In the race to enhance agricultural productivity, irrigation will become more dependent on poorly characterized and virtually unmonitored sources of water. Increased use of irrigation water has led to impaired water and soil quality in many areas. Historically, soil salinization and reduced crop productivity have been the primary focus of irrigation water quality. Recently, there is increasing evidence for the occurrence of geogenic contaminants in water. The appearance of trace elements and an increase in the use of wastewater has highlighted the vulnerability and complexities of the composition of irrigation water and its role in ensuring proper crop growth, and long-term food quality. Analytical capabilities of measuring vanishingly small concentrations of biologically-active organic contaminants, including steroid hormones, plasticizers, pharmaceuticals, and personal care products, in a variety of irrigation water sources provide the means to evaluate uptake and occurrence in crops but do not resolve questions related to food safety or human health effects. Natural and synthetic nanoparticles are now known to occur in many water sources, potentially altering plant growth and food standard. The rapidly changing quality of irrigation water urgently needs closer attention to understand and predict long-term effects on soils and food crops in an increasingly fresh-water stressed world.

A four-decade dataset (1974-2013) of 107,823 nitrate samples in 25,993 wells from western and eastern parts of Nebraska was used to assess long-term trends of groundwater nitrate concentration and decadal changes in the extent of groundwater nitrate-contaminated areas (NO3-N ≥ 10 mg N/L) over the entire state. Spatial statistics and regressions were used to investigate the relationships between groundwater nitrate concentrations and several potential natural and anthropogenic factors, including soil drainage capacities, vadose zone characteristics, crop production areas, and irrigation systems. The results of this study show that there is no statistically significant trend in groundwater nitrate concentrations in western Nebraska, in contrast with the increasing trend (p < .05) to the east. The spatial extent and nitrate concentrations in contaminated groundwater in center pivot-irrigated areas was less than in gravity-irrigated areas. Areas with a thicker vadose zone and larger saturated thickness of the aquifer have relatively lower nitrate concentrations. The results of a classification and regression tree (CART) model indicate the difference in the influence of physical factors on groundwater nitrate concentrations between western and eastern Nebraska, namely that groundwater nitrate concentrations correspond with vadose zone thickness, effective hydraulic conductivity, and saturated thickness in the west, while in eastern Nebraska, concentrations are correlated with average percent sand in the topsoil (0-150 cm), well depth, and effective hydraulic conductivity.

Assessment of pesticide and nitrate contamination of rural private wells was conducted using artificial neural networks. Data from 192 drilled and driven wells and 115 large-diameter dug and bored wells, available from two earlier studies, were used for training and testing purposes. Four separate models, two for the two well types and one each for pesticide and nitrate, were developed. Parameters such as depth to aquifer material from land surface, well depth, and distance to cropland were used as input parameters, and the concentrations of nitrate or pesticides were the outputs. While the training efficiency of the network reached between 95 and 100% for these four models, the prediction accuracy for the four models ranged from a low of slightly above 50% for nitrate in dug and bored wells to a high of 90% for pesticides in drilled and driven wells. Sensitivity analyses were performed to examine the impact of network architecture, training and testing parameters, and the size and type of input parameters on model predictions. Multiple hidden layers with a large number of nodes did not appear to have a significant impact on model predictions in two of the four models. The relative importance of input parameters was tested by adding or removing certain key parameters to the model, and it was observed that the parameters had a different impact on drilled/driven versus dug/bored wells.