Systematic sampling grid resolution?

[Madeline Steele]

Hello all,

I am planning to use the systematic sampling approach to help correct for sampling bias in my observation data. Fourcade et al. (2014) demonstrated that this method performs well, relative to other common approaches, across a range of sampling bias types and severity levels. However, I haven't seen any clear guidance on just how much coarser the systematic sampling grid should be than the environmental covariate rasters. Does anyone know of any studies that explore this issue, or perhaps just have a rule of thumb for determining an appropriate coarsening factor?

Thanks very much,

Madeline