At mesoscale, trade wind clouds organize with various spatial arrangements, shaping their effect on Earth's energy budget. Representing their fine‐scale dynamics even at 1 km scale climate simulations remains challenging. However, geostationary satellites (GS) offer high‐resolution cloud observation for gaining insights into trade wind cumuli from long‐term records. To capture the observed organizational variability, this work proposes an integrated framework using a continuous followed by discrete self‐supervised deep learning approach, which exploits cloud optical depth from GS measurements. We aim to simplify the entire mesoscale cloud spectrum by reducing the image complexity in the feature space and meaningfully partitioning it into seven classes whose connection to environmental conditions is illustrated with reanalysis data. Our framework facilitates comparing human‐labeled mesoscale classes with machine‐identified ones, addressing uncertainties in both methods. It advances previous methods by exploring transitions between regimes, a challenge for physical simulations, and illustrates a case study of sugar‐to‐flower transitions.

In this study, we introduce a self-supervised deep neural network approach to classify satellite images into independent classes of cloud systems. The driving question of the work is to understand whether our algorithm can capture cloud variability and identify distinct cloud regimes. Ultimately, we want to achieve generalization such that the algorithm can be applied to unseen data and thus help automatically extract relevant information important to atmospheric science and renewable energy applications from the ever-increasing satellite data stream. We use cloud optical depth (COD) retrieved from postprocessed high-resolution Meteosat Second Generation (MSG) satellite data as input for the network. The network’s architecture is based on the DeepCluster, version 2, and consists of a convolutional neural network and a multilayer perceptron, followed by a k-means algorithm. We explore the network’s training capabilities by analyzing the centroids and feature vectors found from progressive minimization of the cross-entropy loss function. By making use of additional MSG retrieval products based on multichannel information, we derive the optimum number of classes to determine independent cloud regimes. We test the network capabilities on COD data from 2013 and find that the trained neural network gives insights into the cloud systems’ persistence and transition probability. The generalization on the 2015 data shows good skills of our algorithm with unseen data, but results depend on the spatial scale of cloud systems. Significance Statement: This study uses a self-supervised deep neural network to identify distinct cloud systems from cloud optical depth satellite images over central Europe. Satellite-retrieved products support the physical interpretation of the identified cloud classes and help optimize the number of identified classes. The trained neural network gives insights into cloud systems’ persistence and transition probability. The generalization capacity of the deep neural network with unseen data is promising but depends on the spatial scale of cloud systems.

Ubiquitous low-level cloud fields (LCFs) play a crucial role in solar energy applications by significantly influencing surface solar irradiance (SSI) and its spatio-temporal availability. The short-term variabilities caused by these clouds impact solar power plant dispatch and thus affect the stability of electricity grids. This variability stems from their fragmented nature and the rapid evolution of these small spatial-scale clouds. The present study employs cloud optical depth (COD) retrievals over central Europe from high-resolution meteosat second generation satellite imagery and a self-supervised neural framework to isolate the LCFs. The framework identifies four LCF classes corresponding to different stages in the transition from weak, shallow convection with low COD and cloud fraction to more developed, optically thicker cloud fields. Ground-based cloud radar reflectivity, continuously observed at a reference site, highlights the varying microphysical and dynamical characteristics across the LCFs. Integrating pyranometer ground data with LCFs, we analyze the clear sky index over time to quantify SSI fluctuations from s to 30 min. Our results show that LCFs drive the most severe SSI variability, with two LCFs exhibiting the highest fluctuations, challenging solar forecasting, while one remains comparatively more stable, ensuring better grid stability. Additionally, we assess the diurnal evolution of rapidly developing LCFs in the low-dimensional feature space of the framework. Validation with ground-based cloud radar demonstrates how inherent cloud properties cause changes in neural embeddings. Preliminary results suggest that our framework, applied to satellite data, can infer rapid changes in the mesoscale horizontal distribution of low-level cloud systems.

