Understanding Shifts in Landfall Risk in an Anomalously Warm Atlantic Climate

Catastrophe (“CAT”) risk models were first introduced in the early 1990s, and since then have become a fundamental tool used by (re)insurers and other risk managers to assess and quantify hurricane risk. While most models’ baseline risk is defined by the statistics of observed tropical cyclones (TCs) contained in the historical record, questions around climate change and natural climate variability are incentivizing more robust and innovative solutions.

For example, most CAT modelers would agree that warmer-than-average Atlantic sea-surface temperatures observed since 1995 have increased the frequency and/or severity of landfalling hurricanes in the U.S. However, accurately quantifying the magnitude of regional shifts in landfall risk across the full intensity distribution is a challenge. Furthermore, assigning an objective level of confidence to such shifts provides important guidance, given the array of natural climate oscillations that influence Atlantic tropical activity each season.

This study updates and extends earlier related work to examine risk across the full TC life-cycle from genesis to landfall and dissipation. Climate-induced changes in genesis density across the Atlantic are shown to be non-uniform, which has implications for regional landfall frequency along the U.S. coastline. Tracking patterns from genesis to landfall under warm SST conditions reveal a possible connection between the strength and position of the subtropical high (“Bermuda-Azores High”) and shifts in tracking, especially in the Caribbean passage between the open Atlantic and the Gulf of Mexico. These findings, combined with shifts in genesis, also have important implications on regional landfall risk. Finally, landfall intensity is considered in light of observed increases in rapid intensification cycles (RICs), and in turn, the relationship of RICs to where hurricanes that experience them are most prone to making landfall.

Other aspects of the global climate – including the El Nino Southern Oscillation (ENSO) – are examined as well, noting their influence relative to a locally warmer Atlantic climate, from event-level impacts over days to weeks to seasonal impacts over months to years. The study addresses how these and other factors can be appropriately considered as part of a holistic view of hurricane risk.

The study broadens our understanding of the key components of hurricane risk, enabling decision-makers to refine their risk estimates while reinforcing statistical results with physics-based analysis. The study also extends research that has framed ‘climate-conditioned’ CAT models, allowing for a more robust treatment of climate change in a warming world.