It is proposed that increases in anthropogenic reactive nitrogen (Nr ) deposition may cause temperate and boreal forests to sequester a globally significant quantity of carbon (C); however, long-term data from boreal forests describing how C sequestration responds to realistic levels of chronic Nr deposition are scarce. Using a long-term (14-year) stand-scale (0.1 ha) N addition experiment (three levels: 0, 12.5, and 50 kg N ha(-1)  yr(-1) ) in the boreal zone of northern Sweden, we evaluated how chronic N additions altered N uptake and biomass of understory communities, and whether changes in understory communities explained N uptake and C sequestration by trees. We hypothesized that understory communities (i.e. mosses and shrubs) serve as important sinks for low-level N additions, with the strength of these sinks weakening as chronic N addition rates increase, due to shifts in species composition. We further hypothesized that trees would exhibit nonlinear increases in N acquisition, and subsequent C sequestration as N addition rates increased, due to a weakening understory N sink. Our data showed that understory biomass was reduced by 50% in response to the high N addition treatment, mainly due to reduced moss biomass. A (15) N labeling experiment showed that feather mosses acquired the largest fraction of applied label, with this fraction decreasing as the chronic N addition level increased. Contrary to our hypothesis, the proportion of label taken up by trees was equal (ca. 8%) across all three N addition treatments. The relationship between N addition and C sequestration in all vegetation pools combined was linear, and had a slope of 16 kg C kg(-1)  N. While canopy retention of Nr deposition may cause C sequestration rates to be slightly different than this estimate, our data suggest that a minor quantity of annual anthropogenic CO2 emissions are sequestered into boreal forests as a result of Nr deposition.

No abstract

We report results from long-term simulated external nitrogen (N) input experiments in three northern Pinus sylvestris forests, two of moderately high and one of moderately low productivity, assessing effects on annual net primary production (NPP) of woody mass and its interannual variation in response to variability in weather conditions. A sigmoidal response of wood NPP to external N inputs was observed in the both higher and lower productivity stands, reaching a maximum of ~65% enhancement regardless of the native site productivity, saturating at an external N input of 4-5 g N·m ·yr . The rate of increase in wood NPP and the N response efficiency (RE , increase in wood NPP per external N input) were maximized at an external N input of ~3 g N·m ·yr , regardless of site productivity. The maximum RE was greater in the higher productivity than the lower productivity stand (~20 vs. ~14 g C/g N). The N-induced enhancement of wood NPP and its RE were, however, markedly contingent on climatic variables. In both of the higher and lower productivity stands, wood NPP increased with growing season precipitation (P), but only up to ~400 mm. The sensitivity of the response to P increased with increasing external N inputs. Increasing growing season temperature (T) somewhat increased the N-induced drought effect, whereas decreasing T reduced the drought effect. These responses of wood NPP infused a large temporal variation to RE , making the use of a fixed value unadvisable. Based on these results, we suggest that regional climate conditions and future climate scenarios should be considered when modeling carbon sequestration in response to N deposition in boreal P. sylvestris, and possibly other forests.