As part of the International Decade for Action “Water for Sustainable Development” (2018–2028), assessing the water and irrigation potential of river basin catchment areas is essential for sustainable water use and agricultural development. This study assesses the water and irrigation potential of the Karatal River Basin in the Zhetysu Region of the Republic of Kazakhstan using trend analysis of climatic and hydrological indicators considered as functions of time, providing a basis for forecasting. Source data for this study were obtained from yearbooks published by the “Kazhydromet” and from the regional water management organization “Kazvodkhoz”. Analysis of the spatial and temporal variability of climatic and hydrological indicators revealed both stochastic and deterministic components, with the latter expressed as significant trends, providing a basis for assessing the water and irrigation potential of river basin catchments. For the entire period from 1966 to 2024, the estimated total increase in the average annual air temperature at the meteorological stations under study, determined based on a linear trend, ranged from 1.444 to 6.078 °C. Annual precipitation showed mixed trends, with linear trend coefficients ranging from −2.916 to +1.989 mm/year. The estimated total decrease in the average annual flow of the Karatal River, determined based on a linear trend over the same period, was 15.658 m3/s at the Ushtobe hydrological station and 31.575 m3/s at the Naimansuek station. The river discharge decrease occurred against a backdrop of rising average annual air temperatures and changes in precipitation patterns, which affected the components of the basin’s water balance, including river discharge and evaporation. Based on the identified trends and indicators of the spatial organization of water use, the water availability of the Karatal River Basin was assessed using the water stress index and the per capita water availability indicator—the Falkenmark index. The volume of available water resources per capita ranged from 32.233 to 5.872 thousand m3 per person per year in the Eskeldi District and from 87.826 to 44.814 thousand m3 per person per year in the Karatal District. The water stress index ranged from 21.130% to 13.096% and from 19.10% to 2.848%, respectively, indicating water stress levels ranging from moderately high to moderately low and from moderately low to low. An assessment of trends in the dynamics of climatic and hydrological indicators, as well as criteria for water resource availability, revealed differences in the structure of water use among administrative districts.
To reveal how climate change and reservoir operation differentially affect water temperature in contrasting climate basins, this study applies a coupled hydrodynamic–thermal HEC-RAS model to the Ili River (temperate continental) and the Klamath River (Mediterranean–montane). Using ERA5 reanalysis and CMIP6 projections, we quantify the effects of climate forcing and reservoir regulation on spatiotemporal water temperature patterns, and compare the two drivers side by side. Three main findings emerge. (1) Reservoir thermal effects are climate-dependent. In the Ili, cold-water releases lower spring temperatures (9 °C→20 °C without dam vs. 5 °C→18 °C with dam) and reduce diel fluctuations. In the Klamath, cascade reservoirs cause autumn heat retention and anomalous midstream warming, which disappears in the simulated no-dam scenario. (2) Climate change drives non-uniform warming. Under SSP5-8.5 by 2080, annual mean water temperature increases by 1.5–2.5 °C in both basins, but with strong seasonality. The Ili warms most in spring and autumn (2.0–3.7 °C), while the Klamath shows an autumn peak (+3.7 °C in October). The annual mean increases are lower than the peak monthly increases because winter months contribute little to the annual average. (3) Energy budget analysis reveals the warming mechanism. Both basins share a similar budget: solar radiation (summer peak around 370–375 W/m2 in 2020, increasing to about 385–395 W/m2 by 2080) and atmospheric radiation are primary heat sources, and longwave radiation (summer about −410 W/m2 in 2020, reaching about −430 W/m2 by 2080) is the main heat sink. By 2080, enhanced solar radiation (+5%) and reduced longwave loss (–5%) increase net summer heat gain, driving warm-season warming. These findings provide a basis for considering reservoir operations in relation to regional climate contexts in transboundary rivers, and for identifying where thermal responses are most sensitive to reservoir regulation.
