Variation in Factors Regulating Net Greenhouse Gas Exchange Across Different Vegetation Types at Cape Bounty, Melville Island, Nunavut

Abstract:

Global-scale climate simulations predict significant changes both in temperature and moisture regimes in the high Arctic. This could lead to changes in vegetation community distribution, as vegetation communities are distributed along moisture gradients often determined by snowfall patterns across the landscape. Furthermore, changes in soil moisture and temperature could alter fluxes of greenhouse gases such as carbon dioxide (CO₂), methane (CH₄) and nitrous oxide (N₂O), and the impacts of changes in these controlling factors could vary by vegetation type.

We measured both spatial and temporal variation in CO₂ fluxes using combinations of eddy covariance, auto-chamber, and static chamber techniques at the Cape Bounty Arctic Watershed Observatory (CBAWO). Measurements were performed in three major plant community types: polar semi-desert (PSD), mid-moisture tundra (MM) and wet sedge meadow (WS).

Based on our auto-chamber data collected in all vegetation types, ecosystem respiration (ER) related positively to air temperature, and correlated more strongly with air temperature than soil temperature. Modeled ER based on eddy covariance data and air temperature over MM agreed well with measured ER in the same vegetation type. In the WS community, average net ecosystem exchange (NEE) in 2014 measured by static chambers differed in spectrally separable ‘wet’ and ‘dry’ sedge areas (-0.33 and 0.01 µmol m^-2 s^-1, respectively; p<0.001). Rates of ER also varied across this moisture gradient (p<0.05). Over the entire growing season and multiple years, NEE correlated poorly with air and soil temperature, suggesting that ER is not the dominant processes driving NEE. This can vary, however, over the growing season. In PSD communities measured in 2013, air temperature related positively to NEE early in the growing season, but not during the latter part of the season, when PAR (photosynthesis) became the key factor controlling NEE. Not surprisingly, NEE related strongly (0.93) to vegetation cover during this latter part of the growing season.

Our results suggest critical differences in both the type and timing of key factors that control NEE across High-Arctic landscapes, suggesting simple models based on one vegetation type may not accurately predict the response of these landscapes to changes in climate.