FLOODPLAIN ASSESSMENT WITH THE INTEGRATION OF REMOTE SENSING AND GIS: MERIC RIVER CASE STUDY

Abstract

In this study, Meric River flood that occured on 16 February 2010 is investigated by optical satellite images. The extent of flooding has been established by using multitemporal Landsat 5 TM images taken before, during and after the flood. With the help of these images, land cover and flood maps were generated and several evaluations were accomplished. Supervised and unsupervised classification techniques were applied for before, during and after flooding images. The data of the predominantly agricultural area, obtained by these classification techniques were compared with the data based on ground observations. These observational methods recorded on topographic maps executed by XI. Regional Directorate of State Hydraulic Works (DSI)-Edirne. It is determined that the data of the agricultural area obtained by the supervised classification method has a 91% accuracy compared to DSI ground based data. It is concluded that optical satellite images can be effectively used for flood studies.

Keywords

• Remote Sensing
• Geographic Information System
• Classification
• Flood
• Accuracy Analysis