Identification of Trend in Reference Evapotranspiration Series with Serial Dependence in Iran

Monitoring the temporal variations of reference evapotranspiration (ET_o) and quantifying any trends offer valuable information for regional hydrology, agricultural water requirements and water resources management. This study aimed to examine the temporal trends in the Penman–Monteith ET_o in the west and southwest of Iran by using the Kendall and Spearman tests after eliminating the influence of significant lag-1 serial correlation from the ET_o time series. The magnitudes and starting years of significant ET_o trends were determined by the Mann-Kendall rank statistic and the Theil–Sen’s estimator, respectively. For the study period of 1966–2005, a significant positive lag-1 serial correlation coefficient was observed at almost all the stations. The existence of the positive serial correlation in the ET_o series increased the possibility of the Kendall and Spearman tests to reject the null hypothesis of no trend while it is true. It was found that the Kendall test was more sensitive than the Spearman test to the existence of the positive serial correlation in the ET_o series. After removing the serial correlation effect with pre-whitening method, only three significant increasing ET_o trends were obtained at Khorram-Abad, Shahrekord and Zanjan stations at the rates of 0.16, 0.06 and 0.06 mm/day per decade, respectively. The significant increasing ET_o trends of Khorram-Abad, Zanjan and Shahrekord stations started in 1997, 1994 and 1998, respectively. The stepwise regression method showed that wind speed was the most dominating variable affecting on the significant changes of ET_o.