In this thesis I present a simple, computationally-inexpensive moist turbulence model in order to study the differences between moist and dry turbulence. The model is validated by comparing a moist-bubble simulation with ones presented in Grabowski and Clark (1993) using a more-sophisticated model. We show that the outputs compare well and that our model can easily be extended to higher resolutions due to its simplified equations and uncomplicated implementation. Measurements of liquid water content spectra from the 3843 validation run are shown having shallow slopes, implying that moist processes require high resolution. Consideration is also given to the issue of Gibb's oscillations near sharp gradients, such as at a cloud boundary. It is shown that, due to our high resolutions, the dynamics of our model are not seriously affected if corrections are not made to address them. / The model is used to study the small-scale predictability and dynamics of moist and dry shallow convective turbulence. Although moist flows are less predictable than their associated dry flows, we can account for the differences via a simple scaling. Using large-scale (the root-mean-squared vorticity) and small-scale (the dissipation wavenumber, kd) measures, we can reconcile classical predictability statistics from both wet and dry runs, with different lapse rates and relative humidities. / Finally, I present a more thorough investigation of the dynamical differences between wet and dry convective turbulence, and then consider the very small-scale (ℓ ≲ 10 m) variability of liquid water content and compare our high-resolution simulation results to existing in situ cumulus-cloud observations. We find that there is a small decrease in the spatial intermittency of vorticity in wet runs relative to dry ones. This is consistent with the idea that evaporation of the liquid water in the clouds reduces the instabilities that would lead to the most intense vortices. At the same time, the liquid water content spectra show that in these areas of intense mixing and cloud decay, the characteristic scale of variability is shifted to smaller scales compared to a passive scalar. Further integrations in which the convective forcing is removed show that as the amount of liquid water decreases through evaporation, there is delayed decay of the smallest scales of the cloud. These findings may explain the small-scale shallow liquid water content spectra from cumulus-cloud fly-through measurements reported in Davis et al. (1999).
| Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.103296 |
| Date | January 2007 |
| Creators | Spyksma, Kyle. |
| Publisher | McGill University |
| Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
| Language | English |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Format | application/pdf |
| Coverage | Doctor of Philosophy (Department of Atmospheric and Oceanic Sciences.) |
| Rights | © Kyle Spyksma, 2007 |
| Relation | alephsysno: 002672192, proquestno: AAINR38648, Theses scanned by UMI/ProQuest. |
Page generated in 0.0023 seconds