estimating p_breed

[joelpick]

the estimated and simulated p_breed (probability that female breeds in a given year) do not match well. This is because if a female doesn't breed and then dies, that skipped breeding isn't detected - it has to be followed by a breeding attempt. This will decrease with increasing adult survival, but will always be biased downwards. I think we can correct for this using estimated adult survival