Carbon dioxide emissions from deadwood in forests of the Northeastern US

Background/Question/Methods

Global climate change has emphasized the need to better understand carbon emissions. However, major gaps still exist in our understanding of forest carbon dynamics, including the rate at which trees release stored carbon after they die. Decomposition is a major contributor to CO2 emissions, yet this component of the carbon cycle has often been overlooked. We sought to understand these CO2 emissions by pursuing the following questions: (1) How do CO2 emissions change with time since tree death, (2) How do CO2 emissions from decaying logs compare to CO2 emissions from soils, and (3) How much do CO2 emissions from logs contribute to total ecosystem respiration? To answer these questions, CO2 emissions were measured in chambers attached to stumps, logs, and soil in conifer forests of central Maine using a LiCOR-830. Specifically, (1) A chronosequence of spruce stumps varying from 0 to 32 years since harvest was selected to determine change in CO2 emissions over time, (2) Spruce logs were randomly selected and paired with adjacent soils to compare emissions by substrate type, and (3) Field inventories of logs were used to scale up CO2 emissions to the ecosystem level for comparison with ecosystem-level respiration estimated by an eddy-flux tower.

Results/Conclusions

Our results indicate that stump CO2 emissions change significantly over time, increasing from immediately post-harvest to a maximum at 8 years post-harvest, then decreasing through year 23 and reaching a low asymptote by year 32. These changes were likely primarily due to fungal colonization and changes in fungal activity. CO2 emissions from decaying logs did not differ significantly from those of adjacent soils (p > 0.05). When scaled up to a per-area basis, stumps emitted about 2.2 ± 2.1 g C/m2/yr. In contrast, carbon emissions from decaying logs contributed about 42.6 ± 23.8 g C/m2/yr, or 3.8% of total ecosystem respiration. The difference between stump and log emissions was primarily due to the much greater surface area of logs compared to stumps within the forest. Using these data, we can determine the rate at which trees release stored carbon after they die, and how much this offsets CO2 uptake by live trees. Our results point to the need for further research on deadwood carbon dynamics, as stump and log emissions appear to be overlooked and underrepresented in forest carbon models and budgets. Overall, these empirical data can inform process-based models of forest carbon dynamics.