Search Results (84)

Accurate water level observations are essential for flood forecasting, hydrological analysis, and water resource management; however, sensor malfunctions and telemetry errors frequently introduce anomalous observations that compromise data quality. This study proposes a reconstruction-oriented Bidirectional Long Short-Term Memory (BiLSTM) Autoencoder for river water level anomaly correction and compares its performance with conventional first-, second-, and third-order polynomial and exponential regression models. The proposed framework incorporates a simplified encoder–decoder architecture, a dynamic block masking strategy to emulate contiguous sensor failures in highly autocorrelated water level series, and a threshold-based peak-oriented training scheme to improve reconstruction during high-flow events. Model hyperparameters were optimized using Gaussian process-based Bayesian optimization. The methodology was evaluated using hourly observed water level data from the Han River, Republic of Korea. Results showed that the proposed BiLSTM Autoencoder achieved reconstruction accuracy comparable to conventional regression models during calibration while exhibiting superior generalization to unseen validation datasets and better preserving the temporal continuity and dynamic characteristics of downstream hydrographs. Furthermore, a model calibrated using a relatively short but hydrologically representative period successfully reconstructed a substantially longer unseen record. Synthetic outlier injection experiments further demonstrated that reconstruction accuracy gradually deteriorated with increasing training data contamination, emphasizing the importance of high-quality training data for reliable sequence reconstruction. The proposed framework demonstrates potential as an effective sequence-reconstruction approach for offline river water level quality control. Full article

Floods degrade road networks at the same time as demand for emergency medical services (EMSs) rises, yet national EMS command systems rarely receive any information on flood-induced road barriers. This paper presents Ambulance STARS (SaTellite-assisted Ambulance Routing System), a service-oriented framework that links satellite observation with ambulance dispatch. A cloud-based flood detection service derives flood extent from Sentinel-1 SAR amplitude change detection executed in a cloud-based Earth observation data and compute backend and translates it into road passability layers. A routing engine then maintains an in-memory road graph whose travel times are calibrated with empirical ambulance speed models built from four years (2020–2023) of GPS records of an EMS fleet in southern Poland, with separate speeds for driving with and without emergency signals (61.8 and 37.2 km/h, respectively). An API gateway with single-file tile delivery, a replicated relational data tier, and an observability stack complete the architecture, and a web client offers dispatchers live routing and multi-unit incident simulation. The framework was tested on the September 2024 flood in the Municipality of Nysa, Poland. The SAR module delineated 665 ha of inundation and marked 8.5 km of the 656.7 km routing network as impassable (508 barrier points), and the same procedure applied to a reference optical mask of 18 September yielded 17.3 km and 1006 points. Because the SAR and optical acquisitions captured different phases of the flood wave, agreement on the rare impassable-road class was low, and the two products were, therefore, used to bracket operational uncertainty rather than to define a single ground truth. Applied without local retuning to Lewin Brzeski, the same flood detection workflow showed consistent performance against the CEMS reference product. The routing module produced statutory 8/15/20 min accessibility maps in 12–34 s under warm-cache benchmark conditions. With SAR-derived barriers, the share of the network reachable within 15 min fell from 88% to 80%, and 2 villages with 938 inhabitants lost road access to EMS entirely. With barriers derived from the optical mask, the 15 min share fell to 39.8% and seventeen settlements lost road access entirely, underlining how strongly the barrier source shapes the operational picture. Post-acquisition processing completes in under one minute under warm-cache conditions with road data preloaded, and satellite-derived road passability is fast enough to support near-real-time decision-making, subject to the constellation revisit time and to integration with EMS command systems. Full article

Compound coastal hazards such as flooding, land subsidence, storm surge, and salinity intrusion impose accelerating risks on deltaic communities. This study presents a unified multi-hazard susceptibility mapping framework for Terrebonne Parish, Louisiana, modeling all four hazards from a common 30 m predictor stack, with per-hazard exclusion of label-related predictors. Eight Machine Learning and Deep Learning algorithms were benchmarked per hazard against an ensemble meta-learner. Generalizability was assessed under three designs of increasing spatial rigor: blocked holdout, interleaved block cross-validation, and a strict contiguous-zone design with a 5 km buffer. Best holdout F1-macro ranged from 0.644 (salinity) to 0.923 (flood). Interleaved-block cross-validation was statistically indistinguishable from holdout; only the buffered contiguous-zone design revealed genuine transfer limits, with F1-macro declining 12–54 percentage points by hazard. Ensemble stacking did not improve upon cross-validation-guided single-model selection despite roughly five times the training cost. Salinity labels were derived from 21 kriged monitoring stations (RMSE = 3.40 PSU; R² = 0.82). A composite Multi-Hazard Susceptibility Index (mean = 0.675 parish-wide; 0.674 land-masked) identifies southern coastal Terrebonne as the priority zone for risk reduction, robust to reweighting of any single hazard. To our knowledge, this is the first framework to jointly map these four hazards while quantifying how validation design governs apparent model transferability. Full article

Crowdsourced street-level images offer a practical source of street-scale evidence for urban flood monitoring, yet floodwater-level recognition remains difficult because such images are captured from heterogeneous viewpoints and are often affected by occlusion, reflections, nighttime glare, and cluttered backgrounds. This study proposes a Mask-guided Enhanced Object Detection (MEOD) pipeline for vehicle-based floodwater-level recognition, which first uses floodwater segmentation to suppress unstable water-surface appearance and generate water-masked inputs, and then adapts YOLO11-based object detection to focus on vehicle-submersion cues to obtain floodwater levels. We compiled a four-level vehicle-centered dataset comprising 3110 crowdsourced street-level flood images, and evaluated the effectiveness of MEOD with four controlled settings. Results showed that MEOD achieved an F1-score of 0.89 and an mAP@0.5 of 0.93, compared with 0.77 and 0.84 for the original object-detection baseline, respectively. The class-wise correct recognition rates indicated that MEOD reduced interference-driven ambiguity between neighboring floodwater levels under the evaluated dataset and experimental protocol. These results indicate that floodwater segmentation can serve as a task-oriented intermediate representation for detector-based floodwater-level recognition in vehicle-centered street-level flood images. 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

Flood-illuminated adaptive optics (AO) imaging enables high-resolution quantification of retinal vascular morphology; however, its repeatability in older adults with lens opacities remains poorly characterized. In this prospective study, 40 subjects (66 eyes) underwent flood AO imaging using the rtx1 system (Imagine Eyes, Orsay, France) at three retinal locations across multiple sessions. Two independent masked graders measured lumen diameter (LD), total diameter (TD), wall cross-sectional area (WCSA), and wall-to-lumen ratio (WLR), with repeatability assessed using intraclass correlation coefficients (ICC). Inter-grader agreement was excellent for LD and TD across all regions (ICC range 0.974–0.994) and for WCSA at the superior and inferior regions (ICC range 0.902–0.952). Intrasession and intersession repeatability were similarly excellent for LD and TD across all locations. Repeatability remained excellent across age groups and cataract grades. Image region was the only variable significantly associated with image quality (p < 0.0001), with temporal images showing the lowest proportion of good quality ratings (12%). These findings support flood AO imaging as a reliable tool for quantitative vascular assessment in age-related and vascular-related ocular diseases. Full article

Recurrent and increasingly severe flooding along the Wabi Shabelle River—displacing approximately 184,000 people from Beledweyne, central Somalia, in the 2020 Gu season alone—has made in situ reconstruction untenable and planned resettlement a central adaptation option. Site selection in this agropastoral context must simultaneously avoid the riparian flood corridor and preserve access to the river as the dominant livelihood resource. We develop a transparent, reproducible GIS-based Analytic Hierarchy Process–Weighted Linear Combination (AHP–WLC) framework over a 30 × 30 km region of interest at 30 m resolution. A hard safety mask (Height Above Nearest Drainage > 5 m and slope < 5°) is combined with six normalised criteria, including a non-monotonic, piecewise river-livelihood score, using literature-anchored AHP weights (consistency ratio CR = 0.004). Seven initial criteria were pre-screened with Pearson, Spearman and Variance Inflation Factor diagnostics (maximum |r| = 0.52, maximum VIF = 1.44), removing a perfectly collinear services-distance layer before weighting. Robustness was confirmed by a ±10% one-at-a-time sensitivity analysis with targeted river-weight and HAND-threshold tests, all criteria remaining very robust (|Δ| < 5%). The composite identifies 12,723 ha (14.3%) as highly suitable, resolving into 113 operationally meaningful candidate sites (≥5 ha; 95.6% of the highly suitable area). Candidate areas sufficient to absorb the 2020 displacement land demand (552–1104 ha) lie within 6 km of the city centre. The framework offers an operational, defensible foundation for resettlement planning in flood-exposed agropastoral cities of the Horn of Africa. Full article

