High Resolution Modeling of Cloud Fields at the TWP Site with MM5

The advent of powerful computers has enabled atmospheric scientists to run weather forecast and climate models at ever increasing resolutions. Of particular interest are limited area models with high temporal and spatial resolution that can be run affordably on workstations. These models, initialized and forced at the boundaries with data from large-scale models with coarse resolution, resolve fine-scale features in the interior, a process generally known as dynamical downscaling (e.g., Murphy 1999). However, the success of a high-resolution simulation depends on a number of factors such as the coupling at the boundaries or the representation of physical processes. These processes are treated differently at different resolutions; convection, for example, is parameterized at coarse resolution but cumulus clouds are explicitly resolved at finer scales. Furthermore, the values of model parameters may be optimal at certain scales but inappropriate at others. The selection of the right combination of parameterizations is therefore a difficult task in complex numerical models of the atmosphere.