Proceedings of the
NATIONAL
CONFERENCE ON
DISPOSAL OF
RESIDUES
ON LAND
-------
Proceedings of the
NATIONAL CONFERENCE ON
DISPOSAL OF RESIDUES ON LAND
September 13-15,1976
St. Louis, Missouri
Sponsored by:
Office of Research and Development,
Office of Solid Waste Management Programs
U.S. Environmental Protection Agency
Environmental Quality Systems, Inc.
Information Transfer Inc.
-------
Printed in the United States of America
Library of Congress Catalog No. 77-78970
Copyright © 1977 by
Information Transfer Inc.
1160RockvillePike
Rockville, Maryland 20852
All Rights Reserved
-------
CONTENTS
The Institutional Challenges to Residual Management
Charles V. Wright 1
Residue Disposal Onto Land—An Overview
Harold Bernard 5
Current EPA Guidance on Land Application of
Municipal Sewage Sludges
Robert K. Bastion and William A. Whittington 13
Runoff and Movement of Soil Particles
W. H. Wischmeier 15
Heavy Metals Contained in Runoff From Land
Receiving Wastes
W. E. Larson and R. H. Dowdy 21
Heavy Metal Contents in Runoff and Drainage Waters
from Sludge-Treated Field Lysimeter Plots
Thomas D. Hinesly and Robert L. Jones 27
Transport Model to Predict the Movement of Pb, Cd,
Zn, Cu, and S Through a Forested Shed
J. K. Munro, Jr., R. J. Luxmoore, C. L. Begovich,
K. R. Dixon, A. P. Watson, M. R. Patterson, and D. R. Jackson 45
Leachate From Applications of Fertilizers, Manures
and Sewage Sludges to Land
P. F. Pratt and A. L. Page 59
The Effects of Industrial Sludges on Landfill
Leachates and Gas
D. R. Streng 69
A Preliminary Examination of Vinyl Chloride Emissions
From Polymerization Sludges During Handling and
Land Disposal
R. A. Markle, R. B. Iden and F. A. Sliemers 77
Land Disposal of Organic Hazardous Wastes Containing HCB
W. J. Farmer, M. Yang and J. Letey 83
Sludge Farming of Refinery Wastes as Practiced at
Exxon's Bayway Refinery and Chemical Plant
Robert S. Lewis 87
Radioactivity Uptake by Plants
Ronald G. Menzel 93
Factors Affecting Plant Uptake of Heavy Metals From
Land Application of Residuals
J. A. Ryan 98
Heavy Metal Uptake and Control Strategies Associated
with Sewage Sludge Fertilized Crops
Cecil Lue-Hing, J. R. Peterson, D. R. Zenz and T. D. Hinesly 106
General Technology on Soil-Salt Chemistry and Saline
Subsurface Irrigation Return Flows
Gaylord V. Skogerboe, Wynn R. Walker, David B.
Me Whorter and James E. Ayars 115
Tracing Leachate from Landfills—A Conceptual Approach
Hugh Mullen and Steven 7. Taub 121
Transport of Salts from Disturbed Geologic Formations
David B. Me Whorter and Jerry W. Rowe ,, 127
-------
Disposal of Coal Mining Industry By-Products
Waller E. Grube, Jr., Eugene F. Harris and John F. Martin 133
Phosphate Transport Through Soil
Carl G. Enfield 138
Updating the Nitrogen Cycle
L. M. Walsh 1*6
Fate of Nitrogen From Fertilizer Practices
G. W. Wallingford 152
Fate of Nitrogen From Manure Disposal
W. L. Powers, R. V. Terry, L. S. Murphy and
G. W. Wallingford 156
Fate of Nitrogen From Municipal Sludges
Larry D. King 161
Disposal of Industrial Nitrate Effluents by Means
of Forest Spray Irrigation
W. F. Harris, G. S. Henderson and D. E. Todd 166
Transport and Reaction of Contaminants in Ground-Water
Systems
David B. Grove and Jacob Rubin 174
Field Verification of Hazardous Waste Migration from
Land Disposal Sites
J. P. Gibb and K. Cartwright 179
A Preliminary Assessment of the Effects of Subsurface
Sewage Sludge Disposal on Groundwater Quality
Dale C. Mosher 185
Control Program for Leachate Affecting A Multiple Aquifer
System Army Creek Landfill, New Castle County, Delaware
Walter M. Leis, Abraham Thomas, Roy F. West on,
David C. Clark and Kenneth D. Shuster 189
The Potential for National Health and Environmental
Damages from Industrial Residue Disposal
Emery C. Lazar, Robert Testani and Alice B. Giles 196
Health Effects—Land Application of Municipal
Wastewater and Sludge
Thomas L. Gleason, HI, Frank D. Kover and
Charles A. Sorber 203
Health Effects of Municipal Refuse Disposal
Stephen C. James 211
-------
-------
The Institutional Challenges
to Residual Management
Charles V. Wright
Region VII, U. S. Environmental Protection Agency
Kansas City, Mo.
INTRODUCTION
Residuals are a problem of gargantuan size as well as com-
plexity. The Office of Solid Waste Management Programs has
presented testimony to the U. S. House of Representatives con-
cerning the quantity and composition of the non-hazardous and
hazardous discarded materials stream. This includes 1.783 bil-
lion tons of mining wastes, 687 million tons of agricultural
wastes, 135 million tons of municipal wastes, .260 million tons of
industrial wastes, and 7.3 million tons of sewage sludges. This
imposing number of three to four billion tons of solid waste
must be disposed'of predominantly on land. EPA has further
documented over 23 million tons of potentially hazardous waste
produced by eight groups of industries. The growth of sludges as
a result of air and water pollution control standards may take
dramatic jumps if the 1977 and 1983 mandates of Public Law 92-
500 are met. We also expect an increase in potentially hazardous
waste by 1983 of 31 million tons by dry weight, an increase of
over 33 percent in nine years; and, the growth of all solid wastes
are increasing between 2.5 and 4.5 percent per year.
We must examine the technical issues on residuals manage-
ment and to help narrow the gap between what we know and
what we need to know. This I am confident we can do—R&D on
safe, beneficial methods to dispose of residuals will ultimately
lead to standards and methodology.
Concurrently, we must develop less cumbersome, more effec-
tive institutional processes. The problem from social, political,
and economic influences, in some respects surpass the technical
in difficulty to master. They are complex and often illogical and
frustrating.
The institutional factors which will affect residuals manage-
ment can be segmented into:
• Legal and regulatory
• Social and political—and the bonds of existing practice.
Each of the residuals has a particular set of issues, institu-
tional arrangements and economic considerations which govern
the resolution (or lack of resolution) of problems.
Industrial Residues
Industrial residues represent one of the largest in volume—
they are being produced in increasing quantities and complexi-
ties. They are a result of both public and private sector activities
and their disposal on land is controlled basically by the states.
The role of EPA has been primarily one of technical and finan-
cial assistance to the states. The states have the primary regu-
latory function which a few states have exercised very adroitly,
but others have failed to act.
The institutional framework for controls is complicated by
the media approach of our pollution control efforts—(indepen-
dently). The following discussion by media presents the pri-
mary sources of industrial residues and the roles of different
governmental levels and the private sector in their disposal.
Air
The Federal regulatory program established by the Clean Air
Act of 1974 has considerable impact on production and controls
of residuals by industry. During the past, the actual impact of air
pollution regulations—and technology—have been assessed
poorly if done at all. However, the background documents
"Standards Support and Environmental Impact Statements"
(SSEIS) will in the future contain assessments of residuals and
economic analysis. Fortunately, it is expected that residuals pro-
duction from air pollution control will increase much more
gradually than from water pollution controls.
The disposal of materials collected by air pollution devices
can be divided into sludges or solids basically collected as partic-
ulate matter. The sludges are primarily sulfur oxide sludges
which are not hazardous but represent a voluminous, special
land disposal problem. The particulates are primarily dust, dirt,
and fly ash. In the Region VII states, Iowa has developed guid-
ance for the disposal of fly ash under the solid waste disposal
law. The other three states in Region VII have not developed
guidance for fly ash disposal and none of the Region VII states
have developed guidance for sulfur oxide sludge disposal. In our
Region it is a problem of minor proportions.
Water
On the other hand—industrial residues from water pollution
controls under PL 92-500 already present a challenge.
The production of industrial sludges has been estimated at 35
million tons per year on a dry-weight basis and is greater than
municipal waste water sludges. Ten major industrial sectors
were estimated to produce in excess of 7 million tons of sludge
for disposal on land in 1971. These ten will increase disposal on
land requirements to 13 million tons in 1977 when the best prac-
tical treatment (BPT) is installed. All industries may increase
their disposal on land requirements to 25 million tons by 1983.
The role of the Environmental Protection Agency with regard
to industrial sludge management will follow the same basic pat-
tern as described for municipal sludge, but will impact the pri-
vate sector rather than the public.
The EPA draft residuals sludge management strategy paper
includes two "immediate action" tasks with regard to industrial
sludge, namely: "(1) development of detailed program plan for
industrial sludge management by 12/76 and (2) development of
-------
The Institutional Challenges
a guidance document for the utilization/disposal of industrial
sludges to be used by the regional offices, states, and localities by
12/77." In addition, several issues directly related to industrial
sludges require further evaluation: (1) the determination of the
public health consequences of various disposal mechanisms to
define the risk/benefit relationships and (2) the technological
problems of dewatering, recovery, operations, and mainte-
nance. The effects of sludge are intermedia, therefore must be
investigated on an intermedia basis. The lack of adequate data
on and guidance on adequate practices, market information for
selected sludges, and recognition of the impact of the sludge
management problem created by legislation on the state and
federal level will be overcome through research, investigation
and dissemination of information. These issues are being inves-
tigated by a variety of methods including the definition of the
roles of the EPA, USDA, and FDA with the states and the
establishment of communication channels for information flow.
In addition, issue papers or similar documents outlining EPA's
policy and procedures will be developed as rapidly as possible in
order to provide the needed guidance .to states, local agencies,
and the private sector.
The role of the states includes operation of the water pollution
control permit program in the states which are delegated such
authority. In Region VII, Nebraska, Missouri, and Kansas, the
permit programs have been delegated. The state agencies have
the primary regulatory authority for industrial sludge disposal
on land and the local agencies or private firms have the opera-
tional, financial, and technical responsibilities. The activities of
the Region VII states will be described later in the municipal
sludge section.
Solid Waste
The production of industrial solid waste other than mining
has been estimated by EPA's Office of Solid Waste Manage-
ment Programs to be between 140 and 270 million tons per year.
Unfortunately wide variations in the production of solid wastes
make estimates a dynamic statistic. We have to remember this in
developing control strategies. The dynamics of estimates can be
illustrated; e.g., original estimates had been made that the
hazardous waste stream amounted to 10 million tons/year.
However, further investigations have placed the quantity of
hazardous waste produced at closer to 23 million tons/year. As
a result of the BPT and best available treatment (BAT) require-
ments, this total is expected to increase to over 26 million tons in
1977 and over 30 million tons in 1983. Although the nonhazard-
ous industrial solid waste stream is not well documented or
defined, it is estimated to amount to 270 million tons/year and
will be increasing at a rate of between 2.5 and 4.5 percent per
year.
The role of the Region VII states in regulating industrial solid
wastes has traditionally been to treat them much the same as
municipal solid waste unless it contained a hazardous material
The states of Kansas, Missouri, Iowa, and Nebraska regulate
solid waste disposal sites by requiring a permit for construc-
tion and operation of the sites. The state agencies review the site
design and operation plan which should include a description of
any industrial solid waste to be accepted. If, in the judgment of
the agency the site is designed to adequately handle industrial
waste, approval to receive specific materials is provided with the
permit. If special conditions for receipt of the industrial waste
are judged necessary, the conditions for accepting the wastes are
included in the permit.
Basic data on hazardous waste production in Region VII is
now being developed: In Iowa, the Department of Environmen-
tal Quality is conducting a hazardous waste survey to determine
the extent of hazardous waste generation, handling, and dispo-
sal problems within the state. They have developed and pro-
posed to the state legislature a bill which would provide the state
agency with the authority to control the treatment, disposal, and
storage of hazardous wastes and require reporting on genera-
tion.
In Kansas, a hazardous waste survey of 450 industries has just
been completed by the state solid waste management agency.
The agency is drafting a report on the findings of the the survey
and also plans to develop recommendations for legislation.
State personnel are appearing today (September 13) before the
legislative committee to discuss these needs. State regulations
already provide for limited control of hazardous waste land dis-
posal and can be expanded.
In Missouri, the Department of Natural Resources is also
completing a hazardous waste survey of over 400 industries and
is at the same time working with an ad hoc committee to devel-
op consensus hazardous waste management legislation. The
state agency currently has authority over hazardous waste treat-
ment and disposal operations and three sites are permitted to
receive various types of hazardous wastes on an "as requested"
basis for each waste. The new legislation will be directed toward
insuring that all hazardous wastes generated are reported and
disposed of properly.
In Nebraska, the Department of Environmental Control has
the primary authority over solid waste disposal from first-class
cities and at all private sites. Other agencies, such as the Depart-
ment of Health and Highway Department, also have some juris-
diction over such sites, but agreements are under development
to avoid duplicate jurisdiction. The state solid waste agency has
completed a survey of approximately 90 generators of hazard-
ous wastes and is in the process of developing a survey report
with recommendations for needed legislation. The state envi-
ronmental protection act contains several sections relating to
hazardous and toxic pollutants which require a legal opinion
pertaining to their scope before recommendations on legisla-
tion can be developed.
In the hazardous waste management area, the states in Region
VII are responding to the challenge. Nationally, California has
led the way but Minnesota and Oregon also have relatively new
comprehensive hazardous waste management laws, regu-
lations, and programs. Illinois, Oklahoma, Washington, and
Maryland have hazardous waste management acts in various
stages of implementation, and Pennsylvania, Massachusetts,
and other states are exercising through other acts limited control
over hazardous wastes. Georgia and Kentucky have issued guid-
ance concerning the handling, treatment, and disposal of
hazardous wastes, which is another approach to the problem. As
you can see, our states are developing programs without a fed-
eral mandate, on a heterogeneous pattern. A uniform national
program is needed.
The private sector to be regulated has had numerous oppor-
tunities for input to a national hazardous waste management
program. With the growth and diversity of state programs, little
qyert opposition to the hazardous waste management provi-
sions of Senate Bill 2150 and House Bill 14496 has developed
signaling to EPA that perhaps the time has arrived.
The two bills have no substantial differences: They would
establish criteria and designation of hazardous waste; regu-
lations for disposal, treatment, and storage; criteria for a system
of manifests; and minimum criteria for state programs autho-
rized to operate the permit, regulatory, and enforcement activi-
ties. In addition, the bills would authorize the Administrator to
seek injunctions for imminent hazards caused by disposal of
solid or hazardous wastes and provide for enforcement of the
regulations and standards established under the hazardous
waste management portion of the act.
Implementation of the hazardous waste management provi-
sions would take place over an 18-month period and would
extend program support to those states that met the require-
ments for state programs. In addition, the definition of "hazard-
-------
The Institutional Challenges
ous" would not be limited to solids but would include solids, liq-
uids, and contained gases, making the provision intermedia and
establishing a program concerned with these wastes regardless
of form.
Municipal Sludges
The role of the federal government in control of land disposal
of the municipal sludges is one of financial and technical assis-
tance to states, local government, and industry in the devel-
opment of technology for proper disposal. However, the impact
of the Federal Water Pollution Control Act, P. L. 92-500, goes
considerably further.
The impact of P. L. 92-500 was reported in detail by Harold
Bernard at the Second Annual National Conference on Munici-
pal Sludge Management and Disposal. Briefly, he listed nine
specific sections of the Act which will significantly increase
sludge production by 1977 and 1983, and/or will require strict-
er enforcement.
Shortly after passage of the Act in late 1972, EPA formed a
sludge management working group including USDA and FDA
to develop recommendations on proper municipal sludge prac-
tices.
At the June 9 and June 10 meeting of this year, the regional
requirements subcommittee developed a statement concerning
the major needs in the area of residual sludge management. The
highest priority is the development of a national policy including
an interim policy on industry pretreatment requirements for
heavy metals. Technical guidance which provides limits on con-
taminants discharged to land was the next priority, followed
closely by the need for states to develop sludge management
guidelines.
This imposing list of major recommendations provides a good
summary of the state of the sludge management program. It
requires a major commitment of resources. A draft-strategy pa-
per for residual sludge management has just been completed for
internal review by the working group. Several of you have seen
the draft—we look for its publication in the near future.
Also, on June 3, 1976, in the Federal Register, the
Environmental Protection Agency issued a technical bulletin on
the environmental factors in municipal sludge management.
The objective of the bulletin is to provide design information on
sludge management options and factors on environmental
acceptability.
The state role in the area of municipal sludge disposal on land
will be regulatory. Many state agencies which control solid
waste management have broad control over land disposal which
should be expanded, with the assistance of air and, water pollu-
tion control programs, to encompass disposal of all wastes on
land.
Although several states have advanced further in municipal
sludge guidance than those of Region VII, our states are moving
(at different rates) to establish a regulatory control program
over the land disposal of municipal and industrial sludges. The
state of Iowa has proposed to develop guidance for land disposal
of municipal and selected industrial sludges. The state of Kansas
allows the disposal of municipal waste treatment sludges in sani-
tary landfills. This disposal practice is proposed to be evaluated
under the funds available under section 208 of P. L. 92-500. The
state of Missouri allows the disposal of municipal waste water
treatment sludges in sanitary landfills designed for receipt of
such sludges. They have proposed to evaluate land disposal of
sludges under their section 208 grant also. The state of Nebraska
allows disposal of municipal sludge in sanitary landfills which
are permitted by the state agency. They are examining the possi-
bility of control over municipal sludge disposal through regu-
lations aimed at the source rather than the method of disposal.
Cities must adopt specific pretreatment requirements in order
to be eligible for federal construction grant funds. The national
pollutant discharge elimination system (NPDES) requirements
and permit program will also require communities not involved
in the construction grant programs to meet the pretreatment
requirements for noncompatible wastes.
New (and substantial) funds have been made available for
planning residual management in an intermedia context under
the 208 planning program under P. L. 92-500. Basically, the
state 208 programs will allow state agencies to focus on munici-
pal and industrial sludge management concepts. The state agen-
cies should develop rules and regulations and technical guid-
ance covering these sludges as part of the statewide 208 planning
program and can exercise guidance over institutional arrange-
ments through matching grant provisions of state law. The local
role will be primarily limited to municipal sludge management.
Municipal Refuse
The production of solid waste has been correlated with the
growth of the gross national product by Professor Zaltzman of
West Virginia. The post-consumer solid waste production figure
for 1971 was 125 million tons and for 1973 grew to 135
million tons on a wet basis. Of primary concern is that the vast
maj ority of this waste stream is disposed of on land and that con-
trols on these operations are becoming more intensive.
In 1965 when the federal involvement began with the passage
of the Solid Waste Disposal Act, only five states had solid waste
management programs and employed a total of 13 people. As of
1976, all 50 states had active programs employing over 400 per-
sons. But this is not a measure of progress in itself. A real mea-
sure of progress is in the number of state permitted land disposal
sites, which for Region VII has increased from 27 in 1970 to 380
in 1976, serving 84 percent of the population. This dramatic
growth indicates the increasing concern for the manner in which
we dispose of residuals on land. The initial permitting and con-
trol of disposal practices is being rapidly completed with the
establishment of properly engineered sanitary landfills. How-
ever, we are beginning to ask whether the practices we have
established are environmentally adequate.
National legislation is being considered to require a national
ban on open dumping. It would be implemented in a cooperative
fashion between the federal and state governments with the fed-
eral government providing financial and technical assistance
and the states providing the legal, regulatory, and permitting
activities. The focus of the Environmental Protection Agency
with regard to residuals management has, until very recently,
concentrated on the disposal of municipal solid waste. Designa-
tion of the Office of Solid Waste Management programs as the
lead office concerning all sludge management and their investi-
gations of hazardous waste management problems has
expanded the scope of residuals disposal programs within EPA.
The role of state government in the land disposal of solid
waste is as the regulatory body. The states have accepted the
challenge to develop and operate a variety of comprehensive
solid waste management programs which have resulted in dra-
matic progress. They have developed regulations governing
solid waste disposal sites and are continuing to evolve better
standards of construction and operation. The state programs in
the Region VII states have limitations depending on the political
and social pressures within the state, but, for the most part, they
have evolved dynamic programs for land disposal. The chal-
lenge they now face after their initial success is to continue to
improve and to spur the local and private agencies to upgrade
their operations. Their cooperation and institution of adequate
disposal practices is vital to protection of the environment.
Their investment of resources in adequate facilities has contrib-
uted to the development of sound land disposal practices and
will continue to fare into the future.
Agricultural chemicals represent a special category—at the
federal level, there is an institutional mishmash. The Environ-
-------
The Institutional Challenges
mental Protection Agency has the primary responsibility for the
regulation of pesticides through registration and use control.
The U. S. Department of Agriculture, an advocate for agricul-
ture, assists farmers in pesticide selection and other pest control
methods. The Food and Drug Administration enforces the pes-
ticide residue tolerance limits and regulates the use of veterinary
drugs. Finally, National Institute of Occupation, Safety and
Health (N1OSH) establishes, and Occupational Safety and
Health Administration (OSHA) enforces the health and safety
standards for workers, both farm and manufacturing, from the
adverse effects of pesticides.
The EPA regulatory program established under the new fed-
eral insecticide, fungicide, and rodenticide act (FIFRA) is a far
reaching concept with social as well economic consequences.
The Act makes the user responsible for his action in pesticide use
including proper storage and disposal of pesticides and contain-
ers. Proposed regulations on this subject were published in the
Federal Register on October 15, 1974, and final regulations are
expected by the end of Fiscal Year 1976 (this month). Basically,
the disposal regulations would ban open dumping, including
land disposal; open burning which does not provide adequate
environmental and public health protection; and well injection
which does not have prior state agency approval. The regu-
lations apply or will apply to both containers and pesticides,
which are viewed as equally serious problems. Incidentally, an
estimated 250,000,000 pesticide containers require disposal each
year.
(Without legislation which requires some type of a collection
system and provides an incentive for return of the con-
tainers, the pesticide container disposal problem will continue to
mount. Pressure must be applied: (I) to the manufacturer to de-
velop reusable containers; (2) to dealers or collection centers
requiring them to accept containers for return to the manufac-
turer; and (3) consumers (farmers) must be provided a deposit
return on containers at a level sufficient to make them return the
empties.)
SUMMARY
The definition of residuals for this proceedings focuses on
industrial and municipal sludges, municipal refuse, irrigation
salts, mining residuals and salts, and agricultural fertilizers. The
element common to all is their application to land. Yet the land
varies from sand to clay in proportions which can change in a
few feet. The complexity of developing a set of criteria to protect
the land quality will be enormous. In addition, the constituents
of every sludge can vary with every facility and may change on a
daily basis. The disposition of these by-products should be care-
fully tailored to the receiving soil or, if not suited for this dispo-
sal method, should be diverted to a long-term, secure disposal
site or adequate treatment facility.
To paraphrase Tom Hayden, "The earth is a closed system—
there is no 'away' to throw things." The responsibility for de-
veloping safe and economical disposal methods for residuals
rests with both the public and private sectors. The public (gov-
ernment) must prescribe reasonable requirements for residuals
handling, transport, treatment, and disposal. The private sector
must provide the means to accomplish environmentally accepta-
ble treatment and disposal. The research and development of
methods involves both sectors.
The division of responsibility for the purpose of defining land
use, treatment methods, and standards is divided, as it should
be, among local, federal, and state governments with participa-
tion of the populace at each level. Generally, the treatment
standards for Tand disposal have been established at the state
level of government. However, increasing concern with hazard-
ous wastes and their environmental and health impact have led
to the recommendation that a nationwide hazardous waste man-
agement program be developed. Senate Bill 2150, and House
Bill 14496, incorporate a national hazardous waste management
program to be administered by the Environmental Protection
Agency and would provide—at last—a rational institutional
framework to all levels.
-------
Residue Disposal Onto Land—
An Overview
Harold Bernard
Environmental Quality Systems Inc.
Rockville, Maryland
The purpose of this Conference is to review those aspects of
the residue disposal on land that are known and that can be used
immediately for disposal operations, to determine the confi-
dence level of usefulness, and to indicate the areas for additional
information prior to use.
The aspects that are of concern at this Conference are the
transport, fate and effects of the nitrogen, phosphorus, heavy
metals, inorganic salts, and organic compounds in residues, and
the potential health effects produced by the disposal of these
residues on the land.
There are many sources of waste. They can, however, be
broken down into basic constituents of concern and their respec-
tive impact on the environment. Wastes can be considered in
regard to their pollution potential, such as organic-BOD,
organic-COD, complexed organics, inorganics-high salts,
inorganics-heavy metals, phosphates, and nitrogen compounds.
In addition, these can be considered in terms of surface runoff,
leachate into and through the root zone, uptake and impact on
plants, movement of leachate through the unsaturated zone of
the soil column and finally movement in aquifers as shown in
Figure 1. Looked at in this manner, residues have a commonal-
ity of constituents, transport phenomena and effects that can be
interchanged.
This does not suggest that one can expect an exact exchange
of data from one residue source and site to another. It is
recognized that the combination of a residue and a specific site
are unique, but they are unique to degree, not to kind. That is to
say a site and waste will react similarly from one site to another,
but not exactly the same.
Figure 2 indicates the relative annual contribution of mining,
agriculture, and industrial residues, sludges from the industrial
treatment of industrial wastewaters, sludges from the treatment
of municipal wastewaters, municipal refuse and water plant
sludges. Note that industrial and municipal sources of residue
constitute only 415 million tons of the total annual production
of 2885 million tons, or about 14%. For the agricultural and
mining sources, the major source of waste is the sediment runoff.
The fate of the constituents in the residues, i.e. the nitrogen or
the phosphate, the organics, heavy metals, inorganics, are gener-
ally similar, as shown in Figure 1. Of concern is vaporization,
runoff, leaching into the ground and subsequent uptake through
the roots and back up into the plants which are eventually har-
vested or die off. The percolate continues down through the
unsaturated zone into the saturated zone, or the water table,
where it moves generally horizontally down gradient hydrauli-
cally to resurface in a stream or be extracted in a well.
RunoffO
Vaporization
Residue
Harvested Crops
Residue-soil - Atmosphere interface
Soil and Plant Uptake
Unsatur
S- L ' t
>s (£
one
Satura
Figure 1: Fate of Residues Applied to Land
Leachate
-------
An Overview
Figure 2: Estimated Industrial Versus Other Residual (August,
1970-1974) (Dry Weight in Million Tons per Year)
Nitrogen
One of the limiting factors to unlimited application of
residues onto the land is nitrogen; its uptake by humans and
animals can constitute a potential hazard. Consequently, the
mass distribution of all forms of nitrogen applied to the land is
an important facet of residue disposal that must be known or
accounted for. The factors affecting the fate of nitrogen are
shown in Table I.
Table I: Factors Affecting Fate of Nitrogen
AVAILABILITY OF OXYGEN
FORM OF NITROGEN
BIODEGRADABILITY OF CARBON (OGRANIC) SOURCES
WATER
TEMPERATURE
pH
The various sources of nitrogen applied to the land are
precipitation, rain out, biologically fixed nitrogen, and synthetic
and natural fertilizers applied to the soil. The forms of nitrogen
are determined by the micro-environment surrounding the ni-
trogen. The transport phenomena is closely associated with the
amount and rate of movement of soil water. The factors affect-
ing nitrogen uptake and movement are the characteristics of the
nitrogen in the residue, its fate in and through the soil, and the
utilization of the end-product as shown in Figure 3.
Air:
Other:
Nitrogen Sources
Transformations in Soil
and Possible Losses
Utilization
NH3, N20 & NO
adsorbed by soil
Non-synthetic N
Fertilizers
Manures, Seeds,
Residues, etc.
Figure 3: Soil Nitrogen Sources and Transformations and Fate of the End Products
-------
An Overview
Rain has about a 50% efficiency for scrubbing or scavenging
pollutants from the air. Figure 4 shows a substantial amount of
NH4 and NO3 nitrogen in rainout. A significant fraction of this
source of nitrogen is immediately available. For example, I have
a small 24-foot diameter swimming pool in my yard. I live in the
suburbs of Washington, DC, a relatively industrial-free envi-
ronment, and yet every time it rains, the swimming pool shows
an instant algal bloom, unless it is predoused with extra chlorine
to prevent the algal bloom. I also noticed that when I am caught
in the first 10 minutes of any rainfall, my scalp begins to itch like
crazy (my personal variation of litmus paper). This is due to the
acidity of rainfall as it washes the industrial pollutants from the
air. Figure 5 shows the variation of rainfall pH in the eastern
part of the country, ranging from about 4.4 to about 5.5
.3 kg/ha/yr
1.0 kg 'ha.'y
Figure 4: Nitrogen (NH -N and NO -N) in Precipitation
5.50
Rainfall is a significant source of pollution in industrial
communities and should not be overlooked either as a source of
nitrogen or as a vector for moving constituents through the soil
column. Not all residues deleteriously affect soils. Fertilizer, ni-
trogen and heavy metals are applied to the land on purpose;
crops need them for growth. Table II indicates the tonnage of
these elements annually applied to the soil. Residues can have a
beneficial effect on the soil and crops if applied and managed
judiciously. Please utilize the tonnages as an indication of
values, not as absolute values. They are presented to indicate an
overview.
Table II: Plant Nutrients Applied to the Land from Fertilizers
NITROGEN
PHOSPHATE
POTASSIUM
COPPER
IRON
MANGANESE
ZINC
MOLYBDENUM
9 MILLION TONS/YEAR
2.5 MILLION TONS/YEAR
3.5 MILLION TONS/YEAR
2500 TONS/YEAR
3300 TONS/YEAR
11,000 TONS/YEAR
15,000 TONS/YEAR
80 TONS/YEAR
Figure 5: Predicted pH of Precipitation over the Eastern United
States, 1965-1966
Phosphate
The factors affecting the uptake of phosphate are shown in
Table III. Phosphate is a peculiar sort of a compound in that
most of our soils already have appreciable quantities of phos-
phate associated with them; however, very little phosphate is
available to the plants. Farmers annually apply an excess of
phosphate onto the land to make up the deficiency, but the plant
has only a relatively small percentage of the phosphate that is
applied available for utilization. Apparently the needs of the soil
must first be satisfied before the phosphate is available to the
crops. A detailed treatment of phosphates is included in Mr.
Enfield's paper, "Phosphate Transport Through Soil", which is
included in these proceedings.
Heavy Metals
Table IV indicates the numerous factors affecting the fate and
uptake of heavy metals applied to soils. Type and form of heavy
metals are extremely important. The pH of the soil and sludge
have a tremendous impact on fate of heavy metals in residues. A
low pH will generally release heavy metals to the terrestrial en-
vironment. The micro and macro geology is likewise important.
Whether the soil has an ion exchange capacity of one mili equiv-
alent per 100 grams or 60 mili equivalents per 100 grams of soil
makes a tremendous difference on the amount of heavy metals
that can be adsorbed by the soil, the type of crops harvested, the
rate and total quantity of metals applied yearly and the trans-
port phenomena that can be expected.
-------
An Overview
Table III: Factors Affecting Fate of Phosphates
FORM OF PHOSPHOROUS
pH OF SOIL
TYPE OF SOIL
CLAY FRACTIONS
PERMEABILITY
PRIOR HISTORY OF PHOSPHATE APPLICATIONS
PLANT
CROPPING PRACTICES
PLANT TYPE
Table IV: Factors Affecting Uptake of Heavy Metals
TYPE AND FORM OF HEAVY METAL
pH OF SOIL AND SLUDGE
CLAY FRACTIONS IN SOIL
EXCHANGE CAPACITY OF SOIL
ORGANIC FRACTIONS IN SOIL
PERMEABILITY OF SOIL
CLIMATE
SOIL TEMPERATURE
SOIL MOISTURE
PLANT TYPE
PLANT AGE
TOTAL QUANTITY OF HEAVY METALS
INTERFERENCE OF OTHER HEAVY METALS
PAST HISTORY OF APPLICATION
AGE OF HEAVY METAL (FORM?)
PRESENCE OF HYDROUS OXIDES
Though the Conference will deal with heavy metals in
residues, similar constituents already abound in soils. Table V
indicates the various elements that are both inherently found in
soil and in plants, and which are necessary for the satisfactory
growth of particular crops.
Table VI shows the range of heavy metals in extracts actually
taken from municipal sludges and California soils. Note the sim-
ilarity of the ranges of these elements found in sludges and in
soils. One could ask the question, "If this is a reality then why is
there so much concern for the application of residues to the
soil?" Is it the application rate or dose, or is it the chemical form
of the constituent, or the end-product of the crop, or all of these?
Table V: Total Concentrations of Trace Elements Typically
Found in Soils and Plants
Cone, in Soils (HR/JE)
Element
As
B
Cd
Cr
Co
Cu
Pb
Mn
Ho
Hi
Se
V
Zn
!_/ Toxlclties
Common
6
10
0.06
100
8
20
10
850
2
40
0.5
100
50
listed do not
Range
0.1 -40
2 -100
0.01-7
5 -3000
1 -40
2 -100
2 -200
100 -4000
0.2 -5
10 -1000
0.1 -2.0
20 -500
10 -300
apply to certain
Cone, in Plant
Normal
0.1 -5
30 -75
0.2 -O.8
0.2 -1.0
0.05-0.5
4 -15
0.1 -10
15 -100
1 -100
1
0.02-2.0
0.1 -10
15 -200
accumulator plant
s (ue/e)
Toxic-'
-
>75
-
>20
-
>50
50-100
>10
>200
specie:
Table VI: Range of Heavy Metals in Extracts from Sludges and
California Soils (milligrams per liter)
Element
Ran
Sludges
Molybdenum 0.10 - 0.37
Copper
Zinc
Nickel
Cobalt
Lead
Vanadium
Boron
Cadmium
Silver
0.14 - 6.0
0.5 2.5
0.6 18
0.04 - 0.35
0.3 - 2.0
0.04 - 0.25
2.7 - 17.0
0.05 - 2.0
0.01 - 0.30
Soils
0.01 -22.0
0.01 0.20
0.01 0.40
0.01 0.09
0.01 0.14
0.01 0.30
0.01 1.20
0.10 -26.0
0.01
0.01
Median
Sludges
0.18
1.30
1.25
1.15
0.16
0.35
0.05
6.80
0.12
0.20
Soils
0.01
0.03
0.04
0.01
0.01
0.01
0.01
0.1
0.01
0.01
a) Derived from Bradford (1973), and Bradford, Bair, and Hunsaker (1971).
b) Data for soils represent 68 soil samples obtained from 30 soil series.
Data for sludges are obtained from six metropolitan Southern Cali-
fornia sludges.
c) From Page 1974.
In addition to concern for the heavy metals and elements as
they are applied to the soil, there is also concern for the
magnification of these constituents in biological vectors. When
there is direct uptake by biological factions, there is generally
subsequent biological magnification of these constituents by
each species that is higher in trophic level due to its feeding on
lower trophic level species, therefore ingesting the heavy metals
incorporated in the lower trophic level prey. Since the feeder
generally absorbs much more quantity than its prey, it magnifies
manyfold the heavy metal in its own metabolic processes, as
-------
An Overview
indicated in Table VII. If you've ever eaten a metallically-tasting
oyster, it's probably due to the concentration in the oyster of the
zinc extracted from the local waters. Usually the zinc doesn't do
much for the oyster, but does galvanize you for action.
Organic Constituents
Organic residues is the most difficult category to consider.
Besides the (literally) thousands of commercially available
organic compounds and the variability of these compounds,
industry annually develops some 5000 new organic compounds.
Many of these, like their predecessor compounds, eventually
terminate as a residue requiring disposal. The organic contents
of municipal residues are indicated in Table VIII.
Table VIII: Content of Municipal Residues
Leachate analyses of refuse disposed onto land indicates that
there is a significant fraction of organic compounds that are
released to the surrounding soil. Table IX indicates the percent
of the leachate emanating from a refuse pile that is organic.
Public Health Aspects of Land Disposal
An important aspect of land disposal of residues that have
contained pathogens is the potential effect on public health.
Sewage sludge carrying products of human metabolic waste are
suspect as vectors of pathogenic organisms (Table X). Industrial
and chemical sludges also contain large microbial populations
which may affect the environment (Table XI).
Questions that must be answered are: are pathogens present,
what are the species, what are their survival times and under
what conditions? Tables X and XII indicate the survival times of
various bacteria and pathogens in sludges and in various envi-
ronments. Table XII also indicates the travel distance and time
for various species of bacteria. With this information one can
possibly utilize site selection such that travel times offsite are in
excess of survival times of the various species of concern.
Groundwater
Once the contaminants have penetrated the soil and have
passed the root zone, one must become concerned about any
impact on the receiving groundwater, the travel time and path-
way of a contaminant in the groundwater regime and the even-
tual resurfacing either in a stream or from extraction via a
pumped or artesian well, as shown in Figure 6.
The variables that affect movement and fate of pollutants in
an unsaturated or saturated soil column are (1) depth to zone of
saturation, (2) frequency and amount of precipitation or water
application, (3) nature of zone of aeration, (4) nature of quality
of groundwater, (5) physical and chemical nature of aquifer, (6)
physical and chemical nature of pollutant, and (7) uses of
groundwater.
Table VII: Concentration Factors of Various Elements in the Marine Environment. The Factors were Based on the Live Weight of the
Organisms.
CELLULOSE
LIGNINS
PROTEINS
CARBOHYDRATES
FATS
GREASES
METALS
Concentration
Element Form in in Sea Water
Na
K
Cs
Ca
Sr
Zn
Cu
Fe
N1
Mo
V
Ti
Cr
P
S
I
Sea Water
Ionic
Ionic 3
Ionic
Ionic
Ionic
Ionic
Ionic
Participate
Ionic
Ionic-parti culate
not given
not given
not given
Ionic
Ionic
Ionic
ug/ liter)
107
.8 x 105
0.5
4 x 105
7 x 103
10
3
10
2
10
2
1
0.05
70
9 x 105
50
Algae
(noncalcareous)
1
25
1
10
20
100
100
20,000
500
10
1,000
1,000
300
10,000
10
10,000
Invertebrates
Soft
0.5
10
10
10
10
5,000
5,000
10,000
200
100
100
1,000
10,000
5
100
Skeletal
0
0
1,000
1,000
1,000
5,000
100,000
200
10,000
1
50
Vertebrates
Soft
0.07
5
10
1
1
1,000
1,000
1,000
100
20
20
40
40,000
2
10
Skeletal
1
20
200
50
30,000
1,000
5,000
0
2,000,000
-------
10 An Overview
Table IX: % Leachate from Refuse and Ash
Percentage leached under given conditions*
Material leached ^ 2
Permanganate value 30 min 0-039
Do 4 hr .060 0.037
Chloride .105 .127
Ammonia nitrogen .055 .037
Biochemical oxygen demand .515 .249
Organic carbon .285 .163
Sulfate .130 .084
Sulfide .011
Albumin nitrogen .005
Alkalinity (as CaCOs)
Calcium
Magnesium
Sod i urn
Potassium
Total iron
Inorganic phosphate
Nitrate
Organic nitrogen .0075 .0072
Conditions of leaching:
1. Analyses of leachate from domestic refuse deposited in standing water.
2. Analyses of leachate from domestic refuse deposited in unsaturated environment
and leached only by natural precipitation.
-------
An Overview
II
Table X: Survival Times of Organisms
Organism
Ascaris ova b)
). Typhosa
Cholera vibrios
Co ii form
Endamoeba
histolytica b)
Hookworm larvae '
Leprosptra
3olio virus
Salmonella typhi
Shigella
Tubercle bacilli
Typhoid bacilli
Type of
Medium application
Soil
Vegetables
Soil
Vegetables
Spinach, lettuce
Non-acid vegetables
Grass
Tomatoes
Vegetables
Soil
Soil
Soil
Polluted water
Radishes
Soil
Tomatoes
Soil
Soil
Sewage
AC c>
AC
AC
AC
AC
Sewage
Sewage
AC
AC
Infected Feces
AC
Infected feces
Infected feces
AC
AC
AC
Survival time
up to 7 years
27 - 35 days
29 - 70 days
31 days
22 - 29 days
2 days
14 days
35 days
3 days
8 days
6 weeks
15-43 days
20 days
53 days
74 days
2-7 days
6 months
7-40 days
a) After Pound and Crites 1973
3) Unlikely to move In the unsaturated zone
c) Artificial Contamination
Table XI: Densities of Various Bacteria in Different Chemical
Sewage Sludges2
Sampling
Date
6- 5-73
7-27-73
9- 5-73
Mean:
6- 5-73
7-27-73
9- 5-73
Mean:
6- 5-73
7-27-73
9- 5-73
Mean:
Heterotropbic Fecal
Bacteria Coliform Coliform
(20°C) MF MF
Alum Sludge (Point Edward)
8.8x 109 1.1 x 10° 7 x 104
7.5x 109 8.7 x 105 1 x 105
2.7x 1010 1.5x 10° 5 x 105
1.4x 1010 1.2x 10° 2.2 x 105
Iran Sludge (North Toronto)
2.7x lO10 2 x 104 1 x 104
2.2x 1010 9 x 104 3 x 104
4.1 x 1010 6 x 105 8 x 104
3 x 1010 2.4 x 105 4 x 104
Lime Sludge (Newmarket)
1.2x 1010 2.5 x 10° 4.5 x 105
3.2xl010 7.5 xlO5 1 x 104
l.lxlO10 3 x 105
1.8 x 1010 1.2x 10° 2.3x 105
Fecal
Streptococcus
MF
4 x
2 v
5 x
3.7x
2 x
1 x
3 x
1.1 x
3.4x
5 x
3 x
1.3x
104
104
104
104
104
104
103
104
105
104
103
105
a) Expressed as number of bacteria per 100 ml of sludge.
SLUDGE-DISPOSAL
AREA
GROUND-WATER
DIVIDE
IMPERMEABLE ROCK
f FLOW LINES
EQUIPOTENTIAL LINES
CONTAMINATED GROUND WATER
Figure 6: Flow in a Water-Table Aquifer (Humid Region)
NOTE: DRAWING NOT TO SCALE
CONSIDERABLE VERTICAL
EXAGGERATION
-------
12
An Overview
Table XII: Summary of Reported Distances of Travel of Pollu-
tion in Soil and Ground Water2
Nature of Pollution
Sewage polluted trenches
intersecting ground water
secting ground water
Sewage in bored latrines
intersecting ground water
Sewage in bored latrines
lined with fine soi 1
Sewage in bored latrines
intersecting ground water
Sewage in bored latrines
intersecting ground water
Coliform organisms in-
troduced into soil
Sewage effluents on
percolation beds
Sewage effluents on
percolation beds
Sewage polluted
ground water
Introduced bacteria
Chlorinated sewage
a) After U. S. Environmenta
b) For bacteria, the distance
Pollutant
Coliform bacteria
Cotiform bacteria
Anaerobic bacteria
Coliform bacteria
Coliform bacteria
Coliform bacteria
Coliform bacteria
Coliform bacteria
Bacteria
Bacteria
Bacillus prodigiosus
Fungi
Protection Agency
observed was the e
Observed
distance of Time of
travel k) travel
65 feet 27 weeks
232 feet
10 feet
50 feet
10 feet
35 feet
80 feet
regressed to 20 feet
50 meters 37 days
400 feet
150 feet
a few meters
69 feet 9 days
300 feet
(1973).
xtent of travel .
The total depth from the root zone to the water table, the
physical and chemical nature of the soil column and the amount
of carrier water that is applied to the land determine the spatial
distribution of the constituent species and their respective time
of travel to the aquifer. Similarly, the physical-chemical nature
of an aquifer and its hydraulic parameters will determine the fate
and travel times of the various constituents to the points of re-
surface (Figure 6). Obviously, any artificial extractions will have
to be superimposed onto the natural phenomenon.
A classical example of a discharge and its spatial distribution
is a disposal operation concerning the discharge of low-level
radioactive wastes shown in Figure 7. Following some initial
testing and field monitoring, the transport and distribution of
the hydraulic plume sphere of the discharged liquid and contam-
inants were predicted. Another example of the "detention"
capability of the soil is the operation at the National Reactor
Testing Station near Idaho Falls, Idaho. In that operation,
radioactive wastes have been injected into the ground since
1953. Other than tritium, radioactive metal isotopes have been
detected only a distance of less than one mile from the injection
well.
In other words, though heavy metals may eventually leach
into an aquifer, they may never become available. On the other
hand, organic, chelated, anionic, etc., wastes, may be insignifi-
•_Sile ol nilrole
decomposition plant
Predicted polh
of mqrolion
Figure 7: Migration of Radionuclides from Ground Disposal
Operations at Chalk River, Canada (Initiated in 1951)
cantly detained and therefore can be predicted to be available by
similitude to that determined for the water in the aquifer.
NOTE:
The papers following this overview paper will provide thean-
swers to the following questions: what do we know, what is the
confidence level, what information can we use now, how should
we use the information in the real world, and what needs more
research.
REFERENCES
1. Bernard, Harold, "Everything You wanted to Know About
Sludge But Were Afraid to Ask," Proceedings of the 1975
National Conference on Municipal Sludge Management and
Disposal, pp. 4-13, Anaheim, California, 1975.
2. Report to the National Commission on Water Quality on
the Environmental Impact of the Disposal of Wastewater Resid-
uals, Environmental Quality Systems, Inc., March 1976.
-------
Current EPA Guidance on Land Application of Municipal
Sewage Sludges
Robert K. Bastian and William A. Whittington
Municipal Technology Branch
U.S. Environmental Protection Agency
Office of Water Program Operations
Washington, D.C.
INTRODUCTION
No matter what technologies are applied, there is nearly
always something left over as a result of wastewater treatment.
While treating sewage to acceptable discharge quality levels,
various residual byproducts are inevitably produced, be it
sewage sludges from conventional physical/chemical and bio-
logical treatment plants, algae in pond systems, or crops from
land treatment systems.
The management of the conventional wastewater treatment
residuals (sewage sludges) is a twofold problem. The sludge
must be disposed of to complete the wastewater treatment
efforts and it must be done in an environmentally acceptable
manner.
The requirements of the Federal Water Pollution ControFAct
Amendments of 1972 (P.L. 92-500) emphasize the need to
employ cost-effective and environmentally sound waste man-
agement technology. At the same time its requirements for
improved wastewater treatment will result in a nationwide
increase in the production of greater quantities of sewage
sludges. The current estimated 5 million dry tons of sewage
sludge produced each year will more than double as a result of
upgrading the nation's publicly owned treatment works to meet
secondary treatment standards.
The "Sludge Bulletin"
Since as much as 40% of the construction costs for individual
treatment plants may be required to build adequate sludge man-
agement facilities and EPA may provide as much as 75% of the
capital funding required for these facilities under the Construc-
tion Grants Program, the Office of Water Program Operations
(OWPO) which manages the EPA Construction Grants Pro-
gram has prepared the proposed technical bulletin, "Municipal
Sludge Management: Environmental Factors." This bulletin
was published for public comment in the June 3, 1976 Federal
Register to assist the EPA Regional Administrators and their
staffs in evaluating grant applications for construction of pub-
licly owned sewage treatment works under Section 203(a) of
P.L. 92-500. The document also will provide designers and
municipal engineers with information for selecting optimal
sludge management options, but should not be construed to be a
regulatory document.
The sludge technical bulletin, as it is commonly referred to,
was developed over a three year period (that involved at least
three versions of the document) by an Agency workgroup with
substantial assistance provided by individuals from CEQ,
USDA, FDA and the Department of the Army. The bulletin
addresses only factors important to the environmental accepta-
bility of a particular sludge management option and does so in a
general manner to allow a maximum flexibility in its interpreta-
tion to meet varying Regional needs and site-by-site evaluation
considerations. Detailed information on costs and cost-
effectiveness analysis procedures, environmental assessment
and environmental impact statement procedures, pretreatment
guidelines and regulations, sample collection/preserva-
tion/analysis procedures, as well as in depth reviews of the
somewhat controversial potential impacts of land application
are or will be covered in additional supporting documents.
The sludge technical bulletin is based on current knowledge
and will be modified from time to time as any new regulations
are developed and additional information becomes available
from current and future research, development and demonstra-
tion projects. The document emphasizes land application alter-
natives since no Agency guidance has been issued on this option
in the past, and Agency guidance (and in some cases regu-
lations) is already available on the other major options—
incineration, landfill, and ocean disposal.
The proposed bulletin is divided into two distinct parts, one
including methods in which the sludge is utilized as a resource
and the second including those methods not utilizing the sludge
for any beneficial purpose. Appendixes are also available that
cover the preparation of environmental impact statements,
groundwater requirements of BPT, guidelines for the land dis-
posal of solid wastes, incinerator emission and performance
standards, and criteria established for ocean dumping of
municipal sewage sludges. Requirements for implementation
plans to reduce toxic materials and interim continuation permits
are discussed in light of established ocean dumping criteria. Dis-
cussion of incineration alternatives includes information on pre-
treatment programs, new source performance standards,
destruction of organic compounds, adequate ash disposal, and
monitoring plans. Discussion of sanitary landfill criteria covers
information on stabilization, EPA guidance for Federal facili-
ties, groundwater protection and monitoring plans. Land appli-
cation information includes discussions and suggestions for
acceptable stabilization techniques, application methods, appli-
cation rates, crops and monitoring plans, as well as precautions
for protection of public health, groundwater and surface waters.
During the development of the technical bulletin,
considerable disagreement surfaced on the topic of utilizing sew-
age sludges by application to agricultural lands. Although utili-
zation of sewage sludges as a resource to recover nutrients and
other benefits has been encouraged by P.L. 92-500 and various
advisory groups, the workgroup members and others involved
in developing the technical bulletin identified conflicting opin-
ions concerning the overall merits vs. potential hazards of apply-
ing sludges to cropland. Possible adverse effects upon the
human food chain (e.g., the potential for increasing human cad-
13
-------
14
Current EPA Guidance
mium intake) has remained a major concern expressed whenev-
er this practice is considered. The relative risks of applying
sewage sludges to croplands, when compared to other routes
through which these contaminants enter the human diet, have
yet to be determined. In addition, detailed trade-off analyses
comparing environmental impacts of land application vs. incin-
eration vs. landfill vs. ocean disposal have rarely been under-
taken in a manner that provided meaningful results.
The proposed bulletin placed primary reliance on FDA and
USDA to establish recommendations for acceptable levels of
heavy metals in food crops and best agricultural practices for
sludge application to agricultural lands. Until the necessary
food quality standards are established, strict regulation of food
crop production on sludge amended soils and required design
criteria necessarily will have to be based upon rather arbitrary
values. There are currently no basic standards for control of
most contaminants in sludges when traced through water, soil,
and plants, possibly to livestock, or ultimately to foods grown
on sludge amended soils. The proposed technical bulletin will
apply some control to the design and proper management of this
practice—at least to the extent that eligibility for capital funds
from the Construction Grants Program is concerned. However,
the new toxics substances and hazardous waste legislation may
encourage new regulatory efforts in the sludge management
area.
Land application of sludge remains a controversial area, but
most of the discussion centers around application to agricultur-
al lands, especially for food crop production. However, the use
of sludges on non-agricultural lands (strip mine reclamation,
parks, construction sites, etc.) or non-food chain crop produc-
tion (thus allowing for sod production, etc.) can be used without
much controversy to recover the nutrient and soil building value
of municipal sewage sludge.
Our experience shows that public acceptance and health
effects concerns are two of the major impediments to the use of
sludge in agriculture. Many of the concerns relate to possible or
potential problems that might occur under adverse situations,
rather than known dangers or hazards that can be dealt with
directly. Improved dissemination of information to the public
and governmental officials may help improve the picture, but
certainly will not remove all opposition or concern to the
practice.
One area of current comment is the need for increased pre-
treatment activities to improve the quality of many sludges that
could be applied to the land. Several studies have indicated that
industrial dischargers to publicly owned treatment works are a
significant source of sewage sludge contamination of heavy
metals and other toxic chemicals. However, both residential
wastes and stormwater runoff are known to contribute toxics to
municipal sewage and sewage sludges. It is also important to rec-
ognize that pretreatment of toxic substances is not equal to des-
truction of these materials. Toxic substances removed by indus-
try must be disposed of in sludges resulting from pretreatment
by those industries. In this regard, it is necessary to weigh the
potential impacts of discharge of toxics into the environment
(i.e., in sludge or effluents whether from municipalities or
industries).
In accordance with the rquirements of P.L. 92-500 and sev-
eral recent court rulings, EPA has embarked on an accelerated
program to develop (1) pretreatment standards for the most
significant polluting industries, and (2) standards pertaining to
the discharge of designated toxic pollutants. A concentrated
effort has been initiated to implement an effective Federal Pre-
treatment program to achieve compliance with the provisions of
P.L. 92-500. Additionally, the Agency has revised and is prepar-
ing to issue pretreatment guidelines to assist municipalities in
developing local pretreatment requirements.
Work Needed
Major technical needs to support the Construction Grants
Program involvement in municipal sewage sludge management
activities actually boil down to developing the best design crite-
ria and cost information for the available technology and the de-
velopment of innovative technologies for future implementa-
tion. With the current phase-out attitude toward ocean
dumping, we are dealing with providing guidance to the Regions
on the best available land-based technologies for sludge man-
agement rather than developing regulatory programs.
From our viewpoint, the work most urgently needed in the
municipal sewage sludge management field includes:
• Resolution of health effects issues
• Continued emphasis on innovative technology leading to
beneficial use and resource recovery
• Breakthrough in public acceptance
• Information dissemination to both design engineers/oper-
ators and elected and government officials.
-------
Runoff and Movement of Soil Particles
W. H. Wischmeier
Agricultural Engineering Department
Purdue University
Lafayette, Indiana
INTRODUCTION
The process of soil particle movement by raindrop impact and
runoff is soil erosion. An end-product of erosion is sediment.
Unabated erosion of agricultural lands destroys an essential and
irreplaceable natural resource. The U.S. Department of
Agriculture has been active in research of soil-erosion processes
and control for more than 40 years.
Sediment is the solid material, both mineral and organic, that
has been moved from its original location by the erosive agents.
Sediment impairs the quality of water resources in which it is
entrained and often degrades the location where it is deposited.
It may also carry pesticides, toxic metals, and plant nutrients
absorbed on the soil particles.4
Amount and Sources of Sediment
Sediment concentrations in rivers of the United States range
from 200 to 50,000 ppm, with occasional much higher concen-
trations.3 The amount of sediment moved by flowing water has
been estimated to average at least 4 billion tons per year, with
about 1 billion tons reaching major streams. A ton of sediment is
roughly one cubic yard in volume.
Estimates ascribe about 30% of this country's total sediment
to geological erosion—the erosion that occurs under natural
conditions of climate and vegetation, undisturbed by man and
machines.
About 20% of the total sediment comes from non-agricultural
sources like construction and mining activities, streambank and
channel erosion, and mass wasting from landslides. Per-acre
sediment production from these conditions may be more than
100 times as great as from cropland.
However, because of the large area involved, our 438 million
acres of cropland as a whole produce about 50% of our country's
total sediment. About half of our cropland is estimated to aver-
age between 3 and 8 tons of soil loss per acre per year. About
30% of it averages less than this, and about 20% averages more.
These soil-loss figures are amounts of soil moved from its gen-
eral position on the field slopes by runoff. The major portion of
this eroded soil may be deposited in places like fence rows, the
toe of field slopes, depressional areas, and elsewhere along the
path of the runoff before it reaches a major stream. On the aver-
age, less than one-fourth of the eroded soil reaches a river.
The Erosion Process
Soil erosion is a mechanical process that requires energy.
Much of this energy is supplied by falling raindrops. The impact
energy of the billions of raindrops falling in 30 minutes of a
Midwest thunderstorm may exceed 1,000 foot-tons per acre.
When raindrops strike bare soil at a high velocity, they shatter
soil granules and clods and detach particles from the soil mass.
Splash action moves the detached particles only short distances,
but shallow overland flow transports some of them directly
downslope and others to small depressions where the flow is
more concentrated and provides better transportation for them.
This type of soil movement is referred to as sheet erosion.
Very shallow sheet flow transports soil material that has been
detached by raindrop impact. When the runoff is impeded by an
obstruction, such as an across-slope strip of plant residues, its
flow velocity is reduced and deposition is likely to occur. Since
sheet erosion occurs rather uniformly over the field, it can
remove considerable soil without becoming readily evident, but
it cannot be ignored as a source of sediment.
Rill erosion is a process in which numerous small channels
only several inches deep are formed. In rill erosion soil particles
are detached by the shearing action of the flowing water and by
slumping of undercut sidewalls. The detached particles are
transported by a combination of suspension, saltation and rol-
ling. Suspended particles, which are mostly clay and fine silt,
may travel long distances before being deposited on the earth's
surface. The erosive potential of flowing water depends on its
velocity, depth, turbulence and type and amount of material
being transported. Rill erosion increases rapidly as longer or
steeper slopes increase runoff flow depth and velocity. Under
continued rainfall, sheet erosion continues between the rills and
feeds additional sediment to the rill flow.
A surface cover of plant residues or close-growing vegetation
like grass reduces erosion by protecting the soil from raindrop
impact and by reducing the flow velocity of the runoff and there-
by decreasing its capacity to detach and transport soil material.
However, small pieces of residue that are not anchored may be
carried away by the runoff as sediment.
Field and construction-site soil losses are usually from a com-
bination of sheet and rill erosion. Since sediment from these two
types of erosion comes from near the surface, it is more likely to
contain nutrient and pesticide contaminants than sediment de-
riving from gully, streambank or channel erosion, although the
amount of sediment from the latter sources may be much
greater.
The average-annual sheet and rill erosion from a particular
site, under given conditions, can be predicted by the universal
soil loss equation. This equation is a very helpful tool for evalu-
ating the severity of sheet and rill erosion hazards and determin-
ing the control practices needed to hold average-annual erosion
losses within some prescribed limit. It was designed for cropland
and construction areas, but can also be adapted for other condi-
tions.
15
-------
16
Soil Particles
The Universal Soil Loss Equation (USLE)
The universal soil loss equation computes average annual soil
loss from sheet and rill erosion on a particular tract of land as a
function of rainstorm characteristics, soil properties, topo-
graphic features, land use, and management practices. The
equation is
A =R K L S C P, where
A is soil loss in tons per acre,
Ris a measure of the erosive forces of the rainfall and
associated runoff,
K is a measure of the inherent susceptibility of the particular
soil to erosion,
L and S define the effects of length and steepness of slope on
soil loss per unit area,
C is the cover and management factor, and
P is the factor for supplemental erosion-control practices like
contouring and stripcropping.
The product of the first four factors (R,K,L and S) is the
inherent erosion potential at the site; that is, the soil loss that
would occur in the absence of any surface cover or management
practices. The last two factors reduce this potential loss to com-
pensate for effects of land use, management, and special prac-
tices.
Values of the six factors at a particular site can be obtained
from maps, graphs and tables published in Agriculture Hand-
book No. 28214 and several more recent supplemental reports
available from the Agriculture Research Service at Purdue Uni-
versity.2,8,9,20,12 Updating of AH-282 to include the more recent
material is under way and scheduled for completion in 1977.
The individual USLE factors are briefly discussed below to
give a concept of their significance, range of values, and sources
of specific-site values. Details on the derivation and limitations
of the equation and published factor-value sources have been
published.5,7,"
Factor R.
The capability of a rainstorm to erode soil depends on both
the total energy of the raindrops and the associated runoff. The
impact energy of a unit of rainfall increases with increasing drop
size, and dropsize generally increases with rainfall intensity up
to about 3 inches per hour. The total energy of a rainstorm can
be computed from recording-raingage data. The most accurate
available index of rainfall erosive potential is the El parameter.
For a given storm, this parameter equals the product of the
storm's raindrop energy and its maximum 30-minute intensity.
Storm El values can be summed to obtain seasonal or annual
values of the erosivity of a rainfall pattern.
Figure 1 shows average annual El values computed from 22-
year location rainfall records. The appropriate R value for the
USLE can generally be taken directly from this map, but in the
Northwest a factor for runoff from thaw and snowmelt must be
added to the El. The El lines on this map are the same as those
on the more detailed map given in Agriculture Handbook No.
282 except in the Coastal Plains of the Southeast and in the
eleven western states.
Factor K.
Some soils erode much more rapidly than others under identi-
cal conditions. Those high in silt or very fine sand erode most
readily. Erodibility decreases as the content of either clay or
sand particles (other than very fine) becomes greater. Soil
organic matter improves structure, infiltration, and aggregation
and decreases credibility, but fairly large soil aggregates may be
transported by high-velocity runoff. Profile permeability is
important because it influences runoff.
Factor K for a given soil is the expected soil loss per acre per
unit of El on a unit plot (a 9% slope 72.6 feet long, continuously
in clean-tilled fallow with tillage operations up-and-down
slope). Known values of this soil factor range from 0.03 to 0.69.
Table I indicates the general relation of K to soil texture, but
texture-class alone does not determine a soil's erodibility. More
accurate K. values can be obtained from the Soil Conservation
Service or by use of a published soil-erodibility nomograph.12
The nomograph graphically computes K for a given soil as a
function of the particle-size distribution, organic-matter con-
tent, structure, and profile permeability.
Table I: Indications of the General Magnitude of the Soil-
Erodibility Factor, K,
Texture class
Sand
Fine sand
Very fine sand
^oamy sand
Loamy fine sand
Loamy very fine sand
Sandy loam
Fine sandy loam
Very fine sandy loam
Loam
Silt loam
Silt
Sandy clay loam
Clay loam
Silty clay loam
Sandy clay
Silty clay
Organ
.0.52
K
0.05
.16
.42
.12
.24
.44
.27
.35
.47
.38
.48
.60
.27
.28
.37
.14
.25
1-The values shown are estimated avera
soil values. When a texture is near the
use the average of the two K values. For
graph or Soil Conservation Service K-valu
accuracy.
iu matter cont
27,
K
0.03
.14
.36
.10
.20
.38.
.24
.30
.41
.34
.42
.52
.25
.25
.32
.13
.23
ent
42
K
0.02
.10
.28
.08
.16
.30
.19
.24
.33
.29
.33
.42
.21
.21
.26
.12
.19
ges of broad ranges of specific-
borderline of two texture classes
specific soils, use of the nomo-
e tables will provide much greater
Factor L.
As runoff accumulates from a long slope, its detachment and
transport capabilities increase. On gradients of 5% or more, fac-
tor L =( X /72.6)0.5, where X =slope length in feet. On gradients
of 3% or less, the exponent is 0.3, and for 4% slopes it is 0.4.
Factor S.
For slopes not exceeding 20 percent, slope-steepness effect is
expressed by
S = 430 sin2 6 + 30 sin 9 + 0.43
6.574
where is the angle of slope.
Factors L and S are usually combined in a single topographic
factor, LS. Values of LS for various combinations of slope
length and uniform steepness are given in Table II.
When a slope is appreciably concave, convex, or irregular, it is
evaluated in segments that for practical purposes can be
assumed uniform. If the segments are taken at equal length, the
gradient of each segment is used with the overall slope length to
enter the slope-effect chart or Table II, and the segmental values
are weighted by the factors given in Table III to obtain an LS
value for the entire slope length.10
-------
Soil Particles
17
Figure 1: Average-Annual El Values
Table II: Values of the Topographic Factor, LS, for Specific Combinations of Slope Length and Steepness
Slope Length (Feet)
%
Slope
0.5
1
2
3
4
5
6
8
10
12
14
16
18
20
25
.065
.085
.133
.190
.230
.268
.336
.496
.685
.903
1.15
1.42
1.72
2.04
50
.080
.105
.163
.233
.303
.379
.476
.701
.968
1.28
1.62
2.01
2.43
2.88
75
.091
.119
.185
.264
.357
.464
.583
.859
1.19
1.56
1.99
2.46
2.97
3.53
100
.099
.129
.201
.287
.400
.536
.673
.992
1.37
1.80
2.30
2.84
3.43
4.08
150
.112
.146
.227
.325
.471
.656
.824
1.21
1.68
2.21
2.81
3.48
4.21
5.00
200
.122
.159
.248
.354
.528
.758
.952
1.40
1.94
2.55
3.25
4.01
4.86
5.77
300
.138
.180
.280
.400
.621
.928
1.17
1.72
2.37
3.13
3.98
4.92
5.95
7.07
400
.150
.196
.305
.437
.697
1.07
1.35
1.98
2.74
3.61
4.59
5.68
6.87
8.16
500
.160
.210
.326
.466
.762
1.20
1.50
2.22
3.06
4.04
5.13
6.35
7.68
9.12
600
.169
.222
.344
.492
.820
1.31
1.65
2.43
3.36
4.42
5.62
6.95
8.41
10.0
800
.185
.242
.376
.536
.920
1.52
1.90
2.81
3.87
5.11
6.49
8.03
9.71
11.5
1000
.197
.258
.402
.573
1.01
1.69
2.13
3.14
4.33
5.71
7.26
8.98
10.9
12.9
Factor C.
The greatest deterrent to erosion is soil cover, but a host of
other crop-system and management variables also greatly influ-
ence the ability of the soil surface to resist erosion. All of these
are combined in the cover-and-management factor, C. Agricul-
ture Handbook 282 presents a procedure for computing C for a
given cropping and management system in relation to the local
rainfall pattern. Regional tables of C values derived by this
procedure are available from state offices of the Soil Conserva-
tion Service. C values are also available for construction-site
conditions13 and for range, idle, and woodland conditions.8
-------
18
Soil Particles
The sample segment of a C-value table given in Table III illus-
trates the wide range of this factor and how much it can be influ-
enced by management decisions. This is extremely important to
planners because if any of the equation's factors is reduced with-
out changing the other factors, soil loss is reduced by the same
percentage. The table shows, for instance, that factor C for con-
tinuous corn can range from 0.54 to as low as 0.03 as a result of
management differences. A good grass-and-legume meadow
may at times allow substantial runoff, but sediment in the runoff
will be extremely low as indicated by the C-value of 0.004.
Table III: Factors to Adjust the LS-Chart Values for Successive
Segments of a Slope When Gradient is ^ 5%
Segment No.
(from top)
1.
2.
3.
4.
5.
No.
5
0.45
.82
1.06
1.25
1.42
of equal-length
4
0.50
.91
1.18
1.40
-
segments:
3
0.58
1.06
1.37
-
-
2
0.71
1.29
-
-
-
Factor P.
This factor accounts for effects of supplemental practices like
contouring or stripcropping. Its value depends on the land slope
and can be obtained from handbook tables.5,14 For estimat-
ing off-site sediment contributions, terrace systems are given a
P-value of 0.2 to compensate for deposition in the terrace chan-
nels and outlet.
Application of USLE.
Soil-loss computation with the USLE is a process of factor
selection and direct multiplication. To illustrate, assume a loca-
tion for which the El map shows that R=175, the credibility
nomograph shows that K=0.32, and the slope-effect table gives
an LS value of 1.2. If the field is cropped continuously to corn,
with the residues plowed-under in spring and use of conven-
tional planting methods, C=0.38. If the slope is less than 8% and
tillage and rows are on the contour, P=0.5.
The predicted average-annual soil loss is then the product of
these five values, or 13 tons per acre. Without contouring, it
would have been 26 tons. If the corn had been no-till planted in a
70% cover of shredded cornstalks, C would have been about
0.11 and the predicted soil loss with contouring would be less
than 4 tons per acre.
If a tolerance limit of T tons per acre has been prescribed and
the equation is written in the form C-T/ RK.LSP, the solution
will be the maximum value of C that can be tolerated on the
field. In the foregoing example, if T=5 T/A, the maximum
acceptable C would be 5/175 (0.32)(1.2)(.5) = 5/33 6 = 0.15.
Limitations of the USLE
The capabilities and limitations of the USLE were recently
reported in considerable detail." Several qualities of the
USLE factors are highly important if the equation is used for
prescribing or monitoring sediment-control standards:
1. The USLE predicts long-time average soil losses for speci-
fied physical and management conditions. In any specific season
or year, soil loss may be much more or less than this average—
because of fluctuations in rainfall, planting dates, or any of a
host of uncontrolled random variables which the equation takes
at their average values.
2. The USLE solutions are estimates, subject to usual experi-
mental and extrapolation error.
3. The USLE was developed for field-size areas. If it is used to
estimate sediment yield from a larger watershed, the drainage
area should be subdivided into relatively homogeneous subareas
for which representative values of the USLE can be defined.
This not only facilitates use of the equation; it also shows which
segments of the watershed require the most attention. Sediment
from gully, streambank, and channel erosion must be estimated
separately and added to the USLE estimate to compute the gross
erosion.
4. The amount of sediment discharged to large rivers is usually
less than one-fourth of that eroded from the land surface. The
fraction of soil loss from specific farm fields that is discharged to
a river varies widely with distance and with soil and watershed
parameters. The erosion equation does not credit deposition by
overland flow, and a dependable equation has not become avail-
able. The gross-erosion estimate must be multiplied by an ap-
propriate sediment-delivery ratio to compensate for deposi-
tion. The best available guides for estimating watershed
sediment-delivery ratios are given in Section 3 of the Soil Con-
servation Service National Engineering Handbook. However,
inability to predict delivery ratios for specific subareas with
desirable precision is presently perhaps the greatest weakness in
sediment estimation.
Planning on Large-Area Basis
For section 208 area planning, nonpoint pollution hazards
must to some extent be assessed on a large drainage-area basis,
and some helpful guides are pointed out below. Planners must
bear in mind, however, that the factors that cause erosion are
highly localized and erosion hazards are site specific. Most of
the sediment from a watershed may come from only a small por-
tion of the drainage area. For effective sediment control, the
subareas that are the major sediment contributors must be iden-
tified, and control-practice requirements must be so stated that
they apply to specific soil and topographic features within the
watershed rather than to the entire area.
Sources of information for broad-area appraisal of problems
and control planning include several methods of land classifica-
tion and a recently published manual.
The best source of information on soil characteristics and
associated land features is soil survey maps. These maps usually
include erosion class and slope class, but they cannot provide
detailed information on slope lengths, gradients and shapes
because these characteristics are extremely localized.
All land is also classified in one of eight capability classes,
with generally four subclasses in each.6 A capability unit is a
grouping of soils that are suited to the same kinds of cultivated
crops and have about the same responses to systems of manage-
ment. Capability subclasses recognize four kinds of limitations
or hazards: erosion, wetness, root zone, and climate. Acreage
data for each capability class and subclass are available by
states, land resource areas, and several land-use classifications.1
The conterminous 48 states have also been mapped as 156
major Land Resource Areas (LRA's). Each LRA consists of
geographically associated land units that are characterized by
particular patterns of soil, climate, water resources, land use,
and type of farming. The map and major characteristics of each
of the 156 LRA's are given in Agriculture Handbook No. 296.
In 1974, a panel of five Agricultural Research Service special-
ists was assigned the responsibility of compiling a two-volume
manual for control of water pollution from cropland.5 The
first volume presents systematic procedures to aid the user in
identifying specific potential problems and appropriate control
measures. The second volume reviews the basic principles on
which the instructions in Volume I are founded and provides
documentation of the information presented. Volume I was
-------
Soil Particles
19
printed in 1975 and copies are available from ARS and EPA.
The second volume is now at the printers and will be available
from the same sources.
Forty maps and charts and 21 tables included in the manual
summarize source data for pollution-control planning. The
runoff percolation, and erosion-hazard maps show relative data
for each of the 156 major land resource areas.
Neither actual nor potential erosion rates have been quantita-
tively mapped nationally because they are extremely localized.
The important soil, topographic, and cultural-practice details
are generally available only on a local basis. However, geo-
graphic trends in the levels of these parameters and in rainfall
pattern were estimated in relative terms and used to prepare a
map of "relative potential contributions of cropland to
watershed sediment yields." The mapped cropland-sediment
potentials reflect a combination of potential erosion and crop-
land density. A second map shows cropland acreage as the per-
centage of total land area for each of the 156 major land re-
source areas. A third map shows the distribution of cropland on
which soil erosion is the dominant limitation for agricultural
use, and a fourth shows percentages of rangeland. Since these
maps were published in color and are readily available,51 shall
not reproduce them for this paper but shall refer to the four
maps by number to illustrate their utility.
If the first map ranks the LRA in question as "low" but sam-
plings show excessive sediment in a stream, the fourth map is
consulted for likely contributions from rangeland. If that is also
low, non-agricultural sources should be suspected. Non-
cropland conditions indicative of high sediment-yield potential
that can be identified by observation include: gullies, residential
or commercial construction, highway construction, unstabilized
roadbanks or streambanks, surface mine spoils, and bare areas.
A rating of "moderate" on the first map combined with a
"high" rating on the third map would suggest widespread need
of relatively minor control practices. But a rating of "moderate"
on the first map combined with a relatively low rating on the
third map would suggest that only a small portion of the area
needs attention and that treatments on that portion may need to
be substantial. Conditions of high cropland-sediment yield that
can be visually observed include:
Long slopes farmed without terrace systems or diversions,
Runoff from upslope pasture or rangeland flowing across
cropland,
Rows and tillage up-and-down moderate or steep slopes,
No crop residues on the surface after new crop seeding,
Poor stands or poor quality of vegetation,
Intensively farmed land adjacent to a stream without an inter-
vening strip of vegetation.
The universal soil loss equation can be applied to sub-areas
suspected of being high sediment contributors, to obtain numer-
ical evaluations of the erosion hazards and to determine the spe-
cific combinations of control practices that would meet the spec-
ified goals.
Runoff from long rains at high intensities cannot be pre-
vented. But runoff can be reduced, and sediment content of the
runoff can be controlled by practices that have low values of the
USLE factors C and P. These practices rely largely on five
means of achieving erosion control: vegetation, plant residues,
improved tillage methods, residual effects of crops in rotation,
and mechanical support practices. The principal types of
erosion-control practices are described in the manual. They are
also listed in tabular form together with the highlights, merits,
and limitations of each.
Other tables list the major types of practices for control of
direct runoff, nutrient losses in runoff, and pesticide losses and
summarize the highlights, merits, and limitations of these prac-
tices.
Determining whether a nonpoint pollution problem may exist
and, if so, what measures may be taken to alleviate it most effec-
tively, involves a logical sequence of decisions. This sequence is
illustrated schematically in the manual by a series of flow charts.
The flow charts are designed as guides for assessing potential
erosion, nutrient, and pesticide problems and selecting physi-
cally feasible control practices for them. They are keyed to the
source-data maps and tables.
Sediment-Control Standards
Sediment-control standards that coincide with the soil-loss
tolerances already established for soil conservation have the dis-
tinct advantage that a farmer is in compliance if he follows a
conservation plan approved by his Soil Conservation District.
They are probably the most feasible starting points. But if the
initial standards fail to attain the desired level of water-quality
control, the next step should be variable standards that suit the
requirements of various local conditions rather than successive
lowerings of uniform soil-loss limits.
For controlling water pollution from nonpoint sources, a lim-
it on the potential amount of off-farm sediment may be more ap-
propriate than a limit on soil movement within field areas.
Sediment filters and traps would then also be possible options as
control measures. Soil texture and the locations of the cropland
relative to streams, lakes, reservoirs, or critical areas are also
important in determining how much field soil loss could be tol-
erated. Texture is important because collodial materials remain
in suspension longer than larger particles and are probably the
primary carriers of chemical compounds. Short-time peak sedi-
ment loads may be more important in pollution control than
they are in preservation of the land resource. Merits and lim-
itations of possible standards based on other than longtime-
average field soil loss are discussed at length in the manual on
control of water pollution from cropland.5
SUMMARY
About 30 percent of our country's total sediment is ascribed to
geological erosion, 20 percent to accelerated erosion on non-
agricultural lands, and about 50 percent to our 438 million acres
of cropland.
Soil-particles are detached by both raindrop impact and
runoff, but transportation of the detached particles is primarily
by runoff. Sediment from sheet and rill erosion comes from near
the surface and is more likely to contain nutrient and pesticide
contaminants than that from gully, streambank, or channel ero-
sion. Average annual soil loss from sheet and rill erosion on spe-
cific land areas can be predicted by the universal soil loss equa-
tion as a function of rainfall characteristics, soil properties,
topographic features, land-use, cultural management, and
erosion-control practices. The equation's factors and sources of
specific-location values of the factors are discussed, and use of
the equation is demonstrated.
The universal soil loss equation is designed to predict long-
time average soil losses, not specific events. It does not predict or
credit sediment deposition by overland flow and does not com-
pute gully, streambank or channel erosion. It was designed for
field-size areas but can be adapted for use in estimating
watershed sediment yields.
Helpful guides for Section 208 area planning on a drainage-
area basis are given in a recently published manual, "Control of
Water Pollution from Cropland." Flow charts have been devel-
oped to guide methodical assessment of potential erosion, nu-
trient, and pesticide problems and selection of physically feasi-
ble control practices for them.
For controlling water pollution from nonpoint sources, lim-
its on the potential amount of off-farm sediment may be more
appropriate than uniform limits on soil-loss rates from field
slopes.
-------
20
Soil Particles
REFERENCES
1. Dideriksen, R. I., and Grunewald, A. R. 1974. What is hap-
pening to our soil resources: the current inventory of rural land.
Proc. 29th An. Meeting Soil Conserv. Soc. Amer.:15-17.
2. Foster, G. R., and Wischmeier, W. H. 1974. Evaluating
irregular slopes for soil loss prediction. Trans. Amer. Soc. Agr.
Engin. 17:306-309.
3. Glymph, L. M., and Carlson, C. W. 1968. Cleaning up our
rivers and lakes. Agr. Engin. 49:590, 607.
4. Holt, R. F., Johnson, H. P., and McDowell, L. L. 1973.
Surface water quality. Proc. Natl. Conserv. Tillage Conf., Soil
Conserv. Soc. Amer.: 141-156.
5. Stewart, B. A., Woolhiser, D. A., Wischmeier, W. H.,
Caro, J. H., and Frere, M. H. 1975. Control of water pollution
from cropland. EPA-600/2-75-026, ARS-H-5-1. U.S. Environ-
mental Protection Agency. Vol I, 11 lp.; Vol. II in press.
6. U.S. Department of Agriculture. 1971. Statisbul. No. 461,
Appendix II, USDA
7. Wischmeier, W. H. 1972. Upslope erosion analysis. Envi-
ronmental Impact on Rivers, Chapter 15. Water Resources
Publications, Fort Collins, Colorado.
8. Wischmeier, W. H, 1975. Estimating the soil loss equation's
dover and management factor far undisturbed areas. Proc. Sed-
iment Yield Workshop, Oxford, Miss. ARS-S-40, U.S. Agr.
Res. Serv.
9. Wischmeier, W. H. 1973. Conservation tillage to control
water erosion. Proc. Natl. Conserv. Tillage Conf., Soil Conserv.
Soc. Amer.:133-141.
10. Wischmeier, W. H. 1974. New developments in estimating
water erosion. Proc. 29th An. Meeting Soil Conserv. Soc.
Amer.:179-186.
11. Wischmeier, W. H. 1976. Use and misuse of the universal
soil loss equation. Jour. Soil and Water Conserv. 31:5-9.
12. Wischmeier, W. H., Johnson, C. B., and Cross, B. V. 1971.
A soil erodibility nomograph for farmland and construction
sites. Jour. Soil and Water Conserv. 26:189-193.
13. Wischmeier, W. H., and Meyer, L. D. 1973. Soil erodibil-
ity on construction areas. Highway Research Board, Nat. Acad.
Sci. Spec. Report 135:20-29.
14. Wischmeier, W. H., and Smith, D. D. 1965. Predicting
rainfall-erosion losses from cropland east of the Rocky Moun-
tains. Agr. Handbook No. 282, U.S. Govt. Printing Off. 47p.
-------
Heavy Metals Contained in Runoff From Land
Receiving Wastes*
W. E. Larson and R. H. Dowdy
Agricultural Research Service
U.S. Department of Agriculture
and
University of Minnesota
St. Paul, Minnesota
INTRODUCTION
Heavy metals were measured in runoff waters and eroded sed-
iments on a sewage sludge treated area. The soils on the 8-ha
area are largely Typic Hapludolls on up to 10% slopes, on which
tile-outlet terraces were constructed. The domestic sludge was
low in metals and 28.5 mt/ha was applied over a 2-year period.
Concentrations of Zn, Cu, Cd, Pb, Ni, and Cr in the runoff
water were low and usually near detection limits with the
procedures used. Quantities of Zn and Cu were much greater in
sediment than in waters, except during snowmelt runoff and
during a storm after sludge application.
Heavy metals contained in waters from sludge-amended soils
are of concern because of possible deleterious effects on human,
animal, or plant life, from water ingested directly or from plants
grown on soils or sediments contaminated with these elements.
Because sewage sludge often contains significant quantities of
heavy metals, possible transport of metals by surface waters
from sludge-amended land is an important consideration.
Runoff water and the sediment from land areas are influenced
by many factors, especially: (a) rainfall characteristics (amount,
duration, intensity, and frequency); (b) watershed characteris-
tics (area, shape, slope gradient, length and orientation, and
type of drainage network); (c) ease of soil erodibility; (d) crop-
ping practice; and (e) land management and conservation
practices.
Proximity of the runoff-producing rainfall event to chemical
application date is a major influence on their concentrations in
runoff water (Leonard, Bailey, and Swank14). High concentra-
tions of chemicals contained in sludge may occur in runoff water
if the rainfall event occurs before surface-applied liquid sludge
dries on the soil surface.
Thus, to guard against direct runoff occurring from land
receiving sludge, an engineering system for temporarily retain-
ing the runoff water and sediment may be desirable. A backslope
terrace system with pipe drains has been used to slowly release
surface waters and to retain most of the sediments within agri-
cultural watersheds. Laflen, Johnson, and Reeve12 found that
average annual soil loss from four tile-outlet terrace systems in
Iowa was less than 840 kg/ha with less than 5% of soil erosion
between terraces. Sediment concentrations averaged between
* Contribution from the North Central Region, Agricultural Research
Service, U.S. Department of Agriculture, St. Paul, MN 55108, in
cooperation with the Minnesota Agricultural Experiment Station,
Paper No. 9676, Scientific Journal Series.
Cooperation and financial support from the Metropolitan Waste
Control Commission, St. Paul, MN, is gratefully acknowledged.
800 and 3850 mg/1. Annual average inorganic P concentrations
in surface runoff varied from 0.013 to 0.204 mg/1 P and average
annual inorganic N concentrations in surface runoff were 4 mg/1
N or less at three of four sites and 11 mg/1 N at one site (Hanway
and Laflen10).
Plan of Experiment
Physical Setup:
Eleven grassed backslope terraces with separate surface tile
inlets were constructed on a 16-ha watershed at the Rosemount
Agricultural Experiment Station of the University of Minnesota
near Rosemount, Minnesota (Figure 1) (Larson et al.,13). Port
Byron (Typic Hapludolls), Bold (Typic Udorthents), and Tal-
lula (Typic Hapludolls) are the dominant soil types. The soils
consist of 60 to 240 cm of a silt loam loess cap overlaying com-
pact glacial till. Slopes, after terracing, range from 2 to 10%.
The parallel terraces were constructed by a cut-and-fill tech-
nique leaving a 2:1 backslope that was seeded to bluegrass. The
terrace ridges were spaced 40 m apart and were designed to
impound a maximum runoff of 6.4 cm. Terrace channels were
graded to the inlet pipes, which were placed in natural depres-
sions (Figure 2). Solid PVC pipe individually connected each
inlet to sampling stations located below the bottom terraces. At
the sampling station flow from runoff water was measured and
sampled by automatic sample collectors.
Water drainage from the terraced area was stored in a
reservoir designed to impound a 100-year storm runoff (11.2
cm). Water stored in the reservoir is periodically used to irrigate
a 2-ha area designated for this purpose.
BACK SLOPE TERRACE
SURFACE INLET TO PIPE
. . r i L.L. r nur»i
! BELOW -
^—*"" ~~"~-— —-
RUNOFF WATER
Figure 2: Terrace and Surface Inlet
21
-------
22
Heavy Metals
DAM
TILE
NATURAL
BOUNDARY
RUNOFF WATER
IRRIGATION
TERRACES
SCALE
O 3D BD Rf~l 1Pn
• ™r ^
METERS
^"f^-*- FRESH WATER
- X "
N X__ x ' i
^-' OFFICE
WELL
Q GROUNDWATER WELL
O PIPE INLETS
_, RUNOFF
U SAMPLING STATIONS
— INSTRUMENT SITES
TERRACE #'s
IRRIGATION
LINE
B VALVE
Figure I: Rosemount Sewage Sludge Watershed (engineering design and sample collection locations)
-------
Heavy Metals
23
Sludge Storage and Application:
Liquid digested sludge (approximately 2 to 5% solids) from
four metropolitan wastewater treatment plants was hauled to
the watershed by tank truck and stored in two lagoons. The
combined capacity of the lagoons was 11,400 m3 (3,000,000 gal),
which will store approximately one-half of the annual sludge
production of the four treatment plants it served.
An underground pipeline extended from the sludge lagoons
across the watershed in the approximate midline of the terraces
for sludge application. At appropriate intervals riser pipes and
outlet valves were installed (Figure 1) to which a traveling irriga-
tion gun or subsurface injector could be connected by flexible
hose. A 1700-1/min (450-gal/min) positive displacement electric
pump was used for pumping sludge and runoff water to the tra-
veling gun. The traveling gun capacity was 1230 1/min (325 gal-
/min) with 7.0 kg/cm2 (100 psi) pressure and a 2.7-cm (1.05-in)
nozzle covering a 45-m arc, which can cover a 0.8-ha area with
1.3 cm of sludge in 1 hr.
About 1 to 1.5 cm sludge per application was applied to the
soil surface during 1974 and spring of 1975. In the fall of 1975,
sludge was applied on the corn area using a subsurface injector
mounted on a wide-tracked tractor. Spring-loaded shanks ap-
plied approximately 5-cm sludge/pass pass at 2.1 to 2.8 kg/cm2
(30 to 40 psi) pressure with a flow rate of 2400 1/min (640 gal-
/min). Runoff, soil loss, and metal concentrations are reported
for the corn area of the watershed, since grass was not estab-
lished until the summer of 1975. A control area not receiving
sludge was also included in the watershed (Figure 1).
Before application, sludge in the lagoon was mixed using a
tractor-driven pump with a flexible outlet hose connected to a
floating, movable raft. The intake of the sludge irrigation pump
was connected to another movable raft by a flexible suction
hose. The suction hose inlet was mounted on a movable boom,
which could be lowered from the raft to the bottom of the
lagoon.
During periods of sludge application, tile inlets were covered
to prevent sludge from entering the runoff drainage system. For
protection from aerosols, irrigations were carried out when the
wind speed was less than 8 km/ hr (5 mph).
Table I presents the times of application, amounts, and metal
contents of sludge applied to corn in 1974 and 1975. The concen-
tration of metals in the sludge at the four application times var-
ied depending on their solids content which varied depending
upon the degree of mixing in the lagoon. Low metal concentra-
tions reflect low solids content from liquid in the upper part of
the lagoon, while higher solids contents and higher metal con-
centrations occurred in the lower levels of the lagoon. In all cases
the concentrations of metals in the sludge were relatively low,
reflecting the domestic source of the sewage. The total amount
of metals applied to the corn area is given in Table II.
Runoff samples were collected during peak flow for any given
runoff event. The sediment fraction was removed immediately
by centrifugation and dried at 105 C. The water was acidified to
pH = 1 with distilled HNOsand stored for analyses.
Water samples were prepared for analyses as follows: i) 100 ml
of solution were evaporated to dryness in Vycor| glass crucibles,
ii) ashed at 450 C for 24 hr, and iii) extracted with 2N distilled
HC1. Sediment samples were: i) extracted with 4N distilled
HNO3(1:8 solid to liquid), ii) evaporated to dryness, and iii)
extracted with 1.IN distilled HC1. All water and sediment
extracts had a pH < 1. Glass-distilled deionized water was used
in all procedures. Metal concentrations of final extracts were
determined by atomic absorption spectroscopy utilizing a deute-
rium lamp background corrector.
Results and Discussion
Runoff water during 1975 totaled 2.1 cm for the control and
11.0 cm for the sludge-treated areas, and sediment loss totaled
130 kg/ha for the control and 700 kg/ha for the sludge-treated
areas. The amount of runoff from the control was smaller
because of differences in snowmelt runoff which reflects differ-
ences in snow cover.
The concentrations of trace metals in storm water runoff was
very low (Table III) and was usually lower than the concentra-
tions in surface waters found in rivers and lakes in selected bas-
ins during the period 1962-1967 (Table IV; Kopp and Kroner").
Except for the 6/11 /75 runoff event, Zn levels in runoff waters
from the sludge-treated areas were no greater than those from
t Mention of trade products or companies in this paper does not imply
that they are recommended or endorsed by the U.S. Department of
Agriculture over similar products of other companies not mentioned.
Trade names are used here for convenience in reference only.
Table I: Time, Amounts, and Metal Concentrations of Sludge Applied to the Corn Area.
Maximum Mean, of all Upper - , _/ „/ „
_, ... I/ , 2/ „. . . .2/ Northeast- Southeast- California^' Lake Erie
permissible— surface waters— Mississippi—
Zinc
Copper
Nickel
Lead
Chromium
Cadmium
5000
1000
50
50
10
64
15
19
23
9.7
9.5
— Maximum Permissible in Public Water
Pollution Control Administration, U
2 /Kopp and
Kroner (1967)
45
14
15
33
7
6
Supplies (Report
S. Department of
96 52
15 14
8 4
17 8
14 4
5 5
16
12
10
4
15
—
of the Committee on Water Quality Criteria, Federal
the Interior, Washington, B.C., April (1968))
205
11
56
39
12
50
Water
-------
24 Heavy Metals
the control area. However, the Zn level was four times higher
(125 /xg/1) in runoff water from the sludge-treated land versus
the control area for the 6/11/75 storm, which followed a 3.6-
mt/ ha sludge application on the soil surface the previous month.
Several workers have shown that the concentration of pesticides
in runoff increased as the time between applications and a runoff
event decreased (Leonard, Bailey, and Swank14). Chancy and
Giordano6 described this phenomenon as "reversion," where
metals added to soil slowly become bound in an insoluble form.
Copper and Cr concentrations in runoff water were the same
for control and sludge-amended areas (Table HI), although 3.2
kg of Cu and 1.38 kg Cr/ ha had been applied. Dowdy and Lar-
son9 added sludge containing 610 and 4550 jug/ g of Cu and Cr,
respectively, to a soil, found they were not appreciably taken up
by plants and presumably insoluble. Levels of Cd, Pb, and Ni in
runoff were usually below our detection limits of 1.0,6.5, and 3.5
jug/1, respectively.
Table III: Metal Content of Selected Storm Runoff and
Sediment from the Corn Area.
Table IV: Permissible and Mean Concentrations oT Metals in
Surface Waters from Selected Basins in the United States.
1974
Spring 1975
Total
Metal
Zn | Cu | Cd H
4.15 2.39 0.04 0.
1.37 0.79 0.02 0.
5.52 3.18 0.06 0.
1 Pb 1 Cr
11 2.27 8.96
OS 0.43 4.86
16 2.70 13.82
The low sediment yields (Table V) limited metal analyses to
Zn, Cu, and Cd. The 4N HNOsextractable metals from sedi-
ment, which approximates total metal content (Bradford et
al.3), were always higher from the sludge-treated areas than
from the control areas (Table IV). However, differences in sedi-
ment Zn and Cu levels were much less (control versus sludge ter-
races) for the major 4/27/75 storm of 9.3 cm rainfall, which sug-
gests that small, frequent periods of precipitation cause
proportionally larger losses of sludge-borne metals per unit
weight of sediment. During small runoff events, the less dense
sludge-derived organic colloids and associated metals are prob-
ably preferentially eroded. Also, a proportionally higher Zn
concentration was observed in sediment from sludge-amended
areas after sludge had been applied 1 mo (6/11 /75 storm) than
was observed for longer time intervals after application. This
observation is consistent with the runoff-water data and sug-
gested that Zn losses are more-highly dependent on
application /precipitation timing than are the loss of other
metals. Cadmium losses in sediment followed the same general
trends as those for Zn and Cu, but absolute losses were generally
two orders of magnitude less.
Control*
Sludge
Control
Sludge
Control
Sludge
Control
Sludge
Control*
Sludge
Control
Sludge
Control
Sludge
Control
Sludge
Control
Sludge
4/18/75
4/23/75
4/27/75
6/11/75
3/15/76
4/18/75
4/23/75
4/27/75
6/11/75
Metal
Zn C
7.0> 4
4.5 5
6.5 4
5.0 2
5.0 5
6.0 <2
36 7
125 8
22 36
25 41
32 12
57 36
30 11
53 18
30 11
36 14
60 12
320 38
u Cd 1 Pb 1 Nl 1 Cr
Runoff
.5 <0.8 <6.5 <3.5 6.5
.0 <0.8 <6.5 <3.5 11.0
.5 <0.8 <6.5 <3.5 5.5
5 <0.8 <6.5 <3.5 7.0
.0 <0.8 <6.5 <3.5 <4.5
5 <0.8 <6.5 <3.5 <4.5
5 1.0 <6.5 <3.5 <4.5
0 1.0 <6.5 4.0 5.5
<1.0 <6.5 <6.0 9.0
<1.0 <6.5 <6.0 10.7
Sediment
0.35
0.48
0.18
0.90
0.15
0.53
0.18
1.01
Snowmelt
Table V: Heavy Metals Removed in Selected Stormwater and
Sediment Losses from the Corn Area.
Terrace
Control
Sludge
Control
Sludge
Control
Sludge
Control
Sludge
Date
4/18/75"
4/18/75*
4/23/75
4/23/75
4/27/75
4/27/75
6/11/75
6/11/75
Sample
Utter
Sediment
Water
Sediment
Water
Sediment
Water
Sediment
Water
Sediment
Water
Sediment
Water
Sediment
Water
Sediment
Runoff
Hater
em
0.10
7.40
0.05
0.04
1.70
3.10
0.25
0.48
Sediment
kg/ha
22
18
5
4
640
16
30
Hetal
Zn | Cu [ Cd
70 45 —
260 700 —
3330 3700 —
990 625
35 20 —
145 55 —
15 10 —
215 70 —
•50 810 —
(3400)* —
1860 —
22.970 9000 —
925 190 27
960 190 3
6,000 375 48
9.525 1140 30
Represent! cxmlstiv* movnelt.
Sediaent yield missing,
loss between vster and
for previous events.
Value ••tlM
•ediaent vss i
I by sssuainf that partitioning of !
••a* for thl* runoff event •• It w
Table II: Metal Applied to Corn Through Spring 1975.
Season
Spring 1974*
Pall 1974+
Spring 1975+
Fall 1975$
Total
Applications
number
2
3
3
2
10
Sludge
cm
2.6
2.9
4.5
6.4
16.4
Solids
mt/ha
1.8
5.2
3.6
17.9
28.5
Metal
Zn Cu | Ni | Pb | Cr | Cd
542 374 15 219 500 5
610 330 15 360 1550 6
380 220 14 120 1350 4
1200 840 16 325 4360 9
Applied with tank wagon on soil surface
Applied with traveling gun on soil surface
4-Applled with subsurface injector
-------
Heavy Metals
25
Zinc and Cu losses were calculated for selected runoff events
for the spring of 1975 (Table V). Most of the eroded metals were
associated with the sediment from the sludge-treated areas dur-
ing the major storm of 4/27/75 where~8% of the Zn losses were
attributable to runoff waters. Curnoe8 noted similar findings for
runoff from poorly drained sandy clay loam soils with a 2 to 6%
slope. However, for snowmelt (4/18/75) and storms closely fol-
lowing sludge applications (6/11/75) proportionally more Zn
was associated with the solution phase.
Runoff losses of sludge-borne metals closely paralleled
phosphorus (P) runoff findings. Burwell, Timmons, and Holt4
reported that <6% of the total P loss in surface runoff was asso-
ciated with the solution phase from a continuous corn cropping
system. In contrast, total pesticide losses were generally great-
est in the aqueous phase as compared with sediment losses,
although pesticide concentrations may be greater in the
sediment phase (Leonard et al.14).
The total metal losses for the first four storms (Table V) rep-
resent an approximation of total metal losses in the spring of
1975 as a result of sludge applications of 6 mt/ha the previous
years. By performing these calculations for Zn and comparing
this value with that of total Zn applied (Table II),
a.
O
CJ
TJ
O
cc
CO
a
LU
20
15
10
(after Andersson and Nilsson, 1974)
llliticclay —->
soil 1
6
pH
8
10
Figure 3. Distribution of Cd Between Soil Materials and
Equilibrium Solution as a Function of p H. Solid / Liquid Ratio
was I/50,and 20M eq. of Cd was Added/g Soil Material
(after Andersson and Nilsson2).
The terraces and underground drain system reported herein
were designed to minimize runoff and sediment loss on this slop-
ing permeable soil landscape. Our preliminary data (Table V)
indicated the engineering design has limited sediment movement
to the drainage way to less than 700 kg/ha. Because the metals
are highly sorbed on the sediment, limiting sediment movement
off the land effectively limits total metal loss. The terrace outlet
tile perforated-entrance retains the heavier runoff events and
distributes them over longer time periods, which increases total
infiltration. An engineering design to control sediment move-
ment and slow runoff is desirable on all sloping lands.
Management of the soil and crop are important aspects of
controlling sediment and water loss. By using established con-
servation practices, these losses can be held to acceptable levels.
We used a chisel as the primary tillage tool and kept the soil sur-
face rough with some residue cover at all times when there was
no crop canopy. Sludge was applied only during periods of dry
weather.
-------
26
Heavy Metals
REFERENCES
I. Andersson, A., and K. O. Nilsson, 1974. Influence of lime
and soil pH on Cd availability to plants. Ambio. 3:198-299.
2. Bloomfield, C., and G. Pruden. 1975. The Effects of Aero-
bic and Anaerobic Incubation on the Extractibilities of Heavy
Metals in Digested Sewage Sludge. Environ. Pollut. 8:217-232.
3. Bradford, G. R., A. L. Page, L. J. Lund,and W. Olmstead.
1975. Trace element concentrations of sewage treatment plant
effluents and sludges; Their interactions with soils and uptake
by plants. J. Environ. Qual. 4:123-127.
4. Burwell, R. E., D. R. Timmons, and R. F. Holt. 1975. Nu-
trient transport in surface runoff as influenced by soil cover and
seasonal periods. Soil Sci. Soc. Amer. Proc. 39:523-528.
5. Busman, L. M. 1976. Characterization of metal chelates in
sewage sludge and their effects on metal availability in the soil.
M.S. Thesis. Univ. Minnesota, St. Paul, MN.
6. Chancy, R. L., and P. M. Giordano. 1976. Microelements
as related to plant deficiencies and toxicities. In L. F. Elliott and
F. J. Stevenson (ed.) Soils for management and utilization of
organic wastes and wastewaters. Soil Sci. Soc. Amer., Inc.,
Madison, Wise. In press.
7. Cheng, M. H., J. W. Patterson, and R. A. Minear. 1975.
Heavy metal uptake by activated sludge. J. Water Poll. Control
Fed. 47:362-376.
8. Curnoe, W. E. 1974. Runoff and erosion losses. Sludge
Handling and Disposal Seminar, Conf. Proc. No. 2, Toronto,
Ont., Environment Canada and Ontario Ministry of the Envi-
ronment, pp. 174-184.
9. Dowdy, R. H., and W. E. Larson. 1975. Metal uptake by
barley seedlings grown on soils amended with sewage sludge. J.
Environ. Qual. 4:229-233.
10. Hanway, J. J., and John M. Laflen. 1974. Plant nutrient
losses from tile-outlet terraces. J. Environ. Qual. 3:35-356.
11. Kopp, J. F., and R. C. Kroner. 1970. Trace metals in
waters of the United States. Fed. Water Poll. Contr. Admin.,
U.S. Dept. of the Interior, Cincinnati, Ohio.
12. Laflen, John M., H. P. Johnson, and R. C. Reeve. 1972.
Soil loss from tile-outlet terraces. J. Soil and Water Conserv.
27:74-77.
13. Larson, R. E., J. A. Jeffery, W. E. Larson, and D. R. Dun-
comb. 1976. A closed watershed for applying municipal sludge
on crops. Paper No. 76-2079. American Society of Agricultur-
al Engineers, St. Joseph, Michigan.
14. Leonard, R. A., G. W. Bailey, and R. R. Swank, Jr. 1976.
Transport, detoxification, fate, and effects of pesticides in soil,
water, and aquatic environments. Proc. Symposium, Land
Application of Waste Materials. Soil Conserv. Soc. Amer., Inc.,
Ankeny, Iowa. In press.
15. Silviera, D. J., and L. E. Sommers. 1977. Extractability of
copper, zinc, cadmium, and lead in soils incubated with sewage
sludge. J. Environ. Qual. 6: (In press).
16. U.S. Dept. of the Interior. 1968. Report of the Committee
on Water Quality Criteria. Fed. Water Poll. Contr. Admin.,
Washington, D.C.
-------
Heavy Metal Contents in Runoff
and Drainage Waters from
Sludge-Treated Field Lysimeter Plots
Thomas D. Hinesly and Robert L. Jones
Department of Agronomy
University of Illinois
Champaign-Urbana, Illinois
INTRODUCTION
Beginning in 1969, sludge from high-rate anaerobic digesters
at the Calumet and Stickney wastewater treatment plants near
Chicago was applied by ridge and furrow irrigation on 44
lysimeter plots measuring 3.05 by 15.25 m. Plots were arranged
in two banks of 22 plots each, referred to below as north and
south series. The lysimeter installation was located on a small
watershed on the Northeast Agronomy Research Center near
Elwood, Illinois. Each sludge treatment was replicated three
times on the native Blount silt loam soil occupying the small
watershed. Elliott silt loam soil was simulated in 10 lysimeters by
removing the surface one-foot depth of Blount silt loam and
replacing it with an equivalent depth of surface soil from Elliott
silt loam sites. Also, 10 lysimeter plots were constructed by exca-
vating the complete plot to a depth of 1.52 m and filling with
Plainfield sand. Tile drains were installed through the center of
each lysimeter plot and at the downslope end of each plot a fiber-
glass trough was installed to collect runoff water. Both drainage
water and runoff water were conveyed separately through PVC
pipes from the plots by gravity to the basement of an instrument
house. Tipping bucket equipment was provided to measure rates
and total volume of flows from each of the PVC pipes through
which either runoff or drainage water was delivered from the
plots to the instrument house basement. Automated sampling
equipment was provided for each of the PVC pipes and, for the
most part, a 400ml sample was collected from the second tip of a
tipping bucket and again on each sequential 42nd tip of a bucket
thereafter for the duration of a flow event. Thus, except for some
special studies, a sample represented the instantaneous water
flow that occurred between each 42 volumes of sequential flow.
The equipment was designed to take samples at more frequent
volumes of flow if desired, but the number of water samples col-
lected at the selected flow volume increments were all that could
be conveniently analyzed for heavy metals and several other
parameters not discussed here. Multiple-channel event re-
corders were used to record rates and total flow volume of
runoff and drainage waters relative to clock time. Detailed de-
scriptions of all the flow volume measuring, recording, and sam-
pling equipment, as well as the construction of lysimeter plots
were presented in an earlier report (Hinesly et al.5) and therefore
were not discussed in detail here.
Sludge loading rates and
some constituents of sludge
During the first two years of operations, the north series were
planted to soybeans and the south series to corn each year. Each
year sludge was applied as often as weather conditions permit-
ted, at rates of 2.54, 1.27, and 0.64 cm during and after the grow-
ing season. All plots received an annual application of KC1 at a
rate to supply 111 kg K./ha. Control plots received further addi-
tions of inorganic fertilizer to supply 333kg N/ha and 111kg
P! ha. Except for maximum sludge-treated plots of simulated
Elliott silt loam and Plainfield sand, sludge applications were
terminated on the north 22 lysimeter plots after November,
1973, at which time total sludge applications on a dry weight
basis were 223.8, 111.9, and 56.0 mt/ha on maximum, one-half
maximum, and one-fourth maximum sludge-treated plots,
respectively. On the south 22 lysimeter plots where applications
were continued, total sludge applications at the end of the 1975
application period amounted to 374.3, 187.2, and 93.6 mt/ha of
dry solids on maximum, one-half maximum, and one-fourth
maximum sludge-treated plots, respectively. Sludge loading
rates as well as quantities of several heavy metals applied annu-
ally on lysimeter plots as constituents of the sludge are presented
in Table I. Although, numerous data have been collected with
regard to the accumulations of heavy metals in soils and plant
tissues, the discussion here is limited to the effect of maximum
sludge loading rates, that is, those plots receiving incremental
applications of 2.54 cm, on concentrations of Fe, Mn, Zn, Cu,
and Cd in runoff and drainage waters.
Procedures and methods of analysis
As water samples were removed from automated sample col-
lection equipment they were sealed with a clean cap, labeled to
identify the date, plot number, kind of water (runoff or drain-
age) and sequence of sampling, after which they were placed in
specially prepared shipping crates for transfer to the laborato-
ries within a period of 30 hours after collection. All of the ele-
ments, Fe, Mn, Zn, Cu and Cd, were determined from a water
sample volume of 150 ml to which 2 ml of 6 TV HC1 (redistilled)
was added before evaporating to dryness on a steam-heated
sand bath, whereupon an additional 2 ml of 6 TV HC1 was added
to samples before evaporating to dryness for a second time. To
the dry residues, 10 ml of 1.0 TV HC1 (redistilled) was added and
refluxed with a watch glass cover for one hour (on a sand bath).
Samples were then diluted to 25 ml with deionized water, filtered
and analyzed by atomic absorption spectroscopy.
