openCR survival interpretation

153 views
Skip to first unread message

HL Y

unread,
Jan 14, 2024, 9:32:03 AM1/14/24
to secr

Dear all,

Happy New Year!

I am Hongli Yu. I am currently modelling the survival of hedgehogs with an 11-year SCR dataset using openCR 2.2.5, CJS. I am not quite sure how to interpret the results (shown below), and could anyone help with this please? Thank you in advance!

Basic information: hedgehogs were classed as juveniles (year of born) and adults (survived for at least one winter; the majority were classified as adults when they were first caught). Our search area was about 10 times male adult’s normal home range(~20ha), but most captures were from the central area, and only a small number were from areas close to the edge (despite the comparable search effort). I fitted models for female adults, male adults, female adults and juveniles combined, and male adults and juveniles combined separately. When adults and juveniles combined, they shared all the parameters (as the detection p~age did not work well). Then based on the AIC, across these groups the constant survival models (phi~1) were more supported than year/session-specific (phi ~t), and the Spatial normal movement models were more supported than the Spatial static.

My questions are: can I understand that the Movement kernel scale α is the estimated average distance of activity centers across individuals between years? The α was almost identical for adult males and adult females; but when juveniles were further added, it became higher in males (adult juvenile combined, 354 m, 95% CI 193 - 648) than that in females (adult juvenile combined, 180m, 95% CI 1 - 326), likely indicating the possible natal dispersal in related to males from their born year to their second year? Indeed, all juvenile males were not recaptured after July of their second year, compared to 76% of juvenile females). Further, is the survival from spatial normal (e.g., 0.447 for male adult juvenile combined) minus that from spatial static (0.442) relevant to the migration rate? If so, then this seems to be very low?

Thank you very much! Any help would be very appreciated!

Kind regards,

Hongli Yu, PhD student

Nottingham Trent University, UK

hongli...@my.ntu.ac.uk

hongli...@gmail.com

openCR questions.docx

Murray Efford

unread,
Jan 16, 2024, 1:44:31 PM1/16/24
to secr
Hello Hongli Yu
Open-population SECR is a bit of a minefield as you've found: results from real-world (modest-sized) datasets don't always 'make sense'.

You mention CJS, but there is no direct spatial analogue of CJS. Type CJSsecr produces nonsense unless you set CJSp1 to TRUE to model first captures, and then you may as well fit one of the models with recruitment.

We can be almost certain that a bivariate normal movement model is inferior to one with longer tails e.g. BVE or RDE (see MEE review 2022).

The parameter 'a' of the movement model does not equate with average distance moved - you need to get this later see e.g. ?summary.kernel

The pattern of sex-biased natal dispersal in your hedgehogs seems very plausible. As I interpret it, all locally born males had disappeared by the second year? Modelling long-distance emigration is going to be hard (impossible with this scale of sampling?), and a BVN model doesn't help. I wouldn't make anything of the difference in survival rates between static and bivariate normal models.

Hope this helps despite being a bit gloomy!
Murray

HL Y

unread,
Mar 18, 2024, 5:37:44 AM3/18/24
to secr...@googlegroups.com, Murray Efford
Dear Murray,

Apologies for my late reply! I was on a trip and could not access to gmail for the last month or so. Thank you very much for the paper you sent. It is interesting and helpful - I was just thinking how to write the discussions and frame the draft with my current results!

Thank you very much.

Cheers,

Hongli

Hongli Yu, PhD student

Nottingham Trent University, UK

---------- Forwarded message ---------
From: Murray Efford <murray...@gmail.com>
Date: Fri, 19 Jan 2024 at 01:51
Subject: Re: openCR survival interpretation
To: HL Y <hongli...@gmail.com>


Thanks. It's good to post replies to the group.
Delayed natal dispersal may be very widespread in mammals. I was reminded of work I published on brushtail possums in 1998 (attached). That preceded spatial capture-recapture; maybe someday I'll revisit it with spatial methods!
Cheers
Murray

On Fri, Jan 19, 2024 at 1:46 PM HL Y <hongli...@gmail.com> wrote:
Dear Murray,

Thank you very much for your help! I have set CJSp1 as TRUE, and tested BVN, BVE, and RDE movement models and will further check based on your suggestions.

Good to to know that natal dispersal might occur in our hedgehogs. All juvenile males disappeared by second year seems support this. Despite the scale of dispersal is hard to know with our dataset, the results is encouraging. The overall stable survival trend we found is also positive. 

Thank you very much!

Best regards,

Hongli




--
You received this message because you are subscribed to the Google Groups "secr" group.
To unsubscribe from this group and stop receiving emails from it, send an email to secrgroup+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/secrgroup/16fcba5b-8aea-4774-b5de-19803b98521dn%40googlegroups.com.
Efford 1998 JAE.pdf
Reply all
Reply to author
Forward
0 new messages