Combined retrievals of stratocumulus cloud-top turbulence, radiative flux, and entrainment from ARM measurements

 
Poster PDF

Authors

David B. Mechem — University of Kansas
Sandra Yuter — North Carolina State University
Simon Paul de Szoeke — Oregon State University

Category

Vertical Velocity

Description

Turbulence at the top of a stratocumulus-capped boundary layer entrains air into the cloud. We compute vertical velocity variance and dissipation of turbulent kinetic energy (TKE) near the cloud top from case studies of stratocumulus clouds observed by the W-band ARM Cloud Radar (WACR) at the DOE ARM site at Graciosa, Azores. Turbulence statistics are compared to net radiative flux retrievals computed for stratocumulus clouds. Net radiative divergence cools the cloud top, generating greater TKE at cloud top at night than during daylight. We estimate the turbulent convective buoyancy flux from the turbulence dissipation, and relate the entrainment rate to the buoyancy flux using the common scaling of entrainment rate to the buoyancy flux divided the buoyancy jump at the inversion. Solar and longwave radiative fluxes are calculated by four different methods of specifying the liquid water content (LWC) from ARM radar and microwave observations of stratocumulus clouds. Flux calculations are evaluated against surface downwelling flux observations. In the first method LWC is computed from cloud radar reflectivity from the WACR. Large drops increase radar reflectivity disproportionately compared to solar extinction cross section, so drizzle and virga cause an overestimate of the solar cloud radiative effect. This problem is ameliorated somewhat by the second method, which uses a profile of LWC that decreases linearly from the radar reflectivity-derived LWC at cloud top. The methods that constrain LWC with microwave liquid water path (LWP) perform better than the radar-only methods. Radiative flux calculations from linear LWC profiles based on 10-minute LWP averages agree with surface observations better than calculations based on 10-minute averages of linear profiles of LWC. Calculations from 10-minute average LWP agree as well with surface observations as 10-minute averages of nearly instantaneous (30 s) flux calculations.