992. Bioavailability of arsenic and lead in soils from
the Butte, Montana, mining district. Environ. Sci. Technol. 26:461-468.
Davis, R.D. 1981. Copper uptake from soil treated with sewage sludge and its implications for
plant and animal health, pp. 223-241. ]n P. L'Hermite and J. Dehandtschutter (eds.) Copper in
Animal Wastes and Sewage Sludge. Reidel Publ., Boston, MA.
Davis, R.D. and C.H. Carlton-Smith. 1981. The preparation of sewage sludges with controlled
metal content for experimental purposes. Environ. Pollut. 62:167-177.
Davis, R.D. and C.H. Carlton-Smith. 1984. An investigation into the phytotoxicity of zinc, copper
and nickel using sewage sludge of controlled metal content. Environ. Pollut. B8:163-185.
Davis, S., P. Waller, R. Buschom, J. Ballou and P. White. 1990. Quantitative estimates of soil
ingestion in normal children between the ages of 2 and 7 years: Population-based estimates using
aluminum, silicon, and titaniun as soil tracer elements. Arch. Environ. Health 45:112-122.
Decker, A.M., C.H. Darrah, J.R. Hall, E. Strickling, J.P. Davidson, R.C. Hammond, S.B. Mohanty,
R.L. Chaney, and J.J. Murray. 1979. Feasibility of using sewage sludge for plant and animal
production. Final Report 1976-1977. 195pp. University of Maryland, Department of Agronomy.
Decker, A.M., R.L. Chaney, J.P. Davidson, T.S. Rumsey, S.B Mohanty, and R.C. Hammond.
1980a. Animal performance on pastures topdressed with liquid sewage sludge and sludge
compost, p. 37-41. ]n Proc. Natl. Conf. Municipal and Industrial Sludge Utilization and Disposal.
Information Transfer, Inc., Silver Spring, MD.
Decker. A.M., C.H. Darrah, D.J. Wehner, E. Strickling, M.F. Rothschild, F.R. Gouin, C.B. Link, J.B.
Shanks, J.P. Davidson, R.C. Hammond, S.B. Mohanty, R.L. Chaney, J.J. Murray, D.L. Kern, and
T.S. Rumsey. 1980b. Feasibility of using sewage sludge for plant and animal production. Rnal
Report 1978-1979. 222 pp. University of Maryland, Department of Agronomy.
deVries, M.P.C. 1980. How reliable are the results of pot experiments? Commun. Soil Sci. Plant
Anal. 11:895-902.
deVries, M.P.C. and K.G. Tiller. 1978. Sewage sludge as a soil amendment, with special
reference to Cd, Cu, Mn, Ni, Pb, and Zn - Comparison of results from experiments conducted
inside and outside a greenhouse. Environ. Pollut. 16:213-240.
Diehl, J.F. and U. Schlemmer. 1984. Bestimmung der Bioverfugbarkeit von Cadmium in Pilzen
42
-------
durch Futterungsversuche mit Ratten; Relevanz fur den Menschen (Assessment of bioavailability of
cadmium in mushrooms by means of feeding experiments with rats: relevance for man.). Z.
Ernarhungswiss. 23:126-135.
Domsch, K.H., K. Grabbe, and J. Fleckenstein. 1976. Schwermetallgehalte im Kultursubstrat und
Erntegut des Champignons, Agaricus bisporus (Lange) Singer, beim Einsatz von
Mullklarschlammkompost (Heavy metal contents in the culture substrate and in the mushroom,
Agaricus bisporus, grown in composts mixed with municipal waste and sewage). Z. Pflanzerern.
Bodenk. 1976:487-501.
Doring, H. 1960. Chemical reasons for the "fatigue" of Berlin sewage farm soils and possibilities
for correcting it (in German). Dtsch. Landwirtsch. 11:342-345.
Dressier, R.L., G.L. Storm, W.M. Tzilkowski, and W.E. Sopper. 1986. Heavy metals in cottontail
rabbits on mined lands treated with sewage sludge. J. Environ. Qua). 15:278-281.
El-Aziz, R., J.S. Angle, and R.L. Chaney. 1991. Metal tolerance of Rhizobium meliloti isolated
from heavy metal contaminated soils. Soil Biol. Biochem. 23:795-798.
El Bassam, N. and A. Thorman. 1979. Potentials and limits of organic wastes in crop production.
Compost Sci. 20( ):30-35.
Ellwardt, P.-C. 1977. Variation in content of polycyclic aromatic hydrocarbons in soil and plants
by using municipal waste composts in agriculture, p. 291-298. in Soil Organic Matter Studies,
Proceedings of a Symposium (meeting date 1976). (ed.). Vienna, Austria: IAEA.
Enke, M., M. Roschig, H. Matschiner, and M.K. Achtzehn. 1979. Zur Blei-, Cadmium- und
Quecksilber-Aufnahme in Kulturchampignons (Uptake of lead, cadmium and mercury by cultivated
mushrooms). Die Nahrung 23:731-737.
Epstein, E., R.L. Chaney, C. Henry, and TJ. Logan. 1992. Trace elements in municipal solid
waste compost. Biomass Bioenergy. Submitted.
Evans, K.J., I.G. Mitchell, and B. Salan. 1979. Heavy metal accumulation in soils irrigated by
sewage and effect in the plant-animal system. Progr. Water Technol. 11:339-353.
Fitzgerald, P.R., J. Peterson and C. Lue-Hing. 1985. Heavy metals in tissues of cattle exposed to
sludge-treated pastures for eight years. Am. J. Vet. Res. 46:703-707.
Francois, L.E. 1986. Effect of excess boron on broccoli, cauliflower, and radish. J. Am. Soc.
Hort. Sci. 111:494-498.
Francois, L.E. and R.A. Clark. 1979. Boron tolerance of twenty-five ornamental shrub species. J.
Am. Soc. Hort. Sci. 104:319-322.
Frank, R., J.R. Rainforth, and D. Sangster. 1974. Mushroom production in respect of mercury
content. Can. J. Plant Sci. 54:529-534.
Freeman, G.B., J.D. Johnson, S.C. Liao, P.I. Feder, J.M. Killinger, R.L. Chaney and P.O. Bergstrom.
1991. Effect of soil dose on bioavailability of lead from mining waste to rats. Chem. Spec.
Bioavail. 3:121-128.
Freeman, J.L., M.K. Yousef, S.R. Naegle, and D.S. Barth. 1983. Biokinetics of 74As, 109Cd, and
203Pb: Wild and laboratory rodents. Trace Subst. Environ. Health 17:51-57.
43
-------
Fricke, K., W. Pertl, and H. Vogtmann. 1989. Technology and undesirable components on
compost of separately collected organic wastes. Agric. Ecosys. Environ. 27:463-469.
Gadgil, R.L. 1969. Tolerance of heavy metals and reclamation of industrial waste. J. Appl. Ecol.
6:247-259.
Gaffney, G.R. and R. Ellerston. 1979. Ion uptake of redwinged blackbirds nesting on sludge-
treated spoils, pp. 507-515. in W.E. Sopper and S.N. Kerr (eds.) Utilization of Municipal Sewage
Effluent and Sludge on Forest and Disturbed Land. The Pennsylvania State University Press,
University Park, PA.
Gallardo-Laro, F. and R. Nogales. 1987. Effect of the application of town refuse compost on the
soil-plant system: A review. Biol. Wastes 19:35-62.
Ciller, K.E., S.P. McGrath, and P.R. Hirsch. 1989. Absence of nitrogen fixation in clover grown on
soil subject to long-term contamination with heavy metals is due to survival of only ineffective
Rhizobium. Soil Biol. Biochem. 21:841-848.
Gilliam, C.H. and M.E. Watson. 1981. Boron accumulation in Taxus media. HortSci. 16:340-341.
Gilmour, J.T. and M.D. Clark. 1988. Nitrogen release from wastewater sludge: A site specific
approach. J. Water Pollut. Contr. Fed. 60:494-498.
Giordano, P.M., J.J. Mortvedt and D.A. Mays. 1975. Effect of municipal waste on crop yields and
uptake of heavy metals. J. Environ. Qua). 4:394-399.
Gogue, G.J. and K.C. Sanderson. 1973. Boron toxicity of chrysanthemum. HortSci. 8:473-475.
Gogue, G.J. and K.C. Sanderson. 1975. Municipal compost as a medium amendment for
Chrysanthemum culture. J. Am. Soc. Hort. Sci. 100:213-216.
Gray, K.R. and A.J. Biddlestone. 1980. Agricultural use of composted town refuse, pp. 293-305.
]n Inorganic Pollution and Agriculture. Agr. Develop. Advis. Serv., HMSO, London.
Gupta, U.C. 1979. Boron nutrition of crops. Adv. Agron. 31:273-307.
Gupta, U.C., J.D.E. Sterling and H.G. Nass. 1973. Influence of various rates of compost and
nitrogen on the boron toxicity symptoms in barley and wheat. Can. J. Plant Sci. 53:451-456.
Haan, S. de. 1981. Results of municipal waste compost research over more than fifty years at the
Institute for Soil Fertility at Haren/Groningen, the Netherlands. Neth. J. Agric. Sci. 29:49-61.
Ham, G.E. and R.H. Dowdy. 1978. Soybean growth and composition as influenced by soil
amendments of sewage sludge and heavy metals: Field studies. Agron. J. 70:326-330.
Hansen, L.G., P.K. Washko, L.G.M.T. Tuinstra, S.B. Dorn, and T.D. Hinesly. 1981.
Polychlorinated biphenyl, pesticide, and heavy metal residues in swine foraging on sewage sludge
amended soils. J. Agr. Food Chem. 29:1012-1017.
Harms, H. and D.R. Sauerbeck. 1983. Toxic organic compounds in town waste materials: Their
origin, concentration and turnover in waste composts, soils and plants, p. 38-51. In Environmental
Effects of Organic and Inorganic Contaminants in Sewage Sludge. R.D. Davis, G. Hucker and P.
L'Hermite (ed.). Boston: D. Reidel Pub. Co. Proceedings of a workshop held 25-26 May 1982.
Stevenage.
44
-------
Heckman, J.R., J.S. Angle, and R.L. Chaney. 1986. Soybean nodulation and nitrogen fixation on
soil previously amended with sewage sludge. Biol. Fertil. Soils 2:181-185.
Heckman, J.R., J.S. Angle, and R.L. Chaney. 1987a. Residual effects of sewage sludge on
soybeans. I. Accumulation of heavy metals. J. Environ. Qual. 16:113-117.
Heckman, J.R., J.S. Angle, and R.L Chaney. 1987b. Residual effects of sewage sludge on
soybeans. II. Accumulation of soil and symbiotically fixed nitrogen. J. Environ. Qual.
16:118-124.
Heffron, C.L., J.T. Reid, A.K. Furr, T.F. Parkinson, J.M. King, C.A. Bache, L.E. St. John, Jr., W.H.
Gutenmann and D.J. Lisk. 1977. Lead and other elements in sheep fed colored magazines and
newsprint. J. Agr. Food Chem. 25:657-660.
Hegstrom, L.J. and S.D. West. 1989. Heavy metal accumulation in small mammals following
sewage sludge application to forests. J. Environ. Qual. 18:345-349.
Helmke, P.A., W.P. Robarge, R.L. Korotev, and P.J. Schomberg. 1979. Effects of soil-applied
sewage sludge on concentrations of elements in earthworms. J. Environ. Qual. 8:322-327.
Hewitt, E.J. 1954. Metal interrelationships in plant nutrition. I. Effects of some metal toxicities
on sugar beet, tomato, oat, potato and marrowstem kale in sand culture. J. Ept. Bot. 4:59-64.
Hinesly, T.D., E.L. Ziegler, and G.L. Barret. 1979. Residual effects of irrigating corn with digested
sewage sludge. J. Environ. Qual. 8:35-38.
Hinesly, T.D., E.L. Ziegler, and J.J. Tyler. 1982. Selected chemical elements in tissues of
pheasants fed corn grain from sewage sludge-amended soil. Agro. Ecosyst. 3:11-26.
Hogue, D.E., J.J. Parrish, R.H. Foote, J.R. Stouffer, J.L. Anderson, G.S. Stoewsand, J.N. Telford,
C.A. Bache, W.H. Gutenmann, and D.J. Lisk. 1984. Toxicologic studies with male sheep grazing
on municipal sludge-amended soil. J. Toxicol. Environ. Health 14:153-161.
Holmgren, G.G. S., M.W. Meyer, R.L. Chaney and R.B. Daniels. 1992. Concentrations of cadmium,
lead, zinc, copper, and nickel in agricultural soils of the United States. .J. Environ. Qual. In press.
Hunter, B.A., and M.S. Johnson. 1982. Food chain relationships of copper and cadmium in
contaminated grassland ecosystems. Oikos 38:108-117.
Ireland, M.P. 1983. Heavy metal uptake and tissue distribution in earthworms, pp. 247-265. In
J.E. Satchell led.) Earthworm Ecology: From Darwin to Vermiculture. Chapman and Hall, New
York.
Jacobs, L.W., G.A. O'Connor, M.A. Overcash, M.J. Zabik and P. Rygiewicz. 1987. Effects of
trace organics in sewage sludges on soil-plant systems and assessing their risks to humans, pp.
107-183. ]n A.L. Page, TJ. Logan and J.A. Ryan (eds). Land Application of Sludge. Lewis
Publishers, Chelsea, Ml.
Jing, J. and T. Logan. 1992. Effects of sewage sludge cadmium concentration on chemical
extractability and plant uptake. J. Environ. Qual. 21:73-81.
Johnson, J.C., Jr., P.R. Utley, R.L. Jones, and W.C. McCormick. 1975. Aerobic digested
municipal garbage as a feedstuff for cattle. J. Anim. Sci. 41:1487-1495.
