Downloaded by NORTH CAROLINA STATE UNIV on November 8, 2012 | http://pubs.acs.org Publication Date (Web): November 6, 2012 | doi: 10.1021/bk-2012-1111.ch021
Chapter 21
Using GIS To Assess Pesticide Exposure to Threatened and Endangered Species for Ecological Risk Assessment James L. Cowles,* Kelly McLain, Perry L. Beale, and Kirk V. Cook Washington State Department of Agriculture, 1111 Washington Street SE, Olympia, WA 98504-2560 *E-mail:
[email protected]
With the listing of salmon for protection under the Endangered Species Act, the Washington State Department of Agriculture (WSDA) determined that the traditional environmental data sets for pesticide registration decisions were insufficient to accurately determine potential exposure and subsequent effects of pesticides on salmonids and other listed species in Washington State. WSDA has implemented a program to spatially determine the location and use of pesticides in relationship to salmonid habitat and monitor pesticide residues in salmon-bearing streams in Washington State. Data elements developed include a geographic information system (GIS) incorporating the location of 160 crop types grown in Washington and an estimation of state-specific pesticide use.
Introduction The United States Environmental Protection Agency (EPA) is incorporating consultation under Section 7(a)(2) of the Endangered Species Act (ESA) into Registration Review for pesticides registered in the United States (1). All Federal agencies must consult with either the U.S. Fish and Wildlife Service or National Marine Fisheries Service (NMFS) if they authorize an action that will jeopardize the existence of a listed species or adversely modify a listed species critical habitat (2). Typically ecological risk assessment for pesticide registration is done in a tiered manner where screening level assessments based on conservative exposure scenarios are used to determine if an adverse environmental outcome is not likely © 2012 American Chemical Society In Pesticide Regulation and the Endangered Species Act; Racke, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.
Downloaded by NORTH CAROLINA STATE UNIV on November 8, 2012 | http://pubs.acs.org Publication Date (Web): November 6, 2012 | doi: 10.1021/bk-2012-1111.ch021
(3). It is anticipated that consultation under ESA will necessitate ecological risk assessments conducted for pesticide registration to be more spatially explicit when determining potential exposure to listed species. Recent biological opinions conducted by NMFS have incorporated spatial analysis to determine the co-occurrence of potential pesticide use and salmonid habitat (4–7). The NMFS analysis focused on determining the potential for pesticide use based on the land cover categories found in the 2001 National Land Cover Database (NLCD) (6). However, the 2001 NLCD only has two land cover classifications for agricultural lands: pasture/hay and cultivated crops (8). In a minor crop state such as Washington, two agricultural land cover categories must represent over 200 commodities each with unique pest pressures and pesticide use practices (9). For example the NLCD does not identify the spatial extent of where caneberries, mint or hops are grown nor their relationship to habitat for a listed species. Since Washington’s agricultural lands coincide with threatened and/or endangered species habitat, the Washington State Department of Agriculture (WSDA) has instituted a program to collect state-specific pesticide use data and compile a high resolution land cover dataset of agricultural land for use in ecological risk assessment for pesticide registration. In combination, these datasets allow risk assessors to evaluate the spatial and temporal use of pesticides in relationship to listed species habitat thus reducing uncertainty of exposure estimates and allowing for development of targeted mitigation measures.
Methodology Pesticide Use Information Accurate pesticide use data is invaluable for assessing the potential impacts of pesticides on water resources and ESA listed species. However, comprehensive pesticide use reporting programs such as the one administered by the California Department of Pesticide Registration (10) are rarely implemented. Given the cost and data management infrastructure, and in the absence of stakeholder support needed to implement a program similar to California’s, WSDA has developed and implemented a unique program to develop pesticide use profiles that capture the typical use patterns for pesticides by commodity. Pesticide use data is collected by conducting detailed surveys with farmers, ranchers, land managers, pesticide applicators, and crop consultants for a specific commodity. Typically, surveys are conducted during an on-site interview with a respondent; however, telephone and e-mail correspondence may also be used. To develop the survey WSDA reviews available data for the respective commodity which includes but is not limited to the following: • • •
Pesticide labels Previous pesticide use summaries Washington State University Cooperative Extension recommendations (11) 294 In Pesticide Regulation and the Endangered Species Act; Racke, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.
