The overarching goal of the CARBOEUROFLUX programme is to improve our understanding on magnitude, location, temporal behaviour and causes of the carbon source/sink strengths of terrestrial ecosystems which can be used to improve the negotiation capacity of the European Community in the context of the Kyoto protocol. (for further details refer to CARBOEUROFLUX web site: http://www.bgc-jena.mpg.de/public/carboeur/projects/index_p.html)
Within this framework we are strongly limited in two important issues. First, in identifying and tracing individual fluxes of CO2 associated with photosynthetic assimilation or respiration. Second, in understanding observed large inter-annual variations in the net CO2 exchange between land ecosystems and the atmosphere.
Stable isotopes are among the few tools available to us to improve on these fronts. In the first case, by taking advantage of the specific isotopic signals associated with photosynthesis and respiration. In the second case, by using ecosystem-scale isotopic discriminations as indicators for ecophysiological response to environmental change.
A new isotopic sampling program was developed and applied in 18 tower flux sites across Europe. In this program air samples above and within the canopy are taken in glass flasks. Analysis of these samples is used to examine the co-variations in [CO2] and its stable isotopic compositions (d13C and d18O) in the canopy over time and space. These relationships can, in turn, be used to estimate the isotopic signals associated with net ecosystem CO2 exchange and Ecosystem Discrimination. By complementing such measurements with sampling and analysis of organic materials (leaf, stem and soil) we can potentially constrain the relative contributions of specific CO2 source or sinks in the ecosystem.
One of the common ways to use the co-variations of [CO2] and isotopes in canopy air is to apply the 'Keeling model' (Keeling 1958, 1961). This approach will also be used here among other data analyses. The Keeling model assumes that the observed variation in [CO2] and isotopes reflect simple mixing between a unique CO2 sink/source in the ecosystem and unique CO2 in the PBL. Accordingly, a linear fit to the data, appropriately plotted, provides information on the isotopic signal of the CO2 source/sink in the system (Figure 1).
Figure 1. Two forms of Keeling plot using nocturnal air samples from Yatir forest, Israel on
Similar approach can be used for both d13C and d18O. The results, however, are independent since different processes underlie variations in the two isotopes. For d13C, it is based on the discrimination associated with assimilation by Rubisco, and for d18O, it is based on oxygen exchange between CO2 and water in soil and leaves.
Two recent reviews provide more information on the stable isotope approach and its potential uses (Pataki et al., 2002; Yakir and Sternberg, 2000).