The structure and function of Arctic ecosystems are responding rapidly to climate changes. Here, we describe the results of the ABACUS project, which addressed the question “what controls the temporal and spatial variability of carbon exchange between Arctic ecosystems and the atmosphere?” ABACUS was a multi-scale investigation of ecosystem processes undertaken in Sweden and Finland, 2006-9. Measurements included isotopic analyses, monitoring of foliage and soil profiles, automated chambers, four eddy covariance towers, and dedicated aircraft flights.
Soil organic matter content was highly variable on a range of scales, which complicates upscaling. Further, soil processes differed between dominant vegetation types. The 14C content of CO2 released from the soils of birch woodlands increased substantially in mid-summer, despite the plants being most active in the same time period; we expected the input of recently fixed C to reduce the 14C content of the CO2 released. These data suggest high plant activity in the birch woodlands must have stimulated the decomposition of a 14C-enriched pool of soil organic matter, i.e. a positive priming mechanism, that was not found in nearby tundra.
There is a broadly observed correlation between leaf area index (LAI) and total foliar nitrogen across Arctic tundra ecosystems. As a result, we hypothesized that fine root biomass, a factor in N uptake, would be correlated with LAI. In sparser vegetation, fine root biomass increased linearly with LAI, but saturated at ~0.4 kg m-2 at LAI > 1. Explaining these emergent ecosystem properties is a critical target for process based models.
The partitioning of fixed C into biomass or autotrophic respiration is a critical determinant of ecosystem C balance, often assumed ~50% but rarely measured. 13C pulse labelling in a range of moss communities provided a means to quantify the fate of fixed C. Measurements and modeling of the fate of recently fixed 13CO2 in Sphagnum confirmed the expected 50% partitioning. But in Polytrichum, a more productive moss, autotrophic respiration was ~80% of GPP. These results indicate very different patterns of C dynamics among moss species, with implications for total ecosystem budgets.
Chamber and eddy covariance measurements of CO2 exchange recorded similar seasonal timing over a range of vegetation types, which could be largely modeled as a function of LAI. Aircraft flux measurements during the peak growing season provided an estimate of landscape variability alongside the temporal sampling from fixed tower systems, and a means to constrain upscaling via models.