Spatial characteristics of ecosystem respiration in three tundra ecosystems of Alaska

Abstract

Ecosystem respiration (ER) is a significant source within the carbon budget of tundra 3 themselves vulnerable to permafrost degradation, expansion of shrub extent, and shortening of 4 the snow-covered period in response to a changing Arctic climate. Here, we report on our 5 assessment of the spatial characteristics of ecosystem respiration, using a manual chamber 6 over three different tundra ecosystems in Alaska. Short growing-season ERs were 648, 653, and 653 mg C m2 for normalized air temperature (AT) and 668, 564, and 587 mg C m2 7 for in-situ 8 temperature (ST) at sites in Council, the upland tundra of the North Slope, and within the Arctic 9 National Wildlife Refuge (ANWR), respectively, corresponding to 63, 74, and 65 % (AT), and 10 64, 56, and 71 % (ST) of annual ecosystem respiration at each tundra site. Hence, soil 11 at a depth of 5 cm was an important driver in modulating the respiration over tundra, suggesting 12 soil temperature elucidates > 80 % of air temperature. At the Council site, a total of 81 sampling 13 points for each month are required for the manual chamber system to attain an experimentally 14 spatial representativeness for respiration falling within ±10 % of the full sample mean, with a 15 95 % confidence level. At the North Slope and ANWR sites, the number of sampling points for 16 each species during each month was chosen to yield results within at least ±20 %, with a 90 % 17 confidence level. Further, the respiration measuring frequency for each species must increase to 18 also satisfy at least this ±20 % and 90 % confidence level. These findings suggest that the use of 19 larger volume manual chamber and/or measurement frequency at 50 or more points can 20 logistical constraints and subsequently determine representatively mean ecosystem respiration at 21 each tundra site. Page 2 of 24