A Report on the

Plant Uptake of Metals

from Fertilizers
This report summarizes and makes recommendations based on the
Washington State University report entitled:

Influence of Metal Rates and Forms on Crop Productivity and

Metal Uptake in Some Washington Soils

Submitted to:

Washington State Legislature

Submitted by:

Lead Agency
Washington State Department of Agriculture

Cooperating Agencies
Washington State Department of Health
Washington State Department of Ecology
Washington State University

This report is mandated by the

Fertilizer Regulation Act of 1998 (Chapter 36, Laws of 1998).

December 31, 2001

For additional copies of this report contact:

Washington State Department of Agriculture

Pesticide Management Division
P.O. Box 42560
Olympia, WA 98504-2560
(360)902-1851
or visit our website: www.wa.gov/agr/pmd

PUB 072 (N/12/01)

Inquiries regarding availability of this publication in alternative formats

should be directed to the WSDA Receptionist at (360)902-1976 or
Telecommunication Device for the Deaf (360)902-1996.

Report prepared as required by Section 14, Chapter 36, Laws of 1998

 Table of Contents

Acronyms and Symbols ........................................................................................... 2

Glossary ................................................................................................................... 3
I. Executive Summary ............................................................................................. 5
II. Introduction ........................................................................................................ 7
History .......................................................................................................... 7
III. Study Summaries .............................................................................................. 8
Part 1: Plant Yield and Uptake of Heavy Metals from Fertilizers
and Soluble Metal Salts in Greenhouse-Grown Lettuce, Puyallup Site ....... 9
Part 2: Plant Yield and Uptake of Heavy Metals from Fertilizers
and Soluble Metal Salts in Field-Grown Lettuce and Cucumbers,
Puyallup Site ................................................................................................ 11
Part 3: Uptake of Heavy Metals from Fertilizer and a Soluble Metal
Salt in Field-Grown Potatoes and Winter Wheat, Prosser Site .................... 13
Part 4: Laboratory Analysis of the Availability of Heavy Metals in
Soils Fortified with Fertilizers and Soluble Metal Salts, Pullman Site ........ 15

IV. Implications ...................................................................................................... 16

A. Accumulation of Metals in Soil Over Time ............................................ 16
B. Plant Uptake of Metals from Soil Treated with Fertilizer ....................... 18
C. Relevance to Other Studies ..................................................................... 18
(1) California Department of Food and Agriculture, 1998.
Development of Risk-Based Concentrations for Arsenic ,
Cadmium, and Lead in Inorganic Commercial
Fertilizers ......................................................................................... 19
(2) U.S. EPA, 1999. Estimating Risk from Contaminants
Contained in Agricultural Fertilizers ............................................... 20
(3) The Fertilizer Institute (TFI), 2000. Health Risk Evaluation
of Select Metals in Inorganic Fertilizers Post Application ............... 20
(4) Use of the Tc’s from WSU study................................................. 21
D. Comparison of concentrations of metals in crops to FDA's Market
Basket survey results ............................................................................. 21
V. Conclusions ......................................................................................................... 22
VI. Recommendations ............................................................................................. 23

1
 Acronyms and Symbols

DAP Diammonium Phosphate

DTPA Diethylenetriaminepentaacetic acid.
E DAP Diammonium Phosphate from Eastern United States
PR Phosphate Rock
Tc Transfer Coefficient
TSP Triple Superphosphate
W DAP Diammonium Phosphate from Western United States
WSU Washington State University

2
Glossary
For purposes of this report, the following definitions are used:

Diammonium phosphate – a product composed of ammonium phosphates, principally

diammonium phosphate, resulting from the ammoniation of phoshoric acid. It may
contain up to 2% non-ammoniacal nitrogen. An ingredient in commercial phosphate
fertilizers.

Diethylenetriaminepentaacetic acid – a chelating agent for extracting cadmium, lead,

zinc, and other heavy metals from soils.

Ironite – A fertilizer sold primarily in the home and garden market as a source of iron
and zinc. It contains relatively high levels of arsenic and lead.

Lime – an agricultural product with calcium and magnesium compound that are used for
the purpose of neutralizing soil acidity.

Phosphate rock - a natural rock containing one or more calcium phosphate minerals of
sufficient purity and quantity to permit its use, either directly or after concentration, in the
manufacture of phosphate fertilizers. About 84% of the US production of phosphate rock
comes from North Carolina and Florida. The remaining is from Idaho, Montana, Utah,
and Wyoming.

Sodium bicarbonate – a sodium salt for use in extracting arsenic from the soil.

Transfer coefficient - the ratio of the concentration of metal in a plant to the

concentration of metal in soil. Transfer coefficients can be calculated based on the
concentration of extractable metals in soil or based on the total concentration of metal in
soil. A higher transfer coefficient indicates a greater transfer of metal from soil into
plant.

Triple superphosphate - a product obtained when rock phosphate is treated with sulfuric
acid, phosphoric acid, or a mixture of those acids. Refers to all grades containing 40% or
more available phosphate, which are commonly made by acidulation of natural
phosphatic material primarily with phosphoric acid. An ingredient in commercial
phosphate fertilizers.

Zinc sulfate – a zinc salt of sulfuric acid for use as a zinc fertilizer.

3
THIS PAGE INTENTIONALLY LEFT BLANK

4
I. Executive Summary
The Fertilizer Regulation Act of 1998 directed the Department of Agriculture to conduct a
comprehensive study evaluating plant uptake of metals from fertilizers. As mandated by
the Act, this report, prepared by the Departments of Agriculture, Ecology, Health and
Washington State University (WSU) summarizes and interprets the results of the study.

The WSU study was conducted to determine, in part, if the accumulation of arsenic, lead,
and cadmium in soils and crops may occur from continual agricultural applications of
fertilizers. The four crops used were lettuce, cucumber, wheat, and potato. The two soils
utilized were a Sultan silt loam at the Puyallup site or Quincy-Hezel fine sandy loam at
the Prosser site. Commercial fertilizers that contain relatively high levels of heavy metals
were evaluated.

