loading page

Vertical structure of turbulence and fluxes across cloud mesoscale organizations from the WP-3D Orion aircraft during ATOMIC
  • +4
  • Dean Henze,
  • David Noone,
  • Simon De Szoeke,
  • Gijs De Boer,
  • Richard Fiorella,
  • Adriana Bailey,
  • Peter Blossey
Dean Henze
Oregon State University

Corresponding Author:[email protected]

Author Profile
David Noone
Waipapa Taumata Rau
Author Profile
Simon De Szoeke
Author Profile
Gijs De Boer
National Oceanic and Atmospheric Administration
Author Profile
Richard Fiorella
Los Alamos National Laboratory
Author Profile
Adriana Bailey
National Center for Atmospheric Research
Author Profile
Peter Blossey
University of Washington
Author Profile


In-situ measurements of the trade cumulus boundary layer turbulence structure are compared across large-scale circulation conditions and cloud horizontal organizations during the EUREC4A-ATOMIC campaign. The vertical structure of turbulent (e.g. vertical velocity variance, total kinetic energy) and flux (e.g. sensible, latent, and buoyancy) quantities are derived and investigated using the WP-3D aircraft stacked level legs (cloud modules).The 16 cloud modules aboard the P-3 were split into three groups according to cloud top height and column-integrated TKE and vertical velocity variance. These groups map onto qualitative cloud features related to object size and clustering over a scale of 100 km. This grouping also correlates to the large scale forcings of surface windspeed and low-level divergence on the scale of a few hundred km. The ratio cloud top to trade inversion base height is consistent across the groups at around 1.18. The altitude of maximum turbulence is 0.75-0.85 of cloud top height. The consistency of these ratios across the groups may point to the underlying coupling between convection, dissipation, and boundary layer thermodynamic structure. The following picture of turbulence and cloud organization is proposed: (1) light surface winds and turbulence which decreases from the sub-cloud mixed layer (ML) with height generates clouds with generally uniform spacing and smaller features, then (2) as the surface winds increase, convective aggregation occurs, and finally (3), if surface convergence occurs, convection and turbulence reach higher altitudes, producing higher clouds which may precipitate and create colds pools. Observations are compared to a CAM simulation is run over the study period, nudged by ERA5 winds and surface pressure. CAM produces higher column integrated turbulent kinetic energy and larger maximum values on the days where higher cloud tops are observed from the aircraft, which is likely a factor that influences the development of deeper clouds in the model. However, CAM places the peak turbulence 500 m lower than observed, suggesting there may be a bias in CAM representation of turbulence and moisture transport. CAM also does not capture the large LHFs seen for two of the days in which lower cloud tops are observed, which could result in insufficient lower free tropospheric moistening in the model during this type of cloud organization. A large and consistent bias between the model and observations for all cloud groups is the negative SHFs produced in CAM near 1500 m. This is not observed in the measurements. This leads to a net negative buoyancy flux not observed and provides evidence of a specific shortcoming that can be addressed as part of the needed improvement in the representation of clouds by large-scale models.