Search Results (1,211)

Water flooding in low-permeability, highly heterogeneous reservoirs often causes a rapid increase in water cut and inefficient pressure maintenance because injected water preferentially flows through high-permeability channels. In this study, a field-scale compositional simulation model was established to evaluate CO₂ water-alternating-gas (WAG) enhanced oil recovery in a mature waterflooded reservoir in the Daqing Oilfield. The model was constrained by geological data, experimentally tuned pressure–volume–temperature (PVT) behavior, relative-permeability measurements, and slim-tube tests. The minimum miscibility pressure (MMP) of the CO₂-oil system was estimated to be 19.8 MPa. An 187-month production history was matched using field oil rate, water production, water cut, and reservoir-pressure data. At the current development stage, the reservoir has an oil recovery of 23.6%, an average water cut of 61.34%, and an average reservoir pressure of approximately 6.9 MPa. A 30-year prediction was then performed to compare continued water flooding with several CO₂-WAG development strategies. Sensitivity analyses were conducted for the pressure-restoration level, pre-injection fluid, well-pattern conversion, slug size, and gas/water slug-size ratio. Continued water flooding increased the final oil recovery to only 28.4% and resulted in a water cut of 92.8%. Sequential scenario screening identified a best-performing case among the tested scenarios, consisting of CO₂ pre-injection to restore the average reservoir pressure to 11 MPa, conversion to a staggered line-drive well pattern, a slug size of 0.025 PV, and a gas/water slug-size ratio of 1:1. Under this sequentially selected case, the end-of-forecast oil recovery reached approximately 57.24%, which was the highest value among the cases evaluated in this study and was 28.84 percentage points higher than continued water flooding. The predicted recovery is conditional on the adopted geological, relative-permeability, EOS, and history-matching assumptions. Because the designed average reservoir pressure is below the measured MMP and local pressure above the MMP was not demonstrated, the modeled process is consistently interpreted as immiscible CO₂-WAG. The predicted recovery improvement is interpreted as being associated with pressure support, gas-mobility control, improved sweep efficiency, and compositional CO₂–oil interactions represented by the model, including CO₂ dissolution, oil swelling, and viscosity reduction. The contribution of this work is a field-scale, experimentally constrained workflow for selecting CO₂-WAG operating parameters in mature waterflooded low-permeability reservoirs; CO₂ storage performance should be quantified separately in future work. This study provides an experimentally constrained and history-validated field-scale workflow for identifying a best-performing CO₂-WAG operating case among the tested scenarios in mature waterflooded low-permeability reservoirs. Full article

Disaster-prone urban areas must support transient and vulnerable populations, including foreign tourists who may face language barriers and difficulty interpreting conventional 2D hazard information. This study developed a smartphone-based augmented reality (AR) system designed to support flood-risk interpretation by foreign tourists in Japan. The system integrates four live external APIs providing location, weather forecast, hazard, and designated emergency evacuation site information, and retrieves these data on demand. Its core function converts hazard map-derived flood-inundation category data into 1:1-scale AR flood visualizations using simultaneous localization and mapping (SLAM) and inertial measurement unit (IMU) data. The system also provides a multilingual interface in English, Japanese, and Thai, weather forecast display, and nearby designated emergency evacuation site identification within a 1500 m radius. This study is positioned as a technical feasibility study. The evaluation comprised a verification of the AR height-placement mechanism against a physical reference at the five display heights used by the system, including the smallest category at 0.3 m (21.5 mm mean absolute alignment error, 1.0–2.3% of the target height); a quantitative SLAM drift evaluation under three movement patterns (at most 1.6 cm mean horizontal and 0.7 cm mean vertical drift after 180 s, within a priori thresholds); and an API response time evaluation under 5G and Wi-Fi (50 measurements per API per network; summed per-API means of approximately 3.99 s and 3.89 s at the endpoint level). The results support technical feasibility under the tested conditions, while effects on comprehension, preparedness, and evacuation decision-making remain for future user studies. Full article

