AbstractExtreme waves are often enclosed by other waves which are also higher than the average. These wave groups have to be taken into account for instance for the design of offshore platforms, breakwaters or ships, because successive high waves can cause more damage on those structures than the same waves separated by smaller waves. Further they can excite the resonance frequencies of moored structures like platforms due to non-linear effects or cause capsize. They are therefore of interest for engineers and scientists (e.g. Goda 1983). A method is presented to localize wave groups spatial and spatio-temporal utilizing synthetic aperture radar (SAR) images and nautical radar-image sequences. The approach to detect wave groups is based on the detection of the wave envelope. It is assumed that the sea surface elevation can be treated as a Gaussian process. The method is applied to SAR images acquired by the European satellite ERS-1 and to radar-image sequences recorded by tower-based nautical radars. In contrast to 1D sensors like buoys the SAR records an image and gives therefore a 2D description of the sea surface by measuring the radar backscatter from the sea surface. The measurements taken by a nautical radar provide the possibility to record time series of images and therefore to get a 3D description of the sea surface. Radar-image sequences are acquired by recording the spatial and temporal evolution of the sea surface backscatter, which is modulated through the surface wave field. Nautical radar-image sequences allow to detect wave groups within a time span that makes it possible to start safety programs before the group reaches a platform. The existing data sets are exploited with respect to the recognition of extreme wave events.