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.

-------