Small-scale spatial heterogeneity as a source for uncertainty of methane fluxes in an extensive near-natural bog-ecosystem

Natural and restored peatlands are often a strong source of the greenhouse gas methane (CH4). CH4 fluxes vary greatly between different peatland ecosystems, depending on temperature, water level and vegetation. In addition, peatlands often show high small-scale spatial heterogeneity that strongly influences the magnitude of CH4 production. This heterogeneity potentially induce a sensor location bias and leads to additional uncertainties due to the flux footprint variability over heterogeneous terrain. To account for such uncertainty, we installed two eddy covariance towers 26 m apart (height: 6 m) to measure the CH4 flux, together with latent heat and CO2 fluxes at the bog ecosystem "Schechenfilz" in southern Germany. The study site is a large near-natural bog (111 ha) with heterogeneity that is characterized by patches of bog-pine forest, sedge meadows, peat mosses and open water areas. Ongoing CH4 measurements on one of the towers began in July 2012, and both towers were operated with a LI-7700 to measure the CH4 flux simultaneously from autumn 2014 to early spring 2015. In a second campaign, from mid-October to mid-November 2014, both instruments were operated at the same tower for comparison. Throughout the investigation, 17 water level gauges were used to measure the temporal variability of the water level in the mean footprint area. The water level was interpolated based on a high-resolution digital terrain model, which also allows us to account for the impact of the spatial variability of the water table. A vegetation map focused on the distribution of plants with aerenchymous tissues was used to determine the influence of the vegetation composition on the CH4 exchange. In this study, we estimated the uncertainty of CH4 fluxes induced by the instrument system and the flux footprint variability. The footprint analyses together with the water table measurements and vegetation map were also used to analyze the impact of small-scale spatial heterogeneity on the CH4 flux magnitude.