Gary M. Davis

Chloramination of organophosphorus pesticides found in drinking water sources

The degradation of commonly detected organophosphorus (OP) pesticides, in drinking water sources, was investigated under simulated chloramination conditions. Due to monochloramine autodecomposition, it is difficult to observe the direct reaction of monochloramine with each OP pesticide. Therefore, a model was developed to examine the reaction of monochloramine (NH(2)Cl) and dichloramine (NHCl(2)) with chlorpyrifos (CP), diazinon (DZ), and malathion (MA). Monochloramine was found not to be very reactive with each OP pesticides, (k)NH(2)Cl,OP = 11-21 M(-1)h(-1). While, dichloramine (NHCl(2)) was found to be 2 orders of magnitude more reactive with each of the OP pesticides than monochloramine, (k)NHCl(2),OP = 2000-2900 M(-1)h(-1), which is still three orders of magnitude less than the hypochlorous acid reaction rate coefficient with each OP pesticide. For each pesticide, the reactivity of the three chlorinated oxidants was then found to correlate with half-wave potentials (E(1/2)) of each oxidant. With reaction rate coefficients for the three chlorinated oxidations as well as neutral and alkaline hydrolysis rate coefficients for the pesticides, the model was used to determine the dominant reaction pathways as a function of pH. At pH 6.5, OP pesticide transformation was mostly due to the reaction of hypochlorous acid and dichloramine. Above pH 8, alkaline hydrolysis or the direct reaction with monochloramine was the primary degradation pathway responsible for the transformation of OP pesticides. This demonstrates the ability of models to be used as tools to elucidate degradation pathways and parameterize critical reaction parameters when used with select yet comprehensive data sets.

Climate change and watershed mercury export: a multiple projection and model analysis

Future shifts in climatic conditions may impact watershed mercury (Hg) dynamics and transport. An ensemble of watershed models was applied in the present study to simulate and evaluate the responses of hydrological and total Hg (THg) fluxes from the landscape to the watershed outlet and in-stream THg concentrations to contrasting climate change projections for a watershed in the southeastern coastal plain of the United States. Simulations were conducted under stationary atmospheric deposition and land cover conditions to explicitly evaluate the effect of projected precipitation and temperature on watershed Hg export (i.e., the flux of Hg at the watershed outlet). Based on downscaled inputs from 2 global circulation models that capture extremes of projected wet (Community Climate System Model, Ver 3 [CCSM3]) and dry (ECHAM4/HOPE-G [ECHO]) conditions for this region, watershed model simulation results suggest a decrease of approximately 19% in ensemble-averaged mean annual watershed THg fluxes using the ECHO climate-change model and an increase of approximately 5% in THg fluxes with the CCSM3 model. Ensemble-averaged mean annual ECHO in-stream THg concentrations increased 20%, while those of CCSM3 decreased by 9% between the baseline and projected simulation periods. Watershed model simulation results using both climate change models suggest that monthly watershed THg fluxes increase during the summer, when projected flow is higher than baseline conditions. The present studys multiple watershed model approach underscores the uncertainty associated with climate change response projections and their use in climate change management decisions. Thus, single-model predictions can be misleading, particularly in developmental stages of watershed Hg modeling.