Abstract

Earth Observation data are around for several decades but its use has been limited to security and environmental research domains until few years ago, when Copernicus Sentinel satellites were launched, providing free and open access to multi-spectral and SAR data of such a quality that they started a revolution in the field. Powered by the general accessibility of the cloud computing and open-source machine learning technologies application, developers and data scientists around the world are now able to work with the data that were previously available only to a small number of remote sensing experts. New technologies and statistical approaches allowed for exploration of the data beyond visual interpretation - analysis of time-series.

Copernicus Sentinel satellites provide a plethora of sensors - from three- to five-daily multi-spectral coverage of complete land surface at 10-meter resolution, daily 300-meter global coverage, 6-daily land surface with 20-meter synthetic aperture radar and more. Combined with Landsat mission (30-meter resolution, 12-daily, since 1984) and a number of commercial satellites the volume and variety of data is sufficient to proactively monitor the world. We will present several cases of large-scale monitoring of water resources, agriculture activities and natural disasters such as floods, fires and drought, which opens a door to pro-active planning of our response to these.