The results of the study indicate that cadmium is of greater concern than arsenic and lead
in terms of crop uptake. Cadmium is commonly found in phosphate fertilizers used in
agriculture. The WSU study also shows, that for the conditions tested, the amount of
cadmium taken up by crops is greatly influenced by the source of the cadmium.
Although crops take up some amount of cadmium from the soil, the WSU study indicates
that cadmium could build up in soil over time due to continual fertilizer use. This is more
likely to occur from using fertilizers that contain higher cadmium levels. Increases in soil
cadmium concentrations are of concern since they may lead to increases in cadmium
concentrations in crops. Cadmium levels in soils should be periodically monitored to
ensure that levels do not become a concern in the future.

There are many factors that influence the uptake of metals into plants. They include soil
pH, source of metals, soil type, existing or initial soil metal concentrations, and whether
all plant material is removed before tilling and subsequent planting. Due to the many
uncertainties associated with predicting the uptake of metals into plants and the limited
number of conditions evaluated in this study, projections of metal concentrations in
plants over time cannot be estimated with a high degree of certainty. This is because
soils vary considerably in their ability to bind cadmium. Additionally, there may be
major differences in cadmium availability between soil types even when they contain the
same amount of cadmium. The study would need to be extended to other soil types
before it could be applicable to a broad range of soil conditions.

Arsenic and lead levels in most agricultural fertilizers are very low. Occasionally,
relatively high levels of lead can be found in some micronutrient fertilizers, but the
application rate of those products is so low that it does not significantly contribute to lead
soil levels. In addition, compared to cadmium, the transfer of arsenic and lead from soil
to crop is minimal.

The study indicates the existing Washington State standards for arsenic, cadmium, and
lead in fertilizer appear to be sufficiently protective at this time. However, the study also
indicates there is a potential for the levels of these metals to increase in soil over time.
This is especially true of cadmium, due to current applications of phosphate fertilizers

5
containing relatively high levels of cadmium. The agencies involved in the development
of this report make the following recommendations:

1) WSU should extend its study for a minimum of two years and focus on the
development of a soil-to-plant transfer coefficient for cadmium that is
independent of time. This would greatly improve the accuracy of predicting crop
uptake of cadmium.
2) A long-term program should be developed to monitor soil levels of arsenic,
cadmium, and lead over time to ensure that soil levels do not become a concern in
the future.
3) A process should be initiated to attempt to develop a regional approach dealing
with the issue of metals standards and labeling. This should initially include
Washington, Oregon and Idaho, but could be expanded to include other western
states.

6
II. Introduction
The Fertilizer Regulation Act of 1998 directed the Department of Agriculture to conduct a
comprehensive study evaluating plant uptake of metals from fertilizers. As mandated by
the Act, this report, prepared by the Departments of Agriculture, Ecology, Health and
Washington State University (WSU), summarizes and interprets the results of the study
conducted by WSU regarding potential impacts to public and environmental health from
metals in fertilizers.

History

The issue of heavy metals in fertilizers arose in Washington State when a small group of
citizens living in and around Quincy, Washington became concerned about recycling
hazardous wastes into fertilizers. Several 1997 articles in The Seattle Times addressed
these concerns and brought the issue to national attention.

The state agencies recognized recycling of waste materials as a valid way to conserve
limited natural resources; and that the use of recycled material in fertilizer had grown in
concert with increased industrialization, advancement in technologies, population growth,
and increased costs of production. It was also recognized that fertilizers could contain
contaminants such as heavy metals at varying concentrations, regardless of whether the
fertilizer was waste-derived or not.

In August 1997, the Departments of Agriculture, Ecology, and Health proposed a ten-
point plan to ensure that fertilizers used in Washington did not pose a risk to human
health or the environment. As part of this plan, an advisory group was established which
included representatives from the fertilizer industry, agricultural commodity producers
and processors, environmental interests, county health agencies, the departments of
Agriculture, Ecology, Health, and Labor and Industries and the Governor’s Office. A
primary task of the group was to advise the agencies in developing legislation that would
authorize the Department of Agriculture to adopt standards for heavy metals content for
all fertilizer products.

The advisory group met a number of times in late 1997 and, in January 1998, the
agencies proposed legislation to significantly revise the state's fertilizer law. On March
18, 1998, the Fertilizer Regulation Act was signed into law and Washington became the
first state in the country to adopt metals standards for fertilizers.

The Act adopted the Canadian standards for maximum acceptable heavy metals additions
to soil. The standards limit the levels of nine heavy metals in fertilizers – arsenic,
cadmium, cobalt, mercury, molybdenum, nickel, lead, selenium, and zinc. The following
table converts the Canadian standards (expressed in kilograms/hectare/year) to
Washington standards (expressed in pounds/acre/year).

7
 Table 1. Washington standards for metal additions to
soils.
Metals lbs/acre/yr kg/ha/yr
Arsenic 0.297 0.333
Cadmium 0.079 0.089
Cobalt 0.594 0.667
Mercury 0.019 0.022
Molybdenum 0.079 0.089
Nickel 0.713 0.800
Lead 1.981 2.222
Selenium 0.055 0.062
Zinc 7.329 8.222

III. Study Summaries

This Summary Report is based on the Washington State University (WSU) study entitled,
"Influence of Metal Rates and Forms on Crop Productivity and Metal Uptake in some
Washington Soils". The WSU study was designed to evaluate the various effects that
heavy metals from fertilizers might have on the production of four crops produced in
Washington State soils at two sites. The four crops used were lettuce, cucumber, wheat,
and potato. The two soils utilized were a Sultan silt loam at the Puyallup site or Quincy-
Hezel fine sandy loam at the Prosser site. Commercial fertilizers that contain relatively
high levels of heavy metals were used in this study. The materials used in the production
of commercial phosphate fertilizers such as phosphate rock (PR), triple superphosphate
(TSP), and diammonium phosphate (DAP) were evaluated. For the purpose of
comparison, the effects of soluble metal salts were also determined. Soluble metal salts
are not commercial fertilizers but represent a readily available source of heavy metals to
crops. Treatments containing soluble metal salts were compared with those with
commercial fertilizers. Liming treatments were also evaluated. Liming is the act of
adding lime to the soil for the purpose of neutralizing soil acidity (increasing soil pH).
The availability of metals in soils is dependent on the acidity of the soil. For a full
understanding of the WSU research effort, reviewers of this Summary Report are
encouraged to consult the original study.

The WSU Study consisted of four parts: 1) a greenhouse study conducted in Puyallup, 2)
a field study conducted in Puyallup, 3) a field study conducted in Prosser, and 4) an
incubation study conducted in Pullman. Each part of the study was conducted under
different conditions, and, in most cases, different parameters were measured. The four
parts of the WSU Study are summarized as follows.

