Figure 6. Posterior probability distributions for selected carbon cycle variables from Bayesian inverse analysis.

The first column is the posterior distributions from our nominal model, whereas the second column is distributions from a modified model, where the Michaelis–Menten law is used to describe the pCO₂ dependence of continental weathering. Results for the two parameterizations are similar. Dotted lines represent the median value with 1σ error bars. From the marginal distributions, we conclude that the temperature sensitivity of continental weathering is weak (T_e>15 K), Cretaceous silicate weatherability (1+W) was ∼half modern weatherability, and the average equilibrium climate sensitivity is ∼5.6 K for a CO₂ doubling.