Comparisons when using the Forest Inventory and Analysis variables to quantity forest disturbances

Link To Share This Poster: https://cdmcd.co/QM8kP3

Lucia Alejandra Fitts, Forest Resources, University of Minnesota, Saint Paul, MN, Matthew Russell, University of Minnesota and Grant M. Domke, Northern Research Station, USDA Forest Service, St. Paul, MN

Background/Question/Methods
Forest disturbance regimes play a critical role in ecosystem dynamics, especially in forest health. However, quantifying forest disturbances at aggregated scales generally underestimates the disturbances that affect individual trees and subsequently affects the magnitude and scale of forest health problems. Tree-level disturbance variables have a high potential for early disturbance detection that directly affects forest management plans. This research focuses on reporting different methods to capture disturbances from the USDA Forest Inventory and Analysis database (FIA) using all of the remeasurements available since the start of the annual inventory for the lower 48 US states. The goals of the study are (1) to describe different methods for quantifying disturbances at different scales, (2) to compare the differences between disturbances variables, and (3) to provide a methodology for selecting a variable that represents scale of disturbance depending on the research or management objectives. Variables used included disturbance code, agent of mortality, and damage code. Chi-square tests of independence were used to verify how the choice of the variable that represents disturbance affects the magnitude of disturbance. Disturbed plots, as classified by each disturbance variable, were mapped to observe their spatial distribution. In addition, decision trees were built to provide a methodology for different research objectives.
Results/Conclusions
Chi-square tests were significant when using all the states together and when comparing each state individually, indicating that the different results exist depending on which variable is used to represent disturbance. Results indicate that the condition level disturbance variable captured fewer disturbance events compared to using the agent of mortality quantified at the individual tree level, even when we included a 25% threshold (similar to how the condition level disturbance is captured) for a plot to be considered disturbed. For disturbances such as fire, including the agent of mortality was not enough to capture the disturbances because smaller fires that would not kill a tree would not be represented (e.g., prescribed fires). Therefore, damage agent plus agent of mortality together results in a more comprehensive method of representing a disturbance. Our results will be a useful tool for researchers using the US forest inventory data as it will inform the magnitude and scale of disturbance. The manner in which disturbances are categorized will impact national and international reports of forest carbon stocks and sequestration potential under future global change scenarios.