Artificial Neural Network Optimized with a Genetic Algorithm for Seasonal Groundwater Table Depth Prediction in Uttar Pradesh, India

Groundwater level forecasting with machine learning models: A review

Groundwater, the world's most abundant source of freshwater, is rapidly depleting in many regions due to a variety of factors. Accurate forecasting of groundwater level (GWL) is essential for effective management of this vital resource, but it remains a complex and challenging task. In recent years, there has been a notable increase in the use of machine learning (ML) techniques to model GWL, with many studies reporting exceptional results. In this paper, we present a comprehensive review of 142 relevant articles indexed by the Web of Science from 2017 to 2023, focusing on key ML models, including artificial neural networks (ANN), adaptive neuro-fuzzy inference systems (ANFIS), support vector regression (SVR), evolutionary computing (EC), deep learning (DL), ensemble learning (EN), and hybrid-modeling (HM). We also discussed key modeling concepts such as dataset size, data splitting, input variable selection, forecasting time-step, performance metrics (PM), study zones, and aquifers, highlighting best practices for optimal GWL forecasting with ML. This review provides valuable insights and recommendations for researchers and water management agencies working in the field of groundwater management and hydrology.

Watershed groundwater level multistep ahead forecasts by fusing convolutional-based autoencoder and LSTM models

The development of deep learning-based groundwater level forecast models can tackle the challenge of high dimensional groundwater dynamics, predict groundwater variation trends accurately, and manage groundwater resources effectively, thereby contributing to sustainable water resources management. This study proposed a novel ConvAE-LSTM model, which fused a Convolutional-based Autoencoder model (ConvAE) and a Long Short-Term Memory Neural Network model (LSTM), to provide accurate spatiotemporal groundwater level forecasts over the next three months. The HBV-light and LSTM models are chosen as benchmarks. An ensemble of point data and the corresponding derived images concerning the past (observations) and the future (forecasts from a conceptual model) of groundwater levels at 33 groundwater wells in Jhuoshuei River basin of Taiwan between 2000 and 2019 constituted the case study. The findings showcase the effectiveness of the ConvAE-LSTM model in extracting crucial features from both point and imagery datasets. This model successfully establishes spatiotemporal dependencies between regional images and groundwater level data over diverse time frames, leading to accurate multi-step-ahead forecasts of groundwater levels. Notably, the ConvAE-LSTM model exhibits a substantial improvement, with the R-squared values showing an increase of more than 18%, 22%, and 49% for the R1, R2, and R3 regions, respectively, compared to the HBV-light model. Additionally, it outperforms the LSTM model in this regard. This study represents a noteworthy milestone in environmental modeling, offering key insights for designing sustainable groundwater management strategies to ensure the long-term availability of this vital resource.

Economic evaluation of groundwater resource in the Effutu Municipality: An application of the Gisser-Sanchez effect

This study presents an economic valuation of the groundwater resource in the Effutu Municipality. It tests the validity of the Gisser-Sanchez's position that the benefits derived from implementing a groundwater management intervention are insignificantly small compared to when no intervention is made. Hundred groundwater-user households were sampled by quota, convenience, and simple random sampling techniques. Assuming a quantitative approach, a contingent valuation-based willingness to pay questionnaire was used for data collection. Respondents were asked to value groundwater under two regimes based on quality: (1) unmanaged quality and (2) hypothetically-managed quality regimes. Using the Lancaster demand theory, the values assigned under either regime were assumed as the benefits users would derive from using groundwater. The statistical difference between the benefits of the two regimes was established by the Wilcoxon Signed Rank Test. The findings revealed that groundwater users are willing to pay 20 Pesewas (GH₵ 0.2) and 30 Pesewas (GH₵ 0.3), respectively, for a 10 L bucket of groundwater from the unmanaged quality regime and groundwater from the hypothetically-managed quality regime. The study established a statistically significant difference between the economic values of groundwater under either regime, indicating that the Gisser-Sanchez effect does not hold for groundwater used for drinking and domestic purposes in the Effutu Municipality. It has been expressed that improving groundwater quality will significantly increase the economic value of the resource. It has therefore been recommended that efforts should be made to treat groundwater to assume the quality of the Ghana Water Company's pipe-borne water after drilling projects in the Municipality.

