Trends in the Potential for Environmental Risk:

Pesticide Loss from Farm Fields

Robert L. Kellogg, Natural Resources Conservation Service, USDA
Richard Nehring, Economic Research Service, USDA
Art Grube, Office of Pesticide Programs, EPA
Steve Plotkin, Natural Resources Conservation Service, USDA
Don W. Goss, Texas Agricultural Experiment Station, Temple, Texas
Susan Wallace, Natural Resources Conservation Service, USDA

Poster Presentation for "The State of North America's Private Land,"
a Conference Held January 19-21, 1999, Chicago, Illinois


Environmental indicators of the potential for pesticide leaching and runoff from farm fields were constructed to provide policy makers and national program managers with information to help assess whether existing programs and policies have been effective in reducing externalities from agricultural activities.

Twelve pesticide-related time series were estimated for 1972 through 1997 for watersheds at the 6-digit Hydrologic Unit level:

  • Pounds of pesticides applied
  • Leaching mass loss
  • Dissolved runoff mass loss
  • Adsorbed runoff mass loss
  • Pesticide leaching risk index for protection of drinking water
  • Pesticide runoff risk index for protection of drinking water
  • Pesticide leaching risk index for protection of fish
  • Pesticide runoff risk index for protection of fish
  • Pesticide leaching risk index for protection of algae
  • Pesticide runoff risk index for protection of algae
  • Pesticide leaching risk index for protection of crustaceans
  • Pesticide runoff risk index for protection of crustaceans


A model was constructed to estimate pesticide loss from farm fields using the following data and information:

  • The National Pesticide Loss Database.
  • Annual estimates of pesticide use by crop and state from the Doane Countrywide Farm Panel Survey and USDA pesticide use surveys.
  • NASS annual county estimates of acres planted.
  • Soil distribution from the National Resources Inventory (NRI).
  • Irrigated acreage from the National Resources Inventory.
  • Water quality thresholds corresponding to drinking water standards (or equivalent derived from mammalian chronic toxicity data) and the maximum safe levels for chronic exposure of fish, algae, and crustaceans to pesticides.

The National Pesticide Loss Database was used as the basis for all but the time series for pounds applied. The National Pesticide Loss Database provides estimates of the 95th percentile of annual mass loss and concentrations at the bottom of the root zone and the edge of the field for 120 soils, 55 climate stations, and 243 pesticides. Separate estimates are available for irrigated and nonirrigated conditions. Pesticide leaching and runoff losses were estimated using the pesticide fate and transport model GLEAMS. GLEAMS is a process model that uses as inputs soil parameters, field characteristics, management practices, pesticide properties, and climate to estimate pesticide leaching and runoff losses. (For additional details on the National Pesticide Loss Database, see "The National Pesticide Loss Database: A Tool for Management of Large Watersheds.")

The analytical framework consists of about 5,000 resource polygons representing the intersection of 48 states, 280 watersheds at the 6-digit Hydrologic Unit level, and 1,400 combinations of climate and soil groups. Seven crops are currently included in the analysis: corn, soybeans, wheat, cotton, sorghum, barley, and rice. A total of 94 pesticides were included in estimates for 1972-86, and up to 192 pesticides were included in estimates for 1987-97. The framework was constructed using the NRI after imputing the pesticide loss data onto NRI sample points by crop, soil group, and climate group. Concentration and mass loss estimates in the National Pesticide Loss Database are relative to application rates so that, when multiplied by application rates specific to crops and states, the final estimate of the annual concentration or mass loss is obtained. Estimates of pounds of pesticides applied, mass loss, and annual concentrations were obtained for pesticides used on the 7 crops in each of the 5,000 resource polygons for each of the 26 years.

On average, 75 percent of the cultivated cropland in the 48 states was incorporated into the simulation. Map 1 and map 2 show the spatial distribution of the approximate coverage for 1982 and 1992.

(Note on the map legends: In the process of converting maps to the World Wide Web the legends on certain maps became unreadable due to their small size. As a temporary solution the legends are presented in a separate file until we can update the maps. Simply look up the map number to find the appropriate legend.)

Environmental risk was assessed using Threshold Exceedence Units (TEUs), which measure the extent that the annual concentration exceeds a water quality threshold. For each pesticide used in each of the 5,000 resource polygons, the threshold-concentration ratio was calculated. Where the threshold concentration was exceeded, the ratio was multiplied by the acres treated to obtain estimates of Threshold Exceedence Units (TEUs). Acres treated were estimated for each resource polygon using acres-planted data and state-level estimates of the percentage of acres treated derived from the Doane and NASS survey databases. Thus, TEUs account for the amount the threshold was exceeded and the amount of acres treated in an area. TEUs were derived in this manner specifically for the purpose of comparing risk from multiple pesticides over space and time. They are a relative measure of risk (the higher the TEU score, the higher the risk), but do not measure absolute levels of risk. (TEUs are similar in concept to the acre-feet volumetric measure, since they are a multiple of acres times a measure of magnitude at a point.) TEUs per watershed were obtained by summing TEUs over all pesticides used in each resource polygon, then summing over all resource polygons in each watershed. (For information about another application using TEUs, see "An Information Aid for Assessing Possible NRCS Involvement in the State Management Plan Process for Regulation of Pesticides.")

