Parameter range for validation

[Long Le]

Dear all,

I am new SWAT-CUP user's.

I have done the calibration for my study area. I used SUFI-2 for my calibration. I am confusing about parameter range for validation.

Should I use the same parameter range as calibration for validation or the new parameter range which obtained after calibration for validation?

Thank you very much.

Long Le.

[izhar shah]

For validation, You Should use the same range of parameters as used for calibration

Regards

[Long Le]

Thank you very much!

On Sun, Jul 14, 2019 at 9:16 PM izhar shah <izhar...@gmail.com> wrote:

For validation, You Should use the same range of parameters as used for calibration

Regards

--
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/swat-cup/CA%2BTGk2iZXjp4vj9knmr_DXZ10cWQ%3D--4c5u8TSwWZZSSC5UeYQ%40mail.gmail.com.

[Bendik Hansen]

Just to be clear: For validation you should use the NEW parameter ranges obtained from calibration. Validating with the raw parameter ranges will do nothing. 

Best regards,

Bendik

On Sunday, July 14, 2019 at 4:24:34 PM UTC+2, Long Le wrote:

Thank you very much!

On Sun, Jul 14, 2019 at 9:16 PM izhar shah <izhar...@gmail.com> wrote:

For validation, You Should use the same range of parameters as used for calibration

Regards

--
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...@googlegroups.com.

[Nilkanth Herbha]

Hello Bendik,

After reading this conversation I am still confused about the validation process. can you please stepwise explain things to do for validation after successful calibration? It helps many people like me.

Thanks.

Regards,

Nilkanth

[Bendik Hansen]

I'm afraid I don't know the details, I haven't gotten to that part of using SWAT yet. Sorry! I know there are some posts on the SWAT google groups about this, so try searching around some more.

I do know that you need to use the calibrated parameter ranges for validation, though. You need to verify that those parameter ranges actually reproduce reality satisfactorily on independent data. 

Best regards,

Bendik

[Nilkanth Herbha]

Hello Bendik,

Thank you for your time.

I have other questions also so I hope please have a look there. I replied to you in another conversation. find it below.

https://mail.google.com/mail/ca/u/0/#inbox/FMfcgxwChcrhfPHWhrSzqtsBLjKXZCQd

[izhar shah]

Dear Bendink,

One should not use the new set of parameters obtained during callibration. For validation, the same set of parameters should be use as was used in callibration.

To unsubscribe from this group and stop receiving emails from it, send an email to swat-cup+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/swat-cup/a1a3c8ad-a07b-4cb1-b548-dfaeaf3580f5%40googlegroups.com.

[izhar shah]

Dear Bendik, 

If you haven't reached to that part of SWAT yet, then why are you misleading someone by answering wrong ?

To unsubscribe from this group and stop receiving emails from it, send an email to swat-cup+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/swat-cup/db09889c-1a57-4fff-998d-5a51439fbffc%40googlegroups.com.

[Bendik Hansen]

Hi izhar,

How does that make sense? When trying to validate a model, you use the calibrated model for the validation period data. Using the ranges USED for calibration yields nothing. The calibrated parameter ranges ARE the model, and they must be validated. 

See page 1494 of this paper for a basic description of validation for modelling in general https://swat.tamu.edu/media/90102/azdezasp.pdf. This paper is even specifically about SWAT, but one doesn't have to be a SWAT expert to know that validating a model is done by using calibrated parameters. 

" The final step is validation for the component of interest (streamflow, sediment yields, etc.). Model validation is the process of demonstrating that a given site-specific model is capable of making sufficiently accurate simulations, although “sufficiently accurate” can vary based on project goals (Refsgaard, 1997). Validation involves running a model using parameters that were determined during the calibration process, and comparing the predictions to observed data not used in the calibration"

Best regards,

Bendik

To view this discussion on the web visit https://groups.google.com/d/msgid/swat-cup/db09889c-1a57-4fff-998d-5a51439fbffc%40googlegroups.com.

[izhar shah]

Dear Bendik,

Validation means to check your model output for different time period by using the same set of parameters used for callibration.

Screenshot attached for your kind information..... verified by Dr. Karim Abbaspour sir.

To unsubscribe from this group and stop receiving emails from it, send an email to swat-cup+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/swat-cup/679a2924-9a50-41d8-b306-30d3599603b8%40googlegroups.com.

[izhar shah]

Dear Bendik,

If you callibrate a model for a specific set of parameters and then validate it with diffrent set of parameters, then which fitted values will you put in your model to get fully callibrated output ? 

On Thu, 18 Jul 2019, 12:18 am Bendik Hansen, <marek...@gmail.com> wrote:

To unsubscribe from this group and stop receiving emails from it, send an email to swat-cup+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/swat-cup/679a2924-9a50-41d8-b306-30d3599603b8%40googlegroups.com.

[Bendik Hansen]

This was just a communication problem, then. When I said "calibrated parameter ranges" I meant the ranges that yielded acceptable model results (the input to the last iteration). I thought you were talking about the initial input parameter ranges (i.e. when starting calibration...), which would be nonsense. I can see why you would think I meant otherwise though, I didn't phrase it that well. 

Best regards,

Bendik

To view this discussion on the web visit https://groups.google.com/d/msgid/swat-cup/679a2924-9a50-41d8-b306-30d3599603b8%40googlegroups.com.