A modeling study of the subtropical stratocumulus-to-trade-cumulus transition

Abstract

The transition from the subtropical stratocumulus-capped marine boundary layer to the trade-cumulus boundary layer is studied using numerical simulations. Model results are used to propose a new conceptual model of this transition based on the decoupling of turbulent convective eddies near the sea surface from those within the stratocumulus layer.First, a mixed-layer model is used to analyze the relation between sea-surface temperature (SST) rise, increasing latent heat fluxes, deepening of the boundary layer, and decoupling of the stratocumulus-topped boundary layer. Then a two-dimensional eddy-resolving model (ERM) is used to simulate an idealized 10-day Lagrangian trajectory representative of summertime climatological conditions in the subtropical northeastern Pacific. The sea-surface temperature is increased steadily at 1.5 K day$\sp{-1}$ while the free tropospheric temperature remains unchanged.The results of these experiments indicate the physical mechanisms responsible for the cloudiness transition. Decoupling, similar to the decoupling predicted in the mixed-layer-model, results in the creation of a marine boundary layer with cumulus rising into stratocumulus but the cloud cover still remains extensive. In the second stage, further SST increase causes the cumuli to become deeper and more vigorous, penetrating further into the inversion and entraining more and more dry air from above the inversion. This evaporates liquid water in cumulus updrafts before they detrain, causing the eventual dissipation of the overlying stratocumulus. Diurnal variations of insolation lead to a large daytime reduction in stratocumulus cloud amount, but have little impact on the systematic evolution of boundary layer structure and cloud type. The simulated cloudiness changes are not consistent with existing criteria for cloud-top entrainment instability.Simulations testing the sensitivity of the cloudiness transition to parameterized microphysics and cloud-droplet concentration indicate that the largest sensitivities occur in the later decoupled stages of the transition due to the influence of drizzle on detrained cloud amount.ERM simulations of two Lagrangian field experiments are also compared to observational data to assess the ability of the model to predict boundary layer structure, surface fluxes, and cloud morphology.