Enhancing Long-Term Flood Forecasting with SageFormer: A Cascaded Dimensionality Reduction Approach Based on Satellite-Derived Data

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(3), P. 365 - 365

Published: Jan. 22, 2025

Floods, increasingly exacerbated by climate change, are among the most destructive natural disasters globally, necessitating advancements in long-term forecasting to improve risk management. Traditional models struggle with complex dependencies of hydroclimatic variables and environmental conditions, thus limiting their reliability. This study introduces a novel framework for enhancing flood accuracy integrating geo-spatiotemporal analyses, cascading dimensionality reduction, SageFormer-based multi-step-ahead predictions. The efficiently processes satellite-derived data, addressing curse focusing on critical long-range spatiotemporal dependencies. SageFormer captures inter- intra-dependencies within compressed feature space, making it particularly effective forecasting. Performance evaluations against LSTM, Transformer, Informer across three data fusion scenarios reveal substantial improvements accuracy, especially data-scarce basins. integration hydroclimate attention-based networks reduction demonstrates significant over traditional approaches. proposed combines advanced deep learning, both interpretability precision capturing By offering straightforward reliable approach, this advances remote sensing applications hydrological modeling, providing robust tool mitigating impacts extremes.

Language: Английский

---

VMDI-LSTM-ED: A novel enhanced decomposition ensemble model incorporating data integration for accurate non-stationary daily streamflow forecasting

Journal of Hydrology, Journal Year: 2025, Volume and Issue: unknown, P. 132769 - 132769

Published: Jan. 1, 2025

Language: Английский

---

Estimation of Flood Inundation Area Using Soil Moisture Active Passive Fractional Water Data with an LSTM Model

Sensors, Journal Year: 2025, Volume and Issue: 25(8), P. 2503 - 2503

Published: April 16, 2025

Accurate flood monitoring and forecasting techniques are important continue to be developed for improved disaster preparedness mitigation. Flood estimation using satellite observations with deep learning algorithms is effective in detecting patterns environmental relationships that may overlooked by conventional methods. Soil Moisture Active Passive (SMAP) fractional water (FW) was used as a reference estimate areas long short-term memory (LSTM) model combination of soil moisture information, rainfall forecasts, floodplain topography. To perform modeling LSTM, datasets different spatial resolutions were resampled 30 m resolution bicubic interpolation. The model’s efficacy quantified validating the LSTM-based inundation area mask from Senti-nel-1 SAR images regions topographic characteristics. average under curve (AUC) value LSTM 0.93, indicating high accuracy FW. confusion matrix-derived metrics validate had high-performance ~0.9. SMAP FW showed optimal performance low-covered vegetation, seasonal variations flat regions. estimates show methodological promise proposed framework resilience.

Language: Английский

---

Variational mode decomposition coupled LSTM with encoder-decoder framework: an efficient method for daily streamflow forecasting

Earth Science Informatics, Journal Year: 2024, Volume and Issue: 18(1)

Published: Dec. 13, 2024

Language: Английский

---

Attention-based deep learning framework for urban flood damage and risk assessment with improved flood prediction and land use segmentation

International Journal of Disaster Risk Reduction, Journal Year: 2024, Volume and Issue: unknown, P. 105165 - 105165

Published: Dec. 1, 2024

Language: Английский

---