Workshop Home

WWF/WPVGA Collaboration For Pesticide Risk Reduction
Background and First-Year Results
- Sarah Lynch, Chuck Benbrook
Background on the WWF/WPVGA Collaboration
The goal of the precedent setting collaboration between World Wildlife Fund (WWF) and Wisconsin Potato and Vegetable Growers Association (WPVGA), an environmental organization and an agricultural commodity association, is to promote development and wider use of farming systems that are safer for farm families, consumers and the environment.

The collaboration between WWF and WPVGA grew out of a shared concern over the impact of pesticides on the health of rivers and streams that feed into the Great Lakes Basin. A 1997 report entitled "Reducing Reliance on Pesticides in Great Lakes Basin Agriculture" presented the findings of a long-term study conducted by WWF with funding from the Joyce and Charles Stewart Mott and Johnson Foundations and Great Lakes Protection Fund. The report discussed the impact of agriculture's reliance on high-risk pesticides on basin ecosystems and wildlife diversity and health. The study identified the potential for significant reductions in reliance on and use of highly toxic pesticides from greater adoption by farmers of more biologically based Integrated Pest Management (IPM) systems.

In October 1996, WWF and WPVGA signed a Memorandum of Understanding (MOU) in order to establish a working relationship between the two organizations to undertake mutually desirable activities. The MOU identified several common goals, among them:

WWF and WPVGA established an Advisory Committee to help guide the collaboration on matters related to setting and measuring attainment of pesticide risk reduction and IPM adoption goals. The Advisory Committee, first convened in May 1997, consists of potato farmers, University IPM researchers, food industry representatives, a farm finance expert, representatives of environmental, consumer and sustainable agriculture groups and WWF and WPVGA staff. Members of this group have overseen refinements in collaboration goals, timetables, and measurement methods, and their application in monitoring progress toward project goals.

Attainment of 1997 WWF and WPVGA Industry-wide Pesticide Risk Reduction Goals
The 1996 WWF-WPVGA Memorandum of Understanding established 1995 as the baseline year, and one, three, and five year pesticide use and risk reduction goals applicable to crop seasons 1997, 1999, and year 2001. Two criteria were agreed upon in identifying which of the pesticides potato farmers used in 1995 would be subject to risk reduction goals.

Acute Risk. Pesticides classified as "Extremely Hazardous" or "Highly Hazardous" by the World Health Organization in terms of acute mammalian toxicity fall under the project's acute toxicity reduction target. The first year target was a 25 percent reduction in 1997 compared to 1995, using toxicity units as the basic measure of goal attainment. The three-year acute risk reduction goal set for crop season 1999 was a 50 percent reduction in toxicity units from the 1995 baseline. The five-year goal calls for the use of any pesticides falling in this highly dangerous and disruptive class to be phased out or no detectable residues found by the end of crop season 200 1. Four insecticides triggered the acute toxicity criterion: the organophosphates (OP) azinphos-methyl (Guthion) and methamidophos (Monitor), and the carbamates carbofuran (Furadan) and oxamyl (Vydate).

Chronic Risk. The second risk reduction criterion targets pesticides that pose longer-term risks to humans from low-level, chronic exposures. Any pesticide that is a known endocrine disrupter or a Class A or B2 carcinogen is subject to a 15 percent chronic toxicity reduction goal between 1995 and 1997, again measured as change in toxicity units. Third year goals call for a 30 percent reduction from the 1995 baseline and phase out or no detectable residues of these chemicals by year five of the collaboration in crop season 2001. Seven active ingredients fall under this criterion: the herbicide metribuzin (Sencor), the insecticides permethrin (Pounce) and endosulfan (Thiodan), and the fungicides maneb, mancozeb (Dithane), chlorothalonil (Bravo), and triphenyltin hydroxide (Super Tin).

The chronic risk reduction goal was set at the more modest level of 15 percent for two reasons. First, there is a lack of evidence of human exposure to these seven pesticides under the conditions of use on Wisconsin potato farms. Worker exposure concerns are minimal since there is little hand labor in potato production. Residues of these products are found rarely in fresh or processed potato products, nor do they appear in drinking water. Second, the continuing difficulty growers have faced in managing the outbreak of a new strain of the potentially devastating disease late blight, which hit the industry in 1994 and has grown increasingly virulent since.

Measuring Progress
The goal of the collaboration is to limit human health risks and promote improvement in environmental quality. To incorporate into one assessment tool the environmental and public health impacts, the project developed an index of pesticide toxicity levels encompassing the ecological, environmental and human health risks associated with application of different active ingredients (see Appendix 1). The index was used to calculate a "toxicity factor value" for all pesticides used in potato production. These composite values allow comparisons across active ingredients on a pound-for- pound basis, and are reported in the first column of Table 1.

The pounds applied of each of the 11 active ingredients subject to risk reduction goals were converted to toxicity units. Toxicity units are calculated by multiplying the pounds applied of a pesticide by the pesticide's toxicity factor value. The result of this calculation for 1995 is shown in the fourth column in Table 1. The last column shows 1997 toxicity units, based on pounds applied in 1997 multiplied by the same toxicity factor values. Progress from 1995 to 1997 in reducing risk is then estimated by the reduction in overall toxicity units between 1995 and 1997. Tables 2 and 3 report several indicators of progress, each based on changes in industry-wide toxicity units. Some are expressed in terms of industry-wide totals, others on the basis of averages per planted acre.

Indicators of Progress
Table 2 shows clearly the remarkable progress made in just two years by the Wisconsin potato industry. The last two columns contain the most important measures of progress - "Industry-wide "Toxicity Units" and "Toxicity Units per Planted Acre." The first measure reflects changes in the overall number of toxicity units associated with pesticides applied in the Wisconsin potato industry. It is the best indicator of the overall potential impacts of pesticide use in the six-county region where most potatoes are grown in the state. It does not, however, take into account the annual variation in acres planted in potatoes.

"Toxicity Units per Planted Acre" for 1995 is simply industry-wide toxicity units divided by the total acres planted. In 1995, there were 83,000 acres of potatoes grown in Wisconsin, and in 1997, 78,000 acres. The five thousand-acre reduction in total acres planted accounts for about a 6 percent decrease in industry-wide toxicity units.

Compared to the 1995 baseline on the average acre planted, Wisconsin potato growers were able to -

Table 4 places the accomplishment of Wisconsin potato farmers into a national perspective. In three key measures of pesticide risk, the national trends are rising while in Wisconsin the trends are solidly downward. Overall, national toxicity units in potato production increased 16 percent in 1997 compared to the baseline, while in Wisconsin they fell by 20 percent.

Lessons Learned
IPM Adoption Reduces Reliance on Pesticides
An important factor in the significant reductions in pesticide risk achieved by Wisconsin farmers is their reliance on sound integrated pest management practices. Over the last decade Wisconsin potato farmers, in conjunction with an interdisciplinary team of researchers from the University of Wisconsin, have developed a potato crop management system. The computer-based expert system, WISDOM, helps growers integrate diverse information, thereby aiding pest management decision-making. WISDOM incorporates and assesses data on soil temperature, plant age, and humidity and the growth patterns of insects and diseases in order to make prescriptive versus routine calendar pesticide applications. Crop rotations have lengthened, improving soil quality and lessening the intensity of potato pest pressure. Finally, growers, crop consultants, extension agents, and processors have banded together in a commitment to crop scouting that covers over 90 percent of Wisconsin potato acreage. Scouting helps ensure that pesticide applications are made based on economic thresholds, thereby reducing overall pesticide use.

