SZO Seminar Notes – April 15, 2015: Possibilities for offshore geodesy

Discussion leaders: Emily Roland, David Schmidt

Papers: Ito, Y., T. Tsuji, Y. Osada, M. Kido, D. Inazu, Y. Hayashi, H. Tsushima, R. Hino, and H. Fujimoto (2011), Frontal wedge deformation near the source region of the 2011 Tohoku‐Oki earthquake, Geophys. Res. Lett., 38, doi:10.1029/2011GL048355.
Gagnon, K., C.D. Chadwell and E. Norabuena (2005), Measuring the onset of locking in the Peru-Chile trench with GPS and acoustic measurements, Nature, 434, 205-208.

The discussion was framed around basic questions related to earthquakes and plate boundary processes that offshore geodesy would be helpful to answer. 

• What is the spatial distribution of plate locking, particularly in the shallower portions of the interface?  This needs to be addressed to be able to 1) anticipate future earthquakes, 2) identify segment boundaries, 3) determine the shallowest extent of locking (e.g. especially for tsunami studies), 4) link seismicity with slip behavior.  Emily showed an example  from Alaska where seafloor geodetic data would be very useful. Offshore of Unalaska existing models show an along-strike transition from a creeping to locked zone, consistent with a gap between the boundaries of the 1946 M8.6 and 1957 M8.6 rupture planes.  Paul Bodin suggested that it would also be fruitful to look at the seismicity along this margin, because high rates of small earthquakes often is diagnostic of creep. 
• Are there temporal changes in slip behavior?  What are the temporal and spatial scales of slip transients, and do they occur up-dip of or within the locked zone? David Schmidt noted that for Cascadia, it may be possible to look for temporal changes using GPS data from the last 10 years.  He also commented that current locking models combine tide gage data spanning 80 years with GPS data spanning the last 10, thus the time scales sampled by the two data sets aren't consistent.
• How is strain partitioned between the interface and within the accretionary prism?  What is the failure mode within the prism; e.g., are there splay faults, elastic versus plastic deformation, how porous and what are pore pressure conditions within the prism?
  

Much of the discussion centered around Cascadia, where most of of the locked zone is offshore.  David Schmidt noted that even basic characteristics like the convergence rates, which have to vary along strike, are not known very precisely. Brian Atwater commented that the locking model is much more uniform along the coast than one might infer from 1700 co-seismic and post-seismic  land-level changes along the coast.  Paul Johnson mentioned that a complete trench-perpendicular profile of heat flow should be available soon (latitude of Gray's Harbor), as his group measured one last year and are finalizing the data analysis.

The example of possible premonitory seismic and geodetic deformation prior to the Tohoku earthquake was discussed, and if and how such deformation could be distinguished from cases that did not culminate in a large earthquake.  It was noted that no geodetic signals were detected onshore prior to Tohoku. 

The two papers discussed illustrated several approaches to seafloor geodesy.  The paper by Ito et al. discussed analysis of pressure sensor data, which are only useful for very large vertical displacements.   It also described their use of seafloor benchmark surveys, involving acoustic ranging, which also has very large uncertainties.  These methods were useful in Tohoku because of the enormous displacements, and because measurements exists from before the earthquake. The Gagnon et al. paper used acoustic arrays, which the authors showed can be very accurate for horizontal displacements, but are difficult measurements to make. 

We also briefly brainstormed other possible approaches to making geodetic measurements on the seafloor, particularly using bathymetric measurements of differential uplift.  This currently is being done over axial seamounts.  Standard ship surveys have resolution of meters, so deformations need to be very large.  Surveying done with AUVs can achieve cm resolution, but only over small areas and without being tied to any absolute reference. The need to correct for the water column and know ship lines precisely makes it very difficult to do use legacy surveys.  Other methods noted were seafloor ‘InSar’ of sorts, which involves some sort of multibeam imaging from a ship and correction for internal waves in the water column and thus, is not trivial.  Burial of fiber optic cables used as strainmeters also is being considered, but also is not at all trivial.