An official website of the United States government
Abstract Midwestern cities require forecasts of surface nitrate loads to bring additional treatment processes online or activate alternative water supplies. Concurrently, networks of nitrate monitoring stations are being deployed in river basins, co‐locating water quality observations with established stream gauges. However, tools to evaluate the future value of expanded networks to improve water quality forecasts remains challenging. Here, we construct a synthetic data set of stream discharge and nitrate for the Wabash River Basin—one of the United States’ most nutrient polluted basins—using the established Agro‐IBIS and THMB models. Synthetic data enables rapid, unbiased and low‐cost assessment of potential sensor placements to support management objectives, such as near‐term forecasting. Using the synthetic data, we established baseline 1‐day forecasts for surface water nitrate at 12 cities in the basin using support vector machine regression (SVMR; RMSE 0.48–3.3 ppm). Next, we used the SVMRs to evaluate the improvement in forecast performance associated with deployment of additional nitrate sensors. We identified the optimal sensor placement to improve forecasts at each city, and the relative value of sensors at each candidate location. Finally, we assessed the co‐benefit realized by other cities when a sensor is deployed to optimize a forecast at one city, finding significant positive externalities in all cases. Ultimately, our study explores the potential for machine learning to make near‐term predictions and critically evaluate the improvement realized by expanding a monitoring network. While we use nitrate pollution in the Wabash River Basin as a case study, this approach could be readily applied to any problem where the future value of sensors and network design are being evaluated.
more »
« less
This content will become publicly available on August 1, 2026
Deep Learning Prediction and Interpretation of Riverine Nitrate Export Across the Mississippi River Basin
Excess riverine nitrate causes downstream eutrophication, notably in the Gulf of Mexico where hypoxia is linked to nutrient-rich discharge from the Mississippi River Basin (MRB). We developed a long short-term memory (LSTM) model using high-frequency sensor data from across the conterminous US to predict daily nitrate concentrations, achieving strong temporal validation performance (median KGE = 0.60). Spatial validation—or prediction in unmonitored basins—yielded lower performance for nitrate concentration (median KGE = 0.18). Nonetheless, spatial validation was crucial in quantifying the impact of current data gaps and guiding the model's targeted application to the MRB where spatial validation performance was stronger (median KGE = 0.34). Modeling results for the MRB from 1980 to 2022 showed relatively low riverine nitrate export (19 ± 4% of surplus), indicating large-scale retention of surplus nitrate within the MRB. Interannual nitrate yields varied significantly, especially in Midwestern states like Iowa, where wet-year export fractions (42 ± 24%) far exceeded dry year export (6 ± 6%), suggesting increased hydrologic connectivity and remobilization of legacy nitrogen. Further evidence of legacy nitrate remobilization was noted in a subset of Midwestern basins where, on occasion, annual surplus export fractions exceeded 100%. Interpretable Shapley values identified key spatial drivers influencing mean nitrate concentrations—tile drainage, roadway density, wetland cover—and quantitative, non-linear thresholds in their influence, offering management targets. This study leverages machine learning and aquatic sensing to provide improved spatiotemporal predictions and insights into nitrate drivers, thresholds, and legacy impacts, offering valuable information for targeted nutrient management strategies in the MRB.
more »
« less
- PAR ID:
- 10635654
- Publisher / Repository:
- John Wiley & Sons Ltd
- Date Published:
- Journal Name:
- Water Resources Research
- Volume:
- 61
- Issue:
- 8
- ISSN:
- 0043-1397
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
It is essential to identify the dominant flow paths, hot spots and hot periods of hydrological nitrate-nitrogen (NO3-N) losses for developing nitrogen loads reduction strategies in agricultural watersheds. Coupled biogeochemical transformations and hydrological connectivity regulate the spatiotemporal dynamics of water and NO3-N export along surface and subsurface flows. However, modeling performance is usually limited by the oversimplification of natural and human-managed processes and insufficient representation of spatiotemporally varied hydrological and biogeochemical cycles in agricultural watersheds. In this study, we improved a spatially distributed process-based hydro-ecological model (DLEM-catchment) and applied the model to four tile-drained catchments with mixed agricultural management and diverse landscape in Iowa, Midwestern US. The quantitative statistics show that the improved model well reproduced the daily and monthly water discharge, NO3-N concentration and loading measured from 2015 to 2019 in all four catchments. The model estimation shows that subsurface flow (tile flow + lateral flow) dominates the discharge (70%-75%) and NO3-N loading (77%-82%) over the years. However, the contributions of tile drainage and lateral flow vary remarkably among catchments due to different tile-drained area percentages and the presence of farmed potholes (former depressional wetlands that have been drained for agricultural production). Furthermore, we found that agricultural management (e.g. tillage and