Block 116 has an annual oil production of 10.2 × 10⁴ t, yet the increasing edge-water encroachment has raised the natural decline rate from 7.8% to 11.4%, significantly deteriorating production performance. The field data shows an edge-water rise rate of 0.4 m/a. Since 2023, 11 flank wells have experienced water breakthrough, resulting in a daily loss of 12.6 t and contributing 4.6% to the overall decline. To address these challenges, this study systematically evaluates development characteristics, identifies key influencing factors, and establishes optimal injection–production parameters through an integrated experimental and numerical approach. The research results show that a novel combined flooding formulation—water + foaming agent + polymer (viscoelastic) + N₂—is proposed and validated, achieving the highest recovery efficiency of 63.09%, which is 22.99% higher than conventional waterflooding. This confirms that switching to a multimedium injection strategy at mid- to high-water-cut stages can substantially improve ultimate recovery. Numerical simulation quantitatively delineates the remaining oil distribution across individual sublayers, revealing that residual recoverable reserves (76.6 × 10⁴ t) are predominantly concentrated in five specific sand bodies, providing precise targets for infill drilling. A comprehensive field-scale adjustment strategy is formulated—integrating optimized technical policies, planar and vertical injection–production improvements, and coordinated end-to-end management—which, over the 15-year forecast period, is projected to increase cumulative oil output from 149.4 × 10⁴ t to 168.1 × 10⁴ t, raising the recovery factor by 3.3%. These findings offer both theoretical insights and practical guidance for sidetracking and effective reserve utilization in analogous low-permeability reservoirs within the Changqing Oilfield and beyond. Full article

Accurate rainfall input is critical for reliable flood simulation, particularly in semi-arid watersheds with pronounced spatiotemporal precipitation heterogeneity. This study developed a radar–gauge rainfall fusion framework to improve HEC-HMS (Hydrologic Engineering Center–Hydrologic Modeling System) model performance in the Liulin experimental watershed, Xingtai City, Hebei Province. Radar quantitative precipitation estimation (QPE) was generated via a dynamically optimized Z-I relationship, then fused with gauge observations using three methods—Geographical Differential Analysis (GDA), Conditional Merging (CM), and Random Forest (RF). The fused products drove a calibrated HEC-HMS model, evaluated over five representative flood events. All three methods corrected radar QPE underestimation. Under independent cross-validation, GDA and CM achieved comparable point-scale accuracy (CC ≈ 0.81, RMSE ≈ 5.7 mm), while RF showed lower generalization (CC ≈ 0.48, RMSE ≈ 8.7 mm) due to overfitting. In flood simulations, GDA performed most robustly, followed by RF and CM, all surpassing single-source inputs. Notably, CM’s higher statistical accuracy did not translate into better flood performance, indicating that optimal statistical fidelity does not guarantee optimal hydrological results. Peak discharge deviations persisted for short-duration intense storms and long-duration uneven rainfall events. This study confirms that radar–gauge fusion enhances rainfall input quality and provides a reliable approach for improving flood forecasting. Full article

High-resolution precipitation fields are essential for flash-flood forecasting and hydrological risk management, especially in small and medium-sized basins, yet single-source precipitation products often show limited accuracy over complex terrain. This study develops a two-stage machine-learning framework for 1 km/1 h multi-source precipitation fusion over Shaoxing, China, during the 2025 flood season. In the first stage, a machine-learning classifier identifies precipitation occurrence and reduces zero-inflated noise; in the second stage, an optimized tree-based residual-regression model corrects precipitation estimates for rainy samples. A 61-dimensional feature set was constructed by integrating satellite precipitation estimates, weather-radar precipitation estimates from the Zhejiang radar network, temporal-lag and accumulation statistics, neighborhood descriptors, cyclic time variables, and terrain-derived interaction features, with gauge observations used as the training target. After quality control, the dataset comprised 41,458 hourly station samples from 72 rain gauges. The stations were divided at the station level into a 57-station development set and a fixed 15-station held-out spatial test set containing 8637 hourly samples. Station-blocked fivefold cross-validation within the development set was used for model selection, hyperparameter tuning, and probability-threshold selection, whereas the held-out stations were used only for final performance evaluation. On the fixed held-out test set, the occurrence classifier achieved an overall accuracy of 0.947, with a probability of detection of 0.806, a false alarm ratio of 0.158, a critical success index of 0.700, and an F1 score of 0.823. For quantitative estimation, the two-stage fusion product reduced root mean square error from 2.342 mm for satellite precipitation estimates to 1.189 mm, corresponding to a 49.22% reduction, and decreased mean absolute error from 0.712 mm to 0.262 mm, while increasing the coefficient of determination to 0.685. The fused precipitation product also improved the detection of intense rainfall events, with probability of detection and critical success index reaching 0.511 and 0.442, respectively, for events exceeding 10 mm/h, while reducing false weak precipitation and showing closer agreement with observed station-level spatial variability. By separating precipitation-occurrence identification from rainfall-intensity correction, the framework reduces zero-inflated bias, improves heavy-rainfall representation, and demonstrates predictive skill at gauges excluded from model development during the 2025 flood season. Full article

