Skip to content

ENFA scatter plot problem & habitat suitability map question

3 messages · Michela Giusti, Alexandre Villers, Mathieu Basille

Original
Michela Giusti
# 
Hello,

I have a problem with my ENFA scatter plot. Data are from three 
different sites far from one another along the Mediterranean Sea.
You can find it in the atteched file.
The first scatter is made with the line command :

scatter(enfa_TOT,pts = FALSE, nc = TRUE, percent = 100)


After many tests I realized that the habitat used is not visible because 
it coincides with the x-axis. Do you have an idea about that?

I did the same scatter with the line command:

scatter(enfa_TOT,pts = T, nc = TRUE, percent = 100)....the result is 
below... Any idea about it?




Another question is about the habitat prediction map that I would like 
to do after the ENFA analysis. What method do you saggest to use to do that?


Thank you very much for your help.

Michela
1 day later
Mathieu Basille
# 
Dear Michela,

In addition to Alexandre answer, it is not advised to use the ENFA to 
produce suitability maps. You might be interested in the MADIFA instead, 
also available in adehabitatHS. See [1].

As for your scatter plot, you should investigate your variables 
individually, e.g. with ?hist.enfa (see the parameter 'score'). You may 
have a categorical variable, with only one level used, or something similar.

Hope this helps,
Mathieu.


[1] Calenge, C.; Darmon, G.; Basille, M.; Loison, A. & Jullien, J.-M. 
(2008) The factorial decomposition of the Mahalanobis distances in habitat 
selection studies. Ecology, 89: 555-566 doi 10.1890/06-1750.1


Le 06/27/2013 11:13 AM, Alexandre Villers a ?crit :