The findings of two CNR researchers published recently in the Procedings of the National Academy of Sciences, contain several take-home lessons for conservation biologists and land managers.
One of the primary theories in island biogeography states that the size of an island and its degree of isolation are proportional to the amount of biodiversity the island can support. For oceanic islands such as Hawaii and Easter Island, this theory appears well supported by hard data. Ecologists have also tried to apply the same reasoning to continental ecosystems: Certain patches of land will have features such as a specific plant or certain environmental conditions that make it good habitat for a given species, and these patches are surrounded by relatively inhospitable lands that lack these amenities.
According to the theory of island biogeography, the likelihood of finding a given species on one of these habitat patches should be proportional to the size and degree of isolation of the patch. While biologists have gathered data about patch size and habitat suitability for individual species, no one had ever conducted a large scale meta-analysis (a study based on data from other studies--forgive the talking down, not sure whether you were familiar with this term) of whether this theory actually applies to terrestrial eosystems.
Laura Prugh, a postdoctoral researcher in the laboratory of Justin Brashares, decided to test how well the theory applied by examining studies describing habitat patches for more than 700 species. What she found was that patch size and isolation could account for only 25 percent of the likelihood of a species being found in a certain patch. While this wasn't a shocker per se, as it doesn't contradict any previous findings, that's a pretty paltry number. It indicates that patch size and isolation in and of themselves aren't the greatest predictors of good habitat.
Instead, the factor that was more likely to influence whether a species was present in a certain patch was the quality of the land outside of the designated habitat patch, or the matrix. Species were much more likely to be found in areas where the matrix was of relatively high quality, such as an area that had not been clear cut, or an area with gardens or hedgerows, compared to habitats that are surrounded by clearcuts or parking lots.
It may be that biologists aren't that good at delineating habitat patches. But it also makes intuitive sense. Matrix areas can function as corridors, or as hiding places, or can provide innumerable habitat amenities that biologists might not have noticed in their assessments of the land. For bees, this might mean garden flowers that bloom at lean times of the year; for salamanders this might mean a field where they can hide in summer, because they only need streams and ponds when they reproduce.
All of this makes intuitive sense, yes, but no one had quantified these effects before.
These findings contain several take-home lessons for conservation biologists and land managers. In a world where pristine, undisturbed habitat is becoming ever harder to find, those seeking to preserve biodiversity can maximize the value of existing reserves by working with communities to improve the habitat quality of surrounding land. And when humans go to alter landscape, installing hedgerows and choosing shade-grown coffee plantations and opting against clear cutting can go a long way towards helping species remain on the land. While this effect is obviously more pronounced for smaller species such as songbirds than larger predators such as bears, it could, in the long run, have a significant impact on how we manage our environment and preserve biodiversity.
The paper was important enough to merit a commentary in the same issue of PNAS. The commentary essentially takes Prugh's findings that extra mile to relate it to land management issues.
While not all biologists will necessarily agree with Prugh's findings (everybody has their own theory to uphold) the sheer size of her original data pool suggests her findings are quite reliable.--Kathleen M. Wong