The Sentinel‐1 mission is part of the European Copernicus program aiming at providing observations for Land, Marine and Atmosphere Monitoring, Emergency Management, Security and Climate Change. It is a constellation of two (Sentinel‐1 A and B) Synthetic Aperture Radar (SAR) satellites. The SAR wave mode (WV) routinely collects high‐resolution SAR images of the ocean surface during day and night and through clouds. In this study, a subset of more than 37,000 SAR images is labelled corresponding to ten geophysical phenomena, including both oceanic and meteorologic features. These images cover the entire open ocean and are manually selected from Sentinel‐1A WV acquisitions in 2016. For each image, only one prevalent geophysical phenomenon with its prescribed signature and texture is selected for labelling. The SAR images are processed into a quick‐look image provided in the formats of PNG and GeoTIFF as well as the associated labels. They are convenient for both visual inspection and machine learning‐based methods exploitation. The proposed dataset is the first one involving different oceanic or atmospheric phenomena over the open ocean. It seeks to foster the development of strategies or approaches for massive ocean SAR image analysis. A key objective was to allow exploiting the full potential of Sentinel‐1 WV SAR acquisitions, which are about 60,000 images per satellite per month and freely available. Such a dataset may be of value to a wide range of users and communities in deep learning, remote sensing, oceanography and meteorology.

A three‐state global estimator of marine surface layer atmospheric stratification is demonstrated using more than 600,000 Sentinel‐1 synthetic aperture radar wave mode images at incidence angle ≈36.8°. Stratification is quantified using a bulk Richardson number, Ri, derived from collocated ERA5 surface analyses. The three stratification states are defined as unstable: Ri < −0.012, near‐neutral: −0.012 < Ri < +0.001, and stable: Ri > +0.001. These boundaries are identified by the characteristic boundary layer coherent structures that form in these regimes and modulate the surface roughness imaged by the radar. An automated machine learning algorithm identifies the coherent structures impressed on the images. Data from 2016 to 2019 are used to examine spatial and temporal variation in these state estimates in terms of expected wind and thermal forcing. This new satellite‐based approach for detecting air‐sea stratification has implications for weather modeling and air‐sea flux products.

ESA's two Sentinel-1 satellites collect ~120,000 synthetic aperture radar wave mode vignettes every month over the world's oceans. This provides a new and unique opportunity for routine identification and study of a wide range of oceanic and atmospheric phenomena observed in SAR imagery. To this end, the first challenge is to develop an efficient and accurate method to detect and classify key geophysical phenomena signature among the whole dataset. In this study, the deep-learning-based convolutional neural network architecture of Inception v3 model was adopted. We identified 10 geophysical categories detectable by SAR and selected 320 Sentinel-l wave mode imagettes for each category. The full pre-trained Inception v3 model was then retrained using these images. Preliminary results demonstrate that this deep-learning methodology is quite effective, with overall accuracy each of the 10 classes exceeding 0.93 and clear class differentiation in cluster analysis. This opens perspectives to rely on wave mode vignettes in order to analyze geophysical phenomena at global scale. Further work remains to address ambiguous or unknown images that often include a mix of several air-sea processes.