Results from the stable isotope sampling network in Carboeuroflux

Integrating stable isotopic measurements of canopy air and ecosystem organics with flux tower and ecophysiological data provides a powerful tool to differentiate between carbon sources and sinks, and scale-up processes from plant to ecosystem levels. During the 2001 and 2002 growing-seasons monthly flask samples of nocturnal canopy air and ecosystem organics were collected from selected forest flux sites within the Carboeuroflux network (13 sites in 2001 and 18 in 2002). Flask air was analysed for CO₂ concentration ([CO₂]), and the carbon and oxygen isotopic compositions (δ13C and δ¹⁸O) of this CO₂. The δ¹⁸O of waters distilled from leaf, stem and soil samples, and the δ13C and δ¹⁸O of these dried, homogonised organic samples were also measured. Analytical precisions were ñ0.1ppmv for [CO₂], ñ0.1 permil and ñ0.2 permil for the δ13C and δ¹⁸O of atmospheric CO₂, ñ0.05 permil for water δ¹⁸O and ñ0.1 permil for both the δ13C and δ¹⁸O of organics. The δ13C of ecosystem respired CO₂ (δ13C_R) was determined for each sampling period and location using a Keeling plot approach. Ecosystem discrimination (Δ13C_E) was estimated as the difference between the δ13Cs of background atmospheric CO₂ and ecosystem respired CO₂. The seasonal and spatial variation in these variables, and the δ13C and δ¹⁸O compositions of the organic samples are examined relative to meteorological and ecophysiological conditions. We assessed the potential for using the δ¹⁸O of ecosystem respired CO₂ (δ¹⁸O_R) together with that of soil and leaf waters to partition between the soil and above-ground respired CO₂ sources. At sites where soil δ13C varied significantly from leaf δ13C, we also assessed the partitioning potential in using the δ13C data. More intensive sampling campaigns, including incubations in branch-bags, and leaf, trunk and soil chambers, were also conducted at specific sites to examine the partitioning and scale relationships between individual source CO₂ contributions and δ13C_R. These experiments are still being conducted and their final results will be presented. Significant variations are observed in Δ13C_E and δ13C_R spatially, seasonally and between years. The observed variations in Δ13C_E and δ13C_R follow a general trend with daytime average temperature, such that the most enriched compositions are typically observed during the mid-season and in warmer, more continental climatic locations. The average seasonal range in δ13C_R at individual sites is ~5 permil, between ~-26 permil and -21 permil, although, the average range between sites is also ~5 permil. The average δ¹⁸O compositions of soil and twig waters for both years were -6.6 permil and -6.0 permil respectively (SD=2.5 permil). Throughout each season and between sites these compositions varied from 0 permil to -10 permil. The results of this network highlight some of the applications of stable isotope monitoring for assessing local and regional-scale terrestrial ecosystem dynamics. In particular, the observed regional variations in Δ13C_E may have implications for modelling regional carbon sources and/or sinks from measurements of the Δ13C of atmospheric CO₂, as these estimates currently assume constant Δ13C_E.