[R-sig-eco] package FD--unexpected singular species results?

Etienne Laliberté etiennelaliberte at gmail.com
Fri Apr 16 01:00:13 CEST 2010


Because the bug was posted on r-sig-ecology, I'm now posting a message
to mention that it's now fixed in FD 1.0-9, which I just submitted to
CRAN.

Thanks to Sarah Berke for reporting it.

Etienne


Le vendredi 16 avril 2010 à 08:27 +1200, Etienne Laliberté a écrit :
> Dear Sarah,
> 
> Thanks for your interest in FD. As the posting guide says
> (http://www.r-project.org/posting-guide.html)
>         
>         If the question relates to a contributed package, e.g., one
>         downloaded from CRAN, try contacting the package maintainer
>         first.
> 
> That would be me in that case. Or write a message / bug request on the
> R-Forge page http://r-forge.r-project.org/projects/fdiversity/
> 
> You understand sing.sp correctly. What you've identified is a bug. I've
> now added it to:
> 
> https://r-forge.r-project.org/tracker/?atid=1895&group_id=465&func=browse
> 
> (which is the preferred way of listing bugs).
> 
> Unfortunately I can't guarantee that I'll fix the bug quickly at this
> stage, because I'm trying to finish my PhD thesis in a mad rush before
> our second baby arrives in 3 weeks. In the meantime, if you want to have
> a look at the code behind dbFD
> 
> https://r-forge.r-project.org/plugins/scmsvn/viewcvs.php/pkg/FD/R/dbFD.R?rev=2&root=fdiversity&view=markup
> 
> and try to identify what's wrong and let me know I'd greatly appreciate.
> But please take this off r-sig-ecology; there is no need to tell the
> whole world about my incompetence.
> 
> Cheers
> 
> Etienne
> 
> 
> Le jeudi 15 avril 2010 à 12:00 +0200,
> r-sig-ecology-request at r-project.org a écrit :
> > Send R-sig-ecology mailing list submissions to
> > 	r-sig-ecology at r-project.org
> > 
> > To subscribe or unsubscribe via the World Wide Web, visit
> > 	https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
> > or, via email, send a message with subject or body 'help' to
> > 	r-sig-ecology-request at r-project.org
> > 
> > You can reach the person managing the list at
> > 	r-sig-ecology-owner at r-project.org
> > 
> > When replying, please edit your Subject line so it is more specific
> > than "Re: Contents of R-sig-ecology digest..."
> > 
> > 
> > Today's Topics:
> > 
> >    1. package FD--unexpected singular species results? (Sarah Berke)
> > 
> > 
> > ----------------------------------------------------------------------
> > 
> > Message: 1
> > Date: Wed, 14 Apr 2010 12:28:27 -0500
> > From: Sarah Berke <skberke at gmail.com>
> > To: r-sig-ecology at r-project.org
> > Subject: [R-sig-eco] package FD--unexpected singular species results?
> > Message-ID:
> > 	<p2webcfe7821004141028hf5812d78jeb8d24dff8e9912a at mail.gmail.com>
> > Content-Type: text/plain
> > 
> > Hello R-Ecologists,
> > 
> > I am working with the dbFD function of the FD package to analyze some
> > functional community data, and I'm confused about the value of sing.sp in
> > the output.  Basically, the number of functionally singular species
> > (sing.sp) equals the raw number of species (nbsp) for every site in my
> > community matrix.  However, I know for a fact that many sites have species
> > with identical functional classifications, which in this case are all
> > categorical (there is 1 continuous variable, but I get sing.sp = nbsp
> > regardless of whether the continuous variable is included or excluded).
> > 
> > If two species have identical functional classifications, then shouldn't
> > they be functionally singular?  Or is sing.sp calculated in some
> > non-intuitive way, perhaps along the lines of taxonomic distinctness?  Or
> > does sing.sp just not work for categorical data?
> > 
> > I suspect that I am simply confused about what sing.sp means, in which case
> > this is really more an ecology question than an R question per se.  I have
> > not found anything in the help files or the FD documentation or the
> > literature that explains how sing.sp is calculated, does anyone here know?
> >  It's certainly possible that something in my code is wrong, or that this is
> > a bug in FD, so I hope that someone on this list can shed light on the
> > issue.
> > 
> > Here is a simplified example showing what I mean (this is similar to my data
> > in that functional categories are nominal and the community matrix has only
> > presence/absence data)
> > 
> > fd <- matrix (c("a","b","c","d", "a","b","c","d",
> > "b","c","d","a", "c","d","b","a"), byrow = TRUE, nrow=4, ncol=4)
> >     rownames (fd) <- c("spA", "spB", "spC", "spD")      #  species
> >     colnames (fd) <- c("F1", "F2", "F3", "F4")       # functional categories
> > 
> > fd      # note that spA and spB are functionally identical
> > 
> > comm <- matrix (c(1,0,1,0, 1,0,1,0, 0,1,1,0, 0,1,0,1), nrow=4, ncol=4)
> >     rownames (comm) <- c("site1", "site2", "site3", "site4")
> >     colnames (comm) <- c("spA", "spB", "spC", "spD")
> > 
> > comm      # note that site 1 has only spA and spB, so I'd expect it to
> >                # have sing.sp = 0 (no singular species, or perhaps 1
> >                # if it means the site has 1 type of species??)
> > 
> > require (FD)
> > ex1 <- dbFD (fd, comm, w.abun = FALSE, corr="cailliez", clust.type="ward")
> > ex1      # site 1 comes out with nbsp = sing.sp = 2, which is surprising
> > 
> > Any explanations / insight / advice would be greatly appreciated!
> > 
> > Many thanks,
> > Sarah
> > University of Chicago
> > 
> > 	[[alternative HTML version deleted]]
> > 
> > 
> > 
> > ------------------------------
> > 
> > _______________________________________________
> > R-sig-ecology mailing list
> > R-sig-ecology at r-project.org
> > https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
> > 
> > 
> > End of R-sig-ecology Digest, Vol 25, Issue 8
> > ********************************************
> 
> 


-- 
Etienne Laliberté
================================
School of Forestry
University of Canterbury
Private Bag 4800
Christchurch 8140, New Zealand
Phone: +64 3 366 7001 ext. 8365
Fax: +64 3 364 2124
www.elaliberte.info



More information about the R-sig-ecology mailing list