Data Handling
The scheme by which water data were collected and treated is
conceptually depicted in Figure 1. The physical aspects of col-
lecting water samples were briefly discussed above, but it should
be recognized that the samples were collected by a systematic
sampling plan, as contrasted to random and stratified random
sampling plans. That is to say, the first sample to be collected
during any runoff or drainage event was chosen and then a sam-
27
-------
28 Lysimeter Plots
pie was collected each time an amount of flow equal to 77.87 I
from the lysimeter surface and 26.04 I from the lysimeter drain-
age tiles was delivered to tipping buckets. To our knowledge,
this is the first and may still be the only field lysimeter equipped
to provide observations on discrete volumes of flow. In the past,
most sampling of lysimeter waters was conducted by a more or
less random sampling plan since all or some of the water from a
flow event was collected in containers from which subsamples
were withdrawn. In other cases, lysimeter flows were automati-
cally sampled on a unit time basis. Neither of these methods of
sampling provide opportunities for observing changes in water
quality parameters which change with flow rates or volumes of
flow.
Ranking
Check for Homogeniety
or Discrepancies
Handling Less Than
Detectable Values
Calculate Means
and
Standard Deviations
Figure I: Sequence for Sample Handling and Data Treatment.
Following the preparation and analysis of water samples, all
raw data required some calculations to transform them to stan-
dard reportable units. Most of the calulations were performed
on an electronic desk calculator. After the data were converted
to reportable units, they were punched on cards for storage on
disks. Then an IBM sort-merge routine was used to group data
stored on disks according to kind of water (runoff or drainage),
soil type, soil treatment, and period of time (an observation year
began on May 1 and continued to April 30). After checking for
discrepancies, the data were sorted again to arrange them in
order of lowest to the highest value of a measured metal concen-
tration (data ranking), where below detectable level values were
entered as one magnitude less than the lowest detectable level.
For example, Cd concentrations could not be detected below
0.001 ppm in water samples, but were entered as concentrations
of 0.0001 ppm as a means of includingthem in the total number,
N, observations. From the number of ranked data points, N, the
cumulative percent groups were calculated from lowest to high-
est utilizing the equation cumulative percent = n'/N 100 where
n j is the "i"th rank in a group of N data. Following the ranking
of data within groups and assignment of cumulative percentiles
to each datum, log-normal probability graphs were generated
using a computer program developed in FORTRAN language
and an IBM calcomp plotter. Where the number of observations
were equal or greater than 100, only a representative portion of
the plotted points were depicted on the graphs. That is, symbols
were not drawn for all plotted values but all data were included
in the graph. Less than detectable values were included in the
calculation of percentile distributions, but were not shown in
graphs. Data are log-normally distributed when their loga-
rithms fit a normal (Gaussian) distribution. For the log-normal
probability plots presented here, the ordinate is a logarithmic
scale of measured values and the abscissa is the probability scale
where the frequency of a measured value x is expressed as a per-
cent of occurrences equal to or less than the specified value.
Table I: Annual digested sludge loading rates and quantities of
heavy metals applied as constituents of sludge on maximum
sludge-treated Blount silt loam, Elliott silt loam, and Plainfield
loamy sand lysimeter plots before and after the growing seasons
of 1969 through 1974. Maximum sludge-treated plots were
located in north and south series of plots. Sludge quantities are
expressed as mt/ha dry weight and constituents are kg/ha dry
weight.
Sludge or
Constituent
Sludge
Fe
te
Zn
Cu
Cd
Location
South
North
South
North
South
North
South
North
South
North
South
North
Year
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
1969
1970
1971
1972
1973
1974
Blount
silt loam
19.33
36.15
104.69
32.19
58.75
70.73
16.37
52.78
57.83
44.38
61.08
terminated
455
1723
3967
1239
2041
3339
860
2734
2526
1825
2067
terminated
4
10
25
14
31
37
5
17
17
21
32
terminated
122
312
464
135
224
350
158
427
265
192
248
terminated
27
69
114
22
38
114
36
101
69
33
41
terminated
5.2
22.2
24.0
5.8
6.7
21.7
7.9
22.6
13.2
7.8
6.8
terminated'
Soil
Elliott
- -mt/ha
same
"
"
"
"
"
same
"
"
"
"
69.80
same
"
"
"
"
"
same
11
11
11
"
3190
same
"
11
"
11
same
"
"
11
"
30
same
"
"
"
11
fl
same
"
"
"
359
same
"
"
"
"
"
same
11
"
11
"
117
same
"
"
11
"
"
same
11
"
"
"
21.8
Type
Plainfleld
same
11
11
"
"
11
same
44.60
49.56
52.23
same
"
same
"
"
"
"
"
same
2187
2112
2401
same
"
same
"
"
"
"
same
13
15
25
same
"
same
"
"
"
"
"
same
341
224
229
same
"
same
n
"
"
"
"
same
81
57
39
same
same
"
11
"
11
"
same
18.1
10.7
9.3
same
"
-------
Lysimeter Plots
29
Many kinds of data exhibit log-normally distributed concentra-
tions and can be collectively described as members of a homo-
genous group by a single central value and a variance (Aitchison
and Brown,1), when log x concentrations or other appropriate
units of measured water quality parameters are plotted versus
the percent of the population of observations having val-
ues ^ x, on log-probability paper, a straight line is the best
fit to the plotted points when the distribution of the population
of values is log-normal. From this straight line, the estimated
geometric mean ft is given by the fiftieth cumulative percent
intercept while the estimated standard geometric deviation is
indicated by the slope of the line. The estimated standard geo-
metric deviation 8 is often calculated as the ratio between
values for cumulative percent intercepts of either 84.13 and 50 or
50 and 15.87, where either of the ratios corresponds to a whole
unit of standard deviation. More exactly, the estimated stan-
dard geometric deviation 8 is obtained from
log a = log XQ 841 3 - log XQ 50 = log -
X0.8413
X0.50
or
log a = log XQ 5Q - log XQ 1587 = log
X0.50
X0.1587
In the same way that 68.3, 95.5, and 99.7 percent of the area
under a normal distribution curve are located, respectively, in
the ranges x + s, x + 2s, and x + 3s (where "s" is the standard devi-
ation), confidence intervals for the estimation of the true value
of x can be calculated when the frequency distribution of log x is
presumed to be a normal "error" distribution. The difference in
calculating confidence limits is that the estimated standard devi-
ation for a normally distributed population is an added and sub-
tracted factor, whereas the antilog of the estimated geometric
standard deviation is a multiplying and dividing factor applied
to the antilog of the estimated geometric mean. More specifi-
cally, the intervals between /x x a and fj. 7 3, /u. x ff2
and /a 7 O 2, and /u. x o 3 and /x 7 o 3 are estimates of boun-
daries containing 84.1, 97.5, and 99.9 percent, respectively, of
the total number of observations from a log-normal distributed
population.
Rather than fit a straight line to all the log-normal probability
graphs, it was more convenient to compute both the estimated
geometric mean and geometric standard deviation values for the
several measured water quality parameters. The estimated geo-
metric means were calculated by the formula:
log xj + log
iog/2
+ log X3 . . . log xn
where x,, x,, x, . x are values of individual observations, or
1 2 .1 n
are values one magnitude less than the lowest detectable concen-
tration level. Using the computer generated estimated geometric
mean values, estimated geometric standard deviations were
computer calculated by the formula:
loga =
(logXi-logM)2/N-l
where x., i = I, 2, 3 ... n are the values determined from N
observations.
For purposes of comparison, other measurements of central
tendency, such as the median, arithmetic mean, and the stan-
dard deviation associated with arithmetic mean values for metal
concentrations were also calculated. It may be noted that in the
majority of cases, the arithmetic average or mean values are
larger than either the geometric mean or median values.
Although the geometric mean, like the arithmetic mean, is a
mathematical rather than a positional average, the way in which
individual measured values enter into the calculation of the geo-
metric mean causes less weight to be given to extreme values.
Thus, unless the numbers being averaged all have the same
value, the geometric mean will always be less than the arithmetic
mean. The median value is often an appropriate measure of cen-
tral tendency for random variables which are not symmetrically
distributed. Like the geometric mean, the median is not as sensi-
tive to a small number of extreme values as is the arithmetic
mean. The median is the mid-point value that divides in exactly
one half the random variable values that are higher or lower
values than itself.
Total numbers of observations, N, for each of the metal con-
centrations measured during each of the data years from water
samples collected from the various lysimeter plots were pre-
sented in Table II. Thus, with the availability of total number of
observations in Table II and estimated standard deviations for
metal concentrations presented later in this section, all the infor-
mation is provided for calculating standard error of means for
each of the measured water quality parameters.
Comparisons of geometric means of data from sludge-treated
plots to those from control plots were made to determine possi-
ble significant differences in metal concentrations due to treat-
ment. Data were prepared according to sampling year, kind of
water (runoff or drainage), soil type, and location of plots in
north or south series. Tests for significant differences between
two treatments within each group were based on the t-distribu-
tion. Values of t for a log-normal distribution were calculated
and significant differences in geometric means due to treatment
were determined from a table of the distribution of t for two-
tailed tests (Snedecor and Cochran," p. 549).
Discussion of Results
Relating the metal concentrations in runoff and drainage
waters to water quality criteria for public water supplies seemed
to be the most appropriate method for interpreting the findings
presented here. More specifically, metal concentrations were
compared to water quality recommendations for public water
sources published in Water Quality Criteria 1972, Ecological
Research Series, EPA-R3-73-0003, March, 1973.
Concentrations of Fe and Mn
in Runoff Waters
Iron and Mn concentrations in runoff water were plotted as
log-normal frequency distribution graphs. As examples of such
graphs, Figure 2 for Fe and Figure 3 for Mn concentrations in
runoff water samples collected in 1972-73 from Blount silt loam
plots were presented. Since most of the concentration levels
plotted as frequency distribution graphs could have been repre-
sented fairly well by a straight line, the population was inter-
preted as being log-normally distributed. Thus, geometric
means and geometric standard deviations for Fe and Mn con-
centrations in runoff waters, as presented in Tables III through
VIII, were most useful in assessing results obtained from the
study. In runoff waters from control and sludge-treated Blount
plots mean Fe contents ranged from 7.90 to 17.05 ppm and 5.73
to 10.86 ppm, respectively (Table III). Sludge treatments did not
significantly affect mean concentration of Fe and Mn in runoff
waters from Blount silt loam plots during the three years of col-
lecting samples. However, 1972-73 runoff waters from the
-------
30 Lysimeter Plots
sludge-treated Elliott silt loam plot, located in the north series,
had significantly (Pr<0.05) higher mean levels of Mn than were
found in runoff waters from control plots (Table IV). Both Fe
and Mn concentrations in runoff waters from the sludge-treated
Plainfield loamy sand plot, located in the north series, were
higher than levels in comparable waters from control plots dur-
ing 1972-73 (Pr<0.05) and 1973-74 (Pr<0.0!) (Tables V and
V1I1, respectively). Mean levels of Fe and Mn were also
increased in 1971-72 runoff waters from the Plainfield loamy
sand plot in the south series by sludge applications (Pr<0.01).
If Fe and Mn concentrations have been determined in runoff
waters by others, reported results were not discovered. Since
waters in streams are composed to varying degrees of watershed
runoff and drainage waters these kinds of reported data may be
the best available for comparison with results reported here.
Kopp and Kroner9 pointed out that Fe may occur in stream
waters in both bivalent and trivalent forms, but in natural
streams most of the ferrous-Fe is readily oxidized to ferric-Fe to
form insoluble Fe'(OH)3. Precipitates of Fe(OH)3 tend to ag-
glomerate, flocculate, and settle or be absorbed so that, in well-
aerated waters, soluble Fe concentrations seldom exceed 0.2
ppm. Like Fe, bivalent and trivalent Nn may be present in sur-
face waters but in natural streams most of it exists in the latter
oxidized state and concentrations are generally less than 0.02
ppm as a result of precipitate settling. Probably appreciable
amounts of Fe and Mn applied as constituents of digested sludge
were in more soluble bivalent states and pronounced differences
in mean concentrations of both elements in runoff waters from
control and sludge-treated plots might have been expected. Per-
haps the reason why differences in concentration levels were not
greater is the same as that for the relatively low NO3 -N contents
in runoff waters; that is, the more soluble forms were probably
carried into the soil with infiltrating water before runoff
occurred.
Table II: Number of observations, N, for heavy metal concentrations in runoff and drainage water samples collected from Blount silt
loam, Elliott silt loam and Plainfield loamy sand lysimeter plots during the period May 1 to April 30 for the data years 1971-72,
1972-73, and 1973-74. Water samples were collected from fertilized and irrigated control or check plots, 1, and maximum, 4, sludge-
treated plots located in either a north, N, or south, S, series from an instrument house.
Location
Water Quality and
Parameter Year Treatment
Fe 71-72 SI
HI
S4
N4
72-73 SI
Nl
S4
N4
73-74 SI
Nl
S4
N4
Mn 71-72 SI
Nl
S4
N4
72-73 SI
Nl
S4
N4
73-74 SI
Nl
S4
N4
Zn 71-72 SI
Nl
S4
N4
72-73 SI
Nl
S4
N4
73-74 SI
Nl
S4
N4
Runoff
Water
Drainage Water
Soil Type
Blount
silt loam
40
52
62
42
108
107
124
66
52
34
59
21
38
48
54
38
103
103
118
66
53
35
57
21
40
52
62
42
109
106
123
68
53
34
59
21
Elliott
silt loam
46
32
16
13
83
64
33
37
52
70
17
21
44
30
14
13
86
58
35
35
51
70
18
23
46
30
16
13
91
65
35
38
52
71
18
21
Plainfield
loamy sand
12
16
9
13
54
37
30
36
28
22
19
22
9
13
9
12
49
37
28
34
27
22
20
22
12
16
9
13
54
39
30
37
28
22
20
21
Blount
silt loam
41
15
85
51
126
57
167
314
183
147
268
336
37
14
47
45
127
57
166
321
181
147
269
330
42
14
91
52
124
53
167
324
179
134
265
321
Elliott
silt loam
24
31
35
16
43
216
93
24
51
259
111
33
20
28
17
13
44
221
91
24
51
257
110
32
24
30
33
15
43
221
92
24
48
250
107
30
Plainfield
loamy sand
96
50
3
14
392
214
31
140
209
177
43
126
91
49
3
18
406
230
31
140
203
171
43
123
99
50
3
21
403
233
31
143
208
170
43
124
-------
Lysimeter Plots
31
Table II (Cont.)
Runoff Water
Location
Water Quality and
Parameter Year Treatment
Cd 71-72 SI
Nl
S4
N4
72-73 SI
Nl
S4
N4
73-74 SI
Nl
S4
N4
Cu 71-72 SI
Nl
S4
N4
72-73 SI
Nl
S4
N4
73-74 SI
Nl
S4
N4
Blount
silt loam
26
32
44
20
109
100
124
66
54
34
59
21
40
52
63
42
108
105
124
68
54
34
59
21
Elliott
silt loam
27
18
13
9
90
63
35
37
52
71
18
22
45
32
16
13
91
64
35
37
52
71
18
23
Soil
Plainfield
loamy sand
6
13
8
7
54
36
30
34
28
22
20
22
12
16
9
13
53
39
30
37
28
22
20
22
Type
Blount
silt loam
36
14
83
43
128
57
167
324
184
147
269
337
40
14
91
53
126
57
168
321
180
147
270
332
Drainage
Elliott
silt- loam
16
28
31
14
44
222
93
24
51
261
111
33
24
30
37
16
44
222
94
24
51
261
111
33
Water
Plainfield
loamy sand
56
19
3
10
398
219
31
145
210
177
44
127
97
48
3
21
407
225
31
144
211
176
44
127
Fe
(ppm)
North
Control
Control
A
4-
2 10 30 50 70 90 9B
% of Samples
Figure 2: Frequency distribution of Fe concentrations during
the period May 1, 1972 to April 30, 1973, in runoff water sam-
ples collected from field lysimeter plots containing Blount silt
loam soil. Distributions of Fe concentrations are percent of
samples containing less than amounts shown on the vertical
axis. Median concentrations and standard deviations are pre-
sented in Table III.
30 10 70
% of Samples
Figure 3: Frequency distribution of Mn concentrations during
the period May 1, 1972 to April 30, 1973, in runoff water sam-
ples collected from field lysimeter plots containing Blount silt
loam soil. Distributions of Mn concentrations are percent of
samples containing less than amounts shown on the vertical
axis. Median concentrations and standard deviations are pre-
sented in Table VI.
-------
32 Lysimeter Plots
Table III: Arithmetic mean, 7T, standard deviation, s, median, M, geometric mean, ft , and geometric standard deviation, $, for Fe
concentrations as ppm in runoff and drainage water samples collected from Blount silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-73, and 1973-1974. Water samples were collected from fertilized and irrigated
control or check plots, 1, and maximum, 4, sluge-treated plots located in either a north, N, or south, S, series of plots adjacent to an
instrument house. Summary measurements include data from three replications in each of two series.
Location
Year and
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
32.81
67.28
16.21
27. 17
45.49
55.38
35.28
29.74
24.95
68.93
32.71
16.23
Runoff Water
s
54.09
162.12
24.04
65.83
85.54
77.21
66.30
50.00
49.83
119.41
46.96
17.13
M
10.47
8.95
5.98
4.10
12.04
21.52
9.63
8.53
7.57
11.17
10.91
10.40
u
12.43
12.91
6.45
5.73
13.15
17.05
9.63
10.86
7.90
12.72
10.17
9.36
a
4.25
6.52
3.99
5.70
5.24
6.26
5.96
4.14
4.76
8.18
5.65
3.41
X
3.49
1.75
1.47
3.15
3.23
1.34
1.15
1.18
5.85
4.57
1.06
0.56
Drainage Water
s
7.08
5.56
2.70
8.88
9.13
2.08
1.54
2.97
19.35
24.56
2.12
1.91
M
0.73
0.18
0.19
0.13
0.77
0.48
0.55
0.31
0.67
0.35
0.37
0.21
y
0.52
0.23
0.26
0.24
0.54
0.29
0.46
0.22
0.66
0.30
0.37
0.12
a
9.63
5.52
8.50
8.67
15.30
13.90
5.87
12.00
11.83
17.50
5.81
12.67
In Water Quality Criteria 19722 it was recommended that sol-
uble Fe should not exceed 0.3 ppm and soluble Mn should not
exceed 0.05 ppm in public water sources. Both of these metals
are undesirable at higher soluble concentrations because of their
effect on taste, staining of plumbing fixtures and laundered
clothes, and accumulation of deposits in distribution systems.
To compare runoff water quality with these recommendations,
the soluble proportion of the total Fe and Mn concentration
would need to be determined. Although much less than in drain-
age waters, runoff waters from lysimeter plots generally con-
tained NO j-N at some concentration level indicating that high
levels of reduced or soluble forms of Fe and Mn could not have
existed in the same water samples. Thus, the major portion of
the Fe and Mn transported by runoff water from lysimeter plots
was probably constituents of settleable suspended solids. If true,
most of the Fe and Mn in runoff water could be maintained at
very low levels by soil conservation structures and practices. In
the same way, sediment basins, as employed in connection with
the use of digested sludge to reclaim strip-mined lands, will
probably reduce Fe and Mn contents in surface runoff waters to
levels acceptable for discharge to streams used as public water
sources.
Concentrations of Fe and Mn
in Drainage Waters
Typical log-normal frequency distribution plots for both Fe
and Mn concentrations in drainage water were presented in Fig-
ures 4 and 5. Since these plots of data could be represented fairly
well by a straight line, Fe and Mn concentrations in drainage
waters were discussed in terms of geometric mean concentra-
tions and associated geometric standard deviations. First, data
reported in Tables 111 through VIII show that both Fe and Mn
geometric mean concentrations are severalfold less than levels
found in runoff water. As was the case for runoff waters, geo-
metric standard deviations associated with mean concentrations
of Fe and Mn in drainage waters indicate a considerable range of
contents. Geometric mean concentrations of total Fe ranging
from 0.12 to 0.66 ppm in drainage waters from Blount silt loam
frequently exceeded maximum recommended levels of soluble
Fe in public water sources, regardless of treatment.
Concentrations of total Mn ranging from 0.009 to 0.027 ppm in
drainage water from Blount silt loam (Table VI) were seldom
higher than recommended maximum soluble concentrations for
public water sources. Sludge treatments apparently affected Mn
levels in drainage waters in Blount silt loam soils only in
1971-72. During this one year, drainage waters from the maxi-
mum sludge-treated Blount silt loam plots in the north series
had higher Mn concentrations (Pr
-------
Lysimeter Plots
33
forms and absorbed on soil particle surfaces before they reach
drainage tiles. This speculation has not been verified by results
from analyses of samples from several soil depths because
amounts of Fe and Mn supplied by sludge applications,
although sizable in comparison to amounts of other metals, are
relatively low in comparison to native amounts of Fe and Mn in
soils. In their native state, surface horizons of Blountand Elliott
silt loam soils contain about 2 percent and Plainfield loamy sand
about 0.8 percent Fe. Additions to the surface of these soils
(Table 1, Fe and Mn) through 1974 amounted to 10 percent or
less of the native Fe contents and less than 5 percent of native
Mn contents.
Hem4 discussed the stability of solids and solutes and solubil-
ity of Mn as functions of pH and redox potential. He pointed
out that within the water stability region the three oxidation
states of +2, +3, and +4 must be considered for Mn. From his dis-
cussion it seems probable that the Mn supplied on the soil sur-
face as a constituent of anaerobically digested sludge was pres-
ent as Mn+2 but was readily oxidized to the Mn+4 state soon
after the sludge dried on the well aerated Plainfield sand plot
surface. Evidently, once Mn+2 is precipitated the Mn+2 to Mn+4
reaction goes very readily (Jenne6. As soon as Mn oxides are
present they act as a catalyst for the oxidation of sorbed Mn+2
causing Mn concretions to grow. Perhaps because of better aer-
ation these kinds of processes occurred more readily on the
sandy textured soil than on the heavier textured silt loam soils.
Such reactions must also occur to a lesser extent in the artifi-
cially drained, finer textured soils as evidenced by the fact than
Mn in drainage water was maintained at an unexpectedly low
level.
Table IV: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, p , and geometric standard deviation, & , for Fe
concentrations as ppm in runoff and drainage water samples collected from Elliott silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from two control plots and one maximum sludge-treated plot in each of
two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72 '
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
22.20
50.86
17.06
36.06
24.53
45.83
21.32
51.70
24.44
50.53
10.06
63.95
s
35.61
79.52
33.82
49.42
42.81
70.69
29.71
58.11
51.33
76.32
7.43
102.32
M
6.18
16.05
4.47
16.21
8.45
15.99
12.55
26.35
7.39
11.89
9.29
20.72
y
7.56
15.11
5.92
15.19
9.00
14.30
9.03
23.00
8.43
14.54
5.52
17.10
a
4.76
6.30
4.06
4.47
4.59
5.49
4.41
4.42
4.13
8.76
4.58
6.48
X
0.80
1.77
10.43
0.51
1.94
1.50
1.38
2.70
1.23
5.69
1.39
3.04
Drainage Water
s
1.04
3.77
20.44
0.52
2.22
3.24
3.14
7.95
1.48
21.09
3.71
6.03
M
0.21
0.10
2.39
0.26
1.10
0.36
0.51
0.51
0.70
0.42
0.31
0.44
H
0.29
0.20
2.00
0.21
1.04
0.32
0.43
0.46
0.42
0.40
0.37
0.29
o
5.14
9.10
7.72
6.14
3.50
10.16
7.42
10.39
11.02
17.37
6.68
30.34
Table V: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, £ , and geometric standard deviation, 8, for Fe
concentrations as ppm in runoff and drainage water samples collected from Plainfield loamy sand lysimeter plots during the period
May 1 to April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized
and irrigated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots
adjacent to an instrument house. Summary measurements include data from two control plots and one maximum sludged-treated plot
in each of two series of plots.
Location
Year and
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
3.92
23.37
46.46
25.21
33.09
46.21
44.28
70.53
36.23
24.01
24.47
105.86
Runoff Water
s
3.65
41.16
43.82
46.73
42.76
85.53
49.51
56.84
96.69
50.52
36.38
75.67
M
1.74
5.59
28.08
3.52
15.40
15.74
23.50
74.60
7.70
4.72
9.91
90.96
u
2.13
7.23
22.91
5.13
11.40
11.29
23.01
33.62
9.66
6.92
7.63
64.09
a
4.21
4.74
4.57
6.95
6.03
7.27
3.69
4.86
5.10
4.53
5.84
3.93
X
5.64
1.52
0.31
1.25
1.27
2.65
1.34
0.98
1.80
4.43
3.05
1.12
Drainage Water
s
11.07
1.74
0.48
3.93
3.05
7.57
3.05
2.04
3.87
7.84
16.88
5.23
M
0.56
0.72
0.04
0.07
0.20
0.41
0.47
0.18
0.58
1.78
0.17
0.20
y
0.56
0.80
0.10
0.09
0.28
0.32
0.46
0.20
0.51
1.81
0.05
0.16
0
11.23
3.38
6.50
8.11
7.04
13.41
6.67
8.10
9.14
4.43
37.00
10.31
-------
34 Lysimeter Plots
Although drainage water from sludge-treated plots contained not considered to be a problem by users (Miller et al.10). Like
total Fe contents which generally exceeded maximum soluble total Fe it seems probable that only a small fraction of the total
Fe contents recommended for public water supplies, it seems Mn measured in drainage water was in soluble form. Neverthe-
probable that only a small proportion of the total Fewasinsolu- less, even mean total concentrations of Mn in drainage waters
ble form. At any rate, total Fe contents in drainage water were did not generally exceed maximum recommended levels for sol-
not higher than soluble contents in many ground water sources uble Mn in public water supplies.
Table VI: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, £, and geometric standard deviation, 3 , for Mn
concentrations as ppm in runoff and drainage water samples collected from Blount silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from three replications in each of two series of plots.
Location
Year and
Treatment X
71-72
72-73
73-74
SI
Nl
SA
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.67
1.48
0.34
0.40
1.20
1.09
0.90
0.47
0.95
2.14
0.92
0.71
Runoff Water
s
1.12
3.84
0.54
0.83
2.32
1.68
2.09
0.78
2.09
3.88
1.29
2.03
M
0.21
0.15
0.09
0.09
0.25
0.41
0.19
0.16
0.16
0.35
0.24
0.21
u
0.24
0.25
0.13
0.12
0.30
0.35
0.20
0.18
0.21
0.40
0.24
0.19
a
4.38
6.53
4.10
4.45
6.24
5.52
6.47
4.19
6.10
7.81
6.81
4.48
X
0.050
0.038
0.019
0.039
0.035
0.020
0.022
0.041
0.228
0.071
0.069
0.083
Drainage Water
s
0.096
0.064
0.034
0.053
0.067
0.021
0.022
0.054
0.803
0.267
0.308
0.442
M
0.008
0.007
0.008
0.015
0.013
0.014
0.015
0.015
0.016
0.016
0.013
0.013
u
0.014
0.014
0.009
0.018
0.015
0.012
0.015
0.018
0.027
0.015
0.017
0.017
a
5.00
4.07
3.00
3.67
3.73
3.17
2.40
2.83
5.59
4.27
3.06
3.41
Table VII: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, /* , and geometric standard deviation, 3 , forMn
concentrations as ppm in runoff and drainage water samples collected from Elliott silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from two control plots and one maximum sludge-treated plot in each of
two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.55
1.23
0.52
0.83
0.68
0.95
0.57
1.00
1.04
1.85
0.25
2.56
s
0.93
1.84
0.86
0.91
1.28
1.65
0.77
1.10
2.46
2.94
0.19
5.75
M
0.18
0.26
0.16
0.36
0.23
0.26
0.75
0.42
0.19
0.47
0.23
0.52
u
0.19
0.40
0.20
0.43
0.22
0.22
0.25
0.52
0.21
0.47
0.11
0.57
5
4.40
5.16
4.23
3.78
4.82
6.75
4.22
3.56
7.05
6.71
8.06
6.45
X
.014
.032
.045
.026
.027
.018
.133
.032
.018
.234
.091
.052
Drainage Water
s
.017
.055
.053
.035
.034
.026
.233
.057
.018
.738
.270
.086
M
.007
.006
.024
.014
.014
.009
.038
.014
.012
.017
.015
.013
y
.008
.010
.030
.019
.013
.009
.051
.017
.012
.027
.025
.022
a
3.38
5.10
2.33
1.95
4.15
3.67
3.82
2.65
2.42
6.04
3.28
3.50
-------
Lysimeter Plots
35
Table VIII: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, ft , and geometric standard deviation, & , for Mn
concentrations as ppm in runoff and drainage water samples collected from Plainfield loamy sand lysimeter plots during the period
May 1 to April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized
and irrigated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots
adjacent to an instrument house. Summary measurements include data from two control plots and one maximum sludge-treated plot in
each of two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.07
0.53
1.01
0.42
0.79
0.99
0.73
1.35
1.23
1.01
0.80
2.68
s
0.05
0.82
1.04
0.67
0.94
2.00
0.90
1.17
3.39
2.02
1.55
1.74
M
0.04
0.12
0.87
0.07
0.38
0.19
0.33
0.60
0.16
0.19
0.18
3.06
y
0.05
0.17
0.59
0.10
0.25
0.26
0.33
0.64
0.26
0.29
0.17
1.56
'J
1.98
4.98
3.25
7.40
6.81
2.71
4.02
4.67
5.42
4.55
7.03
4.61
X
0.119
0.041
0.013
0.012
0.022
0.054
0.011
0.009
0.070
0.105
0.086
0.029
Drainage Water
s
0.225
0.045
0.017
0.035
0.061
0.158
0.014
0.010
0.196
0.199
0.486
0.148
M
0.014
0.017
0.006
0.002
0.007
0.012
0.006
0.005
0.018
0.043
0.007
0.007
y
0.018
0.021
0.005
0.003
0.007
0.016
0.006
0.005
0.023
0.046
0.008
0.008
0
7.61
3.57
6.60
4.00
4.00
4.06
4.50
3.20
3.61
3.46
3.63
2.75
Fe
(ppm)
10' L
c
30 50 70 90
% of Samples
(ppm)
10 30 50 70 90 98
% of Samples
Figure 4: Frequency distribution of Fe concentrations during
the period May 1, 1972 to April 30,1973, in drainage water sam-
ples collected from field lysimeter plots containing Blount silt
loam soil. Distributions of Fe concentrations are percent of
samples containing less than amounts shown on the vertical
axis. Median concentrations and standard deviations are pre-
sented in Table III.
Figure 5: Frequency distribution of Mn concentrations during
the period May 1, 1972 to April 30,1973, in drainage water sam-
ples collected from field lysimeter plots containing Blount silt
loam soil. Distributions of Mn concentrations are percent of
samples containing less than amounts shown on the vertical
axis. Median concentrations and standard deviations are pre-
sented in Table VI.
-------
36 Lysimeter Plots
Concentrations of Zn, Cu,
and Cd in Runoff Water
Like the Fe and Mn data previously discussed, frequency dis-
tribution graphs for concentrations of Zn, Cu, and Cd (Figures
6, 7, and 8, respectively) indicated that the results were best de-
scribed in terms of geometric mean concentrations and their cor-
responding geometric standard deviations.
Zn
(ppm)
10°
North
South
Control a
Control u
Maximum X
2 10 30 50 70 90 98
% of Samples
Figure 6: Frequency distribution of Zn concentrations during
the period May I, 1972 to April 30, 1973, in runoff water sam-
ples collected from field lysimeter plots containing Blount silt
loam soil. Distributions of Zn concentrations are percent of
samples containing less than amounts shown on the vertical
axis. Median concentrations and standard deviations are pre-
sented in Table IX.
Inspection of results reported in Tables IX through XVII
show that runoff water from lysimeter plots subjected to more or
less normal agricultural management practices had mean con-
centrations in the range of 0.05- to 0.15- ppm Zn, 0.016- to
0.038- ppm Cu, and 0.003- to 0.008- ppm Cd. Where sludge was
applied at rates several times amounts required to supply N and
P fertility, mean levels ranged from 0.10- to 0.63- ppm.Zn, 0.036-
to 0.170- ppm Cu, and 0.003- to 0.027- ppm Cd. Compared with
control plots, runoff waters from sludge-treated Blount silt loam
plots, located in the south series, had higher total Zn contents
during 1971-72 (Pr<0.05), 1972-73 (Pr<0.01). and 1973-74
(Pr< 0.01), while those located in the north series had higher lev-
els only in 1972-73 (Pr<0.05). Maximum sludge applications
caused increased levels of Zn in runoff waters from Elliott silt
loam plots during 1971-72 (Pr<0.05) and 1972-73 (Pr<0.01)
for the one located in the south series and 1971-72 (Pr<0.05),
1972-73 (Pr
-------
Lysimeter Plots
37
% of Samples
Figure 8: Frequency distribution of Cd concentrations during
the period May I, 1972 to April 30, 1973, in runoff water sam-
ples collected from field lysimeter plots containing Bount silt
loam soil. Distributions of Cd concentrations are percent of
samples containing less than amounts shown on the vertical
axis. Median concentrations and standard deviations are pre-
sented in Table XV.
These data for total Zn, Cu, and Cd contents in runoff waters
from lysimeter plots may be the only ones compiled, since none
were found in the literature. However, concentrations of these
metals have been measured in urban runoff waters by Klein et al.
(1974). They found average concentrations of 1.6-, 0.46-, and
0.025- ppm Zn, Cu, and Cd, respectively, in 35 grab samples of
runoff water collected at several areas in New York. Sampled
runoff events were produced by rains of various intensities fol-
lowing a variety of antecedent dry weather periods. With these
limited numbers of data it appears that urban runoff water can
be expected to contain much higher levels of Zn and Cu and
about the same concentration of Cd as runoff waters from agri-
cultural lands amended with digested sludge at rates which far
exceeded requirements of crop plants for supplementary N.
Acid extractable contents of Zn, Cu, Cd, Pb, Ni, and Hg in
runoff waters from plots treated with various rates of sludge
from the north Toronto sewage treatment plant were measured
during several events in 1972 and 1973 (Environment Canada3).
Unfortunately acid extractable metals in runoff waters were
reported only in terms of kg/ha total losses from lands with dif-
ferent slopes and concentrations of metals for comparison with
data presented here were not included. Nevertheless, heavy
metal losses were not noticeably increased by sludge application
except where sludge was eroded immediately following an appli-
cation. They concluded that "Heavy metal losses tend to be
small particularly where sludge of low metal content is used; the
mobility of metals when at higher concentrations cannot be pre-
dicted from the present work" (Environment Canada3 p. 23).
On the basis of consumer taste preference and because water
treatment processes may not remove appreciable amounts of Zn
and Cu it was recommended in Water Quality Criteria 19722
that levels in public water supply sources should not exceed 5
ppm of Zn and 1 ppm of Cu. It was further recommended, on the
basis of adverse physiological effects, that Cd levels should not
exceed 0.010 ppm. At the maximum recommended level for Cd,
human intake of the element from water would amount to about
one-third of the Cd intake as a constituent of foods. According
to these recommendations mean total concentrations of both Zn
Table IX: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, £ , and geometric standard deviation, ft, for Zn
concentrations as ppm in runoff and drainage water samples collected from Blount silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from three replications in each of two.series of plots.
Location
Year and
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.14
0.25
0.17
0.28
0.24
0.32
0.51
0.89
0.13
0.32
0.31
0.18
Runoff Water
s
0.19
0.51
0.16
0.54
0.42
0.41
1.10
2.19
0.25
0.60
0.40
0.16
M
.066
.103
.107
.101
.068
.141
.147
.137
.047
.070
.131
.141
V
0.08
0.10
0.12
0.12
0.09
0.12
0.16
0.21
0.05
0.07
0.14
0.13
a
2.96
4.07
2.30
3.57
4.44
5.28
4.46
4.76
4.49
7.93
4.19
2.61
X
0.028
0.031
0.052
0.047
0.026
0.027
0.085
0.026
0.041
0.038
0.048
0.025
Drainage Water
s
0.026
0.034
0.043
0.059
0.028
0.035
0.307
0.028
0.085
0.087
0.043
0.047
M
0.019
0.016
0.035
0.022
0.016
0.016
0.035
0.018
0.018
0.017
0.035
0.015
u
0.019
0.021
0.037
0.021
0.016
0.015
0.038
0.018
0.016
0.016
0.033
0.015
a
2.47
2.43
2.41
4.05
3.06
3.80
2.92
2.50
4.50
4.06
2.58
2.93
-------
38 Lysimeter Plots
and Cu were well within the tolerances for public water sources,
even in runoff water from maximum sludge-treated plots. How-
ever, mean total Cd concentrations in runoff waters from
sludge-treated plots may sometimes exceed allowable limits.
Results from several studies in the literature suggest that rec-
ommended maximum Cd levels in public water sources were
established at levels just slightly above those existing in many
streams. Furthermore, it can be seen from the results reported
here that the agreement between maximum recommended lev-
els and the upper range of mean total Cd concentrations
observed in runoff water from control plots were rather remark-
able. Thus, if a major proportion of Cd in runoff waters exists in
soluble forms, no increase over background levels would be
acceptable in water discharged directly to public sources.
However, if sludge applications were limited to rates that supply
N at optimum amounts for crop production and insoluble frac-
tions of total Cd present in runoff waters were retained on the
field or farm by erosion control practices or sediment basins,
soluble contents in runoff water reaching streams may not
exceed recommended maximum levels.
Table X: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, £ , and geometric standard deviation, 8, for Zn
concentrations as ppm in runoff and drainage water samples collected from Elliott silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from two control plots and one maximum sludge-treated plot in each of the
two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72
72-73
73-7A
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.10
0.24
0.17
0.54
0.15
0.25
0.29
0.60
0.13
0.27
0.13
0.58
s
0.12
0.26
0.19
0.55
0.24
0.35
0.39
0.60
0.19
0.33
0.11
0.79
M
0.06
0.16
0.09
0.40
0.06
0.11
0.15
0.39
0.06
0.10
0.10
0.29
-
0.06
0.15
0.12
0.34
0.06
0.09
0.15
0.34
0.07
0.11
0.10
0.25
G
3.49
2.75
2.31
2.90
4.30
5.30
3.07
3.18
3.26
4.10
4.39
4.00
X
.027
.025
1.837
.049
.037
.018
.418
.029
.031
.046
.197
.083
Drainage Water
s
.028
.027
3.155
.037
.037
.063
.444
.031
.036
.135
.263
.100
M
.018
.011
.675
.038
.022
.010
.243
.023
.020
.015
.134
.032
y
.018
.015
.688
.040
.022
.008
.271
.021
.019
.014
.151
.047
a
2.67
2.73
3.91
1.92
3.00
3.75
2.48
2.33
3.11
4.71
1.87
3.00
Table XI: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, ju , and geometric standard deviation, <5 , for Zn
concentrations as ppm in runoff and drainage water sample collected from Plainfield loamy sand lysimeter plots during the period May
1 to April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from two control plots and one maximum sludge-treated plot in each of
two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.07
0.15
0.45
0.35
0.21
0.26
0.37
0.95
0.27
0.15
0.20
1.26
s
0.06
0.18
0.49
0.41
0.25
0.32
0.42
1.21
0.73
0.20
0.27
1.82
M
0.05
0.07
0.26
0.10
0.12
0.13
0.22
0.51
0.07
0.08
0.10
0.94
0.06
0.09
0.31
0. 18
0. 10
0.13
0. 22
0. 44
0.08
0.09
0.10
0.63
5
2.07
2.88
2.37
3.54
4.21
3.27
2.96
4.14
3.91
2.60
3.73
4.14
X
0.032
0.020
0.016
0.019
0.036
0.047
0.019
0.039
0.022
0.038
0.029
0.023
Drainage Water
s
0.034
0.017
0.013
0.018
0.175
0.247
0.011
0.154
0.030
0.043
0.087
0.043
M
0.016
0.013
0.015
0.013
0.009
0.018
0.019
0.020
0.015
0.026
0.016
0.015
V
0.016
0.013
0.012
0.015
0.008
0.018
0.015
0.022
0.013
0.026
0.011
0.014
a
3.69
2.92
2.67
1.93
4.25
2.78
2.20
2.09
3.08
2.31
4.09
3.00
-------
Lysimeter Plots
39
Table XII: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, /x , and geometric standard deviation, d , for Cu
concentrations as ppm in runoff and drainage water samples collected from Blount silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from three replications in each of two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.032
0.061
0.053
0.069
0.059
0.081
0.145
0.204
0.046
0.099
0.099
0.060
s
0.032
0.133
0.027
0.096
0.093
0.159
0.253
0.468
0.073
0.174
0.099
0.032
M
0.021
0.022
0.049
0.039
0.022
0.025
0.055
0.063
0.019
0.031
0.058
0.051
u
0.022
0.025
0.047
0.043
0.028
0.034
0.072
0.080
0.024
0.033
0.063
0.054
a
2.36
3.28
1.70
2.47
3.46
3.65
2.88
3.07
2.83
4.48
2.67
1.57
X
0.011
0.006
0.022
0.014
0.013
0.015
0.104
0.016
0.019
0.023
0.040
0.017
Drainage Water
s
0.020
0.005
9.019
0.017
0.013
0.025
0.834
0.016
0.026
0.051
0.059
0.018
M
0.005
0.003
0.014
0.006
0.010
0.008
0.028
0.012
0.011
0.011
0.027
0.013
v
0.007
0.004
0.016
0.007
0.009
0.008
0.026
0.011
0.012
0.012
0.029
0.013
a
2.43
2.25
2.31
3.29
2.56
3.25
3.00
2.73
2.67
2.67
2.17
2.31
Concentrations of Zn,
Cu and Cd in Drainage Water
As may be seen in example Figures 9, 10, 11 concentrations of
Zn, Cu, and Cd found in drainage water could have been repre-
sented fairly well by a straight line when they were plotted in a
log-normal frequency distribution array. Therefore, results were
most appropriately interpreted in terms of estimated geometric
mean concentrations and geometric standard deviation. These
statistical measurements are presented in Tables IX through
XV11 along with values for other methods of expressing central
tendenices and variations of data. As previously mentioned, a
great deal more confidence was placed on the results from the
Blount silt loam than on results from other soil types. Mean con-
centrations of metals in drainage water from Blount silt loam
plots were compilations of results from three replications of
each treatment in each series, whereas sludge-treated plots of the
other 2 soil types were not replicated. It was also mentioned ear-
lier that during the early years of operations the Elliott silt loam
sludge-treated plot, located in the south series of plots (S4, Table
X), often leaked sludge into the tile system following an irriga-
tion. Thus very little significance is attached to the fact than Zn
and other heavy metal elements appeared in relatively high lev-
els in drainage waters from this particular plot.
Geometric mean concentrations of total Zn in drainage
waters from Blount silt loam control plots varied from 0.015 to
0.021 ppm and from 0.015 to 0.038 ppm in samples from sludge-
treated plots on the same soil type. Drainage waters from Elliott
silt loam had mean Zn contents that ranged from 0.008 to 0.022
ppm in samples from control plots and from 0.021 to 0.047 ppm
in samples from the sludge-treated plot in the north series. Mean
Zn contents in drainage waters from the sludge-treated Elliott
silt loam plot in the south series were 0.688, 0.271, and 0.151
ppm in successive years from 1971 to 1974. These values indi-
cated that the amount of direct contamination of drainage water
by sludge decreased with the succeeding years, but nevertheless
remained at levels too high to have occurred by simple leaching
processes. Plainfield loamy sand discharged drainage water
from control plots having mean Zn concentrations ranging from
0.008 to 0.026 ppm as compared to mean Zn concentrations
ranging from 0.011 to 0.022 ppm in similar waters from sludge-
treated plots. Levels of Zn in drainage waters from sludge-
treated Blount plots in the south series were significantly higher
(Pr 0.01) during all three years as compared to levels in similar
water samples from control plots, whereas, Zn contents of
waters from plots in the north series did not differ with treat-
ment. Differences in Zn contents of drainage waters from Elliott
silt loam plots were similar to those from Blount silt loam plots.
Maximum sludge treatments increase Zn contents in drainage
waters from the Elliott silt loam plot in the south series
(Pr 0.01) during all three years, but not in the north series.
Only in 1972-73 did sludge treatments on Plainfield loamy sand
cause increased Zn levels in drainage waters (Pr 0.05). During
this one year waters from Plainfield loamy sand plots in both the
north and south series were similarly affected by sludge treat-
ments.
2 10 30 50 70
% of Samples
Figure 9: Frequency distribution of Zn concentrations during
the period May 1, 1972 to April 30,1973, in drainage water sam-
ples collected from field lysimeter plots containing Blount silt
loam soil. Distributions of Zn concentrations are percent of
samples containing less than amounts shown on the vertical
axis. Median concentrations and standard deviations are pre-
sented in Table IX.
-------
40
Lysimeter Plots
Cu
(ppm) [
% of Samples
Figure 10: Frequency distribution of Cu concentrations during
the period May I, 1972 to April 30,1973, in drainage water sam-
ples collected from field lysimeter plots containing Blount silt
loam soil. Distributions of Cu concentrations are percent of
samples containing less than amounts shown on the vertical
axis. Median concentrations and standard deviations are pre-
sented in Table XII.
During the three years, mean Cu concentrations in D drain-
age waters from Blount silt loam control plots ranged from
0.007 to 0.012 ppm in samples from the south series and from
0.004 to 0.012 ppm in samples from the north series. Drainage
waters from sludge-treated Blount silt loam plots had mean Cu
concentrations that ranged from 0.016 to 0.029 ppm in those col-
lected from plots in the south series and from 0.007 to 0.013 ppm
in samples from the north series. Mean Cu contents of drainage
waters from Elliott silt loam control plots (Table XIII) ranged
from 0.007 to 0.010 ppm in samples from the south series and
from 0.005 to 0.010 ppm in samples from the north series. Dur-
ing successive years mean Cu contents were 0.074, 0.044, and
0.039 ppm in drainage waters from the sludge-treated Elliott
plot in the south series which initially produced water directly
contaminated with sludge as was discussed earlier. On the other
hand, drainage waters from the sludge-treated Elliott silt loam
plot in the north series contained mean Cu concentrations rang-
ing from 0.013 to 0.022 ppm. Thus, in the absence of direct con-
tamination Cu levels in drainage water from sludge-treated Elli-
ott silt loam were similar to those observed in samples from
Blount silt loam. Regardless of location in north or south series,
Plainfield loamy sands (Table XIV) produced drainage waters
having mean Cu contents ranging from 0.005 to 0.007 ppm in
samples from control plots and from 0.007 to 0.015 ppm in sam-
ples from sludge-treated plots. Sludge applications significantly
(Pr«=O.OI) increased mean Cu levels in drainage waters from
Blount silt loam plots located in the south series during all years
and during the 1972-73 data year (Pr«0.05) in samples from
plots located in the north series. Sludge treatments caused
higher levels of Cu in d rainage water from Elliott plots during all
years (Pr«=r0.01), but for reasons mentioned above evidence is
uncertain with regard to the data collection from the sludge-
treated plot in the south series. Mean Cu levels were significantly
(Pr«c0.01) increased by sludge applications in drainage waters
from Plainfield loamy sand plots in the south series during
1972-73 and 1973-74 and north series during 1971-72 and
1972-73.
Table XIII: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, /x , and geometric standard deviation, a , for Cu
concentrations as ppm in runoff and drainage water samples collected from Elliott silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from two control plots and one maximum sludge-treated plot in each of
two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.027
0.041
0.054
0.123
0.030
0.055
0.071
0.160
0.068
0.043
0.157
s
0.026
0.040
0.046
0.120
0.042
0.078
0.082
0.160
0.080
0.029
0.196
M
0.019
0.026
0.037
0.090
0.018
0.023
0.042
0.097
0.024
0.031
0.038
0.073
V
0.018
0.026
0.043
0.081
0.017
0.027
0.048
0.098
0.034
0.038
0.036
0.085
a
2.44
2.73
1.91
2.72
3.29
3.26
2.26
2.81
3.68
2.92
1.86
3.07
X
0.009
0.006
0.378
0.019
0.013
0.010
0.073
0.022
0.011
0.022
0.054
0.025
Drainage Water
s
0.007
0.005
0.819
0.017
0.011
0.023
0.141
0.014
0.007
0.083
0.093
0.015
M
0.007
0.005
0.048
0.011
0.009
0.007
0.036
0.018
0.011
0.010
0.033
0.023
V
0.007
0.005
0.074
0.013
0.010
0.005
0.044
0.018
0.008
0.010
0.039
0.022
a
2.14
2.20
5.20
2.38
2.00
4.00
2.25
1.89
3.00
3.30
1.85
1.73
-------
Lysimeter Plots
41
Table XIV: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, ft, and geometric standard deviation, O , for Cu
concentrations as ppm in runoff and drainage water samples collected from Plainfield loamy sand lysimeter plots during the period
May 1 to April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized
and irrigated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots
adjacent to an instrument house. Summary measurements include data from two control plots and one maximum sludge-treated plot in
each of two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.018
0.033
0.166
0.083
0.049
0.053
0.096
0.220
0.071
0.045
0.060
0.265
s
0.009
0.043
0.113
0.082
0.058
0.078
0.099
0.257
0.159
0.059
0.062
0.336
M
0.016
0.015
0.137
0.071
0.024
0.034
0.072
0.132
0.019
0.022
0.031
0.193
Vi
0.016
0.016
0.131
0.057
0.025
0.024
0.066
0.125
0.029
0.027
0.041
0.170
0
1.69
3.44
2.13
2.44
3.84
4.54
2.45
3.05
3.07
2.59
2.29
2.65
X
0.010
0.005
0.008
0.009
0.009
0.007
0.011
0.009
0.011
0.013
0.025
0.017
Drainage Water
s
0.010
0.005
0.006
0.006
0.018
0.007
0.007
0.006
0,. 017
0.017
0.040
0.035
M
0.008
0.004
0.005
0.008
0.005
0.005
0.009
0.007
0.008
0.008
0.013
0.010
V
0.007
0.004
0.007
0.007
0.005
0.005
0.009
0.007
0.006
0.008
0.015
0.009
"j
2.43
2.00
2.00
1.86
3.20
2.80
2.11
2.29
4.00
3.13
2.40
3.22
10°
Cd
(ppm)
10-'
North
South
Control
Moximum
Control
Maximum X
2 10 30 50 70 90 98
% of Samples
Figure 11: Frequency distribution of Cd concentrations during
the period May 1, 1972 to April 30,1973, in drainage water sam-
ples collected from field lysimeter plots containing Blount silt
loam soil. Distributions of Cd concentrations are percent of
samples containing less than amounts shown on the vertical
axis. Median concentrations and standard deviations are pre-
sented in Table XV.
Drainage waters from sludge-treated Blount plots in the south
series and significantly lower concentrations of Cu in 1971-72
(Pr^ 0.01) than in all later years. Similarly treated Blount plots
in the north series yielded drainage waters having significantly
higher concentration of Cu in each succeeding year as compared
to the year before, although 1973-74 waters were significantly
higher in Cu content than 1972-73 waters at only the 5 percent
probability level. Drainage waters from the sludge-treated Elli-
ott plot in the north series had higher Cu levels in 1973-74 than
in 1971-72 (Pr-eO.05). Copper levels were also higher in drain-
age waters collected from both north and south series of sludge-
treated Plainfield sand in 1973-74 as compared 1972-73 waters
(Pr-=0.05). Thus, these data indicate that Cu levels in drainage
waters tend to increase with increased years of annual sludge
applications.
Mean Cd concentrations in drainage waters from Blount silt
loam control plots (Table 15) located in the south and north ser-
ies ranged from 0.003 to 0.009 ppm and 0.003 to 0.011 ppm,
respectively. Drainage waters from sludge-treated Blount silt
loam plots located in the south series had mean Cd contents
ranging from 0.006 to 0.013 ppm and a similar range of 0.004 to
0.010 ppm was observed in the north series. Ranges in mean Cd
concentrations in drainage waters from control and sludge-
treated Elliott plots were not appreciably different from those
observed in waters from Blount silt loam except for samples col-
lected from the sludge-treated plot located in the south series.
Drainage waters from the sludge-treated Elliott silt loam plot in
the south series had mean Cd concentrations that decreased
from 0.043 ppm during 1971-72 to 0.017 ppm during 1973-74.
Thus, changes with time in Cd concentrations in drainage waters
from this particular plot were in agreement with changes in Zn
and Cu concentrations and all these data confirm direct contam-
ination of drains with sludge during early years of operation.
Without regard to location, drainage waters from Plainfield
loamy sand control plots had mean Cd concentrations ranging
from 0.001 to 0.003 ppm as compared to mean concentrations
ranging from 0.003 to 0.009 ppm in drainage waters from
sludge-treated plots of the same soil type. Sludge applications
-------
42
Lysimeter Plots
significantly increased Cd concentrations (Pr« 0.01) in Blount
silt loam drainage waters during 1971-72 and 1973-74 from
plots located in the south series. Cadmium concentrations in
drainage waters were also increased (Pr« 0.01) during 1971-72
and 1973-74 by sludge applications on the Elliott plot located in
the north series. Although drainage waters from the sludge-
treated Elliott silt loam plot in the south series always contained
significantly higher concentrations of Cd than waters from con-
trol plots, it was caused by direct contamination and thus is of
little interest with regard to assessing metal transport in soils.
Mean Cd concentrations in drainage waters were increased
(Pr-eO.Ol) by sludge applications on Plainfield loamy sand plots
during each of the sampling years for the plot in the north series
and during 1972-73 from the plot in the south series.
Heavy metal concentrations in drainage waters from field
plots or watersheds have seldom been reported. Where metals
have been applied to soil surfaces as soluble salts and their con-
centrations measured at various soil profile depths after leach-
ing with water, it has generally been conculded that they were
rapidly fixed in the soil in forms not easily transported by perco-
lating water.
Table XV: Arithmetic mean, X, standard deviation, s, median M, geometric mean, £ , and geometric standard deviation, d , for Cd
concentrations as ppm in runoff and drainage water samples collected from Blount silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from three replications in each of two series of plots.
Location
Year and
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4 •
0.014
0.007
0.021
0.029
0.036
0.023
0.040
0.049
0.016
0.011
0.015
0.028
Runoff Water
s
0.031
0.013
0.059
0.039
0.101
0.049
0.082
0.091
0.064
0.012
0.019
0.050
M
0.002
0.002
0.006
0.010
0.006
0.006
0.013
0.015
0.004
0.006
0.009
0.009
V
0.004
0.004
0.007
0.014
0.008
0.007
0.015
0.018
0.004
0.006
0.008
0.011
o
4.25
3.00
3.29
3.43
5.00
4.86
3.80
4.17
4.75
4.17
3.75
3.55
X
0.004
0.003
0.007
0.010
0.035
0.042
0.031
0.038
0.013
0.010
0.012
0.011
Drainage Water
s
0.002
0.003
0.005
0.025
0.081
0.077
0.061
0.112
0.051
0.023
0.022
0.042
M
0.003
0.002
0.006
0.005
0.009
0.008
0.012
0.011
0.003
0.003
0.088
0.004
y
0.003
0.003
0.006
0.004
0.009
0.011
0.013
0.010
0.004
0.004
0.008
0.005
5
2.00
2.00
1.83
3.00
5.89
5.64
3.92
5.50
3.50
3.25
2.25
2.80
Table XVI: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, /t , and geometric standard deviation, $ , for Cd
concentrations as ppm in runoff and drainage water samples collected from Elliott silt loam lysimeter plots during the period May 1 to
April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized and irri-
gated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots adjacent to
an instrument house. Summary measurements include data from two control plots and one maximum sludge-treated plot in each of
two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.005
0.014
0.011
0.067
0.023
0.037
0.038
0.011
0.036
0.022
s
0.007
0.023
0.012
0.156
0.048
0.075
0.036
0.019
0.097
0.022
M
0.003
0.004
0.005
0.017
0.004
0.007
0.014
0.027
0.004
0.005
0.007
0.013
0
0.003
0.005
0.007
0.021
0.006
0.007
0.016
0.024
0.004
0.004
0.008
0.013
c
2.67
4.00
2.43
3.48
6.33
4.57
3.37
2.83
9.00
5.25
5.50
3.31
X
0.003
0.002
0.116
0.006
0.036
0.031
0.046
0.029
0.017
0.008
0.026
0.009
Drainage Water
s
0.002
0.001
0.172
0.004
0.052
0..078
0.050
0.037
0.043
0.033
0.050
0.007
M
0.002
0.002
0.028
0.007
0.009
0.006
0.032
0.014
0.003
0.003
0.014
0.006
u
0.002
0.002
0.043
0.005
0.011
0.006
0.031
0.012
0.005
0.003
0.017
0.007
d
2.00
1.50
4.33
2.20
6.45
6.83
2.77
5.00
3.60
3.33
2.06
1.86
-------
Lysimeter Plots
43
Table XVII: Arithmetic mean, X, standard deviation, s, median, M, geometric mean, ft , and geometric standard deviation, O , for Cd
concentrations as ppm in runoff and drainage water samples collected from Plainfield loamy sand lysimeter plots during the period
May 1 to April 30 during each of the data years 1971-1972,1972-1973, and 1973-1974. Water samples were collected from fertilized
and irrigated control or check plots, 1, and maximum, 4, sludge-treated plots located in either a north, N, or south, S, series of plots
adjacent to an instrument house. Summary measurements include data from two control plots and one maximum sludge-treated plot in
each of two series of plots.
Location
Year and
Runoff Water
Treatment X
71-72
72-73
73-74
SI
Nl
S4
N4
SI
Nl
S4
N4
SI
Nl
S4
N4
0.003
0.016
0.079
0.017
0.014
0.012
0.016
0.041
0.013
0.011
0.007
0.035
s
0.002
0.033
0.136
0.011
0.023
0.021
0.020
0.052
0.020
0.012
0.006
0.035
M
0.003
0.004
0.011
0.017
0.007
0.005
0.010
0.016
0.004
0.008
0.005
0.025
y
0.003
0.003
0.027
0.013
0.007
0.005
0.010
0.020
0.005
0.007
0.003
0.019
0
2.00
6.33
4.52
2.31
3.29
4.20
2.60
3.60
4.60
3.43
5.33
1.68
X
0.003
0.002
0.003
0.004
0.020
0.019
0.036
0.033
0.011
0.013
0.005
0.017
Drainage Water
s
0,003
0.001
0.001
0.001
0.099
0.051
0.056
0.092
0.033
0.051
0.004
0.059
M
0.002
0.001
0.004
0.004
0.002
0.003
0.010
0.006
0.003
0.003
0.004
0.004
y
0.002
0.001
0.003
0.004
0.003
0.003
0.009
0.008
0.003
0.003
0.004
0.005
a
2.50
2.00
1.33
1.50
5.67
7.67
9.00
5.00
4.33
4.67
2.00
3.20
As a method of predicting limits for sludge applications on
land that would assure the protection of groundwaters used for
public water supplies against contamination by heavy metals,
Jorgensen7 determined concentration distributions of several
metals between liquid and solid phases ot sludge-soll-water mix-
tures. Using several soil types he mixed lOOg of soil (wet weight)
with 20g of sludge (wet weight) and an unspecified amount of
rainwater which were shaken for 24 hours before the samples
were filtered. After filtering, contents of Zn, Cu, Cd, Cr, Ni, and
Pb in solution were determined and used to calculate the ratio
between metal concentrations (ppm) in solution to total metal
concentrations in sludge-soil mixtures (ppm) for the several
measured metal species. This calculated ratio was called the dis-
tribution coefficient. With the use of several soils having clay
contents that varied from 2.4 to 34.4 percent and humus con-
tents that varied from 1.7 to 14.3 percent, distribution coeffi-
cients as a function of soil pH were shown as graphs for the sev-
eral elements. Based on the distribution coefficient, water
quality standards and amounts in sludge, Jorgensen7 concluded
that Pb or Cr limits the application of sludge. He further con-
cluded that the distribution coefficients as influenced by soil pH
was independent of the volume of water and sludge/soil ratio.
On this basis he considered adsorption and precipitation as
major processes responsible for binding metals in soils. It was
estimated that if soils having high contents of humus and clay
were maintained at pH 7 or greater, 20 tons per hectare per year
of the sludge used in the study could be applied for 10 years with-
out endangering groundwater supplies. However, it was not
clear how the loading rate limit was derived from distribution
coefficients.
In view of findings reported by others, especially Jorgensen, it
is unfortunate that Pb was not one of the elements measured in
drainage waters during this study. Considering amounts added
to soils as constituents of sludge and the relatively low concen-
trations recommended as maximum limits in public water
sources, Pb, Cr, and Ni should be given priority in further moni-
toring operations. At this time the findings reported here indi-
cate that while mean Zn contents may sometimes be increased in
drainage water where relatively high sludge loading rates are ap-
plied, the levels are well within recommended limits for public
water sources and Zn levels in drainage waters do not appear to
increase with increased years of sludge applications. With
regard to mean Cu contents in drainage waters, all were within
recommended limits for public water sources, but the trend
toward higher levels with increased years of maximum sludge
applications cannot be ignored. Like Zn, mean Cd concentra-
tions in drainage waters do not appear to increase with increased
years of sludge applications. However, because recommended
limits for soluble Cd levels in public water sources are very near
background levels to be expected in drainage waters, the limit in
terms of total Cd contents was exceeded in waters from both
control and sludge-treated fine textured soils in the 1972-73
year. During this particular year mean total concentrations of
Cd exceeded recommended soluble Cd limits in drainage water
from replicated Blount silt loam control plots located in the
north series and sludge-treated plots located in the south series.
Also, during this particular year mean Cd levels in drainage
waters from Elliott silt loam control plots located in the south
series and from the sludge-treated plot in the north series
exceeded recommended limits. Thus, ignoring data collected
from the sludge-treated Elliot silt loam plot where drainage
water was known to have been directly contaminated with
sludge, mean Cd levels in drainage waters from four plots
exceeded recommended limits during one year. Limits were
exceeded by total Cd concentrations in waters from control
plots with the same frequency of occurrences as oberved for
waters from sludge-treated fine textured soils. Contrary to
expectations Cd levels were lower in waters from sludge-treated
Plainfield loamy sand plots than in correspondingly treated
plots of finer textured soils. Mean Cd concentrations in drain-
age waters from all Plainfield sandy loam plots were always well
within recommended limits for Cd levels in public water sources.
However, because of relatively higher amounts of water
percolated through the sandy textured soils concentrations were
diluted and total amounts of the metal removed from the soil
profile were not considerably different.
Summary
In view of the relatively large estimated geometric standard
deviations associated with measurements of heavy metals in
-------
44 Lysimeter Plots
runoff waters, attempts to assess water pollution hazards from
the results of a relatively few grab samples collected over a rela-
tively short time span presents a considerable risk in underesti-
mating or overestimating the problem. Probably the metal con-
centrations in runoff and drainage waters found in this study
were higher than would be found in streams receiving waters
from field-sized areas treated with equivalent sludge and loading
rates, assuming that such areas were protected against excessive
erosion by water. Lysimeter water samples were never subjected
to conditions favoring metal losses by precipitation followed by
sedimentation and biological scavenging of heavy metals from
waters flowing over soil surfaces protected by close-growing
vegetation. Nevertheless, the results from this study will be of
value to those concenred with assessing the environmental
impacts of utilizing municipal sludges on well drained soils in
humid regions. For the metals measured during the study, only
Cu appeared to increase in drainage waters with increased years
of annual sludge applications, but the highest mean concentra-
tion during the last year was relatively low as compared to
recommended limits for public water sources. Because Cd
contents in both runoff and drainage waters from control plots
is so near recommended limits in public water sources, it is likely
to be the metal of major concern in the disposal of municipal
sludges on land.
REFERENCES
I. Aitchison, J. and J.A.C. Brown, 1957. The Log-normal
Distribution. Cambridge University Press, Cambridge, Eng-
land.
2. Committee on Water Quality Criteria 1972. "Water Qual-
ity Criteria 1972", Environ. Studies Board, Nat. Acad. Sci.,
Nat., Acad. Eng., Wash. D.C. 594 pp.
3. Environment Canada. 1975. "Land Disposal of Sewage
Sludge, Vol. II". Research Report No. 24, Ontario Ministry of
the Environment Pollution Control Branch, Toronto, Ontario,
276 pp.
4. Hem, J.D., 1972. "Chemical Factors that Influence the
Availability of Iron and Manganese in Aqueous Systems''. Geo-
logical Society of Amer. Bulletin, Vol. 83. pp. 443^50.
5. Hinesly, T.D., O.C. Braids, R.I. Dick, R.L. Jones, and
J.A.E. Molina, 1974. "Agricultural Benefits and Environmental
Changes Resulting from the Use of Digested Sludge on Field
Crops" Research progress report prepared for the Metropoli-
tan Sanitary District of Greater Chicago and the U.S. Environ-
mental Protection Agency (Grant No. DOI-U1-00080), 375 pp.
6. Jenne, E.A., 1968. "Controls on Mn, Fe, Co, Ni, Cu, and
Zn Concentrations in Soils and Water: the Significant Role of
Hydrous Mn andFe Oxides", pp. 337-387, In Trace Inorganics
in Water, R.A. Baker, Symposium Chairman, Advances in
Chemistry Series 73, Amer. Chem. Soc., Washington, D.C.
7. Jorgensen, S.E. 1975. "Do Heavy Metals Prevent the Agri-
cultural Use of Municipal Sludge?" Water Research, Vol. 9, pp.
163-170.
8. Klein, L.A., M. Lang, N. Nash, and S.L. Kirschner, 1974.
"Sources of Metals in New York City Wastewater". Journ.
Water Pollution Control Federation, Vol. 46, No. 12, pp.
2653-2662.
9. Kopp, J.F. and R.C. Kroner. 1970. "Trace Metals in
Waters of the United States". U.S. Dept. of Interior, Fed. Water
Pollution Control Admin., Div. of Pollution Surveillance, Cin-
cinnati, Ohio, 28 pp.
10. Miller, D.W., F.A. DeLuca, and T.L. Tessier, 1974.
"Ground Water Contamination in the Northeast States." Envi-
ronmental Protection Technology Series, EPA-660/2-74-056,
U.S. EPA, Wash., D.C. 325 pp.
11. Snedecor, G.W. and W.G. Cochran. 1967. Statistical
Methods. 6th Edition, The Iowa State University Press, Ames,
Iowa.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge financial support pro-
vided in part by the Metropolitan Sanitary District of Greater
Chicago and the U.S. Environmental Protection Agency. The
authors are also thankful for the assistance rendered by G. Bar-
rett, J. J. Tyler, E. L. Ziegler and laboratory and field techni-
cians during the collection of data discussed here.
-------
Transport Model to Predict the Movement of
Pb, Cd, Zn, Cu, and S Through A
Forested Watershed*
J. K. Munro, Jr., R. J. Luxmoore, C. L. Begovich, K. R. Dixon,
A. P. Watson, M. R. Patterson, and D. R. Jackson
Oak Ridge National Laboratory
Union Carbide Corporation, Nuclear Division
Oak Ridge, Tennessee
INTRODUCTION
The original motivation for the work to be described in this
paper was to determine what effects Pb might be having on the
forest ecosystem of the New Lead Belt area of southeast
Missouri. As the study got underway, its scope was broadened
to include effects due to Cd, Zn, Cu, and SO2. This study had
two aspects. The first involved obtaining measurements of
quantities for use as input data to a Unified Transport Model
(UTM) and measurements of stream flow, heavy metal concen-
trations and biomasses of various forest components for use in
calibrating and validating the UTM. The second aspect was to
develop a UTM consisting of modules which modeled the
behavior of various subsystems of a forested watershed with
emphasis on the transport and effects of heavy metals. This pa-
per presents a summary of results from the second aspect of the
Crooked Creek study, the development and application of the
UTM.
Some History of the Development of the UTM
Development of the UTM began in late summer of 1972 under
the sponsorship of the National Science Foundation—RANN
Environmental Aspects of Trace Contaminants Program. The
earliest version of the UTM consisted of a merger of the Oak
Ridge National Laboratory (ORNL) version of the Wisconsin
Hydrologic Transport Model (WHTM) as described by Patter-
son et a/.14 and the Atmospheric Transport Model (ATM) of
Mills and Reeves." The WHTM is based on the Stanford
Watershed Model - IV of Crawford and Linsley2 as modified by
Huff.9 The ATM, refined and improved [Culkowski and Patter-
son3] during the course of this study, is built around a Gaussian
plume model which was adapted to treat point, line, and area
sources.
More physically and chemically based mechanistic submodels
were developed to treat atmosphere-soil-plant water flow [Gold-
stein et alJ\ on a daily basis (PROSPR), plant growth (CERES)
which stimulates plant water effects on photosynthesis and
growth [Dixon et a/.4], soil chemical exchange of heavy metals
(SCEHM) [Begovich and Jackson1], and solute uptake by vege-
tation (DIFM AS-DRY ADS) [Luxmoore et a/.10].
All these models mentioned above together form a large
subset of a suite of models* which have been developed to allow
•Work supported by the National Science Foundation—RANN Envi-
ronmental Aspects of Trace Contaminants Program
"These submodels and several different versions of the UTM are avail-
able at cost in FORTRAN IV source form. Documentation for these
programs also is available. For ore information, write J. K. Munro, Jr.,
Computer Sciences Division, Oak Ridge National Laboratory, P.O.
Box X, Oak Ridge, TN 37830.
for flexibility and to provide options for various types of appli-
cations. Two versions of the UTM were used in the Crooked
Creek study.- One, based primarily on the WHTM, was used to
simulate heavy metal transport through the entire watershed
and to generate the outflow hydrograph and mean daily heavy
metal stream concentrations at the watershed outfall. A second
version, based on the Terrestrial Ecology and Hydrology Model
(TEHM) [Huff el a/.9] which used PROSPR, CERES, etc., was
used to simulate heavy metal distribution in the soil profile and
plants; to study heavy metal accumulation in the soil profile and
plants; and to study the effects of heavy metals and SO2 on plant
growth and litter decomposition.
Some Historical Background for the Crooked Creek
Watershed (CCW) Study
Measurements of the distribution and concentrations of the
heavy metals in the soils of the New Lead Belt area of southeast
Missouri were begun in 1971 under the direction of B. G. Wix-
son by an Interdisciplinary team from the University of
Missouri-Rolla (UMR) [Wixson et a/.18]. Most of the area they
studied lies in the Clark National Forest, shown in Figure 1. At
the suggestion of the NSF-RANN managers in the Trace Con-
taminants Program, a group from the Ecology and Analysis of
Trace Contaminants (EATC) program at ORNL began a col-
laboration in 1973 with the UMR team for the purpose of apply-
ing the UTM to a watershed in the New Lead Belt area. A
watershed in the headwaters drainage area of Crooked Creek
was chosen because it had heavy metal sources which could be
characterized and which were located along one boundary of the
watershed and it was easily accessible for making measurements
and for supplying electrical power for various measuring instru-
ments.
Forest ranger stations are located in the towns shown (Figure
1) around the periphery of Clark National Forest. These ranger
stations have recording gauges for measuring hourly precipita-
tion. They also collect several different kinds of climatological
data. Some of the data from these stations were used in the ini-
tial simulations and later as backup data to fill gaps in data
obtained from gauges closer to the Crooked Creek Watershed
(CCW).
The original goals for application of the UTM to CCW were
calibration and validation of the models developed for predict-
ing transport of heavy metals through a forested watershed fol-
lowed by some long term simulations and studies of various pos-
sible scenarios, such as the accidental release of large amounts of
SO2 from the smelter acid plant. The long term simulations
would be used to predict accumulation, distribution, and toxic
effects of heavy metals in the watershed. The UMR team agreed
to supply data needed for calibration and validation of the
45
-------
46 Transport Model
ST. LOUIS
CLARK
NATIONAL FOREST
STEELVILLE..
• ••
—. .BELLEVIEW
CROOKED CREEK
WATERSHED
TREND,
BELT
KANSAS CITY
VIBURNUM
NEW LEAD
SPRINGFIELD
Figure I: Location of Crooked Creek Watershed (inverted block triangle) with Respect to Nearby Cities.
hydrologic and heavy metal transport components of the UTM.
The ORNL group concentrated on development and applica-
tion of the UTM.
Description of Crooked Creek Watershed
As Figure 2 shows, Crooked Creek Watershed (CCW) is a
triangularly-shaped, 466 ha watershed, bounded on two sides by
roads maintained by the state of Missouri or the AMAX Lead
Company of Missouri (in the vicinity of the smelter) and
bounded on the west for the most part by a jeep trail. Roads in
this area of Missouri generally run along the ridgelines, so can be
used to define surface divides of watersheds. The CCW lies on
the Springfield-Salem Plateau, an area characterized by hills al-
ternating with steep valleys [Schwarz and Schwarz16]. Second
growth canopy species found on the watershed are largely black
oak (Quercus velutina Lam.), white oak (Q. alba L.), northern
red oak (Q borealis Michx.), and shortleaf pine (Pinus echinata
Mill.). Seventy-five percent of the total foliage biomass is pro-
vided by oak species; twenty-five percent provided by shortleaf
pine [Watson17].
Soils of the region are thin, stony, limestone-residual, cherty
silt loams. Ridgetops on CCW have a Lebanon soil type with 5-8
cm fragipan; slope soils are dominated by cherty silt loams of the
Wilderness (40%) and Clarksville (25%) series. The valley floors
are occupied by the alluvial Razort and Ashton soils [G. Gott, C.
L. Scrivner, personal communication].
A primary lead smelter located at the apex of the watershed
has been in operation since 1968. Most of the watershed is a
woodland ecosystem and remains largely intact. Current heavy
metal loads on CCW are the result of both stack emissions and
fugitive sources. The location and orientation of the major
sources with respect to the watershed are shown in Figure 3.
Fugitive sources include orehandling processes, yard dusts, and
exposed concentrate piles. These sources contribute mostly
finely divided sulfide dusts. Stack emissions contain metal sul-
fates and oxide particulates less than 2 microns in diameter.
These soluble oxides and sulfates represent a much greater
biological hazard to the watershed than the extremely insoluble
sulfides. For purposes of modeling the atmospheric transport of
the particulates and gases, three sources were considered: a
point source (the smelter stack) and two types of area sources
originating from the smelter yard (vehicular suspension and
wind resuspension of yard dusts).
Figure 3: Heavy Metals Source Locations in Relation to
Crooked Creek Watershed. °
-------
Transport Model
47
Figure 2: Aerial View of Crooked Creek Watershed. Dashed Line Shows Western Watershed Boundary.
Brief Review of the UTM
Overview of General Features
The UTM was developed for the purpose of simulating the
transport of trace toxic substances through an ecosystem con-
sisting of the atmosphere, land surface, vegetation, soil layers,
ground water, and streams. It operates with a time resolution of
one hour when rain is absent and 15 minutes when it rains. Dif-
ferent versions of the UTM have been applied to watersheds
ranging in size from a couple hundred hectares to the order of
ten thousand square kilometers. The ATM imposes an upper
limit in size of an area having about a 70-80 km radius when it is
coupled to the submodels in the UTM.
Input data common to the various versions of the UTM
include the historical hourly precipitation, daily values of aver-
age wind speed, maximum and minimum temperature, mean
dew point, integrated total radiation, values of various
quantities characterizing the watershed topography, soil charac-
teristics, and extend of vegetative cover. Hourly wetfall and
monthly dryfall contaminant values may be required jf the ATM
is not used. If the ATM is used, data on contaminant source
characteristics, physical character of the contaminant, and
hourly wind speed and direction are required.
The current complete set of submodels which can be coupled
together within the UTM framework are shown in Figure 4.
Application of the UTM to a particular watershed or to a partic-
ular problem in a watershed does not usually require consid-
eration of physical and chemical models for all the possible
components which can be treated by the suite of models which
-------
48 Transport Model
are available. So some appropriate subset of models is usually
coupled together for a given simulation. Collecting and assem-
bling input data is always a major project. Some versions and
some options within versions allow for running with less data or
different data. For example, the WHTM version can be run
using daily values of evaportation instead of computing evapo
transpiration from daily values of average wind speed, mean
dew point temperature, integrated incoming solar radiation, etc.
Two versions of the UTM were used in the Crooked Creek
study. They are shown in Figure 5 and differ only for the land
simulation. The WHTM version is presented on the left; the
TEHM version on the right.
ATM
ATMOSPHERIC
TRANSPORT MODEL
SULFUR CONCENTRATION! AM
DEPOSITIONS FROM STACK
OENCRD: GRID OF DEPOSITION POINTS
FOR DETAINING AREA-WEIGHTED
DEPOSITION RATES
KHAR:
(UNIT AREA TERRESTRIAL RESPONSE)
SNOMLT: SNOW PACK ENERGY BALANCE AND MELT
HTM: PARAMETRIC RUNOFF MODEL, ION EXCHANGE OF CONTAMINANTS.
SOIL AND CONTAMINANT EROSION
SLOPE AND ASPECT ADJUSTMENTS TO RADIATION
SOIL-PLANT-WATER-ATMOSPHERE DYNAMICS, EVAPORATION
SOIL CHEMICAL EXCHANGE OF HEAVY METALS
PLANT GROWTH, TRANSLOCATION
DRYADS: LEAF AND ROOT UPTAKE
MFMAS: MASS FLOW AND DIFFUSION TO ROOTS
ONE DIMENSIONAL ANALYTIC MODEL FOR SOIL WATER POTENTIAL
TRACE CONTAMINANT CONCENTRATION
HYDROLOGIC SOURCE AREAS. SATURATED AND UNSATURATED
DRAINAGE AND GROUNDWATER FLOW
SCEHM:
CERES:
CHAW.
STREAMFLOW HYDRAULICS AND TRANSPORT
CHNSED: SEDIMENT AND CONTAMINANT EXCHANGE AND TRANSPORT
SEDTRN: SUSPENDED- AND SED-LOAD TRANSPORT AND COMPOSITION
PNTSRC: POINT SOURCE DISCHARGE INPUTS
WHTM OPTIMAL PARAMETER
SET DETERMINATION
UNIFIED TRANSPORT MODCL OUTPUT
Figure 4: The Submodels that May be Linked to Form a Uni-
fied Transport Model.
Description of the WHTM Version
The WHTM version contains only limited treatment of
vegetation effects. It allows for interception storage, percent
ground cover, and solubility limited leaching of chemicals from
a litter layer. It also treats evaporation and transpiration effects
together as a single process. The soil profile is divided into sur-
face, upper zone, lower zone, and ground water as indicated in
the lower part of Figure 6. The ground water storage volume has
active and inactive components. The inactive volume acts like an
infinite sink or source and can be used to allow for water leaving
or entering a watershed through unobserved aquifers.
Material transport is treated in tandem with the hydrologic
flow as indicated in Figure 7. The transition from surface to sub-
surface soil is represented by a stack of four theoretical ion-
exchange plates. In addition, the top inch of the upper zone is
also singled out to aid in the transition from soil surface effects
to subsurface processes. For each storage volume, contaminant
mass is assumed to be mixed uniformly throughout the volume.
For a particular contaminant the same ion-exchange distribu-
tion coefficient value is used for each storage volume where
mixing of soil and contaminant occurs.
ORIGINAL WHTM PROGRAM
Figure 5: Modular Structure of Submodels Used for the
WHTM and TEHM Versions of the Unified Transport Model
(UTM).
Figure 6: Hydrologic Flow in a Land Area Segment as Repre-
sented in WHTM Version of the UTM.
-------
Transport Model
49
(1-A1IA3P+- BARE « A3!
Figure 7: Material Flow in a Land Area Segment as Repre-
sented in WHTM Version of the UTM.
Surface erosion and transport are treated in terms of thresh-
olds for resuspension of loose soil particles and for eroding new
material. Erosion is driven by the flow power of water moving
over the ground surface. Exchange of contaminants can occur
between the surface water and sediments so the discharge to the
streams includes water, eroded sediments, amounts of contami-
nants sorbed on these sediments, and contaminants dissolved in
the water.
Once the materials discharged from the land surface enter the
stream they are carried through the stream network, with the
proper delays, to the watershed outfall. The version of WHTM
used for the CCW study considered no mixing or chemical
exchange of contaminant with the stream bed. So whatever
enters the stream network rather quickly passes through the
watershed outfall. The total amount of contaminant entering
the streams is used for the channel routing calculations, so
stream concentrations at the outfall include contaminants both
dissolved in the water and sorbed on sediments. Submodels
[Fields5] were available to treat contaminant-sediment interac-
tions in the streams, but measurements necessary to assemble
the input data for these models were not made. So no attempt
was made to use these models.
Description of the TEHM Version
The TEHM version was specifically developed to emphasize
the relationships between plants and their environment. Mecha-
nistic models for the soil-plant-water-solute relationships are
used as much as possible within the basic UTM framework. The
TEHM version is more modular in its treatment of the soil-
water processes and allows for a variable number of soil layers of
varying characteristics. The ion-exchange model (SCEHM)
which couples to this version also works within the modular
framework of the soil layers and is more general than its coun-
terpart in the WHTM version since it allows for a different value
of the ion-exchange distribution coefficient for each soil layer.
Contaminant concentration transition from the soil surface to
subsoil, modeled by means of a stack of theoretical ion-
exchange plates, is not considered in this version. Neither are
surface erosion effects.
Figure 8 shows how the various modules in the TEHM ver-
sion are coupled. Atmosphere-plant-soil moisture relations are
simulated by PROSPR [Goldstein, Mankin, and Luxmoore7].
Plant water potentials generated by this module are used by
CERES [Dixon et a/.4], the submodel which simulates plant
growth and respiration, standing crop biomass of leaves, stems,
fruits, and roots, and storage pools of sugars, live tissue, and
dead or woody tissue. CERES also computes the translocation
of carriers (transpired water and phloem sugar movement) of
chemical substances. Mass flow of solutes from the soil into the
roots is simulated by the submodel DIFMAS and the transport
and accumulation of solutes in vegetation and litter is simulated
by the submodel DRYADS. These last two submodels can sim-
ulate effects of toxic gases and vapors, such as SO2, in addition
to heavy metals.
Since the TEHM version focuses on effects of contaminants
on plant growth and litter decomposition, its results were not
used with the channel routing simulation. The TEHM simula-
tions were carried out for a typical hillside site about 0.4 km
from the smelter stack.
DRYADS
« AM
PLANT CAPACITY FOR CONTAMINANTS -
HJNT OF PLANT UPTAKE OF CONTAMINANTS -
Figure 8: Coupling of the Modules in the TEHM Version of the
UTM: PROSPER, CERES, SCEHM, DIFMAS and
DRYADS.
-------
50 Transport Model
Results of Model Applications
Application of the WHTM Version
The WHTM version of the UTM, more than any other, has
been applied to many different watersheds, so a large amount of
experience has been accumulated about its use. As a result, it
was possible to begin application of this version to CCW long
before all the input data for CCW had been obtained. Calibra-
tion of the model continued as more data became available.
Early simulation results showed [Munro and Wixson12] that
large heavy metal concentrations occurred in runoff from the
first storms following an extended dry period. Concentrations in
runoff from more storms following soon afterward were con-
siderably reduced. These results suggested that special atten-
tion be given to sampling runoff from storms immediately
following long dry periods.
Input data for the Atmospheric Transport Model (ATM)
were available first so most of the early application efforts cen-
tered on calibrating and validating it. The results of this work
are described by Patterson et al. ' 3, so won't be reviewed again. In
summary, the ATM simulated wetfall and dryfall values for Pb
which were factors of 2-5 higher than those measured, depend-
ing on the sampling site location relative to the heavy metal
sources. Dusts suspended by the smelter yard vehicular traffic
constituted the largest single source of heavy metals. Total
monthly deposition rates of heavy metals on the ridge tops are
shown in Table I as an example of values typical for the
watershed. The Pb values shown were simulated for seasonally
averaged wind data; values for the other heavy metals are aver-
ages of measured data.
Table I: Monthly Heavy Metal Deposition Rates (mg/ft2/Mo.)
for Segment One on Crooked Creek Watershed. Lead Values
were Simulated Using the ATM. Values for Other Heavy Metals
were Measured by K. Purushothaman (UMR).
Year
73
74
Month
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Lead
187.
179.
141.
151.
143.
154.
150.
155.
174.
176.
175.
183.
Cadmium
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
Zinc
25.
25.
25.
25.
25.
25.
25.
25.
25.
25.
25.
25.
Copper
10.
10.
10.
10.
10.
10.
10.
10.
10.
10.
10.
10.
Input data for simulating the unit land area responses were
obtained from the UMR team [Wixson and Jennett, eds.18], C.
L. Scrivner and G. Gott [personal communication] of the Uni-
versity of Missouri - Columbia (UMC), and the AMAX Lead
Company of Missouri environmental officer George Carr [per-
sonal communication]. The watershed area was divided as
shown in Figure 9 into three segments consisting of ridgetops,
hillsides and meadow. This division was made on the basis of
land use patterns, variation in soil type and pH, and surface
topography. The channel system was divided according to the
natural nodes of the stream network.
Stream flow data were available for only two storm events,
[Foil6] both of which occurred in the same season of the year.
These two storm events did not provide enough data for a good
calibration and validating. Data for a couple of storm events in
each season of the year would have been much more useful. Pre-
cipitation data for both storms were measured at the watershed
outfall. The remainder of the precipitation data were measured
at Viburnum, 10 km due north of CCW. Gaps in the Viburnum
data were filled using data from one or a combination of ranger
stations surrounding the Clark National Forest (Figure 1).
Hourly precipitation and measured stream flow for the first
storm event are shown in Figure 10. Notice that this event spans
parts of two days and contains several gaps in the flow data.
Stream flows were measured with a nitrogen gas bubbler
placed about five meters upstream from a culvert which defined
the channel cross section geometry. The first set of stream flow
data shows a dip at 5:00 a.m. on 28 August 1974 when a rise due
to the incoming precipitation would be expected. This suggests
that the bubbler arrangement may not have been working prop-
erly at this time.
1.0
o
0.5
10
S 6
O
\
8P I2M
4A
8A 12 N 4P 8P I2M 4A
28 AUGUST 1974
TIME (hr)
Figure 10: Hourly Precipitation and Stream Flow for Storm
No. 1, 28-29 August 1974, Measured by Foil [1975].
The succession of small-to-moderate intensity storms shown
here followed a long dry period. Subsurface water storages are
being replenished by this event. Data from this series of storms
give enough information to be able to adjust parameters govern-
ing watershed response for characteristic times less than two
weeks.
Data for the second storm event (II September 1974) are
shown in Figure 11. This storm was well isolated in time from
storms before and after it. The storm was also intense, so pro-
vided a lot of useful information for calibration and validation
purposes, especially for parameter adjustments affecting short
time response behavior.
These stream flow data limited model calibration basically to
those parameters governing overall infiltration rate of water
into the soil (CB), intermediate time response of flow recession
(IRC, CC), and base flow conditions (K24L). All these parame-
ters were varied except CC. Calibration of the remaining hydro-
logic parameters which were still free to be adjusted would have
been possible if stream flow data for storms in other seasons of
-------
Transport Model
51
the year had been available. Two parameters in this category,
known to govern longer time hydrologic responses, are the nom-
inal storages for the upper (UZSN) and lower (LZSN) zones.
UZSN affects hydrologic response on a time scale of several
weeks; LZSN affects response on a time scale of several months.
These storages had been adjusted in preliminary simulations to
approximately correct values by requiring the values of upper
(UZS) and lower (LZS) zone storages at the end of a simulated
year to be approximately equal to their values at the beginning
of the simulated year. Most of the parameters affecting hydro-
logic response such as fraction of bare soil, overland flow length,
average segment slope, etc. are pretty well determined from
measurements.
Results of varying the parameters CB, IRC, and K24L in the
final stages of hydrologic parameter calibration will be used to
give some idea of the success of the calibration efforts. Figure 12
shows what happens when different values of CB are used. There
seems to be no way to account for the dip in the first segment of
measured flows other than problems with the data themselves.
Agreement looks better for the case CB=4.5 at the higher flows
than at low flows. With the infiltration rate essentially deter-
mined, it was possible to begin adjusting parameters governing
hydrograph shape. The parameter IRC seemed to be the best
candidate to look at first. Effects of different values for IRC are
shown in Figure 13. Analysis of these results suggests a value for
IRC determined by the half-life of the intermediate response
component of the stream flow recession curve. An appropriate
value for IRC lies between 0.05 and 0.1. The value IRC=0.05 was
used for subsequent simulations. Parameter CC was not
adjusted, but probably could have been used to make some
further small adjustment in the time distribution of flow for the
intermediate time response component. With a reasonable value
for the interflow recession constant having been chosen, atten-
tion was finally directed to simulating base flows better. Stream
flow data for the region around the CCW indicated consid-
erable drainage of water occurred through unobservable
aquifers, perhaps as much as 2/3 of the water entering the
watershed. So a value of 0.7 was used for the parameter K24L to
jjjfr'cl^a^jl;
f^. \ , -. NI\ v.
XC.M'..
nXw.Y ^'^-;l5fe^;^^
b>)^ ^b-7 ^'-^-^^v^^-^j^^
&{^^%£^- . ibx \"v.----Cvj\',r^ r.^<:;^
).- "V'\ iiivV"
'•• •- «in •• «•—^ -.. -
TOPS
•j I I RIDGE SIDES
>~ ===
w: tss
Figure 9: Map of Crooked Creek Watershed Showing Segment Boundaries for Segment Areas Considered in the Simulation Calcula-
tions and Showing the Stream Channel Network Used in the Flow Rounting and Timing Calculations.
-------
52
Transport Model
see what would happen. The result is shown in Figure 14, again
using data for the second storm event. Low flows are simulated
much better.
1.5
c
- (.0
z
o
5
Q.
O «
U'
0
(20
(00
80
O
60
UJ
o:
40
20
4A 8A 12N 4P 8P (2M
(1 SEPTEMBER (974
TIME (hr)
Figure 11: Hourly Precipitation and Stream Flow for Storm
No. 2, 11 September 1974, Measured by Foil [1975].
O
IUUU
500
200
too
50
20
(0
5
2
1
0.05
0.02
n m
E i i 1 1 1 r
~ * MEASURED FLOWS, STORM NO. I
SIMULATED FLOWS, CB = 0 8
SIMULATED FLOWS, CB = 2'&
_ SIMULATED FLOWS, CB = 4.5
-
/""\
\
^
Ei >
~(
-1
j
1
j
7
4
a/— X
;/
/'/"
/ I
—f
-V
b""*^
_
\
\
\
\
\
i
1
\ \
Yv
\ *^_
\\
vs=
L
i
A
I •
j
j
i
i
i
i
; /^
\
' ' \
^-''\
/
I
\
, :
\ -'
\ 1
u
s"
. /
v^x x
. ,'
f
I
\
"\ i
\ i
\ i
\j
i
i
i
i
^-\>
'*^\-
I
"\
•\
\
V..
f\
\
\
\
\
\
&-&^. V
^TCM
/ v\
\
; v
/
f
1
---
/^'•v
f N.
i
/
_j
j
' /^^\
f N
i
/
/
/ A
' n
! /
/
^/
^tfi
*•»./
_
-
_
-
~
-
~
-
,\ —
-
-
__
-
-
_
-
8
20
24
12 (6
TIME (hr)
Figure 12: Comparison of Simulated and Measured Stream
Flows for Storm No. 1 forthe Following Values of CB:a)0.8,b)
2.5, c) 4.5.
1000
500 \—
0.5
0.2
O.I
-w>-- MEASURED FLOWS, STORM NO. 2
SIMULATED FLOW, IRC = 0.734
SIMULATED FLOW, IRC= 0.20
- - SIMULATED FLOW, IRC= 0.05
SIMULATED FLOW, IRC= 0.01
e
20
24
12 16
TIME(hr)
Figure 13: Comparison of Simulated and Measured Stream
Flows for Storm No. 2 for the Following Values of IRC- al
0.734, b) 0.20, c) 0.05, and d) 0.01. '
-------
Transport Model
53
o
00
00
50
20
to
5
2
1
0.5
0.2
n (
-
-
—_____-.
=
-
•*— MEA
SIML
SIMl
™— __— *'
1
SURED Fl
LATED F
JLATED F
"^
/ 1
.OWS, ST
_OW, K2<
LOW, K2'
1
f>
ORM NO.
tL = 0.70
»L=0
v
\
2
—
—
—
^
—
=
—
12 16
TIME(hr)
20
24
Figure 14: Comparison of Simulated and Measured Stream
Flows for Storm No. 2 for K.24L Values 0.0 and 0.7.
Once the hydrologic response was calibrated reasonably well,
parameters governing the rate of sediment transport were
adjusted to give the correct annual erosion rate. The annual ero-
sion rate for a forested watershed is about 0.85 g/ft2. Though a
certain degree of latitude in the parameter values affecting mate-
rial transport was possible, every attempt was made to keep
these parameter values within the ranges suggested by Huff8 and
consistent with whatever was known about the watershed. There
was very little room within this framework to play around with
the contaminant and erosion parameter values,
A comparison of simulated and measured stream flows and
heavy metal concentrations for the two storm events described
above is shown in Table II. Overall best agreement is obtained
for the second storm event. Results for the two days of the first
storm event do not look as good, but this should not be surpris-
ing since the measured data available for this comparison is
incomplete. It is perhaps encouraging to observe the consistency
between measured and simulated heavy metal concentrations
except for Pb on 28 August 1974.
Application of the TEHM Version
The TEHM version used considerably more input data than
the WHTM version beyond the large subset of input data com-
mon to all versions of the UTM. The more mechanistic submod-
els of the TEHM version require a larger number of parameters
to characterize the soil-plant subsystems and their behavior.
These parameters also require more effort to obtain because
more detailed physical and chemical measurements must be
made. For the CCW application many of the parameters relat-
ing to the physiological responses of the plants were taken from
existing literature since, for different natural systems, the ranges
of values were fairly consistent. Required biomass values were
taken from measurements made at CCW or the control site 21
km from the smelter stack.
Table II: Comparison of Simulated and Measured Stream Flows and Heavy Metal Concentrations for Several Storms at Crooked
Creek Watershed, New Lead Belt, Missouri.
Heavy
Metal
Pb
Cd
Zn
Cu
Storm
Dates
8-28-74
8-29-74
9-11-74
8-28-74
8-29-74
9-11-74
8-28-74
8-29-74
9-11-74
8-28-74
8-29-74
9-11-74
Daily Flows [cfs]
Measured
1.69
6.75
12.7
1.69
6.75
12.7
1.69
6.75
12.7
1.69
6.75
12.7
Simulated
3.15
10.7
11.8
3.15
10.7
11.8
3.15
10.7
11.8
3.15
10.7
11.8
Simulated
Total Metal
Transported
Per Day
[kg]
58.3
22.4
110.0
2.01
3.01
6.72
14.1
10.4
62.2
0.47
1.08
6.64
Daily Average Metal
Concentration (Unfiltered
Samples) [mg/£]
Measured
3.7
9.0
9.9
1.8
2.8
0.57
4.57
10.01
5.43
0.18
0.50
0.55
Simulated
7.56
0.86
3.81
0.26
0.11
0.23
1.83
0.40
2.15
0.06
0.04
0.23
-------
54
Transport Model
Before models of plant growth, litter decomposition, and
effects of heavy metals were fully coupled together in the TEHM
version, Begovich and Jackson1 [summary by Patterson et al.13]
used a version with only the PROSPR and SCEHM submodels
coupled to simulate heavy metal build-up in the soil. They found
that the simulated amounts of Cd and Zn in the top soil layers
reached equilibrium after several years while Pb and Cu con-
tinued to accumulate. All heavy metals continued to accumulate
in the lower soil layers. Agreement between simulated and mea-
sured (1974) concentrations of heavy metals in the top two soil
layers at the end of the six-year simulation was quite good. The
simulation was started with heavy metals absent from the soil
for 1968 and used measured heavy metal deposition rates [Wix-
son and Jennett, eds.18].
The plant stand growth submodel CERES was calibrated to
give a mass balance and to simulate a steady state for the plants
on an annual basis. The solute uptake, litter decomposition, and
toxic effects submodels DIFMAS and DRY ADS were also first
run with the absence of contaminant input and were calibrated
to simulate litter biomass values for the control site. Calibration
with contaminant input absent was relatively easy because a lot
of data are available for natural plant systems free of toxic
chemicals.
Calibration of the TEHM version to simulate heavy metal
uptake correctly was difficult because reliable data are scarce,
particularly for litter fall rates, litter decomposition rates, and
root uptake rates for heavy metals. Diffusion coefficients for
heavy metal transport across root membranes and various
thresholds for toxic effects on plant growth and litter decompo-
sition had to be adjusted to give reasonable simulated values of
biomass and heavy metal uptake rates and concentrations in
plant parts.
Heavy metal effects on the CCW forest were simulated with
the fully coupled TEHM version for a six year period corre-
sponding to the period of time from the beginning of smelting
operations in 1968 to the time when measurements were made
on CCW (the "present"). The simulated effect of the atmo-
spheric deposition of Pb on the dynamics of an oak forest litter
system is shown in Table III. Year zero values are those simu-
lated for the steady state before heavy metals were introduced
into the system. Plant mortality increased only 3% during the six
years with almost all loss from roots. The steady annual input of
plant litter coupled with a slowing of the litter decomposition
rate resulted in a 44% increase in the litter mass. The litter
decomposition rate appears to increase from year to year
because of the increase in litter mass. On an annual incremental
Table III: Simulated Dynamics of an Oak Forest Liner System
Subjected to Successive Years of Atmospheric Deposition of
Lead
Year
0
1
2
3
4
5
6
Plant
Mortality
(g/m2/yr)
720.0
720.0
741.0
742.0
743.0
744.0
744.0
Total Litter
Mass
(g/m2)
2150.0
2451.0
2670.0
2829.0
2947.0
3034.0
3099.0
Decomposition
Rate
(g/m2/yr)
438.0
438.0
522.0
581.0
624.0
655.0
677.0
basis, however, the decomposition rate is leveling off, suggesting
that the litter mass may be approaching an equilibrium level.
Table IV shows the simulated September plant tissue concen-
trations of heavy metals for each year of the six year simulation.
Heavy metal concentrations at the beginning of the simulation
were assumed to be zero. The results show a rapid rise in concen-
tration during the first year for the leaves followed by a constant
amount for the succeeding years. Concentrations of Pb in stem
and root sapwood increase substantially each year, whereas
concentrations of Cd, Zn, and Cu for these plant parts
increase rapidly the first year and very slowly or reach the maxi-
mum level for succeeding years. Heavy metal concentrations for
heartwood components show a gradual increase in all cases due
to the large amount of biomass present in the components. Con-
centrations of Cd, Zn, and Cu in fruit appear to approach a
maximum level during the simulation. Although annual fruit
production is similar to leaves, transport of solutes to fruits is
slower than in other plant parts. Therefore fruit heavy metal
concentrations level off only after the maximum concentrations
in other plant parts is reached.
Annual uptake of heavy metals by roots and leaves, plant loss
of heavy metals from mortality, and litter content of heavy
metals are shown in Table V. Annual root uptake of lead
increased with the accompanying build-up of Pb in the soil sys-
tem [Begovich and Jackson1]. Root uptake of other metals
drops off after the first year because the maximum concentra-
tion allowed in the model was reached. Both the reduced rate of
contaminant build-up in the plant tissues and the reduced rate of
plant uptake with time is due to the decrease in plant demand as
the concentration of heavy metals in the various plant parts
approached its maximum. Uptake of heavy metals by leaves was
constant through the six-year simulation period since atmo-
spheric deposition was assumed constant from year to year.
Recycling of heavy metals back to litter from mortality of plant
parts increased with time and can be mainly attributed to the fall
of leaves and fruits during autumn.
There was build-up of litter Pb of about 10 g/ m2/ yr as well as
an increase in average litter concentration. The rate of increase
in concentration varies from year to year as a result of toxic
effects slowing the rate of litter decomposition and causing a
build-up of litter mass (Table III). Litter content and minerali-
zation of Pb and Cu increased while that of Cd and Zn decreased
slightly (Table V). Although mineralization results from decom-
position, it is assumed that the metals that are released are
retained by the litter through absorption. There is a decrease in
mineralization rate of Cd even though there is an increase in
plant mortality input into litter. The decline in mineralization is
due to the increased simulated toxic effect of Pb on the rate of
litter decomposition.
Results of the simulation of litter and heavy metal accumula-
tion after six years of smelter operation are given in Table VI.
Simulated values generally are lower than those experimental
values measured on CCW by Watson17. The predicted litter
mass (not including root litter) is about 47% less than the experi-
mental value. This could result from the low levels of Pb pre-
dicted by the model which determines the toxic effect on decom-
position or from the estimate of toxic threshold level being too
low. The predicted low Pb levels in litter are evident from the
comparison of litter concentrations. Since the predicted litter
mass is low, the Pb content also must be low. The predicted low
heavy metal concentrations in litter apparently are not a result
of leaching to the soil and uptake by plants since the plant levels
are a small fraction of the litter levels. A reasonable explanation
of the predicted low levels is that the estimated rates of heavy
metal deposition may be too low and/or that the rate of deposi-
tion may have been higher in previous years.
-------
Transport Model
55
Table IV: Simulated September Tissue Concentration of Heavy Metals ( Mg/g) in Oak Vegetation Following Successive Years of
Heavy Metal Atmospheric Deposition
Years
Lead
1
0
3
4
5
6
Cadmium
1
2
3
4
5
6
Zinc
1
2
3
4
5
6
Copper
1
2
3
4
5
6
Leaf
176.0
172.0
172.0
172.0
172.0
173.0
9.84
10.3
10.4
10.5
10.6
10.5
49.6
49.9
49.9
49.9
49.9
49.9
9.82
9.91
9.97
10.0
10.0
10.0
Stem
Sapwood
1.64
3.81
5.73
9.1
14.3
21.7
7.96
7.98
7.98
7.98
7.99
7.98
29.9
30.0
30.0
30.0
30.0
30.0
14.8
14.9
14.9
14.9
14.9
14.9
Stem
Heartwood
0.010
0.048
0.123
0.231
0.403
0.667
0.032
0.163
0.288
0.408
0.527
0.646
0.142
0.358
0.838
1.30
1.76
2.22
0.067
0.312
0.554
0.705
1.02
1.25
Root
Sapwood
183.0
450.0
524.0
695.0
888.0
1080.0
44.3
47.2
47.5
47.7
47.8
47.7
49.0
49.3
49.3
49.3
49.4
49 A
9.75
9.85
9.87
9.88
9.88
9.88
Root
Heartwood
1.27
9.68
25.6
44.2
68.3
98.1
0.476
1.98
3.51
4.99
6.12
6.09
0.577
2.14
3.64
5.1
6.51
7.87
0.110
0.423
0.727
1.02
1.30
1.58
Fruit
0.180
0.260
0.262
0.459
0.724
1.06
9.49
15.8
17.7
18.9
19.5
18.4
48.6
51.1
53.1
54.8
54.9
55.1
8.76
9.90
10.5
11.0
11.2
11.3
-------
56
Transport Model
Table V: Simulated Heavy Metal Uptake, Content, and Losses for Successive Years of Heavy Metal Atmospheric Deposition
Vegetation (mg/m2/yr)
Years
Lead
1
2
3
4
5
6
Cadmium
1
2
3
4
5
6
Zinc
1
2
3
4
5
6
Copper
1
T
3
4
5
6
Root
Uptake
98.4
171.2
107.7
175.5
216.4
245.6
36.3
12.5
11.8
11.8
11.8
11.6
85.2
16.4
16.7
16.8
16.9
21.4
28.2
4.4
4.5
4.5
4.5
4.5
Leaf
Uptake
85.7
82.8
82.8
82.8
82.8
82.8
1.7
2.0
2.0
2.0
2.0
2.0
15.7
21.6
21.3
21.2
21.2
21.2
5.3
4.6
4.5
4.4
4.4
4.4
Mortality
Loss
84.6
85.2
88.0
91.0
95.2
99.9
5.2
5.8
6.2
6.4
12.6
12.6
26.6
27.1
27.4
27.8
28.0
28.2
5.2
5.4
5.4
5.5
5.6
5.7
End
Content
(g/m2)
18.9
32.8
43.4
51.5
57.7
62.4
0.0038
0.0038
0.0038
0.0038
0.0041
0.0041
0.057
0.057
0.057
0.057
0.057
0.057
0.17
0.26
0.31
0.35
0.37
0.40
Litter
of Year
Concentration
(Mg/g)
16385.0
21003.0
29051.0
38430.0
44362.0
51900.0
2.88
3.69
3.77
4.20
4.94
5.35
53.3
76.0
81.4
94.4
107.6
120.9
117.9
207.5
238.3
267.6
334.4
386.8
Mineralization
Loss
(g/m2/yr)
2.98
7.26
10.51
12.61
14.92
16.37
0.0015
0.0015
0.0015
0.0014
0.0018
0.0013
0.041
0.042
0.039
0.039
0.039
0.039
0.047
0.075
0.092
0.101
0.114
0.118
-------
Transport Model
57
Table VI: A Comparison of Simulated and Experimental
Results for Crooked Creek Watershed
Total Litter mass (g/m2)
Litter Cone. (PPM)
Lead
Cadmium
Zinc
Copper
Leaf Tissue Cone. (PPM)
Lead
Cadmium
Zinc
Copper
Root Tissue Cone. (PPM)
Lead
Cadmium
Zinc
Copper
Stem Tissue Cone. (PPM)
Lead
Cadmium
Zinc
Copper
Soil Cone. Al Layer (PPM)
Lead
Cadmium
Zinc
Copper
Simulated
2448.0
51000.0
5.3
121.0
375.0
170.0
10.0
50.0
10.0
1150.0
52.0
52.0
10.0
27.0
9.0
35.0
18.0
271.0
4.4
52.0
46.0
Experimental
3600.0
72000.0
150.0
2350.0
1400.0
467.0
3.9
47.0
10.0
2400.0
46.0
48.0
6.0
680.0
3.9
30.0
13.0
334.0
3.55
25.0
18.0
CONCLUSIONS/RECOMMENDATIONS
In conclusion the simulation studies indicate or predict the
following:
• Fugitive sources of heavy metals from the mining and
smelting operations are a significant problem. Roadways
around the smelter yard have been paved and are now washed
regularly to suppress suspension of dusts.
• Seventy percent of the water entering the watershed leaves
it through unobserved aquifers. What is the fate of this water
and the amount of heavy metals it may be carrying away?