The escalating frequency of extreme hydrological events under environmental uncertainty poses a severe socio-economic threat to floodplains such as the Brazilian Pantanal, the world’s largest tropical wetland. Mitigating dynamic flooding and drying cycles is highly challenging due to a critical scarcity of in situ monitoring, leaving flood risks poorly understood. To address these data gaps, this study presents an advanced deep learning forecasting framework that integrates multisource environmental data, fusing satellite-derived precipitation (CHIRPS) and global land data assimilation evapotranspiration (GLDAS) data with historical river gauge telemetry. Multi-layered neural network architectures were optimized and combined with progressive moving average filters (10− and 15−day windows) to capture the complex hydrometeorological patterns of the data-scarce Miranda River Watershed. The optimal deep learning configuration, utilizing a robust two-hidden-layer topology (15 and 60 neurons), consistently outperformed standard baselines. Although purely exogenous data blocks successfully minimized satellite noise and captured seasonal trends (NSE ≥ 0.92), structural underestimation of peak flows was observed. When incorporating the previous day’s streamflow (lag t−1) as a physical anchor, this limitation was noticeably alleviated, increasing both the Nash–Sutcliffe Efficiency (NSE) and Coefficient of Determination (R²) values above 0.99. While this performance surge is driven by the strong temporal persistence inherent to the autoregressive lag, it introduces an operational trade-off by restricting the forecast to a reactive 24 h window. In this regard, an evaluation of the operational forecast horizons revealed that the exogenous deep learning blocks maximize warning lead times, providing a vital tool for proactive civil defense and disaster risk reduction. Ultimately, this multisource framework establishes a methodological foundation for automated decision support systems, providing the high-accuracy streamflow forecasting capability required to support future flood mitigation frameworks. Full article

This study aims to deepen the understanding of the spatiotemporal evolution of the Extreme Precipitation Index in the YRD, evaluating the comprehensive flood disaster risk across the region. Existing studies have rarely incorporated the frequency, duration, and intensity of extreme precipitation events as indicators of the hazard of causative factors into comprehensive flood risk assessments. Using daily precipitation records from 107 national meteorological stations spanning 1960–2024, this study employs four extreme precipitation indices recommended by the ETCCDI (PREPTOT, CWD, R95p, and Rx1day) to examine the spatiotemporal characteristics of extreme precipitation in the YRD, one of China’s most densely populated and economically significant regions. Furthermore, a comprehensive flood risk assessment framework encompassing the hazard of causative factors, the sensitivity of disaster-forming environments, the vulnerability of disaster-affected entities, and disaster prevention capabilities is constructed. Based on this framework, an integrated Analytic Hierarchy Process–Entropy Weight method is adopted to evaluate Flood Risks in the YRD. The results showed that: (1) during the period 1960—2024, only the Extreme Precipitation Index (R95p) exhibited a significant upward trend, increasing at a rate of 1.7 mm/10a, whereas PREPTOT, CWD, and Rx1day showed slight declining trends. Nevertheless, all four indices displayed pronounced oscillatory characteristics, characterized by recurring “decrease—increase” cycles over time. (2) In terms of spatial distribution, PREPTOT and R95p exhibited a clear south-to-north gradient pattern, while CWD and Rx1day demonstrated a multicentric distribution. This pattern highlights the combined influence of typhoon landfall frequency and topographic conditions on extreme precipitation across the southern YRD. (3) The Flood Risks in the YRD exhibited a distinct spatial pattern characterized by higher risk levels in the east than in the west and in the south than in the north. Areas classified as moderate-to-high risk accounted for 48.6% of the total study area, with high-risk zones primarily concentrated in the Shanghai—Hangzhou—Ningbo corridor. These findings suggest that Flood Risks in the YRD are not driven by a single factor; rather, they result from the complex interactions among extreme precipitation, topographic and geomorphological conditions, levels of social exposure, and regional buffering capacities. Consequently, under the increasingly frequent occurrence of extreme precipitation events, flood management strategies that rely predominantly on static engineering measures are becoming insufficient. Greater emphasis should therefore be placed on enhancing the resilience of urban lifeline infrastructure, improving high-resolution forecasting and early-warning capabilities for extreme precipitation, and establishing dynamic warning-release mechanisms based on risk thresholds. Such measures are essential for effectively mitigating regional flood risks. Full article

