Variability in vegetation and surface fuels across mixed-conifer-dominated landscapes with over 40 years of natural fire

Studies of historical fire and vegetation conditions in dry conifer forests have demonstrated a high degree of heterogeneity across landscapes. However, there is a limit to the amount of inference that can be drawn from historical fire reconstructions. Contemporary reference landscapes may be able to provide information that is not available from historical reconstructions. In this study, we characterized variability in vegetation structure and composition across two Sierra Nevada landscapes with long-established fire restoration programs. We used tree, shrub, and surface fuel data from 117 initial plots, 86 of which were re-measured 8-12 years later, to identify the mechanisms driving variability in vegetation and fuel conditions. Our analyses identified nine distinct vegetation groups, with mean live tree basal area and density ranging from 0.3 to 72.7 m(2) ha(-1) and 2.5 to 620 trees ha(-1) for individual groups. For all plots combined, mean live tree basal area and density was 28.4 m2 ha(-1) and 215 trees ha(-1), but standard deviations (SD) were 29.1 m2 ha(-1) and 182 trees ha(-1), respectively. These ranges and SDs demonstrate considerable variability in vegetation structure, which was partially related to site productivity and previous fire severity. Fine surface fuel loads were generally low (overall mean, 16.1 Mg ha(-1)), but also exhibited high variability (SD, 12.6 Mg ha(-1)). Surprisingly, surface fuel loads based on initial measurement and change between measurements were not related to fire characteristics. The only statistical relationship found was that surface fuel loads were associated with forest structure and composition. These results capture a contemporary 'natural' range of variability and can be used to guide landscape-level restoration efforts. More specifically, these results can help identify distinct targets for variable forest structures across landscapes. Published by Elsevier B.V.