Assessment of the Conditioning Factor for Flash Flood Susceptibility Potential Based on Bivariate Statistical Approach in the Wonoboyo Watershed in East Java, Indonesia

Abstract

Flash floods that occur suddenly, which cause damage to the weirs or embankments, immediately threaten human life. Identifying the causes of a flash flood is very important to reduce its negative impact. This paper examines changes in flash flood disasters in the Wonoboyo watershed based on estimates of flash flood hazard, land-use changes, and rainfall depth distribution patterns. The method of predicting susceptibility to flash flood hazards is based on various environmental factors that are integrated with GIS. Three bivariate statistics consisting of the Statistical Index (SI), Frequency Ratio (FR), and Predictor Rate (FP-PR) model are applied to select the best Flash Susceptibility Index (FFHSI) model. Changes in land use are then explored based on the conditioning factor for a flash flood. In the final stage, the estimation of areal rainfall uses Inverse Distance Weighting (IDW) to describe the position of rain and flash flood events. The best statistical bivariate statistical approach for the FFHSI is FR. Assessment of environmental factors using the FFHSI shows that 21% of the catchment area has moderate to high until very high vulnerability levels. Changes in land cover significantly affect flash floods, especially changes from forest to agricultural land or settlements. The distribution pattern and intensity of rainfall are closely related to the location of the flash flood. This study results can guide future flood mitigation measures.