Forest fragmentation is ubiquitous, resulting in substantial differences in both micro-environmental conditions and biogeochemistry between the forest edge and interior. The majority of forests occur in fragmented landscapes, yet our understanding of forests and the role they play in carbon cycling is still predominantly based on intact ecosystems. Recent work shows increases in carbon uptake and storage at the forest edge, but this response is heterogeneous across climatic conditions. Individual study sites are often geographically limited and deal only with aboveground carbon dynamics. The variability of the edge-to-interior gradient and the effect on soil respiration are largely unknown. We have attempted to address this question with a multifaceted approach. At the microscale, we characterized the forest physical climate and quantified soil respiration rates along an edge-to-interior gradient in two temperate forests in the eastern United States that vary in climate, species composition, and soil type. To examine macroscale variability, we combined the USFS’s Forest Inventory and Analysis database with 20,000 manually classified land cover maps across 20 states in the northeastern US to quantify the edge effect on tree growth dynamics. This macroscale study allowed us to measure the extent, variability, and mechanistic drivers of the altered forest carbon cycling.
Results/Conclusions
The microscale examination of belowground activity found that rates of growing season respiration are 15-25% higher near the forest edge relative to the interior. These gradients in soil respiration appear to be driven by large gradients in soil temperature, an important driver of soil respiration, which increases significantly from edge to interior. At the landscape scale, aboveground carbon uptake as measured by tree biomass increases by a regional average of 40% near the edge, but the magnitude of this signal varies spatially and by land cover type. The variability is driven by both macro- and microscale differences in environmental conditions across landscapes, including light, temperature, and nutrient availability. Our results collectively show that forest fragmentation leads to significant differences in carbon dynamics at both a local and regional scale. All current resolutions of carbon cycle modelling, from single-site empirical models to global earth-systems models, do not fully account for the unique ecosystems that are fragmented forests. This work demonstrates the need to not only characterize forest edge dynamics but to use this information to enhance our understanding of the carbon cycle.