Analysis of stream flow trend across Karkheh watershed and effect of autocorrelation coefficient on the trend of flow

Trend analysis conducted on stream-flow measured at hydrometric stations across the central part of Karkheh River watershed in monthly, seasonal and annual time scales using statistical methods. The data used were the monthly discharge time series of 11 selected hydrometric stations over a 40 year period (1969 to 2009). The slope of the trend line of discharge time series was estimated using the Theil-Sen approach (TSA) and the effect of the significant autocorrelation coefficient was removed using trend-free pre-whitening (TFPW), pre-whitening (PW) and variance correction approach (VCA) methods, and the discharge time series were pre-whitened. Then, the trend of original and pre-whitened data were analyzed using the Mann–Kendall (MK) test. The ability of TFPW, PW and VCA methods for removing the effect of the autocorrelation coefficient were evaluated. The results showed that the TFPW is the best method for removing the effect of serial correlation from discharge time series. So, trend analysis of monthly, seasonal and annual discharge time series was conducted by using the MK-TFPW method. The results show that at annual time scale, more than 70 percentages of considered stations have experienced a significant decreasing trend at the 5% level.Trend analysis conducted on stream-flow measured at hydrometric stations across the central part of Karkheh River watershed in monthly, seasonal and annual time scales using statistical methods. The data used were the monthly discharge time series of 11 selected hydrometric stations over a 40 year period (1969 to 2009). The slope of the trend line of discharge time series was estimated using the Theil-Sen approach (TSA) and the effect of the significant autocorrelation coefficient was removed using trend-free pre-whitening (TFPW), pre-whitening (PW) and variance correction approach (VCA) methods, and the discharge time series were pre-whitened. Then, the trend of original and pre-whitened data were analyzed using the Mann–Kendall (MK) test. The ability of TFPW, PW and VCA methods for removing the effect of the autocorrelation coefficient were evaluated. The results showed that the TFPW is the best method for removing the effect of serial correlation from discharge time series. So, trend analysis of monthly, seasonal and annual discharge time series was conducted by using the MK-TFPW method. The results show that at annual time scale, more than 70 percentages of considered stations have experienced a significant decreasing trend at the 5% level.Trend analysis conducted on stream-flow measured at hydrometric stations across the central part of Karkheh River watershed in monthly, seasonal and annual time scales using statistical methods. The data used were the monthly discharge time series of 11 selected hydrometric stations over a 40 year period (1969 to 2009). The slope of the trend line of discharge time series was estimated using the Theil-Sen approach (TSA) and the effect of the significant autocorrelation coefficient was removed using trend-free pre-whitening (TFPW), pre-whitening (PW) and variance correction approach (VCA) methods, and the discharge time series were pre-whitened. Then, the trend of original and pre-whitened data were analyzed using the Mann–Kendall (MK) test. The ability of TFPW, PW and VCA methods for removing the effect of the autocorrelation coefficient were evaluated. The results showed that the TFPW is the best method for removing the effect of serial correlation from discharge time series. So, trend analysis of monthly, seasonal and annual discharge time series was conducted by using the MK-TFPW method. The results show that at annual time scale, more than 70 percentages of considered stations have experienced a significant decreasing trend at the 5% level.Trend analysis conducted on stream-flow measured at hydrometric stations across the central part of Karkheh River watershed in monthly, seasonal and annual time scales using statistical methods. The data used were the monthly discharge time series of 11 selected hydrometric stations over a 40 year period (1969 to 2009). The slope of the trend line of discharge time series was estimated using the Theil-Sen approach (TSA) and the effect of the significant autocorrelation coefficient was removed using trend-free pre-whitening (TFPW), pre-whitening (PW) and variance correction approach (VCA) methods, and the discharge time series were pre-whitened. Then, the trend of original and pre-whitened data were analyzed using the Mann–Kendall (MK) test. The ability of TFPW, PW and VCA methods for removing the effect of the autocorrelation coefficient were evaluated. The results showed that the TFPW is the best method for removing the effect of serial correlation from discharge time series. So, trend analysis of monthly, seasonal and annual discharge time series was conducted by using the MK-TFPW method. The results show that at annual time scale, more than 70 percentages of considered stations have experienced a significant decreasing trend at the 5% level.