Methane Flux Influenced by Experimental Water Table Drawdown and Soil Warming in a Dry Boreal Continental Bog

To quantify the effects of water table drawdown and soil warming on CH4 fluxes, we used a static chamber technique during the growing seasons (May-October) of 2011-2013 at hollow and hummock microforms at three sites of a continental bog near the town of Wandering River, Alberta, Canada: (1) Control, (2) Experimental drained, and (3) old Drained. To simulate climatic warming, we used open top chambers to passively warm half of the hollows and half of the hummocks at each of the water level treatment sites. Water table drawdown significantly reduced CH4 flux by 50% in 3 years and 76% in 13 years of drainage. The hollows showed greater reduction of efflux as compared to hummocks. A persistent functional relationship of CH4 flux with water level was found across all sites in all years. The relationship revealed that the contribution of change in vegetation type at hollows and hummocks to CH4 production and emission was relatively less important than that of the water level. Hummocks and hollows responded to warming differently. At the control, experimental and drained sites, warming increased flux at hollows by 16, 21 and 26%, and reduced flux at hummocks by 4, 37, and 56%, respectively. The combined effect of lowered water table and warming on CH4 emission was overall negative, although the interaction between the two contributing factors was not significant. Therefore, whereas climatic warming and subsequent lowering of water table are expected to reduce CH4 efflux from dry ombrotrophic bogs of Alberta, different microforms at these bogs may respond differently with accelerated emissions at warmed, wetter (hollows) and reduced emissions at warmed, drier (hummocks) microforms. Overall, CH4 efflux from Alberta's dry continental bogs that are not underlain by permafrost might be affected only slightly by the direct effect of predicted climate warming, although initial water table position will be an important control on the overall response.