Intraspecific variation in juvenile tree growth under elevated CO2 alone and with O-3: a meta-analysis


Atmospheric carbon dioxide (CO2) concentrations are expected to increase throughout this century, potentially fostering tree growth. A wealth of studies have examined the variation in CO2 responses across tree species, but the extent of intraspecific variation in response to elevated CO2 (eCO(2)) has, so far, been examined in individual studies and syntheses of published work are currently lacking. We conducted a meta-analysis on the effects of eCO(2) on tree growth (height, stem biomass and stem volume) and photosynthesis across genotypes to examine whether there is genetic variation in growth responses to eCO(2) and to understand their dependence on photosynthesis. We additionally examined the interaction between the responses to eCO(2) and ozone (O-3), another global change agent. Most of the published studies so far have been conducted in juveniles and in Populus spp., although the patterns observed were not species dependent. All but one study reported significant genetic variation in stem biomass, and the magnitude of intraspecific variation in response to eCO(2) was similar in magnitude to previous analyses on interspecific variation. Growth at eCO(2) was predictable from growth at ambient CO2 (R(2)aEuro…=aEuro…0.60), and relative rankings of genotype performance were preserved across CO2 levels, indicating no significant interaction between genotypic and environmental effects. The growth response to eCO(2) was not correlated with the response of photosynthesis (PaEuro…> aEuro…0.1), and while we observed 57.7% average increases in leaf photosynthesis, stem biomass and volume increased by 36 and 38.5%, respectively, and height only increased by 9.5%, suggesting a predominant role for carbon allocation in ultimately driving the response to eCO(2). Finally, best-performing genotypes under eCO(2) also responded better under eCO(2) and elevated O-3. Further research needs include widening the study of intraspecific variation beyond the genus Populus and examining the interaction between eCO(2) and other environmental stressors. We conclude that significant potential to foster CO2-induced productivity gains through tree breeding exists, that these programs could be based upon best-performing genotypes under ambient conditions and that they would benefit from an increased understanding on the controls of allocation.