Causal factors and consequences of mixed- severity fire in Black Hills ponderosa pine forests

Abstract

We used the ~ 34,000 ha Jasper fire of 2000 to address the theme of mixed- severity fire in Black Hills ponderosa pine forests. Mixed severity fire resulted in many patches of low, moderate, and high severity, distributed in roughly equal proportion across the burned landscape. There was a significant effect of stand structure and topography on burn severity despite the extreme weather conditions during the Jasper fire. Relative density, tree size, and slope, ranked in order of importance, were influential in predicting burn severity. Relative density < 40 % and slope < 18 % represent critical thresholds where low and moderate severity were more likely to occur. High stand density presented the most likely scenario for crown or high severity fire. Our conclusion that structure does influence burn severity has important implications in terms of fuels management and in understanding patterns of vegetation that influence burn severity even in events occurring during extreme weather conditions. The Jasper fire left a landscape imprint distinct from other recent, large, western wildfires (e.g. the Yellowstone fires, WY, 1988, the Hayman fire, CO 2002, the Rodeo-Chedeski fire, AZ 2002, and the Biscuit fire, OR and CA, 2002). Rather than small patches of surviving vegetation in a matrix of crown fire, we observed small patches of crown fire interspersed among large patches of surviving vegetation. We inferred variable fire effects on aspects of canopy, understory plant communities, forest floor, and soil and differential rates of recovery to be the result of different burn severity. Tree mortality increased from ~ 1 to 16 % and 18 to 59 % in low and moderate severity patches during the first 3 years post-fire. All trees were killed in high severity patches. Greater reductions in plant cover were seen with greater severity. Herbaceous cover and richness recovered to pre-burned levels in all burned areas within three years post-fire. Fire initially removed aboveground shrub components, independent of severity, although shrub cover was ~ 40 % of pre-burn cover in high severity areas three years post-fire. Pre-fire species composition was highly influential in determining post-fire species composition, and we found ~ 70 % of species regenerated vegetatively. We found three times more non-native plant species following fire. Soil nitrogen availability was 21 and 41 times greater in areas of low and high severity than in unburned areas, but decreased two years post-fire. Tree mortality and associated changes in light transmission are ongoing in areas with surviving trees. Tree regeneration was limited and variable in the first three years post-fire. Many small patches of low, moderate, and high burn severity with a high proportion of surviving vegetation accelerated rates of recovery in the Jasper fire compared with other recent and large western fires. Fire histories in ponderosa pine (Pinus ponderosa var scopulorum) forests are reconstructed from fire scar and tree origin information. Fire frequency, severity, and size are inferred from dates of fire occurrence as recorded as scars on individual trees and from cohort origin and structure. However, little is known about rates of fire scar formation in relation to burn severity and stand conditions, and critics have proposed a modern calibration to validate sampling and interpretation of fire history evidence. We examined tree mortality, incipient fire scar formation and ponderosa pine regeneration in patches of low, moderate, and high burn severity 2-3 years post-fire. Two years post-fire, tree mortality was ~ 6, 24 and 100 % in low, moderate and high burn severity. Three years post-fire, tree mortality had increased to ~ 21 and 52 % in areas of low and moderate burn severity. Two years post-fire, we examined ~ 2100 live trees for evidence of dead cambium within low and moderate severity patches. Dead cambium on a significant portion of tree circumference in a tree with live cambium and a vigorous crown was taken as evidence of incipient fire scar formation. Dead cambium was detected on ~ 24 and 44 % of surviving trees in low and moderate burn severity patches. Regeneration densities were ~ 531, 796, and 11 seedlings ha-1 in low, moderate, and high severity patches two years post-fire. Three years post-fire, regeneration densities were ~ 612 and 450 seedlings ha-1 in low and moderate severity patches, and we observed no regeneration in high severity patches. Tree-dominated cover may not return to the interior of large patches where the distance to seed source exceeds ponderosa pine seed dispersal distance for a long time. The pattern of future fire-scarred trees and post-fire cohorts that resulted from the Jasper fire is indicative of a mixed- severity fire regime consistent with fire history reconstructions in Black Hills ponderosa pine forests.