As has been mentioned earlier, the data processing architecture underpinning the different GFM flood mapping algorithms is based on the data cube concept (Wagner et al., 2020, Wagner et al., 2021) whereby all incoming Sentinel-1 SAR satellite images are geocoded, gridded, and stored as analysis ready data (ARD) in an existing space-time SAR data cube. By using a data cube (where the temporal and spatial dimensions are treated in the same way) each Sentinel-1 image can be compared with the entire backscatter history, allowing the implementation of different types of change detection algorithms in a rather straightforward manner. Importantly, the entire backscatter time series for each pixel can be analysed. Therefore, model training and calibration may be carried out systematically for each pixel.
The main advantages of working with data cubes are: