Hello!
I have performed a number of vegetation surveys on alpine pastures, within two treatments (A and B). Namely, I have 64 surveys from treatment A and 178 from treatment B. For every survey, I recorded the presence and abundance of every plant species.
Now, I would like to assess the number of species found exclusively within A and within B, respectively (and consequently the number of species common to both treatments).
Roughly in my 64 surveys (treatment A) I have 59 species not in common with tratment B, and in my 178 surveys (B) I have 81 species not in common with A, while 249 were species found within both treatments.
Do I have to apply any sort of correction linked to sample size (A = 2.8 x B) to assess the actual number of exclusive species for both treatment?
Many thanks for any suggestion.
SR
Hello Simone,
It's an interesting question. Descriptively, you've given the results obtained, and that may be sufficient for your research aims. If you wanted to know whether more surveys (esp. for treatment A) would have increased the number of unique species (and whether the appropriate "blow-up" estimate would be on the order of 2.8X), it would be helpful to know this: How many unique species are identified in each survey (from #1 to #64)?
For example, if the first three surveys picked up 50 unique species, and the subsequent 61 surveys (treatment A numbers here) only added 9 more to this initial set, then the expected yield for additional surveys is on the order of N/61 * 9 or fewer (where N is the proposed additional number of surveys). So, it's not as simple as saying, 64 surveys yielded 59 unique species, so N additional surveys would yield N*(59/64) more unique identifications. And that example presumes that surveys are independent of one another, which may not be the case.
Good luck with your work.
Hello Simone!
I had same problem dealing with insect species richness. Usually when I want to compare species richness between study areas differing in sampling efforts, I estimate rarefaction curves per study area. The curves allow to estimate if the survey was able to detect most of species and if certain areas have more new species detected per each further sampling round. I saw, doing a fast search in Google Scholar, that such method is used in plant studies as well (e.g Article Rarefaction method for assessing plant species diversity on ...
). I do not know if this method could be applied for different species groups (unique species per area/ common species between areas).- Badges
- Science topic
- Similar topics
- Correction