Surface flux parameterization in a numerical model, especially in a coupled ocean-atmospheric model, is related to the entire atmospheric boundary layer including both its surface layer flux and its boundary layer turbulence parameterization. Although various modelling schemes have been used in a variety of models, there is a lack of systematic evaluation mainly due to inadequate observations. During the Japan/East Sea Experiment (JES), intensive aircraft measurements were made using the Twin Otter research aircraft operated by the Center for Interdisciplinary Remote-Piloted Aircraft Studies (CIRPAS) of the Naval Postgraduate School (NPS). The Twin Otter measurements included high-rate sampling of the three-dimensional wind speed, turbulence, temperature and moisture, for nine days (from late January and through February 2000) covering a variety of meteorological conditions. In particular, multiple slant-path soundings were made along a vertical cross-section that can be used to characterize the vertical and horizontal variations of the marine boundary layer. This study will utilize the JES aircraft data for meso-scale model evaluation.

Numerical simulations using the Navy’s mesoscale model, the Coupled Ocean and Atmosphere Prediction System (COAMPS), are made for the entire JES measurement period. The objective of the study is to understand the model behaviour via inter-comparisons with the observations from the JES flights and understand the model sensitivity to various parameters in the surface flux and boundary layer turbulence parameterization. Previous preliminary work has shown that COAMPS predicts the mean wind speed, turbulent kinetic energy and the boundary layer height quite well while the surface fluxes are ill-predicted. The present study involves extended testing that includes inter-comparisons for multiple JES flights between the modeled and measured turbulence statistics as well as budget analysis for the major boundary layer mean and turbulence properties. Inter-comparisons involving a variety of cases enabled us to obtain statistically significant conclusions on the accuracy of COAMPS parameterizations of the marine boundary layer and seek for improvements.