Abstract:
One of the biggest natural disasters is flooding, frequently occurring in Pakistan, due to the excessive rainfall in monsoon and the climate change. These floods cause severe damage to physical and anthropogenic resources. Identification and mapping of flood susceptible areas is essential for rescue, relief, rehabilitation, and timely management related informed decision-making. Creating flood susceptibility maps and conducting assessments are essential elements of flood prevention and mitigation strategies. Geographic Information System (GIS) and remote sensing data are valuable resources for mapping flood susceptibility at different spatial scales. This study focused on utilizing remote sensing data and GIS to evaluate Frequency Ratio Model (FR), Shannon’s Entropy (SE) and Analytical Hierarchy Process (AHP) for the identification and mapping of flood susceptible areas in District Khairpur of Sindh Province. Eleven flood causative parameters were considered for flood susceptibility mapping. The results validation was based on the comparison of the historical flood susceptible zones using Area Under Curve (AUC) Method. The results reveal that precipitation, distance to stream, and soil are the most significant factors in flood generation and elevation is the least. The AUC values were 90.7%, 87.6%, and 79.5% for the FR, SE, and AHP models, respectively. These values indicate that FR model provides the highest accuracy and better results compared to SE and AHP. The results highlight the potential of remote sensing data for generating flood susceptibility maps and their use to devise effective mitigation and formulate efficient flood management plans.
