C-hat is a measure of over dispersion. The primary influence of incorporating c-hat is that it increases variances and readjusts AIC values to be more conservative; larger values indicate less of a fit. For a paper,see Assessing the Fit of Site-Occupancy Models.

Non-explanatory covariates within your model pool or dependent observations may be culprits. Apply the neccessary VIF adjustments and see whether you have informative parameters. A c-hat >2 strikes me as pretty high for a single season occupancy model, but would be less surprising under more complex structures.

Hi all, I am using multi-season occupancy models but there isn't a way to assess the goodness of fit in Presence. I wondered if you think there is any value in running GOF tests on each single season model and using the average c-hat value to adjust the multi-season model?

thanks Giancarlo, that is much appreciated.