estimation of c-hat in CJS models with '.'

[roche042]

Does anybody have any advice as to how one might estimate c-hat for a CJS dataset with many structural zeros? I have 25 encounter history data set for 6 groups (2 species at 3 different locations)... this represents two species of chicks from age 1 day through 25 days... Not each location was visited at each day of chick age so I have many '.' in the encounter history on days during which certain locations were not visited (and of course, there is no real pattern to these '.'...). The goal of the analysis is to estimate daily chick survival and detection. However, I am pretty sure this dataset will have a relative high c-hat value and I don't want to proceed to far with any analysis without estimating a chat value. I have attempted to estimate this using RELEASE GOF and median c-hat in Mark and both seem to crash (and ,median c-hat warns me that '.' are not handled correctly).

Any advice/direction that anybody can offer would be greatly appreciated, thanks!