Long-Term Release of Carbon Dioxide from Arctic Tundra Ecosystems in Northern Alaska

Abstract:

Recent data syntheses and modeling studies of arctic tundra carbon dioxide (CO₂) balance have suggested that the tundra is a CO₂ sink, a source or neutral. Much of this uncertainty arises from a lack of data pertaining to winter CO₂ flux, as well as how these ecosystems have responded to recent warming trends. Due to a harsh, remote environment, long-term, continuous measurements of arctic tundra CO₂ fluxes over the full annual cycle have been non-existent. In September 2007, we began eddy covariance measurements of net ecosystem exchange (NEE, where a negative value denotes a sink) of CO₂ in northern Alaska at two ecosystems, heath and wet sedge tundra. These measurements continue to the present, and represent the longest continuous record of arctic tundra NEE currently available. From January 2008 – December 2013, the ecosystems were annual sources of CO₂, with the wet sedge tundra acting as a greater source (mean ± standard deviation of 50 ± 30 g C m^-2 y^-1) than the heath tundra (16 ± 6 g C m^-2 y^-1). During these same years, the ecosystems were sinks of CO₂ in the summer (June – August), with less variability between the ecosystems, -77 ± 15 g C m^-2 in the wet sedge tundra, and -70 ± 12 g C m^-2 in the heath. Environmental controls over NEE differed between ecosystems and seasons, with the wet sedge tundra acting particularly responsive in terms of CO₂ release during periods with warm air temperatures from fall to early winter. During cold winter periods, CO₂ release from the snowpack in both ecosystems was related to increases in wind speed and drops in atmospheric pressure. Overall, the measured differences in the annual versus summer NEE illustrate how the sink strength of the tundra can be overestimated if data are only collected during the growing season. Furthermore, eddy covariance measurements of methane (CH₄) in the wet sedge tundra during late spring to early fall from 2012 to present show that this ecosystem releases 0.34 ± 11 mg CH₄ m^-2 d^-1. Thus, to understand and predict the sink strength of the arctic tundra, long-term measurements of CO₂, CH₄ and associated meteorological variables across a range of tundra ecosystems over the full annual cycle are critical, and biogeochemical models, which typically underestimate tundra winter respiration, should be more accurately formulated to take into account winter CO₂ release.