Groundwater Withdrawal Prediction Using Integrated Multitemporal Remote Sensing Data Sets and Machine Learning

Abstract

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 multitemporal satellite and modeled data from sensors that measure different components of the water balance and land use 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 data sets (R 2 ≈ 0.99 and R 2 ≈ 0.93, respectively) during leave-one-out (year) cross validation with low mean absolute error (MAE) ≈ 4.31 mm and root-mean-square error (RMSE) ≈ 13.50 mm for the year 2014. Moreover, we found that even for the extreme drought year of 2012, we have a satisfactory test score (R 2 ≈ 0.84) with MAE ≈ 9.72 mm and RMSE ≈ 24.17 mm. Therefore, the proposed machine learning approach should be applicable to similar regions for proactive water management practices. Plain Language Summary Groundwater is an essential component of global water resources and is the largest source of Earth's liquid freshwater. It is extensively used for drinking water and to support global food production. Consequently, groundwater consumption has significantly increased owing to the pressing demands for water, food, and energy primarily driven by the increasing global population. Despite its critical role in the water-food-energy nexus, very few regions in the United States or elsewhere actively monitor their groundwater withdrawals (extraction or pumping) for implementing sustainable water management solutions. We develop a novel approach combining water balance components and land use measured using openly available remote sensing products with a machine learning model to estimate groundwater withdrawals. This framework automatically learns the interrelationships among these variables and groundwater withdrawals. Our study area is a portion of the High Plains Aquifer in Kansas (central United States) where overpumping has caused substantial groundwater storage loss. Also, a large amount of groundwater pumping data are available for validation. The results indicate good accuracy even for extreme drought years. Thus, this approach should be applicable to similar regions having sparsely or moderately available groundwater pumping data, enabling water managers to proactively implement sustainable solutions addressing water security issues.

Evaluation of Data Needs for Assessments of Aquifers Supporting Irrigated Agriculture

“…In a region dominated by central pivot irrigation, t n could be estimated by combining a satellite-based count of pivot circles with an estimate of the average number of wells per pivot. Majumdar et al (2020) demonstrate the effectiveness of machine learning techniques for predicting groundwater withdrawals based on a combination of remote sensing data sets; such techniques could be applied to estimate total extraction in less densely monitored regions for a variety of irrigation practices.…”

“…Majumdar et al. (2020) demonstrate the effectiveness of machine learning techniques for predicting groundwater withdrawals based on a combination of remote sensing data sets; such techniques could be applied to estimate total extraction in less densely monitored regions for a variety of irrigation practices.…”

Machine Learning in Agriculture: A Comprehensive Updated Review

“… Estimation of inland water quality. LSTM, PLSR, SVR, DNN DNN: R 2 = 0.81, MSE = 0.29, RMSE = 0.54 [ 359 ] Groundwater In-situ water quality data, hyperspectral, satellite spectral data Estimation of water quality DT Acc = 81.49%, ROC = 87.75% [ 360 ] Groundwater Weather data, ET, satellite spectral data, land use Estimation of groundwater withdrawals RF R 2 = 0.93, MAE = 4.31 mm, RMSE = 13.50 mm [ 361 ] Groundwater nitrate concentration Various geo-environmental data Comparison of different ML models for estimating nitrate concentration SVM, Cubist, RF, Bayesian-ANN RF: R 2 = 0.89, RMSE = 4.24, NSE = 0.87 Acc: Accuracy; CC: Coefficient of Correlation; ET: Evapotranspiration; ET o : reference EvapoTranspiration; ROC: Receiver Operating Characteristic; ME: Model Efficiency; NSE: Nash-Sutcliffe model efficiency Coefficient; POD: Probability Of Detection. …”

Introduction to Special Section: The Quest for Sustainability of Heavily Stressed Aquifers at Regional to Global Scales

“…The state of Kansas is the most prominent exception with over 95% of its approximately 18,900 irrigation wells metered and subject to regulatory verification (Butler et al, 2016;NASS, 2018). Bohling et al (2021), Lamb et al (2021), and Majumdar et al (2020) use the Kansas pumping data to evaluate how many wells need to be metered, the major influences on pumping volume, and the effectiveness of a new machine learning approach for estimating pumping volumes, respectively. Foster et al (2020) use the pumping data from a heavily monitored area in the state of Introduction to Special Section: The Quest for Sustainability of Heavily Stressed Aquifers at Regional to Global Scales…”