Keivan KHalili; Farshad Ahmadi; Yagub Dinpashoh; Ahmad Fakheri Fard
Volume 1, Issue 10 , October 2013, , Pages 1220-1235
One of the most important hydrological time series task is to determine if there is any trend in the data and how to achieve stationarity when there is nonstationarity behavior in data. ...
One of the most important hydrological time series task is to determine if there is any trend in the data and how to achieve stationarity when there is nonstationarity behavior in data. Detecting trend and stationarity in hydrological time series may help us to understand the possible links between hydrological processes and global climate changes. In this study yearly, monthly and daily streamflow data records of Baranduz Chai, Shahar Chai and Nazlu Chai rivers and Urmia synoptic stattion in the west of Lake Urmia, located in the West Azarbaijan of Iran, used to trend and stationarity analysis. Trend analysis with Mann-Kendall and seasonal Kendall tests showed that most annual and monthly flow series had significant negative trend at 1% and 10% . Five common methods named ADF, DFGLS, ERS, KPSS and PP have been used to examine nonstationarity of river flows. Results demonstrated that most annual, monthly and daily series appear to be stationary after removing trend component from series. Also results illustrated, that mean air temperature of this region increased significantly at 1% level. Increasing air temperature causes changing most of precipitations to rain in the replace of snow that maybe the main reason of river flow decreasing and Lake Urmia depletion in recent years. Furthermore, studied rivers have high dependence on snow melt water, therefore are affected by temperature changes. This showed obviously effect of global warming on the decreasing river flow discharges in the west of Lake Urmia