Temporal and spatial variations of CO<sub>2</sub>, CH<sub>4</sub> and N<sub>2</sub>O fluxes at three differently managed grasslands
Abstract. A profound understanding of temporal and spatial variabilities of CO2, CH4 and N2O fluxes between terrestrial ecosystems and the atmosphere is needed to reliably quantify these fluxes and to develop future mitigation strategies. For managed grassland ecosystems, temporal and spatial variabilities of these three greenhouse gas (GHG) fluxes are due to environmental drivers as well as to fertilizer applications, grazing and cutting events. To assess how these affect GHG fluxes at Swiss grassland sites, we studied three sites along an altitudinal gradient that corresponds to a management gradient: from 400 m a.s.l. (intensively managed) to 1000 m a.s.l. (moderately intensive managed) to 2000 m a.s.l. (extensively managed). Temporal and spatial variabilities of GHG fluxes were quantified along small-scale transects of 16 static soil chambers at each site. We then established functional relationships between drivers and the observed fluxes on diel and annual time scales. Furthermore, spatial variabilities and their effect on representative site-specific mean chamber GHG fluxes were assessed using geostatistical semivariogram approaches. All three grasslands were N2O sources, with mean annual fluxes ranging from 0.15 to 1.28 nmol m−2 s−1. Contrastingly, all sites were net CH4 sinks, with uptake rates ranging from −0.56 to −0.15 nmol m−2 s−1. Mean annual respiration losses of CO2, as measured with opaque chambers, ranged from 5.2 to 6.5 μmol m−2 s−1. While the environmental drivers and their respective explanatory power for N2O emissions differed considerably among the three grasslands (adjusted r2 ranging from 0.19 to 0.42), CH4 and CO2 fluxes were much better constrained (adjusted r2 ranging from 0.41 to 0.83), in particular by soil water content and air temperature, respectively. Throughout the year, spatial heterogeneity was particularly high for N2O and CH4 fluxes. We found permanent hot spots for N2O emissions and CH4 uptake at the extensively managed site. Including these hot spots in calculating the mean chamber flux was essential to obtain a representative mean flux for this ecosystem. At the intensively managed grassland, management effects clearly dominated over effects of environmental drivers on N2O fluxes. For CO2 and CH4, the importance of management effects did depend on the status of the vegetation.