Increasing potential NEP of eastern boreal North American forests constrained by decreasing wildfire activity

Abstract

Wildfire activity has been decreasing over the past few centuries across North America's southeastern boreal forest. This change has caused a gradual shift in the age-class distribution toward a stronger representation of old-growth stands (age > 140 years) in unmanaged forest landscapes. Parallel to these changes, there has been an improvement in growth conditions driven by climate warming. Given the negative feedback of forest aging on net ecosystem production (NEP), a reasonable case can be made about an age-effect constraint on a climate change-induced increase in carbon assimilation. Here we combine empirical information (multicentury tree-ring records, time-since-fire map, and forest inventories) along with model simulations to develop a reconstruction of NEP of jack pine forests (Pinus banksiana Lambert) at the transition zone of the Spruce-feather moss and Balsam fir-white birch bioclimatic domains of eastern boreal North America covering the period A.D. 1830 to 1999. From this reconstruction, we test the hypothesis that currently improving carbon assimilation in unmanaged forests due to climate change is being offset by the aging of the forests resulting from the decreasing activity of wildfires. A sensitivity analysis of results to different assumptions about the carbon sink strength of old-growth forests is also made via a probability distribution derived from 10,000 ensemble members. Our simulation results suggest that increasing moisture availability and lengthening of the growing season contributed to a statistically significant upward trend in carbon assimilation of 0.39 g C m−2 yr−1 over 1901–2009. Meanwhile, mean age of jack pine stands has also changed, shifting from 87 years in the 1920s to 131 years in 1999. This forest aging contributed up to a 12% reduction in carbon assimilation from the 1930s to 1990s, an effect that is found to be sufficient to offset NEP increases due to climate change in 75% of our ensemble members. Great caution should therefore be applied when interpreting future carbon assimilation or forest growth projections that do not account for such second-order feedback from disturbances.