Two-third of the Earth surface and more than 90% of known active tectonic-plate boundaries where plate motion occur are underwater. Yet, due to the difficulty to monitor and measure ground-displacements underwater, little is known about active processes along undersea plate boundaries and associated natural hazards (earthquakes, tsunamis), or on any ground-motions related to human activities (oil pumping). These displacements may occur continuously at a millimeter or centimeter scale over several years (e.g., creeping fault), or instantaneously at a meter or several meter scale  (e.g., rupturing fault).

On-land, the past 20 years have seen great strides in our understanding of faults, landslides and volcanoes, thanks to technological advances allowing high-resolution ground-deformation measurements. Optical correlation (e.g., Michel & Avouac, 2002), Differential Interferometric Synthetic Aperture Radar (DInSAR, e.g., Delacourt et al., 2009), Differential Global Positioning System (DGPS, e.g., Segall & Davis, 1997; Dzurisin, 2003) provide nearly direct information about the strain energy accumulating along locked portions of active faults that can be released by a major earthquakes, plate motions, inflation and deflation at volcanoes.

The ability to detect and monitor tectonically active structures on the seafloor at such high-resolution is a real challenge. Since electromagnetic waves do not propagate in the water, transposing on-land techniques underwater is not possible. The only approach is to deploy sensors on the seafloor that can communicate through acoustics with a surface-ship or through a cable to shore. Furthermore such sensors must stand extreme conditions (e.g. pressure, corrosion) and require long-lasting power supplies. Maintaining such equipment and retrieving the data require repeated cruises and is, thus, a costly and time-consuming approach, particularly in open seas. As a consequence, monitoring ground-deformation of the seafloor is often conducted on shore with land-based sensors, offering little resolution on submarine ground-motions away from the shore, or uses underwater-cabled sensors, which requires costly infrastructures and appropriate sites (e.g. Neptune cabled observatory off Canada).

Nevertheless, overcoming these limitations in the marine environment would open new fields of discoveries and possibilities for a variety of applications including global geodesy, geodynamical studies, natural and industrial risk assessments. Along active seismogenic faults, where displacement rates are relatively low (a few millimeters to a few centimeters per year), several years of geodetic data acquisition are needed to characterize these displacements, which demonstrates the urgency to initiate these measurements in areas of high level of seismic risk. As an example, the seismic hazard at subduction or strike-slip zones has often been calculated based on historical seismicity, plate movement rate, and the estimated size of the seismogenic zone. None of these parameters provides quantitative information about the plate coupling and stress accumulation that powers mega-earthquakes and none of them forecast the 2004 Sumatra or 2011 Japan mega-events.

The objectives of the GEODESEA project were to test and improve innovative methods for an absolute marine geodetic positioning.