Amphibious Array Facility Workshop Abstracts

Back to AAF Workshop

Submitted By:Anne Trehu
Title:Performance of APGs during ChilePEPPER: calibration using tides and examples of earthquake, tsunami and possibly geodetic signals
Authors:Jose Mieres Madrid, Anne Tréhu, Mike Tryon, Dave Chadwell, Andrew Barclay, Spahr Webb, Eduardo Contreras-Reyes, Javier Ruiz, Gary Egbert, Lana Erofeva, Alex de Moor
Category:General Amphibious Array Science
Keywords:APG, seafloor geodesy, tsunami
Abstract:During ChilePEPPER (Project Evaluating Prism PostEarthquake Response), 5 APGs (absolute pressure gauges) were included in small aperture ocean bottom seismometer array that was deployed from May 2012 to March 2013 updip of the patch of greatest slip during the 2010 Mw8.8 Maule earthquake. Three of the 5 APGs performed very well, as indicated by data with large dynamic range, the absence of obvious artifacts, and coherent recordings of teleseismic earthquakes. To evaluate the long period stability of these sensors and their potential for obtaining geodetic data on vertical Earth motion, we are comparing the tidal signal recorded by these instruments to the predicted tide. Preliminary results indicate that the three instruments recorded tides precisely and accurately, and predicted tides are suggested as a possible approach to empirical in-situ calibration of APGs. We also show records across the ChilePEPPER array of a tsunami generated by the Haida Gwaii earthquake. This signal is coherent between the instruments for ~3 hours after its onset and indicates a maximum peak-to-peak tsunami amplitude of ~1 cm. The Haida Gwaii tsunami signal is the only signal with a period of minutes to days, in addition to tides, that is visible by eye in unfiltered data. All three APGs, however, show a long period decrease in pressure that may represent a geodetic signal. The ChilePEPPER data are well suited to this exploratory calibration test because the instruments were closely spaced (<10 km apart) on thick sediment (~7 km) in deep water (2000-3000 m), minimizing the impact of other potential differences between sites. We plan to apply a similar analysis approach to data from the Cascadia Initiative.
Image:
Caption:
Acknowledgements:ChilePEPPER was funded by the National Science Foundation Marine Geology and Geophysics program through grants to OSU (OCE1130013) and SIO. Instrumentation was provided by the NSF-funded Ocean Bottom Seismolograph Pool. Participation in the post-cruise data analysis was provided by Comisión Nacional de Ciencia y Technología under a program to develop research projects between the United States and Chile (grant #USA2012-001).
Reference:
Submitted By:Xiaowei Chen
Title:Analysis of microearthquakes in the Mendocino triple junction area using the Cascadia Initiative focused array
Authors:Xiaowei Chen, Jeff M. McGuire
Category:Cascadia Science
Keywords:stress drop, apparent stress, rupture mechanics, focused array
Abstract:We detect and locate earthquakes in the Mendocino triple junction area (figure A) using OBS data from the Year 2 deployment. We relocate earthquakes within similar event clusters based on waveform cross-correlation. An example of cross-correlation result is in Figure B. Then we select nearly co-located event pairs with similar waveforms for source parameter analysis. We apply the empirical Green's function method to obtain the source spectra for target events (see Figure C) at each available station. The final source spectra are averaged over all stations. A suite of source parameters will be derived from the source spectra: stress drop, radiated energy, moment, and apparent stress. We are currently working on getting a range of different sized earthquakes. The apparent stress measurements are to be compared with predictions from laboratory derived strength envelope for different plates. We hope to have enough moderate earthquakes with well resolved source parameters to investigate the spatial variations of the source parameters between different types of faults within the array.
Image:
Caption:Figure (A): Map view of stations, white dots are earthquakes from NCEDC occurred during YEAR2 deployment. Figure (B): example of waveforms for an repeating earthquake pair with waveform cross-correlation > 0.95, left column is P wave and right column is S wave. Figure (C): example of EGF deconvolution for an event pair. (Ca) source spectra for the M2 target event. (Cb) and (Cd) individual source spectra at two stations (red: target, black: EGF event). (Cc) and (Ce) spectrum at each station.
Acknowledgements:
Reference:
Submitted By:Kerry Key
Title: Mapping fluids along plate margins with amphibious electromagnetic exploration
Authors:Kerry Key
Category:General Amphibious Array Science
Keywords:electrical conductivity, fluids, electromagnetic exploration, magnetotelluric method
Abstract: Recent advances in electromagnetic (EM) acquisition hardware and modeling tools have created new opportunities for studying plate boundary structure using amphibious arrays. This presentation reviews our current capabilities for EM exploration including shallow imaging using the controlled-source electromagnetic (CSEM) method and broader scale imaging using the magnetotelluric (MT) method. Since the bulk electrical resistivity sensed by EM data depends strongly on porosity and fluid content, EM surveys can provide valuable constraints on fluid processes. At subduction zones, offshore EM data can map the hydration of the incoming oceanic plate, fluids released by sediment compaction and dehydration reactions occurring after the plate is subducted, and fluids escaping through the overlying forearm crust. Since excess fluid pressure may play an important role in regulating subduction zone seismicity, EM data have the potential to reveal new insights on the causes of large earthquakes and episodic tremor and slip. Onshore data can map fluids in the mantle wedge and the generation and delivery of melt beneath the volcanic arc, providing key constraints on the processes governing arc volcanism. Results from the Middle America Trench offshore Nicaragua demonstrate how combined CSEM and MT data can be used to map fluids present at upper mantle to shallow crustal depths; a new 2D inversion images reveals a complex system of water rich faults in the incoming oceanic plate and the complete subduction of water saturated sediments into the margin wedge. A newly acquired amphibious 3D MT array collected in Cascadia demonstrates that high quality data can now be collected routinely in both the onshore and offshore environments.
