- 1Max Planck Institute for Meteorology, Hamburg, Germany (lennea.hayo@mpimet.mpg.de)
- 2International Max Planck Research School on Earth System Modeling (IMPRS-ESM), Hamburg, Germany
- 3Monash University, School of Earth, Atmosphere & Environment, Monash University, Australia
The organization of clouds in the Intertropical Convergence Zone (ITCZ) varies markedly from day to day. To investigate how different mesoscale cloud patterns relate to the mean properties of the ITCZ, e.g. precipitation, we have identified and classified recurrent patterns. Focusing on the Atlantic ITCZ, we define five mesoscale cloud patterns: Line, Double Line, Broad, Cluster, and Speckles. We investigate the patterns using two different methods, human labeling and automated identification based on profile-fitting. The human labelers classified a total of ~6,600 images for the seasons July, August and September and December, January and February. The profile-fitting is based on typical signals seen in the human classified labels but automates the detection of the patterns. Both methods show that the preferred location of the most cloudy pattern (Broad) and the least cloudy pattern (Speckles) is seasonally dependent. Additionally, they hint at a connection between the pattern distribution and the regions of highest and lowest precipitation. While the human labeling results are restricted to the afternoon peak in convection, the automated detection method is applied to additional seasons, regions and even times of day.
How to cite: Hayo, L. and Windmiller, J.: Beyond the Mean: the Mesoscale Cloud Patterns of the Atlantic ITCZ, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11016, https://doi.org/10.5194/egusphere-egu26-11016, 2026.