Strong Leadership Can Accelerate Change
Strong university IPM team involvement and commodity association leadership clearly has made a major difference in how swiftly IPM practices and reduced risk alternatives have been adopted in Wisconsin. The best evidence of the importance and impact of a committed and vocal commodity organization is the dramatic drop in the use of two high-risk insecticides that WPVGA targeted well before the 1997 growing season. Two insecticides-- the OP insecticide methamidophos, or Monitor, and the chlorinated hydrocarbon endosulfan, or Thiodan--accounted for 63 percent of the toxicity units stemming from insecticide use in 1995.

Monitor use in Wisconsin dropped from 69,000 pounds applied on 54 thousand acres in 1995, to just 17,000 pounds on 17,100 acres in 1997. The trend was opposite nationwide - Monitor was applied to 29 percent of national acreage in 1996 and 36 percent in 1997. Thiodan was an equally important insecticide in 1995 in Wisconsin, with about 65 percent of acres treated with 60,000 pounds. Recent research and new IPM techniques led WPVGA to advise growers to stop applying Thiodan altogether. Acres-treated with Thiodan dropped to just 10,000 and only 10,000 pounds were applied. Again, the nation wide trend in use was up - 10 percent of acres treated in 1996, rising to 14 percent in 1997.

Safer Chemistry Can Make a Difference
The availability of an effective, affordable, and safer insecticide for Colorado potato beetle control - imidachloprid, or Admire - made possible the wholesale shift away from the much higher- risk OPs and carbamates that were used widely in 1995. In the area of plant disease control, Wisconsin growers did not have any important new fungicides to lessen reliance on older, higher risk products. Nor did any new resistant varieties or biological control techniques become available. As a result, less progress was made in reducing fungicide toxicity units than those associated with insecticide use.

Resistance Management is Critical
The dramatic decline in toxicity units made possible by the introduction of Admire, and other reduced risks pesticides, can be sustained if new tools are managed systematically to limit the emergence of resistance. Two major insect pest management tools that fit well into IPM systems - Admire and Bt - are both in some jeopardy to resistance if not used properly.

The same will hold true as other biopesticides are registered and gain a place in pest management systems. Most biopesticides are much more selective in their mode of action, which reduces adverse impacts on non-target organisms, but this positive trait can also heighten the risk of resistance. For years, the University of Wisconsin IPM research team has placed great emphasis on the aggressive management of resistance. Growers have responded positively, following recommended practices. It will be important to maintain this focus as new pesticides enter the market.

Appendix 1:
Calculating Pesticide Toxicity Units
The project developed a multiattribute index that allows comparison of pesticide active ingredient toxicity across several dimensions of potential impact. The index is composed of four major components: acute mammalian toxicity (AM), chronic mammalian toxicity (CM), ecotoxicity (ECO), and impacts on biointensive IPM systems (BioIPM). The general equation used to calculate the toxicity factor value for a given pesticide is:
Value for Pesticidex = (a)AMx + (b)CMx + (c)ECOx + (d)BioIPMx

Where, (a), (b), (c), and (d) are weights assigned to each component index.

Guidance was sought from the WWF-WPVGA Advisory Committee and technical consultants regarding what weights to use for the purpose of establishing baseline multiattribute toxicity units subject to project risk reduction goals. Several different formulas were calculated and explored, leading to a decision to adopt the following weighting scheme -

Wisconsin Project Risk Index for Pesticidex = (0.5)*AMx + CMx + ECOx + (1.5)*BioIPMx

The Advisory Committee assigned a lower weight to acute mammalian toxicity because of the relative lack of circumstances leading to worker and applicator exposure and low frequency of resides of acutely toxic pesticides in harvested potatoes, especially after washing, peeling, cooking and/or processing. An adjustment was not applied to the chronic mammalian toxicity index because the potential for low-level exposures are more widespread in the region from pesticides in drinking water, the air, and as a result of occupational exposure. The ecotoxicity index was also not adjusted and reflects the importance of impacts on birds, fish and other aquatic organisms. A (1.5) weight was given to the BioIPM component because of the importance of a pesticides impact on soil microorganisms and beneficial arthropods.

Further details on this method and data sources are available in the methodological paper "Monitoring Progress Toward Pesticide Risk Reduction Goals in the Wisconsin Potato Industry," written by collaboration consultant, Dr. Charles Benbrook. The paper is available from WWF or WPVGA.

Table 1. Wisconsin Potato Pesticide Use and Toxicity Units: 1995 and 1997 3.2

 Toxicity
Factor
Values		 1995
Acres
Treated		 1995
Pounds
Applied		 1995
Toxicity
Units		 1997 Acres
Treated		 1997 Pounds
Applied		 1997
Toxicity
Units
Herbicides:






Glyphosate 78.06,640 4,000312,000 7,8009,000 702,000
Unuron 73.17,470 7,000511,700 5,4603,000 219,300
Metolachlor 47.814,940 21,0001,003,800 4,6807,000 334,600
Metribuzin 126.673,870 39,0004,937,400 72,54034,000 4,304,400
Pendimethalin 166.629,880 24,0003,998,400 22,62016,000 2,665,600
Sethoxydim 60.88,300 2,000121,600 00 0
Total: All
Herbicides
141,10097,000 10,884,900113,100 69,0008,225,900
Per Planted
Acre
1.71.2 1311.5 0.9105








Insecticides:
Azinphos-
methyl		 220.321,580 26,0005,727,800 00 0
Carbofuran 383.513,280 13,0004,985,500 00 0
Dimethoate 314.423,240 11,0003,458,400 46,02030,000 9,432,000
Endosulfan 288.554,780 60,00017,310,000 10,14010,000 2,885,000
Esfenvalerate 417.049,800 3.0001,251,000 46,0202,000 834,000
Imidachloprid 115.00 00 36,6608,000 920,000
Methamidophos 199.753,950 69,00013,779,300 17,16017,000 3,394,900
Oxamyl 282.06,640 5,0001,410,000 00 0
Permethrin 264.718,260 4,0001,058,800 00 0
Piperonyl butoxide 48.714,110 3,000146,100 17,9407,000 340,900
Pyrethrins 164.38,300 830136,369 00 0
Total: All
Insecticides
263,940194,830 49,263,269173,940 74,00017,806,800
Per Planted
Acre
2.3 5942.2 0.95228
Fungicides:






Basic copper
sulfate		 42.14,150 13,000547,300 7,0208,000 336,800
Chlorothalonil 79.873,040 08,00032,558,400 75,660591,000 47,161,800
Copper
hydroxide		 51.531,540 40,0002,060,000 31,98052,000 2,678,000
Copper resinate 50.55,810 12,000606,000 10,1402,000 101,000
Cymoxanil 62.0
17,940 17,9405,000 310,000
Mancozeb 180.071,380 412,00074,160,000 52,260287,000 51,660,000
Maneb 153.911,620 76,00011,696,400 16,38062,000 9,541,800
Metalaxyl 172.012,450 4,000688,000 00 0
Propamocarb
hydroch		 51.19,960 9,000459,900 00 0
Triphenylbn
hydroxide		 424.038,180 12,0005,088,000 36,6608,000 3,392,000
Total: All
Fungicides
258,130 986,000127,881,940 248,0401,015,000 115,181,400
Per Planted
Acre
3.111.9 1,5413.2 13.01,477








Herbicides,
Insecticides,
and Fungicides






Total: H+I+F
663,1701,277,830 188,030,109535,080 1,158,000141,214,100
Per Planted
Acre
8.515.4 2,26516.9 14.81,810

Table 2. Progress in Reducing Pesticide Use and Risks in Wisconsin Potato Production: Percent Reduction in 1997 from 1995 Baseline Values


