Bigfoot. Sasquatch. Whatever you want to call the legendary North American biped, it is likely the elusive beast will lose a portion of its existing habitat in the coastal and lowland regions of the northwestern United States as the climate warms. The good news? Bigfoot will be gaining a bunch of new land in the Rockies and up into Canada.
In a paper recently published in the Journal of Biogeography, biologist Jeff Lozier of the University of Illinois at Urbana-Champaign and his colleagues argue that a potential source of error in publicly available data employed in a commonly-used statistical modeling technique to predict the “ecological niche” of species may affect the accuracy of ecological niche models.
Using a database of sightings and footprints for Bigfoot in western North America, the researchers suggest that convincing distributions of an animals range can be generated from questionable data. By comparing the distribution of Bigfoot to that of a black bear, Lozier et al. “suggest that many sightings of this cryptozoid may be cases of mistaken identity.”
In other words, it’s not Bigfoot that is really the focus of the research by Lozier and his team, but rather the methodology used and the assumptions drawn from incomplete data.
Biogeographers are among the many scientific disciplines that have been employing increasingly sophisticated computer algorithms to predict the ecological niche of species. The algorithms take information about sightings or recorded incidences of a species, find commonalities among those sightings against maps of other ecological data (i.e rainfall, forest type, presence of other species, etc.), and produce a geographic distribution for the target species.
The paper, “Predicting the distribution of Sasquatch in western North America: anything goes with ecological niche modelling,” constructs ecological niche models (ENMs) for the elusive Bigfoot. By using a large database of georeferenced sightings and footprints for Sasquatch in western North America, Lozier and his colleagues aim to demonstrate how convincing environmentally predicted distributions of a taxon’s potential range can be generated from questionable site-occurrence data. In this case, a data set that relies on sightings from unreliable sources that might actually be confusing black bears with Bigfoot can produce even more questionable results.
“This Bigfoot paper is really good,” writes anthropologist John Hawks, who notes that the authors intended the piece as a tongue-in-cheek’ example, and an illustration of the problems presented by the ‘garbage in, garbage out’ principle.
Lozier et al. do not take an explicit stance on the existence of Bigfoot, but rather make use of publicly available data sets with questionable records to illustrate the danger of using incomplete data to make statistical correlations.
“In our paper we simply note that the black bear results do suggest a parsimonious explanation for many sightings for those inclined to be skeptical of the creature’s existence” writes biologist Lozier in a follow-up to the paper.
Lozier explains that he and his colleagues are not opposed to the use of niche modeling technology per se, but rather the improper use of—and over reliance on—it as a scientific tool.
via NatureNews
Follow Tim Hurst on twitter
