Digging around in the rarefaction literature even a little will find a decent amount of discussion on unconditional vs. conditional confidence intervals on rarefaction curves. The short version of the argument is given here http://stevencarlislewalker.wordpress.com/2013/02/25/a-common-common-common-mistake-in-rarefaction-analysis/. Essentially, the commonly used conditional variance calculations assume that the sample being analyzed is complete rather than a subsample of a larger population, so CIs close at the largest sample size. Unconditional variance calculators more accurately treat the analytical sample as a subsample of the the larger population, so the CIs remain open (there’s probably undetected species out there). EstimateS will calculate unconditional CIs on what amounts to a shared file and perform multiple randomizations, but I cannot figure out how to specify the size of the subsample it uses. It would be great if this was an option in mothur!
Relevant papers:
Colwell, R. K., C. X. Mao, & J. Chang. 2004. Interpolating, extrapolating, and comparing incidence-based species accumulation curves. Ecology 85, 2717-2727.
Colwell, Robert K., et al. “Models and estimators linking individual-based and sample-based rarefaction, extrapolation and comparison of assemblages.†Journal of Plant Ecology 5.1 (2012): 3-21.