## LSU Historical Dissertations and Theses

1989

Dissertation

#### Degree Name

Doctor of Philosophy (PhD)

D. D. Reible

#### Abstract

Motivated by a need to assess pollutant transport by upslope flows, an investigation has been conducted into the fundamentals of natural convection flow over inclined surfaces. Particular attention was focused on the influence of ambient fluid stability. Field studies were performed using tracer gas releases into the upslope flow over a Southern California mountain range. The field studies served to reveal the presence of a split slope flow recirculation and demonstrated the impact of this recirculation on the transport of pollutants from a valley. In order to pursue a controlled investigation of the phenomena found in the field work, a laboratory model was developed using water as a working fluid. Extensive dye studies demonstrated the presence of this recirculation to varying degrees in nearly every configuration with a stable layer present. Heat transfer experiments were conducted with the laboratory model to refine and validate the experimental techniques used. Comparisons are made with existing theoretical and empirical predictions where available. Existing correlations for inclined surfaces are extended two orders of magnitude lower in Rayleigh number. A modification to vertical theory based on simply replacing g with gcos$\theta$ is shown to be useful for inclinations down to 75$\sp\circ$ from vertical. Transition ranges and empirical correlations are expressed for inclinations of 0$\sp\circ$, 15$\sp\circ$, 30$\sp\circ$, 45$\sp\circ$, 60$\sp\circ$, 75$\sp\circ$, and 90$\sp\circ$. Overall correlations are also reported with apparently far less scatter of data than for any previously reported research with inclined surfaces. Experimentation with stratified ambient fluid resulted in the observation that, with turbulent flow, stratification could apparently be disregarded and heat transfer simply calculated from local conditions. The observed heat transfer coefficients are essentially independent of position along the slope suggesting that an approximate analytical model of upslope flows developed by L. Prandtl in 1942 is applicable. However, quantitative laboratory results showed that Prandtl's one dimensional theory underpredicted the observed boundary layer depths. The results are in good agreement with observed characteristics of atmospheric slope flows.

212

COinS