Determination of flood prone areas with FR, SI and Shannon models in order to reduce flood risks (Case study: Kashkan watershed)

Lorestan province, especially the Kashkan basin, is very flooded and has suffered flood damage many times, and in April 2019, it experienced the biggest flood in the last 200 years. In this study, an attempt has been made to map flood zonation to reduce flood hazards in Kashkan watershed using FR, SI and Shannon models and also using ArcGIS techniques to improve flood decision-making and management in Provide this area. For this purpose, the geographical location of 123 flood-catching points in the region were divided into two groups: calibration and validation. In the implementation of all three models of effective parameters in floods including: slope, slope direction, land curvature, geology, land use, soil science, topographic moisture index, precipitation, waterway density, distance from waterway and digital elevation model of the area used Were. The ROC curve is also used to validate the results of the models. The highest accuracy for this region was attributed to Shannon entropy model, followed by frequency ratio models and statistical index, respectively. Due to the fact that surface water management is very important in order to prevent the recurrence of high flood damage in this area, so the use of flood sensitivity maps to improve management and decision-making in flood management is essential.