Augmented Bowen ratio analysis, II: Ohio comparisons

Abstract

An experiment conducted over a field of maize in Ohio during the growing season of 2015 provided measurements sufficient to compare among three methods for quantifying net ecosystem exchange of sensible heat, latent heat and carbon dioxide: eddy covariance (EC), the familiar Bowen ratio energy balance (BREB) method and a statistical procedure making use of surface temperature measurements by downward-looking infrared thermometry (the augmented Bowen ratio analysis – ABRA). EC and BREB yield run-by-run estimates of fluxes. The third method (ABRA) gives average values over prescribed ensembles of data. The Ohio observations support the conventional understanding that EC is the preferred methodology provided the site is large enough. Both BREB and ABRA can make use of measurements closer to the surface than EC, and hence are more suited for studies of small test plots. The ABRA method provides results agreeing with conventional BREB and EC in daytime except for periods near dawn and dusk when heat storage can be a substantial contributor in the surface heat energy budget. The unattributed heat storage in the morning found in analysis of observations made elsewhere is substantiated. It is concluded that the apparent heat storage term in the surface heat budget is most influenced by heating of crop biomass and the energy demands of photosynthesis. The comparisons provide clear evidence of the difficulties associated with application of all gradient-based methodologies at night (affecting both BREB and ABRA) when atmospheric stability often prohibits reliable association of fluxes with gradients. CO2 measurements reveal substantial differences among the three different methodologies (ABRA, BREB and EC), with EC appearing to be more robust and therefore preferred if the test area is large enough.