Image:
Caption:
Acknowledgements:
Reference:
Submitted By:Samer Naif
Title:Imaging the hydrogeology of the Nicaragua margin with electromagnetic methods
Authors:Samer Naif, Kerry Key, Steve Constable, Rob L Evans
Category:General Amphibious Array Science
Keywords:magnetotelluric, controlled source electromagnetic, electrical resistivity, water budget, crustal hydration
Abstract:Fluids, volatiles, or partial melts below the surface of the Earth decrease bulk electrical resistivity by up to several orders of magnitude. Hence, electromagnetic methods are an ideal detection tool that can be used to constrain porosity and fluid fluxes. In order to map the electrical resistivity structure at the Middle America Trench, we collected marine magnetotelluric and controlled-source electromagnetic data along a 280-km profile that spans the offshore component of the Nicaraguan margin. We performed 2D inversion with MARE2DEM, a free state-of-the-art modeling package that is open-source [1]. Our models image a variety of conductive channels that are indicative of hydration and dehydration associated with subduction zone processes at crustal depths, as well as an anomalous conductive layer at the lithosphere-asthenosphere boundary (LAB) . We use our resistivity measurements to estimate porosity with empirical relationships such as Archie’s law. We find that the conductive anomaly at the LAB beneath the incoming oceanic plate is most likely caused by a volatile-rich partial melt layer [2]. At shallower depths, the porosity of the incoming oceanic crust is doubled at the outer rise as compared with the abyssal plain, suggesting that more pore water is being subducted than previously thought. At the margin toe, we image a conductive channel that likely represents subducting sediment. Our porosity estimates for the underthrust sediment layer decay exponentially with increasing depth of burial to 10-km inland of the trench and agree remarkably well with estimates of compacted sediment porosity from laboratory studies [3]. At 20-km into the forearc, we find an anomalous conductor that extends from the plate interface into the overlying plate below a high density of active fluid seeps and mud mounds on the seafloor [4]. The expected temperature/pressure regime at the conductor is consistent with diagenetic dewatering of subducted hydrous minerals [5].
Image:
Caption:
Acknowledgements:We thank Rob Harris for providing us with the thermal model from ref [5]. This work was supported by the National Science Foundation (grants OCE-08411141 and OCE-0840894) and the Seafloor Electromagnetic Methods Consortium at Scripps Institution of Oceanography.
Reference:[1] Key, K. Marine EM inversion using unstructured grids... SEG Exp. Abs. (2012). [2] Naif, S., et al. Melt-rich channel observed at the LAB. Nature (2013). [3] Spinelli, G. A., et al. Hydrogeologic responses to 3D temperature variability, Costa Rica subduction margin. JGR (2006). [4] Sahling, H. et al. Fluid seepage at the continental margin offshore Costa Rica and southern Nicaragua. GGG (2008). [5] Harris, R. N. et al. Thermal regime of the Costa Rican convergent margin: 2. GGG (2010).
Submitted By:Haiying Gao
Title:Offshore Structure of the Cascadia Subduction Zone from Full-wave Tomography
Authors:Haiying Gao, Yang Shen
Category:Cascadia Science
Keywords:full-wave tomography, offshore structure, serpentinization
Abstract:We construct a preliminary offshore model of the crust and uppermost mantle at the Cascadia subduction zone using a full-wave tomographic method. We include the ocean bottom seismometers deployed by the Cascadia Initiative community experiment and Neptune Canada from 2011-2013, and the available broadband stations on land. We have extracted the empirical Green’s functions from continuous seismic records on the vertical components of the OBS and inland station pairs with a frequency-time normalization method, which provide useful Rayleigh-wave signals within the periods of 7-50 s. We have also selected ~50 regional earthquakes between 2011-2013 offshore of the Cascadia subduction zone, which generated useful surface-wave signals up to 75 s period. We simulate wave propagation within a 3D Earth structure using a finite-difference method to generate a station Strain Greens Tensor database and synthetic waveforms. Rayleigh wave phase delays are obtained by cross-correlating the observed and synthetic waveforms. The sensitivity kernels of Rayleigh waves on the perturbations of Vp and Vs are calculated based on the Strain Greens Tensor database. We then invert for the velocity perturbation from the reference model and progressively improve the model resolution. Our preliminary full-wave tomographic imaging using the EGFs and earthquake Rayleigh waves shows: (1) Segmented low-velocity anomalies along the forearc, which are spatially correlated with the patterns of offshore basins and high slip patches; (2) Low velocities beneath the Blanco fracture zone; (3) The distribution of pseudofaults defines the seismic velocity heterogeneities; and (4) A low-velocity zone beneath the oceanic Moho near the trench, which may indicate serpentinization of the mantle lithosphere.
Image:
Caption:
Acknowledgements:
Reference:
Submitted By:Yen Joe Tan
Title:Tidal triggering of earthquakes in the Alaska-Aleutian subduction zone
Authors:Yen Joe Tan, Maya Tolstoy
Category:Cascadia Science
Keywords:forecasting, earthquake, tidal triggering
Abstract:Significant tidal triggering of earthquakes has been observed precursory to the Tohoku and Sumatra megathrust earthquakes (Tanaka 2010; 2012). The appearance of high correlation between tidally-induced stresses and earthquake occurrence frequency several to ten years before these megathrust earthquakes suggests that such statistical analysis could be useful in improving forecasting of future subduction zone earthquakes. Using this statistical method, we analyzed the Alaska-Aleutian subduction zone which has been known to produce devastating tsunamigenic earthquakes, and specifically the Semidi Segment that is probably late in its earthquake cycle (Davies et. al. 1981). Our study aims to understand if significant tidal triggering of earthquakes were present precursory to historical great earthquakes in this region. We also aim to understand if any segment along the subduction zone is currently displaying statistically significant tidal triggering of earthquakes and whether such observations are indicative of the stress state of the segment. Finally, we test if the strength of tidal triggering captured by this statistical method is sensitive to the tidal stress azimuth used. Such sensitivity could be indicative of the predominant fault slip direction in the specific segment.
