Postfire Tree Mortality in Relation to Wildfire Severity in Mixed Oak Forests in the Blue Ridge of Virginia

Abstract

Listen

Translate article

Abstract Tree mortality in relation to fire severity was investigated in eight mixed oak stands in the second growing season following the 1900 ha Big Run Fire in May 1986, in Shenandoah National Park in the Blue Ridge of Virginia. Plots were located in four stands each burned by high- and low-severity fire. Prefire stands were evenaged (55-60 yr), of medium quality (upland oak 50 yr site index 18-20 m ), and dominated by chestnut oak, scarlet oak, northern red oak and pignut hickory. High-severity fire top-killed 67% of stand basal area and 81% of the trees, whereas low-severity fire top-killed 8% of stand basal area and 15% of the trees. Height of stem-bark char accounted for 93 % and 96% of the variation in fire-killed basal area and number of trees, respectively. Logistic regression models relating probability of top-kill of individual trees with tree dbh and height of stem-bark char were developed for chestnut oak, scarlet oak, northern red oak, black oak, pignut hickory, red maple, blackgum, and serviceberry. Regression diagnostics indicated that the models predicted well for all species, with 79 to 95% concordance between predicted probabilities and observed outcomes. The models demonstrate the interaction of species-specific fire resistance, tree size, and fire severity in determining fire induced top-kill. These models can be used to estimate postfire mortality for a variety of purposes, including planning postwildfire timber salvage and regeneration operations, planning prescribed fires, and ecosystem process modeling. North. J. Appl. For. 11(3):90-97.