The basic task of inventorying biodiversity has actually been under way for many years. Existing natural history museum collections, like those in which we work, can provide major contributions to such inventories in the form of valuable historic organism occurrence records, and their specimens can be used in many ways for basic research and applied conservation planning. Unfortunately, much of the wealth of information stored in natural history collections requires substantial investment to be made accessible and useful to natural resource managers and researchers. We were charged by the GPLCC with providing some of the inventory data that will be required, and to assess what other data may be available and what will be required to make it useful. From databases that we and our collaborators (see Acknowledgments) manage, we compiled extensive, high quality data sets on occurrences of fishes, aquatic reptiles and amphibians (“herps”), freshwater mussels, and caveinvertebrates from the Texas, New Mexico, Colorado and Oklahoma portions of the GPLCC. We here deliver these 76,000 complete, standardized and normalized records, over 55% of them georeferenced and in a format that should make them immediately useful to the GPLCC. We also surveyed our colleagues and otherwise explored availability of other data sets for aquatic organisms in the GPLCC, providing 19 metadata records describing these additional resources. These metadata have been accepted by the National Biological Information Infrastructure (NBII) and will be published by that major metadata aggregating service to assure future availability to interested parties. We also mined the Global Biodiversity Information Facility (GBIF) for organismal occurrence records within the GPLCC and here provide those nearly 2 million records of over 27,000 species ranging from bacteria through fungi, plants and animals. Unfortunately only about 2% are georeferenced with precision estimates and much work would be required to standardize and georeference these records and make them useful to the GPLCC via applications such as those used in this project. Once the GPLCC obtains the extensive biodiversity inventories it requires, it is by no means easy to integrate such massive data sets into management planning. However, we demonstrate how raw occurrences for diverse sets of organisms can be effectively combined in computer models with diverse environmental data (including past, present and future) in ways that greatly facilitate planning at the landscape level. Our methods also allow incorporation of complex information on socioeconomic factors that in practice always complicate on-the-ground management into such planning. We do this by first developing powerful predictive computer models of each species’ distribution. These models provide a continuous coverage of probabilities of occurrence of each species for all cells of a fine-scale grid extending across the landscape of interest (the entire state of Texas in our demonstration), thus “filling in the blanks” between the actual occurrences that are limited by many factors such as historic factors, accessibility, and landowner permission. Our models were developed with recent occurrence records and recent climate data, and were thoroughly tested and demonstrated to be powerful predictors of actual occurrences under current conditions. While our demonstration was done statewide for Texas, it uses species that occur in, and are of particular interest to, the GPLCC, and our methods could be used by the GPLCC for its geographic area once appropriate occurrence data are obtained. However, we know the current conditions on which our models are based are not going to persist; climates are changing globally but, at least for the GPLCC reliable fine-scale predictions of exactly how they will change have not been available. We here provide a solution to the previous lack of high resolution regional climate change predictions by taking the most widely accepted and authoritative, and most recent, global predictions of the International Panel on Climate Change (IPCC) and regionalizing them, at high spatial resolution, for the GPLCC and for all of Texas. We were then able to replace the current climate data that went into our species distribution models with predicted future climate data, and thus compute how species’ distributions would shift if those climate predictions were realized. But, simply knowing how the climate-based habitat suitability for a handful of species might shift under predicted scenarios of climate change does not go a long way toward planning conservation of those and many more species indefinitely into an uncertain future, especially in complex socioeconomic settings that invariably limit management options. To illustrate how substantive progress can be made toward solving such exceedingly complex conundrums, we demonstrate how our species distribution models can be used together with current and predicted future environment and socioeconomic factors as input to a protocol for the selection of priority areas for biodiversity conservation. We use the powerful ConsNet conservation planning program to implement this protocol and produce a portfolio of priority area sets for conservation network planning. Initial results from ConsNet integrate a great diversity of biological knowledge and serve as a baseline starting point from which managers and policy makers can proceed by adding additional levels of multi-criteria analyses of other factors, such as habitat impaction and/or socioeconomic/ecosystem service cost-benefit parameters. With our sample data we demonstrate how, with ConsNet, planners can easily and interactively produce large numbers of variations of such results for diverse criteria of interest, thus providing a large variety of alternatives to consider for potential implementation. In summary, GPLCC support for this project enabled us to utilize fish occurrence data for Texas that we had been compiling, normalizing and improving for many years and apply it in the rigorous modeling, climate change and conservation network planning exercises reported here. These proof-of-concept demonstrations focused on Texas only because that is the area for which our previous projects provided the required high quality data. However, this project has now begun to compile the basic historic, current and future species occurrence and environmental data sets the GPLCC will need to perform such analyses for its own geographic scope, perhaps applying the same methodologies, data sets and tools we developed and provided in this project. We look forward to continuing to work with GPLCC to build and improve its data resources and tool set to help it address the complex issues it will face as it strives to attain its long-term conservation and sustainability objectives.
