[R] Modeling presence only data in R

Clément Calenge calenge at biomserv.univ-lyon1.fr
Wed Apr 23 16:50:50 CEST 2008

Actually, the ENFA (function enfa, in the package adehabitat) is not 
really suitable for modelling environment suitability maps (though it is 
still useful to discover the patterns of the ecological niche with 
presence-only data). However, the package adehabitat also contains other 
functions allowing this kind of modelling: the Mahalanobis distances 
(function mahasuhab) and their factorial decomposition (function madifa) 
could be a useful approach. See:

Calenge C, Darmon G, Basille M, Loison A, Jullien JM. (2008) The 
factorial decomposition of the Mahalanobis distances in habitat 
selection studies. Ecology, 89, 555-566.

and references therein. You may also consider the GNESFA, a 
generalization of the MADIFA/ENFA approach (see the function gnesfa and 
references therein). Other commonly used approaches are also available 
in adehabitat.
Best wishes,

Clément Calenge.

Stéphane Dray wrote:
> Ecological Niche Factor Analysis (ENFA) is implemented in the adehabitat 
> package available on CRAN.
> Cheers,
> Alejandro González wrote:
>> Dear Milton, you have 2 packages for modelling species distribution in 
>> R: grasper and Biomod (by Wilfried Thuiller), but they are all 
>> presence-absence models, so you must generate pseudoabsences for each 
>> species, following recommendations in papers (see Lobo 2007). On the 
>> other side, there are free gis tools to perform presence-only models 
>> (Biomapper, GARP, Openmodeller). Write me if you need if you want some 
>> advice.

Observatoire des galliformes de montagne
Office national de la chasse et de la faune sauvage
Saint Benoist - 78610 Auffargis
tel. (33)

More information about the R-help mailing list