With ever-increasing resolution, geostationary satellites are able to reveal the complex structure and organization of clouds. How cloud systems organize is important for the local climate and strongly connects to the Earth's response to warming through cloud system feedback.Motivated by recent developments in computer vision for pattern analysis of uncurated images, our work aims to understand the organization of cloud systems based on high-resolution cloud optical depth images. We are exploiting the self-learning capability of a deep neural network to classify satellite images into different subgroups based on the distribution pattern of the cloud systems.Unlike most studies, our neural network is trained over the central European domain, which is characterized by strong land surface type and topography variations. The satellite data is post-processed and retrieved at a higher spatio-temporal resolution (2 km, 5 min), enhanced by 66% compared to the current standard, equivalent to the future Meteosat third-generation satellite, which will be launched soon.We show how recent advances in deep learning networks are used to understand clouds' physical properties in temporal and spatial scales. In a purely data-driven approach, we avoid the noise and bias obtained from human labeling, and with proper scalable techniques, it takes 0.86 ms and 2.13 ms to label an image at two different spatial configurations. We demonstrate explainable artificial intelligence (XAI), which helps gain trust for the neural network's performance.To generalize the results, a thorough quantified evaluation is done on two spatial domains and two-pixel configurations (128x128, 64x64). We examine the uncertainty associated with distinct machine-detected cloud-pattern categories. For this, the learned features of the satellite images are extracted from the trained neural network and fed to an independent hierarchical - agglomerative algorithm. Therefore the work also explores the uncertainties associated with the automatic machine-detected patterns and how they vary with different cloud classification types.

At mesoscale, trade wind clouds organize with various spatial arrangements, shaping their effect on Earth's energy budget. Representing their fine-scale dynamics even at 1 km scale climate simulations remains challenging. However, geostationary satellites (GS) offer high-resolution cloud observation for gaining insights into trade wind cumuli from long-term records. To capture the observed organizational variability, this work proposes an integrated framework using a continuous followed by discrete self-supervised deep learning approach, which exploits cloud optical depth from GS measurements. We aim to simplify the entire mesoscale cloud spectrum by reducing the image complexity in the feature space and meaningfully partitioning it into seven classes whose connection to environmental conditions is illustrated with reanalysis data. Our framework facilitates comparing human-labeled mesoscale classes with machine-identified ones, addressing uncertainties in both methods. We highlight the potential to explore transitions between regimes, a challenge for physical simulations, and illustrate a case study of sugar-to-flower transitions.

Shallow cumulus clouds always played an essential role in the uncertainties in climate predictions. The EUREC4A campaign was conceived to tackle the problem of how such clouds will respond to climate change. Recent studies showed that although the research outcomes of the EUREC4A campaign constrained their response to climate sensitivity, open questions remain on the importance of mesoscale processes and the role of precipitation in the cloud organization, both aspects not well represented in climate models. The research vessel (RV) Maria S. Merian, during the campaign, continuously provided high-resolution observations of clouds, precipitation, and atmospheric boundary layer properties in a vast area of the Atlantic Ocean east and south of Barbados island. Here, we exploit such observations to statistically characterize clouds and precipitation properties and the surrounding environment in which they develop. In agreement with the literature, we define shallow clouds with cloud tops within 600 m of lifting condensation level (LCL) and congestus clouds with cloud tops between 600 and 4000m. We characterize their cloud properties, rain rates, and raindrop size distributions. We investigate virga generated from shallow and congestus clouds and describe how humidity and temperature change with the different cloudy conditions. We also display the relation between the W-band radar reflectivity and the radar skewness, revealing insights into the precipitation onset for shallow and congestus clouds and characterizing their cloud lifetime. Finally, we connect the local boundary layer and cloud properties to configurations occurring at the mesoscale, providing additional characterizations of flower, fish, sugar, and gravel in terms of ship-based observations.