8
Part 1: Plant Yield and Uptake of Heavy Metals from Fertilizers and Soluble Metal
Salts by Greenhouse-Grown Lettuce - Puyallup Site

Purpose:

The purpose of the study was to see if fertilizers applied to a Sultan silt loam at rates up
to 40 times the Washington fertilizer standard (or Canadian standard) for arsenic,
cadmium and lead could cause phytotoxicity in lettuce grown in pots in the greenhouse.
Phytotoxicity was measured by lettuce dry and wet yields. In addition, the study
evaluated the uptake of metals by lettuce grown in the greenhouse with varying
application rates of three fertilizers (PR, granular zinc and Ironite) and soluble metal
salts. Fertilizers and soluble metal salts were applied up to 40 times the Washington
fertilizer standard for arsenic (0.333 kg/ha/yr), cadmium (0.089 kg/ha/yr), and lead (2.222
kg/ha/yr). The study evaluated if the uptake of metals into lettuce differs between
fertilizers and soluble metal salts. Concentrations of total and extractable metals in soils
were measured to determine which most closely correlated with plant metal
concentrations. Extractable metals reflect the amount of metal in soil that is in a form
available for plant uptake (i.e. plant availability or bioavailability). Soil metal and plant
metal concentrations were used to develop soil-to-plant transfer coefficient (Tc). A Tc is
the ratio of the concentration of metal in a plant to the concentration of metal in soil. A
higher Tc indicates a greater transfer of metal from the soil into the plant. The study also
evaluated the effect of liming on plant uptake of metals. Greenhouse study results were
compared with the results from the field study.

Methods:

• Crop: lettuce
• Soil: Sultan silt loam
• Application Rates: 0, 1X, 2X, 4X and 8X, repeated after 3-month incubation of the
treated soil for a total of 5 times to reach a maximum application of 40X. 1X is based
on maximum allowable metal addition to soil from the Washington fertilizer standard
(WAC 16-200-7064). Product application rates at 1X: 97.1 kg granular zinc/ha,
2493 kg western PR/ha, 76.5 kg Ironite/ha. Metals addition at 1X rate for soluble
metal salts: 0.09 kg cadmium /ha, 0.33 kg arsenic /ha, 2.2 kg lead/ha.
• Fertilizers and soluble metal salts applied and metals measured: granular zinc
(cadmium and lead), PR (cadmium), Ironite (arsenic and cadmium), and soluble metal
salts (arsenic, cadmium and lead).

Main Results:

Cadmium
• Application of cadmium to a Sultan silt loam up to 40 times the Washington fertilizer
standard had no effect on fresh or dry weight lettuce yield, i.e. phytotoxicity was not
observed at any application rate.
• Lettuce cadmium concentrations were higher from soluble metal salt application than
from application of any of the three fertilizers. Lettuce uptake of cadmium from

9
 different sources was observed in the following order: PR < granular zinc < soluble
metal salts. Ironite did not contribute significant cadmium to lettuce.
• Cadmium added to the soil from soluble metal salts was more extractable compared
to the fertilizers. The extractability increased in the order of PR < granular zinc <
soluble metal salts.
• Lettuce cadmium concentrations correlated better with soil extractable cadmium than
with total cadmium soil concentrations.
• Tc's based on soil extractable cadmium provided a more accurate prediction of uptake
of cadmium by lettuce than Tc's based on the total cadmium concentration in the soil.
• Cadmium soil-lettuce Tc's were higher for soluble cadmium salt than for the
fertilizers. The average soil-plant Tc for cadmium for PR and granular zinc was 4.17.
• Soil cadmium extractability decreased with increasing lime rates. Liming had the
greatest effect on the extractability of cadmium from PR. Cadmium plant
concentrations decreased with increasing rates of lime.
• Lettuce cadmium concentrations and cadmium soil-lettuce Tc's were lower in the
greenhouse study than in the field study.

Arsenic
• Application of arsenic to a Sultan silt loam up to 40 times the Washington fertilizer
standard had no effect on fresh or dry weight lettuce yield, i.e. phytotoxicity was not
observed at any application rate.
• Lettuce arsenic concentrations increased with increasing application rates of Ironite.
• Less arsenic accumulated in soil than would be expected from arsenic soil additions.
• Liming did not affect arsenic uptake into lettuce.
• Extractability of soil arsenic increased with increasing lime rates.
• Soil extractable arsenic was similar between Ironite and soluble metal salts at an
application rate 40 times that allowed by the Washington State standards.
• The average Tc was small (0.012); indicating a low potential for plant uptake of
arsenic.

Lead
• Application of lead to a Sultan silt loam up to 40 times the Washington fertilizer
standard had no effect on fresh or dry weight lettuce yield, i.e. phytotoxicity was not
observed at any application rate.
• Lettuce lead concentrations increased with increasing application rates of granular
zinc.
• Liming did not affect lead uptake by lettuce.
• Soil extractable lead was lower in soil treated with granular zinc than with soluble
lead salt.
• The average Tc was small (0.005); indicating a low potential for plant uptake of lead.

10
Main Conclusions: (Part 1 continued, Greenhouse study, Puyallup Site)

• Application of metals to a Sultan silt loam up to 40 times the Washington fertilizer

standard had no effect on fresh or dry weight lettuce yield, i.e. phytotoxicity was not
observed at any application rate.
• The amount of metals taken up by lettuce depends on whether the metal was added to
a Sultan silt loam soil as fertilizer or as soluble metal salt. Cadmium is more
available to lettuce in the form of soluble metal salts than from the fertilizers.
• Lettuce accumulated more cadmium than arsenic or lead.

Part 2: Plant Yield and Uptake of Heavy Metals from Fertilizers and Soluble Metal
Salts in Field-Grown Lettuce and Cucumbers- Puyallup Site

Purpose:

The purpose of this study was to determine the amount of metals taken up by lettuce and
cucumbers grown in the field on a Sultan silt-loam with varying application rates of three
fertilizers (TSP, PR, and Ironite) and with varying application rates of cadmium and
arsenic in soluble metal salts. Ironite and soluble metal salts were applied to the field up
to 16 times the Washington fertilizer standard for arsenic and cadmium additions to soil
(WAC 16-200-7064). TSP and PR were applied up to 16 times their maximum product
application rate as defined by in WAC 16-200-7063.

