Estimates of occupancy and detection: can I multiply them?

questions concerning analysis/theory using program PRESENCE

Estimates of occupancy and detection: can I multiply them?

Postby rockylxc » Thu Apr 15, 2021 8:07 pm

Dear friends:

What I’ve got:
--Modelling work has done for several species within 141 15km*15km grids, based on detection histories collected from about 500 interviewees. In general, I chose variables from four variable groups to establish models, e.g. 1. geography, like altitude and terrain ruggedness; 2. forage availability, like NDVI, historic NDVI change, and dominant eco-types; 3. competition or disturbances, like local livestock number and road density, and 4. Interviewee related, like education, years in town etc.

My current goal:
--based on those occupancy and detectability estimates, my plan is to "stratify" a national park, then will sample n grids on the ground (e.g. double observer method for density estimates), so as to further develop global abundance estimates for the target species within the park.

My question:
As for the stratification, I am thinking about multiplying the detection rate and occupancy rate from the best model, to assign a value to each grid for each species. The reason I do the multiplication is, for which I do need expert’s thoughts, to my understanding, the result represents “the maximum likelihood that the existing detection history can be explained”.

Would it be safe to say that it also represents “the possibility that there is widely distributed a very large, unmissable population of the species of interest”?

Previously I only planned to use the occupancy estimates for stratification. But when getting the modelling results, several grids that I know target species does not exist came to be of high predicted occupancy. If detectability rate is only applied for stratification, it also seems not very right.

Thanks for your thoughts.
Posts: 1
Joined: Thu Apr 15, 2021 7:33 pm

Return to analysis help

Who is online

Users browsing this forum: No registered users and 1 guest