• • •
Downloaded by NORTH CAROLINA STATE UNIV on November 8, 2012 | http://pubs.acs.org Publication Date (Web): November 6, 2012 | doi: 10.1021/bk-2012-1111.ch021
• •
Pacific Northwest Pest Control Handbooks (12, 13) Washington State University Pesticide Information Center On-Line (PICOL) (14) National Agricultural Statistics Service (NASS) chemical use surveys (15) The Compendium of Washington Agriculture (9) Sales data when available
Information gathered from the data review is incorporated into a survey guide which provides an inclusive pesticide list with abbreviated use summaries that are vetted through the interview. Data collected during the survey interview includes: • • • • • • • •
Beginning and ending application dates Pounds of active ingredient applied per acre per application Number of applications Application interval Percent acres treated Application method Region of application Target pest (optional)
Anecdotal information such as opinions about product availability or trends in use of specific product are noted but are not used in developing a pesticide use profile. WSDA pesticide use data is typically not collected annually. The frequency of data collection for a specific pesticide is determined by agency priorities and limitations in staffing or resources. Generally, WSDA expects to update data by commodity every five years. Recognizing the qualitative nature of the pesticide use surveys, WSDA has also developed a cooperative agreement with NASS to augment their Fruit and Vegetable Chemical Use surveys to include application timing windows for data collected in Washington State. NASS gathers chemical use data based on a Multivariate Probability Proportional to Size design. This sample design accounts for approximately 90 percent of all lands in farms in the United States (16, 17). Rather than typical use data, the NASS surveys are an accounting of the previous year’s pesticide applications in their entirety (product, rate per acre, date of application, application region, percent of acres treated). The farms sampled in each survey are representative of the whole industry and include small, medium, and large acreage operations. The cooperative agreement calls for the final accumulated data to be provided to WSDA by active ingredient and growing region. The reported data includes the mean and median application rates per acre, month and year of application, and the coefficient of variation (CV) of the use estimates. All pesticide use data is compiled in an Access (Microsoft) database for use within a geographic information system (GIS).
295 In Pesticide Regulation and the Endangered Species Act; Racke, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.
Downloaded by NORTH CAROLINA STATE UNIV on November 8, 2012 | http://pubs.acs.org Publication Date (Web): November 6, 2012 | doi: 10.1021/bk-2012-1111.ch021
Land Cover In 2001, WSDA began development of a high resolution land use database to better characterize crop production locations in relation to habitat occupied by federally listed salmonid species. A high resolution land use database is needed to determine the correlation of crop production locations to habitat occupied by listed species. Agricultural land use data is compiled using GIS software developed by Environmental Systems Research Institute, Inc. (ESRI). The geodatabase consists of a feature dataset containing feature classes, tables, and topology rules. Domains have been created in the attribute tables for the crop type, crop group, irrigation type, rotation crops, NLCD category, and county to minimize data entry errors. There are 160 crop types (e.g., wheat, apple, carrot) that fall into 17 crop groups (e.g., cereal grain, orchard or vegetable). Automation has been added to the geodatabase to maximize efficiency of data input. For example the township, range, section (TRS), county, NLCD land use category fields are automatically updated when creating or editing a record. All geospatial data is in Washington State Plane coordinate system with NAD83 (HARN) horizontal datum and coordinate units of meters. Depending on regulatory priorities, fields are surveyed or updated every two to five years. Agricultural land use data collected is based on surveys of individual fields; as such the base land unit is the field boundary. Field borders are verified and drawn based upon imagery. Typically National Agricultural Imagery Program (NAIP) color mosaics or orthoquad photos downloaded from the USDA website (18) are used. Field boundaries are drawn at a minimum scale of 1:8000. Any field smaller than 0.5 acre is not mapped. Timing is critical for accurate crop classification, as there must be physical evidence of the crop at the time of visual inspection. This physical evidence includes but is not limited to all stages of plant growth from seedling to maturity, post harvest crop residue, and seed. Most perennial crops (e.g., orchards, vineyards, hay, hops or mint) can be classified year around. However, annual crops (e.g., potatoes, onions or beans) need to be surveyed during the growing season. Field survey data may be collected directly from producers. This usually occurs via onsite consultation. Electronic or hard copy maps are generated showing field boundaries with reference layers such as aerial imagery to aid in identify specific fields. Crop locations, irrigation methods, and classifications are verified by the producer via an interview as opposed to visually inspecting specific fields. This approach is beneficial when mapping large farms with limited access. Other data sources that have been used to determine field borders or identify crop type include the National Agriculture Statistics (NASS) Cropland Data Layer (19) and field surveys conducted by Washington State Conservation districts. To ensure data integrity, rules are established within the crop geodatabase to maintain consistent data entry/modification and provide for quality control and quality assurances. In addition to the domains previously mentioned, topology rules prevent polygons from overlapping each other or crossing TRS boundaries. Lastly, data is selected at random for review as part of an established quality assurance and control program (QA/QC). The QA/QC review includes validation 296 In Pesticide Regulation and the Endangered Species Act; Racke, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.
of the field survey data and field boarders by individuals who did not collect the original data. The minimum acceptable crop classification accuracy is 90%, with a target goal of 95%. The 2011 field survey work had a 4.3% error rate.
Downloaded by NORTH CAROLINA STATE UNIV on November 8, 2012 | http://pubs.acs.org Publication Date (Web): November 6, 2012 | doi: 10.1021/bk-2012-1111.ch021
Pesticide Use Intensity Combining pesticide use and agricultural land use data allows for the determination of a spatially explicit estimation of pesticide use intensity. Although the field border is the smallest land unit within the agricultural land use dataset, pesticide use intensity is calculated at the section level which is typically one square mile. Pesticide use intensity calculations are aggregated to the section level to normalize for the spatial variability in agricultural land use. For example, orchards, vineyards and hop yards typically remain in fixed locations, while commodities such as corn, potatoes or tomatoes are typically not grown in the same field in successive years. Aggregating use to the section assumes the commodity could be grown within section on any given year but, not necessarily in the same field. Finally, aggregating data to the section level provides continuity with the reporting unit used by the California Department of Pesticide Regulations Pesticide Use Reporting Program (20).
Figure 1. Oryzalin pesticide use within the ESUs of listed salmonids in Washington State. (see color insert) 297 In Pesticide Regulation and the Endangered Species Act; Racke, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.
Downloaded by NORTH CAROLINA STATE UNIV on November 8, 2012 | http://pubs.acs.org Publication Date (Web): November 6, 2012 | doi: 10.1021/bk-2012-1111.ch021
Pesticide use is calculated as follows for each commodity and summed for each section:
Where AR is application rate, AP is number of applications and AT is percent acres treated for the state. Pesticide use is calculated for the minimum, maximum or median rates determined from the pesticide use database. All crop uses are summed within each section, resulting in a loading estimate that is section specific rather than crop specific. The final map shows pesticide use intensity displayed at the section level as pounds per acre for all known uses (Figure 1). Using the application timing data, temporal use intensity by month can also be calculated.