The study evaluated whether uptake of arsenic and cadmium by lettuce and cucumbers
differs between fertilizers and soluble metal salts and whether uptake of metals changed
over the 2-year study period. Concentrations of total and extractable metals in soils were
measured to determine which most closely correlated with plant metal concentrations.
Soil metal (total and extractable) and plant metal concentrations were used to develop
Tc's. The study also evaluated the effect of liming on plant uptake of arsenic and
cadmium from fertilizers and soluble cadmium and arsenic salts. The study also
measured the effect of different application rates of fertilizers and liming on fresh and dry
weight lettuce yields. The amount of the metals removed from the soil due to plant
uptake was calculated to determine the amount of the metals that remains in the soil after
crop harvest to address possible metal accumulation in soil over time. The result for the
field study was compared with that of the greenhouse study.

Methods:

• Crop: lettuce and cucumber

• Soil: Sultan silt loam
• Application rates: 1X, 2X, 4X and 8X for the 1st year; 2X, 4X, 8X and 16X
cumulative rates for the 2nd year. 1X rate for Ironite = maximum arsenic addition per
Washington fertilizer standard (WAC 16-200-7064), 1X for soluble metal salts =
maximum arsenic and =cadmium addition per WA fertilizer standard (WAC 16-200-
7064), 1X rate for TSP and PR = maximum recommended screening product

11
 application per WAC 16-200-7063. Product application rates at 1X rate: 430 kg
TSP/ha, 645 kg western PR/ha, 100 kg Ironite/ha. Metals addition at 1X rate for
soluble metal salts: 0.09 kg cadmium/ha, 0.33 kg arsenic/ha.
• Fertilizers and soluble metal salts applied and metals measured: TSP (cadmium), PR
(cadmium), Ironite (arsenic and cadmium), and soluble metal salts (arsenic and
cadmium).

Main Results:

Cadmium
• The concentrations of cadmium in lettuce were higher than in cucumber (i.e. lettuce
took up more cadmium from fertilizers and soluble salt added to a Sultan silt loam
than did cucumber at the same rate of cadmium). The concentration of cadmium was
higher in the cucumber vine than in the fruit.
• In the first year of the study, the availability of cadmium to lettuce increased in the
following order: PR < TSP < soluble cadmium salt. During the second year of the
study, the availability of cadmium from soluble salt remained high. In the second
year, however, the availability of cadmium to lettuce from TSP decreased while the
availability of cadmium from PR increased.
• Soil cadmium extractability increased in the following order: PR < TSP < soluble
salt.
• The concentration of cadmium in lettuce correlated better with the level of cadmium
extracted from soil than with the total level of cadmium in the soil. Conversely, the
concentration of cadmium in cucumber correlated better with the total level of
cadmium in the soil.
• The two-year average Tc, for cadmium in lettuce, based on the total level of cadmium
in the soil was 16.9. Although, the Tc value of 16.9 appears high, this estimate is not
a reflection of commercial fertilizers only as this number also includes the Tc's for
commercial fertilizers and soluble metal salts. The average Tc for only fertilizers was
6.2. The average Tc for cadmium for cucumber fruit was 0.68.
• Generally, liming caused a decrease in the extractability of cadmium from soil.
Liming decreased the uptake of cadmium from fertilizers by lettuce and had no effect
on the uptake of cadmium by cucumber.
• Application of fertilizers within the rates used did not decrease crop yields. Lettuce
yield increased with increasing TSP rates.
• Lettuce and cucumber accumulated less than 3% of the total cadmium added to soil.
• Tc's for cadmium in lettuce and lettuce cadmium concentrations were higher for field
than for greenhouse for the phosphate fertilizers.

Arsenic
• Lettuce and cucumber accumulated less than 0.001%∗ of the total arsenic added to
soil. The lower accumulation of arsenic compared to cadmium may be attributed to
differential sorption mechanisms of both cucumber and lettuce for these two
elements.

∗
Originally reported incorrectly as 0.1%.

12
• There was no difference between extractable soil arsenic and total soil arsenic for
predicting the concentration of arsenic in lettuce.
• The Tc averaged 0.021 for lettuce and 0.011 for cucumber fruit.

Main Conclusions:

• There were significant year and source effects, as well as lime effects, in regard to the
concentrations of cadmium found in lettuce. Due to differences in plant uptake of
cadmium from different fertilizers added to a Sultan silt loam, extrapolation of plant
uptake from one fertilizer source to other fertilizer sources should not be done
without qualification.
• Soil-to-lettuce Tc's based on extractable soil cadmium concentrations are more
responsive to factors that influence plant uptake of cadmium. Consequently, this Tc
provides a more accurate prediction of potential food chain exposures in human
health risk assessments than Tc's based on the total level of cadmium in the soil.
• The Tc's for cadmium in lettuce or cucumber are higher than those used in EPA’s
biosolids (leafy = 0.6, fruits = 0.11) and fertilizer (lettuce 0.05-1.56) risk assessments.
• The recovery of cadmium in lettuce from any of the added cadmium sources was very
low (3%) from a mass balance point of view. This indicates that increased cadmium
accumulation in soil over time should be expected unless the concentration of
cadmium in the fertilizers is reduced.
• The recovery of arsenic in plants is low (< 0.1%). Consequently, one would predict
an accumulation of arsenic in soil over time.

Part 3: Uptake of Heavy Metals from Fertilizer and a Soluble Metal Salt in Field-
Grown Potatoes and Winter Wheat- Prosser Site

Purpose:

The purpose of this study was to determine the amount of metals taken up by potatoes
and winter wheat grown in the field in a Quincy-Hezel fine sandy loam with varying
application rates of five fertilizers. The five fertilizers used were TSP, PR, DAP from
Idaho (Western DAP), DAP from Florida (Eastern DAP), and zinc sulfate. TSP and DAP
are derived from PR. The phosphate fertilizers manufactured from Florida rock
phosphate deposits are typically low in cadmium (3 to 15 ppm). Conversely, fertilizers
from western phosphate rock contain cadmium levels several times higher than that of PR
deposits in Florida. Washington soils have a greater potential for cadmium accumulation
because western PR has become the material currently utilized in the production of most
commercial phosphate fertilizers used in Washington State. Fertilizers were applied up
to 16 times their maximum product application rate as defined in WAC 16-200-7063.
The crop uptake of cadmium from a soluble cadmium salt was included for comparison.
The study also evaluated if there were differences in cadmium concentrations between
peeled and unpeeled potatoes at selected application rates. The quantity of extractable
metals in soils was measured for each application rate for each fertilizer. The study also
measured the effect of different application rates of fertilizers on crop yields. In addition,

13
this study also determined the concentration of extractable metal in soil for each fertilizer
to a depth of 0-15 and 15-30 cm. The amount of metal removed from the soil due to
plant uptake was calculated to determine the amount of metal remaining in the soil after
crop harvest to address possible metal accumulation in soil over time. This study was
conducted over a 2-year period. Lime treatments were not included in this study.

