From: Bradley Boyle <bboyle@email.arizona.edu>

Subject: Morphospecies and scope

Date: Mon, 6 Feb 2012 16:57:01 -0700

Cc: BIEN Database <bien-db@iplantcollaborative.org>

To: Aaron Marcuse-Kubitza <aaronmk@nceas.ucsb.edu>

Hi all,

Just a quick example to illustrate my concern about morphospecies names.  Below is a flattened view of how we handle morphospecies names in the BIEN2 analytical database.

The fields family, species, and taxon contain various representations of the "standard" part of the scientific name (`taxon` is the lowest taxonomic level to which an organism has been determined).

`morphospecies` is the non-standard part of the name. The morphospecies string enables accurate counting of species within a plot, even if standard species names are not known. `morphospecies` strings in theory are of two types: "dependent" morphospecies must be combined with `taxon` to be unique. Most names in the above example are of this type. "standalone" morphospecies string are unique on their own. In the above example, "Boyle 6155" (a citation of a voucher specimen) is standalone. "Fuzzy Miconia" and Monsterleaf are also probably standalone names as well. To be safe, we always combine the `taxon` plus `morphospecies` to produce a unique string, `taxonMorphospecies`.

This of course is just a view. In terms of a schema, I have modeled morphospecies in the past as a text field in the table taxonDetermination. taxonDetermination still links a taxonOccurrence with a taxonName (taxonConcept, if you prefer), and taxonNames are *only* standard botanical names. The full taxonMorphospecies must be formed by concatenating the taxonName with the (optional) morphospecies string.

As you can see, my approach has drawbacks, among them the fact that the morphospecies name is never indexed as a name, but is added separately to each taxonDetermination. However, I do like keeping informal morphospecies strings separate from formal scientific names. I do not recommend we follow my approach; I just wanted to give a better idea of the issues involved.

Another important challenge with morphospecies names is "scope". Scientific names have global scope. Assuming it is used correctly to refer to the same concept, a scientific name has the same meaning everywhere. By contrast, a morphospecies name has a more local scope; typically a single plot. For example, Cestrum sp.1 in a plot of mine in Mexico almost certainly doesn't mean the same thing as Cestrum sp.1 in a plot of Peter Jorgensen's in Bolivia. However, there are cases where morphospecies names DO have a scope that encompasses more than one plot. This information, if available, is essential for determining species turnover among plots; without it, you cannot assess beta diversity and related measures. An example of morphospecies name scope is all my plots (13 of them) from Braulio Carrillo National Park in Costa Rica. "Miconia sp.3" means exactly the same thing in all plots.

A number of plot datasets have explicit morphospecies name scope. Is there a way we could capture this information in VegBIEN?

Brad