Johnson, N.B., P.H.T. Beckett, and CJ. Waters. 1983. Limits of zinc and copper toxicity from
45
-------
digested sludge applied to agricultural land. pp. 75-81. ]n R.D. Davis, G. Hucker, and P. L'Hermite
(eds.). Environmental Effects of Organic and Inorganic Contaminants in Sewage Sludge. D. Reidel
Publ., Dordrecht.
Kienholz, E., G.M. Ward, D.E. Johnson, J.C. Baxter, G.L. Braude, and G. Stern. 1979.
Metropolitan Denver sewage sludge fed to feedlot steers. J. Anim. Sci. 48:735-741.
King, L.D. and H.D. Morris. 1972. Land disposal of liquid sludge: II. The effect on soil pH,
manganese, zinc, and growth and chemical composition of rye (Secale cereale L.). J. environ Qual.
1:425-429.
Kinkle, B.K., J.S. Angle, and H.H. Keyser. 1987. Long-term effects of metal-rich sewage sludge
application on soil populations of Bradyrhizobium japonicum. Appl. Environ. Microbiol. 53:315-
319.
Konig, W., J. Leisner-Saaber, T. Delschen and C. Berns. 1991. Schwermetallbelastung in Garten
des Raumes Stolberg -Datenauswertung/ Untersuchungsprogramm/ Anbauempfehlungen.
Landesanstalt fur Okologie, Landschaftsentwicklung und Forstplanung, Nordrhein-Westfalen, FRG.
116pp.
Korcak, R.F. and D.S. Fanning. 1985. Availability of applied heavy metals as a function of type of
soil material and metal source. Soil Sci. 140:23-34.
LJsk, D.J., W.H. Gutenmann, M. Rutzke, H.T. Kuntz and G. Chu. 1992a. Survey of toxicants and
nutrients in composted waste materials. Arch. Environ. Contam. Toxicol. 22:190-194.
LJsk, D.J., W.H. Gutenmann, M. Rutzke, H.T. Kuntz and G.J. Doss. 1992b. Composition of
toxicants and other constituents in yard or sludge composts from the same community as a
function of time-of-waste-collection. Arch. Environ. Contam. Toxicol. 22:380-383.
Logan, T.J., and R.L. Chaney. 1983. Utilization of municipal wastewater and sludges on land —
Metals, pp. 235-323. ]n A.L. Page, T.L. Gleason, III, J.E. Smith, Jr., I.K. Iskander, and L.E.
Sommers (eds.) Proc. 1983 Workshop on Utilization of Municipal Wastewater and Sludge on Land.
University of California, Riverside, CA.
Logan, T.J., and R.L. Chaney. 1987. Nonlinear rate response and relative crop uptake of sludge
cadmium for land application of sludge risk assessment, pp 387-389. jn Proc. Sixth Intern. Conf.
Heavy Metals in the Environment. Vol. 1. CEP Consultants, Edinburgh, Scotland.
Lumis, G.P. and A.G. Johnson. 1982. Boron toxicity and growth suppression of Forsythia and
Thuja grown in mixes amended with municipal waste compost. HortSci. 17:821-822.
Lutrick, M.C., W.K. Robertson, and J.A. Cornell. 1982. Heavy applications of liquid-digested
sludge on three ultisols: II. Effects on mineral uptake and crop yield. J. Environ. Qual. 11:2883-
287.
Ma, W.-C. 1982. The influence of soil properties and worm-related factors on the concentration
of heavy metals in earthworms. Pedobiologia 24:109-119.
Ma, W.-C. 1987. Heavy metal accumulation in the mole, Talpa europea, and earthworms as an
indicator of metal bioavailability in terrestrial environments. Bull. Environ. Contam. Toxicol.
39:933-939:
Ma, W.-C. 1989. Effect of soil pollution with metallic lead pellets on lead bioaccumulation and
46
-------
organ/body weight alterations in small mammals. Arch. Environ. Contam. Toxicol. 18:617-622.
Macdonald, D.W. 1983. Predation on earthworms by terrestrial vertebrates, pp. 393-314. ]n
J.E. Satchell (ed.) Earthworm Ecology: From Darwin to Vermiculture. Chapman and Hall, London.
Mahler, R.J. and J.A. Ryan. 1988a. Cadmium sulfate application to sludge-amended soils: II.
Extraction of Cd, Zn, and Mn from solid phases. Commun. Soil Sci. Plant Anal. 19:1747-1770.
Mahler, R.J. and J.A. Ryan. 1988a. Cadmium sulfate application to sludge-amended soils: III.
Relationship between treatment and plant available cadmium, zinc, and manganese. Commun. Soil
Sci. Plant Anal. 19:1771-1794.
Mahler, R.J., J.A. Ryan, and T. Reed. 1987. Cadmium sulfate application to sludge-amended
soils. I. Effect on yield and cadmium availability to plants. Sci. Total Environ. 67:117-131.
Marks, M.J., J.H. Williams, and C.G. Chumbley. 1980. Reid experiments testing the effects of
matal-contaminated sewage sludges on some vegetable crops, pp. 235-251. In Inorganic
Pollution and Agriculture. Min. Agr. Fish. Food Reference Book 326. HMSO. London.
McGrath, S.P., P.C. Brookes, and K.E. Ciller. 1988. Effects of potentially toxic metals in soil
derived from past applications of sewage sludge on nitrogen fixation by Trifolium repens L. Soil
Biol. Biochem. 20:415-424.
McGrath, S.P., P.R. Hirsch and K.E. Ciller. 1988. Effect of heavy metal contamination on the
genetics of nitrogen-fixing populations of Rhizobium leguminosarum nodulating white clover, pp.
164-166. jn A.A. Orio (ed.) Environmental Contamination.
McKenna, I.M., and R.L. Chancy. 1991. Cadmium transfer to humans from food crops grown in
sites contaminated with cadmium and zinc. pp. 65-70. In L.D. Fechter (ed.) Proc. 4th Intern.
Conf. Combined Effects of Environmental Factors; Oct. 1-3, 1990, Baltimore, MD. Johns Hopkins
University School of Hygiene and Public Health, Baltimore.
McKenna, I.M., R.L. Chaney, S.H. Tao, R.M. Leach, Jr. and F.M. Williams. 1992. Interactions of
plant zinc and plant species on the bioavailability of plant cadmium to Japanese quail fed lettuce
and spinach. Environ. Res. 57:73-87. . .
McKenzie-Parnell, J.M., and G. Eynon. 1987. Effect on New Zealand adults consuming large
amounts of cadmium in oysters. Trace Subst. Environ Health 21:420-430.
McKenzie-Parnell, J.M., T.E. Kjellstrom, R.P. Sharma, and M.F. Robinson. 1988. Unusually high
intake and fecal output of cadmium, and fecal output of other trace elements in New Zealand
adults consuming dredge oysters. Environ. Res.46:1-14.
Minagawa, K., T. Sasaki, Y. Takizawa, R. Tamura, and T. Oshina. 1980. Accumulation route and
chemical form of mercury in mushroom species. Bull. Environ. Contam. Toxicol. 25:382-388.
Minnich, M.M. and M.B. McBride. 1986. Effect of copper activity on carbon and nitrogen
mineralization in field-aged copper-enriched soils. Plant Soil 91:231-240.
Minnich, M.M., M.B. McBride, and R.L. Chaney. 1987. Copper activity in soil solution. 2.
Relation to copper accumulation in young snapbeans. Soil Sci. Soc. Am. J. 51:573-578.
Muller, W.P. and F. Korte. 1976. Ecological chemical evaluation of waste treatment procedures:
The behavior of xenobiotics in waste composting. Environ. Qual. Saf. 5:215-236.
47
-------
Neary, D.G., G. Schneider, and D.P. White. 1975. Boron toxicity in red pine following municipal
wastewater irrigation. Soil Sci. Soc. am. Proc. 39:981-982.
Nogales, R., J. Robles and F. Gallardo-Lara. 1987. Boron release from town refuse compost as
measured by sequential plant uptake. Waste Manag. Res. 5:513-520.
NRC (National Research Council). 1980. Mineral Tolerance of Domestic Animals. National
Academy of Sciences, Washington, D.C. 577pp.
O'Connor, G.A., R.L. Chaney, and J.A. Ryan. 1991. Bioavailability to plants of sludge-borne toxic
organics. Rev. Environ. Contam. Toxicol. 121:129-155.
O'Connor, G.A., D. Kiehl, G.A. Eiceman, and J.A. Ryan. 1990. Plant uptake of sludge-borne
PCBs. J. Environ. Qual. 19:113-118.
Oertli, J.J. and E. Grgurevic. 1975. Effect of pH on the absorption of boron by excised barley
roots. Agron. J. 67:278-280.
Osuna, 0., G.T. Edds and J.A. Popp. 1981. Comparative toxicity of feeding dried urban sludge
and an equivalent amount of cadmium to swine. Am. J. Vet. Res. 42:1542-1546.
Page, A.L., T.J. Logan and J.A. Ryan (eds.) 1987. Land Application of Sludge — Food Chain
Implications. Lewis Publishers Inc., Chelsea, Ml.
Page, A.L., T.J. Logan and J.A. Ryan (eds.) 1989. W-170 Peer Review Committee analysis of the
Proposed 503 Rule on sewage sludge. CSRS Technical Committee W-170, Univ. California-
Riverside.
Petruzzelli, G. 1989. Recycling wastes in agriculture: Heavy metals bioavailability. Agric.
Ecosyst. Environ. 27:493-503.
Pimentel, D., T. Culliney, M.N. Burgess, G.S. Stoewsand, J.L. Anderson, C.A. Bache, W.H.
Gutenmann and D.J. Lisk. 1984. Cadmium in Japanese quail fed earthworms inhabiting a golf
course. Nutr. Rep. Int. 30:475-481.
Poole, D.B.R., D. McGrath, G.A. Fleming, and J. Sinnott. 1983. Effects of applying copper-rich
pig slurry to grassland. 3. Grazing trials: Stocking rate and slurry treatment. Ir. J. Agric. Res.
22:1-10.
Purves, D. 1972. Consequences of trace-element contamination of soils. Environ. Pollut.
3:17-24.
Purves, D. and E.J. Mackenzie. 1973. Effects of applications of municipal compost on uptake of
copper, zinc and boron by garden vegetables. Plant Soil 39:361-371.
Purves, D. and E.J. Mackenzie. 1974. Phvtotoxicity due to boron in municipal compost. Plant
Soil 40:231-235.
Quinch, J.P., A. Bolay, and V. Dvorak. 1976. Contamination of plants and soils of French-
speaking Switzerland by mercury. In vitro tests of methylation of mercury by fungi (in French).
Rev. Suisse Agric. 8:130-142.
Reddy, C.S., C.R. Dorn, D.N. Lamphere and J.D. Powers. 1985. Municipal sewage sludge
application on Ohio farms: Tissue metal residues and infections. Environ. Res. 38:360-376.
48
-------
Rinno, G. 1964. A contribution on the cause of sewage-exhaustion of soil. Albrecht Thaer. Arch.
8:699-710.
Rohde, G. 1962. The effects of trace elements on the exhaustion of sewage-irrigated land. J.
Inst. Sewage Purif. 1962:581-585.
Ross, A.D. and C. Short. 1990. Utilisation of sewage sludge on agricultural land — Animal trial
(The effects of sewage sludge applied to pastures on production parameters and residue levels in
sheep.). Final Report on Project P86/03. Macarthur Agricultural Institute, New South Wales,
Australia. 12pp.
Rother, J.A., J.W. Millbank and I. Thornton. 1982a. Seasonal fluctuations in nitrogen fixation
(acetylene reduction) by free-living bacteria in soils contaminated with cadmium, lead and zinc. J.
SoilSci. 33:101-113.
Rother, J.A., J.W. Millbank, and I. Thornton. 1982b. Effects of heavy-metal additions on
ammonification and nitrification in soils contaminated with cadmium, lead, and zinc. Plant Soil
69:253-268.
Rother, J.A., J.W. Millbank, and I. Thornton. 1983. Nitrogen fixation by white clover (Trifolium
repens) in grasslands on soils contaminated with cadmium, lead, and zinc. J. Soil Sci.
34:127-136.
Ruby, M.V., A. Davis, J.H. Kempton, J.W. Drexler, and P.O. Bergstrom. 1992. Lead
bioavailability: Dissolution kinetics under simulated gastric conditions. Environ. Sci. Techno).
26:1242-1248.
Ryan, J.A. and R.L. Chaney. 1992. ]n H.A.J. Hoitink et al. (eds.). Regulation of municipal
sewage sludge under the Clean Water Act Section 503: A model for exposure and risk assessment
for MSW-compost. Proc. International Composting Research Symposium.
Ryan, J.A., H.R. Pahren, and J.B. Lucas. I982. Controlling cadmium in the human food chain: A
review and rationale based on health effects. Environ. Res. 28:251-302.
Sanders, J.R. and T.M. Adams. 1987. The effects of pH and soil type on concentration of zinc,
copper and nickel extracted by calcium chloride from sewage sludge-treated soils. Environ. Pollut.
A43:219-228.
Sanders, J.R., S.P. McGrath, and T.M. Adams. 1986. Zinc, copper, and nickel concentrations in
ryegrass grown on sewage sludge-contaminated soils of different pH. J. Sci. Food Agric.
37:961-968.
Sanderson, K.C. 1980. Use of sewage-refuse compost in the production of ornamental plants.
HortSci. 15:173-178.
Sanson, D.W., D.M. Hallford, and G.S. Smith. 1984. Effects of long-term consumption of sewage
solids on blood, milk and tissue elemental conposition of breeding ewes. J. Anim. Sci.
59:416-424.
Sauerbeck, D.R. 1991. Plant, element and soil properties governing uptake and availability of
heavy metals derived from sewage sludge. Water, Air, Soil Pollut. 57-58:227-237.
