Page 1 of 1

Beginner's Question - Capture probability & Assumptions

PostPosted: Mon Jan 22, 2018 12:46 pm
by smw62
Hi,
This is a very basic question that I hope someone can answer with one word.
From the literature: Most crucial assumption is capture probability remains homogenous over the study.

My best fitting open population model (POPAN) uses a time dependence estimate for capture probability.

Does using this model violate the assumption?

I am trying to get a population estimate of eels passing through a barrier. There are no environmental (short study), behavioural (no avoidance), or biological factors (all young females) that would skew individual capture probabilities, before or after tagging. So I thought the capture probability would remain the same for all individuals.
Does the time dependence apply to the entire population? I.e. the capture probabilities remain homogenous for individuals across the study, but at each sampling point it can change so it doesn't actually violate the assumption?

Re: Beginner's Question - Capture probability & Assumptions

PostPosted: Mon Jan 22, 2018 2:09 pm
by Bryan Hamilton
No the time dependent model does not violate the assumption of homogenous capture probabilities.

It allows varying capture probabilities by time period, but within time periods, capture probabilities are homogenous across individuals.

Re: Beginner's Question - Capture probability & Assumptions

PostPosted: Mon Jan 22, 2018 3:16 pm
by smw62
Awesome, thanks so much

Re: Beginner's Question - Capture probability & Assumptions

PostPosted: Mon Jan 22, 2018 4:49 pm
by murray.efford
Taking this a little further... The assumption is that there is no unmodelled variation in capture probability. Hence the industry supplying models for capture probability, of which the time-dependent model is one example. Rather than representing a violation of the assumption, it increases the chance of compliance.