Evaluation of a lower-powered analyzer and sampling system for eddy-covariance measurements of nitrous oxide fluxes

Abstract. Nitrous oxide (N₂O) fluxes measured using the eddy-covariance method capture the spatial and temporal heterogeneity of N₂O emissions. Most closed-path trace-gas analyzers for eddy-covariance measurements have large-volume, multi-pass absorption cells that necessitate high flow rates for ample frequency response, thus requiring high-power sample pumps. Other sampling system components, including rain caps, filters, dryers, and tubing, can also degrade system frequency response. This field trial tested the performance of a closed-path eddy-covariance system for N₂O flux measurements with improvements to use less power while maintaining the frequency response. The new system consists of a thermoelectrically cooled tunable diode laser absorption spectrometer configured to measure both N₂O and carbon dioxide (CO₂). The system features a relatively small, single-pass sample cell (200 mL) that provides good frequency response with a lower-powered pump ( ∼  250 W). A new filterless intake removes particulates from the sample air stream with no additional mixing volume that could degrade frequency response. A single-tube dryer removes water vapour from the sample to avoid the need for density or spectroscopic corrections, while maintaining frequency response. This eddy-covariance system was collocated with a previous tunable diode laser absorption spectrometer model to compare N₂O and CO₂ flux measurements for two full growing seasons (May 2015 to October 2016) in a fertilized cornfield in Southern Ontario, Canada. Both spectrometers were placed outdoors at the base of the sampling tower, demonstrating ruggedness for a range of environmental conditions (minimum to maximum daily temperature range: −26.1 to 31.6 °C). The new system rarely required maintenance. An in situ frequency-response test demonstrated that the cutoff frequency of the new system was better than the old system (3.5 Hz compared to 2.30 Hz) and similar to that of a closed-path CO₂ eddy-covariance system (4.05 Hz), using shorter tubing and no dryer, that was also collocated at the site. Values of the N₂O fluxes were similar between the two spectrometer systems (slope  =  1.01, r² =  0.96); CO₂ fluxes as measured by the short-tubed eddy-covariance system and the two spectrometer systems correlated well (slope  =  1.03, r² =  0.998). The new lower-powered tunable diode laser absorption spectrometer configuration with the filterless intake and single-tube dryer showed promise for deployment in remote areas.