A Remote Sensing and GIS Approach Toward the Analysis of Patel Milmet Dam Burst, Kenya

On May 9th 2018, Patel’s Milmet dam burst amidst heavy rainfall and reported the death of at least 48 people including large scale damage of infrastructure in the downstream region. The Milmet dam was one of the earthen embankment dams in Kenya with a capacity of about 200,000 cubic meters. This study relies on the use of advanced geospatial techniques, to map and assess the dam break phenomena. Imagery acquired by Sentinel-1 & Sentinel-2 missions during pre- and post-dam burst event as well as daily precipitation data over the study area from global precipitation measurement (GPM) mission was utilized. Elevation information was acquired from a 12.5 m spatial resolution Digital Elevation Model (DEM), generated from ALOS PALSAR data. This DEM has been used for analyzing the elevation pattern and mapping the catchment area of the dam. It is observed that the elevation of the study area ranges from 1632 to 2430 m above Msl with the catchment area covering around 480 ha. The change analysis of pre and post event satellite imagery identified that the flash flood caused due to dam burst damaged agricultural fields of about 40 ha. In addition to this, approximately 1 km length of road network in the downstream which included a major road of about 100 m was also affected. Our analysis also observed that the 90 m burst that extended the breach length affected an area of about 50 ha. Therefore, the study depicts that the huge amount of precipitation in April 2018 along with the steep slope of the dam area may be the major cause for the catastrophic failure. The paper ends with the remark that usage of remotely sensed datasets are essential for planning, mitigating and estimating damage of an area from natural hazards, especially as an early response for disaster mitigation.