Seasonal dynamics and partitioning of isotopic CO2 exchange in a C3/C4 managed ecosystem

Ecosystem-atmosphere fluxes of ¹²CO₂ and ¹³CO₂ are needed to better understand the impacts of climate and land use change on ecosystem respiration (F(R)), net ecosystem CO₂ exchange (F(N)), and canopy-scale photosynthetic discrimination (Δ). We combined micrometeorological and stable isotope techniques to quantify isotopic fluxes of ¹²CO₂ and ¹³CO₂ over a corn-soybean rotation ecosystem in the Upper Midwest United States. Results are reported for a 192-day period during the corn (C₄) phase of the 2003 growing season. The isotopomer flux ratio, d¹³CO₂/d¹²CO₂, was measured continuously using a tunable diode laser (TDL) and gradient technique to quantify the isotope ratios of F(R) (δ(R)) and F(N) (δ(N)). Prior to leaf emergence δ(R) was approximately -26 ppt. It increased rapidly following leaf emergence and reached an average value of -12.5 ppt at full canopy. δ(R) decreased to pre-emergence values following senescence. δ(N) also showed strong seasonal variation and during the main growing period averaged -11.6 ppt. δ(R) and δ(N) values were used in a modified flux partitioning approach to estimate canopy-scale Δ and the isotope ratio of photosynthetically assimilated CO₂ (δ(P)) independent of calculating canopy conductance or assuming leaf-scale discrimination factors. The results showed substantial day-to-day variation in Δ with an average value of 4.0 ppt. This flux-based estimate of Δ was approximately 6 ppt lower than the Keeling mixing model estimate and in better agreement with leaf-level observations. These data were used to help constrain and partition F(R) into its autotrophic (F(Ra)) and heterotrophic (F(Rh)) components based on the numerical optimization of a mass balance model. On average F(Ra) accounted for 44% of growing season F(R) and reached a maximum of 59% during peak growth. The isotope ratio of F(Rh) (δ(Rh)), was -26 ppt prior to leaf emergence, and became increasingly ¹³C enriched as the canopy developed indicating that recent photosynthate became the dominant substrate for microbial activity. Sensitivity analyses substantiated that F(Rh) had a major influence on the seasonal pattern of δ(R), δ(N) and the isotopic disequilibrium of the ecosystem. These data and parameter estimates are critical for validating and constraining the parameterization of land surface schemes and inverse models that aim to estimate regional carbon sinks and sources and interpreting changes in the atmospheric signal of δ¹³CO₂.