Scanlon P.F. 1987. Heavy metals in small mammals in roadside environments: Implications for
food chains. Sci. Total Environ. 59:317-323.
49
-------
Schellmann, B., M.-J. Hilz, and O. Opitz. 1980. Cadmium- und Kupferausscheidung nach
Aufnahme von Champignon-Mahlzeiten (Fecal excretion of cadmium and copper after mushroom
(Agaricus) diet). Z. Lebensm. Unters. Forsch. 171:189-192.
Schellmann, B., E. Rohmer, K.H. Schaller, and D. Weltle. 1984. Cadmium- und
Kupferkonzentrationen in Stuhl, Urin und Blut nach Aufnahme wildwachsender Champignons.
(Concentration of cadmium and copper in feces, urine and blood after ingestion of wild mushrooms
[Agaricus species].) Z. Lebensm. Unters. Forsch. 178:445-449.
Shacklette, H.T., and J.G. Boerngen. 1984. Element concentrations in soils and other surficial
materials of the conterminous United States. U.S. Geol. Surv. Prof. Paper 1270:1-105.
Sharma, R.P., T. Kjellstrom, and J.M. McKenzie. 1983. Cadmium in blood and urine among
smokers and non-smokers with high cadmium intake via food. Toxicology 29:163-171.
Sheaffer, C.C., A.M. Decker, R.L. Chancy, G.C. Stanton, and D.C. Wolf. 1981. Soil temperature
and sewage sludge effects on plant and soil properties. EPA-600/S2-81-069. (NTIS:PB
81-191,199).
Sileo, L., and W.M. Beyer. 1985. Heavy metals in white-tailed deer living near a zinc smelter in
Pennsylvania. J. Wildlife Dis. 21:289-296.
Smith, G.S., D.M. Hallford, and J.B. Watkins, III. 1985. Toxicological effects of gamma-irradiated
sewage solids fed as seven percent of diet to sheep for four years. J. Anim. Sci. 61:931-941.
Spotswood, A. and M. Raymer. 1973. Some aspects of sludge disposal on agricultural land.
Water Pollut. Contr. 72:71-77.
Stanek, E.J., III and E.J. Calabrese. 1991. Methodological considerations in soil ingestion
estimation. Chem. Spec. Bioavail. 3:65-67.
Steele, M.J., B.D. Beck, B.L. Murphy, and H.S. Strauss. 1990. Assessing the contribution from
lead in mining wastes to blood lead. Regulat. Toxicol. Pharmacol. 11:158-190.
Stegnar, P., L. Kosta, A.R. Bryne, and V. Ravnik. 1973. The accumulation of mercury by, and the
occurrence of methyl mercury in, some fungi. Chemosphere 2:57-63.
Stivje, T. and R. Besson. 1976. Mercury, cadmium, lead, and selenium content of mushroom
species belonging to the Genus Agaricus. Chemosphere 5:151-158.
Stivje, T. and R. Roschnik. 1974. Mercury and methyl mercury content of different species of
fungi. Trav. Chim. Aliment. Hyg. 65:209-220.
Stoewsand, G.S., C.A. Bache, W.H. Gutenmann and D.J. Lisk. 1986. Concentration of cadmium
in Coturnix quail fed earthworms. J. Toxicol. Environ. Health 18:369-376.
Strehlow, C.D., and D. Barltrop. 1988. The Shipham Report — An investigation into cadmium
concentrations and its implications for human health: 6. Health studies. Sci. Total Environ.
75:101-133.
Svartengren, M., C.-G. Elinder, L. Friberg, and B. Lind. 1986. Distribution and concentration of
cadmium in human kidney. Environ. Res. 39:1-7.
Terman, G.L. and D.A. Mays. 1973. Utilization of municipal solid waste compost: Research
50
-------
results at Muscle Shoals, Alabama. Compost Sci. 14(1 ):18-21.
Trocme, S., G. Barbier and J. Chabannes. 1950. Chlorosis, caused by lack of manganese, of
crops irrigated with filtered water from Paris sewers (in French). Ann. Agron. 1:663-685.
US-DHEW. 1991. Preventing Lead Poisoning in Young Children. A statement by the centers for
disease control - October, 1991. 105 pp.
U.S. Environmental Protection Agency. 1989a. Development of risk assessment methodology for
land application and distribution and marketing of municipal sludge. EPA/600/6-89/001.
U.S. Environmental Protection Agency. 1989b. Standards for the disposal of sewage sludge.
Federal Register 54(23):5746-5902.
U.S. Environmental Protection Agency. 1990. Technical support documentation for Part I of the
National Sewage Sludge Survey Notice of Availability. US-EPA, Analysis and Evaluation Division,
Washington, DC. Oct. 31, 1990.
Utley, P.R., O.H. Jones, Jr., and W.C. McCormick. 1972. Processed municipal solid waste as a
roughage and supplemental protein source in beef cattle finishing diets. J. Anim. Sci. 35:139-143.
Van Wijnen, J.H., P. Clausing and B. Brunekreef. 1990. Estimated soil ingestion by children. Env.
Res. 51:147-162.
Vimmerstedt, J.P. and T.N. Glover. 1984. Boron toxicity to sycamore on minesoil mixed with
sewage sludge containing glass fibers. Soil Sci. Soc. Am. J. 48:389-393.
Volk, V.V. 1976. Application of trash and garbage to agricultural lands, pp. 154-164. jn Land
Application of Waste Materials. Soil Conservation Soc. Am., Ankeny, Iowa.
Vreman, K., N.G. van der Veen, E.J. van der Molen, and W.G. de Ruig. 1986. Transfer of
cadmium, lead, mercury, and arsenic from feed into milk and various tissues of dairy cows:
Chemical and pathological data. Neth. J. Agric. Res. 34:129-144.
Webber, M.D., Y.K Soon, T.E. Bates, and A.U. Haq. 1981. Copper toxicity resulting from land
application of sewage sludge, pp. 117-135. ]n P. L'Hermite and J. Dehandschutter (eds.) Copper
in Animal Wastes and Sewage Sludge. Reidel Publ., Boston, MA.
Wild, S.R. and K.C. Jones. 1992. Polynuclear aromatic hydrocarbon uptake by carrots grown in
sludge-amended soil. J. Environ. Qual. 21:217-225.
Wild, S.R., J.P. Obbard, C.I. Munn, M.L. Berrow, and K.C. Jones. 1991. The long-term
persistence of polynuclear aromatic hydrocarbons (PAHs) in an agricultural soil amended with
metal-contaminated sewage sludges. Sci. Total Environ. 101:235-253.
Wild, S.R., K.S. Waterhouse, S.P. McGrath and K.C. Jones. 1990. Organic contaminants in an
agricultural soil with a known history of sewage sludge amendments: Polynuclear aromatic
hydrocarbons. Environ. Sci. Technol.24:1706-1711.
Zabowski, D., R.J. Zasoski, W. Littke, and J. Ammirati. 1990. Metal content of fungal sporocarps
from urban, rural, and sludge-treated sites. J. Environ. Qual. 19:372-377.
51
-------
TREATMENT OP SLUDGE FOR LAND APPLICATION:
WHICH PROCESSES ARE ACCEPTABLE UNDER CURRENT FEDERAL REGULATION?
U.S. Environmental Protection Agency's Pathogen Equivalency Committee
ABSTRACT
Current federal regulations require that municipal wastewater sludge be
treated prior to land application by one of several listed technologies or by
an "equivalent" process, in 1985, the U.S. Environmental Protection Agency
created the Pathogen Equivalency Committee (PEC) to provide guidance on
equivalency. The Committee developed criteria for equivalency and initiated
an evaluation process. Any interested party may submit an application for PEC
guidance on equivalency. The PEC has evaluated several different
technologies; many have been found equivalent. The sludge regulations
proposed on February 6, 1989 contain performance-based standards for sludge
land application that are similar to the current equivalency criteria.
- 1 -
-------
INTRODUCTION
Municipal wastewater sludge is used as a soil conditioner and partial
fertilizer in the United States and many other countries. While sludge has
beneficial plant nutrients and soil-conditioning properties, it may also
contain bacteria, viruses, protozoa, parasites, and other pathogenic
»
microorganisms. All land application of sludge creates a potential for human
exposure to these organisms through direct and indirect contact.
In September 1979, under the joint authority of Resource Conservation and
Recovery Act (RCRA) and the Clean Water Act (CWA), EPA promulgated regulations
governing the application of wastewater sludge to land under 40 CFR Part 257 -
Criteria for Classification of Solid Waste Disposal Facilities and Practices
(see also 44 Federal Register 53460, September 13, 1979; and 44 PR 54708,
September 21, 1979). These regulations apply to all municipal sludge destined
for land application, including sludge products that are distributed and
marketed. They protect public health from pathogens in land-applied sludge by
mandating treatment of sludge prior to application to reduce its
disease-bearing potential, and by controlling land use following sludge
application. The regulations list specific sludge treatment technologies that
provide acceptable levels of pathogen reduction.*
The 40 CFR regulation also states that sludge from other treatment
technologies can be applied to land if the alternative treatment controls
*It should be noted that while 40 CFR was promulgated in 1979, it ws not
until the Agency's 1984 municipal sludge policy was developed and the Water
Quality Act of 1987 was passed with its 405(d)(4) program requiring interim
permitting of sludge management programs, that enforcement, via the National
Pollutant Discharge Elimination System (NPDES) permit system, occurred.
Recently, EPA* has issued guidance for writing case-by-case permit
requirements for municipal wastewater sludge.
- 2 -
-------
pathogens and disease vectors (rodents, flies, mosquitoes, etc.) to an extent
equivalent to that provided by the listed technologies. However, the
regulations provide no guidance on determining whether alternative processes
are equivalent. Following promulgation of the regulations, developers,
owners, and operators of sludge treatment technologies began contacting EPA
»
for -guidance on whether their technology was equivalent. To respond to this
need, EPA created a Pathogen Equivalency Committee in 1985 to review
alternative sludge treatment technologies and to provide technical guidance on
whether they are equivalent.-- This paper describes the function of the
Pathogen Equivalency Committee (PEC); the criteria for equivalency developed
by the PEC; and the equivalency guidance process. Additional information can
be found in the EPA guidance document Control of Pathogens in Municipal
Wastewater Sludge.
LISTED PROCESSES UNDER CURRENT FEDERAL REGULATIONS
The treatment processes and operating conditions that must be followed to
ensure appropriate pathogen and vector attraction reduction are listed in
Appendix II of 40 CFR Part 257. These processes are divided into two
categories based on the level of pathogen control they can achieve:
•Processes to Significantly Reduce Pathogens' (PSRPs)(see Table 1), which
reduce pathogens to a level comparable to that achieved by a well-run
anaerobic digestor, and "Processes to Further Reduce Pathogens' (PFRPs)(see
Table 2), which reduce pathogens to below detectable levels. Since PSRPs do
not eliminate pathogens, PSRP-treated sludge still has the potential to
transmit disease. The regulations protect public health by controlling public
access, the growing of human food crops, and grazing by dairy or
meat-producing livestock for specific time periods at sites where sludge has
been applied. PFRPs reduce pathogens to below detectable levels, so there are
- 3 -
-------
no pathogen-related management restrictions for sites where PFRP-treated
sludges have been applied.
THE PATHOGEN EQUIVALENCY COMMITTEE
The Pathogen Equivalency Committee has been providing guidance on
>
equivalency for over 3 years. It currently consists of approximately six
members with expertise in microbiology, wastewater engineering, statistics,
and sludge regulations. The committee reviews and makes recommendations to
EPA management on applications for PSRP or PFRP equivalency for treatment
technologies and stockpiled sludge, and provides guidance to applicants on the
data necessary to determine equivalency. The committee does not recommend
process changes or appropriate uses of sludge products. The PEC's
determinations concerning equivalency are not formal binding Agency
decisions. Rather, they constitute technical guidance and are advisory.
DEFINING EQUIVALENCY
The first task of the PEC was to develop criteria for equivalency. To do
this, the committee examined the rationale behind the selection of the listed
PSRP and PPRP technologies. The PEC then adopted -the criteria used-to list
the technologies as the basis for defining equivalency, as described below.
PSRP Equivalency
As clarified by Whittington and Johnson, the listed PSRPs and the
specified operating conditions for these technologies (see Table 1) were
selected to ensure the processes would (1) consistently reduce the density of
pathogenic viruses and bacteria (measured as the no./g TSS at 5% solids) in
mixed sludge from a conventional plant by equal to or greater than 1 log (base
10), and (2) reduce vector attractiveness to the same degree as properly
- 4 -
-------
conducted anaerobic digestion. This is the reduction achieved by anaerobic
digestion under the operating conditions described in the regulation. This
level of pathogen and vector attraction reduction was viewed as the standard
that any PSRP should meet. The PEC adopted these "listing criteria" as the
basic criteria for PSRP equivalency.
»
The PEC then examined the ways in which this reduction in pathogenic
viruses and bacteria could be demonstrated. It determined that different
approaches were possible, depending on whether the process is conventional or
nonconventional. The PEC also-found that the criteria could be modified
somewhat for sludges produced by no primary/long sludge age (NP/LSA)
wastewater treatment processes, because of the consistently lower pathogen
densities in these sludges. The various criteria for demonstrating PSRP are
described below.
4
Conventional Processes. Data compiled by Farrell et al. and Farrah
et al. indicate that, for conventional biological and chemical treatment
processes (e.g., digestion, lime treatment, chlorine treatment), a reduction
of 1 log (base 10) in pathogenic virus and bacteria density correlates with a
reduction of 1 to 2 logs (base 10) in indicator organisms. On this basis, a
6
2-log (base 10) reduction in fecal indicators is accepted by EPA as
satisfying the requirement to reduce pathogens by 1 log (base 10) for these
types of processes. Specifically, a 2-log (base 10) reduction (measured in
no./g total suspended solids) in either (1) fecal coliforms and fecal
streptococci, or (2) fecal coliforms and enterococci must be demonstrated.
