We defined forests across our study area as those described as a “forest” or “forest and woodland” land cover class in the biophysical setting model. National Forest System lands are typically considered “forest” if they have >10% tree canopy cover, and this generally coincides with forest, and forest and woodland land cover classes

(USDA Forest Service, 2004). Each biophysical setting model is composed of a suite of 3–5 successional/structural stages (s-classes). These classes typically include: (A) Early Development, (B) Mid-Development Closed Canopy, (C) Mid-Development Open Canopy, (D) Late Development Open Canopy, and (E) Late Development Closed Canopy. The definition of SRT1720 datasheet each s-class in terms of species composition, stand structure, and stand age is unique for each biophysical setting (Appendix A.2). The percentage of a biophysical setting in each s-class will differ depending on disturbance frequencies and/or intensities. The LANDFIRE and FRCC conceptual framework assumes that, given natural processes, a biophysical setting will have a characteristic range of variation in the proportion in each s-class and that an effective indicator of “ecological condition” for a given landscape is the relative abundance of each s-class within biophysical settings (Barrett et al., 2010 and Keane Afatinib et al., 2011). NRV reference models describe how

the relative distribution of s-classes for a biophysical setting were shaped by succession and the frequency and severity of disturbances prior to European settlement and provide a comparison to present-day forest conditions (Keane et al., 2009 and Landres et al., 1999). LANDFIRE biophysical setting models are used to develop NRV estimates through the use of state-and-transition

models incorporating pre-European settlement rates of succession and disturbance. Rates were determined through an intensive Mannose-binding protein-associated serine protease literature and expert review process (Keane et al., 2002, Keane et al., 2007, Pratt et al., 2006 and Rollins, 2009). The distribution of s-classes for each biophysical setting which results from running state-and-transition models for many time-steps (Appendix A.3) does not represent a specific historical date, but instead approximates characteristic conditions that result from natural biological and physical processes operating on a landscape over a relatively long time period. NRV is frequently represented by a single value, the mean relative abundance of each s-class from a collection of Monte Carlo state-and-transition model simulations (e.g., Low et al., 2010, Shlisky et al., 2005 and Weisz et al., 2009). However, we extended this method by developing and using ranges for each s-class resulting from the stochastic variation around the mean within the state-and-transition models.