Identifying large-scale drivers of the daily cycle of trade-wind cloudiness from geostationary satellite

Shallow cumulus clouds in the trade-wind region are a major source of uncertainty in the global cloud feedback on climate. Although previous studies have investigated cloud feedback based on daily mean or even monthly data, the time of the day when clouds occur currently and in the future matters for their radiative effect. On top of that, with the availability of high-frequency data comes the opportunity to study time series data of cloud fields rather than relying on snapshots. To quantify the role of the diurnal cycle for the cloud feedback, we study the relationship between the daily cycle in cloudiness and in the large-scale environment. We compile a dataset of Lagrangian satellite observations together with cloud controlling factors (CCFs) along ~30000 ERA5 trajectories, obtained from 925hPa wind fields. The 6-day-long trajectories are centered at the tropical North Atlantic in the winter months (DJF), which is representative of the trade-cumulus regime. We utilize the high temporal resolution of GOES-16 (10-15 min), ERA5, and CERES (both hourly) to fully resolve sub-daily timescales. With this dataset, we explore correlations between the amplitudes and phases of cloudiness and CCFs. We examine which CCFs control the daily cycle of clouds and quantify response times between the drivers and their effects. Our goal is to develop a model that describes the daily cycle in cloudiness based on the most important CCFs and use time series data to constrain the trade cumulus cloud feedback.