• Amounts of Cd and Zn in the top soil layer have reached
equilibrium; Pb and Cu continue to accumulate.
• Solubility limited leaching of heavy metals from forest
floor litter is a significant process in the movement of these
metals through a forested watershed.
• Simulated levels of Pb in litter and predicted litter masses
are 50% lower than measured values, suggesting that measured
Pb deposition rates were too low and/or were higher in years
prior to the time monitoring of deposition rates was begun.
Based on experience gained with the application of the
WHTM and TEHM versions of the UTM to heavy metal trans-
port on CCW, the following recommendations are made:
• The WHTM version is more appropriate for studying water
management, water quality, and land use pattern effects. It
should be well suited for assessing strategies to reduce move-
ment of toxic substances through a watershed and for assessing
effects associated with the siting of a facility that would be a
source of toxic substances.
• The TEHM version is more appropriate for studying effects
of water management and toxic substances on vegetation. It can
be used to assess effects of SO2 damage to plants, CO2 balance in
the atmosphere and effects of fossil fuel burning, and effects of
toxic substances from waste disposal operations. This version
should be well suited for studying the toxic effects of heavy
metals on plants and heavy metal uptake and distribution in
plants resulting from application of sludges to agricultural land.
The WHTM version is generally easier to apply to a problem
than the TEHM and runs faster on a computer. The WHTM is a
shorter program and runs in less time, so is to be preferred when
plant effects are not an important consideration.
REFERENCES
1. Begovich, C. L. and D. R. Jackson, Documentation and
Application of SCEHM: A Model for Soil Chemical Exchange
of Heavy Metals, ORNL-NSF-EATC-16 (1975).
2. Crawford, Norman H. and Ray K. Linsley, Digital Simu-
lation in Hydrology: Stanford Watershed Model IV, Depart-
ment of Civil Engineering, Stanford University. Technical
Report No. 39 (July, 1966).
3. Culkowski, W. M. and M. R. Patterson,y4 Comprehensive
Atmospheric Transport and Diffusion Model, ORNL-NSF-
EATC-17 (April, 1976).
4. Dixon, K. R., R. J. Luxmoore and C. L. Begovich,
CERES - A Model of Forest Stand Biomass Dynamics for Pre-
dicting Trace Contaminant, Nutrient, and Water Effects,
ORNL-NSF-EATC-25 (1976).
5. Fields, D. E., CHNSED: Simulation of Sediment and
Trace Contaminant Transport with Sediment I Contaminant
Interaction, ORNL-NSF-EATC-19 (1976).
6. Foil, James Lee, Aquatic Transport of Lead and Other
Heavy Metals from a Lead-Zinc Smelting Area, Thesis, Univer-
sity of Missouri - Rolla (1975) unpublished.
7. Goldstein, R. A., J. B. Mankin and R. J. Luxrnoore,
PROSPER, A Model of Atmosphere-Soil-Plant Water Flow,
EDFB-IBP-73-9(1974).
8. Huff, D. D., Simulation of the Hydrologic Transport of
Radioactive Aerosols, Thesis, Stanford University, (1968).
9. Huff, D. D., R. J. Luxmoore, J. B. Mankin and C. L.
Begovich, TEHM: A Terrestrial Ecosystem Hydrology Model,
ORNL-NSF-EATC-27 (1976).
10. Luxmoore, R. J., C. L. Begovich and K. R. Dixon,
DRYADS and DIFMAS: FORTRAN Models for Investigat-
ing Solute Uptake and Incorporation into Vegetation and Lit-
ter, ORNL-NSF-EATC-26(1976).
11. Mills, M. T. and M. Reeves, A Multi-Source Atmo-
spheric Transport Model for Deposition of Trace Contami-
nants, ORNL-NSF-EATC-2(1973).
12. Munro, John K., Jr., and Bobby G. Wixson, "First-order
Simulation Run for Lead Transport Through Crooked Creek
Watershed - New Lead Belt, Missouri," in Trace Contaminants
in the Environment, Proceedings of the Second Annual NSF-
RANN Trace Contaminants Conference, Asilomar, Pacific
Grove, California, 29-31 August 1974. LBL-3217 (1974).
13. Patterson, M. R., C. L. Begovich, and D. R. Jackson,
"Environmental Transport Modeling of Pollutants in Water
and Soil," in Proceedings of the Symposium on Nonbiological
Transport and Transformation of Pollutants on Land and
Water, sponsored by the National Bureau of Standards, Gai-
thersburg, Maryland (May 1976).
14. Patterson, M. R., J. K. Munro, D. E. Fields, R. D. Elli-
son, A. A. Brooks and D. D. Huff, A User's Manual for the
FORTRAN IV Version of the Wisconsin Hydrologic Transport
Model, ORNL-NSF-EATC-7, EDFB-IBP-74-9 October
(1974).
-------
58
Transport Model
15. Patterson, M. R., J. K. Munro and R. J. Luxmoore,
"Simulation of Lead Transport on the Crooked Creek
Watershed," Presented at the 9th Annual Conference on Trace
Substances in Environmental Health, University of Missouri,
Columbia, June 9-12, 1975.
16. Schwarz, C. W. and E. R. Schwarz, The Wild Mammals
oj Missouri, University of Missouri Press and Mo. Cons.
Comm. p. 341 (1959).
17. Watson, A. P., Impact of a Mining-Smelting Complex on
the Forest - Floor Litter Arthropod Fauna in the New Lead Belt
Region of Southeast Missouri, Dissertation, University of Ken-
tucky, (1976).
18. Wixson, B. G. and J. C. Jennett (eds.), An Interdiscipli-
nary Investigation of Environmental Pollution by Lead and
Other Heavy Metals from Industrial Development in the New
Lead Belt of Southeastern Missouri, Interim Progress Report
May 1972 to June 1974, University of Missouri-Rolla, Report to
National Science Foundation, (June, 1974).
-------
Leachate From Applications of Fertilizers, Manures
and Sewage Sludges to Land
P. F. Pratt and A. L. Page
Department of Soil Science and Agricultural Engineering
University of California
Riverside, California
INTRODUCTION
The application of fertilizers, manures, and sewage sludges to
lands may affect the concentrations of various elements in leach-
ates from soils in a variety of ways. The leachate composition
will depend upon the solubility of the element in the water drain-
ing through the soil, the volume of water that moves through the
soil, the composition and amount of material applied, the
amount of each element volatilized or removed in harvested
crops, and the amounts adsorbed on soil colloids or precipitated
within the soil. Also, the chemical, physical, and biological
properties of soil have a profound influence on leachate compo-
sition. In the sections which follow we will analyze the properties
of fertilizers, manures, and sewage sludges and evaluate how
applications of these materials will affect leachate composition
under a variety of conditions. Although there are a number of
similarities between the reactions of elements in the three mate-
rials in soil, for the sake of clarity of presentation, they are each
discussed separately.
Fertilizers
Composition
The materials known as fertilizers include organic as well as
organic materials and mixtures of each type. Because manures
and sludges are treated separately in this paper, this section will
deal with inorganic salts as fertilizers in which organic materials
are added only as fillers or conditioners and do not add signifi-
cantly to their supplies of nutrient elements.
Fertilizers are added to soils to supply nutrient elements that
are present in soil in insufficient amounts to meet the needs of
growing crops. Mostly, fertilizers are added to supply nitrogen,
phosphorus and potassium, but in some cases they are added to
supply other essential elements. The usual fertilizers consist of
single salts or various combinations of some of the following:
urea, ammonium sulfate, ammonium phosphates, ammonium
nitrate, calcium nitrate, calcium phosphates, potassium chlo-
ride, potassium sulfate and other soluble salts. Some materials
contain calcium sulfate and others contain small amounts of
borates and the oxides, chlorides or sulfates of trace metals.
When fertilizers are added to soils, the elements supplied are
either 1) removed in harvested crops, 2) precipitated in or sorbed
by the soil and become part of the solid material, 3) removed as
soluble salts in the leachate that leaves the soil-root zone, or 4)
accumulate in the root zone as soluble salts. In addition to these
alternative sinks, nitrogen can leave the soil by volatilization of
ammonia from the surface or by denitrification and loss of
nitrous oxide and dinitrogen gases. The accumulation of soluble
salts in soils usually does not take place in humid regions
because of their removal by leaching waters. In arid regions
under irrigation, deliberate leaching is frequently necessary to
remove soluble materials to prevent depressing effects of soluble
salts on crop yields, but mostly these salts come from irrigation
water and not from fertilizers.
Impact on Leachate Quality
Trace Elements. The reactions of most of these elements with
soils insure that they will not be a problem for groundwater
water quality with usual agricultural practices. Some of these
elements are essential for crop production and others are not
essential but are toxic if present in sufficient concentrations in
the soil. Those elements that are essential can be deficient in the
soil, in which case they are added at low rates to eliminate the
deficiency. However, at the rates used to correct deficiencies the
adsorption by the soil is sufficiently effective that movement to
groundwaters is prevented. Those elements not essential but
toxic to plants are added to the soil only as impurities in other
chemicals; for example, a number of phosphate fertilizers have
some heavy metal impurities, but these impurities are not known
to cause problems for groundwaters.
Phosphorus and Potassium. The reactions of P in most soils is
such that application rates used in crop production will not
cause groundwater problems. The added P is so effectively
adsorbed and/or precipitated that movement beyond 15-20 cm
below the depth of incorporation over a period of years is highly
unlikely. The only exception is on extremely sandy soils in which
added P has been shown to move slowly with drainage.
Sjpencer,30 for example, working with a Lakeland fine sand (99%
silica sand pfus silt and 1% organic matter) found that surface
applied P had moved to the 210-cm depth in a period of 10 years.
Obviously, if this is the order of magnitude of the maximum rate
of movement, the problem of P moving to groundwaters does
not seem serious for most soils which will have rates of move-
ment orders of magnitude slower.
The downward movement of K in most soils, when added at
the rates used in crop production, is usually so slow that
groundwater quality is not a factor. Potassium reacts into the
cation-exchange complex of all soils and in many it is fixed into
slowly leachable forms by micaceous clays. The rate of move-
ment is variable depending on the cation-exchange capacity and
fixation processes, but no problems are to be expected except in
very sandy soils.
Nitrate. A great deal has been written about soil N and its
pollution potential. Until recently very little research by agrono-
mists and soil scientists had dealt with tracing the nitrate (NOj)
form of N through the unsaturated zone beneath the soil-root
system into the groundwater. But recent research has concen-
trated on this important question.
For purposes of discussing groundwater pollution the N cycle
59
-------
60
Leachate
can be described as presented in Figure 1. The organic N consists
of soil humus, residues of plants and the N added in organic
wastes. In aerated soils the organic N that is mineralized goes
rather quickly through the NH+and NOj forms to NO^". Com-
mercial fertilizers usually add N in the NH| or NOjform or as
urea which quickly hydrolyzes to NH^. Nitrite oxidizes rapidly
to NO^in most aerated soils and does not accumulate. Thus, the
N added to aerated soils is either in the NO^form or is converted
to this form. In anaerobic soils (water saturated soils with low
oxygen content), the NH+does not oxidize to NOjand the NO,
in the soil is reduced to N2Oand/or N2 by microbial denitrifica-
tion processes in which organisms use the O from the NOjand
release N2O and N2 as end products. The N2O and N2 thus pro-
duced escape to the atmosphere.
Plant Uptake
Plant Uptake
Organic N - NH-N - NON - NON
y 4 j 3
Volatilization
NH,
Figure 1: Nitrogen Cycle
Volatilization
N20,N2
Leachote
Crop plants use either NH^or NO^and convert the N into
proteins. The N thus used goes to market as feed, food and fibers
or is returned to the soil in plant residues (roots, stems and
leaves). Thus, a simple N cycle can be visualized as a N input that
is balanced by outgoes in I) harvested crops, 2) leaching of the
NOf in drainage water, and 3) volatilization from the soil as NH3
or as N2O and N2.
The removal of N in harvested crop plants is usually about
50% of the total inputs, but this removal varies from about 25 to
75% depending on the crop and the level of N inputs relative to
crop needs. This low efficiency of use leaves a fairly large portion
of the total input for leaching as NOj or loss of NH3, N2O or N2
to the atmosphere. Thus, there is little doubt that NO3 escapes
from the soil-root system and contributes to contamination of
groundwaters. The only question is how much is contributed
under various situations of soil, climate, cropping system and
management.
The direct measurement of losses of NOjfrom the soil-root
system to the ground water requires data for the volume of water
leaving the root zone and the average NO^concentration of this
water for the period of time being studied. The concentration of
NO," in the water moving to the groundwater is not sufficient,
since without an estimate of the volume of drainage, the transit
time for the NO^to move through the unsaturated zone and the
dilution factor when the drainage water reaches the ground-
water the concentration of the groundwater cannot be
estimated.
NOi
N =
ION
IT
:n
in which NOjfN is the concentration of NOj-N in mg per liter in
the drainage water, N is the leached N in kg per hectare per year,
D is the drainage volume in surface cm per hectare per year and
10 is a proportionality constant to convert from kg N per hectare
in a given volume of water, expressed in surface cm, to mg N per
liter, has been found to be useful in calculations. This relation-
ship is expressed graphically in Figure 2.
CONCENTRATION OF
VS LEACHING VOLUME
N03-N
120
100
E
CL
<=>- 80
LU
h-
60
40
20
1 1 r
LEACHING VOLUME IN
SURFACE cm
_ (SURFACE inches)
A
B
C
D
15 (6)
30 (12)
45 (18)
60 (24)
50
100
150 200 250 300
EXCESS N, kg per ha
Figure 2: Relationship Between NOj-N Concentration and
Leached N for Four Drainage Volumes, Calculated Using
Equation [1].
The relationships among yield of harvestable crops, total
available N input, N use efficiency, denitrification or total vola-
tilization loss and the concentration of NO^N in the drainage
water are presented in Figure 3. This is a generalized relation-
ship for a given drainage volume (15 surface inches) and a crop
with a high N requirement. Also, the assumption is made that
denitrification does not reduce fertilizer use efficiency.
N USE EFFICIENCY
1 T
2"0 «
CJi
6
i rO
o
TOTAL N AVAILABLE, kg per ha per year
Figure 3: General Relationships of Yield of Harvested Crops, N
Use Efficiency, and Concentration of NOjN in Drainage Waters
to Total Available Soil N fora Drainage Volume of 15 Surface
Inches. Surves 1,2,3 and 4 represent concentrations of NON in
drainage water for 1) no crop-no denitrification, 2) crop-no den-
itrification, 3) crop-50% denitrification and 4) crop-75% denitri-
fication, respectively. Denitrification is expressed as a percen-
tage of the available N not removed by crops. The efficiency of
removal of N by crops is assumed to vary from 90% at 112 ke
available N to 4 at 672 kg of available N.
I? 224 336 448 560 67^
-------
Leachate
61
Considering the large numbers of combinations of N inputs,
crop removals, leaching volumes and denitrification losses,
NOJ-N concentrations from a few to several hundred mg per liter
appear to be possible and have been found in the field. And con-
sidering all the variables discussed, the accurate prediction of
the NOj-N concentration of drainage water is difficult. Howev-
er, useful estimates can be predicted from knowledge of fertilizer
inputs, water intake and evapotranspiration (ET), removal in
harvested crops and soil properties that influence volatilization
losses.
The main concept in managing nitrogen fertilizer applications
to croplands without increasing unnecessarily the leaching of
nitrate is illustrated in Figure 3. In these relationships, the leach-
ing of nitrate increases slowly with increase in available nitrogen
until the point of maximum yield is reached. In some cases,
where increased available nitrogen and the resultant yield
increase cause greatly reduced volumes of leachate, the
relationship between nitrate leached and available nitrogen
inputs can be negative. In any event, as the point of maximum
yield is exceeded the ratio of leached nitrate to nitrogen input
increases with further increases in available nitrogen. Thus, only
sufficient nitrogen required to reach the maximum yield should
be applied.
The leachable nitrate is not a static value but is a highly varia-
ble and transient value. The actual amount of leached nitrate
depends on how much water leaches, the timing of the leaching
relative to the growth and nitrogen use cycle of the plants or the
efficiency of the root system in absorbing nitrate during leaching
and the amount of denitrification going on during the growth of
the plant and during leaching. Thus, crop use of available nitro-
gen, nitrate leaching and denitrification are mutually competing
processes, the final outcome of which is not easily predicted.
Both leaching and denitrification or a combination of the two
can cause inefficiencies that create the need for higher available
nitrogen inputs to produce maximum yields. Increasing the
yield of crops by better management of factors other than avail-
able nitrogen inputs will also increase the amounts of nitrogen
used by the crop and reduce the potential leachable nitrate.
Thus, the total yield per unit of available nitrogen needs to be
kept at a high level.
The contribution of fertilizer nitrogen to nitrate in the
groundwater depends not only on how much nitrogen is used
per unit of cropland but also on the yield and removal of nitro-
gen in harvested crops, losses to the atmosphere and the volume
of leachate in which the leached nitrate is dissolved (Figure 2).
Thus, the correlations between concentration or amounts of
nitrate leached and amount of fertilizer nitrogen are not high.
Factors that control leaching, such as the amount and distribu-
tion of rainfall or water management in irrigated lands, and soil
properties that influence both denitrification and leaching are
likely to exert more control over the amount of nitrate leached
to groundwater than the amount of fertilizer nitrogen added.
Two possible criteria for the contribution of fertilizers on
croplands to the nitrate in groundwaters are concentrations of
nitrate in leaching waters or mass emissions of nitrate nitrogen
per unit area of land. The more rational of the two is the mass
emission criterion. Methods of control should naturally lead to
greater efficiency of use of nitrogen by crops. Perhaps the most
effective method of increasing this efficiency is to reduce leach-
ing to a minimum by control of the factors that influence water
intake into the soil and evapotranspiration, i.e., control of irri-
gation and control of vegetative cover. But reduced leaching vol-
ume can create leachates with higher nitrate concentrations. In
irrigated agriculture a quality standard based on nitrate
concentration might encourage increased water use to keep the
leached nitrate below some designated concentration. Increased
water use would promote greater leaching losses and greater
inefficiencies of applied nitrogen. A mass emission criterion
would promote lower leaching volumes and greater efficiencies
of use of both nitrogen and water.
Soluble Salts. The inorganic salts that are the main
components of commercial fertilizers become part of the com-
plex of soluble salts in the soil and thus contribute to the salinity
of groundwaters. The contribution depends on what fraction of
these salts are removed from the land in harvested crops and the
fraction that is adsorbed and/ or precipitated or in the specific
case of nitrogen how much is lost to the atmosphere as ammo-
nia, nitrous oxide and nitrogen. In humid regions, the salt from
fertilizers that leaches to the groundwater is probably unimpor-
tant as compared to weathering processes that contribute salin-
ity to groundwater. In arid irrigated regions, the salts from com-
mercial fertilizers contribute to the salinity in the soil, however,
the main source of soluble salts is the irrigation water itself. And,
of course, in newly irrigated land residual salts in the soil and in
underlying sediments can contribute substantial amounts of sol-
uble salts to the groundwater.
Manures
Composition
The safest statements that might be made about the
composition of manures is that they have low concentrations of
the fertilizer constituents nitrogen, phosphorus and potassium
as compared to commercial fertilizers, they have traces to small
amounts of most elements found in soils and plants and the con-
centrations of many elements vary over considerable ranges.
The composition of manure depends on the source, the ration
and the management of manures following their production.
Powers et al.,21 after reviewing available data that could be
expressed on a dry weight basis, found large variations in com-
position of manures but could not relate these variations to cli-
matic or geographical regions in the U.S. A. Table I, taken from
Powers et al.,21 presents data for the ranges in concentrations of
some constituents in manures. Of the manures from beef and
dairy cattle, swine and poultry, which represent the bulk of
animal manures in the U.S.A., poultry manure usually has the
highest and beef and dairy cattle manure the lowest nitrogen
contents. But, fresh manure from cattle can have much higher
nitrogen contents than aged poultry manures.
Table II presents data for a representative chemical
composition of beef cattle feedlot manure from the arid South-
west (Meek et al.18). Many other trace elements could have been
detected but were not included as part of the analyses. Furr et
al.6 found small concentrations of many trace elements and
heavy metals in cattle manure. Most cattle manures that accu-
mulate on corral floors lose 50% or more of their nitrogen con-
tents by volatilization of ammonia before they are collected and
spread on cropland.
Table III, which presents data for the concentration of sev-
eral elements in a number of batches of both dairy corral manure
and beef feedlot slurry manure, illustrates the problem of
dealing with manures. Even though these batches of manures
came from the same areas and were collected at the same times
(fall for odd-numbered samples and spring for even-numbered
samples), the concentrations were not consistent. The last batch
collected was quite different from all other batches.
Impact on Leachate Quality
The effect of any soil amendment fertilizer or waste product
added to the soil on the quality of the leachate that moves from
the soil-root zone is naturally dependent on (1) the amount
added, (2) the constituents contained in the material and their
concentrations, (3) amounts of these that go into harvested
crops and retained in the soil as insolubles as part of the soil sol-
ids. The elements of concern from the application of manures at
recommended fertilizer rates for optimum crop production are
-------
62
Leachate
nitrogen in the nitrate form and the cations of calcium, magne-
sium and sodium and the anions chloride, sulfate and bicarbo-
nate that constitute the usual soluble salts.
Trace tlements. 1 he trace elements and heavy metals added
in manures are of no concern except in unusual cases where such
elements are added as inorganic salts to the rations such as feed-
ing copper salts to swine. The concentrations of these elements
are usually of the same order of magnitude as those in plants,
residues and foods and feeds.
Phosphorus and Potassium. These elements are held in soils
by adsorption and/ or precipitation reactions so that they do not
leach. Exception can be found in very sandy soils or in peat soils
having only small capacities to retain or sorb these elements.
However, when rates of application of manure far exceed the
recommended rates needed for nutrient supplies, the leaching of
phosphorus and potassium can take place depending on the
ratio of inputs of these elements to the capacity of the soil to
adsorb or precipitate them.
Nitrate. Perhaps the most universal leachate quality factor
that results from the application of manures to croplands or to
the spreading or incorporation of manures for disposal is the
leaching of nitrate. Soluble salts are not necessarily a problem
(when manure is used at recommended fertilizer rates) in humid
regions where dilution with low salinity water occurs. But, in
general, both the nitrate and soluble salt problems become more
serious in arid regions. Thus, aside from certain local situations,
the problems of nitrates and salts in leachates are inversely
related to rainfall.
The problem in the use of manures to supply available
nitrogen to meet crop needs is to determine proper application
rates. Rates can be adjusted by conducting field experiments in a
given local area. A number of such experiments can provide an-
swers for the various combinations of manures, crops and soils.
However, a degree of uncertainty always exists because of
variations in composition of manures that appear to be the same
product that has come from the same source.
Table II: Representative Chemical Composition of Beef Cattle
Feedlot Manure from the Arid Southwest (Meek et al., 1974 and
1975).
Constituent
Organic Matter
Soluble ions
Total N
Nitrate-N
Ammonium-N
Ca 1 c i urn
Potassium
Magnesium
Sodium
Phosphorus
Aluminum
Iron
Copper
Lithium
Manganese
Zinc
Boron
Composition, oven dry
basis
°L _mq/kq
64
10
2.0
3.5
0.19
2.8
1.1
1.5
2.8
0.27
0.52
0.48
30
9
153
99
137
Table I: Ranges in Concentrations of Four Elements, Expressed in Percent of the Dry Weight, for Manures from Beef, Dairy, Swine
and Poultry
Animal
Beef
Da i ry
Swine
Poultry
*No reliable
Element
N
0.6 -
1.5 -
2.0 -
1.1 -
data
4.9
3.9
7.5
7.8
availabl
P
0.11 -
0.41 -
0.56 -
0.38 -
e on a
1.6
1.6
2.5
6.3
dry wei
K
0.053 -
1.4 -
1.5 -
0.73 -
ght basis.
3.8
3.3
4.9
4.8
Na
0.05 - 2.8
0.35 - 0.90
_*
0.66 - 0.89
-------
Leachate 63
Table III: Dry Matter and Mineral Element Contents of Batches of Manure Obtained Over a Period of 4 Years in Southern California.
Batch
number**
1
2
3
4
5
6
7
8
Dry
matter
80
61
49
62
75
68
80
77
Element***
N
2.
1.
2.
1.
2.
1.
1.
0.
Dairy
2
3
2
4
1
5
5
75
Beef feedlot
1
2
3
4
5
6
7
8
* From a
of Soil
ornia,
9.
12.
10.
10.
10.
10.
9.
15.
7
2
0
9
2
7
0
8
5.
4.
5.
4.
5.
4.
5.
2.
6
4
0
3
2
1
2
4
P
manure
0.55
0.38
1.25
0.81
0.48
0.54
0.37
0.34
slurry
0.93
0.78
1.9
1.7
1.0
0.81
•0.95
0.51
o/_ _ _
K
from corral
2
2
4
3
3
3
2
2
.6
.7
.8
.5
.1
.4
.4
.6
for slatted
3
2
3
3
4
.9
.3
.4
.4
.1
2.9
3
2
.6
.5
Na
s
1
1
1
0
0
0
0
0
floor
2
2
1
1
1
1
1
0
field experiment by P.P. Pratt and Sterling Davis
Science and Agricultural Engineering, University
Riverside, California 92502.
k* Odd-numbered batches were
the spring.
***E1 ement
concentrations
obtained in
expressed
on a
the fall and
dry wei
.0
.0
.1
.76
.75
.82
.40
.47
.7
.1
.6
.6
.8
.4
.2
.54
Cl
1
1
2
1
1
1
1
2
3
3
2
2
3
2
2
0
.3
.3
.0
.2
.4
.4
.7
.2
.7
.1
.4
.2
.2
.6
.6
.92
, Department
of Calif-
even-numbers
in
ght oasis.
-------
64
Leachate
A more rational approach would be to determine the
availability of nitrogen in the manure based on its properties.
For example, one could analyze each batch of manure for its
nitrogen content before it is applied and then calculate amounts
to add based on previously determined nitrogen availability
coefficients. Pratt et al.22 developed a simple method of calculat-
ing the mineralization of nitrogen added in manures. The miner-
alization rates, referred to as decay series, were related to the ni-
trogen contents of manures as well as the nature of the manure.
Poultry manure and cattle manure do not necessarily have the
same nitrogen mineralization rates at the same nitrogen con-
tents. Dry chicken manure at 3% nitrogen on a dry weight basis
will provide much more available nitrogen than fresh feedlot
slurry manure that has 3% nitrogen on a dry weight basis. Pow-
ers et al.21 developed an approach for estimating manure rates
based on the decay series and the soluble salt content of the
manure.
The decay series is a set of coefficients for the yearly minerali-
zation of a given application of a manure. For example, the
series 0.35, 0.15, 0.10, 0.05—implies that 35% of the nitrogen
will mineralize the first year, 15% of the residual nitrogen will
mineralize the second year, 10% the third year, and 5% of the
residual will mineralize the fourth and each subsequent year. By
keeping records of how much manure is added and the decay
series of each application, one can calculate the amount of nitro-
gen that will be mineralized each year for an indefinite period of
time. This concept is presented in Figures 4 and 5. Figure 4
shows the increase in yearly mineralization over a 20-year period
if manure is added at a constant rate, whereas Figure 5 shows the
downward adjustment of yearly rates of application of manure
to adjust for mineralization of residual organic nitrogen that
accumulates so that various constant rates of yearly mineraliza-
tions are obtained.
600
500
o 400
-C
o>
o
LU
N
cr
LU
300
200
100
DECAY SERIES = 0.35,015,0.10,0.05
DRY MANURE
Mt ho"1 yr"1
50
40
20
10
10 15
TIME, years
20
25
Figure 4: YcarK Mineralization Rate in Relation to Time for
Various Constant Rates of Corral Manure Having 1.59c N and
the Decay Series Indicated
LL)
(r.
=>
O
LJ
a:
LJ
rr
60
50
40
30
20
10
DECAY SERIES = 0.35,O.I5,010,0.05
CONSTANT YEARLY
MINERALIZATION RATE
kg ha"1 yr"1
300
200
100
10
15
20
25
TIME, years
Figure 5: Yearly Rates of Application of Manure Having 1.5%
Nitrogen and the Indicated Decay Series Required to Give Var-
ious Constant Yearly Rates of Nitrogen Mineralization.
Powers et al.21 developed a formula for calculating the rate of
manure based on the nitrogen requirement of the crop, the nitro-
gen content of the manure, the available nitrogen in the soil and
expected inefficiencies of available nitrogen because of volatili-
zation losses and denitrification. The equation, expressed in
English units, is
RJ =
J-l i=J
20CIDJ+E DJ+1(1-S D.)]
J=l i=l
where R, = application rate for the Jth year in tons manure
per acre.
N£ = nitrogen used by the crop in pounds per acre.
Ng = nitrogen available in the soil in pounds per acre.
N^ = nitrogen loss expected from denitrification and
volatilization.
C = concentration of N in the manure in percent.
Dj= first term in the decay series (dimensionless).
DJ+J = the (J+l)th term in the decay series (dimen-
sionless).
Dj = ith term in the decay series (dimensionless).
The limit to Rj was based on a value, Rm, which is a factor for
build-up of inorganic salts and is a function of soil, salinity of the
irrigation water, amount of leaching and the salt content of the
manure.
This decay series approach requires that (1) the manure be an-
alyzed for nitrogen and soluble salts, (2) the farm manager keep
accurate records of manure applications and their decay series,
(3) estimates of losses to the atmosphere by denitrification and
by ammonia volatilization are predictable, (4) available nitro-
-------
Leachate
65
gen in the soil at the start of each season be determined, and (5)
crop needs are known. The practical application of this
approach as a management procedure is doubtful even if it were
possible. An alternative and much simpler approach would be to
analyze the soil each year to determine the available nitrogen
and the expected mineralization during the season, analyze the
manure to be added and estimate how much to add to supply the
needed extra nitrogen based on a first-year mineralization rate
for that material. Unfortunately, we do not have reliable
methods for determining the available nitrogen and expected
mineralization of nitrogen that can be used for all soil-climate-
management conditions but estimates of both can be obtained
for many such combinations.
The decay series approach may not be useful directly in mak-
ing management decisions because of the lack of specific
information and lack of record keeping, but the concept can and
has been used to illustrate improper management principles.
Pratt et al.23 used decay series to calculate the large potential
leaching of nitrate from large applications of manure followed
by a period of years of no applications as compared to smaller
yearly applications. Decay series and calculations based there-
on can be used as models to predict effects of various
management alternatives. Meek et al.18 found in field plots that
more nitrate leached from a fine-textured soil when a large
application was followed by two years with no manure as com-
pared to manure applications each year. The annual application
of manure apparently provided a yearly input of available car-
bon that promoted increased denitrification.
Soluble Salts. The amounts of soluble salts provide an upper
limit to the use of manures on cropland because of reduced crop
yields as a result of increased soil salinity. This upper limit
depends on the salinity of the soil, of the water if the land is irri-
gated and of the manure, the salt tolerance of the crop and man-
agement factors that promotes leaching to remove the salt.
These upper limits in most cases are usually not far above the
point of optimum supplies of plant nutrients. Thus, manage-
ment to use the manure for crop production is not compatible
with using the soil to dispose of the manure.
The amounts of soluble salts contributed by manures to
groundwaters depends on (1) the composition of the manure
and amount added, (2) the amounts of elements removed by
crops, and (3) the quantities of various constituents precipitated
and/ or adsorbed in the soil. These are all highly variable factors
making general predictions rather unreliable.
In acid soils of humid climates, the precipitation of calcium
carbonate and calcium phosphates are effectively prevented
leaving the calcium to leach. On the other hand, in alkaline soils
of irrigated arid regions, both calcium carbonate and calcium
phosphate precipitate as a result of adding manure such that
much less of the added calcium leaches. In fact, the addition of
manures to some irrigated lands may not increase the leaching of
salt in direct proportion to the salt content of the added manure
because of precipitation reactions.
Sewage Sludges
Composition
Sewage sludge is a by-product of the treatment of domestic
wastewater. In the sewage treatment process inorganic and
organic substances in the wastewater are separated, and these in
addition to biomass produced in the treatment process and
excess water make up the bulk of sludges.
The solids content of sludges varies and tends to increase with
the degree of treatment. Primary and secondary sludges have
bonded water and settle out slowly; their solids content ranges
from 1 to 5%. Digested sludges characteristically retain less
water and their solids content commonly ranges from 6 to 12%.
Greater amounts of water can be removed from sludges by heat
treatment, chemical additives and/or mechanical aids. The
degree to which sludges are dewatered has a marked influence
on the concentration of inorganic nitrogen. The inorganic forms
of nitrogen in anaerobically and aerobically digested sludges are
ammonium (NH4) and nitrate (NO3) ions, respectively. Since
these forms of nitrogen are highly soluble, they tend to remain in
the liquid phase, and as the degree of dewatering increases, their
concentrations in sludges tend to decrease.
The chemical composition of sludge is highly variable. Being
an aggregate of impurities originating in wastewaters, the chem-
ical composition of sludges depends upon the source water, the
kind and extent of sewage treatment, the nature of the sewage
treatment process, the nature of the collection and conveyence
system, local conditions relative to consumer products entering
sewers through domestic use, and the kinds and percentages of
industrial input. Because of these factors the composition of
sludges are highly variable within the same treatment plant in
relation to time and among treatment plants in different
communities.
Data considered representative of the chemical composition
of sludges generated in the U.S.A. are presented in Table IV.
Although sludges are seldom treated to reach a completely dry
stage, all data are expressed on a dry weight basis to provide
more convenient comparisons. The data have been derived from
recent published reports (Baker and Chesin,2 Furr et al.,6 Kon-
rad and Kleinert,15 Page,20 Singh, Reefer, and Horvath," Shipp
and Baker,27 Harter,7 Blakeslee,3 and Sommers29). The concen-
trations given may differ from those of other published reports
because we have exercised some judgment in eliminating
extremes at both ends of the scale where abnormally high or low
concentrations for various elements occurred in sludges from
only a very limited number of those surveyed.
As with all materials which in one way or another find their
way into the surface of soils, the extent to which sludges may
contribute to the quality of the leachate depends upon (a) the
chemical composition, (b) the amount applied, (c) amounts
removed by volatilization or in harvested crops, and (d)
amounts precipitated in soil or dissolved from soil. The cation
exchange complex in soils will influence the ionic composition
of electrolyte in the leachate, but will not alter the total composi-
tion when expressed in chemical equivalents.
Impact on Leachate Quality
When sludges are applied to soil, a series of complex chemical
and microbial reactions take place. These reactions are time and
temperature dependent, and are governed largely by the
chemical properties of the soil and sludge. The systems have not
been studied in sufficient detail to predict the time required to
achieve a steady state leachate composition in the sludge-soil
mixtures. The information which is available, however, suggests
that at temperatures which are not extreme, after a few months
changes which occur in leachate composition are gradual. Fol-
lowing sludge application leachate composition will depend
upon the quality and quantity of water which enters and moves
through the mixture.
Trace elements. For major changes in leachate composition in
sludge amended soils most probably occur in a transition zone
immediately below the depth of incorporation of the sludge. The
extent to which the concentrations of trace elements are changed
in this zone are determined largely by the chemical properties of
the soil. The chemical properties of soil are governed largely by
the amount and kind of colloidal clay and organic matter. Solu-
tions passing through soils low in organic matter and with high
percentages of sand-sized particles generally have a lower capac-
ity to attenuate the composition of the solution passing through
them than do soils with high percentages of clay and organic
matter. Generally, those trace elements which occur in solution
in a molecular or anionic form tend to migrate to greater depths
-------
66
Leachate
in the soil profile than do trace elements which occur as cations
in solutions.
Following application of sludge to soil, the composition of the
leachate in relation to depth will depend largely on the factors
which control the solubility of the various trace elements. Soils
and sludges are dynamic complex systems and although prog-
ress has been made in our knowledge of their chemistry, we still
lack the necessary basic information to predict concentration of
trace elements in even the most simple sludge-soil mixtures.
Generally, when sludges are incorporated into soil the solubility
of most trace elements is reduced considerably.
Where sludges are used in agriculture as a plant nutrient
supplement, application rates are commonly based upon the ni-
trogen requirement of the crop grown, and usually would not
exceed 40 metric tons per hectare per year. At annual sludge
application rates of this order of magnitude for periods of at
least 10 years, the available information indicates that move-
ment of elements such as Cd, Co, Cr, Cu, Hg, Mo, Mn, Ni, Pb,
and Zn beyond one meter depth in the soil profile is unlikely.
Kirkham,14 for example, did not observe movement of Cd, Cu,
Ni, Pb, and Zn below 61 cm depth following applications of ap-
proximately 28 metric tons of dry sludge per hectare per year for
a 35-year period. Page20 evaluated data presented by Anders-
son and Nilsson1 and demonstrated that following annual appli-
cations of seven metric tons of sludge per year for twelve years,
essentially all of the applied Mn, Zn, Cu, Ni, Co, Cr, Cd, Hg, As,
and Se were recovered in the surface 20 cm of soil. Only B moved
in substantial amounts to depths lower than 20 cm in the soil
profile. In a study at the Werribee Sewage Farm at Melbourne,
Australia, Johnson et al.9 observed practically no movement of
Zn, Cu, Ni, and Cd below 45 cm in a field which had been irri-
gated with raw sewage for 70 years.
In systems involving long-term, high rate volume operation
for the disposal of sludge, movement of trace elements to depths
of at least three meters has been observed. Lund, Page, and Nel-
son17 presented data which show movement of Cd, Cu, Ni, and
Zn to a depth of at least three meters in a soil beneath a sewage
lagoon which had been used to pond sewage sludge for a period
of at least 20 years.
In summary, soils except for sands have a high capacity to
retain As, Cd, Cr, Cu, Hg, Ni, Pb, Se, and Zn. In situations
where sludges are applied to agricultural soils to supplement
plant nutrients or as a source of irrigation water, the probability
that leachate below one meter depths will show enrichment in
the above elements is light. Boron, however, is an exception.
Except in highly alkaline soils (pH >8.5), B occurs in soil solu-
tions as undissociated boric acid [B(OH) .i). It is sorbed by active
oxides of iron and aluminum in soil, but its affinity for these
solid surfaces is low compared to affinities of other trace ele-
ments for these surfaces. For this reason, boron is quite mobile
in soils. In sandy soils low in organic matter and active iron and
aluminum oxides, B in water passes through soils essentially
unchanged. Organic matter and active iron and aluminum tend
to limit the mobility of B in soils, but generally the equilibrium is
such that substantial percentages of the B remain in solution and
move with the percolating water.
Phosphorus and Potassium
Phosphorus concentrations of sewage sludges range from
about 0.5 to 3.0% with typical concentrations of about 1.5%
(Table IV). In liquid sludges dissolved P occurs as orthophos-
phates, polyphosphates, and organic phosphates. In acid soils
phosphorus is adsorbed or precipitated by crystalline and/or
amorphous iron and aluminum oxides. In neutral or calcareous
soils, orthophosphate reacts with calcium to form, depending
upon ionic composition, a number of slightly soluble or insolu-
ble calcium phosphates.
Table IV: Typical Concentrations of Various Elements in
Sewage Sludges from the U.S.A.
Elements
N
P
S
K
Ca
Mg
Fe
Al
Cl
B
Cu
Mo
Mn
Zn
As
Cd
Co
Cr
F
Hg
Ni
Pb
Se
Range
1.0 -10.0
0.5 3.0
0.5 1.5
0.05- 1.0
1.0 -10.0
0.2 1 0
1.0 6.0
0.5 -10.0
0.1 1.0
- - - - ug/g
5- 150
100-4,000
1-40
100-800
1,000-6,000
5-50
5-500
1-20
200-20,000
2-400
0.5-50
10-4,000
20-20,000
2-20
Median
3.0
1.5
1.0
0.3
4.0
0.4
1.0
0.5
0.4
_____
40
750
5
300
1,800
10
15
10
400
50
7
60
450
4
Mean
4.0
2.0
1.0
0.4
5.0
0.5
1.5
1.5
0.4
_ _
70
1,000
15
500
2,800
25
80
10
2,000
170
10
350
1,800
10
A number of long-term fertility trials and studies involving
land applications of sewage sludges have demonstrated that
phosphorus movement to lower depths in soil profiles is quite re-
stricted. Pratt et al. (1956) reported that following application of
a total of 1756 kg P per hectare (as treble superphosphate) over a
28-year period more than 80% of the applied P remained in the
surface 30 cm of soil and no movement of P was observed at
depths greater than 90 cm. Similarly, William and David31
observed essentially no movement of P below 10 cm depth in a
soil which had received 2500 kg P per hectare over a forty-year
period.
In systems where liquid sludges are applied at high rates,
phosphorus movements to considerable depths in soil profiles
has been observed. Hook et al.8 observed movement of phos-
phorus to a depth of 120 cm after irrigation with a sewage efflu-
ent at a rate of 5 cm per week for a six-year period. The concen-
trations of phosphorus at the depths sampled were greater for a
sandy loam soil than a clay loam soil. Bray extractable phospho-
rus (0.025 N HC1 - 0.03 N NH4F) increased to a depth of 90 cm
in a sandy loam profile, but only to 30 cm in a clay loam profile.
In the sewage farm serving Melbourne, Australia, Johnson et
al.9 reported total concentrations of phosphorus at depths
between 25 and 45 cm in soils irrigated with primary sewage
effluent for a period of 45 to 73 years which were twice those of
the non-effluent treated soils. Most of the phosphorus, howev-
er, accumulated in the surface 25 cm of soil. Lund, Page, and
Nelson17 followed the movement of phosphorus in soil beneath
lagoons used as percolation ponds for liquid sludges. In their
study movement of phosphorus to a depth of at least 3 meters
was observed.
It seems reasonable to conclude that in situations where
-------
Leachate
67
sludges are used in agriculture as a nutritional supplement for
long periods of time (>25 years) the probability that this prac-
tice would cause P contamination of leachates from soils at
depths greater than a few meters is remote.
Potassium concentrations of sludges on a dry weight basis are
usually less than those of animal manures. The rate of move-
ment of K in soil is minimal, and except in very sandy soils, the
probability of a significant K contribution from sludge applica-
tions to leachate below one meter depth in soil is nil.
Nitrogen
Concentrations of nitrogen in sewage sludges range from 1 to
10% (Table IV). In liquid anaerobically digested sludges from 30
to 60% of the total N is present in the ammonical form and the
remainder in the organic form (Metcalf and Eddy19). If liquid
sludges are applied directly to land, a certain percentage of the
ammonical N is lost by volatilization as NH3 as the sludge dries
in the field. The actual amounts lost by volatilization reported in
the literature vary widely and range from a few percent to almost
complete loss (Coker5; Ryan and Keeney25; King12). Incorpora-
tion of liquid sludges below the soil surface serve to minimize N
losses by this process.
Nitrogen in organic forms applied to soils must undergo
mineralization to inorganic forms (NH^~and NOj) before it can
be utilized by plants or to NO^form before it is readily leachable.
The rate of mineralization, a microbial process, depends upon a
number of soil factors such as water content, aeration, pH, and
temperature. Although precise rates for various soil types and
climatic regions have not been completely worked out, the avail-
able data indicate that from 15 to 35 percent of the organic N is
mineralized the first year of the application (Pratt, Broadbent,
and Martin22; Keeney, Lee, and Walsh"; Ryan et al.26; King12).
Lesser percentages of the remaining organic N are mineralized
in subsequent years. For example, Keeney, Lee and Walsh11
suggest 15% mineralization in a sandy loam soil the first year,
6% the second year, 4% the third year, and 2% thereafter.
The first product of mineralization of organic N is NHJ~ions.
If the pH of the system is less than 4.5, further conversion to NO3"
is completely inhibited and the N remains as NH^ The NH^"is
adsorbed on the negatively-charged surfaces of the soil colloids
and, as such, leaching of N is highly restricted. In most soils
(sands are possible exceptions) leaching of N in the NH^form is
so small that for all practical purposes it can be ignored. If the
pH of the soil exceeds 9.0, sufficient NH3 forms to inhibit nitrifi-
cation to NOj~and N can be lost by NH3 volatilization.
At soil pH's in the range 4.5 to 8.5, and under conditions when
the dissolved O2 exceeds 0.2 mg/1, the NH+" formed by
mineralization of organic N is rapidly oxidized to NO3". Nitrate
is quite mobile in soils and that which escapes the zone in soil
under the influence of evapotranspiration will leach with water
to lower depths in the soil profile. In soils where regions of low
dissolved O2 (<0.2 mg/1) occur at depths within or below the
rooting zone, if there is sufficient organic carbon, NO^can be
reduced and lost as N2 gas. Bouwer and Chaney4 state that about
one milligram of organic carbon is required to reduce one milli-
gram of nitrate nitrogen to N2 gas.
Once NO 3-N escapes the zone in soil where it is removed by
plants or denitrified, its leaching in soil parallels quite closely the
movement of water. The amounts of N available for leaching are
therefore the difference between the amount applied and the
sum of amounts removed by harvested crops plus volatilization
losses as either NH3 or N2.
A number of studies have demonstrated substantial
concentrations of NOj at depths in soil well below the root zone.
Lund, Page, and Nelson16, for example, reported concentrations
of NOjN greater than 400 mg/1 at a depth of 12 meters below a
sludge lagoon. The lagoon had been used to dewater sludge at
the treatment plant for about 20 years. On an offsite control at a
comparable depth (12 meters) the concentration of NOjN was
30 mg/1. King and Morris13 and Kardos and Sopper10 have
observed substantial increases in NO3N in soil solutions at a
depth of 1.2 meters beneath fields receiving either sludge or
effluent applications.
Although the above observations show high concentrations
of NO3N in leachate at lower depths in the soil profile, it is not
possible to evaluate the impact on groundwater quality. As men-
tioned previously, mass emissions of NO^, which depend on vol-
ume of flow, must be determined before an evaluation of the
extent of groundwater contamination can be made.
SUMMARY
The amounts of elements in fertilizers, manures, and sewage
sludges which when applied to soils will enter leachates are
determined by the difference between the amounts added plus
amounts solubilized in soils less the amounts removed in har-
vested crops, lost by volatilization, and precipitated or sorbed
by the soil. Organic forms of nitrogen added to soils, regardless
of source, must undergo mineralization to NOj-N forms before
extensive leaching of the added nitrogen will occur. The rates of
mineralization are not precisely known for the various materials
and environmental conditions, but most likely they are from 15
to 35 percent of the applied organic nitrogen during the initial
year following application, with progressively lesser percentages
of the remaining organic nitrogen mineralized each subsequent
year. Once NOjN leaches to depths in soil not influenced by eva-
potranspiration, the NOjN will migrate with water essentially
unchanged. To minimize NOjN leaching, quantities of plant
available nitrogen in excess of that required for optimum crop
production should not be applied.
Nitrogen applied to soils can be volatilized as NH3, N2, or N2O
gas. Ammonia is lost when ammonical sources of N dry on the
surface of soils or in soils where the pH is greater than 8.5. In
soils under anaerobic conditions NOj-N is reduced to N2 and
N2O gas and lost from the soil system.
Commercial fertilizers are applied to soils in amounts
sufficient to meet nutritional requirements of the crop. Potas-
sium and phosphorus are relatively immobile in most soils
(sands are exceptions) and the probability that soil applications
of commercial fertilizer sources of these elements will cause
substantial enrichment of leachates in soils below depths of one
meter in the soil profile is remote. Trace elements (Cu, Zn, Mn,
Fe) are frequently added to commercial sources of N, P, and K
in small amounts to serve as sources of the plant essential
micronutrients. Since amounts added to fertilizers, or those that
occur naturally as contaminates are very small and soils have a
high capacity to attenuate these elements, commercial fertilizers
are not expected to serve as a source of enrichment of trace
elements in leachates from soils.
Leachate enrichment resulting from the practice of spreading
manures is usually restricted to soluble nitrate, chloride and
bicarbonate salts of sodium, calcium, and magnesium. Phos-
phorus and potassium present in manures are retained by the
soil and will not contribute to leachate contamination below one
meter depth in most soils. The concentrations of trace elements
in animal manures are low and, as such, the probability that
manures will cause trace element contamination of leachates is
also low.
Sewage sludges can be considered low analysis fertilizers.
Their composition varies widely and depends largely upon the
composition of the wastewaters entering the treatment plant. In
situations where sludges are applied to soils at rates sufficient to
supply fertilizer nitrogen, as with manures, enrichment of leach-
ate below one meter depth in the soil profile is usually restricted
to soluble nitrate, chloride, and bicarbonate salts of
-------
68
Leachate
sodium, calcium and magnesium. Prolonged use of sludges or
where application rates greatly exceed those required to supply
fertilizer nitrogen, however, may cause trace element and phos-
phorus contamination of leachates at considerable depths in the
soil profile. The extent of contamination will depend upon the
composition of the sewage sludge, the total amounts applied and
the physical, chemical, and biological properties of the soil.
REFERENCES
1. Andersson, A. and K. O. Nilsson. 1972. Enrichment of
Trace Elements From Sewage Sludge Fertilizer in Soils and
Plants. AMBIO, 1(5):176-179.
2. Baker, D. E., and L. Chesnin. 1975. Chemical Monitoring
of Soils for Environmental Quality and Animal and Human
Health. Adv. Agron. 27:305-374.
3. Blakeslee, P. A. 1973. Monitoring Considerations for
Municipal Wastewater Effluent and Sludge Application to
Land. Recycling Municipal Sludges and Effluents on Land
(Washington, D.C.: National Association State University and
Land-Grant Colleges) pp. 183-198.
4. Bouwer, H. and R. L. Chaney. 1974. Land Treatment of
Wastewater. Adv. Agron. 26:133-176.
5. Coker, E. G. 1966. The Value of Liquid Digested Sewage
Sludge. II. Experiments on Rye-grass in Southeast England,
Comparing Sludge with Fertilizers Supplying Equivalent Nitro-
gen, Phosphorus, Potassium and Water. J. Agr. Sci., Camb.
67:99-103.
6. Furr, A. K., A. W. Lawrence, S. S. C. Tong, M. C. Gran-
dolfo, R. A. Hofstader, C. A. Bache, W. H.Gutenmann,andD.
J. Lisk. 1976. Multielement and Chlorinated Hydrocarbon
Analysis of Municipal Sewage Sludges of American Cities. Env.
Sci. and Technol. 10:683-687.
7. Harter, R. D. 1975. A Survey of New Hampshire Sewage
Sludges as Related to Their Suitability for On-land Disposal.
Station Bulletin 503, New Hampshire Agricultural Experiment
Station, University of New Hampshire, Duran, New
Hampshire.
8. Hook, J. E., L. T. Kardosand W. E. Sopper. 1973. Effect of
Land Disposal of Wastewaters on Soil Phosphorus Relations.
In Recycling Treated Municipal Wastewater and Sludge
Through Forest and Cropland. Penn. State University Press,
University Park, Pennsylvania.
9. Johnson, R. D., R. L. Jones, T. Hinesly, and D. J. David.
1974. Selected Chemical Characteristics of Soils, Forages, and
Drainage Water from the Sewage Farm Serving Melbourne, S.
Australia (Dept. of Army, Corps of Engineering).
10. Kardos, L. T. and W. E. Sopper. 1973. Renovation of
Municipal Wastewater Through Land Disposal by Spray Irriga-
tion, pp. 148-163. In Recycling Treated Municipal Wastewater
and Sludge Through Forest and Cropland. Penn. State Univer-
sity Press, University Park, Penn.
11. Keeney, D. R., K. W. Lee, and L. M. Walsh. 1975.
Guidelines For the Application of Wastewater Sludge to
Agricultural Land in Wisconsin. Tech. Null. No. 88, Dept. of
Natural Resources. Madison, Wisconsin.
12. King, L. D. 1973. Mineralization and Gasseous Loss of
Nitrogen in Soil-Applied Liquid Sewage Sludge. J. Environ.
Quality 2:356-358.
13. King, L. D. and H. D. Morris. 1972. Land Disposal of Liq-
uid Sewage Sludge: III. The Effect of Soil Nitrate. J. Environ.
Quality 1:442-446.
14. Kirkham, M. B. 1975. Trace Elements in Corn Grown on
Long-Term Sludge Disposal Site, Environ. Sci. Technol.
9:765-768.
15. Konrad, J. G. and S. Kleinert. 1974. Removal of Metals
From Waste-Waters by Municipal Sewage Treatment Plants. In
Surveys of Toxic Metals in Wisconsin. Tech. Bull. No. 74, Dept.
of Natural Resources. Madison, Wisconsin, pp. 2-7.
16. Lund, L. J., A. L. Page and C.O. Nelson. 1976a. Nitrogen
and Phosphorus Levels in Soils Beneath Sewage Disposal
Ponds. J. Environ. Quality 5:26-30.
17. Lund, L. J., A. L. Page and C. O. Nelson. 1976b. Move-
ment of Heavy Metals Below Sewage Disposal Ponds. J.
Environ. Quality 5:330-334.
18. Meek, B. D., A. J., McKenzie, T. J. Donovan, and W. F.
Spencer. 1974. The Effect of Large Applications of Manure on
Movement of Nitrate and Carbon in an Irrigated Desert Soil. J.
Environ. Quality 3:253-258.
19. Metcalf and Eddy. 1972. Wastewater Engineering.
McGraw-Hill Book Co., New York.
20. Page, A. L. 1974. Fate and Effects of Trace Elements in
Sewage Sludge When Applied to Agricultural Lands. Environ.
Protection Tech. Series, EPA-670/2-74-005. (U.S. Environ.
Protection Agency) Cincinnati, Ohio.
21. Powers, W. L., G. W. Wallingford, and L. S. Murphy.
1975. Research Status on Effects of Land Application of Animal
Wastes. EPA-660/2-75-010. U.S. Government Printing Office,
Washington, D.C. 20402. Stock No. 055-001-01026.
22. Pratt, P. R., F. E. Broadbent, and J. P. Martin. 1973.
Using Organic Wastes as Nitrogen Fertilizers. Calif. Agr.
27(6): 10-13.
23. Pratt, P. F., S. Davis, J. Warneke, and R. G. Sharpless.
1976. A Four-year Field Trial with Animal Manures: II. Miner-
alization of Nitrogen. Hilgardia. In Press.
24. Pratt, P. F., W. W. Jones and H. D. Chapman. 1956.
Changes in Phosphorus in an Irrigated Soil During 28 Years of
Differential Fertilization, Soil Sci. 82:295-306.
25. Ryan, J. A. and D. R. Keeney. 1975. Ammonia Volatiliza-
tion From Surface Applied Sewage Sludge. J. Water Poll. Cont.
Fed. 47:386-393.
26. Ryan, J. A., D. R. Keeney, and L. M. Walsh. 1973. Nitro-
gen Transformation and Availability of an Aerobically Digested
Sewage Sludge in Soil. J. Environ. Quality 2:489-492.
27. Shipp, R. F., and D. E. Baker. 1975. Pennsylvania's Sew-
age Sludge Research and Extension Program. Compost Sci.
16(2):6-8.
28. Singh, R. N., R. F. Keefer, and J. D. Horvath. 1975. Can
Soils Be Used For Sewage Sludge Disposal? Compost Sci
16(2):22-25.
29. Sommers, L. E., D. W. Nelson, J. E. Yahner, and J. V.
Mannering. 1972. Chemical Composition of Sludge From
Selected Indiana Cities. Ind. Acad. of Sci. 82:424-432.
30. Spencer, W. P. 1957. Distribution and Availability of
Phosphates Added to a Lakeland Fine Sand. Soil Sci Soc
Amer. Proc. 21:14-144.
31. William, C. H. and D. J. David. 1973. The Effect of Super-
phosphate on the Cadmium Content of Soils and Plants Aust
J. Soil Res. 11:43-56.
-------
The Effects of Industrial Sludges
on Landfill Leachates and Gas
D. R. Streng
Systems Technology Corporation
Cincinnati, Ohio
INTRODUCTION
Environmental effects from landfilling result from not only
soluble and slightly soluble material disposed of in the landfill,
but also from the products of chemical and microbial transfor-
mations. These transformations should be a consideration in
management of a landfill to the extent that they can be predicted
or influenced by disposal operations. The motivational aspects
of this project were the lack of quantitative data on the decom-
position process under field conditions, and, in particular, the
effects of the introduction of solid and semi-solid industrial
waste materials with municipal solid waste.
Previous reports on this study have dealt mainly with place-
ment of the waste materials and concentration/time relation-
ships. This presentation will deal with total mass flows emanat-
ing from these cells.
Approach
In an effort to evaluate the many variables which may effect
gas and leachate production within a sanitary landfill environ-
ment, nineteen (19) large scale environmental test cells are being
monitored to achieve the following objectives:
I. Assess the effects of varying rainfall regimens, 203.2 to
812.8 mm/yr (8.0 to 32.0 in/yr).
2. Determine the impact of municipal sewage sludge addi-
tions on the rates of decomposition and gas and leachate
production. ,
3. Determine the impact of decomposition rates by the
addition of a pH buffer (limestone) into waste during
landfill construction.
4. Determine the survival of polio virus in a landfill envi-
ronment.
5. Determine if differences in ambient soil temperatures
significantly effect the rates of decomposition and gas
and leachate production.
6. Determine the impact on decomposition in gas and lea-
chate production from the co-disposal of six (6) selected
industrial residuals with municipal solid waste.
7. Determine the ability of duplicate test cells to generate
similar physical and analytical data.
8. Determine the impact on decomposition rates by rapidly
bringing the municipal solid wastes to field capacity.
The test cells (experimental landfills), employed for this study
were epoxy coated steel, 1.8m (6 ft) in diameter and 3.6m (12 ft)
in height; capable of holding approximately 3000kg (6600 Ibs) of
municipal solid waste in a manner comparable to large area
landfills. The size of the test cells was selected to minimize the
problems of scaling factors generally associated with smaller
laboratory lysimeters and to avoid the use of shredded refuse. A
total of fifteen (15) cells were placed in the ground outdoors with
the remaining four (4) cells in the enclosed bay area where higher
ambient temperatures were maintained. Prior to placement of
any solid and/or industrial wastes, a layer of silica gravel,
300mm (1 ft) deep, was placed in all cells as a base for the solid
waste and to allow leachate to permeate to the drain system. All
test cells were coated with coal tar based epoxy paint which was
proven to be resistant to leachate degradation.
All test cells were loaded simultaneously in a period of five (5)
days employing municipal solid waste from the City of Cincin-
nati. The solid waste was placed into experimental landfills in
370kg (800 Ib) increments and compacted to a height of 300mm
(I ft), and a density of 470kg/cu m (800 Ib/cu yd) until each test
cell had received 2.4m (8 ft) of solid waste or approximately
3000kg (6600 Ibs) of compacted material. Industrial residuals
were added to the solid waste as it was placed into the experi-
mental test cells.
The solid waste was then covered with a layer of compacted
clay and all cells were shielded from both moisture and sunlight.
Temperature and gas monitors are installed throughout all cells
and in the soil. Water addition to the cells is at a rate of 406mm
(16 in) per year and is accomplished on a monthly basis in accor-
dance with anticipated net infiltration for the midwestern por-
tion of the country, Figure 1.
12 13 14 15 16 IT IB 19 20 21 22
Figure
TIME (MONTHS)
: Water Addition Regimen
Industrial Residuals Under Study
The industrial process residuals evaluated include a refinery
sludge (RS), a battery production waste (BPW), an electroplat-
ing waste sludge (EW), an inorganic pigment sludge (IPW), a
69
-------
70 Effects of Industrial Sludges
chlorine production brine sludge (CPBS), and a solvent based
paint sludge (SBPS). Physical characteristics and amounts of
each industrial waste added is provided in Table 1.
Refinery sludge is a byproduct of the refining of crude oil. The
waste material obtained is an API bottoms sludge. This is mate-
rial resulting from gravity oil / water separator which is very high
in biological activity.
The battery production waste is a composite from all phases
of a lead/acid storage battery manufacturing operation, with
the exclusion of the battery plate assembly section. This opera-
tion is a closed system from which the waste material is shipped
to a smelter for the recovery of lead. The waste obtained goes to
a neutralization process before entering a settling pond from
which the sludge was obtained.
The electroplating waste was obtained from a large plating
firm which employs a variety of plating processes including:
chromium, nickel, cadmium, copper, iron and zinc. No tin plat-
ing occurs at this location. These wastes were treated in various
categories.
Chromium wastes were converted to tnValent chromium by
the addition of sulfur dioxide in an acid system (pH of 3). The
pH was then raised to approximately 8.0 with caustic to precipi-
tate the trivalent chromium. The cyanide containing wastes were
treated by raising the pH of the water with caustic and then
breaking down the cyanogen system by the addition of sodium
hypochlorite. All other acid wastewaters were treated with base
to elevate the pH to approximately 8 to 9. Wastewaters were
then pumped to a lagoon for settling.
The inorganic pigment waste is a byproduct of the processing
of raw ore and the production of titanium dioxide. Wastewaters
from various manufacturing operations are pumped to a single
facility and treated simultaneously. This waste material was
then pumped to a primary settler during which the pH was
raised with base and an alkaline precipitate formed. The sludge
was then dewatered and the residue removed for disposal.
The chlorine production brine sludge is from a mercury cell
chlorine production plant. In this facility the production of chlo-
rine was accomplished by the electrolysis of sodium chloride in a
mercury cell. The majority of the sludge (60 to 80%) was from
the brine saturator. While the remainder was from the clarifier
after settling.
The solvent based paint sludge used was representative of a
paint sludge produced in industries involved in painting large
numbers of metal products. The paint overspray was caught in a
water curtain and was pumped directly to a holding tank for dis-
posal with little or no pre-treatment.
The chemical analyses of these waste materials, Table 2, indi-
cated that the sludges contained significant levels of potentially
toxic materials. The teachability of these materials, however,
still was not known.
Leachate Composition
Immediately after compaction, small volumes (less than I
liter) of leachate were obtained from the majority of the cells.
Based on the chemical characteristics and time of appearance it
was concluded that these initial volumes of liquid were intersti-
tial waters squeezed from the refuse or refuse/sludge mixtures as
a result of the compactive effort and were not truly leachate
derived from infiltrating water and decomposition processes.
After collecting and the initial squeezings from the test cells, no
further liquid was produced for a period of approximately six
(6) months. T he early occurrence of leachate was attributable to
several factors, first, the number of available moisture retention"
sites was reduced by the addition of the moist industrial wastes
and for this reason field capacity was approached at a much ear-
lier date than originally expected. Secondly, it appeared that
Table I: Waste Stream Characteristics
CELL WASTE
STREAM
9 RSC
10 BPWd
12 EWe
13 IPWf
14 CPBS9
17 SBPSh
aPERCENT BY WET WEIGHT
"Kg
CREFINEBY SLUDGE
dBATTERY PRODUCTION WASTE
eELECTROPLATING WASTE
MOISTURE AMOUNT
CONTENT3 ADDEDb
79.00 1518
89.25 1291
79.53 1191
51.75 1420
24.11 2039
24.75' 1604
ANORGANIC PIGMENT WASTE
9CHLORINE PRODUCTION BRINE SLUDGE
hSOLVENT BASED PAINT SLUDGE
'MAINLY ORGANIC SOLVENTS
CHARACTERISTICS
HIGH BACTERIAL
ACTIVITY; BLACK
MOIST
GREY, HIGH LIQUID
CONTENT
BROWN, SOUPY
BLACK, SOLID, NO
ODOR
VERY DENSE, NO ODOR,
LIGHT BROWN, MOIST
RED TO WHITE COLOR,
PUTTY CONSISTENCY,
STRONG ODOR
-------
Effects of Industrial Sludges
71
there was channelling within the solid waste/industrial waste
test cells. Channelling prolongs the time required to reach field
capacity, but allows the early appearance of leachate.
The behavior of the organic fraction of the solid waste/indus-
trial waste leachate is represented by graphical displays of
chemical oxygen demand (COD), Figures 2-4. The concentra-
tion histories for COD in leachates from the refuse alone and
from the refuse/industrial waste cells are very similar and after
approximately two years of monitoring appear identical. Dur-
ing the initial phases of this study, the COD from several of the
industrial waste cells (electroplating waste, battery production
waste) indicated a significantly higher chemical oxygen demand
than did the solid waste leachates. This is attributable to a chem-
ical demand being asserted on the leachates and not from the
presence of organic carbon analyses. After approximately eight
(8) months from placement of the solid waste or solid waste/in-
dustrial waste mixtures, the graphical displays for COD are
almost identical.
Table II: Industrial Waste Chemical Analysis21
WASTE CELL NUMBER
TOTAL SOLIDS*1
TOTAL VOLATILE SOLIDSh
MOISTURE11
CR
Ni
Cu
FE
As
BE
SE
CD
CN
PB
C1h
ASBESTOSJ
HG
SN
SB
CLAY VOLATILE FIBERSJ
ZN
V
B
Ti
RSb
9
21.00
31.00
79.00
125
23
3500'
5560
1.0
4.8
26.0
0.50
1.0
182
2.35
3.00
10.6
NAk
NA
40.0
NA
7.20
NA
BPWC
10
10.75
7.94
89.25
155
32
1125
2950
72
1.8
180
29.0
4.2
3.48h
1.12
208
4.80
6800
1.32h
720
120
8.10
NA
EWd
12
20.47
8.98
79.53
1.56h
35
100
1.37h
460
0.25
4.50
38.5
460
267
1.35
23.0
14.7
NA
NA
86.0
NA
19.0
NA
IPWe
13
48.25
22.25
51.75
0.50
10
110
1000
3.4
20.2
16.0
10.5
3.4
120
10.0
45.0
7.60
NA
NA
185
40
28.5
NA
CPBSf
14
75.89
1.17
24.11
5.00
65
125
2000
14.5
<1.0
16.5
0.70
14.5
697
20.0
110
227
NA
NA
480-
NA
1.70
<0.1
SBPS9
17
75.25
55.25
24.75
75.0
0.5
2.0
150
12.8
<1.0
7.60
0.50
12.8
12.6
0.75
9.00
16.7
NA
NA
65.0
NA
11.4
NA
aALL VALUES IN PPM UNLESS OTHERWISE SPECIFIED
bREFINERY SLUDGE
CBATTERY PRODUCTION WASTE
dELECTROPLATING WASTE
eINORGANIC PIGMENT WASTE
fCHLORINE PRODUCTION BRINE SLUDGE
9SOLVENT BASED PAINT SLUDGE
hPERCENT BY WET WEIGHT
"UNDERLINED VALUES INDICATE
JFIBERS/100G.
kNOT ANALYZED
MAXIMUM
SLUDGE CONCENTRATIONS
-------
72 Effects of Industrial Sludges
p SOLID WASTE (CELL 1>
~I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I I I
2 3 4 5 6 7 8 9 10 II 12 13 U 15 16 17 18 19 20 21 21
TIME(MONTHS)
Figure 2: Chemical Oxygen Demand
o-o tLLCIRUPLAIIilG HAbIL (CtlL U)
a—a INORGANIC PICxIllll HAS1L (CLLL IJ>
n~o CHtORIIIt PROUIIUIOS BHINt SLUDGt (CtLL 11)
Figure
TIME(MONTHS)
3: Chemical Oxygen Demand
In an effort to show the effects of a co-disposal of industrial
residuals upon the release of the nutrient parameters, the con-
centration histories for total Kjeldahl nitrogen (TK.N), Figures
5-8, are given. Initially the solid waste only was generating more
TK.N than the solid waste/industrial waste admixtures. This is
probably due to the fact that the nitrogenous compounds within
the industrial waste cells were not as amenable to degradation.
After a period of approximately eight (8) months, however, the
mass flows of TK.N are approximately equal.
i SOLli) UASIt (CtLL 1)
-I 1 1 1 1 1 1—1 1 1 1 1 1 1 1 1
234 5 6 7 8 9 10 II 12 13 14 15 16 IT IB 19 20 21 22
TIME (MONTHS)
Figure 5: Kjeldahl Nitrogen
o-« LLLillWLAll.lli UAblt (CLLL 12)
a-a 1,1.1;.,',.; 1C i'IG,"Ul WASIL (CELL li)
I 2 3 4 5 6 T 8 9 10 II 12 13 » 15 16 17 18 19 20 21 22
TIME (MONTHS)
Figure 6: Kjeldahl Nitrogen
hAIILRV PRODUCT 1011 UASIt (CELL 10)
. LHLuRLIL I'h.iln.f II u HKl.NL SLUJ&L (CtLL 11)
123*56789 10
"I 1 1—T 1 1 1 I 1
i« 15 16 17 18 19 20 21 22
"1 > 1 1 1 1 1 1
I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 1C, 17 18 19 20 21 22
TIME (MONTHS)
Figure 4: Chemical Oxygen Demand
Figure 7: Kjeldahl Nitrogen
-------
Effects of Industrial Sludges
73
;,\\\Ltlt rsOUUUIOil WbTL (CtLL 10)
HAI1LHY IWJUluN WAblE ULLL 10)
5 6 7 8 9 10
12 13 14 15 16 17 18 19 20 21 22
TIME (MONTHS)
Figure 8: Kjeldahl Nitrogen
Total solids, Figures 9-11, is one of the few parameters where
variances between the solid waste only cell and those containing
industrial wastes are evident. The mass flow diagrams for the
solid waste; the battery production waste, and the electroplating
waste are almost identical. The total solids leached from the
inorganic pigment waste is considerably lower than that leached
from the other cells while the chlorine production brine sludge is
substantially higher. The solids content within the leachate is
relatable to the concentration of several other pollutants, in par-
ticular metallic ions, but at this time we have not been able to
statistically evaluate the exact nature of this relationship. It does
appear, however, that the higher the solids content the higher
the pollutant potential of the leachate stream.
o—o jILIil HfV.TE if! II 'H
T 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I 1
I 2 3 4 5 6 7 6 9 10 II 12 13 14 15 16 17 IB 19 20 21 22
TIME (MONTHS)
Figure 9: Total Solids
o-o LLtCTROPLATMG rfAbTt (CtLL 12)
a-a INORGANIC PIGHtill UAbTt (CtLL 1>)
o-a CHLORI.IL PROUUCHO.I BRI,I[ bLUDGt (CELL 11)
I 234 56 78 9 10 II 12 13 14 15 16 '7
TIME (MONTHS)
i—i—i—i—i—i—i—r
I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 16 19 20 21 22
TIME (MONTHS)
Figure 11: Total Solids
It has been felt for some time that the co-disposal of industrial
residuals with municipal solid waste would generate a leachate
stream which would be higher in metallic ion content than that
from the municipal solid waste alone. In examining the total
mass flow of pollutants from these experimental landfills that
conclusion does not appear to be always valid.
The solid waste test cell is leaching zinc at a higher rate than
any of the industrial waste/solid waste mixtures, Figures 12-14.
During the initial phases of decomposition, the organic acid
content increased within all test cells and as it did the amount of
metallic ions being leached increased at a substantial rate. After
approximately twelve (12) months, the organic acids content
began to level out as well as the amount of metallic ion leaching.
At this point in time all test cells are leaching zinc at approxi-
mately the same amounts. It is interesting to note the already
solubilized zinc within the electroplating waste. The initial
amounts of zinc leached from the electroplating waste cell were
relatively high and no substantial increase above these values
has occurred.
is1-,
o—o SOL1U HASTE (CELL 1)
~l I I 1 1 1 I I 1 1 1 1 1 1
9 10 I! 12 13 i4 15 16 17 18 19 20 21 22
TIME (MONTHS)
Figure 12: Zinc
o-o LLLCTROPLAIUG WASTE (CtLL W
a-a INORGANIC Pldl'iLHI JAbIL (CLLL ID
Figure 10: Total Solids
IO'1
Figure 13: Zinc
r
7 8 9 10 II 12 13
TIME(MONTHS)
i—i—i—i—i—i—i
15 16 17 18 19 20 21 22
-------
74 Effects of Industrial Sludges
AHEFY P80WJCTIO.I UASTE (CtLL 10)
. BATTERY PRODUCTION WASTE (CELL 10)
16"
Figure 14: Zinc
9 10
TIME (MONTHS)
13 14 15 16 \7 18 19 20 21 22
The increase in lead leached, Figures 15-17, with the increased
organic acids content indicates the solubilization of the lead
within the solid waste. This is also evident within the cells con-
taining the inorganic pigment waste and the chlorine brine
sludge. Initially, the already solubilized forms of lead within the
electroplating and the battery production wastes are present
even though both wastes had been treated with caustic to form
an "insoluble" hydroxide material. However, the amount
leached from the solid waste, the inorganic pigment waste, the
electroplating waste, and the battery production waste all are
presently at the same levels. The one cell which is leaching lead at
a higher rate, is the chlorine production brine sludge which
appears to be leaching many materials in greater amounts. It is
interesting that the industrial residuals containing ten to fifty
times the amount of lead initially present within the solid waste
itself are not leaching lead at any greater rate or duration than
the solid waste alone.
10"'-
lo'-
10'-
lO'"
I 2 3 4 5 6 78 9 10 II 12 13 11 15 16 17 18 19 20 21 22
TIME (MONTHS)
Figure 15: Lead
o—o ELECTROPLATING WASTE (CELL 12)
*—° IHORGArflC PIGHtilT HASTE (CELL li)
o-o CHLORIilE PRODUCT lOil BRUE SLUDGE (CELL It)
Figure 16: Lead
n—i—i—i—i—i—I—I—I—I
9 10 II 12 13 14 15 16 17 18 19 20 21 22
IME (MONTHS)
ID-'-,
i 234 5678 9 10 n 12 13 H is 16 ir IB is 20 21 22
TIME (MONTHS)
Figure 17: Lead
The graphical displays for nickel, Figures 18-20, indicate the
same increased solubilization with time, due to the presence of
organic acids, with the exception of the electroplating waste and
the battery production waste which contained solubilized forms
of nickel. As time progresses, a majority of the concentration
histories appear to equilibrate with the exception of the chlorine
production brine sludge and the electroplating waste. Both
industrial waste cells are leaching higher amounts of nickel than
the solid waste control cell.
• SOLID HASIE (CELL 4}
10 -
"I I I I I 1—I—I—I—I—I—I—I—I—I—I—I—I—I
I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 IB 19 20
TIME(MONTHS)
Figure 18: Nickel
_
£ 10 H
LLlCIROPLAIIilG HASTt (CELL 12)
IHORGAillC PIGIlLilT WASIE (CELL li)
CHLORINE PROUUCJIIM biu.iE SLUJGE (CELL 1
io-'°H
10"
I 2 3 4 5 6 7 8 9 10 II ,2 13 I4 15 16 17 IS S 20
TIME(MONTHS)
Figure 19: Nickel
-------
Effects of Industrial Sludges 75
BATTERY PRODUCTION WASTE (CELL 10)
i.,.llii'V I'.-iUiW I I'M JA.,11 ILLLL 111)
10 n
1234567
Figure 20: Nickel
8 9 10 II 12 13 i* 15 16 17 18 19 20
TIME(MONTH^)
There was an initially high concentration of solubilized chro-
mium leached from the electroplating waste cell, this has con-
tinued to leach at a high level. The chlorine production brine
sludge and battery productionste cells are leaching high levels of
chromium. All these cells are leaching at greater rates than from
the solid waste alone. It is yet too early in this project to deter-
mine if the present rate will continue or whether it will level off,
Figures 21-23.
o—o SOLID HASTE (CELL 1)
10-'-,
2 34 56 78 9 10 II 12 13 14 15 16 IT 18 19 20 21 22
TIME (MONTHS)
Figure 2 1 : Chromium
o—o ZLcCTROPLAi hlG 'JASTc (CELL 12)
a—o INURLANit PMILill UASIL (CLLL 13)
D—o C.iLOIIhlL Plli)i)liUIUri BRIitL SLUUGL (CLLL 11)
I 2 34 56 78 9 10 II 12 13 14 15 16 17
10'
TIME (MONTHS)
ID'1
9 10 II 12
TIME (MONTHS)
Figure 23: Chromium
Figure 22: Chromium
Other leachate parameters were also monitored but on a less
frequent basis. Sustained maximum concentrations of several
parameters were determined but only those concentrations
which occur on a somewhat continual basis are shown in Table
3. It would appear that certain of the waste materials (chlorine
production brine sludge) are leaching materials at a higher rate
and duration than the solid waste only test cell. The chlorine
production brine sludge has shown thirteen (13) contaminants
exceeding those concentrations which appear from the refuse
only cell. Of those thirteen (13) which exceed the solid waste
leachate, ten (10) of those are maxim urns for all the lechates
evaluated. We, at this time, do not have sufficient data on these
parameters to draw statistically significant interpretations and
therefore it is impossible to presently assess the effects of many
of these pollutants.
Microbial Analyses
Monitoring of the bacterial activity within all test cells indi-
cates no survival of the coliform group and increased survival
for the fecal streptococci. Attempts to increase bacterial recov-
ery employing EDTA (ethylenediaminetetracetic acid) at var-
ious concentrations (0.05,0.1 and 0.15 molar EDTA) has shown
inconsistent results. The data generated so far on the bacterial
EDTA suspensions do not indicate increased bacterial recovery
employing this technique. It does not appear that metallic ion
concentrations within the industrial waste test cells is having any
effect upon microbial activity within those cells.
Gas Production
Gas production within the solid waste test cells has been
approximately 55 liters per day. Only one industrial test cell is,
at this time, being monitored for gas production and it is pro-
ducing gas at a substantially lower rate. Major problems have
been encountered in sealing the test cells against gas leakage and
we have only recently been able to collect the gas for quantitative
measurement. For these reasons, it is too early to draw any
definitive conclusions as to the effects of industrial waste upon
gas production in a municipal landfill environment.
Summary
The addition of industrial residuals to the conventional solid
waste landfill appears to have the following effects: (1) A reduc-
tion in field capacity by the addition of solid and semi-solid
industrial waste materials. (2) No difference in the total amounts
of leached organic or nutrient parameters from the solid waste
-------
76 Effects of Industrial Sludges
only or the solid waste/industrial waste mixtures has been seen.
(3) Initially high amounts of leached pollutants from the indus-
trial waste test cells indicates already solubilized forms of these
pollutants present within the waste stream. (4) No substantial
difference has been noted in the amounts of metallic ions
leached from solid waste only or the solid waste/industrial waste
mixtures.
ACKNOWLEDGEMENT
This work was accomplished under EPA contract 68-03-2120
under the direction of the Municipal Environmental Research
Laboratory (MERL), Cincinnati, Ohio.
Table HI: Sustained Maximum Concentrations of Leachate Contaminants'1
WASTE
CELL
VANADIUM'
ARSENIC1
SELENIUM1
ANTIMONY'
TIN1
SULFUR1
ASBESTOSP
CYANIDE1
ALUMINUM'
PHENOL'
BERYLLIUM1
LEAD1
TITANIUM'
MERCURY'
BORON'
NICKEL1
ZINC'
HEXAVALENT
CHROMIUM1
SUMMATION
SWb
4
<0.03
12.2
<5
0.59
150
674
<100
<1
<0.05
4.3
<0.3
1510
<0.1
19.7
18.1
1360
4.3
<2.5
aHIGHER CONCENTRATIONS MAY HAVE
bSOLID WASTE
CSEWAGE SLUDGE
dREFINERY SLUDGE
SSC
7
<0.03
10.7m
<5
(0.7)
NA"
289
(200)
<1
<0.05
[(18.0)]
<0.3
1064
<0.1
[(284)]
(19.0)
(3400)
2.9
<2.5
4 (6) [2]
BEEN DETECTED
RSd
9
<0.03
<3.0
(13)
0.19
(800)
NA
<100
<1
(1.2)
[(22.9)]
(0.9)
1450
0.1
(31.2)
2.8
1010
2J
1(19)]
6 (7) [2]
'mg/1.
BPWe
10
EWf
12
[(0.29) ]Jk <0.03
(14.9)
<5
0.54
(1200)
491
<100
(1.2)
(8.1)
NA
[(45)]
(3380)
<0.1
7.2
13.3
1319
(5.2)
<2.5
4 (8) [2]
6.3
[(230)]
(0.98)
(3000)
68
<100
<1
<0.05
NA
(30)
(1820)
<0.1
(26.9)
(22.0)
(3505)
0.5
<2.5
3 (8) [1]
IPW9
13
[(0.23)]
(21.2)
(10.0)
(0.84)
(2400)
(846)
[(6000)]
[(5.0)]
(5.4)
NA
(4.5)
(2140)
<0.1
(67.1)
(24.0)
(1490)
[(11.2)]
[(12.0)]
0 (16) [5]
CPBSh
14
<0.03
[(214)]
<5
[(1.82)]
[(8000)]
[(3362)]
(3000)
<1
1(11.2)]
NA
[(48.0)]
[(6460)]
[(0.6)]
[(328)]
[(83.0)]
[(6020)]
(8.0)
<2.5
0 (13) [11]
i( ) INDICATES EXCEEDED SOLID WASTE
k [ ] INDICATES MAXIMUM SUSTAINED CONCENTRATION OBSERVED
•BATTERY PRODUCTION WASTE
'ELECTROPLATING WASTE
w
mUNDERSCORE INDICATES LESS THAN SOLID WASTE
"NOT ANALYZED
SINORGANIC PIGMENT WASTE
PFIBERS/1
hCHLORINE PRODUCTION BRINE WASTE
-------
A Preliminary Examination of Vinyl Chloride
Emissions From Polymerization Sludges
During Handling and Land Disposal
R. A. Markle, R. B. Iden, and F. A. Sliemers
Battelle, Columbus Laboratories
Columbus, Ohio
INTRODUCTION
In January, 1974, the deaths of four workers in the vinyl chlo-
ride/poly vinylchloride (VCM/PVC) industry were attributed
to VCM exposure. Since then angiosarcoma of the liver, a rare
and fatal tumor, has resulted in the death of at least 15 workers
in U.S. PVC facilities. In addition, other forms of cancer, certain
nonmalignant liver diseases, and acroosteolysis, a unique
occupational disease, also have been found in workers within
the industry1.
Thus it was decided by the EPA that a need existed to
investigate VCM emissions from PVC waste sludges and typical
disposal sites representing a cross section of climate conditions,
disposal methods, and contiguous population densities. Conse-
quently the present study was initiated as.a preliminary, low-lev-
el effort, to determine approximate VCM concentrations in
landfill air and to perform initial measurements, in the labora-
tory and under controlled conditions, on the rates at which
VCM is released from PVC sludges.
Polyvinyl Chloride Production
PVC, commonly known as vinyl plastic, is produced from
VCM, a colorless, faintly sweet smelling gas. VCM is converted
to solid PVC by one of four different batch polymerization pro-
cesses. U.S. PVC production for 1974 was about 4.75 billion
pounds2. The processes used and the percentages of total pro-
duction they represent are listed in Table I.
Table I: PVC Processes
Process
Type
Suspension
Emulsion
Bulk
Solution
Polymerization
Medium
Water
Water
Monomer
Organic Solvent
Percent of
Total PVC
Production
78
12
6
4
Regardless of the process used, we are interested here only in
those steps of the PVC production process which result in
byproduct wastes containing suspended solid matter and VCM.
Since most PVC is produced in aqueous media these by-prod-
uct . streams consist basically of water suspensions of fine,
particulate PVC containing small amounts of various polymeri-
zation processing aids, and dissolved and/or absorbed VCM. It
is this entrapped VCM which is of concern.
The aqueous waste streams are treated in various ways at
different PVC plants. Basically the processing consists of steps
to concentrate the solids content of the waste as much as possi-
ble while discharging waste water of acceptable quality to the
local water treatment system, or natural outlets such as rivers.
The processing includes chemical treatments to coagulate and
sediment the solids and physical separation procedures such as
large, specially designed settling and concentrating tanks and
specialized centrifuging and filtration procedures. The final
waste material is a water-based sludge ranging from about 15 to
40 percent solids. Physically these sludges range from waterlike,
thin slurries to thick pastes approximating the consistency of a
concrete premix.
At the present time most, if not all, of these sludges, are dis-
carded at municipal or privately owned landfills. Typically the
sludges are transported to the landfill in pressure-controlled
tank trucks or open-bed trucks and dumped into bulldozer-
prepared pits or trenches that are 0.6 to 3 or more meters deep.
They are then covered with compacted layers of trash and soil to
a depth of 0.3 to 1 meter or more.
Sample Collection
Arrangements
At the beginning of this study three PVC plant site/landfill
combinations were selected for sampling purposes. These com-
binations were chosen to provide good cross-sections of geo-
graphical location and climate, PVC plant technology, sludge
type, and landfill practice. The protocol established for sample
collection included the following steps:
( 1) Visit to PVC plant by EPA and Battelle personnel
( 2) Tour of PVC sludge processing, isolation, and storage
facilities
( 3) Observations of PVC sludge collection by waste hauling
company
( 4) Collection of PVC sludge samples for VCM analysis
( 5) Follow sludge hauling truck to landfill with PVC
company and/or hauling company personnel
( 6) Meet landfill operating personnel and gain access to
landfill
( 7) Collect background air grab sample before PVC sludge
disposal
( 8) Observe PVC sludge disposal practice
( 9) Collect air and sludge samples during disposal
(10) Collect air samples after PVC sludge disposal and coverage
(11) Collect air samples at same landfill site approximately 1
day later.
77
-------
78 Vinyl Chloride Emissions
Air Samples
Grab air samples were collected at landfills before, during,
and after sludge disposal and coverage, downwind of the known
VCM emissions sites. Several samples were also collected out-
side the boundaries of one landfill. The samples were collected in
preevacuated (10-8 torr at 150 °C) 3.5-liter stainless steel cylin-
ders by opening the entrance port at normal breathing levels.
The date, time, weather, and wind conditions were recorded for
each sample taken.
Sludge Samples
During this study samples of the PVC sludges were collected
both at the plants and at the landfills. Sludge collections at the
landfill were done during the actual disposal operation except at
Landfill 1 where sludge was collected both after disposal and
after bulldozing and partial coverage. Sludge samples were col-
lected in tightly sealed glass containers to prevent VCM evapo-
ration, returned to the laboratory and stored at 5 °C until anal-
ysis could be performed.
Analytical Methods
The standard equipment used for VCM analysis in this study
was a gas chromatograph-flame ionization detector (GC-FID)
apparatus. Seven crosschecks were performed using a mass
spectrometer (MS), with excellent agreement found. Grab air
samples were analyzed directly, by injection of air aliquots into
the GC-FID or, in one case, the MS. Headspace and liquid
phase portions of PVC sludge samples were also analyzed by
direct injection into the GC-FID or MS. PVC sludges were rou-
tinely analyzed for VCM content by extraction with tetrahy-
drofuran (THF) and injection of an aliquot of the THF extract
into the GC-FID1. One direct analysis of a PVC sludge sample
was also performed using the MS.
VCM Concentrations
VCM concentrations are expressed in parts per million
(ppm). VCM concentrations in air are based on a volume ratio,
or microliters of VCM per liter of air. Thus 1 ppm equals 1 f-1
VCM/liter of air. However, VCM concentrations in PVC
sludges are based on a weight ratio, and micrograms of VCM
per gram of sludge. Thus 1 ppm VCM in sludge equals 1 /A g
VCM/gram of sludge.
VCM Analysis by GC-FID
The analyses were done on a Packard Series 800 gas
chromatography instrument using the following conditions:
Column
Porapak Q in an £' x 316" stainless steel tubing
Temperatures
Column 120 C, Detector 120 C, and Injector 120 C
Flows
Nitrogen Carrier 30 ml/minute; Air 300 ml/minute;
Hydrogen 30 ml/minute
Electrometer
500 volts; 1 x lO-'o amps
Detector
Hypodermic syringe septa and six way gas sampling valve
Detector
Flame Ionization.
The GC was calibrated using commercial (Matheson Gas Co.)
standards of VCM in nitrogen, supplied with a certified analy-
sis. These were reanalyzed in our laboratory by MS. One stan-
dard contained 20.5 ppm VCM in nitrogen and the other 0.45
ppm. The 20.5 ppm standard was used routinely for this work.
In addition, a mixture of saturated and unsaturated
hydrocarbons including methane, ethene, acetylene, ethane,
propane, propene, isobutane, 1-butene and n-butane was chro-
matographed. Also individual samples of dichlorodifluorome-
thane (Freon 12), isobutylene and 1,3-butadiene were chromat-
ographed separately. The total set of compounds chromato-
graphed and their retention times in comparison to VCM are
listed in Table II.
Table II: Resolution of VCM from Potential Contaminants
Compound
Methane
Ethene
Acetylene
Ethane
Propene
Propane
Freon 12
VCM
Isobutane
1-butene
Isobutene
n-hutane
1,3-butadiene
Formula
.CH
C2H4
G2H2
C2H6
C3H6
C3Hg
CCl2F2
CH2=CHC1
CH(CH3)3
CH2=CHCH2CH2
CH2=C(CH3)2
CH3CH2CH2CH3
CH2=CHCH=CH2
Ketent Lou
Time ,
minutes
0.6
1.0
1.0
1.1
2.8
3.1
3.7
5.7
7.9
8.7
8.8
9.9
18.3
Freon 12 and isobutane have the nearest retention times to
VCM of the thirteen compounds listed. However even these two
compounds show differences in retention time of 2 minutes or
longer, which is a substantial difference resulting in total
separation of the elution peaks.
VCM Analysis by MS
The MS used for the crosscheck of the VCM analysis was a
Consolidated Electrodynamics Corporation Model 21-620,
equipped with a calibrated inlet system specially designed for
gas analysis. Inlet sample pressure is measured using a microma-
nometer. Ionization conditions used were 50 volts at 40 milli-
amps. Pure VCM was used for calibration so that standardiza-
tion of the GC and MS were completely independent.
Seven MS verification analyses were performed during this
study to confirm GC analysis of VCM. These are summarized in
Table III.
Table III: Crosscheck VCM Analyses by MS
Sample
1,
2,
3,
8,
9,
K),
Sludp
SI ud}
Land I"
Slue!:
Plant
PI. -ml
Plant
ill
Si
Si
Si
Vapor 1'liase
Vapor I'hase
Air
IH-v Sol ids
roam
rrain
Liquid
Vapor
Vapor
2
2
23
8
30
MS
VCM
,200.
,300.
0.05
210.
, 000
•
2
1
28
8
37
PPin
CC-1"
,700.
, 900.
0.
200.
,000.
/)00.
,400.
ID
07
"^'-•-rr1
-------
Vinyl Chloride Emissions
79
Discussion
In the following sections data obtained on VCM
concentrations in air samples and PVC sludge vapor, liquid and
solid phases are discussed. Also the results of a very preliminary
study of VCM release rates from PVC sludges are discussed.
Finally a brief analysis of the VCM emissions potential of the
sludges is presented.
Grab Air Samples
The results of laboratory analysis of grab air samples
collected at three landfills are listed on the following page in
Table IV.
VCM concentrations ranging from 0.07 to 1.10 ppm were
found at normal breathing levels at three landfills. At Landfill 1,
the levels found were relatively low and the spread in concentra-
tions was quite small (0.07 to 0.11 ppm). At Landfill 2 concen-
trations ranging from 0.13 to 0.49 ppm were found while con-
centrations found at Landfill 3 ranged from 0.16 to 1.15 ppm.
Three important features of these data are noted. First, there
appears to be a VCM background level of about 0.1-0.3 ppm in
the air at all three landfills. Secondly, instantaneous VCM con-
centrations as high as about 1 ppm are on occasion observed,
even as long as 24 hours after the PVC sludge is buried under
compacted soil. The third observation concerns an air sample
which was collected about 5 cm from a stream of liquid sludge
discharging from a truck during landfill disposal. This air sam-
ple was, in effect, "spiked" with extra VCM. The fact that this
particular air sample showed an appreciably higher VCM anal-
ysis (1.90 ppm) than the other air samples collected at the same
landfill provides good indirect proof that the VCM peaks in the
chromatographs of landfill air are correctly identified.
Three grab air samples were also collected outside the
boundary of Landfill 2 in August 1975. The VCM data are pre-
sented in Table V together with a control sample collected at the
landfill. Collection of these air samples was not correlated with
PVC sludge disposal. The instantaneous levels found outside the
landfill ranged from 0.12 to 0.37 ppm. The 0.37 ppm sample was
collected in a residential area approximately 1.3 kilometers from
the landfill disposal area. This random sample concentration is
of the same order of magnitude as those found by the EPA
beyond the fence line at VCM and PVC plants1.
Sludge Samples
The PVC sludges were analyzed to determine VCM contents.
The vapor phase (head space) and liquid filtrate portions of the
sludge samples were analyzed first. It was determined that the.
VCM content of these phases was < 1 percent of the total sludge
VCM content with even moderately high total VCM contents
(>200 ppm, dry solids). Thus the amount of VCM in these two
phases is negligible in terms of potential landfill VCM
emissions.
VCM Contents of Sludge Solids
The PVC sludges were filtered and the sludge solids subjected
to VCM analysis as described earlier. The results obtained are
shown in Table VI.
VCM concentrations found in the sludge samples ranged
from 7 to 520 ppm in the wet filtered sludge, and from 20 to 1260
ppm on a dry solids basis. These concentrations can be com-
pared with the values found in EPA studies (1) in the spring of
1974 at six PVC plants. In that work, VCM concentrations
found ranged from< 1 ppm to 3520 ppm in wet sludge and from
< 1 ppm to 4200 ppm in dry sludge but with most of the samples
containing >10 ppm on either a wet of dry basis. Thus the VCM
concentration range found in the present work is similar to that
observed in the earlier studies, although the highest concentra-
tion level found in the present work is about 7 times lower than
the highest values found by the EPA workers in the spring of
1974.
Table V: VCM Concentrations in Grab Air Samples Taken Outside Landfill Boundaries
Collection Information
Air
Sample
Number
10
18
19
20
(a)
(b)
(c)
Temp,
C
29
27
29
22
Weather
Very cloudy
Cloudy — Hazy
Very cloudy
Cloudy
Wind velocity in kilometers per hour measured
All samples taken at times when the wind was
GC-FID
Wind
Velocity,
8-16
0-8
8-16
0-8
Sampling Location''3'
180 meters inside landfill adjacent to sludge
disposal area (8/18/75)
Residential housing area closest ( ~1.3 kilo-
meters) to the landfill. Also the site is the
lowest 'elevation point in that area (8/15/75)
Beach Club ~ 0.8 kilometers east of the
landfill and lower in elevation (8/18/75)
Public boat landing ~ 900 meters from the
landfill and lower in elevation (8/20/75)
ppm
0.49
0.37
0.19
0.12
with a portable windmeter.
blowing from the landfill toward the site (downwind).
-------
80
Vinyl Chloride Emissions
Table IV: VCM Concentrations in Grab Air Samples Taken at Landfills
Air
Sample
lumber
Controls
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
(a) Wi
(b) (la
(c) Th
Temp ,
C
14
14
14
17
17
16
16
23
23
29
22
23
23
23
21
26
27
ncl veloci.
s chroma I
is sample
Weather
Analysis
Very cloudy
Very cloudy
Very cloudy
Very cloudy,
raining
Very cloudy,
raining
Cloudy
Cloudy
Partly cloudv
Partly cloudy
Very cloudy
Partly cloudy
Partly cloudy
Partly cloudy
Partly cloudy
Partly cloudy
Partly cloudy
Partly t loudy
Collection Information
Wind
Velocity,
kph(a) Sampling Location
of laboratory air grab samples
Landfill 1
0-8 30 meters from disposal site just
before sludge dump (date 4/24/75)
0-8 At leading edge of freshly dumped
sludge (4/24/75)
0-8 30 meters from disposal site after
dumping and dozing (4/24/75)
Landfill 2
5-11 At disposal site before sludge
discharge started (6/12/75)
5-11 Edge of sludge pit as soon as
discharge is completed (6/12/75)
3-11 At previous days disposal site
before fresh discharge (6/13/75)
3-11 About 5 centimeters from sludge
discharge stream (6/13/75)
0-8 Edge of sludge pit during second
dump (6/13/75)
0-8 Edge of sludge pit during third
dump (6/13/75)
8-16 180 meters inside landfill, near
sludge disposal area (8/18/75)
Landfill 3
5-11 30 meters from disposal fjite be-
fore sludge discharge (6/24/75)
5-11 Edge of sludge pit between two
trucks, while both are dis-
charging (6/24/75)
5-11 Same as (12) near the end of the
discharge period (6/24/75)
5-11 30 meters from disposal site after
sludge pit is covered (6/24/75)
0-8 Standing over previous days covered
disposal site (6/25/75)
0-8 Same as. (15) (6/25/75)
0-8 Same as (15) (f>/25/75)
f Q J
ppm
<0.01
0.11
0.10
o.ojM
0.13
0.13
0.27
1.90
0.30
0.12
0.49
0.16
0.40
1.03
0.16
0.17
1.00
1.10
ty in kilometers per hour measured with an anemometer.
o;;r.-i|ihy with flame ionix.alion detector (CC-I-'I I)) .
was also analysed by MS as a crosscheck and 0.05 ppm were found.
-------
Vinyl Chloride Emissions
81
Table VI: VCM Concentrations Found in PVC Sludges
PVC
Sludge
No.
2(g)
300
5(J)
7(k)
Slud-ge
Weight
C'Cb)
34
35
34
36
17
30
Solids,
Percent
" F(c)
Plant 1
42
55
41
42
41
Plant 2
40
Plant 3
60
VCM
W(d)
150
210
520
90
90
7
90
l}
D(e)
360
380
1260
200
200
20
130
(a) VCM analysis of wet, filtered sludge solids by GC-FID anal-
ysis of THF extract.
(b) As collected.
(c) After filtration.
(d) Wet sludge.
(e) Dry solids.
(f) Freshly centrifuged sludge.
(g) Sludge from partly filled truck loader.
(h) Full truck loader.
(i) Sludge just unloaded at landfill.
(j) Sludge (i) after bulldoze burial.
(k) Fluid sludge collected during discharge from tank truck.
VCM Release Rate Studies
Release rate studies were performed using a cylindrical glass
chamber with 14 in. ID stainless steel inlet and outlet tubes for an
air sweep over the sludge. The tubes extended 1 cm below the
apparatus cover. A Teflon®baffle was provided between the air
entrance and exit ports to insure that the air sweep was directed
onto the bottom sludge layer. The baffle extended within about
1 cm of the thickest sludge and overlayer samples studied (2.6
cm). The air exiting from the chamber could be vented to the at-
mosphere or injected into the GC-FID at will.
Release rate experiments were performed at 25 ° C using 13 to
15 grams of sludge No. 2. This provided a layer about 1.3 cm
deep in the release rate apparatus. In release experiment 2 an
equal thickness layer of loam soil was placed over the sludge. A
30 cc/minute air flow was passed over the sludge. This gas flow
rate is equivalent to a calm day at a landfill. It is under these con-
ditions that maximum VCM concentrations might be expected
to accumulate at landfills. The VCM concentrations are instan-
taneous values obtained by injecting a 5.28 cc portion of the con-
stantly outflowing air into the GC apparatus. The release rate
data collected are given in Table VII expressed as percentages of
the total VCM content of the given sludge sample.
The results obtained indicate that a minor portion of the
VCM in PVC sludge may be quickly released at the landfill but
that a major portion of the VCM is released very slowly over a
long time period. This means that as PVC sludge is disposed,
absolute amounts of VCM will probably continue to rise for a
long, indeterminate period until a "quasi-steady-state condi-
tion" is reached. At this point continuous, slow evolution of
VCM will probably occur for a long time after sludge is no
longer disposed of at the landfill. This is consistent with the find-
ing that a VCM air concentration in the range of 0.1 to 0.3 ppm
was found at each landfill.
Table VII: Fraction of VCM Available in Sludge Samples
Released at Specified Time Intervals
Run
1
2
Ca) 1
Sludge
VCM,
Grams mg
13.6 2.86
13,3(a) 2.79
.3 cm loam soil
Percent VCM Released
Time Interval,
2 8 24
5 16 25
258
cover.
hr
110
32
11
VCM Emissions Potential
The VCM emissions potential of the PVC sludges was
calculated based on the VCM concentrations in Table VI and
data supplied by the PVC companies on the amounts of PVC
sludge being disposed of at the landfills. The results of these cal-
culations are shown in Table VIII.
Table VIII: Potential VCM Emissions from Landfills Based on
Company Supplied PVC Sludge Disposal Rates and Analytical
VCM Contents
PVC
Sludge
Number
PVC
Dry
Sludge
Solids
Disposal
Rate,
kg/day
Dry
Sludge
VCM
Content,
mg/kg
VCM,
Daily
Disposal Rate
kg liter(a)
Plant 1
4,626 0.00126 5.83 2,285
4,490 0.00020 0.90 353
Plant 2
3
4
6 2,948 (b) 0.00002 0.059 23
Plant_3
7 172(-c') 0.00013 0.022 9
(a) (kg VCM)(103)(24. 5 I/mole)/(62.5 g/mole)
(b) Company supplied figure • 35,000 gal/wk.
Dry solids based on 159,110 1/wk and
1.1 kg/1.
(c) Company supplied figure - 4,000 gal/mo.
Dry solids based on 18,184 1/wk and
1.1 kg/1.
The amount of VCM being disposed of at the landfills thus
varied between 0.022 and 5.83 kg or 9 and 2,285 liters on a per
day basis. However, it is pertinent to note that PVC production
levels were well below normal during the time this work was
being done. This resulted in less PVC sludge being produced and
disposed. When the PVC industry returns to former high pro-
duction levels the amount of VCM being disposed of in sludge at
-------
82
Vinyl Chloride Emissions
landfills will probably increase. This may be true even though
more efficient removal of VCM from PVC products has now
been accomplished to meet the new OSHA and EPA standards
pertaining to allowable levels of VCM in working place air and
escaping to the atmosphere. The effects of these changing eco-
nomic conditions and PVC technology factors can only be
ascertained through additional sampling and analysis of VCM
concentrations in PVC sludges and in air at or near landfills
where PVC sludge is disposed.
CONCLUSIONS
The following conclusions are indicated by the findings of this
study:
(1) A background air concentration of about 0.1 to 0.3-ppm
VCM appears to be present in air at landfills where PVC
sludge has been disposed of for several years.
(2) Instantaneous VCM air concentrations on the order of 1.0
ppm can occur at normal breathing heights (~1.5 meter)
above ground level at these landfills as long as 24 hours
after PVC sludge deposits are covered.
(3) Instantaneous VCM air concentrations on the order of 0.1
to 0.4 ppm can occur outside landfills where PVC sludge is
disposed.
(4) Time-weighted average sampling (15-minute, 8-hour, 24-
hour) is required to determine whether concentrations of
VCM in air that pose a health hazard occur either at the
landfills or in adjacent residential or public access areas.
REFERENCES
1. "Preliminary Assessment of the Environmental Problems
Associated with Vinyl Chloride and Polyvinychloride", Report
on the Activities and Findings of the Vinyl Chloride Task Force,
Environmental Protection Agency, Washington, D.C., Sep-
tember 1974.
2. Carpenter, B. H., "Vinyl Chloride—An Assessment of
Emissions Control Techniques and Costs", EPA-650/2-74-097,
September 1974.
ACKNOWLEDGMENTS
This work was funded by a grant from the Solid and Hazard-
ous Waste Research Division, Environmental Research Labor-
atory, Cincinnati, Ohio, Mr. Donald A. Oberacker, Project
Monitor.
-------
Land Disposal of Organic
Hazardous Wastes Containing HCB
W. J. Farmer, M. Yang, and J. Letey
Department of Soil Science and Agricultural Engineering
University of California
Riverside, California
and
W. F. Spencer
Agricultural Research Service
Riverside, California
INTRODUCTION
The problem of land disposal of hazardous industrial organic
wastes containing hexachlorobenzene (HCB) affords the unique
opportunity to examine in detail the phenomenon of vapor
phase transport in the absence of any other significant transport
process. Hexachlorobenzene (HCB) is essentially insoluble in
water so that leaching with moving water will be insignificant.
The persistence of HCB is long enough that degradation as a
means of dissipation can be considered insignificant and vapor
phase movement will be the primary means of dissipation of
hexachlorobenzene.
Hexachlorobenzene is present in industrial waste as a by-
product in the commercial production of several chlorinated
solvents, like perchloroethylene and carbontetrachloride7. HCB
is a registered fungicide used as a seed protection chemical for
seed grains. In addition significant quantities of HCB are pro-
duced as impurities or by-products in the production of certain
pesticides, like PCNP, dacthal, mirex, simazine, atrazine, and
propazine. Industry has used several methods to dispose of the
large quantities of HCB-containing waste (hex waste). These
methods include municipal landfill, land burial, lagooning, deep
well injection, incineration, and product recovery. Land dis-
posal in municipal landfills and land burial will be the topic of
this paper. Land burial differs from municipal landfill in that
land burial is a procedure used by the waste manufacturer on his
own property. Lagooning is a method whereby hex waste is tem-
porarily stored under water in an unlined reservoir before being
placed into a landfill. Hence, this project includes the effective-
ness of a water cover in decreasing volatilization.
A typical industrial operation where land disposal of hex
waste is used is shown in Figure 1. The solid phase, remaining
after the water admixture, may be either hauled directly to the
final land disposal site or left temporarily in a lagoon. When
lagoon storage is used, the cooling and crystalization step (water
admixture) and lagooning are the same; i.e., the waste stream
from the production process is fed directly below the water
surface of a lagoon, where it is stored temporarily. Periodically,
the lagoon is emptied and the hex waste carried by truck to the
land disposal site.
This study was initiated because of a specific report of HCB
contamination of beef cattle in December, 1972 in southern
Louisiana. Beef cattle to be slaughtered for human consumption
were quarantined from sale in a 200-square mile area because
high levels of HCB were found in their fat tissue. After extensive
investigations by local, state, and federal agencies and the coop-
eration of the area's HCB-producing industries, the HCB source
was traced to a municipal landfill where waste containing HCB
was disposed. Uncovered trucks were used to haul hex waste
from the industrial source to the landfill. This had caused spill-
age and contamination along the pathways followed by ,the
trucks. Waste material deposited at the landfill sites was left
uncovered. Reportedly hex waste was used as a fly-repellent
covering over municipal waste. However, no hex waste is pres-
ently being disposed of in municipal landfills in affected areas in
southern Louisiana. The uncovered waste at these landfills has
been collected into a small area of the landfill and covered with 4
to 6 feet of soil, with a 10-mil thick sheet of polyethylene film
buried about midway in the soil cover.
