p-factor and r-factor

[GULESER SUFRACI]

Dear Swat-cup users,

I am writing you all to make sure about p and r factor. I have calibrated my watershed and summary stats seems fine. R2=0.71 and NSE=0.71. I just have a confusion about p and r values. P value= 0.63 and R value= 0.48. According to manual (page 18) p factor of 1 and R factor of zero  is good; however, in the group and in several papers I have read r value equal 1 is better. Please take a look attached files and give some feedback. Are my results good enough or not?

I am looking forward to hearing from you.

Thanks,

Guleser Sufraci

Graduate Student

Auburn University

[Iman Fazeli]

Dear Guleser,

You could get p-factor of 1 and r-factor of zero when there isn't any uncertainty while in modeling we have always uncertainties sources. You can increase p-factor and r-factor using increase parameters rangesconsidering your objective function value.

Best,

Iman

--

You received this message because you are subscribed to the Google Groups "SWAT-CUP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to swat-cup+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Iman Fazeli]

Dear Guleser,

You could get p-factor of 1 and r-factor of zero when there isn't any uncertainty while in modeling we have always uncertainties sources. You can increase p-factor and r-factor using increase parameters ranges considering your objective function value.

Best,

Iman

[GULESER SUFRACI]

Hi Iman,

I was wondering that my results are okay or not? These results are from my final calibration and I do not want to make one more iteration. 

[Iman Fazeli]

Your results are acceptable but if I were you try more iterations for better results.

Best,

Iman

[Sushant Mehan]

I do agree with Iman, but make sure the range of best fitted values should remain within acceptable range, as i found in my case with increase in number of iterations, anything can happen, uncertainty may increase or decrease,  objective function sometimes increases or decreases, the acceptable range do become narrower but unacceptable sometimes. So be wise while performing another iteration. Though for me your model is producing good results.

Regards

Sushant 

[GULESER SUFRACI]

Dear Sushant,

After I am done with calibration, I will use the last parameter range for validation right? not the fitted values? I will just use the last iteration parameter ranges for validation, am I right?

[Abbaspour]

Right.

Karim