Flood is considered to be a major natural disaster which affects many parts of the world. Geographical Information System (GIS) can be used to visualize the extent of flood and also to analyze the risk of disasters. The objective of this study is to generate a flood hazard map using satellite image data from Landsat 5. To characterize the land cover types it was used an innovative Semi-automated Classification Processing (SCP) tool. An experimental study was conducted on the downstream of Prut River Romania, more exactly nearby Cârja (Romania) village situated on the right bank and Goteşti (Moldova) village situated on the left bank of Prut river. To monitor the effects of flood was used the supervised classification of two Landsat 5 images acquired on 2010 when a severe flood was affected the above mentioned area. This research presents a very rapid and affordable method of flood monitoring that could be very useful for the emergency management plan of the local authorities.