Insight into ENMeval Package for Maxent

[Abid Ali]

Hi,

I'm in the process of transitioning to ENMeval for modeling my species and I'm new to this package. I came across a nice tutorial https://rsh249.github.io/bioinformatics/ENMeval.html, which provides insights into fitting the model. 

However, in the Model Testing section, the provided code doesn't seem to include options for using random background points. 

My occurrence data is stored in spg, and predictors are in env. Could someone share an R Markdown or a detailed tutorial for running Maxent, MESS, and subsequently calculating niche overlaps between lineages under the ENMeval package?
 

The provided code for model fitting is from the given link. Any help or guidance in this regard would be highly appreciated.

Thanks

eval = ENMevaluate(occ=sp_df[,c('longitude', 'latitude')], env = preds, method='randomkfold', kfolds=10, parallel=TRUE, numCores = 12, fc=c("L", "Q", "LQ"), RMvalues=seq(0.5, 2, 0.5), rasterPreds=T)

[Jamie M. Kass]

This is my vignette for ENMeval 2.0. This ENMeval 2.0 vignette will be updated when new versions are uploaded to CRAN, so it's best to reference this one.

https://jamiemkass.github.io/ENMeval/articles/ENMeval-2.0-vignette.html

The one you linked uses the older version 1, and because it did not specify background records in the call to ENMevaluate, the function will generate random background points over the extent of the input environmental raster. You can make your own random points and input them yourself -- see the vignette above.

Jamie