Explicit forecasting of supercooled liquid water in winter storms using the MM5 mesoscale model

Abstract An explicit microphysical parametrization including ice physics was developed for use in the NCAR/Penn State Mesoscale Model Version 5 (MM5). This scheme includes three options of increasing complexity to represent the hydrometeor species. The scheme is evaluated by comparing model simulations with two well observed winter storms that occurred during the Winter Icing and Storms Project. The evaluation focused on the prediction of supercooled liquid water (SLW), which is of particular importance to aircraft icing. The intercomparisons showed that: The double‐moment microphysical scheme, in which both ice mixing ratios and number concentrations were predicted, performed best, with close agreement to the observed fields. The single‐moment schemes, in which the mixing ratio of ice species are predicted and number concentration specified, performed reasonably well if a diagnostic equation for N o, s , the Y ‐intercept of the assumed exponential snow distribution, is allowed to vary with snow mixing ratio. Accurate microphysical simulations of SLW in shallow upslope clouds and cyclonic storms required accurate simulations of the kinematic and thermodynamic structure and evolution of the storms. Though the two storms were dynamically different, the SLW formed through a balance of the condensational growth of cloud water and the depletion of cloud water by deposition and riming of snow and/or graupel for both storms. The results of this study suggest that accurate prediction of SLW over limited areas of the country may be possible using the current microphysical parametrization and high‐resolution grids (δχ <10 km).