Cloud types are determined using machine learning image classifiers applied to satellite imagery for 1 year in the North Atlantic. This survey of these cloud types shows that the climate impact of a cloud scene is, in part, a function of cloud type. Each type displays a different mix of thick and thin cloud cover, with the fraction of thin cloud cover having the strongest impact on the clouds' radiative effect. Future studies must account for differing properties and processes among cloud types.

Trade wind clouds are often organized on the mesoscale (O(100 km)), forming different cloud patterns. We present C³ONTEXT (a Common Consensus on Convective OrgaNizaTion during the EUREC⁴A eXperimenT), a dataset that contains information about the mesoscale cloud patterns identified during the EUREC⁴A (Elucidating the role of clouds–circulation coupling in climate) field campaign in January–February 2020 and thereby provide the mesoscale context for the campaign's measurements.

The German research aircraft HALO took part in the research campaign EUREC⁴A in January and February 2020. The focus area was the tropical Atlantic east of the island of Barbados. We describe the characteristics of the 15 research flights, provide auxiliary information, derive combined cloud mask products from all instruments that observe clouds on board the aircraft, and provide code examples that help new users of the data to get started.

The shallow cumulus clouds that populate the trade-wind regions can produce substantial amounts of rain. Before reaching the surface, part of the rain can evaporate and form pools of cold air that spread at the surface as density currents. We use 10 years of data from Barbados to show that such cold pools occur on 3 out of 4 d, that cold-pool periods are 90 % cloudier relative to the average winter conditions, and that they are connected to specific patterns of mesoscale cloud organization.

Dropsondes measure atmospheric parameters such as temperature, pressure, humidity and horizontal winds. The EUREC⁴A field campaign deployed 1215 dropsondes during January–February 2020 in the north Atlantic trade-wind region in order to characterize the thermodynamic and the dynamic structure of the atmosphere, primarily at horizontal scales of ~ 200 km. We present JOANNE, the dataset that provides these dropsonde measurements and thereby a rich characterization of the trade-wind atmosphere.