QUANTIFYING HYDROMETEOR ADVECTION AND THE VERTICAL DISTRIBUTION OF CLOUD FRACTION OVER THE SGP CART SITE

A single column model (SCM) is, in essence, an isolated grid column of a general circulation model (GCM). Hence, SCMs have rather demanding input data requirements, but do not suffer from problems associated with balance of a GCM. Among the initial conditions that must be used to describe the initial state of the SCM column are the vertical profile of the horizontal wind components and the vertical profiles of cloud water and ice. In addition, the large-scale divergence and advective tendencies of cloud water and ice must be supplied as external parameters. Finally, the liquid and ice cloud amount as a function of height within the SCM column are required for model evaluation. The scale of the SCM column over which the initial conditions, external parameters, and model evaluation fields must apply is relatively large ({approximately}300 km). To quantify atmospheric structure on this scale, the ARM SGP CART site is located within the NOAA wind profiler network and has boundary and extended measurement facilities in an area compatible with the scale requirements of SCMs. Over an area this size, however, there is often rich mesoscale structure. This mesoscale variability creates a sampling problem that can thwart even the most sophisticated attempts to quantify the initial conditions and external parameters, and to evaluate model performance. There are two approaches that can be used to quantify the time varying quantities required for SCMs: objective analysis and data assimilation. The latter relies on products produced for operational forecasting, while the former involves methods that can be used to combine measurements from various sources to produce synoptic descriptions of the large-scale dynamical and thermodynamic fields. Since data assimilation from operational models introduces the uncertainty of the parameterizations used in the models, most of the focus in the SCM effort has been on developing objective analysis techniques