טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentHilel Roni
SubjectTurbulence of anabatic (up-slope) thermally driven
flow
DepartmentDepartment of Civil and Environmental Engineering
Supervisor Dr. Dan Liberzon
Full Thesis textFull thesis text - English Version


Abstract

Understanding the physical nature of thermally driven upslope (anabatic) and downslope (katabatic) turbulent flows in the atmospheric-boundary layer is of paramount importance for producing accurate weather forecasts, investigating climate processes, and gaining a better understanding of pollution transport in complex terrains. Of particular interest are the microclimates of urban centers which are often positioned in valleys and next to mountain ridges. The large scales of turbulent flows may simply be measured with inexpensive low spatial and temporal resolution instrumentation such as the ultrasonic anemometers (Sonic), while fine scales are generally estimated using Kolmogorov’s  theory due to measuring impediments such as the calibration of hot-wire/film anemometers. However, natural thermally driven turbulent flows are anisotropic and not homogeneous deducing that Kolmogorov’s theory is not sufficient to describe such flows nor to be used as an extrapolation basis; nonetheless, these flows lack comprehensive models to describe them. These models are ought to include several statistical parameters such as turbulence intensity (TI), turbulent kinetic energy (TKE) dissipation rates, velocity derivative skewness, and various length scales with respect to Re. The length scales consist of Kolmogorov length scale (an indication of viscous forcing dominance), Taylor length (an indication of the size at which viscous forcing begins to play a role in the cascade of TKE dissipation), and horizontal length scale (an indication of the size of eddies in which the energy input occurs). Eventually, the fully resolved spectra of all three-velocity field components is required in order to characterize the behavior of the turbulent flow. Furthermore, the often-present phenomenon of short-timed rapid increase in velocity fluctuations intensity, bursting, is virtually impossible to capture in natural setups with low spatial-temporal instrumentation, yet it plays an important role in setting the turbulent flow energy cascade. Therefore, in addressing the need for high accuracy data on the turbulent statistics of anabatic flows, a field experiment was staged to obtain continuous high accuracy measurements of the diurnal cycle of thermally driven flow over moderate slope. The main instrument allowing high-resolution measurements of the velocity field in interest was a recently developed collocated Sonic and hot-film (HF) anemometer?Combo probe. In this instrument, the simultaneously measured slow data from the Sonic is used for in-situ calibration of the HF voltages using Neural Networks. This allowed investigation of both mean and fluctuating components of the upslope flow, and additional instruments were used to record temperature fluctuations at two heights effectively detecting periods of stable and unstable stratification of the air mass on the slope. Strong correlation of the developing flow with the diurnal heating was detected in absence of significant synoptic forcing. Detailed analysis of turbulence statistics is provided; the main products being the fully resolved spectra of all three velocity field components, various turbulence statistics, characterization of the bursting phenomenon (i.e. length scales, frequency of appearance of bursting, and examination of the generation mechanism), comparison of and the flow’s characterizing length scales (Kolmogorov, Taylor, and horizontal) in minutes including and excluding bursting, and empirical fits were made available for future measurements with low-resolution instruments and numerical models.