UC Berkeley

Technological advances in modern agriculture and the application of nitrogen-based synthetic fertilizers and manure to agricultural crops have increased crop yields and food production for the world's growing population. However, a significant portion of the applied nitrogen is in excess of crop needs. This results in leaching of nitrate into the groundwater and eutrophication of surface water systems via surface runoff. Agricultural pesticides are also required to maintain high levels of crop production. As a result of their inherent toxicity, they have adverse effects on the environment when they leave agricultural systems. Wetlands have been offered as Best Management Practices (BMPs) for the treatment of return flows from irrigated agriculture. To investigate nitrate removal kinetics in wetlands receiving agricultural drainage, field studies were conducted and nitrate removal efficiencies were determined in three agricultural wetland sites (Chapter 2). Microcosm studies were conducted to supplement field data and to provide insight into removal kinetics. The results suggest that wetlands constructed for purposes of wildlife habitat can remove nitrate from irrigation return flows, but efficiencies are typically low, with nitrate mass removal efficiencies ranging from 23% to 35% in wetlands examined in this study. Modified areal first-order removal rate constants (k) determined for field sites varied between 4.0 and 12.1 cm d-1. Microcosm studies were used to supplement field studies and to determine saturation kinetics which was practically impossible to measure in the field. The first order nitrate removal rate for the microcosm (12.97 cm d-1) was approximately equal to the observed k-value from Ramona Lake, the source of the sediments used in the microcosm, suggesting the value of microcosms for estimating nitrate removal rates in wetlands. Measurement of saturation kinetics in the microcosm system showed that the apparent half-saturation constant (Km) and maximum removal rate (Jmax) were 43.8 mg L-1 and 4.1 g m-2 d-1 respectively, for these sediments. Estimates of land requirements for wetlands in one agricultural watershed indicated that less than 3 % of the watershed area would need to be devoted to wetlands to achieve an effluent nitrate concentration of 0.5 mg L-1, the target value for limiting the growth of nuisance algae.

Chlorpyrifos is the most widely used organophosphate insecticide in California's San Joaquin Valley and is widely used elsewhere. While several prior studies evaluated the effectiveness of different best management practices (BMPs) for chlorpyrifos mitigation, these studies have mostly focused on sorption of chlorpyrifos to wetland sediments and soils with removal efficiency assessed by measuring inlet and outlet concentrations. To assess the long-term performance of wetlands, it is also important to know the ultimate fate of chlorpyrifos in wetland sediments. Specifically, particle-associated pesticides stored in the sediments can be transported via runoff and other processes to surface water systems. Three different phosphoesterase enzymes; phosphomonoesterase, phosphodiesterase and phosphotriesterase, are involved in chlorpyrifos biotransformation pathway; however, the link between these enzymes and chlorpyrifos biotransformation rates had not been previously addressed. The research presented in this dissertation demonstrated that wetland sites showed temporal and spatial variation in observed chlorpyrifos biotransformation rates, with half-lives ranging from 1 to 35 days under aerobic conditions (Chapter 3). Chlorpyrifos transformation slowed significantly under anaerobic conditions, with a half-life of approximately 92 days. Biodegradation rates decreased significantly in sediments from the Hospital Creek site during 2011 due to flooded conditions that preceded sample collection. These results suggest that allowing a wet-dry cycle can enhance the transformation rates of an organophosphate insecticide in these systems by providing aerobic conditions in sediments. The dry phase would encompass the non-irrigation season in late fall and winter, and the wetland would be flooded again in spring and summer when the irrigation season begins. There was significant correlation between phosphotriesterase activity and the chlorpyrifos biotransformation rates, with this relationship varying among sites. Phosphotriesterase activities may be useful as an indicator of biodegradation potential with reference to the previously established site-specific correlations.

In addition, kinetic parameters obtained in the laboratory studies were used to model agricultural non-point source pollution in California's San Joaquin River (SJR) watershed using a previously developed water quality model WARMF (Watershed Analysis Risk Management Framework) (Chapter 4). The results of the nitrate simulations suggest that a wetland area about 1.8% of the agricultural land in the Orestimba Creek watershed could significantly reduce nitrate concentrations supporting our results from Chapter 2. The results of the chlorpyrifos simulations underlined the importance of the management strategies to enhance the biotransformation rates. The scenario using enhanced biotransformation rates was more effective at reducing chlorpyrifos concentrations to values below regulatory limits compared to the 30% chlorpyrifos use reduction scenario. From a management perspective of view, use reduction should be implemented together with other best management practices if possible, especially in impaired surface waters. Given the difficulty with completely eliminating organophosphate pesticide use in agriculture under present conditions, management strategies for enhanced organophosphate pesticide removal are of crucial importance in efforts to keep the organophosphate pesticide concentrations within regulatory limits.