Analysis and prediction of the effects of temperature extremes on the environment heating in the Central Part of Iran, Iran, Southwest Asia

Abstract

Investigation of temperature extremes is very important as one of the most important climate parameters in different parts. The purpose of this study is to analyze and predict the temperature extremes in central Iran. Therefore, the minimum and maximum data of 15 synoptic stations in the study area for the period (1988–2018) using Hybrid Artificial Neural Network (HANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) models were used. Finally, Multi-Criteria Decision-Making (MCDM) models TOSIS and SAW were used to prioritize the areas exposed to rising temperature. The results showed that according to ANFIS modeling for predicting extreme temperatures, the lowest mean training error and the mean error of validation for the minimum temperature was equal 0.10 for Yazd Station and 1.66% for Damghan Station. The lowest mean training error and the mean error of validation for the maximum extreme temperature obtained 0.016 for Garmsar Station and 9.39% for Shahroud station. The maximum extreme temperature of two stations of Garmsar and Bafgh (1 and 0.9689, respectively) were more exposed to extreme temperatures based on the TOPSIS model. Garmsar and Salafchegan Stations (1 and 0.9873, respectively) were more exposed to extreme temperatures based on the SAW model.