Establishment of best-fitted rainfall intensity-duration-frequency curves for the hydrologic regions of Bangladesh

Abstract

Frequency analysis is a fundamental approach for estimating the magnitude of extreme weather occurrences, including heavy rain, flooding, and other climatic phenomena. Rainfall frequency analysis iswidely used to design storm runoff-controlling engineering works such as agricultural drainage systems, highway and railway culverts, and, most critically, urban storm sewer systems. Establishing a reliable Rainfall Intensity-Duration-Frequency equation for a given location that offers accurate rainfall estimates throughout different durations and recurrence periods is critical for various water resources projects and designs estimating the highest rainfall/design rainfall. A realistic estimation of the Rainfall Intensity-Duration-Frequency equation and trend analysis assist engineers and practitioners in predicting the return period and risk assessments of flooding, cyclones, and other extreme climatic occurrences to design infrastructures. The best-fitted Rainfall Intensity-Duration-Frequency equation was derived in the present study utilizing rainfall data from 34 locations from 1948 to 2017. The data for rainfall have been examined for trend, homogeneity, and randomness first. The Mann-Kendall trend test indicates that four of the 34 stations show significant trends, which have been de-trendedbefore the further examination. The rainfall data for Jessore are identified to be non-homogeneous using both the Standard Normal Homogeneity test and the Pettitt test, and hence, they were excluded from the frequency analysis. In a randomness test, it is determined that the rainfall data at all stations are random. Five of the most frequently used IDF equations and five goodness of fit tests were employed to determine the best-fitting rainfall IDF equation.The lowest sum of the ranks from each station across seven recurrence intervals was used to determine the best-fitted rainfall IDF equation for a station. The rainfall IDF equation with the lowest rank for the maximum number of stations was selected as the best fit for Bangladesh. This analysis found the Sherman Equation as the best-fitted rainfall IDF equation for Bangladesh. Bangladesh's hydrological and climatic zones have also been evaluated for the best-fitted rainfall IDF. In the hydrological zones, the Sherman Equation was found to be the best fit for the North-West (NW), North-Central (NC), South-West (SW),South-East (SE) zone, and Eastern Hills (EH) Zone, while the Bernard was found to be the best fit for the North-East (NE) and South-Central (SC) zones.While considering the climatic regions, it has been found that the Sherman Equation is best-fitted IDF equation in zone A: South-Eastern,Zone C: North-Northern, Zone D: North Western, Zone E: Western Zone, Zone G: South Central; Bernard Equation is best fitted inZone B: North-Eastern and Zone F: South-Western Zone. These findings can be used to plan engineering projects that require stormwater runoff calculations.