Waste
Production
Water Admixture
(Cooling and
Precipitation)
Figure 1. Typical hex waste handling procedures used in land
disposal
The disposal of hex waste in landfill sites in southern
Louisiana has caused a pattern of HCB contamination of resi-
dents in the area, operators of the municipal landfills, beef cat-
tle, and soil, plant and air samples1,6,11,12. The HCB content in
soil and plant samples taken from near landfill areas used for
disposal of hex waste decreased as distance from the landfill
increased6.
Burns and Miller1 reported high HCB levels in the plasma of
individuals exposed to HCB from the transporting and dispos-
ing of hex waste in southern Louisiana. A sampling of 29
households located along the route of trucks containing hex
waste showed that the average plasma residents' HCB level was
3.6 ppb, with a high of 23 ppb. The range for landfill workers
was 2 to 345 ppb plasma HCB. The average plasma HCB-level
from control group was 0.5 ppb, with a high of 1.8 ppb.
Hexachlorobenzene is a stable persistent compound of low
water solubility and moderate vapor pressure. It is a white
powder at room temperature. Its empirical formula is C^.
HCB has a melting point of 230 C, sublimes at 322 C,4 and HCB
is essentially insoluble in water. We have measured its solubility
83
-------
84
Organic Hazardous Wastes
in water as 6.2 Mg/1- Sears and Hopke7 reported a vapor
pressure of 2.10 x 10-'mm Hgat25 Cfor HCB. Wemeasuredits
vapor pressure as 1.91 x 10-J mm Hg at 25 C. HCB is soluble in
several organic solvents, like benzene and hexane, and in fats
and oil. Hence it tends to accumulate in fatty tissues.
Based on the pattern of HCB contamination of soils and
plants in Louisiana, its moderate vapor pressure, its low water
solubility, and its long-term persistence, we concluded that its
volatilization and subsequent transport by moving air currents
would be the principal mechanism by which HCB would move
about the environment. Presently, no information is available to
indicate that degradation of HCB is significant in the environ-
ment.
Therefore, we initiated research on the volatilization of HCB
from hex waste land disposal. The objective of this study was to
determine the effectiveness of various coverings—soil, water,
polyethylene film—in decreasing HCB volatilization from land.
When investigating the effectiveness of a soil cover, the influence
of soil water content, soil compaction and temperature were
included. Recent rereview articles by Spencer, Farmer, and Cli-
ath10 and Letey and Farmer' have detailed discussions of factors
affecting the volatilization of organic compounds from soils and
on vapor phase movement in soils.
Simulated Landfill
To determine the effectiveness of various coverings on HCB
volatilization from industrial waste, the simulated landfill appa-
ratus, as depicted in Figure 2, was constructed and operated in
the laboratory so that critical factors like temperature and air
flow rate could be controlled and monitored.
AIR
R.H.
SENSOR
SOIL
WASTE
>f
T
•n
WATER
FLOW
METER
HEXYLENE
GLYCOL
Figure 2. Closed air flow system for collecting volatilized HCB
from simulated landfill
The soil used for the volatilization experiments had been col-
lected from a municipal landfill in Louisiana where industrial,
hazardous waste containing HCB had been previously depos-
ited. The soil was a silty clay loam with an organic matter con-
tent of 1.4% and a field bulk density of 1.2 g/cm3. Samples of
hex waste were collected directly from two separate manufactur-
ers of chlorinated solvents. Hex waste samples A and B as col-
lected from the manufacturer contained 54.9 and 56.9% HCB,
respectively. These samples were air-dried at room temperature
before being used in the simulated landfill experiment. After air
drying, the HCB contents of samples A and B increased to 65.7
and 90.5% HCB, respectively. A third material used in the vola-
tilization experiment was recrystallized, practical grade, HCB
which was 98+% pure. We found no differences in the volatiliza-
tion of HCB from these materials in the simulated landfill. That
is, HCB volatilized from all the materials as if they were pure
HCB, except for waste not air-dried. This wet hex waste con-
tained an amount of reddish-brown organic liquid. One waste
sample collected before the water admixture step contained as
much as 76.8% of the reddish-brown liquid by volume. The sam-
ples used in the simulated landfill study were collected after
adding water to the waste stream, and they contained only a
small amount of the reddish-brown liquid. Except when stated
otherwise, for the studies reported here the hex waste samples
were air-dried to remove the organic liquid before starting the
experiments.
The volatilization cell consisted of plexiglass sections bolted
together with O-ring seals between the sections. Hex waste was
placed in the bottom section and the soil to be tested placed on
top. The cover (top section) of the volatilization cell contained a
cavity (2 mm deep) which allowed air flow over the surface of the
covered hex waste. An air flow rate of 0.7691/minute was used
in this study which provided an air speed (21.5 cm/sec or 0.48
mile/hr) across the soil surface. The apparatus utilized a closed,
air-flow system with the volatilized HCB collected in hexylene
glycol traps, like that used by Farmer et al.3 to measure the vola-
tilization of insecticides from soils. The entire landfill operation
was simulated inside a temperature-controlled cabinet main-
tained at 25 C, since volatilization processes are extremely de-
pendent on temperature, due to the temperature dependence of
vapor pressure. The hexylene glycol traps were replaced with
fresh traps at suitable intervals and the trapped HCB extracted
into hexane to be analyzed by gas liquid chromatography (GLQ
using an electron capture detector. In order to make possible the
analysis by GLC, most of the HCB samples obtained by volatili-
zation from the industrial wastes required a column cleanup on
activated neutral alumina to remove several interfering com-
pounds present in the industrial waste samples.
Research Findings
Since soil is a naturally porous body, storage of volatile mate-
rials under a soil cover will always result in some volatile loss to
the atmosphere. The extent of this loss can be controlled by con-
trolling the size or number of pores in the soil. The factors inves-
tigated for their effect on soil porosity included soil compaction,
soil water content, soil depth, temperature, source of waste, as
welt' as using as coverings polyethylene film and water. HCB
volatilization through these coverings were compared with HCB
flux from uncovered wastes. The flux from uncovered hex waste
in the simulated landfill apparatus at 25 C and 0.769 ml/minair
flow rate was 8700 ng/cm2/day.
Soil Compaction and HCB Volatilization
A limited number of choices are available in a field situation
with a landfill site on which volatilization of an organic com-
pound from a buried waste must be decreased. One choice is to
compact the soil cover over the landfill, since soil cover can be
used to decrease HCB volatilization. HCB diffusion in the vapor
phase is the major mode of movement through soil. Soil com-
paction or soil bulk density determines the porosity of a soil and
thus affects HCB vapor flux. Data in Figure 3 show that HCB
fluxes from cover soil with a bulk density of 0.96 g/cm3 (low
compaction) are greater than those from cover soil with a great-
er bulk density of 1.15 g/cm3 (high compaction). The final soil
water contents were very similar for these experiments, and the
major effect can be attributed to the effect of bulk density on air-
filled porosity. Calculation of the effect of air-filled porosity on
steady state diffusion showed that increasing the relative air-
filled porosity by 24% increases HCB flux by more than twice.
Similar exponential effects of air-filled porosity on vapor phase
diffusion flux has been shown for lindane2.
Soil Water Content and HCB Volatilization
To obtain maximum compaction of the cover soil, water must
often be added to the soil during compaction. Natural rainfall
also adds water to the soil. The amount of water in a soil affects
the air-filled porosity, or the pore space available for HCB vapor
diffusion. Figure 4 shows the effect of soil water content on HCB
flux. Clearly HCB flux from 1.8-cm cover soil with a water
content of 16.7% is greater than that from cover soil with 23.5%
-------
Organic Hazardous Wastes
85
water content. Calculation of air-filled porosity showed that
decreasing soil water content by 6.8% (dry weight basis)
increases relative air-filled porosity by 21%, but because of the
exponential effect of air-filled porosity on vapor diffusion, the
HCB flux increases almost twice. From previous considerations
of the effect of soil compaction, it is obvious that lower soil
water content will have effects similar to that of lower soil
compaction. The lower the soil water content of the cover soil,
the more rapidly will the HCB flux reach steady state.
Shearer et al.8 studied lindane diffusion in soil and observed a
similar exponential effect of soil water content on vapor phase
diffusion. Increasing soil water content will decrease HCB vola-
tilization flux. When soil is saturated with water, the effect will
be the same as a covering of water.
320
10 20 30 40 50 60 70 80
TIME (days)
Figure 3. Effect of soil bulk density on the volatilization of HCB
from industrial hazardous waste covered with 1.8 cm soil
160
140 -
TTAT T
A A
17.24% W/W A.-^X'"
19.58% (W/W)
A O Q
O .OQA n , „ O
SOIL WATER CONTENT
AND HCB VOLATILIZATION
J_ I I I I I I L
20
30 40 50
TIME (days)
60
70 80
90 100
Figure 4. Effect of soil water content on HCB volatilization
from industrial waste covered with 1 .8 cm soil at a bulk density
of 1.19g/cm3
Water as a Cover Over Hex Waste
A water cover was found to be very effective in preventing
HCB volatilization. Covering the hex waste with 1.43 cm (9/16
in) of water in the simulated landfill apparatus reduces the flux
1000 times compared to uncovered hex waste.
Polyethylene Film and HCB Volatilization
Polyethylene film has been used with soil in a municipal land-
fill as a cover for hex waste to decrease or prevent HCB volatili-
zation. Its effectiveness was tested using the simulated landfill.
Four-mil polyethylene film was placed between two 0.95 cm soil
layers. The film decreased HCB flux by approximately 40%
compared to a soil cover alone and increases the time to reach
maximum HCB flux. However, the amount of decrease is not
very large considering the cost of the film indicating the film is
not very effective as a barrier to HCB volatilization.
In a second experiment polyethylene film (6-mil) was used
alone without soil and was placed next to the waste. Polyethy-
lene film decreased HCB flux 37% as compared with uncovered
hex waste. The film did not seem to be very effective as a barrier
to HCB volatilization. This supports the conclusion obtained
with a composite soil and film cover that the addition of polyeth-
ylene film to a soil cover was not very effective in reducing HCB
volatilization compared to a soil cover alone.
Hex Waste Origin and HCB Volatilization
HCB volatilization from hex waste from two different indus-
trial sources covered with 1.8 cm soil was studied to determine if
different sources of hex waste would have significantly different
HCB volatilization rates. There was little difference in HCB
volatilization from these two wastes. Both had a steady state
flux of approximately 130ng/cm2/day. Experiments with prac-
tical grade HCB also gave the same steady state HCB flux. Thus
the HCB vapor pressure in these two hex wastes must be similar
to that in practical grade HCB.
Temperature and HCB Volatilization
Table I shows HCB saturation vapor densities from practical
grade HCB and hex waste at 15, 25, 35, and 45 C.
Table I: HCB saturation vapor densities from practical grade
HCB and hex waste at four temperatures
Temperature
C
15
25
35
45
HCB
ug/1
0.0630
0.294
0.95
3.007
Hex waste
ug/1
0.0686
0.286
0.92
3.095
The vapor density of HCB at 25C is equivalent to a vapor
pressure of 1.91 x 10-5 mm Hg. Vapor densities from practical
grade HCB and from hex waste are essentially the same.
Increasing the temperature 10 C increases the vapor density
about 3.5 times.
Vapor diffusion rate depends upon the vapor density gradient
and the diffusion coefficient. Both increase as temperature
increases. However, as temperature changes, vapor density
-------
86
Organic Hazardous Wastes
changes are considerably greater than changes in diffusion
coefficient. Since HCB moves through soil by vapor diffusion,
temperature effects on HCB volatilization flux will be about the
same, but somewhat greater than effects on HCB vapor density.
Thus, the effect of temperature on HCB volatilization flux is
exponential. Ehlers et al.2 who studied lindane diffusion in soils
found there was a similar exponential relationship between
temperature and the amount of lindane diffused.
Problems Associated with Liquid Components of Waste
Additional experiments were performed with the simulated
landfill using the hex waste which had not been air-dried and
which contained a small amount of reddish-brown liquid waste.
HCB fluxes from this wet hex waste covered with 6-mil film were
greater by about 20% than those from air-dried hex waste
covered with the same film. This liquid portion of the waste was
observed to deposit on the polyethylene film and to cause the
film to partially dissolve and expand. Thus the liquid waste may
affect the HCB transmission property of the film. Conceivably
this liquid waste may also have deleterious effects on other syn-
thetic membranes and thus decrease their effectiveness as barri-
ers to liquid and gas movement, when used in a landfill.
The liquid portions of the waste contained 1.4% HCB with a
density of about 1.67 g/ml. Because it is heavier than water, it
may move downward in a landfill with the potential of leaching
HCB into groundwater.
SUMMARY
Table II illustrates the effectiveness of various coverings in
decreasing HCB losses from industrial wastes.
Table II: Summary of the effectiveness of various coverings in
decreasing HCB volatilization losses from industrial hazardous
wastes containing HCB
None
Soil (experimental)*
Water
Polyethylene
Soil (predicted)**
Thickness
(cm)
1.8
1.43
0.015
120
Volatilization flux
(kg/ha/yr)
317
4.56
0.38
201
0.066
For a bulk density of 1.19 g/cn and 17% water content.
** Calculated assuming diffusion in the vapor phase as the
mechanism of movement.
The effectiveness of the materials are in the order of water
> soil > polyethylene film. (Polyethylene film and water are
about equally effective in decreasing HCB flux when compared
on an equal layer thickness. The cost of polyethylene film,
however, precludes its use in thick layers.) Increasing soil bulk
density and/or water content decreases HCB flux through soil.
HCB flux decreased proportionally as the thickness of all
materials increased. Thus, soil cover appears to provide an
efficient means of decreasing the loss of volatile materials placed
in a landfill. This assumes that the integrity of the soil cover is
maintained. Factors such as erosion, settling, and cracking of
the soil cover would also have to be evaluated when considering
landfill disposal of volatiles.
REFERENCES
1. Burns, J. E., and F. E. Miller. 1975. Hexachlorobenzene
contamination: Its effects in a Louisiana population. Arch.
Environ. Health 30:44^8.
2. Ehlers, Wilfried, W. J. Farmer, W. F. Spencer, and J.
Letey. 1969. Lindane diffusion in soils: II. Water content, bulk
density, and temperature effects. Soil Sci. Soc. Amer. Proc.
33:505-508.
3. Farmer, W. J., K. Igue, W. F. Spencer, and J. P. Martin.
1972. Volatility of organochlorine insecticides from soil: I.
Effect of concentration, temperature, air flow rate, and vapor
pressure. Soil Sci. Soc. Amer. Proc. 36:443-447.
4. Handbook of Chemistry and Physics. 1973. R. C. Weast,
editor. 534d ed. CRC Press, Inc., Cleveland, Ohio.
5. Letey, J., and W. J. Farmer. 1974. Movement of pesticides
in soils. In: Pesticides in Soil and Water. Guenzi, W. D. (ed.),
Madison, Wise., Amer. Society of Agronomy, pp. 67-98.
6. Louisiana Air Control Commission and Louisiana Divi-
sion of Health, Maintenance and Ambulatory Patient Services:
Summary of sampling results for hexachlorobenzene in
Geismar, Louisiana, vicinity. New Orleans, loose-leaf publica-
tion, Aug. 5, 1973.
7. Quinlivan, S., M. Ghassemi,and M. Santy. 1976. Survey of
methods used to control wastes containing hexachlorobenzene,
U. S. Environmental Protection Agency, Office of Solid Waste
Management Programs, Washington, D.C. (in press).
8. Sears, G. W., and E. R. Hopke. 1949. Vapor pressures of
naphthalene, anthrocene, and hexachlorobenzene in a low pres-
sure range. J. Amer. Chem. Soc. 71:1632-1634.
9. Shearer, R. C., J. Letey, W. J. Farmer, and A. Klute. 1973.
Lindane diffusion in soil. Soil Sci. Soc. Amer. Proc. 37:189-193.
10. Spencer, W. F., W. J. Farmer, and M. M. Cliath, 1973.
Pesticide volatilization. Chapter in Residue Reviews. Vol.
49-47.
11. U. S. Department of Agriculture News Release No.
1105-73, Washington, D.C. 1973.
12. U. S. Environmental Protection Agency, Open Public
Hearing of the Environmental Hazardous Materials Advisory
Committee Meeting chaired by E. Mrak. Aug. 6-7,1973. Wash-
ington, D.C.
-------
Sludge Farming of Refinery Wastes as Practiced at
Exxon's Bayway Refinery and Chemical Plant
Robert S. Lewis
Exxon Company, U.S.A.
Linden, New Jersey
INTRODUCTION
Sludge farming or land farming as we refer to it has been prac-
ticed on a full scale at Exxon's Bayway Refinery and Chemical
Plant Complex since mid-1973. The existing farm is 8 acres and
is utilized for the disposal of up to 3500 tons/year of oily waste
materials. The process has proven to be reliable as well as less
expensive and more environmentally sound than other available
disposal methods.
Background
The Bayway Complex, which occupies an area of 1500 acres,
is located in Central New Jersey in the town of Linden. The facil-
ity is an integrated petroleum refinery and petrochemical plant
which processes about 300 MB/ D of crude. Motor gasoline,
home heating oil, fuel oils,jet fuel, dieselfuel, asphalt,and white
oils are among the products produced by the Refinery. The
Chemical Plant produces alcohols, solvents, industrial
chemicals, and lube oil additives.
In a complex operation such as this, there are many processes
that produce solid residues, often containing oil and other
contaminates, that must be disposed of properly. Solvent
extraction, land filling and land farming are the primary final
disposal techniques used by Bayway. Land farming is the dispo-
sal of oily wastes (oil, water, and solids) by incorporation into
the soil and the subsequent biological degradation of the waste
hydrocarbons. It should be emphasized that this is an ongoing
process and is an important part of our waste disposal program.
We do not grow or plan to grow any crops on the site.
The initial experimental work to develop this technique for
use at Bayway was conducted on three small plots (1 /1000 acre
each) outside our Main Laboratory in 1972. Application rates,
degradation rates, soil analysis, and runoff contamination were
studied for two materials: waste activated sludge (from our bio-
logical water treatment plant) and tank cleanings. The results
indicated that loadings of 150 tons/acre/year of oils and solids
were possible without overloading the soil or producing any oil
runoff contamination. These results were in general agreement
with other published results and those obtained at another
Exxon Refinery where sludge farming of separator bottoms had
been practiced since 1954.
From this pilot work, the process was scaled to~ 2 acres for
testing of application methods, soil cultivation methods, and
overall operability. From there we expanded to the existing 8
acre farm in mid-1973.
Existing Land Farm Site
The existing 8 acre farm of which approximately 7 acres are
farmable is located on a completed sanitary landfill site within
the refinery. Figure 1 shows the layout of the Sludge Farm. Note
the following features of the farm:
• Topography reaches a maximum grade elevation of 21 ft. in
the center and decreases steadily to grade elevation 11 ft. on
each side.
• A network of dirt roads surrounds the site and divides it
into quadrants. These roads provide easy access for
vehicles.
• Earth dikes are located at the low ends of the site to pond
water and control runoff.
• Soil borings, which were required for registration of the site
with the State of New Jersey, show that a 2' layer of
impermeable clay exists below the site.
• Unconfined, non-potable, saline ground water exists above
the clay which interacts with the creek.
• Three (3) monitoring wells have been installed around the
site.
Figure 2 is a photograph of the northwest quadrant of the
farm and shows the farming equipment—a bulldozer and a disc
harrow.
Operating Details of Sludge Farming
Figure 3 shows the types of oily sludges that are farmed.
Cleanings from crude, slop emulsion, distillate, and additives
tanks; API separator bottoms, and other cleaning residues such
as sewer, desalter, and spill cleanings are the bulk of the sludges
disposed of on the farm. Some waste biological sludge and filter
clays have been farmed on an infrequent basis. The average
composition of the farmed sludges are ~25% oil, ~40% solids,
and the rest water.
The key operating parameters and guidelines for the land
farm are shown on the next slide (Figure 4). Soil loading and
waste characteristics are very important parameters since over-
loading the soil with oil can cause loss of biological activity as
well as runoff problems. Likewise, low pH materials could cause
leaching of metals. For these reasons the initial loads going to
the farm from each individual cleaning operation are analyzed
for oil, pH, and metals before they are applied.
Since 1974 an average of 3200 tons/yr of sludge has been de-
posited on the farm. This corresponds to a total loading of 450
tons/ acre of oil, solids, and water. Expressed in terms of oil only
this is equivalent to ~ 100 tons/acre. We believe this applica-
tion rate is moderate and could easily be increased to 150 tons/
acre/year. In fact, since the farm is only operated for 9 months
out of the year we are effectively at a loading rate of 133 tons/
acre/yr. During the winter months the farm is not operated
because the ground is either frozen or too wet. In addition bio-
logical activity is low due to lower temperatures.
87
-------
Sludge Farming
ROADWAYS
SOUTHWEST
MONITORING
Figure I: Bayway Landfarm Layout
Figure 2
-------
Sludge Farming
89
1) TANK CLEANINGS - 20-30% OIL
• CRUDE TANKS
• SLOP EMULSION BOTTOMS
• DISTILLATE TANKS
• PARAMINS TANKS
2) SEPARATOR BOTTOMS - 15-20% OIL
• REFINERY SEPARATORS
• CHEMICAL PLANT SEPARATORS
3)
OTHER CLEANINGS - 10% OIL
• SEWERS
• DESALTERS
• SPILLS
TREATMENT PLANT SLUDGE (INFREQUENT) - 5% OIL
• WASTE BIOLOGICAL SLUDGE
5)
FILTER CLAYS (EMERGENCY BASIS) -
• JET FILTERS
• WHITE OIL FILTERS
Figure 3: Materials Presently Being Land Farmed
1) SOIL LOADINGS AND WASTE CHARACTERISTICS
• OIL. METALS, pH
t PERMIT SYSTEM
2) APPLICATION AND SPREADING
• 6" MAXIMUM
• QUADRANT ROTATION
• MATERIAL STORAGE
3) HARROWING
• INDUSTRIAL DISC HARROW
t FREQUENCY DEPENDS ON LOADINGS
4) VISUAL INSPECTION
• COLOR
• CONSISTANCY
• TEMPERATURE
5) SOIL ANALYSIS
• OIL •
• pH •
6) RUNOFF AND LEACHATE CONTROL
• RUNOFF
OIL
Mutrients
METALS
- DISKING
- EVAPORATION, DIKES, PONDING
t LEACHATE - pH
- MONITORING WELLS
Figure 4: Operating Techniques, Parameters, and Guidelines
On a daily operating basis, loadings are not calculated but the
operator's experience is relied on to spread the loading. In addi-
tion, a permit system is enforced to provide records as well as
control over the dumping of wastes. Waste materials are ap-
plied to the farm by means of vacuum truck or dump truck
depending on the nature of the material. Dump trucks generally
drive onto the farm and dump while moving to spread the mate-
rial. Vacuum trucks either use hoses or spray nozzles from the
roadways to spread the material. Again close operator atten-
tion is required in order not to overload any one area.
The sludges are spread to a final depth of ~3" (6" maximum)
by a bulldozer and then disced into the ground. The quadrants
being worked are rotated so that while one is being used actively
the others are in various stages of recovery. During the winter
and peak generation periods, the oily sludges are stored in an
abandoned, concrete separator for application at a later date. In
general, the farm's capacity has been adequate to avoid double
handling of the sludges since most of the cleanings occur during
the non-winter months. Figure 5 shows the dumping and
spreading of separator bottoms on the farm.
Figure 5
M ixing of the sludges with the soil and aeration of the soil is
accomplished through frequent discing to a depth of 6-8 inches.
A construction type disc harrow is used as shown in Figure 6.
Figure 7 shows the discing operation. The required frequency of
discing is dependent on the loading being applied to the farm.
We began discing at a frequency of once per month but have
progressed to once/wk. Increased frequency results in faster oil
-------
Sludge Farming
decomposition. With moderate loadings and weekly discings a
quadrant is ready for another application in about 2 months.
Figure 6
The actual performance and health of the farm is readily
determined by visual inspection of the color and consistency of
the soil. When sludges are first applied, the soil is black and
tends to clump. As the oil decomposes the soil turns gray and
then brown as well as becoming looser or powdery. We have also
recently begun monitoring temperatures of the soil. Indications
are that during the process, the temperature of the soil rises
~10-I5°F above ambient and then drops off as the oxidation
rate decreases.
The farm soil is sampled and analyzed every two months for
the parameters shown in Figure 8. Typical soil analysis and
guidelines are discussed below:
pH: The target for pH is 7.0-7.5 in order to keep the metals
immobilized as insoluble salts. In order to maintain a high
pH, lime is applied to the farm as needed (2 times/yr).
Nutrients: Nitrates and phosphates are both nutrients for the
bacteria and are necessary for good oxidation. Our guidelines
for nitrates and phosphates are 20-30 mg/kgm. Application
of a commercial fertilizer 2-3 times per year is necessary to
maintain these levels.
Oil and Grease; After an application of sludges the oil content
of the top six inches of the soil is 8-9%. The oil content is
allowed to decrease to 2-4% before the next application of
sludge is made.
Metals: Heavy metals such as lead and zinc have accumulated
in the soil of the farm. This should cause no problems with
leaching as long as a high pH is maintained to immobilize the
metals. We understand that some researchers have found zinc
contents as high as 20,000 ppm with no adverse effects, pro-
vided crops are not grown. As mentioned before no crops are
grown. Experimental work by others has shown that there is
no downward movement of the metals through the soil.
pH
OIL & GREASE
AMMONIA
NITRATE
PHOSPHATE
CADMIUM
LEAD
ZINC
Figure 8: Typical Soil Analysis
7.0 - 7.8
2 - 9*
0.3 - 3.0 MG/KGM
2.0 - 25 MG/KGM
<1.0 - 600 MG/KGM
2.0 MG/KGM
150 MG/KGM
1000-1500 MG/KGM
Figure 7
Runoff from the farm is controlled by discing frequently,
contouring to the slope of the land. This enables the soil to
absorb the maximum amount of water. Dikes have also been
provided at the low points of the farm to pond the runoff until it
evaporates. Figure 9 shows this ponding effect. It should be
noted that sludge is not applied to these ponding areas since the
wetness inhibits oxidation.
Leaching is controlled, as mentioned before, through
maintenance of the proper pH. In addition the 3 monitoring
wells located around the site are sampled quarterly. Figure 10
shows one of our engineers sampling a monitoring well. The
-------
Sludge Farming
NITRATES
SULFATES
METALS
CHLORIDES
(2) DIFFICULT TO INTERPRET DATA
EFFECT OF FORMER LANDFILL OPERATION?
EFFECT OF MORSES CREEK?
NO BACKGROUND DATA?
EFFECT OF DRILLING WELLS?
SAMPLING TECHNIQUES?
ACCEPTABLE LEVELS?
Figure 11: Leachate Testing (Quarterly)
Figure 10
-------
92
Sludge Farming
samples are analyzed for the parameters shown on Figure 11,
namely—pH, TOC, Oil, COD, Nitrates, Sulfates, Metals, Chlo-
rides. In general the quality of the water has been similar to that
of Morses Creek and thus indicates no problem. The data have
been difficult to interpret since no background data are avail-
able and we don't know what effect the former landfill operation
has had. Initially, the actual drilling of the wells and sampling
techniques caused variability in the results but the results have
since stabilized. In addition, acceptable contaminate concentra-
tions for this type of ground water have not been established. We
have plans to run a test program to collect leachate ~ 1' below
the surface to better understand the leaching characteristics of
our operation.
COSTS
• CAPITAL COSTS
+ LAND VALUE
+ EQUIPMENT - HARROW, BULLDOZER
+ MONITORING WELLS
• OPERATING COSTS
+ REGISTRATION FEES
+ SAMPLING & ANALYSIS
+ BULLDOZER OPERATOR
+ SUPERVISION, TECHNICAL SUPPORT
+ FERTILIZER & LIME
+ AREA MAINTENANCE
COST/TON: ~$3.00/TON vs. $10-30/TON
Economic Factors
The final items to be covered are the economic factors of land-
farming. Figure 12 shows some of the cost items for the sludge
farming operation. Capital costs for our operation include the
value of the land (very low), the farming equipment (bulldozer
and disc harrow) and the installation of the monitoring wells.
Future regulations could require leachate collection facilities.
The operating costs include registration fees, sampling and anal-
ysis, the bulldozer operator, technical support, fertilizer and
lime, and area maintenance. For our operation it costs ~$3/ ton
for disposal of these sludges versus a cost of $10-30/ton for
outside landfill disposal. The time a contractor spends dumping
on the site has not been included in the $3/ton cost because it is
included in the cleaning operation. If the sludge is stored and
then farmed later the double handling raises the cost to
~$8/ton. In addition, sludge farming onsite assures us better
control over the disposal of these wastes.
SUMMARY
Landfarming at Exxon's Bayway Refinery and Chemical
Plant has been shown to be a reliable, practical disposal process
for oily wastes such as tank cleanings and separator bottoms. It
has proven to be both more economical and more environ-
mentally sound than other available disposal methods. In light
of potential environmental regulations additional research and
development work is needed; indeed it is being done by Exxon,
the API, and others across the country. We are confident (as wit-
nessed by the interest in this conference) that Sludge Farming is
a viable and sound disposal method which will gain regulatory
acceptance.
OTHER ADVANTAGES:
• CONTROL OF WASTE DISPOSAL
• LOWER CONTRACTOR CLEANING COSTS
Figure 12: Economic Factors
-------
Radioactivity Uptake by Plants
Ronald G. Menzel
USDA-ARS
Water Quality Management Laboratory
Durant, Oklahoma
INTRODUCTION
The potentially wide distribution of radioactive wastes in the
environment has led to much research on their uptake by plants.
This research deals with many, if not most, of the chemical ele-
ments. In this review, the great differences in plant uptake of dif-
ferent chemical elements will be stressed. Soil properties
affecting plant uptake of certain elements will be pointed out.
Finally, the tendency of a few plant species to concentrate spe-
cific elements will be discussed.
Much of the data were presented in an earlier review.80
Uptake data for a few more elements, including some actinides,
have been added.
Relative Concentrations in Soils and Plants
After radioactive nuclides are added to a soil, their uptake by
plants may result in plant concentrations that are much greater
or lower than their soil concentrations. Estimates of this concen-
tration factor for 49 elements have been derived from the experi-
mental data of many greenhouse and field experiments (Table
I). To compare elements I established certain rules to govern the
selection of experimental data and the calculation of concentra-
tion factors.
(1) The elements are added to the soil in water-soluble forms
and mixed with the volume of soil under consideration.
In small pot experiments, the entire volume of soil is
included. In lysimeter or field experiments, only the sur-
face 6 or 7 in is included (2,000,000 Ib/acre of mineral
soil).
(2) The amount of element added does not change the
growth of plants appreciably.
(3) The concentration is measured in plant material harv-
ested within 6 months after the element is added to the
soil.
(4) Soil concentrations are based on dry weight of soil. Only
the amount of element added to the soil is included in cal-
culating this concentration.
(5) Plant concentrations are based on total dry weight of
above-ground plant parts. Only that amount of the ele-
ment estimated to have originated from its addition to the
soil is included in calculating this concentration.
Concentration factors for each element vary over a wide range
with changing soil and plant combinations. The ranges given in
Table I are not absolute limits, but include most observations
for each element. Concentration factors for elements in adjacent
columns of the table overlap considerably. Those elements
tending to have higher concentration factors are listed first in
each column.
Data for calculating concentration factors have been drawn
from many sources, including experiments done without
isotopic tracers. In experiments made with no tracer, the con-
centration of element derived from the soil addition was deter-
mined by subtracting the concentration in plant material grown
in the same soil without addition of that element. A detailed
comparison has been made of isotopic and total-uptake
methods for determining P derived from fertilizer applica-
tions.81 As long as the condition was met for little growth
response, the methods agreed closely. The same agreement was
found for several other elements, including Ca, Mnand Zn. The
isotopic method is more precise, but within the ranges defined in
Table I, the total-uptake method is satisfactory.
Elements Strongly Concentrated
The elements that are strongly concentrated in plants are
alkali metals, halides, or essential nutrient elements. Most of the
salts of these elements are very soluble in water, and their solu-
bilities do not change much with pH of the solution. Some of the
phosphates and sulfates are not very soluble, particularly in
combination with the cations Ca, Al and Fe that are abundant in
soils. However, phosphate and sulfate are involved in important
metabolic processes in plants and may be highly concentrated in
cell nuclei and rapidly dividing plant tissues.
To illustrate the concentration factors observed with these
elements, let us consider a hypothetical soil solution transferred
into a green plant without changing the solution composition.
The soil solution would amount to about one-fifth of the dry
weight of a mineral soil. The solution in the plant amounts to
about 10 times its dry weight. The resulting concentration fac-
tor, according to the conditions stated above, is about 50. The
correspondence between this concentration factor and that
observed for the alkali metals, chloride and bromide, is not
accidental, but results from the permeability of plant tissues to
these ions.82
Elements Slightly Concentrated
Elements that are slightly more concentrated in plant tissues
than in the soil are the lighter alkaline earths, Se, Te, and certain
micronutrients. Calcium and strontium, which have been exten-
sively studied in fission product investigations, are in this group.
Increasing the supply of available Ca in the soil reduces the
uptake of Sr by plants grown on that soil. The Sr/Ca ratio in
plants is nearly proportional to the ratio of these cations in soil
solution.83 Analogous behavior has been noted for the similar
anions, selenate and sulfate.84
Elements Not Concentrated
Elements that are neither more nor less concentrated in the
93
-------
94 Radioactivity Uptake
plant than in the soil are the remaining alkaline earths and
halides, Si, Cd and some micronutrients. As compared with ele-
ments that are concentrated, these elements are either more
strongly adsorbed on soil materials or have less-soluble hydrox-
ides.
Both Ba and Ra are more strongly absorbed in soils than are
the more common alkaline earth cations.85,86 Fluoride may react
with silicates to become largely unavailable, and iodide may
react with soil organic matter.87
Table I: Relative Concentrations of Elements in the First Crop Plants Grown After the Elements Have Been Applied in Water-Soluble
Form and Mixed Into Surface Soil.
Relative concentration
10-1000 1-100
Strongly Slightly
Concentrated Concentrated
Potassium Magnesium
K(l-6)* Mg(26-29)
Rubidium Calcium
Rb(7,8) Ca(29,30)
Nitrogen tStrontium
N(9-12) Sr(8, 31-35)
Phosphorus Boron
P(13-16) 8(26,36-38)
Sulphur Selenium
5(17-20) Se(39-40)
Chlorine Tellurium
Cl(18,20-22) Te(41)
Bromine Manganese
Br(23) Mn(42)
Sodium Zinc
Na(l-3,24) Zn(31, 42-46, 66)
Lithium Molybdenum
Li(25) Mo(19, 47-49)
*Numbers in parentheses refer to
the remaining elements.
Element has radioactive nuclides
fartnr (Ppm
0.1-10
Not
Concentrated
tlodine
1(22,35,50,
tBarium
Ba(8,35)
Silicon
S1(51)
Fluorine
F(52)
Cadmium
Cd(53-55)
tRadium
Ra(56)
Cobalt
Co(57,98)
Nickel
N1(58)
Copper
Cu(59,60)
1n dry plant material*
ppm in dry soil '
0.01-1
Slightly
Excluded
tCesium
76) Cs(8, 31-33, 61)
Iron
Fe(62)
tRuthenium
Ru(32,33,35,63)
Mercury
Hg(64)
Arsenic
As(107)
Beryllium
Be(65)
Chromium
Cr(66,67)
Vanadium
V(68)
Antimony
Sb(69,70)
Tungsten
W(69-72)
literature references. No suitable
significant
in plant growth.
|
<0.1
Strongly
Excluded
tUranlum
U(79)
Yttrium
Y(31,33,35,63)
Scandium
Sc(69,70)
Tantalum
Ta<69,70)
tZirconium
Zr(32,35)
tCerium
Ce(32,33,35,63,79)
tPromethium
Pm(35,63,79)
Tin
Sn(73)
tLead
Pb(72,74,75)
tCurium
Cm(77)
tAmericlum
Am(79)
tPlutonium
Pu(35, 72,78, 79)
data were found for
-------
Radioactivity Uptake
95
Elements Slightly Excluded
The elements that are less-concentrated in plants than in soils
include the essential nutrient, Fe, and the important fission
product, Cs.
Availability of Fe for plant growth is a problem on many cal-
careous soils.88 The availability is affected by valence state and
chelation of Fe, as well as by pH. Nevertheless, it is difficult to
greatly affect the availability by adding ferrous or chelated iron,
since the form changes when these materials are added to cal-
careous soils.
Cesium is strongly fixed on clay minerals89 which accounts for
its lower concentration in plants than in soils.
Elements Strongly Excluded
The elements that are strongly excluded are very limited in
solubility. No data were found for uptake of Al, Ti and many
other elements. Likely these would also be slightly or strongly
excluded.
Effect of Soil Conditions on Uptake
Soil conditions can modify plant uptake of some elements. In
this regard, the most important soil properties are exchange
capacity, pH, and redox potential.
It is not possible to entirely separate the effect of exchange
capacity from that of competing ions on plant uptake of
radioactive elements. A soil with a high exchange capacity will
have more available Ca, K, etc., than a soil with low exchange
capacity. Soils with high exchange capacity generally sorb a
greater proportion of dissolved radioactive elements than do
those with low exchange capacity.90,91 The situation is similar
for both cations and anions, except that soil organic matter
plays a more important role in sorption of iodide and bromide.
Whether the reduced uptake is an effect of exchange capacity or
of competing ions may be immaterial. The reduction seldom
exceeds a factor of 10, which is small as compared with the dif-
ferences between elements noted in Table I.
The pH of soils has a similar effect, seldom exceeding a factor
of 10, on the uptake of certain elements. The uptake of Mn, Zn
and Mo is strongly influenced by pH.92 Low pH increases plant
uptake of elements with slightly soluble hydroxides, like Mn and
Zn. However, high pH increases plant uptake of slightly soluble
anions, like molybdate and dichromate.
The redox potential of soils may vary enough to change the
oxidation state of a few elements, and thus may markedly affect
their plant uptake. Generally, the oxidized state will show great-
er plant uptake for anion forms, and the reduced state will show
greater plant uptake for cation forms. Some examples are Mn,
Fe, Ru, and Cr.
Submerged soils have lower redox potentials and less-
extreme pH values than aerated soils.93 The uptake of radioac-
tive elements would be affected accordingly. Uptakes of Fe and
Mn, for example, would tend to be increased by reduction to
their lower valence state, but would be limited by the nearly
neutral pH. The net result of flooding is usually to increase
uptake of Fe and Mn naturally present in soils.94 However, few
experimenters have measured the effect of flooding on uptake of
soluble Fe or Mn added to soils. In one case95 it was uncertain
whether or not flooding increased uptake of added Fe and Mn
by rice.
Flooding may affect plant uptake of other metallic elements
through solubilizing Fe and Mn oxides.96 This was possibly the
mechanism by which the concentration factor for uptake of ap-
plied Zn doubled when rice was grown under flooded as com-
pared with unflooded conditions.97
Uptake of Cs and I are increased in unique ways from
submerged soils. The fixation of Cs on soil clays is much reduced
by the prevalence of the ammonium form of N in submerged
soils.61 Because of this, rice sorbed 20 times more radioactive Cs
when grown under flooded conditions. Uptake of iodide by rice
was also 10-100 times greater from flooded than from unflooded
soils, apparently because bound I was released from soil organic
matter upon flooding.76
Effects of Chelation on Uptake of Radioactivity
The uptake of radioactive elements may be considerably
increased by adding chelating agents to the soil. The uptake of
certain rare earth elements was increased by two or three orders
of magnitude31,63 which would give them concentration factors
similar to those of Sr or Ba. Less-striking changes were found
for the uptake of certain micronutrient elements, like Zn, Mg,
and Fe.31,62,65,95
Recent research shows that plant uptake of actinides is great-
ly increased by adding chelating agents to soil99,100 The
magnitude is similar to that observed with the rare earth ele-
ments, and has led to speculation that natural chelating agents in
soils may increase the uptake of actinides, which behave sim-
ilarly to the rare earths over a long period of time. This effect, if
persistent, will be extremely important in determining the per-
missible concentrations of actinide elements in soils because of
their very long life and low permissible intake for humans. The
natural distribution of U in plants and soils should include the
effects of natural chelating agents for this element. The average
concentration in dry plant material is 0.05 times that in dry
soil,106 indicating about the same concentration factor as was
observed when soluble U was added to soils.79
Uptake of Radioactivity by Different Plant Species
Various plant species are known as accumulators of certain
elements. This subject was thoroughly reviewed recently by
Peterson,101 who cited as the most extreme example the
accumulation of Se by some Astragalus species. Although the
concentration factors have not been documented according to
the criteria established for this paper, plant uptake of insoluble
and native soil Se may be 100 times greater for accumulator
species than for most crop species.102,103 Among crop plants, the
legumes are generally known as accumulators of alkaline earth
elements, and the grasses and cereals as accumulators of Si and
alkali elements. However, concentration factors for different
species do not generally differ by more than a factor of 10.
There may be significant accumulator species for the acti-
nides. Brazil nuts are known to accumulate Ba and rare
earths.104 Tea and wormwood are reported to accumulate Th
and rare earths.105
CONCLUSIONS
Plant concentrations of various radioactive elements, after
they have been added to soils in water-soluble form, may be sev-
eral orders of magnitude higher or lower than concentrations in
soil. The concentration factor depends mainly on chemical
properties of the elements, but it may vary one or two orders of
magnitude, depending on soil properties or plant species.
Important soil properties include exchange capacity, pH, redox
potential, and presence of chelating agents in the soil solution.
REFERENCES
1. T. B. Hutcheson, Jr., W. G. Woltz and S. B. McCaleb, Soil
Sci. 87, 28(1959).
2. R. L. Wehunt, M. Stelly and W. O. Collins, Soil Sci. 83,175
(1957).
3. J. J. Lehr, Soil Sci. 72, 157 (1951).
4. E. O. McClean and R. H. Simon, Soil Sci. 85, 324(1958).
5. A. L. Page, F. T. Bingham, T. J. Ganje and M. J. Garber,
Proc. Soil Sci, Soc. Am. 27, 323 (1963).
6. W. D. Laws, Soil Sci. 94, 230 (1962).
-------
96 Radioactivity Uptake
7. M. Fried, G. Hawkes and W. Z. Mackie, Proc. Soil Sci.
Soc. Am. 23, 360(1959).
8. R. G. Menzel, Soil Sci. 77, 419 (1954).
9. Ye. A. Andreyeva and G. M. Shcheglova, Pochvovedeniye
No. 12, p. 47 (1964).
10. J. O. Legg and F. E. Allison, Proc. Soil Sci. Soc. Am. 23,
131 (1959).
11. K. B. Tyler and F. E. Broadbent, Proc. Soil Sci, Soc. Am.
22,231 (1958).
12. T. W. Walker, A.F.R. Adams and H. D. Orchiston, Soil
Sci. 81,339(1956).
13. W. H. Fuller, T. C. Tucker, E. W. Carpenter and J. L.
Abbott, Proc. Soil Sci. Soc. Am. 27, 397 (1963).
14. K. Simpson, Soil Sci. 92, 1 (1961).
15. G. Stanford and L. B. Nelson, Soil Sci 68, 1957 (1949).
16. J. R. Webb, J. T. Pesek and K. Eik, Proc. Soil Sci. Soc.
Am. 222(1961).
17. C. M. McKelland A. M. Wilson, Agron, J. 55,134(1963).
18. K. F. Nielson, R. B. Carson and I. Hoffman, Soil Sci. 95,
315(1963).
19. H. M. Reisenauer, Proc. Soil Sci. Soc. Am. 27,553 (1963).
20. L. F. Seatz, A. J. Stergesand J. C. Kramer, Proc. Soil Sci.
Soc. Am. 22, 149(1958).
21. H. M. Reisenauer and W. E. Colwell, Proc. Soil Sci. Soc.
Am. 15,222(1951).
22. H. P. Newton and S. J. Toth, Soil Sci. 73, 127 (1952).
23. J. P. Martin, G. K. Helmkamp and J. O. Ervin, Proc. Soil
Sci. Soc. Am. 20, 209(1956).
24. J. Lunin, M. H. Gallatin and A. R. Batchelder, Soil Sci.
97,25(1964).
25. F. T. Bingham, A. L. Page and G. R. Bradford, Soil Sci.
98,4(1964).
26. T. C. Tucker and F. W. Smith, Proc. Soil Sci. Soc. Am. 16,
252(1952).
27. C. D. Foy and S. A. Barber, Proc. Soil Sci. Soc. Am. 22,
145(1958).
28. K. E. E. Johnson, J.F.Davis and E.J. Benne, Soil Sci. 91,
203(1961).
29. H. Al-Abbas and S. A. Barber, Soil Sci. 97, 103 (1964).
30. W. R. Hourigan, R. E. Franklin, Jr., E. O. McLean and D.
R. Bhumbla, Proc. Soil Sci. Soc. Am. 25, 491 (1961).
31. E. Essington, H. Nishita and A. Wallace, Soil Sci. 94,96
(1962).
32.1. V. Gulyakinand Ye. V. Yudintseva, Radioactive Fission
Products in Soils and Plants (in Russian). Gosatomizdat, Mos-
cow (1962).
33. E. M. Romney, J. W. Neel, H. Nishita, J. H. Olafson and
K. H. Larson, Soil Sci. 83, 369 (1957).
34. G. M. Milbourne, F. B. Ellis and R. S. Russell, J. Ncul.
Energy (Parts A and B: Reactor Sci. Technol.) 10, 116 (1959).
35. J. H. Rediske, J. F. Cline and A. A. Selders, Hanford
Atomic Products Operations, HW-36734, p. 9 (1955).
36. F. W. Schaller.'Soil Sci. 66, 335 (1948).
37. C. H. Stinson, Soil Sci. 75, 31 (1953).
38. W. L. Powers and J. V. Jordan, Soil Sci. 70, 99 (1950).
39. G. Gissel-Nielsen and B. Bisbjerg. Plant and Soil 32, 382
(1970).
40. J. W. Hamilton and O. A. Beath, Agron. J. 55,528 (1963).
41. O. A. Beath, H. F. Eppson and C. S. Gilbert, Wyo. Agric.
Exp. Sta. Bull. No. 206, p. 38 (1935).
42. E. M. Romney and S. J. Toth, Soil Sci. 77, 107 (1954).
43.E. J. Langin, R. C. Ward, R. A. Olson and H. F. Rhoades,
Proc. Soil Sci. Soc. Am. 26, 574 (1962).
44. E. Shaw, R. G. Menzel, and L. A. Dean, Soil Sci. 77, 205
(1954).
45. R. J. Speer, S. E. Allen, M. Maloney and A. Roberts, Soil
Sci. 74, 291 (1952).
46. L. C. Boawn, F. G. Viets, Jr., C. L. Crawford and J. L.
Nelson, Soil Sci. 90, 329 (1960).
47. L. E. Haley and S. W. Melsted, Proc. Soil Sci. Soc. Am.
21,316(1957).
48. H. M. Reisenauer, Soil Sci. 81, 237 (1956).
49.L.E. Ahlrichs, R. G. Hanson and J. M. MacGregor,
Agron. J. 55,484(1963).
50. J. B. Orr, F. C. Kelly and G. L. Stuart, J. Agric. Sci. 18,
159(1928).
51. C. J. Schollenberger, Soil Sci. 14, 347 (1922).
52. A. L. Prince, F. E. Bear, E. G. Brennan, I. A. Leone and R.
H. Daines, Soil Sci. 67, 269 (1949).
53. F. Haghiri, J. Environ. Qual. 3, 180 (1974).
54. M. K. John, C. V. VanLaerhoven, and H. H. Church,
Environ, Sci. Tech. 6, 1005 (1972).
55. J. E. Miller, J. J. Hassett, and D. E. Koeppe, J. Environ.
Qual. 5, 157(1970).
56. O. N. Popova, R. P. Kodaneva, and P. P. Vavilov, Fizio-
logiya Rastenii, 11, 436 (1964).
57. I. V. Gulyakin and Ye. V. Yudintseva, Izv. Timiryazev.
Sel'.-khoz. Akad. 5, 114 (1960).
58. A. T. Chang and G. D. Sherman, Hawaii Agric. Exp. Sta.
Tech. Bull. No. 19, (1953).
59. J. C. Brown, R. S. Holmes, R. E. Shapiro and A. W.
Specht, Soil Sci. 79, 363 (1955).
60. M. R. Pack, S. J. Toth and F. E. Bear, Soil Sci. 75, 433
(1953).
61. K. Tensho, K-L. Yeh, and S. Mitsui, Soil and Plant Food
6, 176(1961).
62. J. C. Brown and R. S. Holmes, Soil Sci 82, 507 (1956).
63. E. Essington, H. Nishita and A. Wallace, Soil Sci. 95, 331
(1963).
64. C. A. Bache, W. H. Gutenmann, L. E. St. John, R. D.
Sweet, H. H. Hatfield, and D. J. Lisk. J. Agr. Food Chem. 21,
607(1973).
65. E. M. Romney and J. D. Childress. Soil Sci. 100, 210
(1965).
66. J. J. Mortvedt and P. M. Giordano. J. Environ. Qual. 4,
170(1975).
67. A. Wallace, S. M. Soufi, J. W. Cha, and E. M. Romney.
Plant and Soil 44, 471 (1976).
68. R. Sbremark. J. Nutrition. 92, 183 (1967).
69. E. M. Romney and W. A. Rhoades, Proc. Soil Sci. Soc.
Am. 30, 770(1966).
70. E. M. Romney, personal communication (1976).
71. E. M. Romney and J. D. Childress, Radioecology (Edited
by V. Schultz and A. W. Klement, p. 521. Reinhold, New York
(1963).
72. D. O. Wilson and J. F. Cline. Nature. 209, 941 (1966).
(also Pb and Pu).
73. E. M. Romney, A. Wallace, and G. V. Alexander. Plant
and Soil 42, 585(1975).
74. A. J. McLean, R. L. Holstead, and B. J. Finn. Canad. J.
Soil Sci. 49, 327(1969).
75. M. K. John and C. Van Laerhoven. J, Environ. Qual 1,
169(1972).
76. K. Tensho and K-L. Yeh, Soil Sci. and Plant Nutrition 16,
30(1970).
77. W. A. Thomas and D. G. Jacobs. Soil Sci. 108,305 (1969).
78. S. L. Cummings and L. Bankert. Radiolog. Health Data
12,83(1971).
79. W. H. Adams, J. R. Buchholz, C. W. Christenson, G. L.
Johnson, and.E. B. Fowler. Studies of Pu, Am, and U in envi-
ronmental matrices. AEC Report LA-5661 (1974).
80. R. G. Menzel. Health Phys. 11, 1325 (1965).
81. L. A. Dean, Proc. Soil Sci. Soc. Am. 18, 462 (1954)
82. S. B. Hendricks, Am. Scient. 52, 306 (1964).
-------
Radioactivity Uptake
97
83. R. S. Russell, R. K. Schofieldand P. Newbould, Proceed-
ings of the 2nd International Conference on the Peaceful Uses of
Atomic Energy, Geneva, 1958, Vol. 27, p. 146. United Nations,
New York (1958).
84. A. M. Hurd-Karrer, Am. J. Bot. 25, 666 (1938).
85. L. Wiklander, Chemistry of the Soil (edited by F. E. Bear),
(2nd Ed.), p. 163. Reinhold, New York (1964).
86. R. O. Hansen, R. D. VidalandP. R. Stout, Radioisotopes
in the Biosphere (Edited by R. S. Caldecottand L. A. Snyder), p.
23. University of Minnesota Center for Continuation Study,
Minneapolis (I960).
87. M. E. Raja and K. L. Babcock, Soil Sci. 91, 1 (1961).
88. J. C. Brown, Adv. Agron. 13, 329 (1961).
89. T. Tamura and D. G. Jacobs, Health Phys. 2, 391 (1960).
90. J. R. McHenry, Proc. Soil Sci. Soc. Am. 22, 514 (1958).
91. H. Nishita, B. W. Kowalewsky, A. J. Steen, K. H. Larson.
Soil Sci. 81:317(1956).
92. J. F. Hodgson, Adv. Agron. 15, 119 (1963).
93. F. N. Ponnamperuma, Adv. Agron. 24, 29 (1972).
94. M-t. M. Nhung and F. N. Ponnamperuma. Soil Sci. 102,
29, (1966).
95. E. C. Cherian, G. M. Paulsen, and L. S. Murphy. Agron.
J. 60, 554(1968).
96. E. A. Jenne, Trace Inorganics in Water (edited by R. A.
Baker), p. 337, American Chemical Society (1968).
97. P. M. Giordano and J. J. Mortvedt. Agron. J. 64, 521
(1972).
98.N. V. Kulikov and N. A. Timofeyevz. Pochvovedeniye,
1965, no. 4, 70(1965).
99. A. Wallace. Health Phys. 22, 559 (1972).
100. E. M. Romney, H. M. Mork, and K.H. Larson. Health
Phys. 19,487(1970).
101. P. J. Peterson. Sci. Prog., Oxf. 59, 236 (1971).
102. O. A. Beath, H. J. Eppson, and C. S. Gilbert. J. Am.
Pharm. Assoc. 26, 394 (1937).
103. T. Walsh and G. A. Fleming. Intern. Soc. Soil Sci. Trans.
(Comm II and IV), 2, 178 (1952).
104. W. O. Robinson and G. Edgington. Soil Sci. 60,15 (1945).
105. C. A. 74, 1081231(1971).
106. H. J. M. Bowen. Trace Elements in Biochemistry, p. 207,
Academic Press, New York (1966).
107. E. A. Woolson, J. H. Axley, and P. C. Kearney. Proc. Soil
Sci. Soc. Am. 37, 254(1973).
-------
Factors Affecting Plant Uptake of
Heavy Metals From Land Application of Residuals
J. A. Ryan
Wastewater Research Division
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio
Pressure induced by population density has given soil the
dubious honor of being a convenient waste depository. The
waste products, when placed in soil, have four fates: (1) decom-
position of organic material, (2) leaching of soluble products, (3)
volatilization of gaseous products, and (4) accumulation of the
remaining residue as part of the soil matrix. This residuals man-
agement approach is a simple disposal process for some waste
products. For other waste products it can be considered a recyc-
ling process when some of the components of the waste are util-
ized. For example, numerous experiments have shown that
municipal sewage sludge has significant fertilizer value
(Lunt,31,32 Croker, " Braids, et al.,7 Hinesly et al.,21 King and
Morris,27 Kelling et al.,16 and Cunningham, et al.12) and the
application of this material to soil is considered to be a recycling
process. However, sewage sludge may contain, in addition to
those elements utilized by plants, relatively large amounts of ele-
ments that are not essential to plants or are essential at low con-
centration but may be phytotoxic at higher concentrations in the
soil (Le Riche,28 Berrow and Webber,4 Hinesly etal.,21 Chancy,8
and Cunningham et al.12,13,'4). This paradox confronts the
agronomist and dictates that when sewage sludge is utilized as a
fertilizer there must be consideration given to the accumulation
of residues in the soil matrix which may result in undesirable
accumulation of elements in the crop and/or impair crop
growth.
The method used to define plant uptake of elements from
waste additions to soil is the classical approach which has been
used by the agronomist to solve problems associated with plant
nutrition. Emphasis is placed on plant analysis for determining
the effects of waste addition on plant growth and quality. To
attack the problem various fractions of the plant (leaf, grain,
root, stem, and entire plant) are used for analysis. These anal-
yses are then correlated with the chemical properties of the
waste amended soil (total soil analysis, strong and weak acid
extractions, extractions with chelating agents, analysis of
exchangeable ions and fractional analysis of organic and inor-
ganic compounds) in search of the key parameters controlling
plant uptake and accumulation.
As would be anticipated, the data collected have resulted in
controversial and contradictory explanations and hypothesis.
Many of the interrelationships between elements in the soil and
in the plant will be controversial for some time. However, as the
data base is expanded, analytical capability increased, and con-
trol of environmental conditions reported, a better understand-
ing of the problem will develop and these interactions will begin
to be more clearly comprehended.
Plant uptake (entry into the above-ground portions of plants,
excluding aerial contamination) of an element is a function of
four processes and their interrelationships. These processes
include availability of the element in the soil, movement of the
element to the root, absorption of the element by the root (roots
absorb only ionic forms of elements) and translocation 'of the
element in the plant. For a given system, any one of the factors
may be rate limiting and thus control the uptake of an element.
In order to understand the system, each of these processes and
their interactions must be understood.
Metals in Soil
The chemical matrix of soil consists of numerous primary and
secondary minerals in various stages of weathering. Because of
the amorphous nature and lack of knowledge of the solid phases
in soils, we tend to discount their importance and consider
exchange reactions (cation exchange) of added constituents as
the major reaction in soils. Recent advances in soil chemistry
have demonstrated, however, that consideration must be
granted to the specific ionic species in the soil solution and the
many precipitation and dissolution reactions that are involved.
The effects of metal waste additions take on new meaning when
the stability of metal complexes, solid phase precipitates, and
metal chelates are considered.
A general overview of the reactions of metals in soil indicates
the complexity of the problem (Figure 1). Mechanisms for re-
moval of the metals from the soil include plant uptake, leaching
with soil water and volatilization of gaseous products. With time
the added compounds are broken down into soluble constitu-
ents which become part of the soil solution. The soluble com-
pounds can be exchanged for other ions on the soil exchange
complex. When the limit of ions in solution exceeds the solubil-
ity of solid phase compounds and minerals, these compounds
can precipitate. The inverse is also true that as the soil solution
becomes unsaturated to any element of the solid phase or min-
eral that is present, that solid phase or mineral can dissolve. The
ions in soil solution are subject to ingestion by soil microorga-
nisms which then incorporate the ion into soil organic matter
and/or chelation with soil organic matter.
The soil solution is affected by all the reactions that occur as
the various constituents are added or taken from it. As such, its
composition is ultimately controlled by the solubility of the var-
ious mineral phases. Since the rates of precipitation and dissolu-
tion for many of the reactions are slow, both kinetic and therm o-
dynamic factors must be considered.
From these considerations two facts become apparent: (I)
The total amount of an element in a soil is not indicative of the
environmental risk associated with further additions of that ele-
ment, (2) The cation exchange capacity (CEQ of a soil is not the
controlling factor in determining the fate of elements added to
soil. This is not to say that these are not important consid-
erations for determining the amount of an element which can be
98
-------
Land Application
99
added, but that other factors may be more important. Our
knowledge of these factors is only general and thus confidence in
the factors is lacking. This indicates the dire need for basic
research to evaluate the chemistry of elements in the soil.
EXCHANGEABLE IONS
AND SURFACE
ADSORPTION
MICROORGANISMS!
AND
CHELATES
I CROP REMOVAL]
Figure 1: Transformations of Inorganics in Soil
Factors Controlling Metal Solubility in Soil
Numerous reviews of metal reactions in soil are available
(Hodgson,22 Jenne,24, Ellis and Knezek,l8 Lindsay,29,30 Steven-
son and Ardakani,42.) From these reviews it is apparent that
many soil components (clays, organic matter, hydrous metal
oxides, carbonates, and inorganic compounds of the individual
elements) have the ability to bind metals in soil and thus reduce
the amount of soluble metal present. Superimposed are the pH
and oxidation state of the soil.
The form of metal added (sulfide, hydroxide, carbonate,
phosphate, etc.) will have a significant effect on its initial solubil-
ity in soil and therefore its initial impact on plant growth (Figure
I). Cunningham et al.,14 showed that yield was more adversely
affected by inorganic metal additions than the addition of an
equivalent amount of metals by a sewage sludge which was first
equilibrated with the inorganic metals (Table I). These results
show only the initial response to the total metal content while
the real questions are: "What is the difference in soluble metals
added by the two systems?" and "Did the addition of sewage
sludge add a sink for metals which was not present in the soil ini-
tially?" At least from Figure 1 it would seem logical to assume
that, given sufficient time (weeks, months, years), the amount of
metals in the soluble fraction would be dependent upon the
amount added and not on the initial form. This may not be true,
however, if the waste product added a sink for metals which was
not as important in the soil initially. For example, sewage sludge
adds organic material which reduces the availability of the
metals.
The form of metal added would be a function of the waste
product. In the case of anaerobically digested sewage sludge the
metals are in a reduced form whereas with some waste products
they could be in an oxidized form. Each waste and metal will
have to be evaluated independently. Chromium is a good exam-
ple of a metal where form is important. The hexavalent ion is
more phytotoxic than the trivalent ion. With time however the
hexavalent form is reduced to the trivalent form since it is more
stable, and thus after this conversion the phytotoxicity of the
two would be the same.
The total metal content added may not be significant in an ini-
tial evaluation of the impact of the added waste product.
However, it has a significant impact on the ability of a soil to
maintain a specific solution concentration. This can be illus-
trated by looking at the equilibrium between the insoluble and
soluble forms of metals in soil. This reaction can generally be
characterized by Langmuir or Freundlich isotherms which
imply that there is an inverse relationship between bonding
strength and amount of metal bound. Therefore, as the total
amount of metal increases the equilibrium soil solution concen-
tration will increase.
Table I: Dry Matter Yield as Affected by Inorganic or Sludge
Sources of Heavy Metals(Cunningham,Keeney,andRyan,197S)
Treatment Metal added*
No. CJr Cu Zn
Inorganic source
1 700 75 300
2 350 150 300
3 350 75 600
4 350 75 300
Sludge source
5 697 88 327
6 358 160 327
7 358 88 618
8 358 88 327
Ni
15
15
15
30
22
22
22
36
Corn(l)
yield
1.7
2.6
2.2
3.5
4.9
4.7
5.0
3.6
Crop
-------
100 Land Application
of its involvement in metal chemistry of soil. Production prac-
tices (crop rotation, manure crops, organic matter addition,
etc.) which maintain or add organic matter to soil should reduce
the solubility of metals.
Inorganic solid phases which contain heavy metals may be the
controlling factor in establishment of solubility limits. For ex-
ajnple, the oxides of Fe and Mn play a dominant role in
governing their solubility (Lindsay29). Solid phase Zn, Cu, Cd,
Pb, and Ni in soils have not been greatly explored. Jenne24
postulated that Zn and Cu were bound as trace constituents in
the more abundant hydrous oxides. In fact, Thomas and Swo-
boda43; Stanton and Du T. Burger41; and Gadde and Laitinen"
have shown that artificially prepared Fe oxides sorb Zn. Shu-
man38 demonstrated that soil Fe oxides sorbed Zn when he
found that the removal of the Fe oxides from clay systems
caused a decrease in their sorption capacity (Table II). However,
it should be noted that this decrease in sorption capacity was not
true for all the clay materials tested and in fact for some clay
materials there was an increase in sorption capacity when Fe
oxides were removed. This increased sorption capacity
appeared to be a result of the change in surface area and CEC of
the clay when the Fe oxides were removed. This interaction
between sorption capacity and Fe oxide content and clay con-
tent indicates that the Fe oxide content must be considered when
heavy metals are added to soil and it could be further postulated
that the other metal oxides and hydroxides in soils may be
important considerations in heavy metal chemistry in soil.
Table II: Langmuir Coefficient, Surface Area, and Cation
Exchange Capacity for Soil Clays From Two Horizons With
and Without Fe Oxides (Shuman, 1976)
Soil Horizon
Decatur A
B2t
Cecil A
B2t
Norfolk A
B2t
Leefield A
B2t
Fe Oxide
with
without
with
without
with
without
with
without
with
without
with
without
with
without
with
without
Langmuir
Coefficient
Adsorpt ive
Capacity(b)
mg/g
4.30
4.29
3.78
1.86
2.27
4.12
1.85
1.91
1.39
1.15
1.96
2.16
3.74
4 .70
».82
5.69
Surface Area
»2/g
264
287
162
204
111
71
156
517
132
124
150
504
223
220
256
232
Cation Exchange
Capacity
'1eq/100g
23.8
27.4
13.7
17.6
6.4
10.1
15.7
19.1
17.8
13.9
14.8
18.2
25.8
23.1
--
28.5
Superimposed on the aforementioned soil variables is the pH
variable. Superimposed means that the variable has a significant
effect on the other variables. It has been demonstrated numer-
ous times that soil pH has a significant effect on heavy metal
uptake (Chaney,8 and Chancy, et al.9). This pH effect could be
the result of the decrease in solubility of metals as the pH
increases. Andersson and Nilsson2 demonstrated the interaction
between soil material and pH on sorption of Cd (Figure 2).
These data support the observations that the Cd content of
crops is a function of pH in that as the pH is increased the con-
centration of Cd in the plant tissue decreases. The interesting
feature is the difference in response between soil material and
pH. For example, with Fe oxide there was no sorption of Cd
below pH 4.5 but at pH above 6 most of the soluble Cd was
sorbed. In contrast, pH had very little effect on the sorption of
Cd by the organic soil. It can be concluded that not only is pH
important but the importance of pH on sorption of Cd is a func-
tion of the soil material responsible for the sorption. Generally,
the higher the pH the lower the solubility, regardless of the type
of solid phase. This same general conclusion can be reached for
most of the heavy metals.
25 -
20 ••
1
3 15 +
u
Q 10 i
o
1/1
5 •
CONTROL
ILLITIC CLAY SOIL
Figure 2: Distribution of Cd Between Soil Material and
Equilibrium Solution as a Function of pH( Andersson and Nils-
son, 1974)
The oxidation-reduction status of the soil has a great impact
on the production of organic complexing agents, the oxidation
state of S compounds and the oxidation state of the hydrous
oxides of Fe and Mn. Therefore, it should have an effect on the
sorption of heavy metals by these compounds. The net result of
forming reduced conditions in the soil should be to increase the
solubility of Co and Cu (Mitchell33) and Mo (Jenne24); and a
decrease in the solubility of Zn and Cd (Bingham et al.6).
Movement to Root Surface
Even when the element is in solution it is not subject to uptake
by the plant root unless it first reaches the region of the roots
where the ions are absorbed. The general mechanisms to
accomplish this have been described by Barber3 and have been
classified as mass flow, diffusion and root interception. Barber3
defined mass flow as the movement of dissolved nutrients in the
water to the root. Water moves to the roots as a result of water
absorption by the roots which is induced by transpiration. If
mass flow does not supply as much nutrient to the root as is
absorbed by the root, the solution concentration at the root sur-
face is lowered and a concentration gradient between the bulk of
the soil solution and the soil solution in contact with the root is
developed. Movement of nutrients to the root surface as a result
of this gradient has been defined as diffusion.
Root interception (contact exchange) is perhaps the least
important of the three mechanisms and has been defined as the
interception of the nutrients by the root as it grows through the
soil. Olsen and Kemper35 questioned the relevance of root inter-
ception as a separate process, but there is no question that the
intensity of root growth will have a large influence on the
amounts of nutrients which can be moved to the root surface as a
result of mass flow and diffusion.
The plant stimulates nutrient movement to the root by the
growth of roots in a region (contact exchange), movement by
mass flow as a result of transpiration, and movement by diffu-
sion if its rate of nutrient absorption is greater than the rate of
movement of ions to the root by mass flow. It is evident then that
-------
Land Application
101
any factors which alter the concentration of ions in solution, the
ability of the soil to maintain the solution concentration, the
amount of root developed by the plant, the plant's ability to
accumulate ions, or the rate of transpiration of the plant can
affect the movement of ions to the root surface.
The knowledge about movement of ions to roots is at best
qualitative. The data which are available to develop a quantita-
tive concept have been developed under conditions where the
element under study was limiting plant growth. Thus the appli-
cation of this knowledge to systems where the element is in
excess is tenuous. It would seem logical to assume that when
excessive amounts of an element have been applied to the system
the rate of movement to the root would not be a rate limiting
step. However, where the element is in the upper layer (0-6") of a
soil profile and the uptake of nutrients is below this depth, it may
be that movement to the root is the rate limiting step and thus
should not be ignored. For example, it is known that plant root
development is strongly influenced by soil pH with a neutral pH
being favorable. Thus, if the pH of the lower soil profile was neu-
tral and the 0-6 inch layer was acid, root development and plant
uptake could be from the lower portions of the profile.
It can be argued that movement to the root is a function of the
concentration of ion in solution and the ability of the soil to
maintain the solution concentration and thus the chemistry of
the element in the soil is the most important consideration.
However, given the same growth medium the plant species and
variety become important because of their impact on the rooting
and growth characteristics of the plants.
Absorption of Elements by Roots
The plant is able to take up only those elements that exist
either as cations or as anions. For this reason, nutrient uptake is
commonly referred to as ion uptake. Probable form in which
they are absorbed by roots is shown in Table III. Plants
maintain nutrients in their tissue at concentrations much higher
than exist in the soil solution. For example, K in the soil solution
amounts to about 5 ppm, whereas within the plant the K content
is 1,000 to 3,000 ppm. This phenomenon is referred to as ion
accumulation.
Table III: Ionic Form of Ions Absorbed by Plants
Element
Arsenic
Cadmium
Chromium
Copper
Iron
Mercury
Nickel
Lead
Molybdenum
Selenium
Zinc
Absorbed Form
As04=
Cd++
Cr04=
Cu++
Fe++
Hg++
Ni++
Pb++
Mo04=
Se04=
Zn++
Ion accumulation occurs with all the nutrients and cannot be
explained by the "soaking up" of nutrients from the solution sur-
rounding the roots. A sponge or dead root will soak up nutrients
from a solution, but the concentration within the sponge or dead
root will never exceed the concentration in the nutrient solution.
Therefore, the phenomenon of ion accumulation is dependent
upon living tissue. Obviously any factor which affects the meta-
bolic activity of the plant will influence ion absorption. Such
factors as temperature, O2, and metabolic inhibitors affect
metabolism and thus ion uptake (Rains36).
The p H of the solution has a marked effect on both cation and
anion absorption. Cation absorption reaches its maximum rate
at pH 5 to 7, and maintains that rate of absorption up to pH 10.
Anion absorption is less affected by low pH, but as the pH is
increased above 6 the rate of absorption decreases. Changes in
the p H of the solution have been observed to occur when anion-
cation absorption is not balanced (Hiatt23). Thus, there is a
release of H+ to the solution when cation uptake exceeds anion
uptake and a release of OH- or HCO-, when the reverse occurs.
These changes in solution pH will be most drastic in the region
of the root (rhizosphere) and the chemistry of the soil solution
may be radically different in the rhizosphere as compared to the
bulk of the soil solution.
For a number of ions and tissues it has been shown that as the
concentration in solution increases the absorption by tissue
increases. This increase in absorption rate levels off at 0.1 mM
and remains essentially constant up to 1 mM. A second plateau
is reached at 10 mM. This has been interpreted to show there are
two mechanisms of absorption, one at low concentrations and
one at high concentrations.
Phytotoxicity from heavy metals has been observed to cause
Fe deficiencies which suggest that the heavy metals compete
with Fe at the absorption step. Schmid, et al.37 showed that Cu
strongly inhibited the absorption of Zn whereas Mn had no
effect. This has led to the hypothesis that there may be nutri-
tional interactions involved in the absorption of heavy metals.
Thus, not only is the concentration of the metal ion under study
important, but the concentration of other ions in solution is
important.
High phosphate concentration has been associated with elimi-
nation of metal toxicities (Spencer40). Smilde, et al.39 studied
additions of P to soil which contained phytotoxic concentra-
tions of Zn. For some crops, P alleviated Zn toxicity and
reduced Zn uptake. The involvement of P could be in soil solu-
tion chemistry, ion uptake by roots or in translocation from
roots to shoots.
Translocation of Elements Within the Plant
In the early seedling stage stored nutrients are moved from the
seed to the roots and shoot. In the vegetative stage there is move-
ment from roots to leaves, leaves to roots, and leaf to leaf.
Finally, roots and leaves of the maturing plant translocate nu-
trients to new seed.
The process of upward ion movement in xylem tissue appears
to require metabolic energy even though within the nonliving
xylem channels nutrients apparently move passively in transpi-
ration and root pressure streams. Metabolic energy is involved
in the release of nutrients into the root xylem and in their release
from the xylem into surrounding leaf tissue. Movement of ions
in the phloem tissue appears to be related to metabolism. This
requirement of metabolic energy for the transport of elements
within plants indicates that any factor which influences plant
growth and development will influence the translocation of ions
within plants.
The transport of ions can be influenced by competing metal
ions, particularly where the extraneous metals at high concen-
trations preempt ligands or displace metals normally chelated in
crucial reactions. For example, it has been shown the P can dis-
rupt the Fe nutrition of plants even though the Fe concentration
seems adequate for normal growth. It is clear that high P will
cause deranged transport and metabolism of Fe and Zn and it is
presumed that it will alter the translocation of other heavy metal
cations as well.
-------
102 Land Application
CONCLUSION
From the above discussion of factors controlling the entry of
an element into the above-ground portion of a plant it becomes
apparent that any number of variables (soil properties, waste
properties, and plant properties) can be the controlling factor
for a given experiment. However, in general it appears that
movement to the root will be controlled by solution concentra-
tion if all the ions moved to the root are absorbed by the root.
Further, from the data on ion uptake by roots from solution cul-
ture (at least for the major nutrient ions) it appears that the con-
centration in soil solution would be 1 to 2 orders of magnitude
less than the concentration in solution cultures where absorp-
tion by roots reaches the first plateau. Therefore, it becomes evi-
dent that the chemistry of metals in soil as related to the soluble
fraction of metals is a major controlling factor in absorption of
ions by the roots and thus lends validity to the relationship
between soil composition and plant uptake. However, one must
not fail to recognize the importance of the other processes
involved and their potential impact on the analysis of the data.
What Elements Should Be Considered?
Any element can be toxic to plants and / or animals if it is pres-
ent in sufficient quantities. Therefore, the obvious question is
what are sufficient quantities? The answer lies in knowing what
is in the material to be placed on the soil and what is the normal
concentration of the elements in the soil. With this information
it can be ascertained if the waste product will have a significant
impact on the concentration of the element in the soil. Table IV
gives the ranges and common concentrations of trace metals
found in soils.
Table IV: Total Concentration of Trace Elements Typically
Fo-nd in Soils and Plants (Allaway, 1968)
Element
As
B
Cd
Cr
Co
Cu
Pb
Mn
Mo
Ni
Se
V
Zn
•Toxicities
Cone . in
Common
6
10
0.06
100
8
20
10
850
2
40
0.5
100
50
listed do not
Soils fxg/g)
Ra
0.1
2
0.01
5
1
2
2
100
0.2
10
0.1
20
10
apply to
nge
-40
-100
-7
-3000
-40
-100
-200
-4000
-5
-1000
-2.0
-500
-300
certain
Cone.
Diagnostic
Normal
0.1
30
0.2
0.2
0.05
4
0.1
15
1
1
0.02
0.1
15
-5
-75
-0.8
-1.0
-0.5
-15
-10
-100
-100
-2.0
-10
-200
accumulator plant
in Plant
Tissue tVB/8)
Toxic*
>75
>20
>50
50-100
>10
>200
species.
Assuming that the waste product has a greater concentration
of one or more of the metals than is typically found in the soil,
then the movement of the element through the soil to crop to
man becomes an issue and has to be examined to determine
whether toxic or undesirable concentrations are produced.
From this point the discussion will center around one waste
product (anaerobically digested sewage sludge), but the same
general approach can be used for other waste materials.
The first question to be answered concerns the value of the
material to agriculture production. As has been shown numer-
ous times, sewage sludge has significant quantities of the major
plant nutrients (N, P. and K.) and thus land application can
benefit crop production as well as being a disposal operation.
The point at which land application stops being a utilization
process and becomes a disposal process will not be discussed. It
will be assumed that the material is used as an N or P fertilizer
and yearly application rates adjusted accordingly.
From a comparison of Tables IV and V, it becomes apparent
that for a given sewage sludge any metal can be in concentra-
tions greater than that normally found in the soil. In general, it
has been found that Cd, Cu, Pb, Zn, Cr, Hg, and Ni will be the
metals which increase the greatest from sludge application.
From Table VI it can be seen that Cu and Ni are more phyto-
toxic than Zn and Cd and that Pb, Hg and Cr are the least phyto-
toxic. Therefore, for crop production Cu, Ni, Zn, and Cd
become the most important considerations. However, if the
impact of this practice results in increased levels of Cd and Pb in
the crop then the concern shifts from effects on crops to effects
on animals that consume the crops because of the cumulative
effects of Cd and Pb on the animal system. Thus, the concentra-
tion of Cd and Pb in the crop can impose limits on the utilization
of the crop and thus on the amounts of these materials applied.
Table V: Total Concentrations of Trace Elements Typically
Found in Sewage Sludge1
Elem
Cone, in Anaerobically Digested Sewage Sludge
enl: Median Mean
Range
As
B
Cd
Cr
Co
Cu
Hg
Pb
Mn
Mo
Ni
Zn
CD
10 43
36
16
1,350 2
7
1,000 1
5 1
540 1
280
30
85
1,890 3
From L. E. Soiraners, Personal
data provided by members of
NC-118.
97
106
,070
8.8
,420
,100
,640
400
29
400
,380
6
12
3
24
3
85 -
.5
58
58
24
2
108
communication
Regional Research
230
760
3,400
28,800
18
10,100
10,600
19,730
7,100
30
3,520
27,800
based on
Committee
Table VI: Potential Toxicity of Heavy Metals
Element
Cd
Cr
Cu
Hg
Ni
Pb
Mo
Se
Zn
•When
"When
Essential!/
Plant
NO
NO
YES
NO
NO
NO
YES
NO
YES
metal is
metal is
Animal
NO
YES
YES
NO
YES
NO
YES
YES
YES
applied to soil
fed to animal
•••When animal is fed crop which
••••Cumulate effects
Plant*
MODERATE
LOW
HIGH
LOW
HIGH
LOW
LOW
LOW
MODERATE
on which crop
is grown on
Toxicity
Animal**
HIGH**-*
LOW
MODERATE
HIGH****
MODERATE
HIGH*"*
HIGH
HIGH
LOW
is grown
contaminated
Animal***
HIGH****
LOW
SLIGHT
LOW****
LOW
LOW*"*
MODERATE
MODERATE
LOW
soil.
In greenhouse experiments to evaluate source and rate of
anaerobically digested sewage sludge on crop growth, it was
found that source of anaerobically digested sewage 'sludge
application rate of anaerobically digested sewage sludge, and
their interaction all had significant effects on yield (Table VII)
-------
Land Application
103
In general, there was an increase in yield up to the 125 or 251
mT/ha rate and then a decrease in yield with higher rates of
application. Multiple regression analysis using yield as the de-
pendent variable, and the amount of N, P, K added by 0,63, and
125 mT/ha rates of the four anaerobically digested sewage
sludges as the independent variable, indicated that these varia-
bles explained 63 percent of the yield increases.
Table VII: Effect of Source and Rate of Anaerobically Digest
Sewage Sludge Application on Yield of Corn (Cunningham,
Keeney, and Ryan, 1975a)
RATE
netric tons/ha
0
63
125
251
502
LSD 0.05
SOURCE
JANESVILLE
8.0
13.4
15.2
1.3
1.0
1.4
FOND DU LAC
8.0
10.6
13.6
11.9
1.1
1.6
WISC. RAPIDS
8.0
13.4
15.7
15.9
1.3
1.6
WAUKESHA
8.0
13.2
12.2
13.1
6.6
4.9
Regression estimates to explain the decreases in yield using
yield as the dependent variable and amount of metals added by
the 125, 251 and 502 mT/ ha rates of anaerobically digested sew-
age sludge as the independent variable, suggested that Cu, Zn,
and Ni were important but there were significant interactions
between these metals and other metals added which complicated
the interpretation.
In further experiments to evaluate the interaction between
metals, a 34 factorial experiment using four metals (Cr, Cu, Zn,
Ni) at three rates were equilibrated with anaerobically digested
sewage sludge. The anaerobically digested sewage sludge was
added to soil at a rate of 63 mT/ha and the metal loading was
comparable to the previous experiment.
The findings of this experiment are presented in detail in
Cunningham et al.13 All treatments had adverse effects on plant
yields with phytotoxicity being due to Cu and Zn. Rates of Ni
addition were relatively low and Ni had no adverse effect, while
Cr increased yield (decreased the phytotoxicity of Cu and Zn).
Statistical analysis of the tissue concentration of metals
showed that a number of interactions, including increases in the
Zn and Cd concentration of plants tissue, occurred when more
Cu was added (Table VIII). This resulted in changes in the
Cd:Zn ratio of the plant tissue (Table IX). Since Cd was added at
a constant rate (4.5 kg/ha) from the anaerobically digested sew-
age sludge, the increase in Cd:Zn observed with increasing Cu
indicates that complex interactions occurred.
Table VIII: Cadmium and Zn in Plant Tissue at Varying
Levels of Cu and Zn (Cunningham, Ryan and Keeney, 1975b)
Zn Added
ppm
410
707
*First
Tissue Zn Tissue Cd
120 Cu 194 Cu 343 Cu 120 Cu 194 Cu 349 Cu
ppm of plant tissue*
343 356 478 3.6 6.2 9.7
520 422 744 5.2 5.3 8.4
corn crop
Table IX: Cd:Zn Ratio in Plant Tissue as Affected by Cu Addi-
tions (Cunningham, Ryan, and Keeney, 1975b)
Metal Added,
ppm of Soil
Cu
120
194
343
Corn
0.88
1.10
1.47
Crop
Rye
0.57
0.66
0.97
Corn
0.57
0.81
1.31
The effect of soil pH on crop yield and ion uptake cannot be
ignored. In a study of sites where sludge had been applied over
long periods of time. Chaney et al.'° demonstrated the impor-
tance of soil pH (Table X). It should also be noted that plant spe-
cies and plant part has a significant effect on the concentration
of Cd in the tissue. From similar studies it has been found that
even though there was considerable increase in the Cd content of
the soil there was minimal increase in Cd content of corn grain
(Table XI). It has also been found that the metals of concern
(Cu, Ni, Zn, Cd, and Pb) have remained in the upper horizons of
the profile where tillage has occurred.
Table X: Cd Content of Crop Tissues as Influenced by Soil
Cd Content and pH (Chaney, Hornick and Simon, 1976),
Treatment
Control
Sludged
Soil Cd
DTPH
Extraction
ppm in Soil
.13
.10
1.13
1.19
Soil
pH
5.3
6.7
4.8
6.6
Chard
Leaf
— - ppm
3.6
1.2
73.0
5.5
Soybean
Leaf
in Tissue
1.04
0.55
10.7
1.87
Grain
0.36
0.28
3.70
1.51
Table XI: Cd Content of Soil in Relation to Cd Content of
Plant Tissue (Ryan Old Site Studies)
Soil Cd
Soil
PH
Total
.IN HC1
Ex tractable
Plant Cd
Chard
Corn
Leaf Leaf
Grain
6
7
6
6
6
6
6
.4
.0
.8
.8
.5
.5
.5
35
7
9
6
4
4
56
55
14
32
33
40
32
7
9
5
4
4
32
20
82
31
30
30
--
--
2
2
1.51 3
.50 1
--
--
.06
.52
.15
.96
.20
.38
.22
.36
.10
--
--
Field experiments involving use of anaerobically digested
sewage sludge for production of corn have led to conflicting
results (Table XII). Kellingetal.26 found that applications of 4.3
kg Cd/ha did not cause a change in the Cd content of the corn
grain. Hinesly et al.21 found that applications of 88 kg Cd/ha
caused an order of magnitude increase in the Cd content of corn
grain (0.06 to 0.59 ppm). In contrast Decker et al.15 have
reported that applications of 2 kg Cd/ ha have caused a fivefold
increase in Cd content of corn (0.12-0.68). Simlar findings have
been reported by Giordano et al.20
The resolution of these results is extremely difficult and indi-
cates our lack of complete knowledge about the systems. The
difference could be a result of:
Varietal differences—The variety of corn used by Decker et
al.15 may translocate more Cd to the grain than the varieties used
by Kelling et al.26 and Hinesly.21 The differences between sweet
corn and field corn are unknown (i.e., data of Giordano et al.20
vs. others).
Forms of metals—The sewage sludges used by Decker et al.ls
and Giordano et al.20 may have contained Cd in a much more
available form than the sewage sludges used by Kelling et al.26
and Hinesly.21
-------
104 Land Application
Table XII: Field Studies on Zn and Cd Content of Corn
Sludge
Rate
MT/ha
0
3.7
7.5
15
30
60
0
92,9
184.3
386.6
0
50
100
200
400
0
55
110
220
Soil
pH
6.0
4.9
5.3
5.3
5.6
5.2
5.4
5.3
5.3
Soil
Kg/ha
11.
22.
45
90
180
510
1011
2122
90
180
360
720
64
129
258
Zn
Grain
ppm
20.1
2 20.3
5 20.6
22.0
26.3
26.9
28.9
38.4
47.6
60.0
35
47
48
52
61
22
34
49
60
Cd
Soil
Kg/ha
0.26
0.53
1.07
2.15
4.30
21
41
88
2.5
5
10
20
.5
1.0
2.0
Grain
ppm
0.09
0.08
0.10
0.09
0.09
0.10
0.06
0.17
0.33
0.59
0.3
0.7
0.8
1.0
1.0
0.12
0.41
0.59
0.68
Data from Kelling et al. (1976), Hinesley (personal
communication), Giordano, Mortvedt and Mays (1975)
and Decker, Chaney and Wohf (1975)
Soils—There could be completely different complexing
agents involved in the chemistry of Cd in the soils. It would seem
that the pH difference could explain part of the results. This
could be a direct influence on the soil chemistry of Cd or it could
be an indirect effect (i.e., root distribution of the plant). There
may also be plowpans or fragipans in the soil which influenced
root distribution and thus the zone in the soil where ion uptake
occurred.
Results such as these are to be expected until we have suffi-
cient knowledge to completely define the variables involved.
This does not imply that the practice is bad and should not be
allowed to continue, but that there can be problems and one
must be aware of the potential problems and concerned about
their impact on the practice.
Limitations
As has been demonstrated, our knowledge about the chemis-
try of metals in soils and its impact on crops is not complete, but
we do know enough to make some rational judgments. For ex-
ample, we know the crop (species as well as variety) have signifi-
cant differences in their tolerance to metals and thus we should
grow crops which are least affected by metals. If the crop is to be
a general agriculture crop, then corn grain appears to be the logi-
cal choice. Grass may be the choice where one is not concerned
about a marketable crop. Because of their ability to accumulate
metals, the growth of vegetable crops, particularly leafy vegeta-
bles, should be discouraged. However, if the material is to be ap-
plied to farmer-owned land then consideration should be given
to the types of crops normally grown in the area. If the farmer
grew a crop which accumulated the metals then there should be
an effort to reduce the metal content of the waste product being
applied to the land.
The soil pH is perhaps the most important soil variable and
should be maintained at pH^6.5. The impact of pH difference
with depth in the profile is unknown but it would seem logical to
assume that a soil with a neutral profile would be a better choice
than a soil which had an acid profile.
The soil properties which are involved in the removal of
metals from solution are diverse and the impact of each is not
conclusive. Therefore, it would seem logical to use soil CEC as a
method of grading soils on their affinity for metals until more is
known about the chemistry of metals in soils. The land grant
universities and the USDA-ARS have proposed interim limits
on metal application to agriculture soils (Table XIII). These lim-
its seem appropriate and should give sufficient lead time for de-
velopment of the basic information required to develop a more
rational basis for land application.
Table XIII: Maximum Accumulative Amount of Metal Which
Can Be Added to Privately-Owned Farmland
Zn
Cu
Ni
Cd
Pb
0-5
250
125
50
5
500
Soil CEC
5-15
k a/ha
500
2SO
100
10
1000
> IS
1000
500
200
20
2000
REFERENCES
1. Allaway, W. H., 1968. Agronomic control over
environmental cycling of trace elements, Advan. Agron.
20:235-274.
2. Andersson, A., and K. O. Nilsson, 1974. Influence of lime
and soil pH on Cd availability to plants. Ambio 3:198-200.
3. Barber, S. A., 1974. Influence of plant root on ion move-
ment in soil. p. 525-564. In E. W. Carson (ed), The plant root
and its environment. Univ. Press of Virginia, Charlottesville,
Va.
4. Berrow, M. L.,and J. Webber, 1972. Trace elements in sew-
age sludges, J. Sci. Food Agr. 23:93-100.
5. Bingham, F. T., A. L. Page, and J. R. Sims, 1964. Retention
of Cu and Zn by H-Montmorilonite. Soil Sci. Sco. Amer. Proc.
23:351-354.
6. Bingham, R.T., A. L. Page, R. Mahler, and T. Ganje, 1975.
The effects of soil Cd on flooded- and nonflooded-rice, 1975
Agronomy Abstracts p. 21.
7. Braids, O. C, M. S. Ardakani and J. A. E. Molina, 1970.
Liquid digested sewage sludge gives field crops necessary nutri-
ents, 111. Research 12(3):6-7.
8. Chaney, R. L., 1973. Crop and food chain effects of toxic
elements in sludges and effluents. In Recycling municipal
sludges and effluents on land. National Assoc. of State Universi-
ties and Land-Grant Colleges, Washington, D.C., pp. 129-141.
9. Chaney, R. L., M. C. White, and P. W. Simon, 1975. Plant
uptake of heavy metals from sewage sludge applied to land, pp.
169-178. In Proc. 2nd Natl. Conf. Municipal Sludge Manage-
ment, Information Transfer, Inc.
10. Chaney, R. L., S. B. Hornick, and P. W. Simon, 1976.
Heavy metal relationships during land utilization of sewage
sludge in the northeast. In Proc. 8th Annual Waste Manage-
ment Conf., Cornell University, Rochester, N.Y.
-------
Land Application
105
11. Croker, E. G., 1966. The value of liquid digested sewage
sludge. III. The results of an experiment on barley. J. Agr. Sci.,
Camb. 67: 105-107.
12. Cunningham, J. D., D. R. Keeney,and J. A. Ryan, 1975a.
Yield and metal composition of corn and rye grown on sewage
sludge-amended soil. JEQ 4:448-454.
13. Cunningham, J. D., J. A. Ryan, and D. R. Keeney, 1975b.
Phytotoxicity in and metal uptake from soil treated with metal-
amended sewage sludge. JEQ 4:455-460.
14. 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 sludge. JEQ 4:460^62.
15. Decker, A. M., R. L. Chancy, and D. C. Wolf, 1975.
Effects of sewage sludge and fertilizer application on yield and
chemical composition of corn and soybeans. In Crops and Soil
Res. Vol. 9, University of Maryland College Park, Md.
16. DeMumbrum, L. E., and M. L. Jackson, 1956. Infrared
absorption evidence on exchange reaction mechanism of copper
and zinc with layer silicate clays and peat. Soil Sci. Soc. Amer.
Proc. 20:334-337.
17. Elgabaly, M. M., and H. Jenny, 1943. Cation and anion
interchange with zinc montmorilonite clays, J. Phys. Chem.
47:339-408.
18. Ellis, B. G., and B. D. Knezek, 1972. Adsorption reactions
of micronutrients in soils, pp. 59-78. In J. J. Mortvedt, P. M.
Giordano, and W. L. Lindsay (ed.), Micronutrients in agricul-
ture. Soil Sci. Soc. Amer., Madison, Wise.
19. Gadde, R. R., and H. A. Laitinen, 1974. Studies of heavy
metal adsorption by hydrous iron and manganese oxides. Anal.
Chem. 46:2022-2026.
20. Giordano, P. M., J. J. Mortvedt, and D. A. Mays, 1975.
Effect of wastes on crop yields and uptake of heavy metals. JEQ
4:394-399.
21. Hinesly,T. D., R. L. Jones and E. L. Ziegler, 1972. Effects
of corn by application of heated anaerobically digested sludge.
Compost Sci. 13:26-30.
22. Hodgson, J. F., 1963. Chemistry of the micronutrient ele-
ments in soil. Adv. Agron. 15:119-159.
23. Hiatt, A. J., 1967. Relationship of cell-sap pH to organic
acid change during ion uptake. Plant Physiol. 42:291-298.
24. Jenne, E. A., 1968. Controls on Mn, Fe, Co, Ni, Cu, and
Zn concentrations in soils and water: the significant role of
hydrous Mn and Fe oxides. Adv. Chem. 73:337-387.
25. Kelling, K. A., A. E. Peterson, J. A. Ryan, L. M. Walsh,
and D. R. Keeney, 1973. Crop response to liquid digested sew-
age sludge. Proc. of the International Conference on Land for
Waste Management, Ottawa, Canada, pp. 243-252.
26. Kelling, K. A., D. R. Keeney, L. M. Walsh, and J. A. Ryan
(1976). A field study of the agriculture use of sewage sludge: III.
Effect on uptake and extractability of sludge-borne metals. Sub-
mitted to JEQ.
27. King, L. D., and H. D. Morris, 1972. Land disposal of
liquid sewage sludge. I. The effect on yield in vivo digestibility
and chemical composition of coastal bermuda grass (Cynodon
dactylon L. Pers). JEQ 1:325-329.
28. LeRiche, H. H., 1968. Metal contamination of soil in the
Woburn Market garden experiment resulting from the applica-
tion of sewage sludge. J. Agr. Sci.,-Camb. 71:205-208.
29. Lindsay, W. L., 1972. Inorganic phase equilibria of micro-
nutrients in soils, pp. 41-57. In J. J. Mortvedt, P. M. Giordano
and W. L. Lindsay (ed.), Micronutrients in agriculture. Soil Sci.
Soc. Amer., Madison, Wise.
30. Lindsay, W. L., 1973. Inorganic reactions of sewage
wastes with soils, pp. 91-96. In Proc. Joint Conf. on Recycling
Municipal Sludges and Effluents on Land.
31. Lunt, H. A., 1953. The case for sludge as a soil improver.
Water and Sewage Works 100:295-301.
32. Lunt, H. A., 1959. Digested sewage sludge for soil
improvement, Conn. (New Haven) Agr. Exp. Sta. Bull. 622.
33. Mitchel, R. L., 1964. Trace elements in soils, pp. 320-368.
In F. E. Bear (ed.), Chemistry of the Soil. 2nd ed., Van
Nostrand-Reinhold, Princeton, New Jersey.
34. Mortensen, J. L., 1963. Complexing of metals by soil
organic matter. Soil Sci. Soc. Amer. Proc. 27:179-186.
35. Olsen, S. R., and W. D. Kemper, 1968. Movement of nu-
trients to plant roots. Advance Agron. 20:91-151.
36. Rains, D. W., 1969. Sodium and potassium absorption by
bean stem tissue. Plant Physiol. 44:547-554.
37. Schmid, W. E., H. P. Haag, and E. Epstein, 1965. Adsorp-
tion of zinc by excised barley roots, Physiol. Plant 18:860-869.
38. Schuman, L. M., 1976. Zinc adsorption isotherms for soil
clays with and without iron oxides removed. Soil Sci. Soc.
Amer. Jour. 40:349-352.
39. Smilde, K. W., P. Koukoulakis, and B. Van Luit, 1974.
Crop response to phosphate and lime on sandy soils high in zinc.
Plant Soil 41:445-457.
40. Spencer, W. F., 1966. Effect of copper on yield and uptake
of phosphorus and iron by citrus seedlings grown at various
phosphorus levels. Soil Sci. 102:296-299.
41. Stanton, D. A., and R. Du T. Burger, 1967. Studies on
zinc in selected Orange Free State Soils: V. Mechanisms for the
reaction of zinc and iron with aluminum oxides. Agrochemo-
physica 2:65-75.
42. Stevenson, F. J., and M. S. Ardakani, 1972. Organic mat-
ter reactions involving micronutrients in soils, pp. 79-114. In J.
J. Mortvedt, P. M. Giordano and W. L. Lindsay (ed.), Micro-
nutrients in Agriculture. Soil Sci. Soc. Amer., Madison, Wise.
43.Thomas, G. W., and A. R. Swoboda, 1963. Cation
exchange in kaolinite-iron oxide systems. Clays and Clay Miner
11:321-326.
-------
Heavy Metal Uptake and Control Strategies
Associated with Sewage Sludge Fertilized Crops
Cecil Lue-Hing, J. R. Peterson
and
D. R. Zenz
The Metropolitan Sanitary District
of Greater Chicago
Chicago, Illinois
and
T. D. Hinesly
University of Illinois
Champaign-Urbana, Illinois
INTRODUCTION
The use of human and animal waste products for plant nutri-
tion is a very old practice. However, during the last decade there
was been a renewed interest in the use of municipal sewage
sludge for its nutrient value. Concurrent with this interest in the
nutrient value of municipal sludges is a growing interest in their
metals content—particularly, for those sludges produced in
highly industrialized areas. Quite frequently the drawback cited
with using sewage sludges from industrialized areas is that they
might be high in non-plant essential heavy metals. It is true that
many metals from industry are discharged to municipal sewer
systems; and as the wastewater treatment plants improve their
treatment of those wastes, the more metals can be expected to
accumulate in the resulting sewage sludges. It is true also that
metals are present in varying amounts in all sewages, and some-
times in relatively high concentrations in situations where indus-
trial discharges are not the prime source, as was demonstrated
bv Klein13, for New York City.
A review of the more recent literature leads to the conclusion
that the current strong interest in the metals content of munici-
pal sewage sludges is generated in great part by the belief of some
workers that the applications of sewage sludges for productive
agricultural purposes, is undesirable. By productive agriculture
here is meant the production of crops such as corn, soybeans,
wheat and barley, and fodder for cattle and swine production.
While the proposals for controlling sludge-borne metals may
vary in their specifics, they usually fall into two general catego-
ries. One category aims at controlling the discharge of metal-
bearing industrial wastes to municipal sewer systems, which in
turn would reduce the metals content of the resulting sludges.
The other category aims at "protecting the land from additional
heavy metals" by restricting the application of sludges to only
the lowest metal-content sludges available from small residen-
tial communities which have no industrial activities.
This latter approach is at best extreme and could be inter-
preted to prohibit the use of rock phosphate which has been
shown to range as high as 100 Mg CD/g (Williams and David,
1976). In a study of 21 commercial fertilizers used in Sweden the
Cd concentration was found to range from 0.1 to 225)ug/ g and
the Pb from<0.1 to 225/ig/g.2' These authors concluded that
"for Swedish conditions and with application of present Swed-
ish recommendations, commercial fertilizers will probably in
the long run have an appreciably greater influence on the
cadmium contents of agricultural products and food than will
sewage sludge."
To further illustrate this point, the work of Furr et al.8 is also
germane. They analyzed the sludges of 16 American cities for 68
elements, dieldrin, and polychlorinated biphenyls (PCB's), and
selected the sludge of Cayuga Heights, N.Y., as an example of a
domestic sludge. Of the 16 sludges reported this domestic sludge
was in the upper 50 percentile in concentrations of Hg, Pr, Pt,
Re, Y and Ca.
There is little doubt that industrial control is feasible and
effective. This topic will be discussed later. However, the sec-
ond type of control defeats the very concept of sludge utilization
for productive purposes since the bulk of municipal sludges is
produced in the nation's large industrial urban areas. Studies by
Zenz et al.23 and Konrad et al.M show that approximately 85% of
the nation's municipal sludges is produced in the large urban
industrial centers.
After many years of successful use of sludge in England, Paris,
Berlin, the United States, and other places, the question of phy-
totoxicity has been satisfactorily addressed. Unless the soil itself
simultaneously has very high metal levels and low pH, phyto-
toxicity is not a significant cause for concern.
Control Strategies
The United States Environmental Protection Agency (EPA)
during the last three years has released various drafts of a policy
statement which proposed guidelines for the utilization and dis-
posal of sludges for publicly-owned treatment works. The drafts
released before the most recent version, which appeared in the
June 3, 1976 Federal Register (EPA, 1976) contained limita-
tions on the levels of heavy metals in sludges applied to land,
and also presented an equation which limited the total
application of sludge based upon its Zn, Cu, and Ni content, or
based upon its Zn equivalent toxicity with respect to the com-
bined concentrations of Zn, Cu, and Ni. This concept of Zn
equivalent toxicity was introduced by Chumbley5 in Great Brit-
ain, picked up and recommended by Chancy4 to the U.S.
scientific community in 1973. This concept has since been
attributed to the United States Department of Agriculture
(USDA). As of this writing, neither Chancy nor Chumbley nor
the USDA has presented any acceptable scientific data to
support this concept. In fact, it has recently been concluded that
the Zn equivalent equation was based upon inadequate informa-
tion and that data currently available show that the toxicity of
these elements is not additive, and that the use of the equation
greatly underestimates the amounts of sludge-borne metals
which can be safely applied to near neutral, neutral and
calcareous soils. Furthermore, the equation does not apply
uniformly over a broad spectrum of plant species (CAST 1976).
106
-------
Heavy Metal Uptake/ Control Strategies
107
The early response of the interested scientific community to
Chaney's 1973 recommendations was very disappointing. For
example, only a very few individuals thought it necessary to
question the scientific basis for the recommendations or to
inquire into the availability or existence of data in support of
these concepts. Instead, the interested scientific community was
asked to prove the recommendations and concepts wrong while
they were rapidly being incorporated into guideline documents
of the USEPA (1974) and the States of Ohio (1975) and Wiscon-
sin (1975), and more recently, the State of California (1976).
Although the most recent version of the EPA (June 3, 1976)
guidelines does not contain specific limitations on heavy metals
or the above mentioned equation, the document implies that the
USDA and FDA are now preparing guidelines and recom-
mendations to reintroduce these limitations and also the
"sludge" equation. Although the early EPA drafts which did
contain such limitations were never published nor proposed
officially, they nevertheless have had a pronounced adverse
impact on state agencies formulating their own guidelines. Sev-
eral states which felt the need of national direction to establish
their own policy on agricultural use of municipal sludges imme-
diately incorporated the USEPA endorsed Zn equivalent con-
cept, the loading formula, and the Zn:Cd ratio concept, pro-
posed by Chaney, into their respective state guidelines.
Wisconsin Guidelines
The Wisconsin Department of Natural Resources in Techni-
cal bulletin No. 88 (Kenney et al.,1') has published guidelines on
sludge application to land. The "guidelines can be used for
screening the land application alternative, evaluation of envi-
ronmental effects . . . and for developing a land application
program. . . ." The guidelines restrict total applications of
sludge by limiting the total "metal equivalent" applications on a
pounds-per-acre basis to not more than 65 times the soil cation
exchange capacity (CEC) in meg/100 g of soil. Metal equiva-
lents of a specific sludge are calculated using the following
formula:
Metal equivalents
ton sludge
(ppm Zn) + 2 (ppm Cu) + 4 ppm (Ni)
500
In addition, yearly applications are limited to a maximum of 2
Ibs/A/yr of cadmium with a total application site lifetime
maximum of 20 Ibs/ A. The lifetime limitation which calculates
to be shorter either by the above equivalent methods or the max-
imum application of 20 lb/ A of cadmium takes precedence.
Ohio Guidelines
A task force appointed by Roy M. Kottman, Director of the
Ohio Agricultural Research and Development Center, pub-
lished in July of 1975.15 The guideline was designed "to assist
landowners in making decisions as they consider the application
of sludge to their land."
The Ohio guide quotes drafts of USEPA guidelines published
prior to June 3, 1976 and suggests that the following equation
"should be used for situations of slightly acid to neutral soils
where pH can be maintained at 6.5 or higher at all times."
Total amount of
CEC x 32.700
sludge (dry tons/acre) = ppm Zn + 2 (ppm Cu) + (ppm Ni) - 200
CEC = Cation exchange capacity of soil before sludge appli-
cation (meg/100 g)
ppm = Parts per million or mg metal/ kg dry weight of sludge
For soils with a pH below 6.5, the 32,700 figure is changed to
16,350 to calculate maximum total loading while for forest soils
with a pH below 6.5, this figure is reduced to 8,175.
Zenz et al.23 presented a study of the impact on municipal
agencies of various drafts of EPA regulations or guidelines con-
cerning heavy metal levels in sludges applied to land. He con-
cluded that most cities could not meet the guidelines while the
restrictive loading rates would virtually eliminate the land
utilization option. For example, it was concluded from a 30-city
survey that only 14.7% of the sewered population of Illinois
could apply sludge to land if sludge with a Zn/Cd ratio of less
than 100 was not allowed to be applied. Even if the Zn/Cd ratio
limitation were not applied, sludge application rates to land
would be limited to a lifetime total of 100 dry tons/ acre for 50%
of the Illinois communities surveyed with soils having a cation
exchange capacity (CEC) of 10 meg/100 g.
Although the most recent guideline of the EPA regarding
sludge application to land (EPA, 1976) does not have limita-
tions on heavy metals contained in sludge applied to land, it does
call for review of all sludge application projects by the United
States Dept. of Agriculture (USDA). The USDA has also
formulated drafts of regulations or guidelines regarding sludge
application, but these have not been officially released.
USDA Regulations/Guidelines
Here we will discuss the USDA regulations or guidelines and
their impact on municipal wastewater treatment agencies.
Basically, the various drafts of USDA regulations contain the
following limitations regarding the heavy metals content of
municipal sludges applied to land:
[No greater amount of sludge-borne metals (kg/Ha)
may be applied than shown in the Table below.]
Table I
METAL
Soil Cation
0-5
Exchange Capacity (racq/] 00 q)
5-15
lb
kg/ha
Zn
Cu
Ni
Cd
Pb
250
125
50
5
500
500
250
100
10
1000
1000
500
200
20
2000
The USDA documents recommend the following:
(1) That sludges with a cadmium content>25 mg/kg (dry wt.)
should not be applied to privately owned land unless their
Cd/Znis^0.015.
(2) Cadmium loadings on land should not exceed 1 kg/ha/yr
from liquid sludge and not more than 2 kg/ha/yr from
dewatered sludge.
(3) Sludge having a cadmium content >1.5 percent of its zinc
content should not be applied on a continuing basis unless
there is an abatement program to reduce cadmium in the
sludge to an acceptable level.
As mentioned earlier, the USDA has offered no direct scien-
tific data to support these recommendations.
