Environmental Risk Indicator For Pesticides

The use of pesticides has been a major concern in environmental health. To properly assess the impact of pesticide use on the environment, hundreds of pesticide environmental risk indicators were developed throughout the world. This study evaluated and compared qualitatively and quantitatively ten popular pesticide environmental risk indicators models, with the expectations to offer guidance in the application and development of these environmental risk indicators. The compartments of surface water, groundwater, soil and/or air were taken into account, and the methodologies utilized in the ten environmental risk indicators had large variations. The associated results differed from indicator to indicator which may have affected by their related approaches. For both the aggregated and separate risk indices, similar methodologies generally resulted in closer outputs. Surface water was the most frequently studied compartment, which was included in all the ten indicators. Additionally, a new validation attempt was proposed, using three banned or high restricted pesticides (i.e. lindane, parathion and parathion-methyl) as "marked pesticides" to validate the indicators. The results generated by some of the ten indicators showed that these three pesticides had the highest risks. Furthermore, this thesis also discussed three major issues: filling in data gaps, aggregating risks of separate compartments, and incorporating application rate. Several major features (i.e. coherent, modularized, user-friendly and extendable) were recommended as the direction for future development of pesticide environmental risk indicators.By comparing and modifying the existing indicators, the thesis proposed a new indicator which was more transparent, flexible and extendable for assessing pesticide environmental risks. Through the literature review and the results of the indicator comparisons, it was recognized that risk ratio methodology and fuzzy logic expert system can be best used to address the impact on non-target groups. Risk ratio methodology was appealing because the relationship between predicted concentration and toxicity concentration could be evaluated mathematically in similar units; for non-target groups where existing data and/or simulation models were not considered feasible, the fuzzy logic expert system will be employed as a robust method of aggregation of dissimilar, uncertain and imprecise data. The non-target groups including long and short term impact on soil, long and short term impact on surface water through both drift and runoff, and impact to the operator, picker, mammals, birds, bees, beneficial insects, and resident (risk from groundwater) could be analyzed by the proposed methods. For all organisms (operator, picker, mammals, birds, bees, and beneficial insects), both acute and chronic impact would be evaluated.