Groundwater level modeling framework by combining the wavelet transform with a long short-term memory data-driven model

Evaluate effect of 126 pre-processing methods on various artificial intelligence models accuracy versus normal mode to predict groundwater level (case study: Hamedan-Bahar Plain, Iran)

The estimation of groundwater levels is crucial and an important step in ensuring sustainable management of water resources. In this paper, selected piezometers of the Hamedan-Bahar plain located in west of Iran. The main objective of this study is to compare effect of various pre-processing methods on input data for different artificial intelligence (AI) models to predict groundwater levels (GWLs). The observed GWL, evaporation, precipitation, and temperature were used as input variables in the AI algorithms. Firstly, 126 method of data pre-processing was done by python programming which are classified into three classes: 1- statistical methods, 2- wavelet transform methods and 3- decomposition methods; later, various pre-processed data used by four types of widely used AI models with different kernels, which includes: Support Vector Machine (SVR), Artificial Neural Network (ANN), Long-Short Term memory (LSTM), and Pelican Optimization Algorithm (POA) - Artificial Neural Network (POA-ANN) are classified into three classes: 1- machine learning (SVR and ANN), 2- deep learning (LSTM) and 3- hybrid-ML (POA-ANN) models, to predict groundwater levels (GWLs). Akaike Information Criterion (AIC) were used to evaluate and validate the predictive accuracy of algorithms. According to the results, based on summation (train and test phases) of AIC value of 1778 models, average of AIC values for ML, DL, hybrid-ML classes, was decreased to -25.3%, -29.6% and -57.8%, respectively. Therefore, the results showed that all data pre-processing methods do not lead to improvement of prediction accuracy, and they should be selected very carefully by trial and error. In conclusion, wavelet-ANN model with daubechies 13 and 25 neurons (db13_ANN_25) is the best model to predict GWL that has -204.9 value for AIC which has grown by 5.23% (-194.7) compared to the state without any pre-processing method (ANN_Relu_25).

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.

Groundwater level forecasting using ensemble coactive neuro-fuzzy inference system

A modeling framework utilizing the coactive neuro-fuzzy inference system (CANFIS) has been developed for multi-lead time groundwater level (GWL) forecasting in four different wells located in Texas and Florida, USA. Various model input combinations, including GWL, precipitation, temperature, and surface water level variables, have been derived based on proposed correlation analysis using singular spectrum analysis (SSA) remainders. The models have been trained on data subsets of varying lengths to identify the optimal training data duration. Additionally, we have introduced the bagging ensemble learning method to enhance the performance of the CANFIS model. As part of a comprehensive model evaluation process, the best-performing CANFIS model for each forecasting scenario has undergone uncertainty analysis using bootstrap sampling. Our results reveal that the CANFIS model performs satisfactorily for daily forecasting but leaves room for improvement in monthly forecasting, particularly for two-month and three-month ahead forecasts. Moreover, we have identified several optimal input combinations, highlighting the significance of the temperature variable in monthly forecasting. Furthermore, our findings indicate that additional training data does not necessarily lead to improved performance. The ensemble CANFIS model has demonstrated significant performance enhancement, particularly for monthly forecasting. Finally, the CANFIS model uncertainty analysis has shown satisfactory results for daily forecasting scenarios, while monthly forecasting models exhibit higher uncertainties, particularly during periods with distinctly different GWL fluctuation patterns.

Land subsidence prediction in Zhengzhou's main urban area using the GTWR and LSTM models combined with the Attention Mechanism