Machine-learning landslide-susceptibility maps are almost always judged by inventory-split skill (the area under the receiver-operating-characteristic curve, AUC, and Cohen’s κ), not by model-independent physical observation of where an event caused ground disturbance. For the May 2023 Emilia-Romagna event (>80,000 landslides; RER2023 inventory), we confront an open-data, event-conditioned susceptibility model (trigger rainfall is among its predictors) with a co-event disturbance footprint built from two satellites: phenology-matched Sentinel-2 change in the Normalized Difference Vegetation Index (ΔNDVI) and Normalized Burn Ratio (ΔNBR), and a 12-day Sentinel-1A C-band coherence-and-backscatter layer used as a cloud-independent coverage check (C-band 12-day decorrelation is the a priori expectation in this setting); neither enters the model. Apparent geomorphic plausibility depends critically on how the independent footprint is constructed. Against a raw co-event footprint (contaminated by flooding and agriculture), a model with AUC ≈ 0.945 is indistinguishable from a random mask. On a landslide-relevant footprint, the same model captures optically detectable disturbance at roughly twice the chance (bootstrap median 1.94×, 95% CI 1.25–2.72, excluding 1; ≈1.46× above a vegetation-phenology null), but the aggregate is driven by an upper-tail minority of 10 km blocks—the majority of blocks have per-block median lift 0.67× (below chance). The map is therefore a regional, not a per-place, statement. Results are a within-event consistency test; cross-event transferability is not claimed. Footprint construction is decisive and currently neglected. The open-source pipeline is released upon acceptance. Full article

Satellite-based precipitation products such as the Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (IMERG, V07) provide global coverage but exhibit systematic biases in daily accumulations, particularly for extreme events. This study presents a hybrid bias-correction framework (LSEQM+DL) for daily satellite precipitation that sequentially integrates Linear Scaling (LS) for mean bias, Empirical Quantile Mapping (EQM) with a Generalized Pareto Distribution (GPD) tail adjustment for distributional alignment, and a Convolutional Neural Network (CNN) refinement that targets extreme-precipitation pixels. A station-density confidence mask scales the deep-learning influence with gauge density, so the CNN refinement is strongest where the reference, the CPC Unified Gauge-Based Analysis of Daily Precipitation (CPC-UNI), is best constrained. The framework targets the IMERG Late Run (IMERG-L), whose roughly 14 h latency suits near-real-time flood monitoring. It is applied over Indonesia (2001–2025) and evaluated against CPC-UNI and 171 independent stations of the Meteorological, Climatological, and Geophysical Agency (BMKG) through three pillars: adjusting values, aligning distributions, and preserving extremes. At independent stations, the correction brings the standard deviation ratio from 0.71 (LS) to 1.00, the relative bias from $-$11.4% to $-$0.6%, and the 99th-percentile ratio from 0.71 to 1.01, and reduces a 21% over-estimation of wet-day frequency to within 5% of that observed. These gains carry a designed cost: the probability of detection falls from 0.78 to 0.65, while pixel-level temporal metrics (correlation, root-mean-square error, Nash–Sutcliffe efficiency) remain largely unchanged, confirming that the framework improves statistical properties rather than day-to-day timing. Relying only on globally available satellite and gauge-analysis data, and degrading gracefully where gauges are sparse, the framework is portable in principle with regional recalibration of its three tuning parameters. The corrected near-real-time product, with its station-density mask as a spatially explicit quality indicator, is intended to support flood monitoring, water resource management, and climate risk assessment in Indonesia and other gauge-sparse tropical regions. Full article

Floodplain wetlands are globally important ecosystems, yet altered hydrological regimes increasingly disrupt the balance between woody and non-woody vegetation. In Australia’s regulated Murray–Darling Basin, it remains unclear whether woody plant encroachment represents a persistent shift toward terrestrialisation or a dynamic process that can be periodically reversed by flooding. This study quantified long-term patterns of woody-vegetation encroachment and retreat across 32,000 ha of mapped wetlands in the mid-Murrumbidgee River floodplain from 1988 to 2023, and assessed how hydrological variability and floodplain connectivity mediate these dynamics. Using open, analysis-ready Earth observation data from Digital Earth Australia (DEA) within the Open Data Cube (ODC) framework, we combined DEA Land Cover for transition mapping, Water Observations for hydrological masking, Landsat surface reflectance for Enhanced Vegetation Index (EVI)-based spectral plausibility testing, and the Wetlands Insight Tool for qualitative temporal context. Woody-vegetation dynamics were strongly non-linear and closely linked to alternating drought and flood phases. During the Millennium Drought (2001–2009), mapped woody-cover decline exceeded 50% of wetland area in some sub-regions, whereas the post-drought recovery interval (2008–2013) produced encroachment exceeding 40% in the most affected areas. Across the full 35-year record, mean encroachment rates ranged from 85 to 250 ha yr⁻¹ among sub-regions, summing to approximately 865 ha yr⁻¹ of woody expansion across the floodplain, while retreat rates were lower overall (approximately 634 ha yr⁻¹), resulting in a net expansion of woody cover. Local hydrological connectivity strongly mediated these responses: infrequently inundated wetlands showed persistent terrestrialisation, whereas more frequently inundated, better-connected wetlands experienced periodic flood-driven retreat. Landsat-derived EVI broadly supported the mapped transitions, indicating general consistency with canopy greening and canopy decline, supporting the ecological plausibility of the detected changes. This open DEA–ODC workflow provides a transparent, transferable framework for operational wetland monitoring and demonstrates that maintaining natural flood frequency, duration, and connectivity is essential for sustaining the resilience of regulated floodplain systems. Full article

