Hydraulic models are essential for simulating river water levels, but their results often deviate from reality due to systematic errors and modelling uncertainties. These are commonly addressed through roughness calibration, which can obscure underlying model issues and limit reliability under varying flow conditions. This study explores alternative strategies that align simulation results with measurements. We apply: (i) a mesh elevation modification technique to correct spatial representation errors, and (ii) an upstream boundary condition adjustment to refine flow inputs. The approach is tested on the Waal River in the Netherlands, with both adjustments implemented without roughness calibration. The model successfully reproduces observed water levels across the full discharge range, achieving a mean absolute error below 10 cm over two hydrologically distinct years. The findings demonstrate that targeted structural corrections and informed boundary adjustments can improve model performance, providing a complementary approach to traditional calibration-based methods.
The study aims to evaluate the suitability of gridded and observed rainfall data for detecting droughts in areas with limited rainfall information. We utilized a time-series historical rainfall dataset spanning from 1990 to 2022. To evaluate dry and wet conditions during this period, we utilized the Standardized Precipitation Index (SPI). The ARIMA model was employed to predict future drought trend in the study area. Analysis revealed a low bias of 3.89% between gridded and ground-based rainfall datasets. This corresponded to a strong coefficient of determination (R2 = 0.707) and correlation coefficient (r = 0.84), indicating robust agreement between the datasets. A 50% chance of near-normal drought is predicted to occur in the study area over the next decade. Policymakers in Bontanga should therefore prioritize rainwater harvesting, water-saving irrigation technologies, and drought-resilient crop varieties to strengthen food production resilience across northern Ghana.
Flood and drought disasters are often assessed independently, resulting in fragmented adaptation strategies and limited consideration of their interactions. This study presents an integrated drought-flood risk assessment framework that combines GIS-based multi-criteria evaluation (MCE) with hydrologically derived indicators to jointly assess both disasters. The framework integrates hazard, vulnerability, and exposure components, and introduces a transition layer to quantify the temporal linkage between drought and subsequent flood events. Applied to the Muda River Basin in Malaysia, the results identify the southwestern basin as a key hotspot driven by intensive land use activities and high population exposure. Incorporating the transition layer increases high-risk areas from 14.60% to 25.5% of the basin, highlighting the importance of sequential hazard dynamics. This framework addresses a critical gap in multi-hazard assessment and supports climate adaptation, spatial planning, and disaster risk reduction.
This study undertakes an uncertainty analysis for two regional flood frequency analysis techniques (random forest regression (RFR) and multiple linear regression (MLR)). This uses data from 201 gauged catchments of southeast Australia. To quantify uncertainty, a synthetic dataset comprising 15 000 samples from the original 201 catchments' data was generated via Monte Carlo simulation. Model performance was evaluated using Monte Carlo cross-validation. Key statistical indicators, including relative error, absolute relative error, and median relative error (REr) were used to assess the model performance. Results show that RFR consistently outperforms MLR with narrower uncertainty bounds. Median REr for MLR (original data) ranges 40.32 +/- 5.09% to 54.94 +/- 7.51%, MLR (synthetic data) 38.69 +/- 0.64% to 51.79 +/- 0.88%, and RFR (synthetic data) 36.81 +/- 0.68% to 50.84 +/- 0.89%. These results demonstrate that artificial intelligence (AI)-based RFR provides a lower error range and more consistent uncertainty estimates compared to the MLR.
Estimating the suspended sediment load (SSL) and understanding the impacted factors can aid in watershed management. This study evaluates the potential of ElasticNet linear regression (ENLR), random forest (RF), AdaBoost, CatBoost, and extreme gradient boosting (XGB) for SSL estimation at Ziway catchment (Ethiopia). The Shapley additive explanation method was used to determine the contribution of different driving factors for river SSL generation. Then, hybrid models, namely ENLRXGB, ENLRRF, ENLRAdaBoost, and ENLRCatBoost, were proposed by coupling the ENLR model with ensemble tree models to capture both the linear and non-linear patterns of SSL. The finding revealed that XGB outperformed other individual models. In addition, combining linear and non-linear models produced better performance, i.e. ENLRXGB demonstrated the highest accuracy Nash-Sutcliffe efficiency (NSE = 0.967), improving the base model's NSE by 3.755-21.33%. Overall, the results demonstrate the efficiency of hybrid models for estimating river SSL in this data-limited agricultural catchment.
Historical water management practices often hold lessons for the sustainability of contemporary environmental systems, especially in arid regions where freshwater resources are declining under anthropogenic stresses. This study examines 740 years of water management (1171-1911 CE) at the Longzici karst spring in North China, focusing on how integrated infrastructure, governance, and cultural practices addressed semi-arid water scarcity. Based on historical literature analysis and field surveys, this study has found that: (1) gradual channel expansion paired with flood-control measures were successful in balancing water supply demands and disaster-risk reduction; (2) a hierarchical organizational structure created equitable water allocation and assisted in conflict mediation; and (3) religious and cultural practices served to consolidate community consensus on water reverence and utilization. The system for the utilization of Longzici spring discharge provides actionable lessons for contemporary water management practices. It highlights shared values among diverse stakeholders through ritualization and collective community extending beyond immediate individual benefits.
