Calibration on biweekly data

[wdo...@gcp.usgs.gov]

Hello,

I have some biweekly data that I'd like to use for model calibration. Specifically, it is biweekly wetland storage data, with data missing during winter periods.

Is there a way to calibrate the model on the wetland storage data at biweekly timesteps using SWAT-CUP? Or does it have to be monthly data? 

Thank you for your advice.

[omc...@gmail.com]

Hello,

The main issue is that SWAT prints the results at daily, monthly, or yearly time steps, If there was an option to print the results at weekly or biweekly time steps, then it would be possible to use your data without issues. Your only alternative right now is to try to convert your biweekly data to monthly data. Also, as a comment, I am not completely sure if it is possible to calibrate directly ponds in SWAT-CUP. I calibrate reservoirs in SWAT, which is similar to ponds, but I have never seen a case of pond calibration.

Kind regards,

Oscar

[wdo...@gcp.usgs.gov]

Hi Oscar,

That makes sense, thank you! When you calibrate reservoir data, do you do it in SWAT-CUP or a different calibration software?

Thank you,

Wayana

[omc...@gmail.com]

Hi,

I used SWAT-CUP to calibrate reservoirs! In general, SWAT represent very simplistic reservoirs (it was develop for small reservoirs) but it is possible to replicate reservoir behavior accurately on a monthly scale. Now, in your case, SWAT do not create an output file in which it writes specifically pond results, so calibrating it would probably be challenging.

Regards,

Oscar

[wdo...@gcp.usgs.gov]

Hi Oscar,

Gotcha, thank you! My model outputs an output.wtr file that includes total surface water stored within each subbasin in either ponds or wetlands (PNDVOLm^3 or WETVOLm^3 columns in the output.wtr file) at each time step, so I am hoping to be able to calibrate based on that file - fingers crossed I can do it using SWAT-CUP!

Thanks,

Wayana

[omc...@gmail.com]

You are completely right! I just tried with a model that I worked in the past, and the output.pot and output.wtr refers to ponds! However, I was reviewing the SWAT-CUP program, and I do not see that the programs to extract either output.pot or  ouput.wtr are available. So that is going to be a challenge!

I wish you the best, you can contact me through my email omc...@gmail.com by any chance, I really find your project very interesting!

Kind regards,

Oscar

[Karim Abbaspour]

Hi,

One possible approach to calibration is to convert your data to a monthly format. However, this method may result in the loss of important information from your measured data. An alternative approach is to run SWAT on a daily basis, extract daily data using the No_Observation option in SWAT-CUP, and then convert the daily output to a bi-weekly format. You can then compare the bi-weekly data with your measured data. To accomplish this, you will need a program that converts the daily SWAT output to a bi-weekly format.

In my first SWAT application (the Thur Watershed in Switzerland), I also worked with bi-weekly data and applied the method described above. 

It would also be valuable to calibrate using both methods (monthly and bi-weekly) and compare the results. This comparative analysis would make the study more robust and could enhance the quality of the paper.

Best, Karim

--
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 visit https://groups.google.com/d/msgid/swat-cup/9325fb38-f72f-495f-b4b1-f176bf314bb1n%40googlegroups.com.