Manual calibration in SWAT+

792 views
Skip to first unread message

NMC 96

unread,
Apr 24, 2021, 1:12:56 PM4/24/21
to SWAT+
Dear all,

I am trying to calibrate SWAT+ using monthly discharge, however, the simulated discharge is much higher than the observed.

Which parameters can I change in order to decrease substantially the discharge?

Thank you in advance.

Capture.PNG

Celray James

unread,
Apr 24, 2021, 4:18:25 PM4/24/21
to NMC 96, SWAT+

Are you using channel_sd_mon.txt or channel_sdmorph_mon.txt to extract the flows? And what revision are you running?

 

Celray James


--
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/97eebf73-7157-4ba5-9a80-2521631aa970n%40googlegroups.com.

 

NMC 96

unread,
Apr 24, 2021, 5:00:12 PM4/24/21
to SWAT+
Hi James,

I am using channel_sd_mon.txt and the revision of 59.3 of SWAT+. I also tried to calibrate the model using SWAT+ Toolbox v0.4.5 but it did not improve.

celray.chawanda

unread,
Apr 25, 2021, 6:04:52 AM4/25/21
to NMC 96, SWAT+
1. Try SWAT+ Toolbox 0.7.6
2. You must NOT use sd files from 59.3, flows from that file are not correct for monthly.
3. Use revision 60.5 if you want monthly, and the file should be channel_sdmorph_mon.txt.

Hope this helps. let me know how it goes.

James

NMC 96

unread,
Apr 25, 2021, 12:58:55 PM4/25/21
to SWAT+
Hi James, 

I downloaded the updated version for SWAT+ and SWAT+ Toolbox. The results are better, however, my discharge flow is much lower than the observed, and also the ET is underestimated. 

Toolbox says that it is caused because the ratio of 'Surface runoff' to 'Total flow' is too high, causing that surface runoff is excessive... How can I decrease this ratio? I don't see the  'surq' or 'srr' variables in Toolbox...

Thank you for your help.

Celray James

unread,
Apr 25, 2021, 1:20:00 PM4/25/21
to NMC 96, SWAT+

This is happening because of 2 possible reasons.

  1. Calibration: - you need to adjust parameters that affect surface runoff such as curve number. By, calibrating your model, you improve its representation of reality and hence its performance. However, sometimes, model calibration does not fix your model if it has structural issues.
  2. Model structure or bad data: - if the model has problems in data or configuration, this may happen. You may know this if the model is way off and is ‘impossible to calibrate’ because if the problem is with data of model configuration, no amount of calibration will fix it.

 

Concerning the warning SWAT+ Toolbox gave you, that is just to let you know where your model is producing potentially problematic results. If your catchment produces high surface runoff, you can ignore the warning. Also, there is no ratio based calibration in SWAT+ Toolbox. Soft calibration is scheduled for version SWAT+ Toolbox v1.0 and is still in the works.

 

           With Kind Regards,

           Celray James CHAWANDA

--
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/97eebf73-7157-4ba5-9a80-2521631aa970n%40googlegroups.com.

 

--
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.

--
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.

NMC 96

unread,
Apr 27, 2021, 4:34:27 AM4/27/21
to SWAT+
Hi James,

Thank you, I am calibrating now with the new update... 

However, I have another doubt... when you calibrate using monthly discharge, do you take the monthly average or the sum of the discharge?

Best regards,

Nuria

Rishabh Srikar

unread,
Aug 12, 2024, 10:58:38 AM8/12/24
to SWAT+
Hi, I was facing the same issues so although this thread seems to be dead, I'll reignite it by answering what I can. 

Nuria: All flow variables have their units in m3/sec which means that their values (daily,  monthly, yearly) need to be averages. Summing a rate (discharge per time) makes no physical sense. However, water balance variables are usually representative of the mass balance equation and therefore have their units in mm (depth in a certain area) and these can be both averaged and summed. As most of us have observed data in the form of rate of discharge, the former applies to your situation. 

I too have been observing high simulations of discharge using the 59.3 Rev. I wished to know whether anyone can confirm whether this issue is arising out of the version rather than a calibration/bad data/model-setup.
Reply all
Reply to author
Forward
0 new messages