Refining the alkenone-pCO2 method II: Towards resolving the physiological parameter ‘b’

Abstract

Haptophyte algal biomarkers called alkenones are widely used to reconstruct atmospheric CO2 in Earth’s Cenozoic history. This method is based on the notion that the algal carbon isotope fractionation during photosynthesis, as represented by εp37:2, is a function of seawater CO2 concentration and algal physiology. Constraining the algal physiological parameter, known as the ‘b’ term, is the key for successful applications of the alkenone-CO2 method. Using sensitivity analyses, we show that the growth rate (μ), cell size (r), and membrane permeability (P) are the most important variables to determine b. For all life on Earth, body size is a key factor that regulates metabolic rates. Exploiting the interdependence between phytoplankton cell size and growth rate, and specifically, the r – μ relationship for coccolithophores, we show that the length of fossil coccoliths (Lcoccolith) produced by ancient alkenone-synthesizers can be used to estimate r and therefore μ. Combining our new Lcoccolith data and published εp37:2 from the South China Sea Site MD01-2392, existing results from ODP Site 925, and ice core CO2, we determined the cell membrane permeability (P = 5.09 × 10−5 m s−1) for the Pleistocene community employing a bootstrap resampling technique. These new methods of constraining r, μ and P, combined with proxy-derived temperature (T), led us to rebuild b as a variable for each sample individually, which is subsequently used for alkenone-CO2 calculations. Application of this approach established pCO2 of the last 3 glacial-interglacial cycles, which turns out to be comparable with the ice core data in both the amplitude of changes and absolute values. It also reconciles the published Eocene – Oligocene alkenone-CO2 data which showed large geographical differences, with the new estimates much more consistent among different sites, and in line with other proxy-based results and ice sheet model predictions.