PREDICTING THE LOCAL IMPACT OF URBAN FLOODING IN MAJOR CITIES

Flood are one of the unfavorable naturaldisasters. A flood can result in a huge loss of humanlives and properties. It can also affect agricultural landsand destroy cultivated crops and trees. The flood canoccur as a result of surface-runoff formed from meltingsnow, long-drawn-out rains, and derisory drainage ofrainwater or collapse of dams. Today people havedestroyed the rivers and lakes and have turned thenatural water storage pools to buildings andconstruction lands. Flash floods can develop quicklywithin a few hours when compared with a regularflood. Research in prediction of flood has improved toreduce the loss of human life, property damages, andvarious problems related to the flood. Machinelearning methods are widely used in building anefficient prediction model for weather forecasting. Thisadvancement of the prediction system provides cost- effective solutions and better performance.This paper, a prediction model is constructed using rainfall data topredict the occurrence of floods due to rainfall. Themodel predicts whether “flood may happen or not” based on the rainfall range for particular locations.Indian district rainfall data is used to build theprediction model.