Handout (656.1 kB)
The simulations indicate that the cellular structure in the convective region depends on how the MAUL is perturbed. When the environment is horizontally homogeneous and free of perturbations, the model often produces regularly spaced plumes in the convective region. These plumes tend to have undilute cores and long lifetimes, align with the low-level shear vector, and have an average spacing of ~3km. This ~3 km spacing is unexpected since theory indicates that the smaller scales should grow faster in a statically unstable layer. Hence the simulations suggest that another process is acting to perturb the MAUL at scales of ~3 km. To test the latter hypothesis and the robustness of the simulated plumes, random, 0.1 K temperature perturbations are added to the pre-squall-line environment; in this case the simulations show that the convective region is instead dominated by small-scale, turbulent eddies, with a scale consistent with the smallest resolvable by the model. Furthermore, there is no indication of plumes in this case. We also find that entrainment is more effective, which leads to lower cloud tops (by several km), less rainfall (by 70%), and weaker near-surface winds.
We surmise that the runs with small random perturbations are more representative of the actual atmosphere, which is constantly perturbed by variations in terrain, land use, and boundary-layer eddies. On the other hand, there is some evidence that plumes exist in some observed MCSs (although they have spacings larger that ~3 km). The present simulations suggest that preexisting perturbations are likely to be a determining factor in the existence, structure, and spacing of plume-like overturning in squall lines with MAULs.