Deep learning surrogate models trained with mean-squared-error loss produce statistically accurate but physically unconstrained flood predictions: water may flow uphill, appear spontaneously, or smooth over street-level corridors. In this work, a physics-informed training framework is developed for CNN-LSTM models that predict urban flood depths at 15 min intervals over a $128\times 128$ spatial grid. Three differentiable penalty terms are embedded directly into the loss function: (i) a gravity loss that penalizes depth increases against the water-surface-elevation gradient, (ii) a continuity loss enforcing local mass conservation with rainfall-adaptive thresholds, and (iii) a topography-aware false-alarm penalty modulated by the topographic wetness index (TWI). The framework is evaluated on the Norfolk, Virginia, flood dataset spanning two major storm events (August 2017 and September 2022) comprising 300 samples, with all variants trained on identical splits and robustness assessed over repeated random splits and leave-one-storm-out tests. A road-proximal evaluation restricted to a TWI-derived street mask quantifies street-level skill. The physics-constrained model achieves near-zero gravity violations (∼10⁻⁶) and the highest street-channel recall (0.77 ± 0.09 versus 0.44 ± 0.10 for the unconstrained baseline), the capability most relevant to downstream traffic routing, and its recall advantage more than doubles on a held-out storm, while a uniform false-alarm variant attains 16% lower mean absolute error but suppresses street recall to 0.25. The proposed TWI-modulated penalty reconciles this trade-off: it improves upon the uniform variant on every metric measured, recovering 60% higher street recall at the lowest MAE among all constrained variants and the best street-level F1 score. These results expose a fundamental tension between aggregate pixel-level error metrics and application-specific physical plausibility, and demonstrate that terrain-aware loss modulation offers a principled resolution. Full article

Accurate runoff prediction is an essential foundation for water resource management, flood prevention, and drought warning. Despite the superior performance of deep learning models in runoff prediction, the high-dimensional hyperparameter optimization limits their widespread application. To address this challenge, this study proposed a hyperparameter optimization strategy that integrated Uniform Design (UD) and Meta-Learning (ML) within the Transformer framework (UD-ML-Transformer) for daily runoff prediction. Performance of the proposed model was systematically evaluated against five benchmark models, including the UD-Transformer, Particle Swarm Optimization (PSO)-Transformer, and three Receptance Weighted Key Value (RWKV)-based models (PSO-RWKV, UD-RWKV, and UD-ML-RWKV), using hydroclimatic data spanning 1980 to 2014 from the Rio Pueblo de Taos watershed in USA. Results showed that the UD-ML-Transformer model performed the best in both prediction accuracy and peak flow, with the highest Nash-Sutcliffe Efficiency (NSE) of 0.906, and the lowest Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) of 0.004, 0.062, and 0.034, respectively. The UD-Transformer ranked second in performance, followed by the PSO-Transformer. The integrated UD-ML hyperparameter optimization strategy also improved the performance of RWKV-based models. Compared with the PSO-RWKV and UD-RWKV models, the UD-ML-RWKV model exhibited an NSE improvement of 0.45–7.65% and an RMSE reduction of 1.47–21.18%, respectively. Moreover, cross-watershed validation conducted in the Ford River watershed, USA, also demonstrated the satisfactory performance of the proposed UD-ML-Transformer model, with the highest NSE of 0.890, and the lowest MSE, RMSE, and MAE of 0.088, 0.296, and 0.141, respectively. These findings highlight the superiority of integrating UD and ML for hyperparameter optimization in runoff forecasting. Full article

