Issue with parameter changes in SWAT+ Rev59.3 projects generated with the SWAT+ Editor 1.2.0 (and higher?)

[Christoph Schürz]

Dear all,

 

I am maintaining the SWATplusR R package and get more and more requests that parameter changes in SWAT+ rev59.3 projects that were set in the run_swatplus() function in R are ineffective. I was able to identify the issue and I want to make clear that the issue is not to be found in the R code, but is an issue of an erroneous line in the file.cio of the SWAT+ project that was generated with the SWAT+ editor. As the number of requests concerning that issue increase, I think it is important to post the issue and the solution (at least how it worked for everyone that contacted me) here.

 

Line 22 in the file.cio controls where to look up the parameter changes defined by the user and where to look up the parameters that can be changed and their boundaries. Apparently, the file names for the lookup are not correctly set when the project is generated.

 

I was able to fix the issue by replacing the line 22 in the file.cio with the following:

 

chg               codes.cal         cal_parms.cal     calibration.cal   ls_parms.cal      ls_regions.cal    null              null              null              null             

 

I hope that the fix above helps other SWAT+ users as well. @ Srini, Jeff, and Nancy I know that I already addressed that issue directly via email. But I think there is no solution implemented yet, other than replacing the line in the file.cio? Please correct me if I am wrong on that matter.

 

Best wishes

Christoph

[Jaclyn Tech]

Everyone, we checked and the correct line 22 in file.cio will need to be:

chg               cal_parms.cal     calibration.cal   codes.sft         wb_parms.sft      water_balance.sft  null              null              null              null

This is compatible with SWAT+ rev. 59 and up. Please note the different order and renaming of some of the files. 

[Christoph Schürz]

Thank you Jaclyn!

I tested both line replacements in a small example with SWAT+ rev59.3.

The line Jaclyn posted works. Mine had no effect. I accidentally posted the wrong line. So please use the one Jaclyn posted!

[belay asmare]

Hey all, How to calibrate and validate SWAT+ Output?

--
You received this message because you are subscribed to the Google Groups "SWAT+" group.
To unsubscribe from this group and stop receiving emails from it, send an email to swatplus+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/swatplus/67636072-4c93-4e57-b6bc-5363ea2087df%40googlegroups.com.

[Christoph Schürz]

Hi Belay,

I think your post is not really related to the issue posted above. But I saw the topic you posted in the google group and planned to respond to that and hopefully give some info that could provide a good starting point to you.

Cheers

Christoph

To unsubscribe from this group and stop receiving emails from it, send an email to swat...@googlegroups.com.

[laurie boithias]

Hi Jaclyn,

Does this apply to the SWATplusR only, or does it apply to any SWAT+ model ?

Thanks,

Laurie

[Christoph Schürz]

Hi Laurie,

it is (or was) specifically an issue with SWAT+. SWATplusR is only an R package that links the SWAT project with the analysis in R. It does not provide any specific SWAT executables. The issue was a result of an erroneous line in the file.cio written by the SWAT editor

[Rafael Navas]

Dear Jaclyn and Christoph,

I replace the line 22 in file.cio. It works for ".hru" and ".sol" files.

Unfortunately, I can't change parameters in ".aqu" and ".rte" files using SWATplusR.

I'm currently using a model written in SWAT+ editor v1.2.3, and running with rev59_3_64rel.exe

Rafael

[Christoph Schürz]

Dear Rafael,

this might be an issue with the current version of SWAT+. I had already several discussions about this matter, that parameter are not modified in a model run, even though the parameters are listed in the cal_parms.cal file and a change is defined in the caliration.cal file. Katrin Bieger Jeff are currently working on it, as far as I know. Katrin and I did several test runs to identify parameters where a change did not have any effect in the model run (see attached figure). We also identified several parameters for rev 59.3 with no effect that you have identified as not changeable. This issue should be fixed in future revisions of SWAT+ (we are currently running these tests for rev60.4). For now, please be very careful with parameter changes and check, if parameter changes do have an effect, before using these parameters in your model calibrations. Otherwise, you might run into totally wrong interpretations of your simulation results.

Best,

Christoph

[MK]

Hi,

I am trying to calibrate my SWAT+ model using SWATplusR. While running the model, I have noticed that the parameters (i.e. cn2.hru , lat_ttime.hru,lat_len.hru, k.sol, z.sol , esco.hru, epco.hru)  are not changing at all. I have made changes in the file file.cio  as suggested in the above conversation. But, I am still having the issue. I am attaching my file.CIO file. I am using SWAT+ editor v1.2.3 & rev59_3_64rel

I would really appreciate any help to address the issue.

Best,

Manas

[Seth Younger]

I'm having the same problem with SWAT 2012 in SWATplusR. Parameters do not change. I see there is a CALFILE argument in the file.cio but there is nothing in the SWAT documentation about how to build one. Has anyone successfully run SWAT2012 calibrations with SWATplusR?

Thanks,

Seth