systematic covariate testing to ID salient parameters

[Pentatome]

I apologise if this query is basic, but I am very new to this field. I would greatly appreciate any feedback on the soundness of my methods and reasoning.

I am running a single-season, single-species model on Presence using camera trapping data (19 cameras, 969 total trap nights). 21 environmental covariate data were collected at each site, our goal is to identify the micro-habitat preferences of the studied species (Arctictis binturong).

I decided to individually test each covariate for detection and occupancy to identify which parameter would be more salient.



I selected the most prevalent covariates or best proxies and ran multiple models between them:



Here are my interpretations:

- All covariates tested individually for occupancy with a high ΔAIC are not good predictors for binturong occupancy.
- My best models are:

1) psi(Theight),p(Theight), ΔAIC= 0.00
2) psi(Theight+GC),(Theight), ΔAIC= 0.82

- Hence the best predictors for binturongs are tree height (Theight) and connectivity between trees (GC).
- The biggest influence for detection is tree height (Theight), with a minor effect of camera trap height (CTheight).

Thank you for your help!