, 2004, Van Pelt, 2008 and Shuffield, 2011). Using CVS data to estimate current forest conditions, abundance of large trees decreased by almost 50% while basal area in large trees decreased by 64% since the time of the timber inventory (1914–1922, Table 5). The percentage AZD5363 cost of the area inventoried that supports at

least 25 large-diameter tph (>53 cm dbh) decreased by 70%, and the mean proportion of ponderosa pine in large-tree basal area decreased by 53% on Dry Mixed sites and 44% on Moist Mixed sites (Table 5). The contemporary estimates of large tree abundance contrast markedly with both the population levels of large trees and the collective area supporting at least 25 tph > 53 cm dbh that we found in the historical forests (Table 4 and Table 5). One important artifact of the scale at which the data were recorded (1.6 ha transect for 1920–1922 or four 1.6 ha transects from 1914 to 1919) is the ubiquitous mix of tree sizes which might lead one to infer that large areas of single-story older forest were absent. Unfortunately, at the coarse scale of this inventory,

any fine-scale patterning would not p38 MAPK inhibitor be apparent. The majority of the variability in structure in frequent-fire forests has been observed at spatial scales smaller than 0.4 ha (Larson and Churchill, 2012). The scale at which the inventory data were recorded homogenizes this patchiness, which has been shown to include widely spaced individuals, clusters of large trees, dense patches of regeneration, and

small openings (Franklin and Van Pelt, 2004 and Larson and Churchill, 2012). This fine-scale patchiness is still evident today in ponderosa pine sites on the 3-mercaptopyruvate sulfurtransferase Reservation that have not been either intensively logged or burned (Johnson et al., 2008). The capacity for records and reconstructions of historical forests to represent conditions on a larger landscape has been questioned due to potential subjectivity in site selection and limited spatial extent (Bell et al., 2009). This timber inventory, consisting of transects systematically located to provide a 10–20% sample of the Reservation forests from lower to upper timberline, overcomes both of those limitations and is a record – not a reconstruction – of tree density by diameter and species for trees ⩾15 cm dbh. A landscape overwhelmingly occupied by low-density forests and dominated by large trees and fire- and drought-tolerant species is evident from these records. This historical landscape is consistent with most of the other reconstructions and records of historical forest conditions in central Oregon (Munger, 1917, Perry et al., 2004, Youngblood et al.