Abstract:
Long-term carbon cycle models and independent proxies indicate that the atmospheric CO2 concentration was substantially higher in the Mesozoic and early Tertiary than now. The responses of tree growth to a high CO2 environment must therefore be understood if present-day relationships between climate and tree growth are to be used with confidence for palaeoclimate reconstructions from Mesozoic and Tertiary fossil woods. A brief review of experimental data shows that stimulation of leaf photosynthesis, suppression of plant respiration and lower transpiration rates under elevated CO2 (500–700 ppm) can reduce the ratio of wood basal area to tree height, but increase total basal area growth in wood by 20–25%. To assess the consequences of these plant CO2 responses for tree growth during ancient ‘greenhouse’ episodes, we developed a mathematical model explicitly incorporating the physiological mechanisms underpinning the observed CO2 effects. Application of this model to predict the height of coniferous forests in the Late Jurassic (150 Ma) and mid-Cretaceous (100 Ma) indicated maximum heights in the mid-latitudes of over 60 m and 40–60 m respectively. Corresponding wood growths of these forests, expressed as basal area growth rates, were 1.5–2.5 cm2 and 0.75–1.75 cm2 sapwood m−2 ground area yr−1, values exceeding those of modern conifers in warm temperate sites. These results were highly sensitive to the concentration of atmospheric CO2, and this CO2-sensitivity depended strongly on climate. Consequently, direct application of present-day growth–climate relationships to fossil woods may be inappropriate for the Mesozoic and Tertiary. However, models that explicitly consider the mechanisms of plant CO2 responses may provide a useful method for accounting for this CO2-sensitivity, allowing more secure palaeoclimate reconstructions from material of this age.