Methods:

• Crop: winter wheat and potato (tuber only)

• Soil: Quincy-Hezel fine sandy loam
• Application rates: 1X, 2X, 4X and 8X for the 1st year; 2X, 4X, 8X and 16X
accumulative for the 2nd year for TSP, PR and zinc sulfate. 1X, 2X 1st year; 2X and
4X 2nd year for W DAP and E DAP. 1X for all fertilizers = maximum recommended
product screening application per WAC 16-200-7063. 1X for soluble cadmium salt
(cadmium chloride) = maximum cadmium addition per Washington fertilizer standard
(WAC 16-200-7064). Product application rates at 1X: 435 kg TSP/ha, 653 kg PR/ha,
426 kg W DAP/ha, 426 kg E DAP/ha, and 48.3 kg zinc sulfate/ha. Metals addition at
1X for soluble metal salts: 0.09 kg cadmium/ha.
• Fertilizers and soluble metal salts applied and metals measured: TSP, PR, DAP (W
DAP, from Idaho), DAP (E DAP, from Florida), and zinc sulfate. Products, soils at
all treatments and crops at all treatments were analyzed for both cadmium and lead.
DTPA extractable cadmium and lead in soils were measured at 0-15 and 15-30 cm
depths. Total metals in soils were only measured at the beginning of the study and
were not measured for treatment groups.

Main Results:

Cadmium
• Wheat stover had higher cadmium concentrations than wheat grain. Stover is the
dried stalks and leaves of a cereal crop, used as bedding or incorporated into the soil
as residue after the grain has been harvested.
• TSP added the most cadmium to the Quincy-Hezel fine sandy loam of any fertilizer in
this study. Cadmium potato tuber concentrations increased at the higher TSP
application rate of 8 and 16X in the 2nd year.
• In the 1st year of the study (1999), the average cadmium concentration in peeled
potato tuber was significantly lower than in unpeeled tubers. For those treatments
analyzed in the 2nd year of the study (2000), the average cadmium concentration, in
unpeeled and peeled tuber was not significantly different. Additional analysis of the
2000 data is being conducted to verify this inconclusive result.
• For all treatments, the cadmium concentrations in wheat grain and potato tuber
correlated with the levels of cadmium extracted from soil to a depth of 0-15 cm. This
indicates that soil extractability is good predictor of cadmium uptake in these crops.
• In 1999, there was a lower grain yield with PR compared to other phosphate
fertilizers. The wheat grain yield in 2000, however, was not significantly affected by
fertilizer source or rate. Potato tuber yield was unaffected by fertilizer source or rate
for both 1st and 2nd years of study.

14
• It was estimated that 1% of total soil cadmium was recovered in wheat, indicating
that most cadmium remains in the soil.
• The plot design in this study did not allow the differences in cadmium availability to
be sorted out between years. Since crops were rotated between plots; the year effect
and plot effect could not be separated from the variability in yearly uptake.
Increasing the application rate in the second year increased uptake of Cd in both
crops.
• The Tc's obtained at Prosser were much smaller than those obtained in Puyallup.
This could be explained by differences in the crops and soils used at both locations.
• The Tc's based on total soil cadmium in Prosser were as follows: The Tc for wheat
grain in 1999 was 0.34. The Tc for wheat grain in 2000 was 0.24. The Tc for potato
tubers in 1999 was 0.35. The Tc for potato tubers in 2000 was 0.5.

Lead
• The rate and type of fertilizer had little effect on the level of lead found in grain or
potato tuber. Concentrations in both grain and tubers were very low.

Main Conclusions from Prosser study

• The Quincy-Hezel fine sandy loam data indicates that even with rototilling and
digging of potatoes, most of the cadmium remained in the top 0-15 cm soil layer.
• As expected, most of the cadmium uptake was related to the upper part of the soil
profile where soil cadmium levels had increased.

Part 4: Laboratory Analysis of the Availability of Heavy Metals in Soils

Supplemented with Fertilizers and Soluble Metal Salts- Pullman Site

Laboratory experiments determined the capacity of the two soils used in the field studies
to retain zinc, lead, and cadmium from soluble metal salts of these metals. The Sultan silt
loam had a higher capacity for all metals than the Quincy-Hezel fine sandy loam, but the
retained metals were easier to extract by dilute salt solutions, water, and DTPA from the
Sultan soil. The difference in extractability of metals was also determined by comparing
the incremental addition of fertilizers to both soils versus an “all-at-once” addition of
each soluble metal salt to both soils. The DTPA extractability for cadmium even after 45
additions was similar to cadmium added all at one time. For both lead and zinc, the
metals were much less extractable after 30 and 45 additions than the one-time
applications. This observation suggests that metal addition over time in a cropping
situation could render lead and zinc less available than a single application of the same
amount of metal.

Soil samples taken from the field sites in Puyallup (Sultan) and Prosser (Quincy-Hezel)
were also amended with fertilizers in the laboratory. Soils fortified with fertilizers up to
50X an estimated agronomic rate, were incubated at field moisture content for up to one
year. Over the first 30 weeks of incubation, the DTPA-extractability of metal was erratic
with time. In the following 20 weeks, there was a general decrease in extractability with

15
time, but many samples did not reach a steady state. These results imply that the
availability of lead, zinc, and cadmium to crops decrease over time.
IV. Implications

A. Accumulation of Metals in Soil Over Time

The WSU study was conducted to determine, in part, the accumulation of arsenic, lead,
and cadmium in soils and crops that may occur from continual agricultural applications
of fertilizers. The predictions and comparisons presented in the following sections focus
on cadmium. The results of the study indicate that cadmium is of greater concern than
arsenic and lead in terms of soil accumulation and crop uptake. The WSU study also
shows, that for the conditions tested, the amount of cadmium taken up by crops is greatly
influenced by the source of the cadmium. Although crops take up some amount of
cadmium from the soil, the WSU study indicate that cadmium could build up in soil over
time due to continual fertilizer use. This is more likely to occur from using fertilizers that
contain higher cadmium levels. Western PR which is used to produce a majority of the
phosphate fertilizers used in Washington agriculture, has considerably higher levels of
cadmium that eastern PR. Increases in soil cadmium concentrations are of concern since
they may lead to increases in cadmium concentrations in crops.

