e-mails from Brad Boyle <bboyle@email.arizona.edu> on 2013-3-8:

-----

We are hoping to finish loading and validating the BIEN3 database in the next couple of weeks so we can re-run the species distribution models in April. For this first release of BIEN3, we are fine with using the same summary CTFS extract used for BIEN2, as long as it does not contain any known errors. I seem to recall there were doubts as to whether counts of individuals were aggregated correctly by species. Attached is an example of the first few lines from that data dump. If you wish to review the entire file, it is on vegbiendev.nceas.ucsb.edu in /home/bien/svn/inputs/CTFS/_src/ctfs_extract.tab. If you do not already have a shell account on that server, Aaron can set you up.

If you think this extract is error-free, then we will go ahead and use it for the first release of BIEN3. If not, then we would like to request a new extract.

After April, we would very much like to replace the CTFS data in BIEN with an updated and more complete CTFS extract containing raw stem measurements. Let us know if that is a possibility.

-----

One more clarification. One problem with BIEN2 was that the method used by Steve's students to aggregate individuals by species excluded morphospecies. Therefore, counts of species (and in some cases, of individuals) were incorrect. I'm not certain if this problem extends to the raw CTFS extract used for BIEN2, but if it does then we cannot use it. If it does not, and the extract is error-free, then I see no reason not to use it for the first release of BIEN3.

-----