SWAT Model for future and baseline scenario

656 views
Skip to first unread message

Saima Nauman

unread,
Jun 28, 2018, 7:54:45 AM6/28/18
to SWAT-CUP
Hi,

I have calibrated and Validated my SWAT-CUP project. Now i want to use the calibrated model results to run baseline and future scenarios in SWAT. I have made changes to the parameters in SWAT folder and run the model for my historical period but the flow-out results for the calibrated model are coming very different to the simulated flow results in SWAT CUP and there is a huge difference between the observed data and calibrated model data. How to check the comparison between observed data and calibrated model discharge data in this case?

Regards,
Saima

Saima Nauman

unread,
Jun 28, 2018, 8:06:01 AM6/28/18
to SWAT-CUP
is there any need to rerun the SWAT Model with calibrated parameters using same meteorological(observed dataset) after successful calibration in SWAT-CUP?

Karim Abbaspour

unread,
Jun 28, 2018, 11:15:11 AM6/28/18
to swat...@googlegroups.com

The swatcup folder you calibrated your model in contains your calibrated swat model. What does it mean you made changes to the parameters in the swat folder? You have calibrated parameters ranges, not just single valued parameters. To do climate change analysis, create the pcp and tmp files in ArcSwat and copy it in the calibrated swatcup folder and do an “iteration” with the “parameter ranges”. If you change any other swat files in ArcSwat (say management or other files), you can copy those in the swatcup folder and again do an iteration to get your results, which include model uncertainty as well.

Hope this is clear.


Karim

Ps, if you get different swat results, it could be because of different swat versions.

--
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/d/optout.

Saima Nauman

unread,
Jun 29, 2018, 3:12:37 AM6/29/18
to SWAT-CUP
Thanks a lot for your response Sir. 

Sir the iteration should be of same number of simulations as the last calibration iteration or should the starting and ending of this iteration should be the best simulation number as attained for the calibrated model? Also the parameter ranges should be from the best parameter ranges?

I did 500 simulations for my last calibrated iteration and best simulation number is 79 for my case.

Karim Abbaspour

unread,
Jun 29, 2018, 3:21:42 AM6/29/18
to swat...@googlegroups.com
Use the calibrated range and do as many simulations as before.

Karim

Saima Nauman

unread,
Jun 29, 2018, 3:47:45 AM6/29/18
to SWAT-CUP
Sir only the best parameter ranges could only be used for the future and historical scenario but not the fitted values in the best simulation?

Should I make the pcp and tmp files of the same format as already exit in swat cup project as the one we input in swat are individual for stations and the one in swat cup project is combined for all stations?

Saima Nauman

unread,
Jul 3, 2018, 5:32:18 AM7/3/18
to SWAT-CUP
Dear Sir, 
I need to change file.cio, number of yeras of simulations and start year of simulation in this case, also I have only temperature and precipitation data shuld I remove solar, humidity and wind data from SWATCUP folder? 

Should I make any changes to observed.rch as I am running future scenarios so what should I input for observed.rch?

Regards.


On Friday, June 29, 2018 at 12:21:42 PM UTC+5, Abbaspour wrote:

Natalja C.

unread,
Jul 3, 2018, 6:31:52 AM7/3/18
to SWAT-CUP
To run the future scenarios you do not need SWAT-CUP. SWAT-CUP is used for calibration.

To setup your future scenario - follow the SWAT I/O documentation, make the changes to the files (file.cio, wgn, cst, sub) either by hand, or by the means of SWAT-Editor. Be sure that you are using the calibrated version of your model.
After you set up everything, run ONLY the swat executable. You do not need to run the swat-cup or anything else.

Best,
Natalja

Saima Nauman

unread,
Jul 3, 2018, 7:33:29 AM7/3/18
to SWAT-CUP
hi Natalja,
Thanks a lot for your response. 

By calibrated version you mean I should make amendments in the ".sol", ".gw" , "hru" parameters as per calibrated ranges. My question is how to incorporate those ranges in arcSWAT it only accepts single values how can i use calibrated ranges from SWAT CUP to arcSWAT?
Best Regards.

Karim Abbaspour

unread,
Jul 3, 2018, 8:41:10 AM7/3/18
to swat...@googlegroups.com
you cannot ignore the uncertainty in the parameters. I have explained here before how you do that.