Arsenic and lead levels in most agricultural fertilizers are very low. Occasionally,
relatively high levels of lead can be found in some micronutrient fertilizers, but the
application rate of those products is so low that it does not significantly contribute to lead
soil levels. In addition, compared to cadmium, the transfer of arsenic and lead from soil
to crop is minimal.

Although the study indicates cadmium may increase over time due to the application of
cadmium-containing fertilizer, the prediction of future concentrations of cadmium in the
soils depends on the annual input of cadmium and how much cadmium is removed via
the harvested plant part. Because not all soils have the same initial cadmium
concentration or contribute the same amount of cadmium to crops, the rate of increase in
the total level of cadmium in soil from the applications of fertilizers will vary with soils.
The number of years it will take to double the concentration of cadmium initially in the
soil from continuous applications of the fertilizers is estimated with the following
assumptions:

1. The same crops are grown and the yields of the crops are the same throughout
the entire period of this prediction;
2. Plant materials other than the top growth of the lettuce, cucumber fruit, potato
tuber, and wheat grain are left in the same field;
3. Tillage depth is less than15 cm;
4. No loss of cadmium occurs other than that removed in the harvested plant
material and there is no addition of cadmium into the soils happens other than
the amount of cadmium added from the phosphate fertilizers;
5. The same soil conditions and same phosphorus fertilizers are used; and

16
 6. The soil-to-plant Tc's calculated based on the total level of cadmium in the
soil for each crop are the same throughout the entire period of this prediction
and are considered satisfactory as a measure of the transfer of soil cadmium to
the plants.

In most instances, the above assumptions would result in an estimate that is higher than
the actual levels of metal in soils. This is because monocropping systems are not what
always occurs in Washington. For instance, potatoes require the highest levels of
phosphate fertilizers and are normally grown on a three-year rotation. Because the total
level of cadmium in soils was not determined for the field experiment in eastern
Washington that involved potato and winter wheat, it was estimated from the amount of
cadmium applied.

The following section provides an estimate of the time it may take for the soil to double
the level of cadmium for each of the four crops based on average fertilizer application
rates. This estimate is referred to as the doubling time. The doubling time is based on the
concentrations of cadmium in soil, in fertilizer, as well as the amount of cadmium
removed in the harvested plant parts. The average fertilizer rates are rough estimates.
They were determined by consulting a variety of experts that included WSU personnel
and fertilizer distributors. Growers are strongly encouraged to perform an annual soil test
for each field and modify their use of fertilizer accordingly.

Doubling Time of Cadmium in Soil Cropped to Leaf Lettuce:

Initially, the Sultan soil contained 0.44 ppm cadmium. The estimated number of years
for total cadmium in this soil to double its initial level at the average application rate (248
kg/ha for TSP and 339 kg/ha for PR) will be 38 years for the TSP and more than 100
years for the PR.

Doubling Time of Cadmium in Soil Cropped to Cucumber:

At the average application rate, the total cadmium present in the Sultan soil will double
from its initial value in about 38 years for TSP and about 98 years for the PR.

Doubling Time of Cadmium in Soil Cropped to Potato:

The initial cadmium concentration for the Quincy fine sandy loam was 0.1 ppm. The
number of years for the Quincy fine sandy loam to double its initial total cadmium
concentration will be 17 and 36 years for the TSP and PR, respectively, at their average
application rates (373 kg/ha for TSP and 509 kg/ha for PR).

Doubling Time of Cadmium in Soil Cropped to Winter Wheat:

The estimated number of years for the Quincy fine sandy loam to double its initial
cadmium concentration will be 46 and more than 100 years for the TSP and PR,

17
respectively, if applied at their average application rates (50 kg/ha for TSP and 68 kg/ha
for PR).

The observations in the WSU study are consistent with the findings of a report entitled,
"Final Report, Screening Survey for Metal and Dioxins in Fertilizer Products and Soils in
Washington State, April 1999, Washington State Department of Ecology Publication No.
99-309". The findings in Ecology Publication No. 99-309, include:
• Agricultural practices over the last 50 years may have increased cadmium
concentrations to a small degree over background levels in the Columbia Basin,
although further investigation is necessary to confirm this.
• The increased cadmium soil concentrations in the agricultural fields sampled suggest
no potential soil quality impairment because the values detected are within the lower
range of background comparison studies.
• Increased agricultural cadmium levels over background levels indicate a need to
periodically monitor cadmium concentrations in soils to determine if any rate of
increase is occurring and to ensure levels do not become a concern in the future.

B. Plant Uptake of Metals from Soil Treated with Fertilizer

There are many factors that influence the uptake of metals into plants. They include soil
pH, source of metals, soil type, existing or initial soil metal concentrations, and whether
all harvested plant material is removed before tilling and subsequent planting. Due to the
many uncertainties associated with predicting the uptake of metals into plants and the
limited number of conditions evaluated in this study, projections of metal concentrations
in plants over time cannot be estimated with a high degree of certainty. However, there
are some comparisons that can be made with the data generated from the present study
with other recent studies.

C. Relevance to Other Studies

Since the time the WSU study was initiated, three fertilizer risk assessments have been
completed (California Department of Food & Agriculture, 1998; US EPA, 1999; and The
Fertilizer Institute, 2000). Each of these assessments evaluated possible human
exposures, primarily to metals, in fertilizers.

Risk assessment is a process used to evaluate potential health risks associated with
contaminants in the environment. Human health risk assessments estimate how much
people may be exposed to the contaminants of interest via different possible pathways,
such as breathing dust or drinking water containing contaminants. Exposure estimates
are combined with information on the toxicity of the contaminants to calculate health
risks. Risk assessment can be used to develop risk-based environmental standards, set
environmental cleanup levels at hazardous waste sites, and provide general or
comparative information on health risks associated with contaminants in the
environment.