In recent years, due to urbanization and human activities, groundwater overexploitation has become increasingly severe, resulting in some degrees of land subsidence and, consequently, causing a series of geological disasters and other environmental issues. Therefore, large-scale and high-precision land subsidence prediction is of great importance for the prevention and control of geological disasters. However, the existing prediction models and methods ignore the effects of the spatiotemporal non-stationary relationships between the influencing factors and the accumulated land subsidence, causing the poor accuracy of the predicted land subsidence results. In this context, a Geographically and Temporally Weighted Regression combined with the Long Short-Term Memory (LSTM)-multivariable and Attention Mechanism (AM) (GTWR-LSTMm-AM) was proposed to more accurately predict the deformation of time series land subsidence in this study. The small baseline subset-interferometric synthetic aperture radar (SBAS-InSAR) was used to reveal the temporal deformation information of Zhengzhou's main urban area, then the GTWR model was used to assess the spatiotemporal non-stationarity relationships between the accumulated land subsidence and its influencing factors monthly groundwater stability level, monthly precipitation and Normalized Difference Vegetation Index (NDVI) data, and to determine the corresponding weight matrix. In addition, we introduced an LSTM model with AM to extract key information from the time-series land subsidence data and adjusted the dynamic weights of the three selected influencing factors to predict the land subsidence in Zhengzhou's main urban area. The prediction accuracy R2 of the GTWR-LSTMm-AM model reaches 0.972, which is higher than 0.929 of the LSTMm model. The prediction accuracy RMSE is less than 3 mm and reaches 2.403 mm. In addition, we determined the importance of the impact factor on the subsidence results by randomly interrupting the impact factor time series, disclosuring that the monthly groundwater level contributed the most to the land subsidence in Zhengzhou's main urban area.

Hysteresis response of groundwater depth on the influencing factors using an explainable learning model framework with Shapley values

Machine learning has been widely used for groundwater prediction. However, the hysteresis response of groundwater depth (GD) to input features has not been fully investigated. This study uses an interpretation method to reveal the interplay between climate, human activity, and GD while considering the response of groundwater to multiple factors. Six factors [precipitation (P), wind speed (WS), temperature (T), population (POP), gross domestic product (GDP), and effective irrigated area (EIA)] were selected to analyze the hysteresis response of GD in terms of the lag correlation coefficient and lag time. The correlation between climatic variables and GD was weaker than that of anthropogenic variables. The lag time between variables and different types of GD was less than four months at most sites, except for EIA and WS in deep groundwater. The SVM model achieved satisfactory performance in 89 % of the sites. If there were sharp changes in GD during the testing period or significant variations in its seasonal patterns at different times, the SVM model performed poorly. The model was interpreted using the Shapley additive explanation method. The impact of POP and GDP on deep groundwater in irrigated areas was higher than that of shallow groundwater. In urban areas with intensive human activities, anthropogenic variables were the main factors affecting shallow groundwater while the impact of climate was gradually increasing in the suburbs. The influence of precipitation on shallow groundwater was decreased after water transfer from the South-to-North Water Diversion project. Furthermore, this study proposed a multifactor-driven conceptual model that can provide recommendations for analyzing groundwater dynamics in similar areas.

Long short-term memory models to quantify long-term evolution of streamflow discharge and groundwater depth in Alabama

Long short-term memory (LSTM) models have been shown to be efficient for rainfall-runoff modeling, and to a lesser extent, for groundwater depth forecasting. In this study, LSTMs were applied to quantify the spatiotemporal evolution of surface and subsurface hydrographs in Alabama in the Southeastern United States, where water sustainability has not been fully quantified across spatiotemporal scales. First, the surface water LSTM model with extensive dynamic (precipitation and other weather variables) and static (basin characteristics) inputs predicted the main characteristics of streamflow for six years at 19 gauged basins in Alabama. The model tended to underestimate extremely high streamflow but adding drainage density as an input feature slightly improved the predictions of extreme events. Second, to predict the groundwater depth evolution, a groundwater LSTM (GW-LSTM) model was proposed and applied using static inputs capturing the aquifers' hydrogeological properties and dynamic inputs of meteorological information. Three precipitation scenarios were also explored to evaluate the groundwater hydrograph evolution in the next two decades. The GW-LSTM model predicted the general trend of daily groundwater depth fluctuations (at 21 wells distributed across Alabama from 1990 to 2021) including most extremely high groundwater levels, and recovered groundwater depth for locations withheld from model training and validation. This study, therefore, extended the application of LSTMs in quantifying the spatiotemporal evolution of surface water and groundwater, two manifestations of a single integrated resource.

