Document Type : Original Article
Authors
1
Ph.D. Student, Department of Reclamation of Arid and Mountainous Regions, Faculty of Natural Resources, University of Tehran
2
Professor, Department of Reclamation of Arid and Mountainous Regions, Faculty of Natural Resources, University of Tehran
3
Research Expert, Khuzestan Water and Electricity Organization
4
Assistant Professor in Research, Bushehr Agricultural Research and Training Center
Abstract
Global warming, through its impact on precipitation and temperature, has intensified natural hazards such as floods and droughts, leading to significant damage to fisheries, agriculture, infrastructure, and human health. In this study, the landscape vulnerability of Khuzestan province to flood inundation and dust emission was simultaneously analyzed using the maximum entropy method. All driving factors were prepared in raster format, and training samples were collected separately from areas of flood and dust occurrence. These training points were verified through land surveying. The modeling was performed using the MaxEnt GUI software. Model validation was conducted using the ROC curve (AUC), and the importance of variables was assessed through the jackknife approach. The results indicated that, for both dust emission and floods, the AUC values for both training and validation were categorized as very good and good, respectively, demonstrating high model performance. Effective factors influencing dust emission included AOD (Aerosol Optical Depth), precipitation, elevation, population density, and NDVI. In contrast, variables such as distance from rivers, surface coarse fragments, NDWI, surface temperature, and land use had significant impacts on increasing vulnerability to floods. The integration of flood and dust storm maps revealed that 4% of the province's surface exhibited high vulnerability to both hazards, primarily in the southwestern regions. Additionally, 27% of the province was classified as having medium vulnerability. The northern half of the province, covering an area of 42,000 km², showed low or no vulnerability to flooding and dust storms. Given the accuracy of the model used, the findings of this study could be useful for decision-making and management, aiding in the organization and design of appropriate engineering and operational projects.
Keywords
Subjects
