Direct measurement of biosphere-atmosphere isotopic CO2 exchange using the eddy covariance technique

Quantifying isotopic CO2 exchange between the biosphere and atmosphere presents a significant measurement challenge, but has the potential to provide important constraints on local, regional, and global carbon cycling. Past approaches have indirectly estimated isotopic CO2 exchange using relaxed eddy accumulation, the flask-based isoflux method, and flux-gradient techniques. Eddy covariance (EC) is an attractive method because it has the fewest theoretical assumptions and the potential to give a direct measure of isotopic CO2 flux, but it requires a highly sensitive and relatively fast response instrument. To date, no such field measurements have been reported. Here we describe the use of a closed-path tunable diode laser absorption spectroscopy and eddy covariance (EC-TDL) system for isotopic (C(16)O2, (CO2)-C-13, (COO)-O-18-O-16) flux measurements. Results are presented from an intensive field experiment conducted over a soybean canopy from 18 July to 20 September 2006. This experiment represents a rigorous field test of the EC-TDL technique because the transport was dominated by relatively high frequency eddies. Net ecosystem CO2 exchange (FN) measured with the EC-TDL system showed strong correlation (r(2) = 0.99) in the half-hourly fluxes with an EC open-path infrared gas analyzer (EC-IRGA) over the 60-d period. Net CO2 flux measured with the EC-IRGA and EC-TDL systems agreed to within 9%. Flux loss associated with diminished frequency response beyond 1 Hz for the EC-TDL system was approximately 8% during daytime windy (> 4 m s(-1)) conditions. There was no significant evidence of a kinetic-type fractionation effect related to a phase shift among isotopologues due to tube attenuation. Investigation of isotopic spectral similarity in the flux ratio (delta(x)(N)) for both (CO2)-C-13 and (COO)-O-18-O-16 transport showed that delta(x)(N) was relatively independent of eddy scale for this ecosystem type. Flux loss, therefore, did not significantly bias delta(N)x. There was excellent agreement between isofluxes (F-delta(x)) measured using the flux-gradient and eddy covariance methods. Application of the EC-TDL technique over rougher surfaces or below canopy, where the flux-gradient approach is difficult to apply, appears promising for obtaining continuous long-term measurements of isotopic CO2 exchange.