Zenz et al.23 did an extensive survey of the metal levels in the
sludges from 30 communities in Illinois. Presented in Figure 1 is
a cumulative frequency plot of the allowable loadings for these
30 Illinois communities for the USDA limitations of 5, 10, and
20 kg Cd/ha. As can be seen, 50% of the communities in Illinois
could not apply over 150 dry tons per acre of sludge to soils with
a CEC of 5 meq/100 g. Since most crops require 10-20 tons of
-------
108 Heavy Metal Uptake/Control Strategies
sludge per acre to supply nitrogen requirements, this means that
application sites would have only a 7.5 to 15 year life span.
Of these 30 Illinois communities, only 8.3% of those with a
sludge cadmium content of over 25 ppm had a Cd/Zn of less
than .015, while 36% of the sludge had Cd contents greater than
1.5% of the zinc content.
OS
0.2
O.I
0.05
150 200 250 300
DRY TONS SLUDGE/ACRE
Figure 1: Cumulative Frequency Distribution for Sludge Load-
ing Necessary to Achieve 5, 10, and 20 kg Cd/ha—Illinois
Zenz et al.23 also presented data on the heavy metal levels of
sludges from five midwestern states and two eastern states of the
United States. These data were collected by the North-Central
Regional Agricultural Experiment Stations Committee on Util-
ization and Disposal of Municipal, Industrial and Agricultural
Processing Wastes on Land. The seven states studied were Wis-
consin, Michigan, Indiana, New Jersey, Minnesota, New
Hampshire and Ohio.
Figure 2 contains a cumulative frequency graph of the allowa-
ble loadings for the sludges from these seven states for the
USDA limitations of 5,10, and 20 kg Cd/ha. In contrast to the
Illinois data, 50% of the sludges from these seven states could
not be applied at rates exceeding 30 dry tons of sludge to soils
with a CEC of 5 meq/100 g over the life of the site. Even soils
with CEC values of 20 would be restricted to total sludge appli-
cations of 120 dry tons or less for sludges from 50% of the cities
in the States surveyed.
Table II contains a listing of the annual application rates per-
missible by the USDA regulations for liquid sludge (1 kg Cd/ha)
and dewatered sludge (2 kg Cd/ha) for the seven States men-
tioned above. As can be seen, only two States could apply over 7
dry tons per acre of sludge to land in any given year. Again, since
most crops will require about 10-20 dry tons of sludge to supply
nitrogen needs, this limitation would virtually eliminate the
option of land utilization.
DRr TONS SLUDGE/ACRE
Figure 2: Cumulative Frequency Distribution for Sludge Load-
ing Necessary to Achieve 5, 10, and 20 kg Cd/ha—Seven States
Table II: Effects of Sludge Application Limitations (dry tons/
acre). Proposed by U.S.D.A. on the Annual Application Rates
of Seven States
STATE
Wisconsin
Michigan
New Jersey
New Hampshire
Indiana
Minnesota
Ohio
* To determine the
muat be known.
Annual Sludge Application Limit*
Annual Loading f,ijni.t for Cadmi.wi
2 kg/ha 1 kg/ha
dry tons per acre
11.6 5.8
5.48 2.8
30.4 15.2
86.6 43.3
5.5 2.74
13.8 6.91
4.5 2.3
life of a site, the CEC of the soil
For example, if the soil has a Cf:C
of 10 meq/lOOg and an annual application rate of
2 kg Cd/ha, the life of this site is 5 years.
Case Histories of Large-Scale and Long-Term
Projects
While long-term data may not be readily available on many
older projects for review and analysis, there is evidence that
sludge utilization in agriculture has been practiced in the United
States for many decades, particularly in the midwestern region.
However, during the last decade, interest at the federal, state and
local levels has generated several large-scale projects which have
produced very valuable data. The following are brief summa-
-------
Heavy Metal Uptake/Control Strategies
109
ries of long-term projects, all of which are ongoing with the ex-
ception of that of the city of Dayton, Ohio.
Hanover Park Experimental Corn Plots
The most extensive demonstration study undertaken within
Metropolitan Chicago is the MSDGC's Hanover Park experi-
mental corn plots.
In 1968, a 7-acre field at the Hanover Park Water Reclama-
tion Plant was selected for the development of experimental
corn plots for furrow irrigation, the topsoil was stripped, the
sub-soil graded to desired slope, and the topsoil was replaced.
Table HI: Average Chemical Concentration of the Liquid Di-
gested Sludge Applied at the Hanover Park Plots from 1968 to
1975 and at the Fulton County Site in 1975.
CONSTITUENT
KJELDAHL N
NH3-N
P TOTAL
K
~A
MG
FE
ZN
UN
^li
Cu
NA
PB
CR
CD
HG (PG/G)
HANOVER PARK
%I1D V
LJKY
4,8
2.8
2,0
0,40
3.4
1,3
2,0
0,075
0,048
0,022
0,070
0,46
0,028
0,026
0,0056
—
FULTON Co,
EASIS
5,7
2,5
3.8
0.39
2.8
1,2
4,7
0,39
0,038
0,042
0,18
0,23
0,093
0,38
0,030
6,95
The soil is a disturbed Drummer silty clay loam with poor natu-
ral drainage. The original experimental design was a random-
ized block with five replications having grades of 0.5, 1.5, and
2.5% confounded in the replications.
Methods
Soil samples were taken in spring and were composites of 20
cores, 0-6 inches deep. The samples were air-dried and then an-
alyzed for pH (Peech17) and electrical conductivity (Bower and
Wilcox2); 1 /3 bar moisture content (Richards19); organic carbon
content (Allison1); cation exchange capacity and exchangeable
K, Na, Ca, Mg (Chapman, 1965); 0.1N HC1 extractable Zn, Cd,
Cu, Cr, Fe, Ni, Pb, Mn, and Al (1:10 soil to 0.1 TVHC1 solution);
and available P (Olsen & Dean16).
Corn grain analysis for trace metal content began with the
1973 corn crop. Grain subsamples were taken during corn yield
determinations. Grain was dry-ashed at 450° C, taken up in 0.1N
HC1, and analyzed by atomic absorption spectroscopy.
Each spring the plots were plowed and harrowed. Ridges and
furrows were formed parallel to field slope. Sludge was applied
from gated irrigation pipes at the upper end of the plots and
allowed to flow down the furrows.
The plots have been fertilized with sludge every year begin-
ning in 1968. Sludge application has not been replicated accord-
ing to the original design. Most of the sludge applied came
directly from heated anaerobic digesters at the Hanover Park
Water Reclamation Plant. In a few instances, lagooned di-
gested sludge was also applied. The average analyses of the di-
gested sludge applied is presented in Table III.
Effect of Sludge Application on Soil
Analyses of the soil samples in 1970, 1973, and 1975 are pre-
sented in Table IV. The control plot showed a depletion of some
major nutrients and organic carbon over this time period. No
nitrogen or phosphorus fertilizer has been applied to the control
plots. Potash fertilizer was applied to all plots in 1968 and
1972-1975. There was no influence on soil pH from the sludge
application.
Table IV: Chemical and Physical Properties of a Disturbed Drummer Silty Clay Loam Collected in 1970,1973 and 1975
from the Hanover Park Experimental Corn Plots. These Plots have Received Liquid Sludge from 1968-1975.
Parameter
PH
CEC meq/lOOg
1/3 bar H2O %
C-organic %
P-available
pg/g
Exchangeable
K pg/g
Na pg/g
Ca pg/g
Mg pg/g
0.1 N HC1
Extractable
Fe yg/g
Mn pg/g
Zn pg/g
Cd pg/g
Cu pg/g
Cr pg/g
Ni pg/g
Pb pg/g
1970
1973
Sludge Applied to
0
7.9
33.4
38.0
3.08
N.A.*
138
95
5440
1520
813
253
30.5
0.7
18'. 3
1.7
11.3
16.2
15
7.7
34.0
38.0
3.21
N.A.
158
131
5660
1600
827
281
39.0
0.7
66.6
.2.30
10.9
11.8
25
7,7
32.2
36.8
3.15
N.A.
153
117
5530
1600
835
290
37.5
0.5
33.8
5.2
10.9
13.3
0
7.3
33.9
32.2
2.61
9.8
126
25
5140
1660
301
198
24.4
0.68
6.72
0.93
8.60
8.25
37
7.1
33.7
33.1
3.37
51.5
149
30
6300
1700
339
200
23.4
0.58
13.3
1.26
9.11
9.21
1975
Date (T/A)
62
7.1
33.3
32.2
3.33
79.8
126
44
5650
1690
346
195
27.6
0.61
15.8
1.47
8.93
9.68
0
7.8
31.4
32.0
2.66
31.3
106
22
4300
1060
247
211
22.9
0.39
9.3
0.42
6.78
9.1
52
7.5
34.5
34.7
3.35
102.9
122
84
4680
1100
409
222
50.0
0.49
23.2
1.02
8.15
12.0
91
7.6
32.3
33.8
3.01
113.9
117
78
4590
1060
412
235
51.7
0.50
30.2
1.46
7.39
13.9
*NA = No Analysis
-------
110 Heavy Metal Uptake/Control Strategies
The 1975 soil analysis showed that Zn, Cd, Fe, available P,
NO2 + NO. = N, organic carbon, and 1/3 bar water holding
capacity all increased with the increase in the levels of applied
sludge. The level of NI and cation exchange capacity (CEQ
indicates that if higher levels of sludge application were to be
used that the NI and CEC level would not increase any further.
The soil concentration of Cu, Pb and Cr increased with
increased sludge application.
Effect of Sludge Application on Corn
Metal uptake in the grain was determined in the 1973 and
1975 crops. These data are presented in Table V.
The only significant increases in metal content in 1973 were
with K and Ca. These increases occurred with the 52 T/ A sludge
application, and the reasons for these increases are not appar-
ent. The 1975 grain crop showed significant increases in Zn, Cd,
Cu, and Fe content with the addition of sludge. The Fe increase
was linear in behavior for the sludge application rates used,
while the Cu response appears to be approaching a maximum.
The Cd concentration of the grain grown in 1975 on the 85
and 110 T/A total sludge applications were 0.091 and 0.096
fjglg, respectively.
A survey conducted by Pietz, Peterson, Lue-Hing and Welch
(Unpublished Data, 1976) on 15 corn fields located throughout
Illinois, where all fields were fertilized with inorganic fertilizer,
yielded a mean Cd concentration of 0.038 Mg/g ano< a range of
0.001 to 0.294/ig/g in tne grain. A summary of this survey is
presented in Table VI. The concentrations of Zn, Cu, Cd and Fe
at the Hanover Park farm were all within the range reported by
Pietz etal.18
The corn yields in 1973 were 29,78, and 99 bushels per acre for
the 0, 52, and 79 T/ A (accumulative) of sludge applications,
respectively. In 1975 the corn yields were 49,102, and 82 bushels
per acre for 0, 85 and 110 T/A of sludge applications, respec-
tively.
Using the USDA guidelines or the Wisconsin guidelines max-
imum lifetime allowable sludge application at the Hanover
Park site is 89 dry T/A. This application limit is due to the Cd
content of the sludge which was 56/ug/g dry sludge.
Table VI: Elemental Analysis of Corn Grain from Fifteen Loca-
tions in Illinois (243 Samples)
ELEMENT
IA
FE
K
16
^
YA
ID
CR
Cu
Hi
PB
ZN
tSTANDARD
MEAN
50.6
20,3
2,970
1,080
5.60
7.80
0,038
0,087
2,62
0,87
0,168
19,6
DEVIATION
* SD*
+ 8.1
* 9.4
* 954
• 105
* 1,01
' 5.44
+ 0,030
± 0,059
+ 0.88
< 0,34
* 0.065
• 3,2
OF THE SAMPLE
RANGE
33,3
10,4
1,930
617
2,00
1,24
0.001
0,002
1,32
0,17
0,033
13,0
98,5
84,0
4,000
1,480
9,26
35,5
- 0,294
0.429
6.31
2,50
0,519
32,6
COEFFICIENT
OF
VARIATION,
16,6
42,2
32,1
9,77
18.1
69.7
31,3
68.9
33.5
39.6
33.5
16,6
Table V: Chemical Concentration of 1973 and 1975 Corn Grain Grown at the Hanover Park Experimental Corn Plots which
Have Received Liquid Sludge from 1968-1975.
Sludge
Applied
to Date
T/A
1973
0
52
79"
1975
0
85
110
*Values
**Values
Zn Cd
25
25
23
18
25
25
in
in
.0 <0.5
.8 <0.5
.8 <0.5
.2* 0.067**
.8 0.091
.8 0.096
a column are
a column are
Cu
1.74
2.01
1.70
1.81**
2.32
2.29
Cr
0.77
0.76
0.71
0.076
0.058
0.067
significant at
significant at
Fe
39.2
34.4
36.9
23.6**
29.8
30.3
the 0.
the 0.
Ni
g dry bas
<0.5 0
< 0.5 0
< 0.5 0
0.63 0
0.43 0
0.48 0
05 level
01 level
Pb K
.81 3030
.83 3600
.79 3158
.54 3260
.53 3490
.51 3380
Na
26.2
27.2
32.1
9.71
2.78
3.39
of probability for
of probability for
Ca
57.9
70.6
55.0
61.4
57.3
54.2
that
that
Mg
1330
1340
1240
1070
1140
1160
year.
year.
Mn
5.07
5.51
5.65
5.70
6.52
6.52
-------
Heavy Metal Uptake/Control Strategies
Dayton, Ohio Sewage Treatment Plant Farm
At Dayton, Ohio, the sewage sludge has been spread on War-
saw silt loam for 35 years. The annual sludge application rate
has been calculated at 28 metric tons/ha (12.5 T/A). A study of
the trace metals in the corn grown on this land was conducted by
Kirkham12 in 1973. She reported the total concentration of
metals in the 0-30 centimeters depth of soil was increased by the
following factors: Cd, 35; Cu, 16.5; Ni, 2; Pb, 16.5; and Zn, 13.
Even with these large increases in the metals present in the soil,
these metals did not accumulate appreciably in the corn plants
when compared to unsludged corn plants. For example, the Cd
content of corn grains was 0.9 and 0.8 ppm dry wt. for sludge-
grown and control corn, respectively.
The Dayton, Ohio, sludge contained from 800-830 mg Cd/ kg,
4,100-6,020 mg Cu/kg, 400 mg Ni/kg, 4,000-6,940 mg Pb/kg,
and 8,390-10,500 mg Zn/kg dry wt. basis. This calculates to a
total 35 year per hectare application of approximately 780 kg
Cd, 4,020 kg Cu, 390 kg Ni, 3,800 kg Pb, and 8,200 kg Zn.
These loadings for Cu, Ni, Pb, and Zn, as reported by Kirk-
ham, are about 40-50 times higher, and loadings of Cd about 100
times higher than those proposed by Chancy in 1973. However,
it is clear that the Cd uptake in the sludge amended corn grain is
very small. Using the Wisconsin guidelines where the total
sludge application is limited by the lifetime maximum of 20 Ib
Cd/A (22.4 kg/ha). The Dayton sludge which contained 0.8 kg
Cd/metric ton of solids would be limited in Wisconsin to a
lifetime application of 28 metric tons/ha (12.5 T/ A), or field
operational life of 1 year per field.
Kirkham estimated the total application at Dayton, Ohio, to
be 980 metric tons/ha over the 35 year period of record.
University of Illinois Experimental Field Plots—
Soybeans
Except for the 1975 data, most of the results reported in Table
VII were presented previously by Hinesly et al.9 By regression
analyses they concluded that sludge-borne Cd incorporated into
soil just prior to planting soybeans was more efficiently assimi-
lated than sludge-borne Cd applied in years previous to the crop
year. Results of their analyses suggested that if annual sludge
applications were terminated, Cd contents in plant tissues would
be drastically reduced. To further test the hypothesis, annual
sludge applications were continued on one-half and terminated
on the other half of split plots, beginning in 1973. Another rea-
son for making the split was because soybeans growing on maxi-
mum sludge-treated plots had suffered from P toxicity in 1972.
The P toxicity was not observed in 1973, when sludge was not
applied to any plots after 2 applications in the early part of the
1972 growing season until after harvesting operations in 1973. In
1974, on split-plots where maximum sludge applications were
resumed in 1973, soybeans had a Cd content of 0.92//g/g as
compared to 0.35 Mg/g in soybeans grown on plots where no
further applications were made. Also, it may be noted that
where sludge applications were terminated, Cd contents of soy-
beans from plots previously treated with 1 /4-maximum sludge
applications were not significantly different from Cd contents in
soybeans from control plots. In 1975, Cd contents of soybeans
from plots where sludge applications were terminated were sim-
ilar to those of soybeans in 1974. But in 1975, Cd contents in
beans from previously treated I/4-maximum and 1/2-
maximum treated plots were not significantly different from
contents of soybeans from control plots. Soybeans growing on
Table VII: Mean Dry Weighted Contents of Soil Samples from 0- to 6-Inch Depths and Soybean Samples Collected from
Blount Silt Loam Plots9
SLUDGE APPLICATION RATES
Cumm.
YEAR Solids
mt/ha
1969 43
1970 102
1971 237 <
1972 242 0
1973 255 0
1974 (A) 314 0
1974 (B) 255 0
0
(inches)
1/4 1/2
1
-~^ My/y
Soil 0.1 N HC1 Extractable
. 10
.19
.16
.20
.15
2.94 8
3.63 9
3.86 8
4.03 8
4.43 8
.11
.53
.51
.14
.74
13.33
15.24
13.66
14.69
14.13
SOYBEANS
1970
1971
1972
1973
1974 (A)
1974(3)
1975(A)
1975(B)
A — Sludge applications
B — Sludge applications
*Significant at the 0.
**Signif icant at the 0.
.06
.27
.29
.18
.31
.13
.07
.12
.32
.41
.49
.31
.31
.23
.34
.18
.54
.90
.94
.55
.57
.35
.72
.34
were resumed after harvest in
were suspended following
05
01
level of probability.
level of prabability.
one
Soil
PH
5.7
5.4
5.2
5.8
6.0
6.1
6.0
6.3
1973.
application
1.02
1.55
3.00
0.75
0.92
0.35
1.69
0.46
in 1972.
Cumni.
Cd
kg/ha
34.7
58.5
59.4
62.0
77.6
62.0
89.2
62.0
LSD
2.29**
1.43**
1.38**
4.53**
4.45**
0.30**
0.30**
0.93**
0.13**
0.39**
0.15*
0.31**
0.29**
-------
112 Heavy Metal Uptake/Control Strategies
maximum sludge-treated plots in 1975 where application had
been resumed, suffered from P toxicity again. Cadmium con-
tents were highest in 1972 and 1975 in soybeans harvested from
maximum sludge-treated plots where plant growth was ad-
versely affected by P toxicity. Thus, it is concluded that Cd
contents in soybeans will decrease with time after sludge appli-
cations have been terminated and that Cd contents will be
increased in plants subjected to conditions adversely affecting
growth.
University of Illinois Experimental Field Plots—Corn
Except for the 1975 results, all of the data presented in Table
VII were previously reported by Hinesly et al.10 Regressing Cd
contents of corn grain on annual and accumulative sludge-borne
Cd applications, they concluded that application just prior to
planting the crop was the main determinant of Cd levels in corn
grain. However, even in the absence of results from regression
analyses, an examination of the data presented in Table VIII
shows that Cd levels in corn grain varied with the amount of
sludge-borne Cd applied by furrow irrigation during the pre-
vious growing season and incorporated just prior to the crop
year of interest. On a per unit basis of annually applied sludge-
borne Cd, levels of the metal in corn grain remained fairly con-
stant regardless of amounts incorporated by spring plowing in
years previous to a particular crop year. These data suggest that
if Cd levels in grain are within acceptable concentrations for a
particular annual loading rate of sludge-borne Cd, the situation
will not change with accumulative years of sludge applications.
From both the soybean and corn data presented in Tables VII
and VIII, it appears that availability of Cd is considerably more
important in determining levels in plant tissues than soil pH.
Availability of sludge-borne Cd for assimilation by plants is
greatest immediately after incorporation into soils and then
decreases with time.
These data from Hanover Park, Illinois; Dayton, Ohio; and
the University of Illinois prove that the type of restrictions being
proposed on heavy metals and on Cd in particular are not nec-
essary. Bodman and Baker (1975) presented evidence that the
Cd in the sewage sludge was not readily available to soybean and
corn leaves or grain. They also found that the Zn:Cd ratio
increased over the check plot with the use of 10 dry T/ A/yr of
sewage sludge, and they concluded that the Zn:Cd ratio did not
provide a reliable index for evaluating the suitability of a sludge
for application on cropland. Bodman and Baker did get
increased metal concentrations of Zn and Cd in the leaves and
grain of corn and soybeans, if the soil was treated with the 10
T/ A sludge equivalent level of a Zn- or Cd- salt. Thus showing
the higher availability when metals are applied directly as salts
rather than in combined forms such as sewage or sludge.
Alternatives to Empirical Sludge Equations—
Industrial Waste Control
A very effective way to reduce the level of metals contained in
municipal sewage sludges is to limit the industrial input of
metals at the source; namely, at the point where industry
discharges into the sewer system. While industrial waste control
Table VIII: Mean Dry Weight Cd Contents of Soil Samples from 0- to 6-Inch Depths and Corn Grain Samples Collected
from Blount Silt Loam Plots10
SLUDGE APPLICATION
YEAR
0
1/4
RATES (inches)
1/2
1
..„/„
1971
1972
1973
1974
1970
1971
1972
1973
1974
1975
Cumm.
Solids
mt/ha
152.5
280.9
306.5
368.6
417.4
449.9
<0
0
0
0
0
0
0
0
0
0
A — Sludge applications
**3ignificant at the 0.
.25
.27
.29
.58
.30
.14
.14
.08
.09
.06
*-" My/y -
Soil 0.1 N HC1 Extractable
1.3 3.6 6.8
3.1 4.7 12.1
3.3 6.7 12.9
3
0
0
0
0
0
0
.8
7.3
CORN GRAIN
.60 0.79
.70
.45
.15
.18
.17
0.65
0.83
0.35
0.40
0.28
were resumed after ^harvest in
01 level of probability.
16.4
Soil
4.9 1.00
5.
5.
5.
6.
6.
1973.
2 0.92
0 1.10
3 0.61
2 0.81
6 0.51
Cumm.
Cd
kg/ha
48
77
81
88
101
108
LSD
4.0**
3.3**
3.6**
4
0
0
0
0
0
0
.9**
.20**
.40**
.52**
.06**
.22**
.08**
-------
Heavy Metal Uptake/Control Strategies
113
has been shown to be effective, it may be emphasized here that
metals including cadmium and mercury will always be present in
sewage sludges as a result of their natural content in food and
human wastes, and from surface runoff, etc.
A successful case in point is the experience of the Metropoli-
tan Sanitary District of Greater Chicago, where an industrial
waste control ordinance has been in effect since 1969. To evalu-
ate the effects of this ordinance, the metal content of anaerobi-
cally digested sludges from the Calumet plant which were placed
in storage lagoons before 1969 were compared with sludges from
this same plant through the summer of 1974, as shown in Table
IX. It is clear from the data that the metal content of the sludge
was lower after passage of the ordinance than before. The con-
centrations of metals of greatest environmental concern (nickel,
copper, lead, and cadmium) were reduced in the digested sludges
by 92,81,73, and 72%, respectively, during the period from 1969
to 1974. Data for 1976 are similar to those for 1974, which indi-
cates that industrial waste control has reached its limits of
effectiveness for this plant with respect to sludge metals levels.
Although an industrial waste control program can effect sub-
stantial reductions in the metal content of sludges in industrial
areas, it must be emphasized that each sewage treatment plant or
the area served by one has an indigenous concentration for all
metals which may not be further reduced by more stringent
industrial pretreatment. For example, Chicago's industrial
waste control ordinance has had no appreciable effect on the
metal content of sludge from the Hanover Park plant, which is
subject to the same ordinance and which serves a nonindustrial
area. Digested sludges from this plant had a cadmium content of
60 ppm in 1969 and 56 ppmin 1975. The cadmium content of the
Hanover Park sludge before and after the 1969 ordinance is sim-
ilar to that of the Calumet sludge in 1974.
A successful industrial pretreatment program implies the
existence of effective monitoring and enforcement procedures.
In view of the political ramifications of these procedures, the
effectiveness of an industrial pretreatment program cannot be
generalized from one area to another until such time that the
commitment becomes national.
Genetic Modification
There is ample data that the soil-root barrier restricts the pas-
sage of metals into the leaves, stems and reproductive parts of
plants; and that metals taken up by roots accumulate preferen-
tially in the stems and leaves rather than being translocated to
the fruits or grains as is typified by the cereals. These natural
characteristics of- the cereal plants, particularly corn with its
high nitrogen demand, makes it a prime candidate for possible
genetic modification of hybrids to reduce the accumulation of
metals in the grain.
This type of plant genetics research is currently being con-
ducted at the University of Illinois. Preliminary results of green-
house studies using 6 inbred lines and 2 crosses of corn have indi-
cated that some lines are capable of greatly reducing the Cd
uptake to the grain.
REFERENCES
1. Allison, L. E. 1965. Organic carbon. Chap. 90. In Methods
of Soil Analysis. C. A. Black (Ed.), Am. Soc. Agron., Madison,
Wisconsin.
2. Bower, C. A. and L. V. Wilcox. 1965. Soluble salts by elec-
trical conductivity. Chap. 62. In Methods of Soil Analysis. C. A.
Black (Ed.), Am. Soc. Agron., Madison, Wise.
Table IX: Effect of Industrial Waste Control Ordinance on Metals Content of the Calumet Sewage Treatment Plant's
Sludge
METAL
CD
MAX, LIMIT
BY
ORDINANCE
wflr/A
2.0
CR (Tot.) 25.0
Cu
FE
HG
Ni
PB
Z?j
•MEAN
•tflEAN
±MEAN
FROM
3.0
50.0
0.0005
10.0
0.5
15.0
OF 10 SAMPLES
OF 22 SAMPLES
SLUDGE SOURCE
LAGOONS ANAEROBIC DIGESTER
BEFORE
1969* 1972 1 1974±
190 100 54
2100 1100 790
1500 900 282
53700 36800 24200
3.3 3.0 2.15
1000 200 77
1300 1800 486
5500 3500 2800
BEFORE
1969/1974
RATIO
3.51
2. 55
5.32
2.22
1.53
12.99
3.70
1.95
FROM THE MSDGC CALUMET SEWAGE TREATMENT PLANT
FROM JUNE TO OCTOBER 1972 (PETERSON,
LUE-HlNG,
METAL REDUCTION
PERCENTAGE
72
62
81
55
34
92
73
49
SLUDGE LAGOON
AND ZENZ, 1973)
OF 6 SAMPLES FROM THE MSDGC CALUMET SEWAGE TREATMENT PLANT ANAEROBIC DIGESTERS
JUNE TO OCTOBER 1974
-------
114 Heavy Metal Uptake/Control Strategies
3. California, State of, Dept. of Health. 1976. Proposed regu-
lations on handling, disposal and application to land of sewage
solids. To be included in the California Administrative Code—
Title 22, Div. 4, Environmental Health. Sacramento, Calif.
4. Chancy, R. L. 1973. Crops and food chain effects of toxic
elements in sludges and effluents, pp. 129-152. In Proceedings
of the Joint Conference on Recycling Municipal Sludges and
Effluents on Land. National Assoc. of State Univ. and Land
Grant Colleges, 1 DuPont Circle N.W., Washington, D.C.
5. Chumbley, C. G. 1971. Permissible levels of toxic metals in
sewage used on agricultural lands. In ADAS Advisory Paper,
No. 10. Ministry of Agriculture, Fisheries, and Food. Wolver-
hampton, England.
6. Council for Agricultural Science and Technology. 1976.
Report No. 64. Application of sewage sludge to cropland:
Appraisal of potential hazards of the heavy metals to plants and
animals. November.
7. EPA. 1976. Municipal sludge management, environ-
mental factors, Part IV. Federal Register 41:22532-22536. June
3.
8. Furr, A. K., A. W. Lawrence, S. S. C. Tong, M. C. Gran-
dolfo, R. A. Hofstader, C. A. Bache, W. H. Gutenmann, and D.
J. Lisk. 1976. Multielement and chlorinated hydrocarbon anal-
ysis of municipal sewage sludges of American cities. In Environ.
Sci. and Tech. 10:683-687.
9. Hinesly, T. D., R. L. Jones, J. J. Tyler, and E. L. Ziegler.
1976A. Soybean yield responses and assimilation of Zn and Cd
from sewage sludge-amended soil. In Journ. Water Pollution
Control Federation, Vol. 48, No. 9, pp. 2137-2152.
10. Hinesly, T. D., R. L. Jones, E. L. Ziegler, and J. J. Tyler.
I976B. Effects of annual and accumulative applications of sew-
age sludge on the assimilation of zinc and cadmium by corn (Zea
mays L.). In press Environmental Sci. and Tech.
11. Keeney, D. R., K. W. Lee, and L. M. Walsh. 1975. Guide-
lines for the application of wastewater sludge to agricultural
land in Wisconsin. Tech. Bull. No. 88. Dept. of Natural Re-
sources, Madison, Wise.
12. Kirkham, M. B. 1975. Trace elements in Corn Grown on
long-term sludge disposal site. In Environmental Science and
Technology, Vol. 8, No. 8. August, pp. 765-768.
13. Klein, L. A., M. Lang, N. Nash, and S. L. Kirschner. 1974.
Sources of metals in New York City wastewater. In JWPCF
Vol. 46, No. 12. December, pp. 2653-2662.
14. Konrad, I. G. and S. J. Kleinert. 1974. Removal of metals
from waste waters by municipal sewage treatment plants. Tech.
Bull. No. 74, Dept. of Natural Resources. Madison, Wisconsin.
15. Ohio Cooperative Extension Service. 1975. Ohio guidefor
land application of sewage sludge. Bull. 598. Cooperative
Extension Service. The Ohio State University. Columbus.
16. Olsen, S. R. and L. A. Dean. 1965. Phosphorus. Chapt.
73. In Methods of Soil Analysis. C. A. Black (Ed.), Am. Soc.
Agron., Madison, Wise.
17. Peech, M. 1965. Hydrogen-ion activity. Chapt. 60. In
Methods of soil analysis. C. A. Black (Ed.), Am. Soc. Agron.
Madison, Wise.
18. Pietz, R. I., J. R. Peterson, C. Lue-Hing, and L. F. Welch.
1976. Variability in the concentration of 12 elements in corn
grain. March. Unpublished Data. Dept. of Research and De-
velopment, The Metropolitan Sanitary District of Greater Chi-
cago.
19. Richards, L. A. 1965. Physical condition of water in soil.
Chapt. 8. In Methods of Soil Analysis. C. A. Black (Ed.), Am.
Soc. Agron. Madison, Wise.
20. Sewage and waste control ordinance as amended. 1975.
Adopted bythe Board of Trustees of the MSDGCon September
18, 1969, and as amended to and including February 24, 1972
and October 2, 1975.
21. Stenstrom, T. and M. Vahter. 1974. Cadmium and lead in
Swedish commercial fertilizer. Ambio 3:91-92.
22. USEPA notice of intent to issue a policy statement on
acceptable methods for the utilization or disposal of sludge from
publicly-owned wastewater treatment plants. Dated Feb. 20,
1974. Comments and recommendations of the MSDGC.
December 6, 1974.
23. Zenz, D. R., et al. 1975. USEPA guidelines on sludge
utilization and disposal—a review of its impact upon municipal
wastewater treatment agencies. Presented at the 48th Annual
WPCF Conf., Miami Beach, Fla. October 5-9.
-------
General Technology on Soil-Salt
Chemistry and Saline Subsurface
Irrigation Return Flows
Gaylord V. Skogerboe, Wynn R. Walker and
David B. McWhorter
Agricultural Engineering Department
Colorado State University
Fort Collins, Colorado
James E. Ayars
Agricultural Engineering Department
University of Maryland
College Park, Maryland
INTRODUCTION
There has been considerable interest in recent years regarding
water quality problems resulting from irrigated agriculture.
Most everyone involved with agriculture in the West is keenly
aware of EPA's efforts to include irrigation return flows in the
permit program under the National Pollutant Discharge Elimi-
nation System (NPDES).
This paper discusses the most prevalent water quality prob-
lem resulting from irrigation—namely salinity—which results
from subsurface irrigation return flows. These subsurface flows,
consisting of seepage losses from canals and laterals along with
deep percolation losses from over-irrigation of croplands, result
in increased ground-water salinity concentrations which fre-
quently reach downstream river channels. Generally, subsurface
saline irrigation return flows must be considered as a non-point
source of pollution.
That portion of the water supply which has been diverted for
irrigation but lost by evapotranspiration (consumed) is essen-
tially salt-free. Therefore, the irrigation return flow will contain
most of the salts originally in the water supply. The surface irri-
gation return flows will usually contain only slightly higher salt
concentrations than the original water supply. Thus, the water
percolating through the soil profile contains the majority of salt
left behind by the water returned to the atmosphere as vapor
through the phenomena of evaporation and transpiration. Con^
sequently, the percolating soil water contains a higher concen-
tration of salts. This is referred to as the "concentrating" effect.
As the water moves through the soil profile, it may pick up
additional salts by dissolution. In addition, some salts may be
precipitated in the soil, while there will be an exchange between
some salt ions in the water and in the soil. The salts picked up by
the water in addition to the salts which were in the water applied
to the land are termed salt "pickup." The total salt load is the
sum of the original mass of salt in the applied water as the result
of the concentrating effect plus the salt pickup.
The most significant improvements in reducing downstream
salinity problems will potentially come from improved on-farm
water management. This will be particularly true for areas con-
taining large quantities of natural pollutants, such as salts, in the
soil profile. In such situations, the key is to minimize the subsur-
face return flows, thereby minimizing the quantity of salt
pickup. Poor irrigation practices on the farm are the primary
cause of overly large water diversions, as well as being the pri-
mary source of present return flow quality problems. Due to the
nature of irrigated agriculture, whereby salts must be leached
from the root zone, an optimum solution will, in most cases,
require improvements in on-farm water management.
A term frequently used to characterize irrigated agriculture is
"site specific", which means that each irrigated area has its own
unique physical characteristics, that in turn require a unique set
of solutions for alleviating the irrigation return flow quality
problems peculiar to that area. In order to arrive at the unique
set of technological solutions to reducing salt loads from a par-
ticular irrigated area requires a process of: (a) using an inflow-
outflow analysis to determine if there is a salinity problem, and if
so, its magnitude; (b) determining the sources of salinity by de-
veloping a hydro-salinity model of the area; (c) determining the
soil chemistry characteristics of the area including the devel-
opment of a predictive soil chemistry model; (d) developing
cost-effectiveness relationships for a variety of technological al-
ternatives, as well as combinations of appropriate technologies;
and (e) assessing the cost-effectiveness relationships in an opti-
mizational format in order to arrive at the least cost combina-
tion for achieving a desired salinity reduction.
Inflow-Outflow Analysis
An inflow-outflow analysis is usually the first step in analyz-
ing water management and salinity problems. A schematic ex-
ample of irrigated tracts is shown in Figure 1 wherein an inflow-
outflow analysis is represented by the difference in.the two
stream gaging stations, A and B. The well-known hydrologic
equation,
1-O = AS
(1)
where 1 is the inflow, O is the outflow, and A S is the change in
storage, can be used.
Representing the volume of salts passing Stream Gaging Sta-
tion A over any desired period of time as As,
As-Bs- AS
(2)
Initially, the analysis is undertaken using annual data over the
time length of available data for the two stream gaging stations.
(A word of caution—unmeasured subsurface flows underneath
the two stream gaging stations could have a significant effect on
115
-------
116 Soil-Salt Chemistry
the analysis.) If A S is positive, then salts are being accumulated
in the soil profile of the irrigated lands. If A S = O, a salt balance
is being maintained. A long-term (e.g., 10, 20 ... 50 years) salt
balance is necessary for irrigated lands in order to maintain agri-
cultural productivity.
If A S is negative in Eq. 2 (B s > A s), then salt pickup is occur-
ring and the difficulty comes in determining the source of this
salt pickup. Also, this would indicate that at least the shallow
groundwater flows would be expected to be highly saline. A
long-term annual time history of salt pickup will disclose (pro-
vided sufficient data are available) whether or not the salt
pickup rate is increasing, remaining fairly constant, or declin-
ing. A declining salt pickup rate indicates that the natural salts in
the soil profile (either irrigated lands or watershed lands) are
being taken into solution and conveyed into the groundwater
reservoir, and then returned to the river. A relatively constant
salt pickup rate indicates a large source of salts in comparison
with the amount of water percolating through the soil profile
and moving through the groundwater reservoir; however, addi-
tional analysis (e.g., hydro-salinity modeling) will be required to
determine how much of this salt pickup is the result of irrigated
agriculture and how much occurs from natural runoff. Also,
mineralized springs could be a significant contributor of salts.
An increasing salt pickup rate indicates man-made activities
(e.g., increasing irrigated acreages or poorer water management
on existing irrigated lands).
X
V^* I
V i Stream Gaging
~-^JW Station A
where
I =
0 Outflow
AS Change in Storage
Figure I: Schematic Example of Inflow-Outflow Analysis
The schematic in Figure 1 shows two tracts of irrigated lands.
Thus, an inflow-outflow analysis between stream gaging sta-
tions A and B does not yield any information regarding the dif-
ferences between the two tracts. The installation of an additional
stream gaging station on the river between the two tracts may
allow an inflow-outflow analysis which has only one tract of irri-
gated land inside the natural surface boundary, but this depends
upon the geology and its effects upon subsurface return flows.
An inflow-outflow analysis is primarily useful for identifying
which areas are contributing salts to a river, or whether or not an
area is maintaining a salt balance. This analysis will not provide
the solutions to indicated problems, but will disclose which
areas require additional study in order to arrive at necessary
technological solutions.
Hydro-Salinity Model
Salinity problems from irrigated agriculture are the result of
subsurface return flows. Therefore, it becomes highly important
to model subsurface, or groundwater, flows. The capability of a
hvdro-salinity model to provide the necessary information for
arriving at technological solutions is highly dependent upon the
accuracy of the ground-water field data and analysis.
A difficulty often encountered while preparing water and salt
budgets is the variability in the accuracy and reliability with
which the hydrologic and salinity parameters are measured.
Usually, the measurement precision varies with the scope of the
investigation and the area of the study.
Since the hydrologic system is difficult to monitor and pre-
dict, it is impractical to expect models to operate without apply-
ing some adjustments in order that all components will be in bal-
ance. In short, the budgeting procedure is usually the
adjustment of the segments in the water and salt flows according
to a weighting of the most reliable data until all parameters rep-
resent the closest approximation of the area that can be achieved
with the input data being used. The vast and lengthy computa-
tion procedure of calculating budgets is facilitated by a mathe-
matical model programmed for a digital computer. A schemat-
ic diagram of a general hydro-salinity model is shown in Figure
2.
Figure 2: Schematic of Generalized Hydro-Salinity Model
-------
Soil-Salt Chemistry
117
The model was developed in three general sections:
I. All diversions from the canals through small turnouts
into the lateral network are distributed onto the farm-
land after taking into account lateral seepage losses;
2. Flow within the root zone including evapotranspira-
tion, tailwater runoff, and deep percolation losses; and
3. Groundwater return flows resulting from seepage and
deep percolation return to the river system with their
salt loads through both surface and subsurface drain-
age routes.
The details of the computer program are listed by Walker.3
(Note that the hydro-salinity model between sections 2 and 3
does not make any computations between the root zone and the
underlying groundwater, which must be analyzed using the soil
chemistry modeling techniques that will be discussed later.)
The mathematical model attempts to simulate the hydrologic
conditions of an agricultural system. The concepts are general
and can be extended to many areas with some modifications
likely required for each area. The program is written in individ-
ual but interconnected subroutines that provide a measure of
flexibility during operations by separating the calculation phase
from either input or output phases. Thus, several of the subrou-
tines become optional if their functions can be replaced by input
data, or if certain outputs are not desired.
The main portion of the program is used to read necessary
input data and to control the order of water and salt budget cal-
culations. There are certain advantages in separating the input,
output and computational stages of a program including:
I. Input order is not important as the data are completely
available at all stages of computation.
2. Variable sets of data can be utilized in the model when
several budgets are desired, or when some form of in-
tegration is desired. This is especially useful when an
area can be broken down into smaller dependent areas
(e.g., Tracts 1 and 2 in Figure 1).
3. The functions of the subroutines are independent of
input, thereby making each subroutine a unit that can
be implemented in other programs.
4. Corrections and adjustments are easily made without
detailed consideration to other segments of the pro-
gram.
In controlling the computational order of the program, the
main program separates the calculation of the water and salt
budgets. Consequently, the modeling procedure involves only
the water phase of the flow system. This is possible when suffi-
cient field data have been collected. Once the water flow system
has been simulated, the individual flows are multiplied by mea-
sured salinity concentrations and converted to units of tons per
month. At this point in the formation of the budgets, careful at-
tention must be given to the salt flow system since irregularities
may be present, thereby necessitating further model adjust-
ments. Thus, when the final budgets have been generated, the
salt system, groundwater system, and surface flow system must
be reasonably coordinated and additional reliability is assured.
Soil Chemistry Modeling
The hydro-salinity model describes the present situation in an
agricultural area regarding water and salt flows. However, the
only method for predicting the reduction in salts returning to the
groundwater and river through implementation of any salinity
control measure(s) is by assuming a one-to-one relationship
between water (reduction in subsurface return flows) and salt
pickup. That is, if the subsurface return flow is reduced by 50%,
the salt pickup is also reduced by 50%.
The primary objective is to model the transport of salts
through the soils. (See Ayars1 for a more thorough discussion.)
The first portion of the flow of water and consequent transport
of salts is through the root zone, which is usually a zone of par-
tial saturation. A numerical model of the moisture flow and
chemical and biological reactions occurring in the root zone has
been developed by Dutt, et al.2 (This reference contains a com-
plete listing of the computer program.) This is the basic model
which will be used to describe salt transport.
The model consists of three separate programs. The first pro-
gram describes the soil moisture movement and distribution
with time. The second program interfaces the soil moisture
movements with the chemical-biological model. This is needed
because the horizons used in the calculations of soil moisture
and chemistry differ. The third program computes the chemical
and biological activity occurring in the soil profile. Figure 3 is a
block diagram of the overall model. A brief description of the
moisture flow and chemical-biological models is included to
serve as a basis for understanding the data collection require-
ments.
INPUTS - WATER
APPLICATION S
CONSUMPTIVE USE
DATA
INPUTS = FERTILIZER
81 ORGANIC-N
APPLICATIONS,
CROP TYPES
INPUTS et INIT
WATER CONTENT
OF SOIL
Al
a
INPUTS =
CHEMICAL
Figure 3: Generalized Block Diagram of the Model
The flow is one dimensional and was developed using the
Richards equation with a sink term. Schematically, the model is
given in Figure 4. Mathematically, the flow is described using
Richards equation in the form:
90
aT
w « I 1x0(7 •
(3)
where
9 = volumetric water content
t = time
x = length
K = hydraulic conductivity
S = sink term
D = diffusivity
This is the diffusivity form of the equation which means that
only flow in the partially saturated zone of the soil profile can be
described. The sink term, S, is computed using the Blaney-
Criddle equations for evapotranspiration (other equations
could be adapted to the model) with the loss due to evapotrans-
piration being distributed through the soil profile by assuming a
specific root distribution for the crop. The root distribution and
coefficients for the Blaney-Cnddle equations are supplied by the
user. Actual values of evapotranspiration can be used in the sink
term when they are known.
-------
118
Soil-Salt Chemistry
C
START MOISTURE
FLOW PROGRAM
1
PROGRAM MOISTRE
READ CONTROL AND INPUT DATA
COMPUTE MOISTURE CONTENT AND
FLUX FOR EACH DEPTH NODE AND
TIME STEP
WRITE ON MAGNETIC TAPE OR
PRINT OUTPUT
1
SUBROUTINE THEDATE
COMPUTE CALENDAR DATE FROM
DAY NUMBER
1
SUBROUTINE CONUSE
COMPUTE VALUE OF MACROSCOPIC
SINK TERM
STOP MOISTURE
FLOW PROGRAM
Figure 4: Generalized Block Diagram of Moisture Flow Pro-
gram
Salt transport is described by the following equation in one
dimension.
3c 3
where
c = solute concentration
t = time
D = dispersion coefficient
z = depth
v = flux or darcy velocity
By assuming the term 3/3z (D 9c/3z) is negligible com-
pared to v 3c/3z the equation reduces to 3c/3t = -v 3c/9z.
This assumption implies that transport due to dispersion in
partially saturated soils is negligible compared to the con-
vective transport which occurs. This is generally a good
assumption.
The model computes the moisture flow (v) and couples
the flow with the chemical changes 3c/3z computed in the
biological-chemical program to give the salt transport. This
technique is the basis for the mixing cell concept.
The chemical exchange model computes the equilibrium
chemistry concentrations for calcium, magnesium, gypsum,
sodium, bicarbonates, carbonates, chlorides, and sulfates. The
nitrogen chemistry including ammonium, nitrates, and urea-
nitrogen uses a kinetic instead of an equilibrium approach. The
kinetic approach is needed since microbial activity involved in
nitrogen transformation occurs over a period of weeks and days
instead of minutes and seconds. The equilibrium chemistry for
inorganic salts is a good approximation since the reactions de-
scribing their chemistry occur in a matter of minutes or seconds
in a flow regime which is changing very slowly. A block diagram
of the biological-chemical model is given in Figure 5. A block
diagram illustrating the chemical reactions that are considered
in one of the subroutines is shown in Figure 6.
/''START BIOLOGICAL-^
^CHEMICAL PROGRAM^/
i
PROGRAM MAIN
READ CONTROL AND INPUT DATA
STORE INITIAL SOIL-CHEM DATA
PRINT CONTROL AND INPUT DATA
(OPTIONAL)
1
SUBROUTINE EXECUTE
MAKE ANY FERTILIZER AND/OR
ORGANIC MATTER APPLICATIONS
INITIALIZE OR UPDATE SOIL
TEMPERATURES (WEEKLY)
READ MOISTURE FLOW DATA FROM
MAGNETIC TAPE
1
SUBROUTINE COMBINE
FOR EACH SEGMENT:
CALL EXCHANGE SUBROUTINE
CALL NITROGEN SUBROUTINE
CALL SOLUTE REDISTRIBUTION
SUBROUTINE
CALL PLANT-N UPTAKE SUBROUTINE
SUM CHEMISTRY CHANGES AND
UPDATE VALUES IN STORAGE
PRINT OR WRITE SPECIFIED VALUES
\
P
(STOP BIOLOGICAL-"\
VCHEMICAL PROGRAM/
Figure 5: Generalized Block Diagram of Biological-Chemical
Program
Once the necessary field data have been collected, equations
can be developed to predict the variation in chemical quality
(including ionic constituents) of the moisture movement
through the soil profile, as well as the salt pickup (or salt precipi-
tation) resulting from movement of subsurface irrigation return
flows. These results, when combined with the hydro-salinity
model, will allow an evaluation of various salinity control mea-
sures upon salinity reaching the groundwater and returning to
the river.
-------
Soil-Salt Chemistry
119
( Enter
Vrp
Calculate
Specified
\
Call
EQEXCH (First Time)
1
CaC03 Solubility Constant at
Moisture Content
Consider
Consider
CaS04 st
Consider
2Na*+Ca
Solubility Reaction CaS04 x2H20 —
Undissociated
the Exchange
- R ^ Ca* * +
Ion Pair Reaction
Reaction
I
Consider
Mg** + Co
Consider
NH* + Na
the Exchange
-R — Ca** + K
the Exchange
Consider
MgS04 =
Consider
Undissociated
the Solubility
Reaction
Reaction
Ion Pair Reaction
.
Reaction
(
Return to Combine if Equilibrium
Constraints Satisfied
~\
J
Figure 6: Generalized Block Diagram of Subroutine Exchange
Cost- Effectiveness of Technological Solutions
The results from the hydro-salinity model will provide con-
siderable insight as to which technologies might be most appro-
priate in order to reduce subsurface return flows (since salinity is
a problem associated with subsurface return flows). The goal in
reducing subsurface return flows might be: (a) to lower ground-
water levels in order to alleviate waterlogging, thereby allowing
the leaching of salts in the root zone, which in turn facilitates
increased crop production; (b) reducing downstream water
quality degradation resulting from salt pickup; (c) improved on-
farm water management to increase crop production; or (d) any
combination of the above three goals.
Field studies must be undertaken to evaluate the effectiveness
of any particular technology, or combination of technologies, in
achieving the desired goal(s). An example will be used to illus-
trate the development of cost-effectiveness functions. If seepage
loss measurements are made along many sections of each canal
in an irrigated area, there will usually be a considerable variation
in seepage loss rates. If the cost of canal lining is compared with
seepage loss reduction, then the greater benefits accrue by first
lining those sections of canal which have the least cost per unit of
seepage loss reduction. Finally, the last sections of canal lining
may be quite costly per unit of seepage loss reduction. This con-
cept is illustrated in Figure 7a. The maximum seepage loss
reduction shown on the ordinate in Figure 7a represents the
total seepage loss from all canals; but, since canal linings still
have some seepage losses, there will still be some seepage losses
after lining all canals. The lowering of groundwater levels result-
ing from canal lining is illustrated in Figure 7b. In order to com-
pute the downstream salt load reduction resulting from various
levels of seepage loss reduction, the chemical-biological model
described in the previous section must be utilized. This same
model can be utilized to determine the reduction in ionic constit-
uents making up the total salt load reduction, thereby allowing
the prediction of downstream ionic chemical quality. The anal-
ysis could be carried forward another step if data are available
regarding the effect of groundwater levels on production of var-
ious crops.
Note.
Max. Seepage Loss Reduction is
the Total Seepage Losses from Canals
Max
100
Percentage of Total Canal Length Lined
Percentage of Total Canal Length Lined
100
Figure 7: Illustrative Example of Development of Cost-
Effectiveness Function for Canal Lining
The above example illustrates one of the simpler technologies
for developing a cost-effectiveness function. Lateral lining
becomes somewhat more complex because of the combined
effects of seepage loss reduction and providing water control,
which may be highly effective in improving irrigation applica-
tion efficiencies, which in turn reduces deep percolation losses
and increases crop production. The addition of flow measuring
devices at all junctions in laterals, as well as at farm inlets, can
prove highly effective in achieving improved water manage-
ment. The development of improved on-farm water manage-
ment technologies can require considerable field data collection.
One of the most appropriate on-farm technologies is "tuning
up" the present irrigation practices.
Cost-effectiveness functions should be developed for each ap-
propriate technology that can be identified. Then, cost-
effectiveness functions should be developed for each combina-
tion of appropriate technologies. These cost-effectiveness
functions, which relate subsurface return flow reductions to the
desired goal(s) resulting from a specified investment, would then
be assessed in an optimizational format to arrive at the least cost
combination for achieving a desired level of the stated goal(s).
Thus, this analysis details the optimal strategy for implementing
various levels of individual technological improvements into a
comprehensive technology package. An example of such an
analysis considering only the lining of farm head ditches, lining
laterals, and lining specific canals in Grand Valley is shown in
Figure 8."
-------
120
Soil-Salt Chemistry
5.Or
4.0
E 3.0
o
o
2.0
o
o
1.0
Grand Valley Canal and
Grand Valley Mainline
Orchard Mesa#l,
Independent Ranchmen s
Grand Valley Highlme
Farm Head Ditch
Lateral System
Kiefer Extension and Orchard Mesa
Price, Stub, and Mesa County Ditches
Redland
Conveyance Late
0 1.0 2.0 3.0
Conveyance Channel Lining Salinity Control, metric ton x|0~B
Figure 8: Optimal Conveyance Channel Lining Policy for Irri-
gation System in the Grand Valley, Colorado
CONCLUSIONS
Much emphasis will be given in the future towards improved
irrigation water management to reduce downstream water qual-
ity degradation and minimize diversions in order to provide
water supplies for new demands. In most cases, the key will be
improved on-farm water management. However, technology
alone will not usually bring about the necessary improvements.
A combination of technological changes and institutional modi-
fications will usually be required to effectively alleviate salinity
problems resulting from irrigated agriculture.
REFERENCES
I. Ayars, J. E. 1976. Salt Transport in Irrigated Soils.
Unpublished Ph.D. Dissertation, Agricultural Engineering
Department, Colorado State University, Fort Collins, Colo-
rado. August.
2. Dutt, G. R., M. J. Shaffer, and W. J. Moore. 1972. Com-
puter Simulation Model of Dynamic Bio-Physiocochemical
Processes in Soils. Technical Bulletin 196, Department of Soils,
Water and Engineering, Agricultural Experiment Station, Uni-
versity of Arizona, Tucson! October.
1. Walker, W. R. 1970. Hydro-salinity model of the Grand
Valley. M.S. Thesis CET-71WRW8. Civil Engineering Depart-
ment, College of Engineering, Colorado State University, Fort
Collins, Colorado. August.
4. Walker, W. R. 1976. Integrating Desalination and Agri-
cultural Salinity Control Technologies. Paper presented at
International Conference on Managing Saline Water for Irriga-
tion, Texas Tech. University, Lubbock, Texas, August 16-20.
-------
Tracing Leachate from Landfills
A Conceptual Approach
Hugh Mullen and Steven I. Taub
IU Conversion Systems, Inc.
Philadelphia, Pennsylvania
INTRODUCTION
Of the 60 million tons of fly ash produced annually in the
United States, only about 8 percent is utilized. The other 55 mil-
lion tons are discarded; 80 percent of the discarded material is
being sluiced. Add to that the SO2 scrubber sludge that is being
and is expected to be generated and we are confronted with one
of our biggest residual disposal problems. All of this material
will be deposited in landfills, either by sluicing or dry placement.
The environmental impact of such landfills is, therefore, a major
concern. How to determine the potential impact of such residue
before it is placed in a landfill is critical. A great deal of work is
being done in this area and the purpose of this paper is to point
out how some of the data that are being developed can be mis-
leading and to offer a concept that we believe can better deter-
mine the true impact of landfilled residues.
There is nothing very profound or unique about the concept,
but we offer it here because we find that some studies of stabil-
ized materials are only partial analyses which lack detailed inter-
pretation and explanation. The results are, therefore, mislead-
ing. Simply stated, the concept is to determine:
1. The physical characteristics of the material to be tested.
2. The method of placement and the character of the land-
fill.
3. The principal mechanisms for water movement through
and around the material.
4. The laboratory tests that will best demonstrate the total
mass transport of pollutants from the landfill to the
water system.
For illustration, we will compare an unstabilized, sluiced SO 2
sludge and our stabilized material, Poz-O-Tec®, which we con-
sider to be at the opposite ends of the spectrum. We acknowl-
edge that there are many variations in between and that no two
materials are identical but our purpose is to demonstrate princ-
ples, not to make specific comparisons. We have selected Poz-O-
Tec because we have a greater knowledge of our own material
and can discuss it more freely. However, we believe the concept
is generally applicable and should be used to evaluate all land-
filled materials.
The Poz-O-Tec Process
The Poz-O-Tec process, as shown in the diagram on Figure 1,
is a method of stabilizing SO2 sludge, fly ash and bottom ash
with lime and other additives. The product is environmentally
acceptable and its engineering properties can be controlled so
that it is highly suitable for landfill, road embankments, mine
reclamation, impermeable liners and other similar uses.
The complex pozzolanic reactions of the sludge, fly ash, lime
and other additives produce cementitious materials which give
Poz-O-Tec its exceptional physical properties and low permea-
bility. These, coupled with precise moisture control, proper
placement, compaction and good engineering practices result in
a landfill with superior structural integrity which will virtually
eliminate all of the environmental problems of the original
materials: ash and sludge. However, it is not the purpose here to
review the benefits of Poz-O-Tec or the polluting properties of
ash and sludge, other papers have covered those subjects thor-
oughly. Rather, our purpose is to discuss how a completed
landfill can and should be evaluated before the fact. How do you
determine before placing 5 to 10 million tons of material in a
landfill that it will not cause an environmental problem?
Part of our job at Conversion Systems, as suppliers of waste
management systems, is to demonstrate to our customers and
regulatory agencies that the stabilized material we produce,
Poz-O-Tec, is environmentally acceptable, structurally sound
and will be of value to them in the future. As in every business,
we find part of this job simple and part complex and difficult.
The easy part is where there are long established and accepted
test methods to demonstrate specific properties. The difficult
part is where:
1. There are no standards or established test procedures.
2. A multitude of people, including regulatory agencies
and consultants, are testing materials, each with his own
test method for determining a specific property without
relating it to the final disposal method.
3. Comparisons are made between specific properties
without regard to or recognition of the different nature
of the materials.
To illustrate this problem, let us make a hypothetical compar-
ison between our stabilized material and unstabilized sludge
using the test procedures that we use regularly in our laboratory.
A description of these procedures will be given later.
It is conceivable, depending on when the tests are run, that the
results could be as follows:
Hypothetical Comparison
permeability
leachate
runoff
Sludge
1 x 10-5
3000 ppm
1000 ppm
Poz-O-Tec
1 x 10-5*
3000 ppm
1000 ppm
In the face of such results, stabilization is a waste of time and
money.
Those results say that the two materials are environmentally
identical. But that conclusion is absolutely erroneous because
121
-------
122 Tracing Leachate
the data are incomplete and not related to the physical proper-
ties of the materials. A look at some of the properties of the two
materials will indicate why the comparison is fallacious.
" Normal range: 5 * 10-6 to 1 *
10-
Sludge
1. Settled solids from slurry
2. Saturated when placed
3. Continuous recharge
4. Permeates immediately and
continuously
5. Thixotropic
6. Unconfmed compressive
strength 0
7. Completed fill useless
Poz-O-Tec
Placed dry
No surface water
Requires years to saturate,
probably never
Will not permeate until
saturated
Monolithic mass
Unconfined compressive
strength—50 - 1000 psi
Completed fill-
valuable land
Obviously, the two materials will act differently in natural en-
vironment and the identical results from the initial laboratory
tests require explanation and interpretation. The in vitro and in
vivo conditions vary drastically and it is the latter with which we
are concerned. A valid test program should be designed to eval-
uate actual field conditions and not for laboratory convenience.
Physical Properties
The demonstration of physical properties of a material, the
first requisite of our concept, is straightforward and there are
well estabished test methods to determine:
1. Unconfined compressive strength
2. California Bearing Ratio
3. Triaxial shear strength
4. Consolidation
5. Shear strength
6. Slope stability
There is no point in a detailed discussion of these test methods
since they are readily available and we find them completely
acceptable. We are, therefore, only presenting here comparative
results for three different materials.
Unconfined Com-
pressive Strength
California Bearing
Ratio
Consolidation
Sluiced Stabilized
fly ash sludge ash/sludge
0
0
High
0
0
High
50-1000psi
100
Incompressible
The physical and environmental properties of the Poz-O-Tec
material improve with time as the pozzolanic reaction proceeds.
PRIMARY CONVERSION PROCESS
FLY ASH
(WITH OR WITHOUT
BOTTOM ASH)
DRY SOURCE OR
DEWATERED SLURRY
IddJLtivU
SOX SLUDGE
DEWATERED
- PLASTIC
CONSISTENCY
OPTIONAL
PROCESS
BULK DISPOSAL
EMBANKMENTS
STABLE LANDFILL
STRIP MINE AND
QUARRY RECLAMATION
DEEP MINE PROTECTION
ETC
P40CC44
RECYCLE
WATER
PRODUCT DISPOSAL
DISPOSAL LINERS
ROAD BASE
SYNTHETIC AGGREGATES
STRUCTURAL SHAPES
ETC
Figure 1: Schematic Diagram of IU Conversion System's Waste Conversion Process
-------
Tracing Leachate
123
The cementitious nature of the reactions produces a landfill that
is monolithic in nature, of low permeability and is subject to
leaching from the external surface only. All of the internal sur-
faces and, therefore, the mass of the potential pollutants are
unavailable. Sinc.e it is the total mass transport of soluble salts
that is of environmental concern and not the concentrations in
initial leachate, our test procedures are designed to evaluate that
long term mass transport.
Placement
The method of placement, our second requisite, is of extreme
importance because it establishes some of the critical properties
and conditions of the material and the nature of the completed
landfill. A sluiced material, because of its nature, must be
ponded. Since it is saturated, all of the internal surfaces are
exposed and all of the soluble salts are available for leaching.
The leachate will, therefore, always be at the solubility limit or
maximum concentration. Untreated sludge will leach indefi-
nitely. The duration of leaching from treated material will
depend on the type of treatment. If an impermeable liner is used,
you have, in effect, created a perched water table that must some
day overflow and is exposed to the hazard of a ruptured liner. If
not an immediate problem, it is a permanent potential for pollu-
tion.
The characteristics of a material that is placed dry will be
entirely different and we cannot use the same parameters to eval-
uate both.
Mechanisms for Water Movement
Our third requisite is to determine which mechanisms of water
movement, considering the type of material and method of
placement, are likely to result in sufficient flow to cause a prob-
lem. There are six categories of water movement that theo-
retically could occur. They are hygrometric, capillary, thermal,
freeze/thaw pumping, osmotic, gravitational and runoff.
Hygrometric Flow:
Hygrometric flow releases water vapor to the atmosphere as
the result of differences in relative humidity. Since only distilled
water is produced there is no transport of pollutants.
Capillary Flow:
Capillary action could cause flow of water from Poz-O-Tec to
the ground if the ground is unsaturated and contains pores
smaller than those of the landfilled material. That, generally
means that the permeability of the soil must be below 1 x 10-7. If
so, the soil is impermeable and ground water pollution is
unlikely.
Thermal Flow:
Thermal flow is dependent on a temperature gradient with
vapor flow from the higher to lower temperature area; a condi-
tion that is unlikely to have any significant effect on a landfill.
Freeze/ Thaw Pumping:
Freeze/thaw pumping occurs when water in the material
expands as it freezes forcing it deeper into the material. When
thawing occurs, the surface area is recharged setting the stage for
a repetition of the cycle. This condition is easily controlled by
proper cover on the completed fill area and expected to have lit-
tle or no effect on the transport of salts from the fill.
Osmotic Flow:
Osmotic flow occurs when solutions of different concentra-
tions come into contact with each other resulting in the flow of
ions from the higher to lower concentration by diffusion. If
ground water does contact the bottom of the fill material such
flow will occur.
Gravitational Flow:
Gravitational flow or movement of water due to its own
weight is considered to obey Darcy's law. As applied to soils, it
quantifies flow and establishes permeability coefficients.
Whether it can be applied directly, at hydraulic gradients close
to one, is a matter of contention. Work done by the Russians
indicates that at a permeability of 1 x 10-7, a gradient of 20 is
required to initiate flow. However, that is not a matter for dis-
cussion here.
For a sluiced sludge of relatively high permeability and con-
stant recharge, the overriding mechanism is gravitational flow;
assuming, of course, that there is no runoff or discharge and the
supernatant is returned for transport of fresh material. The per-
meate will be at the saturation limits and the maximum concen-
trations of pollutants will be carried to the ground water.
For a material such as Poz-O-Tec, we must consider three
conditions that could result in pollution of the water system:
permeation, surface runoff and ground water contact.
Permeability:
Permeability by definition can only be measured when there is
flow. To have flow, the material must be saturated in some
areas, i.e., some of the channels through the pores of the mate-
rial must be completely filled with water so that pressure exerted
at the top by recharge will result in flow from the bottom. In the
laboratory, the sample is force saturated and the falling head
permeability test is run at a hydraulic gradient of about 10.
Those conditions bear no resemblance to what takes place in a
commercial landfill. A hydraulic gradient of 10 on a 10 ft. deep
landfill requires 90 feet of standing water, a wholly unlikely
situation, in the field, unless you are creating a lake and not a
landfill. Our landfills are designed for rapid runoff and imme-
diate dissipation of surface water. Recharge, therefore, only
occurs during a rain.
Accepting for the moment that the coefficient of permeability
measured at a hydraulic gradient of ten is valid at a gradient of
one, a simple calculation will indicate the time required to satu-
rate a 10 ft. deep landfill.
Assume:
1. 30 inches of rain/yr.
2. No drying or evaporation between rains
3. Permeability of 1 x 10-6 cm/sec.
4. Average rain I/4"/24 hrs.
.000001 cm/sec x 60 min x 24 hr x 120 dys = 10.37 cm/yr
or 4.1 inches/yr
or 30 years to saturate
This time will be greatly extended due to loss by evaporation. It
would, therefore, appear that we can eliminate permeation as a
problem if the permeability is 1 x 10-6 or lower.
Thus, we are left with two mechanisms for the transport of
pollutants from Poz-O-Tec; runoff and ground water contact or
surface washing and diffusion. We have done a great deal of
work to quantify the effects of these phenomena on Poz-O-Tec
material and the results demonstrate that it has minimum pol-
luting potential.
Runoff
Surface runoff water can be contaminated by dissolution of
the surface salts and diffusion of the subsurface salts to the bulk
solution, if there is sufficient contact time. In a properly
designed landfill, contact time will be extremely short and diffu-
sion is, therefore, of minor consequence. Hence, it is not consid-
ered in our runoff analyses.
Two different test procedures were utilized to determine the
quality of runoff from Poz-O-Tec and the effects of repeated
washings. Those tests established that:
-------
124
Tracing Leachate
1. Surface salts wash off rapidly
2. Succeeding runoff contains substantially reduced total
dissolved solids
3. The average weight of surface salts available is .05 gr/sq
in.
The first series consisted of successive 48 hour shake tests in
which a sample of known weight and superficial surface area is
placed in a container of distilled water at a ratio of 1 gram of
material to 4 milliliters of water (typically 500 gr. to 2000 ml).
The container is oscillated (60-70 one inch strikes per hour) for
48 hours. The water is then filtered and analyzed. The results of
such tests of Poz-O-Tec from the Shawnee test pond are shown
in Table I.
Table I
POZ-O-TEC LEACHATE
FROM
SUCCESSIVE SHAKE TESTS
TDS GRAMS GRAMS/
(PPM) LEACHED IN.2
974 1.918 .046
338 .676 .015
268 .536 .012
194 .388 .009
214 .428 .010
SURFACE AREA: 42.4 IN2
DILUTION RATIO (iN2/L): 21.2 : 1
As can be seen, after 3 or 4 washings, the TDS reaches an
equilibrium of approximately 200 ppm. That equilibrium rep-
resents the diffusion controlled leaching process of soluble salts
encapsulated below the immediate surface. Thus, diffusion is
significant only when water is in contact with the material for
extended periods. The washing and diffusion controlled phases
are indicated on the curve in Figure 2. The effect of dilution on
leachate concentration and mass transport is shown on Table II.
As demonstrated by these data, an average surface loading of
.05 gr/in2 is available for leaching. Applying this to a one acre
site, 634 Ibs. of teachable material are available for the first
washing. For a one inch rain, approximately 208,000 Ibs. of
water will contact the one acre surface. That would produce a
maximum loading of 3000 ppm TDS in the runoff, assuming a
48 hour contact. Subsequent rainfalls will produce lower TDS,
with the third approaching 100 ppm for stabilized high sulfite
and 500 ppm or less for stabilized high sulfate sludges.
These shake test results have been substantiated by data
derived from a runoff test that we have developed. This test is
flexible enough to simulate any amount of rainfall over any time
period and any length of flow. As shown in Figure 3, the appara-
tus consists of a box 1 ft. long x 1 ft. deep x '/2 ft. wide. Clean,
inert material of high permeability is placed in the bottom and
the material to be tested is placed on top and compacted as in the
field. Water is sprayed on top of the material to simulate any
desired rainfall, usually a one inch rain, once a week. The box
can be tilted to any angle to conform with the expected slope of
the landfill. The runoff is collected for analysis and there are
drains in the bottom to remove any permeate for testing. The
data that can be gathered by use of this procedure are:
1. Runoff analysis
2. Amount and quality of permeate
3. By weighing, the amount of water absorbed and the loss
through evaporation between tests.
4. By introducing water into the bottom, the effect of
ground water contact.
The test accurately simulates field conditions and produces a
wealth of data. Table III shows some of the comparative results
obtained.
Table II
SURFACE
34.77
31.77
31.77
31.77
60,78
POZ-O-TEC
EFFECT
DILUTION
RATIO i.N2/L
307,6
171
84,6
58,2
30.39
OF DILUTION
TDS
PPM
4812
3536
34.18
3282
1932
GRAMS
LEACHED
0,547
0,778
1.442
1,959
3.864
GRAM/ IN2
,0157
.0288
.0415
.0563
.0636
FROM SHAWNEE-AEROSPACE POND
TDS
(ppm)
Solid Piece
Broken Piece
^-T:
\ SHORT TERM
DIFFUSIOK EFFECT
O
•
WASH NO. 2
Figure 2: Poz-O-Tec from Aerospace Shawnee Pond (Leachate
TDS vs Wash No.)
-------
Tracing Leachate
125
Figure 3A: Leachate Run-off Test Box
|||» | | | i
Figure 3B: Ground Water Simulation Test Box
Table III
RUNOFF RESULTS
STABILIZED
FILTER CAKE
IMMEDIATE 2658
7 DAYS 581
28 DAYS 2138
SULFATE SLUDGE
FILTER CAKE
AND FLY ASH
2132
2296
1800
POZ-O-TEC
2240
588
139
Ground Water Contact
Before a landfill is started, the site is carefully selected, the
hydrology of the area, (height of water table surface flow, etc.) is
determined. The site is selected to prevent ground water contact
and must be acceptable to the local authorities. Where neces-
sary, a bottom drainage system, consisting of French drains or
piping, is installed. In spite of all of that, we must answer the
what if question. What if ground water does contact the mate-
rial?
It is our opinion that for a material of low permeability, such
as Poz-O-Tec, the only water movement mechanism that can
transport pollutants from the bottom surface to the ground
water is diffusion. We have, therefore, performed numerous dif-
fusion tests to determine the effects.
Because ground water moves very slowly and the flow is
laminar and not turbulent, we have discounted any mixing effect
the flow might have. The tests were, therefore, performed in
static water. Specimens were immersed in deionized water and
TDS was measured as a function of time and distance. Assum-
ing a maximum TDS of 5000 ppm, the solubility limit, as deter-
mined from the shake tests, the diffusion coefficient was calcu-
lated to be 2 x 10-5 cm2/ sec. Water must, therefore, be in contact
with Poz-O-Tec for extremely long periods of time before
enough salts are transported to the ground water to be of con-
cern. This theoretical diffusion coefficient was substantiated by
plotting it against long term data obtained from the supernatant
on the Poz-O-Tec pond at Shawnee. The results are shown on
curves in Figures 4 and 5.
SUMMARY
We believe that it is the mass transport of polluting salts to the
water system that is of concern. The magnitude of that problem
cannot be determined by the initial leachate or any test of a sin-
gle property without regard for the physical characteristics of
the material, the manner of placement and the pertinent water
movement mechanisms. The release of partial data, without ex-
planation or interpretation from programs that are limited in
scope, is misleading and a disservice to the industry. An addi-
tional problem is the time frame. Some on-going programs will
not be completed for one to four years. The problem is now, not
four years from now. We need a fast and effective method of
evaluating residues to be landfilled. We believe that our concept,
coupled with the test procedures we have outlined, offers that
capability.
We have spent a great deal of time and effort to develop our
test program. We work with it constantly and continuously. It is
important to us, because sludge stabilization is our livelihood.
We offer our procedures to anyone who is interested and we are
receptive to any that can be shown to be better.
With our Poz-O-Tec landfill material and procedures, we
have greatly reduced the potential for the transport of pollutants
to the water by:
1. Increasing the pH to reduce solubility
2. Placing a dry, unsaturated material.
3. Isolating the internal surfaces by making the material
effectively impermeable.
4. Protecting the external surfaces through rapid runoff
and good bottom drainage.
It must be remembered that we are talking about landfill oper-
ations of 500,000 to 1,000,000 tons per year, and extremes in dis-
posal methods from sluicing untreated sludge and ash to unlined
ponds to placing stabilized material in a well-engineered and
well-managed landfill. There are many variations in between but
proper evaluation shows the Poz-O-Tec system to be an effec-
tive, economically viable means of protecting the environment.
Our test procedures enable us to develop all of the data neces-
sary to design the most environmentally effective landfill.
-------
126 Tracing Leachate
TDS
(ppm)
5000
4000 -
3000 -
2000
1000
0
Figure 4
TDS
(ppm)
5000 •
4000 •
Diffusivity
(D=2 x 10~ cm /sec.)
__ O —
369
ELAPSED TIME IN MONTHS
Supernatant Aerospace Pond—Shawnee (Predicted TDS)
3000
2000
1000
0
12 15
Shawnee Supernatant
Calculated
(D=2 x 10~5cm2/sec.)
T
369
ELAPSED TIME IN MONTHS
Figure 5: Supernatant Aerospace Pond—Shawnee (TDS/Diffusion vs. Predicted TDS)
12
15
-------
Transport of Salts from
Disturbed Geologic Formations
David B. McWhorter and Jerry W. Rowe
Department of Agricultural Engineering
Colorado State University
Fort Collins, Colorado
INTRODUCTION
The observed chemical composition of natural runoff is a
reflection of combined chemical and hydrologic processes that
are highly dependent upon the distribution and physical- >
chemical characteristics of the geologic material in the
watershed. Dissolved solids in watershed drainage result from
dissolution of minerals contacted by the water and depend upon
both the distribution of soluble minerals and the opportunity for
contact of these minerals by precipitation as it moves through
the watershed. Therefore, large scale disturbance' of geologic
strata, particularly those within about 50 meters of the ground
surface, provides the potential for significantly modifying the
water-quality hydrology of the watershed.
This paper is a discussion of factors effecting the concentra-
tion of inorganic dissolved solids in runoff from watersheds that
have been partially disturbed by surface mining. Hydrologic fac-
tors, including surface runoff, runoff by interflow, and ground
water runoff are combined with the corresponding concentra-
tions of dissolved solids to provide a quantitative description of
observed water quality in an area disturbed by surface mining.
The basis for the material in this paper is the observation of the
water-quality hydrology on a surface-mined area in northwest-
ern Colorado.
Study Area and Monitoring Program
The study area is a surface-mined watershed located between
the towns of Steamboat Springs and Oak Creek, Colorado. The
coal mining operation is near the southeast limit of the Yampa
coal field. The disturbed strata are part of the Williams Fork
member of the Mesa Verde group, a Late Gretaceous sequence
of thin sandstones, shales and sandy shales.' The portion of the
watershed investigated lies above 7000 feet in elevation and
extends to elevations in excess of 8000 feet. Annual precipitation
exceeds 20 inches, about half of which occurs as snow. Snow-
melt begins in late March and extends into May on the study
area. The mined area is located on a relatively uniform slope that
dips to the west and northwest (Figure 1). All drainage from the
mined area is received by Trout Creek, a perennial stream that
flows in a generally northward direction at the base of the mined
slope.
The study area was subdivided into four watersheds desig-
nated as C 3, C 5, C 9, and CIO, the boundaries of which are
shown by the heavy lines in Figure I. Each of these watersheds is
drained by a perennial stream tributary to Trout Creek. The
shaded areas in Figure 1 represent the disturbed Williams Fork
rocks.
A total of thirteen water-quality monitoring stations were
established. The location and designation of the most important
of these stations are indicated by the black dots in Figure 1. Sev-
eral of these stations are also gauging stations for the determina-
tion of discharge. A description of the monitoring stations is
contained in Table I. The two discharge stations (C 2 and C 6) on
Trout Creek consisted of rated cross-sections on which the stage
was monitored. The stage was monitored at C 2 with an auto-
matic stage'recorder, and by reading a staff gauge at C 6. All
other discharge stations were equipped with flumes, three of
which (C3, C9, and CIO) we're also equipped with stage re-
corders.
kilometers
Figure 1: Map of the study area showing monitoring stations,
subdivision into watersheds, and disturbed ground.
127
-------
128 Transport of Salts
Table I: Description of Monitoring Stations
Station
No.
Cl
C2
C3
C4
C5
C6
C7
C8
C9
CIO
Cll
C12
C13
Type
Discharge Qua
X
X
X
X
X
X
X
lity
X
X
X
X
X
X
X
X
X
X
X
X
X
Description
One mile above mine on Trout Creek
On Trout Creek immediately above mine
On mine drainage (interflow & surface water)
On Trout Creek
On mine drainage (interflow & surface water)
On Trout Creek near downstream limit of mine
Ground water seep
On Trout Creek near downstream limit of mine
On mine drainage (interflow & surface water)
On mine drainage (interflow & surface water)
On Trout Creek below downstream limit of mine
On ephemeral drainage from undisturbed area
On Little Trout Creek above mine
Grab samples for the determination of water quality were col-
lected periodically at each of the stations. The frequency of sam-
pling varied from station to station and with the time of year.
During the winter months, late summer, and fall, monthly sam-
ples were collected. The interval between samples at selected sta-
tions was decreased to one or two days during peak runoff.
Water samples on which complete chemical analyses were to be
performed were handled in accordance with USEPA Standard
Methods2. On many of the samples temperature, pH, and spe-
cific conductance were the only parameters measured, and these
measurements were made directly in the field.
Evidence of Salt Pickup
Graphic evidence of salt pickup in the portion of the
watershed contributing between stations C2 and C6 on Trout
Creek is shown in Figure 2. Monthly averaged (discharged
weighted) concentrations of dissolved solids at the downstream
station, C6, are substantially greater than at the upstream
Figure 2: Dissolved solids concentration at C2 and C6 on Trout
Creek.
Table II: Monthly Discharge1, Salt Load2, and Concentration3 By Station For 197S
Month
Jan
Feb
March
April
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
TOTALS
C2
OOP
ds t
75.9 83 109
71.0 62 87
75.9 91 120
118.0 137 116
479.9 481 100
1434.5 876 61
517.9 367 71
126.0 148 117
86.5 89 103
86.5 84 97
125.9 128 102
98.3 118 120
3296.3 2664 69
'Q volume of water (rrv^xlQ
2V
3pt
total weight (kgxlO'J
average concentration
C3
Q Q P
ds t
0.444 6.4 1441
0.494 7.9 1599
0.494 10.0 2024
3.555 57.4 1615
11.851 233.5 1970
4.481 86.5 1930
1.284 23.2 1807
0.506 9.9 1957
0.296 5.9 1993
0.370 7.3 1973
0.444 9.3 2095
0.444 9.3 2095
24.664 466.4 1891
~4) discharged during
Q
136.6
127.7
136.6
204.6
590.1
1682.4
635.3
192.8
153.4
161.3
137.7
153.4
4311.8
C6
Qds
273
160
259
1184
2061
1888
689
345
249
282
298
265
7953
Pt
200
125
190
579
349
112
108
179
162
175
216
173
184
Q
0.494
0.494
0.309
3.987
7.962
3.074
1.901
1.358
0.506
0.704
0.704
0.704
22.195
C9
Qds
3.1
3,1
4.3
62.6
104.3
38.5
24.9
18.5
5.6
8.0
6.4
6.4
285.5
Pt
628
628
1392
1570
1310
1252
1310
1362
1107
1136
909
909
1286
Q
0.148
0.148
0.494
5.899
9.740
4.518
0.802
0.296
0.185
0.160
0.148
0.148
22.788
CIO
Qds
2.7
2.7
11.0
145.2
176.1
57.2
12.5
5.9
3.2
2.8
2.7
2.7
424.7
pt
1824
1824
2227
2461
1808
1266
1559
1993
1730
1750
1824
1824
1864
the month.
) of salt contained in Q.
of dissolved sol ids
(mg/i).
-------
Transport of Salts
129
station, C2. All of the indicated accumulation of dissolved sol-
ids is not attributable to the mined area, however, because the
undisturbed portions of the watershed also make a contribution.
Estimates of the contribution attributable to mining are dis-
cussed subsequently.
Monthly total salt loads, discharges, and dissolved solids con-
centrations for several stations during 1975 are presented in
Table 11. The peak concentration at C6 on Trout Creek occurs in
April, but the greatest salt loading from C3, C5, C9, and CIO
occurs in May. This observation is explained by the time distri-
butions of runoff contributing at station C6. More specifically,
the April discharges at stations C3, C5, C9, and CIO are approx-
imately 10 times greater than the corresponding March
discharges, indicating a very significant snowmelt on the dis-
turbed watershed. On the other hand, the discharge at C2 in
April is only 55 percent greater than in March, indicating a
much less significant snowmelt on the high watershed above C2.
Thus, the April concentration at C6 reflects the relatively high
salt concentrations in the inflow from the C3, C5, C9, and CIO
watersheds. In May, however, the discharge at C2 increased by a
factor of four, and this influx of water with low concentration
was sufficient to reduce the concentration at C6 in spite of more
saline inflow from C3, C5, C9, and CIO.
The total salt pickup between stations C2 and C6 for 1975 is
5289 metric tons, the large fraction of which is contributed by
spring runoff. Similar observations were made in the spring of
1974 and 1976.
Factors Effecting Observed Water Quality
The water at monitoring stations C3, C5, C9, and CIO is a
combination of overland flow from both undisturbed and dis-
turbed ground and shallow ground water (interflow) that has
returned to the surface at various points along the water courses.
The observed water quality at these stations depends upon the
concentration of dissolved solids in each contributing compo-
nent and upon the volumetric contribution from each source.
The dependence of the observed water quality on the concentra-
tions and volumes of each component can be estimated by for-
mulating a combination of water and dissolved solids budgets.
This approach has been used successfully in hydrograph analy-
sis.3,4,5
The Water Budget
The water budget for a watershed can be expressed in terms of
the volumes in each budget component over a specified time
interval A t. For example.
(1)
Ve-Vs-Vg-Vd.
where
V_ = volume of precipitation in At
V = volume of evapotranspiration in At
C
Vs = volume of surface runoff in At
V = volume of shallow ground water runoff in At
Vd = volume of percolation to deep aquifers that do
not contribute to stream flow in the watershed
under consideration.
AS = the change in storage in the watershed in At.
Under conditions of dynamic equilibrium, the total inflow
equals the total outflow and A S = 0. This condition is approxi-
mated over a time interval of one year in some cases and will be
assumed to apply in the subsequent developments. It is further
assumed that deep percolation is zero. This is a reasonable
assumption in the study area because it is underlain by a
sequence of shales that are expected to have a small permeability
relative to that of the disturbed ground and natural soils.
Abundant evidence of a water table perched on the interface
between the disturbed material and the underlying shale exists in
the study area. For example, many interflow seeps were
observed where a mine cut or natural drainage intersected this
interface.
With AS and Vd equal to zero, Eq: 1 can be rearranged
so that
Vt = Vp- Ve = Vs + Vg (2)
where Vt is the total volume of watershed drainage. Equa-
tion 2 can be rewritten in terms of volumes per unit area
for the undisturbed area and the disturbed area. Putting
Am equal to the area of the disturbed ground and An equal
to the area of the natural ground, Eq. 2 becomes
Vt = qtAt = qmAm + qnAn =
(3)
where q represents volume per unit area, the subscripts m
and n r^fer to disturbed and natural, respectively, and the
subscripts p and e refer to precipitation and evapotranspira-
tion, respectively. Rearrangement of Eq. 3 yields
It = (
Where Fm = Am/At and 1 - Fm = An/At- In other words,
Fm is the fraction of the total watershed area that is mined
or otherwise disturbed.
A uniform distribution of precipitation is assumed so that
1p = 1pm = Ipn and> therefore,
Fm+
qen
Let fem = qem/qp and fen = qen/qp so that
The parameter fem represents the fraction of the precipita-
tion that is consumptively used on the disturbed land and
fen is the fraction of precipitation that is consumptively
used on the natural portion of the watershed. Differences
in the physical characteristics of soils, topography, and vege-
tation make it unlikely that fem is equal to fen.
Dissolved Solids Budget
The dissolved solids budget for a watershed partially dis-
turbed by mining can be written
where
(7)
Pf. = average TDS concentration in the total drain-
age from the watershed,
Pm = average TDS concentration in drainage from
the disturbed area,
-------
130 Transport of Salts
Pn = average IDS concentration in drainage from
the natural portion.
Equation 7 can be expressed in terms of volumes per unit
area to give
after considerable rearrangement. Let
1-f,
en
1-f,
and
em
1-F
R =
m
m
(13)
(8)
But since qm = qp - qme and qn = qp - qne, then
It
pt = Id - fem> Fmp
em mm
- Fm> PJ
Data at the study site, both under conditions of natural runoff
and runoff induced by rainfall simulators, indicate a large dis-
parity between the TDS in overland flow and in the shallow
ground water on the disturbed area. This is explained by the fact
that the volume of material contacted by overland flow is very
small relative to the volume of surface runoff. Therefore, the
surface layer of material is rapidly leached, and overland flow
contains a relatively small concentration of dissolved solids. On
the other hand, the volume of infiltrated water is quite small
compared to the volume of disturbed material through which it
must pass before reaching the streams. Therefore, leaching of
soluble materials from the bulk of the disturbed ground is a
much much less rapid process than in the surface layer, and
interflow contains a much greater concentration of dissolved
solids. To some degree, the surface runoff from undisturbed
areas is also less than the shallow ground water in the same area.
The difference is less pronounced, however, because leaching of
the bulk of the soil has been occurring for hundreds of years.
Because of the significant difference between the quality of
overland flow and interflow on the disturbed area, the variable
P in Eq. 9 is expressed as the sum of the contributions from sur-
face runoff and interflow:
p = f P
rm 1smrsm
gm>
(10)
wherein f is the fraction of the total runoff from the dis-
sm
turbed ground that is surface runoff (overland flow). Substitu-
tion of Eq. 10 into Eq. 9 yields the following expression for the
dissolved solids budget:
pt =
- f
ern) {
sm + d '
Pgm} Fm
D
d-fen)d-Fm)Pn].
Combined Water and TDS Budgets
Equation 6 from the water budget is substituted into Eq. 1 1 to
yield
Pt =
*smpsm
gm
1 +
-f
en
1-f,
ern
m
which permit Eq. 12 to be written as
impsm + d - fsm)' Pgm
1 +KR
(14)
Equation 14 is the final expression for the concentration of
dissolved solids in watershed drainage as affected by the vol-
umes and concentrations of overland flow and interflow con-
tributed by the disturbed and natural portions of the watershed.
The parameter R is simply the ratio of the undisturbed area to
the disturbed area. R is zero for a watershed that is completely
disturbed and becomes very large for watersheds that are min-
imally disturbed. The parameter K. is the ratio of the total drain-
age (including overland flow and interflow) on the natural por-
tion of the watershed to the total drainage on the disturbed
portion. K, therefore, is a hydrologic parameter. Likewise, the
parameter f , being the fraction of the total runoff from the dis-
turbed areasthat is overland flow, is a hydrologic parameter.
Application of the Budget Equation to Study Area
The application of the foregoing budget equation requires
determination of three average dissolved solids concentrations:
P , P , and P . The TDS concentration in overland flow on
the disfurbed ground, PS , was estimated from samples of natu-
ral runoff and from runoff on four experimental plots subjected
to simulated precipitation. PS for the natural runoff averaged
149 mg/jt(l5 kg/ha-cm) andfrom the experimental plots 158
mg/ J- (16 kg/ ha-cm). The value of Psm is, therefore, taken as 150
mg/ A (15 kg/ ha-cm) in the subsequent calculations.
The experimental plots were equipped with subsurface drains
at depths ranging from 1.5 m to 2.2 m below the surface of the
disturbed ground. The average TDS concentration in 57 sam-
ples of water taken from these drains is 3025 mg/ ,£(303 kg/ha-
cm). This average concentration agreed closely with the average
of 3126 mg/jjj from 11 water samples extracted from saturated
pastes prepared from the disturbed material.6,7,8 The concentra-
tion of dissolved solids in ground water seeps at station C5
also agreed well with the above values. Therefore, the concentra-
tion of dissolved solids in the ground water beneath the dis-
turbed ground was taken as 3025 mg/j£(303 kg/ha-cm).
The remaining water-quality parameter, Pn, is the TDS con-
centration in combined overland flow and interflow runoff from
the undisturbed watershed. Measurements of concentration in a
natural water course draining undisturbed land were used to
estimate Pn. During periods of base flow, the concentration
averaged 462 mg/ J^ (46 kg/ ha-cm) and substantially less when
overland flow was contributing. Thus, Pn was written as the sum
of a ground water and overland component. It was also assumed
that the fraction of total runoff that is overland flow was the
same as on the disturbed ground, an assumption that is not
required if Pn is measured directly. With the above assumption,
P is given by
1-f,
em
-f,
en
(12)
(15)
where Pn is in kg/ha-cm.
-------
Transport of Salts
131
Substitution of the values for Pn, Psm, and Pgm into
Eq. 14 yields
{46 fsm + 15(l-fsm)} KR + 15 fsm + 303(l-fsm)
1 +KR
(16)
The value of R for each watershed was determined by measuring
the disturbed area and the total area of each catchment above
the monitoring stations. The values for F and R are presented
in Table III.
TABLE III: R and Fm For Study Area
Watershed
C3
C5
C9
CIO
F
m
0.68
1.00
0.35
0.44
R
0.47
0.00
1.86
1.27
All parameters in the budget equation are now determined,
save the two hydrologic parameters f and K. It is expected
that the values for these hydrologic parameters are different for
each of the four watersheds, because the catchments differ in
respect to vegetation, topography, etc. However, it was not pos-
sible to measure fgm and K for each watershed, nor is it possible
to determine fsm and K by fitting Eq. 16 to the data for each
individual watershed. The latter statement is true because R is a
single value for a particular watershed, causing the situation to
be one of a single equation and two unknowns. On the other
hand, if it is assumed that K and fsm do not vary appreciably
from watershed to watershed, then average values of K and fsm
can be calculated by determining the best fit of Eq. 16 to the data
from all four watersheds.
The best fit of Eq. 16 to annual data from the four watersheds
is shown in Figure 3. The dashed line is the limiting value of
observed TDS concentration for an undisturbed watershed. All
of the data points for the C3 watershed fall below the curve. This
is probably explained by the fact that the disturbance of the
strata in C3 occurred some 30 years ago, providing sufficient
time for significant leaching to have occurred. All other distur-
bances are relatively recent. The values for K and f determined
by fitting Eq. 16 to the observed data are 1.04 and 0.06, respec-
tively. K equal to 1.04 implies that the total drainage from the
disturbed portions of the watershed is practically equal to that
from the natural portion. A value for fsm of 0.06 means that only
6 percent of the total drainage occurs as overland flow. It has, in
fact, been observed that surface runoff is very small. This results
from highly permeable materials in the study area, and the fact
that most of the water available for runoff is from melting snow
and not from high intensity rainfall.
Inspection of Eq. 14 reveals that the observed concentration
P should vary linearly with the fraction Fm of the watershed that
has been disturbed in the special case in which K = 1.0. Because
the average value of K as determined above is nearly unity, the
measured values of Pt should exhibit a nearly linear relationship
to F . A plot of Pt versus F^ for the study area is shown in Fig-
ure i". The straight line is a linear regression line. Extrapolation
of the regression line to Fm = 0 yields P( = 450 mg/ (45 kg/ ha-
cm), a value that should approximate the annual average TDS
concentration in watershed drainage prior to disturbance of the
area. It is possible, therefore, to estimate the increase in TDS
concentration due to mining. The results are shown in Table IV
for 1975.
s
isodt
I'D 2-0 30 to 5-0 BO 70 80
R
Figure 3: Best fit of the budget equation to observed data - K =
1.04, fm = 0.06.
200
100
.CIO
'C3
C9+Ci
0-2
0-4
0-6
0-8
Figure 4: Relationship between observed water quality and the
fraction of the watershed that is disturbed.
TABLE IV: Estimated Increase In Average TDS Concentration
Attributable to Mining—1975
,, . , , r Percent Increase
Watershed Fm -n JDS
C3
C5
C9
CIO
0.68
1.00
0.35
0.44
320
529
187
313
The percentages indicated in Table IV apply to the individual
watersheds. The percent increase in the TDS concentration in
the receiving waters (Trout Creek), attributable to mining, is
much less than that shown in Table IV.
CONCLUSIONS
Three years of monitoring the water-quality hydrology on
-------
132 Transport of Salts
four small watersheds, partially disturbed by surface mining
operations, have shown that shallow ground water runoff (inter-
flow) from the disturbed lands is the largest contributor to
observed dissolved solids concentration in watershed drainage
on the study area. Concentrations of dissolved solids in inter-
flow exceed those in overland flow by a factor of approximately
20. Highly permeable materials and the gradual release of water
by snowmelt, result in small surface runoff relative to infiltra-
tion. The variation of observed average annual TDS concentra-
tion from different watersheds is satisfactorily explained by a
simple algebraic model derived from material balance consid-
erations. Comparison of the model with annual data from the
four watersheds indicates that the percent of the catchment area
that is disturbed is the most significant parameter with respect to
accounting for the differences in TDS concentration observed
from one watershed to another. This was true in the study area,
only because the chemical characteristics of the disturbed mate-
rial were practically the same in all watersheds, with the possible
exception of the C3 watershed.
The most important factors influencing the observed water
quality in a particular watershed are the chemical characteristics
of the disturbed area and the distribution or routes by which per-
cipitation makes its way to the observation point. The material
balance model presented herein accounts for differences in the
quality of surface and ground water runoff, the division of total
drainage into overland flow and interflow on the disturbed and
natural portions of the watershed, and the chemical characteris-
tics of the two areas. The model should, therefore, be useful for
anticipating, in a quantitative manner, the effects of disturbed
areas on water quality prior to the occurrence of the distur-
bance.
REFERENCES
1. Bass, N. W., J. B. Eby, and M. R. Campbell, 1955, Geol-
ogy and Mineral Fuels of Parts of Routt and Moffat Counties,
Colorado, USGS Bui. 1027-D.
2. U.S. Department of Interior, 1970, Environmental Protec-
tion Agency Methods for Chemical Analysis of Water and
Wastes.
3. Finder, G. F. and J. F. Jones, 1969, Determination of the
Ground Water Component of Peak Discharge from the Chemis-
try of Total Runoff. Water Resources Research, Vol. 5, No. 2,
pp. 438-445.
4. Visocky, A. P., 1970, Estimating the Ground Water Con-
tribution to Storm Runoff by the Electrical Conductance
Method. Ground Water, Vol. 8, No. 2, pp. 5-10.
5. McWhorter, D. B., R. K. Skogerboe, and G. V. Skoger-
boe, 1974, Water Pollution Potential of Mine Spoils in the
Rocky Mountain Region. Proceedings 5th Symposium on Coal
Mine Drainage Research, Louisville, Kentucky, pp. 25-38.
6. McWhorter, D. B., R. K. Skogerboe, and G. V. Skoger-
boe, 1974, Potential of Mine and Mill Spoils for Water Quality
Degradation. Proceedings of the National Symposium on
Water Resources Problems Related to Mining, AWRA,
Golden, Colorado.
7. McWhorter, D. B., R. K. Skogerboe, and G. V. Skoger-
boe, 1975, Water Quality Control in Mine Spoils—Upper Colo-
rado River Basin. Environmental Protection Technology Ser-
ies, EPA-670/ 2-75-048, USEPA, Cincinnati, Ohio. 100 p.
-------
Disposal of
Coal Mining Industry
By-Products
Walter E. Grube, Jr., Eugene F. Harris and John F. Martin
U.S. Environmental Protection Agency
Cincinnati, Ohio
This discussion describes the materials remaining as wastes or
process by-products from mining, processing, and utilization of
coal. Materials considered include coal washing plant wastes,
sludges resulting from acid mine drainage (AMD) neutraliza-
tion, flyash recovered from precipitators in coal-burning power
plants and bottom ash, and solid reaction products recovered
from flue gas scrubbers on coal burning power plants.
A study of the geologic formation of coal is the first step in
understanding its composition and in handling the drainage
from coal wastes. Swamps of the Mesozoic and Paleozoic eras
account for most of the Earth's coal reserves. Vegetable material
falls into the swamp, is protected from immediate decay by the
high water table, and begins to accumulate and form a peat bog.
Natural acidity in the vegetation is probably responsible in part
for the depressed rate of biological decomposition. In the
absence of erosive forces, the bogs are filled, eventually covered
by protective layers of sediment, and slowly converted to lignite,
bituminous, and anthracite coals.
The conditions required for coal formation are: an interior or
coastal swamp environment, little erosion, rapid deposition of
material, and subsidence or subsequent flooding to allow
further deposition and covering. Considering this type of forma-
tion, one can easily visualize layers of silt and other foreign mat-
ter periodically flooding into the swamp, covering the vegetable
material, and eventually forming partings in the fully developed
coal seam.
The vegetable material carries with it various minerals and
trace metals accumulated during the growth process. Among
these constituents is organic sulphur derived from the swamp
water. Studies show that sulphur and pyrite are present in mod-
ern peat deposits, as they must have been in the geologic past.
According to Simon and Hopkins" seawater in the coastal
swamps may have been a primary contributor of sulphur in the
form of lenses, nodules and bands in the coal seam and adjacent
strata. In addition to the impurities formed in the coal during its
deposition, mineral impurities are carried by the groundwater
into the porous layers of fully developed coal seams.
The composition of the coal, and thereby the composition of
washing plant wastes, power plant flyash and bottom ash, and
stack scrubber precipitates, is the determining factor in consid-
ering which elements may be a pollutional problem with respect
to disposal of any of these materials. Materials removed from
raw coal during cleaning and washing processes will not be pres-
ent in flyash bottomash, or scrubber precipitates. However,
washing plant wastes will be enriched in elements removed as
contents of unsalable materials. Ruch, et al."> presented data
suggesting the extent of elemental segregation in cleaned coal,
and conversely elements that may be expected to concentrate in
the cleaning waste products, based on elemental analyses of over
100 coal samples. Their data showing the composition of var-
ious specific gravity fractions show which elements tend to be
associated with the organic fraction of coal, which remains after
the most intensive heavy-media separation of minerals, and
which elements tend to be contained in highest proportions in
the highest specific gravity fractions, indicating the composition
of the most inorganic or mineral-laden portion of the coal. Data
in Table I from analysis of one coal sample illustrate the trend
of elemental composition in coal fractions. Other analyses
reported by Ruch, et al. lo follow the same general pattern. The
elements germanium, beryllium, and boron consistently
appeared in higher amounts in the lightest fractions, suggesting
greatest affinity for or association with the organic matrix of
coal. Elements with the least affinity for, or association with, the
organic coal matrix, and therefore most associated with the min-
eral portion of coal include cadmium, mercury, lead, zinc, and
arsenic. These elements are readily recognized as cations found
with sulfides, and therefore an association with the pyritic
impurities of coal can be inferred.
Table I: Davis Coal
Clean coal - lightest
specific gravity fraction
(elements in "organic
combination")
Mineral matter - specific
gravity greater than 1.60
(elements in "inorganic
combination")
B
Ge
Be
Ti
Ga
P
V
Cr
Sb
Se
Co
Cu
Ni
Mri
Zr
Mo
Cd
Hg
Pb
Zn
As
133
-------
134 Coal Industry By-Products
Elements within the central part of Table I do not appear to be
specifically associated with either the heavy or light portions of
coal with any degree of reliability, therefore, they cannot in a
general way be assumed to be related to any specific components
of coal; likewise, these "'central" elements are ones which may
not be removed in proportionately greater or lesser amounts by
flotation cleaning techniques, and, therefore, also could not be
expected to be concentrated in cleaning plant wastes. However,
some individual elements may predominate in certain coal frac-
tions in coals from specific geographic and/or geologic envi-
ronments.
Metallic elements readily associated with sulfide minerals can
be assumed to be solubilized into the environment as these min-
erals oxidize after cleaning plant separation and subsequent
exposure. Acid generated through oxidation of the sulfides
allows wash water and percolate through coarse refuse disposal
piles to dissolve large quantities of other metals contained in the
waste. The current emphasis of both the coal cleaning industry
and regulatory bodies to produce a zero-discharge of cleaning
plant wash waters will tend to negate a significant contamina-
tion problem from these elements in slurry waters as closed cir-
cuit systems are developed. A long range situation may arise,
however, where these elements accumulate in settling ponds,
coarse and fine refuse dumps, or other areas around cleaning
plants where recycling waters are stored.
Martin6 has presented data showing the elemental content of
some typical coal refuse pile drainages. Silt, heavy metals, and
acid constitute the major water pollution hazards of coal refuse.
His data tables list the substantial amounts of sodium, alumi-
num, magnesium, calcium, manganese, iron, zinc, copper, lead,
and nickel that were found in many samples. Oxidation of
reduced sulfur compounds, mainly pyrite, present in many east-
ern coals and associated rock strata, contributes to the wide
range of acidities contained in waters from coal operations.
Table II illustrates the wide variety of water quality emanating
from coal refuse areas. The amount of pollutants generated by
coal refuse piles can also be related to the compaction of the
refuse, angle of side slopes, soil cover or surface water control, in
addition to the basic geochemical constitution of the coal.
The EPA currently is funding an active project at the Greene-
Sullivan State Forest, where the Indiana Department of Natural
Resources is demonstrating an economical method to abate acid
mine drainage from surface coal mine refuse. The project area
consists of a large pile and adjacent acid lake. The restoration
program includes grading, covering, revegetating and perma-
nently inundating a portion of the refuse. These procedures are
designed to break the cycle of erosion, exposure of pyritic mate-
rial, oxidation, acid formation, and erosion. Eventually, the
area will be restored to recreational use.
The coal mining industry, under severe pressure from
pollution abatement legislation, has in some cases adopted
effective measures to deal with large volumes of acid mine drain-
age waters resulting from various facets of the coal extraction
process. Currently, acid mine drainage treatment employs the
use of lime or limestone as the neutralizing agent. These neutral-
izing agents contribute to the production of substantial quanti-
ties of sludge or insoluble precipitates. In extreme cases, as much
as one-third of the treatment plant inflow can remain as sludge.
This precipitated sludge may contain over 99% water, which
compounds handling and disposal problems. Currently, sludge
is either perpetually stored in ponds, pumped underground into
mined-out workings, trucked to abandoned surface mines for
disposal, or dumped into drying lagoons.
The chemical composition of AMD neutralization sludge is
variable. Lovell4 reported that sludge is generally composed of
hydrated ferrous and/or ferric oxides, gypsum, hydratedalumi-
num oxide, varying amounts of sulfates, carbonates, bicarbo-
nates, calcium, and trace amounts of silica, phosphate, manga-
nese, titanium, copper and zinc. Akers and Moss1 provide
several sludge analyses, along with their investigations of dewa-
tering techniques. Table III presents the means of analyses of
effluents from four AMD neutralization plants discussed in
their report.
Table II: Water Quality Data from Selected Refuse Sites
Parameter
pH
Conductivity*
Acidity+
Alkalinity
S04
Na
Mg
Al
K
Ca
Mn
Fe
Ni
Cu
Zn
Pb
* All Values expressed
in umhos/cm.
+ Acidity to pH 7.3.
Luzerne
County,
Penna.
3.0
4400
690
-
3000
100
250
87
4.8
340
50
30
1.7
0,14
2.8
-
in mg/1
Pike
County,
Ky.
6.9
880
7
135
690
115
26
1.8
8.1
50
3.5
6.2
-
-
0.1
-
except pH in standard
Muhlenburg
County,
Ky.
2.5
6800
7020
-
7800
270
195
440
13
300
72
3400
3.0
-
8
0.12
units and
Sullivan
County,
Ind.
2.4
6400
6500
_
9500
200
285
340
3.0
350
120
2600
1.6
0.16
7.2
0.30
conductivity
-------
Coal Industry By-Products
135
Table III: Average Composition of Four AMD Neutralization Plant Effluents
Treatment Plant
Ca (ppm)
Mg (ppm)
Al (ppm)
Si (ppm)
S0= (ppm)
Total Fe (ppm)
Nonf ilterable Solids (ppm)
Shannopin
Mean
1,600
190
140
40
4,600
550
4,300
Banning
Mean
860
130
45
25
2,700
260
2,400
Norton
Mean
310
30
25
10
790
90
140
Edgell
Mean
920
75
35
15
4,200
260
2,200
Reports of investigations of land disposal of neutralization
sludge have been sparse in the literature. Conrad3 reported that
neutralization sludge was used to neutralize minewaste piles in
Illinois. He also reported beneficial effects resulting from appli-
cation to corn fields. Zaval and Robins13 proposed application
of sludge from the neutralization of acid strip mine pit waters
onto acid spoils in western Kentucky. Lovell4 developed initial
design criteria from preliminary studies on the use of drying
beds for dewatering of AMD sludge.
The Environmental Protection Agency (EPA) has an active
research program currently under way on spray-irrigation and
drying bed dewatering of AMD sludge. The studies are being
conducted at the EPA Crown Mine Drainage Control Field Site
near Morgantown, West Virginia. The sludges for the studies
were obtained from pilot plant neutralization of a moderately
severe ferrous iron AMD source (e.g., 640 mg/1 acidity, 300
mg/1 total iron, 270 mg/1 ferrous iron, 370 mg/1 calcium, 110
mg/1 magnesium, 480 mg/1 sodium, and 3000 mg/1 sulfate).
Hydrated lime, rockdust limestone and a combination of lime-
stone/lime were used as the neutralizing reagents at various
times throughout the research study. A flow diagram of the EPA
neutralization facility is shown in Figure 1.
_Pjessure
Regulator
Strainer Turbine
Flow
Meter
PROCESS A
„ . Sludge Recycle
Reactor ^. — J -i .. • • •- .
pN (control pH I
j j, -
•
U Waste |
SLUDGE JJiverter Valve
Limestone
Feeder
L^» JIV_Y____<^__'
Magnetic )
Flow Meter/
_<¥_
,f,
ACID MINE
DRAINAGE
I T I Composite
1—'sampler
PROCESS
„
Composite
Sampler
Sludge Recycle Diverter Valve
•o-
pH |co,ntrol| pH
XL...—.„„ J*
'HICKENER
Strainer Turbine
Flow
Meter
:i
TO DEWATERING
AND DISPOSAL
FACILITIES
Limestone
Feeder
Magnetic
Flow Meter'
Figure 1: Schematic Flow Diagram for the EPA Neutralization Facility
Composite
Sampler
-------
136 Coal Industry By-Products
Sludge produced from neutralization of acid mine water with
calcium was spray-irrigated onto a plot area composed of
weathering minewaste. Plot runoff water quality, soil moisture
status, and soil surface temperature effects were influenced by
sludge application. The most significant factors affecting plot
runoff water quality were rainfall intensity, vegetative cover,
and plot slope.
