How to code for nests found during the nest building period?

[HS40]

Hello everyone,

I have read chapters 1-7, 11, and 17 in the MARK “Gentle Introduction,” as well as a number of the references listed at the end of chapter 17, including Dinsmore et al. 2002. I have run some models successfully in MARK, but still have some questions I couldn’t find answers to by searching the forum. Hopefully someone could clarify a couple of things for me.

1) I found a number of nests (almost 1/3rd of my sample) during the nest building period. It seems the examples in chapter 17 and Dinsmore et al. 2002 did not take the nest building period into account when calculating DSR. Therefore, when I am coding the information for each nest in MARK, should the day of the nesting season on which the nest was found (referred to as time “i” in chapter 17) be listed as the day the first egg was found in the nest, not actually the day I found the nest?

2) What if I want to consider the laying period when calculating survival and true nest success? I am working with altricial species, the majority of which lay one egg per day until the clutch is complete before beginning the incubation period. Thus far, I have considered the incubation to fledging period for each species as the time required for the nest to reach completion. If I would like to consider the laying period, as well, would adding one day for each egg for the mean known clutch size for each species to the number of days required for the nest of each species to go from incubation to fledging be an appropriate approach?

Thanks in advance for your input!

[birdman]

I realize this is a month-old post, but since no one has responded, I thought I'd offer my thoughts.

For both questions, you have to decide what you are trying to model/estimate. I assume that you, like most of us, are really interested in what happens to a nest once it has contents. What happens before that can vary based on the behavior of the species (e.g.Do they build courtship nests that aren't necessarily used for actual breeding? Do they regularly start multiple nests before deciding on a final one in which to lay?). Unless you want to know something about how likely an initiated nest is to remain active through building and laying, then you would bias your estimates of nest survival (really potential offspring survival) by including all those days prior to there being anything a predator might be interested in. Further, and I think importantly, you shouldn't include days not under observation; i.e. don't add observation days to nests found with eggs, even if you knew they were active.

Example: So you find three nests during the building period, watch them at intervals for a couple of weeks. One suddenly has 3 eggs, the two others are never seen with eggs. You could count back to the initiation of the first, adding days of activity to account for that laying period. This would be fine if you never have nests that receive eggs and are depredated before you find them. So assume those other two nests you thought never had eggs actually did, but were eaten before you rechecked them. Excluding them removes two 'failures' from your sample and fails to fill in a couple of exposure days while they were being laid, prior to failure. This can have a substantial effect on your results if you add data for enough nests, so you only want to work with nests from the point they were observed active. Thus i = the first day you observed contents, regardless of what those contents were.

Hope this helps.
SP

[HS40]

Thanks very much for your reply, birdman. That helps clarify things.