Laserfiche WebLink
<br />C IT X2 (2 U(z) tS) <br />Q = f R ' Re t ' --u- T <br />g <br /> <br />where 2, H are parameters depending on geometric similarity. <br />Ret depends upon turbulence similarity and U(z)/Ug depends <br />upon kinematic or flow similarity. ts/T refers to time <br />similari ty . <br />These nondimensional variables along with the appro- <br />priate upstream boundary conditions, such as, a proper <br />velocity and turbulence profile make up the essential elements <br />of the neutral airflow model. <br />In this physical model the airflow was aerodynamically <br />rough which allows relaxing the requirement for Reynolds number <br />duplication (Refs. 30 and 33). Usually the turbulent Reynolds <br />or Peclet numbers cannot be evaluated because of the lack of <br />proper field data. However, a test of the 'similarity of tur- <br />bulent Reynolds number between field and model airflow was <br />evaluated in the Eagle River Valley-Climax study (Ref. 30). <br />Since the field and laboratory measurements were <br />seldom obtained for all three space dimensions, i.e., x, y , <br />z , the vector notation was dropped and the similarity relations <br />were evaluated in terms of one or two dimensions. <br />2. Barostromatic airflow model <br />a. Thermal similarity <br />In this particular physical model one attempts to <br />simulate the normal temperature stratification observed in <br />the atmosphere, i.e., an increase of potential temperature with <br />height. The similarity relation can be written in terms of <br />the potential temperature and air density, as <br /> <br />2-5 <br /> <br />.', <br />.~ <br /> <br />1:.. ~ > 0] == -' l. ~ > 0] <br />tr az F p az M <br /> <br />2-6 <br /> <br />. .' <br /> <br />The above statement infers that the physical model requires <br />temperature or density stratification. <br />b. Airflow and dispersion similarity <br />The principal difference between this airflow and <br />the neutral airflow is the temperature stratification and low <br />airflow velocities. An examination of the similarity criteria <br />(Appendix) indicated that the relevant nondimensional variables <br />would be, <br /> <br />C IT X2 2 U(z) ts <br />Q = f(R ' Re , Ri , Pr , --u- ' ~) <br />g <br /> <br />where the Richardson and Prandtl numbers are required for <br />dynamic or thermal similarity. These variables, along with the <br />upstream boundary conditions, such as velocity and temperature <br />profiles, constitute the principal elements of the barostromatic <br />airflow model. <br />The present airflow models only attempted to simulate the gross <br />features, therefore, many of the variables representing the fine <br /> <br />2-7 <br /> <br />'" <br /> <br />,. <br /> <br />22 <br />