Application of robust deep learning models to predict mine water inflow: Implication for groundwater environment management

Traditional mine water inflow prediction is characterized by a high degree of uncertainty in model parameters and complex mechanisms involved in the water inflow process. Data-driven models play a key role in predicting inflow mechanisms without considering physical changes. However, the existing models are limited by nonlinearity and non-stationarity. Thus, the principal objective of this study was to propose two robust models, the DIFF-TCN model and the DIFF-LSTM model, for predicting the average water inflow per day. The models consist of three methods, namely Difference Method (DIFF), Temporal Convolutional Neural Network (TCN), and Long Short-Term Memory Neural Network (LSTM). When applied to the Tingnan Coal Mine, Shanxi Province, China, the DIFF-TCN performs better in predicting the average daily water inflow, the model has a MAE of 5.88 m³/h, RMSE of 6.85 m³/h and R² of 0.96 in the test stage of the water inflow event. Comparison with the other deep learning models (with similar complex structures) and traditional time series model shows the superiority of our proposed DIFF-TCN model. The SHAP value is used to explain the contribution of each model input to the predicted values, and it indicates that the historical time of water inflow data are the most important input, and the advance distance and the groundwater level data also contribute to the model predictions, but groundwater level data for some periods in the past may have a detrimental effect on the model. The findings of this study can provide better understanding about potential of robust deep learning models for smart hydrological forecasting, and they can also provide technical guidance for mining safety production and protection of water resources and water environment around the mining area.

A wavelet-assisted deep learning approach for simulating groundwater levels affected by low-frequency variability

Groundwater level (GWL) simulations allow the generation of reconstructions for exploring the past temporal variability of groundwater resources or provide the means for generating projections under climate change on decadal scales. In this context, analyzing GWLs affected by low-frequency variations is crucial. In this study, we assess the capabilities of three deep learning (DL) models (long short-term memory (LSTM), gated recurrent unit (GRU), and bidirectional LSTM (BiLSTM)) in simulating three types of GWLs affected by varying low-frequency behavior: inertial (dominated by low-frequency), annual (dominated by annual cyclicity) and mixed (in which both annual and low-frequency variations have high amplitude). We also tested if maximal overlap discrete wavelet transform pre-processing (MODWT) of input variables helps to better identify the frequency content most relevant for the models (MODWT-DL models). Only external variables (i.e., precipitation, air temperature as raw data, and effective precipitation (EP)) were used as input. Results indicate that for inertial-type GWLs, MODWT-DL models with raw data were notably more accurate than standalone models. However, DL models performed well for annual-type GWLs, while using EP as input, with MODWT-DL models exhibiting only minor improvements. Using raw data as input improved MODWT-DL models compared to standalone models; nevertheless, all models using EP performed better for annual-type GWLs. For mixed-type GWLs, while using EP as input, MODWT-DL models performed well, with substantial improvements over standalone models. Using raw data as input, improvement of MODWT-DL models is marginal compared to that of standalone models; nevertheless, they perform better than standalone models with EP. The Shapley Additive exPlanations (SHAP) approach used to interpret models highlighted that they preferentially learned from low-frequency in precipitation data to achieve the best simulations for inertial and mixed GWLs. This study showed that MODWT-based input pre-processing is highly suitable to better simulate low-frequency varying GWLs.

Mapping of groundwater salinization and modelling using meta-heuristic algorithms for the coastal aquifer of eastern Saudi Arabia

