Unraveling extreme high-temperature events in South Asia: insights from ensemble learning models and geophysical drivers

Suravi Ghosh, Priyanko Das (), Zhenke Zhang, Jianzhong Lu, Brian Odhiambo Ayugi and Zhi Gao
Additional contact information
Suravi Ghosh: Institute of Atmospheric Physics (IAP)
Priyanko Das: Princeton University
Zhenke Zhang: Institute of African Studies, School of Geography and Ocean Sciences, Nanjing University
Jianzhong Lu: Wuhan University
Brian Odhiambo Ayugi: University of Bern
Zhi Gao: Wuhan University

Climatic Change, 2025, vol. 178, issue 8, No 9, 27 pages

Abstract: Abstract The average global temperature has risen and is expected to continue increasing due to the emission of greenhouse gases. South Asia (SA) has seen a notable rise in both hot days and nights over the past few decades, leading to numerous fatalities. This research investigates historical and future extreme high temperature (EHT) occurrences in SA, identifying their causal connections among geophysical drivers (GD) using causal discovery (CD). An Ensemble Machine Learning (EML) model was created to merge bias-corrected CMIP6 GCMs, enhancing the precision of future EHT event projections. The findings show that the EML algorithm performed exceptionally well (CC = 0.98) compared to conventional ensemble models, accurately capturing EHT events in SA. The intensity of hot days and nights (TXx and TNx) increased during the first (1991–2000) and second (2001–2010) decades examined, with warmer climate conditions (0.01–0.6 °C). Future projections suggest that the intensity of hot days and nights will rise by up to 10 °C by the end of the twenty-first century, with the frequency of hot days and nights (TX90p and TN90p) increasing by approximately ~ 50% in SA countries. Causal discovery (CD) findings imply that soil moisture, specific humidity, solar flux, aerosol optical depth, and cloud cover significantly influence EHT events in SA countries. This research offers valuable insights for crafting mitigation strategies and reducing future health risks in SA.

Keywords: High temperature; South Asia; Machine learning; Causal relation (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10584-025-03993-6 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:climat:v:178:y:2025:i:8:d:10.1007_s10584-025-03993-6

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/10584

DOI: 10.1007/s10584-025-03993-6

Access Statistics for this article

Climatic Change is currently edited by M. Oppenheimer and G. Yohe

More articles in Climatic Change from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().