UC Davis

Nitrate contamination in groundwater, driven by nitrogen intensive cropping practices and low nitrogen use efficiency (NUE), poses a significant risk to water quality in California's Central Valley. This study presents the calibration of a vadose zone crop model to simulate nitrate (NO3-N) leaching in response to a best management practice (BMP), consisting of High Frequency Low Concentration (HFLC) fertilizer applications, aimed at improving NUE in a 57-ha commercial almond orchard in Modesto, CA. Using a comprehensive dataset from landscape monitoring and vadose zone observations, a novel approach is taken to optimize soil hydraulic parameters, root solute uptake parameters, and atmospheric boundary conditions. The water balance is calibrated separately from model parameters by comparing ensemble estimates of precipitation and evapotranspiration (ET) from multiple sources with measured changes in soil moisture storage. Simulations are validated using groundwater NO3-N concentrations measured from a dense network of 20 monitoring wells installed in the orchard. Results indicate that simulated leaching concentrations and travel times are most sensitive to changes in the water balance driven by uncertainties in the measurement of ET and precipitation. Observed heterogeneity of NO3-N in groundwater was explained by sub-orchard scale spatial variability in NUE and cycles of orchard-block tree replantings. Soil-type heterogeneity affected modeled solute travel times but did not significantly affect measured NO3-N groundwater concentrations. Simulations indicated the BMP can reduce NO3-N leaching concentrations by up to 40%, with observable benefits in groundwater expected over a 10-year period due to slow solute movement in the vadose zone and mixing in groundwater.