Forecasting of stage-discharge in a non-perennial river using machine learning with gamma test

Knowledge of the stage-discharge rating curve is useful in designing and planning flood warnings; thus, developing a reliable stage-discharge rating curve is a fundamental and crucial component of water resource system engineering. Since the continuous measurement is often impossible, the stage-discharge relationship is generally used in natural streams to estimate discharge. This paper aims to optimize the rating curve using a generalized reduced gradient (GRG) solver and the test the accuracy and applicability of the hybridized linear regression (LR) with other machine learning techniques, namely, linear regression-random subspace (LR-RSS), linear regression-reduced error pruning tree (LR-REPTree), linear regression-support vector machine (LR-SVM) and linear regression-M5 pruned (LR-M5P) models. An application of these hybrid models was performed and test to modeling the Gaula Barrage stage-discharge problem. For this, 12-year historical stage-discharge data were collected and analyzed. The 12-year historical daily flow data (m³/s) and stage (m) from during the monsoon season, i.e., June to October only from 03/06/2007 to 31/10/2018, were used for discharge simulation. The best suitable combination of input variables for LR, LR-RSS, LR-REPTree, LR-SVM, and LR-M5P models was identified and decided using the gamma test. GRG-based rating curve equations were found to be as effective and more accurate as conventional rating curve equations. The outcomes from GRG, LR, LR-RSS, LR-REPTree, LR-SVM, and LR-M5P models were compared to observed values of daily discharge based on Nash Sutcliffe model efficiency coefficient (NSE), Willmott Index of Agreement (d), Kling-Gupta efficiency (KGE), mean absolute error (MAE), mean bias error (MBE), relative bias in percent (RE), root mean square error (RMSE) Pearson correlation coefficient (PCC) and coefficient of determination (R²). The LR-REPTree model (combination 1: NSE = 0.993, d = 0.998, KGE = 0.987, PCC(r) = 0.997, and R² = 0.994 and minimum value of RMSE = 0.109, MAE = 0.041, MBE = -0.010 and RE = -0.1%; combination 2; NSE = 0.941, d = 0.984, KGE = 0. 923, PCC(r) = 0. 973, and R² = 0. 947 and minimum value of RMSE = 0. 331, MAE = 0.143, MBE = -0.089 and RE = -0.9%) performed superior to the GRG, LR, LR-RSS, LR-SVM, and LR-M5P models in all input combinations during the testing period. It was also noticed that the performance of the alone LR and its hybrid models (i.e., LR-RSS, LR-REPTree, LR-SVM, and LR-M5P) was better than the conventional stage-discharge rating curve, including the GRG method.

Evaluating the BFAST method to detect and characterise changing trends in water time series: A case study on the impact of droughts on the Mediterranean climate

Mediterranean climate regions are facing increased aridity conditions and water scarcity, thus needing integrated management of water resources. Detecting and characterising changes in water resources over time is the natural first step towards identifying the drivers of these changes and understanding the mechanism of change. The aim of this study is to evaluate the potential of Breaks For Additive Seasonal and Trend (BFAST) method to identify gradual (trend) and abrupt (step- change) changes in the freshwater resources time series over a long-term period. This research shows an alternative to the Pettitt's test, LOESS (locally estimated scatterplot smoothing) filter, Mann-Kendall trend test among other common methods for change detection in hydrological data, and paves the way for further scientific investigation related to climate variability and its influence on water resources. We used the monthly accumulated stored water in three reservoirs, the monthly groundwater levels of three hydrological settings and a standardized precipitation index to show BFAST performance. BFAST was successfully applied, enabling: (1) assessment of the suitability of past management decisions when tackling drought events; (2) detection of recovery and drawdown periods (duration and magnitude values) of accumulated stored water in reservoirs and groundwater bodies after wet and dry periods; 3) measurement of resilience to drought conditions; (4) establishment of similarities/differences in trends between different reservoirs and groundwater bodies with regard to drought events.

