摘要
Recently, the Urmia Lake located in northwestern Iran which is the second largest hyper saline in the world suffers from the significant fluctuations of water level and surface area. The current study tries to investigate the spatiotemporal trends of mean (Tmean), maximum (Tmax) and minimum (Tmin) temperatures of monthly, seasonal and annual time-series. To do so, the data of 15 temperature gauge stations within the Urmia Lake basin, for the period 1972–2011 was employed. The pre-whitening approach was applied to remove the effects of serial correlation in the air temperature series based on the Mann-Kendall (MK) test. The results of Ljung-Box test showed positive serial correlation in the Tmean and Tmax series for all of the stations at the 0.05 significance level. In the monthly series, the significant warming trends in the Tmean series were more perceptible than the same ones in Tmax series; however, Tmax trend was found more than Tmin series. The Mann–Whitney (MW) test detected a significance upward shift changes in the annual Tmean, Tmax and Tmin series of about 86, 73 and 80 % of the stations, respectively. The average magnitude of significant warming trends in annual Tmean, Tmax and Tmin series were ( ) 0.58 °C, ( ) 0.52 °C and ( ) 0.69 °C per decade, respectively. Furthermore, the interpolation maps showed that warming trends in the east and west of Urmia Lake were more than southern area. Therefore, the results showed that the basin has suffered from increasing trends in the Tmean, Tmax and Tmin over the recent decades. Finally, significant changes were found in 1980s and 1990s based on the Mann-Kendall ranks and change point tests. In this study, it is interesting that the period of significant changes in warming trends were close to the beginning of decreasing water level of the Lake. Keywords Climate variability Trend Serial correlation Urmia Lake Temperature series