Flooding is a destructive natural hazard intensified by climate change, posing challenges to sustainable disaster risk management. This study developed and evaluated machine learning models for flood severity prediction within a hexagonal grid system (H3, resolution 8) under rainy season conditions in Maha Sarakham, Thailand. Four models Random Forest (RF), XGBoost, Gradient Boosting (GB), and Support Vector Machine (SVM) were trained using 11 environmental variables across historical years (2018, 2021, 2022) and tested on a projected year (2025) under SSP scenarios. XGBoost demonstrated the most stable performance (accuracy > 0.95 across all years), while SVM achieved high historical accuracy (0.970 average) but failed to detect positive flood cases in 2025 (recall = 0), highlighting the importance of temporal validation. Topographic variables were the most consistent predictors, but NSMI (soil moisture) emerged as the top SHAP predictor in 2025 (r = 0.52), suggesting a shift in flood-generating mechanisms under climate change. A polarization pattern was observed: flood-affected area declined to 6.8% in 2025 (79% reduction from 2022), yet maximum flood point counts remained high at 14.0, indicating more concentrated but intense flooding. Under SSP projections, using the historical baseline (27.5%), SSP1-2.6 (45.2%) and SSP2-4.5 (45.0%) indicate increased flood risk relative to the historical baseline through 2040. The SSP5-8.5 projection (3.4%) is identified as a model extrapolation artifact through formal out-of-distribution assessment (Mahalanobis distance = 8.72, p < 0.001) and is therefore excluded from policy recommendations. Although GRU and LSTM achieved marginally higher AUC values in retrospective validation, we recommend XGBoost for operational forecasting due to its temporal stability, computational efficiency, and interpretability. We further recommend integrating real-time soil moisture monitoring into early warning systems and shifting to hotspot-targeted adaptation strategies. Full article

Reliable daily runoff forecasting supports flood risk mitigation and sustainable basin water management, but the added value of external information is difficult to identify when antecedent runoff strongly constrains prediction. This study develops a diagnostic framework to examine whether multi-source remote sensing and gridded precipitation provide additional value beyond runoff memory. The framework was applied to 1-, 3-, 5-, and 7-day runoff forecasting in the Beidao and Baijiachuan catchments of the Yellow River Basin. An external-forcing reference model (M1) used meteorological–remote sensing variables, precipitation statistics, and static catchment attributes, whereas a runoff memory reference model (M2) used only antecedent runoff features. Their validation-based combination formed a fusion benchmark. A residual correction model based on an SRCNN–Transformer architecture (M3) was then used to examine whether the remaining fusion errors could be corrected using gridded precipitation fields and multi-source temporal states. Results show that runoff memory dominated short-lead forecasts but weakened with lead time, while external forcing became more useful at medium and longer leads. M3 produced positive but lead time-dependent Nash–Sutcliffe efficiency gains, with the most stable improvements at 3–5 days. These results describe the potential value of remote sensing and gridded precipitation under known forcing conditions rather than operational forecast skill. These findings provide evidence from two contrasting Yellow River catchments, while their broader applicability remains to be tested under more diverse hydrological and operational forecasting conditions. Full article

Accurate streamflow prediction is an important part of disaster management, as forecasting flow rates is an indispensable element of early warning systems, yet it remains challenging due to the complexity and nonlinearity of climatic inputs. This study evaluated the performances of Long Short-Term Memory (LSTM) and random forest (RF) machine learning algorithms for daily streamflow prediction across sixteen hydroclimatically diverse basins in the contiguous United States using the CAMELS dataset. The models were trained on five climatic features augmented with antecedent streamflow lag features at 7-, 14-, and 30-day intervals. The random forest algorithm demonstrated a better accuracy, achieving an average Nash–Sutcliffe efficiency (NSE) of 0.755 compared to 0.632 for the LSTM model. Both models performed well in the snowmelt-dominated basins and were least effective in flashy humid regimes. Additionally, both models exhibited high precision for flood detection, with accuracy rates exceeding 88% for distinguishing flood events and F1 scores of 0.734 and 0.797 for LSTM and RF, respectively. These results recommend RF for operational streamflow forecasting across hydroclimatically diverse settings and LSTM for perennial snowmelt- and groundwater-influenced catchments where long-range temporal dependencies govern runoff generation. Full article

