In this presentation it is argued that ensemble prediction systems can be devised, in such a way that physical parameterizations of sub-grid scale motions are utilized in a stochastic manner, rather than in a deterministic way as it is typically done. This can be achieved within the context of current parameterization schemes. Parameterizations are typically used to predict the evolution of grid-mean quantities due to unresolved sub-grid scale processes. However, parameterizations can also provide estimates of higher moments of sub-grid scale distributions that could be used to constrain the random determination of the future state of a variable.

We discuss results of the implementation of a stochastic convection parameterization, based on these ideas, in the Navy global model (NOGAPS) ensemble system. It is shown that this method is able to generate substantial tropical perturbations that grow and “migrate” to the mid-latitudes as forecast time progresses, while moving from the small scales where the perturbations are forced to the larger synoptic scales.

We also utilize satellite observations, from the Atmospheric Infrared Sounder (AIRS), of temperature and humidity variance to constrain the stochastic parameterizations. In particular we discuss the climatology of variance as well as results of variance length scaling within the troposphere. The implications for parameterization development, stochastic parameterizations in particular, of the AIRS observations of higher moments of the distributions of temperature and humidity are explored in detail.