Summary of a Mini Workshop on Cumulus Parameterization for Mesoscale Models

Representing the effects of cumulus parameterization in numerical models, known as cumulus parameterization, is of fundamental importance in atmospheric sciences. Despite a significant amount of effort devoted to improving the understanding of the parameterization problem, a general theory of cumulus parameterization does not exist. The problem is particularly difficult for a mesoscale model with a grid size of 1-20 km, because at such a resolution, a numerical model introduces an artificial spectral gap by splitting an organized mesoscale convective system into resolvable and subgrid scales of motions. To provide a forum for a focused discussion on the problem of cumulus parameterization for mesoscale models, the Mesoscale and Microscale Meteorology Division of NCAR conducted a mini workshop on 11-12 September 1995 in Boulder, Colorado. Approximately 50 scientists participated in this workshop. The workshop was organized into six sessions: 1) overview of the cumulus parameterization problem, 2) cloud model perspectives, 3) model sensitivity to convective parameterization, 4) mesoscale organization and momentum transport, 5) mesoscale observations and verification of schemes, and 6) summary and general discussion. The scientific presentations and the discussion are summarized from reports submitted by the session chairs.