Search

Dissolved oxygen (DO) is a critical water quality constituent that governs habitat suitability for aquatic biota, biogeochemical reactions and solubility of metals in streams. Recently introduced high‐frequency sensors have increased our ability to measure DO, but we still lack the capacity to understand and predict DO concentrations at high spatial resolutions or in unmonitored locations. Machine learning (ML) has been a commonly used approach for modelling DO, however, conventional ML models have no representation of the limnological processes governing DO dynamics. Here we implement and evaluate two process‐guided deep learning (PGDL) approaches for predicting daily minimum, mean and maximum DO concentrations in rivers from the Delaware River Basin, USA. In both cases, a multi‐task approach was taken in which the PGDL models predicted stream metabolism and gas exchange rates in addition to the DO concentrations themselves. Our results showed that for these sites, the PGDL approaches did not improve upon baseline predictions in temporal and spatially similar holdout experiments. One of the approaches did, however, improve predictions when applied to spatially dissimilar sites. Although this particular PGDL approach did not improve predictive accuracy in most cases, our results suggest that process guidance, perhaps a more constrained approach, could benefit a data‐driven DO model. [ABSTRACT FROM AUTHOR]

Salinity dynamics in the Delaware Bay estuary are a critical water quality concern as elevated salinity can damage infrastructure and threaten drinking water supplies. Current state‐of‐the‐art modeling approaches use hydrodynamic models, which can produce accurate results but are limited by significant computational costs. We developed a machine learning (ML) model to predict the 250 mg L−1 Cl− isochlor, also known as the "salt front," using daily river discharge, meteorological drivers, and tidal water level data. We use the ML model to predict the location of the salt front, measured in river miles (RM) along the Delaware River, during the period 2001–2020, and we compare predictions of the ML model to the hydrodynamic Coupled Ocean–Atmosphere‐Wave‐Sediment Transport (COAWST) model. The ML model predicts the location of the salt front with greater accuracy (root mean squared error [RMSE] = 2.52 RM) than the COAWST model does (RMSE = 5.36); however, the ML model struggles to predict extreme events. Furthermore, we use functional performance and expected gradients, tools from information theory and explainable artificial intelligence, to show that the ML model learns physically realistic relationships between the salt front location and drivers (particularly discharge and tidal water level). These results demonstrate how an ML modeling approach can provide predictive and functional accuracy at a significantly reduced computational cost compared to process‐based models. In addition, these results provide support for using ML models in operational forecasting, scenario testing, management decisions, hindcasting, and resulting opportunities to understand past behavior and develop hypotheses. [ABSTRACT FROM AUTHOR]

Managed aquifer recharge (MAR) can increase groundwater supply and, under some conditions, improve groundwater quality simultaneously. However, the conditions under which water quality improvements can be achieved during infiltration for MAR have not been systematically examined in a spatially explicit manner. This work aims to address that gap by synthesizing observations from laboratory tests, field experiments and operational facilities at four MAR sites within the Pajaro Valley, California, USA to develop a predictive model of nitrate removal during infiltration. We compare two machine learning approaches; random forest and boosted regression trees, to multiple linear regression. The preferred model uses boosted regression trees, with four predictor variables (initial nitrate concentration of infiltrating water, residence time in the soil, soil organic carbon content, and soil percent clay). We apply this model to simulate the spatial distribution of potential nitrate removal (NRp) across a heterogeneous and mixed‐use landscape, finding that >87% of the region has the potential to remove nitrate during infiltration. To link potential nitrate removal capacity to available water for MAR, we combine a map of NRp with independently simulated hillslope runoff, and find that potential load reduction is highest in urban areas (median = 18.8 kg‐N/yr) where large runoff volumes are collocated with soils having high nutrient cycling capacity compared to forested and agricultural areas (median = 1.6 and 3.5 kg‐N/yr, respectively). We analyse potential load reduction across dry, normal, and wet precipitation scenarios, which shows that urban areas have the potential to yield large load reductions (median = 11.3 kg‐N/yr) even under relatively dry conditions. However, much of the generation of elevated nitrate in runoff is associated with agricultural activity. These results suggest that the urban–agricultural interface represents a crucial link between hydrologic and biogeochemical cycles where water supply and quality goals may be met simultaneously. The technical approach used in this study is highly flexible and can help guide decisions in resource management and identify promising MAR sites. More broadly, this approach also elucidates heterogeneity in biogeochemical cycling across complex landscapes. [ABSTRACT FROM AUTHOR]