Atmospheric boundary layer dynamics in regions of complex terrain

Data from four complex terrain field experiments in the atmospheric boundary layer have been studied. In particular, unique flow phenomena associated with morning and evening transition periods, as well as temporally unsteady stable boundary layers, are discussed. A non-local heat flux model was developed and implemented to improve modeling of the breakdown of the stable morning atmospheric boundary layer. The focus of this work was to collect, analyze and understand complex flow phenomena data and develop models based on this understanding.; Data from the Phoenix Air Flow Experiment (PAFEX-I) field campaign was compared with existing laboratory experiments, Large Eddy Simulations (LES) and field experiments of the evening decay of the convective boundary layer. The decay of various dynamic variables, including temperature fluctuations, vertical velocity fluctuations, turbulent kinetic energy and sensible heat flux, were examined. The decay of turbulent kinetic energy was modeled with a simplified equation that includes the time-dependent heat flux term. The model was found to perform better than previous models and compared well with both LES results and field data. It was found that various time scales dominate during different periods of the decay of the convective boundary layer. Once the sensible heat flux near the surface becomes negative, the turbulent kinetic energy decay time scale was found to be consistent with h/ u_*. Data are also presented that show the direct effect of the various mixing periods on Particulate Matter (PM) pollution and carbon monoxide (CO) pollution during the experiment. It is shown that the decay of the heat flux correlation coincides in the highest levels of pollution during the decay period.; The flux Richardson number Ri_f (or the mixing efficiency) for the stably stratified atmospheric boundary layer was investigated as a function of gradient Richardson number Ri_g using two different field experiments: the Vertical Transport and Mixing Experiment (VTMX) in Salt Lake City, Utah and a long-term rural field data set from Technical Area 6 at the Los Alamos National Laboratory in Los Alamos, New Mexico. The results show an increasing trend of Ri_f until a critical gradient Richardson number Ri_g,cr is reached; wherein Ri_f becomes a maximum (≈0.45). The results agreed well with the laboratory stratified shear layer measurements, but were at odds with some commonly used Ri_f parameterizations, particularly under high Ri_g conditions. The observed swings of the buoyancy flux and turbulent kinetic energy production were consistent with the concept of global intermittency of stratified shear flows. A multi-layer slope flow example using the PAFEX-II data set is presented in conjunction with this work as a real world example of how this type of Richardson number parameterization might be applied in predicting high-pollution episodes.