 Acres
Treated		 Pounds
Applied		 Pounds
Per Acre
Planted		 Industry-
wide
Toxicity
Units		 Toxicity Units Per
Acre
Planted
11 Active Ingredients Subject to
Reduction Goals		 36% 10% 4% 29% 25%






All Pesticides




Herbicides 20% 29% 24% 24% 20%
Insecticides 34% 61% 59% 64% 61%
Fungicides 4% -3% -10% 10% 4%






All Herbicides, Insecticides and
Fungicides		 19%9% 3%25% 20%

Table 3. Pesticide Risk Reduction Goals and Accomplishments: 1995 to 1997

Wisconsin Acres 1995 = 83,000
Wisconsin Acres 1997 = 78,000 		Pounds Applied
1995		 1995 Toxicity Units
Subject to
Reduction Goals		 Pounds Applied
1997		 1997
Toxicity Units
Acute Goal



Methamidophos 69,00013,779,300 17,0003,394,900
Azinphos-methyl 26,0005,727,800 --
Carboftiran 13,0004,985,500 --
Oxamyl 5,0001,410,000 --
Total: 4 Acute Pesticides 113,000 25,902,600 17,000 3,394,900
Per Planted Acre 1.36 312 0.22 43.5





Reduction Needed to Meet
25% Industry-wide Goal
6,475,650

Actual Reduction: 1995 to
1997
22,507,700

Muliple of Needed Reduction
3.5






Chronic Goal



Mancozeb 412,00074,160,000 287,00051,660,000
Chlorothalonil 408,00032,558,400 591,00047,161,800
Endosulfan 60,00017,310,000 10,0002,885,000
Maneb 76,00011,696,400 62,0009,541,800
Triphenyftin hydroxide 12,0005,088,000 8,0003,392,000
Metribuzin 39,0004,937,400 34,0004,304,400
Permethrin 4,0001,058,800 --
Total: 7 Chronic Pesticides 1,011,000 146,809,000 992,000 118,945,000
Per Planted Acre 12.2 1,769 12.7 1,525
Reduction Needed to Meet
15% Industry-wide Goal
22,021,350

Actual Reduction: 1995 to
1997
27,864,000

Muliple of Needed Reduction
1.3

Table 4. Wisconsin and National Trends in the Toxicity of Pesticides Used in Potato Production: 1995 to 1997
Herbicides, Insecticides, and Fungicides

1995 Toxicity Units Per
Acre		 1997 Toxicity Units Per
Planted Acre		 Percent Change
1995 to 1997
National1,763 2,05116%
Wisconsin 2,2651,810 -20%




Insecticides
National 825878 6%
Wisconsin 594228 -62%




Fungicides
National 7531,009 34%
Wisconsin 1,541 1,477 -4%

Monitoring Progress In Meeting the Pesticide Risk Reduction
Goals of the WWFIWPVGA Collaboration

In 1996 World Wildlife Fund (WWF) and the Wisconsin Potato and Vegetable Growers Association (WPVGA) launched a collaborative effort to achieve ambitious industry-wide goals of increased adoption of biologically based pest management systems and pesticide risk reduction. The WPVGA represents about 250 growers accounting for nearly all potato production in the state, and has played an active role for more than a decade in supporting IPM research and on-farm innovation designed to lessen the environmental effects of potato productions systems.

WWF is an international environmental organization working to protect species, enhance wildlife habitat, and prevent pollution through the reward of private sector commitment to environmental stewardship. Now in its fourth decade, WWF works in more than 100 countries around the globe and has over 1.4 million members in the United States.

Major goals of the collaboration between WPVGA and WWF include developing and promoting adoption of biointensive Integrated Pest Management1 (IPM) systems and demonstrating the linkages between IPM and pesticide use and risk reduction. A critical step in demonstrating such a linkage is developing a method to take into account the highly variable toxicity of pesticides on a pound-for-pound basis in terms of the public health and environmental consequences of changes in pesticide use. In this technical paper we summarize the pesticide toxicity adjustment methodology developed by WPVGA and WWF, with the assistance and support of scientists at the University of Wisconsin -- Madison. The toxicity adjustment index described in this paper will be used to monitor over time the progress made by WPVGA in meeting the pesticide risk reduction targets established in the collaboration between WPVGA.

The toxicity adjustment index, and the database supporting its application to potato pesticides, will continue to evolve over the life of the project. A number of researchers and organizations, both in the U.S. and abroad, are working to refine methods to quantify pesticide impacts on human health, wildlife, and the environment. We will incorporate the data and methodological improvements generated by these efforts over time and issue revised versions of the toxicity adjustment index.

A. The Need for a Toxicity Adjustment Methodology
Interest is growing worldwide in the potential of IPM to reduce the direct and indirect health and environmental costs associated with heavy reliance on pesticides. Both state and federal agencies have initiated or are participating in programs designed to promote adoption of IPM. Environmental and consumer groups, including the World Wildlife Fund, are also exploring ways to promote and reward IPM adoption through marketplace initiatives (for a review of WWW's involvement, see Hoppin, 1996).

Better measures of IPM adoption, linked to pesticide use and risk reduction, are needed in part because of the rapid pace of innovation in the design and practice of pest management systems. Several factors are driving change -

The ability of many pest managers - and policy-makers - to project and manage the consequences of change in pest management systems is not keeping pace with pests or IPM systems and technology. Two traditional measures of pesticide use -- pounds of active applied per acre, and number of applications -- are increasingly inadequate when used to estimate the agronomic, environmental and public health consequences of pesticide use.2 A review of several efforts to develop new measurement tools appears in Pest Management at the Crossroad3 (PMAC); the PMAC website contains several documents describing and critiquing various risk index and indicator methodologies (see http://www.pmac.net).4

New and improved analytical tools are needed to address tradeoffs between environmental protection, public health, and crop production that are inherent outcomes whenever pest management systems change. Tradeoffs are unavoidable, increasingly complex, and must be better understood for farmers, researchers, and policy-makers to gain a firmer sense of direction and priority in shaping pest management systems and regulatory policies. Farmers recognize that alternative approaches pose new risks and create demands for new sources and kinds of information. Finding new ways to provide such information and insight is one of the basic goals of the WWF/WPVGA potato IPM project and the focus of this paper.

1. Adjustment Methodology Attributes
Toxicity adjusted methods need to be dynamic and pliable, and offer a structured framework within which to pose and answer questions about the human health, environmental and ecological impacts of changes in pest management systems. The structure should make explicit gaps in knowledge, data sources, statistical formulations, and assumptions, while also facilitating the integration of expert judgment and new information. The necessary components within a pesticide risk index will vary as a function of the types of pesticides being used, the cropping systems they are used within, soil and climatic conditions, and the dominant risk and environmental concerns in the geographic region under study. The toxicity factors described herein, and the methods used to estimate values for application to pesticide use in central Wisconsin, are not necessarily the most relevant to assess potato production in the Pacific Northwest.

Measurement tools should support assessment of different categories of pesticide risks, singly or in combination. Adjustment factors will fit some but clearly not all situations. The broader the application, the more challenging the task of coming up with indicators that are equally applicable to different circumstances.5

There is no "right way" to construct and use toxicity adjustment factors. Most risk indices are too crude to distinguish reliably between two pesticides that are related and cause comparable effects at similar dose rates. Risk indices are most reliable and useful in monitoring changes in average levels of toxicity over time, and in comparing the toxicity and risks associated with groups of pesticides. Such comparisons provide insights into the magnitude of environmental and public health gains possible from progress along the EPM continuum.