Sustainable groundwater management through an optimal water supply system using a coupled simulation-optimization approach

Meeting people's water supply needs in cities is an urgent challenge. Especially in coastal cities, surface water sources are usually inadequate and so groundwater is frequently over-exploited, resulting in its rapid decrease and intrusion of saltwater. Conjunctive use of surface water and groundwater is the best alternative approach to mitigate the overuse of groundwater through effective distribution of surface water sources. It requires numerous runs of the simulation-optimization model to select an optimal pattern of water distribution by keeping the groundwater levels under control. In this study, a new simulation-optimization model is developed using wavelet support vector regression (WSVR) and genetic algorithm (GA) to propose an optimal water distribution system to Visakhapatnam city on the East Coast of India, fulfilling the constraints of surface water quantity, aquifer pumping, and drawdown. Estimation of groundwater pumping and groundwater level variation in spatial context is challenging in urban environment. To overcome this, the calibrated modular finite-difference flow (MODFLOW) model for this study area has been used to prepare the spatial and temporal variation of model inputs such as groundwater pumping and groundwater levels. The WSVR-GA model's performance to reduce the groundwater pumping is evaluated in three distinct cases. The surface water resources from three sources are distributed to different wards in the city source-wise in case I and centralized in cases II and III, while the source-wise surface water constraints are limited to monthly in cases I and II and annual in case III. The WSVR-GA management model suggested ward-wise groundwater pumping restrictions, resulting in 9.57 MCM, 11.64 MCM, and 12.54 MCM increase in total groundwater storage capacity in cases I, II, and III respectively. Cases II and III offer 21% and 25%, respectively, more storage than case I. Thus, centralized distribution systems have increased the sustainability of groundwater supplies by preventing overdrafts caused by a lack of surface water resources. Validation of results using MODFLOW indicates a substantial rise in groundwater levels in the study area.

Application of artificial intelligence models for prediction of groundwater level fluctuations: case study (Tehran-Karaj alluvial aquifer)

The nonlinear groundwater level fluctuations depend on the interaction of many factors such as evapotranspiration, precipitation, groundwater abstraction, and hydrogeological characteristics, making groundwater level prediction a complex task. Groundwater level changes are among the most critical issues in water resource management, which can be predicted to effectively provide management solutions to conserve renewable water resources. Understanding the aquifer status using numerical models is time-consuming and also is associated with inherent uncertainty; therefore, in recent decades, the application of artificial intelligence methods to predict water table fluctuations has significantly gained momentum. In this study, artificial neural network (ANN), fuzzy logic (FL), adaptive neuro-fuzzy inference system (ANFIS), and least square support vector machine (SVM) methods were utilized to predict groundwater level (GWL) with 1-, 2-, and 3-month lead time in Tehran-Karaj plain. Several input scenarios were developed considering groundwater levels, average temperature, total precipitation, total evapotranspiration, and average river flow on a monthly interval. The four error criteria, the correlation coefficient (R), root mean squared error (RMSE), Nash-Sutcliffe efficiency (NSE), and mean absolute error (MAE), were the basis to evaluate the models. Results showed that all the applied methods could provide acceptable GWL prediction, but the ANFIS was the most accurate. However, the ANFIS model showed slightly better performance by yielding R = 0.98 for the training stage and R = 0.98 for the testing stage in the P84 observation well and the second combination of inputs and 1-month lead time. The outcomes also revealed that all the approaches mentioned above could appropriately predict GWL for the leading time of 1 and 2 months, but the models provided unsatisfactory results for a 3-month leading time.

