Data assimilation for prediction, understanding and parameterization of "small scale" atmospheric motions

Abstract:

Parameterization of cloud and boundary-layer processes in atmospheric models remains a thorny problem. Because of improvements in observations and assimilation techniques, applying data assimilation as tool in this problem is increasingly feasible. Data assmilation allows direct comparison between numerical solutions and atmospheric realizations for small-scale motions and provides a framework to quantify uncertianties in other crucial elements, such as the mesoscale environment and the land (or sea) surface. I will review recent work in data assimilation for convective clouds and discuss both its promise and some potential limitations for illuminating the parameterization problem.