www.phidot.org

Hi all,

I'm trying to generate psi and p estimates for a multi-season model psi(.)gam(.)eps(.)p(.). I have changed my #sites (24), #surveys (8), #surveys/season (2,2,2,2) psi (0.6), p(0.25), eps (0.2 for each season) and gam (0.1 for each season).

I have selected multi-season and the model type and used 'analyse w/ expected values'. The notepad output however contains the error: **** error in getParam (mat=2,row=4,nrows=3) ***, and does not give any beta or real output values.

I've uninstalled and reinstalled newest versions of Presence, Genpres and Mark. I've tried a single-season model and it appears to work.

Any clues as to what might be happening? could be user error or something in the program?
Cheers.

Hi Penny,

I tried to replicate your problem, but couldn't get the same result. Here's what the GENPRES screen should look like:

Code: Select all

#sites 24      #surveys  8               #surveys/season  2,2,2,2
psi   0.6   
p   .25  .25  .25  .25  .25  .25  .25  .25
eps   0.0  0.2  0.2  0.2  0.0  0.2  0.0
gam  0.0  0.1  0.0  0.1  0.0  0.1  0.0

Note that every other eps and gam should be zero to reflect the fact than no extinction or colonization takes place within a season.

The error message looks like PRESENCE was expecting a 4th gam parameter when there should only be 3. Make sure you have the zeros in the gam/eps rows and the #surveys/season is exactly as shown. If you still have trouble, please send me the genpres.pao, genpres.dm and mark.out files from your documents folder.

Cheers,

Jim (jhines at usgs.gov)

Hi Jim,
Thanks for the reply. I've got it set up as you say (except the 3rd eps in yours, should that be 0? if it is set up this way my surveys/season change to 2,1,1,2,2), anyhow I am still encountering issues. I'll send you the files.

Thanks,
Penny.

You're right about the 3rd eps -- it should be zero.

My automated compile-and-copy-to-web script didn't work for GENPRES, so the copy on the web was not the latest. I've fixed it, so please try downloading and running again.

Jim

All fixed! Brilliant, thanks a lot Jim