United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens GA 30613
Research and Development
EPA-600/S3-82-060 Sept. 1982
Project Summary
Retention and
Transformations of Selected
Pesticides and Phosphorus in
Soil-Water Systems: A Critical
Review
P. S. C. Rao and J. M. Davidson, Editors
The current state-of-the-art for
measuring or estimating pesticide
retention and transformation param-
eters required in non-point source
pollution models was reviewed. A
data base of sorption partition coeffi-
cients, degradation rate coefficients,
and half-lives for a broad spectrum of
pesticides was compiled from a
literature survey. Adsorption partition
coefficients normalized with respect
to soil organic carbon content were
approximately constant across soils
for a given pesticide. Octanol-water
partition coefficients were "good"
predictors of pesticide adsorption
parameters. Chemical persistence in
soils for a large number of pesticides
has been measured under a variety of
soil environmental conditions. These
data were used to calculate first-order
decay coefficients and half-lives. The
variability of these parameters for a
given pesticide across several soils
was within a factor of two. Multiple re-
gression properties could not be devel-
oped from the literature data because
of inadequate information regarding
the physical, chemical and environ-
mental conditions of soil during the
pesticide degradation studies. Sea-
sonal losses by runoff from agri-
cultural fields were generally less than
0.5% - 1.0% of the total amount
applied. Although pesticide concen-
trations on the sediment phase of the
runoff are larger than those in the
water phase, pesticide carried in the
water phase accounted for more than
90% of the total mass emission during
a given runoff event.
Phosphate sorption parameters
(primarily Langmuir constants) were
collected from the literature or com-
puted from published adsorption iso-
therms. Statistical analysis showed
that Langmuir sorption parameters
Smax and k, each .normalized with
respect to extractable Fe and Al, were
significantly correlated to the extract-
able metals. The correlations gave
higher R2 values and lower probability
levels of significance for oxalate ex-
tractable Fe and Al than forcitrate-di-
thionite-bicarbonate extractions. Cor-
relations for other parameters with ex-
tractable Fe and Al were less signif-
icant. The composition and degree of
crystallinity of Fe, Al oxhydroxides ap-
pear to be the dominant factors in con-
trolling phosphate sorption. Lack of
uniformity in exerimental methods
used for determining Langmuir sorp-
tion parameters was noted during the
literature survey. Development of
standardized methodology (protocols)
for this purpose appears essential for
quantification of appropriate sorption
parameters.
This Project Summary was devel-
oped by EPA 's Environmental Research
Laboratory, Athens, GA, to announce
-------
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back).
Introduction
The Federal Water Pollution Control
Act Amendments of 1972, Public Law
No. 92-500, specifies that the Adminis-
trator of the U.S. Environmental Protec-
tion Agency (EPA) shall, in cooperation
with other agencies, provide guidelines
for identifying and evaluating the nature
and extent of nonpoint pollution sources.
Because fertilizers and pesticides play
such a major role in today's agriculture,
runoff from fields involved with agri-
cultural production has long been
suspected of being a major nonpoint
pollution source. Although it is difficult
to conceive of a situation in which all
possible environmental risks associated
with the use of agricultural chemicals
could be eliminated, management
practices can be used that will signifi-
cantly reduce these risks. Nonpoint
agricultural pollutants of primary con-
cern are sediment, nitrogen, phosphorus,
and pesticides. The latter two pollutants
can be transported from an agricultural
field by both the water and sediment
phases.
Several models (stochastic, empirical
and deterministic) exist for estimating
water and sediment transport from
small fields and watersheds. More than
45 years of research on erosion by the
U.S. Department of Agriculture in
cooperation with state agricultural
experiment stations has resulted in the
development of numerical relationships
for estimating annual soil losses from
fields. Because of the success of these
models and their previous calibration
for specific regions, most agricultural
chemical transport models were devel-
oped by "piggybacking" the compo-
nents for chemical transport to the
hydrologic and sediment transport
models. The sediment and chemical
transport models were developed in
order to simulate the impact of agri-
cultural production on water quality.
The models also have the potential to be
used by state and local agencies in
developing and/or identifying land use
management practices that will provide
the least risk to the environment.