Image:
Caption:
Acknowledgements:NSF grant OCE‐0961594
Reference:Tanaka, S. (2010), Tidal triggering of earthquakes precursory to the recent Sumatra megathrust earthquakes of 26 December 2004 (Mw 9.0), 28 March 2005 (Mw 8.6), and 12 September 2007 (Mw 8.5), Geophysical Research Letter, 37. L02301, doi:10.1029/2009GL041581. Tanaka, S. (2012), Tidal triggering of earthquakes prior to the 2011 Tohoku-Oki earthquake (Mw 9.1), Geophysical Research Letter, 39, L00G26, doi:10.1029/2012GL051179.
Submitted By:Roy Hyndman
Title:The Arctic Beaufort Sea Alaska, Yukon, NWT margin
Authors:Roy Hyndman, PGC, Geol. Surv. Canada
Category:Cascadia Science
Keywords:Arctic, Beaufort Sea, Alaska, Yukon
Abstract:The Arctic margin of the Beaufort Sea, Alaska, Yukon, and western NWT provides an outstanding opportunity for the Amphibious Array OBSs, land stations, and multichannel surveys. USArray-Alaska stations now being installed are very complementary. There have been extensive surveys by the petroleum industry and by the US and Canadian Geological Surveys on the coastal continental shelf but little on the margin deep crustal structure and tectonics. Until recently sea ice severely limited marine operations but arctic warming now allows offshore work with less difficulty. There is a passive rifted margin off western Alaska and the Canadian Arctic Islands. However, for the intervening 400 km from the sinistral Canning River Displacement Zone in eastern Alaska to the dextral Richardson-Eskimo Lakes zone at the eastern edge of the Mackenzie Delta there is a well-developed margin fold and thrust belt of Cretaceous to Recent age. This is an extraordinary “subduction zone” where the continental crust appears to be overthrusting the oceanic lithosphere driven by the Yakutat terrane collision in the Gulf of Alaska 1000 km to the southwest (see Steep Project). Cordillera crustal temperatures appear hot enough for detachment in the lower crust, allowing this long-distance transport. The Canning and Richardson fault zones are seismically active with motions of a few mm/yr. However, the thrust belt has little seismicity so it is not clear if convergence is aseismic or in large ~M8 events like Cascadia. There is a Beaufort Sea seismicity cluster that may result from downward bending of the oceanic lithosphere as it is overthrust by the continent. Some questions: (1) Deep structure of the passive and convergent Beaufort Sea margins? (2) Origin and tectonic history of the Beaufort Sea and arctic margins? (3) Model of tectonic transport from the Yakutat collision and earlier motions? Is this type of lower crust detachment and transport more common globally than previously recognized?
Image:
Caption:Left: Model of long-distance crustal transport from the Gulf of Alaska Yakutat terrane collision; Right: Seismicity defining northwestward moving crustal block over-riding the Arctic ocean.
Acknowledgements:
Reference:
Submitted By:Spahr  Webb
Title:Lessons from the Cascadia Deployment
Authors:Spahr C. Webb
Category:Cascadia Science
Keywords:Cascadia Initiative, Seafloor observations
Abstract:The Cascadia deployment was remarkable in many ways. The instruments were built over a very short time by groups at SIO, WHOI and LDEO with funding from the ARRA program. It was the first large community experiment using broadband instruments managed by an volunteer expedition team. It involved an enormous number of OBS deployments with very short turn arounds between OBS recoveries and redeployments during a huge number of cruises. The new SIO Abalone and the LDEO trawl shielded OBSs demonstrated for the first time that significant reduction in horizontal component noise levels could be obtained through the use of current shielding. The trillium compact sensor deployed with all the instruments proved to be a very successful sensor for marine work, although like most broad band sensors, it will clip on the larger local events. This could be a concern for deployments in seismically active regions (Alaska). The first extensive deployment of high resolution absolute pressure gauges (APGs) was conducted. Comparisons with DPGs and hydrophones reveal tradeoffs between resolution, sensor response stability, and dynamic range that should be considered for future deployments. A review of the results from the first three years of the Cascadia experiment can provide some guidance to planning future deployments with the Cascadia instruments both with regard to logistics and planning and with regard to improvements to the instruments to optimize seismic observations.
Image:
Caption:
Acknowledgements:NSF OCE- 1154795
Reference:
Submitted By:Miles Bodmer
Title:SKS shear-wave splitting observations from the first two years of Cascadia Initiative OBS data.
Authors:Miles Bodmer, Douglas Toomey, Emilie Hooft
Category:Cascadia Science
Keywords:anisotropy, Shear-wave splitting,
Abstract:We present SKS splitting measurements from over 100 ocean bottom seismometers (OBS) deployed during the first two years of the Cascadia Initiative (CI) amphibious array. The CI dataset is unique because it includes several regions that can distinctly influence upper mantle anisotropic fabric development, such as: upwelling beneath the Juan de Fuca (JdF) and Gorda ridges, the young and evolving oceanic lithosphere-asthenosphere system of the plate interior, the Blanco transform, Mendocino triple junction, and the Cascadia subduction zone. To address the high noise levels present on OBSs at the frequencies used for splitting, we implement a rigorous quality control scheme. Our analysis takes into account the response of common splitting methods to high noise data and addresses known issues such as cycle skipping, false minima, low transverse energy, and near-null measurements. Events are filtered at 0.03-0.1 Hz and individual measurements are manually checked for quality and stacked. For the JdF plate interior, results show a coherent fast axis roughly aligned with absolute plate motion (APM), however, there are some systematic offsets. Near the JdF ridge and the subduction zone north of Cape Blanco, a few measurements are rotated parallel to the plate boundaries. Measurements near the Gorda Ridge, including the Pacific and Gorda plates, exhibit a remarkably coherent fast-axis orientation (N70˚W) that is not aligned with APM and is not affected by plate boundaries. From southern Gorda toward the JdF plate, the fast-axis orientations transition from N70˚W toward the APM direction. We also discuss the quality of the CI data, highlighting differences between instrument types, Year 1 versus Year 2 data, and the effect of OBS depth. Further, we consider how the CI as a large-scale community experiment facilitates analyses such as this, involving regional scale phenomenon in complex, noisy environments.
