Published November 2026.

Summary

Tropical Cyclone Arnab was one the most powerful tropical cyclones near India in several decades. For this rapid study the IRIS model estimates that climate change increased the maximum wind speed by 5%, eyewall rain by 11% and economic damage by 33%. About 25% of the economic damage can be attributed to climate change.

  Maximum wind speed Eyewall rainfall Economic damage
+1.3°C (2025) +5% +11% +33%
+2.6°C +8% +22% +56%

Table 1. Changes of maximum wind speed and  eyewall rainfall rate at landfall and  economic damage relative to the pre-industrial baseline for Tropical Cyclone Arnab.

Background

Tropical Cyclone Arnab was Category 4 with the US agency reporting a life-time maximum wind speed of 220 km/h in the Bay of Bengal. Its wind and rainfall caused loss of life and extensive damage in East India. The IRIS model (Sparks and Toumi, 2024) is used to infer the additional strengthening of this type of storm that can be attributed to recent warming or more specifically to changes in potential intensity (PI). For the warming scenarios the observed changes in potential intensity are scaled by the global mean surface temperature as described by Sparks and Toumi (2025). 

Figure 1. Zonal mean potential intensity, PI, difference for October between now (“2026”, +1.3°C) and pre-industrial.

Maximum Wind Speed

Figure 1 shows the zonal mean difference in October PI between 2026 and the pre-industrial estimate. The PI difference between 2026 and pre-industrial is about +18 km/h in the Bay of Bengal. The frequency of maximum wind speed at landfall is the next consideration. The IRIS model calculates the intensity along the observed storm tracks for a 10,000 year simulation.

Figure 2 shows the model return period plot for wind speeds at landfall. Arnab was reported by the US Agency to have reached a lifetime maximum intensity of Category 4 (220 km/h)  and was a Category 3 (202 km/h) at landfall. We estimate that this type of event is about 19% more likely compared to pre-industrial times. For the same return period, the current wind speed compared to pre-industrial events has increased by 9.3 km/h or +5 %. In a +2.6°C warmer world, the wind speed increases by a further +5.5 km/h or +8% compared to a pre-industrial climate.

Figure 2. The wind speed vs return period as calculated for the current climate “2026” (+1.3°C, orange line), pre-industrial (blue line) and +2.6°C climate (red line). FAR is the Fractional Attributable Risk (Equation 1) with a maximum wind speed of 202 km/h in the current climate. ΔVmax is the change in Vmax between pre-industrial and current climates.

In the climate change attribution literature the fractional attributable risk, FAR, is frequently used. FAR is here defined as 

FAR = (Pnow - PPre-Ind) / Pnow  (1)

where Pnow and PPre-Ind  are the probabilities of an event of the minimum intensity for the current (now) and pre-industrial (Pre-Ind) climate respectively. The FAR for “Arnab ” type is 0.16. This means that at least 16% of the likelihood of this type of event can be attributed to climate change.

Eyewall Rain Rate

Figure 3. Eyewall (maximum azimuthal mean) rain rate for tropical cyclone at landfall vs return period as calculated for the current climate “2026” (+1.3°C, orange line), pre-industrial (blue line) and +2.6°C climate (red line). FAR is the Fractional Attributable Risk (Equation 1) of the maximum azimuthal mean rain rate. Δ shows the increase in rain rate compared to pre-industrial “Arnab” type events.

We also calculate the change of the eyewall rain rate as described by (Lau et al., 2025). The observed value of 11.3 mm/h is calculated as the maximum azimuthal mean rain rate at landfall from IMERG. Figure 3 shows the model return period of eyewall rain rate (maximum azimuthal mean) for the region. We estimate that this type of event was about 18% more likely compared to pre-industrial times. The return period has decreased from 16.7 yrs to 14.2 yrs. For the same return period the current rain rate compared to pre-industrial events has increased by 1.1 mm/h or 11%. In a +2.6°C degree warmer world, the eyewall rain rate increases by a further 1.2 mm/h, which amounts to a 22% increase compared to pre-industrial events. The FAR for “Arnab” type rain rate is 0.15. This means that at least 15% of the likelihood of this type of event can be attributed to climate change.

Attributing Economic Damage

We make an estimate of direct economic damage to physical assets. Wind is used as a proxy for tropical cyclone hazard and the damage function indirectly accounts for the other perils such as storm surge and flood. We combine reported wind fields with a damage function and gridded exposure. To estimate the damage uncertainty, we simulate thousands of damages following the track by varying the damage function with a half-damage wind speed (range of 70 - 110 m/s), wind speed (+/-10%), and the exposure (+/- 25%).

Figure 4. The track, wind speed (km/h) and mean economic damages (US$B). Assumed total asset value of India in 2026 is US$ $2,528B (World Bank Wealth Accounts).

Figure 4 shows the wind footprint and the mean economic damages estimated. We then scale the land wind speeds fractional differences for pre-industrial conditions and for a +2.6 °C global warming (see Figure 2). The pre-industrial and +2.6°C simulations are for counterfactual scenarios assuming constant current exposure and vulnerability, only the wind speed is different. We find that the wind speed increase due to climate change causes substantially more damage. To communicate the effect of climate change on loss we define the fractional attributable loss, FAL:

FAL = (Lnow-LPre-Ind)/Lnow (2)

where L is the economic loss for the current (now) and pre-industrial climate (Pre-Ind). For the mean damage, we estimate the FAL is 0.25. This means about 25% of the economic damage can be attributed to climate change compared to the pre-industrial baseline. The damage was increased by +33%. In a +2.6°C warmer world, the damage would increase by 56% compared to preindustrial as shown in Figure 5. 

Figure 5. Estimated mean economic damages (US$M) for pre-industrial, current and +2.6°C global warming scenario. Uncertainty of damages (1��) is based on uncertain damage functions (half damage wind speeds), exposure value and wind speed.

Lau, K.H, Czernichow, S., Sparks, N., Toumi, R.,. A global parametric rain model for landfalling tropical cyclones: a case study for the U.S.. Nat Hazards 122, 415 (2026). https://doi.org/10.1007/s11069-026-08150-5 

Sparks, N., Toumi, R. The Imperial College Storm Model (IRIS) Dataset. Sci Data 11, 424 . https://doi.org/10.1038/s41597-024-03250-y, 2024.

Sparks, N., Toumi , R., The impact of global warming on U.S. hurricane landfall: a storyline approach. Environ. Res. Lett. 20 114006 https://doi.org/10.1088/1748-9326/ae0956, 2025.