Autocorrelation and variable selection in multi-species studies

[bkara...@gmail.com]

Hello everyone,

In the papers and guides a common step of sdm's is removing highly correlated variables while keeping the ecologically sensible ones. I have been using the values of the environmental variables at the occurrence points in the correlation analysis. Recently, I've begun to tinker with niche identity analyses to understand and learn them. In some multi-species studies it is stated that the authors used the same uncorrelated variables for all the species' in the study. However, in my case if I use the values at the occurrence points, correlations between variables are different for each species to the extent that it is impossible to use the same environmental variables for the species' . My question is, do they use correlation analysis for the whole study region and not just use the values at the occurrence points? Or is there any other way? It is not stated in the papers.

Thank you

[Sanjo Jose]

Hi,

Using SDM toolbox in ArcGIS you can do correlation for the whole study region.

Regards, 

ᐧ

Sanjo Jose V,

Junior Research Fellow,

Centre for Climate Change and Adaptation Research,

Anna University,

Chennai - 600025

Mob: 9400824393

Research Gate

--
You received this message because you are subscribed to the Google Groups "Maxent" group.
To unsubscribe from this group and stop receiving emails from it, send an email to maxent+unsubscribe@googlegroups.com.
To post to this group, send email to max...@googlegroups.com.
Visit this group at https://groups.google.com/group/maxent.
For more options, visit https://groups.google.com/d/optout.