Ocean surface boundary extraction using Hypertemporal Classification Analysis


The hyper-temporal algorithm (developed through the FP7-SAFI project and the H2020 Co-ReSyF project) enables users to extract an idea of where more permanent, or stronger boundary regions exist on the ocean’s surface. These boundary regions could be related to a number of factors. For example, when deployed on Sea Surface Temperature data, the high boundary areas most likely represent areas which experience many, and strong, SST fronts.

The Co-ReSyF researcher is refining the algorithm using deployments on SST and measures of photosynthetic activity from the ocean surface. In essence, it analyses a 3-dimensional (Longitude, Latitude and Time axes) datacube of parameters measurements, and extracts boundary areas where as we transition across them, the temporal profile of the parameter at either side, differs. The developers of this algorithm would welcome deployments of the algorithm on different areas of the Earth’s surface, and looking at different satellite-derived parameters (surface winds, surface waves, different parameters concerning photosynthetic activity, water transparency, or others).

If you plan to use this application, have some doubts, or want to share your problems or your results using it, write a comment on this topic thread, and ask for some feedback from the hyper-temporal researchers and the Co-ReSyF community.