Towards a New Paradigm of Parameterizing Ice-Phase Microphysics in Atmospheric Models
Atmospheric and Oceanic Sciences Departmental Seminar Series
presents
Towards a New Paradigm of Parameterizing Ice-Phase Microphysics in Atmospheric Models
a talk by
Jason Milbrandt
Research Scientist,
Atmospheric Numerical Weather Prediction Research Section, Meteorological Research Division, Environment and Climate Change Canada
The representation of cloud and precipitation microphysics in atmospheric models – both for research and numerical weather prediction – has advanced considerably in the past four decades. In 3D models, the most common approach is to use a bulk microphysics scheme (BMS), in which the size distributions of each of the various hydrometeor categories is represented by an analytic function and through the prediction of one or more moments of the distribution, where the moments are related to bulk physical quantities (e.g. the mass mixing ratio). The early BMSs were single-moment schemes with a limited number of hydrometeor categories and microphysical processes. Over the years, these schemes have become more complex and with many more categories, prognostic moments, and parameterized processes. This includes the detailed, triple-moment Milbrandt-Yau BMS, developed at McGill nearly 20 years ago.
In the past decade, there has been a shift in thinking amongst model developers on how best to represent ice-phase hydrometeors. The shift has been a move away from the traditional notion of including more and more representative categories (e.g. graupel) and towards increased emphasis on the prediction of physical properties (e.g. density). As part of this shift, we have developed a new BMS that completely abandons the use of pre-defined categories and uses one or more “free” categories to represent all ice-phase hydrometeors. Each category has between four and six prognostic variables from which a variety of physical properties of ice can be computed. Unlike traditional schemes in which such properties are prescribed and often fixed, the ice properties in the new scheme evolve continuously and more realistically due to microphysical processes. As such, we refer to this new parameterization as the Predicted Particle Properties (P3) microphysics scheme.
A general background on the concepts of representing cloud microphysics in models will be given along with some historical details of the evolution of the development towards detailed traditional microphysics schemes. An overview of the P3 microphysics scheme will be provided, along with illustrations of the benefits of the “free” category approach.