Skip to Main Content

HBV and HQsim were calibrated and validated using manual calibration under a split-sample approach (Klemeš 1986). The models were both calibrated and validated for the same periods (see Table 2), although it has to be mentioned that for the calibration and validation of HBV a warming-up period (of 1 year) was needed in order to fill the reservoirs in the model. We chose to start the calibration in the late 1990s, because of the low data quality and availability in the 1980s and the beginning of the 1990s, making this period less feasible for calibration. For the VA, it was however not possible to start the calibration in the late 1990s since observed discharge time series were not available for the period 2003–2006. Therefore this period was deemed as unsuitable for the calibration of HBV and HQsim. Further, we chose to have two validation periods in order to identify differences between the model performances that are accompanied with the validation periods.

Table 2

Calibration (Cal. P.) and validation periods (Val. P. I and II) (ÖA = Ötztaler Ache, GA = Gurgler Ache, and VA = Venter Ache)

  HBVHQsim
ÖA Cal. P. 1998–2007 1998–2007 
Val. P. I 1987–1997 1987–1997 
Val. P. II 2008–2012 2008–2012 
GA Cal. P. 1998–2007 1998–2007 
Val. P. I 1987–1997 1987–1997 
Val. P. II 2008–2012 2008–2012 
VA Cal. P. 1993–2002 1993–2002 
Val. P. I 1987–1992 1987–1992 
Val. P. II 2008–2012 2008–2012 
  HBVHQsim
ÖA Cal. P. 1998–2007 1998–2007 
Val. P. I 1987–1997 1987–1997 
Val. P. II 2008–2012 2008–2012 
GA Cal. P. 1998–2007 1998–2007 
Val. P. I 1987–1997 1987–1997 
Val. P. II 2008–2012 2008–2012 
VA Cal. P. 1993–2002 1993–2002 
Val. P. I 1987–1992 1987–1992 
Val. P. II 2008–2012 2008–2012 

Close Modal

or Create an Account

Close Modal
Close Modal