multiple detection methods in occumulti

[Divyajyoti Ganguly]

Hi,

I have detection/non-detection data using camera traps and sign surveys from non-overlapping sites for four species. By integrating data from these, I am trying to look at co-occurrence patterns between species pairs using the Rota et al. 2016 multi species model (package occumulti). I plan to use 'effort', calculated as a proportion for both detection types, and 'detection type' as detection covariates. I further plan to incorporate an interaction between effort and detection type to account for the differences in the sampling process. Does it make sense to have the efforts for the two detection types under the same column? The other option is to have two columns for effort, whereby for sites with camera traps, sign survey effort would be zero and vice versa. However, the problem with this is that it may erroneously return zeros for the effort*detection type interaction. Would very much appreciate some views on this.

Best,

Divyajyoti Ganguly

NCBS, Bangalore