B13G-0277:
Ebullition of CO2 and CH4 from an Upland Stream Network in Northeastern Siberia

Monday, 15 December 2014
Seth Spawn1, Samuel Dunn2, Greg J Fiske1, John D Schade3 and Nikita Zimov4, (1)Woods Hole Science Center Woods Hole, Woods Hole, MA, United States, (2)Colorado State University, Fort Collins, CO, USA, Graduate Degree Program in Ecology, Fort Collins, CO, United States, (3)St. Olaf College, Northfield, MN, United States, (4)Northeast Scientific Station, Cherskiy, Russia
Abstract:
While ebullition from arctic thermokarst lakes is a globally significant source of CH4 to the atmosphere, to our knowledge, no study has quantified ebullitive emissions from arctic streams. We assessed bubble emissions from an upland stream network (watershed area = 16.84km2) in the Kolyma River region of Northeastern Siberia. We deployed 24 inverted, funnel-type bubble traps along selected reaches of our study streams for the month of July. Ebullition rates were highly variable in both space and time, ranging from 0 – 2.3 L m-2d-1 (mean = 116 mL m-2d-1). Bubbles were enriched with methane, ranging from 1 to 22% mole fraction (mean = 9%) while CO2 mole fractions were relatively low, ranging from <1 – 5% (mean = 2%). Combined, these data resulted in ebullitive fluxes of 3.3 μmol CH4 m-2 d-1 and 0.9 μmol CO2 m-2 d-1. We scaled these results to the entire stream network for the duration of our study using a Monte-Carlo approach that utilized GIS derived stream lengths, measured stream widths, and measured fluxes. From this, we estimate that our study network emitted 66g CO2-C and 270g CH4-C via the ebullition during the month of July. Comparing our results with literature values on ebullitive methane emissions from thermokarst lakes in this and neighboring watersheds, we estimate that streams account for only 0.05 – 0.12% of the total ebullitive methane flux from our watershed during the open-water season (June-September). Further, comparing our results to the four studies published to date on ebullitive emissions from streams (all from temperate regions), we find our rates to be lower by 1-4 orders of magnitude. We hypothesize that upland, arctic stream sediments may not be favorable environments for bubble production as they are cold, relatively dense and contain mostly mineral substrates. Our results highlight both the importance of sediments as a local control on ebullition and the potential for extreme variability in stream ebullition rates at a global scale.