Historically, this has been the standard reduction to demonstrate equivalency
to PSRP for conventional processes.
However, according to Farrell,7 a substantial amount of data have been
generated recently to indicate that sludge produced by conventional wastewater
- 5 -
-------
treatment and anaerobic digestion at 35°C (95°F) for more than 15 d contains
fecal coliforms and fecal streptococci at average log (base 10) densities
(no./g TSS) of less than 6.0. Thus, for processes or combinations of
processes that do not depart radically from conventional treatment (gravity
thickening, anaerobic or aerobic biological treatment, dewatering, air drying,
»
and "storage of liquid or sludge cake), or for any process where there is a
demonstrated correlation between pathogenic bacteria and virus reduction and
indicator organism reduction, the PEC accepts an average log (base 10) density
(no./g TSS) of fecal coliforms and fecal streptococci of less than 6.0 in the
treated sludge as indicating adequate viral and bacterial pathogen reduction.
(The average log density is the log of the geometric mean of the samples
taken.)
Nonconventional Processes. For nonconventional sludge treatment
processes, such as radiation, for which no data are available or data indicate
an inappropriate correlation between pathogen reduction and indicator organism
reduction, indicator organism data are not acceptable. Instead, it must be
demonstrated that the process is capable of causing at least a 1-log (base 10)
reduction in the least susceptible organism (i.e., total enteroviruses or
Salmonella spp.). Presumably, if the process adequately reduces the most ,
resistant organism, it will also adequately reduce the more sensitive
organism.
Processes Treating Sludges Generated by No Primary/Long Sludge Age
(NP/LSA) Wastewater Treatment. The original PSRP criterion of a 1-log (base
10) reduction in pathogenic viruses and bacteria was based on reductions
achieved by processes treating mixed sludge produced by conventional
Q
wastewater treatment. Recent data, compiled by Farrell et al., indicate
- 6 -
-------
that sludges produced by no primary/long sludge age wastewater treatment
processes,* such as extended aeration and oxidation ditch treatment, have
pathogen densities that are approximately 0.3 log (base 10) lower than sludges
produced by conventional primary and waste-activated wastewater treatment
processes. Therefore, if NP/LSA sludges are treated by processes that provide
v
an additional 0.7 log (base 10) reduction in pathogenic bacteria and viruses,
they will have achieved a pathogen reduction equivalent to that achieved in a
conventional sludge treated by a PSRP. Thus, to be considered equivalent to
PSRP, processes that are treating-NP/LSA sludges need only demonstrate a
0.7-log (base 10) reduction in either pathogenic bacteria or viruses (i.e.,
total enteroviruses or Salmonella spp.), whichever is the least susceptible
organism. If the sludge treatment process is a conventional process, then
indicator organism data can be used to demonstrate pathogen reduction. For
NP/LSA sludges, a conventional process must achieve a 1.4-log reduction in
either (1) fecal coliforms and fecal streptococci, o£ (2) fecal coliforms and
enterococci,
NP/LSA plants generally use treatment processes that do not depart
radically from conventional treatment. In such cases, these plants can also
use an average log density of less than 6.0 for fecal coliforms and fecal
streptococci in the treated sludge to demonstrate adequate viral and bacterial
pathogen reduction. This option is discussed in Conventional Processes
above. Since this approach involves half the sampling and analytical effort
of the indicator organism reduction approach, it is expected that most NP/LSA
plants will choose the log density option.
*No primary/long sludge age treatment processes are processes where
wastewater directly enters a secondary treatment system and sludge circulates
through the system (i.e., 'ages") for 20 or more days.
- 7 -
-------
Reduction of Vector Attractiveness. To be equivalent to PSRPs, a
process must reduce vector attractiveness to the same degree as properly
conducted anaerobic digestion. This requirement can be satisfied in several
ways depending on the type of sludge. Table 3 summarizes the equivalency
criteria for vector attractiveness.
V
PPRP Equivalency
As clarified by Whittington and Johnson, the PFRP technologies and
operating conditions listed in the regulations were selected to ensure that
pathogens (as represented by Salmonella spp., total enteroviruses, and
helminth ova) would be reduced to below the detection limits of the methods in
use in 1979 when the regulations were promulgated. These detection limits
were 3 most probable number (MPN)/100 ml sludge at 5% solids for Salmonella
spp., 1 plaque-forming unit {PFO)/100 ml sludge at 5% solids for total
enteroviruses, and 1 viable ovum/100 ml sludge at 5% solids for Ascaris spp.
In addition, PFRPs had to reduce vector attraction to the same extent as the
reduction achieved by good anaerobic digestion. The PEC adopted these listing
criteria as the basis the equivalency criteria.
One problem with the listing criteria for PFRP is that any particular
batch of sludge may contain few or no Salmonella spp-. or Ascaris ova prior to
treatment. Therefore, a finding of Salmonella spp. and Ascaris ova at the
levels specified for PFRP does not necessarily indicate that the treatment
process is capable of adequately destroying these organisms. For this reason,
the PEC specified that, to demonstrate PFRP equivalency, the untreated sludge
must contain 1,000 MPN salmonella spp./g TSS; 1,000 PFU total enteroviruses/g
TSS; and 100 viable Ascaris spp. ova/g TSS prior to treatment. If the
untreated sludge does not naturally contain these densities, it must be spiked
to achieve these levels.
- 8 -
-------
However, if it can be demonstrated that one organism is more susceptible
than others, it may be sufficient to test only for the least susceptible
organism. For example, viruses are much less sensitive to radiation than
bacteria and helminth ova. For radiation-based processes, it is sufficient to
demonstrate that the process reduces viruses to the required level.
The PEC recognized that there are some processes where a correlation has
been demonstrated between indicator organism reduction and reduction of
pathogenic viruses and bacteria (for example, thermal processes using
temperatures of sufficient degree-and duration to anticipate pathogen
destruction, e.g., 3 d at 53°C, 30 mins at 70°C). The PEC determined that, in
such cases, it may be possible to substitute indicator organism data for total
enterovirus and Salmonella spp. data. Processes that qualify for this
substitution must demonstrate the capability to reduce either fecal coliforms
and fecal streptococci o_£ fecal coliforms and enterococci to densities below
100/g total suspended solids.
Stockpiled Sludge
Some wastewater treatment plants have accumulated stockpiled sludge from
past treatment operations. Land application may be one option for disposal of
this sludge if it can be demonstrated that pathogen levels and attraction to
vectors have been adequately reduced. The problem with stockpiled sludge is
that no pretreatment measurements can be taken. Thus, it is not possible to
demonstrate PSRP equivalency by showing a reduction in pathogens or indicator
organisms. Demonstrating PSRP equivalency is an option only if it can be
shown that the treatment process used did not depart radically from
conventional treatment, or that there was a demonstrated correlation between
pathogenic bacteria and virus reduction and indicator organism reduction. In
- 9 -
-------
such cases, an average log density of fecal coliforms and fecal streptococci
of less than 6.0 in the treated sludge can be used to demonstrate appropriate
pathogen reduction, as described above.
In all other cases, stockpiled sludge must meet the PFRP criteria for
Salmonella spp., total enteroviruses, and Ascaris spp. ova. However, a
»
finding of Ascaris spp. ova at PFRP levels does not necessarily provide
confidence that all types of helminth ova were destroyed. It could simply
mean that the stockpiled sludge did not contain these ova prior to treatment.
To provide greater confidence that helminth ova have been destroyed, the PEC
specified that, for stockpiled sludge, applicants must also demonstrate that
Toxocara spp. ova and Trichuris trichiura ova have also been reduced to 1
viable ovum/100 ml sludge at 5% solids.
DEMONSTRATING EQUIVALENCY
Developers, owners, and operators or sludge treatment technologies can
demonstrate equivalency based on the criteria described above in different
ways: directly, by measuring microbe levels and vector attraction in sludge,
or indirectly by relating process parameters to reduction of pathogens and
vector attraction. The most appropriate choice, will.depend on the-particular
technology. Three basic approaches can be taken to demonstrate equivalency,
as described below.
Comparison to Operating Conditions for Existing PSRPs or PFRPs
If the process is similar to one of the listed PSRPs or PFRPs, it may be
possible to demonstrate equivalency by providing performance data showing that
the process consistently meets or exceeds the conditions specified in the
regulations.
- 10 -
-------
For example, a process that consistently produces a pH of 12 or greater
for 2 hrs of contact (the conditions required in the regulations for lime
stabilization) but uses a substance other than lime to raise pH would qualify
as a PSRP. In such cases, microbiological data would not be necessary.
Use of Literature Data to Demonstrate Adequacy of Operating Conditions
»
If scientific data from the literature establish a reliable relationship
between operating conditions (time, temperature, pH, etc.) and pathogen
reduction, well-maintained operating records verifying that the necessary
operating conditions were satisfied may be acceptable as a substitute for
actual microbiological sampling and analysis. In such cases, adequate
supporting operational and literature data must be included with the
equivalency application.
Process-specific Performance Data and Microbiologic Data
In all other cases, both performance data and microbiological data are
necessary to demonstrate process equivalency. Specifically, the following
information must be provided:
o A description of the various parameters (e.g., sludge characteristics,
process operating parameters, climatic factors, etc.) that influence
(1) the microbiological characteristics of the sludge product and (2)
the attractiveness of the product to vectors.
o Sampling and analytical data to demonstrate that the process has
reduced pathogens and vector attraction to the required levels.
o A discussion of the reliability of the treatment process in
consistently operating within the parameters necessary to achieve the
appropriate reductions.
Stockpiled Sludge
Stockpiled sludge from a past process can be found equivalent to PSRP or
PPRP. PFRP equivalency can be demonstrated by either (1) providing
microbiological data to show that pathogens are reduced to the PFRP limits
- 11 -
-------
throughout the stockpiled sludge (see PFRP Equivalency in previous section),
or (2) showing that the treatment process (including, if relevant, the storage
time) that produced the sludge was sufficient to reduce pathogens to the
required PFRP levels (for example, it may be sufficient to submit indicator
organism and parasite data for a sludge pile produced by a thermal process,
since data indicate a correlation between indicator organism reduction and
reduction of viruses and pathogenic bacteria when heat is used as the method
for disinfection).
PSRP equivalency can be demonstrated by providing microbiological data to
show that the average log density (no./g TSS) of fecal coliforms and fecal
streptococci is less than 6.0 throughout the stockpiled sludge and by
providing data to show that the treatment process either did not depart
radically from conventional treatment or was a process for which there is a
demonstrated correlation between pathogenic bacteria and virus reduction and
indicator organism reduction.
Reduction of vector attraction must also be demonstrated for both PSRP and
PFRP equivalency.
APPLYING FOR EQUIVALENCY
Who Should Apply?
All municipal wastewater sludge or sludge-derived products applied to land
must be treated by a PSRP or a PFRP. No demonstration of equivalency is
necessary for listed processes that consistently meet the specified operating
conditions (Tables 1 and 2). Processes that deviate in any way from the
specified operating conditions or novel processes or process combinations not
described in the regulations must reduce pathogens and vector attraction to an
extent equivalent to a PSRP or PFRP; anyone who markets, owns, or operates
- 12 -
-------
sucn a process may wish to obtain guidance on whether the process is
equivalent to either PSRPs or PPRPs before the sludge product is applied to
land.
preparing an Application
To obtain guidance on equivalency, an application must be submitted to the
Pathogen Equivalency Committee. EPA has recently published a document —
Control of Pathogens in Municipal Wastewater Sludge — which provides detailed
guidance on preparing applications for equivalency. There is no required
outline or form to fill out; however, each application must contain sufficient
information to enable the PEC to evaluate the equivalency of the process based
on the criteria discussed earlier. Suggested information includes a brief
fact sheet summarizing key information about the process; descriptions of the
process, the sludge product, the sampling and analytical techniques used, the
analytical results, measures taken for quality assurance, and the reduction of
vector attraction; and a rationale for why the process should be determined
PSRP or PFRP.
Data quality is an important factor in EPA's equivalency determination.
Data quality can be assured by using accepted, state-of-the-art sampling and
analytical techniques; obtaining samples that are representative of the
expected variation in sludge quality; developing and following quality
assurance procedures for sampling; and using an independent, experienced
laboratory to perform the analysis. Applicants with questions about how to
obtain the necessary microbiological data may submit a work plan to the PEC
describing the proposed approach to sampling and analysis of the sludge
product. The PEC or a designated representative will review the plan and
indicate whether the approach would be expected to yield acceptable and
complete data.
- 13 -
-------
Since processes differ widely in their nature, effects, and processing
sequences, the experimental plan to demonstrate that a process meets the
requirements for PSRP or PFRP must be tailored to the process. Field
verification and documentation by independent or third-party investigators is
desirable.
»
The "Application Process
The first point of contact in the application process for obtaining
guidance on equivalency is the Regional Sludge Coordinator (RSC) in the EPA
Water Management Division of the EPA regional office, or the State Sludge
Coordinator (SSC) in the state environmental agency that regulates land
application of sludge. Either the SSC or the RSC can be contacted to answer
questions. Applications are submitted to the RSC who solicits comments on the
application from other regional personnel and the SSC. The RSC then forwards
the application and any comments to the PEC. The PEC forwards a copy of the
application to the EPA Office of Water Enforcement and Permits and the EPA
Office of Water Regulations and standards (OWRS).
The RSC and the SSC may participate with the PEC in the equivalency
evaluation if they are familiar with the process (e.g., through site visits,
research activities, etc.). For each application, the PEC evaluates the study
design, data accuracy, and the adequacy of the results for supporting the
conclusions drawn in light of the current state of knowledge concerning sludge
treatment and pathogen reduction.
The PEC documents its recommendation concerning the application and
includes any supporting information. A copy of the final recommendation is
forwarded to OWRS for review and approval. OWRS forwards the PEC's
conclusions to the RSC, who forwards a copy to the SSC. The SSC forwards a
copy to the applicant.