Many of the existing chemical trans-
port simulation models have been
calibrated to describe the behavior of a
given plant nutrient or pesticide at a
given location and for a specific cultural
system. The feasibility of continuing to
calibrate these simulation models for a
wide range of chemicals and manage-
ment practices is not practical. Therefore,
general relationships for estimating the
basic coefficients required to describe
adsorption and transformations of
agricultural chemicals in the soil
surface region subject to erosion are
needed. Also the confidence that can be
attached to the independently measured
or estimated coefficients used in the
chemical transport models for describ-
ing adsorption and transformation pro-
cesses must be quantified.
Numerous equilibrium adsorption
studies have been conducted using
various pesticides and phosphorus
sources as well as different soils. The
validity of the equilibrium adsorption
assumption based on relatively short-
term experiments (less than 72 hours)
and the reversibility of the adsorption-
desorption process has been questioned
by several researchers. Adsorption
associated with short-term laboratory
experiments may not be relevant for the
long contact periods encountered
under natural field conditions. Also,
recent experiments involving "bound"
pesticide residues point out the problem
associated with assuming reversible
adsorption, especially that occurring
during sediment and water transport
from an agricultural field or water shed.
The bound pesticide residue question
suggests that some of the pesticide
transformation or disappearance data
available in the literature may be in error
and not suitable for estimating trans-
formation rate coefficients of the
original parent compound.
The primary emphasis of the present
report was to present an extensive and
reliable data base of the principal
coefficients for describing adsorption
and transformation characteristics of
phosphorus and a broad spectrum of
pesticides used across a range of soil
types. This information was obtained
from an extensive literature search
' using various computer information
retrieval packages (data banks such as
CANE, BIOSIS, etc.). The dependence of
these retention and transformation
coefficients on selected soil properties
was evaluated. The information pre-
sented in this report should be helpful in
estimating the values of the retention
and transformation parameters required
in various non-point source pollution
models.
Conclusions
A large data base exists for esti mat i ng
partition coefficients for pesticide
retention in soils. An analysis of these
data indicated that errors associated
with various simplifying assumptions
(e.g., linear and singular isotherms;
instantaneous equilibrium) appear to be
within a factor of 2 or 3. Such errors
may be tolerable for most nonpoint
source pollution modeling applications.
For a given pesticide, adsorption parti-
tion coefficients based on soil organic
carbon were fairly constant regardless
of soil type. Furthermore, octanol-water
partition coefficients were good predic-
tors of pesticide adsorption partition co-
efficients.
The persistence in soils of a large
number of pesticides under a broad
range of soil environmental conditions
has been reported. A data base was
compiled for first-order decay constants
(k) and half-lives (11,2) for pesticide
disappearance in soils from these
reports. Over the wide range of soil and
environmental conditions under which
degradation was measured, the coeffi-
cient of variation of the average k and ti 2
values for a given pesticide was
surprisingly small (<100%). Thus,
pesticide disappearance rates can be
estimated within a factor of 2 to 4 for
most pesticides using presently available
data. In most cases, observed half-lives
of pesticides under field conditions
were shorter than those measured in
laboratory incubation studies. This was
attributed to the fact that under field
conditions a multitude of factors and
processes contribute to pesticide disap-
pearance, while laboratory studies are
performed under controlled conditions
and therefore measure fewer, if not a
single process.
Efforts to develop multiple regression
equations correlating degradation rates
with soil properties were unsuccessful.
Part of the problem arises because many
reports reviewed failed to give soil
physico-chemical properties and incu-
bation conditions (temperature and soil-
water tension). Pesticides were placed
into the following three groups based
upon their half-lives in soils: non-
persistent (ti/2 < 20 days), moderately
persistent (20 < ti/2 < 100 days) and per-
sistent (ti/2 > 100 days). Pesticides in
the first group are 2,4-D, 2,4,5-T,
dicamba, dalapon, methyl parathion,
malathion, and captan. Moderately per-
sistent pesticides are atrazine, simazine,
terbacil, linuron, TCA, glyphosate, para-
thion, diazinon, fonofos, phorate, carbo-
furan, carbaryl, aldrin, dieldrin, endrin,
heptachlor and PCP. Persistent pesti-
-------
cides are trifluralin, bromacil, picloram,
paraquat, DDT, chlordane, and lindane.