Spatiotemporal Distribution and Statistical Analysis of Abnormal Groundwater Level Rising in Poyang Lake Basin

Studies on groundwater have traditionally been based on declining groundwater levels and associated ecological, environmental, and geological problems. However, due to extreme hydrometeorological events and human activities, rising groundwater levels have been observed in many areas. The daily groundwater levels from 2018 to 2020 for the Poyang Lake Basin (PLB) in Jiangxi Province were recorded. The statistical characteristics of abnormal groundwater level rising (AGLR) events and the factors influencing the dynamic changes in groundwater level were analyzed using geostatistical methods and outlier identification methods. The groundwater level in the lower terrain of the PLB has increased significantly in recent years. AGLR events identified by the median absolute deviation and interquartile range methods showed that AGLR events mainly occurred in the spring and summer and were mainly distributed near the surface water bodies. Correlation analysis of the factors influencing the groundwater level revealed that the correlation between precipitation and groundwater level was related to topography. In contrast, the correlation between river stage and groundwater level was related to runoff volume.

Evaluation of groundwater quality for irrigation water supply using multi-criteria decision-making techniques and GIS in an agroeconomic tract of Lower Ganga basin, India

Groundwater irrigation has evolved the monocropping cultivation pattern to multi-cropping, especially in many arid/semi-arid tracts globally. Irrigation practices with the groundwater of poor quality can limit the selection of the crop, reduce crop yields and degrade the soil quality. The present study has been undertaken to identify the hydrogeochemical phenomena of groundwater systems in the south-western Birbhum district, India and to analyze groundwater suitability for irrigation during the pre-and post-monsoon cycles by adopting the Irrigation Water Quality Index (IWQI) using Multivariate Factor Analysis along with some traditional methods viz. sodium adsorption ratio, sodium percentage, magnesium hazards, residual sodium bicarbonate (RSBC) and carbonate (RSC), Wilcox's and USSL diagrams, permeability index and Kelly's index. The hydrogeochemical analysis revealed that chemical weathering and evaporation are predominant in the aquifer systems. Groundwater quality reflected soil salinity, sodicity and magnesium hazards risks and water toxicity to the sensitive plants at 0-46.4% of the post-monsoon samples and 0-38.4% of the pre-monsoon samples based on the individual traditional methods whereas about 97.73-98.88% of the total area was classified as moderate to severely unsuitable for irrigation during both seasons when integrated multiple parameters using the IWQI method. Prolonged use of such groundwater for irrigation is susceptible to causing moderate to severe infiltration problems at a greater extent of the study area. The study recommends adaptation of salinity, sodicity and RSC/RSBC reduction procedures (e.g., the use of acid and gypsum amendments in the irrigation lands and through water blending) and advanced irrigation practices (viz. drips, sprinklers and micro irrigations) to prevent soil degradation and increase crops productivity. Adopting Managed Aquifer Recharge procedures as well as rainwater harvesting in the areas bearing unsuitable water quality can dilute the ionic concentrations of the groundwater facies which in turn will improve the groundwater quality for irrigation.

Predictive analysis of the value of information flow on the shop floor of developing countries using artificial neural network based deep learning

To facilitate the continuous improvement of performance and the management of information flow (MIF) for production and manufacturing purposes on the shop floor of developing countries, there is a need to characterize information flow that will be shared during the process. MIF provides a key performance shop floor metric called the value of information flow (VIF). Previous methods have been used to analyze VIF in developed countries. However, these methods are sometimes limited when applied to developing countries where the shop floor is disorganized. It then renders the MIF with the imported software inefficient because of the gap between the user environments. Taking Cameroon as a case study, this study proposes a new method of modeling and analyzing the information flow and its value based on the characteristics of information flow (CIF) for developing countries. In addition, a predictive analysis of the VIF based on CIF using an artificial neural network (ANN) on one hand and optimized ANN with particle swarm optimizer (PSO) and genetic algorithms (GA) on the other is performed. The ANN model of regression developed has the following performance: coefficient of determination: 0.99 and mean squared error (MSE): 0.00043. For the PSO-ANN, the MSE decreased to 0.00011, and this model result was similar to that of the deep learning model used for regression. The GA-ANN model results were not as satisfactory as those of the PSO-ANN model. A predictive system to analyze VIF is proposed for managers of companies in developing countries.