Model fitting to divergence and diversity dynamics.

AIC is Akaike's information criterion with a second order correction for small sample sizes [61] and SQ is the sum of squared errors; subindex F indicates the full model; S1 submodel 1; and S2 submodel 2. The parameters dS and dmax are estimated from the full model (Figure 3C), along with 95% CIs (brackets) obtained by bootstrapping divergence and diversity dynamics 103 times. The preferred model has the lowest AIC value.