Search Results (609)

This paper advances machine learning (ML)-based streamflow prediction by strategically selecting rainfall events, introducing a new loss function, and addressing rainfall forecast uncertainties. Focusing on the Iowa River Basin, we applied the stochastic storm transposition (SST) method to create realistic rainfall events, which were input into a hydrological model to generate corresponding streamflow data for training and testing deterministic and probabilistic ML models. Long short-term memory (LSTM) networks were employed to predict streamflow up to 12 h ahead. An active learning approach was used to identify the most informative rainfall events, reducing data generation effort. Additionally, we introduced a novel asymmetric peak loss function to improve peak streamflow prediction accuracy. Incorporating rainfall forecast uncertainties, our probabilistic LSTM model provided uncertainty quantification for streamflow predictions. Performance evaluation using different metrics improved the accuracy and reliability of our models. These contributions enhance flood forecasting and decision-making while significantly reducing computational time and costs. Full article

Mitigating the consequences of flash rainfall has become essential for the safety of populations and the promotion of local tourism. A non-structural measure could involve a sensor-based nowcasting system to detect increasingly frequent and intense rainfall events driven by climate change. Therefore, developing wide-range, connected, cheap, small, and easy-to-install rain gauges is desirable. To achieve a useful network of monitoring, a set of technologies such as electrodynamic sensor devices supported by real-time processing and the Internet of Things is proposed. This comparative investigation aimed to evaluate the implementation-friendly network of small, low-cost, solid-state pluviometers for near-real-time monitoring of an early warning system. The ability of a recent patent to provide cumulative rainfall estimates every ten seconds was evaluated for river system flooding, which extends the warning time by 3–4 min in a 1 km² basin. Our results found that even with a rainfall uncertainty of 10%, a network of these new instruments reduced errors in flood wave severity and time estimations. Moreover, intensity–duration thresholds of landslide triggering and debris movements can be modified by flash rainfalls. Specifically, coastal areas with high-density populations can greatly benefit from this solution. Full article

The sustainable use and management of groundwater resources is a challenging issue due to population growth and climate change. Accurate quantification of groundwater recharge is a basic requirement for effective groundwater resource management, yet it is still lacking in many areas around the world. The study was designed to estimate recharge to groundwater from natural rainfall in the Gilgel Gibe and Dhidhessa catchments in southwestern Ethiopia, employing the water table fluctuation (WTF) and chloride mass balance (CMB) techniques. These methods are being applied for the first time in the study area and have not previously been used in these catchments. Given the region’s data scarcity, a community-based data collection program was implemented and supplemented with additional field measurements and secondary data sources. Groundwater level, spring discharge, and rainfall were monitored over the 2022/2023 hydrological year. Groundwater level fluctuations were found to be influenced by topography and rainfall patterns, reaching 8.2 m in amplitude in the upstream part of the catchments. Chloride concentrations were determined in groundwater samples collected from hand-dug wells and springs, and rainwater was also collected. Rainwater exhibited a mean chloride concentration of 2.46 mg/L, while groundwater chloride concentrations ranged from 3 mg/L to 36.99 mg/L. The estimated recharge rates varied spatially, ranging from 170 to 850 mm/year using the CMB method (11% to 55% of annual rainfall, mean recharge rate of 454 mm/year) and from 76 to 796 mm/year using the WTF method (4% to 43% of annual rainfall, mean recharge rate of 439 mm/year). Notably, recharge estimates were lowest downstream in the lowland areas and highest upstream in the highland regions. Rainfall amount, local lithology, and topography were identified as major influences on groundwater recharge across the study area. Both CMB and WTF methods were deemed applicable in the volcanic aquifers, provided that all the respective assumptions are followed. This study significantly contributes to the groundwater dataset for the region, in addition to recharge estimation and the research conclusions, emphasizing the importance of long-term monitoring and time series analysis of chloride data to reduce uncertainties. The work serves as a valuable reference for researchers, policymakers, and regional water resource managers. Full article