Image:
Caption:
Acknowledgements:NSF Award: 209751
Reference:
Submitted By:Yang Zha
Title:Variation of crustal structure along the Eastern Lau back-arc Spreading Center constrained by seafloor compliance
Authors:Yang Zha, Spahr C. Webb, Robert A. Dunn
Category:General Amphibious Array Science
Keywords:Seafloor compliance; Lau Basin. Mid-ocean ridge. OBS
Abstract:Measurements of seafloor compliance, the deformation under long period (typically 30-300 s) ocean wave forcing, are primarily sensitive to crustal shear velocity structure. We analyze seafloor compliance from data collected from a subset of 50 broadband Ocean Bottom Seismographs (OBS) deployed at the Eastern Lau spreading center (ELSC) from 2009 to 2010. We calculate seafloor compliance functions by taking the spectral transfer function between the vertical displacement and pressure signal recorded by the 4-component OBSs. In the ridge perpendicular direction, compliance amplitude vary by more than an order of magnitude from the ridge crest to older seafloor covered by sediment. Along the spreading ridge, compliance measured from on-axis sites increases southwards, indicative of a decrease in the upper crustal shear velocity possibly due to increasing porosity and a thickening extrusive layer [Jacobs et al., 2007; Dunn et al., 2013]. We apply a Markov Chain Monte Carlo method to invert the compliance functions for crustal shear velocities at various locations along the ELSC.
Image:
Caption:
Acknowledgements:
Reference:Dunn, R. A., F. Martinez, and J. A. Conder (2013), Crustal construction and magma chamber properties along the Eastern Lau Spreading Center, Earth and Planetary Science Letters, 371, 112-124, doi: Doi 10.1016/J.Epsl.2013.04.008. Jacobs, A. M., A. J. Harding, and G. M. Kent (2007), Axial crustal structure of the Lau back-arc basin from velocity modeling of multichannel seismic data, Earth and Planetary Science Letters, 259(3-4), 239-255, doi: Doi 10.1016/J.Epsl.2007.04.021.
Submitted By:Zhao Chen
Title:Microseism Source Direction in Cascadia Using Cross-correlation and Array Coherence
Authors:Zhao Chen, Peter Bromirski, Peter Gerstoft, Ravishankar Menon
Category:Cascadia Science
Keywords:microseism, cross-correlation, coherence
Abstract:The ongoing Cascadia Initiative (CI) is providing an increasing quantity of ocean bottom seismometer (OBS) data. These data are used to investigate the microseism source directions and to determine the surface wave tomography of the Cascadia region. Spatial asymmetry in the amplitude of cross-correlations between receiver pairs is observed in both primary microseism and double-frequency (DF) microseism bands. This asymmetry indicates that there are preferential microseism propagation directions, either because of source location and/or upper oceanic crustal structure. Ambient noise group velocity tomography was determined from cross-correlation of DF microseisms between receiver pairs. A low-velocity region oriented roughly north-south was identified, apparently associated with a thick sediment zone off the continental shelf, reflecting the strong influence of ocean bottom sediment layer thickness on DF microseism propagation. The coherence of DF microseism signals between different stations was lowest in the low-velocity thick-sediment region, consistent with the thick sediment layer strongly affecting microseism signal propagation.
Image:
Caption:
Acknowledgements:Thank the NSF for the funding.
Reference:[Bromirski et al., 2013]; [Gerstoft and Tanimoto, 2007]; [Hasselmann, 1963]; [Harmon et al., 2007]; [Kay, 1998]; [Kedar et al., 2008]; [Oliver, 1962]; [Sabra et al., 2005]; [Yang et al., 2008]
Submitted By:Koichiro Obana
Title:SeaJade Earthquake Observations in Cascadia Subduction Zone off Vancouver Island
Authors:Koichiro Obana, Shuichi Kodaira, Yojiro Yamamoto, Tsutomu Takahashi (JAMSTEC) Kelin Wang, Michael Riedel, Honn Kao (GSC) Martin Scherwath, George Spence, Jesse Hutchinson, Ayodeji Paul Kuponiyi (University of Victoria) Michael Bostock (UBC)
Category:Cascadia Science
Keywords:Seismogenic Zone, Subduction, OBS
Abstract:Offshore seismicity at the northern Cascadia margin is poorly constrained because previous earthquake monitoring was conducted mostly using onshore seismograph networks. Project Seafloor Earthquake Array Japan-Canada Cascadia Experiment (SeaJade) is designed to fill this knowledge gap (Scherwath et al., 2011). Under this project, we deployed 33 ocean bottom seismographs (OBSs) in early July, 2010 and recovered 32 in late September, 2010 off Vancouver Island using Canadian research vessel CCGS John P. Tully. Seismic events were detected from the continuous OBS records, and P- and S-wave arrivals were picked manually. We estimated hypocenter locations using a one-dimensional velocity structure with station corrections for low velocity sediments beneath the OBSs. Our results show that during the recording period most of the offshore seismicity occurred in the subducting plate along the Nootka fault zone. Otherwise seismicity was extremely low, with no earthquakes located along the shallow, seismogenic part of the Cascadia megathrust. The lack of interplate seismicity may indicate complete healing and locking of the megathrust over three centuries after the great earthquake of 1700 and a lower degree of structural heterogeneity due to the large amount of seafloor sediments and the absence of large geometric irregularities such as subducting seamounts. Building on the observations in 2010, we are currently conducting earthquake monitoring using a combination of offshore and onshore stations, with a greater focus on the Nootka fault zone and its landward extension. We installed 8 temporary land stations around the Nootka Island in October 2013 and deployed 35 OBSs off the Vancouver Island in December 2013. This amphibious network is complemented by the permanent land stations of the Canadian National Seismograph Network. The OBSs will be recovered in September 2014. The new data will further improve our understanding on earthquake processes at the Cascadia margin.
