Predicting Potential Changes in Suitable Habitat and Distribution by 2100 for Tree
Contributor(s)
The Pennsylvania State University CiteSeerX ArchivesKeywords
Climate changeKey words
Migration
Predictive vegetation mapping
Tree species
Eastern United States
Full record
Show full item recordOnline Access
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.538.8193http://www.fs.fed.us/ne/newtown_square/publications/other_publishers/OCR/ne_2005_iverson001.pdf
Abstract
We predict current distribution and abundance for tree species present in eastern North America, and subsequently estimate potential suitable habitat for those species under a changed climate with 2 x COz. We used a series of statistical models (i.e., Regression Tree Analysis (RTA), Multivariate Adaptive Regression Splines (MARS), Bagging Trees (BT) and Random Forests (RF)) via our model, DISTRIB, for this purpose. These techniques were evaluated on several tree species, and advantages and disadvantages of each method were noted. R F provides the best prediction maps of potential suitable habitat. Overall, a combination of RTA, BT, and RF may yield the best information and most interpretable maps of suitable habitat. Using these tools, we provide statistics on potential changes in suitable habitat for 135 tree species of eastern North America. A suitable habitat does not guarantee the presence of a species, as many barriers for the species still exist before it will be able to colonize that new suitable habitat. Dispersal ability, abundance of the colonizing species, and the nature of fragmented landscapes also influence migration and are modeled with OUT cellular automata model, SHIFT. For each cell outside a species ' current boundary, SHIFT creates an estimate of the probability that each unoccupied cell will becomeDate
2005Type
textIdentifier
oai:CiteSeerX.psu:10.1.1.538.8193http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.538.8193