Spatial variation in landscape‐level CO2 and CH4 fluxes from arctic coastal tundra: influence from vegetation, wetness, and the thaw lake cycle

Abstract

Regional quantification of arctic CO2 and CH4 fluxes remains difficult due to high landscape heterogeneity coupled with a sparse measurement network. Most of the arctic coastal tundra near Barrow, Alaska is part of the thaw lake cycle, which includes current thaw lakes and a 5500-year chronosequence of vegetated thaw lake basins. However, spatial variability in carbon fluxes from these features remains grossly understudied. Here, we present an analysis of whole-ecosystem CO2 and CH4 fluxes from 20 thaw lake cycle features during the 2011 growing season. We found that the thaw lake cycle was largely responsible for spatial variation in CO2 flux, mostly due to its control on gross primary productivity (GPP). Current lakes were significant CO2 sources that varied little. Vegetated basins showed declining GPP and CO2 sink with age (R(2) = 67% and 57%, respectively). CH4 fluxes measured from a subset of 12 vegetated basins showed no relationship with age or CO2 flux components. Instead, higher CH4 fluxes were related to greater landscape wetness (R(2) = 57%) and thaw depth (additional R(2) = 28%). Spatial variation in CO2 and CH4 fluxes had good satellite remote sensing indicators, and we estimated the region to be a small CO2 sink of -4.9 ± 2.4 (SE) g C m(-2) between 11 June and 25 August, which was countered by a CH4 source of 2.1 ± 0.2 (SE) g C m(-2) . Results from our scaling exercise showed that developing or validating regional estimates based on single tower sites can result in significant bias, on average by a factor 4 for CO2 flux and 30% for CH4 flux. Although our results are specific to the Arctic Coastal Plain of Alaska, the degree of landscape-scale variability, large-scale controls on carbon exchange, and implications for regional estimation seen here likely have wide relevance to other arctic landscapes.