Sensitivity analysis of pesticides contaminating groundwater by applying probability and transport methods



The use of pesticides is a potential threat to local groundwater. Once groundwater is contaminated, it is very difficult to clean. Thus, it is of importance to assess the risk of contaminating local groundwater at an early stage when pesticides are found in soils. This knowledge will also help in remediation strategies. Traditional methods of deterministic analysis cannot explicitly account for the sometimes large uncertainties that exist at this stage in the work, whereas probabilistic analyses are better suited for dealing with these problems. In this paper, we have combined contaminant transport with a 1st-order reliability approach. Pesticide concentrations in soil have been studied to estimate the probability of failure—that is, of pesticides exceeding established critical levels in groundwater. Results indict that failure probability increases rapidly within a certain range of pesticide concentrations in soil for different critical levels. In given aquifer conditions and contaminants, probabilities of contaminants exceeding particular critical levels can easily be obtained according to various water usage scenarios. The distribution of importance factors among variables indicates the contribution their relative weights make to the failure probability. Hence, authorities can easily form sensitivity factors to take action and reduce the risk of contaminating the groundwater.