For the design of stormwater pumping stations, there is often uncertainty regarding the selection of an appropriate rainfall event to determine the required pumping capacity and temporary storage volume for managing extreme events of a given magnitude. To account for the risk of system failure, the return period is considered, as recommended based on the size of the catchment’s drainage area or other considerations, depending on the local regulations of a country. This study focused on analysing the direct runoff volume from the catchment, the storage volume required for the operation of the pumping system, and the order of magnitude of the design flow rate. The results indicate that a rainfall event with a duration of at least twice the time of concentration should be used. The design flow rate should range between 50% and 70% of the peak discharge, and designing for flow rates near the peak is not advisable, as it can lead to intermittent pump operation and result in an oversized installed capacity. The methodology developed in this research was applied to the Coastal Protection Project located in the city of Cartagena, Colombia, which includes a 2045.6-m-long box culvert with a cross-sectional area of 2 × 2 m, and three pumping stations, each equipped with three pumps rated at 0.75 m³/s, for a total installed capacity of 6.75 m³/s. Full article

Ground-based weather radars are essential to better understand precipitation systems, to improve the Quantitative Precipitation Estimation (QPE), and to subsequently provide input to hydrological models. However, reflectivity measured by radars is typically affected by various sources of uncertainty, including attenuation and calibration errors. Due to these limitations, the two ground-based X-band weather radars of Cyprus, namely, at Rizoelia (LCA) and Nata (PFO), have not yet been employed for QPE. This study presents a dual neural network framework with the ultimate goal of converting the ground-based radar raw reflectivity to rainfall rate, using satellite and in situ observations. The two ground-based radars are aligned with GPM DPR using the volume-matching method. Preliminary results demonstrate the feasibility of converting raw ground-based radar reflectivity to rainfall estimates using neural networks trained with spaceborne and in situ observations. Full article

Amid growing urban climate uncertainty and complex water demand, conventional standalone rainwater harvesting (RWH) systems often fail to ensure supply reliability and overflow control. Most existing studies focus on single-function building clusters, leaving a gap in understanding how functionally diverse groups with complementary demand patterns can be coordinated. This study addresses this gap by applying an hourly water balance model to compare decentralized and coordinated modes for an integrated RWH system serving a campus and adjacent student dormitories in Kitakyushu, Japan. Five performance metrics were evaluated: potable water supplementation, reliability, non-potable replacement rate, overflow volume, and overflow days. The results show that coordinated operation reduced annual potable supplementation by 14.1%, improved overall reliability to 81.7% (a 9.6% gain over decentralized operation), and increased the replacement rate to 87.9%. Overflow volume decreased by 295 m³ and overflow days by five, with pronounced benefits during summer rainfall peaks. Differential heatmaps further revealed distinct spatiotemporal advantages, though temporary disruptions occurred under extreme events. Overall, the study demonstrates that cross-functional coordination can enhance system resilience and operational stability, while highlighting the need for adaptive scheduling and real-time information systems for broader urban applications. Full article

Accurately estimating the time of concentration (Tc) is critical for hydrological modelling, flood forecasting, and hydraulic infrastructure design. However, conventional methods often overlook the combined effects of rainfall intensity and antecedent soil moisture, thereby limiting their applicability under changing climates. This study presents a novel approach that integrates data-driven techniques with remote sensing data to improve Tc estimation. This method was successfully applied in the Kalu River Basin, Sri Lanka, demonstrating its performance in a tropical catchment. While an overall inverse relationship between rainfall intensity and Tc was observed, deviations in several events underscored the influence of initial soil moisture conditions on catchment response times. To address this, a modified kinematic wave-based equation incorporating both rainfall intensity and soil moisture was developed and calibrated, achieving high predictive accuracy (calibration: R² = 0.97, RMSE = 1.1 h; validation: R² = 0.96, RMSE = 0.01 h). A hydrological model was developed to assess the impacts of Tc uncertainties on design hydrographs. Results revealed that underestimating Tc led to substantially shorter lag times and significantly increased peak flows, highlighting the sensitivity of flood simulations to Tc variability. This study highlights the need for improved $T_C$ estimation and presents a robust, transferable methodology for enhancing hydrological predictions and climate-resilient infrastructure planning. Full article

