occu() with unequal occasion length

[Marc Kery]

Dear Ken,

I am helping some colleagues who have a survey of long-eared owls with three occasions of unequal length (2, 2.5 and 5 mins or something). We want to do occupancy modeling and it would be good to account for this unequal occasion length. I know now how to specify the likelihood in R and could adapt that for unequal length. But before I do that I was wondering whether you would have code somewhere lying around for a beta version of function occu() which would allow for this ?

Thanks and best regards  --- Marc

[Ken Kellner]

Hi Marc,

No, I don't have any code for that, sorry. I'd be interested to see your solution if you're willing to share it, for potential future use in unmarked.

Ken

> --
> *** Three hierarchical modeling email lists ***
> (1) unmarked (this list): for questions specific to the R package unmarked
> (2) SCR: for design and Bayesian or non-bayesian analysis of spatial capture-recapture
> (3) HMecology: for everything else, especially material covered in the books by Royle & Dorazio (2008), Kéry & Schaub (2012), Kéry & Royle (2016, 2021) and Schaub & Kéry (2022)
> ---
> You received this message because you are subscribed to the Google Groups "unmarked" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to unmarked+u...@googlegroups.com.
> To view this discussion visit https://groups.google.com/d/msgid/unmarked/ZR0P278MB0869C90CE7A63F612D488C8AEBF02%40ZR0P278MB0869.CHEP278.PROD.OUTLOOK.COM.

[Josh Twining]

Perhaps there is an advantage of coding up an alternate likelihood in terms of some saved statistical power to estimate other parameters, but if you wanted an easy life - would it not be simpler for your colleagues to account for unequal occasion lengths by fixing a simple site-by-occasion "duration" covariate on p? 

I wouldn't even give this a second thought.... so I wonder what am I not seeing? 

Josh

To view this discussion visit https://groups.google.com/d/msgid/unmarked/Z6eEnyy62azXvatA%40BADGER.

--

Joshua P. Twining, PhD,

Assistant Professor,

126 Nash Hall,

Oregon State University,

Oregon, 

97331

Email: joshuap...@gmail.com / joshua....@oregonstate.edu 

Mob: (607) 220 7547

[Ken Kellner]

I guess one advantage is it's one less parameter to estimate?

I just remembered that we actually do support defining unequal period lengths in the gdistremoval function (specifically the removal part) where we have

p_period[i] <- 1 - (1-p[i])^period_length[i]

where p[i] is the probability of detection for 1 unit of time at site i. I assume it would be a similar approach for occupancy.

Ken

> To view this discussion visit https://groups.google.com/d/msgid/unmarked/CA%2Bg%2BShnZc54KMLqxPa4yWT6y2BVH3gEWhvzoo0vujrqtYkSXtg%40mail.gmail.com.

[Marc Kery]

Dear Ken, Josh,

thanks for the answers. I like Josh's idea, but I think it's more of an approximate solution. In contrast, the solution that Ken reminds us is in gdistremoval() is the mathematically exact one, where we express detection per occasion as a function of detection per suitable length unit (e.g., 1 minute) and occasion length as a multiple of that unit.

I will probably give it a try and will share with the group if I get to it.

Best regards  --- Marc

---

From: unma...@googlegroups.com <unma...@googlegroups.com> on behalf of Ken Kellner <con...@kenkellner.com>
Sent: Saturday, February 8, 2025 19:30
To: unma...@googlegroups.com <unma...@googlegroups.com>
Subject: Re: [unmarked] occu() with unequal occasion length

 

To view this discussion visit https://groups.google.com/d/msgid/unmarked/Z6eiudqw4tGUIvc2%40BADGER.