- 14 -
-------
The Equivalency Determination
The committee recommends one of five decisions about the process or
process sequence:
o It is equivalent to PFRP.
o It is not equivalent to PFRP.
»
'o It is equivalent to PSRP.
o It is not equivalent to PSRP.
o Additional data or other information are needed.
Most processes considered for equivalency have been found equivalent on a
site-specific basis only. That is, the equivalency applies only to that
particular operation run at that location under the conditions specified. For
site-specific PSRP or PFRP determinations, equivalency cannot be assumed for
the same process performed at a different location, or for any modification of
the process.
The PEC has considered applications for national equivalency status. To
show national equivalency, the applicant must demonstrate that the process
will produce the desired reductions in pathogens and vector attraction under
the variety of conditions that may be encountered at different locations in
the country. Processes affected by local climatic conditions or that use
materials whose properties may vary significantly from one part of the country
to another are unlikely to be found equivalent on a national basis.
If the members of the PEC determine, based on the information submitted,
that a process is equivalent to PSRP or PFRP, they specify the operating
parameters and any other conditions critical to adequate disinfection and
reduction of vector attraction. These conditions are communicated to the
applicant in the equivalency determination letter. The process then is
- 15 -
-------
considered equivalent to PSRP or PPRP only when operated under the specified
conditions.
If the Committee determines that a process is not equivalent, the
committee will provide an explanation for this recommendation. If additional
data are needed, the committee will describe what those data are and work with
»
the Applicant, if necessary, to ensure that the appropriate data are gathered
in an acceptable manner. The committee then will review the revised
application when the additional data are submitted.
Table 4 lists those processes that were found by the PEC to be equivalent
to PSRP or PFRP during its first 2 yrs of operation.
FUTURE REGULATORY DIRECTIONS
The EPA is currently revising its technical regulations for all municipal
sludge use and disposal practices, including land application and distribution
9
and marketing of sludge products. The new regulations were proposed by EPA
on February 6, 1989. The proposed land application regulations incorporate
much of the knowledge and experience that has been gained in implementing 40
CFR Part 257. Thus there are many similarities between the proposed
regulations and the guidance on equivalency that has been developed by the
PEC.
The proposed 503 land application regulations are performance-based: They
specify reductions and densities of pathogens that must be achieved in sludges
before they are applied to land. The proposed regulations define three
classes of sludge: Class A, Class B, and Class C. There is a close
correspondence between the requirements for Class A sludges and the criteria
for PRFP equivalency, Class B sludges and the criteria for PSRP equivalency,
- 16 -
-------
and Class C sludges and the PSRP criteria for no primary/long sludge age
(NP/LSA) treatment. Like 257, the proposed 503 regulations also specify some
restrictions concerning access to and use of land where sludge has been
applied, depending on sludge quality.
By proposing performance-based standards rather than continue with the
»
tecHnology-based standards of 257 (PSRPs and PPRPs), EPA has essentially
incorporated the equivalency requirement as an implicit rather than explicit
component of the regulations. The new regulations replace the equivalency
requirement of 257 with an explicit statement of the performance requirements
that all sludge treatment techologies must meet.
Land application will continue to be governed by the 40 CFR Part 257
regulations, as described here, until the final 503 regulations are
promulgated. The proposed 503 regulations will undergo changes in response to
comments, but it is likely that the land application regulations will continue
to be performance-based, and will continue to resemble the equivalency
criteria described here. Final 503 regulations are expected to be promulgated
in October 1991. It is expected that the PEC will provide guidance in
interpreting and implementing the new land application regulations.
ACKNOWLEDGEMENTS
Authors
The U.S. EPA's Pathogen Equivalency Committee consists of Robert Bastian,
Chief, Technical Review Section, Performance Assurance Branch, Office of
Municipal Pollution Control, U.S. EPA, Washington, DC; Joseph Farrell, Chief,
Sludge Technology Section, Risk Reduction Engineering Laboratory, U.S. EPA,
Cincinnati, Ohio; Larry Fradkin, Senior Environmental Engineer, Office of the
- 17 -
-------
Senior Official for Research and Development, U.S. EPA, Cincinnati, Ohio;
Walter Jakubowski, Chief, Parasitology and Immunology Branch, Microbiology
Research Division, Environmental Monitoring Systems Laboratory, U.S. EPA,
Cincinnati, Ohio; James E. Smith, Jr., Senior Environmental Engineer, Center
for Environmental Research Information, U.S. EPA, Cincinnati, Ohio; and Albert
Vencrsa, Research Microbiologist and Chairman of the PEC, Risk Reduction
Engineering Laboratory, U.S. EPA, Cincinnati, Ohio. Jan Connery, Vice
President of Eastern Research Group, Inc. prepared this paper under the
Committee's direction and from information and data supplied by the Committee.
REFERENCES
1. EPA, 'Guidance for Writing Case-by-Case Permit Requirements for Municipal
Sewage Sludge." Permits Division, U.S. EPA Office of Water Enforcement
and Permits, Washington, DC (1989).
2. EPA, 'Control of Pathogens in Municipal Wastewater Sludge.' Center for
Environmental Research Information, U.S. EPA, Cincinnati, Ohio (1989).
3. Whittington, W.A., and Johnson, E., 'Application of 40 CFR Part 257
Regulations to Pathogen Reduction Preceding Land Application of Sewage
Sludge or Septic Tank Pumpings.' Memorandum to EPA Water Division
Directors. U.S. EPA Office of Municipal Pollution Control (Nov., 1985).
4. Farrell, J.B., Stern, G., and Venosa, A.D., 'Microbial Destructions
Achieved by Full-scale Anaerobic Digestion." Workshop on Control of
Sludge Pathogens. Series IV. Water Pollution Control Federation,
Alexandria, Virginia (1985).
- 18 -
-------
5. Farrah, S.R., Bitton, G., and Zan, S.G., "Inactivation of Enteric Pathogens
During Aerobic Digestion of Wastewater Sludge." EPA Pub. No.
EPA/600/2-86/047. U.S. EPA Water Engineering Research Laboratory,
Cincinnati, OH. NTIS Publication No. PB86-183084/A5. National
Technical Information Service, Springfield, Virginia (1986).
»
6. 'EPA, "Technical Support Document for Pathogen Reduction in Sewage Sludge."
Publication no. PB 89-136618. National Technical Information Service,
Springfield, Virginia (1989).
7. Farrell, J.B., "Evaluating Performance of Processes for PFRP." Memorandum
to Larry Fradkin, Chairman, Pathogen Equivalency Committee. O.S. EPA
Risk Reduction Environmental Laboratory, Cincinnati, Ohio (Sept., 1988).
8. Farrell, J.B., Salotto, G.V., and Venosa, A.D., "Reduction in Bacterial
Densities of Wastewater Solids by Three Secondary Treatment
Processes." Submitted to J. Water Poll. Control Fed, for publication
(1989).
9. EPA, "Standards for the Disposal of Sewage Sludge; Proposed Rule."
Federal Register 54,23,5746-5902 (1989).
2238W
- 19 -
-------
TABLE 1
REGULATORY DEFINITION OP PROCESSES TO SIGNIFICANTLY
RBDDCB PATHOGENS (PSRPs)
Aerobic Digestion: The process is conducted by agitating sludge with air or
oxyg*en to maintain aerobic conditions at residence times ranging from 60 d at
15°C to 40 d at 20°Cf with a volatile solids reduction of at least 38%.
Air Drying: Liquid sludge is allowed to drain and/or dry on underdrained
sand beds, or on paved or unpaved basins in which the sludge depth is a
maximum of 23 cm (9 in.). A minimum of 3 months is needed, for 2 months of
which temperatures average on a daily basis above 0°C.
Anaerobic Digestion: The process is conducted in the absence of air at
residence times ranging from 60 d at 20°C to 15 d at 35°C to 55°C, with a
volatile solids reduction of at least 38%.
Composting: Using the within-vessel, static aerated pile, or windrow
composting methods, the solid waste is maintained at minimum operating
conditions of 40°C for 5 d. For 4 hrs during this period the temperature
exceeds 55°C.
Line Stabilization: Sufficient lime is added to produce a pH of 12 after 2
hrs of contact.
Other Methods: Other methods or operating conditions may be acceptable if
pathogens and vector attraction of the waste (volatile solids) are reduced to
an extent equivalent to the reduction achieved by any of the above methods.
Source: 40 CFR 257, Appendix II.
2238W
- 20 -
-------
TABLE 2
REGULATORY DEFINITION OP PROCESSES TO FURTHER REDUCE PATHOGENS (PPRPs)
Composting: Using the within-vessel composting method, the solid waste is
maintained at operating conditions of 55°C or greater for 3 d. Using the
static aerated pile composting method, the solid waste is maintained at
operating conditions of 55°C or greater for 3 d. Using the windrow composting
method, the solid waste attains a temperature of 55°c or greater for at least
15 d during the composting period. Also, during the high temperature period,
there will be a minimum of five turnings of the windrow.
Heat Drying: Dewatered sludge cake is dried by direct or indirect contact
with hot gases, and moisture content is reduced to 10% or lower. Sludge
particles reach temperatures well in excess of 80°C, or the wet bulb
temperature of the gas stream in contact with the sludge at the point where it
leaves the dryer is in excess of 80°C.
Heat Treatoent: Liquid sludge is heated to temperatures of 180°C for 30
minutes.
Thermophilic Aerobic Digestion: Liquid sludge is agitated with air or
oxygen to maintain aerobic conditions at residence times of 10 d at 55°C to
60°C, with a volatile solids reduction of at least 38%.
Other Methods: Other methods or operating conditions may be acceptable if
pathogens and vector attraction of the waste (volatile solids) are reduced to
an extent equivalent to the reduction achieved by any of the above methods.
Any of the processes listed below, if added to a PSRP, further reduce
pathogens.
Beta Ray Irradiation: Sludge is irradiated with beta rays from an
accelerator at dosages of at least 1.0 Mrad at room temperature (ca. 20°C).
Gamma Ray Irradiation: Sludge is irradiated with gamma rays from certain
isotopes, such as 60Cobalt and 137Cesium, at dosages of at least 1.0 Mrad
at room temperature (ca. 20°C).
Pasteurization: Sludge is maintained for at least 30 mins at a minimum
temperature of 70°C.
Other Methods: Other methods or operating conditions may be acceptable if
pathogens are reduced to an extent equivalent to the reduction achieved by any
of the above add-on methods.
Source: 40 CFR 257, Appendix II.
2238W
- 21 -
-------
TABLE 3
REDUCTION IN VECTOR ATTRACTIVENESS;
CRITERIA FOR DEMONSTRATING EQUIVALENCY
TYPE OP SLUDGE
CRITERIA
All types
Sludges from aerobic processes
(aerobic digestion or extended
aeration)
Anaerobic sludges
Sludges that contain
no raw primary sludge
High pH sludges
Stockpiled sludge
Reduction of volatile solids content
of the sludge by at least 38% during
treatment.
Treated sludge has an oxygen intake of
less than 1 mg oxygen/hr.g TSS as
demonstrated by the Specific Oxygen
Uptake Rate (SOUR) test at 20°C.
Volatile solids reduction in treated
sludge after 40 d additional
batch mesophilic digestion is less
than 15%.
Total suspended solids content of
treated sludge is 75% or greater
and remains at this level until
the point of land application.
Treated sludge maintains a pH
of 11.5 or greater up to the
time of land application.
Lack of odor throughout the sludge pile.
2238W
- 22 -
-------
TABLE 4
PROCESSES DETERMINED TO BE EQUIVALENT TO PSRP OR PFRP
OPERATOR
PROCESS DESCRIPTION
STATUS
Town of Telluride,
Colorado
Combination oxidation ditch, aerated storage, and
drying process. Sludge is treated in an oxidation
ditch for at least 26 d and then stored in an
aerated holding tank for up to a week. Following
dewatering to 18% solids, the sludge is dried on a
paved surface to a depth of 61 cm. The sludge is
turned over during drying. After drying to 30% solids,
the sludge is. stockpiled prior to land application.
Together, the drying and stockpiling steps take
approximately 1 yr. To ensure that PSPP requirements
are met, the stockpiling period must include one full
summer season.
PSRP
Comprehensive Materials
Nanaqpuient, Inc.,
Houston, Texas
N-Viro Energy Systems Ltd.,
Toledo, Ohio
Use of cement kiln dust (instead of lime) to treat sludge PSRP
by raising sludge pH to at least 12 after 2 hrs of
contact. Dewatered sludge is mixed with cement kiln dust
in an enclosed system and then hauled off for land
application.
Use of cement kiln dust and lime kiln dust (instead of National
lime) to treat sludge by raising the pH. Sufficient PSRP
lime or kiln dust is added to sludge to produce a pH
of 12 for at least 12 hrs of contact.
Public Works Department,
Everett, Washington
Anaerobic digestion of lagooned sludge. Suspended
solids had accumulated in a 12-ha aerated lagoon
that had been used to aerate wastewater. The lengthy
detention time in the lagoon (up to 15 yrs) resulted
in a level of treatment exceeding that provided by
conventional anaerobic digestion. The percentage of
fresh or relatively unstabilized sludge was very small
compared to the rest of the accumulation (probably much
less than 1% of the whole).
PSRP
Haikey Creek Wastewater
Treatment Plant, Tulsa,
Oklahoma
Oxidation ditch treatment plus storage. Sludge is
processed in aeration basins followed by storage in
aerated sludge holding tanks. The total sludge aeration
time is greater than the aerobic digestion operating
conditions specified in the federal regulations of 40 d
at 20°C to 60 d at 15°C. The oxidation ditch
sludge is then stored in batches for at least 45 d
in an unaerated condition or 30 d under aerated
conditions.