A fundamental limitation of maximum likelihood and Bayesian methods under model misspecification is that the asymptotic covariance matrix of the pseudo-true parameter vector ${\mathsf{\theta }}_{*}$ is not the inverse of the Fisher information, but rather the sandwich covariance matrix ${\displaystyle \frac{1}{n}}{\mathbf{A}}_{*}^{-1}{\mathbf{B}}_{*}^{\phantom{-1}}{\mathbf{A}}_{*}^{-1}$, where ${\mathbf{A}}_{*}$ and ${\mathbf{B}}_{*}$ are the sensitivity and variability matrices, respectively, evaluated at ${\mathsf{\theta }}_{*}$ for training data record ${\omega }_1,\dots ,{\omega }_n$. This paper makes three contributions. First, we review existing approaches to robust posterior sampling, including the open-faced sandwich adjustment and magnitude- and curvature-adjusted Markov chain Monte Carlo (MCMC) simulation. Second, we introduce a new sandwich-adjusted MCMC method. Unlike existing approaches that rely on arbitrary matrix square roots, eigendecompositions or a single scaling factor applied uniformly across the parameter space, our method employs a parameter-dependent learning rate $\lambda \left(\mathsf{\theta }\right)$ that enables direction-specific tempering of the likelihood. This allows the sampler to capture directional asymmetries in the sandwich distribution, particularly under model misspecification or in small-sample regimes, and yields credible regions that remain valid when standard Bayesian inference underestimates uncertainty. Third, we propose information-theoretic diagnostics for quantifying model misspecification, including a strictly proper divergence score and scalar summaries based on the Frobenius norm, Earth mover’s distance, and the Herfindahl index. These principled diagnostics complement residual-based metrics for model evaluation by directly assessing the degree of misalignment between the sensitivity and variability matrices, ${\mathbf{A}}_{*}$ and ${\mathbf{B}}_{*}$. Applications to two parametric distributions and a rainfall-runoff case study with the Xinanjiang watershed model show that conventional Bayesian methods systematically underestimate uncertainty, while the proposed method yields asymptotically valid and robust uncertainty estimates. Together, these findings advocate for sandwich-based adjustments in Bayesian practice and workflows. Full article

Groundwater level (GWL) prediction in arid regions faces two fundamental challenges in conventional numerical modeling: (i) irreducible parameter uncertainty, which systematically reduces predictive accuracy; (ii) oversimplification of nonlinear process interactions, which leads to error propagation. Although machine learning (ML) methods demonstrate strong nonlinear mapping capabilities, their standalone applications often encounter prediction bias and face the accuracy–generalization trade-off. This study proposes a hybrid TCN–Transformer–LSTM (TTL) model designed to address three key challenges in groundwater prediction: high-frequency fluctuations, medium-range dependencies, and long-term memory effects. The TTL framework integrates TCN layers for short-term features, Transformer blocks to model cross-temporal dependencies, and LSTM to preserve long-term memory, with residual connections facilitating hierarchical feature fusion. The results indicate that (1) at the monthly scale, TTL reduced RMSE by 20.7% (p < 0.01) and increased R² by 0.15 compared with the Groundwater Modeling System (GMS); (2) during abrupt hydrological events, TTL achieved superior performance (R² = 0.96–0.98, MAE < 0.6 m); (3) PCA revealed site-specific responses, corroborating the adaptability and interpretability of TTL; (4) Grad-CAM analysis demonstrated that the model captures physically interpretable attention mechanisms—particularly evapotranspiration and rainfall—thereby providing clear cause–effect explanations and enhancing transparency beyond black-box models. This transferable framework supports groundwater forecasting, risk warning, and practical deployment in arid regions, thereby contributing to sustainable water resource management. Full article

