Missing Pieces

Want to know if that Amazonian orchid you love so much is likely to survive a warming climate? Don’t hold your breath. Efforts to create models that predict how distributions of tropical species might shift due to climate change are “potentially crippled by a lack of basic data” on equatorial species, concludes a new analysis.

In recent years, researchers have turned to a technique called species distribution modeling (SDM) to predict how much habitat a particular species might gain, or lose, in a warming world. One recent study, for instance, found that a 2 degree Centigrade increase in temperature could drive up to one-third of the Amazon’s tree species to extinction. But that study rested on analysis of just nine tree species, Kenneth J. Feeley of Florida International University in Miami and Miles R. Silman of Wake Forest University in North Carolina note in Global Change Biology. That’s probably too few to get robust predictions, they argue. “It is clear that we must expand studies… to include a much broader and representative sample of species if we hope to understand, predict, and eventually mitigate the effects of global change on tropical biodiversity.”

That “representative sample,” however, may be hard to get. To see just how much data on tropical species modelers could get their hands on, Feeley and Silman collected every record on tropical vascular plants they could find in the Global Biodiversity Information Facility, the world’s largest online depository of the holdings in the world’s museums and herbaria. Then, they added records from SpeciesLink, which holds information on Brazilian collections.

Overall, they ended up with nearly 800,000 records that showed where more than 100,000 species had been collected. Nearly 40% of the species, however, were represented by just a single record, and the typical plant had just two records. And plants from “large expanses” of tropical Africa, Asia and South America, “including entire countries,” did not show up at all. “These underrepresented areas contrast with other regions, such as Ecuador, where relatively large numbers of collections are available online.”

The troubling reality, they concluded, is that “nine out of 10 species from the three principle tropical realms are so poorly collected (less than 20 records) that they are essentially invisible to modern modeling and conservation tools.”

What would it take to fill the data void?  “We argue that the answer lies principally in increasing the amount of raw data available on species occurrences,” the authors write. “This involves the unglamorous but crucial work of increasing the number of herbarium collections from the tropics.”  More and better-documented tropical specimens could make “biodiversity management in the tropics a predictive exercise,” they note, “one clearly preferable to the alternative of having future ecologists examine extinctions retrospectively.” David Malakoff | January 27, 2011

Source: FEELEY, K., & SILMAN, M. (2011). The data void in modeling current and future distributions of tropical species. Global Change Biology, 17 (1), 626-630 DOI: 10.1111/j.1365-2486.2010.02239.x

Image © Troscha | Dreamstime.com