Arctic ecosystems are characterized by a wide range of soil moisture conditions and thermal regimes and contribute differently to the net methane (CH₄) budget. Yet, it is unclear how climate change will affect the capacity of those systems to act as a net source or sink of CH₄. Here, we present results of in situ CH₄ flux measurements made during the growing season 2014 on Disko Island (west Greenland) and quantify the contribution of contrasting soil and landscape types to the net CH₄ budget and responses to summer warming. We compared gas flux measurements from a bare soil and a dry heath, at ambient conditions and increased air temperature, using open‐top chambers (OTCs). Throughout the growing season, bare soil consumed 0.22 ± 0.03 g CH₄‐C m⁻² (8.1 ± 1.2 g CO₂‐eq m⁻²) at ambient conditions, while the dry heath consumed 0.10 ± 0.02 g CH₄‐C m⁻² (3.9 ± 0.6 g CO₂‐eq m⁻²). These uptake rates were subsequently scaled to the entire study area of 0.15 km², a landscape also consisting of wetlands with a seasonally integrated methane release of 0.10 ± 0.01 g CH₄‐C m⁻² (3.7 ± 1.2 g CO₂‐eq m⁻²). The result was a net landscape sink of 12.71 kg CH₄‐C (0.48 tonne CO₂‐eq) during the growing season. A nonsignificant trend was noticed in seasonal CH₄ uptake rates with experimental warming, corresponding to a 2% reduction at the bare soil, and 33% increase at the dry heath. This was due to the indirect effect of OTCs on soil moisture, which exerted the main control on CH₄ fluxes. Overall, the net landscape sink of CH₄ tended to increase by 20% with OTCs. Bare and dry tundra ecosystems should be considered in the net CH₄ budget of the Arctic due to their potential role in counterbalancing CH₄ emissions from wetlands – not the least when taking the future climatic scenarios of the Arctic into account.

An improvement in our process-based understanding of carbon (C) exchange in the Arctic and its climate sensitivity is critically needed for understanding the response of tundra ecosystems to a changing climate. In this context, we analysed the net ecosystem exchange (NEE) of CO₂ in West Greenland tundra (64° N) across eight snow-free periods in 8 consecutive years, and characterized the key processes of net ecosystem exchange and its two main modulating components: gross primary production (GPP) and ecosystem respiration (R_eco). Overall, the ecosystem acted as a consistent sink of CO₂, accumulating −30 g C m⁻² on average (range of −17 to −41 g C m⁻²) during the years 2008–2015, except 2011 (source of 41 g C m⁻²), which was associated with a major pest outbreak. The results do not reveal a marked meteorological effect on the net CO₂ uptake despite the high interannual variability in the timing of snowmelt and the start and duration of the growing season. The ranges in annual GPP (−182 to −316 g C m⁻²) and R_eco (144 to 279 g C m⁻²) were > 5 fold larger than the range in NEE. Gross fluxes were also more variable (coefficients of variation are 3.6 and 4.1 % respectively) than for NEE (0.7 %). GPP and R_eco were sensitive to insolation and temperature, and there was a tendency towards larger GPP and R_eco during warmer and wetter years. The relative lack of sensitivity of NEE to meteorology was a result of the correlated response of GPP and R_eco. During the snow-free season of the anomalous year of 2011, a biological disturbance related to a larvae outbreak reduced GPP more strongly than R_eco. With continued warming temperatures and longer growing seasons, tundra systems will increase rates of C cycling. However, shifts in sink strength will likely be triggered by factors such as biological disturbances, events that will challenge our forecasting of C states.