A Comparative Study on Parameterization of Vertical Turbulent Exchange Processes

Abstract

A one-dimensional version of the Techniques Development Laboratory boundary layer model has been adopted to simulate the O'Neill fifth period and the Wangara day 32 experiment data. Fourteen parameterization schemes for the vertical turbulent exchange processes in the atmospheric boundary layer are examined. These include 1) conventional K theories based on Blackadar's mixing length with wind shear and stability functions, 2) O'Brien K formula with time-varying mixed layer heights governed by the surface heat and momentum fluxes based on the work of Deardorff, and 3) a turbulent energy closure model. Numerical results were compared to observations. In general, it was found that all the models perform well in the simulation of the Wangara day 32 data but less satisfactorily in the O'Neill fifth-period simulation. The most satisfactory simulation in wind speeds was reproduced by the turbulent energy model. For temperatures the O'Brien K formula topped by the mixed layer heights performs the best, particularly during the convective hours.