Are precipitation concentration and intensity changing in Bangladesh overtimes? Analysis of the possible causes of changes in precipitation systems

Md Siddiqur Rahman; Abu Reza Md Towfiqul Islam

doi:10.1016/j.scitotenv.2019.06.529

Are precipitation concentration and intensity changing in Bangladesh overtimes? Analysis of the possible causes of changes in precipitation systems

Sci Total Environ. 2019 Nov 10:690:370-387. doi: 10.1016/j.scitotenv.2019.06.529. Epub 2019 Jul 3.

Authors

Md Siddiqur Rahman¹, Abu Reza Md Towfiqul Islam²

Affiliations

¹ Department of Disaster Management, Begum Rokeya University, Rangpur 5400, Bangladesh.
² Department of Disaster Management, Begum Rokeya University, Rangpur 5400, Bangladesh. Electronic address: towfiq_dm@brur.ac.bd.

PMID: 31299571
DOI: 10.1016/j.scitotenv.2019.06.529

Abstract

A comprehensive understanding of the changing behaviors of precipitation concentration and intensity plays a pivotal role in water resource management. Hence, we investigated the spatiotemporal changing behaviors of frequency and intensity of 13 precipitation indices and their probable causes of changes in precipitation systems. This study used daily precipitation datasets from 23 sites in Bangladesh and six atmospheric circulation indices during 1975-2017. The results showed that the precipitation concentration index (LCI) varies between 0.57 and 0.63, and the highest value was found in the southeastern region. The precipitation days frequency indices such as AD, LPD, MPD, HPD, and WPD have significantly increased in Bangladesh while precipitation intensity indices such as AII, LPI, MPI, HPI, and WPI have significantly declined; all types of indices have clear rapid changes. The results of detrended fluctuation analysis (DFA) exhibit long-term correlations among all precipitation indices, suggesting that these indices will sustain their present trend line in the upcoming period. The Sunspot (SS) and East Asian Summer Monsoon Index (EASMI) had a negative influence on ACI and South Asian Summer Monsoon index (SASMI) had a strong positive influence on precipitation days frequency indices. The significance analysis using the random forest (RF) algorithm showed that SS is the largest contributing factor affecting the precipitation systems in Bangladesh. ECMWF ERA5 reanalysis datasets revealed that elevating summer geopotential height, higher anticyclonic anomaly, increasing low and decreasing high cloud covers and lower solar radiation with adequate moisture divergence fluxes contributed to variations in precipitation extremes in Bangladesh.

Keywords: Bangladesh; Detrended fluctuation analysis; Precipitation concentration indices; Random forest algorithm; Sunspot influence.