Rarefaction reduced the number of observed taxa


Both my feature table and taxonomy table have the same number of observed taxa (62521). However, I only had 33616 observed taxa after rarefaction. I thought the rarefaction only removes the samples that have fewer reads not removing the observed taxa. Why a half of my taxa were removed after rarefaction? Any suggestions would be very helpful.


I think there is a mis-understanding of what rarefying does. See here.

Also see this text from Schloss 2024:

With rarefaction, the investigator selects a desired threshold of sequencing effort and removes any samples below that threshold. They then randomly select that many sequences, without replacement from each sample.