Item

Abstract

As part of the European Project on Cloud Systems in Climate Models (EUROCS), the stratocumulus-topped boundary layer has been simulated using the Max Planck Institute Large-Eddy Simulation (LES) model and the European Centre Hamburg Version Single Column Model (ECHAM-SCM). We have addressed the full diurnal cycle of stratocumulus off the coast of California based on observations of the First International Satellite Cloud Climatology Project Regional Experiment (FIRE). The results of the LES model demonstrate that the model is capable of reproducing the observed diurnal cycle of the boundary-layer structure reasonably well. In particular, the LES model reproduces the distinct diurnal variation in liquid-water path and of turbulence profiles due to the forcing imposed by the short-wave heating of the cloud layer. In addition, we have examined the sensitivity of our LES results with respect to the assumed values of various external environmental conditions. We found that the largest contribution to the variance of the LES-derived data products is due to the uncertainties in the cloud-top jumps of liquid-water potential temperature and total-water mixing ratio and to the net radiative forcing.

To evaluate the quality of the representation of stratocumulus in a general circulation model, results from the standard ECHAM-SCM are contrasted with diagnostics from LES simulations. Results of the standard ECHAM-SCM reveal the following deficiencies: values of the liquid-water path are too low, and unrealistically large levels of turbulent kinetic energy within the cloud layer are due to a numerical instability arising from a decoupling of radiative and diffusive processes. Based on these findings, the SCM has been revised. The modifications include the vertical advection scheme, the numerical treatment of diffusion and radiation, and the combination of the 1.5-order turbulent closure model with an explicit entrainment closure at the boundary-layer top in combination with a front tracking/capturing method. It is demonstrated that, with these modifications, the revised SCM produces a fair simulation of the diurnal cycle of the stratocumulus-topped boundary layer which is significantly improved compared to the one performed with the standard SCM