Identifying Indicators of State Change in Alaskan Boreal Ecosystems: Testing Previous Hypotheses and Conclusions with Long-term Data

Dr. Michelle Mack | Northern Arizona University

RC-2754

Objective

The research team collected, analyzed and synthesized additional long-term data on the resilience and vulnerability of Alaskan boreal forests to ecosystem state change following wildfire or fuel reduction management. This research revisited site networks established during the completed SERDP project  RC-2109, which focused on identifying factors controlling ecosystem vulnerability to state change after disturbance in black spruce (Picea mariana) forests, the vegetation type that dominates Department of Defense lands in Interior Alaska. At the time of initial measurement, sites were relatively recently disturbed (<10 years after disturbance events). The overarching objective was to determine whether the mechanistic links among fire, soils, permafrost, and vegetation succession predictive of state change immediately following disturbance remain robust predictors over longer timescales. 

Back to Top

Technical Approach

In 2017, the research team remeasured tree regeneration across a regional wildfire network of 90 sites that burned in the record 2004 wildfire year, which they established in 2005 and re-measured in 2006, 2008 and 2011. In 2018, the research team remeasured tree regeneration and permafrost degradation across 36 fuel management network sites that they established in 2012. Over the two years of the project, the research team continuously monitored permafrost soil temperature regimes in three paired burned and unburned sites that were established in 2013.

Back to Top

Results

Resampling tree seedling regeneration in the wildfire network showed that spruce seedling density in 2006 was the strongest predictor of spruce density in 2017. Out of the fire, environment, and biotic drivers that the research team examined, only pre-fire spruce density had a small, positive, direct effect on spruce seedling density in 2017, likely indicating that higher density stands sustained seed rain over a longer post-fire period than low density stands. For deciduous tree seedlings, density in 2006 similarly explained most of the variation in density in 2017. However, organic layer depth had a modest, negative, direct effect on density in 2017, suggesting that deeper organic layers may have suppressed episodic recruitment of deciduous seedlings. Across the soil temperature network, the research team found that most of the increase in soil temperature and permafrost degradation occurred immediately following fire, with site attributes affecting overall warming to a greater degree than time after fire. All sites showed a trend toward increasing soil temperature over time, showing that fire-induced changes in soil temperature are proceeding against a background of regional warming. Across the fuel management network, the research team found persistent conifer recruitment across the first decade after treatment. Although deciduous tree seedlings dominated in severely disturbed (shearbladed) treatments, recruitment of conifers, likely from surrounding undisturbed stands, established mixed composition stands that will likely undergo complex successional trajectories in the future. Active layer depth and permafrost degradation increased over time: this was large in the shearbladed and small in the thinned treatments. New measurements of fire rate of spread (ROS) indicated that the initial benefits of reduced ROS diminished to pre-treatment levels by the first decade after treatment. Additionally, shearblading of black spruce-lichen woodlands increased ROS because vegetation composition shifted towards plant species with a high abundance of fine fuels.

Back to Top

Benefits

The data confirm that managers can use early post-fire data to predict vulnerability to ecosystem state change after wildfire because tree seedling regeneration and soil temperature regime are entrained soon after fire. In fuel management treatments, however, the prolonged window of conifer recruitment will require longer-term (e.g., decadal) observations of composition to predict successional trajectory. New measurements showed that fuel management effects on ROS returned to prefire levels about a decade after treatment. Because shearblading treatments increase ROS in open canopy black spruce woodlands, these ecosystems are not effective candidates for fuel brakes using this method.

Back to Top

Points of Contact

Principal Investigator

Dr. Michelle Mack

Northern Arizona University

Phone: 928-523-9415

Program Manager

Resource Conservation and Resiliency

SERDP and ESTCP

Share