Groundwater is the largest source of Earth’s liquid freshwater and plays a critical role in global food security. With the rising global demand for drinking water and increased agricultural production, overuse of groundwater resources is a major concern. Because groundwater withdrawals are not monitored in most regions with the highest use, methods are needed to monitor withdrawals at a scale suitable for implementing sustainable management practices. In this study, we combine publicly available datasets into a machine learning framework for estimating groundwater withdrawals over the state of Arizona. This extends a previous study in which we estimated groundwater withdrawals in Kansas, where the climatic conditions and aquifer characteristics are significantly different. Datasets used in our model include energy-balance (SSEBop) and crop coefficient evapotranspiration estimates, precipitation(PRISM), and land-use (USDA-NASS Cropland Data Layer), and a watershed stress metric. Random forests, a widely popular machine learning algorithm, are employed for predicting groundwater withdrawals from 2002-2018 at 5 km spatial resolution. We used in-situ groundwater withdrawals available over the Arizona Active Management Area (AMA) and Irrigation Non-Expansion Area (INA) from 2002-2010 for training and 2011-2018 for validating the model respectively. The results show high training (R2 ≈ 0.98) and good testing (R2 ≈ 0.82) scores with low normalized mean absolute error ≈ 0.28 and root mean square error ≈ 1.28 for the AMA/INA region. Using this method, we are able to spatially extend estimates of groundwater withdrawals to the whole state of Arizona. We also observed that land subsidence in Arizona is predominantly occurring in areas having high yearly groundwater withdrawals of at least 100 mm per unit area. Our model shows promising results in sub-humid and semi-arid (Kansas) and arid regions (Arizona), which proves the robustness and extensibility of our integrated approach combining remote sensing and machine learning into a holistic, automated, and fully-reproducible workflow. The success of this method indicates that it could be extended to areas with more limited groundwater withdrawal data under different climatic conditions and aquifer properties.

Effective monitoring of groundwater withdrawals is necessary to help mitigate the negative impacts of aquifer depletion. In this study, we develop a holistic approach that combines water balance components with a machine learning model to estimate groundwater withdrawals. We use both multi-temporal satellite and modeled data from sensors that measure different components of the water balance at varying spatial and temporal resolutions. These remote sensing products include evapotranspiration, precipitation, and land cover. Due to the inherent complexity of integrating these data sets and subsequently relating them to groundwater withdrawals using physical models, we apply random forests- a state of the art machine learning algorithm- to overcome such limitations. Here, we predict groundwater withdrawals per unit area over a highly monitored portion of the High Plains aquifer in the central United States at 5 km resolution for the years 2002-2019. Our modeled withdrawals had high accuracy on both training and testing datasets (R≈ 0.99 and R≈ 0.93, respectively) during leave-one-out (year) cross-validation with low Mean Absolute Error (MAE) ≈ 4.26 mm and Root Mean Square Error (RMSE) ≈ 13.57 mm for the year 2014. Moreover, we found that even for the extreme drought year of 2012, we have a satisfactory test score (R≈ 0.79) with MAE ≈ 10.34 mm and RMSE ≈ 27.04 mm. Therefore, the proposed hybrid water balance and machine learning approach can be applied to similar regions for proactive water management practices.

Many of the world’s major aquifers are under severe stress as a result of intensive pumping in support of irrigated agriculture. The question of what the future holds for these aquifers and the agricultural production they support is of paramount importance in a world of burgeoning populations, dietary shifts, and climate change. Addressing that question requires a better understanding of the how and why of a particular aquifer’s response to pumping. One important, but largely underutilized, source of information is the data from monitoring well networks that provide near-continuous records of water levels through time. Although many regions have such networks operated by local, state, or Federal entities, the vast majority of efforts are, by fiscal necessity, focused on keeping the networks up and running. Little, if any, time is spent on interpreting the acquired hydrographs. The index well network in the High Plains aquifer (HPA) in central and western Kansas is an exception, as hydrograph interpretation is an important program emphasis. Examination of multiyear hydrographs has resulted in the development of profound insights concerning, for example, the frequency of episodic recharge, the magnitude and variability of net inflow, characteristics of the monitored aquifer (continuity, hydraulic regime, etc.), and the impact of extreme meteorological events. These insights have allowed us to develop a significantly better understanding of how the aquifer will respond to proposed management actions; such an understanding is critical for charting more sustainable paths for aquifers across the globe. We will demonstrate these points through an examination of two multiyear hydrographs from the HPA in western Kansas with an emphasis on the insights that shed light on the prospects for the sustainability of this heavily stressed system and the agricultural production that it supports.