Urban land cover and morphometric analysis for flash flood vulnerability mapping and riparian landscape conservation in Kebena River watershed, Addis Ababa

Abstract

Addis Ababa city has a degraded stream ecosystem and redundant flash flooding that can destroy the existing urban infrastructure and utilities. This research aimed to map flash flood vulnerability of the Kebena watershed inside Addis Ababa and evaluate the status of the riparian landscape. Methodologically, the study employed the Biophysical Composition Index (BCI) to detect impervious surfaces and the Normalized Difference Vegetation Index (NDVI) to classify the vegetation cover. The Arc-hydro tool was used to identify micro-watersheds and measure the morphometric factors, then principal component analysis (PCA) identified the surrogate factors. Fuzzy overlay analysis combined land cover and morphometric analysis results to produce the final flash flood vulnerability map. Moreover, riparian buffering at 15, 30, and 90 m distances were defined to measure the degree of imperviousness, greenness, and vulnerability to flash flooding. Accordingly, 969 ha of land was depicted within the watershed as flash flood vulnerable areas. These areas are primarily found in the southeastern and southwestern parts where impervious land cover prevailed, and the northwestern portion which has extremely rugged terrain and has a sparse vegetation cover. For all buffering distances, the proportion of impervious surface is greater than the vegetation cover. Even within 15 m buffering distance, which was set as the national urban planning standard. It is concluded that Kebena watershed is vulnerable to flash floods as the riparian landscape is dominated by impervious and depleted vegetation cover. Therefore, integrated geospatial and statistical techniques are helpful to devise a method for sustainable riparian landscape monitoring.