PSPP
Ned K. Burleson &
Associates, Inc.,
Fort Worth, Texas
Aerobic digestion for
at 35°C.
20 d at 30°C or 15 d
PSRP
Scarborough Sanitary
District, Scarborough,
Maine
Mount Holly Sewage
Authority, Mount
Holly, New Jersey
.Static pile aerated "composting" operation that uses
Dy ash from a paper company as a bulking agent. The
procesri creates pile temperatures of 60° to 70°C
within 24 hrs and maintains these temperatures
Tor up to 14 d. The material is stockpiled after
7 to 14 d of "composting" and then marketed.
Zimpro 3 L/s low-pressure wet air oxidation process.
The process involves heating raw primary sludge to
]77° to 204°C in a reaction vessel under pressures
of 250 to 400 psig for 15 to 30 mins. Small
volumes of air are introduced into the process to
oxidize the organic solids.
PFRP
PFRP
-------
TABLE 4 (Cent.)
PROCESSES DETERMINED TO BE EQUIVALENT TO PSRP OR PFRP
OPERATOR
PROCESS DESCRIPTION
STATUS
N-Viro Energy Systems
Ltd., Toledo, Ohio
V
Miami-Dade Water and
Sewer Authority,
Miami, Florida
Advanced alkaline stabilization with subsequent National
accelerated drying. PFRP
o Alternative 1: Fine alkaline materials (cement
kiln dust, lime kiln dust, auicklime fines,
pulverized lime, or hydrated lime) are uniformly
mixed by mechanical or aeration mixing into
liquid or dewatered sludge to raise the pH to
greater than 12 for 7 d. If the resulting sludge
is liquid, it is dewatered. The stabilized
sludge cake is then air dried (while pH remains
above 12 for at least 7 d) for at least 30 d and
until the cake is at least 65% solids. A solids
concentration of at least 60% is achieved before
the pH drops below 12. The mean temperature of
the air surrounding the pile is above 5"C for the
first 7 d.
o Alternative 2: Fine alkaline materials (cement
kiln dust, lime kiln dust, quicklime fines,
pulverized lime, or hydrated lime) are uniformly
mixed by mechanical or aeration mixing into
liquid or dewatered sludge to raise the pH to
greater than 12 for at least 72 hrs. If the
resulting sludge is liquid, it is dewatered. The
sludge cake is then heated, while the pH exceeds
12, using exothermic reactions or other thermal
processes to achieve temperatures of at least
52"C throughout the sludge for at least 12 hrs.
The stabilized sludge is then air dried (while pH
remains above 12 for at least 3d) to at least
50% solids.
Anaerobic digestion followed by solar drying. Sludge Condi-
is processed by anaerobic digestion in two well-mixed tional
digesters operating in series 'in a 'temperature range of PFRP
35" to 37-C . Total residence time is 30 d.
The sludge is then centrifuged to produce a cake
of between 15 to 25% solids. The sludge cake is dried
for 30 d on a paved bed at a depth of no more than
46 cm. Within 8 d of the start of drying, the sludge
is turned over at least once every other day until
the sludge reaches a solids content of greater than 70%.
The PFRP approval was conditional on the microbiological
quality of t.he product.
2240W
-------
FECAL PATHOGEN CONTROL DURING COMPOSTING
by
Joseph B. Parrel 1
ABSTRACT
The need for pathogen control in compost produced from Municipal
wastewater sludge is self-evident. Just how much control is needed is
much harder to establish. Objectives of EPA's present and proposed
regulations are that viruses, helminth eggs, and pathogenic bacteria are
below detection limits in the final product and vector attraction is
adequately reduced. The present rule, promulgated in 1979, does not
require monitoring of microorganism densities. It is only necessary to
operate the composting process at conditions that produce the desired
microorganism destruction established during previous demonstrations of
the process. The new regulation, to take effect in 1992, will require
periodic monitoring of microbiological quality as well as operation at
conditions that have been demonstrated to produce satisfactory pathogen
reductions. The requirement to monitor composted product has been
introduced because data have shown that salmonellae are frequently
detected in composted sludge. Viruses and viable helminth eggs are rarely
found. As an alternative to monitoring for salmonellae, the new
regulation will allow monitoring of fecal coliform density which is not to
exceed 1000/g dry solids. The development of this standard from data
collected on Los Angeles County and Philadelphia composts is presented.
Means for modifying composting procedures to improve pathogen reductions
are proposed. There is an important need for research and field
demonstration of ways to produce adequate pathogen reduction and limit
regrowth of bacterial pathogens in compost.
Keywords: Pathogens, salmonellae, enteroviruses,
helminths, sewage sludge, composting,
Federal regulations
Jy fa Cfr/i ffak C'
-------
INTRODUCTION
The need for pathogen control in composting depends on the substance
that is being composted. When agricultural products or yard wastes are
composted, there is little need for concern about fecal pathogens unless
pathogen-contaminated wastes such as sewage sludge or animal manures have
been added to the process to provide nutrient. The fecal matter may
contain viruses, bacterial pathogens, protozoan cysts, and animal and
human helminth eggs, which cause a variety of human diseases. If solid
wastes collected from households are composted, there is reason for
concern, because fecal wastes from pets and from infant's disposable
diapers are often present. Obviously, the concern will be even greater if
sewage sludge is composted, because the bulk of the sludge is comprised of
fecal wastes. Untreated sludge has the greatest potential for containing
pathogens. Most of the time when sludge 1s to be composted, the sludge is
stabilized by lime treatment or anaerobic digestion before the composting
step. Pathogen densities are substantially reduced by this pretreatment,
but enough pathogens remain to be a serious concern.
The significance of the problem is related to how much composting of
wastes that contain fecal matter is being carried out. In the sewage
sludge area, there has been a vigorous growth in processing approaches and
numbers of plants in recent years. Goldstein and Riggle (1990) reported
that there were 133 operating plants, 23 under construction, and 80 in
earlier stages. These numbers are no doubt considerably higher now. At
most of these plants, the composted product is being made available to the
public which assumes it to be non-infectious. For solid waste, there are
not many plants in operation but several are anticipated to be constructed
in coming years. Even with fewer plants, the problem will be substantial
because plants are expected to be larger than the sewage sludge plants
simply because there is much more potential raw material in a given
collection area. The average sludge production per capita is about 0.15
pound per day whereas the average amount of compostable solid waste is
about 10 times that amount.
THE 1979 FEDERAL REGULATION
In September of 1979, the U. S. EPA, as required by the Resource
Conservation and Recovery Act (RCRA) and the Clean Water Act (CWA),
published regulations (Federal Register, 1979) controlling the use of
sewage sludge on the land. To protect the public from the risk of
disease, two requirements were developed for sludge that was to be applied
to the land. The sludge had to be treated by either a "Process to
Significantly Reduce Pathogens (PSRP)" or a "Process to Further Reduce
Pathogens (PFRP)". The PSRPs did not eliminate all pathogens so they had
to be used in combination with access and crop restrictions. The PFRPs
eliminated all pathogens of concern and thus required no access or use
restrictions. Both types of processes also had to change the sludge so
that vectors of disease were not attracted to the land use site. The
regulation specified a number of qualified processes of each type that
accomplished both of these goals. Essentially, EPA examined the available
sludge stabilization processes and divided them into two groups - those
that reduced pathogens to below detection limits and those that only
reduced animal virus and pathogenic bacterial densities. All of these
-------
processes stabilized sludge, that is, made it less subject to putrefaction
and less malodorous. The regulation redirected this objective to the dual
goals of pathogen reduction or elimination and reduction of vector
attraction, which are more significant to public health than aesthetics.
Composting of sludge was among the qualified processes, and was
classified in both the PSRP and PFRP categories, depending on how the
process was carried out. All of the processes were defined not by
microbiological goals, but by technology standards. EPA had a clear
understanding of the microbiological goals (U. S. EPA, 1989), but they
were not expressed in the regulation or its preamble. The goals for PSRP
processes were to reduce bacterial pathogens and enteric viruses by at
least one log; for PFRPS, the goal was to reduce salmonellae to less than
3 MPN (most probable number) per 4 grams of sludge solids, viruses to less
than 1 PFU (plaque forming unit) per 4 grams of sludge solids, and
helminth ova to less than 1 per 4 grams of sludge solids.
The technological standards specified in the regulation for
composting to meet the PSRP and PFRP requirements were as follows:
For PSRP, "Using the within-vessel, static aerated pile, or
windrow composting methods, the solid waste is maintained at
minimum operating conditions of 40 C for five days. For four
hours during this period the temperature exceeds 55°C."
For PFRP, "For windrow composting, the sludge must attain a
temperature of 55 C or greater for at least 15 days during the
composting period. In addition, during the high temperature
period, the windrow must be turned at least five times. If the
static aerated pile or the within-vessel method is used, the
sludge must be maintained at operating temperature of 55 C or
greater for three days."
Those familiar with composting will notice that these conditions are unlikely
to produce a stabilized compost. They were set with the pathogen reduction
needs in mind. It is doubtful that material composted for such short periods
of time would be satisfactorily reduced in vector attraction. Despite the
deficiency of the regulation on this point, no difficulty has been reported.
To the author's knowledge, no compost producer is supplying a compost
processed to these minimal conditions. Societal demand for a plant that is
not a source of objectionable odors and customer demand for a well-stabil ized
compost evidently overrule any latent desire of an operator to just meet
minimum standards.
PATHOGEN OCCURRENCE IN SLUDGE COMPOST
The thermal conditions specified in the PSRP and PFRP descriptions are
adequate to produce the desired microbial reductions if all of the material
composted is subjected to them. However, if the conditions are not met and
sufficient nutrient remains in the compost, regrowth of bacterial pathogens is
possible. Subsequent handling of the compost may introduce contamination, and
regrowth by contaminating bacterial pathogens can occur if nutrient supply
permits. A visit to many composting facilities will convince the visitor that
opportunities exist for contamination and that some of the processed material
-------
may not have reached the desired temperature-time requirements. It might be
expected that, out of concern for disease risk to customers, plant managers
would have the microbial condition of the product checked at least
occasionally. Because the regulation does not require it, there has been
little reporting of data from composting facilities about the microbiological
quality of their products. Consequently, the EPA sponsored a survey to
determine whether sludge composts and similar sludge-derived products made
available to homeowners were free of fecal pathogens.
The research was conducted at the Los Angeles County Sanitation
Districts (LACSD) (Yanko 1988). Representative composting facilities were
selected from eight of the ten Federal Regions. Weekly samples were taken of
several products at a large western and a large eastern facility, for a total
of 350 samples. Bimonthly samples were taken at twelve other sites (13
sampling locations) for a year. Microorganisms selected for analysis included
total enteric viruses, total parasites, Ascaris ova, Yersinia. Camovlobacter.
Salmonella, total enteric bacteria, enterotoxigenic Escherichia coli.
bacteriophage, total and thermophilic fungi, anaerobic and aerobic plate
counts, fecal streptococci, and total and fecal coliforms.
Yanko (1988) concluded that there is no significant health hazard
associated with the composted products tested from Camovlobacter. enteric
viruses, or parasitic helminth ova. Yersinia occurred in high densities in
some samples, but, based on a small number of tests, it appeared to be
avirulent. The fungus Asperqillus fumiqatus was detected in most samples, but
it occurred at highest densities in composts from static pile facilities.
Health risks from this fungus have been addressed elsewhere (Clark et al.,
1984). Isolations of salmonellae were "reasonably frequent." Yanko noted
that bacterial and fungal densities were considerably higher in composts to
which amendments were added than in the base composts, suggesting a nutrient-
related regrowth phenomenon. Based on regression analysis, Yanko suggested
that total or fecal coliforms or fecal streptococci may be suitable indicators
for monitoring.
The frequency of detection of salmonellae is shown in Table 1.
Salmonellae were detected in 173 of 428 samples, or 40% of the time. Of the 19
sampling locations (15 sites), salmonellae were detected in more than 10% of
the samples at 11 locations. At five locations, they were detected in more
than 20% of the samples. When detected, the median density was 100 MPN per
gram, and density exceeded 3700 MPN per gram for 10% of the detections. The
relatively high fraction of salmonellae detections suggest the need to
periodically check the product for microbiological quality. For certain of
the locations (IX-A-3 to 6), other materials such as wood chips and rice hulls
were added to the composts. They may have introduced fecal bacteria and
nutrients to the mixture, which may have contributed to regrowth of
salmonellae and fecal indicator organisms.
Yanko's results show a need to check for bacteriological quality but not
for viruses and helminths. Evidently adequate temperatures are reached at
least at one point in the composting operation to reduce these organisms to
below detectable limits. Since they have no capacity to regrow, they are no
longer a problem. This removes a major impediment to microbiological
monitoring since tests for these organisms are expensive and difficult to
perform properly.
-------
The bacteria of primary concern are salmonellae, but testing is
relatively expensive and requires higher skills than, for example, the fecal
indicator organism tests commonly run at wastewater treatment plants.
(Shigellae may be of equal importance, but data are lacking - for now, we must
assume that shigellae densities correlate with salmonellae). Since
salmonellae are occasionally not present in the sludge (Farrell et al., 1990),
testing of the incoming sludge as well as the composted product is advisable.
Determining that salmonellae are absent in the compost does not prove the
ability of the process to kill salmonellae when there are none in the incoming
sludge. For these reasons, an alternative to salmonellae testing is
desirable, preferably a simple test like the fecal indicator tests.
