Environmental controls on the carbon isotope composition of ecosystem-respired CO2 in contrasting forest ecosystems in Canada and the USA

We compared the carbon isotope composition of ecosystem-respired CO2 (delta13C(R)) from 11 forest ecosystems in Canada and the USA and examined differences among forest delta13C(R) responses to seasonal variations in environmental conditions from May to October 2004. Our experimental approach was based on the assumption that variation in delta13C(R) is a good proxy for short-term changes in photosynthetic discrimination and associated shifts in the integrated ecosystem-level intercellular to ambient CO2 ratio (c(i)/c(a)). We compared delta13C(R) responses for three functional groups: deciduous, boreal and coastal forests. The delta13C(R) values were well predicted for each group and the highest R2 values determined for the coastal, deciduous and boreal groups were 0.81, 0.80 and 0.56, respectively. Consistent with previous studies, the highest correlations between delta13C(R) and changes in environmental conditions were achieved when the environmental variables were averaged for 2, 3 or 4 days before delta13C(R) sample collection. The relationships between delta13C(R) and environmental conditions were consistent with leaf-level responses, and were most apparent within functional groups, providing support for our approach. However, there were differences among groups in the strength or significance, or both, of the relationships between delta13C(R) and some environmental factors. For example, vapor pressure deficit (VPD) and soil temperature were significant determinants of variation in delta13C(R) in the boreal group, whereas photosynthetic photon flux (PPF) was not; however, in the coastal group, variation in delta13C(R) was strongly correlated with changes in PPF, and there was no significant relationship with VPD. At a single site, comparisons between our delta13C(R) measurements in 2004 and published values suggested the potential application of delta13C(R) measurements to assess year-to-year variation in ecosystem physiological responses to changing environmental conditions, but showed that, in such an analysis, all environmental factors influencing carbon isotope discrimination during photosynthetic gas exchange must be considered.