Sensitivities of Near-Ground Simulated Parcel Trajectories
Vandeguchte, Andrew Peter
MetadataShow full item record
Parcel trajectory analysis has become commonplace in the study of simulated severe convection, particularly that which deals with the development and maintenance of near-ground vertical vorticity. However, there are a number of unsolved problems with analyzing simulated trajectories that exist near the ground. One of these unsolved problems is how to deal with parcels that pass beneath the lowest scalar model level. Horizontal velocity and scalar variables must be extrapolated from the model grid to a parcel's position along a trajectory that is beneath the lowest scalar model level. However, there is no substantiated way to do this. Using an idealized cloud model with a free-slip lower boundary, an example of two different methods--out of an infinite number of possibilities that satisfy the free-slip boundary--of accomplishing this are shown. It is determined that it is extremely difficult to determine the true, physically and dynamically consistent trajectory of any sub-domain parcel in the free-slip case. The sensitivity of near-ground parcels to the choice of lower boundary condition is also not well understood. Recent research has revealed that in some supercell simulations where friction is included, there is a prevalence of inflow trajectories in the near-ground tornado-like vortex. However, it is unclear whether the presence of such inflow trajectories is sensitive to the inclusion of friction or not. Herein, it is shown that a supercell simulation with friction can result in a substantially different storm structure than an otherwise identical simulation with no friction. Because of this, in the current study the direct sensitivity of near-ground parcels could not be established. However, it is shown that the inclusion of friction has a number of other impacts. One of which is that the trajectories of parcels that are initialized in the simulations will be dramatically different due to the inclusion of friction because the storm evolves so differently. Another impact is that a warmer cold pool and weaker outflow were present in the simulation with friction compared to the simulation without friction. Finally, inflow trajectories were present in the simulations without friction, a result that is seemingly disparate with recent research. Further analysis is required to fully understand the presence of such inflow trajectories in a non-friction simulation. Ultimately, when parcel trajectories are considered as an analysis tool for future research, it is important to also consider the effects that the treatment of the region beneath the physical model domain and lower boundary condition may have on the results of that analysis.