Image:
Caption:
Acknowledgements:This work was supported by Geological Survey of Canada, University of Victoria, and JSPS KAKENHI Grant Number 23253005.
Reference:Scherwath, M., G. Spence, K. Obana, S. Kodaira, K. Wang, M. Riedel, J. McGuire, and J. Collins (2011), Seafloor seismometers monitor northern Cascadia earthquakes, EOS Trans. AGU, 92(47), 421-422, doi:10.1029/2011EO470001.
Submitted By:Takashi Tonegawa
Title:Shear wave anisotropy within marine sediments in the Cascadia subduction zone derived from Ps converted waves
Authors:Takashi Tonegawa, Koichiro Obana, Yojiro Yamamoto, Shuichi Kodaira (JAMSTEC), Kelin Wang, Michael Riedel, Honn Kao (Pacific Geoscience Centre), George Spence (University of Victoria)
Category:Cascadia Science
Keywords:marine sediment, anisotropy, Ps conversion
Abstract:The Seafloor Earthquake Array-Japan-Canada Cascadia Experiment (SeaJade) Project has deployed ocean bottom seismometers (OBSs) on the Cascadia subduction zone, including 32 short period OBSs with three components from JAMSTEC. Although the direction of the horizontal component of OBS is determined from airgun shots and long-period components of P and surface waves, we could not use the both approaches for this observation. In this study, we used Ps phases converted from the bottom of marine sediments to estimate both the direction of the horizontal component and shear wave anisotropy within marine sediments, using near-field earthquakes. Among 1,247 earthquakes occurring at a period from Jul. to Sep. on 2010 (Obana et al., 2014, in revision), we selected events that show clear P wave in the vertical component and Ps wave in the horizontal components at a frequency band of 2-8 Hz. The number of the selected events is totally 1,732. P and Ps waves are emerged in the vertical and radial components, respectively. For an event, rotating the two horizontal components, we can search a radial direction at which the cross-correlation coefficient between the P and Ps waves is maximum, and estimate the horizontal direction using back-azimuth information of the P wave. However, this processing can be performed in the case of marine sediments with isotropic medium. Since we assumed that marine sediments have an anisotropic structure, we first estimated splitting parameters relevant to Ps wave. Correcting seismograms of Ps waves using them, the back-azimuth of P wave can be obtained, i.e., the direction of the corrected particle motion of Ps waves correspond to the back-azimuth. As a result, the fast axes were estimated at the NE-SW direction in the southern part of the array. Perhaps this reflects the crack-orientation within marine sediments, which is formed by the compressional field associated with the northeastward subduction of the Juan de Fuca Plate.
Image:
Caption:Figure 1. (a) Map showing the horizontal direction at each station. (b) Map showing the direction of fast axis at each station. The ellipse indicates the stations at which the fast axis is oriented to the NE-SW direction.
Acknowledgements:
Reference:
Submitted By:Aaron Wech
Title:Amphibious tremor observations
Authors:Aaron G Wech
Category:General Amphibious Array Science
Keywords:tremor, OBS, slow slip
Abstract:Widespread observations of tremor and slow slip in different tectonic environments reveal that these newly recognized processes play fundamental roles in accommodating and marking fault motion. Understanding the relationship between fault conditions, slow slip and rapid earthquake rupture, however, requires a reliable method of identifying and characterizing slow-slip phenomena. These tasks are particularly challenging in offshore environments where the investigation of plate boundary processes, especially within subduction-zone megathrusts, typically relies, at least in part, on ocean bottom seismometers (OBSs). OBS deployments have improved our ability to detect earthquake seismicity offshore, but their noisy settings create uncertainty about their ability to constrain more emergent signals. Here we investigate the utility of OBSs for detecting and locating emergent seismic sources using examples of both tectonic and volcanic origin. For tectonic tremor, we demonstrate that signals associated with the deep extension of the Alpine fault have been recorded on an OBS network lying off the west coast of New Zealand’s South Island. Similarly, offshore Hawaii’s Big Island, we identify deep, volcanic tremor using a Loihi OBS network. In both cases, we focus on time windows known to contain tremor based on land observations and identify coherent tremor signals on OBS stations. Using existing tremor methods, the envelopes from OBS channels were correlated against each other and with land stations to constrain tremor epicenters. There are special noise considerations, but in many cases OBS data correlated at least as well as land stations at similar distances, and Alpine Fault tremor was faithfully recorded at distances of as much as 150 km. These results highlight the performance of OBS stations in the tremor frequency band (∼1–10 Hz) and demonstrate the potential effectiveness of OBS networks in constraining slow slip processes offshore.
Image:
Caption:
Acknowledgements:The NSF funded both OBS deployments. Thanks to Anne Sheehan, Jackie Caplan-Auerbach and Cliff Thurber for sharing data; and thanks to Carolin Boese, Weston Thelen, John Townend and Tim Stern for helpful discussion and distraction.
Reference:Wech, A. G., A. F. Sheehan, C. M. Boese, J. Townend, T. A. Stern, and J. A. Collins (2013), Tectonic Tremor Recorded by Ocean Bottom Seismometers, Seismological Research Letters, 84(5), 752–758, doi:10.1785/0220120184.
