Long-term nitrous oxide (N2O) fluxes in the upper Midwest USA: A comparison between annual and perennial systems

Ilya Gelfand, Iurii Scherbak, Neville Millar, Alexandra N. Kravchenko, and G. Philip Robertson
W.K. Kellogg Biological Station; Great Lakes Bioenergy Research Center; Department of Plant, Soil, and Microbial Sciences, Michigan State University; School of Natural Resource Sciences, Queensland University of Technology

Presented at the All Scientist Meeting (2015-04-15 to 2015-04-16 )

The spatial and temporal variability of soil nitrous oxide (N2O) fluxes makes their evaluation and prediction difficult. To help address these issues we examined a long-term dataset, containing more than 20 years of biweekly measurements of soil N2O emissions, together with measurements of multiple environmental and management parameters, including soil N content and soil moisture, from eleven ecosystems: four annual cropping (corn-soybean-wheat) rotations (conventional, no-till, reduced input, and biological); perennial: alfalfa and tree plantations (coniferous and deciduous); and four successional systems: never tilled grassland, early and mid-successional, and deciduous forest. We used these measurements to determine the effect of different agricultural and land management practices on soil N2O emissions. We found a strong relationship (p < 0.05) between soil NO3- availability and soil NO3- production potential and annual cumulative N2O fluxes. Management and specific crop types had a strong influence on cumulative N2O emissions, with cover crops reducing emissions by ~40% during the wheat portion of the rotation. Corn and soybean crops under conventional management had lower emissions than under biological management, while wheat exhibited the opposite trend. Among the perennial ecosystems, alfalfa emitted ~six times more N2O than poplar, ~four times more than the unmanaged successional communities, and ~two times more than the coniferous plantation and the deciduous forest, which emitted similar amounts. Daily soil N2O emissions were poorly predicted by any individual or combination of the measured variables, and had non-normal flux distributions greatly skewed toward high flux extremes.