Groundwater Prediction Using Machine-Learning Tools

Abstract

Predicting groundwater availability is important to water sustainability and drought mitigation. Machine-learning tools have the potential to improve groundwater prediction, thus enabling resource planners to: (1) anticipate water quality in unsampled areas or depth zones; (2) design targeted monitoring programs; (3) inform groundwater protection strategies; and (4) evaluate the sustainability of groundwater sources of drinking water. This paper proposes a machine-learning approach to groundwater prediction with the following characteristics: (i) the use of a regression-based approach to predict full groundwater images based on sequences of monthly groundwater maps; (ii) strategic automatic feature selection (both local and global features) using extreme gradient boosting; and (iii) the use of a multiplicity of machine-learning techniques (extreme gradient boosting, multivariate linear regression, random forests, multilayer perceptron and support vector regression). Of these techniques, support vector regression consistently performed best in terms of minimizing root mean square error and mean absolute error. Furthermore, including a global feature obtained from a Gaussian Mixture Model produced models with lower error than the best which could be obtained with local geographical features.

Groundwater Potential Zone Mapping Using Analytical Hierarchy Process and GIS in Muga Watershed, Abay Basin, Ethiopia

“…Literatures reveal that several researchers have been using GIS to delineate groundwater potential zones with the integration of statistical approach such as simple additive weight (SAW) and analytic hierarchy process (AHP) [ 25 , 26 , 27 , 28 , 29 , 30 , 31 ] and, machine learning. [ 32 , 33 , 34 , 35 , 36 , 37 , 38 ] The combination of GIS and remote sensing technologies reduce the ambiguity of hydrogeological data various aspect.…”

A Machine Learning Approach to Predict Groundwater Levels in California Reveals Ecosystems at Risk

“…In contrast to traditional process-based numerical modeling methods, which require large, observationally based datasets and input parameters, machine learning approaches are also increasingly being used in hydrogeological applications since they are data-driven and require less calibration than processbased models (Shen et al, 2018). For example, machine learning is being used to predict temporal groundwater dynamics (Daliakopoulos et al, 2005;Banerjee et al, 2009;Adamowski and Chan, 2011;Yoon et al, 2011;Shiri et al, 2013;Gholami et al, 2015), groundwater availability (Hussein et al, 2020), map the water table (Fienen et al, 2013;Koch et al, 2019a), model groundwater quality conditions (Winkel et al, 2011;Nolan et al, 2015;Erickson et al, 2018;Koch et al, 2019b), and strengthen traditional physical-based numerical groundwater models (Shen et al, 2018).…”

Basic Statistical Estimation Outperforms Machine Learning in Monthly Prediction of Seasonal Climatic Parameters

“…For the climatic parameters that were examined in this paper, using previous month (denoted as feature [11]) was not effective, and could even degrade predictive performance when added. However, this conclusion is not applicable to other parameters such as groundwater [7,57], which involves conditions that last over multiple months. The slight improvements seen when adding [11] to evaporation may be due to this effect.…”

“…From a practical point of view, usually the most important policy decisions involving climate require monthly predictions [7]. Relatively few studies exist which use image data to make monthly predictions [57,58]. When time scales on the order of months or longer are involved, datasets are typically much smaller than those involving shorter time scales.…”

“…We transformed the image pixels to greyscale (0-255) and re-sized the images to 70 × 40 to further reduce their complexity. In view of the fact that extreme values are a common occurrence in geophysical parameter data, pixel values were regularized by replacing them with their square roots, following the example of [57,80,81]. Since our study is concerned with relative performance of different algorithms rather than absolute performance, for simplicity we did not remove over-ocean pixels, which are constant in all images and hence perfectly predicted.…”