Re: HOW TO VALIDATE A MODEL IN SWATCUP(SUFI-2)

[Jochen]

Hi,

basicallly you have to change the following files:

File.cio --> change your number of years, your starting and your end year (and have a look at the Julian Day of the last year, could be a leap year).
observed.txt and observed.rch --> add in new observed data (flow, nutritients, whatever)
SUFI2_extract_rch.def --> Change your start / ending years

And then use the best parameter ranges given in your last and best simulation from the calibration run and just do another iteration for the validation period.

Best
Jochen

[Maulana Ibrahim Rau]

Thank you Mr. Jochen for your information.

I was just reading your post, and I want to ask several questions related to SWAT-CUP Validation:

1. Firstly, is it necessary to only change files/information in  File.cio, observed.txt, and observed.rch? Because as I know simulation data in the validation process differs from the calibration. If we put the process as you mentioned (without inputting any simulation data for the validation, e.g.: flow out), might the software will not be able to read the "raw" simulation data for validation?

2. Secondly, concerning to the validation process, as we know we have to put the best calibration ranges, where consequently SWAT-CUP will do the same iteration for the validation. But what I want to ask is, it is also necessary? Because, when SWAT-CUP do another iteration, even though it has the same ranges, the fitted value will not be the same. And since it is not the same input value for both calibration and validation, is it valid?

Thank you very much for your kind attention, I am looking forward for your valuable information.

Regards,

Maulana Ibrahim Rau

[Karim Abbaspour]

1. Firstly, is it necessary to only change files/information in  File.cio, observed.txt, and observed.rch? Because as I know simulation data in the validation process differs from the calibration. If we put the process as you mentioned (without inputting any simulation data for the validation, e.g.: flow out), might the software will not be able to read the "raw" simulation data for validation?

Please note that your calibrated parameters are the ranges you obtained in the last iteration (or any iteration you select) of the calibration.
For validation you want to propagate these ranges (using as many simulations as you did for calibration), get the 95ppu and compare it with your observed data of the validation period. That is all.
So, it is clear which files you need to change to achieve this. The model will then make the simulations, calculate the 95PPU, and compare with observed values, and give you the statistics.
I am not sure what "raw" simulation is?

2. Secondly, concerning to the validation process, as we know we have to put the best calibration ranges, where consequently SWAT-CUP will do the same iteration for the validation. But what I want to ask is, it is also necessary? Because, when SWAT-CUP do another iteration, even though it has the same ranges, the fitted value will not be the same. And since it is not the same input value for both calibration and validation, is it valid?

In the validation there is no fitted values. It just propagates your calibrated parameter ranges.
If you do a large enough number of simulations, then the slight changes in the exact parameter values derived from LH sampling will not matter.
If you want to use the exact parameter values, you can copy the par_val.txt from your last calibration run into SUFI2.IN when you do validation and not run the pre-possessing step. If you do just a few runs, then must do copy the same parameter set.

Hope these are clear...


 

-------------------------------------------------
Dr. K.C. Abbaspour
Eawag, Swiss Federal Institute for Aquatic Science and Technology
Ueberlandstr. 133, P.O. Box 611, 8600 Duebendorf, Switzerland
email: abba...@eawag.ch
phone: +41 44 823 5359
fax: +41 44 823 5375
http://www.eawag.ch/index_EN

---

From: Maulana Ibrahim Rau <maulan...@gmail.com>
To: swat...@googlegroups.com
Cc: arc...@googlegroups.com; SWAT-user <swat...@googlegroups.com>; Cole Green Rossi <cole....@ars.usda.gov>; Georgie Mitchell <georgie....@ars.usda.gov>; Jeff Arnold <jeff....@ars.usda.gov>; Mauro Di Luzio <mdil...@brc.tamus.edu>; Mike White <mike....@ars.usda.gov>; Nancy Sammons <nancy....@ars.usda.gov>; Srini <r-srin...@tamu.edu>
Sent: Tuesday, April 23, 2013 11:17 AM
Subject: Re: HOW TO VALIDATE A MODEL IN SWATCUP(SUFI-2)

--
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/groups/opt_out.
 
 

[Maulana Ibrahim Rau]

Dear Dr. Karim,

Thank you very much for your kind attention. It is very nice of you for giving me with such brief explanation.

Personally, your information is completely clear to me.

I hope you do not mind to be contacted if in the near future I still have enquiries. 

Thank you for sharing your valuable experiences in this email group. 

Best Regards,

Maulana Ibrahim Rau

[Manox Sandy]

DEAR FRIENDS AND DR. KARIM,

       I AM SIMULATING SEDIMENT FROM 1988-2017. 

       I ONLY HAVE MEASURED SEDIMENT FROM 2016-2017. SO THAT MEANS I CAN ONLY CALIBRATE DURING THIS PERIOD. HOW DO I DO VALIDATION SINCE I ONLY HAVE MEASURED                         SEDIMENT FOR TWO YEARS? WILL REALLY APPRECIATE YOUR QUICK RESPONSES. 

REGARDS!

MANOAH