PREDICTION OF SALMONELLAE DETECTION FROM FECAL COLIFORM DENSITY
Yanko showed by regression analysis that densities of the fecal
indicators (total coliform, fecal coliform (FC) and fecal streptococcus)
correlated well with salmonellae density. These correlations cannot be used
for predictive purposes at the low densities because salmonellae frequently
were not detected at low fecal indicator densities. Yanko also obtained a
good correlation by plotting the fraction of measurements in which salmonellae
were not detected versus indicator organism density. He grouped all of his
data from the sites that were monitored weekly into 1 log FC density intervals
starting at 0.699 -1.699, etc., and determined the fraction (f) of times
salmonellae were detected in each interval. He then plotted this fraction
against the average log FC density of the interval and drew the best straight
line through the data points (excluding the intervals below log FC density of
2.699 where f was zero). Extrapolating this line to f«0 (zero probability
level) gave an FC density of 48 MPN per gram. Examination of this curve
showed that it was not linear but had a definite S shape. This curve has been
re-plotted (Figure 1) using a Probit scale (arithmetic probability scale) on
the ordinate versus log fecal coliform density. The triangles of Figure 1 are
points developed from Yanko's data set but the 1 log FC brackets start at 0.
These data are presented in Table 2. This method of presentation sacrifices
the experimentally determined points where f=0 and f=l, but the S-curve
obtained with the linear ordinate is now a good straight line. Interpolation
and limited extrapolation can be done with more confidence.
The curve indicates that at log FC equal to 3, the likelihood of
salmonellae detection is about 9%. If we assume that any log FC measurement
from 0 to 3 is equally probable (this was approximately the case for this data
set - see Table 2), the likelihood of detection of salmonellae in a sample can
be shown to be about 2% or 1 in 50. Based on these data, the log FC density
of 3 (1000 MPN per gram of dry solids) seems to be a reasonable maximum
allowable density for a composting facility to be assured of rare occurrence
of salmonellae in its product.
The data set used in Figure 1 groups together data from two distinctly
different composting processes. The products were manufactured with different
bulking agents and in some cases materials such as wood chips or rice hulls
were added after composting. There is substantial overlap of the data, but
for some of the runs the data are concentrated in the lower log FC range and
for others the data are concentrated in the high log FC range. An attempt was
made to construct f versus log FC curves for the individual products to see if
they resembled each other. Unfortunately, there were too few points for any
-------
one product so the data points were too scattered to allow representation by
smooth curves. A check was made of the probability of obtaining the observed
frequency of detection for each FC bracket for every product, using the
binomial theorem and the predicted fraction of detection obtained from Figure
1. The likelihood of obtaining the observed fraction of detection was less
than 5% twice out of 27 tests, which is slightly more than expected. The check
does not indicate that the frequency of detection for any of the products
could not reasonably be represented by Figure 1.
The data for the bimonthly sites are shown in Table 3. The results for
the various sites have been pooled together into two groups. All plants at
which the sludge was pre-treated with lime are in one group. In the other
group, the sludges were pre-treated by anaerobic digestion or were not pre-
treated. Data from Site VII-A-2 was not included because the stockpile from
which the samples were drawn was over 1.5 years old. The pooling was
necessary because there were only 6 samples taken at each site. The limed
sludges showed no salmonellae detections and only 2 out of 18 measurements
showed log FC densities above 3. For the un-limed sludges, salmonellae
detections were frequent and fecal coliform densities generally much higher
than for the limed sludges. There was one detection of salmonellae at a log FC
value between 2 and 3. The FC density was 230 MPN per gram and the Salmonella
sp. density was 1 MPN per gram. The compost sample was from a site where the
composting process is described by Yanko as proprietary, with additives to the
product. This is the only log FC measurement below 3 with a salmonellae
detection in all of Yanko's data. Considering the low value of the
salmonellae density and the unknown nature of the process, it is difficult to
become overly concerned about this single detection at log FC below 3.
The observed frequencies of detection for the bimonthly sites for the
various log FC intervals were checked for the un-limed sludges to see if they
were reasonable compared to the values obtained from Figure 1. The data for
the limed sludges appeared to give atypically low salmonellae detections and
are not included in the comparison. Using the probability from Figure 1 in
the binomial theorem, the probability of obtaining by chance the observed
frequency was less than 5% only for the 4 - <5 log FC interval. Considering
this and the fact that the f-values for 3 out of 4 brackets of the bimonthly
sites are lower than the Figure 1 values indicates that Figure 1 predicts a
higher rate of salmonellae detection than was actually the case for the
bimonthly sites. More data are needed to verify this suggestion; however,
Figure 1 can be used to give a conservative estimate of the probability of
detection of salmonellae for these sites.
Although it appears safe to say that the detection of salmonellae will
be "rare" when the log fecal coliform density is less than 3, this is not
enough guidance for an operating plant. If for example, one obtains three
successive log FC densities of 2.9, there is evidently more risk of the
presence on salmonellae in the product than if there were three successive
measurements of 1.0. One approach that appears satisfactory is to calculate a
running average of risk for the last half-dozen FC monitoring results, using
Figure 1 to determine the individual risk factors. When the risk exceeds some
predetermined value, such as 0.06, corrective action should commence. When a
value of 0.08 is exceeded, then drastic action such as recycling of the
product for reprocessing should be considered. Plant managers should
-------
institute quality control procedures and use helpful techniques such as
Shewhart or cumulative sum charts (Davies et al., 1976) to track
microbiological performance of their processes.
THE 1992 REGULATION
Sewage sludge use and disposal in the United States is now to be
regulated under the Clean Water Act of 1977 as amended by the Water Quality
Act of 1987. Use and disposal standards are required by Section 405d of the
act, and will be incorporated as 40 CFR Part 503 of the Code of Federal
Regulations. Publication of the new regulation is expected in July 1992. All
final use and disposal methods will be covered by the new regulation, but not
co-disposal with other wastes. For example, sludge disposal in landfills will
be covered for sludge-only landfills, but co-disposal of sludge and municipal
solid waste is covered by another regulation promulgated under RCRA (Federal
Register, 1991). Composting of sludge and solid waste together or of solid
waste alone is not covered by any Federal regulation. If a substantial
portion of the solid waste stream eventually is processed by composting,
regulation will inevitably come about.
The pathogen portion of the 1992 sludge regulation is still undergoing
revisions although it is probably close to its final form. It resembles the
1979 regulation in its general approach although there are some important
differences. There are two classes of treatment for sludges. Class A has
replaced the PFRP classification, and Class B has replaced PSRP. The
requirement that a single named processing step accomplish both pathogen
reduction and vector attraction reduction has been changed. For pathogen
reduction, performance standards have been established for Class A and Class B
treatment, replacing a long list of named processes. Both Class A and Class B
treatment require vector attraction reduction. It is still necessary to use
named processes for vector attraction reduction, although in almost all cases
the requirement is to meet a performance goal rather than adhere to
technological process descriptions identified in the regulation.
The performance-based pathogen reduction standard for Class B treatment
is simple. The sludge or sludge product must have a fecal coliform density
less than 2,000,000 per gram of sludge solids. Periodic monitoring will be
required to demonstrate conformity with the standard. Access to sludge-
application sites and types of crops grown are restricted, because some
pathogenic organisms are present in sludge after Class B treatment. Since
facilities that convert sludge into compost usually do not wish to have their
product subject to the restrictions on disposal required for Class B sludges,
the rationale supporting the selection of Class B pathogen reduction
requirements is of lesser interest and will not be discussed. This subject
will be covered in the Technical Support Document for the pathogen part of the
rule, which will be released when the rule is published in the Federal
Register.
To meet Class A pathogen-reduction requirements, a sludge or sludge
product must be reduced in pathogen densities to the same degree as in the
1979 regulation, that is, to less than 1 PFU of enteric viruses per 4 grams of
sludge solids, less than 1 viable helminth ova per 4 grams of sludge solids,
and less than 3 MPN of salmonellae per 4 grams of sludge or sludge product
solids. The sludge or sludge products are monitored to prove that they are in
-------
conformity with the requirements. Since viruses and helminths are sometimes
not present in sludge, the requirement must be met when these microorganisms
are present in the incoming sludge.
For processes where the mechanism of destruction is elevated temperature
held for a sufficient time, it is possible to monitor performance by observing
whether there is a sufficient reduction in fecal coliform densities. The
critical density is 1000 MPN fecal coliform per gram of solids. This choice
is based on the foregoing analysis of Yanko's research. This requirement has
caused consternation among staff at some sludge composting facilities who feel
that they cannot meet this requirement and also feel that their product is
safe. A provision has been introduced into the regulation that should provide
them with relief. If they choose, they can monitor for salmonellae. If
salmonellae are absent in the final product and were present in the incoming
unprocessed sludge, the monitoring requirement is satisfied.
The temperature-time conditions that must be achieved if monitoring
fecal coliform densities instead of pathogen densities is to be allowed are
0.5 hour at 70°C, 3 days at 55 C, or 5 days at 53 C. Temperatures during
these intervals must be continuously maintained except for momentary changes
as might occur in mixing a windrow. These conditions are compared in Figure 2
to data drawn from the literature by Feachem et al. (1983) for the organisms
of concern in sludge, and to the U. S. Department of Health and Human Services
requirements for eggnog (1989). The EPA curve for sludge requires about a 5 C
higher temperature than suggested by Feachem et al. and is safely
conservative. It is similar to the requirements for eggnog, a food product
with flow characteristics similar to sludge and which, like sludge, contains
ingredients that might protect organisms against heat. There is no reason for
sludge to have a thermal requirement more severe than such a food product.
Figure 2 can be used for situations when the time-temperature conditions are
not identical to the three specified conditions in the regulation.
VECTOR ATTRACTION REDUCTION
The 1992 regulation will require that reduction in vector attraction
(attractiveness to flies and rodents) be demonstrated for either Class A or
Class B treatment. As was mentioned earlier, materials composted according to
the minimum time-temperature requirements of the PSRP and the PFRP
descriptions of composting in the 1979 regulation are unlikely to produce a
product adequately reduced in vector attraction. Despite the inadequacy of
the regulation, no problems have surfaced with compost products not adequately
reduced in vector attraction. There have been complaints about odors from
processing plants and odors probably reflecting the origin of the bulking
agents, but the products presented to the public have evidently not attracted
vectors. Obviously, emphasis at composting facilities has been in producing a
good product rather than just meeting standards. It would seem that no
explicit requirement to reduce vector attraction for these processes is
needed. Nevertheless, the regulation calls for an explicit goal.
One of the qualified means of demonstrating vector attraction reduction
that is useful for many sludges is to show a 38% or greater volatile solids
reduction. This approach is possible for sludges composted with recycled
compost or with inert bulking agents that can be removed from the compost. It
could also be used for composts made with large wood chips that can be
-------
separated from the product. A well composted product is expected to show
greater than 38% volatile solids reduction.
For sludges composted with materials such as sawdust, rice hulls or wood
chips, which fragment and become inseparable from the compost, reduction in
volatile solids content (comparing original sludge to the final product) of
38% cannot be demonstrated because the sludge and the additive become
inseparable. Generally the materials added are higher in volatile solids
content than the sludge, so the final composted product frequently has nearly
the same volatile solids content as the original sludge feed. Volatile solids
reduction is inappropriate as a measure of vector attraction reduction, and
another method must be sought. Specific oxygen uptake rate (SOUR - mg 0, /hr/g
of solids) by a method similar to the one used by Will son and Dalmat (1986)
has been used as a control test to indicate approach to stability in some
composting plants and could serve as a test of vector attraction reduction,
but the method has not been thoroughly investigated. Another alternative is
to designate a time-temperature requirement such as the following: composting
must take place with an adequate air supply for at least 14 days during which
the temperature is not below 40 C and mean temperature is above 45°C but below
60 C. One or both of these approaches will be used in the regulation as an
alternative method for demonstrating vector attraction reduction. The values
selected as standard requirements are not expected to be good control targets
for someone trying to produce a product satisfactory for horticultural
applications. Much further stabilization and maturing will probably be needed
for such uses.
MEANS TO ACHIEVE COMPLIANCE
It is likely that some composting facilities will have difficulty in
meeting the pathogen reduction requirements of the new regulation for Class A
sludge. What means can they take to achieve compliance? At least for the
short term, they can revert to the requirements of the previous regulation.
The PSRP and PFRP requirements will be "grandfathered" for two years to give
wastewater facilities time to develop ways to conform with the regulation.
After that time they will have to meet either the Class A or the Class B
requirements. There are many positive steps that can be taken that will help
facilities achieve compliance. The first step that should be taken at any
facility is to promptly start monitoring bacteriological quality, particularly
fecal coliform density, of their freshly made and matured compost.
Information gained can direct corrective actions such as giving more attention
to meeting or exceeding time-temperature requirements, and reducing
contamination of finished compost with fresh or partially processed material.
Steps should also be taken to isolate "dirty" operations like mixing with wood
chips from "clean" operations like transporting final product to storage.
Steps taken will depend on the process, but improvements can be made in all
processes.
The next obvious step to take is to be certain that the compost is
processed long enough so that it is not a good substrate for survival or
regrowth of fecal bacterial pathogens. Storage or maturation time should be
adequate and product should be removed on a "first in first out" basis. Long
term storage is a reliable way to reduce salmonellae and viruses to negligible
densities. Caution should be exercised not to mix fresh material with aged
material.
-------
Uniform processing is needed if a product uniformly reduced in pathogens
and vector attraction is to be produced. Very few of the available processes
treat the composting mass uniformly. In most processes, forced air flow is
unidirectional, so the mass near the air flow entry is cooler and drier than
the mass near the air exit. The aerated deep-pile process (U. S. EPA 1985),
which is the most commonly used sludge composting process, offers three
imposing obstacles to uniform treatment: flow is unidirectional, sludge and
bulking agent are not mixed or turned during processing, and air supply must
be distributed into a large pile of non-uniform height. Workers at Montgomery
County (Hentz et al. 1991) have demonstrated that improvements in air
distribution and proper pre-mixing of sludge and bulking agent have greatly
reduced odor problems with the aerated deep-pile process. Doubtlessly, these
process improvements have had an equally desirable effect on pathogen and
vector attraction reduction.