Karim

Karim Abbaspour

unread,
Jul 3, 2018, 9:04:48 AM7/3/18
to swat...@googlegroups.com

Steps to do climate change analysis with your calibrated parameter ranges:

1-      Create the pcp and tmp files for the future period and copy it to the swatcup project directory (where you calibrated your model)

2-      Edit file.cio with future dates

3-      Adjust CO2 concentrations in the .sub files

4-      Setup the no_obderved.def files to extract the variables of interest

5-      Edit 95ppu_No_Obs.def file

6-      Check the appropriate extract_no_observed.exe programs in Sufi2_Extrcat.bat file

7-      Check 95ppu_No_Obs.exe in Sufi2_Post.bat

 

Then make as many simulations as you did in the last iteration of calibration run. You will have future results as well as its uncertainties.

 

Saima Nauman

unread,
Jul 5, 2018, 7:04:49 AM7/5/18
to SWAT-CUP
Dear Dr Karim,

Thanks a lot for your detailed response and guidance.

Regards,
Saima

Saima Nauman

unread,
Aug 3, 2018, 10:01:38 AM8/3/18
to SWAT-CUP
Dear Dr karim,

When we have done simulation for future data ,we get 95 PPU for no observed and a file for R_Flow. Is the result of flow in the last simulation (#500) considered for further working.  

Regards.

You received this message because you are subscribed to a topic in the Google Groups "SWAT-CUP" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/swat-cup/z82nEfyTXbo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to swat-cup+u...@googlegroups.com.

NAGESH MISHRA

unread,
Aug 9, 2019, 2:56:21 PM8/9/19
to SWAT-CUP
I have the same doubt. What should be considered as streamflow at a sub basin for future projections (simulated with calibrated  parameter ranges), R_FLOW or M95PPU? 


On Friday, August 3, 2018 at 7:31:38 PM UTC+5:30, Saima Nauman wrote:
Dear Dr karim,

When we have done simulation for future data ,we get 95 PPU for no observed and a file for R_Flow. Is the result of flow in the last simulation (#500) considered for further working.  

Regards.

To unsubscribe from this group and all its topics, send an email to swat...@googlegroups.com.

Karim Abbaspour

unread,
Aug 9, 2019, 3:54:43 PM8/9/19
to swat...@googlegroups.com

P P Choudhari

unread,
Aug 10, 2019, 2:30:16 AM8/10/19
to swat...@googlegroups.com
Dear Karim sir,
                       Good explanation, can you make one small video that will be useful for every one. please...........this is humble request. 








With Regards,

Dr. P P Choudhari
Assistant Professor
Department of Geography
University of Mumbai, Mumbai, Maharashtra, India
Mobile No - 08275308064


Message has been deleted

Karim Abbaspour

unread,
Aug 10, 2019, 8:16:22 AM8/10/19
to swat...@googlegroups.com
I will try to make several videos explaining different aspects of the program.

Best, Karim.

NAGESH MISHRA

unread,
Aug 10, 2019, 10:21:18 AM8/10/19
to swat...@googlegroups.com
Dear Dr. Karim,  
Thanks for your quick response to our doubts. I have further doubt on the same line. 
If I am Interested in the Annual cycle of Streamflow at a sub-basin outlet and I want to plot it with parametric uncertainties (95PPU BAND) in the future projections. Can I take U95PPU and L95PPU along with M95PPU for the Annual cycle calculation?  Does this make the same sense of taking the long term average of monthly simulations as uncertainty explained at a monthly scale?  
Thanks 

Sincerely,
Nagesh Mishra

NAGESH MISHRA

unread,
Aug 13, 2019, 4:44:21 AM8/13/19
to SWAT-CUP
Dear Dr. Karim,

In the previous discussion, you have justified about using M95PPU as Stream flow for future projections explaining hydrologic uncertainties.But if I am using the calibrated range of parameter for historical (Baseline) or Climate model validation for hydrologic perspective, what should i take streamflow as M95PPU or Best Simulation calibrated on best parameter ranges?

Thanks 

sincerely, 
Nagesh Mishra
To unsubscribe from this group and stop receiving emails from it, send an email to swat-cup+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
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+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
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+unsubscribe@googlegroups.com.

--
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+unsubscribe@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages