COS 106-5 - Managed wildfire promotes native plant diversity and fuels reduction: Evidence from California’s Sierra Nevada Wilderness Areas

Tuesday, August 8, 2023

2:30 PM – 2:45 PM PDT

Location: E141

Presenting Author(s)

RF

Raphaela E. Floreani Buzbee

PhD Candidate
University of California, Berkeley
BERKELEY, California, United States

Coauthor(s)

ZS

Zachary L. Steel

USDA Forest Service - Rocky Mountain Research Station and University of California, Berkeley, United States
SS

Scott Stephens, PhD

Professor of Fire Science
University of California, Berkeley
BERKELEY, California, United States
BC

Brandon M. Collins

University of California, Berkeley, United States
DA

David D. Ackerly

Dean of Rausser College of Natural Resources
UC Berkeley, California, United States
NM

Natalie MacMillan

University of California, Berkeley
BERKELEY, California, United States

Abstract: Fire is a fundamental ecological process shaping ecosystems and plant communities. After over a century of fire suppression and exclusion of Indigenous fire stewardship in the Western United States, there is a concerted effort to reintroduce beneficial fire. Managed wildfire (allowing naturally ignited lightning fires to burn in remote wilderness areas) is one approach for restoring fire regimes at scale that has potential to support native plant diversity while increasing the pace of fuels treatments. Over time, this strategy creates a mosaic of patchy fire histories, increasing overall landscape heterogeneity.

To assess the effects of managed wildfire, we compare three sites within wilderness areas of the central Sierra Nevada in California with varied implementations of this management approach in Yosemite National Park and Sierra and Inyo National Forests. In 2021 and 2022, we measured downed surface fuels using Brown’s transects and plant species diversity using transects and a timed search at 105 sites stratified across different levels of time since last fire (0-40 years) and previous burn severity (unchanged, low, moderate, high) determined from composite burn index.

Plot-level plant species diversity ranged from < 10 to >50 species and was highest in the site that burned most recently. We found that the two areas where managed wildfire has been implemented had higher species richness (200-250 species) compared to unburned or fire-excluded areas (71 species). Areas with managed wildfire had significantly lower average total fuel loads (125 Mg/ ha) compared to unburned or fire exclusion areas (320 Mg/ ha), where the total fuel load is the sum of ground fuels and surface fuels. Finally, we find that plot-level plant species richness is negatively correlated with total woody debris (p = 0.0005).

We used AIC model selection to distinguish among possible linear models for predicting plant species richness. The favored model (97% cumulative model weight) included the following parameters: duff depth (p< 0.02), number of fires (p< 0.01), canopy cover (p= 0.07), total fuel load (p< 0.01), and years since last fire (p< 10-5).

Given the scale of recent wildfires these findings provide evidence for a potential win-win for managers: managed wildfire under less-than-extreme burning conditions, increases native plant diversity and decreases woody debris thereby also reducing the risk and severity of future wildfires. Results from this study may guide policy and fire restoration strategies intended to increase ecosystem resilience in a changing world.