Forest structure and fire susceptibility in volcanic landscapes of the eastern high cascades, Oregon

Multidimensional forest health initiatives-those designed to restore and protect forest integrity in the large sense-require, among other things, a better understanding of the relationship between the structure of forests and (1) their susceptibility to wildfires outside the range of natural variability and (2) the level of treatment (if any) required to lower susceptibility into an acceptable range. Within the ponderosa pine zone (Pinus ponderosa Dougl. ex Laws.) on the eastern slopes of Oregon's Cascade Range, an average of 86% of trees >5 cm in diameter at breast height (dbh) were less than 101 years old, the density of young trees correlating negatively with the density of old trees (> 150 yrs). Species composition differed significantly between age classes: 72% of trees older than 150 years were ponderosa pine, whereas 83% of trees younger than 101 years were either grand fir (Abies grandis [Dougl.]Lindl.) or lodgepole pine (Pinus contorta Dougl. ex). Loud Crown bulk density calculated to reflect the multiple canopy layer structure of these stands (CBDm) averaged 0.049 kg/m(3), compared to 0.093 kg/m(3) using the standard approach (p < 0.001), indicating that the latter significantly overestimates risk of active crown fire in stands with complex canopy structure. Nevertheless, modeling with CBDm predicts that under severe but plausible weather conditions (late summer fuel moisture and 48-kph winds) crown kill would exceed 70% on 5 of 14 plots and 50% on an additional 5 plots. Models also predict that thinning trees of <20 cm dbb, coupled with controlled burning to reduce logging slash, would prevent torching (fire moving from the ground into crowns) on all plots, even under extreme fire conditions (low fuel moisture and 80-kpb winds). The amount of thinning required to prevent spread from crown to crown (active crown fire) varied widely among plots and depended on the degree of risk deemed acceptable. The relationship between stand basal area and the critical wind speed required to maintain active crown fire followed a negative power law (r(2) = 0.92), implying that over certain ranges small changes in basal area result in relatively large changes in susceptibility (assuming that changes in basal area are due to addition or removal of smaller trees). Landsat thematic mapper correlates reasonably well with CBDm (r(2) = 0.80) and may provide a tool for rapid risk assessment.