A broad range of langmuir isotherm
parameters were compiled for phosphate
sorption by soils and other solid adsorb-
ents. Based upon the limited amount of
data available, the sorption parameters
were found to be significantly correlated
with extractable "active" Fe and Al. A
lack of uniformity in experimental meth-
ods used in determining the Langmuir
sorption parameters or phosphate sorp-
tion indices was noted. Development of
standardized methodology (or protocols)
for this purpose appears to be essential
for quantification of phosphate sorption
parameters. "Active" Fe and Al appear
to be measured by oxalate extraction
rather than by citrate-dithionite-bi-
carbonate extraction. Crystallinity of Fe
and Al oxyhydroxides play a dominant
role in determining inorganic phosphate
sorption. A "universal" partition func-
tion for inorganic phosphorus retention
in soils can be developed provided
proper input parameters are measured.
These parameters include the measure-
ment of "active" Fe and Al, time-de-
pendence of phosphate adsorption-de-
sorption and phosphate sorption iso-
therm parameters.
Recommendations
Measurement of octanol-water parti-
tion coefficients for a broad spectrum of
pesticides should be continued. Special
emphasis should be given to recently
developed high-pressure liquid chro-
matography (HPLC) methods for esti-
mating octanol-water partition coeffi-
cients. Also, this concept should be ex-
tended for ionic and ionizable organic
compounds. The relative contributions
of various particle-size fractions of soils
to pesticide retention should also be
measured. Such particle size partition-
ing data are now available for only a
limited number of organic compounds
and not for a wide-range of soil-pesti-
cide combinations. This information is
needed to evaluate the significance of
runoff sediment "enrichment" by fines
in estimating total pesticide losses.
The disappearance rate of solvent-
extractable parent compound should
not be used as the only measure of
pesticide degradation rate in soil-water
systems. The potential for significant
accumulations of toxic metabolites
(especially under anaerobic conditions)
and formation of "bound residues"
must be taken into account. Mineraliza-
tion rate (i.e., total breakdown of
pesticides to carbon dioxide, water, and
inorganic ions) should be used as an
index of pesticide degradation rate
because it represents total detoxification
of the pesticide. Because mineralization
rates are generally smaller than parent
compound disappearance rates, the
former provides a more conservative
estimate of the pesticide degradation
rate. The rates and mechanisms of
bound pesticide residue formation in
soils as well as the release character-
istics and environmental toxicity of
these residues should be characterized.
Special attention should also be given to
rates of formation and release of bound
residues under anaerobic environments
(encountered by sediment-bound pesti-
cide residues in streams, rivers, lakes,
etc.).
Standardized methodology (protocols)
should be developed for measuring
phosphate sorption parameters, "active"
soil components (Fe and Al oxyhydrox-
ides and solubilized Ca) involved in
phosphate retention, and the time-
dependence of phosphate sorption-
release in soil-water systems. Further
testing of various phosphate sorption
models is required in order to develop
appropriate sorption parameters. A
quantitative index needs to be defined
for the "crystallinity" of the Fe and Al
oxyhydroxides in soils. Such an index
will be used in developing a "universal"
partition function for phosphate retention
by soils, similar in concept to pesticide
partition coefficients based on soil
organic carbon.
P. S. C. Rao andJ. M. Davidson are with tne University of Florida. Gainesville, FL
32611.
C. N. Smith is the EPA Project Officer (see below).
The complete report, entitled "Retention and Transformations of Selected
Pesticides and Phosphorus in Soil- Water Systems: A Critical Review," (Order
No. PB 82-256 884; Cost: $25.50, subject to change) will be available only
from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
College Station Road
Athens, GA 30613
* US. GOVERNMENT PRINTING OFFICE- 1W2-559-017/0820
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
US ENVIR PROTECTION AGENCY
REGION 5 LIBRAKY
230 S DEARBORN STREET
CHICAGO IL 60604
------- |