A hybrid machine learning model to predict and visualize nitrate concentration throughout the Central Valley aquifer, California, USA

Abstract

Intense demand for water in the Central Valley of California and related increases in groundwater nitrate concentration threaten the sustainability of the groundwater resource. To assess contamination risk in the region, we developed a hybrid, non-linear, machine learning model within a statistical learning framework to predict nitrate contamination of groundwater to depths of approximately 500m below ground surface. A database of 145 predictor variables representing well characteristics, historical and current field and landscape-scale nitrogen mass balances, historical and current land use, oxidation/reduction conditions, groundwater flow, climate, soil characteristics, depth to groundwater, and groundwater age were assigned to over 6000 private supply and public supply wells measured previously for nitrate and located throughout the study area. The boosted regression tree (BRT) method was used to screen and rank variables to predict nitrate concentration at the depths of domestic and public well supplies. The novel approach included as predictor variables outputs from existing physically based models of the Central Valley. The top five most important predictor variables included two oxidation/reduction variables (probability of manganese concentration to exceed 50ppb and probability of dissolved oxygen concentration to be below 0.5ppm), field-scale adjusted unsaturated zone nitrogen input for the 1975 time period, average difference between precipitation and evapotranspiration during the years 1971-2000, and 1992 total landscape nitrogen input. Twenty-five variables were selected for the final model for log-transformed nitrate. In general, increasing probability of anoxic conditions and increasing precipitation relative to potential evapotranspiration had a corresponding decrease in nitrate concentration predictions. Conversely, increasing 1975 unsaturated zone nitrogen leaching flux and 1992 total landscape nitrogen input had an increasing relative impact on nitrate predictions. Three-dimensional visualization indicates that nitrate predictions depend on the probability of anoxic conditions and other factors, and that nitrate predictions generally decreased with increasing groundwater age.

Analysis of Groundwater Nitrate Contamination in the Central Valley: Comparison of the Geodetector Method, Principal Component Analysis and Geographically Weighted Regression

“…Recently, machine learning methods have been used to predict the groundwater nitrate contamination in the CV [27,28]. Nolan et al applied a statistical learning framework to domestic wells data to optimize the over prediction of three machine learning methods: Artificial Neural Network (ANN), Boosted Regression Tree (BRT) and Bayesian Network (BN).…”

Assessment of Groundwater Nitrate Pollution Potential in Central Valley Aquifer Using Geodetector-Based Frequency Ratio (GFR) and Optimized-DRASTIC Methods

“…Ransom et al [69] also found higher nitrate concentration in wells of San Joaquin and Tulare Basin than Sacramento valley. The Sacramento Valley is composed of fine grained sediments compared to southern part of the CV, which could reduce the susceptibility of Sacramento Valley.…”

A Bayesian Approach to Infer Nitrogen Loading Rates from Crop and Landuse Types Surrounding Private Wells in the Central Valley, California

“…A study focused on the TLB (within the CV) geospatially related areas near rivers with lower nitrate concentra-35 tion in wells. The study highlighted areas where major rivers flow into the TLB from the Sierra Nevada Mountains and found that these areas were also characterized by wells with lower relative nitrate concentrations (Boyle et al, 2012 analysis of CV groundwater nitrate recently confirmed that proximity to major streams is a significant controlling factor for a wells' nitrate concentration (Ransom et al, 2017). Boyle et al (2012) point out that agricultural areas near rivers are also more likely to receive surface water irrigations.…”

“…Nitrate loading concentrations vary within specific crop or landuse 35 on shallow private wells for drinking and household purposes. Private wells are not regulated in California and it is difficult to know how many may be contaminated by nitrate: a significant portion of the CV has been estimated to contain shallow groundwater with elevated nitrate concentration (Nolan et al, 2014;Lockhart et al, 2013;Ransom et al, 2017). …”

“…These include dilution with older groundwater, mixing with recharge water from streams, and denitrification or ammonium volatilization in the vadose zone or in groundwater (Ransom et al, 2017). Dilution of nitrogen in recharge water is most likely to occur through mixing along the well screen with older, low nitrogen containing, water.…”