Hoitink and Kuter (1986) have shown the non-uniform temperature
distributions caused by unidirectional air flow and have shown the value of
flow reversal in correcting this problem. Design and operational changes of
this type should be attempted if monitoring shows that microbiological
standards are not being met.
The new regulation unfortunately pushes composting in the direction of
less efficient operation. It proposes a relatively simple monitoring
requirement (measuring fecal coliform or salmonellae densities) if specified
time-temperature requirements are met. If they are not, virus and helminth
egg densities must also be monitored. Compost facility operators will try to
achieve the time-temperature requirements to avoid the more difficult
monitoring requirement. To do this, they generally will operate for periods
of time at much higher average temperatures than the 55 C in order to assure
that all the material exceeds 55 C for 3 days. A more stable product could be
produced in shorter time at lower temperatures. The consensus in the
literature (Finstein et al. 1985, Hoitink and Kuter 1986, Vestal and McKinley
1986) is that composting at temperatures in the range of 40 to 50°C removes
more water, destroys more volatile solids, and produces a stable compost in
less time than at higher temperatures.
If close temperature control of each portion of material processed could
be achieved, it might be possible to operate in the temperature range of 45 to
50 C and destroy all pathogens. Extrapolation of the curve in Figure 2
developed from recommendations by Feachem et al. (1983) indicates that
operation in this temperature range for about 12 days would be adequate. The
advantage of this mode of operation would be a more stable compost that would
be less likely to support regrowth of fecal bacteria. The degree of control
needed appears attainable with some in-vessel composting systems, although
process design and control systems would have to be improved.
Pretreatment of sludge can probably solve the problem of residual
pathogens without unreasonable expense. It could even reduce costs by
reducing the amount of microbial monitoring that has to be done. As noted
earlier, compost made with limed sludge may be lower in both fecal coliform
density and salmonellae detections than compost made with anaerobically
digested sludge. Treatment of sludge to a pH of 12 destroys salmonellae and
reduces fecal coliform by at least 2 logs (Counts et al. 1975). Eliminating
salmonellae in advance of composting greatly reduces the problem of surviving
-------
organisms since now only subsequent contamination needs to be dealt with.
Other pre-treatments are possible. It is possible to pasteurize the incoming
liquid sludge before it is dewatered by one of several techniques: by
thermophilic anaerobic or aerobic digestion, by pre-pasteurizing before
anaerobic digestion (Huber et al., 1984), by dual digestion (1986), or by
pasteurization (0.5 hour at 70°C). Dewatered sludge (sludge cake) can be
pasteurized by indirect heating in jacketed heaters or in microwave heaters.
The cost of energy when sludge cake is brought to pasteurization temperature
is so low that energy recovery is probably not warranted. If the sludge is
disinfected before composting, there can be a reduced concern for temperature
control in the composting step. As noted above, overall composting time could
be shortened because decomposition rate would be higher at lower temperatures.
The problem of contamination of compost by pathogenic bacteria has been
exposed by EPA's monitoring investigation (Yanko, 1988). It must be addressed.
Although requirements will exist only for products derived from sludge,
ultimately all composted solid waste and manure products are likely to
eventually face similar requirements.
SUMMARY AND CONCLUSIONS
Composting of sewage sludge produces a product with superior
agricultural and aesthetic qualities. As a consequence, the use of composting
to provide an end-use for sewage sludge has grown rapidly in recent years.
Microbiological quality of the product has been controlled indirectly
requiring adherence to operational standards that specify minimum composting
time and temperature. An extensive EPA survey (Yanko et al, 1988) of
microbiological quality of composts from sewage sludge revealed frequent
occurrence of salmonellae in the product. Consequently, in its new
regulation, to be published in July 1992, EPA will require that sludge
composters meet a microbiological standard as well as adhering to specified
composting times and temperatures.
EPA's microbiological standard was developed from the results collected
in its survey. It requires that composted sludges have less than 1000 fecal
coliform (MPN) per gram of compost (dry basis) or that salmonellae be absent
from the product.
Many of the municipalities that compost sludges will have to upgrade
their operation if they are to meet the required microbiological standards.
The regulation will allow combined use of time-temperature requirements alone
for two years. This will give ample time to institute measures that will
bring their processes into conformity with the standard.
To prepare to meet the new requirements, composters should start
measuring fecal coliform densities of the product as soon as possible.
Corrective actions should then be taken as needed. Suggested corrective
actions include improved housekeeping to prevent contamination of product with
feed, increased processing time, and greater attention to uniform treatment of
every portion of the feed material.
The use of fecal coliforms densities to indicate potential for the
presence of bacterial enteric pathogens in sludge compost is an improvement
over the time-temperature requirements of the previous regulation. It would
-------
be better to be able to enumerate the specific pathogens of concern but lack
of methods or their high cost make this approach a practical impossibility.
The great advances in biotechnology that are being made make it likely that
economical and accurate methods for detecting these specific pathogens at low
densities in sludge and composts can be developed. Such efforts should be
encouraged. Once specific pathogens can be enumerated, their fate through the
entire treatment train can be determined. Processes can be improved to
maximize pathogen destruction and insure pathogen-free products.
The regrowth of fecal indicator organisms and salmonellae in some
composts brought to light by Yanko's survey could cause these products to be
downgraded so that they cannot be used without restrictions. Research should
be undertaken to determine the causes of the regrowth phenomenon and develop
ways to minimize or eliminate it.
-------
Literature Cited
APHA-AWWA-WPCF, 1989. "Standard Methods for the Examination of Water and
Wastewater", 17th ed., pub. APHA, Wash., D.C.
Appleton, A.R., Jr., C.J. Leong, and A.V. Venosa, 1986. Pathogen and
indicator organism destruction by the dual digestion system, Jour. WPCF, 58,
No. 10, 992-999 ~~
Burge, W.E., P.O. Millner, N.K. Enkiri, and D. Hussong, 1987. "Regrowth of
Salmonellae in Composted Sewage Sludge", EPA No. 600/2-86/106, (NTIS PB 87-
129532/AS)
Clark, C. S., H.S. Bjornson, J. Schwartz-Fulton, J.W. Holland, and P.S.
Gartside, 1984. Biological health risks associated with the composting of
wastewater treatment plant sludge, Journal WPCF, 56, No. 12, 1269-1276
Counts, C.A., and A.J. Shuckrow, 1975. "Lime Stabilized Sludge: Its Stability
and Effect on Agricultural Land", EPA No. 670/2-75-012, (NTIS No. PB 241809)
Davies, O.L., and P.L. Goldsmith (ed), 1976. "Statistical Methods in Research
and Production", 4th revised ed., Longman Group Ltd., Essex, England
Farrell, J.B., B.V. Salotto, and A.D. Venosa, 1990. Reduction in bacterial
densities of wastewater solids by three secondary treatment processes. Res.
Journal WPCF, 62. No. 2, 177-184
Feachem, R.G., D.J. Bradley, H. Garelick, and D.D. Mara, 1983. "Sanitation and
Disease: Health Aspects of Excreta and Wastewater Management", Pub. for World
Bank by J. Wiley & Sons, NY
Federal Register, 1979. "Criteria for Classification of Solid Waste Disposal
Facilities and Practices" (as corrected in FR of Sept. 21, 1979), 44 No. 179,
Sept. 13, 53438-53468. See also Code of Federal Regulations, 40 CFR 257
Federal Register, 1991. "Solid Waste Disposal Criteria; Final Rule" (40 CFR
Parts 257 and 258). 56, No. 196, Oct. 9, 50978-51119
Finstein, M.S., F.C. Miller, S.T. MacGregor, and K.M. Psarianos, 1985. "The
Rutgers Strategy for Composting: Process Design and Control", EPA Rept. No.
600/2-85/059, (NTIS No. PB 85-207538/AS)
Goldstein, N. and Riggle, D., 1990. Sludge composting maintains momentum.
Biocycle, December, 26-32
Hentz, Jr., L. H., C. M. Murray, J. L. Thompson, L. L. Gasner, and J. B.
Duncan, Jr., 1992. Odor control research at the Montgomery County regional
compost facility. Water Environment Research, 64. 13-18, January/February
Hoitink, H. A. J., and G. A. Kuter, 1986. "Factors Affecting Composting of
Municipal Sludge in a Bioreactor", U. S. EPA Report No. EPA-600/2-86-014,
(NTIS NO. PB 86-155579/AS)
-------
Huber, J., and E. Mihalyfy, 1984. Experiences with pre-pasteurization of
sewage sludge with heat recovery, pp 381-398, In "Sewage Sludge Stabilization
and Disinfection, A. Bruce (ed), pub. Water Res. Centre/Ellis Norwood Ltd.,
Chichester, England
Millner, P.O., K.E. Powers, N.K. Enkiri, and W.D. Burger, 1987. Microbially
mediated growth suppression and death of salmonellae in composted sewage
sludge. Microbial Ecology, 14. 255-265
U.S. Dept. Health & Human Services, 1989. "Grade A Pasteurized Milk
Ordinance, 1989 Revision, Public Health Service/Food and Drug Administration
Publication No. 229
U.S. EPA, 1985. "Seminar Publication: Composting of Municipal Wastewater
Sludges", Report No. EPA/625/4-85/014
U.S. EPA, 1989. "Environmental Regulations and Technology: Control of
Pathogens in Municipal Wastewater Sludge", See Section 6. Report No.
EPA/625/10-89/006
Vestal, J. R., and V. L. McKinley, "Microbial Activity in Composting Municipal
Sewage Sludge", U. S. EPA Rept. No. EPA/600/2/86/025, (NTIS No. PB 86-166
014/AS)
Willson, G. B. and D. Dalmat, 1986. Measuring compost stability. Biocycle,
34-37, August
Yanko, W. A., 1988. "Occurrence of Pathogens in Distribution and
Marketing Municipal Sludges", Report No. EPA/1-87/014,
(NTIS #PB 88-154273/AS)
-------
Table 1 Fraction of Salmonellae Detections in Compost
From Sites Sampled by LACSD
Compost Sampling Type of Salmonellae Detections
Location1 Composting Number of Number of Fraction
Detections Measurements Detected
(D) (H) (f-D/M)
Sampled Weekly
IX-A-1,2 windrow 3 54 0.06
I-A-3 windrow 9 52 0.17
IX-A-4,6 windrow 58 102 0.57
IX-A-5 windrow 36 52 0.69
III-B-1 static pile 23 45 0.51
III-B-2 static pile 36 45 0.80
Sampled bimonthly1'3
I-B-1 static pile 06 0
III-B-3 static pile 06 0
III-B-4 static pile 1 6 0.17
IV-B-1 static pile 3 6 0.50
V-B-1 static pile 06 0
IX-B-1 static pile 1 6 0.17
II-C-1 In-vessel 1 6 0.17
X-C-1 In-vessel 1 6 0.17
III-J-1 aerated windrow 06 0
VIII-J-1 aerated windrow 06 0
VII-A-2 windrow 06 0
IX-A-10 windrow 06 0
VIII-H-1 proprietary 1 6 0.17
1. Refer to text and to Yanko (1988) for a description of the sampling sites.
2 All "A" locations were Los Angeles County Sanitation Districts compost. A-l and
' A-2 were in bulk, A-3 was bagged, and A-4, 5, and 6 had wood chips or rice hulls
added The "B" locations were Philadelphia compost. B-l is unscreened at a
giveaway bin, B-2 is screened before bulk distribution.
3. Samples were from various parts of the United States. The Roman numeral indicates
the Federal region where the site was located.
-------
Table 2. Salmonellae Detections as a Function of Average Log Fecal
Coliform Density (log MPN/g) from Yanko's Weekly Sample Data
Log FC Density Average
Log FC Density
Interval
Salmonellae Detections
n.d.
0 - <1
1 - <2
2 - <3
3 - <4
4 - <5
5 - <6
6 - <7
7 - <8
8 - <9
9 - <10
Totals
0.5
1.5
2.5
3.5
4.5
5.5
6.5
7.5
8.5
9.5
Number of
Measurements
(M)
40
8
16
22
31
56
61
47
56
25
_3
Number of
Detections
(D)
0
0
0
0
5
21
30
32
50
24
_3_
Fraction
(f=D/M)
0
0
0
0
0.16
0.375
0.54
0.76
0.89
0.96
1.00
365
165
1. Yanko's population of fecal coliform measurements for his weekly samples was
arranged in a frequency distribution with intervals of 1.0 Log FC density
(Col. 1 - intervals, Col. 3 - number of measurements in each interval). Number
of salmonellae detections in each interval was determined (Col.4) and the
fraction detected in each interval calculated (Col. 5). The average Log FC
density for each interval (Col. 2) was plotted against fraction detected in
Figure 1.
-------
99
T r
CO
0
O
LU
h-
LU
Q
LU
i
LU
O
_J
CO
o
o
r~
X
M—
90
80
60
40
20
10
5
1
0.5
0.1 •
jf\
f\
m ^f
/
'
: / :
' I
/
s
/
s
s
-
t . 1 . 1 , 1 1 _1 1_ 1,1.1.
012345678
AVERAGE LOG FECAL COLIFORM DENSITY
Figure 1; Relationship between Average Log Fecal Coliform
Density and Fraction of Salmonellae Detections
(Yanko et al. Weekly Data)
-------
Table 3. Fraction of Salmonellae Detections (f) as a Function of
Average Log Fecal Coliform Density (log MPN/g) from LACSD
Bimonthly Samples
Log FC
anqe
Average
Loo FC
Sludae Tvoe
n
1
2
3
4
5
6
7
8
.d-
-------
100
90
O
•
UJ
tr
13
cc
UJ
Q.
S
UJ
40
• Ascaris (Feachem)
+ Salmonellae (Feachem)
)K Enterovirus (Feachem)
• EPA thermal process
X PHS - FDA, Eggnog
2345
LOGARITHM OF TIME (LOG SEC)
Figure 2: EPVs Time-Temperature Relationship for Thermal Disinfection
Compared with Others.
-------
|