|Abstract or Summary
- Though the mixed-evergreen forests of the Klamath Siskiyous have a long history of large, mixed-severity fires, most research in this region has concentrated on the impacts of high-severity fire. Knowledge of the ecological effects of low- and moderate-severity areas within mixed-severity fires is important because such areas may account for over half the landscape affected by a fire. The purpose of this study was to understand the relationship of fire severity with dead fuels and understory vegetation across a full range of fire severities. Study sites were located within and just outside the boundary of the 2002 Biscuit Fire, which burned 200,000 hectares in a mosaic of burn severities.
Six years after the Biscuit Fire, the biomass and depth of litter and duff was lower on burned sites than unburned sites, and lowest on high-severity sites. This relationship was reversed for woody fuels >7.62 cm in diameter, where quantities were highest in high- and extreme-severity areas, though there was no evidence that quantities differed between low-severity and unburned sites. There was no evidence of a relationship between woody fuels 0.64-7.62 cm in diameter and fire severity, 6 years post-fire. There was no evidence that fuel quantities differed between sites that burned only in the Biscuit Fire with sites that also burned 15 years earlier in the 1987 Silver Fire. Fuel quantities and composition differed between burned and unburned sites, but these differences disappeared if litter and duff were not
considered. Fuel classes were correlated with each other within three general size classes: small (litter, duff, and fuels <2.54 cm), medium (fuels 2.54-30 cm), and large (fuels >30 cm). There was little correlation between these size classes.
Vegetation response also varied by fire severity, species, and height. Generally, density for tree seedlings <0.5 m was highest on low-severity sites and lowest on high-severity sites. For seedlings 0.5-1.37 m the relationship was reversed, with the highest seedling densities in high-severity areas. Specific seedling relationships to burn severity and other explanatory factors (e.g. shrub cover, elevation, precipitation, maxiumun August temperatures) varied by species and seedling height. Average seedling densities were above the minimum acceptable stocking levels of 333 trees per hectare (135 seedlings/acre) as identified in federal plans for the fire area. Shrub species richness and diversity did not vary with burn severity; however, shrub species and ground cover composition did differ with burn severity six years after the Biscuit Fire. The relationship of understory cover composition with burn severity aligned with species life history traits; fire-adapted, nitrogen-fixing species were more prevalent on high- and extreme-severity sites while species associated with low-severity fire or old-growth forests were more prevalent on low-severity or unburned sites. I found no evidence of difference in total graminoid or forb cover based on fire severity, but shrub quantities were lower on low-severity sites than on unburned, high-, and extreme-severity sites.
This study reveals how fuel and vegetation vary across the full range of fire severities. It demonstrates that mixed-severity fires create a mix of ecological responses. It also provides a baseline for future studies, as the relationships between fire severity and fuels or vegetation may disappear or change in subsequent decades.