161 Real-time Global Property Loss Modeling During 2023 Tropical Cyclone Season

Thursday, 9 May 2024
Regency Ballroom (Hyatt Regency Long Beach)
Annes V Haseemkunju, CoreLogic, Inc, Livermore, CA; and J. M. Brolley and D. Smith

The 2023 Tropical Cyclone (TC) season was above normal in the North Atlantic Ocean with 20 storms reaching tropical storm or hurricane intensity. With seven hurricanes and three attaining major hurricanes status (Franklin, Idalia, and Lee), the 2023 season comes fourth in terms of most named storms since 1950 in the Atlantic. Only one hurricane made landfall in the mainland United States, and it was named Hurricane Idalia. Eastern Pacific (EP) basin was not an exception in the overall TC activity with 17 named storms. Out of 10 hurricanes that formed in the EP basin, 8 became major hurricanes. Idalia made landfall near Keaton Beach, western Florida as a Category 3 hurricane on August 30 with a sustained wind speed of 125 mph or 201 kmph. Hurricane Otis made landfall as a Category 5 hurricane with a sustained wind speed of 165 mph or 266 kmph causing catastrophic property damage and death in the metropolitan city of Acapulco, southern Mexico. Otis broke record as the strongest landfalling hurricane of the EP ocean. The season was near normal in the Central Pacific basin with four named storms with no direct hits. The Western North Pacific (WNP) basin exhibited below average TC activity with just 16 named storms of which 10 intensified into typhoon strength (maximum sustained winds of at least of 74 mph (119 kmph). Four of those typhoons intensified to reach super typhoon (maximum sustained winds >=150 mph or 240 kmph) status with most typhoon landfalls in China and none in the Philippines. In the North Indian Ocean exceptional TC activity was noted with eight TCs of which two formed in the Arabian Sea and six in the Bay of Bengal. Super cyclone Mocha made landfall in central Myanmar as a Category 4 hurricane causing damage never seen in decades with an estimated total or partial loss to 200,000 homes. On an average annually about five to six tropical cyclones form in the North Indian Ocean of which four of them forms in the Bay of Bengal and one to two in the Arabian Sea. Arabian Sea cyclones Tej and Biparjoy were major storms of the season and equivalent to Category 3 hurricanes. Significant damages were reported in Bangladesh (Midhili, Hamoon), India (Biparjoy, Michaung), and Yemen (Tej) during this season.

In terms of overall basin activity, the impact of El Nino was not evident in the Atlantic and WNP basins. With above normal sea surface temperature (SST), the effect of El Nino inhibiting intensification of TCs due to high wind shear was negated by the warmer than normal SST in the North Atlantic Ocean. In the EP, Central Pacific (CP), and WNP basin El Nino favors more TCs with above normal activity in terms of genesis and severity. With 16 named storms, the WNP basin produced well below the annual average of 26 named storms. Multiple factors may have played for a below normal activity in the western Pacific during an El Nino year ranging from warm SST anomalies, mid-tropospheric dryness, high wind shear, and low-level ridging to increased anthropogenic aerosols from the southeast Asian region especially from China resulting weakening of the monsoon trough.

Insurance and re-insurance companies widely use catastrophe risk models (CAT models) to estimate property damage from low frequency but high damaging natural perils. One advantage of catastrophe models is their effective use as real-time property damage estimation tools. TC risk models can be used to project insured loss from major landfalling TCs across the globe. Based on the track and intensity forecast from National Meteorological Centers, the model uses a set of analogue events selected out of thousands of simulated tracks and estimate damage to properties and infrastructures affected by extreme wind and flooding. The event selection process identifies footprints of winds, storm surge, and rainfall induced flooding based on landfall characteristics forecasted by National Centers and estimate property damage. In the present study we summarize how TC risk models are utilized to find best TC analogues to project economic and insured loss before and after the landfall of major hurricanes (Idalia and Otis in the Atlantic and Eastern Pacific), and typhoons (Saola, Haikui, and Koinu in Southeast Asia. In the US, overall insured loss from tropical cyclones were moderate with just one major hurricane landfall in a less exposed region in western Florida. Typhoons generally steered away from the Southeast Asian coastlines triggered below normal direct landfall of major typhoons. Exceptions were Typhoon Doksuri with major damage in China and typhoons Saola and Haikui hitting Taiwan, China, and Hong Kong with moderate insured loss.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner