Mesoscale Convective Systems contribute to a large amount of rainfall in the midlatitudes. Cloud processes in the convective regions can affect the size distributions of cloud hydrometeors, the evolution of convective systems, and total precipitation.
In order to improve the simulations of Mesoscale Convective Systems, we compare in-situ observations from the National Science Foundation (NSF) Deep Convective Clouds and Chemistry (DC3) campaign in 2012 with a cloud-resolving model (CRM) simulation. The CRM we use is the NCAR CM1 model. A unique comparison method we use is to insert artificial aircraft tracers into the CM1, which helps to mimick the flight track of the real research aircraft at various pressure levels.
Our results show that the double-moment microphysics scheme [Bryan and Morrison, 2010] can successfully capture the overall evolutionary trends of ice microphysical properties and relative humidity for different phases of ice particle formation and evolution [Diao et al., 2017]. One recommendation we have is to increase the current relative humidity threshold of activating the Cooper parameterization [Cooper, 1986] from the default 108% to a higher value of 130%.
A similar approach is applied to evaluate the performance of the Weather Research and Forecasting (WRF) model simulations of a real convective event during the DC3 campaign. In the work of D’Alesssandro et al. [2017], we found that the WRF simulations have stronger dependence on higher updrafts for generating and maintaining ice supersaturation than the observations do. Ice supersaturation (RHice > 100%) is the condition that facilitates ice particle formation and growth. These results indicate that the spatial variability of relative humidity has been underestimated by the WRF simulations.
