Evaluating the use of seafloor pressure data for the study of slow slip earthquakes; insights from the 2011-2015 Cascadia Initiative deployment
Abstract
The Cascadia subduction zone produces M8-9 megathrust earthquakes with a recurrence interval of 500 years. While land-based geodetic measurements indicate a large degree of locking offshore, these observations cannot resolve the extent of locking nearest the trench. One method for detecting displacement at shallow depths on the megathrust is through the use of seafloor pressure to track uplift and subsidence of the seafloor, a technique that shows potential for both constraining long term plate locking behavior and searching for slow slip transients. Past studies using seafloor pressure for geodesy have used differenced pairs of pressure records to eliminate oceanographic noise, a primary noise source of seafloor pressure, on the assumption that oceanographic signals are uniform between stations. These studies have identified vertical displacements associated with slow slip on the order of 1-5 cm over instrument separations from 1-50 km in subduction zone settings across the globe. We present an analysis of pressure records from 30 stations in the 2011-2015 Cascadia Initiative experiment and regional physical oceanographic hind cast models developed using the Regional Ocean Modeling System, which have been validated with oceanographic observations, but not previously analyzed for seafloor pressure. We study the root mean square (RMS) amplitude of time series of pressure and pressure differences at periods of 5-30 days to assess the scale, spatial dependence, and temporal dependence of seafloor pressure oceanographic signals. The results indicate that these signals are strongly depth dependent, with filtered pressure RMS values decreasing with depth from >4.5 cm on the continental shelf to <1.5 cm on the abyssal plane for the pressure observations and from >2.5 cm to <1 cm for the model. In contrast, oceanographic signals vary more slowly along depth contours and both data and model show RMS values varying <1 cm at separations >100 km. Based on our noise analysis, we infer that experiments that search for slow slip events should deploy pressure sensors along strike, rather than solely in across strike profiles. We will also explore using temporally and spatially coincident oceanographic models and physical data to correct pressure signals and assess the impact on the threshold for slow slip event detection.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFM.T51E0530F
- Keywords:
-
- 1209 Tectonic deformation;
- GEODESY AND GRAVITY;
- 3006 Marine electromagnetics;
- MARINE GEOLOGY AND GEOPHYSICS;
- 3050 Ocean observatories and experiments;
- MARINE GEOLOGY AND GEOPHYSICS;
- 8170 Subduction zone processes;
- TECTONOPHYSICS