Extreme event analysis of pesticide loads to surface waters

Surface water pollution by pesticide runoff is influenced by random hydrologie events. Risk assessment of such pollution is based on estimates of the probabilities of large or extreme pes ticide runoff loads. A previous paperl described a general Monte Carlo simulation approach for generating pesticide runoff esti mates. The approach was used to determine mean annual pes ticide loads and probability distributions of annual and monthly loads for two pesticides in Georgia and Iowa. However, annual and monthly loads may understate the severity of a pesticide pollution problem, particularly for streams or small rivers. A large runoff even may produce a sudden shock load of chemical which severely impacts aquatic life. Risk assessment of these situations requires probabilities estimates of short-term (daily) pesticide runoff. This paper extends the Monte Carlo approach to the analysis of daily pesticide runoff. The work has three objectives: to il lustrate Monte Carlo simulation to produce information for pesticide pollution risk assessment; to estimate the magnitudes of extreme pesticide runoff events for a range of chemicals and locations; and to determine the general forms of probability dis tributions for annual and monthly maximum daily pesticide runoff loads.