[Biomod-commits] BIOMOD EXAMPLE

Wilfried Thuiller wilfried.thuiller at ujf-grenoble.fr
Mon May 14 11:58:59 CEST 2012


Dear Joseph,

At the moment, BIOMOD is not able to extract pseudo-absence data (PA hereafter) from raster data (the up-coming update by the end of May will). It means that you have to extract a large range of potential pseudo-absence yourself (either using sampleRaster from the raster package or in ArcGIS-type of software). It means that you have to define your domain/study area first, and then select random points within this domain by extracting the environmental variables for these points. That will constitute your "potential absence". Let's say you extract 100,000 potential background data. You put this data with your presence data (0 for absence and 1 for presence). And then you give this matrix to BIOMOD. 
Within BIOMOD you can then chose the method to extract a sub-set of pseudo-absence data (from the large set you have selected already), and the number (e.g. 10,000). 

Is it clearer now? If not, just let me know. 

As an information, the up-coming version of BIOMOD (called BIOMOD2) will do that these two steps in one step. This will be more logical and easier to explain. 

Hope it helps,


Wilfried



Le 7 mai 2012 à 16:25, Joseph Maina a écrit :

> Hi Wilfred,
> 
> I am following up on this thread..
> http://lists.r-forge.r-project.org/pipermail/biomod-commits/2009-September/000019.html
> 
> what is still not clear to me is, from which data will the
> pseudo-absences/background data be selected? in the above thread you
> say:  "When you set the NbRepPA,  two selections
> of pseudo-absence will be carried out. This means than  1000
> pseudo-absences will be drawn from the original dataset to build the
> presence-absence matrix. That will done twice with the algorithm you
> selected (random, or other)."  Is this 'original dataset' the
> occurrence .csv file that is loaded at the start? can I define a
> 'geographical space' (my study area) from where BIOMOD
> could extract the pseudo-absences/background data? if not what spatial
> extent does BIOMOD use to select pseudo-absences? I have followed the
> manual and  I am at a point where I have run the Model function using
> the occurrence .csv file; up to now I have not loaded any raster data
> set and wondering how BIOMOD extracts background points/pseudoabsences
> without prior info on the study extent..and wondering how I can define
> this upfront.
> 
> Thanks,
> Joseph
> _______________________________________________
> Biomod-commits mailing list
> Biomod-commits at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/biomod-commits

--------------------------
Dr. Wilfried Thuiller
Laboratoire d'Ecologie Alpine, UMR CNRS 5553
Université Joseph Fourier
BP53, 38041 Grenoble cedex 9, France
tel: +33 (0)4 76 51 44 97
fax: +33 (0)4 76 51 42 79

Email: wilfried.thuiller at ujf-grenoble.fr
Personal website: http://www.will.chez-alice.fr
Team website: http://www-leca.ujf-grenoble.fr/equipes/emabio.htm

ERC Starting Grant TEEMBIO project: http://www.will.chez-alice.fr/Research.html
FP6 European EcoChange project: http://www.ecochange-project.eu







-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20120514/f630d44e/attachment.html>


More information about the Biomod-commits mailing list