The growing increase in groundwater (GW) salinization in the coastal aquifers has reached an alarming socio-economic menace in Saudi Arabia and various places globally due to several natural and anthropogenic activities. Hence, evaluating the GW salinization is paramount to safeguarding the water resources planning and management. This study presents three different scenarios viz.: real field investigation, experimental laboratory analysis (using ion chromatography (IC) and inductively coupled plasma mass spectrometry (ICP-MS), etc.), and artificial intelligence (AI) based metaheuristic optimization (MO) algorithms in Saudi Arabia. The main purpose of this study is to validate the obtained experimental-based analysis using hybrid MO techniques comprising of adaptive neuro-fuzzy inference system (ANFIS) hybridized with genetic algorithm (GA), particle swarm optimization (PSO), and biogeography-based optimization (BBO) for identification of GW salinization in the coastal region of eastern Saudi Arabia. Additionally, ArcGIS 10.3 software generates the prediction map based on ANFIS-GA, ANFIS-PSO, and ANFIS-BBO. Feature selection was assessed using the PSO algorithm, and four indices evaluated the estimated models, namely, root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and standard deviation (SD). The simulated results are based on three variable input combinations, which showed that the ANFIS-PSO (MAE = 0.00439) algorithm had the highest accuracy (99 %), followed by the ANFIS-GA (MAE = 0.00767) and ANFIS-BBO (MAE = 0.0132) algorithms. Besides, Ca²⁺, Na⁺, Mg²⁺, and Cl^- were the most influential parameters. The accuracy also demonstrated the potential reliability of MO algorithms based on spatial distribution mapping. The employed approach proved to be merit and reliable tool for water resources decision-makers in the coastal aquifer of Saudi Arabia. This approach is believed to improve water scarcity as one of the essential targets for Goal 6 of Sustainable Development Vision 2030 and the Kingdom in general.

Groundwater level response identification by hybrid wavelet-machine learning conjunction models using meteorological data

Due to its heterogeneous and complex nature, groundwater modeling needs great effort to quantify the aquifer, a crucial tool for policymakers and hydrogeologists to understand the variations in groundwater levels (GWL). This study proposed a set of supervised machine learning (ML) models to delineate the GWL changes in the Zarand-Saveh complex aquifer in Iran using 15-year (2005-2020) monthly dataset. The wavelet transform (WT) procedure was also used to improve the GWL prediction ability of ML models for 3-month horizons using input datasets of precipitation, evapotranspiration, temperature, and GWL. The four well-accepted standalone ML methods, i.e., artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), group method of data handling (GMDH), and least square support vector machine (LSSVM), were implemented and compared with the hybrid wavelet conjunction models. The methods were compared based on root mean square error (RMSE), mean absolute error (MAE), correlation coefficient (R), and Nash-Sutcliffe efficiency (NSE). Comparison outcomes showed that the hybrid wavelet-ML considerably improved the standalone model results. The wavelet transform-least square support vector machine (WT-LSSVM) model was superior to other standalone and hybrid wavelet-ML methods to predict GWL. The best GWL predictions were acquired from the WT-LSSVM model with input scenario 5 involving all influential variables, and this model produced RMSE, MAE, R, and NSE as 0.05, 0.04, 0.99, and 0.99 for 1 month ahead of GWL prediction, while the corresponding values were obtained as 0.18, 0.14, 0.95, and 0.90 for 3 months ahead of GWL prediction, respectively.

A Data-Driven OBE Magnetic Interference Compensation Method

Aeromagnetic compensation is a technology used to reduce aircraft magnetic interference, which plays an important role in aeromagnetic surveys. In addition to maneuvering interferences, the onboard electronic (OBE) interference has been proven to be a significant part of aircraft interference, which must be reduced before further interpretation of aeromagnetic data. In the past, most researchers have focused on establishing linear models to compensate for OBE magnetic interference. However, such methods can only work using accurate reference sensors. In this paper, we propose a data-driven OBE interference compensation method, which can reduce OBE interference without relying on any other reference sensor. This network-based method can integrally detect and repair the OBE magnetic interference. The proposed method builds a prediction model by combining wavelet decomposition with a long short-term memory (LSTM) network to detect and predict OBE interference, and then estimates the local variation of the magnetic field to remove the drift of the interference. In our tests, we construct 10 semi-real datasets to quantitatively evaluate the performance of the proposed method. The F1 score of the proposed method for OBE interference detection is over 0.79, and the RMSE of the compensated signal is less than 0.009 nT. Moreover, we also test our method on real signals, and the results show that our method can detect all interference and significantly reduce the standard deviation of the magnetic field.

