Incorporating abundance in single-species multi-method model

[kcpregler]

I'm seeking advice as to how people have incorporated abundance into their design matrix. I have areas of high, low, and no abundance at my sampling sites. It was recommended to me to use a binary coding of 1 for high abundance and 0 for low abundance but I have the issue of a few sampling sites having the species not detected at any of the surveys leading me to believe it is indeed not present. I do not think it would be good to lump these no abundance sites with my low abundance sites. Any advice would be much appreciated!

Thanks,

- Kasey Pregler
University of Connecticut