عنوان مقاله [English]
نویسندگان [English]چکیده [English]
An increase, even moderate, in global temperature is expected to result in a change in frequency of extreme weather events like drought, heavy rainfall and
storms. The study of extreme events is difficult due to the fact that it is difficult to find long-term homogeneous data series. Also the delimitation of extreme events is not univocal since a parameter value that would be defined as an extreme event in one place, might still be considered a normal event in another. In this study, maximum, minimum and mean daily air temperature (Tmax, Tmin, and Tmean) data over a 44 years period (1961-2004) of four synoptic stations of Iran namely; Kerman, Kermanshah, Mashhad, and Shiraz were collected. These stations represent different climates of Iran based on Koppen climatic classification. Required data were obtained from the Islamic Republic of Iran Meteorological Organization (IRIMO). Data were used to calculate extreme temperature values including the magnitudes of the lower (1st, 5th, 10th) and upper (90th, 95th, 99th) percentile threshold values for each year and number of days below the lower threshold values and above the upper threshold values. All time series were checked for normality with the Kolmogorov-Smirnov test. Time trends for all variables were analyzed using parametric and nonparametric techniques (Least squares linear regression, Pearson, Spearman and Kendall's ?-significance test). Kerman (Desert climate) showed significant positive trend in all minimum temperature percentile threshold values except the 95th percentile and a number of upper percentiles (90th and 95th). In maximum temperature, except 1st percentile and the number of days above 90th percentile, other percentiles and the number of days were significant. Results for mean temperature trend were similar to minimum temperature. Mashhad (Temperate humid climate) showed a significant positive trend in all minimum temperature percentile threshold values and the number of days below the 10th, upper 90th and 99th percentile threshold values. Maximum temperature showed a significant positive trend just in upper thresholds and all number of days except the number of days below the 1st and upper 95th percentile threshold values. In mean temperature, results were also similar to the minimum temperature.