Figure 2. Oryzalin use in relationship to salmonid habitat (Lower Yakima, WA). (see color insert)
Integrating Spatial Data into Ecological Risk Assessment Understanding the spatial and temporal distribution of pesticide use allows for refinement of exposure scenarios and improved evaluation of the spatial relevance of environmental monitoring data used for a registration risk assessments. Habitat data for threatened and endangered species can be evaluated for co-occurrence of pesticide use. Figure 2 shows the relationship of pesticide use in the lower Yakima Valley to salmonid habitat. Knowing the temporal aspect of pesticide use 298 In Pesticide Regulation and the Endangered Species Act; Racke, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.
Downloaded by NORTH CAROLINA STATE UNIV on November 8, 2012 | http://pubs.acs.org Publication Date (Web): November 6, 2012 | doi: 10.1021/bk-2012-1111.ch021
and habitat utilization allows for a refined exposure assessment for listed species. Spatially explicit use information also allows for determination of the relevance of environmental data such as surface water monitoring detections. Knowing the locations of sampling sites used for pesticide monitoring and pesticide use intensity within a watershed reinforces the relevance of both pesticide detections and nondetections in evaluating environmental exposure for listed species. NMFS has noted the uncertainty of predicting pesticide use over a 15 year duration of a pesticide registration (7) which is the length of the federal action evaluated during consultation. Programs that continually survey and measure changes in agricultural land cover, pesticide use and environmental concentrations of pesticides can be incorporated into an adaptive management strategy for the registration granted by EPA and incorporated into the reasonable and prudent alternatives (RPA) referenced in the biological opinions to avoid jeopardy. An adaptive management approach would evaluate the exposure assumptions used during the registration and consultation process and evaluate the effectiveness of mitigation measures put into place to protect listed species. If changes of pesticide use occur or environmental exposure exceeds levels of concern identified during registration, mitigation could be tailored to meet local conditions.
Conclusions As EPA incorporates assessments for ESA listed species into Registration Review there will be a greater need for high resolution spatial datasets that identify the relationship of pesticide use to habitat of threatened and endangered species. In minor crop states where there is high variability of commodities grown across the landscape it may be beneficial to develop high resolution land cover data that augments the generic land cover designations of national datasets such as the NLCD. Coupling high resolution land cover data with state-specific pesticide use information allows further refinement of the temporal and spatial use of pesticides and subsequently reducing uncertainty surrounding exposure assessments for listed species. Lastly, temporal and spatial data characterizing pesticide use can be integrated into adaptive management plans to tailor mitigation for local conditions. This allows risk managers to focus limited resources on areas where protection is most needed.
References 1.
2.
Odenkerchin, E. Advancements in Endangered Species Act Effects Determination for Pesticide Registration Actions. In Pesticide Regulation and the Endangered Species Act; Racke, K., McGaughey, B. D., Cowles, J. L., Hall, A.T., Jackson, S. H., Jenkins, J. J., Johnston, J. J., Eds.; ACS Symposium Series 1111; American Chemical Society: Washington, DC, 2012. United States Fish and Wildlife Service and National marine Fisheries Service. Endangered Species Consultation Handbook; March 1998 Final. 299 In Pesticide Regulation and the Endangered Species Act; Racke, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.
3.
Downloaded by NORTH CAROLINA STATE UNIV on November 8, 2012 | http://pubs.acs.org Publication Date (Web): November 6, 2012 | doi: 10.1021/bk-2012-1111.ch021
4.
5.
6.
7.
8. 9. 10. 11.
12.
13.
14.
