Search Results (14)

Groundwater quality assessment in urban and peri-urban environments is often constrained by incomplete monitoring records, irregular sampling frequencies, and heterogeneous environmental datasets. The primary objective of this study is to predict the Water Quality Index (WQI) in the Attica River Basin, Greece, using advanced machine learning (ML) techniques. A groundwater quality dataset comprising 958 observations from 80 monitoring stations was analyzed using six physicochemical parameters, namely electrical conductivity, ammonium, nitrate, nitrite, chloride, and sulphate. Three modeling approaches, namely TabNet (with Winsorization), SVM, and Gradient Boosting Machines (GBM), were implemented to estimate groundwater quality conditions. To address the challenge of missing data, Multiple Imputation by Chained Equations (MICE) with Predictive Mean Matching (PMM) was implemented and systematically compared against conventional imputation approaches, including smoothed averages, interpolation, and forward-fill methods. The novelty of this study lies in the integration of open-access groundwater chemistry data, advanced multivariate imputation (MICE-PMM), and attention-based deep learning (TabNet) for groundwater quality prediction in a Mediterranean peri-urban area under data-scarce conditions. Using a multi-year groundwater monitoring dataset, the results indicate that the integrated MICE-PMM and TabNet framework achieved the highest predictive performance, with R² = 0.91, NSE = 0.91, RMSE = 52.21, and MAE = 25.68. Feature importance and sensitivity analyses identified nitrate as the dominant driver of WQI variability, highlighting the strong influence of anthropogenic nutrient loading on groundwater quality. Overall, the proposed framework provides a transferable, data-driven approach for groundwater quality prediction, environmental monitoring, and groundwater resource management in urban and peri-urban aquifer systems characterized by incomplete environmental datasets. Full article

Accurate and efficient rock mass quality classification is a prerequisite for assessing slope stability, designing support schemes, and ensuring mining safety in open-pit mines. However, traditional empirical classification methods rely heavily on expert judgment and often struggle to capture the complex, nonlinear relationships among factors influencing slope stability. Existing intelligent classification models also suffer from limitations, including sensitivity to incomplete data, insufficient feature interaction learning, and unstable performance on small-scale datasets. To address these issues, this study develops a deep forest (DeepForest) model optimized by three metaheuristic algorithms—brown bear optimizer (BBO), tuna swarm optimizer (TSO), and sparrow search algorithm (SSA)—to intelligently classify slope rock mass quality. A rock mass quality dataset containing 204 groups of slope and non-slope cases was established to train and evaluate the classification performance of the DeepForest models. Six influencing factors were set as input parameters: uniaxial compressive strength (UCS) of rock, rock quality designation (RQD), spacing of discontinuities (Sd), rock mass integrity coefficient (Kv), groundwater conditions (W), and site type (St). Multivariate imputation by chained equations (MICE), isolation forest (IsoForest), and synthetic minority over-sampling technique (SMOTE) were used to handle missing values, outliers, and imbalance in the dataset, respectively. The performance of the proposed models was evaluated using five metrics: accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC). The experimental results indicate that the BBO-DeepForest model performed best on the independent test set, with accuracy, precision, recall, F1-score, and average AUC values of 0.878, 0.682, 0.678, 0.678, and 0.961, respectively. A comparison with seven well-known imputation algorithms revealed the superiority of the selected imputation algorithm in recovering incomplete rock mass quality datasets. Model interpretation results showed that RQD and UCS are critical feature parameters for classifying slope rock mass quality. At last, the proposed BBO-DeepForest model was employed to verify the rock mass quality of three slopes at the Luming molybdenum mine, resulting in classifications consistent with on-site observations. It demonstrates that combining DeepForest with metaheuristic optimization algorithms is a feasible and accurate approach for intelligently classifying the rock mass quality of slopes. Full article

Rivers and groundwater supply 88% of Europe’s freshwater and are critical for public health. We examined whether cross-country differences in arsenic, lead, mercury, and nickel concentrations in groundwater and rivers are associated with disease burden. In an ecological cross-sectional study of 24 European countries, nationally aggregated concentrations from the European Environment Agency’s Waterbase Water Quality (2016–2019) were linked to cause-specific disability-adjusted life years (DALYs) from the Global Burden of Disease 2021 for six disease groups. Variables were z-standardized. Associations were assessed using Pearson correlations and linear regression with Benjamini–Hochberg correction. Missing concentrations were addressed via multiple imputation by chained equations using 1980–2025 monitoring records, and models were sequentially adjusted for health system, demographic, and economic indices. In groundwater, lead was positively associated with diabetes and kidney disease DALYs and remained significant after imputation and adjustment (β = 0.60, p = 0.011). In rivers, arsenic was positively associated with all-cause, cardiovascular, and neoplasm DALYs in unadjusted analyses but attenuated after adjustment and/or imputation. No consistent associations were observed for mercury or nickel. These continent-wide, non-causal findings can help prioritize monitoring and risk management and support progress toward Sustainable Development Goal 6. Full article

This study examines the spatio-temporal evolution of precipitation in the State of Guanajuato, Mexico, from 1981 to 2016 by analyzing monthly series from 65 meteorological stations. A rigorous data quality protocol was implemented, selecting stations with more than 30 years of continuous data and less than 10% missing values. Multiple Imputation by Chained Equations (MICE) with Predictive Mean Matching was applied to handle missing data, preserving the statistical properties of the time series as validated by Kolmogorov–Smirnov tests ($p=1.000$ for all stations). Homogeneity was assessed using Pettitt, SNHT, Buishand, and von Neumann tests, classifying 60 stations (93.8%) as useful, 3 (4.7%) as doubtful, and 2 (3.1%) as suspicious for monthly analysis. Breakpoints were predominantly clustered around periods of instrumental changes (2000–2003 and 2011–2014), underscoring the necessity of homogenization prior to trend analysis. The Trend-Free Pre-Whitening Mann–Kendall (TFPW-MK) test was applied to account for significant first-order autocorrelation (${\rho }_1 > 0.3$) present in all series. The analysis revealed no statistically significant monotonic trends in monthly precipitation at any of the 65 stations ($\alpha =0.05$). While 75.4% of the stations showed slight non-significant increasing tendencies (Kendall’s $\tau$ range: 0.0016 to 0.0520) and 24.6% showed non-significant decreasing tendencies ($\tau$ range: −0.0377 to −0.0008), Sen’s slope estimates were negligible (range: −0.0029 to 0.0111 mm/year) and statistically indistinguishable from zero. No discernible spatial patterns or correlation between trend magnitude and altitude ($\rho =-0.022$, $p>0.05$) were found, indicating region-wide precipitation stability during the study period. The integration of advanced imputation, multi-test homogenization, and robust trend detection provides a comprehensive framework for hydroclimatic analysis in semi-arid regions. These findings suggest that Guanajuato’s severe water crisis cannot be attributed to declining precipitation but rather to anthropogenic factors, primarily unsustainable groundwater extraction for agriculture. Full article

Streamflow in the Upper Williamson Basin of the Klamath Watershed is groundwater dominated with year-to-year fluctuations in both volume and duration, including multi-year periods with no streamflow. The relationship between precipitation, groundwater, and streamflow is difficult to characterize because of the limited number of monitoring wells, large data gaps, and a unique geologic structure that controls flow. To understand why surface flow has ceased entirely, we use the Groundwater Data Mapper Tool to impute gaps in the well data using machine learning and open-source Earth observation data and then compute changes in groundwater storage over time. Our research confirms that groundwater storage is correlated to streamflow and finds that there is a control groundwater storage below which flow does not occur. Furthermore, we find that groundwater storage is correlated to rainfall with a three- to four-year delay. This lag and the geologic structural control mean that even with several years of above-average precipitation, live flow may take years to resume. This insight allows water managers to understand and adjust for this highly irregular streamflow for better management decisions. Full article

The reconstruction of missing groundwater level data is of great importance in hydrogeological and environmental studies. This study provides a comprehensive and sequential approach for the reconstruction of groundwater level data near Lake Uluabat in Bursa, Turkey. This study addresses missing data reconstruction for both past and future events using the Gradient Boosting Regression (GBR) model. The reconstruction process is evaluated through model calibration metrics and changes in the statistical properties of the observed and reconstructed time series. To achieve this goal, the groundwater time series from two observational wells and lake water levels during the January 2004 to September 2019 period are used. The lake water level, the definition of the four seasons via the application of three dummy variables, and time are used as inputs in the prediction of groundwater levels in observation wells. The optimal GBR model calibration is achieved by training the dataset selected based on data gaps in the time series, while test-past and test-future datasets are used for model validation. Afterward, the GBR models are used in reconstructing the missing data both in the pre- and post-training data sets, and the performance of the models are evaluated via the Nash–Sutcliffe efficiency (NSE), Root Mean Square Percentage Error (RMSPE) and Performance Index (PI). The statistical properties of the time series including the probability distribution, maxima, minima, quartiles (Q1–Q3), standard error (SE), coefficient of variation (CV), entropy (H), and error propagation are also measured. It was concluded that GBR provides a good base for missing data reconstruction (the best performance was as high as NSE: 0.99, RMSPE: 0.36, and PI: 1.002). In particular, the standard error and the entropy of the system in one case, respectively, experienced a 53% and 35% rise, which was found to be tolerable and negligible. Full article

