Assessment of Spatial-temporal Lake Victoria Shoreline Variations using Synthetic Aperture Radar

Abstract

In the context of their dynamic and linear character, shorelines feature as among the most important features on the earth’s surface. They change shape and position over multiple spatial and temporal scales, with their water levels serving as a key indicator to characterize the expansion or shrinkage of the water body in question. In the context of Lake Victoria, the central focus of this article, there has been an uncontrolled increase in its water levels which has ultimately contributed to variations in the lake shoreline. This variation has in turn led to unpredictable flooding along the shores of the lake, claiming human property. Therefore, environmental management authorities, such as NEMA, require accurate and up-to-date information about shoreline changes. The main objective of this study was to assess the spatial-temporal variations of the Lake Victoria shoreline in the Southern Buganda sub-region for the period 2015 - 2021, by employing microwave remote sensing. Sentinel 1 and Sentinel 2 imagery were used. The study also assessed the performance of HH and VH polarizations in shoreline delineation. Different image processing techniques such as thresholding and band math were used in both SNAP and ArcGIS software. Based on the DSAS evaluation statistics, VH polarization performed a better delineation of the shoreline than HH polarization. The study also found that the lake shoreline in the Southern Buganda sub-region was subject to entirely low erosion rates, ranging from 0.5m/yr to 2m/yr, as observed in the sub-counties of Buwunga, Kyanamukaaka, and Kabira. High erosion rates of above 5m/yr were observed in some areas in the Bukakata and Kyebe sub-counties. This study recommends that VH polarization be used. Further studies could integrate predictive analytics to attain future shoreline positions.