Rapid and reliable flood segmentation utilizing optical remote-sensing imagery is critical for effective flood disaster response and risk assessment. Nevertheless, current models frequently struggle with imprecise boundary delineation and fragmented predictions in complex environments, especially where floodwater displays high spectral variability and closely resembles shadows, dark pavements, or wet soil. To overcome these challenges, we introduce FloodSeg, an innovative Mamba-CNN encoder–decoder network incorporating two lightweight yet highly effective components: a Shift module and a sequence-shuffle module. The spatial Shift module leverages spatially shifted feature aggregation to fortify boundary-aware representations, thereby ensuring the continuity of inundation contours even under varying illumination and cluttered backgrounds. Meanwhile, the sequence-shuffle module reorganizes multi-scale features via sequence-wise mixing and cross-regional interaction, significantly enhancing long-range dependency modeling. This facilitates the generation of globally consistent flood masks while mitigating local overfitting to dataset-specific textures. Evaluated on the Kaggle and FloodNet benchmark datasets, FloodSeg achieves outstanding mIoU scores of 81.85% and 91.21%, respectively. By outperforming various state-of-the-art CNN-, Transformer-, and Mamba-based baselines, our model demonstrates a superior accuracy-efficiency trade-off. These results substantiate that FloodSeg significantly advances boundary recognition and overall segmentation completeness, establishing it as a robust and practical solution for real-world remote-sensing flood mapping applications. Full article

The Inner Mongolia reach of the Yellow River faces compound “low flow, high sediment” extremes under reservoir regulation, threatening flood and ice-flood safety in ways that traditional mean-based or correlation-based methods fail to quantify. This study integrates POT-GPD extreme value theory with a vine copula-CoVaR framework using daily data (1951–2023) from four stations. The financial CoVaR concept was adapted to rivers through three hydrological modifications: a 5-day hydrodynamic lag, redefinition of the baseline to the downstream unconditional VaR, and semi-parametric tail modeling. Bootstrap confidence intervals (n = 1000) and a sensitivity analysis to the upstream–downstream lag (τ = 3–7 days) and the period cutoff (1984–1990) were used to assess robustness. Bayangol exhibits the highest Expected Shortfall (ES₉₅ = 0.0329 kg·s·m⁻⁶). The Bayangol → Toudaoguai path is the only persistent positive risk transmission link, with ΔCoVaR showing a directionally consistent increase of 253% from the natural period (1951–1986) to the regulated period (1987–2023); by contrast, ΔCoVaR from Dengkou to Toudaoguai remains near zero or negative when assessed under the conventional bivariate framework. A three-dimensional vine copula analysis, conducted independently for the pre- and post-reservoir periods, reveals a qualitative reversal of compound extreme spillover that is masked when the two periods are pooled. While the bivariate analysis identifies Bayangol → Toudaoguai as the only persistent positive spillover route at the annual scale, the 3D vine analysis unpacks the compound extreme mechanism at the daily scale. Under the joint compound extreme condition (upstream Q and S each ≥ Q₉₀), the conditional VaR₉₅ of downstream sediment concentration shifts from systematically negative in P1 (ΔVaR₉₅ = −4.75 kg·m⁻³ at the 90th-percentile threshold, indicating natural attenuation) to systematically positive in P2′ (ΔVaR₉₅ = +4.70 kg·m⁻³, +86.9% relative increase, indicating amplification). The same reversal is observed for the tail mean (ΔES₉₅), is preserved across four compound extreme thresholds (Q₇₅–Q₉₀), and is robust to the choice of period cutoff (28/28 cases reverse across seven candidate cutoffs). Bidirectional counterfactual simulations indicate that the copula shift from tail independence (Clayton) to tail dependence (Gaussian) alone elevates extreme concurrence probability by 58% (from 2.21% to 3.49%), while marginal distribution changes contribute negligibly (≤0.1 percentage points). Structural deterioration of water–sediment coordination therefore dominates risk amplification. The copula-CoVaR framework offers a candidate tool that requires further validation with large samples for tail risk assessment in heavily regulated fluvial systems. Full article