B. Four Major Component Indices

The WWF-WPVGA multiattribute toxicity index is composed of four component indices: acute mammalian (AM), chronic mammalian (CM), ecological (ECO), and impacts on beneficial organisms and EPM systems (BioIPM). The sub-indices within each of the four major indices are based largely on pound-for-pound-applied comparisons of toxicity to a common set of organisms.

The methodology does not include many important factors affecting exposure and hence risk. These include formulation differences, application methods, pre-harvest intervals, safety equipment used, and environmental fate, nor do they encompass some of the ways pesticides can harm non-target species (loss of habitat, food-chain disruption, multigeneration endocrine effects). In addition, the indices do not encompass economic impacts, although future applications are likely to include tradeoffs between reductions in toxicity units and the cost and reliability of control interventions.

An important note about terminology - in each component index, toxicology and environmental fate data on individual pesticide active ingredients are used to calculate "toxicity factors," or "toxicity factor values." When two or more component indices of such values are combined, the result is an estimate of multiattribute toxicity factors. These factors, unique to each pesticide active ingredient, are used in producing estimates of the "toxicity units" associated with the application of a given pesticide. Use of the term toxicity "values" or "factors," then, applies to estimates of toxicity comparable on a pound-for-pound basis; the term toxicity "unit" refers to an estimate of the toxicity units associated with the actual amount of a pesticide applied in a given region.

1. General Methodological Issues
In the process of calculating component indices and combining them into a multiattribute measure of pesticide toxicity suitable for comparisons across active ingredients, several structural and statistical issues arise. Decisions have to be made regarding scaling, outlier values, dealing with missing data, and integrating data from many different, but related studies into a single value. The validity of an index rests in part on how wisely and consistently these decisions are made and then adhered to as the various component sub-indices and indices are estimated and combined.

Selecting a Scaling Factor. Scaling is critical in assuring that the range of numbers in component indices is roughly the same. If they are not, the index with the larger values and/or the greater variability will tend to drive the values in a multiattribute index. We applied scaling factors to remove this implicit weighting, and instead use explicit weighting factors to then alter the importance placed on individual indices.

For example, in estimating Acute Toxicity Factors in this project, a scaling factor of 500 is used. The range and maximum values of Inverse LD-50s, and other component indices, need to be adjusted through multiplication by scaling factors so that each component index varies across roughly the same range of numbers.

Water Leaching Index. A second adjustment was made in acute and chronic mammalian toxicity index values in light of the dominance of drinking water exposure as a potential source of human health risk. The EPA model SCI-GROW was used to produce the water leaching index by active ingredient, which is then multiplied by the Scaled Inverse LD-50s (see the first column of values in Table 6).

Adjusting Outlier Values. Any value in an index that is more than twice as large as the next highest value should be reviewed to determine if the difference accurately reflects actual differences in toxicity, or an artifact of the data accessible to estimate a given index value.

Deciding whether outlier values should be adjusted requires a review of other toxicological data on comparable endpoints from a set of similar studies, or from similar studies in related species. If the difference between an outlier and the next most toxic pesticide is consistently large, the value should not be adjusted. But when the data show that there are equally or more toxic pesticides, based on different studies or endpoints, an adjustment is probably warranted. In cases where an adjustment was found necessary, the mean and standard deviation of unadjusted values within the index were used as the basis for setting maximum values, following guidance from the WWF-WPVGA project advisory committee.

2. Acute Mammalian Toxicity
The first component index reflects acute mammalian toxicity, and is based on oral LD-50s. Data on LD-50 values are derived predominantly from the WHO Recommended Classification of Pesticides by Hazard and Guidelines to Classification 1996-1997" (International Programme on Chemical Safety, 1996). LD-50 values for recently registered active ingredients are derived from EPA tolerance documents appearing in the Federal Register. In a few cases, LD-50 values were derived from "Farm Chemicals Handbook '98" (Meister Publishing Company, 1998), company, or other sources.

The inverse of LD-50 values are calculated so that rising index values correlate with rising toxicity and a scaling factor of 500 is used. At the direction of the WWF-WPVGA project advisory committee, a second adjustment was made in acute and chronic mammalian toxicity index values in light of the dominance of drinking water exposure as a potential source of human health risk. The EPA model SCI-GROW was used to produce the water leaching index by active ingredient, which is then multiplied by the Scaled Inverse LD-50s (see the first column of values in Table 6).

Table 1 presents data on the pesticides used in Wisconsin potato production in 1995 ranked by pounds applied within each major class of pesticides. Table 2 presents pesticide use ranked by "Acute Toxicity Units." The fourth column in Table 2 reports leaching adjusted Scaled Inverse LD-50 values.

A comparison of results in these Tables 1 and 2 highlights the importance of taking toxicity into account when evaluating changes in pesticide reliance and pest management systems -

3. Chronic Mammalian Toxicity
The second component index is most relevant for assessing longer-term drinking water, occupational, and dietary risks. It encompasses chronic mammalian toxicity (CM) - the capacity of an active ingredient to cause adverse health impacts (cancer, birth defects, impaired immune system function, reproductive impacts) as a result of long-term, low-level exposures. It is based largely on a pesticide active ingredient's Reference Dose (RfD). Other factors in the algorithm include oncogenic potential and potency, and the capacity to disrupt endocrine system mediated functions.

The index is calculated using a composite variable "Mam Tox Score" that was first calculated to evaluate long-term trends in pesticide chronic toxicity as part of the analysis reported in the Consumers Union book Pest Management at the Crossroads (Benbrook et al., 1996). Consultations with experts and sensitivity analysis were relied upon in choosing the formula that best represented an estimate of comparative chronic mammalian risks, drawing upon readily available EPA toxicological data. The Mam Tox Score variables and formula are:

RfD: EPA reference dose (or other available estimate)
ED: Endocrine disrupter -- if yes, value=3; if no information or no evidence from appropriate assays6, value=l

Q*: EPA cancer potency factor (or "best" estimate available)
CLASS: EPA Oncogenicity Class--if A or B/2, value=10; if C, value=5; if D, value=2.

Mam Tox Score Pesticidex = [(0.01/RfDx EDx] + [Q*x 50 x CLASSx]

Table 3 presents data on the range and distribution of water leaching adjusted "Scaled Mam Tox Score" values for pesticides applied in potato production in Wisconsin in 1995. The unadjusted Mam Tox Score value of the fungicide triphenyltin hydroxide was clearly an outlier - 2,400 compared to the next highest values of any currently registered pesticide (metiram and ethropop at 1,000). In several chronic toxicology studies, many pesticides are more toxic than triphenyltin. Accordingly, the Scaled Mam Tox Score value of triphenyltin hydroxide was set at 286.7, one-half a standard deviation above the mean value across the pesticides included in Table 3. This value is about 60 percent greater than the next highest, dimethoate's value of 176. Triphenyltin was the only pesticide with an outlier value in this index.

In Table 3, active ingredients within each major class of pesticides are ranked by share of total chronic toxicity units. Key insights include -

4. Ecological Toxicity and Risks7
The third major index is designed to capture the relative ecological toxicity of pesticides. It integrates avian, aquatic and small invertebrate ecological risks. Impacts on organisms that can play a role in biointensive IPM (beneficial arthropods) or enhancing soil quality (worms and soil microorganisms) are included in the Biointensive IPM index, described next.

The ecological risk index (ECO) is composed of sub-indices covering toxicity to two fish species (Rainbow trout and Bluegill), several bird species, and daphnia, a small aquatic organism. The potential impacts captured in this index are important given the environmental attributes of the Central Wisconsin region and the goals of the project.

