I have a set of 331 point counts of nine different species across six sites. At some of the sampling points the total observations equals zero, so calculating the Shannon-Weiner diversity index is not appropriate. I wish to compare diversity among sites, so I need to account for the zero-count sampling points in the comparison among sites. Is there a preferred method for dealing with zero-count data in diversity comparisons? All suggestions are appreciated!
First I thought: What an odd question. Of course, it will be zero.
But facing that Shannon diversity is already zero with only one species in a sample, brings me to think about it a little bit more ...
Apart from this, the Simpson's Index is said to be more appropriate for species diversity. Shannon-Wiener is from information theory and has no connection to biological communities.
But there are even more interesting diversity measures, like taxonomic distinctness or rarefaction curves.
Hi Elizabeth,
I think you should go with Simpson index, I find it more suitable that Shannon and is easier to calculate in excel or by hand! Sites with zero individuals of any species means there is no diversity at all, so you write zero.
Cheers
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools