Is it Possible to Use the Surface Renewal Technique to Estimate GHG Emissions?

Micrometeorological observations are crucial for measuring GHG emissions as surface fluxes. Those methods that function when measurements are taken close to the canopy top (i.e., in the roughness sub-layer) are of special interest because they reduce the fetch requirements and avoid the need to modify the measurement level as the vegetation grows (Castellvi and Snyder, 2009). Recent advancement in the surface renewal techniques (SR) are important for their capacity to independently measure sensible (H) and latent heat (LE) fluxes while avoiding some of the shortcomings of the eddy covariance method (EC) (Suvočarev et al, 2014). Unlike EC, SR avoids orientation limitations, leveling requirements and instrumentation separation and shadowing issues. We applied the recent approach as suggested by Castellvi et al. (2008) over three months (May to August, 2015) of high-frequency data collected by EC equipment from a rice field in Arkansas. The main goal was to extend this SR application to GHG fluxes, CO2 (Fc) and CH4 (Fm), as rice is shown to be one of the biggest water consumers with the highest GHG emissions in cultivated agriculture.

The results show high correlation between three EC and SR fluxes (H, LE and Fc) when they are compared for all atmospheric stability conditions (R2 > 0.87). Some overestimation is observed for SR with respect to EC fluxes, similar to the findings of Castellvi et al. (2008) for rangeland grass. For all the data, SR analysis results were about 9%, 7% and 18% higher than the EC results for H, LE and Fc, respectively. These higher flux estimates resulted in better energy balance closure. Fm estimates demonstrate more scatter between the two methods, so for a preliminary investigation we focus on positive flux only. The SR estimation of positive Fm under all atmospheric conditions showed a similar trend but more scatter when compared to EC results (R2=0.59). This reduced correlation may be the consequence of different transport mechanisms for this gas that are not well captured by SR technique – these effects may include the bursts of concentration from ebullition and different convection relationships for the relatively light CH4 molecule. More research should be done to address the potential of SR in Fm estimation as there are no previous results reported on this topic and the benefits of SR could reduce time series gaps and lower flux uncertainty for this environmentally important scalar.