Accurate real-time information about soil moisture (SM) at a large scale is essential for improving hydrological modeling, managing water resources, and monitoring extreme weather events. This study presents a framework using convolutional long short-term memory (ConvLSTM) network to produce short- (1, 3, and 7 days ahead) and mid-term (14 and 30 days ahead) forecasts of SM at surface (0–10 cm) and subsurface (10–40 and 40–100 cm) soil layers across the contiguous U.S. The model was trained with five-year period (2018–2022) datasets including Soil Moisture Active Passive (SMAP) level 3 ancillary covariables, North American Land Data Assimilation System phase 2 (NLDAS-2) SM product, shortwave infrared reflectance from Moderate Resolution Imaging Spectroradiometer (MODIS), and terrain features (e.g., elevation, slope, curvature), as well as soil texture and bulk density maps from the Soil Landscape of the United States (SOLUS100) database. To develop and evaluate the model, the dataset was divided into three subsets: training (January 2018–January 2021), validation (2021), and testing (2022). The outputs were validated with observed in situ data from the Soil Climate Analysis Network (SCAN) and the United States Climate Reference Network (USCRN) soil moisture networks. The results indicated that the accuracy of SM forecasts decreased with increasing lead time, particularly in the surface (0–10 cm) and subsurface (10–40 cm) layers, where strong fluctuations driven by rainfall variability and evapotranspiration fluxes introduced greater uncertainty. Across all soil layers and lead times, the model achieved a median unbiased root mean square error (ubRMSE) of 0.04 cm³ cm⁻³ with a Pearson correlation coefficient of 0.61. Further, the performance of the model was evaluated with respect to both land cover and soil texture databases. Forecast accuracy was highest in coarse-textured soils, followed by medium- and fine-textured soils, likely because the greater penetration depth of microwave observations improves SM retrieval in sandy soils. Among land cover types, performance was strongest in grasslands and savannas and weakest in dense forests and shrublands, where dense vegetation attenuates the microwave signal and reduces SM estimation accuracy. These results demonstrate that the ConvLSTM framework provides skillful short- and mid-term forecasts of surface and subsurface soil moisture, offering valuable support for large-scale drought and flood monitoring. Full article

Arid mountainous regions are vulnerable to extreme hydrological events such as floods and droughts. Providing accurate and continuous rainfall records with no gaps is crucial for effective flood mitigation and water resource management in these and downstream areas. Satellite data and geospatial interpolation can be employed for this purpose and to provide continuous data series. However, it is essential to thoroughly assess these methods to avoid an increase in errors and uncertainties in the design of flood protection and water resource management systems. The current study focuses on the mountainous region in northern Oman, which covers approximately 50,000 square kilometers, accounting for 16% of Oman’s total area. The study utilizes data from 279 rain gauges spanning from 1975 to 2009, with varying annual data gaps. Due to the limited accuracy of satellite data in arid and mountainous regions, 51 geospatial interpolations were used to fill data gaps to yield maximum annual and total yearly precipitation data records. The root mean square error (RMSE) and correlation coefficient (R) were used to assess the most suitable geospatial interpolation technique. The selected geospatial interpolation technique was utilized to generate the spatial distribution of annual maxima and total yearly precipitation over the study area for the period from 1975 to 2009. Furthermore, gamma, normal, and extreme value families of probability density functions (PDFs) were evaluated to fit the rain gauge gap-filled datasets. Finally, maximum annual precipitation values for return periods of 2, 5, 10, 25, 50, and 100 years were generated for each rain gauge. The results show that the geostatistical interpolation techniques outperformed the deterministic interpolation techniques in generating the spatial distribution of maximum and total yearly records over the study area. Full article

Accurate sub-seasonal-to-seasonal (S2S) forecasts are critical for mitigating extreme weather impacts and supporting development in West Africa. This study evaluates real-time ECMWF S2S rainfall forecasts during the 2020–2021 West African monsoon (March–October) and uses corresponding hindcasts for comparison. We verify forecasts at 1–4 dekads lead against two satellite-based rainfall datasets (TAMSAT and GPM-IMERG) to cover observational uncertainty. The analysis focuses on spatio-temporal monsoon patterns over the Gulf of Guinea (GoG) and Sahel (SAH). The results show that ECMWF-S2S captures key monsoon features. The forecast skill is generally higher over the Sahel than the GoG, and peaks during the main monsoon period (July–August). Notably, forecasts achieve approximately 80% synchronization with observed rainfall-anomaly timing, indicating that roughly 4 out of 5 dekads have correctly predicted wet/dry phases. Probabilistic evaluation shows strong reliability. The debiased ranked probability skill score (RPSS) is high across thresholds, whereas the average ROC AUC (~0.68) indicates moderate discrimination. However, forecasts tend to under-predict very low rains in the GoG and very high rains in the Sahel. Using multiple datasets and robust metrics helps mitigate observational uncertainty. These results, for the first real-time S2S pilot over West Africa, demonstrate that ECMWF rainfall forecasts are skillful and actionable (especially up to 2–3 dekads ahead), providing confidence for early-warning and planning systems in the region. Full article