Water quality forecasting based on data decomposition, fuzzy clustering and deep learning neural network

Water quality forecasting can provide useful information for public health protection and support water resources management. In order to forecast water quality more accurately, this paper proposes a novel hybrid model by combining data decomposition, fuzzy C-means clustering and bidirectional gated recurrent unit. Firstly, the original water quality data is decomposed into several subseries by empirical wavelet transform, and then, the decomposed subseries are recombined by fuzzy C-means clustering. Next, for each clustered series, bidirectional gated recurrent unit is applied to develop prediction model. Finally, the forecast result is obtained by the summation of the predictions for the subseries. The proposed forecast model is evaluated by the water quality data of Poyang Lake, China. Results show that the proposed forecast model provides highly accurate forecast result for all of the six water quality data: the average of MAPE of the forecast results for the six water quality datasets is 4.59% for 7 day ahead prediction. Furthermore, our model shows better forecast performance than the other models. Particularly, compared with the single BiGRU model, MAPE decreased by 32.86% in average. Results demonstrate that the proposed forecast model can be used effectively for water quality forecasting.

Can ensemble machine learning be used to predict the groundwater level dynamics of farmland under future climate: a 10-year study on Huaibei Plain

Accurate and simple prediction of farmland groundwater level (GWL) is an important aspect of agricultural water management. A farmland GWL prediction model, GWPRE, was developed that integrates four machine learning (ML) models (support vector machine regression, random forest, multiple perceptions, and the stacking ensemble model) with weather forecasts. Based on the GWL and meteorological data of five monitoring wells (N1, N2, N3, N4, and N5) in Huaibei plain from 2010 to 2020, the feasibility of predicting GWL by meteorological factors and ML algorithm was tested. In addition, the stacking ensemble model and future meteorological data after Bayesian model averaging were introduced for the first time to predict GWL under future climate conditions. The results showed that GWL showed an increasing trend in the past decade, but it will decrease in the future. The performance of the stacking ensemble model was better than that of any single ML model, with RMSE reduced by 4.26 ~ 96.97% and the running time reduced by 49.25 ~ 99.40%. GWL was most sensitive to rainfall, and the sensitivity index ranged from 0.2547 to 0.4039. The fluctuation range of GWL of N1, N2, and N3 was 1.5 ~ 2.5 m in the next decade. Due to the possible high rainfall, the GWL decreased in 2024 under RCP 2.6 and 2026 under RCP 8.5. It is worth noting that although the stacking ensemble model can improve the accuracy, it is not always the best among ML models in terms of portability. Nevertheless, the stacking ensemble model was recommended for GWL prediction under climate change.

A Hybrid Triple Collocation-Deep Learning Approach for Improving Soil Moisture Estimation from Satellite and Model-Based Data

Satellite retrieval and land surface models have become the mainstream methods for monitoring soil moisture (SM) over large regions; however, the uncertainty and coarse spatial resolution of these products limit their applications at the regional and local scales. We proposed a hybrid approach combining the triple collocation (TC) and the long short-term memory (LSTM) network, which was designed to generate a high-quality SM dataset from satellite and modeled data. We applied the proposed approach to merge SM data from Soil Moisture Active Passive (SMAP), Global Land Data Assimilation System-Noah (GLDAS-Noah), and the land component of the fifth generation of European Reanalysis (ERA5-Land), and we then downscaled the merged SM data from 0.36° to 0.01° resolution based on the relationship between the SM data and auxiliary environmental variables (elevation, land surface temperature, vegetation index, surface albedo, and soil texture). The merged and downscaled SM results were validated against in situ observations. The results showed that: (1) the TC-based validation results were consistent with the in situ-based validation, indicating that the TC method was reasonable for the comparison and evaluation of satellite and modeled SM data. (2) TC-based merging was superior to simple arithmetic average merging when the parent products had large differences. (3) Downscaled SM of the TC-based merged product had better performance than that of the parent products in terms of ubRMSE and bias values, implying that the fusion of satellite and model-based SM data would result in better downscaling accuracy. (4) Downscaled SM of TC-based merged data not only improved the representation of the SM spatial variability but also had satisfactory accuracy with a median of R (0.7244), ubRMSE (0.0459 m3/m3), and bias (−0.0126 m3/m3). The proposed approach was effective for generating a SM dataset with fine resolution and reliable accuracy for wide hydrometeorological applications.