A comprehensive biotic inventory of the Gulf of Mexico (GMx), sponsored by the Harte Research Institute for Gulf of Mexico Studies, Texas A&M University-Corpus Christi, was completed in late 2007 after a four-year effort. A team of 140 expert taxonomists from 80 institutions in 15 countries compiled a checklist of 15,625 species encompassing all biodiversity, from viruses to vertebrates, found living in the Gulf of Mexico.
In addition to an updated taxonomy, the checklist also has information on habitats, ecology, depth, global and regional distribution and major references. Also included were data on endemic, endangered, and non-indigenous species. However, in order to attract such a broad cadre of taxonomists, a compromise had to be done, and, unfortunately, distribution was reported as quadrants in the Gulf of Mexico, instead of occurrence (point) data in a GIS. The compilation of the checklist and publication in book format was Phase I.
The checklist is currently being converted into a searchable online database to be hosted by GulfBase (www.gulfbase.org) and linked to the Ocean Biogeographic Information System (OBIS; www.iobis.org). Because OBIS currently is only interested in distribution data (the GulfBase interface will display other data), the distribution data for OBIS will be reported as coordinates for the centroid of each GMx quadrant in the first iteration (Phase II). Phase II is sponsored by a grant from the Alfred P. Sloan Foundation. Once the database is completed in 2009 and made freely available online, experts will be invited to review the accuracy of the database, and to submit additional habitat and/or occurrence data.
To improve the coarse resolution of distribution, a habitat prediction model will be constructed using GIS, bathymetry, habitat and ecological data (Phase III) to restrict potentially suitable habitats to polygons smaller than the quadrants reported in the original checklist. For example, the distribution of a mollusk species associated with coral reefs and living in a certain depth range could be restricted to a number of small polygons with potentially suitable habitat. The centroid of each polygon can then be used as a proxy for occurrence data for OBIS.
To test the predictive model, datasets with occurrence data will be compared with the results of the habitat prediction model. Another comparison will be made with AquaMaps (www.AquaMaps.org), a predictive model that uses life history data such as depth, water temperature, salinity, etc., to produce maps of habitat usage per species.