Flood susceptibility and probability mapping using frequency ratio ‎method in Pol-Doab Shazand Watershed

Because of data lacking in the most of watersheds, many researchers applied spatial analysis in GIS environment to study the hydrological and flooding condition. This study aims to map flood susceptibility through frequency ratio technique using some parameters such as digital elevation technique, slope, curvature, topographic wetness index, stream power index, average rainfall, distance from river, geology, soil type and land use in Pole-Doab, Shazand Watereshed, Makazi Province, Iran. First, digital map of all of the parameters were prepared based on raster format using Arc GIS 10.1 and SAGA GIS2 softwares. To prepare land-use map IRS-1C satellite image, ENVI 4.8 software and maximum likelihood algorithm were applied. Then flood inventory map was produced by mapping 95 flood prone locations in the study area using documented information on the May-2003, May-2004 and March-2004 floods. These 95 locations divided into two groups including 67 points (70%) and 28 points (30%) for calibration and validation, respectively. For calibration, flood prone locations defined as dependent variable and ten parameters that are affecting flooding condition were introduced to frequency ratio as independent variables. Then flood probability was determined for each class of each parameter. Finally, obtained weights for each class in GIS were implemented in corresponding layers and using the overlay algorithm, susceptibility and probability maps were prepared. Based on the susceptibility map, study area was divided into 5 classes as very high, high, medium, low and very low sensitivity. The findings of the assessment of frequency ratio histogram indicated that the likelihood of very high and very low flooding classes are equal to 67.86 and zero percent, respectively. Therefore, the obtained frequency ratio histogram confirms the adequacy of the implemented method in flood susceptibility and probability mapping for the case study.