18
The fertilizer risk assessments evaluated several ways in which people might be exposed
to metals from fertilizers, including estimating human exposures through the food chain
via crops grown using different fertilizers. A significant component of this analysis
includes estimating crop uptake of metals using Tc's. Tc is the ratio of the concentration
of metal in a plant to the concentration of metal in soil. A higher Tc indicates a greater
transfer of metal from the soil into the plant. The three risk assessments that have
evaluated health risks associated with fertilizer use are briefly summarized.

In the following section, the Tc's used in each of these risk assessments are compared to
Tc's derived in the WSU study. Plant uptake of cadmium is of most concern because
plants take it up to a greater extent than arsenic or lead. Therefore, the Tc values
discussed in the following section are primarily for cadmium. Tc's are reported on a dry
weight basis.

(1) California Department of Food and Agriculture, 1998. Development of Risk-

Based Concentrations for Arsenic, Cadmium, and Lead in Inorganic Commercial
Fertilizers.

California’s risk assessment evaluated possible human exposures to arsenic, cadmium

and lead from fertilizer use. Three pathways of human exposure were evaluated in the
final tier of the risk assessment: ingestion of fertilizer-amended soil, dermal contact with
fertilizer-amended soil, and ingestion of plant crops grown using fertilizers. The risk
assessment indicated that the crop ingestion pathway contributed the most of any
pathway to total risk. The results of this risk assessment were used to generate allowable
levels of the three metals in fertilizers on a per nutrient basis for phosphate and
micronutrient fertilizers. California has finalized standards for arsenic, cadmium and
lead in fertilizer based, in part, on the results of this risk assessment.

California's risk assessment evaluated soil-to-crop uptake for three groups of crops: root
crops (which includes potatoes), vegetable crops (which includes lettuce and cucumbers)
and grains (which includes wheat). A range of Tc's were used for predicting crop
uptake for each of these groups for each metal. For the California assessment, the Tc's
were based on available Tc's in the literature compiled for this risk assessment and
included data from fertilizer and non-fertilizer sources of metals and data primarily from
greenhouse pot-studies and some field studies.

The Tc for cadmium for lettuce based on total soil cadmium in the WSU field study
averaged 6.2 for two phosphate fertilizers. This is within the range of Tc’s used for the
vegetable group (including lettuce) in California's risk assessment. The Tc increased to
16.9 with the inclusion of soluble cadmium salt, which implies that the source of
cadmium is also a major factor affecting the Tc for cadmium. The Tc for cadmium for
cucumber averaged 0.68 in the WSU study, which is within the range of Tc’s for the
vegetable group in the California risk assessment.

It is not apparent how using the Tc's derived from the WSU study would impact the final
results of California's risk assessment. This is because California's risk assessment relies

19
on using ranges of Tc's as well as ranges of other exposure parameters (such as crop
ingestion rates and bodyweights) in calculating risks in a process called probabilistic risk
assessment. Therefore, including a Tc value of 16.9 into the risk assessment does not
necessarily mean that the final calculated risk will be higher, and there is no easy way to
predict if including this Tc value would have any appreciable effect on the risk
calculation.

(2) U.S. EPA, 1999. Estimating Risk from Contaminants Contained in Agricultural
Fertilizers.

EPA’s fertilizer risk assessment evaluated human exposures to nine metals, including
arsenic, cadmium and lead, as well as dioxins. Five exposure pathways were evaluated:
direct ingestion of fertilizer products during application, ingestion of fertilizer-amended
soil, inhalation of particles and vapors associated with fertilizer application, ingestion of
plant crops and animal products produced on soil amended with fertilizers, and ingestion
of fish from streams located adjacent to fertilizer-amended fields. As with the California
risk assessment, exposures through the food chain were found to be one of the main
contributors to the total (albeit small) risks. However, the EPA concluded from their risk
assessment that for most fertilizers "hazardous constituents in fertilizers generally do not
pose harm to human health or the environment."

EPA's risk assessment evaluated five categories of crops: grains, forage, fruit, herbage,
and roots. EPA also collected and compiled a database of Tc values specifically for their
risk assessment. The Tc database used was based on field data only for most metals and
included data from fertilizer and non-fertilizer sources. A range of Tc's were used for
predicting crop uptake for each metal for each category of crops. The Tc range used for
cadmium in lettuce is 0.05-1.56. This range is lower than the average lettuce Tc for the
WSU study of 16.9 (based on total soil metals) for all sources and 6.2 for fertilizer
sources only. Because EPA calculated risks using a probabilistic approach, as did the
California risk assessment, it is unclear what impact including a Tc value of 16.9 or 6.2
would have on the final risk results.

(3) The Fertilizer Institute (TFI), 2000. Health Risk Evaluation of Select Metals in
Inorganic Fertilizers Post Application.

The TFI risk assessment evaluated possible human exposures to twelve metals associated
with fertilizer use, including arsenic, cadmium and lead, and radium 226, a radionuclide.
Three routes of human exposures were evaluated: ingestion of fertilizer-amended soil,
dermal contact with fertilizer-amended soil, and ingestion of crops grown with fertilizers.
Again, the crop ingestion pathway contributed the greatest amount of risk to the total risk
levels. The risk assessment calculated allowable levels of metals in fertilizers based on
acceptable risk levels, similar to the California risk assessment. The Association of
American Plant Food Control Officials, made up of state fertilizer regulators, has
tentatively adopted the allowable metals levels developed in TFI’s risk assessment as
national fertilizer guidelines.

20
TFI's risk assessment evaluated three groups of crops: vegetable, root, and grain crops.
Tc values for the crop groups evaluated in the risk assessment were derived from studies
using fertilizer and non-inorganic fertilizer sources (such as biosolids) and include data
from field studies and a limited number of greenhouse(pot)-studies. The Tc values used
in TFI's risk assessment for cadmium are 1.7, 0.93 and 0.12 (dry weight) for vegetable,
root and grain, respectively. These Tc values are the 90% upper confidence limit of the
geometric mean of the Tc values determined for each metal and each crop group. These
values are lower than the average Tc of 6.2 for cadmium in lettuce across the other
fertilizer treatments. The impact of adding these values to the Tc database used in the
TFI risk assessment on the final allowable metals levels calculated for fertilizers cannot
be easily calculated.

