sensitivity analysis

[neda65]

hello everyone
I would like test sensitivity analysis for my model. I used simulation control window, but in the parameter section, i can only select constant parameters, and when I use the auxiliary and driving variables, it gives an error.
RROR: NDVI not found or wrong type must be constant.
NDVI=RANDOM_UNIFORM(0.01,2).
^.
Line 2 of file sensitivitymodel.vsc.
ERROR: No sensitivity inputs marked in sensitivity file.

why??

[Administrator]

Your profile has you listed as a PLE user, PLE does not have sensitivity analysis. Can you update your profile please?

You'll need to upload your model/files for us to be able to tell you what is wrong.

[tomfid]

This is by design - you can't specify complex behavior for an auxiliary via a simple distribution.

[tomfid]

There are many possible ways to vary an auxiliary, but if you tell us what you're trying to do we might be able to suggest one.

[neda65]

Thanks for your reply.I attached my model. In this model Phytomass is a level variable, temperature and precipitation are driving variables, NDVI is an auxiliary, Stock and Consumption rate are constant variables. I would like to do a sensitivity analysis for Phytomass, Which variables have the most effect on phytomass.
My question: Is the sensitivity analysis only for constant variables or can it be done for other types of variables (i.e., auxiliary, driving )?

[neda65]

Thanks for your reply. I update my profile. I attaches the model image.

[Administrator]

Thanks for your reply. I update my profile. I attaches the model image.

Please complete the serial number part of your profile.

[tomfid]

Looks like you could either vary temperature and precipitation directly, or add a third term to NDVI representing variation around the f(temp,precip) and add the variation to the voc.