The development of civilization and the preservation of environmental ecosystems are strongly dependent on water resources. Typically, an insufficient supply of surface water resources for domestic, industrial, and agricultural needs is supplemented with groundwater resources. However, groundwater is a natural resource that must accumulate over many years and cannot be recovered after a short period of recharge. Therefore, the long-term management of groundwater resources is an important issue for sustainable development. The accurate prediction of groundwater levels is the first step in evaluating total water resources and their allocation. However, in the process of data collection, data may be lost due to various factors. Filling in missing data is a main problem that any research field must address. It is well known that to maintain data integrity, one effective approach is missing value imputation (MVI). In addition, it has been demonstrated that machine learning may be a better tool. Therefore, the main purpose of this study was to utilize a generative adversarial network (GAN) that consists of a generative model and a discriminative model for imputation. Although the GAN could not capture the groundwater level endpoints in every section, the overall simulation performance was still excellent to some extent. Our results show that the GAN can improve the accuracy of water resource evaluations. In the current study, two interdisciplinary deep learning methods, univariate and Seq2val (sequence-to-value), were used for groundwater level estimation. In addition to addressing the significance of the parameter conditions, the advantages and disadvantages of these two models in hydrological simulations were also discussed and compared. Regarding parameter selection, the simulation results for univariate analysis were better than those for Seq2val analysis. Finally, univariate was employed to examine the limits of the models in long-term water level simulations. Our results suggest that the accuracy of CNNs is better, while LSTM is better for the simulation of multistep prediction. Therefore, the interdisciplinary deep learning approach may be beneficial for providing a better evaluation of water resources. Full article

Obtaining and managing groundwater data is difficult as it is common for time series datasets representing groundwater levels at wells to have large gaps of missing data. To address this issue, many methods have been developed to infill or impute the missing data. We present a method for improving data imputation through an iterative refinement model (IRM) machine learning framework that works on any aquifer dataset where each well has a complete record that can be a mixture of measured and input values. This approach corrects the imputed values by using both in situ observations and imputed values from nearby wells. We relied on the idea that similar wells that experience a similar environment (e.g., climate and pumping patterns) exhibit similar changes in groundwater levels. Based on this idea, we revisited the data from every well in the aquifer and “re-imputed” the missing values (i.e., values that had been previously imputed) using both in situ and imputed data from similar, nearby wells. We repeated this process for a predetermined number of iterations—updating the well values synchronously. Using IRM in conjuncture with satellite-based imputation provided better imputation and generated data that could provide valuable insight into aquifer behavior, even when limited or no data were available at individual wells. We applied our method to the Beryl-Enterprise aquifer in Utah, where many wells had large data gaps. We found patterns related to agricultural drawdown and long-term drying, as well as potential evidence for multiple previously unknown aquifers. Full article

This study compares three imputation methods applied to the field observations of hydraulic head in subsurface hydrology. Hydrogeological studies that analyze the timeseries of groundwater elevations often face issues with missing data that may mislead both the interpretation of the relevant processes and the accuracy of the analyses. The imputation methods adopted for this comparative study are relatively simple to be implemented and thus are easily applicable to large datasets. They are: (i) the spline interpolation, (ii) the autoregressive linear model, and (iii) the patched kriging. The average of their results is also analyzed. By artificially generating gaps in timeseries, the results of the various imputation methods are tested. The spline interpolation is shown to be the poorest performing one. The patched kriging method usually proves to be the best option, exploiting the spatial correlations of the groundwater elevations, even though spurious trends due to the the activation of neighboring sensors at times affect their reconstructions. The autoregressive linear model proves to be a reasonable choice; however, it lacks hydrogeological controls. The ensemble average of all methods is a reasonable compromise. Additionally, by interpolating a large dataset of 53 timeseries observing the variabilities of statistical measures, the study finds that the specific choice of the imputation method only marginally affects the overarching statistics. Full article

Sustainable groundwater management requires an accurate characterization of aquifer-storage change over time. This process begins with an analysis of historical water levels at observation wells. However, water-level records can be sparse, particularly in developing areas. To address this problem, we developed an imputation method to approximate missing monthly averaged groundwater-level observations at individual wells since 1948. To impute missing groundwater levels at individual wells, we used two global data sources: Palmer Drought Severity Index (PDSI), and the Global Land Data Assimilation System (GLDAS) for regression. In addition to the meteorological datasets, we engineered four additional features and encoded the temporal data as 13 parameters that represent the month and year of an observation. This extends previous similar work by using inductive bias to inform our models on groundwater trends and structure from existing groundwater observations, using prior estimates of groundwater behavior. We formed an initial prior by estimating the long-term ground trends and developed four additional priors by using smoothing. These prior features represent the expected behavior over the long term of the missing data and allow the regression approach to perform well, even over large gaps of up to 50 years. We demonstrated our method on the Beryl-Enterprise aquifer in Utah and found the imputed results follow trends in the observed data and hydrogeological principles, even over long periods with no observed data. Full article

