Search
Browse by Subject
Contact Information

Northern Research Station
One Gifford Pinchot Drive
Madison, WI 53726
(608) 231-9318
(608) 231-9544 TTY/TDD

Sustaining Forests

Landscape Ecological Modeling: Species Habitat Modeling under Future Climates - DISTRIB

Research Issue

[image:]  Diagram shows modeling process.  Click on image to see larger version.

Combining spatially distributed data related to organisms, climate, soils, and topography (as collected from ground surveys, satellites and geographic information systems) with statistical modeling techniques provides ever-increasing capabilities in modeling past, present, and potential future ecosystems at multiple scales. This research team specializes in this type of modeling, including understanding the potential consequences of climate change on the habitat of 134 tree and 147 bird species across the eastern United States. Using empirical data, we derive statistical species-based models of current habitat associations based on climate and landscape characteristics. These results provide a statistically based prospective of the potential future habitat for trees and birds.

Our Research

We adopt an empirical approach using three decision-tree based ensemble techniques (here called TriMod) to model the abundance of tree species (derived from Forest Inventory and Analysis (FIA) data) using 38 climate, elevation and land-use predictors. We model the current distribution as well as potential suitable habitat changes under three Global Circulation Models (GCMs) - HadCM3, PCM, GFDL under scenarios of high and low emissions as defined by Inter-Governmental Panel on Climate Change (IPCC). We used the TriMod approach to relate these predictors to the abundance of tree species, including RandomForest for prediction, Bagging trees to determine model stability and reliability, and a single regression tree (if a reliable model) to map which variables were important in different geographic areas. The future climate predictions were modeled by swapping current climate with GCM derived future climates.

DISTRIB enabled us to predict the potential suitable habitats via abundance changes for 134 tree species for current and modeled future climates. From these, we derived forest type change maps, hotspot changes, ranked species tables and a host of other relevant outputs as compiled in our climate change tree atlas.

Research Results

Web Products

Research Publications

Iverson, Louis R.; Prasad, Anantha M.; Matthews, Stephen N.; Peters, Matthew. 2008. Estimating potential habitat for 134 eastern US tree species under six climate scenarios. Forest Ecology and Management. 254: 390-406.

Prasad, Anantha M.; Iverson, Louis R.; Liaw, Andy. 2006. Newer classification and regression tree techniques: Bagging and Random Forests for ecological prediction. Ecosystems. (9): 181-199.

Iverson, Louis R; Prasad, Anantha M.; Schwartz, Mark W. 2005. Predicting Potential Changes in Suitable Habitat and Distribution by 2100 for Tree Species of the Eastern United States. J. Agric. Meteorol. 61(1): 29-37.

Iverson, Louis R.; Prasad, Anantha M. 2001. Potential Changes in Tree Species Richness and Forest Community Types following Climate Change. Ecosystems. 4: 186-199.

Iverson, Louis R.; Prasad, Anantha M.; Hale, Betsy J.; Sutherland, Elaine Kennedy. 1999. Atlas of current and potential future distributions of common trees of the eastern United States. Gen. Tech. Rep. NE-265. Radnor, PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station. 245 p.

Research Participants

Principal Investigators

Research Partners

 

Some documents linked to from this page are in PDF format. You may obtain a free PDF reader from Adobe.

Last Modified: 10/21/2010

About this Research Area
DISTRIB Model Resources
Featured Product
Related Links
More Information

Did not find what you were looking for? You may search our Station's data base for free online publications related to:


You may also search the entire NRS publications database by author, keyword, or title.