Submitted By:Helen Janiszewski
Title:Imaging the downgoing Juan de Fuca crust using receiver functions from the Cascadia Initiative
Authors:Helen A Janiszewski, Geoffrey A Abers, Jim Gaherty, Helene Carton
Category:Cascadia Science
Keywords:receiver function, ocean bottom seismometer, low velocity zone, subduction
Abstract:The Cascadia subduction zone is a hot end-member system that is characterized by the subduction of young, thickly sedimented lithosphere. Previous receiver function studies have observed a low velocity zone (LVZ) with strong contrasts along the thrust up to 40 km depth. It is hypothesized that this may be created by a channel of either near-lithostatic pressure fluids or stronger metasediments, implying a weak thrust zone. These studies have been limited to data from onshore stations, and thus have not imaged the shallower, geodetically locked portion of the thrust zone, which is located offshore. The ocean bottom seismometers (OBS) from the Cascadia Initiative (CI), which are among the first broadband instruments successfully deployed in shallow water using low-profile Trawl-Resistant-Mounts (TRM), offer the opportunity to extend receiver function studies of the LVZ offshore. Calculation of receiver functions from OBS data is difficult due to water column noise. Fortunately, the TRM housing yields quieter horizontal-component signals, and receiver functions are calculated at all of the successfully deployed TRM OBS from CI Year 1, as well as at some deep water stations. We use velocity models from the previous onshore receiver function studies to generate synthetic receiver functions to compare with our data. The shallow-water stations deployed off the coast of Grays Harbor, Washington record a high-amplitude asymmetric arrival consistent with reverberations off the top and bottom of the LVZ. This high-amplitude arrival is not as evident at other stations farther south along the margin region. This along strike variation may be evidence for segmentation along the thrust zone. A careful analysis of these complex signals will be needed to determine the extent of the LVZ offshore, but these results indicate that receiver functions calculated from the CI OBS will be useful in imaging structural heterogeneity in the subducting crust along the margin region in Cascadia.
Image:
Caption:
Acknowledgements:Funding provided by the National Science Foundation Graduate Research Fellowship, and awards EAR-1147622.
Reference:
Submitted By:Songqiao Wei
Title:Seismic Attenuation Structure of the Lau Back-arc Basin from a Local Amphibious Array
Authors:S. Shawn Wei, Douglas A. Wiens, Spahr C. Webb, Robert A. Dunn, James A. Conder
Category:General Amphibious Array Science
Keywords:Tonga Subduction Zone, Lau Back-arc Basin, Body Wave Attenuation
Abstract:We investigate seismic attenuation beneath the Lau Backarc basin and adjacent regions using data from the 2009 – 2010 Ridge2000 East Lau Spreading Center project, which consisted of 49 ocean bottom seismographs (OBSs) and 17 land-based seismic stations deployed in Fiji, Tonga, and the Lau basin for about one year. 3D attenuation structures are inverted from t* measurements of P and S waves from local earthquakes. In order to avoid the trade-off between t* and corner frequency (fc), we analyze the spectral ratio of S coda to independently constrain the fc for each event. The Qp and Qs structures are inverted separately, whereas Qp/Qs is jointly inverted from Qp and t*(S). Tomographic results show strong signals of high attenuation within the upper 100-km of the mantle beneath the back-arc basin, suggesting perhaps the highest seismic attenuation (Qp < 35 and Qs < 25) known in the mantle. These anomalies require not only the abnormally high temperature but also the existence of partial melt. The inferred partial melt aligns with the spreading centers at shallow depths, but shift westwards at depth away from the slab, implying a passive decompression melting process governed by the mantle wedge flow pattern. The Tonga volcanic arc does not display as strong of velocity or attenuation anomalies as the spreading centers, suggesting less magmatism associated with the arc compared to the back-arc.
Image:
Caption:Seismic attenuation of P wave beneath the Lau Basin. Upper panels show the map views at the depth of 25 and 75 km, respectively. The lower panel shows a cross-section perpendicular to the Tonga trench. Black curves indicate the spreading centers: CLSC, Central Lau Spreading Center; ELSC, East Lau Spreading Center; VFR, Valu Fa Ridge.
Acknowledgements:We thank Patrick J. Shore and all other field workers of the ELSC project. Heather N. Relyea and Catherine A. Rychert provided great help on data processing. William Menke gave valuable advice on tomography inversion. In addition, we thank Terry Plank and Philip Skemer for constructive discussions on seismic interpretations. This research was funded by NSF Grant OCE-0426408 and EAR-0911137.
Reference:
Submitted By:Jesse Hutchinson
Title:Results and analysis from the SeaJade oceanic array of the Casacdia fore-arc
Authors:Jesse Hutchinson, Honn Kao
Category:Cascadia Science
Keywords:Cascadia subduction zone, OBS, Nootka fault zone, fore-arc, converted phase, relocation
Abstract:SeaJade is a collaborative project between Japan Agency for Marine-Earth Science and Technology (JAMSTEC) and the Geological Survey of Canada (GSC) to study the Cascadia Subduction Zone and Nootka Fault Zone off the west coast of Vancouver Island. The purpose of the experiment is to better understand the plate boundaries and associated faults, and the earthquakes that result from the interaction and build-up of stress among them. The first phase was of SeaJade initiated in July 2010, with the deployment of thirty-five ocean bottom seismometers, and recorded data for three months, while phase II began in October 2013 with the deployment of eight broadband seismometers in the Nootka Sound region of Vancouver Island and 35 OBS’ in December of the same year. For this experiment, we have calculated hypocentral locations, double-differential relocations, magnitudes, and analyzed converted phases. The results have allowed us to identify that most seismic activity occurred within or near the Nootka Fault Zone and we have also interpreted at least two refractive interfaces within the Cascadia fore-arc.
Image:
Caption:
Acknowledgements:Research possible due to cooperation with JAMSTEC and with grants provided by NSERC and the University of Victoria.
Reference:
Submitted By:Daniel Bowden
Title:Ambient Noise Tomography with the ALBACORE OBS Array
Authors:Daniel Bowden, Monica Kohler, Victor Tsai
Category:General Amphibious Array Science
Keywords:Ambient Noise, Velocity Structure, Signal Processing
Abstract:The ALBACORE array, deployed off the coast of Southern California, is providing an unprecedented opportunity to image seismic velocities across the region’s continental shelf and the Pacific-North America plate boundary. Our study focuses on measuring seismic velocities from ambient noise cross correlations, collected from a 12 month period between 2010 and 2011. Particular attention has been paid to improving signal to noise ratios in the noise correlations, by removing the effects of instrument tilt and infragravity waves. These sources of non-elastic noise are measured in the transfer function between horizontal and vertical components for the tilt, and between DPG and vertical component for the infragravity oceanic signals. While removing these effects has a noticeable effect for observing earthquake events in the timeseries, we find that they have only a small effect on the year long stacks of ambient noise cross correlations. We also examine the effectiveness of other processing techniques including time domain normalization and spectral whitening of the signals, and the time required for a stable noise correlation function. Standard time domain and frequency domain methods are used to examine surface wave dispersion curves for group and phase velocity between 5 and 50 seconds period, and these will be simultaneously inverted for a 3D velocity structure. We are using 595 station pairs which include a number of on-land stations from the Southern California Seismic Network, and hope that our modeling will show a compatible extension of community velocity models which previously could not extend across the transition between continental and oceanic lithosphere. Properly constraining crustal properties across the transitions across the continental shelf and the coastline will improve earthquake locations and rupture studies in Southern California.
