Flood inundation modelling using GIS spatial analyst :case study of Pekan sub-district, Pahang

Abstract

A flood modelling study was carried out focusing on flood prone area, Pekan sub-district, Pahang. 3-dimensional inundation map model was used to simulate and predict flood affected zone within the area. The observation of inundation pattern mainly highlights the relativity between land topography with surface water level from Department of Irrigation and Drainage (DID) gauge station of Pahang River located at Pulau Jawa pump house. Methods involved of processing multiple types of spatial information gathered from remotely sensed data within a GIS environment. Data was collected from close and open source provider for the input during modelling process. A set of satellite imagery and digital elevation data was used to build a 3D spatial features of Pekan sub-district. Normal gauge reading of Pahang River was identified from DID infobanjir website at 1.0m and the hazard level increase at 2.44m, 3.05m, and finally 3.66m. Multiple water body layers were populated based on these data and flood pattern was observed until 5.50m gauge reading projection. 3D terrain model was created with assimilation of terrain information from digital elevation model. Water body offsets projection were generated and flood was observed in stereo mode within GIS software. Pekan sub-district was tessellated into square grid networks of 250m2 each. The technique allows identification of flood affected region in relation with water level offsets in simulation. Affected grids were assigned with different color codes for user recognition purpose. A set of feature class were created which represent important infrastructures especially main routes, community halls, hospital, and schools. This supplementary information could be used to assess the rate of habitability and accessibility during flood. Finally, DID historical datasets comparison together with on-site surveys were carried out to determine and validate the outcome of the simulated inundation model with real flood situation. According to SRTM raster data, it is found that Pekan sub-district has about 50% of terrain between 0-10m level. These areas are the most prone to flooding due to accumulation of stagnant water which naturally will be concentrated on low lying ground verified using watershed analysis. With simulation and supporting historical datasets and digital images, Pahang River expansion is not the real caused of flood spread over this region. This study was aimed to produce a flood prediction model that deliver interface that can be understood easily by any level of audience. It will help in earlier preparation especially for the locals if flood were to be discovered in the near future. Information provided on affected region and infrastructures that were emphasized will allow systematic planning and support flood relief committee in emergency period.