Avian Toxicity. Substantial data confirm that the organophosphate and carbamate insecticides are by far the most toxic major groups of pesticides to birds. For this reason, organophosphate and carbamate avian toxicity values were sought from Dr. Pierre Mineau, an avian toxicologist working for the Canadian Fish and Wildlife Service. With colleagues in Germany, Netherlands, and the United States, Dr. Mineau has developed a comprehensive avian toxicity model, drawing on a database with over 1,300 values for 146 pesticides and 61 species (Baril and Mineau, 1998).

Avian toxicity values providing by Mineau assume a 95 percent protection threshold - i.e. dosage levels that would result in 95 percent of the exposed birds surviving. Mineau values for nine OP and carbamate insecticides appear in column three, Table 4, and apply to birds weighting 20 grams (mid-size song birds). Carbofuran accounted for 7 percent of the total pounds of insecticides applied in 1995, and 44 percent of total insecticide avian toxicity units (data not shown). Endosulfan accounts for another 25 percent, methamidophos 16 percent and both azinphos-methyl and oxamyl, 7 percent. No other insecticide accounts for more than I percent of total avian insecticide toxicity units.

For non-OP and carbamate insecticides, herbicides, fungicides, and other chemicals applied in Wisconsin, a method was developed to calculate an approximation of acute avian toxicity levels comparable to the OP-carbamate values provided by Mineau. The method relied on Bobwhite and Mallard duck avian toxicity data provided by EPA in the "Ecotoxicology Database" (EPA-Ectox) managed by Dr. Brian Montague (Montague, 1998). For each active ingredient, avian studies were extracted from the EPA-Ecotox database.

Fish Toxicity. No team of scientists has developed a comprehensive model to synthesize and interpret existing fish acute toxicity data comparable to the work by Mineau and colleagues on avian risks., There is, however, extensive data on pesticide impacts on several fish species available from studies in EPA files, much of it from a comprehensive 1986 report by U.S. Fish and Wildlife biologists.8

Through review of the fish studies in EPA databases, rainbow trout and bluegill were identified as the two most widely used test species. Valid studies were selected by active ingredient, most of which involved between 75 percent and 100 percent active ingredient in the test substance. When there were two or more studies, LC-50 values were averaged. Both the rainbow trout and bluegill data were inverted, so that rising values were associated with increasing toxicity, and a scaling factor was used so that the maximum value in each inverted index was roughly the same.

The most toxic pesticides to fish are several thousand times more toxic than the least toxic pesticides. The range in unadjusted fish toxicity values Table 4 is over 500,000-fold. Esfenvalerate, a synthetic pyrethroid insecticide, is by far the most toxic pesticide, with an unadjusted fish toxicity sub-index value of 500. Since this value was four times higher than the next most toxic pesticide to fish (endosulfan), the esfenvalerate value was adjusted 164.7, two standard deviations above the mean fish toxicity value. As in the case of avian toxicity, the majority of herbicides and fungicides are considerably less toxic than most insecticides.

Small Aquatic Invertebrates. Pesticides can place fish and bird species at jeopardy through impacts on aquatic food chains. In order to capture this potential in the Ecotoxicity index, a small aquatic organism sub-index was calculated, drawing on data in the EPA-Ecotox database on Daphnia, the most frequently treated crustacean species.

Using the approach previously described, comparable studies were selected out of the EPA database, and average values were calculated when there was more than one study. The values were then inverted and scaled, to produce the numbers in column 1, Table 4. Again, esfenvalerate was by far the most toxic pesticide, with an unadjusted value of 166.7, compared to a value of 13.3 for the next most toxic pesticide - another synthetic pyrethroid (permethrin). This value was adjusted downward to 52.7, two standard deviations above the mean. Not surprisingly, esfenvalerate still has the highest overall Ecotoxicity index value - 219, about 25 percent higher than the Ops carbofuran and phorate.

Calculating Ecotoxicity Values. To calculate Ecotoxicity values, the Daphnia, fish, and avian values for each pesticide active ingredient were summed. Note that the high Ecotoxicity values for esfenvalerate and endosulfan reflect largely aquatic toxicity, while the relatively high values for the carbamates carbofuran and aldicarb are largely from avian toxicity.

Significant variability is evident across classes of organisms in Table 4. For all active ingredients, the lowest value in one of the three sub-indices is 10 to 100-fold less than the value in the sub-index with the highest value. This variability complicates the selection of pesticides but also makes it possible, with good information and field knowledge, to choose pesticides less likely to harm the dominant species sharing an agricultural landscape with crop fields.

While there are substantial data on the acute toxicological effects of pesticides on several non-target organisms, there are scant data to evaluate most sub-chronic and chronic effects involving immune system development and function, or neurological integrity and behavior. Scientists are reasonably confident they have identified and studied the acute toxicological properties of most commonly used pesticides that can, and sometimes do poison fish or birds. They are much less confident in their ability to identify pesticides reducing fish and bird populations through other chronic and indirect mechanisms, such as:

There is also relatively little data on ecosystem-scale and multigeneration impacts, yet growing concern about such effects as a result of the increasing homogeneity of agricultural cropping patterns and pesticide use increases.

4. Impacts on the Viability of Biointensive IPM Systems The fourth major index - BioIPM -- encompasses impacts of vital significance to farmers, their neighbors, and the agricultural industry as a whole. Progress along the IPM continuum cannot be sustained without reducing adverse pesticide impacts on a range of beneficial organisms and biodiversity. As above- and below-ground biodiversity increases, new options emerge to manage species interactions in ways that disadvantage pests and strengthen a plant's ability to cope with or out-grow pest pressure. Indeed, the structured management of biodiversity and species interactions is the nuts and bolts of biointensive IPM.

The data needed to calculate BioIPM sub-indices is far from universally available. Assistance was sought from University of Wisconsin IPM experts in compiling the preliminary estimates reported here. The estimates 5 reflect pesticide use patterns, soils and agroecosystems in central Wisconsin, and should not be accepted as automatically relevant to other regions and cropping systems.

Beneficial Arthropods. The first component is impacts on beneficial arthropods that play direct roles in biological control processes. In the case of insecticides, values in Table 4 are derived from the "Toxic Effect" index developed by Dr. Brian Croft and Karen Theiling at Oregon State University.9 For insecticides, "Scaled Impacts on Beneficial" values were derived from "Toxic Effect' values using the formula:

"Scaled Impact on Beneficials Pesticidex" = 100/(5-Toxic Effect Pesticidex)
This formula was selected as a way to expand the relatively narrow range of values from those reported by Theiling and Croft, and to increase the maximum values comparable to other BioIPM sub-indices. "Toxic Effect" values were not available for most other pesticides, so other sources of data were sought. Values in column two, Table 4 for these other classes of pesticides are derived predominantly from the work of Dr. Joe Kovach and colleagues at Cornell University.10

Soil Microorganisms. Many species of soil microorganisms play an important positive role by improving soil quality, enhancing nitrogen cycling and availability in the soil, promoting healthy root development, and suppressing nematodes and related plant pathogens.

"Scaled Impacts on Soil Microorganisms" values in column four were derived from estimates provided by the University of Wisconsin IPM team. As fanning and IPM systems evolve in Wisconsin, it is likely that soil microbial communities will grow more diverse and important in assuring success with biointensive IPM. A major goal of ongoing research at the University is to promote microbial biocontrol of soil pathogens, as an alternative to periodic fumigation with metam-sodium. As progress is made toward this goal, the team will revisit these values, since the impacts of different pesticides on soil microorganisms is bound to change as the diversity of species increases.