Image:
Caption:
Acknowledgements:Supported by grants from the NSF-OCE and the USGS NEHRP.
Reference:
Submitted By:Allen Husker
Title:Subduction in Mexico
Authors:UNAM Seismology Group
Category:Future Array Targets
Keywords:Mexican Subduction Zone, SSE, Tectonic Tremor, Large Earthquakes
Abstract:The Mexican Subduction zone is very complex with many large earthquakes, SSE’s and tremors, flat slab geometry and many unanswered questions. It is the only place in the world where there are 1000+ km of coastline within 80 km of the trench allowing GPS measurements directly above the seismogenic zone of large earthquakes. Since 1990 there have been ten M7+ thrust events in the subduction zone as compared to one in Cascadia in that same time period. There are also M7+ SSE’s that occur every four years within the Guerrero Seismic Gap where no M6.5+ earthquake has occurred for 100+ years. The SSE’s diminish the accumulated slip in the Gap to four times less than the accumulated slip outside of the Gap. The subducting slab is one of only 3 truly horizontal slabs in the world. Unlike the other two, Chile and Peru, there is no evidence of a ridge to raise the slab and the volcanism is pushed farther from the trench instead of being absent. At 150 km from the trench and 40 km depth the slab turns to become flat and stays there until nearly 300 km from the trench where it turns again to dive into the mantle. In addition, the Cocos plate rotates with respect to the North American plate so that the rate of subduction, the length of the flat portion of the slab and the distance from the trench to the volcanoes generally increases with distance from the Euler Pole. The SSE’s are divided into short term and long term. In the Gap where the flat slab extends the furthest from the trench, the long term SSE’s partially invade the seismogenic zone and the short term SSE’s are further downdip in the flat section of the slab. The tremor is separated into three zones within the flat slab: 1) the Sweet Spot has the largest amount (190 km – 230 km from the trench); 2) the Buffer Zone has almost none (170 km – 190 km from the trench); and 3) the Transient Zone has tremor during SSE’s (140 km – 170 km from the trench).
Image:
Caption:Large Mexican Earthquakes
Acknowledgements:
Reference:
Submitted By:Joseph Byrnes
Title:Structure of the Lithosphere-Asthenosphere System Beneath the Juan de Fuca Plate: Results of Body Wave Imaging Using Cascadia Initiative Data
Authors:Joseph S. Byrnes, Douglas R. Toomey, and Emilie E. E. Hooft.
Category:Cascadia Science
Keywords:Tomography, Body waves, Juan de Fuca, Gorda, Lithosphere, Asthenosphere, upwelling, mid-ocean ridge
Abstract:The Cascadia Initiative (CI) of NSF deployed ocean bottom seismographs across two oceanic plates from the mid-ocean ridges where they are created to their subduction beneath a continent, providing the opportunity to study the lithosphere-asthenosphere system during each stage of the evolution of a tectonic plate. We present body wave tomographic images of the isotropic and anisotropic seismic structure beneath the Juan de Fuca and Gorda plates that extend from the North American coast out to the Pacific plate. Our results provide constraints on mantle structure beneath several geologic features of interest, including the Juan de Fuca and Gorda spreading centers, the Blanco Transform, near ridge hotspots such as Axial Seamount, and oceanic upper mantle structure beneath the Cascadia Subduction Zone. We measured the delay times of teleseismic P and S phases by cross correlation of seismograms from the first three years of the CI. While the signal to noise ratio is generally higher for P waves, S waves show a greater range of delay times. We inverted for seismic velocity perturbations with a damped least-squares method that accounts for finite-frequency effects, perturbations to the ray path by 3D heterogeneity, and seismic anisotropy. We construct velocity models under the assumption of both isotropy and hypothesized orientations of seismic anisotropy. Preliminary results indicate that the velocity structure of the Juan de Fuca spreading center is asymmetric, with lower velocities beneath the Pacific plate. The velocity anomaly beneath the Gorda spreading center also appears isolated beneath the spreading center, with low velocities not extending into the adjacent, older plates. We compare the results with geodynamic models for a deeper understanding of how the oceanic lithosphere and asthenosphere behave. Future work includes inverting body and surface wave data together, and testing orientations of seismic anisotropy motivated by SKS splitting measurements.
Image:
Caption:
Acknowledgements:
Reference:
Submitted By:Weisen Shen
Title:Directionality of ambient noise on the Juan de Fuca Plate: Implications for source locations of the primary and secondary microseisms
Authors:Ye Tian, Michael H. Ritzwoller, Weisen Shen
Category:Cascadia Science
Keywords:Seismology, Surface waves, Wave propagation, Interferometry
Abstract:Based on cross-correlations of ambient seismic noise computed from 61 ocean bottom seismometers (OBSs) within the Juan de Fuca plate from the Cascadia Initiative experiment and 42 continental stations near the western US coast, we investigate the locations of generation of the primary (11-20 sec period) and secondary (5-10 sec period) microseisms in the northern Pacific Ocean by analyzing the directionality of the microseism signals received in this region. (1) Ambient noise observed across the array is much different in the primary and secondary microseism bands, both in its azimuthal content and seasonal variation, indicating different source generation locations. (2) The principal secondary microseism signals propagate toward the east, consistent with their generation in deep waters of the North Pacific, perhaps coincident with the region of body wave excitation observed by recent studies. (3) Evidence is presented both for distant primary microseism sources probably in the southern hemisphere as well as local sources within and near the Juan de Fuca plate. Observations of the azimuthal dependence of the amplitude of the fundamental mode Rayleigh wave as well as observations of precursory arrivals in the cross-correlations establish the strongest local generation region lies northwest of the Juan de Fuca plate near the coast of British Columbia perhaps near Graham Island with weaker local sources appearing oceanward of Vancouver Island and southern Oregon. (4) High quality Green’s functions are derived from cross-correlations between deep water OBSs and continental stations illustrating that deep water generated signals can efficiently propagate onto the continent and are well recorded by continental seismic stations, at least at periods longer than about 5 sec. In conclusion, the primary and secondary microseisms are generated at different locations.
