Heat waves in India: patterns, associations, and subseasonal prediction skill
Raju Mandal,Susmitha Joseph,Shubham Waje,Anurag Chaudhary,Avijit Dey,Mahesh Kalshetti,A. K. Sahai
DOI: https://doi.org/10.1007/s00382-024-07539-x
IF: 4.901
2024-12-18
Climate Dynamics
Abstract:Heat waves (HWs) in India during March–June and from 1951 to 2023 are thoroughly analysed in this work, emphasising trends, decadal variations, and related large-scale features. Average HW days per decade and anomalies are computed using HW criteria based on high-resolution maximum temperature (Tmax) data. The findings indicate a notable increase in HW occurrence in the central, southeast, and northwest regions after the year 2000. Month-wise analyses reveal detailed patterns, showcasing increased HW days in non-traditionally hot months, like March in southern regions. This suggests an intensification of extreme summer conditions over both time and across different regions. Examining the spatial HW trends exposes a notable increase in total HW days/year over northwest, central and south-eastern regions, while few others witness decreasing trends. The study reveals significant increasing trends in the total number of HW days in the two HW-prone regions, Northwest (NW) and Southeast (SE), from 1951 to 2023, where HW spells have also become more persistent. Three types of HW spells are analysed: NW-spells and SE-spells, defined by area-averaged daily Tmax exceeding 43 °C and 40 °C, respectively, for six consecutive days, and NWSE-spells, where HW periods overlap between the two regions. The analysis of large-scale characteristics associated with these HW spells emphasises the possible role of oceans and atmospheric variables in HW patterns. These findings highlight the importance of improving predictive capabilities for HWs. To this end, the extended range prediction system version 2 (ERPSv2) is introduced in this study, assessing its subseasonal prediction skill. The results demonstrate that ERPSv2 performs better than its predecessor, ERPSv1, particularly with a three-week lead time. Validation through a case study on the June 2023 HW disaster showcases ERPSv2's efficacy in forecasting real-time events with a four-week lead time. Incorporating ERPSv2 adds a practical dimension, enhancing HW predictions and facilitating timely responses to extreme heat events, crucial for public health measures and climate resilience planning in the face of escalating HW occurrences.
meteorology & atmospheric sciences