Resistance. The emergence of resistance can undermine the efficacy of a pesticide, and hence undermine the ability of farmers to use otherwise effective pest management tools. In the case of relatively safe pesticides like Bt and glyphosate, the emergence of resistance is likely to lead to increased reliance and use of higher-risk products.

An estimate of each pesticide's likelihood of triggering the evolution of resistance was provided by the University of Wisconsin EPM research team. It should be noted that the very low value for Bt is based on foliar applications of the formulated insecticide, not the planting of Bt-transgenic varieties, which would have a higher likelihood of triggering resistance.

Bee Toxicity. Bees play a critical role in the pollination of both agricultural crops and native species. Pesticide impacts on bees are among the most significant economic losses associated with pesticide use, and are a growing concern worldwide. In some regions, vegetable and fruit crop yield reductions of 40 percent or more have been attributed to poor pollination caused by pesticide impacts on bees.

Bee toxicity data was extracted from the EPA-Ecotox database and appears in column five, table 5. In some cases, values were extrapolated from the three acute bee toxicity ratings in "Farm Chemicals Handbook '98" - "Practically Non-toxic," "Moderately Toxic," and "Highly Toxic." For several active ingredients falling in each "Farm Chemicals" category, average bee toxicity values were calculated from data in the EPA database. These average values were then used for pesticides in the "Farm Chemicals Handbook" classes, but not in the EPA database. A scaling factor of 10 was used in producing final bee toxicity factor values.

The bee toxicity value for imidacloprid required adjustment because of the very large difference in bee toxicity between the granular and foliar formulations of this insecticide. When applied as a liquid foliar spray, imidacloprid's scaled bee toxicity value is by far the highest of any pesticide used in Wisconsin potato production. But a significant portion of imidacloprid is applied at planting time as a granular, essentially eliminated exposure to bee. Plus, there are rarely foraging bees in potato fields later in the season when foliar imidacloprid products are applied, according to University of Wisconsin entomologist Dr. Jeffrey Wyman. Accordingly, the imidacloprid scaled bee toxicity value was adjusted downward to 20.

BioIPM Values. Final BioIPM values were derived by summing the values of the four sub-indices, as reported in column 7, Table 5. These values were then scaled using a factor of 0.4 to produce a maximum value and range comparable to the other major indices. As expected, the average value of insecticides is substantially higher than the average value for other classes of pesticides. But like other indices, there is large variability within and across categories of pesticides.

C. Integrating the Component Indices into a Single Index

Index values and toxicity-adjustment factors are intended to more accurately characterize the tradeoffs inherent in the selection of pesticides and pest management systems. While researchers are often interested primarily in a single dimension of risk, farmers and society as a whole have to live with all risk, efficacy, and economic impacts, both positive and negative. Hence, the complexity of the challenge inherent in identifying the least disruptive and dangerous product across all categories of risk and impacts from those products available to address a given problem pest.

The general functional form of the equation used to calculate multiattribute index values is:

Value for Pesticide. = (a)AMx + (b)CMx + (c)ECOx + (d)BioIPMx

Where, (a), (b), (c), and (d) are weights assigned to each component index.

1. Assigning Weights: An Example Involving Potato Production in Wisconsin

Decisions regarding the importance to place on the four component indices must be made and incorporated in the equation through the weighting factors (a), (b), (c), and (d). Guidance was sought from the WNW-WPVGA Advisory Committee and technical consultants regarding what weights to use for the purpose of establishing baseline multiattribute toxicity units subject to project risk reduction goals. The committee recommended that four different formulas be calculated, reflecting different environmental and public health concerns. The formulas are -

Wisconsin Project Risk Index = (0.5)*AMx + CMx + ECOx + (1.5)*BioIPMx
Equal Weight Index = AMx + CMx + ECOx + BioIPMx
Human Health Focus Index = (1.5)*AMx + (2.0)*CMx + (0.5)*ECOx + BioIPMx
Environment Focus Index = (0.3)*AMx + (0.5)*CMx + (2.0)*ECOx + (2)*BioIPMx

After reviewing the results, the "Wisconsin Project Risk Index" was chosen for use in setting the toxicity unit baseline (pounds applied of pesticide times the toxicity factor value of pesticidex). It will also be used in monitoring risk reduction progress because, in the judgement of the advisory committee, it best reflects the balance of concerns associated with pesticide use on central Wisconsin potato farms.

The weight assigned the acute mammalian toxicity component is set at (0.5), reducing its significance relative to other component indices. This adjustment reflects the relative lack of circumstances leading to worker and applicator exposure and the low frequency of residues of acutely toxic pesticides in harvested potatoes, especially after washing, peeling, cooking and/or processing. The adjustment was not applied to the chronic mammalian toxicity index because low-level exposures are more widespread in the region from pesticides in drinking water, the air, and as a result of occupational exposure. To the limited extent the general public faces risks from pesticide residues in potatoes, they are likely to be chronic in nature.

A (1.5) weight has been assigned to the BioIPM component because BioIPM index impacts, especially resistance management and impacts on soil microorganisms and beneficial arthropods, are particularly important as Wisconsin potato producers progress along the EPM continuum toward more biologically based methods to manage pests. In recent years, secondary pests have been a recurrent concern and pest managers have invested considerable effort in devising and implementing resistance management plans. Efforts are also underway to build soil quality by raising organic matter content. Progress in enhancing soil quality is seen by many growers as critical in their efforts to improve nitrogen management efficiency, a key goal in reducing production costs.

Results using these four formulas are presented in columns 7, 8, 9, and 10 in Table 6 for pesticides used in potato production in 1995 in all states according to National Agricultural Statistics Survey (NASS) data. There are about 15 pesticides in Tables 4, 5, and 6 that were not used in Wisconsin in 1995, and hence these active ingredients do not appear in Tables 1, 2, 3, and 7. They are included in Table 6 to allow comparison with the pesticides used in other states by potato farmers.

Note the significant differences in the rankings of pesticides across the four formulas. Under the "Environmental Effects Focus Index," pesticides that score high on "Scaled Ecotoxicity" top the list - esfenvalerate at 695, carbofuran at 566, and phorate, 560. But under the "Health Effects Focus Index," esfenvalerate drops to 11th out of 18 insecticides, aldicarb rises to number one, and carbofuran drops to number 14. Under an environmental focus, carbaryl scores almost five times higher than in the health focus index. The botanical pyrethrins score about seven times higher.

Some fungicides, on the other hand, score significantly higher in the human health focus index than the environmental effects index. The differential is greatest in the case of metiram, which has a human health focus value of 439 (driven by its maximum value under chronic effects) and an environmental focus value of only 139. Among herbicides, all environmental focus index values are higher than health focus values mostly because of their relatively high "Scaled BioIPM Index" values. The environmental index value for the herbicide pendimethalin exceeds its human health focus value by 6.5 fold. While these indices are not accurate enough to express such differences to two significant digits, they are generally reliable in highlighting the often significant differences that exist in the relative hazards posed by different pesticides.

1. Attainment of WWF-WPVGA Project Risk Reduction Goals

WWF-WVGA risk reduction goals apply to high-risk pesticides that trigger an acute or chronic toxicity trigger, as shown in Table 7. In the case of acute risk, any active ingredient appearing in the WHO categories Class la, "Extremely Hazardous" and Class Ib, "Highly Hazardous" is subject to the 25 percent reduction goal between crop years 1995 and 1997. Four pesticides used in Wisconsin potato production in 1995 meet this criterion: azinphos-methyl, carbofuran, oxamyl, and methamidophos (see the fourth column in Table 7).

