Locally narrow droplet size distributions in stratocumulus clouds: Insights from ACE-ENA and LES

Description

Gamma-like distributions are widely used for cloud droplet size distributions in coarse-resolution climate models and remote sensing algorithms. This parametric representation is based on cloud measurements from instruments with large path lengths and hence constitutes an averaged quantity. However, cloud droplets interact locally, so sub-grid correlations of number concentration and droplet size distribution shape are needed. We ask whether such averaged gamma-like distributions are found for “local” cloud volumes at small spatial scales. If not, what distribution shapes occur and what are their properties? We devised and tested an algorithm that combines hypothesis testing and machine-learning data clustering to explore these questions. The algorithm identifies statistically similar cloud regions defined by the same distribution shape called the characteristic droplet size distribution. The identified characteristic distributions combine to give back the averaged gamma distribution. When implemented on cm-scale cloud samples taken using Holographic Detector for Clouds (HOLODEC) deployed during Aerosol and Cloud Experiments in the Eastern North Atlantic (ACE-ENA), the algorithm reveals that local characteristic distribution types are ubiquitous in stratocumulus clouds. These distribution types are generally narrow with distinct modes and do not resemble the gamma-like averaged shape. Each characteristic distribution represents identical-looking local cloud volumes which tend to occur in spatial blocks of varying extent, usually of order 1s to 10s of km. These similar-looking cloud blocks may represent regions with similar microphysical history. The characteristic distributions are found to contain cloud samples with the same distribution shape but diluted by different amounts, indicating that these may be resulting from individual inhomogeneous mixing events. Subsequently, we show first results from an investigation of whether these characteristic distributions are present for LES of stratocumulus with Lagrangian and bin microphysics schemes. We also explore consequences for remote sensing retrieval of microphysical properties.