SWAT-CUP SUFI2 "Cannot open input file" and adjusting parameter ranges

[Rosie Records, CA Cooperative Fish & Wildlife Research Unit]

Hi all,

I have two SUFI2 questions that I would appreciate your help with:
1) I had a successful run of SUFI2 sensitivity analysis but have
encountered the following error during sufi2.post.bat: "Cannot open
input file \sufi2.out\no_beh_sims.sf2." All my calibration output
files from the iteration appear to be opening fine, and so it's likely
not a cause for concern...but I am not sure what the no_beh_sims.sf2
file is, and whether the fact that it is missing could cause me
problems later. Does anyone know?

2) I would like to include some parameter ranges that are outside the
limits of Absolute_swat_values.txt. (Because I have a high
concentration of wetlands and ponds in some subbasins, the actual
observed WET_MXVOL, etc. are higher than the upper limit allowed by
SUFI). I assumed that all I need to do is locate the
"Absolute_swat_values.txt" file in the SUFI2 project TxtInOut folder,
and change the upper limit for these parameters to accommodate the
values I want to set. Do you know if this will work, or if there is
anything else I need to change for the run to be successful?

Thanks so much!

Rosie Records

[Gregory S. Karlovits - CEMVR]

Hi Rosie -

I'm not sure about your first question since I'm not sure how much
different the SUFI2 I/O is from GLUE, but if you define a "behavioral"
threshold for your objective function that could have part to do with
it.

As for the second part, yes it is that easy to edit that file to
change the thresholds. I would keep a backup of the original but
editing that file will allow the bounds to change.

Greg


On Mar 30, 12:39 pm, "Rosie Records, CA Cooperative Fish & Wildlife

[Rosie Records, CA Cooperative Fish & Wildlife Research Unit]

Hi Greg -

Thanks so much for the response. I did define a "behavioral"
threshold for the objective function and the threshold was not met in
the number of simulations I specified for this run...Do you think this
could be why SUFI2 can't find the "beh" file?

Also thanks for the confirmation for the SWAT-CUP min/max values.
It's good to know that will work.

Thank you!
Rosie

On Mar 31, 7:30 am, "Gregory S. Karlovits - CEMVR"

[Gregory S. Karlovits - CEMVR]

Rosie -

I'm pretty sure that's why it didn't create that file. The GLUE
algorithm does something similar, either if you set your objective
function behavioral threshold too high or the parameterization you
chose to try and calibrate wasn't really sensitive enough to get a
good result. Usually if I have that problem I try to evaluate which is
the case before I make any changes. If you look at your outputs (if
it's anything like GLUE) you'll have one that shows the relative
changes for each parameter along with the value for the objective
function, so you can get an idea of where the results are landing.
That can help you diagnose why you're not getting any "behavioral"
results.

Greg

On Apr 1, 2:16 pm, "Rosie Records, CA Cooperative Fish & Wildlife

> > > Rosie Records- Hide quoted text -
>
> - Show quoted text -