Reliable precipitation projections are essential for water-resource management, flood-risk assessment, and drought preparedness in hydroclimatically complex regions such as Mexico, where uncertainty remains high due to monsoon dynamics, complex topography, and tropical moisture transport. This study evaluates paired CMIP5 and CMIP6 global climate models in simulating the historical (1940–2005) precipitation annual cycle across four regions of Mexico (NW, NE, SW, SE). Model outputs were compared against ERA5 and cross-validated with CRU using complementary metrics assessing error magnitude, variability, temporal phase, and spatial coherence. Results indicate that CMIP6 provides moderate but regionally heterogeneous improvements rather than a uniform advance. The most consistent gains occur in NE and SE Mexico, where dry biases are reduced and seasonal amplitude is better represented. In contrast, SW Mexico exhibits persistent summer wet biases linked to monsoon–topography interactions, while improvements in NW Mexico are mainly confined to selected individual CMIP6 models and are not consistently reflected in the ensemble median. A marked SW–SE summer dipole bias highlights ongoing deficiencies in representing moisture transport and convection. These findings demonstrate that increased model complexity does not guarantee improved regional skill and that ensemble medians may mask individual model performance, underscoring the need for targeted model selection, multi-dataset validation, and bias-correction strategies. Full article

Accurate reservoir-fluid identification and water saturation prediction are essential for remaining-oil evaluation and water-flooding adjustment in heterogeneous oilfields. However, in the Youshashan Oilfield, southwestern Qaidam Basin, China, thin interbeds, strong reservoir heterogeneity, complex oil–water transitions, and inter-well logging-response differences make conventional single-task interpretation difficult. To address these problems, this study proposes a joint prediction method based on the Youshashan Fluid Prediction Network (YSF-Net) for six-class reservoir-fluid identification and continuous water saturation (S_w) prediction. A total of 200 wells were used and strictly divided by well into 140 training wells, 30 validation wells, and 30 independent test wells to avoid data leakage. Conventional logs were first processed through depth matching, outlier correction, robust standardization, and missing-value masking. Then, sliding-window logging sequences, regional stratigraphic embeddings, and reservoir-prior parameters, including shale volume, porosity, and permeability, were jointly input into the YSF-Net. The model uses a shared feature encoder with classification and regression branches to simultaneously identify oil layers, oil–water layers, water layers, and weakly, moderately, and strongly water-flooded layers, while predicting continuous S_w. A modified Simandoux-based physical consistency constraint was further introduced during training to improve the geological rationality of S_w prediction. Experimental results show that YSF-Net outperforms the CNN, BiLSTM, CNN-BiLSTM, and Transformer. It achieves an Accuracy of 0.926, Macro-F1 of 0.913, Macro-AUC of 0.968, S_w RMSE of 0.061, S_w MAE of 0.047, and S_w R² of 0.947. In direct cross-well testing without fine-tuning, YSF-Net obtains a Cross-well Accuracy of 0.918, Cross-well Macro-F1 of 0.904, and Cross-well S_w RMSE of 0.064. Ablation, transition-boundary, and typical well-interval analyses further demonstrate that regional constraints, reservoir-prior inputs, multi-task learning, and physical consistency improve class-boundary discrimination and S_w prediction reliability. The proposed method provides an accurate, consistent, and practical workflow for intelligent reservoir-fluid interpretation in heterogeneous reservoirs. Full article

Social networks such as Twitter have grown rapidly and are now flooded with sarcastic comments, both in text and in images. Detecting sarcasm in multimodal data has significant social value and is attracting increasing research attention. However, most studies overlook the role of sentiment, even though sentiment information in text is closely linked to clues of sarcasm. Additionally, few consider how text and images align semantically. To address these issues, we propose a sentiment-guided mutual learning network (SGMLN) for multimodal sarcasm detection. SGMLN utilizes sentiment information to inform the combination of text and image features, and employs mutual learning to facilitate knowledge sharing among classifiers. We design a sentiment-guided attention layer that injects sentiment into both modalities, producing features that capture sarcasm more effectively. Sentic-BERT extracts sentiment-aware vectors from text, using word-level sentiment as a mask. In mutual learning, a logistic distribution function measures differences between classifiers, improving knowledge transfer between modalities. This step boosts multimodal understanding and model performance. By introducing sentiment-aware representations and semantic alignment, SGMLN bridges the gap between text and images, making them more consistent. Experiments on public datasets demonstrate that our model is effective and outperforms alternatives. Full article