Micrometeorological and chamber methods for measurement of nitrous oxide fluxes between soils and the atmosphere: Overview and conclusions

Emissions of N 2 O from agricultural grassland fertilized with 185 kg N ha −1 have been measured using a range of chamber and micrometeorological methods at a common site in the lowlands of Scotland. Measurements were made over similar periods (30 to 60 min) by all methods, but the areas over which the fluxes were measured ranged from 0.008 m 2 for the smallest chambers to 10 4 to 10 5 m 2 for the micrometeorological methods. The fluxes measured using chambers ranged from less than 10 to more than 1700 ng N 2 O‐N m −2 s −1 ; they were a factor of 3 larger from a previously grazed area than from the remainder. Fluxes showed great spatial variability and also a general decline with time following a peak after fertilization. Micrometeorological measurements were made by eddy correlation using fast‐response tunable diode laser absorption spectroscopy (TDLAS) and by aerodynamic (flux gradient) methods using Fourier transform infrared spectroscopy (FTIR), gas chromatography (GC), and TDLAS. All of the flux gradient methods provided similar fluxes of N 2 O over the ungrazed section of the field, with averages over the measurement period in the range 52 to 55 ng N 2 O‐N m −2 s −1 . Eddy correlation measurements with the TDL averaged 85 and 43 ng m −2 s −1 on the two days when measurements were made. Mean fluxes from the chamber methods, using GC, FTIR, and long‐path infrared spectroscopy to detect N 2 O, were larger than those from the micrometeorology and ranged from 280 ng N 2 O‐N m −2 s −1 for the smallest chambers to 210 ng N 2 O‐N m −2 s −1 for the 0.13/0.49 m 2 chambers and 300 ng N 2 O‐N m −2 s −1 for the 62 m 2 chamber. The different techniques employed averaged over different spatial scales, and the measurements related to different areas of the field. Nonetheless, the different micrometeorological methods gave similar fluxes. The higher values obtained by the chamber methods may have been due either to the spatial variability of the fluxes at the site, with the chambers being located in regions of relatively greater source strength, or to factors associated with the methods themselves.