2.2. Synthetic TC Dataset
To overcome the limitation of historical data, we estimate TC-induced coastal flood and inundation hazard based on large numbers of synthetic storms. We apply a synthetic TC dataset generated by the deterministic-statistical TC model of Emanuel et al. (2008). This model uses thermodynamic and kinematic statistics of the atmosphere and ocean derived from reanalysis data or climate-model estimation to produce synthetic TCs; it has been widely used to study hurricane wind, storm surge, and rainfall hazards (e.g., Emanuel 2017, Marsooli et al. 2019). Specifically, Xu et al. (2020) applied the synthetic TC model to study typhoon wind hazard for the Shanghai area. We apply the synthetic TC dataset generated in Xu et al. (2020) for the climate over the period of 1979 ~ 2015 based on the National Centers for Environmental Prediction (NCEP) reanalysis (Kalnay et al. 1996). The dataset includes 5018 synthetic storms. Each synthetic storm passes within a 350-km-radius circle centered at a point near Shanghai (latitude 29.86° and longitude 121.56°), and all storms pass within this circle with the maximum wind intensity (1-min wind speed at 10 m above sea level) greater than 21 m/s (40 knots). Each storm is characterized by 2-hourly time series of the storm parameters (i.e., time, center position, maximum wind speed, pressure deficit, and radius of maximum wind) required for storm tide modeling. The annual storm frequency associated with the storm set is about 2 storms per year. Xu et al. (2020) found that the estimated wind hazard based on the NCEP synthetic dataset compared well with historical wind observations in the Shanghai region.