Winter fluxes of greenhouse gases from snow‐covered agricultural soil:intra‐annual and interannual variations

Abstract

Abstract. Despite the length of winter in cold temperate climates, few studies refer to greenhouse gas emissions from soils during the nongrowing season. In this study, N20 and CO2 fluxes from agricultural and forest soils in southeastern Quebec (Canada) were measured during winter and spring from 1994 to 1997, and the influences of climate, soil, and snow properties on the gaseous emissions were examined. N20 fluxes were far greater from the agricultural soil (2-187 ng N20 m -2 s 'l) than from the forest soil (< 3 ng N20 m -2 s-I), but CO2 fluxes were equivalent for both soil systems (2-102 gg CO2 m -2 s-i). The higher N20 concentrations in the lower soil horizons could be explained by positive temperature gradients with depth and concomitant negative gas solubility gradients. However, the higher N20 concentrations could also be explained by variations in the expression of N20 reductase with depth, which can modify the N2/N20 ratios in relation to the availability of 02. Calculated N20-N fluxes showed that N losses by gaseous emissions from soils during winter and spring were comparable to, or exceeded, similar reported N losses during the growing season. The highest winter fluxes observed in 1997 were interpreted to be due to favorable meteorological conditions that prevailed for denitrification through high soil water content in summer and fall of 1996. Although interannual and interseasonal variations of fluxes are important, this study shows that wintertime losses of N20 from agricultural soil can be up to 2 to 4 times greater than emissions measured during the growing season in similar agroecosystems.

Wavelet analysis of wintertime and spring thaw CO2 and N2O fluxes from agricultural fields

“…Therefore, the CO 2 flux was equally dependent on surface conditions as subsurface conditions. The CO 2 is likely produced throughout the soil profile when there is sufficient available substrate but unlike N 2 O, the CO 2 does not become consumed or altered as it travels to the surface (Van Bochove et al, 2000). This also supports the suggestion of surface N 2 O production, since if N 2 O was trapped and released upon thaw, CO 2 would be undergoing the same process.…”

“…Conditions at depth in the soil profile (!20 cm) are also conducive to denitrification and N 2 O production, as evidenced by high concentrations measured during winter (Burton and Beauchamp, 1994;Van Bochove et al, 2000). Such high N 2 O concentrations were also observed at our site during January to March 2003(Hu, 2006.…”

“…However, WCO analyses did not show a high coherence between N 2 O surface flux and sub-surface soil temperature. The suggestion of large spring thaw N 2 O flux being due to release of trapped N 2 O stems from studies that found high N 2 O concentrations at depth (Burton and Beauchamp, 1994;Van Bochove et al, 2000). During non-freezing conditions, studies using independent methods to measure concentration at depth and surface flux, have found high N 2 O concentrations at depth that did not always correlate with high surface flux of N 2 O (Burton et al, 1997;Van Groenigen et al, 2005).…”

“…Spatial and temporal variation of N 2 O fluxes is large as N 2 O production is often episodic, occurring at times of soil wetting and thawing, and is dependent on the spatial distribution of soil characteristics which control N 2 O production through denitrification (Christensen and Tiedje, 1990;Christensen and Christensen, 1991;Nyborg et al, 1997;Lemke et al, 1998;Lark et al, 2004;Pennock et al, 2005;Yates et al, 2006). Van Bochove et al, 2000;Dö rsch et al, 2004). Spring thaw emissions can represent up to 70% of annual N 2 O emissions (Lemke et al, 1998;Wagner-Riddle and Thurtell, 1998;Dö rsch et al, 2004;Teepe et al, 2004;Wagner-Riddle et al, 2007).…”

Upscaling chamber-based measurements of N2O emissions at snowmelt

“…These emissions can be a major component of annual N 2 O emissions but the magnitude of emissions can differ greatly between years at a given site. For example, Van Bochove et al (2000) found that measured fluxes ranged from 0.4 kg N 2 O-N ha -1 in 1995-1996 to 10 kg N 2 O-N ha -1 in 1996-1997 for their study on winter fluxes from snow-covered agricultural fields in Quebec. Differences in snowmelt emissions among land uses and topographical positions have also been documented.…”