Image:
Caption:The azimuthal variations of SNRs averaged across all months for the secondary microseisms are shown in (a) for OBS stations, (b) for continental stations to the north of 44°N, and (c) for continental stations to the south. The back-projected great circle paths of these three sets of peaks are plotted in (d) with the same colors.
Acknowledgements:The authors are grateful to the Cascadia Initiative Expedition Team for acquiring the Amphibious Array Ocean Bottom Seismograph data and appreciate the open data policy that made the data available shortly after they were acquired. The IRIS DMS is funded through the US National Science Foundation under Cooperative Agreement EAR-0552316. This work utilized the Janus supercomputer, which is supported by the National Science Foundation (award number CNS-0821794)
Reference:Ye, T., W. Shen, and M.H. Ritzwoller, Crustal and uppermost mantle shear velocity structure adjacent to the Juan de Fuca Ridge from ambient seismic noise, Geochem. Geophys. Geosyst., 14(8), 3221-3233, doi:10.1002/ggge.20206, 2013. Ye, T. and M.H. Ritzwoller, Directionality of ambient noise on the Juan de Fuca Plate: Implications for source locations of the primary and secondary microseisms, Geophys. J. Int., submitted, 2014.
Submitted By:Samuel Bell
Title:Comparison of Long-Period Noise on Vertical Component OBS Records from Cascadia
Authors:Sam Bell, Don Forsyth, Youyi Ruan
Category:Cascadia Science
Keywords:OBS, noise, instrument design, compliance
Abstract:Although most of the Earth's surface lies beneath the ocean, ocean bottom seismology (OBS) is plagued by high noise, which complicates many analyses. Two main approaches have been taken to address this problem. First, vertical noise can be removed by using the horizontal displacement data to predict the vertical tilt noise caused by bottom currents and using the pressure gauge record to predict the vertical infragravity wave compliance noise. Second, advances in instrument design have been made. Instrument packages have been shrunk (Trillium compact) or buried (Japanese design) to present a lower profile to the bottom currents, shielding has been applied, and the instrument architectures have been improved to reduce sensor tilt angles. To better understand the performance of the various instrument designs, we assessed the success of the noise removal process in two older instrument types with the sensor in large spherical pressure chambers, WHOI-Keck and SIO-traditional, and two newer designs with compact Trillium sensors, the unshielded WHOI-ARRA and shielded SIO-TRM (Abalone). We measured the typical noise amplitude at each station in the 50s-100s band. Due to large glitches, the measurements were conducted by hand examination of the seismograms. Because of high noise in shallow water, we excluded stations shallower than 500 m. The smaller, unshielded Trillium compact reduces the horizontal noise levels relative to the older, larger spherical designs. The shielding further reduces horizontal noise significantly. However, the current-induced noise on the vertical component is much less correlated with the horizontal components in these new designs, making the noise removal process less effective. The older designs actually tend to have lower vertical noise after noise removal. The geometric mean noise amplitude in the 50s-100s band after noise removal is 210 nm for SIO-traditional, 330 nm for WHOI-Keck, 340 nm for SIO-TRM, and 920 nm for WHOI-ARRA.
Image:
Caption:
Acknowledgements:
Reference:
Submitted By:Aubreya Adams
Title:Seismic structure of the Tonga Arc and Lau Backarc Spreading Center from joint inversion of local and teleseismic body wave arrivals recorded on OBS and land-based stations
Authors:Aubreya Adams, Douglas Wiens, Donna Blackman, Spahr Webb, Rob Dunn, James Conder, and Dapeng Zhao
Category:General Amphibious Array Science
Keywords:tomography, subduction, melt
Abstract:The Lau Backarc Spreading Center and Tonga Arc comprise and excellent location in which to the study the dynamics and variations in melt production and migration. The Lau Backarc Spreading Center is comprised of three segments, Central Lau Spreading Center, Eastern Lau Spreading Center, and the Valu Fa Spreading Center. The segments show very different spreading and magma production rates as a function of their distance to the arc and slab, which decreases from north to south. Geochemical studies also indicate a systematic change in magma composition along strike of the Lau Back-Arc Spreading Center with more water and fluid-mobile elements in the south. From 2009-2010, 16 broadband seismometers and 51 ocean bottom seismometers (OBS) were installed for one year across the Lau backarc spreading centers and in Fiji and Tonga to image the along-strike structural variation of the subduction zone and the dynamics of the melt production region. We present preliminary results from the joint inversion of P and S wave arrivals from local and teleseismic events recorded by this network. Preliminary results show a clearly defined westward-dipping low velocity region beneath the Eastern Lau and Valu Fa Spreading Centers, which is most pronounced at depths less than 150 km. The low velocities beneath the Eastern Lau and Valu Fa Spreading Centers are more pronounced relative to those associated with the Tonga Arc.
Image:
Caption:Location of study region, showing segmentation of the back arc spreading centers along strike of the subduction system. (CLSC - Central Lau Spreading Center; ILSC - Intermediate Lau Spreading Center; ELSC - Eastern Lau Spreading Center; VFSC - Valu Fa Spreading Center)
Acknowledgements:
Reference: