New Reports Released on Predictors of Forest Harvest Behavior in Two Distinct US Regions
The Forest Carbon and Climate Program (FCCP) released two reports on the predictors of family forest harvest behavior in the Northeastern US and US Forest Service Ecoregion 212.
Kylie Clay, Associate Director for the FCCP, Chad Papa, Research Assistant for the FCCP, and Hunter Stanke, MSU Forestry PhD student, developed a random forest (RF) model to conduct the plot-level analysis key predictors of harvest likelihood (HL) and harvest intensity (HI) in two US regions. Their research focused on non-industrial private forestland in the Northeastern US (Connecticut, Maine, Massachusetts, New Hampshire, New York, Rhode Island, and Vermont) and USFS Ecoregion 212 which includes northern Michigan, Minnesota, and Wisconsin.
The objectives of these analyses were to:
- Identify covariates that best predict harvest likelihood (HL) and harvest intensity (HI) on non-industrial private forestland in Ecoregion 212 and the Northeastern US
- Identify appropriate subregions for analysis and Family Forest Carbon Program donor pool selection
- Identify tiers of plot-level carbon potential (i.e., predicted harvest intensity) according to key indicators to inform FFCP participation requirements or funding tiers (cap impact analysis)
- Assess the impact of selected caps on available donor plots by Forest Type Group and ecoregion (cap feasibility testing).
All data inputs are derived from the Forest Inventory and Analysis (FIA) database, the US Census, forest ownership data, and mill location data. Only plots encompassing privately-owned forestland were considered in the analyses (including tribal lands). This work was supported by the Family Forest Carbon Program which is a joint forest carbon project of the American Forest Foundation and the Nature Conservancy. These analyses help in better understanding and predicting family forest harvest behavior in two important regions of the US.
Learn more about the project below.
Click here to access the reports.