A novel statistical-dynamical model, the Columbia HAZard model (CHAZ) is presented for estimating the long-term risk of a rare, high-impact landfall event. The model generates ensembles of synthetic tropical cyclones whose properties depend on the climates in which they occur. The model contains three essential components: a genesis model, a beta-advection track model, and an auto-regressive TC intensity model with deterministic and stochastic components. A model for the structure of the surface wind field is also included, and can be used to drive storm surge models. The genesis, track, and intensity models all are dependent on the large-scale environment through empirically determined relationships. Environmental predictors are chosen carefully to allow the model to handle climate change scenarios. CHAZ, using 400 realizations of a 32-year with environment conditions from ERA-Interim, captures many aspects of TC statistics. Of particular note, it simulates the observed number of rapidly intensifying storms, a challenging issue in tropical cyclone modeling and prediction.

In this talk, I will first introduce CHAZ and present its results from global and basin-wide perspective, including estimates of wind speed return periods and other metrics of landfalling storm hazard. I will then use subsets of synthetic storms associated with areas of interest, such as TX, FL, and Puerto Rico, to investigate the probability of the recent high-impact events. Changes in the probabilities of the occurrences of high-impact storms at these area due to a warming cliamte will be discussed in the end.