targeted background locations and environmental variable cell size variation

[colin.g...@gmail.com]

Hello Group,

I have what I hope will be a simple question. Briefly, I am hoping to model environmental correlates of a species' distribution with maxent. My data comes from citizen science photos, which are certainly a spatially biased sample. To try to account for this bias I'd like to use background points derived from photos of closely related species which I assume have the same sampling bias as my species of interest. 

My question is: If I am supplying targeted background locations in this way, is it necessary to project my environmental variables (e.g., bioclim and MODIS data) with an equal area projection? 

My intuition is that cell size variation should only be an issue when randomly sampling environmental correlates for background points, but I am new to maxent and would like to check-in in case I am missing something. I would like to avoid processing the data more than is needed if possible. 

  

Best wishes,

Colin