Several factors appear significant when considering the
applicability of spray-irrigation of neutralization sludge:
I) Erosion of the sludge during high and medium intensity
rainfall is readily apparent and undesirable.
2) No gross pollution was observed in the runoff from the
sludge irrigated areas during mild precipitation events.
3) It appears essential to have virtually flat topography on
sites proposed for spray irrigation of neutralization sludge
in order to prevent erosive loss of sludge during precipita-
tion events.
4) Sludge application appeared to have a slight beneficial
effect upon establishment and maintenance of a vegetative
cover. This was difficult to substantiate, however, in this
initial study.
Drying bed sludge dewatering was complicated by winter
freezing problems. Summer operation using lime-neutralized,
coagulant-treated sludge provided the following observations:
I) Approximately 20-percent by volume of the influent
sludge quantity (2.5-percent solids) was retained on the
sand bed as a 20-percent solids gell-like mass, 50 percent
drained through the sand as a clear effluent, and the
remaining 30 percent was assumed to be lost to the atmo-
sphere by evaporation.
2) The drainage rate through the sludge and sand averaged 26
liters/day/sq m (0.6 gal/day/sq ft).
3) Sludge solids appeared to stabilize near 20 percent within
20-days drying time.
Plans are under way to landfill the gell-like sludge removed
from the drying beds. Casual observations indicate that plants
species tolerant of alkaline soils may be successful on sludge
landfills.
Several organizations exist whose members have been
engaged in research into many aspects of coal ash disposal. The
National Ash Association (1819 "H" St. NW, Washington, D.C.
20006) sponsors regular Ash Utilization Symposia within which
soil amendments of flyash and bottom ash have been discussed.
The Dairyland Power Cooperative (Box 855, LaCrosse, Wis-
consin 54601) has sponsored one conference on soil amend-
ments and disposal of flyash, and has also assembled a bibli-
ography on the characterization and utilization of coal ash.
The geographical occurrence of coal, associated processing
industries, and the materials involved limit the availability of by-
products to the immediate locality where they are generated.
Therefore, disposal or utilization of these by-products occurs
nearby to the site of formation. The expensive hauling costs of
powerplant flyash, primarily because of the tremendous vol-
umes, dictate that flyash disposal be nearby to the powerplant.
Current flyash disposal is either by dumping on the floor of
nearby surface mines, placement into abandoned underground
mines, placement onto large disposal piles in remote (to the pub-
lic) neighborhoods, or disposal with bottom ash or SO
scrubber sludge.
Potential soil and groundwater contaminants from disposal
of flyash depend upon the specific properties of the local flyash
involved. The operating conditions of the powerplant boilers
have an effect on the physical nature of the flyash particles, and
also on the degree of chemical recombination of elements during
the coal burning process. Probably of greater significance,
however, is the nature and composition of the particular coal
being burned. Chemical analyses of coal samples taken from
within and among various seams show tremendous variability;
even one seam can vary in composition geographically. How-
ever, it is usually found in practice that a particular powerplant
has developed a long-term contract with a specific mine, or a few
mines, that provides a consistent quality of coal, and the power-
plant is then operated in adjustment for the particular feedstock,
so that the flyash generated by a specific powerplant can be well
characterized. These properties are depended upon as long as
the coal supply and plant operation are consistent. Surveys have
found that particular powerplants are a source of flyash of
distinctive properties that could be selected if needed to accom-
plish some particular purpose, such as strip mine reclamation
where a limey amendment is needed.
Many studies of flyash, such as those of Townsend and
Hodgson12 have expressed concern over the high amounts of
boron contained in flyash applied to soil. The boron is consid-
ered to be present as a complex glass-like borosilicate, and thus
potentially available for soil release over high levels of boron in
coal are not universally found, suggesting a regional ordeposi-
tional variation in coal boron composition. Some workers
believe that higher boron levels are a result of coal depositional
swamp development under heavy influence of sea water (marine
environment) while lower coal boron levels are found where
coals formed under continental or less saline conditions.
Martens5 has extensively reviewed literature on the
implications of agricultural use of flyash. Flyash particles are silt
or fine-sand sized, and can be considered as amorphous silica
with crystalline impurities of many elements in varying
amounts. The soluble salt content was found to range between 1
and 3% with pH ranging from below 5 up to 12. Neutralizing
capacity ranged up to almost one-fifth that of calcium carbo-
nate. Studies were made of the capability of flyash to supply the
plant nutrients boron, molybodenum, phosphorus, and zinc.
Results varied with the source of flyash and the type of soil to
which it was applied. He concluded that application of flyash
may be beneficial for crop production where the material
increases the pH of acidic soil, the availability of an essentialele-
ment in soil, or the the amount of available water in soil. Indis-
criminate application of flyash to soil was deemed undesirable
because of (a) possible decreased plant growth due to boron tox-
icity, (b) injury from a high concentration of soluble salt, (c)
phosphorus deficiency to plants induced by reaction of soil
phosphorus with hydroxides and oxides of aluminum and iron
in flyash, (d) possible nutrient deficiencies induced at high lev-
els of soil pH. Monitoring of the elemental composition, includ-
ing plant-available elements, in flyash prior to incorporation
into agricultural soils was considered essential to forestall possi-
ble future problems.
Solids generated by stack scrubbers are an increasingly
significant waste disposal problem for coal-burning power-
plants and other industrial plants that currently exhaust flue-
gasses unchecked. Rossoff, et al.9 and Proceedings7 have pre-
sented the most recent review of the status of flue-gas-desulfuri-
zation (FGD) product disposal. These volumes contain detailed
papers which present industry and regulatory experience to date
in this field.
The disposal of scrubber sludges presents a potential water
pollution problem and land reclamation problem because of sol-
uble species and sludge dewatering difficulties. Sludges contain-
ing leachable chemical pollutants will require containment
within the disposal site or chemical fixation to reduce leaching
potential. Permeability of the sludge to leachate waters varies
tremendously depending on sulfate/sulfite composition, mois-
ture content, compaction, and flyash content. Characterization
of input materials such as coal, flyash, make-up water, lime, and
limestone showed that most of the constituents of concern,
which without scrubbing are discharged in the flyash and flue
gas, originate in the coal.
-------
Coal Industry By-Products
137
Chemical analysis of five untreated scrubber sludge liquors
showed mercury, selenium, boron, chloride, sulfate and TDS
considerably in excess of water quality criteria. Each of the
sludge liquors exceeded EPA proposed criteria in at least one of
the following trace elements: arsenic, cadmium, chromium,
lead, mercury, and selenium. Particular excesses existed for
chloride, sulfate and TDS. Fluorine is also becoming a major
concern. Soil attenuation of selenium, boron, and chloride is
known to be ineffective. Mechanical stability of the sludge at a
given moisture content is affected by the sulfite content; sludges
proportionately higher in sulfite than sulfate tend to be more
thixotropic, leading to compaction and stability problems in de-
veloping completed landfills for structural or recreational uses.
SUMMARY
Coal mining industry by-products subject to land disposal
include washing plant wastes, AMD neutralization sludges,
powerplant flyash and bottom ash, and flue-gas-desulfurization
sludges. The geochemical composition of the raw coal deter-
mines the elements and compounds that are removed in coal
cleaning operations, and thus are concentrated in cleaning plant
wastes. Likewise, elements carried within the coal itself accumu-
late in the combustion exhaust products, flyash, bottom ash,
and flue-gas-desulfurization sludges. Elements solubilized as
groundwater passes through coal seams and associated rock
strata contaminate mine drainages, and are found in mine drain-
age treatment residues.
Ultimate disposal of most of these solid wastes has heretofore
been into ponds and dumps for perpetual storage. Recent stud-
ies of elemental constitution, leachate water composition, and
other environmental aspects of disposal of these materials indi-
cate that greater notice must be taken of the potential reactions
between the product, containment medium, and associated
waters, in order to prevent long-term degradations to the envi-
ronment. Various research and demonstration projects are
showing short-term improvements; but study efforts must be
continued to resolve the long-term problems concerning the
leaching and percolation of acid, heavy metals, and other toxic
materials associated with coal.
REFERENCES
1. Akers,D. J., Jr., and E. A. Moss. 1973. Dewatering of Mine
Drainage Sludge. Environmental Protection Technology Series
EPA-R2-73-169. Office of Research and Monitoring, U.S.
EPA, Washington, D.C. 20460.
2.Coalgate, J. L., D. J. Akers, and R. W. Frum. Gob Pile Sta-
bilization, Reclamation, and Utilization. Office of Coal
Research, DOI, Report No. 75. Coal Research Bureau School
of Mines, West Virginia University, Washington, D.C. May
1973. 127 pp.
3. Conrad, J. W., Proc. 111. Mining Inst. Ann. Mtg., Spring-
field, 111., pp. 104-108(1966).
4. Lovell, H. L. "The Control and Properties of Sludge Pro-
duced from the Treatment of Coal Mine Drainage Water by
Neutralization Processes," Third Symposium on Coal Mine
Drainage Research, Mellon Institute, Pittsburgh, Pennsylvania,
pp. 1-11 (May 1970).
5. Martens, D. C. 1976. Agricultural Uses of Fly Ash. Depart-
ment of Agronomy, VPI & SU, Blacksburg, VA 24061.
6. Martin, J. F. 1974. Quality of Effluents from Coal Refuse
Piles. Proceedings of the First Symposium on Mine and Prepa-
ration Plant Refuse Disposal, Louisville, Ky. National Coal
Association, Washington, D.C.
7. Proceedings: Symposium on Flue Gas Desulfurization,
New Orleans, March 1976. EPA-600/2-76-136a, b; Environ-
mental Protection Technology Series. IERL, USEPA, RTP,
North Carolina 27711.
8. Radian Corporation. 1975. Environmental Effects of Trace
Elements from Ponded Ash and Scrubber Sludge. Final Report
EPRI 202, EPRI, Palo Alto, Calif.
9. Rosoff, J., R. C. Rossi, and L. J. Bornstein. 1974. Disposal
of By-Products from Non-Regenerable Flue Gas Desulfuriza-
tion Systems, in Proceedings: Symposium on Flue Gas
Desulfurization-Atlanta, November 1974. EPA-650/2-74-l26a,
ORD, U.S. EPA, Washington, D.C. 20460.
10. Ruch, R. R., H. J. Gluskoter, and N. F. Shimp. Occur-
rence and Distribution of Potentially Volatile Trace Elements in
Coal. Env. Geol. Notes No. 72, Aug. 1974 111. State Geol. Surv.,
Urbana, 111. 61801.
11. Simon, J. A., and M. E. Hopkins. Geology of Coal. pp. 11-
39 inS. M. Cassidy. ed., Elements of Practical Coal Mining. The
American Institute of Mining, Metallurgical, and Petroleum
Engineers, Inc., New York. 1973.
12. Townsend, W. N. and D. R. Hodgson. 1973. Edaphologi-
cal Problems Associated with Deposits of Pulverized Fuel Ash.
in Ecology and Reclamation of Devasted Land, R. J. Hutnik
and G. Davis, ed. Gordon and Breach, New York.
13. Zaval, F. J. and J. D. Robins, 1973, Revegetation
Augmentation by Reuse of Treated Active Surface Mine Drain-
age. EPA-R2-72-119, Environmental Protection Technology
Series, U. S. Environmental Protection Agency, Washington,
D. C.
-------
Phosphate Transport Through Soil
Carl G. Enfield
Wastewater Management Branch
Robert S. Kerr Environmental Research Laboratory
U.S. ENVIRONMENTAL PROTECTION AGENCY
Ada, Oklahoma
Phosphorus in water readily reacts with the available calcium
or iron and aluminum in soil to form relatively insoluble com-
pounds, and must soils have a large capacity for immobilizing
applied phosphorus. The phosphorus concentration in naturally
occurring ground waters is low, typically 0.05 mg/1 or less. It is,
thus, easy to see why designers ignore phosphorus when design-
ing land application wastewater treatment systems. Soils do,
however, have a limited capacity for immobilizing applied phos-
phorus. The objective, herein, is to discuss factors affecting
phosphorus transport in relation to land application of waste-
water.
Land application of waste material containing phosphorus
can be thought of in two different ways. First, the land can be
considered a treatment facility where the objective is to remove
phosphorus from a waste stream prior to discharge. Secondly,
the land can be considered as a media by which phosphorus can
be recycled through crop production. When the application rate
is low, plant removal can play a significant role in the fate of ap-
plied phosphorus; but, as the application rate increases the sig-
nificance of plant removal is reduced.
Table I gives some typical phosphorus removals which can be
expected by plants.1 The impact of the plant on the treatment
system can be seen by following some example calculations
using the following simplified balance equation for phosphrous
transport
L - A-(OS)
where L = Loss of phosphorus from the land treatment facil-
ity;
A = application of phosphorus to the treatment facil-
ity;
C - crop removal of phosphorus;
S net gain in storage of phosphorus by the soil pro-
file.
C can be determined from the values in Table I; A can be calcu-
lated with a knowledge of the phosphorus concentration in the
wastewater being supplied to the treatment system and a knowl-
edge of the wastewater's application rate.
Example I. If the phosphorus concentration in the applied
wastewater is 10 mg/1 as phosphorus; the proposed application
rate is I ft/year; and corn is to be grown at the treatment facility
which can be expected to remove 31 pounds of phosphorus per
acre per year.
then A (Ibs/acre/yr) = 2.7 x concentration of P in wastewater
(mg/1) x annual application of wastewater (ft)
A 27 Ibs/acre/yr
L 27-(3l+S)
L + S = -4 pounds/acre/yr.
This would indicate no phosphorus would be lost from the treat-
ment system, and supplemental phosphorus would be required
to maintain the level of production.
Example 2. If the phosphorus concentration in the applied
wastewater is 10 mg/1 as phosphorus; the proposed application
rate is 50 ft/yr; and rice is to be grown at the treatment facility
which is expected to remove 20 pounds of phosphorus/
acre/year, then using equation (1)
L = A-(C+ S)
L 1,350 - (20 + S)
L + S - 1,330 pounds/acre/yr.
Few soils would have a capacity to remove this amount of phos-
phorus without losing some from the treatment facility.
With this introduction, it is clear that a knowledge of the fate
of phosphorus (S) in soils and the transport of phosphorus (L)
through soils is desirable when considering the design of a land
treatment system.
There is a large variability in a soil's ability to sorb phospho-
rus. A soil's capacity to sorb phosphorus is a function of pH and
the concentration of phosphorus in the applied waste material.
In other words, as the applied phosphorus concentration
increases, the total capacity of the soil to react with phosphorus
also increases. Simultaneously, passing water through the pro-
file reduces the total capacity of the profile to react with phos-
phorus. This is because the elements which have the potential to
react with phosphorus are washed out of the profile.
Figure 1 shows the relative end points of the applied phospho-
rus as a function of pH; thus, in acidic soils, we can conclude that
iron and aluminum play a predominate role in a soil's capacity
for sorbing phosphorus. In basic profiles, calcium is the pre-
dominate actor, while neutral soils give a confused picture with
many compounds influencing the fate of phosphorus.
SOLUBLE FORMS
0.
IT
o
_J
UJ
ir
SORPTION BY
CALCIUM
SORPTION BY
HYDROUS OXIDES
OF IRON ALUMINUM
AND MANGANESE
SORPTION
BY SOLUBLE
Ft, Al, 8 Mn
Figure I. Fate of phosphorus applied to soils.
138
-------
Phosphate Transport
139
Table I: Plant Phosphorus Uptake by Group
;orn:
180 bu. grain
8,000 Ibs. stover
Cotton:
1,500 Ibs. lint and 2,250 Ibs. seed
stalks, leaves, burrs
Wheat:
80 bu.
8,000 Ibs. straw
Oats:
100 bu.
straw
Barley:
100 bu.
straw
Rice:
7,000 Ibs. grain
7,000 Ibs. straw
Sugar beets:
30 teas roots
16 tons tops
Sugar cane:
100 tons stalks
tops and trash
Tobacco (burley):
4,000 Ibs. leaf
3,600 Ibs. stalks, tops, suckers
Soybeans :
60 bu.
7,000 stalks, leaves, pods
Peanuts:
4,000 Ibs. nuts
5,000 Ibs. vines
Apples:
600 boxes (42 Ibs.)
Grapes:
12 tons fruit
vines
Oranges:
600 boxes (90 Ibs.)
trees (70/acre)
Tomatoes:
40 tons fruit
4,400 Ibs. vines
Potatoes :
500 cwt
vines
Celery:
75 tons tops
roots
Sweet potatoes:
400 bu.
vines
Cabbage:
35 tons
23 tons stem and leaf
Snap beans:
4 tons
plants
Table beets:
25 tons roots
20 tons tops
Flax:
30 bu.
2,100 Ibs. straw
Cucumbers :
10 tons
vines
>eas:
3 tons
pods and vines
Onions:
30 tons
Lespedeza:
3 tons
Johnson grass:
12 tons
'ara grass:
12 tons
7.0, assumes one mole of cal-
cium will fix one mole of phosphorus such as would be the case for the
formation of dicalcium phosphate dihydrate, Ca H PO4» 2H2O. It is
assumed that the bulk density of the soil is 1.5 g/cc or 93.6 lb/ft.3 Thus,
the "total capacity" in pounds/acre foot of soil can be calculated as
31,600 x (percent calcium in the soil sample.)
-------
140 Phosphate Transport
The second soil to be considered is from the Maumee series in
Indiana. This is an acidic soil, pH 5.3, where the total capacity
should be a function of the iron and aluminum. Using break-
through data, one would estimate under the experimental condi-
tions when wastewaters were continuously applied, an estimated
290 pounds of phosphorus per acre foot of soil could be applied
prior to any measurable phosphorus leaving the system. Allow-
ing the system to rest will increase the system's overall apparent
detention capacity. A total capacity term for acidic soils is diffi-
cult to estimate. If all of the aluminum is considered in the
capacity term, it would require dissolving all of the clays in the
soil. This would require many centuries. Thus, at the present
time, it is not possible to reasonably estimate a total capacity for
acidic soils.
10 15 20 25 30 35
ELAPSED TIME (DAYS)
45 50
Figure 3. Experimental breakthrough curve for a 5 cm column
of Maumee soil when simulated wastewater, with a phosphorus
concentration of 9.2 mg/1, is applied at an application rate of 88
ft/yr.
A third soil comes from the Anway soil series of Arizona with
a pH of 7.2. This is a finer textured soil than either of the first
two samples. The experimental data shown in Figure 4 indicate
this soil has a much greater sorption capacity than either of the
two precious samples. Here, one would estimate the soil's capac-
ity before any measurable discharge at 670 pounds of phospho-
rus per acre foot of soil. Based on the calcium in the profile, the
capacity is 2,200 pounds of phosphorus per acre foot of soil.
- 10
I 8
15 20 25
ELAPSED TIME
30 35
(DAYS)
40 45 50
55
Figure 4. Experimental breakthrough curve for a 5 cm column
of Anway soil when simulated wastewater, with a phosphorus
concentration of 9.2 mg/1, is applied at an application rate of 88
ft/yr.
In each of the cases discussed thus far, the first detectable
phosphorus is considered. If some point other than break-
through is desired, the situation becomes more complex. Not all
breakthrough curves have the same shape. Figure 5 is experi-
mental data from a coarse textured soil from the Chigley soil ser-
ies of Oklahoma. With a p H of 8.0 one would anticipate a capac-
ity of 240,000 pounds of phosphorus per acre foot of soil because
of the calcium in the soil sample. Here, the phosphorus was
detectable after about 8 days of application. However, after 54
days of application, the concentration of P leaving the soil was
still approximately 1 mg/1. Apparently, this particular soil is
capable of releasing calcium at a rate sufficiently fast to form a
relatively insoluble phosphorus compound, Ca HPO4
(dicalcium phosphate dihydrate). This indicates there is more
calcium available for "immediate" reaction in this soil than in
the other soils tested.
10-
15 20 25 30 35 40
ELAPSED TIME (DAYS)
45 50 55
Figure 5. Experimental breakthrough curve for a 5 cm column
of Chigley soil when simulated wastewater, with a phorsphorus
concentration of 9.2 mg/1, is applied at an application rate of 88
ft/yr.
These examples point out some of the complexities of phos-
phate chemistry. Before an effective transport model can be de-
veloped, a firm understanding of the chemistry involved is
required. At the present time, there is no accepted phosphate
transport model in the literature. There are several attempts at
developing transport models for phosphorus, and with addi-
tional work the accuracy and reliability of these models will be
improved to an acceptable level.
Phosphate Transport
The dominance of the rate of reaction on the soil's ability to
remove phosphorus has created the need for developing trans-
port models capable of predicting how phosphorus will move
through soils. In the following paragraphs, three methods based
on laboratory sorption data are presented to predict how phos-
phorus will move through soils. Each of these procedures is
compared with the previously discussed experimental break-
through data to determine the reliability of the procedure.
It was mentioned that the higher the concentration of phos-
phorus in the applied waste, the greater the capacity for the soil
to remove the applied phosphorus. It has been observed by
many researchers that if soil is equilibrated with a solution con-
taining phosphorus, experimental data can be generated which
follows adsorption isotherm theories. Figure 6 shows experi-
mental data, for the soils previously discussed, plotted as
adsorption isotherms. The data and experimental procedures
were previously presented by Enfield and Bledsoe.2
Adsorption data indicate what level of phosphorus must be
supplied to the soil before a given solution concentration can be
supported. It is generally easier to use the log transformed data
as shown in Figure 7 than the linear form in Figure 6. This per-
mits expanding the lower concentrations and obtaining a more
detailed description of the sorption. One can then use these data
to project how wastewater might be transformed as it passes
through a soil profile. Figure 8 projects a breakthrough curve
for the Lakeland soil by calculating the time that would be
required for sufficient phosphorus to be applied to the soil to
support the effluent concentrations plotted in the figure. These
values are based on the flow rate of the applied solution and the
concentration of phosphorus in the applied waste. Figures 9,10,
and 11 show the same predictive approach for the other three
soils presented. Notice that in each case, this method of predict-
-------
Phosphate Transport
14!
ing phosphate transport underestimates the soil's ability to
assimilate phosphorus.
Using the adsorption isotherm approach, just discussed,
assumes the soil and the phosphorus in solution were in equilib-
rium prior to the experimental measurements. Researchers have
shown this to be an incorrect assumption. The more time
allowed for equilibration, the greater the amount of
phosphorus removed by the soil. Figure 12 shows how time
plays an important part in the sorption phenomena. Here, the
Lakeland isotherm is plotted on log transformed axis for eight
different times of equilibration. The equilibration times approx-
imate a geometric progression. This timed series of measure-
ments has made it possible to generate a sorption surface which
is a function of (1) equilibrating solution concentration, (2)
amount of phosphorus sorbed by the soil, and (3) time allowed
for equilibration.
400t-
10 2O 3O 40 50 6O 70 80 90 100
EQUILIBRATING SOLUTION CONCENTRATION (mg/l)
Figure 6. Adsorption isotherms for the four soil samples
included in the study. Gruve m is the Lakeland soil, E the Chig-
ley soil, Z the Maumee soil, and I the Anway soil. The time for
equilibration is 10 hours. The letter key corresponds to the letter
key in reference 2.
O.I 0.5 1.0 5.0 10
EQUILIBRATING SOLUTION CONCENTRATION
50 100
( mg/l)
Figure 7. Log transormed adsorption isotherms for the four soil
samples included in the study. Curve E is the Chigley soil, 1 the
Anway soil, M the Lakeland soil, and Z the Maumee soil.
~ 10
o»
— 8
-------
142 Phosphate Transport
100-
o
UJ
(D
I
a.
V)
o
0.5 I 5 10 50 100
EQUILIBRATING SOLUTION CONCENTRATION (nig/I)
Figure 12. Time dependence of the sorption of phosphorus by
the Lakeland soil. This family of curves thus develops a sorption
surface.
These sorption surfaces suggest an alternate method of pre-
dicting phosphorus transport through soils. By using an
approach similar to the equilibrium adsorption isotherm
method just presented, but taking into consideration time of
application, it is possible to develop breakthrough curves by
interpolating times between actual measurements. This is
accomplished by selecting an effluent solution concentration;
determining the amount of phosphorus which must be supplied
at several measurement times, based on flow rate and applied
solution concentration; then determining the actual time by lin-
ear interpolation. Figure 13 is the breakthrough for the Lake-
land soil developed in this manner. Figure 14 is the time depen-
dent sorption surface for the Maumee soil, and Figure 15 is the
calculated breakthrough curve. Figure 16 is the time depen-
dent sorption surface for Anway soil, and Figure 17 is the
calculated breakthrough curve. In Figure 18 the geometry of the
sorption surface for the Chigley soil is quite different from the
surfaces for the other soils presented. Here, a radial series of
curves were measured with time while, in all the other experi-
mental data, the curves appeared to be more or less parallel. This
difference in geometry in the sorption surface is apparent also in
both the calculated and measured breakthrough curves pre-
sented in Figure 19.
9 2
/
15 20
ELAPSED
25
TIME
30 35
(MRS )
Figure 13. Projected breakthrough curve for the Lakeland soil.
This projected curve is based on the sorption surface in Figure
E
a
a.
Q
UJ
CO
CO
-------
Phosphate Transport
143
15 20 25
ELAPSED TIME
30 35
( DAYS )
40 45
55
Figure 17. Projected breakthrough curve for the Anway soil
based on the sorption surface in Figure 16.
50000-
E
a.
a.
m
a:
o
O
I
a.
in
o
CL
O.I 0.5 1.0 5 10
EQUILIBRATING SOLUTION CONC.
50 100
(mg/l)
Figure 18. Phosphorus sorption surface by the Chigley soil.
10
15 20 25 30 35
ELAPSED TIME (DAYS)
Figure 19. Projected breakthrough curve for the Chigley soil
based on the sorption surface in Figure 18.
The two predictive methods thus far presented assume the
concentration every place in the profile is the same. This is an
obvious limitation of the procedures. Neither approach consid-
ers the history of application. The general approach which has
been proposed to take these two factors into consideration fol-
lows miscible displacement theory. This requires a solution to
the differential equation
3c_
9t
_3c p 9s
(2)
where
c = solution phase concentration of phosphorus
(mg/l)
V = average pore-water velocity (cm/hr)
\ = distance from the beginning of the flow path
p = bulk density of the soil (g/cm^)
0 = fractional solution-filled-volume in the por-
ous media
s = solid phase concentration of phosphorous
(Mg/g)
A miscible displacement approach should accurately predict a
breakthrough curve if an accurate sink term (ds/dt) is available.
The sink term gives an indication of the rate at which the phos-
phorus can be sorbed as a function of the equilibrating solution
concentration and the amount of phosphorus already sorbed.
One sink term which has been proposed is
9s , .
—=acb.sd (3)
where a, b, and d are constants. This is an empirical function
which has been shown to conform to many sorption surfaces.
Figure 20 is a graphical example of the sink term. Note the
similarity between this function and the sorption surfaces pres-
ented in Figures 12, 14, and 16. Thus, we could expect this sink
term to reasonably predict breakthrough curves for these three
soils. Figures 21, 22, and 23 are the predicted breakthrough
curves using the above miscible displacement approach. These
breakthrough curves were obtained using a finite difference
solution to the differential equation. Only part of the elapsed
time could be projected using an IBM 1130 computer because of
core limitations of the machine.
It appears that a reasonable fit will be obtained for the two
soils with "low capacity" but the fit is unsatisfactory for the
Anway soil sample. Apparently the sink term 6s/dt does not
adequately describe the sorption of phosphorus by this soil. In
Figure 24, the projected breakthrough curve for the Chigley soil
samples is presented with the experimental data. A reasonable
fit is obtained over the projection period even though the shape
of the sorption surface is quite different from the one described
by the sink term.
CONCLUSIONS
None of the approaches presented are capable of accurately
predicting how phosphorus will move through a profile every
time. None of these methods are capable of predicting the phos-
phorus concentration on the desorption side of the break-
through curve. We can, thus, conclude that additional work is
still required to develop more refined predictive methods and
-------
144 Phosphate Transport
increase our confidence in our ability to understand phosphate
transport through soils.
The time dependent sorption surface approach or the
miscible displacement approach gives an adequate first approxi-
mation to the transport of phosphorus through soils. Using one
of these approaches is much better than ignoring the problem
altogether and gives a basis for a design philosophy.
Since the models are not capable of predicting the desorption
of phosphorus, it is felt a more reliable estimate can be obtained
for the entire life of a proposed treatment system by considering
the average application rate, thus, taking into consideration the
"rest period" and an adjusted applied solution concentration to
permit including plant removals and dilution effects of rain-
water.
Through use of these design procedures, the confidence lim-
its can be developed and reliability of a given approach deter-
mined. It is through application that an approach will be veri-
fied. The two above methods are good starting points to develop
these design procedures.
10
too
EQUILIBRATING CONCENTRATON (mg/l)
Figure 20. Regression of the empirical rate equation to experi-
mental data of a Maumee soil sample. In the figure, the rate of
sorption is presented in parts per million per hour. The experi-
mental data points are plotted numerically with the data point
located at the decimal point of the value (a = 7.3; b = 2.28; d =
-1.63).
ELAPSED TIME
30 35
(DAYS)
50 55
Figure 21. Projected breakthrough curve for the Lakeland soil
based on the miscible displacement approach (a = 183; b = 2.40;d
--2.91; p = l.5g/cc; 0 - 0.18; V- 1.69 cm/hr).
15 20 25 30 35 4O
ELAPSED TIME (DAYS)
45 50
55
Figure 22. Projected breakthrough curve for the Maumee soil
based on the miscible discplacemeirt approach (a = 280; b = 2.08;
d = -2.08; P = 1.5g/cc; 6 - 0.18; V = 1.69 cm/hr).
10
20 25
35 40
ELAPSED TIME (DAYS)
Figure 23. Projected breakthrough curve for the Anway soil
based on the miscible displacement approach (a = 612000; b =
2.81; d = -3.87; P=1.5g/cc; d = 0.18; V = 1.69 cm/hr).
- 10
8
15 20 25 30 35
ELAPSED TIME (DAYS)
40 45 50 55
Figure 24. Projected breakthrough curve for the Chigley soil
based on the miscible displacementapproach (a = 4.88; b = 1.92;
d--1.05; p-l.5g/cc; 9 = 0.18; V = 1.69 cm/hr).
REFERENCES
I. U.S. Department of Agriculture, Soil Conservation Ser-
vice. Agricultural Waste Management Field Manual. August
1975.
2. Enfield, Carl G., and Bert E. Bledsoe. Kinetic Model for
Orthophosphate Reactions in Mineral Soils.
EPA-660/2-75-022. U.S. Environmental Protection Agency,
Ada, Oklahoma, 1975. 133 pp.
BIBLIOGRAPHY
I. Dean, L. A. 1949. Fixation of Soil Phosphorus. In: Advan-
ces in Agronomy, Vol. I, Norman, A. G. (ed.). New York, Aca-
demic Press, pp. 391-411.
-------
Phosphate Transport
145
2. Enfield, Carl G., and Bert E. Bledsoe. 1975. Fate of Waste-
water Phosphorus in Soil. Journal of the Irrigation and Drain-
age Division, ASCE, 101:145-155.
3. Enfield, Carl G., Curtis C. Harlin, Jr., and Bert E. Bledsoe.
1976. Comparison of Five Kinetic Models for Orthophosphate
Reactions in Mineral Soils. Soil Sci. Soc. Am. J. 40:243-249.
4. Enfield, Carl G., and Lowell E. Leach. 1975. Phosphorus
Model of Muskegon Wastewater System. Journal of the Envi-
ronmental Engineering Division, ASCE, 101:911-916.
5. Enfield, Carl G., and D. C. Shew. 1975. Comparison of
Two Predictive Non Equilibrium One-Dimensional Models for
Phosphorus Sorption and Movement through Homogeneous
Soils. J. Environ. Qual., 4:198-202.
6. Larson, Sigurd. 1967. Soil Phosphorus. In: Advances in
Agronomy, Vol. 19, Norman, A. G. (ed.). New York, Academic
Press, pp. 151-210.
7. U.S. Department of Agriculture, Agricultural Research
Service. July 1974. Factors Involved in Land Application of
Agricultural and Municipal Wastes. 200 pp.
-------
Updating the Nitrogen Cycle
L. M. Walsh
Soil Science Department
University of Wisconsin
Madison, Wisconsin
INTRODUCTION
In a publication discussing disposal of residues on land, it is
appropriate that we spend some time reviewing the nitrogen
cycle and expanding our concepts about it based on recent
research information. Since the fate of the nitrogen (N) applied
in waste material often dictates or prescribes what an environ-
mentally acceptable waste management practice would be, it is
essential that everyone involved in management decisions con-
cerning the application of wastes on land thoroughly under-
stand the N cycle. Considering this fact, the first part of this pa-
per will deal with a review of where N is found in the biosphere
and the N cycle, with emphasis on the physical, chemical and
biological processes that affect the transport of N from one pool
to another. The remainder of the paper will be devoted to using
known principles and research findings to make estimates on
how additions of N or changes in soil management can affect the
N balance in soils and the amount of N which may be found in
various pools or compartments within the biosphere.
Nitrogen in the Biosphere
Nitrogen is found in many chemical and biological forms but
most of it resides in the atmosphere, the soil, and the water.
Although some N is cycled through man, animals and plants, the
amount found in these biological organisms is very small as
compared to the amount in the atmosphere, soil and water'. The
rate of chemical and biological transfers and the rate of hydro-
logic transport are influenced by man's activities, by climate,
and by other environmental factors. As a result, the amount of N
in the various compartments or pools will change from time to
time. In general, however, our overall goal should be to develop
management systems which will result in maximum utilization
of available N by plants and minimum loss of N to surface or
ground waters.
Soils contain substantial amounts of organic N, generally in a
range of 1000 to 3000 kg/ha, but most of this N is resistant to
microbial decomposition and mineralizes slowly. Mineraliza-
tion rates of only 1 to 3% per year are common; at these rates
mineral soils would supply only 10 to 90 kg/ha of N annu-
ally",14. This means that we usually have to compensate for lim-
ited supplies of available soil N by adding fertilizer, manure,
sewage sludge or other nitrogeneous waste materials. As the
amount of N applied increases, the percent of the applied N re-
covered by the crop decreases. Some of the nitrogen not used by
the crop could become a potential environmental hazard if
excess nitrate nitrogen (NO3-N) enters drinking water supplies.
However, it is important to emphasize that the residual NO3-N
constitutes only a potential hazard because it may be converted
to atmospheric N by denitrification, immobilized by crop
residues, or carried over for use by a succeeding crop. With any
of these possibilities the N not recovered by the crop would not
be considered an environmental hazard because it would not
enter surface or ground water. An exception to this statement
would be that nitrous .oxide (N2O), one of the products of denit-
rification, may cause depletion of the ozone layer in the upper
atmosphere2.
Nitrogen Cycle
Organic-Inorganic Nitrogen Transformations
Nitrogen is transferred between soils, plants, animals, and
man so it often does not enter or leave the pools of N in the at-
mosphere or water. However, the N cycle properly includes all
exchanges and transformations noted in Figure 1. A basic
understanding of this cycle is necessary in order to predict the
fate of all forms of N in the ecosystem.
Ammonification (Reaction 1, Figure 1) is the biological
conversion of organic N into ammonium nitrogen (NH 4-N). It
occurs over a wide range of soil and climatic conditions, howev-
er, the rate of reaction is favored by warm temperatures and
good aeration of the soil. Ammonium N is available to plants,
but it generally does not leach because the positively charged
cation (NH4+) is held on the surface of the negatively charged
soil particles. On soils with a low cation exchange capacity, espe-
cially irrigated sandy soils, some downward movement of
NH4-N may occur.
Nitrification (Reaction 2, Figure 1) is the biological
transformation of NH 4-N to NO 3-N. Nitrate is readily available
to plants. However, it is a negatively charged anion (NO,-) so it
remains in solution in the soil and may be leached below the root
zone as water percolates through the soil. This fact gives rise to
most of the concern being expressed about N in relation to water
quality. In warm, well-aerated, and properly-limed soils (pH
5.6-8.0), much of the NH4-N is changed to NO3-N within one to
three weeks after application. Recent efforts to improve the re-
covery of NH4-N fertilizer have been directed toward reducing
the solubility of fertilizer, slowing down the rate of nitrification
by the use of bacterial inhibitors, or maintaining environmental
conditions which are not favorable for the nitrifying bacteria.
Immobilization (Reaction 3, Figure 1) is the process whereby
bacteria which decompose carbon-rich crop residues or wastes
immobilize or "tie-up" the available N (NH4-N + NO3-N). The
role immobilization plays in the fate of residual available N must
be carefully evaluated because the N not recovered by the crop
may be utilized by soil microorganisms as they decompose car-
bonaceous residues such as straw and corn stover14. Even
though N immobilized as microbial protein is again mineralized
after decomposition takes place, immobilization reduces leach-
ing of NO3-N during that part of the year when most crops are
not assimilating N, i.e., the late fall and early spring.
146
-------
Updating Nitrogen '47
crop, resKfu^-^.
r
Figure 1: The Nitrogen Cycle and Transfers of Nitrogen Between Various Compartments in the Environment [The Encircled Numbers
Refer to Reactions Discussed in the Text l6].
-------
148 Updating Nitrogen
Additions of Nitrogen to the Soil
Fixation of Atmospheric N (Reaction 4, Figure 1) by bacteria
results in the addition of considerable quantities of N each year.
In calculating N budgets, perennial legumes contribute a net
addition to the soil and should be credited accordingly when
determining the N requirement for a succeeding crop. Howev-
er, annual legumes probably assimilate as much available soil N
in the year they are grown as they provide to the succeeding
crop. In fact, recent estimates indicate that growing soybeans
results in a net removal of N from the soil system, especially
when the soil contains residual inorganic N from a preceding
crop6.
As indicated by Reaction 4 in Figure 1, some lower plants,
primarily blue-green algae, fix atmospheric N in surface water
and may contribute significantly to the N budget in some lakes
and streams. For example, 14 percent of the N in Lake Mendota,
Madison, Wisconsin is estimated to be fixed by algae8
Wastes excreted by animals and man can be returned to the
soil as organic N or as some form of available N. Animals pro-
duce large quantities of wastes but as much as 80 percent of the
N in these wastes probably is lost in the collection handling, and
spreading processes and is not returned to cultivated land15.
Humans excrete significant quantities of N in wastes but only a
small portion of these wastes, probably less than 20 percent, are
being returned to cropland at the present time. Through better
management of animal manure and development of systems to
apply more human wastes on land, the amount of N recycled
from these wastes could easily be doubled or tripled. More effi-
cient utilization of organic wastes would significantly reduce fer-
tilizer N requirements in some localized areas but nationwide
fertilizer needs could be reduced only 10 to 15 percent, even with
very efficient waste management systems.
Fertilizer N is added to soils in relatively large amounts, often
in the range of 50 to 200 kg N/ha, to replace the N removed by
crops or lost by various physical, chemical, and biological pro-
cesses. As indicated in Figure 1, atmospheric N is converted into
NH 4-N or NO 3-N in the manufacture of fertilizer N. However,
most of the N fertilizer presently added to the soil is in the form
of NH 4-N. Ammonium nitrate contains half of its N as NO 3-N
but its usage is declining rapidly because it is more expensive
than other forms of N fertilizer. When urea, a soluble organic
form of N, is added to the soil, it is quickly converted to NH 4-N
by the urease enzyme.
Precipitation (Reaction 5, Figure 1) generally adds 5 to 10
kg/ ha acre of available N annually4. This is a small addition on a
per hectare basis but, since the entire land and surface water area
receives precipitation, it is a significant contribution to the total
N budget.
Losses of Nitrogen from the Soil
Leaching of nitrate (Reaction 6, Figure 1) can be a serious
problem, especially in humid areas on light-textured soils
because the water holding capacity of these soils is low and rela-
tively small amounts of rain or irrigation water readily move
NO 3-N below the root zone10. Loss of NO 3-N by leaching is of
environmental concern because it adds to the pool of nitrates in
the surface and groundwater. As shown in Figure 1, once the
NO 3-N is in the water compartment very little will reenter the N
cycle. Only small quantities of water would be consumed by man
and animals, or applied to irrigated soils and taken up by plants.
Denitrification, a process by which soil bacteria change
NO j-N into unavailable atmospheric N (Reaction 7, Figure 1),
occurs primarily in poorly aerated, waterlogged soils, but some
denitrification can take place within the aggregates of well-
drained soils, especially when heavy rains occur. Denitrification
takes place rapidly. If water stands on the soil more than three or
four days during the growing season, much of the NO 3-N will be
lost by denitrification. For denitrification to occur, decomposa-
ble organic matter must be present as a source of energy. In
general, denitrification will not take place deep in the subsoil or
in the groundwater because of inadequate sources of energy
material to support the growth of denitrifying bacteria. An
important exception is that denitrification, and to a lesser extent
immobilization, of NO 3-N occurs as groundwater moves
through sediments into a lake or stream8.
Volatilization of ammonia (NH,) is another way in which N is
lost to the atmosphere (Reaction 8, Figure 1). When urea fertil-
izer, sewage sludge, or fermented manure is surface applied and
not incorporated into the soil, relatively large amounts of
NH4-N can be lost as NHr Injection or immediate incorpora-
tion of these materials eliminates most of the volatilization loss.
Some NH 3 volatilization loss also occurs when anhydrous
ammonia or low-pressure solutions containing NH 3 are not
properly applied. The total losses are thought to be small but in
localized situations as much as 50 percent of the NH 4-N, such as
from surface applied urea or sewage sludge, may be volatilized
and lost as NHy
Erosion (Reaction 9, Figure 1) may account for loss of a con-
siderable amount of organic N. Runoff water tends to selectively
carry fine soil particles and organic matter from the land. Thus,
sediment contains a higher concentration of total N than the soil
which remains behind. Runoff water generally contains very low
concentrations of NH 4- and NO 3-N, even though fertilizer use
and the production of cash grain crops has increased
dramatically in recent years.
Crop removal accounts for the transport of much N from the
soil to other N pools (Figure 1). Some of the N is returned to the
soil in crop residues and manure. Even so, removal of 50 to 150
kg N / ha in the harvested portion of many agronomic and horti-
cultural crops is common.
Nitrogen Balance
In recent years we have been concerned with producing more
food and this generally means using more N and following a
more intensive crop rotation. Concomitantly, we have been con-
cerned about the addition of excessive amounts of N to natural
waters, depletion of soil organic N, erosion losses, and other
undesirable side effects which often accompany efforts to
increase crop production.
One way of trying to arrive at an answer to these questions,
especially the concern about increased leaching of N, is to look
at the N budget, i.e., compare N inputs with N removals. Such a
comparison has been made on a nationwide basis15 and is pres-
ented in Table I. Even though some of these estimates are specu-
lative, certain trends are apparent. When fertilizer N consump-
tion was low in the 1930's and 1940's, removals of N far exceeded
inputs; thus little leaching of fertilizer N could have occurred.
Since 1969 most estimates indicate that inputs have equalled or
exceeded removals, and some leaching of fertilizer is likely to
have taken place in areas of high fertilizer N use.
Even though the balance sheet for 1969 (Table I) shows that
inputs and removals were equal, considerably more N was
removed from harvested crops than was applied with N fertil-
izer. This observation has led some to believe that N fertilizer
cannot be contributing significantly to the pollution of the natu-
ral waters with NO 3-N. However, progressive farmers, espe-
cially those producing "cash crops," are often adding more N
fertilizer than their crops recover17.
A more recent estimate for the midwestern states is presented
in Table II16. It is based on 1975 nutrient removals and inputs
and shows that N fertilizer additions exceed removals in several
states. When other sources of N, such as manure and legumes
are credited, total additions exceed total removals in most of the
states surveyed. However, this does not necessarily mean that
excess N is being applied. Some N is lost to the atmosphere by
-------
Updating Nitrogen
149
NH 3volatilization and denitrification. In the case of denitrifica-
tion, losses will be substantial whenever well fertilized soils
become waterlogged.
Table I: Balance Sheet of N in the USA: Estimated Changes
from 1930 to 1969 on Harvested Cropland.15
Nitrogen — Millions of Tons
Item 1930
Inputs of Nitrogen from:
1. Fertilizer N 0.3
2, M fixed by legumes 1.7
3. N fixed (nonsymbiotic) 1.0
t. Barnyard manure ±.y
a. Roots of unharvested
6. Rainfall 0.8
Total D.H
Removals of Nitrogen by:
7. Harvested crops 4.b
8. Erosion b.O
9. Leaching of soil N 4.0
10. Leaching of fertilizer Na 0
11. Denitrification ?
Total 13.6
aZeros and question marks indicate that little is
of these losses.
19U7 1969
0.7 6.8
1.7 2.0
-L.U 1.0
1.3 1.0
1.5 2.5
1.0 _1.5
7.2 14.8
b.5 9.5
4.0 3.0
3.0 2.0
0 ?
JL_ ^_
13.5 14.5
known about the magnitude
Another important concept relating to this question is that
some leaching loss of N is inevitable whenever drainage occurs
because crops cannot remove all of the NO 3-N from percolating
water. Optimum fertilization of deep rooted annual crops such
as corn or sudangrass grown on medium- and fine-textured soil,
usually results in drainage water containing from 2 to 15 ppm of
NO 3-N' ,3. Other researchers have observed that a concentration
of at least 40 ppm NO 3-N may be required in the soil water in the
root zone of sudangrass12 and irrigated potatoes13 in order to
produce maximum yields. It should also be pointed out that
some leaching occurs when plants are not growing. In fact,
midwestern researchers18 have found that of the total N leached,
about 70% was leached before corn was planted. These observa-
tions indicate that some leaching loss of N will occur even under
the best management systems.
Nitrogen Model for Wisconsin
If reasonable estimates can be made for all N inputs and
losses, and if rates of transfer between the various N pools can be
estimated, it ought to be possible to construct a N model which
would help predict the effect of many soil, crop and waste man-
agement practices on the fate of N in the environment. In some
cases, accurate estimates can be made, such as the amount of N
fertilizer added or the amount of N exported in crop and animal
products, but only crude estimates can be made for other parts
of the model such as leaching, denitrification and volatilization
losses. Because of these uncertainties, an N model should not be
used to try to precisely define the actual amount of N in a pool.
However, a model would be effective in predicting general
trends, especially when rather large changes in N inputs or out-
puts are made. A model should find its greatest use as an aid to
decision-makers and researchers plan for the future.
Table II: Estimate of Nitrogen Additions and Removals on Cropland in Selected Midwestern States.16
State
Arkansas
Illinois
Indiana
Iowa
Kentucky
Michigan
Minnesota
Missouri
Tennessee
Wisconsin
Approx.
Cropland
Acreage
6
x 10 acres
7.7
22.7
11.5
23.8
4.2
6.5
20.9
12.5
4.6
9.5
N Additions
N
Removal
72241
535400
371723
645237
65410
155250
386300
228705
73273
210625
Fertilizer
105870
724109
345762
712576
117000
120000
401000
171215
115808
125000
Other
Sources
7867
46755
54508
112825
26200
36980
58500
46500
51800
85000
Total
113737
770864
400270
845401
143200
156980
459520
217715
172283
210000
Net Addi- Ratio of
tion (+) Fertilizer
or Remov- N Applied to
al (-) N Removed
+41496
t235464
+28547
+180164
+77790
+1730
+73200
-10990
+99010
-625
weighted regional aver. =
1.47
1.35
.93
1.10
1.79
.77
1.10
.75
1.58
.59
1.10
-------
150 Updating Nitrogen
Agricultural N Balance
An N model was recently developed for the state of Wiscon-
sin7 and will be used as an example of how an N model can be de-
veloped and employed as an aid in decision-making.
Data for 1974 were used in constructing this model19. The
flow of N in Wisconsin agriculture is depicted in Figure 2. Ni-
trogen inputs to agricultural land included fertilizer (127 x 106
kg N), precipitation (46 x 106 kg N), symbiotic fixation (310 x 106
kg N), nonsymbiotic fixation (39 x 106 kg N), and imports of
animal feed (60 x IO6 kg N), with a total input of 582 x 106 kg N.
(Paniculate N in precipitation is assumed to be mainly soil
derived so its net contribution is zero). Nitrogen losses from
agricultural land were estimated to be as follows: leaching, 84 x
106 kg N; denitrification and volatilization of available soil N,
130 x I06 kgN; direct NH3 volatilization from fertilizer, 5 x 106
kg N; direct NH3 volatilization from manure and other wastes,
130 x I O6 kgN; erosion of organic N into surface water, 18 x 106
kg N; human waste not returned to land, 27 x 106 kg N; exports
of crop and animal products out of the state, 86 x 106 kg N; and
unaccounted losses in the handling and utilization of crop and
animal products, 62 x 106 kg N. Losses total to 542 x 106 kg N
which is slightly less than the estimated inputs but this is reason-
ably good agreement considering the lack of good data for many
of the estimates.
Some of the estimates used in determining the inputs and
outputs listed previously are as follows. The soil organic N pool
was calculated assuming 0.15% N (4000 kg N/ha 20 cm; bulk
density, 1.33 g/cm3) and net mineralization was estimated at 1%
year. Inputs to the dynamic available soil N pool total to 393 x
106 kg N/yr (70 kg N/ha avg.). Outputs due to crop uptake are
estimated by the difference between total crop removal and
residual and symbiotic N2 fixation. Leaching lesses were esti-
mated at 10 mg/liter of NOs-N in 15 cm of drainage (15 kg
N/ha/yr). The gaseous loss of N (denitrification plus volatiliza-
tion) of 130 x 106 kg N or 23 kg N/ha/yr was probably largely
due to denitrification. An additional 130 x 106 kg N is estimated
volatilized from animal wastes, and 5 x 106 kg N is estimated to
be volatilized directly from urea, NH 3and N solution fertilizers.
Most of the crop N is transferred to the animal, with the
majority of this N transferred on to wastes. Human consump-
tion and exports account for 113 x 106 kg N (27% of the har-
vested plant N). Consumed N is approximately equal to fertil-
izer N.
Input into the soil organic pool is 263 x 106 kg N/yr, or 1.2%
of the total. This is slightly more than the estimated annual min-
eralization rate. Erosion output to streams are estimated at 3 kg
N/ha/yr. On balance, the model predicts that Wisconsin agri-
cultural soils are not being depleted in organic N. More detailed
information on how each of the estimates for inputs and outputs
were arrived at cannot be provided here but they are included in
the original report7.
Response of Wisconsin Agriculture to N Fertilizer Inputs
The Wisconsin agricultural N balance is less sensitive to
fertilizer N changes than might be expected with a strictly cash
STREAMS
130 (NH3)
SOIL
ORGANIC
NITROGEN
22,500
WISCONSIN
PEOPLE
WASTE
(MINERALIZATION)
1%/YR
CROP REMOVAL
AND RESIDUAL
489
AVAILABLE
SOIL N
393
84?
130?
LEACHING
Figure 2: The Flow of Nitrogen in Wisconsin Agriculture in 1974. (Values Are Given in I06 kg of N.)
DENITRIFICATION,
VOLATILIZATION
-------
Updating Nitrogen 151
grain system, since a relatively large input from soil organic N
mineralization exists. The area in corn silage has remained rela-
tively constant at about 400,000 ha. However, area devoted to
corn for grain has increased 59% in a decade while average fertil-
izer rates have more than doubled. The state-wide average corn
yields reflect weather more than they do N fertilizer rates. In
1974 (the year chosen for the N balance in Figure 2) N removal
by corn was considerably below average, due to a late spring and
an early frost.
The model in Figure 2 can be tested by extreme pertubations
on the N fertilizer input, holding corn acreage constant (i.e.,
changing only the average rate of fertilizer N on corn). If fertil-
izer N inputs were reduced 50% it would constitute 17% rather
than 31 % of the available soil N pool. Using N response data for
Wisconsin,5 average corn grain yields would decline by 18%
(1,070 kg/ha or 17 bushels/A). At present prices, this decline
would result in a loss of $ 115 million in crop value or a net loss of
$100 million due to $15 million less investment in fertilizer.
Leaching and denitrification would decrease by about 23%.
On the other hand, a doubling of N fertilizer use would
increase this input to 48% of the total available soil N pool, and
increase leaching and denitrification by 58%. Crop uptake
would increase by 20 x 106 kg N (11%). A yield increase of 10%
would be expected, increasing gross returns by about $61 mil-
lion but costing $30 million in extra fertilizer N for a net return
of $31 million. This analysis, however, is highly dependent on
the balance between corn and fertilizer prices.
ACKNOWLEDGMENTS
Appreciation is expressed to the Soil Conservation Society of
America for their permission to reprint Figure 1 and Table II.
Also, the author is indebted to Dr. D. R. Keeney for much of the
information included in the last section of this paper on use of a
"Nitrogen Model for Wisconsin."
REFERENCES
1. Bolton, E.R., J.W. Aylesworth, and F.E. Hare. 1970.
Nutrient losses through tile drains under three cropping systems
and two fertility levels on a Brookstone clay soil. Can. J. Soil
Sci. 50: 275-279.
2. CAST, 1976. Effect of increased nitrogen fixation on stra-
tospheric ozone. Council for Agricultural Science and Technol-
ogy, Report No. 53. Jan. 19, 1976. Iowa State Univ., Ames. 33
pp.
3. Erickson, A.E. and B.C. Ellis. 1971. The nutrient content of
drainage water from agricultural land. Research Bui. 31, Michi-
gan State University.
4. Hoeft, R.G., D.R. Keeney and L.M.Walsh. 1971. Nitrogen
and sulfur in precipitation and sulfur dioxide in the atmosphere
in Wisconsin. J. Environ. Qual. 1: 203-208.
5. Ibach, D.B. and J.R. Adams. 1968. Crop yield responses to
fertilizer in the United States. USDA—Economic Research Ser-
vice and Statistical Reporting Service. Stat. Bull. 431. Wash.,
D.C. 295 pp.
6. Johnson, J.W., L.F. Welch and L.T. Kurtz. 1975.
Environmental implications of N fixation by soybeans. J.
Environ. Qual. 4: 303-306.
7. Keeney, D.R. 1976. Nitrogen balance in Wisconsin.
Unpublished Report, Dept. of Soil Science. Univ. of Wisconsin,
Madison. 15 pp.
8. Macgregor, A.N. and D.R. Keeney. 1975. Nutrient reac-
tions. In N.F. Stanley and M.P. Alpers, (ed.). Man-made lakes
and human health. Academic Press, London, pp. 237-257.
9. National Academy of Sciences. 1972. Accumulation of
nitrate. National Academy of Sciences, Wash., D.C. 106 pp.
10. Olsen, R.J., R.F. Hensler, O.J. Attoe, S.A. Witzel and
L.A. Peterson. 1970. Fertilizer nitrogen and crop rotation in
relation to movement of nitrate nitrogen through soil profiles.
Soil Sci. Soc. Amer. Proc. 34: 448^52.
11. Parr, J.F. 1973. Chemical and biochemical consider-
ations for maximizing the efficiency of fertilizer nitrogen. J.
Environ. Qual. 2: 75-84.
12. Pratt, P.P., S. Davis and J. Warneke. 1975. Yields of for-
ages leaching of nitrate and nitrogen balance in an irrigated field
treated with dry and liquid manures. Annual Report National
Science Foundation. Grant GI 34733X and GI43664. Univer-
sity of California.
13. Saffigna, P.O., D.R. Keeney, and C.B. Tanner. 1973.
Water and nitrogen balance on potatoes grown on irrigated
sandy soils. In Proc. of the 1973 Pert, and Aglime Conf., Dept.
of Soil Science, University of Wisconsin, Madison: 30-37.
14. Stanford, G. 1973. Rationale for optimum nitrogen fertili-
zation in corn production. J. Environ. Qual. 2: 159-165.
15. Stanford, G., C.B. England, and A.W. Taylor. 1970.
Fertilizer use and water quality. U.S. Dept. Agr., Agr. Res.
Serv., ARS 41-168.
16. Walsh, L.M., M.E. Sumner and R.B. Corey. 1976. Con-
sideration of soils for accepting plant nutrients and potentially
toxic nonessential elements. In Land Application of Waste
Materials. Soil Conservation Society of Omeried, Onkeny,
Iowa. pp. 22-47.
17. Welch, L.F. 1972. More nutrients are added to soil than
are hauled away in crops. Illinois Research 14: 3-4.
18. Welch, L.F. and A.A. Bomke. 1969. Potential for nitrogen
loss through tile drains. In Proceedings of the 1969 Illinois Fer-
tilizer Conference, Dept. of Agronomy, University of Illinois:
37-39.
19. Wisconsin Agricultural Statistics. 1975. H.M. Walters,
statistician in charge. Wis. Dept. Agric., U.S. Dept. Agric. Mad-
ison, WI. 83 pp.
-------
Fate of Nitrogen
From Fertilizer Practices
G. W. Wallingford
Department of Soil Science
University of Minnesota
Crookston, Minnesota
The fate of fertilizer nitrogen in the soil-plant system has con-
cerned agriculturalists for many years. Lack of nitrogen limits
crop production on United States soils more than any other
essential plant nutrient. Methods to most economically correct
nitrogen deficiencies have received intensive investigation by
researchers throughout the nation. The use of inorganic nitro-
gen fertilizers has emerged as the most economical and most
widely used corrective measure.
The use of nitrogen fertilizers by US farmers to satisfy their
crop's nitrogen needs increased dramatically in the last 45 years
(Table I) and has contributed more than any other single factor
to the concomitant increases in grain and forage yields. Corn
receives more nitrogen than any other crop (Table II).
Table I. Nitrogen fertilizer use
(Statistical Reporting Service, USDA)
Year
1930
1950
1960
1968
1973
US
-10QO
343
912
2484
6157
7565
Minnesota
Metric Tons-
-
-
50
228
385
Table II: Nitrogen fertilizer use by crop (USDA Statistics)
(1973 - US Total)
Crop
Corn
Wheat
Soybeans
Cotton
Hectares
(Millions)
25.1
21.9
22.7
A. 9
Rate
(Kg N/HA)
119
34
3
61
Total
(Million KG)
2990
745
68
299
The availability and continued use of nitrogen fertilizers is
essential if food production is to keep abreast of population
increases. But unnecessary and excessive use of nitrogen fertiliz-
ers can waste scarce energy sources and contaminate surface and
subsurface water supplies. Determining the most economical
and least environmentally damaging methods to use nitrogen
fertilizers is the goal of all those involved with agriculture.
Forms of Fertilizer Nitrogen
The production of fertilizer nitrogen begins with the fixation
of atmospheric nitrogen gas (N^ into anhydrous ammonia
which can be transformed by further chemical processes into
other nitrogen fertilizers such as ammonium nitrate and urea.
Because anhydrous ammonia (NH 3, 82% N) is produced in the
first step of the chemical synthesis, it is the least expensive form
of nitrogen available. Being a compressed gas, it is more difficult
to use than dry fertilizers and can be dangerous if improperly
handled. Its application requires special equipment that can
inject it at least four, inches below the soil surface to prevent
escape of ammonia. Nevertheless, its low cost generally over-
rides these disadvantages and its extensive use verifies its accep-
tance by farmers to satisfy their crop's nitrogen needs (Table
III).
Table III: Nitrogen use by material
(Crop Reporting Board, USDA)
Material
1970
1974
Million Metric Tons of N
Anhydrous Ammonia
Nitrogen Solutions
Ammonium Nitrate
Urea
3.1
2.9
2.6
0.5
3.8
3.7
2.9
0.9
Ammonium nitrate (NH
-------
Fertilizer Practices
153
cost, urea is being used in larger quantities and prospects for an
increasing share of the fertilizer market are good because the
newer nitrogen fertilizer plants produce urea only. In the future,
urea is also likely to compete well with anhydrous ammonia due
to its ease of handling and its ability to mix with other solid fer-
tilizers.
Solutions of urea-ammonium nitrate mixtures (28 to 32% N)
are an important part of the fertilizer market. Non-pressure liq-
uids are easier to handle than solids, can be more accurately ap-
plied, and allow the addition of trace nutrients. The slightly
higher prices of liquids have not slowed their increasing popu-
larity.
Nitrogen fertilizers which have more limited use include cal-
cium nitrate and sodium nitrate. Various nitrogen-phosphorus-
potassium blends are also produced in both solid and liquid
forms.
Application Techniques
The method of application of nitrogen fertilizers depends on
the physical form of the material. Solid, granular fertilizers can
be broadcast by various types of spreader machines which is
usually done prior to planting which allows the application of
fertilizer in close proximity to the seed which can increase the
efficiency of the fertilizer compared to the broadcast technique.
Fertilizer solutions which contain no anhydrous ammonia are
called nonpressure solutions and can be applied to the soil sur-
face without significant volatilization loss of nitrogen. This
allows the broadcast application of solutions by spreader trucks
prior to the time of seeding or they can be applied by planter at-
tachments.
Anhydrous ammonia and pressure nitrogen solutions which
contain ammonia must be injected beneath the soil surface to
prevent gaseous loss of ammonia. The various types of applica-
tors designed for this injection generally include a shank which
penetrates the soil at least four inches with a tube near the bot-
tom of the shank in which the solution is discharged into, and
immediately covered by, the soil.
Foliar applications of nitrogen to an established crop can be
an effective method of satisfying the nitrogen requirement, but
its use is limited because only small quantities of nitrogen can be
applied per application to prevent salt burning of the leaves.
Another method of applying nitrogen solutions is the metering
of the solution into irrigation water and allowing the fertilizer to
be distributed across the field through the irrigation system.
Soil Pathways of Fertilizer Nitrogen
The nitrogen cycle has been examined in detail by other pa-
pers in this conference and the specifics will not be dealt with
here. The key to understanding what happens to nitrogen after
soil application is a solid grasp of the relationships between
ammonium and nitrate-nitrogen forms and how they are biolog-
ically and chemically related.
The ammonium ion is a cation (positively charged) and is
attracted to the negatively-charged clay particles that are pres-
ent in all soils. This attraction is a very effective barrier against
movement of the ammonium ions through the soil. In other
words, the ammonium ion will stay where it is placed until it is
changed to other nitrogen forms or removed by plant roots.
Nitrification is a biological process which converts ammo-
nium to nitrate. If soil conditions exist which promote active
biological activity (warm, moist soils), the conversion occurs
rapidly within a few days or weeks. In the form of nitrate, the ni-
trogen is more available for plant uptake—desirable for good
plant nutrition. This nitrate; however, is subject to other less
beneficial losses.
Nitrate is an anion (negatively charged) and is repulsed by
negatively-charged clay particles in the soil. Thus it remains in
the soil solution and is susceptible to leaching losses. How much
is lost due to leaching depends on several factors such as the
amount of moisture received and the permeability of the soil.
Generally, sandy soils are more prone to leaching losses than are
clay soils.
Nitrogen may also be lost from the nitrate form by volatiliza-
tion of nitrogen gases through denitrification. This biological
process converts nitrate into nitrogen gases, primarily N2 and
N2O, which are lost into the atmosphere. Soil conditions which
encourage denitrification are water logging, the presence of eas-
ily oxidizable organic matter, warm soil temperatures and the
presence of nitrate ions. An example of when significant denitri-
fication losses could occur is after heavy and continuous rains
on a warm, poorly-drained soil in the late spring or summer.
Losses of nitrogen due to denitrification are hard to measure
precisely because the main end product, N2, is so abundant in the
atmosphere. It has been shown in carefully controlled labora-
tory experiments, and indirectly shown by nitrogen balance cal-
culations in field experiments, that denitrification can account
for significant losses of nitrogen and contributes detrimentally
to the efficiency of plant use of fertilizer nitrogen.
Ammonia volatilization is another source of potential loss of
nitrogen. The most extreme ammonia volatilization losses occur
when ammonium containing fertilizers are applied on the soil
surface and allowed to remain without incorporation for several
days. Hot, windy, drying conditions and soils of high pH en-
courage ammonia volatilization losses. Urea is particularly sus-
ceptible to this loss because it is readily hydrolyzed by the urease
enzyme to give off ammonia gas.
Fertilizer Nitrogen Losses in Surface Runoff
Movement of fertilizer nitrogen with surface runoff has
received attention by researchers because of the high water-
solubility of most nitrogen fertilizer materials and concern that
nitrogen on or close to the soil surface would tend to dissolve or
diffuse into surface water. There are several factors that could
affect nitrogen loss in runoff water: intensity of rainfall; slope of
the land; total amount of runoff; soil texture and infiltration
rate; vegetative cover; antecedent soil moisture levels; and depth
of fertilizer incorporation.
Moe et al7,8 found that nitrogen loss in rainfall runoff was
greatest when ammonium nitrate applied without incorporation
to wet silt loam soils or to fallow soils with a surface seal. The
maximum nitrogen loss was 15% of the 224 kg N/ha applied.
When runoff losses from soil treated with urea and ammonium
nitrate were compared, it was shown that less ammonium was
lost from the urea plots, probably because urea was leached
deeper in the soil before hydrolysis than the more highly ionized
ammonium nitrate, thus making the ammonium (formed after
hydrolysis of the urea) less susceptible to dissolution into surface
waters.
When the nitrate content of runoff from cultivated soils in
South Dakota was compared to the nitrate in the runoff from a
virgin prairie, White and Williamson14 found no differences.
The quality of the runoff was influenced by how long the precipi-
tation remains on the surface and the amount of precipitation,
but not by agricultural practice. Similar results were recorded by
Thomas and Crutchfield'2 in Kentucky. They reported that the
nitrate content of streams draining agricultural watersheds was
correlated to local geology and not land use.
After working with an irrigated soil in Nebraska, Edwards et
al2 reported that once nitrate moves below the soil surface it usu-
ally will not enter surface water unless it moves with erosion or
with plant residues. If the soil is initially saturated, however,
nitrate will not move downward as rapidly as the water.
Knight5 found very little nitrate in tailwater off fertilized, per-
meable soils in Kansas. After the first one and one-half hours of
irrigation there were no differences between the nitrate in tail-
waters off the control and fertilized plots.
-------
154
Fertilizer Practices
Nitrate runoff losses from a simulated 12.7 cm rainfall
reported by White et al13 were only 0.15 percent of the 224 kg
N/ha applied as ammonium nitrate to plots in sod on a sandy
loam soil in Georgia. On fallow plots the loss was still a low 2.3
percent. It was suggested that the first few minutes of rainfall
leached the fertilizer into the soil and made it less available to
runoff loss.
Research has shown that fertilizer nitrogen will not be lost in
surface runoff except during the most severe runoff events in
which soil erosion occurs. Adequate soil incorporation of the
fertilizer, a vegetative cover and dry antecedent soil conditions
will tend to reduce nitrogen losses in surface runoff.
Leaching of Fertilizer Nitrogen Into Groundwater
Most of the inorganic nitrogen in soils is in the form of
nitrate—an anion with no electrostatic attraction to the nega-
tively charged soil colloidal particles and susceptible to move-
ment with water. Factors which could affect movement of
nitrate into groundwater are several: amount of nitrate present
in the soil; soil conditions which are conducive to denitrifica-
tion; the leaching volume which is determined by precipitation
and the hydraulic properties of the soil; plant uptake; and
microbial immobilization.
Gentzsch4 found less nitrate in poorly-drained Illinois soils
with impermeable horizons. The levels of nitrate in soils without
impermeable horizons were related to the amount of fertilizer
used or to animal activity. Impermeable horizons could enhance
conditions conducive to denitrification and lower losses of fertil-
izer nitrogen.
Layers which restrict drainage were also found by Pratt10 to
lower the leaching of nitrate in soils of California. Concentra-
tions of nitrate of up to 123 and 70 ppm were found in the water
in the unsaturated zone beneath row crops and citrus, respec-
tively. There were three main factors influencing these nitrate
levels: the difference between nitrogen input and crop removal;
drainage volume; and soil profile characteristics (restricted
drainage) that apparently control denitrification. In soils with-
out restricted drainage the nitrate levels could be predicted by
the difference between nitrogen input and crop removal of nitro-
gen.
In a deep coring study in the South Platte River Valley in
Colorado, Stewart et al'' found increased total nitrate in the top
6.1 m of irrigated and dryland-cultivated fields compared to
native grassland. Alfalfa fields had somewhat less nitrate than
native grassland fields. They estimated that 28 to 34 kg N/ha
were lost annually to the groundwater beneath irrigated fields.
Significant amounts of nitrate were found beneath the root zone
in dryland-cultivated fields even though the average rainfall is
only 38 cm/yr. It was noted that much of this nitrate could have
come from mineralization of native organic nitrogen.
Bower and Wilcox' reported that even though nitrogen fertili-
zation increased greatly over a 30-year period on irrigated land
adjacent to the Upper Rio Grande River, the average nitrate
content of the river did not increase. Denitrification in saturated
soil close to the water table was suggested as an explanation for
the apparent lack of nitrate movement to the river. Willardson
and Meek15 were able to show that nitrate reduction can indeed
occur at the water table.
Linville and Smith6 studied the fate of fertilizer nitrogen as
affected by rate of application and soil texture on continuous
corn plots in Missouri. Large accumulations of nitrate were
found in the soil at rates of 168 kg N/ha and higher, while rates
of 112 and 134 kg N/ha caused no movement of nitrate below
244 cm. Generally, the coarser textured soils had deeper move-
ment of nitrate than the finer textured soils. It was concluded
that nitrogen applications greatly in excess of crop requirements
increase the potential for nitrate leaching.
Gast et al1 measured nitrogen loss through tile lines in plots
that had received up to 448 kg N/ha on a Minnesota soil. Total
nitrogen loss through the tile lines under the plots which
received 112 kg N/ha (the recommended rate) was only slightly
greater than under the plots that received no nitrogen. The 224
and 448 kg N/ha rates significantly increased the total loss of
nitrogen in one year, but the increased loss was much less than
the additional fertilizer applied over the recommended rate.
In an investigation of factors influencing the nitrogen content
of waters of Nebraska, Muir et al9 found little correlation
between nitrogen content of waters and the use of fertilizers.
Only where there was irrigation on sandy soils with shallow
water tables was there contamination of groundwater by fertil-
izer nitrogen.
Research has shown that applications of fertilizer nitrogen in
great excess of the nitrogen removed by the crop will result in
nitrate accumulations in the soil and with sufficient leaching vol-
umes movement to the groundwater. There are factors that can
lower nitrate accumulations, such as layers of restrictive drain-
age in the soil which encourage denitrification. Most research
showing leaching loss of fertilizer nitrogen has involved the
extreme cases of high application rates. These high rates are
rarely used by farmers, who recognize the diminishing returns in
yields and profits from the higher rates. As nitrogen fertilizer
prices continue to increase, farmers will become even more con-
scious of the inefficiency of applying nitrogen in great excess of
that removed by the crop.
Farming Practices That Reduce Nitrogen Losses
Numerous researchers have shown that there is very little ni-
trogen loss from soils that have received the application rate rec-
ommended by state extension agencies. These rates are deter-
mined by field trials which compare different ratesof nitrogen
and are correlated to soil test results. The optimum rate is the
one which gives the maximum yield from the least amount of
nitrogen applied. Even though most recommended rates are
higher than that removed by the crop, the balance is made up by
biological immobilization in the soil and gaseous losses through
denitrification.
Nitrogen losses can be reduced by proper timing of the fertil-
izer application. The most efficient timing would be to add ni-
trogen only as the plant needs it or feeding the plant periodically
throughout the growing season. On a field-scale basis this is
impractical, however, because of the high cost of the energy
required to spread the fertilizer and because of the crop damage
when the plants are large. The most common practice now is to
apply all of the projected crop fertilizer needs prior to planting.
This leaves the nitrogen in the soil, where it is susceptible to
leaching and denitrification losses, for several weeks before the
plants are large enough to take up significant quantities of the
nitrogen. Applying the fertilizer as close to planting time as pos-
sible reduces chances for nitrogen loss. Some corn producers are
now optimizing yields and nitrogen use efficiencies by applying
side-dress applications of nitrogen after the corn plants are well
established. Split applications of nitrogen have also been a com-
mon practice on pastures for many years because several small
applications throughout the growing season are more efficient
and produce higher yields than a single large application early in
the season.
In an effort to reduce the spring workload, many farmers
apply their fertilizers in the fall. If applied late enough in the fall
and if the weather cooperates, no significant losses of nitrogen
will occur. If there is a late warm fall or an early warm spring,
however, in which the nitrogen is exposed to soil conditions con-
ducive to nitrification, there can be losses of nitrogen prior to the
growing season. This makes fall application of nitrogen fertiliz-
ers generally less efficient than spring applications. When fall
applications are made, it is best to apply ammonium-based fer-
tilizers such as urea or anhydrous ammonia which are less sus-
-------
Fertilizer Practices
155
ceptible to loss. If ammonium fertilizers are applied late enough
in the fall when the soil temperatures are cool enough to prevent
biological nitrification of the ammonium into nitrate, there will
be little or no movement of the nitrogen until the following
spring.
Slow-release nitrogen fertilizers are available which have
lower water solubilities due to surface coatings on the fertilizer
prill or to combination with materials which lowers the water
solubility. Sulfur-coated urea is one type of slow-release fertil-
izer. The sulfur coating restricts water penetration and subse-
quent release of the nitrogen into the soil solution, thereby
extending the nitrogen availability over longer periods. This
material has shown to be an effective method to increase nitro-
gen efficiency on soils that are subjected to high leaching losses
such as highly permeable irrigated sands. The higher cost of this
material restricts its use to soils which have high leaching losses
or when more frequent applications of nitrogen are not practi-
cal.
Preventing the conversion of ammonium to nitrate by nitrifi-
cation can lower leaching and denitrification losses. Chemical
additives to ammonia-based fertilizers are available which tem-
porarily lower the activity of the soil organisms that carry out
nitrification. These chemicals, called nitrification inhibitors, can
be mixed prior to application with anhydrous ammonia and
urea. Upon application, they stop the nitrification process in the
zone immediately around the fertilizers, thus keeping the nitro-
gen in the ammonium form longer.
Good soil conservation practices that prevent soil erosion can
lower significantly the nitrogen lost into surface waters. If ero-
sion is controlled, nitrogen loss in runoff is negligible.
REFERENCES
1. Bower, C.A., and L.V. Wilcox. 1969. Nitrate content of
the upper Rio Grande as influenced by nitrogen fertilization of
adjacent irrigated lands. SSSAP. 33:971-973.
2. Edwards, D.M., P.E. Fishback, and L.L. Young. 1972.
Movement of nitrates under irrigated agriculture. Trans.
ASAE. 15:73-75. In Fertilizer Abstracts, 15:68, 1974.
3. Cast, R.A., W.W. Nelson and A.C. Caldwell. 1975. A
Report on Field Research in Soils. Soil Series 95. Dept. Soil
Science, Univ. of Minnesota.
4. Gentzsch, E.P., E.C.A. Runge, and T.R. Peck. 1974.
Nitrate Occurrence in some soils with and without natric hori-
zons. J. Environ. Quality. 3:89-94.
5. Knight, C.W. 1971. Nitrate accumulation in soils and
water as affected by nitrogen fertilization. M.S. Thesis, Kansas
State University.
6. Linville, K.W., and G.E, Smith. 1971. Nitrate content of
soil cores from corn plots after repeated nitrogen fertilization.
Soil Science. 112:249-25?.
7. Moe, P.G., J.V. Mannering, and C.B. Johnson. 1967.
Loss of fertilizer nitrogen in surface runoff water. Soil Science.
104:389-394.
8. . 1968. A comparison of nitrogen losses from
urea and ammonium nitrate in surface runoff water. Soil Sci.
105:428-433.
9. Muir, J., E.G. Seim, and R.A.Olsen. 1973. A study of fac-
tors influencing the nitrogen and phosphorus contents of
Nebraska waters. J. Environ. Quality. 2:466^70.
10. Pratt, P.P. 1972. Nitrate in the Unsaturated Zone Under
Agricultural Lands. Water Pollution Control Res. Series. 16060
DOE.
11. Stewart, B.A., F.G. Viets, and G.L. Hutchinson. 1968.
Agriculture's effect on nitrate pollution of groundwater. J. of
Soil and Water Conser. Vol. 23.
12. Thomas, G.W., and J.D. Crutchfield. 1974. Nitrate-
nitrogen and phosphorus contents of streams draining small
agricultural watersheds in Kentucky. J. Environ. Quality.
3:46^9.
13. White, A.W., A.P. Barnett, and W.A. Jackson. 1967. Ni-
trogen fertilizer runoff from crop land tested. Crops and Soils.
19:28.
14. White, E.M., and E.J. Williamson. 1973. Plant nutrient
concentrations in runoff from fertilized cultivated erosion plots
and prairie in eastern South Dakota. J. Environ. Quality.
2:453-5.
15. Willardson, L.S., and B.D. Meek. 1969. Agricultural
nitrate reduction at a water table. In Collected Papers Regard-
ing Nitrates in Agricultural Waste Water. Water Pollution Con-
trol Res. Ser., 13030 ELY. p. 41-52.
-------
Fate of Nitrogen From Manure Disposal1
W. L. Powers, R. V. Terry and L. S. Murphy
Agronomy Department
Kansas State University
Manhattan, Kansas
G. W. Wallingford
Northwest Experiment Station
Crookston, Minnesota
INTRODUCTION
Nitrogen from animal manure has polluted our water level
several times during the past ten to fifteen year years and caused
eutrophication of our lakes and streams. Some livestock deaths
and even infant deaths have been attributed to excessive nitrate
concentrations in drinking water. Such nitrogen pollution has
warranted the U.S. Public Health Service to set
concentration limits of 10 parts per million (ppm) nitrate-
nitrogen for drinking water. To abide by these limits, we must
control the fate of the nitrogen in the animal manure we add to
the soil.
Controlling the fate of nitrogen involves understanding the
process by which nitrogen from decaying manure is released to
the soil; how it can be returned to the atmosphere; how it can
accumulate in the soil; or, how it can be leached from the soil to
the ground water below. Most of all, we need to know the
amount and form of nitrogen being added to the soil.
Knowing the form of nitrogen is necessary because nitrogen in
the form of nitrate (NOj) is highly soluble and moves easily with
the water in the soil; however, nitrogen in the ammonium form
(NHp is attracted to the negative soil articles and doesn't move
easily with the soil water.
The following are the objectives of this presentation: (1) Dis-
cuss the nitrogen concentration in animal manure, (2) discuss
the factors affecting the pathways nitrogen will follow in the ni-
trogen cycle and (3) review publications on management prac-
tices to control the fate of nitrogen in animal manure.
We will discuss first the nitrogen concentration in manure
because this probably has the greatest influence on its fate.
Concentration and Forms of Nitrogen
in Manure
The chemical and physical properties of manure vary widely
across the United States. Important factors affecting the proper-
ties of manure are the physiology, ration, and environment of
the animals. Most nitrogen consumed by the animal is in the
form of plant protein. These proteins vary in digestibility, and
most plant nitrogen excreted by the animal in the solid form is
that which resists digestion. Except for those used to build flesh
in the animal, the digested proteins are absorbed, cycled from
one protein to another, and are later excreted.22 Table I shows
'Contribution No. I607a Agronomy Department. Kansas Agricultur-
al Experiment Station, Manhattan and Department of Soil Science,
University of Minnesota, St. Paul.
the percentage of nitrogen recovered in manure from various
types of animals.
Table 1: Percent of Fed Nitrogen Recovered in Animal Manure'
Type of Animal
Nitrogen Recovered
(Percent)
Dairy Cattle
Beef Cattle
Sheep
Swine
74
75 - 89
68
72
Adapted from Table 4 of Azevedo and Stout (2).
Because nearly all nitrogen excreted is protein, the decay pro-
cess and the factors affecting the rate of decay are paramount
when considering the availability of nitrogen in animal manures.
The rate and process of decay are dynamic and highly depen-
dent upon the nitrogen content of the manure. This
decomposition which converts the nitrogen in the organic form
to the inorganic form such as nitrate (NOj) is called mineraliza-
tion. (Many of the terms we will use here are found in the Re-
source Conservation Glossary published by the Soil Conserva-
tion Society of America.5) It is this rate of nitrogen
mineralization that determines the availability of the nitrogen
and its potential to pollute.
At the present time, the best method to express the rate of ni-
trogen mineralization is one using a decay series. As described
by Pratt et al.,18 the decay series concept is based on mineraliza-
tion of organic nitrogen into inorganic or available nitrogen.
The rate of mineralization will be most rapid in the first year
after application and will decrease in subsequent years. For ex-
ample, 40% of the applied nitrogen might become available in
the first year, 25% of the residual nitrogen in the second, 6% of
the residual nitrogen in the third year, 3% of the residual nitro-
gen in the fourth, and 3% all subsequent years. That decay series
would be expressed as0.40,0.25,0.06, and 0.03. The percentages
after the first year refer to the organic nitrogen remaining in the
soil and not the original amounts of nitrogen applied. Manure
higher in nitrogen content will decay faster. For example, a
manure with 3.5% nitrogen may have a decay series of 0.75,0.15,
0.10,0.05; a manure with 1.0% nitrogen may have a decay series
of 0.20, 0.10,0.05.
The effect of the nitrogen on the decay series demonstrates the
156
-------
Manure Disposal
157
need to know the nitrogen concentration of the manure. How-
ever, upon examining Table II, we see that the nitrogen concent
of manure from a given species and handling system can range
from a relatively low to a high value.
Table II: Maximum and Minimum Values for Total Nitrogen
Content of Animal Manures"
Species and Type
of Manure
Beef
Solid
Runoff .
Slurry(D)°
Slurry(U)D
Dairy
Solid
Runoff
Slurry(D)
Slurry(U)
Swine
Solid
Runoff
Slurry(D)
Slurry(U)
Poultry
Solid
Runoff
Slurry(D)
Slurry(U)
Minimum Value
(percent)
0.6
0.0011
11.00
1.9
1.5
4.8
1.8
2.0
0.4
8.8
3.4
1.1
_
2.2
7.4
Maximum Value
(percent)
4.9
0.86
11.0
9.0
3.9
4.8
_
5.1
7.5
4.8
12.0
19.0
11.0
-
7.8
7.4
aBeef runoff data is on wet weight basis, all other is on a dry
weight basis. Single
values indicate
Table adapted from Table 1 of Powers
D refers to digested
and U refers to
only one value found.
et. al. (16).
undigested.
The variability of the nitrogen content of manure is due to fac-
tors such as climate, species, ration, and manure handling. Fig-
ure 1 is a schematic diagram of factors affecting the nitrogen
concentration of the manure.
Climate can affect the nitrogen composition in many ways. If
the manure is stored where it is exposed to rainfall, the natural
precipitation can dilute the manure when it is in the liquid form
or it can remove soluble nitrogen compounds by leaching them
from the solid manure. This process will proceed at greater rates
in regions of higher rainfall. In climatic regions characterized by
hot drying conditions, manure can lose large quantities of ni-
trogen by ammonia volatilization. It has been estimated that
under conditions similar to those of Southern California, about
50% of the nitrogen in dairy manure is lost to the air by ammonia
volatilization and denitrification between the time of excretion
and the time manure is incorporated into the soil.1
The species and animal size also contribute to the variable ni-
trogen content of the manure. Poultry and swine are considered
to produce manure highest in nitrogen.
Modern rations used in today's feedlots can be altered, there-
by increasing the variability of the nitrogen content.of the
manures. Rations used today range from 70 to 80% in digestibil-
ity.2 This means that the materials excreted by the animals are
less resistant to decomposition and thus speed decaying to
release the nitrogen.
The nitrogen content, more than any other constituent, is sen-
sitive to management. The method of manure handling and stor-
age between the time of excretion and application can have a
major effect on the quantity of nitrogen contained in the
manure. Nitrogen can be lost by ammonia volatilization, leach-
ing of soluble nitrogen compounds, dilution, and microbial util-
ization. Several researchers have shown that up to 50% of the
nitrogen excreted by cattle can be lost through ammonia volatil-
ization.7,8,19 Leaching of solid waste can remove much of the
water soluble nitrogen, which can comprise as much as 50% to
60% of the total nitrogen in steer manure.19 Up to 80% of the ni-
trogen in swine manure can be removed by treatment in an oxi-
dation ditch.20
Figure 1 Schematic summary of factors affecting waste composition and influencing the efficiency of its use on cropland.
-------
158
Manure Disposal
Pathways for Nitrogen Movement in the
Nitrogen Cycle
To control water pollution from nitrogen, we need to under-
stand the various pathways the nitrogen will follow after the
manure is spread on the soil. Earlier in this conference we dis-
cussed the nitrogen cycle. Here we will review some of the path-
ways in this cycle. Once the factors which determine the path-
way the nitrogen will follow are known, we can control the ni-
trogen movement to surface and ground water.
The variability in the nitrogen content of manure causes diffi-
culty in establishing application guidelines for all areas of the
United States. The above factors indicate that sampling to deter-
mine nitrogen content should be done as close as possible to the
time of application so that the sample will represent the nitrogen
content of the manure as it is applied to the soil.
In the next section we will see how the nitrogen content and its
availability affects the pathway the nitrogen takes in the nitro-
gen cycle.
Figure 2 is a schematic diagram of the various pathways the
nitrogen can follow once it has been added to the soil. From this
figure we see that when the manure is first added to the soil,
some of the nitrogen might be volatilized as ammonia (NH,) and
returned to the atmosphere. Warm, dry and alkaline conditions
favor the volatilization of ammonia gas.2 The nitrogen move-
ment in the runoff can be of any chemical form and usually
occurs on sloping land in connection with heavy precipitation,
snow melt on frozen ground, or overirrigation.
Once the manure is incorporated into the soil and starts to
decay, the nitrogen can follow several different pathways
depending upon the soil aeration, texture, water content, and
temperature. When the soil is well drained and aerated, the
decomposition is aerobic. The first step in aerobic decomposi-
tion is the hydrolysis of the nitrogen-containing proteins in the
manure to yield polypeptides, amino acids, and some ammo-
nia.2 This step is referred to as amination and makes available
the amine compounds which may be synthesized by micro-
organisms into their own protoplasm. Amine groups not used
by microbes or releasetions of the microbes die are converted to
ammonia (NH3). This process is termed ammonification.
Ammonia, if not lost by volatilization, is rapidly converted to
the ammonium ion (NHJ). As the nitrogen appears in the ammo-
nium form, it may be absorbed by plants or microorganisms or it
may be held by the negatively charged soil particles. The ammo-
nium ion is usually converted rather rapidly to the nitrate (NOj)
form. This is called nitrification. Nitrate is the end product of
aerobic decay. All compounds of nitrates are water soluble, and
the negative nitrate ion is not attracted to the negatively charged
soil particle. Thus, it moves easily with the soil water to be
leached to the groundwater. The nitrate form is also the most
easily taken up by the plant and so nitrogen in this form can be
recycled through cropping.
When the soil is cool and poorly drained, the decomposition is
usually aerobic and the end products are often gaseous com-
pounds of nitrogen such as nitrogen (N2) and oxides of nitrogen.
The nitrate from aerobic digestion can also be converted to gas-
eous compounds if anerobic conditions develop. This process is
called denitrification. However, denitrification is not likely to
occur below the root zone of the crop because this process needs
a source of oxidizable carbon not generally found at such depths
in the soil. Therefore, once nitrate gets below the root zone, it
moves to the ground water in that form.
If precipitation or irrigation water is insufficient, nitrogen
compounds can accumulate in the soil profile. The accumulated
nitrogen can be found in the form of organic compounds such as
proteins of plants or bacteria or in the inorganic forms such as
the nitrates. Figure 3 shows some research results where nitro-
gen has been concentrated in the soil profile.
Explaining the decay process has been simplified for the pre-
sentation here because it would take too long to cover in detail.
Many research papers have been written on each of the many
GASEOUS N COMPOUNDS
VOLATILIZED NH.
MANURE
DECOMPOSED
V
LEACHED
Figure 2. Pathways for nitrogen movement from decaying manure applied to cropland.
-------
Manure Disposal
159
pathways of the nitrogen cycle and their accompanying chemi-
cal reactions. However, our version should provide sufficient
background to understand some management practices to con-
trol water pollution from nitrogen in manure.
FRLL 1972
60-
;i20-
1180-
240-
300
H—n CONTROL
230 MT/Kfl
481 MT/Hfl
5 10 15 20
NITRflTE-NITROGEN. PPM
Figure 3. Nitrate nitrogen in soil profile after a single application
of beef feedlot manure in the fall of 1969, Wallingford (25).
Management Practices to Control Nitrogen Pollu-
tion
Nitrogen pollution can best be controlled by sending it down
the least polluting pathway of the nitrogen cycle. If we return to
Figure 2, we can see several pathways we do not want the nitro-
gen to follow. Obviously we do not want it to flow into surface
streams in runoff nor do we want it to leach into our ground
water. It could accumulate in the soil if it didn't become toxic to
plant and animal life, and if it didn't later leach downward to the
ground water. It appears difficult to control nitrogen accumula-
tion so this may not be a good pathway either.
We could leave the manure on the soil surface to volatilize as
ammonia, but then the manure would be subject to movement to
streams in runoff. Also, Azevedo and Stout2 feel that volatilized
ammonia could be transported to other areas and reabsorbed by
plants or surface water downwind from the application site.
Of the remaining pathways of denitrification and plant
intake, the latter is most preferred because we have a wealth of
information on nitrogen use by various crops. We also have
information about denitrification and anerobic decay process
although it is not so easily controlled as the uptake by plants.
However, if we have enough experience with anerobic decompo-
sition and denitrification to know how much nitrogen can be
returned to the atmosphere as a gas, then we could count on this
process to remove some nitrogen added to the soil.
What this all boils down to is the question of what manage-
ment practices we can use to move the nitrogen down the path-
way to plant uptake, or possibly, denitrification and anerobic
decomposition. These management practices are timely incor-
poration of manure into the soil to reduce runoff and volatiliza-
tion and application of manure in amounts which allow all of the
nitrogen applied to be recycled by the soil-plant system so no ni-
trogen would accumulate in the soil profile or leach to the
ground water.
Now we have to determine the proper application rate of
manure. If we know the nitrogen content of the manure and the
rate at which it decays to release the nitrogen, i.e., the decay ser-
ies, we can calculate the application rate. This sounds simple
enough except that we saw earlier how much the nitrogen con-
tent of the manure varies. Why not use an average nitrogen con-
tent? Would you buy yourself a pair of shoes the size of the aver-
age foot in the United States? Of course you wouldn't. You must
match the pair of shoes to your foot just as we must match the
manure to our soil and climate.
Large errors can be made if average nitrogen concentrations
are used to calculate application rates. For example, assume
that the average nitrogen content of solid poultry manure is 5%.
If we applied the poultry manure of Table II with the maximum
value of 11 % nitrogen, based on the 5% average, the application
rate would be 2.2 times larger than what it should be. If the
actual nitrogen content were 1.1% (the lowest value in Table II)
then the application rate based on the 5% average value would
be less than one quarter of what could be recycled by the system.
This illustrates the importance of knowing the nitrogen content
of the manure applied to your soil and that you must know your
soil and crop at the application site.
The application rate affects substantially the pathway the ni-
trogen will follow in the nitrogen cycle. As seen in Figures 4 and
5 of, 24 overapplication of manure can reduce the ability of the
crop and soil system to recycle nitrogen. Not only is the potential
for nitrogen pollution increased by the additional manure
added, but the system's ability to recycle nitrogen has been
impaired. Some operators like to apply heavy amounts of
manure every three or four years and let the system recover
between applications. It has been shown that such a practice
may temporarily overload the system's ability to recycle the ni-
trogen and thus increase the potential for groundwater pollution
(see Figure 3 and Ref. 17).
FflLL 1971
"0 250 500 750 1000 1250
flCCUMUlflTIVE MRNURE flPPLICflTION. MT/Hfl
Figure 4. Corn forage yields as affected by application of beef
feedlot manure that began in the fall of 1969, Wallingford (24).
We have discussed considerably the variability of the nitro-
gen content of manure and the factors affecting the pathways the
nitrogen will take in the nitrogen cycle. We also have decided to
send the nitrogen down the less polluting pathway. We have not
-------
160
Manure Disposal
discussed specific management practices for given areas of the
United States because of the limited time and space. However,
we should now have enough background to look at some of the
recent guidelines and manuals on the application of manure to
land.
Several publications on animal waste disposal have appeared
the past few years. Most are based on the concept of trying to
recycle the nitrogen and secondly, of trying to avoid a buildup of
toxic substances such as ammonium, inorganic salts, and heavy
metals. Good guidelines must consider the area climate, soil
type, crop and nitrogen release of the manure. Several states,
some of which are Indiana,3 North Carolina,9,10,",12 Maine23
and Kansas13,14 have developed guidelines on animal waste
applications. Formulas have been developed for applying
manure to land based on the decay series for a given manure and
region.15,16 Two publications give the general concept of manure
application to land.2,16 There are also regional guidelines which
may help you.4,7,21
32CH • ' • ' • • ' —
a:
:n
UJ
cr
f—
Q_
o
o
cc
GO
0-
KflLL 197
'0 250 500 750 1000 1250
RCCUMULflTIVE MflNURE flPPL I CRT ! ON , HT/H.P
Figure 5. Nitrogen uptake of forage corn as affected by applica-
tions of beef feedlot manure that began in the fall of 1969, Wal-
lingford (24).
SUMMARY
We have discussed the variability of nitrogen concentration in
manure and its effect on the pathway the nitrogen will take in the
nitrogen cycle. We have also found that the soil, climate, and
application rate influence this pathway. The best management
system apparently is one which recycles the nitrogen in the
manure either through plant uptake or denitrification and ane-
robic decomposition. Further, we found no universal manage-
ment system to control nitrogen.pollution, but because of the
variable nitrogen content, each application of manure must be
considered an individual case. Several publications are now
available or in process to provide guidelines for sound manage-
ment decisions.
REFERENCES
1. Adriano, D. C, P. F. Pratt and S. E. Bishop. Fate of
organic forms of nitrogen and salt from land-disposed manures
from dairies. Livestock Waste Management and Pollution
Abatement, Proc. Int. Symposium on Livestock Wastes Amer.
Soc. Agric. Engr. St. Joseph, Michigan. 1971 pp. 243-246.
2. Azevedo, J. and P. R. Stout. Farm Animal Manures: An
overview of their role in the agricultural environment. Agric.
Expt. Sta. Man. 44. University of Calif., Berkeley. 1974.
3. Cooperative Extension Service, Purdue University. Waste
handling and disposal guidelines for Indiana beef producers.
1D-84, Lafayette, Ind. 1972.
4. Gilbertson, C. B., (private communication), Dept. of Agric.
Engr., University of Nebraska, Lincoln. 1976.
5. Hutchinson, D. E. (editor). Resource Conservation Glos-
sary. J. Soil and Water Conservation 31: lg-63g. 1976.
6. Mathers, A. C. (private communication), USDA South-
western Great Plains Research Center. Bushland, Texas. 1976.
7. Meek, B. (editor) Guidelines for manure use and disposal in
the western region, U.S.A. College of Agriculture Research
Center, Washington State University, Bui. 814, 1975.
8. Midgley, A. R.and Vr L. Weiser. Effect of superphosphates
in conserving nitrogen in cow manure. Vermont Agric. Expt.
Sta. Bui. 419, 1937.
9. North Carolina Agricultural Extension Service, North
Carolina State University. Beef cattle waste management alter-
natives. Cir. 571, Raleigh. 1973.
10. North Carolina Agricultural Extension Service, North
Carolina State University. Dairy waste management alterna-
tives. Cir. 568, Raleigh, 1973.
11. North Carolina Agricultural Extension Service, North
Carolina State University. Poultry waste management alterna-
tives. Cir. 570, Raleigh, 1973.
12. North Carolina Agricultural Extension Service, North
Carolina State University. Swine waste management'alterna-
tives. Cir. 569, Raleigh, 1973.
13. Powers, W. L., R. L. Herpick, L. S. Murphy, D. A. Whit-
ney, H. L. Manges and G. W. Wallingford. Guidelines for land
disposal of feedlot lagoon water. Coop Ext. Serv. Circular C-
485. Kansas State University, Manhattan, 1973.
14. Powers, W. L., G. W. Wallingford, L. S. Murphy, D. A.
Whitney, H. L. Manges and H. E. Jones. Guidelines for apply-
ing beef feedlot manure to fields. Coop. Ext. Serv. Circular C-
502, Kansas State University, Manhattan, 1974.
15. Powers, W. L., G. W. Wallingford and L. S. Murphy. For-
mulas for applying organic wastes to land. Journal of Soil and
Water Conservation. 30:286-289, 1975.
16. Powers, W. L, G. W. Wallingford and L. S. Murphy.
Research Status on effects ofland application of animal wastes.
Report number EPA-660/2-75-010. Natl. Environ. Res. Center
Office of Res. and Development. U. S. Environmental Protec-
tion Agency, Corvallis, Oregon, 97330. 1975.
17. Pratt, P. F. (private communication) Dept. of Soil Science
and Engineering, University of California, Riverside, Califor-
nia. 1976.
18. Pratt, P. F., F. E. Broadbent, and J. P. Martin. Using
organic wastes as nitrogen fertilizers. California Agriculture.
June 1973. pp. 10-13.
19. Salter, R. M. and C. J. Schollenberger. Farm Manure.
Ohio Agric. Expt. Sta. Bui. 604, Wooster. 1939.
20. Smith, R. J., T. E. Hazen and J. R. Miner. Manure man-
agement in a 700-head swine-finishing building; two approaches
using renovated waste water. Livestock Waste Management and
Pollution Abatement, Proc. Int. Symposium on Livestock
Wastes. Amer. Soc. Agric. Engr., St. Joseph, Michigan. 1971.
pp. 149-153.
21. Stewart, B. A. (editor) Control of water pollution from
cropland, Volume I a manual for guideline development.
Office of Research and Development U. S. Environmental Pro-
tection Agency, Corvallis, Oregon. 1975.
22. Taiganides, E. P. and T. E. Hazen. Properties of farm
animal excreta. Transactions of the ASAE 9:374-376. 1966.
23. University of Maine. Maine guidelines for manure and
manure sludge disposal on land. Misc. Rpt. 142, Orona, 1972.
24. Wallingford, G. W., Effects of solid and liquid beef feedlot
wastes on soil characteristics and on growth and composition of
corn forage. Ph.D. thesis, Kansas State University, Manhattan,
1974.
25. Wallingford, G. W., L. S. Murphy, W. L. Powers and H.
L. Manges. Disposal of beef feedlot manure. Effects of residual
and yearly applications of corn and soil chemical properties J
Environmental Quality. 4:526-531, 1975.
-------
Fate of Nitrogen
from Municipal Sludges
Larry D. King
Department of Soil Science
NC State University
INTRODUCTION
The increasing interest in land application of sewage sludge in
the past ten years has led to a considerable amount of research
on the effects of agricultural use of sludge. Although the heavy
metal content of sludge will determine the total quantity that can
safely be applied over the lifetime of a site, it is generally agreed
that the nitrogen content will dictate the annual rate of sludge
application. Consequently, it is important to know the nitrogen
content of sludge, the forms of nitrogen present, the nitrogen
transformations that occur when sludge is applied to soil, and
the pathways the nitrogen follows after application.
Nitrogen Content
One of the first steps taken in determining a suitable applica-
tion rate for a particular sewage sludge is the analysis of a sample
for total nitrogen content. However, an important characteristic
of sludge is the fact that the nitrogen content can vary consid-
erably among treatment plants and with time at the same treat-
ment plant. An example of variation with time is shown in Fig-
ure 1. The data depicted are the total nitrogen content of liquid
digested sludge from Chatham, Ontario (unpublished data).
During the six month period nitrogen content fluctuated by a
factor of three. During a one week period in October there was a
factor of two fluctuation.
TOTAL N
mg/ I
3000.
2500.
2000.
1500 -
1000.
JUNE ' JULY AUG SEPT OCT NOV
1971
Figure 1: Variation of total nitrogen content of liquid sewage
sludge from Chatham, Ontario (unpublished data).
In an effort to obtain sludge with a constant nitrogen content
for field experiments, sludge was withdrawn over a two day
period from the same depth in a digester at the wastewater treat-
ment plant in Guelph, Ontario.2 The nitrogen content of the
sludge ranged from 2900 to 3580 mg/1 or 23% of the lower value.
Since it is important to apply sludge nitrogen at a reasonably
precise rate to preserve environmental quality and to assure
acceptable crop production, it is evident that management steps
must be taken to produce a more homogeneous sludge at treat-
ment plants where land application is practiced.
Nitrogen Forms
Since sludge undergoes perhaps 30 days of anaerobic diges-
tion, nitrogen mineralization stops at the ammonium step
because oxygen is required to proceed from ammonia to nitrate.
Consequently, liquid sludge from the digester contains ammo-
nium nitrogen and a host of protein, amino acid, etc. nitrogen
forms which we lump together and call organic nitrogen. Table 1
shows values of total nitrogen and soluble ammonium nitrogen
gathered from various references. The organic material in sludge
has undergone decomposition in the digester and has developed
some of the same properties of soil organic matter. One of these
properties is cation exchange capacity. Since ammonium nitro-
gen exists as a cation (NH4+) we find that in addition to the
ammonium nitrogen in solution there is also some held on the
cation exchange complex of the sludge. The organic nitrogen
content can be determined by subtracting the soluble and
exchangeable ammonium nitrogen from the total nitrogen.
The data in Table I point out the variation in nitrogen among
treatment plants. Also of interest is the variation in ammonium
nitrogen expressed as a percentage of total nitrogen. This varia-
tion is probably a function of such factors as digestion time, type
of inputs to the sewer system, and type of secondary treatment
(e.g. activated sludge vs trickle filter sludge).
Transformations and Subsequent Pathways
Ammonia
Under alkaline conditions, ammonium ions (NH4+) give up a
hydrogen ion (H+) and become ammonia (NH3). Ammonia
volatilizes readily and thus can be lost as a gas. This is commonly
what happens to a portion of the ammonium nitrogen in sludge
because sludges are generally slightly alkaline.
Ryan and Keeney17 found that when containers of liquid
sludge were allowed to air dry in the laboratory approximately
90% of the ammonium nitrogen was lost via volatilization. Since
the ammonium ion is positively charged it can be adsorbed onto
the cation exchange complex in the soil. Consequently, ammo-
nia loss from soil-applied sludge is less than that from sludge
161
-------
162
Municipal Sludges
dried in the absence of soil. Ryan and Keeney17 reported lower
ammonia losses with soil-applied sludge than with container-
dried sludge. The losses were inversely proportional to the clay
content of the soil—an indirect measure of cation exchange
capacity:
Soil
Sand
Sandy loam
Clay loam
%NH-Nloss
29
18
13
Other laboratory studies", l5 have shown ammonia losses rang-
ing from 4 to 36% of the applied ammonium nitrogen. Through
periodic soil sampling, Stewart et a/.19 estimated a 30% loss of
applied ammonium nitrogen from liquid sludge applied for corn
production.
The more intimately sludge is mixed with soil after applica-
tion the lower the loss of ammonia. Thus, ammonia is lost most
readily from surface applications and less so when sludge is ap-
plied and then plowed or disked into the soil. Subsurface injec-
tion behind some type of tillage implement essentially eliminates
ammonia loss.
Generally, the volatilization occurs rather rapidly after the
sludge is applied. Figure 2 shows that most of the loss from a sur-
face application occurred during the first week after application.
The ammonium nitrogen that is not lost via volatilization can
move through several pathways:
I. Direct uptake by crops.
2. "Fixed" within the crystal structure of certain clays thus,
rendering it unavailable for crop uptake or further bio-
logical action.14
3. Temporarily adsorbed onto the cation exchange sites in
the soil.
Table I. Nitrogen content of anaerobically digested liquid sewage sludge.
Total
N
mg/1
2250
2180
2710
3250
2620
2090
1620
2240
2350
3160
2300
Range
1620-3250
Mean
2430
Soluble
mg/1
520
340
550
490
690
1020
950
650
950
840
700
340-1020
700
NH4-N
% of total N
23
16
20
15
26
49
59
29
40
27
30
15-59
29
Exchangeable
NH4-N Reference
mg/1
7
7
10
11
12
15
17
17
18
110 19
60 L. D.
(unpublished
85
King
data)
-------
Municipal Sludges
163
4. Converted by nitrifying bacteria to nitrate which may
then be taken up by plants, leached out of the soil, or
converted to nitrogen gas via denitrification.
WEEKS
Figure 2: Ammonium nitrogen loss from surface-applied sewage
sludge."
It should be pointed out that nitrification is an acidifying pro-
cess:
NH4+ + 202 -> NO3"
K,O
the application of sludge will therefore increase soil acidity8
unless the sludge contains sufficient cations (Ca, Mg, etc.) to
pffset the acidifying effect of nitrification. Consequently, an
adequate liming program is essential in managing a land appli-
cation site.
Denitrification occurs when nitrate moves into an environ-
ment that is devoid of oxygen but contains sufficient available
carbon for anaerobic bacteria to use as an energy source to
convert the nitrate to nitrogen gas. The application of sludge to
soils can bring about these conditions. We don't usually think of
nitrification (an aerobic process) occurring simultaneously with
denitrification (an anaerobic process). However, it has been
found that in a normally aerobic soil there can be small areas or
microsites which are anaerobic. The addition of organic mate-
rial such as sludge can produce anaerobic zones within the soil
due to oxygen depletion caused by decomposition of the organic
material. Thus, sludge additions supply nitrogen to be nitrified
in aerobic sites and organic material which can deplete oxygen
and provide available carbon in other sites. As a result, some of
the sludge-applied nitrogen can be lost by denitrification.
Ryan et al.18 found that where high rates of sludge were incu-
bated with soil from 24 to 38% of the applied nitrogen was lost
via denitrification. In an incubation study10 and a greenhouse
leaching study" I found denitrification losses ranging from 15 to
27% of the applied nitrogen.
Organic Nitrogen
The anaerobic digestion process used at wastewater treatment
plants converts most of the readily available organic nitrogen in
the sludge to ammonium nitrogen. The remaining organic mtro-
een is relatively inert and resistant to further mineralization.
