How Many Species Are There?
How many species are there? While the question has an almost childlike simplicity, the answer has proven elusive. Biologists agree that the list of known species—numbering about 2 million—is woefully incomplete. Estimates of how many more species exist range from 5 to 50 million, a practically meaningless span.
Biologists have long sought to identify areas where effective conservation could save the most species. Biodiversity hotspots—places with extreme rates of habitat loss as well as unusually high numbers of endemic species—are priorities. But with so many species as yet unknown, we have to wonder: could and would their discovery change those priorities?
Moreover, those unknown species are likely to have small geographic ranges and to be rare within their habitats, and thus they would be prime candidates for extinction. So by figuring out how many species are “missing” from the record, Lucas Joppa, an ecologist in the Computational Science Laboratory at Microsoft Research—together with Stuart Pimm, a world-leading expert in conservation ecology at Duke University, and David Roberts at University of Kent—are providing quantitative estimates for how many more species might be threatened and endangered.
The unique insight that the team brings is the acknowledgment of an inherent social dimension to the process of species description. People (taxonomists) describe species, so the number of species described must surely depend on the number of people actively describing them. By incorporating human effort into our statistical model, we can predict, with measured confidence, the numbers of species remaining to be discovered.
Stuart Pimm and Lucas Joppa defined and refined the species model from a quantitative viewpoint, providing deep insight into how one might potentially account for unknown species by using a novel proxy parameterization around the number of taxonomists in a given field. Then, by using new scientific software tools and technologies being developed by Microsoft’s Computational Science Laboratory, they are taking massive amounts of hugely dispersed data, bringing them together in a computationally powerful manner, and applying statistical models to make predictive assessments of the total number of species.
The new approach to conservation science is helping to push the boundaries of current spatial database technology. The project uses Microsoft SQL Server in novel ways that change how we think about analysing spatial data, helping to bring new insights across a range of application scenarios. The researchers are working with the product team to push the limits of SQL Server’s capabilities, while their feedback is contributing to the development process for future versions, including SQL Azure.
The future of human life depends upon the consequences of the massive extinction crisis we currently face. Our unknown-species estimates present quantitative evidence that can improve the effectiveness of our environmental regulation and conservation—and help the planet.
Learn more about this research:
- Biodiversity hotspots house the majority of missing species
- How many species of flowering plants are there?
- The population ecology and social behaviour of taxonomists