Another seven active ingredients fall under the chronic toxicity reduction goal, for a total of 11 pesticides out of a total of 31 applied in 1995. Any pesticide that is a known endocrine disrupter or a B2 carcinogen is subject to the 15 percent chronic toxicity reduction goal between 1995 and 1997.

The goal of the WWF-WPVGA project is to promote adoption of biointensive IPM as a means to reduce reliance on pesticides posing risk to humans, wildlife and IPM systems. To assure that risks are reduced comprehensively, the pounds applied of the11 active ingredients subject to the reduction targets are converted to Wisconsin Project Toxicity Units by multiplying pounds applied by multiattribute toxicity index values (shown in the third column in Table 7). Toxicity units are then summed across the four active ingredients in the case of acute risks, and seven active ingredients in the case of chronic risks. These totals are then divided by the acres planted - 83,000 - to produce an estimate of per acre planted toxicity units. The 25 percent acute and 15 percent chronic reduction goals are then applied to these estimates, as shown in the bottom portion of Table 7.

In 1995, there were about 25.9 million toxicity units associated with the four pesticides triggering the acute risk reduction criterion, or 312 per planted acre. To meet the reduction goal in crop season 1997, the toxicity units associated with active ingredients meeting the acute risk trigger must not exceed 234 per planted acre.

There were on average 1,769 chronic toxicity units per acre associated with pesticide use in 1995. In order to reach the 15 percent reduction target, the average number of toxicity units per acre associated with the application of pesticides meeting the chronic toxicity trigger must be reduced by 265, to 1,504 per acre in 1997.

It is important to note that the reduction goals apply to any and all pesticides applied in 1997 that meet one or both toxicity criteria, possibly including active ingredients not applied in 1995, and hence not contributing to the baseline of toxicity units.

3. Challenges and Next Steps

A number of activities are underway to collect better data and incorporate more sophisticated concepts and methods into the methodology used to calculate toxicity factor values. Some of the major areas in need of further work are -

Future Applications. With the benefit of these and other refinements, the system described herein will provide a solid basis to monitor progress toward biointensive IPM and pesticide risk reduction goals set as part of the WWF-WPVGA potato IPM project. Other partnerships involving farm groups and consumer and environmental organizations are likely to use or build on the system. Government agencies evaluating the impacts and accomplishments of IPM programs are also likely to make use of pesticide toxicity indices.

Current and future applications will provide valuable insights into long-term trends in pesticide toxicity levels, as well as the benefits of progress along the IPM continuum in reducing the average toxicity units associated with the production of major crops. One longer-term goal is establishment of an information base useful in forging consensus on future regulatory policies and research and education priorities. Crisper recognition of problems, constraints, and opportunities should help the nation take the steps needed to provide farmers with progressively safer and more effective IPM tools, and the public with safer and higher quality foods grown with fewer adverse impacts on the environment.

Bibliography and Further Information

Baril, A., and P. Mineau. 1998. "A Distribution-Based Approach to Improving Avian Risk Assessment," Canadian Wildlife Service (in press).

Baril, A., Jobin, B., Mineau, P., and B.T. Collins. 1994. "A Consideration of Inter-Species Variability in the Use of Lethal Dose (LD-50) in Avian Risk Assessment," Technical Series Report No. 216, Canadian Wildlife Service, Ottawa, Ontario.

Barnard, C., Daberkow, S., Padgitt, M., Smith, M.E., and N.D. Uri. 1997. "Alternative Measures of Pesticide Use," The Science of the Total Environment," Vol. 203: 229-244.

Benbrook, C. 1996a. "Adoption of Integrated Weed Management Systems by Com and Soybean Farmers in 1994: An Application of a New Methodology to Measure Adoption of IPM and Pesticide Use and Reliance," Paper presented at the Weed Science Society of America Annual Meeting, February, 1996, Norfolk, Virginia. Available on the Intemet at http://www.pmac.net Go to the World Wildlife Fund section in "Measuring IPM Adoption."

Benbrook, C. 1996b. "Primer on Pesticides Known to Disrupt the Endocrine System," Campaign for Pesticide Policy Reform, Washington, D.C..

Benbrook, C., Groth, E., Halloran, J., Hansen, M., and S. Marquardt. 1996. Pest Management at the Crossroads, Consumers Union, Yonkers, New York. See also PAFAC website at http://www.pmac.net

Burnet, M.W.M., Hart, Q., Holtum, J.A.M., and S.B. Powles. 1994. "Resistance to Nine Herbicide Classes in a Population of Rigid Ryegrass (Lolium rigidium)," Weed Science, Volume 42, Number 3: 369-377.

Colborn, T., vom Saal, F.S., and A.M. Soto. 1993. "Developmental Effects of Endocrine- Disrupting Chemicals in Wildlife and Humans," Environmental Health Perspectives, Volume 101, Number 5: 378-384.

Croft, B.A. & K. Theiling. 1990. "Pesticide effects on arthropod natural enemies: a database summary," Chap. 2. p. 1746. In Arthropod Biological Control Agents & Pesticides. Ed. B. A, Croft. J. Wiley, N.Y. 723 pp.

Gressel, J. 1997. Review of Glyphosate Resistance in Resistant Pest Management, Volume 8, Number 2, Winter 1996.

Hager, A.G. 1996. "Weed Resistance to Herbicides: Understanding How Resistance Develops in Weeds is the First Line of Defense," Weed Control Manual: Volume 30. Meister Publishing Company, Willoughby, Ohio.

Hoppin, P. 1996. "Reducing Pesticide Reliance and Risk Through Adoption of IPM: An Environmental and Agricultural Win-Win," Presentation February 27, 1996 before the Third National IPM Symposium/Workshop.

International Programme on Chemical Safety, "The WHO Recommended Classification of Pesticides by Hazard and Guidelines to Classification 1996-1997," World health Organization, Geneva, Switzerland, 1996).

Kovach, J., Petzoldt, C., Degnil, J. and J. Tette. 1992. "A Method to Measure the Environmental Impacts of pesticides," New York's Food and Life Sciences Bulletin, Volume 139:14.

Landy, D. 1995. "Multiattribute Ranking Systems for Pesticides," Masters of Science Thesis, Energy and Resources Department, University of California, Berkeley.

Leivtan, L. 1997. "An Overview of Pesticide Impact Assessment Systems (a.k.a. 'Pesticide Risk Indicators') based on Indexing or Ranking Pesticides by Environmental Impact," Background Paper, OECD, Paris, France, accessible on the Intemet at http://www.pmac.net/loispart.htm.

Lewis, W.J., van Lenteren, J.C., Phatak, S.C., and J.H. Tumlinson. 1997. "A Total System Approach to Sustainable Pest Management," Proc. Nat. Acad of Sci., Vol. 94, pp. 12243-12248, November 1997. (Accessible via the PNAS website at http://www.pnas.org, search for article by lead author's name).

Meister Publishing Company. 1998. "Farm Chemicals Handbook '98," Willoughby, Ohio.

Mineau, P., Jobin, B., and A. Baril. 1994. "A Critique of the Avian 5-Day Dietary Test (LC-50) as the Basis of Avian Risk Assessment," Technical Report Series No. 215, Canadian Wildlife Service, Ottawa, Ontario.

Montague, B. 1998. "EPA Ecotoxicological Database," available from EPA headquarters. Palling, E.D. 1992. "Evidence for Pesticide Synergism in the Honeybee (Apis mellifera)," Aspects of Applied Biology Vol. 31, pages 43-47.

Theiling, K.M., Croft, B. 1988. "Pesticide Side-Effects on Arthropod Natural Enemies: A Database Summary." Unpublished Manuscript, Oregon State University Department of Entomology.

