Optimum pattern to identify flooded areas using remote sensing techniques (Houiza)

Journal of Radar and Optical Remote Sensing and GIS

مقاله 4، دوره 6، شماره 2، شهریور 2023

نوع مقاله: مقاله پژوهشی

شناسه دیجیتال (DOI): 10.30495/jrors.2023.1993648.1195

نویسندگان

seyede Razieh Keshavarz^{* 1}؛ Seyed Aghil Ebrahimi²؛ Jalal bayati³؛ Ali Hasan Abadi⁴

¹Master of Remote Sensing and Geographic Information Systems

²Islamic Azad University Tehran Branch

³Expert of Remote Sensing and Geographic Information Systems

⁴Governor's employee

چکیده

Several methods have been developed to represent flood-related hazards using ground-based measurements. Satellite remote sensing data have been used for flood assessment due to their spatial resolution and capacity to provide information for areas with poor access or lack of ground measurements. Identifying flood prone areas is one of the basic strategies in planning to reduce the destructive effects of floods. Flood is considered as one of the events that causes damage to human societies. Therefore, the importance of estimating the damages caused by flood and determining its extent in planning to reduce these damages and determining high risk points is very important. For this purpose, the OLI sensor images of Landest 8 satellite, before and after the flood of April 2018 in Khuzestan, Hoyze region, were used. Then NDWI water index was used for identification and SVM (Support Vector Machine) method was used for classification and it was found that before the flood there were 36999.99 hectares of water areas in the region and after the flood this amount reached 274279.95 hectares. The results show that the south and south-west parts are in a very severe situation and the central and south-eastern parts are in a severe flood risk situation, which are among the most prone to flood areas in the province. Also, the monitoring of flood maps in Khuzestan province shows that there is a perfect match between the flood zone map and the recent flood.

کلیدواژه‌ها

Flood؛ Landsat؛ Hoyze؛ remote sensing؛ backup vector machine