Program on Mathematical and Statistical Ecology Working Group - Tipping Points with Forestry Applications

Group leaders: Philip Dixon, Lou Gross, Ron McRoberts

Description: The working group will consider both theoretical aspects of tipping points, critical thresholds and resiliency, as well as applications to forest degradation and other fields of interest. Topics are (i) connecting theory of resiliency to data sets; (ii) dealing with heterogeneous data at multiple scales to provide insight in resiliency of ecological systems; in particular, for forest systems the combination of plot data, imagery (Landsat, Lidar, Modis, etc.), topo data, soils data, etc. (iii) there is a new field of trajectory analysis that looks at very long time-series image data at the pixel level to try to identify when a change occured accurately; (iv) change detection in image data sets, emphasizing detection of fire and insect disturbances in forest systems. For any kind of large area approach, particularly in the tropics, a data combination that includes remotely sensed data is probably essential. Spectral data such as from Landsat or finer resolution RapidEye are possibilities as are lidar data where it can be acquired. For forestry applications, the American inventory data set has many possibilities. In some upper Midwestern and southern states, 4-5 measurements of the same trees on the same plots at 5-year intervals are available. However, all kinds of issues emerge when attempting to combine data from multiple sources, at multiple dates, and with multiple resolutions. Within the remote sensing community, some interesting work characterized as trajectory analysis is underway at Corvallis and also at Boston. This work follows individual pixels over time and focuses spectral thresholds indicating disturbance and recovery trajectories following disturbance.

Return to Ecology Program Home Page.