Spatial and Temporal Variability of Temperature in Iran for the Twenty-First Century Foreseen by the CMIP5 GCM Models

Abstract

The changes and fluctuations in climate variables, especially temperature, that subsequently affect human activities and natural environments are one of the critical topics in scientific societies. Therefore, time variability analysis of average temperature is an important concept in climate studies, particularly in environmental planning and management at various levels. The main purpose of the present study is to detect the temporal and spatial variability of average monthly temperature in Iran during the period 2015–2060 based on the RCP4.5 and RCP 8.5 scenarios simulated by the CMIP5 atmospheric general circulation models. For this purpose, the monthly average temperature data relative to four CMIP5 models, including CMCC-CM, CESMI-BGC, CCSM4, and MRI-CGCM3 models, for the period 1987–2060, and the observed monthly average temperature for the period 1987–2014 measured at 88 synoptic stations distributed all over Iran were used. The accuracy of the CMIP5 models in simulating the historical data for the period 1987–2005 was evaluated against the observed data at the synoptic stations using R, R2, RMSE, bias, EF, NARMSE, slope, and IA statistics. To study the temporal and spatial variation of temperature relative to the historical and future periods across Iran, the statistical spatial variance model was applied. The result showed that the historical temperatures estimated by all CMIP5 models are highly correlated with the observed temperatures all over Iran, but the accuracy of the MRI-CGCM3 model was found slightly higher than the other three models in most areas of Iran. In general, the results showed that the temperature estimated by the selected models and scenarios have a very high correlation with the observed data in most parts of Iran, especially in the mountainous areas of the western country. In the coastal areas of southern and northern Iran, the accuracies of the models somehow decreased which can be attributed to the complex topographical structure of these areas and/or the other effective local features that have not been incorporated in the models. The results showed that the highest temporal variability of temperature has occurred in winter months and partly in the autumn, and the spatial variability of temperature has been observed primarily in the mountainous areas of Iran. The investigation of the temperature variability in the future decades (2015–2059) was found in parallel with the temporal variations of temperature in the present period, considering that the highest temperature variability will occur in winter and somehow in the autumn, mostly in the mountainous areas of Iran. Generally, in most parts of the country, the air temperature in future decades will have an increasing tendency in all four seasons of the year.