Dear Dr. Abbaspour
I want to calibrate SWAT model using SUFI2 algorithm. My case study area is 850 km2 .this watershed has 3 hydrometric gauges for streamflow measurement. At attachment, showed map of case study and number of gages.I think, there are 2 methods for using 3 gauges in multi-site calibration:
1) At first method, I calibrate upstream subbasins of each hydrometric gauges that donot have overlap with another (upstream subbasins of 2nd and 3rd gauges). Then, I calibrate another subbasins parameters existing at downstream of 2nd and 3rd gauges and mutually exclusive at upstream of 1st gauge, when used from calibrated parameters range of upstream of 2nd and 3rd gauges, simultaneously. The results of the above explained calibration method are 3 sets of any sensitive parameters (equilibrium number of hydrometric gauges).
2) In second method, some weights are assigned to any gauge and so are build an objective function containing goodness of fit (gof) criteria for each hydrometric gauge and its weight. So result of calibration contain 1 set of sensitive parameters relating with average of whole watershed.
Please witch method is correct? If you have any other suggestion please tell me.
Thank you