A comparison of physiologically equivalent temperature based on model simulation and site observation data for Iran

Climate change will have a great impact on thermal bioclimate conditions and thermal perceptions, and statistical downscaling methods may introduce uncertainties to climate projection which should be considered in future planning. Therefore, the present study aims to evaluate the statistical downscaling method of Cumulative Distribution Function–transform (CDF-t) in simulating the thermal comfort characteristics of Iran based on the physiologically equivalent temperature index (PET). The daily maximum and minimum temperature, wind speed, and relative humidity of ERA5 climate reanalysis as well as the observational data of 91 synoptic stations during 30-year period (1987–2017) were used. The model was validated using Pearson correlation, Kolmogorov–Smirnov test, climatic RMSE, and Moran’s I statistic. RMSE values (2–4.9) indicated a slight deviation of the simulated series compared to the observed PET of the stations. Correlation values of 0.6–1 indicate a linear and positive relationship between the comfort series. In addition, adherence to the same statistical distribution between the mentioned series was confirmed by Kolmogorov–Smirnov test. The results showed that the observed and simulated PET values were consistent in the nine climatic regions of Iran except for the temperate regions of dry and hot summers (Csb) as well as snowy climate with dry and hot summers (Dsb). Moran’s statistics at a 95% confidence level showed that this method is suitable for simulating the data without making any unexpected changes in the geographical distribution of it. In general, the CDF-t has a good performance in downscaling of the above-mentioned variables in this study area for future.