Theiling K. & B.A. Croft. 1988. "Pesticide Effects on Arthropod Natural Enemies: A Database Summary," Agric., Ecosys. & Environ. 21:191-218.

Vandeman, A., Femanedez-Comejo, J., Jans, S. and B.H. Lin. 1994 Adoption of integrated Pest Management in US Agriculture. Agriculture Information Bulletin No. 707, Economic Research Service, Washington, D.C.

Table 1. Pesticides Applied in Wisconsin Potato Production, 1995
(Wisconsin Acres
Planted: 83,000)		 Area
Applied		 Acres
Treated		 Number of
Applications		 Rate per
Application		 Rate per
Crop year		 Total
Applied

(percent)
(number) (#s per acre) (#s per acre) (pounds)
Herbicides:





Metribuzin 8973,870 1.10.46 0.5239,000
Pendimethalin 36 29,880 1 0.8 0.81 24,000
Metolachlor 18 14,940 1 1.44 1.44 21,000
Linuron 9 7,470 1.1 0.89 0.97 7,0001
Glyphosate 8 6,640 1.0 0.62 0.62 4,0001
Sethoxydim 10 8,300 1.2 0.17 0.2 2,000
Total: All Herbicides 170 141,100


97,000
Per Planted Acre




1.17







Insecticides:





Methamidophos 65 53,950 1.4 0.88 1.26 169,000
Azinphos-methyl 26 21,580 2.1 0.57 1.19 26,000
Carbofuran 16 13,280 1 0.89 0.93 13,000
Dimethoate 28 23,240 1.3 0.38 0.48 11,000
Oxamyl 8 6,640 1.1 0.77 0.85 5,000
Permethrin 22 18,260 1.4 0.15 0.21 4,000
Esfenvalerate 60 49,800 1.7 0.04 0.06 3,000
Piperonyl butoxide 17 14,110 1 0.2 0.21 3,000
Pyrethrins 10 8,300 1 0.02 0.0216
Total: All
Insecticides		 318 263,940


194,016
Per Acres Planted




2.341







Fungicides:





Mancozeb 86 71,380 4.7 1.24 5.76 412,000
Chlorothalonil 88 73,040 5.9 0.95 5.61 408,000
Maneb 14 11,620 5.3 1.25 6.65 76,000
Copper hydroxide 38 31,540 2.4 0.54 1.26 40,000
Basic copper sulfate 5 4,150 1.9 1.57 2.92 13,000
Copper resinate 7 5,810 1.9 1.11 2.1 12,000
Triphenyltin hydrox. 46 38,180 2.9 0.11 0.31 12,000
Propamocarb hydroch. 12 9,960 1.1 0.9 0.96 9,000
Metalaxyl 15 12,450 1.5 0.21 0.31 4,000
Total: All
Fungicides		 311 258,130


986,000
Per Planted Acre




11.88







Other Chemicals:





Sulfuric acid 13 10,790 1.1 142 150.35 1,632,000
Metam-sodium 8 6,640 1 152 152 970,000
Diquat 80 66,400 1.4 0.3 0.42 28,000
Maleic hydrazide 8 6,640 1 1.91 1.91 13,000
Endothall 11 9,130 1.1 0.76 0.82 7,000
Paraquat 7 5,810 1.2 0.44 0.54 3,000
Total: Other Chemicals 127 105,410


2,653,000
Per Planted Acre




31.96







Total Herbicides,
Insecticides, and
Fungicides		 799 663,170


1,277,016
Per Planted Acre




15.39







Totals: All
Chemicals		 926 768,580


3,930,016
Per Planted Acre




47.35

============

Table 2. Pesticides Used in Wisconsin Potato Production by Acute Toxicity Units, 1995

(Wisconsin Acres Planted: 83,000)

Acres Treated

Total Pounds Applied

LD-50 Value

Scaled Inverse LD-50

Acute Toxicity Units

Share of Acute T.U. by Type of Pesticide
             

Herbicides:

            

Metribuzin

73,870

39,000

2,200

0.26

10,140

40.2%

Pendimethalin

29,880

24,000

1,050

0.42

10,057

39.9%

Metalochlor

14,940

21,000

2,780

0.17

3,570

14.2%

Linuron

7,470

7,000

4,000

0.11

770

3.1%

Glyphosate

6,640

4,000

4,230

0.1

400

1.6%

Sethoxydim

8,300

2,000

3,200

0.14

280

1.1%
             

Total: All herbicides

141,100

97,000

    

25,217

  

Per Planted Acres

  

1.17

    

0.3

  

Insecticides:

            

Methamidophos

53,950

69,000

30

14.6

1,007,400

29.3%

Carbuforan

13,280

13,000

8

68.56

891,280

25.9%

Azinphos-methyl

21,580

26,000

16

27.69

719,940

21.0%

Oxamyl

6,640

5,000

6

86.05

430,249

12.5%

Endosulfan

54,780

60,000

80

5.53

331,800

9.7%

Dimethoate

23,240

11,000

150

2.93

32,230

0.9%

Esfenvalerate

49,800

3,000

67

6.55

19,650

0.6%

Permethrin

18,260

4,000

500

0.88

3,520

0.1%

Piperonyl butoxide

14,110

3,000

5,000

0.09

270

0.0%

Pyrethrins

8,300

16

500

0.88

14

0.0%

Total: All Insecticides

263,940

194,016

    

3,436,353

  

Per Planted Acre

  

2.34

    

41

  
             

Fungicides:

            

Mancozeb

71,380

412,000

5,000

0.09

37,080

23.8%

Chlorothalonil

73,040

408,000

5,000

0.09

36,720

23.6%

Triphenyftin hydrox.

38,180

12,000

156

2.81

33,720

21.6%

Basic copper sulfate

4,150

13,000

300

1.46

18,980

12.2%

Copper hydroxide

31,540

40,000

1,000

0.44

17,600

11.3%

Maneb

11,620

76,000

5,000

0.09

6,840

4.4%

Metalaxyl

12,450

4,000

670

0.75

3,000

1.9%

Copper resinate

5,810

12,000

5,000

0.09

1,080

0.7%

Propamcarb hydroch

9,960

9,000

5,000

0.09

810

0.5%

Total: All Fungicides

258,130

986,000

    

155,830

  

Per Planted Acre

  

12

    

2

  

Other Chemicals:

            

Metam-sodium

6,640

970,000

285

1.54

1,493,800

64.0%

Sulfuric acid

10,790

1,632,000

1,000

0.44

718,080

30.7%

Endothall

9,130

7,000

51

8.62

60,340

2.6%

Diquat

66,400

28,000

231

1.90

53,200

2.3%

Paraquat

5,810

3,000

150

2.92

8,760

0.4%

Maleic hydrazide

6,640

13,000

5,000

0.11

1,430

0.1%

Total: Other Chemicals

105,410

2,653,000

    

2,335,610

  

Per Planted Acre

  

32

    

28

  
             

Per Planted Acre

  

15

    

44

  
             

All Chemicals:

768,580

3,930,016

    

5,953,010

  

Per Planted Acre

  

47.35

    

72

  

Table 3. Chronic Toxicity of Pesticides Used in Wisconsin Potato Production, 1995

(Wisconsin Acres Planted: 83,000)

Total Pounds Applied

Percent Pounds Applied

Scaled Mam Tox Score

Chronic Toxicity Units

Share of Chronic Toxicity Units by

Type of Pesticide

Herbicides:

          

Metribuzin

39,000

40%

26.77

1,044,139

89.9%

Linuron

7,000

7%

11.25

78,750

6.8%

Metolachlor

21,000

22%

0.93

19,530

1.7%

Pendimethalin

24,000

25%

 <