A new boundary layer cloud parameterization is presented. This scheme contains a new type of turbulence closure that represents subgrid-scale variability in temperature, moisture, and vertical velocity using a joint probability density function (PDF). The PDF is used to close higher order turbulent moments as well as predict cloud fraction and cloud water, all in a manner consistent with the PDF. This leads to a unified framework capable of representing both very low cloud fraction layers and solid cloud decks.

Results from various boundary layer regimes, including clear convective layer, stratocumulus-topped layer, trade-wind cumuli and fair-weather cumulus clouds over land are shown and compared to large-eddy simulation model output. The scheme is capable of predicting reasonable values of cloud fraction and cloud water for the various cloudy regimes without the need for case-specific adjustments.