Changes in snow cover alter nitrogen cycling and gaseous emissions in agricultural soils

[Display omitted] •A field study with removal, ambient, and accumulation snow treatments was conducted.•Snow manipulation altered snow depth, soil frost intensity, and soil moisture.•Greatest gas fluxes were in removal (year 1) and ambient (year 2) snow treatments.•Overwinter soil moisture and temperature were drivers of N cycling and N2O fluxes.•Winter soil conditions influenced overwinter and spring thaw N2O production. Climate change-related increases in winter temperatures and precipitation, as predicted for eastern Canada, may alter snow cover, with consequences for soil temperature and moisture, nitrogen cycling, and greenhouse gas fluxes. To assess the effects of snow depth in a humid temperate agricultural ecosystem, we conducted a two-year field study with (1) snow removal, (2) passive snow accumulation (via snow fence), and (3) ambient snow treatments. We measured in situ N2O and CO2 fluxes and belowground soil gas concentration, and conducted denitrification and potential nitrification laboratory assays, from November through May. Snow manipulation significantly affected winter N2O dynamics. In the first winter, spring thaw N2O fluxes in snow removal plots were 31 and 48 times greater than from ambient snow and snow accumulation plots respectively. Mid-winter soil N2O concentration was also highest in snow removal plots. These effects may have been due to increased substrate availability due to greater soil frost, along with moderate gas diffusivities facilitating N2O production, in snow removal plots. In the second winter, spring thaw N2O fluxes and soil N2O concentration were greatest for ambient snow plots. Peak fluxes in ambient snow plots were 19 and 24 times greater than in snow accumulation and snow removal plots, respectively. Greater soil moisture in ambient snow plots overwinter could have facilitated denitrification both through decreased O2 availability and increased disruption of soil aggregates during freeze-thaw cycles. Overall, results suggest that effects of changing snow cover on N cycling and N2O fluxes were not solely a direct effect of snow depth; rather, effects were mediated by both soil water content and temperature. Furthermore, the fact that treatments with greatest mid-winter belowground N2O accumulation also had greatest spring thaw N2O fluxes in both years suggests the hypothesis that high spring thaw fluxes were due not only to spring soil conditions, but also to an effect of soil conditions in frozen soils that had