INVESTIGATION OF CHARACTERISTICS, DRIVERS, AND PREDICTABILITY OF COMPOUND DRY AND HOT EXTREMES

Abstract

Compound drought and hot extremes (CDHE) are periods of prolonged dry and hot weather exhibiting adverse impacts on nature and humankind than their counterparts. Understanding compound extremes is in its infancy due to complex dynamical climate systems involving interactions and feedback with the different processes at different scales. A multitude of studies have investigated the characteristics, drivers, and future trends of CDHE. However, most of this research has focused on the hottest three-month period, with limited attention given to the Indian Summer Monsoon and its intraseasonal variability. This study has investigated the characteristics, drivers, and predictability of CDHEs during the ISM, considering intra-seasonal variation. Our detailed investigation of the last seven decades of CDHE during the Indian Summer Monsoon has shown alarming observations. Our results confirmed a threefold increase in CDHE frequency for the recent period (1977–2019) relative to the base period (1951–1976), exhibiting a strong spatial pattern. Further investigation revealed that CDHE likelihood and spatial diversity in CDHE occurrence is a function of the strong negative association between precipitation and temperature and soil moisture-temperature coupling, respectively. Investigation into the temporal evolution of CDHE confirms the strengthening of the negative association between precipitation and temperature, indicating a higher number of CDHEs in the future. Our complexity-based approach outperformed traditional and contemporary approaches in predicting CDHE at a seasonal scale on a grid resolution across India’s mainland. This thesis has employed a comprehensive approach, utilizing statistical methods and modelling techniques to investigate the various climatological and hydrological elements contributing to CDHEs. The overall contribution of this thesis demonstrates that CDHEs are the new normal and explains the spatial diversity of their likelihood with improved prediction accuracy.