Premi and Cornfield", '« reported that only about 3% of the
organic nitrogen was mineralized during laboratory incubation
studies with sludge. Ryan et al. '8 reported 4 to 48% mineraliza-
tion of organic nitrogen during a 16-week incubation study.
They found that as the sludge application rate increased, the per-
cent mineralization decreased. Larson et al.l3 used relative plant
response to estimate an annual mineralization rate of organic
nitrogen of 5%. Stewart et al.20 found very little mineralization
of organic nitrogen in solids separated from liquid sludge or in
solids collected from plots which had received liquid sludge 6
months earlier.
In a laboratory incubation study, I found that about,40% of
the organic nitrogen in sludge was mineralized during an 18
week period.10 In a field study9 samples of sludge crust that had
accumulated from surface sludge applications the previous two
summers were collected. From these samples it was determined
that 33% of the applied organic nitrogen had been mineralized.
It should be remembered that the availability of organic nitro-
gen in sludge will depend on how well the sludge has been stabil-
ized in the digestor. Poor stabilization will result in lower
ammonium/organic nitrogen ratios and consequently organic
nitrogen which is more available for mineralization.
From the above data, I think we can draw the general conclu-
sion that the nitrogen in liquid sludge exists in two distinct
phases: (I) The ammonium form which can be readily taken up
by plants, volatilized, nitrified, etc., (II) The organic form which
mineralizes slowly. Once the organic nitrogen is mineralized it
follows the same pathways mentioned earlier for ammonium
nitrogen.
The low availability of organic nitrogen in sludges leads one
to question the practice of dewatering the sludge prior to land
application. This practice greatly reduces the amount of
nitrogen available for crop production. It also imposes an
increased nitrogen load on the treatment plant since the high ni-
trogen supernatent (or filtrate) must be recycled back through
the plant. Removal of the liquid fraction also reduces the nitro-
gen/heavy metal ratio and makes the sludge less suitable for
crop production.
Application of Research Results
One of the purposes of the research efforts that have been
cited (plus a vast number that have not been cited) is to aid regu-
latory agencies in setting up guidelines for use of sludge on land.
Hopefully, these regulations will not only protect environ-
mental quality but will also allow maximum utilization of the re-
sources in the sludge.
Various agencies have or are developing guidelines for sludge
application. I have used the nitrogen guidelines from several
sources to construct Table II. It should be stressed that all but
the Wisconsin guidelines are provisional and could be revised.
Also the guidelines have special restrictions based on heavy
metal levels. These restrictions could limit sludge application to
lower rates than might be calculated from Table II. However,
the data are presented for comparison purposes with the
assumption that heavy metals are not a limiting factor.
All the guidelines except those from Ontario take into con-
sideration ammonia volatilization in determining ammonium
nitrogen availability. However, there is considerable variation
among agencies. In addition to the effect of soil incorporation
the Illinois guidelines also consider the effect of cation exchange
capacity as evidenced by the soil texture variable.
Organic nitrogen availability is described as a decreasing ser-
ies over a period of 4 to 5 years. The total availability over a five
year period is the same for EPA and Wisconsin but Illinois sug-
gests a much higher availability.
The Wisconsin and Illinois guidelines state that sludge will be
applied at a rate such that the available ammonium and availa-
ble organic nitrogen will supply nitrogen equal to the recom-
mended fertilizer nitrogen requirement for the specific crop. The
EPA guidelines suggest twice the nitrogen requirement of the
crop. This would be approximately the recommended fertilizer
nitrogen requirement. Ontario limits sludge application to the
-------
164
Municipal Sludges
amount that would apply 120 Ib/acre of ammonium nitrogen.
This could be applied once every four or five years depending on
the crop, or it could be split into a series of smaller applications
not to exceed a total of 120 Ib/acre over the specified time
period. This low application rate is an effort to distribute the
sludge over a larger land area to better utilize the sludge phos-
phorus and to limit heavy metal accumulation. This more con-
servative approach negates the need to be concerned about
availability of the organic nitrogen.
Using the data in Table II calculations were made to deter-
mine the volume of liquid sludge and the resulting amounts of
sludge-applied total nitrogen and ammonium nitrogen that
could be applied to a corn crop (nitrogen requirement: 120 Ib
nitrogen/acre/year) over a five year period. The results are
shown in Table III. Although EPA, Wisconsin, and Illinois
allow the same available nitrogen application rate, the more
rapid mineralization rate suggested by Illinois limits the sludge
application rate to 63% of that permitted by EPA and Wiscon-
sin. At first glance one might expect the application rate permit-
ted by Ontario to be 1 / 5 of that permitted by the other agencies.
Since Ontario does not consider mineralization of organic ni-
trogen it turns out that the Ontario rate is actually 55% of that
permitted by Illinois.
Table III. Total volume of sludge and pounds of nitrogen that
could be applied over a five year period using the regulations
for nitrogen application shown in Table II.
US EPA
Wisconsin
Illinois
Ontario
Sludge
Volume
gal/acre
120,000
61,200
38, WO
20,700
Assumptions: Sludge
Sludge-applied N
Total N NH^-N
Ib/acre Ib/acre
2^30 mg/1 total
1565
1590
1000
540
nitrogen
3^5
350
220
120
700 mg/1 ammonium nitrogen
1730 mg/1 organic nitrogen
Soil texture
Crop
loam
corn with 120
requirement
Ib/acre N
Table II: Guidelines for use of sewage sludge on agricultural land.
Agency
Annoniiin
Crfcanic nitrogen (CM)
availability
Kitrop;en(NH^-!") Year after application
Availability 1 2 3 4 5
5 year Recommended nitrogen
Total application rate
''3 Environmental Surface -50 15-20 333
Protection Agency ('4-)* Incorp. -100
'.'isconsin, Department Surface 50
of natural Resources Incorp. 100 15 (-. \\. 2
25
25
Illinois 2nvlronrnentf.il
Protection Agency (5;*
Surface 0
Shallow incorp.
sanely soil-50
non-sandv -30
Subsurface ap-
plication
sandy noil-5"
non-sandy -100
.'•0 ?0
10
2.5
rntario rinistries: 1.00
Agriculture and food
Snvironnent (l)*
i'Tot considered
Available (NH^-N+CN) + N
mineralized from soil not
to exceed tv?ice the soil
requirement
Available (NH;,-N 4CN) not
to exceed the 'fertilizer
N requirenent
Available (r^-NtCN) not
to exceed the ''fertilizer
1C requirement
Corn and Hay: 120 Ib/A
IIIij,"N once every 5 years
or 'equivalent.
Commercial sod: 1201b/A
KK^-N once every b years
or equivalent.
^Guidelines are provisional.
-------
Municipal Sludges
165
SUMMARY
Sludges applied to land have usually undergone an extended
period of anaerobic digestion. This process results in two basic
nitrogen forms: (1) ammonium nitrogen which is readily availa-
ble to plants; (2) organic nitrogen which mineralizes slowly and
is thus less available. The concentration of these forms can vary
greatly among treatment plants and with time at the same treat-
ment plant.
Much of the ammonium nitrogen can be lost via volatilization
if the sludge is not incorporated with the soil or subsurface ap-
plied. The remaining ammonium nitrogen and mineralization of
organic nitrogen provide nitrogen for crop uptake or loss by
leaching if excessive sludge is applied. A sizeable fraction of the
applied nitrogen may also be lost via denitrification.
Guidelines being developed by various regulatory agencies
generally base the sludge application rate on the amount of
available nitrogen the sludge will supply. The available nitrogen
is defined as a fraction of the ammonium nitrogen (a function of
volatilization) plus a fraction of the organic nitrogen (a function
of the rate of mineralization).
REFERENCES
1. Ad Hoc Joint Committees of Ontario Ministry of Agricul-
ture and Food and Ontario Ministry of the Environment. 1975.
Provisional guidelines for sewage sludge utilization on agricul-
tural lands.
2. Bates, T. E. 1972. Land application of sewage sludge.
Ontario Ministry of the Environment, Toronto, Ontario, p. 5-6.
3. Coker, E. G. 1966. The value of liquid digested sewage
sludge. I. The effect of liquid sewage sludge on growth and com-
position of grass-clover swards in South-east England. J. Agric.
Sci. Comb. 67:91-97.
4. Environmental Protection Agency [FRL 552-7] Munici-
pal sludge management. 1976. Fed. Register.
41(l08):22532-22536.
5. Illinois Environmental Protection Agency. Design criteria
for municipal sludge utilization on agricultural land. Tech. Pol-
icy WPC-3, draft copy.
6. Keeney, D. R., K. W. Lee and L. M. Walsh. 1975. Guide-
lines for the application of wastewater sludge to agricultural
land in Wisconsin. Tech. Bull. No. 88, Dept. of Natural Re-
sources, Madison, WI.
7. King, Larry D. and H. D. Morris. 1972. Land disposal of
liquid sewage sludge: 1. The effect on yield, in vivo digestability,
and chemical composition of Coastal bermudagrass (Cynodon
dactvlon L. Pers.). J. Environ. Quality 1:325-329.
8. King, Larry D. and H. D. Morris. 1972. Land disposal of
liquid sewage sludge: II. The effect on soil pH, manganese, zinc,
and growth and chemical composition of rye (Secale cereale L.).
J. Environ. Quality 1:425^29.
9. King, Larry D. and H. D. Morris. 1972. Land disposal of
liquid sewage sludge: III. The effect on soil nitrate. J. Environ.
Quality 1:442^46.
10. King, Larry D. 1972. Mineralization and gaseous loss of
nitrogen in soil-applied liquid sewage sludge. J. Environ. Qual-
ity 2:356-358.
11. King, Larry D. and H. D. Morris. 1974. Nitrogen move-
ment resulting from surface application of liquid sewage sludge.
J. Environ. Quality 3:238-243.
12. King, L. D., L. A. Rudgers, and L. R. Webber. 1974.
Application of municipal refuse and liquid sewage sludge to
agricultural land: I. Field study. J. Environ, quality 3:361-366.
13. Larson, W. E., C. E. Clapp and R. H. Dowdy. 1972. Inte-
rim report on the agricultural value of sewage sludge. USDA,
ARS and the Dept. of Soil Sci. Univ. of Minnesota, St. Paul.
14. Nommik, H. 1965. Ammonium fixation and other reac-
tions involving a non-enzymatic immobilization of mineral ni-
trogen in soil. In Bartholomew, W. V. and F. E. Clark (ed.) Soil
nitrogen. Agronomy 10:198-258. Amer. Soc. Agron., Madison,
WI.
15. Premi, P. R. and A. H. Cornfield. 1969. Incubation study
of nitrification of digested sewage sludge added to soil. Soil Biol.
Biochem. 1:1-4.
16. Premi, P. R. and A. H. Cornfield. 1971. Incubation study
of nitrogen mineralization in soil treated with dried sewage
sludge. Environ. Pollut. 2:1-5.
17. Ryan, J. A. and D. R. Keeney. 1975. Ammonia volatiliza-
tion from surface applied wastewater sludge. J. Water Poll.
Control Fed. 47:386-393.
18. Ryan, J. A., D. R. Keeney and L. M. Walsh. 1973. Nitro-
gen transformations and .availability of an anaerobically di-
gested sewage sludge in soil. J. Environ. Quality 2:489-492.
19. Stewart, N. E., E. G. Beauchamp, C. T. Corke and L. R.
Webber. 1975. Nitrate nitrogen distribution in corn land follow-
ing applications of digested sewage sludge. Can. J. Soil Sci.
55:287-294.
20. Stewart, N. E., C. T. Corke, E. G. Beauchamp and L. R.
Webber. 1975. Nitrification of sewage sludge using miscible dis-
placement and perfusion techniques. Can. J. Soil Sci.
55:467-472.
-------
Disposal of Industrial Nitrate Effluents by Means
of Forest Spray Irrigation*
W. F. Harris, G. S. Henderson, and D. E. Todd
Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee
INTRODUCTION
Trends in water pollution control practices recognize, in addi-
tion to technological solutions to effluent processing, that eco-
logical criteria to maintain environmental quality are neces-
sary.15 The ecological approach relies on maintenance of the
physical, chemical, and biological integrities of the landscape,
whereas the technological solution involves removal of mate-
rials to some maximum permissible discharge level based on
water quality criteria. Applicable legislation, which imposes a
timetable for compliance,1 also requires changes in pollution
treatment practices to meet new water quality standards.
Broadly categorized, these new treatment procedures involve:
(1) total elimination of waste discharge following development
of industrial recycle techniques; (2) imposition of land-use con-
trols for nonpoint pollution sources; and (3) recycling of
materials by land disposal.15 Because of low initial and annual
maintenance costs,2 land disposal of wastes is receiving wide-
spread attention as a means to meet higher water quality stan-
dards. Where land is available and engineering criteria can be
met, land disposal is justifiably a logical choice.
Land disposal utilizes the terrestrial ecosystems as a "living
filter" to renovate waste water by incorporating effluent mate-
rial into natural biogeochemical cycles. Land disposal requires
careful management because the introduction of materials into
the terrestrial environment must take into account the biologi-
cally determined capacity of the ecosystem to assimilate effluent
material, particularly when the effluent material occurs in the
environment in small quantities such as in the case of nitrate.
'Research sponsored by the Energy Research and Development
Administration under contract with Union Carbide Corporation.
Environmental Sciences Division Publication No. 959.
1 The amendments to the 1970-71 legislation (S. 2270 and H.R. 11896)
require elimination of point source discharge by 1981, implementation
of "best practicable" technology by 1976 and "best available" technol-
ogy by 1981. "Practicable" refers to the average of best available tech-
nologies and "best available" refers to the best performer.
: Based on data from a Corps of Engineers study of the Chicago met-
ropolitan area, initial costs of land disposal are from 50-80^ less than
phystcal-chemical and advanced biological treatment of domestic sew-
age and urban runoff. Maintenance costs of land disposal were esti-
mated to be less than 50(7 those of other systems (Sopper and Kardos
1973).
Environmental degradation resulting from improperly man-
aged land disposal has consequences similar to those presently
experienced in direct effluent discharge to aquatic ecosystems
because both procedures are limited by a discrete assimilative
capacity of the system for introduced materials. Numerous ex-
amples of land disposal of domestic sewage have been reported;
disposal efficiency of nitrate (efficiency being determined as the
proportion of effluent material incorporated in the biological
system compared to the total added) ranges up to 85%.13 Factors
affecting disposal efficiency are not well documented. Limited
information is available on the success of land disposal of highly
concentrated industrial nitrate wastes.
Nitrate-nitrogen accounts for a small amount of the total ni-
trogen in a terrestrial ecosystem. Nitrates are readily leached by
percolating soil water, as well as immobilized by vegetation and
microorganisms. Once introduced to the soil ecosystem either
through decomposition of in situ organic nitrogen or as nitrate
salts, nitrate can be: (1) rapidly leached to groundwater, (2)
absorbed by plants ( 0.1% annual of the total pool of soil ni-
trogen), (3) immobilized through assimilation and conversion to
organic forms by soil microbes, and (4) denitrified to gaseous
form, N2, by anaerobic soil microbes with diffusion to the at-
mosphere. Under predominately aerobic soil conditions, dissim-
ilatory denitrification accounts for no more than 10-15% of
annual nitrogen losses.2 Therefore, land disposal of nitrates
relies on conversion of inorganic nitrogen to an organic form
which is subsequently mineralized (transformed from organic to
inorganic form), recycled, and lost from the system at a slow
rate. However, a finite capacity for assimilatory denitrification
can be expected; the consequences of exceeding this capacity can
be severe.
Addition of nitrogen, an essential element for microbes as
well as plant growth, improves the nutritional quality of the
growth substrate (dead plant parts and plant residues compris-
ing forest litter) for microbes. Decay of organic residues (humic
compounds), [which are the principal "disposal sinks" for nitro-
gen], by microbial activity is accelerated. Transformation of
organic nitrogen contained in humic compounds to NO,
exceeds the rate of utilization by other ecosystem components
and loss of environmental quality occurs when groundwater is
contaminated by NOj-rich soil water. Degradation of the chemi-
cal and structural integrity of the disposal system could also
occur as the result of accelerated decay of both dead plant parts
and humic compounds.
This paper summarizes field experiments initiated in June
1972 to evaluate the feasibility of land disposal of high nitrate
salvage solutions. Pollution abatement procedures for Energy
Research and Development Administration operations at Oak
Ridge, Tennessee, currently are being updated in accordance
166
-------
Forest Spray Irrigation
167
with federal legislative directives for higher water quality stan-
dards. Among the local waste practices examined was the
disposal of acidic effluents resulting from the operations of the
Oak Ridge Y-12 plant. As a consequence of many processing
operations, large quantities of waste solutions with high nitrate
content require disposal. These wastes were being transported to
disposal ponds (Figure 1). Seepage from the storage ponds and
entry into headwaters of Bear Creek constituted an immediate
problem to be solved while longer term concerns involved: (1)
reduction in total volume of acidic effluent by means of inplant
recycling, and (2) elimination of any point source discharge by
means of alternative disposal practices.
Land disposal was suggested as both: (1) a backup system to
inplant recycling procedures, and (2) a permanent system to dis-
pose of wastes receiving inplant treatment to reduce total efflu-
ent volume and nitrate concentration. The purpose of these
experiments was to determine:
1. the overall annual efficiency of landscape disposal at sev-
eral application rates;
2. major biotic and abiotic accumulators of effluent NO3-N;
3. rates of nitrogen loss; and
4. the optimum rate of nitrate application without loss of
water quality or degradation of the terrestrial ecosystem.
Experimental Procedures
A series of 16 circular plots (each 0.088 acre, — 0.04 ha) was
established in a mixed deciduous forest along the 1010-ft con-
tour north of Bear Creek (Figure 1). In order to evaluate the
capacity of the forest to incorporate nitrate, nitrogen concentra-
tion values were determined for the terrestrial ecosystem com-
ponents (represented by boxes) in Figure 2. The nitrogen budget
for the forest ecosystem, which is calculated from these data,
serves as the basis for determining the assimilative capacity of
the forest ecosystem for additional nitrogen. Periodic sampling
of ecosystem components was used to determine seasonal and
annual rates of nitrogen transfer (represented by arrows in Fig-
ure 2) about the system. The existence of many extant data from
previous and ongoing Environmental Science Division (ESD)
research projects on terrestrial element cycling minimized the
amount and frequency of additional sampling needed.10 Direct
measurements of litter organic matter pools and nitrogen con-
centrations of both litter and soil were required, since these com-
ponents are known to have large spatial variability.
Forest biomass was estimated from allometric relationships
of tree diameter and weight of tree components (e.g., bole,
branch, foliage, stump). Estimates of atmospheric input of ni-
trogen (~7.8 kg/ha annually) were obtained from the ongoing
,^5S>^^( / \ ^3^=^ei
^**\^^' -^-"
-^^.^'Z^fnSty^-*^^ -i eeeNp
noo< ."'• 'fft™"* J---' •^^"^ wqir ex >«V*W0 *o«r
. — — ;.' W.rxA'^ f*~^ t )""iu*! innoH • Htsituru i
•."«'* .x1, '//N ""'•» J—-' • fif»vr ex wemtrnt t*
^— — •;!•' V-/«MW* f*^ • )""iu*l innoH- Httt,
*~.t~.j<^53'*^^ '• / $> *fi*4TttttviiNQ are
Valley Showing Location of Hardwoods Spray Irrigati
•^
Figure 1: Area of Uppei
the 1010-foot Contour
North of Bear Creek Road Along
-------
168
Forest Spray Irrigation
Walker Branch Watershed Project located about one-half mile
southwest of the irrigation site.1 Nitrate, ammonia, and total-N
(NH4 and organic N) losses from each plot at Bear Creek were
measured by extracting samples of percolating soil water at 75-
cm depth with ceramic cup lysimeters. The porous ceramic cups
have an entry value of 1 bar and, therefore, sample the
gravitational soil water leaving the zone of major biological
activity. Nitrate flux in soil water could be estimated from simu-
lation models of water movement in the plant soil system as
influenced by climatic variables, precipitation, temperature and
solar radiationl4 or from known values of subsurface drainage in
similar forested environments. Measured drainage from Walker
Branch Watershed, a long-term environmental watershed study,
was used in the present analysis. The experimental design pro-
vided for four replications of a control and three treatments (50,
125, and 200 kg NO,-N per ha). The maximum application was
that required for disposal of current salvage solution discharge
on 1000 acres (404 ha), while the minimum rate represented the
amount annually accumulated in vegetation. The intermediate
rate was included to evaluate the nature of the forest responses
to NO3-N amendments.
Irrigation involved neutralization of salvage solution (con-
taining 12,500 ppm nitrate) with lime hydrate to precipitate
aluminum and other elements. The resultant supernatant of cal-
cium nitrate was transported by truck to the irrigation site (efflu-
ent composition is summarized in Table I). Three applications
(spray irrigation) of similar proportion were required to achieve
the desired total N-amendment. Operations began on June 27,
1972, and the final application was delivered on August 24,
1972. Similar irrigation regimes were followed in 1973. Applica-
tions coincided with periods of depleted soil moisture to mini-
WATER
mize rapid leaching of the highly soluble nitrate. By increasing
the residence time of nitrate amendments in the upper soil zone,
an increased amount of assimilatory denitrification (microbial
immobilization) could be attained.
Table I: Summary of Effluent Composition and Application
Utilized in the Bear Creek Irrigation Experiment.
June 27
July 27
August 14D
pH
10.2
10.8
3.6
200
27
89
84
Treatment
(Ib NOi per acre)0
500
161
177
168
800
292
265
250
d
b
L
Element
Al
Ca
Cr+s
K
Li
Mg
Na
Cl
CO 3
F
N03
SO,
P0»
One pound N03 per acre
Lower pH is the result
salvage solution.
Transport tanker water
ppm
59.0
5000.0
0.1
40.0
10.0
0.3
148.0
21.0
546.0
0.5
11514.0
<15.0
< 0.1
= 0.25 kg N03-N per ha.
of addition of nitric acid neutralized
sample 7/27/72.
CHEMICALS
ATMOSPHERE 1
ATMOSPHERE 1
PRECIPITATION! EVAPOTRANSPIRATION RAIN SCAVENGED
1 ^
i i_ P
VEGETA
THROUGHFALL, TRANSPIRATION # | LEACHATE,
z in
rnl, 1 0 fc.
DECOM
PERCOLATION
EXCHANGE R
"" """" t-rnrAun m« l«—
.. oTRCAMFLOW | SOJL SOLUT,ON
DRY PARTICULATES
, LEAF FALL
LITTER V*
POSITION,
SOIL
IACTIONS,
1
1
MAMMAL
CONSUMPTION
UPTAKE ]
SOIL- WATER I
SYSTEM 1
SEEPAGE! , SEEPAGE,
j~ DEEP LOSSES "1 ^SfflST1 |
CONSUMER j
FOOD WEB I
r_..L...,
EXIT |
r ^
DEEP LOSSES 1
1 __l
ALLOCHTHONOUS I
LEAF FALL-LEACHATE
MATERIAL 1 "
^ ECOSYSTEM !^~"
Figure 2: Compartment Diagram of Major Components (boxes) and Flows (arrows) in a Terrestrial Element Cycle. The present stud-
ies focus on the enclosed portion of the figure. Estimates of atmospheric input, vegetation leachate, etc., were obtained from Walker
Branch Watershed studies. The exit point from the irrigated ecosystem is soil water percolating beyond the limits of the rooting zone
(approximately 60 cm).
-------
Forest Spray Irrigation
169
Results and Discussion
Forest Nitrogen Cycle - Pretreatment
Apportionment of nitrogen in the forest ecosystem depends
on both biomass and nitrogen concentration in ecosystem com-
ponents. The mean biomass of living vegetation (1.8 x 10s
kg/ha) is of similar magnitude as other intensively studied local
forests and is summarized in Table II. Estimates of root bio-
mass pool size, mean annual net production, and turnover
through annual root death were based on general forest growth
patterns determined in other ongoing BSD research programs.
Values of N-concentration were determined for the Bear Creek
disposal site following standard procedures or based on local
values reported.10 Limited additional sampling of foliage and
litter for N-concentration at the Bear Creek site was necessary to
accurately establish baseline conditions since some
enhancement of nitrogen uptake by plants was anticipated,
which subsequently will be reflected in litter input and accumu-
lation of N in woody material. Estimates of forest litter weight
are averages (N = 16) for the irrigation area; no significant
differences existed among plots.
Table II: Summary of Forest Organic Matter Pools; Net,
Annual Accumulation, and Nitrogen Content (% Dry Weight)
Forest Standing pool
component (kg/ha)
I content
(*)
Net annual biomass
accumulation (kg/ha)
Foliage
Branch
Bole-stump
5,500a
32,450
123,800
Lateral root
<0.5 cm 10,050C
>0.5 cm 10,050
Litterfall
Leaf
Branch
Litter
Oi
0;
Wood
4,090f
4009
6,875
11,900
3,250
1.26
0.42
0.17
0.84
0.42
0.60
0.42
1.42
2.43
0.54
3,536°
5,970"
(l,450)e
% carbon
(Reichle et a). 1972)
46.5
38.7
47.0
dEstimates of forest biomass density were calculated from the diameters
on irrigation plots.
Net annual aboveground biomass accumulation based on average for similar
forests on Walker Branch Watershed.
cBased on analyses on Walker Branch Watershed, lateral root biomass
density is equal to stump biomass density (20,100 kg/ha).
Total root biomass accumulation equals annual root death of roots
<0.5 cm (^0% of standing pool <0.5 cm) plus 0.16 of net aboveground produc-
tion (including foliage); proportions are based on extensive data collected
in conjunction with Walker Branch Watershed.
^Represents net biomass accumulation (0.16 of total aboveground production).
fLitterfall is i<74% of peak leaf biomass density due to tronslocation of
material from the leaf prior to abscission.
^An average value for similar forests on Chestnut Ridge.
Forest biomass represented only 8.7% of the total terrestrial
nitrogen; dead organic matter in forest litter accounted for
another 6.5%. The nitrogen associated with mineral soil com-
prised the bulk of the terrestrial nitrogen (Table III). Of the 6290
kg/ha of total nitrogen, nitrate-nitrogen accounts for 2.4% and
is almost exclusively associated with soil. Nitrate-nitrogen
amendments during effluent irrigation therefore were 0.32,0.80,
and 1.28 times as large as the mean pool of NO3-N.
While N-pool sizes are based on biomass parameters specific
to the Bear Creek site, atmospheric input, rate coefficients for
vegetative uptake, root decay, and total system losses are based
on forest responses determined during the course of other
related research on forest ecosystems (Table III). Calculation of
forest budgets is based on conservation of mass; the sum of
inputs and initial compartment size equals the sum of final com-
partment value and all losses from that compartment. For ex-
ample, estimates of litter decay assume that the size of litter
compartments and annual decay fluxes change very slowly from
year to year. Other assumptions involved in budget calculations
are:
1. all vegetative N-uptake is from the soil compartment;
2. dissimilatory denitrification loss equals 15% of total
system losses and is equal to microbial N-fixation. Both
processes proceed slowly in the forest ecosystem;3 and
3. the rate coefficient for measured system losses in soil water
is assumed to be proportional to soil nitrogen; this implies
that the mineralization of soil-N is substrate dependent
and mineralization reactions follow the Law of Chemical
Equilibrium.
Analysis of annual N-transfers among ecosystem components
indicates that the largest fluxes of nitrogen are associated with
decay of forest litter and root organic matter (129 kg N per ha
per yr), and uptake by vegetation (120 kg N per ha per yr). Decay
and vegetation uptake account for 93% of total annual nitrogen
transferred within the system (Table IV). The forest ecosystem
represents a close coupling of chemical and biological processes
such that the overall ecosystem behavior is extremely conserva-
tive with respect to nitrogen; only 2.5 kg N per ha per yr of a total
transfer among compartments of 266 kg N per ha per yr is lost
from the system. The natural forest ecosystem conserves 99% of
this annual N-flux (Table IV). Land disposal of nitrate requires
manipulation of this efficient set of processes. Total inputs to
litter annually amount to 38 kg N per ha per yr; NO3-N
amendments in current experiments elevate this input to 88,163,
and 238 kg N per ha per yr—up to a 619% increase over natural
input rates.
Soil Water Quality
The efficiency of forest nitrogen cycling following effluent
application was determined from analysis of nitrogen content of
water exiting the rooting zone as gravitational soil water. No
sustained significant elevation in NO3-N concentration follow-
ing effluent applications up to 200 kg NO3-N per ha per yr was
observed; Figure 3 summarizes average irrigation plot NO3-N
concentrations of soil water and control data from Bear Creek
control plots and similar forest plots on Walker Branch
Watershed. Elevation of NO3-N concentration was coincident
with each date of effluent application; concentrations did not
exceed 0.5 ppm and occurred during a period of depletion of free
soil water due by seasonally high evapotranspiration. Each
application was closely followed by significant rainfall (Figure
3). The effect of summer rainfall, while adding insignificantly to
the amount of free soil water, would readily leach nitrate into the
soil profile. Elevation of NO3-N in soil water from control areas
also occurred (June 23) following rainfall totaling about 1.10 in
(2.8 cm). The periodicity of elevated NO3-N content closely
follows control plot responses. The responses observed follow-
ing effluent irrigation was confounded with typical responses
following significant amounts of summer rainfall which fol-
lowed irrigation operations.
3 Broadbent and Clark (1965) have summarized results from lysime-
ter studies where unaccounted nitrogen averaged 15%. Since measured
losses in deep seepage from the forest ecosystem are small in comparison
to other N-fluxes and Todd (1972) reports similarly small N-
amendments in forest soils due to fixation of atmospheric nitrogen,
these two flux pathways are assumed to be small, equal, and therefore
offsetting.
-------
170
Forest Spray Irrigation
Table HI: Forest Nitrogen Cycle. Compartment listing summarizes the average values for total-N (forest-litter components)
and NOj-N and NH,-N for soil. Total organic soil nitrogen equals total-N less NH4-N. Matrix to right summarizes estimated
annual N-fluxes among components
Pool - 6/72 (kg ha'1)
Total-N N03-fl NH^-N
2 3
Destination Compartment
10
Atmosphere
1. Wetfall
N-fixation
Vegetation
2.Foliage3
3. Branch
4. Bole-Sump
5. Lateral Roots
Subtotal
Litter
6. Oi
7. 02
Q; Twig Branch
Subtotal
S£ll
6-12
12-18
18-24
Total Soil
40.6C
7.8 0.33b
24.6 - - -
4.4d 1.68s -
£
|
5. -
10.
Losses
Total
2164
1174
1020
837
5195
2.25
6294.7
76.0 248
33.8 100
27.1 81
19.6 52
156.5 481
7. —
9. 0.339
10. -
- 21.5
2.16
69.6h 50.15
41.O1 -
62.59 -
2.16 -
2.251
aLitterfall amounts to 4090 kg/ha.
bNitrogen fixation by soil-litter microbes assumed to be equal to dissimilatory denitrification.
cNet N-accumulation is equal to foliage uptake less amounts leached in throughfall am'returned to annual litterfall
This calculation assumes no net change in N-concentration of woody components from year-to-year.
dThroughfall estimates from Walker Branch Watershed (Henderson and Harris 1975).
eTwig-branch litterfall estimate is for similar forests on Walker Branch Watershed (Henderson and Harris 1975).
fRoot death/production estimates from other research (Harris et al., unpublished ms,).
9Denitrification losses estimated to be 15% of annual amount leached.
hAll nitrogen uptake assumed to come from soil; seasonal (within year) distribution within tree not differentiated.
losses -ire the mean of 1969-70 losses in streamflow from Ualker Branch Watershed.
Table IV: Summary of nitrogen transfers in undisturbed forest
ecosystem. Numerical code refers to source-destination legend
in Table HI.
Compartment
input
(kg ha-^ yr'1)
9 -
9 -
1 -
2
2-5
3
5
7
8
9 ->
1 0.325
2 69.60
50.15
6 7.80
6 24.60
6 4.40
8 1.68
9 41.00
9 62.59
9 2.16
10 2.25
Compartment
Atmosphere
Vegetation
Litter
Soil
Groundwater
Total 266.55
Note:
Efficiency
100 x total
100 total
of terrestrial
system efflux
Compartment
loss
(kg ha"1 yr'1)
7.80
1 -» 6
4.40 2-5 7
24,60
1.68
41.00
69.59
2,16
2.25
69.59
50.15
?
266,55
nitrogen cycling
(2.25) _
annual N-transfers (^266.55)
2 6
3 8
5 9
7 9
8 9
9 ->• 10
9 ->• 2
9-5
Total
99+%
-CONTROL AREA
I I I
J-U
Figure 3: Summary of average NO3-N concentrations for
treated plots. There was no free soil water at the Bear Creek site
from August 25-October 10, 1972; control plot data are for
Walker Branch (when available). From October 18 on, control
plot data are from Bear Creek. Effluent applications occurred
on June 27, July 27 and August 14
-------
Forest Spray Irrigation
171
As soil moisture recharge occurred beginning in late Sep-
tember, elevated NO3-N concentration probably was due to
small amounts of nitrate produced during mineralization of
orgamc-N (augmented by salts introduced in effluent) and
accumulating in the soil profile. Due to dry soil conditions
which eliminate movement through the soil and sharply reduce
biological activity, this material is retained until sufficient water
is available for transport or immobilization by stimulated
microbial activity. The period of elevated NO3-N concentration
was short. Average NO3-N concentration the following week
was 0.3 ppm. Irrigation plot responses indicate that nitrate
losses following irrigation of up to 200 kg NO3-N per ha per yr
can be expected to increase NO3-N concentration in soil water
by small amounts during the year following summer irrigation.
A similar pattern of NO3-N concentration was observed
through 1974 (the termination of the monitoring program):
temporarily elevated NO3-N in free soil water coincident with
irrigation returning quickly to less than 0.2 ppm. Averaging
some 15 sampling dates from January 1973 to April 1974, NO3-
N concentrations (ppm) for 0, 50, 125, and 200 kg ha per hr
applications were: 0.007, 0.069, 0.066 and 0.160, respectively.
While these values are well below any regulatory limit or known
damage threshold, such an increase could forebode another
class of problem—non-point pollution inputs. Assuming drain-
age flux characteristics remain the same, this irrigation practice
could increase the N export (and potential input to streams) by
as much as a factor of 20. This increase in itself is not a problem
but emphasizes the need to plan such operations recognizing
other land-use impacts, watershed size and need for long-term
monitoring.
Organic Matter Losses
The carbon-to-nitrogen ratio (C/N) of an organic substrate
(e.g., litter) is a main determinant of microorganism activity. At
a C/N ratio above 20 to 25, microorganisms are nitrogen limit-
ed; as the C/N ratio is lowered below 20 to 25 (e.g., addition of
nitrate salts), microbial activity is accelerated since nitrogen no
longer limits metabolic processes. As the C/N ratio approaches
15, other factors (probably the amount or chemical form of
substrate carbon and/or phosphorus) become limiting to
further increases of microbial activity. Thus, a land disposal
system may be limited by the amount of organic substrate
available to support assimilation of nitrogen into the natural
biogeochemical cycle. Adding wastes beyond the assimilation
capacity which is limited in part by the carbon-nitrogen balance
could produce deleterious effects on the disposal system
attributed to the removal of litter organic matter.
The layer of forest litter organic matter appeared to undergo
accelerated decay with the initiation of irrigation operations. In
order to evaluate the effect of NO3 irrigation on litter decay,
litter compartment sizes were compared with initial values
(Table V).
Table V: Percent of initial (6/72) litter organic matter following
irrigation with NO3-N at rates of 0, 50, 125 and 200 kg ha yr.
Irrigation applications were made in three equal treatments
between June 27 and August 14,1972.
Irrigation rate
(kg/ha/yr)
0
50
125
200
September 1972
89
84
82
66
Date
February 1973
90
96
85
73
Hay 1973
85
87
79
68
Treatment levels up to 125 kg NO3-N ha-1 yr-1 had little effect
on litter compartment size compared to the control. Only at an
application rate of 200 kg NO3-N ha-1 yr-1 was litter appreciably
decreased (Table V). Much of the observed decrease was the
result of accelerated disappearance of woody litter. Woody litter
nitrogen content at this treatment level also exhibited a nearly
twofold increase from pre-irrigation levels (0.5% pre-irrigation
to 0.97% in September following completion of irrigation). This
increase in woody litter N content combined with the acceler-
ated disappearance of this substrate type suggests that one
impact of high nitrogen wastes could be on the C/N ratio or
organic substrates essential to microbial immobilization.
Longevity of Land Disposal System—Simulation
Models
Experimental data indicated some reduction in litter organic
matter at the highest rate of application. However, any addition
of nitrogen gradually changes the initial conditions. Annual N-
amendments could gradually lower C/N ratios of litter compo-
nents and thus promote accelerated decay. Within a few years
the nitrogen balance within the forest ecosystem could change
such that any subsequent addition of NO3-N could result in
deleterious effects on the litter and soil nitrogen balances.
In order to evaluate the long-term response of land disposal
system without further experimentation, which was not feasible
in this case, linear compartmental simulation models were con-
structed for both nitrogen (control and minimum treatment of
50 kg NO3-N per ha per yr) and carbon (revised after ).12 The
results of model simulation are not a substitute for a monitoring
program, but they do facilitate selection of ecosystem parame-
ters essential to an effective monitoring program. Model simula-
tion also assists in cost analysis since capital cost outlay for relo-
cation of the irrigation area can be projected from a minimal
program of further ecosystem analysis. Model simulation can
further assist in cost analysis by allowing for a priori study of
disposal system longevity and efficiency for alternative types of
ecosystems.
The basis for calculation of rate coefficients for N cycling was
the budget and annual flux summary (Tables III and IV). A dia-
grammatic summary of the nitrogen model and the equations
used in calculations are shown in Figure 4. Simulation of NO3-N
irrigation involved changing the input constant to leaf litter
from 7.85 to 57.85 kg NO3-N per ha per yr. The simulation
further assumed that a 50 kg NO3-N amendment would increase
foliage N-content by 67%." Data on effects of nitrogen
fertilization on hardwood forest production are inconclusive;
therefore, a constant amount of forest production was assumed
in the simulation on the basis of current growth rates (leaves -
2000, wood 1750, and roots = 3750 kg C per ha per yr, Figure
5). Other transfers were calculated from extant data on forest
carbon cycling.12 The criterion used to estimate the maximum
disposal period was depression of total litter C/ N to the value of
the C/ N ratio (pre-irrigation) of the fragmented layer (O2) of
litter organic matter.
The results of simulation suggest that the hardwood forest
could withstand a 3-4 yr chronic input (summer) of 50 kg NO3-N
per ha per yr (200 Ib NO3 per acre). In the simulation, nitrogen
accumulation in litter results in depression of the C/N ratio to a
level generally observed to accompany accelerated organic mat-
ter decay (Figure 6). The time required for recovery of the
carbon-nitrogen balance of litter in the hardwood forest was not
determined in this case. Where effluent disposal does affect the
ecosystem, recovery time should be determined in order to
establish a rotation cycle for disposal.
Since the amount of organic matter and its carbon-nitrogen
balance appear to be the factors largely determining the capacity
of the forest ecosystem to assimilate nitrate-nitrogen, we sum-
marized nitrogen and carbon budgets for pine forests. Pine typi-
cally has a lower nitrogen content and therefore a higher C/N
ratio. These characteristics result in greater accumulation of
organic matter in the litter (Table VI). A typical pine forest has a
-------
172
Forest Spray Irrigation
total litter C / N ratio of >31. Based on the compartment designs
in Figures 5 and 6, but with transfer coefficients specific for pine
forests (Table VI), nitrogen cycling of this forest ecosystem was
simulated using a chronic additional nitrogen input of 50 kg
NOj-N per ha per yr by utilizing a forest system with a large litter
C/N ratio, disposal on the same forest area could continue for 6
yr before nitrogen accumulation would depress the litter C/N
ratio to a level accompanying accelerated decay of litter organic
matter.
INPUT
0.00042
NITROGEN
* I = 0.013 (X6I-0.583 (XI) -0.353 (XI 1-0.063 (XI)
X2 = 0.583 (XI)-0.00491X2)-0.0951X2)
X3= INPUT + 0.353(XI 1+O.O63 (XO-O.220 (X3)
X4 = 0.0049 (X2)-0.0951X4)
X5=0.220(X3)+0.095 (X4) -0.215 (X5) + INPUT
X6 = 0 216 (X5) + 0.328 (X7)-0.0032 (X6)-0.013 (X6)-0.00042 (X6)
X7= 0.0032 (X6)-0.328(X7) +00951X2)
Figure 4. Summary of linear compartment model of annual ni-
trogen cycling. Initial conditions for each compartment are
based on compartment values listed in Table III. Rate constants
are the quotient of annual N-flux (kg N ha-1 yr-1) and compart-
ment size (kg N yr-1). The leaf compartment (XI) includes a
transfer associated with litterface (0.353) and foliar leaching by
rainfall (0.063)
CARBON: NITROGEN RATIO OF FOREST LITTER
-.noi\3roror\3oji,
3>0i\j.&a>c»oi\
, NO 1
\
\
\
\
i__NO_
—\rCHI
RRIGATION
HRONIC IF
)
\
^1
_RR]GATIO
^ONIC IRF
\._
t
H
-PINE
RIGATION
V
M-HARDV\
1
IGATION-
/
f
P
-PINE
^pop_
HARDWOI
)D
4 6
TIME (yeors)
10
Figure 6: Effects of annual NO3-N irrigation on forest litter
carbon-nitrogen ratios. Results of model simulation indicate
that pine forests have twice the longevity as disposal areas
compared to hardwood forests for an annual application of 50
kg NO,-N ha-1
Ps 2000 u
Ps 1370 u
Ps 3750
X1 LEAVES
2000
X2 WOOD
75,000
X3 ROOT
10,000
0.2925
TRs
'•° ^
0.00217^
LEAF
X5 LITTER
3910
WOOD
X4 LITTER
200
0.580
0.430 *ni
0.099 °| 0.048
5750
0.0325
CARBON
XI =200 -1.0 (XI) X5=1.0(X1 -0.580(X5)
X2 = 1370-0.00217 (X2) X6= 0.580 (X5)-0.430C
SOIL
122,500
0.000005
i
F
<6 )- 0.048 (X6)
•• 3750-0.2925 (X3)-0.0325(X3)
X4 = 0.00217(X2)-0.099(X4) ' X7= 0.033 (X6)+0.0325 (X3)-0.000005(X6)
Figure 5: Summary of linear compartment model of carbon accumulation in the forest ecosystem. Compartment values are
for the Bear Creek site based on 50% carbon content. Rate constants were determined from data summarized by Reichle el al.
(1972). Carbon pool sizes are expressed as kg C ha-1. Allocations of net photosynthate (Ps) to tree components were assumed
to be constants. Heterotrophic respiration of decay organisms (Rs) is included where compartment carbon losses are largely
as CO,
-------
Forest Spray Irrigation
173
Table VI: Summary of Carbon and Nitrogen Pools Used in Pine
Forest Simulation
Compartment
Living vegetation
Wood
Roots
Forest litter
Oi
Oi-wood
02
Soil
al.!2 kg/ha = 1 Ib/acre.
bl kg N03-N per ha = 4 Ib
Carbon
(kg/ha)a
2,000
75,000
6,050
2,000
5,550
122,500
N03 per acre.
Nitrogen
(kg/ha)b
36
201
159
9
237
5,350
SUMMARY AND RECOMMENDATIONS
Land disposal of high nitrate-bearing wastes was evaluated
using a limited experimental program and extant data on forest
ecosystems With effluent applications limited to dry summer
periods, NO3-N content of soil water exiting the root zone was
increased on the average but content remained low relative to
the content of effluent input. Even small changes in water qual-
ity such as those observed in the present study need to be evalu-
ated relative to existing water quality, other land-use practices
or options and potential long-term impact.
First year responses of the litter compartment of the treated
ecosystems suggested that over repeated applications litter
decay could be appreciably accelerated. Altered balance of car-
bon-nitrogen appeared to be the controlling factor. Simple
linear compartment models of the forest carbon and nitrogen
cycles were used to conservatively project the dynamics of the
litter C/N ratio. Assuming constancy of biomass production
and decay rates, hardwood and pine forests were estimated to
have a longevity of from 3 to 6 years respectively before
significant narrowing of the C/N ratio occurred. This may be an
underestimate of longevity since the only nitrogen content
appreciably increased was woody litter—a small fraction of the
total litter.
A mass balance of the disposal system could not be deter-
mined because nitrogen increments added in effluent were a
small fraction of the total nitrogen in the ecosystem, particu-
larly of the soil compartment. Determination of mass balance is
essential to determine the actual disposal efficiency, fate and
potential for deleterious effects of the disposal, but represents a
cost and time allocation well beyond the interest and resources
of most clients. In the present case, land disposal and in vitro
denitrification (packed column and stirred reactor types) were
simultaneously evaluated as possible means of disposal. Consid-
ering the prompt effects on forest litter, the relatively short lon-
gevity of particular disposal sites, the large land area required
and the very successful trials with denitrification processes, land
disposal was not selected as a primary means of disposal.
ACKNOWLEDGMENTS.
The authors would like to thank H. C. Francke, Union Car-
bide Corporation, Nuclear Division (retired) for his support and
assistance in this study.
REFERENCES
1. Auerbach, S. I. et al. 1972. Walker Branch Watershed: A
study of terrestrial and aquatic system interaction. In: Ecologi-
cal Sciences Div. Ann. Progr. Rep. Sept. 30, 1971, pp. 30-48.
ORNL/4759. Oak Ridge National Laboratory, Oak Ridge,
Tennessee.
2. Broadbent, F. E. and F. E. Clark. 1965. Denitrification.
In: W. F. Bartholomew and F. E. Clark (eds.) Soil Nitrogen, pp.
344-359. American Soc. of Agronomy, Inc. 615 pp.
3. Edwards, N. T. and P. Sollins, 1973. Continuous mea-
surements of carbon dioxide evolution from partitioned forest
floor components. Ecology 54(2):406-412.
4. Goldstein, R. A. and J. B. Mankin. 1972. PROSPER: A
model of atmosphere-soil-plant water flow. In: Proceedings of
the 1972 Summer Simulation Conference, June 14-16,1972, San
Diego, California, pp. 1176-81.
5. Harris, W. F., R. A. Goldstein and P. Sollins. 1971. Net
aboveground production and estimates of standing biomass on
Walker Branch Watershed. Eastern Deciduous Forest Biome
MemoRept. 71-80. 12 p.
6. Harris, W. F. and D. E. Todd. 1972. Forest root biomass
and turnover. Eastern Deciduous Forest Biome Memo Rept.
72-156. 17 p.
7. Harris, W. F., G. S. Henderson and D. E. Todd. 1972.
Measurement of turnover of biomass and nutrient elements
from the woody component of forest litter on Walker Branch
Watershed. Eastern deciduous Forest Biome Memo Rept.
72-146. 11 p.
8. Henderson, G. S. 1971. Nutrient concentrations in woody
components of sixteen tree species. Eastern Deciduous Forest
Biome Memo Rept. 71-90. 10 p.
9. Henderson, G. S., W. F. Harris and D. E. Todd. 1972.
Validation of foliage biomass prediction equations using litter-
fall data on Walker Branch Watershed. Eastern Deciduous
Forest Biome Memo Rept. 72-158. 8 p.
10. Henderson, G. S. and W. F. Harris. 1975. An ecosystem
approach to characterization of the nitrogen cycle in a decidu-
ous forest watershed. In: Bernier, B. and C. H. Winget (eds.)
Forest Soils and Forest Land Management. Les Presses de
FUniversite. Laval, Quebec, Quebec, Canada, pp. 179-193.
11. Jones, H. C. and J. W. Curlin. 1968. The role of fertilizers
in improving the hardwoods of the Tennessee Valley. In: Forest
Fertilization: Theory and Practice, pp 185-190. Tennessee Val-
ley Authority, Knoxville, Tennessee. 306 p.
12. Reichle, D. E., B. E. Dinger, N. T. Edwards, W. F. Harris
and P. Sollins. 1972. Carbon flow and storage in a forest ecosys-
tem. In: Carbon and the Biosphere, AEC CONF 720510 (in
press).
13. Sopper, W. E. and L. T. Kardos, eds. 1973. Symposium
proceedings: Recycling treated municipal wastewater and
sludge through forest and cropland. August 21-24, 1972. The
Pennsylvania State University Press, University Park, Pennsyl-
vania. 479 p.
14. Todd, R. L. 1972. Terrestrial microbial populations and
their role in nutrient cycling on experimental watersheds with
contrasting vegetation: A progress report. Eastern Deciduous
Forest Biome Memo Rept. 72-145. 22 p.
15. Westman, W. E. 1972. Some basic issues in water pollu-
tion control legislation. Amer. Scientist 60:767-773.
-------
Transport and Reaction of
Contaminants in Ground-Water Systems
David B. Grove and Jacob Rubin
U.S. Geological Survey, Denver, Colorado and
Menlo Park, California
INTRODUCTION
Problems of mineral-resources utilization and waste-product
disposal include assessment of the environmental impact of
waste-product incorporation into the subsurface layers of the
earth. Assessment of this kind concerns itself primarily with the
possible pollution of useable ground-water resources. Among
other things, it must consider the possibility of moderating or
eliminating such dangers by using the natural dispersive,
pollutant-fixing, and pollutant-degrading capabilities of the
subsurface. Accurate assessment of such problems requires the
capability to predict quantitatively the transport of solutes and
the resulting concentrations of pollutants in the subsurface en-
vironment, including saturated and unsaturated zones. It is the
purpose of this paper to discuss brieflythe present status of tech-
niques for making such predictions in the saturated zone.
Theory
The most often used and probably the most promising of the
currently available tools for predicting solute transport in the
subsurface environment are mathematical models that utilize
digital computers. Of these, the ones that now are being devel-
oped extensively are based on adaptations, generalizations, and
extensions of an approach (the analytical solutions of the defin-
ing differential equations) that has been used for some time in
connection with certain chemical applications, such as chemical
engineering and chromatography (Bird2 Helfferich9). These gen-
eralizations, extensions, and modifications are needed because
the systems dealt with in connection with environmental prob-
lems are often more complex than those in the chemical-
processing applications. For instance, environmental models
that cannot handle layered or other types of nonuniform sys-
tems would have little general utility. On the other hand, most
operational chemical-engineering models assume uniformity of
the porous media analyzed.
The models of solute transport in subsurface layers ordinarily
consist of two parts. One part, the waterflow submodel, com-
putes water seepage velocities in the subsurface region investi-
gated. The second part, the solute-transport submodel, uses
these velocities (as well as other parameters) to compute the con-
centrations of the solutes of interest.
The seepage velocity necessary to the solution of the solute-
transport equation is computed using Darcy's law and can be
written as
v,-^ £
1 e ox;
where V , is the seepage velocity in the x , direction, LT -'. The
solute-transport submodel is usually considerably more com-
plex than the waterflow submodel. The waterflow submodel de-
scribes a single, purely physical process involving a single,
mobile, system component (water). On the other hand, the
solute-transport submodel may be concerned with many pro-
cesses, each one of which may involve many, usually miscible,
system components.
Solute transport includes at least two processes that are physi-
cal in nature: convection (solute transfer by and with flowing
water) and hydrodynamic dispersion (the mixing between the
incoming solutes and those originally present in the pore water).
In addition, the submodel under consideration very often must
takeinto account chemical and biological processes. Reddell
and Sunada"and Bredehoeft and Finder4 present the following
equation to prepresent solute transport with reactions through
porous media. The present authors have generalized and added
into such an equation the chemical reactions of the solute species
both within the void space and on the solid surface.
9c_
?3t
where CHEM equals
9c
"Pb ai-
s
E
k=l
jg—J +C'W* = CHEM (3)
J /
for equilibrium-controlled
exchange reactions
lon-
- X (ec
for s chemical rate controlled re-
actions
) for radioactive decay
where
C'
is the concentration of the solute (mass of
solute/volume of liquid), ML"-*;
is the coefficient of hydrodynamic disper-
sion, L2T'] ;
is the concentration of the solute in the
source or sink fluid, ML"3;
is the bulk density of the solid (mass of
solid/volume of sediment), ML"3;
174
-------
Ground-water Systems
175
c is the concentration of the species adsorbed
on the solid (mass of solute/mass of sedi-
ment), ML"3;
Rjf is the rate of production of the solute in
reaction k of s different reactions, ML'3 T'l;
X is the radioactive-decay constant (equal to
In 2/half life), T-l.
The solute-transport equation contains several flux and
sink/source terms that merit further consideration and
definition.
Physical Processes
The second and third terms on the left-hand side of Equation
3 define solute transport due to dispersive and convective fluxes,
respectively. The dispersive flux is proportional to D.., the coef-
ficient of hydrodynamic dispersion, and the concentration gra-
dient, and accounts for the mixing process that occurs within
pores between the native and contaminant fluid. The dispersion
coefficient is assumed to be a function of the seepage-velocity
vector V. and the characteristic lengths, L for the longitudinal
direction and T for the transverse direction. Bear1 gives in
detail the formula for calculating the various components of the
dispersion tensor if the velocity and characteristic lengths are
given. Bear1 also indicates that Dj. includes both diffusion and
mechanical dispersion, and is equal to the latter when diffusion
is negligible.
Chemical Processes
Two types of chemical reactions often occur in contamination
problems and are given as examples; these are ion-exchange
reactions that may be considered equilibrium controlled and
irreversible rate reactions. Because several field contamination
problems have been modeled, a discussion of the chemistry that
affects the analysis of these processes follows:
Equilibrium-controlled ion exchange
A typical reaction might be written as follows (Bolt,
1967; Helfferich, 1962)
+ a
(4)
where 1 and 2 are chemical exchanging species with a and
b valences respectively.
The adsorbed species are given as cj and the dissolved
species as q, i = 1,2. The use of a selectivity coefficient
may be used to relate the concentration of products and
and reactants at equilibrium:
phase of the major ion is then nearly equal to the cation-
exchange capacity (CEC), and the solution phase of the major
ion is equal to the total concentration (CQ), which remains con-
stant. Equation 5 can then be rewritten as,
(6)
J£ =
s (CEC)b (c2)a
or
K --
-~-
K, (CEC)b
(Cn)b
I/a
(7)
where K^ = ion-exchange distribution coefficient.
Equation 7 postulates a linear relationship between the
adsorbed species and the solute species where the slope of the
equilibrium ion-exchange isotherm is the distribution coeffi-
cient. Equation 7 thus provides the relationship between the
adsorbed species and dissolved species necessary to solve the
solute-transport equation.
Rate-controlled chemical reactions
The Rk term in Equation 3 denotes all possible rate-con-
trolled chemcial reactions occurring within the pore space,
including both homogeneous (liquid-liquid) and heterogeneous
(liquid-solid) reactions. Homogeneous reactions could include
ion-complex dissociation and formation, and heterogeneous
reactions, precipitation-dissolution, and ion exchange.
Radioactive decay
The subsurface disposal of radioactive products presents an
example where a first-order irreversible rate reaction occurs.
This reaction is the radioactive decay of the adsorbed or dis-
solved species. The rate constant can be derived in the following
manner. The disappearance of a species by a first-order, irre-
versible reaction is given by the equation
dc %
—= -Xc
dt
(8)
when X is the rate constant, T . This equation can be
integrated with integration limits chosen as the time neces-
sary for the initial concentration to decrease by half
(•=0/2 dc f'1,2
J T-» J «
c0 0
(9)
where
(5)
where Ks is the ion-exchange selectivity coefficient.
Equation 5, when incorporated into the solute-transport
equation, results in nonlinear terms. Rubin and James21 treat
this case, as well as cases involving two or more exchanging spe-
cies. A simplified equation results from the common situation
that occurs when the exchanging ion is very low in concentration
relative to the other ions. Then the exchange process will not
materially affect the concentration of the other exchanging ions,
either in solution or adsorbed on the matrix. The adsorbed
cn = initial concentration, ML , and
t = half life of species, T.
The equation is integrated and solved for X to obtain
0.693
X = -
t
1/2
(10)
These reaction terms can be incorporated into the solute-
transport equation in,the folio wing manner: Assume that during
its transport a radioactive chemical species is disappearing by a
-------
176
Ground-water Systems
first-order irreversible rate reaction and is being exchanged
reversibly by an equilibrium-controlled process in which the
exchange isotherm is linear.
Equations 7 and 8 give the relationships between c and c.
The term c is Kdc; and 9c/9t is Kd (9c/9t).
Substitution of these terms into Equation 3 gives
9c
e —
9t
-eXc
_JL(ecVi)-c.w*
(ID
A retardation factor, Rf, which appears in equilibrium-
controlled trace-exchange reactions, is defined as
Rf=l+Kdpb/e. (12)
Equation 11 is divided by this retardation factor to obtain
9t Rf 9xj
9c
19
1 c'W*
(13)
The retardation factor thus retards the flux of the solute species
with respect to the fluid flow. This reduced velocity of ions effec-
tively provides a safety factor that prevents excessive movement
of harmful pollutants.
Numerical Techniques and Field Studies
The equations discussed in the preceding section describe how
changes in concentration, resulting from various processes, are
related to each other. These partial differential equations may be
solved for concentrations provided that certin necessary data
about the system in question are available to describe the system
properties, sources and sinks, and initial and boundary
conditions. Several ways of solving the equations are
available—analytical techniques, analog computers, and
numerical techniques. Under conditions usually met in envi-
ronmental problems, it seems that only the combined use of
numerical techniques and digital computers can provide suffi-
cient generality for routine performance of the task under con-
sideration.
Many numerical methods are known (Remson111 von Rosen-
berg,2'). In all of these the basic transport equations must be
approximated by other questions (usually algebraic) that subse-
quently are used to write the computer program. This transfor-
mation reduces the original mathematical problem to the solu-
tion of large systems of simultaneous algebraic equations. Also,
note that in all numerical methods concentrations are computed
only for certain discrete points in space. This is done for
successive time values, the intervals between these values usually
being relatively small. The points in space and time for which
computations are made are called grid points, and the intervals
between them are called grid intervals. Numercial methods yield
concentration values that are approximate. However, by
decreasing the size of the grid intervals, it is possible to increase
the accuracy of numerical results as much as one wishes (within
the limitations of the conceptual model used and the data availa-
ble). The numerical accuracy of the solution usually exceeds the
measurement accuracy of the system parameters. Use of smaller
grid intervals requires more time-consuming (and hence more
expensive) computer calculations. Therefore, the grid intervals
may be too small to be practical. Various numerical methods
differ in the degree of accuracy they may yield for any particular
problem with a given set of grid intervals. They also may differ in
computer size and speed requirements, in the degree of mathe-
matical sophistication needed to use them, and in other fea-
tures. Because of such differences and because of the variety of
needs the models must satisfy, it is usually better to select the
most efficient numerical method for a given type of a problem
than to try to develop one general method applicable to all the
expected problems.
The numerical methods used earliest in solute-transport mod-
els were the finite-difference methods. Although the finite-
difference methods performed very satisfactorily in connection
with the waterflow submodel, their use in the solute-transport
submodel produced considerable difficulties whenever the
effects of convection were large in comparison with those of dis-
persion. These difficulties are caused by numerical
dispersions—artifacts that distort the model's predictions and
that result from the nature of approximation made by the differ-
ence methods. They express themselves by the smearing of sharp
fronts or by oscillations.
These difficulties can be rectified by using either large artifi-
cial dispersion coefficients or small space increments. However
these techniques produce either inaccurate results or involve
excessive computer storage and time requirements.
Peaceman and Rachford13 presented a finite-difference
scheme to solve the two-dimensional transport equation. They
illustrated the oscillation problem, caused by small dispersion
coefficients, for the one-dimensional case and presented a
"transfer of overshoot or undershoot" correction technique.
Shamir and Harleman22 studied a steady-state water-flow situa-
tion and solved the solute-transport equation in terms of stream-
lines and velocity potentials. Because velocity is parallel to a
streamline, one-dimensional flow along a streamline could be
assumed and cross-produce terms of the dispersion tensor omit-
ted. Unfortunately, both of these approaches have restrictions.
The method of Peaceman and Rachford13 is not rigorous and
the use of streamlines assumes steady-state flow conditions and
unrealistic one-dimensional flow.
In an effort to avoid difficulties associated with finite-
difference methods a numerical technique was developed by ap-
propriately modifying the method of characteristics (MOC).
Garder and others6 were among the first to use the MOC to solve
the solute-transport equation. Their modification of the method
included a point-tracking technique in which particles were
assigned initial concentrations and allowed to move with the
velocity of water to new locations during successive time incre-
ments. Concentrations were averaged over the various grid
domains and the dispersion changes calculated by an explicit
finite-difference method. Reddell and Sunada17 and Bredehoeft
and Pinder4 expanded the one-dimensional case of Garder and
others'' to two-dimensions. Pinder and Cooper15 as well as Red-
dell and Sunada17 utilized the MOC to include density depend-
ence and solved a salt-water encroachment problem.
The method and the computer program developed by Brede-
hoeft and Pinder4 have been extensively used within the U.S.
Geological Survey to solve field problems involving contamina-
tion. Bredehoeft and Pinder investigated a contamination prob-
lem in Georgia, where saline water upwelled into a fresh-water
aquifer, and computed the effects of discharge or barrier wells to
limit the migration of the contaminant. Hughes and Robson10
investigated contamination from sewage lagoons and industrial
cleaning areas, and predicted the results of various ground-
water-quality containment practices. Konikow and Brede-
hoeft" investigated the effects of irrigation and return flow on
water quality in the ground-water system and in the adjacent
river. Robertson and Barraclough," at the National Reactor
-------
Ground-water Systems 177
Testing Station (now the Idaho National Engineering Labora-
tory), Idaho, modeled the movement of injected pollutants into
a basaltic aquifer and predicted the concentration profile for a
period of 20 years. Robertson20 modified the solute-transport
equations to include the chemical-reaction terms that accounted
for ion exchange and radioactive decay.
The Galerkin finite-element method has been utilized recently
to solve transport equations of all types. It can be a highly accu-
rate and efficient method, but is difficult to program. In the fore-
seeable future this method probably will be used with increasing
frequency whenever the hydrodynamic-dispersion effects are
small in comparison with the convection effects.
Price and others16 first showed the superiority of the Galerkin,
finite-element method over the standard, finite-difference
method to solve the one-dimensional mass-transport equation.
Guymon and others" solved the multidimensional convection-
diffusion equations using finite-element methods.
Finder14 used a Galerkin finite-element technique with iso-
parametric quadrilateral elements to solve a ground-water con-
tamination ploblem on Long Island, New York. In this case, ele-
ments could take on a variety of configurations and be reduced
to rectangles by a mapping procedure. The resulting set of linear
equations was solved by a direct-matrix technique. This applica-
tion was conservative in that chemical reactions involving the
solute were not considered.
Grove7 used the Galerkin finite-element technique to solve the
field problem of Robertson20 that included the chemical-
reaction effects of ion exchange and radioactive decay. Since
1952 chemical and low-level radioactive wastes were disposed of
into the ground-water system at the Idaho National Engineering
Laboratory in southeastern Idaho. These wastes migrate down
the hydraulic gradient and, unless removed from the aquifer by
physical or chemical means, discharge into the Snake River.
Figures I and 2 present typical modeling results matching the
numerical and field data for groundwater contamination by tri-
tium and strontium-90. The model closely approximates the
movement of these two ions. The tritium decays radioactively
but moves with the velocity of the ground water, while the
strontium-90 is greatly retarded by ion exchange as well as being
influenced by radioactive decay.
EXPLANATION
EQUAL TRITIUM CONCENTRATION IN pCi/ml FOR 1966
50 WELL SAMPLES
50 DIGITAL MODEL (FINITE ELEMENT)
EXPLANATION
.001 WELL SAMPLES
0.01 DIGITAL MODEL (FINITE ELEMENT!
Figure 1: Comparison of waste tritium plumes for 1968-1969
based on well sample data and computer model
Figure 2: Comparison of waste strontium-90 plumes for 1964
based on well sample data and computer model
Some Unsolved Problems
The lack of data concerning chemical parameters that control
solute concentrations during movement through porous media
is a weak link in the modeling process. The independent determi-
nation of chemical-reaction parameters that are applicable to
field situations is indeed a most difficult task. The concentration
of moving solute may be influenced by concentrations and reac-
tions of other adsorbed and dissolved chemical species. The cap-
ability to model multiple reactions may be limited by computer
size and economics. Another matter that must be considered is
the development of methods to use field geochemical data avail-
able at point locations to generalize the chemical characteristics
of the aquifer matrix through which reacting solute species pass.
The determination and understanding of the dispersion coef-
ficient is another research need. Although the dispersion process
is idealized by analogy with Pick's law of diffusion it most likely
results through permeability variations in the aquifer. A more
complete understanding of dispersion through field studies that
involve fracture flow and anisotropic systems would be of value.
The unsaturated zone as a source to the ground-water aquifer
must be more fully defined concerning solute transport. Disper-
sion phenomena, chemical reactions, and possible biological
reactions need additional attention. Proper verification of mod-
els with independently determined chemical and physical
parameters and carefully selected field locations are the ultimate
test.
SUMMARY
There are currently available mathematical formulations and
numerical techniques that can be used to predict the transport
and reaction of solutes in the saturated ground-water environ-
ment. These techniques use a water-flow submodel with a con-
centration computing submodel. The numerical techniques
have been verified with analytical solutions and in a number of
cases the models as a whole have been verified with field tests. In
the latter verification predictions of solute transport involving
irreversible rate reactions and equilibrium-controlled ion-
exchange reactions were successfully made.
-------
178
Ground-water Systems
Future research needs include chemical-parameter identifica-
tion and more insight into the physical process of dispersion.
Unsaturated zone research that provides information concern-
ing the sources to the saturated ground-water system should
receive increased emphasis.
REFERENCES
I. Bear, J., 1972, Dynamics of fluids in porous media: New
York, Am. Elsevier Publishing Co., 764 p.
2. Bird, R.B., Steward, W.E., and Lightfoot, E.N., 1966,
Transport Phenomena: New York, John Wiley and Sons, 780 p.
3. Bolt, G.H., 1967, Cation exchange equations in soil
science: A review, Netherlands Jour. Agr. Sci., v.15, p.81-103.
4. Bredehoeft, J.D. and Pinder, G.F., 1973, Mass transport
in flowing groundwater: Water Resources Research, v.9, no. 1,
p. 194-209.
5. Cooper, H.H., Jr., 1966, The equation of groundwater
flow in fixed and deforming coordinates: Jour. Geophys.
Research, v.71, no.20, p.4785^790.
6. Garder, A.O., Peaceman, D.W., and Pozzi, A.L., 1964,
Numerical calculation of multi-dimensional displacement by
the method of characteristics: Soc. of Petroleum Eng. Jour., v.4,
no. I, p.26-36.
7. Grove, D.B., 1976, The use of Galerkin finite element
methods to solve mass transport equations: Ph. D. thesis, Colo.
School of Mines, Golden, Colo., 152 p.
8. Guymon, G.L., Scott, V.H., and Herrman, L.R., 1970, A
general numerical solution of the two-dimensional diffusion-
convection equation by the finite element method: Water Re-
sources Research, v.6, no.5, p.1611-1617.
9. Helfferich, Friedrich, 1962, Ion Exchange: New York,
McGraw Hill, 624 p.
10. Hughes, J. L. and Robson, S. G., 1973, Effects of waste
percolation of groundwater in alluvium near Barstow, Califor-
nia in Underground waste management and artificial recharge:
v. l,p.9l-129.
11. Konikow, L. F., and Bredehoeft, J. D., 1974, Modeling
flow and chemical quality changes in an irrigated stream-aquifer
system: Water Resources Research, v. 10, no. 3, p. 545-562.
12. Lantz, R. B., Pahwa, S. B., and Grove, D. B., 1976, De-
velopment of a subsurface waste disposal simulation model
(abs.): EOS, Am. Geophys. Union Trans., v. 57, no. 4, p. 249.
13. Peaceman, D. W. and Rachford, H. H., Jr., 1962, Numer-
ical calculation of multi-dimensional miscible displacement:
Soc. of Petroleum Eng. Jour., v. 2, no. 4, p. 327-339.
14. Pinder, George F., 1973, A Galerkin finite element simu-
lation of ground-water contamination on Long Island, New
York: Water Resources Research, v. 9, no. 6, p. 1657-1670.
15. Pinder, G. F. and Cooper, H., 1970, A numerical tech-
nique for calculating the transient position of the salt water
front: Water Resources Research, v. 6, no. 3, p. 875-882.
16. Price, H. S., Cavendish, J. C, and Varga, R. S., 1968,
Numerical methods of higher order accuracy of diffusion-
convection equations: Soc. of Petroleum Eng. Jour., v. 8, p.
293-303.
17. Reddell, D. L. and Sunada, D. K., 1970, Numerical simu-
lation of dispersion in ground-water aquifers: Colorado State
Univ. Hydrology Paper 41, Fort Collins, Colo. 79 p.
18. Remson, 1., Hornberger, G. M., and Molz, F. J., 1971,
Numerical methods in subsurface hydrology: New York, Wiley
Interscience, 389 p.
19. Robertson, J. B. and Barraclough, J. T., 1973, Radioac-
tive and chemical-waste transport in groundwater at National
Reactor Testing Station, Idaho: 20 year case history, in Under-
ground waste management and artificial recharge: v.I,
p.291-322.
20. Robertson, J. B., 1974, Digital modeling of radioactive
and chemical waste in the Snake River Plain Aquifer at the
National Reactor Testing Station, Idaho, 41 p.; available only
from U.S. Dept. of Commerce, Nat'l. Tech. Inf. Service, Spring-
field, Va. 22151.
21. Rubin, J. and James, R. V., 1973, Dispersion-affected
transport of solutes in porous media; I The Galerkin method ap-
plied to equilibrium-controlled exchange in unidirectional,
steady, saturated waterflow: Water Resources Research, v.9,
no. 5, p. I 332-1357.
22. Shamir, U. Y. and Harleman, D. R. F., 1967, Numerical
solutions for dispersion in porous mediums: Water Resources
Research, v.3, no.2, p.557-581.
23. von Rosenberg, D. V., 1969, Methods for the Numerical
Solution of Partial Differential Equations: Elsevier, New York.
-------
Field Verification of Hazardous Waste
Migration from Land Disposal Sites
J. P. Gibb
Illinois State Water Survey
Urbana, Illinois
and
K. Cartwright
Illinois State Geological Survey
Urbana, Illinois
INTRODUCTION
In Illinois, 62 land disposal sites are permitted by the State
Environmental Protection Agency to receive hazardous chemi-
cal wastes. In addition, more than 2000 active or abandoned
landfill sites and private industrial disposal sites have received
large but unknown quantities of all types of wastes, which may
have included toxic chemicals. Some of these are adjacent to or
directly underlain by shallow aquifer systems vulnerable to pol-
lution from surficial sources.
The amount and areal extent of hazardous material migration
from these disposal sites is not known. Few sites are monitored
for possible pollution of contiguous aquifers, and most that are
monitored appear to be ineffectively instrumented. Tradition-
ally, monitoring wells are installed and water samples collected
and analyzed periodically. However, these wells generally can-
not monitor very large vertical segments of an aquifer, and the
water samples are not always analyzed for the many different
organic or inorganic chemical compounds that may originate
from the disposal sites.
Existing air and surface-water pollution regulations are forc-
ing an ever increasing volume of hazardous chemical waste to
the land for ultimate disposal. This is true particularly in the
heavily populated and industrialized regions of the United
States where most of these wastes are generated, used, and even-
tually discarded. In the humid parts of the country, shallow
groundwater reservoirs are recharged by precipitation infiltrat-
ing through the land surface. As a result, some shallow aquifers
may be in danger of serious water quality degradation if soils are
not effective in keeping hazardous wastes from migrating
downward to the aquifers.
Purpose of Study
The primary purpose of this study is to develop effective
investigative and monitoring techniques for detecting and eval-
uating quantitatively the extent of groundwater pollution from
surface toxic waste disposal sites. The study also is designed to
verify in the field the effectiveness of glaciated region soils and
associated surface deposits in retaining specific hazardous
chemicals.
The findings of this study should provide an invaluable tool
for predicting possible long-term effects of hazardous chemical
disposal on land. Also, the methodologies used should be imme-
diately applicable in evaluating the extent of hazardous chemi-
cals migration from disposal sites situated in similar geohydro-
logic environments throughout Illinois and other humid-region
states in the country.
Site Selection
Four sites were selected for study on the basis of geology, the
types and quantities of hazardous wastes generated, and the
manner of waste disposal. Sites were selected where the uncon-
solidated materials range from about 15 to 75 feet thick. Three
consist predominantly of low-permeability silt and clay soils,
and the fourth, selected later, is sandy. Pennsylvanian-age shales
or sandstones also lie beneath the glacial deposits at all four
sites. Thus, from the geology, at least three of the sites would
theoretically be desirable disposal areas with little resulting
groundwater pollution.
Three sites (A, B, and D) are secondary zinc smelting plants
located in central Illinois. A large quantity of waste from these
plants has been disposed of in solid form on the plant property
over many years of operation. Site C is a chlorinated hydrocar-
bon processing plant. Approximately half of the liquid waste
from this plant is stored and pretreated in lagoons or pits on the
plant property and then injected into a deep disposal well. The
possible effects of the surface lagooning and storage pits are the
principal concern in this study. This discussion will be limited to
work accomplished at site A.
Site A:
Site A is a secondary zinc smelter located in southcentral
Illinois. The plant, which started operations between 1885 and
1890, initially processed zinc ore. It was converted to a sec-
ondary zinc smelting facility about 1915. Wastes from the smelt-
ing operations during the first 85 years were principally heavy
metals-rich cinders and ashes. During the early years large quan-
tities of cinders were used as road fill or surfacing for secondary
roads and farm lanes in the plant area. The remainder was used
as fill material around the plant buildings and as surfacing over
the property. As a result of these disposal practices, there now is
a 1 - to 10-foot thick layer of metals-rich cinders covering about
12 acres of the plant property.
In compliance with air pollution control regulations, a
scrubber was installed on the plant stack in 1970. Prior to that
time, wind-blown ash, rich in zinc and other heavy metals, was
deposited on the plant site and on approximately 100 acres of
surrounding farmland. This source of pollution has now been
minimized, but wastewater from the scrubber (about 14,500 gal-
lons per day) is deposited in a seepage pit constructed on the
cinder materials that form the land surface at the site. Several
hundred tons of high zinc content sludge have accumulated
from the frequent cleaning of this pit and are now being repro-
cessed for zinc and lead recovery. Most of the water from the pit
infiltrates into the ground underlying the plant property.
Prior to the study, limited data suggested that groundwater
pollution might be occurring from three possible sources: (1) the
large volume of solid waste materials (cinders and stored junk to
be processed) at the plant site; (2) highly mineralized liquid
wastes from the stack scrubbers, and (3) wind-blown ash from
the smelter furnaces prior to installation of the scrubbers.
179
-------
180
Field Verification
Because of the long period of operation at this facility and the
various sources and forms of pollution likely to be present, this
site appeared to be most desirable to study in detail. The deci-
sion was made to devote maximum time, effort, and money at
this site to develop the study methodology and optimize its
application to other sites.
Method of Investigation
An extensive drilling program to obtain unconsolidated sedi-
ment cores and groundwater samples has been undertaken at
each site. Chemical analyses of the core and groundwater sam-
ples are being used to define: (I) the vertical and horizontal
migration patterns of chemical pollutants through the shallow
glacial deposits and aquifer profiles, (2) the seasonal variation of
toxic chemical levels in groundwater near these sites, and (3) the
residual toxic chemical buildup in the unconsolidated materials
in the vicinity of the sites.
Coring:
Continuous vertical core samples for geologic study and
chemical analyses have been obtained with conventional Shelby
tube and split spoon sampling methods through hollow stem
augers. These dry drilling techniques were used to minimize
chemical alteration of samples from drilling fluids or external
water during excavation.
Coring has been done with a truck-mounted Central Mining
Equipment (CME) 55 and a CME 750 rig mounted on an all-
terrain vehicle. The drilling crew consisted of an equipment
operator and helper, assisted if necessary by the principal
investigators. For the first few test holes drilled at each site, a
geologist from the State Geological Survey assisted in collecting
samples and made preliminary soil identifications for use in sub-
sequent testing.
Shelby tube and split spoon samples were extruded in the field
as collected, cut into 6-inch lengths, placed in wide-mouth glass
jars, and delivered to the State Geological Survey and Environ-
mental Analytical Research Laboratory at the University of Illi-
nois for processing and analyses. One 6-inch length of core from
each 5-foot segment and/or change in formation was taken by
the drilling contractor for moisture content determinations
before being sent for geological and chemical analyses.
Core Analysis:
Core samples for heavy metals determinations are being ana-
lyzed at the Environmental Analytical Research Laboratory
with zinc as a target element. Previous experience in determin-
ing heavy metal contaminants in soil showed that digestion of a
dried soil sample in 3W HC1 at slightly elevated temperatures
effectively releases the heavy metals without destructing the sil-
icate lattice of the soil. The heavy metals so released are deter-
mined by atomic absorption spectroscopy, the principal metho-
dology used to date.
For a limited number of soil samples, the multi-element
capability of optical emission spectroscopy has been used to
determine the Cd, Cu, Pb, and Zn concentrations. Semi-
automated multi-element analysis for a larger number of ele-
ments has shown these four elements to be the main metal pres-
ent.
Preliminary tests using atomic absorption measurement of
small spot samples indicate that the 6-inch long samples are too
heterogeneous to permit reproducible analysis. Reproducible
results have been attained by homogenizing the samples and
subdividing them to sample weight levels of 1 gram for atomic
absorption and 50 milligrams for emission spectroscopy. Pellet-
ized samples of approximately 2 grams have been prepared for
X-ray emission spectroscopy.
Well Construction:
Analyses of water samples from observation wells has been
the traditional method for monitoring groundwater pollution.
To demonstrate the effectiveness of such an approach and the
relative cost of using wells as compared to coring, a number of
small-diameter (2-inch) observation wells were constructed. At
site A, where heavy metal contaminants were expected, plastic
casing, screen, and pumping equipment were used.
Observation wells at site A were constructed in the following
manner. A 7-inch diameter hole was constructed, and a 2-inch
diameter PVC pipe (bottom 2 feet slotted with a hacksaw) was
placed in the hole. Gravel was placed from the bottom of the
hole to a level about 1 foot above the slotted portion of the pipe
followed by about 6 inches of sand. The remainder of the annu-
lus was filled with bentonite slurry to land Surface (see Figure 1).
A shallow well (10 to 15 feet deep) was constructed at each loca-
tion (see Figure 2), and deeper wells just above the bedrock were
added at nine locations.
""- Shrader valve
1/2" discharge
pipe
7" bore hole -*-
clotted PVC
well casing
3"
N -"
o oo
>•
0 O
00
0 0
1
OD\)
r=
=
=
—
—
~
—
n
M
—
=
==
^^
—
6o'o
ss°
°°0°0
O 0
0 0
u°a
oun
°0°
0
on
°a
y°0
bentonite
"slurry
1/4" airline
„ 6" sand
12"
-gravel
Figure 1. Typical Well and Pumping Mechanism.
Observation wells are equipped with individual pumping
devices to minimize possible contamination of well samples
from other wells. The pumping device consists of a '/4-inch
diameter PVC discharge pipe that extends from above the 2-
inch well cashing to the bottom of the well. A tee fitted with
short nipples and removable caps is placed at the top of this pipe
(Figure 1). The cap on the vertical segment can be removed to
allow for water level measurements within the !/2-inch pipe. The
cap on the horizontal segment (water discharge outlet) is vented
to permit stabilization of the water level within this pipe.
A '/i-inch plastic air line also is installed in each well. The air
line is attached to a Shrader valve at the top of the well casing
and extends the entire depth of the well. . The lower end of the
air line is bent up into the bottom of the '/2-inch discharge pipe
for a distance of about 3 inches.
-------
Field Verification 181
SCALE OF FEET
0 100 200 300
Figure 2. Location Map, Site A.
Water is pumped from the wells by removing the cap from the
horizontal portion of the '/2-inch pipe and applying air to the sys-
tem through the Shrader valve. Pumping from depths as great as
70 feet is possible with only a bicycle-type hand pump. A
gasoline powered 4-cylinder air-compresser capable of deliv-
ering about 5 cubic feet per minute at pressures up to 60 psi is
being used. An activated charcoal filter has been placed in the
discharge line from the compressor to insure that air from the
compressor is not introducing airborne contaminants.
Water Sample Analysis:
Water level measurements are made and water samples col-
lected from each well once a month. Samples are collected in 6-
ounce plastic containers and placed on ice until they can be
refrigerated in the laboratory. Each well is pumped for a period
adequate to insure that all stored water in the well casing has
been removed. The wells are allowed to recover and a sample is
then collected from the water that has just entered the well. This
procedure insures that the water sample collected is representa-
tive of the water flowing through the aquifer at the time of col-
lection.
Heavy metals are determined by atomic absorption spectro-
scopy with appropriate correction for blank and background
absorption. Concentrations of Zn, Pb, Cd, or Cu below the
detection limit of atomic absorption spectroscopy are
determined by anodic stripping voltammetry.
Secondary Methods:
A number of qualitative methods of groundwater pollution
detection and mapping have been used with varying degrees of
success. In an effort to evaluate the relative effectiveness of a few
of these methods, electrical earth resistivity surveys, soil temper-
ature surveys, soil and vegetation mapping, and color-thermal
infrared and normal color photography have been employed at
site A.
Results
To date, 47 wells at 34 locations have been completed at site
A. Cores were taken at each well and at 19 additional sites (Fig-
ure 2). Total well and core sampling footages are about 1300 and
1275 feet respectively.
The glacial materials at this site are about 60 to 75 feet thick.
The stratigraphic units recognized are essentially uniform in
character and thickness and generally flat. The elevation of the
surface of the Pennsylvanian bedrock dips from 447 feet above
sea level on the east to about 432 feet on the west. A brief de-
scription of the units, from top down, follows:
(A) Peoria Loess (4 to f>")—Brownish gray clayey silt, non-
calcareous, with iron stains
(B) Roxanna Silt (3 to 4")—Dark brown clayey silt with up to
34% sand (av. 20%), noncalcareous Glassford Formation
(C) Berry Clay Member (3 to 4^—Dark gray sandy silty clay
with trace gravel, up to 40% sand, noncalcareous (an
accretion gley)
(D,E) Vandalia Till Member (2 to 80—Yellowish brown,
oxidized in upper portion, little yellowish gray at base,
-------
182
Field Verification
(F) Smithboro Till Member (25 to 31 *)—Dark gray sandy silt
(G) Lierle Clay Member (1 to 2")—Dark olive brown sandy
silty clay, with little gravel, noncalcareous, soft. Banner
Formation
(H) Unnamed Till Member (1 to T)—Gray clayey silty till,
sandy, with some gravel, mottled yellowish brown,
brown and gray in upper part, strongly calcareous, hard,
dry, a few scattered silt and sand lenses, predominantly
dry Bond
Formation (Pennsylvanian bedrock)—Greenish gray
shale with abundant mica; hard, dry
The stratigraphic units and the textural mineralogical data
are shown in Table I. The expandable clay minerals, generally
referred to as montmorillonite (M), make up more than 80% of
the clay minerals within the Peoria Loess, Roxana Silt, and
Berry Clay Member, thereby suggesting a high base exchange
potential in the upper 13 feet of the materials encountered. The
thin, continuous silty sand zone at the top of the Vandalia Till
would appear to be the only "permeable" unit to allow ground-
water to travel laterally at any moderate rate away from the site.
Although there is probably some downward movement of
ground water through the remainder of the Glasford Formation,
it would be expected to be extremely low.
Water level measurements taken in the shallow wells were
used to construct monthly water table contour maps for the site.
These wells are from 10 to 15 feet below land surface and are all
finished in the sandy unit (D) at the top of the Vandalia Till
Member. Data collected to date show no significant change in
the shallow water levels at this site. The principal groundwater
flow paths from the plant site are illustrated in Figure 3. Water
from the eastern half of the plant, including the liquid disposal
pond, appears to be moving east southeast; water from the west-
ern half of the plant site seems to be moving west southwest.
Therefore, any pollutants in the shallow groundwater system
should be moving in the same general directions.
Table I: Textural and Mineralogical Data.
Units
(A)
(B)
(C)
(D)
(E)
(P)
(G)
(H)
Sd
M
Peoria
Loess
Roxana
Silt
Berry
Clay
Vandalia
Till
Smithboro
Till
Lierle
Clay
Banner
= sand, St
Thick-
ness
(ft)
4-6
3-4
3-4
.5-1.5
2-8
25-31
1-2
1-2
= silt,
%
Sd
3
20
28
51
34
29
21
28
Cl
%
St
58
49
28
31
35
42
40
41
=
% % M
Cl in Cl
39
31
44
18
31
29
39
31
clay
86
91
87
38
28
53
71
56
5
Total
% M
33
28
38
7
9
15
28
17
= montmorillonite
Figure 3. Water Table Contour Map, Site A (November 76).
-------
Field Verification
183
No piezometnc surface maps have been prepared for the
deeper units at this site. Measurements in the wells indicate that
water levels have not stabilized even after 5 to 6 months of non-
pumping (Figure 4). This is due to the low permeability of the
unit (Banner Formation) in which these wells are finished.
510
500
490
£ 480
470
460
450
440
LAND SURFACE
SHALLOW WELL
DEEP WELL
APR MAY JUN JUL AUG SEP OCT NOV
1975
Figure 4. Water Level Elevations.
Single element analyses for zinc have been completed on sam-
ples from most of the core holes at this site. Figure 5 illustrates
zinc concentrations in the soil along two vertical cross sections
through the study area. An explanation of the physical and
chemical phenomena controlling the movement of zinc into the
soil cannot be made until multi-element chemical analyses and
textural and mineralogical data are available for all soil samples.
From general observations, it is noted that the higher zinc
concentrations in soil are limited to the area directly beneath the
cinder fill covering the plant property. The deepest vertical pene-
trations also seem to lie beneath the areas of thicker cinder fill
and prefill topographic depressions.
The migration of zinc into the soil seems to be controlled in
large part by modern and ancient soils. Figure 6 is a detailed de-
scription of boring 2, showing the stratigraphic section, grain
size, clay mineralogy, carbonate content, and zinc concentra-
tions. The leaching of carbonates from the soil is the result of
soil-forming processes. There are several soil horizons repre-
sented in this boring. Soils are represented by horizons A, B, C,
rj), G, and at the Bedrock surface. The upper soil zone repre-
sents general soil-forming periods (including modern) superim-
posed on slowly aggregating materials. The accretion gley zones
in the tills (especially common in unit F, the Smithboro till) are
shear blocks of the underlying soil incorporated into the till as
the glacial ice advanced across the preexisting land surface.
NORTHEAST
a'
4qn _ SCALE OF FEET
0 100 200 300
Figure 5. Soil Zinc Concentrations.
Zinc has relatively high mobility in the soil zone which has a
pH around 7. Background values of zinc as high as 200 to 400
ppm may sometimes occur naturally in soils. Below the leached
soil zone, the pH increases to around 8 and free carbonate is
abundant; here the zinc will precipitate as ZnCO3 with very low
solubility. Examination of the borings for which analyses are
available at this time suggests that the zinc is concentrated
within the leached zone of the soils, with the exception of the
area directly below the scrubber waste pond. Either the massive
concentrations of zinc in the scrubber waste or leaching by the
acid water has caused migration into the underlying tills.
Analyses of water samples from the wells at this site are now in
progress. In general, zinc concentrations of water from all wells
are less than 1 ppm and do not appear to vary significantly from
month to month. Preliminary data also suggest that mineral
concentrations in the samples stabilize after the second or third
month of sampling.
-------
184
Field Verification
Complete analysis of all soil an.d water samples should pro-
vide the needed data to explain the phenomena controlling the
movement of heavy metals from this site.
SITE A
BORING S-2
DEPTH
OF
SPL SAMPLE UNIT GRAPHIC ZINC
No. ft DESCRIPTION LOG ppm
1 1.0
2 2.0
3 3.0
4 4.0
4.5
5 4.9
6 6.0
7 8.0
8 9.0
9 10.0
10 11.0
11 11.8
12 13.0
13 14.0
14 15.0
15 16.0
16 18.0
17 19.0
18 20.0
-- 20-20',
19 21.1
20 22.0
21 23.0
22 24.0
-- 24.5
23 25.0
24 26.0
25 27.0
- 28.0
26 30.0
27 32.0
28 33.0
29 34.0
30 35.0
-- 35.5
31 36.5
32 37.0
33 38.0
34 39.0
35 40.0
36 42.0
37 4?.0
-- 43.5
38 44.0
39 45.0
40 46.0
- 46.5
41 47.0
42 48.0
43 49.0
44 50.4
45 51.0
46 52.0
47 53.0
-- 53.5
48 55.0
49 5$. 5
50 57.5
-- 58-58',
51 59.0
Peoria
(6-60-34)
A
4.7
Roxana
(24-40-36)
Berry
(37-31-32)
C13.0
Pearl
^ D 14.5
Vandal i a
(38-35-27)
E 18.7
Smi thboro
(30-41-29)
F
Smithboro
(29-40-31)
36.7
accretion
qley
38.5
accretion
gley
accretion
gley
44.1
Lierle
(25-40-35)
G
48.2
(33-42-25)
Banner
(31-45-34)
H
59.5
Bedrock
t
81
iH
N>
KSC'
~\ , \""
:oi
7c->
-^
&
i^/-'
^v
^
v-/
'^'
W
(^
^/\^
^L
T/T~-""\-'
]/£\f
\/ - \-
r/'/f
1
y-v
_'_\_/
~\~
i
M'3
I? ''
7 \
'//////
1600
71,67
670,640
430
500
270
1100
45
37
130.110
33
39
47
41, 43
420
260
58
41
57
79
(130)
GRAIN SIZE
1 1 i 3
total
1 7 71 22
0 4 57 39
1 6 56 38
1 5 57 38
1 17 52 31
1 19 44 37
2 30 33 37
2 31 30 39
1 28 36 36
2 33 31 36
10 39 29 32
3 51 27 22
« » 38 18
7 37 37 26
-41 33 2o
4 35 36 29
5 30 45 25
3 32 '.S 28
3 28 43 29
4 28 39 33
3 35 37 28
3 30 39 31
4 30 41 29
5 30 40 30
8 27 42 31
3 30 38 32
3 26 38 36
5 29 40 31
5 29 41 30
3 30 38 32
3 27 42 31
1 22 41 37
1 20 40 40
14 33 37 30
1 18 49 33
4 29 41 30
3 27 40 33
5 31 43 26
2 24 39 37
2 25 38 37
2 26 41 33
4 25 44 31
4 30 41 29
8 32 41 27
0 36 45 19
1 21 52 27
3 21 44 35
2 18 47 34
4 22 44 34
4 22 45 33
4 22 44 34
MONTMORILLONITE
ILLITE
CHLORITE-KAOLINITE 5
-a T 0
!» 3 CALCITE 3
£ § DOLOMITE
82 12 6 - --
84 11 5 -- --
81 13 6 -- --
82 11 7 - --
81 12 7 -- -
87 8 5 - -
83 11 6 -- -
75 15 10 -- --
77 13 10 - —
65 26 9 — -
57 31 1C - —
25 56 19 -- -
27 57 16 - .' 25 1
29 54 17 26 15
22 53 25 25 18
43 38 10 19 14
"15 36 19 16 14
50 31 19 27 15
53 31 16 15 20
51 32 17 19 17
51 29 20 13 17
43 38 19 22 18
47 34 19 15 15
52 30 18 30 - !
50 31 19 17 16
55 26 19 15 16
46 34 20 18? --?
45 35 20 17 11
49 33 18 21 18
51 31 18 22 IS
76 14 10 12? --;
80 U 8 15 16
31 44 25 30 19
57 26 17 15 20
37 39 24 IS 23
82 9 9 -- 12?
49 36 15 34 28
43 39 18 16 30
60 28 12 -? 15?
28 51 21 -? 18?
31 47 22 30 28
21 54 25 34 24
25 54 21 61 25
24 53 23 33 18
21 51 28 33 11
20 52 28 37 20
18 51 31 46 22
19 53 28 50 22
17 52 31 29 20
Preliminary Cost Analysis:
As previously stated, it was decided in the beginning of this
project to devote maximum time, effort, and money at site A to
gain the needed experience and understanding of the methodol-
ogies being applied. Therefore the total expenditures at this site
are much higher than would be necessary at other sites of this
type. The cost to study a similar site using a minimum monitor-
ing approach would be considerably less.
As of January 1, 1976, approximately 1275 linear feet of
piezometer-tube observation wells at 34 different locations have
been completed at a total cost of about $6,150 ($4.80/foot).
Approximately 1300 linear feet of coring has been done for
about $9,150 ($7.05/foot).
As time permits and we gain better understanding of the mon-
itoring methods being tested, the cost effectiveness of the boring
technique combined with a minimum number of wells will be
determined. Preliminary data suggest that soil coring is a relia-
ble monitoring technique and definitely should be used to locate
the proper vertical horizons for installing monitoring wells.
SUMMARY
Work completed at all sites to date has demonstrated that soil
coring is an effective tool for mapping the migration patterns of
chemical pollutants through the earth materials. This approach
also provides field data to verify the effectiveness of various soil
types to adsorb or retain different chemical pollutants.
Sufficient comparative data are not available at this time to
determine the relative usefulness of the supplemental methods
being tested in this study. Temperature surveys, electrical earth
resistivity surveys, and vegetation mapping and sampling may
prove to be useful in certain shallow geologic regimes. Although
color infrared photography is very limited in detecting
groundwater pollution, it is a useful tool for soil and vegetation
mapping and for detecting surface water pollution.
ACKNOWLEDGMENTS
This study is sponsored by the Solid and Hazardous Waste
Research Division of the Federal Environmental Protection
Agency (Grant R/803216-01) under the general guidance of
Mike Roulier, Project Officer.
Figure 6. Description of Boring S-2, Site A.
-------
A Preliminary Assessment of the Effects of
Subsurface Sewage Sludge Disposal on
Groundwater Quality*
Dale C. Mosher
Office of Solid Waste Management Programs
U.S. Environmental Protection Agency
Washington, D.C.
INTRODUCTION
The environmental effects of subsurface sewage sludge dis-
posal have not been thoroughly investigated. This is particularly
true with respect to effects on ground water quality. This paper
presents the results of a preliminary evaluation of the effects of
sewage sludge disposal on groundwater quality. The term "sub-
surface sewage sludge disposal" as used here means the burial of
sewage sludge with or without municipal solid waste in trenches,
pits or area type landfills.
A study conducted at Oceanside, California1 on leachate from
municipal solid waste and mixed sewage sludge/ municipal solid
waste was completed in 1973. This study concluded that other
than a lower pH and higher BOD in the leachate from mixed
sewage sludge/municipal solid waste no other differences
occurred. A continuing study of sewage sludge only disposal was
initiated in 1972 by the U.S. Department of Agriculture, Agri-
cultural Research Service at Beltsville, Maryland. The data
from this study2 show groundwater has been affected after only
19 months. The testing has been limited to chloride and nitro-
gen compounds.
These studies did not allow projection of the total effect of
subsurface sewage sludge disposal on groundwater resources.
The Oceanside study, for example, was limited to leachate anal-
ysis only. The data then could only yield a determination of
potential pollutants. The USDA study did not include analysis
for toxic metals. Toxic metals analysis at the USDA field plots
will be started in the near future.
Sewage sludges, however, are being generated in ever increas-
ing quantities and state regulatory agencies must determine
what methods of sludge disposal are acceptable. In order to pro-
vide a broader data base on which to issue guidance to state reg-
ulatory agencies, Environmental Protection Agency's Office of
Solid Waste awarded a contract to SCS Engineers in Long
Beach California to monitor the effects of subsurface sewage
sludge disposal on groundwater quality. This paper will briefly
describe and report the results of that contract, drawing com-
parisons where appropriate to a similar and somewhat more
comprehensive study of municipal solid waste only sites.
Project Description
One objective of this project was to determine differences in
groundwater contamination resulting from subsurface sewage
* Based in part on the results of a study conducted for the Environ-
mental Protection Agency by SCS Engineers of Long Beach, California,
under Contract No. 68-01-3108
sludge disposal at eight land disposal sites where location and
operational conditions ranged from optimal to unacceptable.
Further, a similar and somewhat more comprehensive study
of ten solid waste sites was being conducted. Thus the second
objective of the study was to compare these two sets of data to
determine if differences in groundwater contamination occurred
between solid waste only and sewage sludge land disposal prac-
tices.
Site Selection and Description—The acceptability of any
given site relative to groundwater protection is dependent upon
many interrelated factors such as permeability, depth to ground-
water, subsurface soil, geologic characteristics, climatology, etc.
The major factors considered in site selection for this study were
as follows:
Factor
Soils/Geology
Precipitation
Operation
Age
Range
Sand to clay
59 to 115 cm/year
poor to excellent
3 to 5 years
It was expected that all sites would generate some leachate
given a minimum annual precipitation of 59 cm. It was also
expected that sites with clayey soils and excellent operations
would show less groundwater contamination (if any) compared
to sites with sandy soil and/or poor operations. Other factors
such as quantity of waste received, depth of wastes disposal and
depth to groundwater would also be expected to influence the
affect a specific site could have on groundwater quality. In
selecting sites to cover the above factors, a wide geographic dis-
tribution was necessary. The approximate location of study sites
is shown in Figure 1. The sludge quantities received at mixed
sites varied from 1 to 18% of the solid waste received on a weight
basis (dry weight for sludge and as received weight for solid
waste). At sludge only disposal sites the loading rates varied
from 2241-5603 tonnes/ha (1000-2500 tons/ac) on a dry weight
basis. Further description of the sludge and solid waste received
at each site is given in Table 1.
Monitoring Program—Wells were drilled at each site to sam-
ple leachate (if present) and groundwater downgradient from
the disposal site. Samples for background water quality were
obtained from nearby existing wells. The general well locations
are shown in Figure 2. Two wells, one shallow and one deep,
were established to sample grandwater downgradient, local con-
ditions permitting.
185
-------
186 Subsurface Sewage Sludge
IDAHO
•I mo
SIOWA
TKHH
Figure 1: Location of Sludge Monitoring Sites
/HHN
/MISS
U.K
Table I: Waste Characteristics for Sites Studied
Site
Criteria 1
Sludge Type Raw
primary
and
secondary
Solids 25-30%
Contents
Annual 13,555m3
Sludge
Quantity
Total 76,500m3
Annual Solid
Wastes Quantity
Proportion 17%
of Sludge to
Solid Waste
Received
(Volume basis)
2 3
Raw Raw
primary primary
and and
secondary secondary
paunch
manure
20-25% 25-40%
26,224m3 Variable
354,100m3 None
7% Sludge
Only
4
Raw
primary
and
secondary
18-25%
7,300m3
None
Sludge
Only
S 6
Raw Digested
Digested and
and septic
incinerated tank
sludge ash pumpings
20-2^% 3-5%
125,700m3 8,000m3
760,420m3 127,700m3
10% 6%
7
Raw
primary
and
secondary
an,d
Zimpro
sludge
40% Zimpro
20% where
down
15,200m3
585,000m3
26%
8
Raw
primary
and
secondary
15%
19,900m3
None
Sludge
Only
-------
Subsurface Sewage Sludge
187
. Leachate Sampling Well
_Shallow Well
— *r X .!
v
Fill Area
T
1!
•~Y ••
Gtoundwater Floi
Figure 2: Generalized Location of Monitoring Wells
Starting one month after the wells were established three sam-
ples of leachate and groundwater downgradient and one sample
of background water were taken during a five month period in
1975 from May to November. An additional sample was
obtained in June of 1976 from most wells. The relationship of
samples taken during this time period to the hydrologic cycle is
unknown. This sampling program was admittedly limited,
however, it must be remembered that the objectives were limit-
ed to (1) determining types of contamination occurring, (2)
compare contamination from sewage sludge to contamination
from solid waste only. The data for the latter being generated by
a somewhat more comprehensive monitoring program. In addi-
tion to the water samples, sludge at each site was analyzed, once
in 1975 and once in 1976.
Before presenting the results obtained, it may be beneficial to
briefly discuss the method of data presentation. Figure 3 shows a
theoretical distribution over time for some contaminant
concentration in both background and downgradient ground-
water. The variation over time shown is a function contaminant
source. The curve representing downgradient groundwater var-
ies then, according to the quality and quantity of leachate pro-
duced over time. The variation shown for background water
quality is exaggerated compared to "normal" uncontaminated
groundwater. This was done for illustrative purposes. From ex-
amination of this figure, it is obvious that when only three or
four samples downgradient are compared to one or two samples
of background groundwater, as in this study, examination of
individual data points can be confusing. In order to avoid a con-
fusing picture, I have averaged all available data points from
each well. In this manner it is possible to subtract background
levels from downgradient levels and display the "average"
increase. This method is obviously subject to inaccuracy, how-
ever, since normal background variation is generally quite small
the probability of showing an increase which is not real is min-
imal.
Downgradient
J J
TIME
Results and Discussion—The presentation of results is being
limited to heavy metals. This is because contaminants such as
sulphates, chloride, etc., have rarely exceeded their suggested
drinking water standard. Secondly, although increases in such
parameters often exceeded industrial standards, discussion of
groundwater contamination is most often related to drinking
water standards. Because standards on groundwater do not
exist, I will in general only, refer to increases above background
levels. Where it is necessary for comparative purposes, increases
will be compared to drinking water standards.
The data generated in this study will be presented in the
following sequence: the source of metals, the leachate
characteristics; and the nature of groundwater contamination.
Where appropriate the discussion will compare these data to
unpublished data from a similar but somewhat more co: ipre-
hensive study of solid waste only land disposal sites.
Source of Contaminants—The total content of selected
metals for the sludge presently entering each site is given in
Table II. Due to the lack of historical sludge data we are forced
to assume that the metals content of the sludge analyzed was
representative of that entering the site during its lifetime. This
assumption is probably reasonable for sites one thru five where
industrial input is apparently limited.
Table II: Metals Content of Sludges
Site
1
2
3
4
5
6
7
8
Cd
4
9
3
10
22
3
23
1
ppm {dr
Cr
111
65
150
120
780
2,750
43,300
33,300
r weight basi
Fe
4,100
13,000
4,700
23,800
67,000
75,000
6,700
2,000
)
Pb
170
220
170
110
j.,100
100
1,000
90
The range of leachate concentrations of selected parameters is
shown in Table III. These are compared to leachate
concentrations from municipal solid waste only sites. The
municipal solid waste only leachate analysis comes from the
draft final report of U.S.E.P.A. contract 68-01-2923 submitted
by SCS Engineers of Long Beach, California. The ranges for
sewage sludge only and mixed sewage sludge/ solid waste are not
separated as no differences were apparent. The only apparent
difference is a higher upper limit of metal contents at sludge
sites. No conclusion, however, can be drawn due to the limited
amount of data and site to site variability. In general, it appears
that the effect of leachate on groundwater quality should be the
same whether the leachate is from municipal solid waste, sewage
sludge or a combination of both.
Table III: Range of Leachate Characteristics
Parameter
TKN
Cl
COD
Cd
Cr
Fe
Pb
* Municipal Solid Wa
+ Sewage Sludge w/Ho
Leachate Source
MSW * ss +
11-758
139-568
165-13,000
.007-. 05
.65-. 33
15-679
.09-. 29
te Only - 5 sites SCS Contract
MSW - 6 sites SCS Contract
115-2513
3-1201
3,000-20,000
.009-. 1
.14-21
14-172
.1-1.55
No. 68-01
No. 68-01-3108
Figure 3: Theoretical Variation of Contaminant Levels in
Groundwater Over Time
This, however, does not appear to be the case. Data from this
study and the previously mentioned study of solid wastes only
show contamination of groundwater has occurred at all sites
based on indicator prameters such as specific conductive and
chemical oxygen demand (COD). Unlike solid waste only sites,
-------
188
Subsurface Sewage Sludge
however, the subsurface sewage sludge sites show a definite
trend toward contamination of groundwater with heavy metals.
Data for selected metals are shown in Table IV.
Table IV: Average Increase Over Background Level of Selected
Parameters
Parameters
Site
1
2
3
4
5
6
7
8
* Increase is
Cd Cr
.02
.65*
.65*
.02
.10*
.10* .38*
at or above applicable Drinking Water
Fe
.46*
539*
28*
1.10
.29
3.8*
374*
Standards
Pb
.03
.79*
.09*
.07*
.02
.34*
In examining the data in Table IV, a logical conclusion would
be that sites 5 and 6 are sanitary landfills and all other sites are
dumps. This, however, is not the case. Using "Sanitjiy Landfill
Design and Operation" as a guide, site 5 would oe called a sani-
tary landfill and site 6 a dump. The latter is largely due to poor
operation. Further, site 3, which shows rather significant con-
tamination, would also be classified as a sanitary landfill. I
should point out that the reference used discusses factors related
to the potential for ground water contamination but does not
give specific recommendations. Classifying any site then is a
matter of professionaljudgment. I believe that few people would
disagree with the above classification of sites 3,5 and 6. Because
metals contamination as shown in Table IV was found at dis-
posal sites accepting sewage sludge and generally not at solid
wastes only sites, suggests that some factor or factors other than
concentration are responsible for the observed results. Such fac-
tors could be chelation or other changes in chemical equilib-
rium brought about by the presence of sludge. It must be remem-
bered that the groundwater samples obtained here were within
61 meters (200 feet) of the landfills studies. The data do not pre-
dict what effect this contamination could have on groundwater
users further downgradient. This question will require further
study.
This study was designed to determine the types of contami-
nants found in groundwater as the result of subsurface sewage
sludge disposal. In developing this study it was assumed that no
greater effect would be observed than would occur from
municipal solid waste only. This was not found to be the case.
The limited scope of this study, however, does not justify
abandonment of subsurface sewage sludge disposal. It does,
however, justify greater emphasis in groundwater resource eval-
uation for sites which will accept sewage sludge.
