Identifying Indicators of State Change and Forecasting Future Vulnerability in Alaskan Boreal Ecosystems

Dr. Edward Schuur | University of Florida

RC-2109

Objective

The boreal region of Interior Alaska is changing. Greater occurrence of high severity fires is shifting forests from mono-dominant black spruce stands to deciduous and mixed forest ecosystems. This shift is altering the relationships between vegetation, soils and permafrost stability which may lead to long-term changes in ecosystem structure and function.

An extensive suite of empirical studies was completed with modeling to improve understanding of the mechanistic linkages among fire, vegetation, the soil organic layer (SOL), and permafrost thaw across interior Alaska. The primary project objectives were to: (1) determine mechanistic links among fire, soils, permafrost, and vegetation succession in order to develop and test field-based ecosystem indicators that can be used to directly predict ecosystem vulnerability to state change, and (2) forecast landscape change in response to projected changes in climate, fire regime, and fire management.

Back to Top

Technical Approach

Fieldwork was conducted across interior Alaska in numerous burned sites of varying ages that spanned a wide range of forest tree species and understory composition to address specific tasks outlined in the original grant proposal. These empirical studies included quantification of above ground tree biomass, tree seedling germination and establishment, understory moss growth, understory vegetation composition, SOL mass and carbon (C) and nutrient cycling and pool sizes and fluxes, permafrost thaw, and impacts of forest management on ecosystem processes. These analyses relied primarily on existing methodologies that have been shown previously to be effective in evaluating similar research questions. Results of these studies informed the development of a predictive model of post-fire reduction in SOL thickness, and incorporation of this model into an existing model (Dynamic Organic Soil version of the Terrestrial Ecosystem Model, DOS-TEM), to improve understanding of the combustion of the SOL and permafrost degradation following fire, vegetation succession following fire, and postfire soil organic layer recovery and relationship to permafrost thaw. A model framework was then developed to couple the Alaska Frame-Based Ecosystem Code (ALFRESCO) with DOSTEM. ALFRESCO was then used to investigate how changing fire management planning options within military training lands would influence the future fire regime and concurrent boreal forest vegetation dynamics.

Back to Top

Results

Results of this study support and expand upon the understanding of the linkages among fire, soils, permafrost, and vegetation succession across multiple analytical scales. The seedling establishment study showed that seedling germination post-fire is affected by complex interactions among moisture, seed size, elevation and latitude, post-fire SOL thickness, and time since the last fire. Seed germination trials showed consistently high germination on mineral soil substrate.

This project also highlighted large differences in ecosystem C and nutrient distribution and cycling in deciduous and conifer stands. Black spruce and Alaska paper birch forest exhibited similar ecosystem C stocks, however black spruce contained greater C in soils while birch contained larger quantities in aboveground tree biomass. Generally, nutrient cycling was much faster in the birch stands relative to spruce. For the moss transplants and litter manipulation study conducted at this site, it was observed after one year that the marked mosses had adapted their growth form to the environmental conditions associated with their new stand, and after two years of treatments impacts of the leaf litter treatments were detected in both forest types, with larger litter inputs resulting in less moss growth. This result suggests that shifts towards greater deciduous tree species dominance may reduce understory moss cover, reducing SOL accumulation and increasing vulnerability of permafrost to thaw. Further, a divergence in total bryophyte cover across a mix of stand types and successional stages showed a divergence between the coniferous (spruce) and deciduous (birch and aspen) successional trajectories that occurred between 20 and 40 years since fire. This divergence seemed to coincide with major changes in leaf litter cover associated with the canopy type.

A decrease in the proportion of spruce relative to total trees from the pre-fire to the post-fire stand was observed in nearly all examined stands, indicating shifts from black spruce dominated stands to mixed or deciduous dominated stands. Forest stands most likely to show the largest change in black spruce dominance were those that had relatively thin post-fire SOL depths and showed dendroclimatic indications of recent drought stress. These sites were generally located at warmer and drier landscape positions, suggesting a landscape pattern of lower resilience to disturbance compared to sites in cool and moist locations.

Management practices designed to reduce the likelihood of fire spread into inhabited areas reduce ecosystem C pools, especially when shearblading methods were used. The greatest C losses were from aboveground pools, however the practice was also associated with high deciduous tree species establishment in areas previously dominated by black spruce and an increase in thaw depth. Thinning also impacted ecosystem processes, but with moderate impacts. These results suggest increased use of shearblading could contribute further to shifts successional trajectories across interior Alaska.

The predictive model of SOL combustion identified landscape drainage as the primary driver of the relative amount of SOL burned during fire, with permafrost thaw initiating with smaller SOL losses in upland sites than lowland sites. When SOL was allowed to re-accumulate post-fire, less permafrost degradation occurred and at the lowland site, full recovery of initial thermal state was regained within 40 years. Wildfire modeling showed that fire frequency and extent are projected to increase across the majority of future climate combinations for the northern boreal forest region of Alaska, despite the increase of less flammable deciduous vegetation and the reduction of more flammable late successional spruce forest. Concurrently, the distribution of near surface permafrost is projected to decrease in the region, driven primarily by the effect of warming and fire disturbance on permafrost thaw. Modeling an experimental alternative fire management planning scenario showed that changing all military lands within the study area from limited to full protection led to a consistent increase in the number of fires, yet a decrease in the amount of area burned through 2100 compared to the status quo. This led to an increase in the amount of late successional coniferous forest present on the landscape, in contrast to maintaining the current fire management scenario, which leads to more early successional deciduous forest on the landscape through the end of the century

Back to Top

Benefits

Collectively, the results support the understanding that increased wildfire severity and loss of the SOL increases permafrost thaw in the near-term and influences seedling establishment and successional trajectories, shifting forest composition from black spruce dominated ecosystems to those comprised of deciduous tree species. In the longer-term, this shift has consequences for understory moss growth and SOL re-accumulation, thereby directly influencing permafrost recovery post-fire. Change in tree species composition also alters C and nutrient cycling rates and the partitioning of ecosystem C stocks into above and belowground pools. Finally, forest management approaches employed will further impact successional trajectories and vulnerability of permafrost to thaw.

Back to Top

Points of Contact

Principal Investigator

Dr. Edward Schuur

University of Florida

Phone: 352-392-7913

Program Manager

Resource Conservation and Resiliency

SERDP and ESTCP

Share