Skip to content
Prev 276669 / 398506 Next

ordination in vegan: what does downweight() do?

On Mon, 2011-11-07 at 10:24 -0800, kelsmith wrote:
That is not what downweight() is for. If you want this sort of
selection, see chooseTaxa() in my analogue package

diat.sel <- chooseTaxa(diatoms, max.abun = 1, type = "AND")

or, if proportions not percent

diat.sel <- chooseTaxa(diatoms, max.abun = 0.01, type = "AND")

This sort of indexing is trivial (but I made a typo in the current
version so type = "OR" won't work) so you can study the code of
analogue:::chooseTaxa.default once I fix the CRAN version or on R-Forge
now:

https://r-forge.r-project.org/scm/viewvc.php/pkg/R/chooseTaxa.R?view=markup&root=analogue

Jari has addressed the other part of your question.

Jari also mentioned the issue about whether you should be removing or
downweighting rare species. Many people, especially diatomists, do this
for practical purposes in a routine fashion because their data sets are
especially speciose and have a large proportion of low abundance taxa.
As a general matter of routine practice, I don't think this is a very
good way of working, especially as we have no good ecological grounds
for doing so and who knows what information these species could be
telling us if we just listened to them instead of discarding them.

HTH

G