The El Niño–Southern Oscillation (ENSO) is inherently asymmetric, a primary characteristic where its warm phase (El Niño) and cold phase (La Niña) differ in amplitude, spatial pattern, and temporal evolution. This review synthesizes over two decades of research to provide a comprehensive overview of ENSO asymmetry. It systematically examines the observed manifestations, evaluates the competing physical mechanisms, and analyzes the ongoing challenges in climate modeling. The key findings in the literature indicate that this asymmetry is driven by complex interactions of nonlinear processes, where atmospheric mechanisms such as state-dependent westerly wind bursts and threshold responses of deep convection are now considered dominant driving factors, which are subsequently amplified and modulated by oceanic feedback. The main challenge in this field is that most of the current state-of-the-art climate models underestimate ENSO asymmetry, which is related to mean-state bias and brings uncertainty to future predictions. Furthermore, a key finding from recent projection studies is that while the asymmetry in ENSO’s sea surface temperature is expected to weaken in a warmer climate, the asymmetry of its global rainfall impacts may paradoxically be amplified. Future research should focus on balanced improvements in ocean and atmospheric model components, development of new diagnostic tools to clarify the roles of different feedbacks, or establishment of a framework that clearly links asymmetry to the full spectrum of ENSO diversity. By consolidating the current state of knowledge and highlighting key unresolved questions, this work provides an essential roadmap to improve the prediction and projection of Earth’s most far-reaching mode of climate variability. Full article

Climate change is increasingly altering rainfall regimes and hydrological processes, posing major challenges to reservoir operation, flood control, and hydropower production. Understanding its impacts on the Ban Chat reservoir in Northwest Vietnam is therefore crucial for ensuring reliable water resource management under future uncertainties. This study aims to assess potential changes in streamflow to the Ban Chat reservoir under different climate change scenarios. The study employed nine Global Climate Models (GCMs) from the Coupled Model Intercomparison Project Phase 6 (CMIP6) under three Shared Socioeconomic Pathways (SSP1-2.6, SSP2-4.5, and SSP5-8.5). Future climate projections were bias-corrected using the Quantile Delta Mapping (QDM) method and used as input for the Hydrological Engineering Center–Hydrological Modeling System (HEC-HMS) to simulate future inflows. Streamflow changes were evaluated for near- (2021–2040), mid- (2041–2060), and late-century (2061–2080) periods relative to the baseline (1995–2014). Results show that under SSP1-2.6, mean annual discharge and flood-season flows steadily increase (up to +6.9% by 2061–2080), while storage deficits persist (−27.7% to −13.1%). Under SSP2-4.5, changes remain small, with flood peaks limited to +4.5% mid-century, but severe dry-season deficits continue (−29.5% to −24.4%). In contrast, SSP5-8.5 projects strong late-century increases in mean flows (+7.5%) and flood peaks (+8.2%), though early-century flood flows decline (−2.1%). These findings provide essential scientific evidence for adaptive reservoir operation, hydropower planning, and flood risk management, underscoring the significance of incorporating climate scenarios into sustainable water resource strategies in mountainous regions. Full article

Reliable rainfall prediction is essential for effective climate adaptation yet remains challenging due to complex atmospheric interactions that vary across regions. This study investigates next-day rainfall predictability in tropical and temperate climates using daily atmospheric data—including pressure, temperature, dew point, relative humidity, wind speed, and wind direction—collected from topographically similar sites in Alor Setar (tropical) and Vercelli, Williams, and Ashburton (temperate) between 2012 and 2015. Logistic regression and random forest models were used to predict rainfall occurrence as a binary outcome. Key variables were identified using Wald’s statistics and p-values in the logistic regression models, while the random forest models relied on mean decrease accuracy for ranking variable importance. The results reveal that rainfall in temperate climates is significantly more predictable than in tropical regions, with the Williams model demonstrating the highest accuracy. Atmospheric pressure consistently emerged as the dominant predictor in temperate regions but was not significant in the tropical model, reflecting the greater atmospheric variability and complexity in tropical rainfall mechanisms. Crucially, the study highlights that as global warming continues to alter temperate climate patterns—bringing increased variability and more convective rainfall—these regions may experience the same predictive uncertainties currently observed in tropical climates. These findings underscore the urgency of developing robust, climate-specific rainfall prediction models that account for changing atmospheric dynamics, with critical implications for weather forecasting, disaster preparedness, and climate resilience planning. Full article