Huang, Chin-hua, authorNickerson, Everett C., authorColorado State University, Fluid Dynamics and Diffusion Laboratory, College of Engineering, publisher2020-03-312020-03-311972-03https://hdl.handle.net/10217/201679CER71-72CH-ECN23.March 1972.Includes bibliographical references (pages 152-160).Prepared under Office of Naval Research contract no. N00014-68-A-0493-0001, project no. NR 062-414/6-6-68(Code 438) U.S. Department of Defense, Washington, D.C.Airflow in the atmospheric surface layer over nonhomogeneous surfaces with discontinuities in surface roughness and temperature is investigated by numerical techniques. A computational scheme is developed for solving the steady state two-dimensional boundary layer equations. Several theorems of convergence are proved. A successful numerical test, which has been compared to the exact solution, is achieved. Some iterative schemes, which have already enjoyed considerable success without theoretical support are here shown to be convergent. The variations in pressure and buoyancy force associated with changes in surface roughness have been neglected by previous investigators whose work is included in the present study. The numerical results of velocity and shear stress are compared with wind tunnel and field data. The roughness and temperature discontinuities are shown to have an effect on the upstream as well as the downstream flow conditions. Significant variations in the horizontal velocity, vertical velocity and shear stress profiles near the roughness discontinuity occurred between those cases neglecting and those retaining the pressure terms in the governing equations. The predicted physical quantities for diabatic conditions also show significant differences in those two cases; thus, the pressure terms should be retained in the governing equations. No inflection point in the wind profile for neutral conditions has been observed in the mixing length model; however, it has been observed in both the turbulent energy model and the model presented in this study. The field and wind tunnel observations also confirm the presence of an inflection point. The inflection point is less visible in the presented model as compared with the turbulent energy model. For a small change in surface roughness, the wind profiles simulated by the numerical method are in good agreement with wind tunnel data. The distribution of the surface shear stress predicted by the presented theory is in better agreement with Bradley's field data than previously existing theories. A proposed mechanism of turbulent energy transfer is developed, based upon the results of numerical experiments that explain the distribution of shear stress, and, hence, the distribution of velocity profiles in the atmospheric surface layer. Two different theories, the mixing length theory and the turbulent energy theory, are modified, and examined in detail; a theory is developed to remove some weaknesses of previously existing theories.technical reportsengCopyright and other restrictions may apply. User is responsible for compliance with all applicable laws. For information about copyright law, please see https://libguides.colostate.edu/copyright.Dynamic meteorologyAtmospheric circulationAir flow -- Mathematical modelsText