Mountain rivers are characterized by steep gradients, high flow velocities, and flash flood regimes, and the interaction between in-stream bridge pier groups and flood flows is a persistent challenge in bridge hydraulics. This study investigates the flood hydrodynamic characteristics at a mountain river bridge section and clarifies the coupled pier–flow–debris interaction mechanism, using the 19 July 2024 collapse of Yanpingcun Bridge No. 2 over the Jinqian River in Shangluo, Shaanxi Province, China, as a case study. High-resolution topographic data (0.5 m) derived from three-dimensional laser scanning point clouds were used to construct a two-dimensional hydrodynamic model (IFMS) and a three-dimensional air–water two-phase flow model (volume-of-fluid method, approximately 8 million cells) of the bridge site. Six scenarios were simulated to quantify the effects of pier arrangement, tie beam elevation, and flood protection works on water level rise, velocity distribution, and main flow direction, and the models were validated against surveyed water marks and video-based velocity measurements. The results show the following. (1) The pier group produced a marked flow-blocking effect: the maximum water level upstream of Pier 10 rose by 1.28 m relative to the natural channel, the main flow direction deflected clockwise by about 18°, and a triangular confluence zone formed between the pier group and the right-bank retaining wall. (2) The existing skewed pier arrangement raised the upstream water level by 0.78 m compared with a flow-aligned arrangement and produced a disordered flow pattern unfavourable to debris passage. (3) Three-dimensional analysis showed local water level rises of 0.52–1.20 m on the upstream faces of the piers, with maximum local velocities of 5.5–6.3 m/s along the pier flanks; the maximum velocity occurred at a water depth of 1.0–2.0 m rather than at the free surface. (4) Tie beams exposed above the bed intensified local flow blockage and increased inter-pier velocities (0.3–0.8 m/s), so that debris was readily trapped beneath the tie beams, forming a blockage–backwater–load positive feedback loop. The findings indicate that the large angle between the piers and the main flow, the dense pier group, exposed tie beams, and debris blockage were the key hydrodynamic factors in the collapse, and that the amplifying feedback among piers, flow, and debris was the core mechanism of structural failure. The results provide a scientific basis for the design optimization and flood safety assessment of bridges aligned along mountain rivers.
Enhancement harvesting and filter-feeding fish stocking are increasingly applied in shallow eutrophic lakes, but their water-quality effects are difficult to evaluate because management signals overlap with seasonal and hydrometeorological variability. We analyzed dense Sentinel-2 observations of Taibo Lake during 2019–2026 using three management stages, phase-specific event windows, and matched forbidden-fishing controls. Regressions adjusted for antecedent precipitation and month used 25 scenes from the 2024 and 2025 events. Chlorophyll-related proxies were descriptively lower during enhancement harvesting than in the 2019 reference year, with no consistent advantage over forbidden fishing. The contrast with 2019 remains confounded by year-specific environmental conditions. Integrated optical and turbidity-related indicators showed no uniform directional change, and representative September scenes revealed substantial within-lake heterogeneity. Event-window comparisons showed no consistent immediate directional response to harvesting or stocking. Exploratory models suggested higher post-stocking normalized difference chlorophyll index (NDCI) values and high-NDCI ratios; only NDCI remained nominally significant after Bonferroni adjustment (adjusted p = 0.036 versus 0.204). Limited event replication and unmodeled temporal dependence precluded confirmatory inference. These findings support phase-specific water-quality monitoring with seasonally matched baselines and continued observation through the post-stocking period to inform adaptive enhancement-fishery management.
Extreme streamflow forecasting remains challenging because hydrological time series are highly imbalanced and meteorological forcings are non-stationary. These characteristics can cause conventional data-driven models to underestimate flood peaks. While wavelet-coupled LSTM models can extract multi-scale hydrological features to boost forecasting performance, their static feature fusion limits the model’s ability to balance peak-flow and baseflow prediction, limiting adaptive forecasting for variable flow scenarios. Moreover, the Gaussian Mixture Model (GMM) adopted for probabilistic prediction suffers from gradient explosion and numerical collapse when fitting heavy-tailed extreme hydrological data, degrading prediction. To address these limitations, we propose a Frequency-Gated Wavelet Transform LSTM (FG-WT-LSTM) framework. The model uses the stationary wavelet transform (SWT) to decompose the input into multiple frequency components. A frequency-domain gating mechanism then assigns time-varying weights to these components, enhancing rainfall-related high-frequency signals while reducing flood-induced high-frequency features noise and suppressing redundant spectral noise. A LogSumExp stabilization strategy is further embedded into GMM to eliminate numerical instability. Validated on the CAMELS dataset, the proposed model achieves a median NSE of 0.872 and reduces FHV bias to −1.108%, with probabilistic prediction intervals reaching 93.9% coverage. These results indicate that the proposed framework can improve peak-flow prediction while maintaining stable baseflow simulation for accurate extreme flood forecasting and hydrological uncertainty quantification.
