| Popis: |
The forest-floor represents an important interface for various carbon dioxide (CO2) fluxes, however, our knowledge of their variability and drivers across a managed boreal forest landscape is limited. Here, we used a three-year (2016-2018) data set of biometric- and chamber-based flux measurements to investigate the net forest-floor CO2 exchange (NEff) and its component fluxes across 50 forest stands spanning different soil types, tree species, and age classes within a 68 km2 boreal catchment in Sweden. We found that the forest-floor acted as a net CO2 source with the 10th-90th percentile (used hereafter for describing reported variations) ranging from 149 to 399 g C m- 2 yr-1. Among the key landscape attributes, stand age strongly affected most NEff component fluxes, whereas tree species and soil type effects were weak and absent, respectively. Specifically, forest-floor net CO2 emissions increased with stand age due to declining understory gross and net primary production, ranging between 77-275 and 49-163 g C m- 2 yr- 1, respectively. Furthermore, we observed higher understory production rates in pine than in spruce stands. Across the 50 stands, the total forest-floor respiration ranged between 340 and 549 g C m- 2 yr-1 and its spatial variation was primarily regulated by its autotrophic components, i.e., understory and tree root respiration, which displayed divergent increasing and decreasing age-related trends, respectively. Furthermore, heterotrophic soil respiration remained within a relatively narrow range (154-290 g C m- 2 yr- 1), possibly owing to compensating gradients in forest-floor properties. We further identified tree biomass as the major driver of the landscape-scale variations of CO2 fluxes, likely attributable to modulating effects on forest-floor resource availability and growing conditions. This implies that tree growth responses to forest management and global change will be particularly important for regulating magnitudes and spatial variations of forest-floor CO2 fluxes in boreal forests. |
Full Text from ScienceDirect