Assessment Of Flood Risk Using GIS and Remote Sensing Techniques: A Case Study of River Mubuku Uganda
Loading...
Date
2023-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Kampala International University
Abstract
Floods are regarded to be the most common and dangerous natural hazard in the entire world. In Uganda, floods have affected more than hundreds of thousands of people across central, western, eastern, and northern region. A case in time is the recent 2020 Kasese flood scenario that resulted in over 1,252 people displaced and property worth billions of shillings or dollars destroyed. However, despite the above devastation, no comprehensive study has been undertaken to assess flood risk and roles played by different flood drivers in Mubuku watershed. The current study aimed at assess flood extents and roles played by different drivers using remote sensing techniques. Relative frequency Ratio (FR) Model was utilized to assess flood risk and the importance of each flood causative factor. A total of (8) flood causative factors were analyzed and sub class performance of each factor indicate altitude (FR = 8.27) for class 1000 – 1500, drainage density (FR = 2.77) for class 10 – 15, slope <5 (FR = 2.97), commercial farming (FR = 25.86), SPI (FR = 1.93) for class <5, peat or peated sand and clays soil type (FR = 48.45), precipitation (FR = 2.44) for class 1500.01 – 2000.00 and finally TWI (FR = 11.89) for class 16.01 - 23.74. Overall influence of selected causative factors indicated soil type (18%) as the largest influencer, followed by altitude (17%), drainage density (13%) and slope (10%). Validation of the flood model indicated good performance by ROC receiver operating characteristic method Area under Curve (AUC = 0.899). Therefore, these results suggest that relative frequency Ratio Model can be utilized as a quick alternative method for analyzing flood risk. Moreover, flood control measures targeting soil improvement such as bamboo and tree planting, flow management such as gabions are equally important. Key Words: GIS, Flood risk, Remote sensing Relative Frequency Ratio Model, Mubuku.