The “felt temperature” is the preferred measure of hotness or coldness expressed to depict human sensory. However, to date, our perception of its spatial pattern with fine spatiotemporal data remains incomplete. Here, we demonstrated an empirical statistical approach incorporating atmospheric dynamics theory with aerodynamic parameters capable of developing hourly datasets at a high spatial resolution (0.01ᵒ x 0.01ᵒ). This fusion mechanism model, named the Humidex Reconstruction Model based on Numerical Simulation Data (HRMNSD), employed reanalysis data and satellite data for both near surface temperature(Tair) and the dew point temperature(Tdew) to combine their respective advantages in the correct representation of a turbulent exchange between the surface and the atmosphere. We showed the good performance of this model in each season using the Yangtze River Delta, China as an example. The RMSEs of the Humidex were 2.47°C (in winter), 2.49°C (in spring), 2.80°C (in summer) and 2.56°C (in autumn), respectively.