Tropical cyclones (TCs) cause catastrophic loss in many coastal areas of the world.TC wind hazard maps can play an important role in disaster management.A good representation of local factors reflecting the effects of spatially heterogeneous terrain and land cover is critical to evaluation of TC wind hazard.Very few studies,however,provide global wind hazard assessment results that consider detailed local effects.In this study,the wind fields of historical TCs were simulated with parametric models in which the planetary boundary layer models explicitly integrate local effects at 1 km resolution.The topographic effects for eight wind directions were quantified over four types of terrain (ground,escarpment,ridge,and valley),and the surface roughness lengths were estimated from a global land cover map.The missing TC parameters in the best track datasets were reconstructed with local regression models.Finally,an example of a wind hazard map in the form of wind speeds under a 100-year ret period and corresponding uncertainties was created based on a statistical analysis of reconstructed historical wind fields over seven of the world’s ocean basins.