Climate change increases risks from tropical cyclone compound events to densely populated coral atolls in the Northern Indian Ocean.

Regions that experience frequent extreme weather events are assumed to have better monitoring, assessment and preparedness for future events. This is particularly true in the case of tropical cyclones (TC) that have a degree of predictability in their genesis, trajectory and intensity, especially in the Caribbean and the Pacific. However, in basins such as the Northern Indian Ocean, especially the Arabian Sea sub-basin, where cyclone occurrences are few, assessing and attributing cyclone risks to climate change suffers from a lack of observational data and low confidence in any detectable change. For low-lying coral atolls with dense human habitation, any climate change management and adaptation plan would likely underestimate risks from TC, potentially leading to increased vulnerability in the future. We focus our study on the densely populated Lakshadweep archipelago, situated in the Arabian Sea and address three main gaps. Using renalysed weather variables from ERA5 datasets, we first assess how two main abiotic drivers of cyclonic activity i.e., SST and wind shear have changed since 1940 across the Northern Indian Ocean and relate these with cyclones that have impacted the archipelago. Second, we examine wave heights and rainfall intensity related to TC and assess their joint probability of occurrence across 10 inhabited islands of the Lakshadweep since 1940. Finally, we use a storylines approach to attribute anthropogenic climate change to the occurrence of the most impactful TC in Lakshadweep, compared to a counterfactual scenario of no anthropogenic climate change. Our results highlight that i) seas around the Lakshadweep are becoming conducive for cyclone formation and propagation with an increase in SST and decrease in wind shear; ii) co-occurring extreme wave and rainfall events are associated with cyclonic conditions with larger return-periods (1-500+ year events) compared to their univariate distributions, and iii) based on our storylines approach, intense cyclones in Lakshadweep can overwhelmingly be attributed to anthropogenic climate change, while weaker storms have limited evidence. We conclude by highlighting how climate change mitigation plans in the global south require long-term analysis and attribution studies of rare events to climate change, without which, any plans may underestimate and potentially increase vulnerability of an already vulnerable population.