It is obvious that more work is needed in the area of
subsurface sewage sludge disposal to better define the extent of
the problems and provide solutions. A contract has been
awarded to SCS Engineers to conduct further study of the sites
discussed. Detailed information is being obtained on soils, geol-
ogy, etc. to better define the cause and effect relationships that
exist.
At the present time, good site selection practices should
include a comprehensive ground and surface water resource
evaluation. This should include quality, quantity, direction and
rate of flow, present and potential use downgradient from the
disposal site. Where the water resource has a "high" value site
design and operation should offer protection of the water re-
sources.
Groundwater investigations are expensive to conduct, diffi-
cult to evaluate and can provide answers only after years of
study. We, therefore, cannot expect definitive solutions rapidly.
We can, however, periodically adjust our activities as more
information becomes available.
REFERENCES
1. "Sewage Sludge Disposal into a Sanitary Landfill"—Ralph
Stone and Company, SW-71d, U. S. Environmental Protection
Agency, 1974.
2. Trench Incorporation of Sewage Sludge in Marginal Agri-
cultural Land. J. M. Walker et al. Agricultural Research Ser-
vice, U.S. Department of Agriculture, Environmental Protec-
tion Technology Series, PB 246 561, U.S. Environmental
Protection Agency, 1975.
3. Sanitary Landfill Design and Operation—D. R. Brunner
and D. J. Keller. Environmental Protection Publication SW-65
ts Washington, U.S. Government Printing Office, 1972.
-------
Control Program for Leachate Affecting
A Multiple Aquifer System
Army Creek Landfill, New Castle County, Delaware
by
Walter M. Leis, Roy F. Weston
David C. Clark,
Dept. Pub. Works
New Castle County, Delaware
Abraham Thomas, Roy F. Weston
Kenneth D. Shuster,
US Environmental Protection Agency
Site History
Between 1960 and 1968, New Castle County, Delaware util-
ized an abandoned sand and gravel pit as the primary disposal
site for municipal and industrial wastes. The landfill location in
Northern Delaware is shown on the map in Figure 1.
From records and test borings, it was determined that a mod-
ified area fill method was used to emplace the refuse material.
During the land-filling operation, daily covering was probably
not practiced. This, coupled with less than adequate compaction
within the areas, caused differential settling and an uneven fin-
ished surface when the fill was closed.
The final lift was emplaced in 1968, and the completed fill
encompassed a volume of 1.9 million cubic yards (1.42 million
cu m) of refuse, about 30 percent (or 600,000 cubic yards,
450,000 cu m) of which was beneath the seasonal high water
table. Refuse thicknesses over the 47-acre (19 ha) site ranged
from about 6 to over 35 feet (2 to 11 m).
Late in 1971, water in a residential well, located 900 feet
southwest of the landfill, developed quality problems such as a
distinctly disagreeable odor and permanent staining of porce-
lain fixtures. Gradually, this condition became more pro-
nounced and the water supply was abandoned. Further analyses
conducted by the Delaware Geological Survey and New Castle
County Public Works Department indicated the presence of lea-
chate in the ground water. In early 1972, intensive field study
was begun by the county, through its consultant Roy F. Weston,
Inc., and the Delaware Geological Survey to determine the
hydrogeologic controls responsible for this problem.
Hydrogeologic Description
The surficial geology in the old gravel pit consists of medium
and coarse-textured, channel-bedded, unconsolidated sands
that demonstrate substantial textural variability within a short
vertical section. The origin of these sands has been related to gla-
cial meltwaters having seasonal discharges.2 The deposits were
placed in shallow, relatively narrow braided channels with
numerous ice floes. These deposits have been formally named
the Columbia Formation and within the study area form a con-
tinuous surficial layer up to 60 feet (18 m) in thickness with the
formational base varying between +20 to -20 feet (+6 m), present
mean sea level. Dragline excavations in the gravel pit during
quarrying had terminated at a basal quartz conglomerate. It has
been determined from air photos that in at least two places
within the gravel pit this basal conglomerate was removed by
excavation and the bottom of the gravel pit was continued into
the underlying aquifer sands of the early Cretaceous Potomac
Formation.
The Potomac Formation of Early to Late Cretaceous age1
consists of unconsolidated sands, silts, and clays genetically
emplaced as channel deposits by southerly flowing meandering
streams.3 Figure 2 is a "hydrogeologic-stratigraphic" column for
this area. A fairly consistent Potomac layer underlying the post-
Cretaceous erosion surface consists of variable thicknesses of
clay. Due to the lensatic depositional nature of the Potomac
Formation, this upper clay layer varies in thickness and textural
character throughout a cross section of the landfill to a north-
erly direction. In this area the clay layer is sandy or completely
absent, while immediately south of the landfill the layer varies in
thickness from 30 feet to over 100 feet (9 to over 20 m).
Hydrologically, the Potomac upper clay functions as a con-
fining zone for the underlying aquifer, which has been named
the upper Potomac aquifer.4 Pump tests conducted in this upper
50-100 feet (15-30 m) thick aquifer zone yield transmissivity
values in the range of 45,000 to 70,000 gpd/ft (560 to 870 sq
m/ day), and storativity values in the 10-4 to 10-5 range. Leakage
coefficients of 0.028 to 0.033 ft (0.85 cm to 1.0 cm) have been
computed for the confining clay. The lower aquifer, although a
coarser sand, is relatively thin in this area, and transmissivity
values of less than 1,000 gpd/ft (12 sq m/day) are common.
The upper aquifer zone has been heavily developed for
ground water since the early 1960's. In the immediate area
pumpage had exceeded 8-10 mgd (30,000 to 38,000 cu m/day)
during this decade. Consequently, the natural southerly ground-
water gradient has been steepened to the south, and the rate of
ground-water movement has been greatly accelerated. As a
result, Potomac potentiometric heads have declined to about 20
feet (6 m) below sea level.
Contamination History
In mid-1972, upon discovery of contimination in a residential
well, an initial investigation was launched to determine the
extent of the leachate-affected area. Figure 3 is a map showing
the leachate plume as first determined by monitoring several
drilled observation wells. This earliest monitoring well network
was soon expanded about the periphery of the landfill and
included a significant down-gradient area. These wells served to
facilitate the monitoring of leachate migration in selected areas
of the aquifer near existing residential and municipal wells,
including a major well field belonging to a water company. As a
result of the expanded monitoring network, by early 1973 the
189
-------
190
New Castle, Delaware
Figure 1: Location Map of Army Creek Landfill
Pleistocene
Columbia Formation
Cretaceous /•—
Potomac Formation ^
T
100ft
1
Water Table Aquifer
Basal Conglomerate
Unconformity
Upper Aquifer
Lower Aquifer
Crystalline Basement
Figure 2: Hydrogeolic-Stratigraphic Column for Army Creek
Area
migration of the leachate plume was found to be farther down-
gradient than was originally thought.
Based on aquifer characteristics of the Potomac, the rate of
contaminant migration was computed. The permeability of the
channel sands of the Potomac is greater 500 gpd/sq ft (20 cu
m/day sq m). The hydraulic gradients, as can be inferred from
the potentiometric surface map in Figure 4, vary from about
0.015 ft/ft immediately sough of the landfill to 0.02 ft/ft, as the
gradient steepens nearer the production wells. An average rate
of movement of 0.5 feet per/day (15 cm/day) is reasonable for
the area close to the landfill and increases to 1.5 ft/day (46
cm/day) down-gradient closer to the pumping centers of the
nearby wells.
As leachate enters the upper Potomac aquifer, barring atten-
uation of select constituents, it was expected to reach the water
company's pumping wells within 10 years if no corrective proce-
dures were undertaken.
Leachate Quality
The quality of landfill leachate generated by saturated decom-
position of domestic refuse depends upon such variables as
waste composition and sorting, compaction, moisture capacity,
temperature, and age of the refuse. Thus, the actual concentra-
tions of chemical species in various leachates vary greatly. A
young landfill will generally produce a leachate characterized by
very high concentrations of dissolved organic and inorganic
substances. Both the Chemical Oxygen Demand (COD, a mea-
sure of dissolved organics) and the Total Dissolved Solids con-
centration (TDS, a measure of dissolved inorganics) are typi-
cally in the thousands of parts per million for the raw leachate.
-------
New Castle, Delaware
191
The concentration of most environmentally significant metals
greatly exceeds the recommended EPA limits for such
substances in potable water. The leachate concentration rapidly
decreases as it migrates down-gradient from the landfill by such
processes as dilution, bioassimilation, chemical oxidation, and
adsorption.
The background water quality in the Potomac aquifer has a
low dissolved solids concentration. The range of inorganic con-
stituents in natural Potomac waters is listed in Table 1, along
with comparison values from selected observation wells located
within and down-gradient of the landfill.
Counter-pumping Program
The installation of a network of monitoring wells was com-
pleted in early 1973, so that the ground-water area affected by
migration of leachate could be fully identified.
Figure 5 shows the line of contaminant migration identified in
late 1973. This line is an approximate location based upon spe-
cific parameters used as standard indicators. These parameters
included physical appearance of water, e.g. color and odor, and
COD, total iron, manganese, and chlorides. The concentration
of iron and manganese in aquifer waters is such that they are use-
ful as quality parameters. Their gradual increase over back-
ground conditions has, therefore, indicated: (1) remobilized
metals under nonequilibrium conditions and/or (2) fresh dis-
solved metals carried in solution. The COD levels within the
landfill were normally high (see Table I). A few hundred feet
south in monitoring wells nearest the landfill the COD levels had
dropped significantly, indicating that organic decomposition of
the landfill was subsiding or, more likely, that fresh, oxygenated
recharge waters were fully oxidizing the organics within the
Potomac aquifer through interformational leakage.
Through an analysis of the initial flow network in the
Potomac aquifer and computed well interference from selected
observation wells, an interim contaminant control was
designed. The control consists of an artificially depressed drain-
age divide within the upper Potomac aquifer, down-gradient
from the landfill at about a third of the linear distance to the
water company well field. The drainage divide, as designed, was
composed of a two-part well interference system. Counter-
pumping wells were systematically located within the channel
sands of the upper Potomac aquifer to create the initial divide.
Secondly, the water company wells were throttled back to
initiate head buildup down-gradient. The counter-pumping re-
covery wells have been in operation since November 1973. At
the time of completion in early 1974, the wellfield was pumping a
total of 3.54 million gallons of contaminated water per day
(13,400 cum/day).
The divide has been effective in preventing contaminated
waters from flowing into the steepened gradient of the water
company's wells. In plan view, the divide consists of roughly an
L-shaped ground-water mound south and east of the landfill as
shown in Figure 6. The flow lines again indicate the preferred
flow path of a particle within this network.
Figure 3: 1973 Contamination Front
-------
192
New Castle, Delaware
Map showing the theoretical ground water flow pattern in the
upper Potomac Aquifer based on piezometric and stratigraphic
information in the vicinity of the Llangollen Landfill, September 1972
10 ft
Production Well
Observation Well
Piezometric Contour (feet below mean sea level)
10 Foot Contour Interval
Flow Lines
0 1000 2000 3000 4000
!=-=-] t=- I I
Scale in Feet
Figure 4: Potentiometric Surface Map of the Upper Potomac Aquifer Prior to Installation of Control Measures
Figure 5: Extent of Contamination Migration as of August, 1973
-------
New Castle, Delaware
193
An added feature to the divide is the physical removal of con-
taminants at that point. The leakage characteristics through the
Columbia Formation south of the counter-pumping network
then allow fresh recharge to enter the aquifer. The recovered
contaminated water is allowed to settle within a perched reten-
tion pond south of the landfill. Through mild oxidation, certain
contaminants are removed and the overflow infiltrates the
Columbia Formation and the Delaware Estuary.
In cross section, the development of the drainage divide dur-
ing the history of the leachate containment is shown in Figure 7.
Line A shows the piezometric gradient form north to south
between the landfill and the community wells. Line B shows the
same traverse line after initiation of the control measures in late
1973. Since that time, the piezometric gradient has risen and
declined with respect to increased recharge, seasonal demand,
chemical and bacteriological encrustation of the recovery wells,
and later rehabilitation; still it has remained fairly constant in
position and effectiveness to this day.
Table I: Concentration of Specific Chemicals and Contaminants Near Army Creek Landfill
Background Quality Water table Beneath
Potomac Aquifer Landfill
COD (ppm)
IDS (ppm)
NH,N (ppm)
NO,N (ppm)
Cl (ppm)
2Fe (ppm)
Phenols (ppm)<
Sp C
(fj. mhos cm'1)
14
26
.55
1.75
9.2
.20
-20 ppb
1.20
5617
3606
113.4
—
436
550
18
2500
Potomac Aquifer 400
ft. Downgradient from
Landfill
170
933
72.2
—
236
47
01
1900
Piezometric surface
3 1976
contours are in feet below sea level
V
Figure 6: Location of Recovery Wells at Army Creek
-------
194 New Castle, Delaware
LANDFILL
Well 56
Horizontal Scale
0:800
Vertical Exageration
SOX
COMMUNITY-WELLS
Well 40
Since October 1974
Prior to Initiation
of Controls
Well 22
Figure 7: Cross Section through the Piezometric Surface "Drainage Divide"
• Wells having improved
water quality
Figure 8: Extent of Contaminant Migration as of May, I976
-------
New Castle, Delaware
195
CONCLUSION
The counter-pumping measures at the Army Creek Landfill
have been successful. Figure 8 shows an interpretive line of con-
taminants based upon spring 1976 data. In certain fringe area
monitoring points, the groundwater quality has actually been
shown to improve, although it is still below U.S. Public Health
Service limits and EPA interim standards for potable water.
Although successful, the present containment program is just
that—a containment program. No resulting aquifer rehabili-
tation in the affected area can be expected unless the source of
the contaminant problem is removed or effectively neutralized.
To date, approximately 1.35 million dollars have been spent
in an effort to contain contaminant migration and preserve the
ground water which, in coastal Delaware, is the primary source
of drinking water. Although the present counter-pumping mea-
sures have been successful, it is evident that a more permanent
solution must be reached. To this end, New Castle County has
directed efforts toward an economically sound and environ-
mentally acceptable solution.
REFERENCES
1. Groot, J. J. (1955) Sedimentary Petrology of the Creta-
ceous Sediments of Northern Delaware in Relation to Paleogeo-
graphic Problems. Del. Geol. Survey Bull. No. 5.
2. Jordan, R. R. (1964) Columbia (Pleistocene) Sediments of
Delaware. Del. Geol. Survey Bull. No. 12.
3. Spoljaric, N. (1967) Quantitative Lithofacies Analysis of
the Potomac Formation, Delaware. Del. Geol. Survey Rept.
Invest. No. 12.,
4. Sundstrom, R. W., et al. (1967) The Availability of Ground
Water from the Potomac Formation in Chesapeake and Dela-
ware Canal Area, Delaware. University of Delaware Water Re-
sources Center.
-------
The Potential for
National Health and Environmental Damages
from Industrial Residue Disposal
Emery C. Lazar, Robert Testani and Alice B. Giles
Hazardous Waste Management Division
Office of Solid Waste Management Programs
U.S. Environmental Protection Agency
Washington, D.C.
INTRODUCTION
According to current estimates by the Office of Solid Waste
Management Programs, 344 million metric tons of industrial
wastes are produced annually in the United States, with a yearly
growth rate of about three percent. This is more than twice the
combined quantities of municipal solid wastes and sewage
sludge. We also estimate that at least ten percent of all industrial
wastes are potentially hazardous. This, of course, raises a
serious environmental issue, which received special attention
with the publication of an EPA Report to Congress in 1973.'
The Report concluded that the prevailing methods of hazardous
waste management are largely inadequate and result in the
uncontrolled discharge of hazardous residues into the environ-
ment, thereby threatening the public health and welfare.
Types of Documented Damages
The Office of Solid Waste Management Programs has com-
piled hundreds of case studies of damages resulting from
improper industrial residuals management; some of these have
been published.2,3,4,5 In the course of this data gathering effort,
we have recognized six major routes of environmental transport
through which the improper land disposal of hazardous wastes
can result in damage: (1) groundwater contamination via leach-
ate; (2) surface water contamination via runoff; (3) air pollution
via open burning, evaporation, sublimation, and wind erosion;
(4) poisoning via direct contact; (5) poisoning via the food chain;
(6) fire and explosion. These forms of damage were discussed,
and illustrative case studies were presented in a previous publi-
cation.6 As examples of potential human health injuries from
industrial waste disposal, we would like to mention only two of
the more recent incidents that have come to our attention:
• A case of suspected chronic mercury poisoning on a farm in
Illinois is currently under litigation. It involves a family of
six who exhibited the typical symptoms of Minamata dis-
ease after prolonged drinking of well water from an aquifier
which had been contaminated by the disposal of industrial
wastes three quarters of a mile away.
• A previously reported case of solvent disposal in a Mary-
land sand and gravel quarry has recently resulted in litiga-
tion. The suit involves three members of a family living near
the quarry who have sustained pancreatitis, a condition
associated with chronic solvent exposure.
Tabulation of Damage Data
The Office of Solid Waste Management Programs has com-
piled an inventory of over 400 cases of damage resulting from
waste disposal practices. The majority of case studies in the
inventory relate to industrial processing waste disposal; how-
ever, damages from the disposal of pesticides and pesticide con-
tainers have also been incorporated. The primary sources for
this data gathering effort were State environmental regulatory
agencies.
Based on 421 industrial and pesticide waste-related damage
case studies compiled to date, we have prepared a number of
tabulations which may help in reaching some preliminary con-
clusions about prevailing damage trends.
Table I summarizes the damage mechanisms involved in the
analyzed case studies by disposal method. It indicates that
groundwater contamination is the most common type of dam-
age reported, followed by surface water contamination.
Moreover, in most cases of established groundwater contamina-
tion, actual water supply wells (as compared to monitoring
wells) have been affected. The table also shows that "other land
disposal," which generally refers to promiscuous dumping or
dumping on land not designated for this purpose, is the most sig-
nificant source of damage. Surface impoundments and landfills
contribute in about equal measure; however, these disposal
methods together account for about as many damage incidents
as "other land disposal." It should be noted that the data sum-
marized in the table are not nationally representative since 65
out of the 421 cases studied were obtained from an incomplete
survey of one State that already has a permit system for landfills
and surface impoundments. The most flagrant environmental
offenses generally occur in those States that do not have regu-
latory programs for industrial waste disposal. Further, such
States generally do not have adequate documentation of dam-
ages.
Table II shows the contaminants identified in damage inci-
dents by disposal method. It should be emphasized that in most
documented damage cases chemical analysis of the contami-
nants is incomplete. This is mainly due to the expensive nature
of thorough laboratory analysis, especially when organic con-
taminants are involved. Despite its incompleteness, the table is
informative because it identifies a wide range of harmful and
potentially harmful contaminants. The largest category, miscel-
laneous organics (identified in 86 separate incidents), includes
some known and suspected carcinogens.
Finally, two other interesting observations derived from the
tabulation of case studies should be noted. One is that in 63 per-
cent of the incidents of damage, the causative waste disposal
action occurred on the property of the waste generator,
although in many instances the damage had spread off-site when
it was discovered. The second observation relates to the time
frame of discovery of damage. Sixty percent of the available
damage incidents were discovered during the past five years;
however, the acts of waste disposal responsible for the damage
may have occurred years or even decades earlier.
196
-------
Health and Environmental Damages
197
Table I: Mechanisms Involved in Incidents of Damage by Disposal Method8
Disposal Method
Number of Cases
Damage Mechanism
(number of cases)
Groundwater
(259)
Surface Water
(170)
Air
(17)
Fires, Explosions
(14)
Direct Contact Poisoning
(52)
Wells Affectedc)
(140)
Smeltings ,
Surface Landfills, Other Land Storage Slag, Mine
Impoundments Dumps Disposal b) of Wastes Tailings
89 99 203 15
57 64 117 10
42 49 71
3 5 9 -
11 3
1 6 40 5
32 28 74 4
a) The tabulation refers to 421 cases studied thus far. The numbers in the matrix add up to
than 421, because several damage incidents involved more than one damage mechanism.
b) Haphazard disposal on vacant properties, on farmland, spray irrigation, etc.
c) Not included as a damage mechanism.
Note: The data presented in
this table have been derived solely from case studies associated
15
11
8
-
_
-
2
more
with
land disposal of industrial wastes.
The Potential Nationwide Impact
Most hazardous waste-related human health injuries result
from months and years of chronic exposure at trace concentra-
tions of the toxicant. The health effects are insidious and almost
impossible to trace back to the causative agents. Yet, even in the
absence of positive identification—and especially
quantification—of health injuries, there should be serious con-
cern about any large-scale release of toxicants, carcinogens, and
other harmful materials into the environment. In the following,
we shall relate industrial waste generation data to information
on waste disposal practices, in order to indicate the potential for
nationwide health and environmental damage.
The Office of Solid Waste Management Programs has studied
14 key industries to gather data on waste generation rates and
disposal practices. These industries, which are thought to gener-
ate most of the nation's potentially hazardous wastes, are indi-
cated in Table III, together with a partial listing of their waste
components. The 14 industry categories generate about 206 mil-
lion metric tons (wet) of process and pollution abatement resid-
uals, of which about 35 million metric tons can be considered
potentially hazardous, chiefly on the basis of toxicological con-
siderations. The annual growth in the generation of potentially
hazardous wastes is estimated at three percent for the next
decade.
In order to gain an understanding of the nationwide potential
for industrial waste disposal-related health and environmental
damages, we have made an assessment of the environmental
adequacy of current disposal practices, as these relate to the esti-
mated 35 million metric tons of potentially hazardous residuals
indentified in the aforementioned industry categories. The
assessment of environmental adequacy was based on the dispo-
sal practices employed by typical manufacturing facilities, i.e.,
those which employ average disposal technology relative to
other plants in the same industry. The modes of disposal have
been categorized as environmentally adequate or inadequate,
whenever possible on the basis of each contractor's judgment,
which was largely arrived at by numerous plant visits. However,
in many instances contractors were not able to make such a
judgment but simply categorized the prevailing disposal practice
into a series of progressively less adequate methods. An exam-
ple of such a progression would be "secure landfill—approved
landfill—landfill—dump." In such instances the authors made
the determination of environmental adequacy. The judgment
was made conservatively for the purposes of this analysis. In the
example cited above, disposal at "approved landfills" was
termed environmentally inadequate, because the Office of Solid
Waste Management Programs has on file numerous case stud-
ies demonstrating damage at landfills which have been approved
by State and local authorities only on the basis of a lack of alter-
natives.
-------
198 Health and Environmental Damages
Table II: Contaminants Involved in Damage Incidents by Disposal Method (Cases Studied: 421)
Disposal Method
Contaminant
As
Cd
Cr
Cs
Cu
Fe
Hq
Mn
Ni
Pb
Zn
Cl
CN
F~
NH3
NOj-
S0^=
Inorganic Acids
Vlisc . Inorganics
PCBs
Petrochemicals
Phenols
yiisc. Organics
Bacteria
Pesticides
Radioactive
Unspecified Leachate
Itotal
*Disposing on vacant {
.H
$
&
19
5
33
1
20
40
11
26
13
22
22
27
19
8
14
16
18
27
83
3
27
31
88
11
71
9
25
689
Drope
Impoundments
5
3
11
6
10
1
3
5
5
9
11
6
5
6
6
9
9
21
10
9
19
1
1
2
5
178
rties, on farm
TjanH-Fi 1 1 o
Dumps
4
1
9
1
4
20
1
15
2
6
5
6
4
2
2
2
4
25
1
5
10
25
2
6
3
18
183
land, spray .
O-f-VioT" Tj=mr1
Disposal*
10
1
12
7
6
9
4
5
8
5
9
9
3
6
7
5
10
29
2
10
12
39
8
57
1
1
275
irrigation, e
Smelting,
Tailings
1
3
4
4
1
3
3
1
2
4
6
1
1
34
JtC.
Q-f-Oy^fT^
1
2
2
5
7
2
19
Table IV categorizes the prevailing disposal practices on the
basis of estimated environmental adequacy and indicates the
percent of total wet weight attributable to each disposal prac-
tice. The table shows that the prime disposal method employed
for potentially hazardous industrial wastes is lagooning in
unlined surface impoundments, accounting for nearly half of the
total of these wastes disposed. Lined impoundments which were
considered adequate receive less than 0.1 percent of the total.
Land disposal in dumps or on other non-secure land surfaces
receives the second largest quantity. Together, these forms of
land disposal account for over 80 percent of the total. Incinera-
tion is the third major disposal practice now in use, with uncon-
trolled incineration accounting for almost twice the amount
adequately handled through controlled incineration. The result
is that over 90 percent of the approximately 35 million metric
tons (wet weight) of potentially hazardous wastes generated
yearly by 14 key industries is handled by disposal practices
which do not seem adequate to provide protection of public
health and the environment. This estimate may be somewhat
pessimistic, because a small portion of unlined surface
impoundments and non-secure landfills may be located in areas
which preclude the escape of pollutants into the environment.
However, this estimate may not be exaggerated if one considers
that the locations for landfills and dumps have been tradition-
ally selected on the basis of economic rather than environ-
mental considerations. Waste disposal usually takes place on
land that has little or no value for other uses, such as marsh-
lands, abandoned sand and gravel pits, old strip mines, lime-
stone sinkholes, etc. Most of these sites have hydraulic connec-
tions with natural waters. Similarly, most industrial surface
impoundments are unlined and were not sited on the basis of
hydrogeological considerations. Therefore, one could venture
to say that up to 90 percent of potentially hazardous wastes are
disposed of by questionable methods and are ultimately suscept-
ible to escape into the environment.
-------
Health and Environmental Damages
199
Table III: Partial Listing of Waste Components of Fourteen Industries
a
c
8
Id G^
o -3
o fd
(fl -H i — 1
(D £ Q,
•H o a
(d iH H
m w w
Ammonium salts
Antimony
Arsenic
Asbestos
Barium
Beryllium X
Biological wastes
Cadmium XXX
Chlor. hydrocarbons X
Chromium X X
Cobalt
Copper XXX
Cyanide X X
Ethanol waste, aqueous
Explosives (TNT)
Flammable solvents X
Fluoride
Halogenated solvents X
Lead X X
Magnesium
Manganese X
Mercury X
Molybdenum
Nickel X
Oil X
Organics, misc.
Pesticides (organo-
phosphates)
Phenol x
V
Phosphorus
Selenium
Silver X X
Vanadium
Zinc XXX
*Including pesticides and explosives
CO
•H
•S
CO fa
,H *
Bt£ CO
H
TO J~!
U EH U
•H
Q ra tn
G CD M
H i-q o
X
X
X
X
X X
X X
X
X
X
X
X
X X
X
X
X
X
X
X X
Paints
X
X
X
X
X
X
X
X
X
X
X
X
s
jjj
0)
Petrol
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
CO
1
-jJ
d)
Q
CM
x-
X
X
X
X
X
X
X
X
CO
H
-S
fll
«*>•
^1
•AJ
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
CO
o
•H
-p
a
CM
t-8
1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
B1
•H
£!
CO
•H
•1 S1
CM -H
c
1 1 1
jd i)
•S ' — '
CO -H
H 0) O
(0 H
•H -H 0)
cu "S eg
jl m ra
w IH IS
X X
X X
X X
X X
xxx
xxx
X
X X
X
X X
X
X X
X X
X
X X
xxx
-------
200
Health and Environmental Damages
Table IV: Estimated Environmental Adequacy
of Disposal Practices
for Potentially Hazardous Wastes*
DISPOSAL PRACTICE
ENVIRONMENTALLY INADEQUATE **
Unlined Surface Impoundments
Men-Secure Landfills
Uncontrolled Incineration
Deep-Well Injection
Landspreading
Use on Roads
Sewered
Total
ENVIRONMENTALLY ADEQUATE
Controlled Incineration
Secure Landfills
Recovery
Lined Surface Impoundments
Wastewater Treatment
Autcc laving
Total
*Based on annual generation in 14 key
period 1973 1375.
PERCENT OF TOTAL WET
WEIGHT OF POTENTIALLY
HAZARDOUS WASTES
.48.4
30.3
9.7
1.7
.3
<.l
<.l
90.4
5.6
2.3
1.7
< .1
< .1
< .1
9.6
industries during the
**Criteria for environmental adequacy defined in text.
There are other circumstances which underscore the signifi-
cant potential for nationwide damages from industrial waste
disposal. One is that most manufacturing industries are located
in the "wet regions" of the nation, where groundwater and sur-
face water contamination are most likely. Figure 1 illustrates
this point by superimposing the major areas of industrial activ-
ity on a contour map combining rainfall and evapotranspiration
data. The areas of maximum rainfall and minimum evaporation
are generally in the eastern third of the nation, where most
manufacturing and consequently industrial waste disposal takes
place.
A similar map (Figure 2) shows where the nation's principal
underground aquifiers are located in relation to industrial con-
centration. One can draw three general conclusions from this
map. The first is that most areas of high industrial concentration
are underlain by principal aquifers. The second is that some of
the most heavily used aquifers are located in dry regions of the
nation, where the risk of groundwater contamination from
land disposal practices is relatively small. The relatively small
risk is counterbalanced by the fact that any contamination of
these scarce water resources would result in particularly severe
environmental and economic damage. The third conclusion
derived from this map is that many groundwater aquifers in
highly industrialized areas are not currently exploited as major
water resources. This interpretation is somewhat misleading,
since the map does not designate those areas where groundwater
usage is only moderate now but is expected to increase signifi-
cantly. From an environmental perspective, of course, the risk
of groundwater contamination should be viewed with concern,
regardless of current usage rates.
In spite of the large rate of generation of potentially hazard-
ous wastes and the prevailing questionable disposal practices.
the documented cases of health and environmental damage are
relatively small. However, for every waste disposal site docu-
mented as a source of environmental contamination, we
estimate that there are thousands which are situated and oper-
ated in a similar manner. The vast majority of such sites are not
monitored for the escape of pollutants into the environment.
Virtually all of the case studies in our damage inventory were
discovered after damages had already occurred. Specifically in
the case of groundwater contamination, the problems usually
manifest themselves only after years—and sometimes decades—
of environmental insult. The abatement of damage—if techni-
cally feasible—may take even longer and could be too costly to
implement. A good case in point is the groundwater contamina-
tion in Colorado described in Reference 3, where just a compre-
hensive study of the problem—excluding abatement—may cost
up to 78 million dollars.
CONCLUSIONS
• The prevailing hazardous waste disposal practices have
often resulted in the release of harmful (e.g., toxic, carcino-
genic) materials into the environment, causing injury to
human health and also environmental and economic dam-
ages.
• The area of most serious concern is the impact of these dis-
posal practices on human health. The effects are mostly
chronic, following years of exposure at trace amounts of
the toxicants.
• Of over 400 hazardous waste disposal-related damage inci-
dents documented by the Office of Solid Waste Manage-
ment Programs, the majority relate to groundwater con-
tamination. Most cases of detected groundwater
contamination affect water supply wells.
• In groundwater contamination cases, years or decades may
elapse before the damages become evident. After-the-fact
abatement is extremely costly, if at all feasible.
• Of the approximately 35 million metric tons of potentially
hazardous wastes generated in 14 key industry may not be
adequate to prevent their escape into the environ ment.
• Over 80 percent of potentially hazardous wastes go to land-
fills, dumps, and surface impoundments. Most of these
were located and designed on the basis of economic rather
than environmental considerations.
• Most areas of industrial concentration, and consequently
disposal sites, are located in wet regions of the nation,
where precipitation exceeds evapotranspiration potential,
increasing the likelihood of soil infiltration and runoff.
• Most areas of industrial concentration are underlain by
groundwater aquifers which are vulnerable to pollutants.
REFERENCES
1. U.S. Environmental Protection Agency, Office of Solid
Waste Management Programs. Disposal of hazardous wastes;
report to Congress. Environmental Protection Publication SW-
115. Washington, U.S. Gov Printing Office, 1974. 110 p.
2. Hazardous waste disposal damage reports. Environmental
Protection Publication SW-151. [Washington], U.S. Environ-
mental Protection Agency, June 1975. 8 p.
3. Office of Solid Waste Management Programs: Hazardous
waste disposal damage reports; document no. 2. Environmental
Protection Publication SW-151.2. [Washington], U.S. Environ-
mental Protection Agency, Dec. 1975. 12 p.
4. Office of Solid Waste Management Programs. Hazardous
waste disposal damage reports; document no. 3. Environmental
Protection Publication SW-151.3. [Washington], U.S. Environ-
mental Protection Agency, 1976. 13 p.
-------
Health and Environmental Damages
201
-50
-60
KEY
Industrial Centers of the U. S.
(Based on Manufacturing Employment: 1963)
( J 120,000 - 479, 999
o
o
> 1,000, 000
400,000 -999,999
o
48, 000 -119, 999
24, 000 - 47, 999
Contours represent mean annual precipitation
(inches) minus potential evaptranspiration.
^National Atlas. 1970
^Proc. of joint conference on recycling
municipal sludges and effluents on
land. Champaign, U. of III., 1973. p. 117.
Figure I: Precipitation—Evapotranspiration Potential Contours and Industrial Centers of the Conterminous U.S.
5. Ghassemi, M. [TRW Systems Group]. Analysis of a land
disposal damage incident involving hazardous waste materials,
Dover Township, New Jersey; final report. Washington, U.S.
Environmental Protection Agency, Office of Solid Waste Man-
agement Programs, May 1976. I2l p. (Unpublished report.)
6. Lazar, E.G. Damage incidents from improper land dispo-
sal. Journal of Hazardous Materials, I(2):l57-l64, Jan. 1976.
ACKNOWLEDGMENT
The authors wish to express their appreciation to Messrs.
Arnold Edelman and Barton Ives and other colleagues at the
Hazardous Waste Management Division who helped compile
the data base utilized in this paper.
-------
202 Health and Environmental Damages
KEY
Industrial Centers of the U.S.
(Based on Manufacturing Employment: 1963)*
C ) 120, 000 -479,999
( 1 > 1,000,000
^ — s r~\
\ ) 48,000—119,999
O480, 000 - 999, 999 ,-.
(_) 24, 000 - 47 999
Each dot represents withdrawal of si.ooo acre -feet
or 100 million cubic meters annually, equivalent
to 71 million gallons per day.
Shaded areas represent groundwater areas
(major aquifers) of the U.S.
* National Atlas, 1970
T Water Atlas of the United States. 1973
Figure 2: Major Aquifers, Well Withdrawals, and Industrial Centers of the Conterminous U.S.
-------
Health Effects-
Land Application of Municipal Wastewater and Sludge
Thomas L. Gleason, III and Frank D. Kover
U.S. Environmental Protection Agency
Washington, D. C.
Charles A. Sorber
University of Texas at San Antonio
San Antonio, Texas
INTRODUCTION
The number, size and complexity of wastewater and sewage
treatment plants is increasing constantly. Whether or not there
are potential health hazards arising from the treatment process
itself and various proposed methods of disposal of the sludges
produced is largely unknown. The problem is nationwide.
Currently, 12 billion gallons of wastewater and 17,000 dry
tons of sludge will be treated and processed daily. It is estimated
that by 1985,22 billion gallons and 23,000 dry tons of sludge will
be treated and processed daily (Table I).1
The Federal Water Pollution Control Act (FWPCA) as
amended in 1972 (PL 92-500) requires that all publicly owned
treatment works install secondary treatment by July 1, 1977.
Consequently, the quantity of wastewater to be treated and
sludge to be disposed of will increase. The Marine Protection
Research and Sanctuaries Act of 1972 (PL 92-532) requires per-
mits for dumping of wastes in the ocean, including municipal
sludge. Section 102(g) requires that permit applicants consider
appropriate locations and methods of disposal, including land-
based alternatives. This impacts sludge disposal practices and
identifies a need for R&D to improve sludge disposal alterna-
tives.
The FWPCA also requires that an assessment of alternative
treatment technologies be made before a construction grant is
awarded (Sec. 20 lg).2 Alternatives that must be considered are
land treatment of wastewater and sludge recycling (Sec. 201 (d)).
As a result, there will be many new technologies that may pose
health risks and must be addressed.
The Safe Drinking Water Act (PL 93-523), Section 1444 (a)'
and 1444(a)2 requires R&D associated with any project which
will demonstrate health implications involved in the reclaiming,
recycling and reuse of wastewaters for drinking water. This too
impacts on sludge disposal. Methods used in the past are now re-
stricted by specific laws or regulations. Current research pro-
grams have emphasized a shift of philosophy from ocean dump-
ing to the development of improved technologies for returning
the sludge to the environment (land) in an ecologically accepta-
ble manner with minimal health risks. Many State agencies are
reluctant to approve use of agricultural land for the treatment
and disposal of wastewater and sludge, because of the lack of
information on the health hazards associated with such practice.
It is essential that the health risks associated with sludge pro-
cessing, utilization and disposal be identified and adequately
studied.
Types of Sewage Treatment, Disposal and/ or
Utilization Methods
Sewage treatment can be divided into three phases: primary,
secondary, and advanced. In the primary phase, the waterborne
constituents are separated from the water stream by screening
and sedimentation or flotation. Conventional secondary treat-
ment removes up to 90% of the organic matter in sewage by
making use of a biological active mass. The two principal tech-
niques used in the secondary phase are trickling filters and the
activated sludge processes. Stabilization ponds can be used to
treat sewage to the secondary phase of treatment or they can be
used to supplement other processes. Physical-chemical methods
may be used to remove colloidal matter, color, odor, acids, alka-
lis, heavy metals and oil. To return water of more usable quality
to receiving lakes and streams, new methods for removing pollu-
tants are being developed. These advanced waste treatment
technologies can include coagulation and sedimentation which
will increase the removal of solids from effluent after primary
and secondary treatment. Filtration, microscreening, ammonia
stripping, carbon adsorption, electrodialysis, ion exchange and
reverse osmosis can be used to remove nutrients, refactory
organics and dissolved solids. Disinfection can be associated
with primary, secondary and advanced waste treatment.
The sludge produced during the treatment can be handled by
any of a number of processes, including anaerobic and aerobic
digestion, lagooning, pasteurization, centrifugation, drying by
sand beds, heat (thermal) drying, composting, incineration; lim-
ing and chlorination. Thermoradiation and radiation are also
being developed as disinfection processes.
Many alternatives are available for the disposition of the liq-
uid phase. Most treatment works effluents are directly dis-
charged to surface waters. Other means for disposal or use of
effluents include: (1) discharge to the land for further treatment;
(2) recharge of the groundwater; (3) irrigation of agricultural
crop land; (4) wastewater effluent discharge into waters used for
recreational purposes; and (5) subsequent industrial reuse.
Direct reuse of effluents for potable water supplies is not practi-
cal currently due to the many potential public health factors.
Of the many methods of processing and disposing of sludges
from wastewater treatment plants, no single method is capable
of solving all problems. Each disposal method has advantages;
each works well under a given set of local conditions while
another approach may be better suited to a different set of cir-
cumstances.
The major decision in selecting sludge disposal alternatives
relates to the desired or allowable final form of the material
when introduced into the receiving environment. All sludge pro-
cesses are directed at economically converting the material with-
drawn from the liquid treatment phase into a form satisfactory
for release to the environment. The alternatives currently availa-
ble are: (1) incineration; (2) composting; (3) landspreading for
land reclamation or agricultural enhancement; (4) landfilling;
and (5) by-product recovery.
203
-------
204
Health Effects
Table I: Sludge Information Summary
1. Quantities of sludge (estinatcd)
dry tons/day
Domestic
Industrial users of
municipal plants
Total municipal
sludge
Current
10,000
7,000
17,000
2. Current disposition of sludge
Method £ Total Sludge
Landfill
Ocean dump
Incineration
Land application
Croplands
Others
25',
15'
35"
25
(20' )
( 5")
Secondary Tre.ltnont
(next 10 years)
13,000
10,000
23,000
Reliability of Estimate
Good
Good
r,ood
r,ood
Pooi-
Poor
•Derived from background information to the Technical Bulletin on
Municipal Sludge Management (1).
3. 1972 Land Spreading Survey (Liquid Sludge Only)
EPA Regions II, III, IV, V, and IX
Mailed 1909, Responded 715 (39',)
Region
II (NJ.HY)
III (DE.MD.PA
Vfl.WV)
IV (AL,FL,GA,
KV.MS.NC,
SC.TH)
V (IL,1N,MI,
MN.OM.WI)
IX (CA.HI.NV)
Total
Si ze
MGD*
Currentl v
Used
wn: use
18'-
Do Hot Use
88".
70'.:
25
Currently
Used
1-1!) 27 M
10-100 15 1
Greater than 100 7 13
l Costs for Various Sludge Methods
Includes operating and construction costs
S/Drv Ton
I MGD 10 MGD
67".
Ppjlptjjse
65
76
30
Land Application
Landfi11
Incineration
Ocean Dumping
127-168
171-203
250-320
376-417
53-71
77-116
111-174
93-134
100 MGD
57-34
63-98
79-120
56-93
MGD - million gal lons/d,i"
Present Knowledge Concerning Presence
and Survival of Pathogenic Agents and Toxic
Chemicals in Various Phases of Sewage Treatment,
Disposal, and/or Utilization Methods
A. Biological: Information on pathogen removal typical of
various municipal wastewater treatment processes is presented
in Table 2.-V.V
Primary treatment by settling, which still constitutes the
whole of municipal wastewater treatment for many communi-
ties, removes little of the pathogens in wastewater. Secondary
treatment by trickling filters, activated sludge, or stabilization
ponds can markedly reduce the number of pathogenic enteric
bacteria, viruses, tubercle bacilli, and parasites in sludge
depending on the process used. Resulting effluents will contain a
quantity of each kind of microorganism present in raw sewage
depending on the efficiency of the particular process.
Table II: Percent Removal of Pathogens by Sewage
Treatment Processes
Total Counts
Coli form
Fecal Strep
Typhoid Group
Shi gel la
Cholera
Microbacterium
Salmonel la
M. Tuberculosis
Polio
Coxsackie
ECHO
Infectious
Hepatitis
Tapeworn Ova
E. Histolytica
Cysts
Ascari s
Lumbncol ides Ova
Taenia Saginata
Trichuris
Schistosoma
Japonicum
S. Mansoni
Hookworm
*Derived from data
Trick! ing
Filter
70-95
82-97
84-94
84-99+
-
66-99-
84-99+
Survive:
66-99
Slinlit-90-i
Slight-90+
Slight-9fJ<-
18-26
11-99+
_
62-70
_
Reduced:
.
100
reported by
Activated
Sludge
Enteric Bacteria
70-99
91-98
Present:
95-99.2
97-98
Not round
Slight to 37
96-99
M. Tuberculosis
Survi vr :
88
Enter_o_ Vmis_e$_
60-90-"
60-90+
60-90+
Pflrasi U",
Not Renoved
No Reduction
_
Little Effect
91.8-100
Excel lent
Hatching
Medium
_
81.5-96
Kabler (1959) and
Anaerobic
Digestion Cli lorin.Hion
96-19
99-ggf
-
Not Eou-id: 90-99
25-92.4
Survi ve : Survi ve
69-90 99+
Sliqht-99i Slight
Sl1ciht-90H to
Slight-99* Consiclpr.iblc
97 No rrfircr
Removed
45; Reduce
Very Slow
_
90(25-35 days)
others; (?,3,4 A 5)
The efficiency of disinfection (chlorination being the most
widely used method) of the effluents discharged varies consid-
erably under different physical and chemical conditions. The
data available have shown that sewage effluents after secondary
treatment and after chlorination contain reduced numbers of
viruses and intestinal parasites.
There are insufficient data regarding the survival and destruc-
tion of pathogens during sewage sludge incineration. It should
be expected that, under conditions of good incineration practi-
ces and equipment, pathogens would not survive and be found
in the sludge residue or emitted from the incinerator stacks.
However, under conditions of poor combustion, pathogens may
survive.
Studies have shown that the pathogen content of sludge is
similar to that of municipal solid waste, thus the pathogen survi-
val in land-filled sludge can be considered similar to that in
landfilled municipal solid waste providing that environmental
factors—temperature, moisture, and pH are comparable.6
B. Presence of Persistent Toxic Chemicals: The concentra-
tion of toxic chemicals in municipal wastewater effluents and
sludge are related to the kinds and amounts of industrial dis-
charges as well as street runoff directed into the municipal treat-
ment systems.
Conventional methods of wastewater treatment do not com-
pletely remove many potentially hazardous chemicals. The con-
-------
Health Effects
205
centration of contaminants in receiving waters has been shown
to increase in downstream communities as water is used and
reused and as the percentage of unremoved residual refractory
chemicals grows.
When sludge is properly incinerated, small but measurable
quantities of specific metals can be found in the stack emissions,
scrubber water and residue of incinerators. The metals identified
are arsenic, beryllium, cadmium, chromium, copper, lead mer-
cury, nickel and vanadium. Aside from the potential hazards
associated with atmospheric exposure to .these metals (for
mercury—performance standards exist), other hazards may be
posed by these metals upon ultimate disposal of the scrubber
water and residue. Many of these metals are known to accumu-
late in the human system.
Small but measurable quantities of specific organic com-
pounds dieldrin, chlordane, DDT and polychlorinated biphen-
yls are also found in some sludge and would be destroyed under
optimum combustion conditions, but under conditions of poor
combustion could be emitted from the stacks of municipal
sludge incinerators. These compounds are known to accumulate
in the human system and many are known to be mutagenic and
carcinogenic.7
Health Implications of Sewage and Effluent
Disposal
The methods presently employed or under consideration for
the reclamation and/or disposal of sewage effluent and sludge
may contribute to adverse health effects unless measures are car-
ried out to curtail exposure to pathogens and toxic chemicals.
A. Biological: Application of sewage sludge to soil has been
shown to increase the number of soil microbes.8
Application of sewage (primary or secondary) effluents or
sewage sludge to the land may present a potential health hazard
because of the presence of human or animal pathogens. Among
the most common pathogens found in sewage effluents are the
bacterial pathogens Salmonella, Shigella, and Mycobacterium;
the entero-viruses, adenoviruses, and possibly hepatitis virus;
the amoeba cysts, and helminth ova. Some studies have sug-
gested that many micro-organisms are retained at or near the
soil surface. Their movement through the soil is not considered
to be a problem.9,10 Only recently has quantitative information
on virus survival appeared in the literature.
Irrigation with sewage has been found to be associated with
vector-borne disease such as malaria and encephalitis; severe
annoyance by blood-sucking insects; as well as typhoid fever,
cholera, and other gastrointestinal diseases." Vegetables irri-
gated with either raw sewage or primary-treated and chlorinated
effluents have been found to contain Salmonella and Ascaris
ova.12 Grass samples sprinkled with raw sewage yielded Salmo-
nella.13 Enteric viruses such as polio-, coxsackie-, echo-, and
infectious hepatitis may be spread through sewage contami-
nated vegetables.l4
Where animals have access to raw sewage and treatment plant
effluents, the possibility of their contracting certain diseases
may exist.'5 In cattle, these diseases include bovine tuberculosis,
anthrax, tapeworm, etc.
Health implications may be associated with: (1 )the municipal
wastewater effluent as it passes over or into the soil —on crops,
animals, or man; (2) groundwater contamination; (3) aerosoli-
zation of wastewater and particulate matter containing patho-
gens; and (4) ponded areas, open lagoon storage sites, or con-
taminated air/soil interface. Where not adequately designed,
mosquitoes may breed, flies and other insects, migrating birds
and rodents may have access and serve as mechanical carriers of
disease organisms.16
Pathogen Concentrations in Wastewater and Sludge
The number of pathogens found in municipal wastewater and
sludge varies widely depending on the time of year and the com-
munity. Estimates of virus concentrations from field studies
have indicated concentrations up to 7000 plaque forming units
(PFU)/liter for raw sewage to about 50-1600 PFU/liter for
chlorinated secondary effluents.10
It has been demonstrated repeatedly that many pathogens,
especially viruses, survive conventional waste treatment pro-
cesses including chlorination, although they are reduced in
number. Berg17 points out that primary and secondary treat-
ment processes reduce the numbers of viruses only slightly.
Although some laboratory studies have shown good reduction,
field evaluations are, at best, erratic.
Current research being conducted by the U.S. Department of
Agriculture, Agricultural Research Station at Beltsville, Mary-
land, has shown that potentially infective stages in the life cycles
of parasites persist in sewage.18
In actual practice, disinfection probably plays the most
important role in reducing pathogen concentrations in waste-
water. However, much of the classical data on disinfection does
not reflect real world conditions. Disinfection as practiced for
wastewaters is often less effective for pathogens, especially vir-
uses, than it is for indicator bacteria. Table 3 contains data
which supports this contention.3, ",20
Table III: Effect of Wastewater Treatment on Various
Organisms3
Total Reduction
in Treatment
Plant (Logic)
Reduction by
Chlorination
Only (Log10)
PLANT HI *
-ecal Coliform
-ecal Streptococci
Salmonella t
Interic Viruses n
PLANT =2 S
Total Coliform
recal Coliform
Klebsiella a
Bacteriophage
3.7
2.4
1.4
0.4
6.8
5.8
5.2
2.1
2.0
0.5
0.3
5.5
3.6
3.7
0.9*
Standard-rate trickling filter; chlorine residual approximately
1.5 mg/1
t Presumptive. M-bismuth sulfite broth with MF procedure
fl Three days on BGM cells after concentration (15)
6 High-rate trickling filter; chlorine residual approximately
4 mg/1 with improved mixing
|a Large number confirned as Klfbsiella on differential media
After Kruse, et_ aj_. (20)
Pathogens in Soils
The application of municipal wastewater and sludge to soil
has been studied to some extent with regard to pathogen mobil-
ity and destruction in soils.21 Studies by Sagik, et al., provide
current information.22 Basically, pathogen removal in soils is a
function of the characteristics of the soil. Heavily textured clay
soils, through adsorption and filtration, can remove viruses,
bacteria and the larger pathogens, on or near the soil surface.
Desorption and transport of viruses may present a more long-
range problem.23
Those pathogens which are collected on or in soil can be inac-
tivated after land application. Exposure to ultraviolet light, oxi-
dation, and desiccation are the most important destructive
mechanisms. Temperature and soil moisture also appear to be
important factors.24 Conversely, there is ample literature
-------
206
Health Effects
indicating that some pathogens can survive these natural de-
structive effects in soil for relatively long periods.
The extent that pathogenic organisms can survive on soil and
vegetation is important.25 It has been shown that coliform orga-
nisms survive longer in soil than on vegetation .l8 Virus survival
on soil is currently receiving attention by Sagik el a/.26 and
Duboise el a/.23,-4 It is unlikely, however, that coliform orga-
nisms can survive longer than viruses.
It is also important to consider pathogen concentration on or
near the soil surface. Consideration should be given to the
effects of long-term application of non-degradable organic
materials that may clog the soil surface or tie-up soil adsorption
sites; the effect of cation species and concentration with respect
to soil tightness and adsorptive capacity; the effects of high pH
on soil filterability and reduced adsorption capability; and
lastly, the reduced treatment capabilities of soil for pathogens
after shock loading of a toxic effluent (from spills).
These possibilities suggest some potential public health impli-
cations. Human contact with organisms on soil surfaces may
result if winds, machinery, or other human activity such as walk-
ing reaerosolize the organisms and make them accessible for
inhalation. Runoff, either during intense effluent application or
precipitation, may cause significant numbers of concentrated
pathogens to enter surface waters. This problem can be mini-
mized by proper design of the application site.
Pathogenic Aerosols
In evaluating the potential problems associated with patho-
genic aerosols consideration must be given to wastewater
pathogen concentration and treatment effectiveness, the
amount of wastewater and the degree of aerosolization, the dis-
tance from receptor populations and the prevailing meteorolog-
ical conditions.
It is reasonable to postulate that, if disinfection of sewage is
not relatively efficient and if pathogenic organisms are aerosol-
ized, even very low numbers of these organisms may be a poten-
tial public health hazard. In order to evaluate this situation,
models have been developed describing the potential human risk
for a variety of wastewater pathogen conditions, meteorological
conditions, and treatment schemes.3 Field validation of these
predictions are becoming available.27,28,29
Although the potential problem of pathogenic aerosols have
been identified, research is necessary to allow better definition
and assessment of the potential problems.
It is interesting to relate the potential aerosol problem to
spray area design criteria. Predictions through modeling indi-
cate a 3-log reduction in virus concentration can be achieved
with a buffer zone of less than 200 meters.30 For comparative
purposes, it can be shown that filtration followed by disinfection
with adequate mixing will provide approximately a 3-log reduc-
tion in the virus concentration.
B. Chemical Considerations: The presence of toxic substan-
ces both inorganic and synthetic organic chemicals in waste-
water and sewage sludge is related to the kinds and amounts of
urban and industrial discharges to the sewage treatment systems
as well as the persistence of the compounds with respect to both
chemical and biological degradation. The composition and
character of urban activities and industrial processes are ever
changing and, so too, municipal sewage wastes vary in their
chemical composition.
I. Inorganics: Applying sludge as a fertilizer or soil condi-
tioner entails the risk of increasing the metals content of crops.
This may be detrimental or beneficial to the quality of the crop
for food. Direct ingestion of contaminated crops, or drinking
water derived from groundwater polluted by leaching or surface
water polluted runoff are possible exposure modes.
Most of the elements are found in wastewater, and if no con-
trols are exerted on what is discharged into sewers, they could be
found at undesirable levels in sludge. The inorganic toxic sub-
stances that may be found in sludge that are of greatest concern
to health are the heavy metals—arsenic, cadmium, lead, mer-
cury, selenium and zinc.
Arsenic, Lead, Mercury and Selenium
Contamination of soils with arsenic, lead, mercury, selenium,
much beyond the normal range in soils is unlikely if the applica-
tion of nitrogen, cadmium, zinc, copper, and nickel is regulated
to control sludge application rates.31
Cadmium and Lead
Cadmium is the element of greatest concern from sludge use
on agricultural land because it is toxic to man and accumulates
in the kidneys and liver, with deleterious effects after a certain
threshold has been reached. Animal studies have indicated that
an accumulation of 200 ppm in the renal cortex is the level at
which susceptible individuals would experience kidney dam-
age. The World Health Organization established a provisional
tolerance of 400-500 /igCd/week; this is equivalent to 57-71
MgCd/day. Estimates of the daily dietary intake of Americans
vary widely, but the most recent FDA surveys indicate that the
level is almost up to the World Health Organization (WHO) tol-
erable dietary intake. The cadmium level in the diet does not
appear to pose a short-term health problem, because the daily
intake at present estimated levels will not result in accumulation
of 200 ppm in the renal cortex for the average life span. Neither,
does it seem likely that judicious use of sewage sludge will signif-
icantly increase the average daily dietary intake of the U.S. pop-
ulation as a whole. However, it could increase for certain local
populations who primarily consume local crops grown on
sludge treated land. Also, the margin of safety is small (safety
factor 4) and parts of the population may not be average in
regard to dietary intake and sensitivities. In addition, heavy
cigarette smoking can add greatly to the daily intake of
cadmium.31
Another concern is for cadmium and lead ingested directly by
animals grazing on pastures contaminated with sludge. Experi-
ments are under way to evaluate the effects on cattle grazing on
sludge-amended pastureland and consuming sludge directly.
Preliminary results indicate that there is some increase in lead
and cadmium levels in the liver and kidneys but no indication
that there is any imminent danger of significantly increasing
man's intake of the metals. It would seem like a reasonable pre-
caution to avoid sludge spreading on pastures pending receipt of
data from research on the effects of sludge application. Restric-
tions against animals grazing sludge-treated pastures until the
sludge has been thoroughly washed off the grass or the levels of
the lead and cadmium in the sludge exceed 1,000 mg/ kg dry
sludge and 20 mg/ kg dry sludge, respectively, are described in
the U.S. EPA proposed Municipal Sludge Management Techni-
cal Bulletin.1 These restrictions appear to be providing some
protection against excessive increase of cadmium and lead in the
daily diet.
The application rate of sludge-borne cadmium to agricultur-
al crops should be controlled until the permissible levels in food
crops have been established and until research can define the
plant uptake by various species and varieties of plants under var-
ying environmental and management conditions. The rates of
application that are being advocated by the U.S. Department of
Agriculture, Agricultural Research Service, or by the Wisconsin
State guidelines appear to minimize accumulations of cadmium
in foods. They are probably restrictive enough that many
industrial cities will find that land spreading is not an econom-
ically feasible disposal method for them, but some large cities
may be able to comply. Dedicated land could have less stringent
restrictions than those for agricultural lands as long as products
to be marketed meet any present or future standards of FDA,
USDA, or State departments of agriculture.31
2. Organics: Numerous efforts have been made to gain a bet-
ter understanding of toxic hazards of modern synthetic organic
-------
Health Effects
207
chemicals that find their way into wastewater and sludge
However, there are many difficult sampling and analysis
problems in concentrating, extracting, and identifying such
compounds many of which still may be unknown breakdown
products of more complex chemicals that have undergone par-
tial biodegradation. In 1960, researchers at the Taft Sanitary
Engineering Center were able to identify fewer than 40 percent
ol the soluble orgamcs remaining in biologically treated sewage
and these were described only in general terms such as ether
treated extractable matter, protein, tannin, lignin or alkyl ben-
zene sulfonate.^z No such systematic effort has been reported in
the literature since then, except for certain specific compounds
like PCBs and organochloride pesticides. However, EPA is
presently conducting a study to identify toxic materials in the
effluent.
The conventional methods of wastewater treatment leave
many potentially hazardous chemicals unchanged. Many of the
organics found in drinking water may also be found in sludge.
EPA has compiled a list of about 300 specific compounds identi-
fied in drinking water." The concentration of some contami-
nants has been shown to increase in downstream communities as
water is used and reused and as the percentage of unremoved
residual refractory chemicals grows.
Petroleum products and refinery wastes are among the
sources of many possible mutagenic and carcinogenic substan-
ces that may appear in municipal sewage sludge. Various poly-
nuclear aromatic hydrocarbons have been recovered from sew-
age sludge among them the known carcinogens 3,4,—benzpy-
rene and 1 ,-2—benzathracene. Carcinogenic aromatic amino
compounds such as beta-naphthylamine and benzidine origi-
nate in dye and rubber works and may be released to sewer sys-
tems together with nitroanalogs used in the production of amino
compounds such as aminoazodyes, amino stilbenes and tri-and
diphenyl methane dyes. Pharmaceutical factories, textile dying
plants, plastic production and similar industries are sources of
these organics. Some of the chemical intermediates such as
orthochloronitrobenzene have been found in the Mississippi
River in appreciable quantities and appear to be quite stable.
Chloronitrobenzene for example, discharged to the Mississippi
atSt. Louis, Mo. was still detectable in the drinking water drawn
at New Orleans, Louisiana hundreds of miles and many days
flow downstream.34 Very little research has been done to deter-
mine the fate of toxic chemical substances in the pasteurization
of sludge. Pasteurization has been proposed as the method for
rendering sewage sludge microbiologically acceptable for
increased land use. However, until recently, the possible adverse
health effects from toxic chemicals which may persist in pas-
teurized sludge has been ignored.
Few data are available on the concentrations of synthetic
organic chemicals in municipal sludges except for PCBs and
chlorinated pesticides. Table 4 summarizes data obtained in the
few studies available.35 Using routine pesticide analytical proce-
dures, all the chlorinated organics examined for were found,
with levels ranging up to 352 ppm for PCBs, 1.1 ppm for the
DDT family, 32.2 ppm for chlordane, 1.4 ppm for dieldrin and
16.2 ppm for aldrin. More up-to-date and extensive data on
PCBs, pesticides and other persistent organics in sludge are
needed.
Disposal of sludge by incineration where persistent organics
and PCBs are found would result in effective destruction under
optimum combustion conditions, but under conditions of poor
combustion could result in emissions from the stacks of sewage
sludge incinerators.
The utilization/disposal of sludge on croplands and pasture-
lands emphasizes the need for definitive data on plant up'take of
synthetic organic chemicals. Uptake studies have shown that
edible parts of plants contain these organics, but at levels 5 to 20
percent of the levels in the soils used. In general, root crops take
up more chlorinated orga'nics from the soil than other types of
crops. However, studies have shown that chlordane, heptachlor
and dieldrin are translocated from the soil into soybeans and
stored in the oil of the seed.36 Although the levels found were
low, these data show that sludge can be a source of recycling of
organic contaminants back into the food chain.
Table IV: Pesticide and PCB Content, Dry Sludges35(ppm)
Contaminant
Aldrin ' '
Dieldrin'1 )
Chlordane'1)
DDT + DDD'1'
PCBs(2)
' ' ' Examined in 1971
/ P \
Jelinek et al . , 1976
Min_.
ND
0.08
3.0
0.1
ND
1973,
(35)
Range
Max.
16.2
1.4
32.2
1.1
352.0
1975
Sludges
E_xamined_
5
7
7
7
69
Uptake of PCBs by plants was recently reported in the docu-
ment Background to the Regulation of Poly chlorinated Biphen-
yls (PCB) in Canada.37 The levels of PCB found in corn grain as
a result of sludge application ranged from less than 5 ppb to 95
ppb (Aroclor 1254).
Direct contamination of growing plants above ground is also
a potential problem with sludge application to croplands and
pasturelands. Recent USDA work for FDA showed that dried
grass contained about 5% by weight of sludge, when the grass
had been mowed 80 days after it had been sprayed with the
sludge. This means about 30% of the applied sludge remained on
the grass. It is noteworthy that the grass in this study was grown
in the East, not an arid section of the country. This work indi-
cates a significant potential for contamination by persistent syn-
thetic organics such as organochlorine pesticides and PCBs as a
result of application of sludges containing these organics.35
The recent discovery (April 1976) of PCBs in a family cow's
milk in Bloomington, Indiana is illustrative of the potential con-
sequences of sewage sludge application to pasture when indus-
trial discharges contaminate the sludge. The family cow grazed
on pasture to which 12 tons per acre of city sewage sludge had
been applied in November 1975. Subsequent analysis of the
sludge samples from that plant showed 105 ppm and 240 ppm
PBBs (Aroclor 1016 dry weight basis). The cow's milk contained
5 ppm PCBs on a fat basis38 This is well above FDA's tolerance
of 2.5 ppm PCBs in milk on a fat basis. Transfer of PCBs to the
cow was probably related to grazing habits resulting in con-
sumption of the contaminant without uptake of pasturage.
Problems may arise when sludges high in PCBs are applied to
pastures that are grazed soon after the application or to forage
crops before harvest. Applying sludge to hay crops after cutting
may not present a problem because much of the new growth
takes place at the top not from the bottom. Technology for pre-
treatment control of PCBs in sludge is available and can be
accomplished by source control. Research now in progress
through an interagency agreement with the U.S. Department of
Agriculture, Agricultural Research Service, should provide
more information on the fate of PCBs in soil. Feeding experi-
ments in progress will also provide more information on the
effect of ingested PCB on the potential contamination of beef
tissues.
-------
208
Health Effects
Health Implications of Municipal Wastewater and
Sludge Disposal
The evidence that intestinal pathogens and parasites are
transmitted by municipal wastewater irrigation indicates the
possible danger for pathogens as it passes through or over the
soils.39 Cases of cysticercosis have been reported near Tucson,
Arizona where cattle were grazed on pastures that had been irri-
gated with municipal wastewater effluent.40
Municipal wastewater contaminated drinking water has been
proven to cause outbreaks of infectious hepatitis.41
Aerosols from around municipal wastewater treatment
plants42 and irrigation fields43 containing intestinal bacteria
have been found to be dispersed over distances of up to 1300
meters.44
Many diseases have been related to municipal waste treat-
ment water and many insects of public health importance have
been found in aquatic habitats associated with municipal waste-
water irrigation.
Pasteurization or equivalent treatment of wastewater sludge
and effluents appear to be the only effective method for patho-
gen and parasite kill. Although the process is used in European
countries, R&D is needed to define temperatures for effective
kill.45 Energy consumptions appear to be minimal for the heat
requirements for pasteurization. However, until such treatment
becomes available in the United States a comprehensive
research program is needed to assess the impact of municipal
wastewater sludge and wastewater effluent utilization upon
public health.
Consequences of Sewage Sludge and Wastewater
Recycling to the Land
Current practices result in about 60 percent of the municipal
wastewater and sludge produced being placed on or in the land,
FWPCA (PL 92-500) and Agency policy require that land appli-
cation be considered as a viable alternative for municipal
wastewater treatment. This emphasis on use of the land places
a burden on EPA, States, and local agencies and municipalities
to ensure that the methods used do not result in creation of
health hazards and are otherwise ecologically sound. Many of
the questions raised concerning potential health problems are
currently unanswerable with existing information. They can
only be answered by an R&D effort which carefully coordinates
the control technology, health and ecological efforts and socio-
economic aspects. Stated in broad terms, the following
questions need to be answered.
1. What is the fate, effect and persistence of metals, organics
and micro-biologicals (pathogens, parasites, and viruses)
applied to the land (aerosols, leachates) under various cli-
matic, meteorological, and soil compositions (clay, sand,
etc.)?
2. What are the potential health and ecological risks involved
when municipal wastewater and/or sludge are applied to
agriculture lands? Constituents of concern are trace
metals, organics, micro-biologicals (pathogens, viruses,
parasites) and nutrients, such as nitrogen.
3. How much municipal wastewater and sludge can be ap-
plied to agricultural land before health and ecological
hazards result?
4. What is the extent and significance of ecological impact on
the environment from applying municipal wastewaters
and/or sludge to the land (insects, wildlife and migratory
birds, grazing animals, groundwater and aquatic life, both
marine and fresh)?
5. What are the environmental benefits to be gained and risks
we are willing to accept by applying municipal wastewater
and sludges to the land?
6. What are the long-term environmental insults to the land,
man, and the ecosystem?
7. Can pretreatment of industrial discharges to municipal
systems be effective in managing metal content of sludge
and effluents? If so, how can the resulting industrial sludge
be managed in an environmentally safe manner?
8. What is the fate, persistence and effect of aerosol transport
of pathogens (bacteria, viruses, parasites) and chemicals
from a municipal treatment plant under various meteoro-
logical and climatic conditions? (Includes both conven-
tional facilities and land application.)
9. What is the scientific basis for establishment of safe buffer
zones for protection to populations living near municipal
wastewater treatment plants?
The Office of Solid Waste Management Programs (OSWMP)
has been designated as the lead office within EPA for coordi-
nating the development of policy, planning, and guidance in the
area of the utilization and disposal of residual sludge. Sheldon
Meyers, Deputy Assistant Administrator for Solid Waste Man-
agement Programs, is the Chairman of the newly-formed Resid-
ual Sludge Working Group. This working group is composed of
five subcommittees (see Table 5). Each subcommittee was
charged with basically the same mission.
Table V: Residual Sludge Working Group Organization
U.S. EPA
Deouty Administrator
Mr. John Quarles
1
Residual Sludge Working Group
Chairman
Mr. Sheldon Meyers
DAA/OSWMP
Residual Sludge
Hanage^nt Subcommittee
Cha i man
Kr. Cruce Ueddle
Municipal Sludge
Subcommittee
Chairran
Mr. Don Ehreth
!
i
Health Effects
S'jbcowii ttee
Chai rman
'•V. ~honas Gleason
k
j P.egional Renuirements
1 Subcommittee
j Chairman
j Mr. Al Montague
Industrial Residuals
Subcommittee
Chairman
Mr. H. George Keeler
-------
Health Effects
209
' • Define the problem(s) facing the municipalities (urban and
rural), the regulatory Agency, the farming community, etc.
2- Establish what is known by EPA, USDA, FDA, and oth-
ers toward resolving the problem(s).
•>• Identify the gaps requiring research.
4. Develop a program plan leading to a successful solution
for sludge disposal (guidelines, policy, models, etc.).
The Health Effects Subcommittee addressed the problems of
heavy metals uptake by crops, parasites, viruses and the trans-
mission of other pathogens from sludge to man. It identified the
current level of R&D effort by the Agency in this area; the prob-
lems to be resolved and the policy issues that need to be
addressed by the Agency. The final output of the subcommittee
was a preliminary assessment of the public health problems for
the Administrator; identification of what Federal involvement is
needed and recommendations for guidance on municipal sludge
utilization. These recommendations were reviewed by the
Residual Sludge Working Group in October of this year and for-
warded to the Administrator for approval.
CONCLUSION
It is obvious that a potential for producing adverse human
health effects is associated with the practice of applying waste-
water or sludge to agricultural lands, and that the primary areas
of concern relate to potentially toxic chemicals and to patho-
genic organisms. The extent of the risk relates to the many varia-
bles relating to application such as soil/ sludge factors, plant fac-
tors and climatic effects, and to the treatment the wastes have
received.
The uncertainties associated with the practice of applying
wastewater and sludge to the land, especially for agricultural
purposes, make it imperative that a continuing vigorous
research program be pursued.
REFERENCES
I. U.S. Environmental Protection Agency, Technical Bul-
letin on Municipal Sludge Management: Environmental Fac-
tors. Fed. Reg. 41 (108): 22532-22543 (June 3, 1976).
2. Kabler, P., ''Removal of pathogenic microorganisms by
sewage treatment processes," Sewage and Industrial Wastes 31:
1373(1959).
3. Sorber, C.A., S.A. Schauband H.T. Bausum, "Anassess-
ment of a potential virus hazard associated with spray irrigation
of domestic wastewaters" in Malina, J.F. and Sagik, B.P. (eds.),
Virus Survival in Water and Wastewater Systems. Center for
Research in Water Resources, The University of Texas at Aus-
tin, Austin, Texas (1974).
4. Foster, D.H. and R.S. Engelbrecht, "Microbial hazards
in disposing of wastewater on soil" in Sopper, W.E. and Kardos,
L.T. (eds.) Recycling Treated Municipal Wastewater and
Sludge through Forest and Cropland, The Pennsylvania State
University, University Park, PA (1973).
5. Clark, N.A. and S.L. Chang, "Removal of Enteroviruses
from Sewage by Bench-Scale Rotary-Tube Trickling Filters,"
Applied Microbiology, 30,2: 223-228 (August 1975).
6. Gaby, W.L., "Evaluation of health hazards associated
with solid waste sewage sludge mixtures," EPA 670/2-75-023
(1975).
7. U.S. Environmental Protection Agency, Report of the
Task Force on Sewage Sludge Incineration, Rept. No. EPA-R2-
72-040, (1973).
8. Miller, R.H., "The Microbiology of Sewage Sludge Dis-
posal in Soil", EPA, 670/2-74-074, (Nov. 1974).
9. Krone, R.B., "The Movement of disease producing orga-
nisms through soil", Proc. of the Symposium on Municipal
Sewage Effluent for Irrigation. Louisiana Polytechnic Institute,
(1968).
10. Sorber, C.A., S. A. Schaub, and K..J. Outer, "Problem
Definition Study Evaluation of Health and Hygiene Aspects of
Land Disposal of Wastewater at Military Installations" , USA-
MEERU Report No. 73-02, AD No. 755122, U.S. Army Medi-
cal Environmental Engineering Research Unit, Edgewood Arse-
nal, MD, (1972).
11. Rainey, M.B. and A.D. Hess, "Public health problems
related to irrigation", Chapter 56 of Monograph 11, Irrigation
of Agricultural Lands, Am. Soc. Agronomy (1963).
12. Dunlop, S.G. and Wen-Lan Lou Wang, "Studies on the
use of sewage effluent for irrigation of truck crops", J. Milk and
FoodTechn. 24:44(1961).
13. Muller, G., "Pollution of irrigated grass with bacteria of
the typhoid-paratyphoid group", Komm. Wirtschaft, 5:409
(1955).
14. Murphy, W.H. et al., "Absorption and translocation of
mammalian viruses by plants: I. Survival of mouse encephalo-
myelitis and poliomyelitis viruses in soil and plant root environ-
ment", Virology (5:612 (1968).
15. Snyder, C.W., "Effects of sewage on cattle and garbage
and hogs", Sewage and Industrial Wastes 2.3:1235 (1951).
16. Straub, C.P., "Public health implication of land disposal
wastewater and sludges", Presented at the Symposium - Land
Application of Municipal Waste Waters, University of Minn.,
St. Paul, Minn. (March 22-23, 1973).
17. Berg, G., "Reassessment of the Virus Problem in Sewage
in Surface and Renovated Waters", in Progress in Water Tech-
nologv, Vol. 3, Oxford-New York: Pergamon Press Limited
(1973).
18. Burge, W., USDA-ARS Beltsville, Personal Communi-
cation (1976).
19. Schaub, S.A. and C.A. Sorber, "Virus and Solids in
Water", Presented at the International Conference on Viruses in
Water, Mexico City (June 1974).
20. Kruse, C.W., K. Kawata, V.P. Olivieriand K.E. Longley,
"Improvement in Terminal Disinfection of Sewage Effluents",
Water and Sewage Works, 120: (June 1973).
21. Sorber, C.A. and K.J. Outer, "Health and Hygiene
Aspects of Spray Irrigation" Amer. J. Public Health 65:1 (Jan.
1975).
22. Sagik, B.P. and C.A. Sorber, University of Texas at San
Antonio "Human Enteric Virus Survival in Soil Following Irri-
gation with Sewage Plant Effluents", EPA Grant No. R-8038
44010 (July 1975-July 1978).
23. Duboise, S.M., B.E. Moore and B.P. Sagik, "Poliovirus
Survival and Movement in Sandy Forest Soil" App. and Env.
Microbiology 31 (4): 536-543 (April 1967).
24. Duboise, S.M., B.P. Sagik, B.E. Moore and C.A. Sorber,
"The Effects of Temperature and Specific Conductance on
Poliovirus Survival and Transport in Soil", Abstracts for:
National Conference on Environmental Engineering Research,
Development and Design, University of Washington, Seattle,
Washington (July 1967).
25. Larkin, E.P., J.T. Tierney and R. Sullivan, "Persistence
of Virus on Sewage - Irrigated Vegetables" Journal of the Envi-
ronmental Engineering Division, ASCE, 102, (1): 29-36 (Feb.
1976).
26. Sagik, B.P. et al., "Surface Application of Municipal
Effluents", Proceedings of the Symposium on Virus Aspects of
Applying Municipal Wastes to Land, University of Florida,
Gainesville (June 1976).
27. Sorber, C.A., H.T. Bausum, S.A. Schaub and M.J.
Small, "A Study of bacterial aerosols at a wastewater irrigation
site", Jour. Water Pol. Control Fed. 48 (10):2367 (Oct. 1976).
-------
210
Health Effects
28. Katzenelson, E., B. Teitch and H.I. Shuval, "Spray Irri-
gation with wastewater: The problem of aerosolization and dis-
persion of enteric microorganisms", presented at the 8th Confer-
ence of the International Association on Water Pollution
Research, Sydney, Australia (1976).
29. Bausum, H.T., S.A. Schauband C.A. Sorber, "Viraland
bacterial aerosols at a wastewater spray irrigation site " pre-
sented at the 76th Annual Meeting, American Society for
Microbiology, Atlantic City, N.J. (1976).
30. Sorber, C.A., "Viruses in Aerosolized Wastewater", Pro-
ceedings of the Symposium on Virus Aspects of Applying
Municipal Wastes to Land, University of Florida, Gainesville
(June 1976).
31. Dotson, K. (EPA) and G. Braude (FDA), Personal Com-
munication (1976).
32. Middleton, P.M., Conference on Physiological Aspects
of Water Quality USPHS; Washington, D. C. I960.
33. U.S. Environmental Protection Agency, Preliminary
Assessment of Suspected Carcinogens in Drinking Water,
Report to Congress, Washington, D.C. (Dec. 1975).
34. Shuval, H. and N. Gruener, "Health Considerations in
Renovating Waste", Environmental Science and Technology,
7:600-609, (July 1973).
35. Jelinek, C.F., G.L. Braude, and R.B. Read, Jr., "Man-
agement of Sludge Use on Land, FDA Considerations", pre-
sented at AMSA Conference on Sludge Management, Houston,
(April 1976).
36. Moore, S., H.B. Petty, and W.N. Bruce, "Insecticide
Residues in Soybeans in Illinois, 1965-1974"; 28th Illinois Cus-
tom Spray Operators Training School, Summary of Presenta-
tions, 226-230 (1976).
37. Environment Canada and Health & Welfare Canada,
"Background to the Regulations of Polychlorinated Biphenyls
(PCB) in Canada", A report of the Task Force on PCB, to the
Environment Contaminants Committee TR 76-1, (April 1976).
38. Jordan, D., "PCBs discovered in family cow's milk",
Bloomington Herald Telephone (Indiana), (April 21, 1976).
39. Rudolfs, W. el. ai, "Literature review on the occurrence
and survival of enteric, pathogenic, and relative organisms on
soil, water, sewage, and sludges, and on vegetation. I. Bacterial
and virus diseases." Sewage and Industrial Wastes, 22(10): 1261
(1950).
40. Mclntosh, A. and D. Miller, "Bovine cysticercosis with
special reference to the early developmental stages of Taenia
Saginata", Am. J. Vet. Res. 21: 169(1960).
41. M osley, J. W., "Transmission of viral diseases by drinking
water" In: Transmission of Viruses by the Water Route, Berg.,
G., ed., Interscience Publishers, N.Y. (1967).
42. Kenline, P. A., "The emission, identification, and fate of
bacteria airborne from activated sludge and extended aeration
sewage treatment plants", Ph.D. thesis, University of Cincin-
nati, Cincinnati, Ohio (1968).
43. Shtarkas, E.M. and D.G. Krasilschikov, "On the sanitary
zone around sewage farms irrigated by sprinkling", Abstract.
Hygiene and Sanitation Vol. 35, Mos. 7-9, July-Sept., 1970.
44. Hickey, L.S. and P.O. Reist, "Health Significance of Air-
borne Micro-organisms from Wastewater Treatment Process,
Part I Summary of Investigation", Jour. Water Pol. Control
Fed. 47 (12): 2749 (Dec. 1975).
45. Gleason, T.L., J. Trax, C. Myers, M. Peterson and J.
French, "Report of work group on problems related to land use
of sewage effluent" U.S. Environmental Protection Agency,
(Jan. 1974). Unpublished report.
-------
Health Effects of
Municipal Refuse Disposal
Stephen C. James
Office of Solid Waste Management Programs
U.S. Environmental Protection Agency
Washington, D.C.
A wide variety of waste from industries, residences, and
municipalities has been and will continue to be disposed of on
the land. Current practices range from simple dumping of refuse
on a readily available piece of property to controlled disposal of
waste on designated sites which are designed to minimize the
potential for contamination of local water resources.
Solid waste land disposal sites (SWLDS) are sources of local
water resources contamination because of the generation of
leachate caused by water percolating through the solid wastes.
Precipitation falling on a SWLDS either becomes runoff,
returns to the atmosphere via evaporation and transpiration, or
infiltrates into the SWLDS. This infiltrating water ultimately
will form leachate (water that has percolated through the wastes
and, through leaching, picked up soluble and suspended con-
taminants).
The process of leachate formation and subsequent contami-
nation is dependent upon the amount of water which passes
through the solid waste. Water which infiltrates the surface of
the cover material, assuming daily and final cover are applied,
will first be used for soil evaporation and plant transpiration.
Any excess water will percolate into the layers of solid waste.
Additional surface runoff from surrounding land, moisture con-
tained in the solid or liquid wastes placed in the SWLDS, mois-
ture from solid waste decomposition, water entering during
waste placement in the SWLDS, and ground water entering
through the bottom or sides of the SWLDS also contribute to
the generation of leachate.'
According to the latest available estimates, 120 million tons of
residential and post-consumer commercial solid wastes are dis-
posed of on the land in the United States annually.2 The largest
component of municipal solid waste is paper, but substantial
food waste, yard waste, glass, metals, plastics, rubber, and liquid
wastes are also included. Pesticides containers, paint cans, bat-
teries, various cleaning agents, dead animals, disposable dia-
pers, grease and oils, and wastes from the health professions are
among the typical potentially hazardous wastes received at
municipal sites. Not included in this 120 million tons are sewage
sludge, industrial chemicals and tailings, septic tank pumpmgs,
street sweepings, discarded automobiles, construction and
demolition wastes, and landclearing wastes. Many of these
wastes, such as industrial chemicals, are disposed of in the
SWLDS without the operator's knowledge.
As noted before, the amount of infiltration from precipitation
that falls on a landfill is the major factor affecting the quantity of
leachate that can be generated, unless, of course, poor site
location was at fault. Therefore, the extent of the potential prob-
lem of water resources contamination resulting from leachate is
greatest in areas where average annual precipitation exceeds the
loss by evapotranspiration. Such areas are generally found east
of the Mississippi River and in the coastal region of the Pacific
Northwest. About 70 percent of the municipal SWLDS found in
the United States are located in these water surplus areas. These
are also areas of high industrial and residential density. Thus,
the local water resources of many cities are located near areas
where potential contamination due to land disposal sites might
occur.
Leachate generation information has been compiled over a
one-year period at five different municipal land disposal sites
under an EPA study conducted by SCS Engineers (1975).3 The
principal criterion for their selection was that they received only
residential and commercial wastes. It should be noted, however,
that many land disposal sites do accept some industrial wastes
and/or sewage sludge, and many others fail to monitor or rec-
ord what they do accept. Site location, age, size, type, and
annual precipitation were recorded for each study site (Table I).
Site age and method of operation were the major differences
between the five sites. Raw leachate was easily obtained in large
quantities, which, if not collected and treated, could pose a
serious threat to ground and surface waters. Organic, heavy
metal, and conductivity data for undiluted, surface leachate are
presented for each study site (Tables 2-6). These tables indicate
the mean and the range, and compare the number of parameter
values, which were above the standard, to the total number of
analyses for each parameter.
Table I: Leachate Study-Site Characteristics
Site location
Washington
Pennsylvania
Indiana
Tennessee
California
*SLF - sanitary
to public health and
+Conversion - a
Age
(years)
2
2.5
3.3
5
20
landfill, following
safety.
site, which has bee
Size
(acres)
83
17
20
7
52
the definit
previously
Type*
SLF
SLF
SLF
conversion*
conversion*
ion described by
operated as a du
Annual precipitation
(inches)
57
41
43
44
40
the ASCE with regard
mp and now, mainly
From the data presented, certain factors concerning the com-
position of leachate are evident. First, younger sites (less than
five years old) exhibit strong organic content (chemical oxygen
demand) and high conductivity. As the site ages and as decom-
position declines. COD and conductivity decrease. COD is an
important parameter for its usefulness in determining the rela-
tive degree of solid waste decomposition, leachate treatment
211
-------
212 Health Effects Municipal
technique, detection of contaminant migration, and organic
contamination. Conductivity is important in determining the
relative degree of solid waste decomposition and detection of
contamination migration.
Table II: Washington
Table V: Tennessee
.J Fe 495 12/12
1.0 Cu .45 1/9
.05 Fb .15 7/7
-ui Cd .007 1/C
•Ob Cr (total) .073 1/5
-5.0 2n 13.6 8/9
.002 Bg .0034 3/5
•01 Se .027 4/5
COO 14,258
84-1 .126
.01-1.13
.07-. 31
.001-. 012
.025-. 22
3.2-33
.001-. 008
.005-. 1
4.963-30,933
Table III: Pennsylvania
Table IV: Indiana
primary
.05
.01
.05
.002
.01
3 Fe 679 10/10 18-1.550
1 -° Cu 2 0/8 .01-. 65
Pb .18 5/6 .017-. 31
Cd .02 4/6 001-.073
Cr (total) .19 6/6 .09-. 29
-5-0 Zn 9.1 4/8 .53-28.2
Hg .018 5/6 .001-. 061
Se .08 5/C .01-. jj
COO i I.274 7,650-17,654
conductivity 5.795 3,730-8,300
9 . P y 1
3
1 .0
.05
.01
.OS
-5.0
.002
.01
•9
Fe
Cu
Pb
Cd
Cr (total)
Zn
Mg
Se
COO
conductivity
618
.43
.26
.015
.11
1 62
.0635
127
12.292
7.148
' '
* of values
above standard
10/10
2/7
7/7
3/5
4/5
0/7
3/5
4/5
Range
208-1 .237
.01-1.54
.2-. 33
.006- .038
.01-. 175
.88-2.25
.001- 008
01-. 3
10,341-16.721
5.875-9.300
Jrlmary
.05
.01
.05
.032
.01
•All
secondary Parameter
.3 Fe
1.0 Cu
Pb
Cd
Cr (total)
-5.0 Zn
H9
Se
COD
conductivity
values In mo/1, except conductivity (m
Average
15.2
.25
.16
.008
.10
7.2
.036
.116
2,900
8,035
Icromhos/cm).
above standard
10/10
0/8
7/8
2/5
6/6
3/7
5/6
5/5
Range
6.5-34.5
.016-. 48
.02-. 33
.004- .01
.06-. 133
.54-27.2
.0006-. 16
.02-. 21
845-8,669
6,550-10,080
Table VI: California
:PA drinking water standards
irlmary
.05
.01
.05
.002
.01
•All
.3 Fe
1.0 Cu
Pb
Cd
Cr (total
-5.0 Zn
Ho
Se
COO
conductivity
values In ng/1 , except conductivity
25.0
.19
1
.006
.034
3.7
.012
.06
168
2,666
mlcromhos/cm).
1 of values
11/11
0/B
5/7
2/6
1/6
1/8
4/5
4/5
Range
9.9-120
.004-. 38
.01-. 16
.0013-. Ul
.01-. 09
.07-20.8
.001-. 032
.001-. 112
79-420
2,050-6,300
The leachates contain measurable quantities of specific
metals. The Environmental Protection Agency Interim Primary
Drinking Water Standards4 and proposed secondary standards
are presented in the tables for comparison against reported
values of contaminants in SWLDS leachates.
From an analysis of the data from this study, several conclu-
sions can be drawn. First, copper is not present in concentra-
tions in this leachate above EPA maximum contaminant levels.
Second, the zinc concentration found in this leachate decreases
as the age of the SWLDS increases. Third, the amount of cad-
mium and chromium appear to vary greatly from site to site.
Fourth, iron is present at each site above secondary standards.
And fifth, lead, selenium and mercury are present at each site in
concentrations above EPA maximum contaminant levels,
regardless of site age.
Since maximum contaminant levels are exceeded by certain
heavy metals present in leachate, consideration must be given to
the overall damage and the individual effect that each contami-
nant presents. As leachate is generated by solid waste land dis-
posal sites, contamination of ground and surface waters may
occur. This leads directly to the problem of contamination of
drinking water supplies either by contamination of surface
supplies or by contamination of an aquifer or well field.
When water supplies are contaminated with leachate contain-
ing heavy metals, the mechanism leading to health hazards is
bioaccumulation. Many living organisms, including man, are
known to accumulate specific toxicants (chemical pesticides,
industrial organics, heavy metals) from the environment. This
capability is widespread and the amount accumulated may
-------
Health Effects/ Municipal
213
range from barely detectable concentrations to concentrations
that greatly exceed the amount present in the environment,
depending on the contaminant and organism involved.
Bioconcentration is considered to be accumulation at a rate
which results in a concentration of contaminant greatly
exceeding that to which the organism was exposed. The degree
of biocpncentration can be described by a concentration factor:
the ratio of the concentration of chemical recovered from the
organism to the concentration to which the organism has been
exposed. This concentration factor may be influenced by several
factors such as (1) nature of the contaminant, (2) amount of con-
taminant in the water,5 (3) duration of exposure to the contami-
nant,6 (4) rates of storage and excretion of the species, (5) size
and age of the organism, (6) reproductive condition of the
animal,7 (7) the organ analyzed, and (8) the manner in which the
contaminant is accumulated—directly from the water or from
the food chain.
It is important to note here that the health effects from
SWLDS leachates are not limited to drinking water but may
also occur through the food chain due to the ingestion of other
organisms (fish, aquatic plants) that habitate an environment
contaminated by SWLDS leachates.
Classic examples of the effect of bioconcentrated toxicants
are the painful and fatal Itai-Itai disease, caused by chronic cad-
mium poisoning, and Minamata disease, causes by chronic mer-
cury poisoning. In both diseases, the contaminants are concen-
trated by fishes from industrial wastes discharged into coastal
waters. Continued consumption of the contaminated fishes by
man permitted accumulation of concentrations sufficient to
produce these diseases.
In relation to general water quality, the initial entry of con-
taminants into a body of water not only degrades it, but also
may weaken or kill the biota. Persistent contaminants, which
could result from long-term discharge of SWLDS leachate into
a stream, may concentrate and increase over time in the biota or
sediments. Some portion of the contaminant may be accumu-
lated and concentrated in living organisms; another portion may
be sorbed onto sediments from which it is accumulated by ben-
thic organisms. Death and decay of these organisms would per-
mit recycling and re-entry of contaminants into the water.
In addition to bioaccumulation, certain metals, such as iron,
may coat the bottom sediment so that feeding on the bottom is
not permitted. SWLDS leachate can thus cut off the food supply
from benthic organisms.
Cd, Hg, Pb, Cr+6, and Se are the metals found in SWLDS
leachate which are important in the areas of toxicity and dietary
essentiality (Table 7). The normal healthy adult levels for these
metals vary widely (Table 8). Here, it is important to note that
the effects of toxic dosages of many of the heavy metals are evi-
dent only after long periods of exposure as in the Minamata Bay
mercury poisonings.
Chromium
Chromium is the seventeenth most abundant element in the
earth's crust. Although chromium has oxidation states ranging
from Cr-2 to Cr+6, only the trivalent and hexavalent forms are
found in nature. Chromium is found in air, soil, some food, and
most biological systems, it is rarely found in natural waters. It is
also recognized as an essential trace element for humans.
Trivalent chromium, at normal blood pH, exists in large,
insoluble macromolecules which precipitate and become biolog-
ically inert. Hexavalent chromium, on the other hand, is irritat-
ing and corrosive to the mucous membranes, is absorbed via
ingestion, through the skin, and by inhalation. Hexavalent chro-
mium forms complexes with organic compounds which may,
unlike trivalent chromium, become cumulative in human tissue.
Knowledge of the harmful human health effects of hexavalent
chromium come almost entirely from the occupational area.
Symptoms of excessive dietary intake of chromium in man are
unknown, and marginal chromium deficiency is of greater
nutritional concern than overexposure. Since lifetime tolerable
levels of chromate ion are not known for man, it is recom-
mended by the Committee on Water Quality that public water
supply sources for drinking water contain no more than .05 mg/1
total chromium.8
Table VII: Toxicity and Dietary Essentiality of Metals*
Toxic only in
exceptional cases
Toxic in certain
occupational cases
Toxic at levels well
below most metals
Serious cumulative
body poisons
Dietary
essentiality
established
Cr+3
Fe
Zn
Dietary Dietary
essentiality essentiality
suspected unknown
Se
*Clark, T. P. Hydrogeology, geochemistry, ant
aspects of environmental impairment at an abandoned
Austin, Texas. M.A. Thesis. university of Texas
p.55.
Cr«
Cd, Hg, Pb
public health
landfill near
at Austin, Hay 1972.
Table VIII: Average Level and Primary Physiological Sites of
Concentration of Metals
Metal
Cr«
Cr+6
Fe
Zn
Se
Hg
Pb
Cd
*Clark,
Average level in healthy
adult human body, 154 Ibs.
.02-. 04 ppra
unknown
4-5 g
1.4-2.3 g
.1-.2 ppm
.05-. 2 ppm
.3 ppm
.2-. 8 ppm
Hydrogeology, geochemistry,
Primary
sites of concentration
cell components
liver, kidney, respi-
ratory problems
blood hemoglobin
kidney, liver, eyes
liver, kidney
kidney, liver, lungs
kidney, liver, lungs
liver and kidney
p.56-60.
Levels of total chromium in the SWLDS study leachates
ranged from .01 to .29 mg/1 with a mean of. 1 mg/1. Since most
of the reported levels are below the recommended standard, the
possibility of an adverse health effect due to total chromium
from this leachate appears remote. But since lifetime tolerable
levels of chromium are not known, caution should be used when
a drinking water supply may be affected.
Iron
Iron is the fourth most abundant element in the earth's crust.
It is an essential trace element required by both plants and ani-
mals. In some waters it may be a limiting factor for the growth of
algae and other plants. Iron is vital to the oxygen transport
mechanism in the blood of all vertebrate and some invertebrate
animals.
-------
214 Health Effects/ Municipal
Granick9 has estimated that the normal adult man contains
about 5 grams iron. Most body iron exists in complex forms
bound to proteins, as in the heme compounds, hemoglobin and
myoglobin. Outside of several rare excess iron storage diseases,
iron toxicity in man is seldom encountered. Iron deficiency is
one of the most frequently diagnosed deficiency diseases in man
and usually is brought about by chronic blood loss or dietary
insufficiencies.
Limitations are placed on the iron content of water mainly to
prevent objectional staining of laundry, cooking utensils and
porcelain fixtures such as sinks and bathtubs. Iron also imparts
objectional tastes and favors the growth of bacteria. Mainly
because of the effect on taste and staining, the EPA has pro-
posed a secondary level of .3 mg/1 iron.
Levels of iron in the SWLDS study leachates ranged from 6.5
to 1,550 mg/1 with a mean of 366 mg/1. The reported values at
each site were high enough to cause concern about staining of
fixtures and laundry and giving a bad taste to the water. Data
indicate that younger sites have very high values while older
sites have been flushed out by infiltrating water. Because high
iron values are detectable by color and taste, iron serves as an
indicator of contamination from SWLDS.
Zinc
Zinc ranks as the twentieth most abundant element in the
earth's crust. It is usually found in nature as the sulfide; often
associated with sulfides of other metals, especially lead, copper,
cadmium, and iron.
Zinc is an essential and beneficial element in human metabo-
lism, and is relatively non-toxic to man.10 The daily adult intake
averages 0 to I5mg. Because of undesirable aesthetic effects pro-
d uced by zinc in water, an E PA standard of 5 mg /1 has been pro-
posed.
Levels of zinc in the SWLDS study leachates ranged from .07
to 33 mg/1 with a mean of 7 mg/1. Since these levels are below
reported levels found in drinking water supplies which caused
no harmful effects, it is believed that the zinc in this leachate
does not pose a health problem."
Selenium
Elemental selenium is amorphous, crystalline or metallic in
form and usually occurs in nature in the sulfide ores of heavy
metals.
Any consideration of the toxicity of selenium to man must
consider the essential dietary requirement of the element in
amounts estimated to be .04 to .1 mg/kg of food. The daily
intake of I mg selenium per kilogram of body weight may pro-
duce chronic toxicity in man. The continued ingestion of food
selenium as low as .2 mg per kilogram of body weight may be
harmful.12
The mechanisms of selenium toxicity is uncertain, and the
precise ways in which selenium interferes with tissue structure
function remain unclear. A water supply containing9 mg/1 pro-
duced chronic selenosis in humans. Parents showed some mild
symptoms, but the children showed partial loss of hair, discol-
oration of finger nails, and loss of metal alertness. Upon discon-
tinuation of the water supply, the hair and finger nails regrew
and the children showed increased metal alertness. '3 Thus, EPA
primary standards recommend a maximum contaminant level
of .01 mg I of selenium in drinking water supplies.
There is also evidence that selenium can pass through the food
chain and accumulate in fish. Based on studies on a reservoir in
Colorado that contained bottom sediments high in selenium,
Barnhart reported that mortalities offish stocked in the reser-
voir were caused by selenium.14
Levels of selenium in the SWLDS study leachates ranged
from .001 to .33 mg 1 with a mean of .08 mg 1. Levels greatly
exceeding the standard (.01 mg/1) were found at each study site.
Since these reported levels exceed the standard, concern for this
toxic element should be a factor when considering a water
supply well in the area of a SWLDS, especially since .07 mg/1
gives rise to selenium toxicity.
Mercury
Mercury is a silver-white, mobile liquid metal at ordinary
temperatures. Biologically, it is a nonessential, nonbeneficial
element recognized as an element of high toxic potential. The
toxic properties of mercury and its compounds have formed the
basis for its widespread use for medical and agricultural pur-
poses.
There have been clinically recognized poisonings from mer-
cury or its compounds and, although its toxic properties are well
known historically, dramatic instances of toxicosis in man and
animals have occurred. Recently, the Minamata Bay poisonings
is a prime example.15 Toxic effects vary with the form of mer-
cury and its mode of entry.
Mercury intoxication may be acute or chronic. The mercu-
rous salts are less soluble than the mercuric salts and conse-
quently are less toxic. For man, the fatal oral dose of mercuric
salts ranges from 20 mg to 3 grams.16 Symptoms of acute inor-
ganic mercury poisoning include gastoenteritis and vomiting
followed by symptoms of systemic poisoning and circulatory
collapse. Chronic mercury poisoning results from exposure to
small amounts of mercury over extended time periods. Chronic
poisoning from inorganic mercurials most often has been asso-
ciated with industrial exposure, whereas poisoning from the
organic derivatives has been the result of accidents or environ-
mental contamination. Alkyl compounds are the derivatives of
mercury most toxic to man, producing illness, irreversible neu-
rological damage, or death from the ingestion of only a few mil-
ligrams.17
Since mercury added to water is subject to biochemical
change, the total mercury level should be the basis for a mercury
criterion, instead of any particular form in which it may be
found within a sample. Also, due to bioconcentration, mercury
limits must accommodate the fgod chain transport path from
lower organisms to man. Therefore, EPA primary standards
recommend a maximum contaminant level of .002 mg/1 of mer-
cury in drinking water supplies.
Levels of mercury in the SWLDS study leachates ranged from
.0006 to .16 mg/1 with a mean of .027 mg/1. Mercury levels
exceeding the standard (.002 mg/1) were found at each study
site. Because mercury effects result from exposure over
extended periods of time, harmful effects could take place if
mercury in this concentration reached drinking water supplies.
Lead
Lead is a toxic metal that tends to accumulate in the bones of
man and other animals. As far as is known, lead has no benefi-
cial or desirable nutritional effects. Although seldom seen in the
adult population, irreversible damage to the brain is a frequent
result of lead intoxication in children. Lead intoxication most
often is from eating lead-containing paint still found in older
homes.
The concentration of lead required to produce symptoms of
plumbism are variable, but in general are more precisely known
than for other cumulative body poisons. Chronic lead poisoning
symptoms are loss of appetite, weakness, lesions, neuromuscu-
lar damages, brain damage, and circulatory damage.
Like other heavy metals, lead may act as an enzyme inhibitor.
There are suggestions that lead may disrupt biochemical func-
tions such as glucose transport and detoxification mecha-
nisms.18 Lead appears to operate at a large number of biochem-
ical sites, and experiments with lead distribution patterns in
-------
Health Effects/Municipal
215
higher animals have indicated similarities to the distribution
observed in lower species. Lead accumulation is most noted in
the liver, bones, kidneys, spleen, lungs, and hair. For these rea-
sons, the EPA maximum contaminant level of lead permissible
in drinking water is .05 mg/1.
Levels of lead in the SWLDS study leachates ranged from
.017 to .33 mg/1 with a mean of .17 mg/1. Reported values are
higher than the standard (.05 mg/1) but are less than the esti-
mated maximum safe level of lead intake (300 micrograms/
day), assuming a daily water intake of 2 liters/day.19
Cadmium
Biologically, cadmium is a nonessential, nonbeneficial ele-
ment which is recognized to be an element of high toxic poten-
tial.
Airborne cadmium has been reported to be a promoter of
human hypertension and possibly related to cardiovascular dis-
ease. Waterborne cadmium has been related to hypertension.
Despite the amount of evidence amassed showing chronic toxic
effects of cadmium, the prognosis of chronic cadmium poison-
ing remains poor, and as yet no specific treatment of the disease
is available. The fact remains that there is little if any significant
improvement in persons already affected with chronic poison-
ing, and once taken up in the body cadmium is practically never
excreted. The most significant case of cadmium poisoning was
the recent outbreak of Itai-Itai disease in a small Japanese farm-
ing and fishing village.20 The cause has been linked to cadmium
uptake by major food staples which were contaminated by irri-
gation water from a river where cadmium wastes had been dis-
charged.
Along with selenium antf mercury, cadmium shares the place
of the most stringently restricted metals in drinking water stan-
dards. The EPA maximum contaminant level for cadmium is
.01 mg/1.
Levels of cadmium in the SWLDS study leachates ranged
from .001 to .07 mg/1 with a mean of .01 mg/1. Some reported
values were above the standard (.01 mg/1), but most values are
below the reported daily intake of .026 mg/day, assuming a
daily water intake of 2 liters/day.19
CONCLUSION
The study indicates that undiluted, raw leachate contains sub-
stances that are potential threats to human health. Although
raw leachate contains concentrations of heavy metals in excess
of the drinking water standards, it is not clear how likely it
would for these recorded levels to be found in drinking water
supplies or for contamination to reach the human body. Before
leachate reaches an aquifer, it is subject to the purifying affect of
the unsaturated zone (attenuation or dilution). The usual odor
and / or color of municipal leachate would be a significant deter-
rent to using water contaminated by it for human needs. Further
study of the environmental affects of leachate are being under-
taken by the EPA.
It is clear that if solid waste is placed directly into ground
water, or if leachate is allowed to drain directly into surface
water, it can cause severe damage. It can destroy life in a water
resource by coating the bottom sediment so that feeding by the
animal population is precluded—iron is a prime example. Toxic
metals can build up in the aquatic life and destroy their use by
man—selenium and mercury are examples. Therefore leachate
generated by improperly located and managed solid waste dis-
posal sites can destroy drinking water resources making it neces-
sary to locate new sources, at considerable cost.
Finally, in evaluating the data from this study, it should be
kept in mind that data were obtained from the monitoring of five
municipal solid waste disposal sites that received only municipal
and commercial wastes. Many SWLDS never question or docu-
ment types of incoming waste. Many sites accept some type of
hazardous waste (inorganic and organic chemicals, electroplat-
ing sludges, petroleum wastes, pharmaceuticals, etc.) and/or
municipal sewage sludges, which tend to be high in organics and
heavy metals. This study can only report values that have been
observed in strictly municipal SWLDS. The monitoring of
SWLDS which accept hazardous wastes and/or sewage sludge
is presently being undertaken by EPA.
REFERENCES
1. Fenn, D. G., K. J. Hanley, and T. V. DeGeare. Use of the
water balance method for predicting leachate generation from
solid waste disposal sites. Environmental Protection Publica-
tion SW-168. [Washington], U.S. Environmental Protection
Agency, Oct. 1975. 40 p.
2. U.S. Environmental Protection Agency, Office of Solid
Waste Management Programs. Resource recovery and waste
reduction; third report to Congress. Environmental Protection
Publication SW-161. Washington, U.S. Government Printing
Office, 1975. 96 p.
3. ASCE manuals of engineering practice no. 39; sanitary
landfill. New York, American Society of Civil Engineers, 1959.
61 p.
4. U.S. Environmental Protection Agency, National interim
primary drinking water regulations. Federal Register, 40(248):
59566-59588, Dec. 24, 1975.
5. Hansen, D. J., P R. Parrish, and J. Forester. Aroclor
1016: toxicity to and uptake by esturine animals. Environmental
Research, 7(3):363-373, June 1974.
6. Nimmo, D. R., et al. Polychlorinated biphenyl absorbed
from sediments from by fiddler crabs and pink shrimp. Nature,
231(5297):50-52, May 7, 1971.
7. Holland, H. T., and D. L. Coppage, Sensitivity to pesti-
cides in three generations of sheepshead minnows. Bulletin of
Environmental Contamination and Toxicology, 5(4):362-367,
July-Aug. 1970.
8. National Academy of Sciences—National Academy of
Engineering. Water quality criteria, 1972. Washington, U.S.
Government Printing Office, 1972. 594 p. (Distributed by
National Technical Information Service, Springfield, Va., as
PB-236 199.)
9. Granick, S. Iron metabolism in animals and plants. In C.
A. Lamb, et al., eds. Trace elements; Proceedings, Conference
held at the Ohio Agricultural Experiment Station, Wooster,
Ohio, Oct. 14-16, 1957. New York, Academic Press, Inc. 1958.
p.365-382.
10. Valle'e, B. L. Zinc and its biological significance. A.M.A.
Archives of Industrial Health, 16(2): 147-154, Aug. 1957.
11. Public Health Service Drinking Water Standards; revised
1962. Public Health Service Publication No. 956. Washington,
U.S. Government Printing Office, 1969. 61 p.
12. Rosenfeld, I., and O. A. Beath. Selenium: geobotany, bio-
chemistry, toxicity, and nutrition. New York, Academic Press,
1964.411 p.
13. Selenium poisons Indians. Science Newsletter, 81(16):254,
Apr. 21, 1962.
14. Barnhart, R. A. Chemical factors affecting the survival of
game fish in a western Colorado reservoir. Colorado Coopera-
tive Fisheries Research Unit Quarterly Report, 4:25-27, July
1957-June 1958. (Thesis abstract.)
15. Smith, W. E., and A. M. Smith. Minamata. New York,
Holt, Rinehart and Winston, 1975. 192 p.
16. Stokinger, H. Mercury, Hg. In F. A. Patty, ed. Industrial
hygiene and toxicology. 2d rev. ed. v.2. New York, Interscience
Publishers, 1963. p. 1090-1104.
17. Berglund, F., and M. Berlin. Risk of methylmercury cu-
mulation in man and mammals and the relation between body
-------
216 Health Effects/Municipal
burden of methylmercury and toxic effects. In M. W. Miller,
and G. G. Berg, eds. Chemical fallout; current research on per-
sistent pesticides. Proceedings; Rochester Conference on Toxic-
ity. University of Rochester, June 4-6, 1968. Springfield, 111.,
Charles C. Thomas, 1969. p.258-273.
18. Ulmer, D. C.,and B. L. Vallee. Effects of lead on biochem-
ical systems. In D. D. Hemphill, ed. Trace substances in envi-
ronmental health-11; Proceedings; University of Missouri's 2nd
Annual Conference on Trace Substance in Environmental
Health, Columbia, No., July 16-18, 1968. p.7-27.
19. Pier, S. M.The role of heavy metals inhuman health. Tex-
as Reports on Biology and Medicine, 33(l):85-106, 1975.
20. Nilsson, R. Cadmium effects. In Metals and ecology;
Symposium, Stockholm, Mar. 24,1969. p.56-63. (Swedish Nat-
ural Science Research Council (Statens Naturvetenskapliga
Forskningsrod), Ecological Research Committee Bulletin no.
5.)
------- |