Habitat responses of fossil plant species to palaeoclimate – Possible interference with CO2?

Abstract

One important tool for palaeoclimate reconstruction uses climate space data of extant species representing the Nearest Living Relatives of fossil taxa, under the assumption of taxon-specific climate demands remaining unchanged through time. This is problematic because the calibration of these proxies is based on current vegetation that has grown under current CO2. Atmospheric CO2 may interfere with taxon-specific climate demands, particularly water availability, because stomatal conductance is negatively correlated with CO2. In order to assess direct effects of CO2 on gas exchange, results of numerous experimental studies are available to provide benchmark data. Experiments will, however, always provide data which are restricted to response ranges that living plants are able to express. Simulation models of plant gas exchange offer additional possibilities to explore the influence of CO2 on gas exchange and to consider evolutionary adaptation to gas exchange.In this study, the potential effects of a changing CO2 level on photosynthesis performance under simultaneously changing water availability and temperature is evaluated by using a model based on optimized gas exchange, including different scenarios of photosynthesis regulation that may result from evolutionary adaptation. The results illustrate the impact of changing CO2 on water and temperature demands under these various conditions, indicating 1) a particularly substantial influence of CO2 under low CO2 (<380ppm); 2) for higher levels of CO2 (>380ppm) a decreasing influence under further increasing CO2, and 3) the relevance of possible evolutionary adaptation of the photosynthetic machinery to a changing CO2 level. Downregulation of photosynthesis, for instance, leads to a distinctly enhanced response of gas exchange to increasing CO2, particularly under low to moderate levels (180ppm–380ppm). Changes in atmospheric CO2 therefore adds an additional source of uncertainty to those already considered for NLR based methods. It may, however, also offer new possibilities for palaeoecophysiological research.