(4) Use of the Tc’s from WSU Study

Although the Tc's developed in the study cannot be directly applied to the existing risk
assessments, they contribute to the database of Tc's that can be used in human health risk
assessments. The Tc's developed in this study more accurately reflect potential crop
uptake of metals from fertilizers since they are based on fertilizer data and were derived
from field data.

The WSU study evaluated the uptake of metals into four crops (lettuce, cucumber, wheat,
and potato) grown using a limited number of fertilizers on two different soil types.
People generally eat a variety of vegetables grown in many different locations either from
commercial farms or from home gardens. Therefore, it would not be realistic to predict
how much a person would be exposed to if their only source of vegetables came from the
locations and conditions tested in this study.

D. Comparison of Concentrations of Metals in Crops to FDA's Market Basket

Survey Results

The U.S. Food and Drug Administration (FDA) conducts routine sampling of food for
various pesticides, contaminants, and nutrients as part of their Market Basket Study. The
main purpose of the Market Basked Study is to estimate intakes of these substances in
representative diets for the U.S. population. Testing of different crops for arsenic,
cadmium and lead is included in the Market Basket Study and these data can be used to
compare to the results of the WSU study.

In general, the concentrations of cadmium measured in the four crop of the WSU study
fall within the ranges of cadmium concentrations reported in FDA Market Basket data.
However, some cadmium concentrations in lettuce in the WSU study at the higher
fertilizer application rates are higher than values reported in the FDA data. This is
expected since the higher fertilizer application rates of the WSU study are much greater
than the normal fertilizer application rates used on a per year basis. The concentrations
of arsenic and lead measured in crops in the WSU study are all within concentration
ranges reported in FDA data.

21
V. Conclusions

The WSU study was conducted to determine, in part, if the accumulation of arsenic, lead,
and cadmium in soils and crops may occur from continual agricultural applications of
fertilizers. The results of the study indicate that cadmium is of greater concern than
arsenic and lead in terms of soil accumulation and crop uptake. The WSU study also
shows, that for the conditions tested, the amount of cadmium taken up by crops is greatly
influenced by the source of the cadmium. Although crops take up some amount of
cadmium from the soil, the WSU study indicate that cadmium could build up in soil over
time due to continual fertilizer use. This is more likely to occur from using fertilizers that
contain higher cadmium levels. Western PR which is used to produce a majority of the
phosphate fertilizers used in Washington agriculture, has considerably higher levels of
cadmium that eastern PR. Increases in soil cadmium concentrations are of concern since
they may lead to increases in cadmium concentrations in crops.

Arsenic and lead levels in most agricultural fertilizers are very low. Occasionally,
relatively high levels of lead can be found in some micronutrient fertilizers, but the
application rate of those products is so low that it does not significantly contribute to lead
soil levels. In addition, compared to cadmium, the transfer of arsenic and lead from soil
to crop is minimal. Other pathways are important for arsenic and lead, therefore, the state
standards for these metals must remain in effect.

The relatively high availability of naturally occurring cadmium and cadmium from the
phosphate fertilizers used in the study led to relatively high accumulation of cadmium in
the lettuce and wheat crops grown, respectively, in the Sultan and Quincy-Hezel soil
types. While the results indicate a need to monitor soil cadmium levels to prevent excess
soil cadmium accumulation, the extrapolation of the results to other soil types cannot be
made with any degree of certainty. This is because soils vary considerably in their ability
to bind cadmium. Additionally, there may be major differences in cadmium availability
between soil types even when they contain the same amount of cadmium. The study
would need to be extended to other soil types before it could be applicable to a broad
range of soil conditions.

There are other important issues that have to be resolved in order to improve the accuracy
of predicting the impact of continuous inputs of cadmium to soil from the application of
the phosphate fertilizers. First, a soil-to-plant Tc is needed that is independent of time.
The availability of cadmium from the TSP decreased in the 2nd year of the study, whereas
the availability of cadmium from the PR increased. This indicates that within the two-
year period of this investigation, the Tc's changed with time. An extension of this
research for two more years would be very useful in determining a Tc value after enough
time has elapsed to allow the majority of the cadmium from fertilizer to be well
equilibrated with the soils. Secondly, the model developed based on the first two years of

22
the field research needs to be validated. Confidence in making long-term predictions
regarding soil loading and plant uptake will be strengthened by such validation and a two
year extension of this research would make that possible.

The current Washington metal standards established by the legislature in 1998, which
limits the input of cadmium and other metals from fertilizers or other soil amendments to
agricultural soils, should be sustained until after Tc values for crops have been further
refined. If the input of cadmium to the soil far exceeds what is removed by the crop then
future increases in the level of cadmium in Washington State soils are inevitable. Given
the high availability of this metal to the vegetable and grain crops in the soils, allowing
the level of cadmium in Washington State soils to rise much higher than their background
level is not justified.

VI. Recommendations
The existing Washington State standards that govern the levels of heavy metals in
fertilizers appear to be sufficiently protective at this time. Future levels of heavy metals
in soils, especially cadmium, may be of concern due to the recent use of phosphate
fertilizers with relatively high levels of cadmium. Cadmium levels in soils should be
periodically monitored to ensure that levels do not become a concern in the future.
Additional studies are necessary to refine the Tc values derived from this study.
Additional data are also required to determine the heavy metal loading in other
Washington soils. Finally, the existing model needs to be validated with a second two-
year study.

The study indicates the existing Washington State standards for arsenic, cadmium, and
lead in fertilizer appear to be sufficiently protective at this time. However, the study also
indicates there is a potential for the levels of these metals to increase in soil over time.
This is especially true of cadmium, due to current applications of phosphate fertilizers
containing relatively high levels of cadmium. The agencies involved in the development
of this report make the following recommendations:

1. WSU should extend its study for a minimum of two years and focus on the
development of a soil-to-plant Tc for cadmium that is independent of time. This
would greatly improve the accuracy of predicting crop uptake of cadmium.

2. A long-term program should be developed to monitor soil levels of arsenic,

cadmium, and lead over time to ensure that soil levels do not become a concern
in the future.

3. A process should be initiated to attempt to develop a regional approach dealing

with the issue of metals standards and labeling. This should initially include
Washington, Oregon and Idaho, but may be expanded to include other western
states.

23