pcountopen/validation/model fitting

[Ashley S]

Hi again

I am using the DM model for my data now and I have a final model, at least I think I do. But once I put it through the Nmix.gof.test  it does not fit well. I fit the model by running a forward stepwise regression and AIC to choose the best covariates in order of detection, lambda, omega, then gamma. I did this because if I used the forward stepwise regression while fitting gamma before omega, most of my covariates fit the model. I am using an interaction term in gamma because my question is about seasonality and change in abundance over seasons. My max K is 21, so I have been using K=41, but I increased that to 100 to see if that changes anything. 

Essentially I want to know about the utilization abundance of a site vs absolute abundance. Am I doing this correctly? 

**I have been referring to AHM2 and other paper sources, but can't seem to find the best method.

Thanks!

Ashley