Accurate characterization of groundwater resources is required for sustainable management. Due to the cost of installing monitoring wells and challenges in collecting and managing in situ data, groundwater data are sparse—especially in developing countries. In this study, we demonstrate an analysis of long-term groundwater storage changes using temporally sparse but spatially dense well data, where each well had as few as one historical groundwater measurement. We developed methods to synthetically estimate groundwater table elevation (WTE) times series by clustering wells using two different methods; a uniform grid and k-means-constrained clustering to create pseudo-wells. These pseudo-wells had a more complete groundwater level time history, which we then temporally and spatially interpolated to analyze groundwater storage changes in an aquifer. We demonstrated these methods on the Beryl-Enterprise aquifer in Utah, USA, where other researchers quantified the groundwater storage depletion rate, and the wells had a large number of historical measurements. We randomly used one measurement per well and showed that our methods yielded storage depletion rates similar to published values. We applied the method to a region in southern Niger where wells had only one measurement per well, and showed that our estimated groundwater storage change trend reasonably matched that which was calculated using GRACE satellite data. Full article

Ammonium is one of the main inorganic pollutants in groundwater, mainly due to agricultural, industrial and domestic pollution. Excessive ammonium can cause human health risks and environmental consequences. Its temporal and spatial distribution is affected by factors such as meteorology, hydrology, hydrogeology and land use type. Thus, a groundwater ammonium analysis based on limited sampling points produces large uncertainties. In this study, organic matter content, groundwater depth, clay thickness, total nitrogen content (TN), cation exchange capacity (CEC), pH and land-use type were selected as potential contributing factors to establish a machine learning model for fitting the ammonium concentration. The Shapley Additive exPlanations (SHAP) method, which explains the machine learning model, was applied to identify the more significant influencing factors. Finally, the machine learning model established according to the more significant influencing factors was used to impute point data in the study area. From the results, the soil organic matter feature was found to have a substantial impact on the concentration of ammonium in the model, followed by soil pH, clay thickness and groundwater depth. The ammonium concentration generally decreased from northwest to southeast. The highest values were concentrated in the northwest and northeast. The lowest values were concentrated in the southeast, southwest and parts of the east and north. The spatial interpolation based on the machine learning imputation model established according to the influencing factors provides a reliable groundwater quality assessment and was not limited by the number and the geographical location of samplings. Full article

Accurately assessing groundwater storage changes in Niger is critical for long-term water resource management but is difficult due to sparse field data. We present a study of groundwater storage changes and recharge in Southern Niger, computed using data from NASA Gravity Recovery and Climate Experiment (GRACE) mission. We compute a groundwater storage anomaly estimate by subtracting the surface water anomaly provided by the Global Land Data Assimilation System (GLDAS) model from the GRACE total water storage anomaly. We use a statistical model to fill gaps in the GRACE data. We analyze the time period from 2002 to 2021, which corresponds to the life span of the GRACE mission, and show that there is little change in groundwater storage from 2002–2010, but a steep rise in storage from 2010–2021, which can partially be explained by a period of increased precipitation. We use the Water Table Fluctuation method to estimate recharge rates over this period and compare these values with previous estimates. We show that for the time range analyzed, groundwater resources in Niger are not being overutilized and could be further developed for beneficial use. Our estimated recharge rates compare favorably to previous estimates and provide managers with the data required to understand how much additional water could be extracted in a sustainable manner. Full article

Groundwater resources are expensive to develop and use; they are difficult to monitor and data collected from monitoring wells are often sporadic, often only available at irregular, infrequent, or brief intervals. Groundwater managers require an accurate understanding of historic groundwater storage trends to effectively manage groundwater resources, however, most if not all well records contain periods of missing data. To understand long-term trends, these missing data need to be imputed before trend analysis. We present a method to impute missing data at single wells, by exploiting data generated from Earth observations that are available globally. We use two soil moisture models, the Global Land Data Assimilation System (GLDAS) model and National Oceanic and Atmospheric Administration (NOAA) Climate Prediction Center (CPC) soil moisture model to impute the missing data. Our imputation method uses a machine learning technique called Extreme Learning Machine (ELM). Our implementation uses 11 input data-streams, all based on Earth observation data. We train and apply the model one well at a time. We selected ELM because it is a single hidden layer feedforward model that can be trained quickly on minimal data. We tested the ELM method using data from monitoring wells in the Cedar Valley and Beryl-Enterprise areas in southwest Utah, USA. We compute error estimates for the imputed data and show that ELM-computed estimates were more accurate than Kriging estimates. This ELM-based data imputation method can be used to impute missing data at wells. These complete time series can be used improve the accuracy of aquifer groundwater elevation maps in areas where in-situ well measurements are sparse, resulting in more accurate spatial estimates of the groundwater surface. The data we use are available globally from 1950 to the present, so this method can be used anywhere in the world. Full article