fertilizer management) and catchment characteristics (e.g. soil properties, farmed potholes, and tile drainage) play important roles in predicting the spatial distributions of NO3-N leaching and loading. The simulated results reveal that the model improvements in representing water retention capacity (snow processes, soil roughness, and farmed potholes) and tile drainage improved model performance in estimating discharge and NO3-N export at a daily time step, while improvement of agricultural management mainly impacts NO3-N export prediction. This study underlines the necessity of characterizing catchment properties, agricultural management practices, flow-specific NO3-N movement, and spatial heterogeneity of NO3-N fluxes for accurately simulating water quality dynamics and predicting the impacts of agricultural conservation nutrient reduction strategies.more » « less
-
Kaplan, J (Ed.)The Mississippi River Basin (MRB), the fourth-largest river basin in the world, is an important corridor for hy- droelectric power generation, agricultural and industrial production, riverine transportation, and ecosystem goods and services. Historically, flooding of the Mississippi River has resulted in significant economic losses. In a future with an intensified global hydrological cycle, the altered discharge of the river may jeopardize commu- nities and infrastructure situated in the floodplain. This study utilizes output from the Community Earth System Model version 2 (CESM2) large ensemble simulations spanning 1930 to 2100 to quantify changes in future MRB discharge under a high greenhouse gas emissions scenario (SSP3–7.0). The simulations show that increasing precipitation trends exceed and dominate increased evapotranspiration (ET), driving an overall increase in total discharge in the Ohio and Lower Mississippi River basins. On a seasonal scale, reduced spring snowmelt is projected in the Ohio and Missouri River basins, leading to reduced spring runoff in those regions. However, decreased snowmelt and spring runoff is overshadowed by a larger increase in projected precipitation minus ET over the entire basin and leads to an increase in mean river discharge. This increase in discharge is linked to a relatively small increase in the magnitude of extreme floods (2 % and 3 % for 100-year and 1000-year floods, respectively) by the late 21st century relative to the late 20th century. Our analyses imply that under SSP3–7.0 forcing, the Mississippi River and Tributaries (MR&T) project design flood would not be exceeded at the 100-year return period. Our results harbor implications for water resources management including increased vulnerability of the Mississippi River given projected changes in climate.more » « less
-
Accurate hydrologic modeling is vital to characterizing how the terrestrial water cycle responds to climate change. Pure deep learning (DL) models have been shown to outperform process-based ones while remaining difficult to interpret. More recently, differentiable physics-informed machine learning models with a physical backbone can systematically integrate physical equations and DL, predicting untrained variables and processes with high performance. However, it is unclear if such models are competitive for global-scale applications with a simple backbone. Therefore, we use – for the first time at this scale – differentiable hydrologic models (full name δHBV-globe1.0-hydroDL, shortened to δHBV here) to simulate the rainfall–runoff processes for 3753 basins around the world. Moreover, we compare the δHBV models to a purely data-driven long short-term memory (LSTM) model to examine their strengths and limitations. Both LSTM and the δHBV models provide competitive daily hydrologic simulation capabilities in global basins, with median Kling–Gupta efficiency values close to or higher than 0.7 (and 0.78 with LSTM for a subset of 1675 basins with long-term discharge records), significantly outperforming traditional models. Moreover, regionalized differentiable models demonstrated stronger spatial generalization ability (median KGE 0.64) than a traditional parameter regionalization approach (median KGE 0.46) and even LSTM for ungauged region tests across continents. Nevertheless, relative to LSTM, the differentiable model was hampered by structural deficiencies for cold or polar regions, highly arid regions, and basins with significant human impacts. This study also sets the benchmark for hydrologic estimates around the world and builds a foundation for improving global hydrologic simulations.more » « less
-
Abstract In contrast to its productive coastal margins, the open-ocean Gulf of Mexico (GoM) is notable for highly stratified surface waters with extremely low nutrient and chlorophyll concentrations. Field campaigns in 2017 and 2018 identified low rates of turbulent mixing, which combined with oligotrophic nutrient conditions, give very low estimates for diffusive flux of nitrate into the euphotic zone (< 1 µmol N m−2d−1). Estimates of local N2-fixation are similarly low. In comparison, measured export rates of sinking particulate organic nitrogen (PON) from the euphotic zone are 2 – 3 orders of magnitude higher (i.e. 462 – 1144 µmol N m−2d−1). We reconcile these disparate findings with regional scale dynamics inferred independently from remote-sensing products and a regional biogeochemical model and find that laterally-sourced organic matter is sufficient to support >90% of open-ocean nitrogen export in the GoM. Results show that lateral transport needs to be closely considered in studies of biogeochemical balances, particularly for basins enclosed by productive coasts.more » « less
