"Post-monsoon Bay of Bengal cyclones are increasingly marked by stalling, rapid intensification and system mergers. Examine the implications for India's early warning architecture." (Cyclone Arnab can serve as the example.)

India's "zero-casualty" record rests on accurate forecasts of where and when a cyclone will make landfall. Cyclone Arnab (October 2026) stalled near the coast, interacted with a nearby low-pressure system, and was followed by forecasts of a second storm. It shows that post-monsoon cyclones now strain a warning system built for one predictable storm track.

Emerging traits: the Arnab case

• Stalling: Arnab lingered off the Andhra coast, raising fears of extreme rainfall and flooding [1].
• System interaction: wind data showed Arnab interacting with a separate low-pressure area, and analysts flagged possible rapid intensification [2].
• Clustering: weather models showed another storm possibly heading towards Odisha soon after [3].

Implications for the early warning architecture

• Forecasting: a single landfall time fails when storms stall or merge, so ensemble (probabilistic) forecasts become essential. IMD's new Decision Support System already displays a cone of uncertainty [4].
• Hazard shifts from wind to water: a stalled storm rains for longer. Arnab's remnants threatened heavy rain as far as northern India and Nepal [2], so inland districts need warnings as much as the coast does.
• Communication: bulletins must convey both uncertainty and likely impacts. This builds on IMD's district-level impact-based warnings, introduced in 2020 [5].
• Last mile: stalls keep evacuees in shelters for longer. About 58,000 people were evacuated in Andhra Pradesh [6]. Back-to-back storms strain the shelters and warning networks built under NCRMP [7].
• Livelihood blind spot: lives were saved, but about 70,900 ha of paddy were damaged during the kharif harvest [6]. Warnings do not yet trigger advisories to protect crops.
• Regional stakes: as RSMC New Delhi, IMD issues advisories to 13 member countries [8], so forecast errors affect the whole region.

Way forward

• Make ensemble guidance and the naming of nearby interacting systems a standard part of bulletins.
• Expand geo-targeted alerts through NDMA's SACHET platform [9].
• Link warnings to anticipatory farm advisories and to power shut-down protocols for electricity distribution companies.
• Plan for multi-storm seasons under the Disaster Management Act, 2005, keeping relief stocks ready.

Arnab shows that stalling, merging and clustering storms demand early warning that is probabilistic, impact-based and inland-aware, not only accurate about landfall. Making this shift would protect livelihoods as well as lives. It would strengthen India's role as a regional forecaster and advance the UN's Early Warnings for All goal and the Sendai Framework.