Climate change refers to the changes in the mean or variations of the climatic characteristics wich will exist for a long period of time. These changes are both of natural and man-made causes. In this study, we used the data on minimum temperature and the atmospheric general circulation models in order to simulate the minimum temperature variations. We also used correlation coefficients and root mean square methods to evaluate the performance of these models. In this study, the years between 1961-1990 were used as the base period to study the changes in the minimum temperature in the future decadal periods of 2011-2040, 2041-2070, and 2071-2099. The results showed that in the coming decades the minimum temperatures tend to decrease. The estimated simulated minimum temperatures using SDSM and artificial neural network model differ some 1.8 and 2.3 C˚ in January compared to the data observed. According to estimates of the two models in the years 2011-2040 in January temperature of 3.3 C˚ temperature rise for years by 2041 to 2070 an estimated 4.7 and for the period 2071 to 2099 will increase by about 5.05 C˚. This study showed that the results of the SDSM model were closer to real values than those of the neural networks models.

Keywords: climate change, artificial neural network model, SDSM model, minimum temperature, urmia.

Full-Text [PDF 1362 kb]   (2495 Downloads)