Development of Intensity-Duration-Frequency Curves for Future Climate Change Scenarios in Some Major Cities of Bangladesh

Abstract:

In the context of climate change, Bangladesh is regarded as one of the most vulnerable countries in the world. It is anticipated that more frequent extreme rainfall events will occur in the country in near future that will severely impact the urban stormwater infrastructure. In this study, we will present the rainfall intensity-duration-frequency (IDF) relationships for some major cities in Bangladesh considering future climate change scenarios. We will utilize the data of predicted daily rainfall from available 107 General Circulation Models (GCMs) for Representative Concentration Pathways (RCP) 4.5 and 78 GCMs for RCP 8.5 under Coupled Model Inter-comparison Project 5 (CMIP5) project for the future period of 2021-2100. The rainfall data from the GCMs will be statistically downscaled and then bias corrected to represent the data at a station level. The historical rainfall data for the available raingauge stations in Bangladesh for the period of 1961-2018 will be collected from the Bangladesh Meteorological Department. Two commonly used frequency analysis techniques such as Gumbel and Log Pearson type III distribution will be used to derive IDFs for various return periods. Deep neural network technique and an empirical equation by Indian Meteorological Department (IMD) will be used to estimate the rainfalls for duration ranging from 10 minutes to 24 hours from yearly maximum rainfall data. The comparative analysis and uncertainties associated with the IDFs will also be discussed. This research is sponsored by the University Grants Commission of Bangladesh.