Spatiotemporal variations of thunderstorm frequency and its prediction over Bangladesh

Abstract

Understanding the spatiotemporal variations of thunderstorm (TS) frequency over Bangladesh under the changing climate is of immense importance but remains a challenging task due to the lack of persistent and reliable weather data set. This study demonstrates a statistically significant increasing trend in the occurrence of monthly, seasonal, and annual TS frequencies except for decadal periods across Bangladesh during the past 6 decades. Further, land use influences and seasonal variabilities in convective available potential energy (CAPE) are also explored in various regions to understand the spatial variabilities among different parts in Bangladesh. The results show the highest occurrence of TS in monsoon and the lowest occurrence in the winter season. The results of Ward’s Hierarchical Cluster analysis (WHCA) illustrate four clusters for the whole country, where the higher TS frequency was observed in northeastern and northern regions and the lower in the southern part of Bangladesh due to the variation of CAPE and development of the low-pressure system. The occurrence of TS frequency maxima and lower coefficients of variation are found in the pre-monsoon and monsoon seasons. The variation coefficient of the winter season is high, while the pre-monsoon season is low. The results of Mann–Kendall (MK) and Spearman’s rho (SR) tests reveal the highest increase and decrease trends in M. Court and Tangail stations. The ARIMA model outcomes are consistent with the findings of MK and SR tests. Overall, this study reveals that elevated CAPE and climate warming may be reasoned of increasing the TS frequency over Bangladesh.