How to deal with inconsistent recapture events
Posted: Fri Jul 08, 2016 12:09 amHello all,
I have a data set of chick banding and adult recapture for a seabird colony. The whole data set covers a long period (over 20-years), during which chicks were banded in most years. However, adults were only recaptured in a few years (including in 2015). The reason for this is that chick banding only takes 1-day of field work and can be done by 1 or 2 people, whereas adult recapture takes nearly a week and requires more people. Unfortunately time and money limitations have meant that adult recaps could not be undertaken very often.
I know that I can change the time intervals between occasions when setting up my models. But I am not sure how to deal with data where there are occasions when marking occurred, but recapture effort did not. I don't want to exclude those occasions from the data set. But if I don't somehow 'tell the model' that recapture effort wasn't expended at each occasion, I feel like this will affect the model estimates (perhaps I am wrong?). Should I fix the recapture probability to 0 for the years when we didn't attempt to catch adults?
It's not directly relevant to my question, but in case it is of interest or makes a difference to specific advice on this issue, I use RMark for my analyses.
Any advice would be much appreciated.
Thanks,
Alice
I have a data set of chick banding and adult recapture for a seabird colony. The whole data set covers a long period (over 20-years), during which chicks were banded in most years. However, adults were only recaptured in a few years (including in 2015). The reason for this is that chick banding only takes 1-day of field work and can be done by 1 or 2 people, whereas adult recapture takes nearly a week and requires more people. Unfortunately time and money limitations have meant that adult recaps could not be undertaken very often.
I know that I can change the time intervals between occasions when setting up my models. But I am not sure how to deal with data where there are occasions when marking occurred, but recapture effort did not. I don't want to exclude those occasions from the data set. But if I don't somehow 'tell the model' that recapture effort wasn't expended at each occasion, I feel like this will affect the model estimates (perhaps I am wrong?). Should I fix the recapture probability to 0 for the years when we didn't attempt to catch adults?
It's not directly relevant to my question, but in case it is of interest or makes a difference to specific advice on this issue, I use RMark for my analyses.
Any advice would be much appreciated.
Thanks,
Alice