http://www.fws.gov/endangered/esa-library/pdf/esa_section7_handbook.pdf (accessed July 13, 2012). U.S. Environmental Protection Agency. Overview of the Ecological Risk Assessment Process in the Office of Pesticide Programs; January 23, 2004. http://www.epa.gov/oppfead1/endanger/consultation/ecorisk-overview.pdf (accessed April 30, 2012). National Marine Fisheries Service Endangered Species Act Section 7 Consultation Biological Opinion for Environmental Protection Agency Registration of Pesticides Containing Chlorpyrifos, Diazinon, and Malathion; November 2008. http://www.nmfs.noaa.gov/pr/pdfs/ pesticide_biop.pdf (accessed April 30, 2012). National Marine Fisheries Service Endangered Species Act Section 7 Consultation Biological Opinion for Environmental Protection Agency Registration of Pesticides Containing Carbaryl, Carbofuran, and Methomyl; April 2009. http://www.nmfs.noaa.gov/pr/pdfs/carbamate.pdf (accessed April 30, 2012). National Marine Fisheries Service Endangered Species Act Section 7 Consultation Biological Opinion for Environmental Protection Agency Registration of Pesticides Containing Azinphos methyl, Bensulide, Dimethoate, Disulfoton, Ethoprop, Fenamiphos, Naled, Methamidophos, Methidathion, Methyl parathion, Phorate and Phosmet; August 2010. http://www.nmfs.noaa.gov/pr/pdfs/final_batch_3_opinion.pdf (accessed April 30, 2012). National Marine Fisheries Service Endangered Species Act Section 7 Consultation Biological Opinion for Environmental Protection Agency Registration of Pesticides Containing 2,4-D, Triclopyr BEE, Diuron, Linuron, Captan, and Chlorothalonil; June 2011. http:// www.nmfs.noaa.gov/pr/pdfs/consultations/pesticide_opinion4.pdf (accessed April 30, 2012). National Land Cover Database 2001 Product Legend; http://www.mrlc.gov/ nlcd01_leg.php (accessed May 7, 2012). The Compendium of Washington Agriculture; http://69.93.14.225/wscpr/ index.cfm (accessed May 7, 2012). California Department of Pesticide Regulation Pesticide Use Reporting; http://www.cdpr.ca.gov/docs/pur/purmain.htm (accessed April 30, 2012). Washington State University Extension Pesticide Safety. http://extension.wsu.edu/agriculture/pests/safety/Pages/default.aspx (accessed April 30, 2012). Oregon State University, Washington State University, and University of Idaho Extension. Pacific Northwest Weed Management Handbook; Oregon State University: Corvallis, OR; 455 pages. Oregon State University, Washington State University, and University of Idaho Extensions. Pacific Northwest Insect Management Handbook; Oregon State University: Corvallis, OR; 665 pages. Washington State University Pesticide Information Center Online Databases; http://cru66.cahe.wsu.edu/LabelTolerance.html (accessed April 30, 2012). 300 In Pesticide Regulation and the Endangered Species Act; Racke, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.
Downloaded by NORTH CAROLINA STATE UNIV on November 8, 2012 | http://pubs.acs.org Publication Date (Web): November 6, 2012 | doi: 10.1021/bk-2012-1111.ch021
15. National Agricultural Statistics Service Agricultural Chemical Usage Reports; http://www.nass.usda.gov/Statistics_by_Subject/Environmental/ index.asp (accessed April 30, 2012). 16. National Agricultural Statistics Service Vegetable Chemical Usage Survey and Estimation Procedures; http://www.nass.usda.gov/Surveys/ Guide_to_NASS_Surveys/Chemical_Use/ChemUseVegetableStatistical Methodology.pdf (accessed April 30, 2012). 17. National Agricultural Statistics Service Fruit Chemical Usage Survey and Estimation Procedures; http://www.nass.usda.gov/Surveys/ Guide_to_NASS_Surveys/Chemical_Use/ChemUseFruitStatistical Methodology.pdf (accessed April 30, 2012). 18. United States Department of Agriculture Geospatial Data Gateway; http:// datagateway.nrcs.usda.gov/ (accessed April 30, 2012). 19. National Agricultural Statistics Service Cropland Data Layer; http:// nassgeodata.gmu.edu/CropScape/ (accessed May 7, 2012). 20. California Department of Pesticide Registration CalPIP Help/User Guide; http://calpip.cdpr.ca.gov/infodocs.cfm?page=aboutpur (accessed May 7, 2012).
301 In Pesticide Regulation and the Endangered Species Act; Racke, K., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.