The estimated indexes of potential risk change over space and time because of: 1) changes in acres planted, 2) changes in percent acres treated, and 3) changes in the suite of pesticides used, and 4) changes in application rates of the more mobile and/or persistent pesticides on vulnerable acres. The estimated indexes are based on national-level databases, and so caution should be exercised in making local interpretations. Management factors are not included in the indicators.

The algorithm for the potential risk index is:

TEUt = Sigma Sigma Sigma exceedence per acre treatedt,r,k,p * acres treatedt,r,k,p
  r k p  
exceedence per acre treated = [(RELCONC*APPRATE)/THRESH)-1], negative values discarded.


TEUt = Threshold Exceedence Units for a given year. Separate estimates are made for leaching and for runoff, each with a) thresholds for drinking water, b) thresholds for protection of fish, c) thresholds for protection of algae, and d) thresholds for protection of crustaceans.
RELCONC = Relative concentration associated with the 95th percentile pesticide loss per acre for a specific pesticide on a specific soil group. Separate values are used for leaching and for runoff.
APPRATE = Application rate for a specific pesticide.
THRESH = Threshold concentration below which the pesticide loss concentration is defined to be "safe" for chronic exposure.
ACRES = Acres of crop planted in the resource polygon.
PCTTREAT = Percentage of acres treated with a specific pesticide.
PCTSOIL = Percentage of resource polygon in a specific soil group.
k = 7 irrigated and 7 nonirrigated crops.
p = Pesticides
r = 5,000 resource polygons
t = Years from 1972 through 1997.


The 8 pesticide risk indexes indicate that the Nation's pesticide policies during the last twenty six years have succeeded in reducing overall environmental risk, in spite of slight increases in acres planted and pounds of pesticides applied. Nevertheless, there are still areas of the country where there is no evidence of progress, and areas where risk levels for protection of drinking water, fish, algae and crustaceans remain high. Observations supporting this conclusion are:

  • Acres planted for the 7 crops are about 20% higher in recent years than in 1972 (figure 1).
  • While pounds per acre of pesticides applied were highest in the late 1970s and early 1980s, in recent years the average application rate is close to the 1972 level (about 1.1-1.2 pounds per acre for pesticides used on the 7 crops included in the study) (figure 2). Total pounds of pesticides applied are about 20% higher in recent years (figure 3). The spatial distribution of pounds applied for the 7 crops and changes over the 26 years are shown in maps 3-5.
  • Total mass loss of pesticides from farm fields averaged 5.1% of the amounts applied: 0.5% in leachate, 3.3% dissolved in runoff, and 1.3% adsorbed to soil particles removed by sheet and rill erosion. Mass loss in leachate and dissolved in runoff increased during the 1970s, but then returned to levels at or near the 1972 level for the remainder of the time series. Mass loss adsorbed to soil particles, however, remained high throughout the 1980s and then increased steadily through the 1990s to levels over twice as high as in 1972 ( figure 4). The spatial distribution of pounds applied for the 7 crops and changes over the 26 years are shown in maps 6-8, 9-11, and 12-14.
  • Potential risk to drinking water for pesticides dissolved in runoff fell about 75% over the 26 years; risk for pesticides in leachate fluctuated over time but in recent years is about equal to the 1972 level (figure 5). Nevertheless, potential risk to drinking water showed significant increases in some watersheds in the South for leachate, and in the lower Mississippi, Pennsylvania, and California for runoff (maps 15-17, and 18-20).
  • Potential risk to fish fluctuated since 1972, probably because it is influenced most by insecticide use. In some years the risk level was nearly 4 times the 1972 risk level. In 1997, however, the risk level was similar to the 1972 level for both leachate and runoff (figure 6). Significant increases in the potential risk to fish occurred throughout most of the South, while significant decreases occurred in the Midwest (maps 21-23, and 24-26).
  • Highest potential risk was estimated for algae, which is influenced most by herbicide use. Risk in recent years is about 40% lower than in 1972 (figure 7). The greatest reductions in potential risk occurred in the Midwest, Northern Plains, and the Southeast, while significant increases in potential risk occurred in the Central Plains and the South Central states (maps 27-29, and 30-32).
  • Although potential risk for crustaceans increased dramatically in the 1970s, risk levels in recent years are 40-70% below the 1972 level (figure 8). Potential risk in leachate decreased throughout the Midwest and the Southeast, but showed increases in the South Central states. For runoff, however, substantial increases in potential risk occurred in the Southeast and Mid-Atlantic states and California, while significant decreases occurred in the Midwest and most of the Great Plains states (maps 33-35, and 36-38).

The authors plan to extend the indexes to include additional crops in the future, but including additional crops is unlikely to alter the broad trends shown here because of the small acreage involved.

Appendix: List of Figures and Maps