Flux Variance Partitioning: A New Approach to Advance Eddy Covariance Observations for Greenhouse Gas Emissions.

Eddy covariance (EC) is a well-established, non-intrusive observational technique that has long been used to measure the net carbon balance of numerous ecosystems including crop lands for perennial crops such as orchards and vineyards, and pasturelands. While EC measures net carbon fluxes well, it cannot directly observe component fluxes of the carbon cycle including Gross Primary Production (GPP) and Respiration (Re) and therefore must be combined with other approaches to estimate photosynthesis and respiration changes in response to management strategies. Most approaches to EC flux partitioning involve parameterizing Re as function of nighttime temperature and extrapolating to estimate daytime Re or parameterizing GPP as a light response function. Both of these approaches have significant potential weaknesses especially with determining GPP and Re under extreme or unusual environmental conditions. In this talk, we present a relatively novel approach to carbon flux partitioning, the flux variance partitioning method (FVP), which combines leaf level water use efficiency with high frequency CO₂ and H₂O observations to partition net carbon flux into GPP and Re. We apply FVP to multiple EC sites with different cropping systems including a furrow irrigated peach orchard and multiple sugarcane fields. Results from the peach orchard show a strong potential for FVP, but with larger differences in approaches for Re compared to GPP. The results indicate that the FVP method can provide a complimentary approach to existing temperature based parameterizations for partitioning EC fluxes and can provide independent validation for biophysical controls on GPP and Re.