A spatially explicit analysis to extrapolate carbon fluxes in upland tundra where permafrost is thawing

One of the most important changes in high-latitude ecosystems in response to climatic warming may be the thawing of permafrost soil. In upland tundra, the thawing of ice-rich permafrost can create localized surface subsidence called thermokarst, which may change the soil environment and influence ecosystem carbon release and uptake. We established an intermediate scale (a scale in between point chamber measurements and eddy covariance footprint) ecosystem carbon flux study in Alaskan tundra where permafrost thaw and thermokarst development had been occurring for several decades. The main goal of our study was to examine how dynamic ecosystem carbon fluxes [gross primary production (GPP), ecosystem respiration (R-eco), and net ecosystem exchange (NEE)] relate to ecosystem variables that incorporate the structural and edaphic changes that co-occur with permafrost thaw and thermokarst development. We then examined how these measured ecosystem carbon fluxes responded to upscaling. For both spatially extensive measurements made intermittently during the peak growing season and intensive measurements made over the entire growing season, ecosystem variables including degree of surface subsidence, thaw depth, and aboveground biomass were selected in a mixed model selection procedure as the 'best' predictors of GPP, R-eco, and NEE. Variables left out of the model (often as a result of autocorrelation) included soil temperature, moisture, and normalized difference vegetation index. These results suggest that the structural changes (surface subsidence, thaw depth, aboveground biomass) that integrate multiple effects of permafrost thaw can be useful components of models used to estimate ecosystem carbon exchange across thermokarst affected landscapes.