Accurate streamflow forecasting is critical for water resource management and disaster mitigation, yet conceptual models often suffer from systematic biases while machine learning lacks physical interpretability. This study proposes a coupled XAJ-LSTM model to integrate the mechanistic strengths of the Xin’anjiang (XAJ) model with the predictive capability of Long Short-Term Memory (LSTM) networks. Applied in the Woluo River basin using an 11-year daily record from 2012 to 2022, the framework was evaluated with leave-one-year-out validation. Results indicate that XAJ-LSTM achieved the highest full-period NSE of 0.885 and the lowest full-period RMSE of 19.48 m³/s, compared with NSE values of 0.867 for XAJ and 0.724 for LSTM. XAJ-LSTM also performed best during the flood period, with NSE of 0.820 and RMSE of 28.81 m³/s. The SHAP framework indicated that XAJ-simulated flow and precipitation remained the dominant predictors, suggesting that the LSTM relied strongly on the conceptual-model baseline and rainfall information. These findings indicate that XAJ-LSTM can improve selected full-period and flood-period diagnostics for daily streamflow simulation in the Woluo River Basin, while further tests with longer records and additional catchments are needed before broader application. Full article

The floods triggered by the 29 October 2024 cut-off low in Valencia (Spain) expose a persistent challenge in disaster risk reduction: extensive meteorological and territorial data do not automatically become timely, trusted or actionable public guidance. This conceptual synthesis uses the Valencia event as a diagnostic case and reconstructs selected evidence on rainfall, hydrological escalation and alert timing to develop a data-driven framework spanning observation, modelling, impact assessment, communication, decision-making and post-event learning. Here, “data-driven” denotes an end-to-end governance and translation process, not the development of a new forecasting model. The framework integrates four dimensions: data governance, user-centred visualization, uncertainty communication and school-based education. It also introduces territorial translation as the link between impact forecasts and place-specific infrastructures, routines, vulnerabilities and responsibilities. Its novelty lies in connecting the Early Warnings for All pillars and impact-based, people-centred warning approaches with an explicit educational and territorial learning loop. Its practical contribution is a responsibility matrix, a minimum governance package and an implementation roadmap with indicators for latency, reach, comprehension and protective action. The framework is intended for adaptation, rather than statistical generalization, across Mediterranean and other fast-onset flood contexts. Improved forecasts remain necessary but insufficient: loss reduction requires interoperable records, accessible impact-based messages and inclusive educational programmes that convert scientific information into situated collective competence. Full article

The paper addresses the intersection between modern built heritage preservation and Climate-Aided Design (CADe) processes in fragile coastal Mediterranean contexts. The study focuses on the city of Beirut in Lebanon, part of the Eastern Mediterranean and Middle East (EMME) region and considered a climate change hotspot facing extreme challenges. Rapid urbanization and socio-political instability, especially during the twentieth century, have undermined the city’s ability to mitigate and adapt to future climate change scenarios. Moreover, Beirut’s modern built heritage faces a constant threat of demolition due to the absence of protective legislation, compounded by aggressive real-estate development ambitions. The hypothesis is that the integration of climatic data and regenerative design with modern cultural heritage classification frameworks can aid the preservation process, drive a more adaptive and inclusive approach to urban regeneration, and inform legislative integration of climate adaptation in conservation frameworks. To test this hypothesis, a multi-scalar case-study-based methodology is adopted using a combination of digital tools to assess and analyze the current and future impacts of climate change on two main case studies. First, the St. George Hotel & Bay, one of the first reinforced concrete recreational buildings in the city, was built during the French Mandate (1920–1946) and is vulnerable to sea-level rise, flooding, and demolition. Second, the Beirut City Center “The Egg”, a Brutalist structure built during Beirut’s modernist “golden era”, which is prone to structural deterioration and demolition. The main objective is to highlight 20th-century built heritage as part of Beirut’s spatial narrative worthy of conservation and rehabilitation by analyzing their capability to adapt to, mitigate, or benefit from future environmental risk. Ultimately, the study explores their potential to catalyze climate-resilient urban regeneration practices in the city. Results show that the integration of current and future forecast environmental analyses informed early preservation and intervention decision-making stages to position 20th-century modern built heritage as an asset to climate action in addition to being a socio-cultural and economic asset. Full article