Theoretical predictions of lift-off heights, including the three levels of approximations discussed above, are based on the premise that the flame tip locates itself on the stoichiometric surface at a point where the flow velocity is close to the value SLo.
The flow velocity measured in the simulations at the flame base is indeed very close to SLo.
However, the lift-off height calculated on this assumption does not match the DNS results very well.
This is only possible, if the flow perturbation due to the edge flame significantly alters the velocity and species distribution of the cold flow theory changing the location of the stoichiometric surface and/or the position where the flow velocity equals SLo. Since the simulated flames are observed upstream from their expected theoretical positions (Fig. 11 and 12), we must conclude that the edge-flame must modify the mixture locally in order to establish this stabilization point