[R-sig-eco] Error with "raup" distance in metaMDS

Jari Oksanen jari.oksanen at oulu.fi
Sun Mar 27 17:07:36 CEST 2011


On 26/03/11 17:35 PM, "Manuel Spínola" <mspinola10 at gmail.com> wrote:

> Dear list members,
> 
> I got an error message working with Raup–Crick distance in metaMDS.
> 
>> flora.mds = metaMDS(indicadora, distance = "raup", trace = FALSE, zerodist
> = "ignore")
> Error in isoMDS(dist, k = k, trace = isotrace) :
>   zero or negative distance between objects 1 and 2
> In addition: Warning message:
> In distfun(comm, method = distance, ...) :
>   data have empty species which influence the results im method raup
> 
Manuel,

As I read it, the error message comes from isoMDS() that does not accept
zero distances, and you have decided to ignore this (by setting zerodist =
"ignore"). Either you or isoMDS must give up, and I think it is easier for
you: try setting 'zerodist = "add"', and confess that isoMDS() wins your
dispute.

Another message you got is from vegdist() which only tells you that empty
species (= that do not occur in your data) will influence your data. That is
only informational, and not an error.

Cheers, Jari Oksanen



More information about the R-sig-ecology mailing list