Modelling and short term forecasting of flash floods in an urban environment

Rapid urbanization, climate change, and extreme rainfall have resulted in a growing number of cases of urban flash floods. It is important to predict the occurrence of a flood so that the aftermath of it can be minimized. As the name suggests, an urban flash flood occurs in an urban area in a very short span of time. To reduce the impact of these events, short-term forecasting or nowcasting is used for prediction of the very near future incident. In orthodox methods of flood forecasting, current weather conditions are examined using conventional methods such as the use of radar, satellite imaging and calculations involving complicated mathematical equations. However, recent developments in Information and Communication Technology (ICT) and Machine Learning (ML) has helped us to study this hydrological problem from a different perspective. The aim of this paper is to design a theoretical model considering the parameters causing the urban flash flood and predict the event beforehand. To test the soundness model, data syntheses is performed and the results are checked using the artificial neural network.