Flood Risk Assessment of Kamala River using Remote Sensing and GIS

The objectives of this project were: 1) To integrate Remote Sensing, GIS and flood modeling to create flood maps, 2) To map flood hazard and vulnerability.

      The study area of this project was Kamala river basin in Nepal. Remote sensing and GIS was used in addressing the flood problem in the study area. ALOS/AVNIR-2 data of November 2006 was used for the landuse/landcover classification of the study area. The landuse classes revealed that about 85% of the study area was agricultural land. This landuse classes were utilized in assigning the Manning’s roughness values to the HEC-RAS model for simulating flood maps. Total population per village was converted to population per pixel by first resampling and then multiplying with the weighted landuse class. The peak discharge data in the Kamala river from 1955 to 2002 were taken for the flood frequency analysis at the Chisapani gauge station. As ALOS/PRISM data of the study area was not available, the DEM of the study area was generated from the contour data digitized from 1:25,000 topographic sheets and spot heights. This DEM was converted to Triangulated Irregular Network (TIN) and given as one of the inputs to the HEC-RAS model. The model was used to simulate flood for a return period of 10, 25, 50 and 100 years. Results showed that for a 10 year flood event, around 143,000 people will be affected by the flood. Out of this population, about 58,000 are in low vulnerable class, while 65,000 are in moderate vulnerable class and 20,000 are in high vulnerable class. Similarly, 80% of the study area are under low vulnerability class, 17% in moderate vulnerability class and 3% is in high vulnerability class.