The sensitivity of convective initiation to the lapse rate of the active cloud-bearing layer

Numerical experiments are conducted using an idealized cloud-resolving model in an attempt to explore the sensitivity of deep convective initiation (DCI) to the lapse rate of the active cloud-bearing layer (ACBL – the atmospheric layer above the LFC). Clouds are initiated using a new technique that involves a preexisting airmass boundary initialized such that the (unrealistic) adjustment of the model state variables to the imposed boundary is disassociated from the simulation of convection. Reference state environments used in the experiment suite have identical mixed-layer values of CIN, CAPE, and LFC as well as identical profiles of relative humidity and wind.

Of the six simulations conducted for the experiment set, only the three environments with the largest ACBL lapse rates support DCI. The simulated deep convection is initiated from elevated sources (parcels in the convective clouds originate near 1300m) despite the presence of a surface-based boundary. DCI failure in two of the three experiments that do not produce deep convection can be explained by the fact that the LFC for parcels lifted from the layer between 1000m and ~2000m are lower in the larger ACBL lapse rate environments. In one experiment, DCI failure occurs even though thermal instability is released. Results from this experiment along with the results from a heuristic Lagrangian model reveal the existence of two convective regimes dependent on the environmental lapse rate. In one regime, parcel dilution is subcritical: the rate of increase in buoyancy due to parcel ascent exceeds the reduction in buoyancy due to dilution. DCI is likely in this regime. In the second regime, parcel dilution is supercritical: the rate of increase in buoyancy due to parcel ascent is outpaced by the rate of reduction in buoyancy from dilution. DCI is unlikely in this regime.