The decline in groundwater storage (GWS) poses a critical threat to water security in semi-arid regions where increasing agricultural water demand and climate variability are increasing pressure on aquifers. This study presents a novel hybrid modeling framework integrating multi-source satellite and climate data (GRACE, GLDAS, TerraClimate, and MODIS) with machine learning and explanatory artificial intelligence techniques for the long-term assessment and interpretation of GWS anomalies in the data-poor Iğdır Basin. Three different modeling approaches were developed: XGBoost, Long Short-Term Memory (LSTM) networks, and their combined model, and interpreted using the Shapley Additive Explanations (SHAP) method. The results showed a significant long-term decreasing trend in groundwater storage anomalies at a rate of −0.87 mm per month during the 2002–2016 period, indicating continuous depletion. The LSTM model demonstrated the best performance with R2 of 0.59, RMSE of 19.5 mm, and MAE of 15.1 mm, revealing the dominant role of temporal dependencies in groundwater systems. SHAP analysis identified lagged groundwater anomalies (especially GWS_lag3) as the most effective predictors; this may reflect the memory effect and lagged response specific to semi-arid aquifer systems, but this interpretation needs to be validated in different study areas. Snow water equivalent and total water storage anomalies also emerged as significant determinants, while the direct effect of instantaneous precipitation was found to be limited. This study addresses significant gaps in the literature by combining sequence-based modeling with model interpretability in a semi-arid closed basin. The findings highlight the necessity of using system memory and explainable artificial intelligence together for reliable groundwater prediction. While the proposed hybrid approach has the potential for application in other semi-arid regions, its broader usability needs to be supported by independent validation studies under different hydrogeological and climatic conditions.
Mehmet Ali Çelik, Adile Bilik, Yasin Paşa· Hydrology· 0 citations
Long-term exposure to ambient fine particulate matter (PM2.5) is the leading environmental risk factor for premature mortality worldwide, yet comprehensive province-level evidence quantifying its health burden across Türkiye remains limited. This study investigated the spatial relationship between long-term PM2.5 exposure and all-cause attributable mortality across all 81 Turkish provinces in 2022 using province-level annual mean PM2.5 concentrations and World Health Organisation (WHO) AirQ+ estimates of PM2.5-attributable deaths among adults aged ≥30 years, assuming a counterfactual concentration of 5 µg/m3. The association between PM2.5 exposure and mortality was evaluated using Pearson and Spearman correlation analyses, ordinary least squares (OLS) regression, a log–log elasticity model, and population-weighted regional and exposure-quartile comparisons, while national temporal indicators for 2010–2023 were reported solely as supplementary context for the primary single-year 2022 cross-sectional analysis. The population-weighted annual mean PM2.5 concentration was 27.0 µg/m3, exceeding the WHO Air Quality Guideline by a factor of 5.4, and all 81 provinces exceeded the recommended threshold. The bivariate OLS model accounted for 41% of the between-province variation in attributable mortality rates (OLS slope = 3.23 additional deaths per 100,000 population for each 1 µg/m3 increase in PM2.5; 95% CI: 2.37–4.10; R2 = 0.41; p < 0.001), while the log–log elasticity model indicated that a 1% increase in PM2.5 concentration was associated with a 0.80% increase in the attributable mortality rate (95% CI: 0.65–0.95). The attributable fraction of natural-cause mortality increased progressively from 8.8% in the lowest exposure quartile to 24.6% in the highest. Nationwide, an estimated 68,440 premature deaths, representing 14.2% of all natural-cause deaths among adults aged ≥30 years, were attributable to PM2.5 exposure. These findings quantify a steep, spatially graded PM2.5-attributable mortality burden across Türkiye. As the attributable estimates derive from the WHO AirQ+ concentration–response function, the gradient describes the magnitude and spatial distribution of the modelled burden rather than an independently estimated exposure–response relationship, and on that basis the results support the adoption of WHO-aligned air-quality standards and accelerated decarbonization strategies to reduce the national health burden attributable to ambient air pollution.
Nebile Özmen, V. Duran, Fatma Şencan et al.· Toxics· 0 citations
