资源环境科技发展态势分析平台

High-resolution climate model simulations and a tropical cyclone damage model are used to simulate the economic damage due to tropical cyclones. The damage model produces reasonable damage estimates compared to observations. The climate model produces realistically intense tropical cyclones over a historical simulation, with significant basin scale correlation of the inter-annual variability of cyclone numbers to observed storm numbers. However, the climate model produces too many moderate tropical cyclones, particularly in the N. Pacific. Annual mean cyclone damage with simulated storms is similar to estimates with the damage model and observed storms, and with actual economic losses. Ensembles of future simulations with different mitigation scenarios and different sea surface temperatures (SSTs), as well as societal changes, are used to assess future projections of cyclone damage. Damage estimates are highly dependent on the internal variability of the coupled system. Using different ensemble members or different SSTs affects damage results by +/- 40 %. Experiments indicate that despite decreases in storm numbers in the future, strong landfalling storms increase in E. Asia, increasing global storm damage by similar to 50 % in 2070 over 2015. Little significant benefit is seen from mitigation, but only one ensemble is available. Projected increases in vulnerable assets increase damage from simulated storms by more than threefold (similar to 300 %, assuming no adaptation) indicating future growth will swamp potential changes in tropical cyclones.