Calculating overall occupancy

[murray.efford]

I regularly see references to point count surveys on this forum. I would like to put out a serious health warning: point counts do not generally yield data suitable for occupancy analysis. The primary reason is that plot size is ill-defined. Further argumentation in Ecosphere (2012) 3:32.

[darryl]

I agree with Murray that ill-defined plot sizes can cause problems, especially around interpretation of results. How you want to interpret the results should be something you spend a good amount of time on well before you set foot in the field, regardless of whether you're doing point counts or some other field sampling method, then design your field methods accordingly to get the right type of data for your purpose. In some situations I think point count-type protocols are reasonable to use to get you that detection-nondetection data, and in other cases they won't be. The real problem is when people are trying to shoe-horn data that was collected for some other purpose or have not thought things through and been unwilling/unable to modify field methods to get the type of data they should be collecting given their question of interest.
Darryl