Attribution analysis of urban social resilience differences under rainstorm disaster impact: Insights from interpretable spatial machine learning framework

Sustainable Cities and Society, Journal Year: 2024, Volume and Issue: unknown, P. 106029 - 106029

Published: Dec. 1, 2024

Language: Английский

---

Flood scenarios vehicle detection algorithm based on improved YOLOv9

Multimedia Systems, Journal Year: 2025, Volume and Issue: 31(1)

Published: Jan. 17, 2025

Language: Английский

---

Flood change detection model based on an improved U-net network and multi-head attention mechanism

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 26, 2025

This work aims to improve the accuracy and efficiency of flood disaster monitoring, including monitoring before, during, after flood, achieve accurate extraction change information. A modified U-Net network model, incorporating Transformer multi-head attention mechanism (TM), is developed specifically for characteristics Synthetic Aperture Radar (SAR) images. By integrating TM, model effectively prioritizes image regions relevant disasters. The trained on a substantial volume annotated SAR data, its performance assessed using metrics such as loss function, accuracy, precision. Experimental findings demonstrate significant improvements in value, precision compared existing models. Specifically, algorithm this reaches 95.52%, marking 3.46% improvement over baseline network. Additionally, achieves an 90.11% while maintaining value approximately 0.59, whereas other algorithms exceed 0.74. Thus, not only introduces novel technical approach but also has potential enhance response procedures provide scientific evidence management risk assessment processes.

Language: Английский

---

Assessment of the Impact of Extreme Hydrological Conditions on Migratory Bird Habitats of the Largest Freshwater Lake Wetlands in China Based on Multi-Source Remote Sensing Fusion Approach

Sustainability, Journal Year: 2025, Volume and Issue: 17(5), P. 1900 - 1900

Published: Feb. 24, 2025

Poyang Lake, the largest freshwater lake of China, serves as a crucial wintering site for migratory birds in East Asian–Australasian Flyway, where habitat quality is essential maintaining diverse bird populations. Recently, frequent alternation extreme wet years, e.g., 2020, and dry 2022, have inflicted considerable perturbation on local wetland ecology, severely impacting avian habitats. This study employed spatiotemporal fusion method (ESTARFM) to obtain continuous imagery Lake National Nature Reserve during seasons from 2020 2022. Habitat areas were identified based classification water depth constraints. The results indicate that both conditions exacerbated fragmentation shallow habitats showed minor short-term fluctuations response levels but more significantly affected by long-term hydrological trends. These exhibited interannual variability across different affecting their proportion within overall distribution area. demonstrates ability ESTARFM reveal dynamic changes responses conditions, highlighting critical role analysis. outcomes this improve understanding impact habitats, which may help expand knowledge about protection other floodplain wetlands around world.

Language: Английский

---

Flood vulnerability assessment in the Ili River Basin based on the comprehensive symmetric Kullback–Leibler distance

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 3, 2025

In vulnerability assessments, accurately determining the indicator weights is essential to ensure results' precision and reliability. This paper proposes an optimized comprehensive symmetric Kullback–Leibler (K–L) distance weighting method, in which K–L for each calculated using a grid-based approach, normalized serves as weight indicator. ArcGIS software was employed assess Ili River Basin flood case study. The results reveal following: (1) method facilitated variable processing disaster where it offered scientific adaptable approach indexing vulnerability, thus improving both evaluation accuracy practicality. (2) spatial distribution of levels uneven, with higher observed northwestern, southwestern, southeastern regions, lower eastern northeastern areas. Yining County, City, certain southern regions Cocodala City were particularly vulnerable due multiple influencing factors, including population, economy, society. These areas require focused attention preventive measures.

Language: Английский

---

Risk assessment of flood disasters in the Loess Plateau using the Hazard–Sensitivity–Vulnerability–Recoverability framework

International Journal of Disaster Risk Reduction, Journal Year: 2025, Volume and Issue: unknown, P. 105379 - 105379

Published: March 1, 2025

Language: Английский

---

Disaster Management Systems: Utilizing YOLOv9 for Precise Monitoring of River Flood Flow Levels Using Video Surveillance

SN Computer Science, Journal Year: 2025, Volume and Issue: 6(3)

Published: March 14, 2025

Language: Английский

---

Fine-Scale Identification of Agricultural Flooding Disaster Areas Based on Sentinel-1/2: A Case Study of Shengzhou, Zhejiang Province, China

Atmosphere, Journal Year: 2025, Volume and Issue: 16(4), P. 420 - 420

Published: April 4, 2025

Flood disasters are one of the major natural hazards threatening agricultural production. To reduce disaster losses, accurately identifying flood-affected areas is crucial. Taking Shengzhou City as a case study, we proposed refined method for by integrating microwave and optical remote sensing data with deep learning techniques, GIS, pixel-based direct differencing method. Complementary advantages can effectively solve problem difficulty in detecting floods due to thick clouds before after flood disasters. Deep technology identify farmland areas, pixel difference analyze Analyzing three typical rainfall events along topographical geomorphological characteristics City, results indicate that exhibit significant spatial heterogeneity. The primary influencing factors include intensity, topography, drainage infrastructure. northern, eastern, southwestern regions particularly peripheral adjacent mountainous hilly terrains, contain most farmland. These characterized low-lying highly susceptible Therefore, optimizing systems near essential enhance resilience.

Language: Английский

---

Holistic risk assessment using a hybrid approach in a flash flood disaster-prone area of the poyang lake basin

Natural Hazards, Journal Year: 2025, Volume and Issue: unknown

Published: May 5, 2025

Language: Английский

---

From lake to fisheries: Interactive effect of climate and landuse changes hit on lake fish catch?

Environmental Research, Journal Year: 2024, Volume and Issue: 258, P. 119397 - 119397

Published: June 12, 2024

Language: Английский

---