Explicit Forecasts of Winter Precipitation Using an Improved Bulk Microphysics Scheme. Part I: Description and Sensitivity Analysis

TLDR

The study evaluates how winter precipitation responds to various aspects of a bulk, mixed‑phase microphysics scheme in the MM5, RUC, and WRF mesoscale models. The authors conduct sensitivity experiments by varying ice initiation, autoconversion, CCN spectra, graupel handling, and snow/rain size distribution parameters while simulating different cloud depths over an idealized two‑dimensional mountain to generate updrafts matching documented freezing‑drizzle cases. The sensitivity tests show that ice‑initiation choice matters mainly for shallow warm clouds, while snow and graupel size distributions and rimed‑snow transformation strongly influence cloud water mass and freezing drizzle, and that increasing the rain intercept parameter is needed to simulate drizzle, with a deep precipitating snow cloud produced when the moisture profile is saturated at low temperatures.

Abstract

This study evaluates the sensitivity of winter precipitation to numerous aspects of a bulk, mixed-phase microphysical parameterization found in three widely used mesoscale models [the fifth-generation Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model (MM5), the Rapid Update Cycle (RUC), and the Weather Research and Forecast (WRF) model]. Sensitivities of the microphysics to primary ice initiation, autoconversion, cloud condensation nuclei (CCN) spectra, treatment of graupel, and parameters controlling the snow and rain size distributions are tested. The sensitivity tests are performed by simulating various cloud depths (with different cloud-top temperatures) using flow over an idealized two-dimensional mountain. The height and width of the two-dimensional barrier are designed to reproduce an updraft pattern with extent and magnitude consistent with documented freezing-drizzle cases. By increasing the moisture profile to saturation at low temperatures, a deep, precipitating snow cloud is also simulated. Upon testing the primary sensitivities of the microphysics scheme in two dimensions as reported in the present study, the MM5 with the modified scheme will be tested in multiple case studies and the results will be compared to observations in a forthcoming companion paper, Part II. The key results of this study are 1) the choice of ice initiation schemes is relatively unimportant for deep precipitating snow clouds but more important for shallow warm clouds having cloud-top temperature greater than −13°C, 2) the assumed snow size distribution and associated snow diffusional growth along with the assumed graupel size distribution and method of transforming rimed snow into graupel have major impacts on the mass of cloud water and formation of freezing drizzle, and 3) a proper simulation of drizzle using a single-moment scheme and exponential size distribution requires an increase in the rain intercept parameter, thereby reducing rain terminal velocities to values more characteristic of drizzle.