Modelling the influence of rimer surface temperature on the glaciation of intense thunderstorms: The rime–splinter mechanism of ice multiplication

Abstract

AbstractA modelling study is presented to look at the potential importance of ice multiplication in a deep Cb cloud observed during the second Egrett Microphysics Experiment with Radiation, Lidar and Dynamics (EMERALD‐II). Simulations with both a cloud‐resolving model (CRM) and an explicit‐microphysics model (EMM) are used. The CRM has a two‐moment bulk microphysics scheme and resolved cloud dynamics, while the EMM has detailed size‐resolved microphysics, but is driven by the CRM cloud dynamics. The simulations consider the sensitivity of the storms to the rime‐splinter, secondary ice production mechanism (or Hallett–Mossop process).Recent literature on modelling this process in Cb clouds suggests that it is of low importance to the cloud glaciation in intense storms. However, until now, numerical simulations in the literature have investigated this process by either switching it on or off in the model microphysics scheme. We show that if the rate of splinter production is doubled from that proposed by Hallett and Mossop's experiment—a reasonable factor—then for this simulated model cloud, there is a large direct impact on anvil glaciation. The approximate net radiative impact of this calculated with the CRM is 10 W m−2. Similar, ‘nonlinear’ results can also be expected from other ice nucleating mechanisms. Simulations are performed for both continental and maritime cloud condensation nuclei concentrations. In stark contrast to previous literature, we find that the Hallett–Mossop process is more important in the continental (‘polluted’) case than in the maritime (‘clean’) case. This is because of the extremely warm cloud base for this particular storm and consequently—because of the rapid production of rain—the scarcity of small cloud droplets for riming in the clean case. Copyright © 2006 Royal Meteorological Society