|
Supplementary material to “Probing the Hawaiian Hot Spot With New Broadband Ocean Bottom Instruments”
13 October 2009
Gabi Laske, Scripps Institution of Oceanography, La Jolla, California
John A. Collins, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts
Cecily J. Wolfe, University of Hawai'i at Manoa, Honolulu
Sean C. Solomon, Carnegie Institution of Washington, Washington, D. C.
Robert S. Detrick, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts
John A. Orcutt, Scripps Institution of Oceanography, La Jolla, California
David Bercovici, Yale University, New Haven, Connecticut
Erik H. Hauri, Carnegie Institution of Washington, Washington, D. C.
Citation:
Laske, G., J. A. Collins, C. J. Wolfe, S. C. Solomon, R. S. Detrick, J. A. Orcutt, D. Bercovici, and E. H. Hauri (2009), Probing the Hawaiian hot spot with new broadband ocean bottom instruments, Eos Trans. AGU, 90(41), 362–363. [Full Article (pdf)]
Data Return and an Example
This supplement provides an example of data collected for the Hawaiian PLUME project. Because the OBSs were not buried, it was expected that ambient noise conditions would substantially restrict analyses at short periods (about 1 s and less), especially on the horizontal components. At longer periods, the horizontal components are much noisier than the vertical, but enough earthquakes can be analyzed to obtain consistent and reliable data sets to determine component orientation relative to geographic north, as well as receiver functions and shear-wave splitting measurements. The majority of the vertical components have high signal-to-noise ratios at periods between 100 and 10 s that are comparable with those for land stations.
Fig. S1. Example of vertical-component data from a moderately large shallow earthquake. The water depth for each OBS is given on the left of the seismogram. For comparison, the data at permanent observatory instruments are also shown. The data were low-pass filtered with a -40-dB roll-off between 40 and 50 mHz and aligned on 50-s Rayleigh wave arrival times predicted by reference model PREM [Dziewonski and Anderson, 1981]. The data from the land stations were obtained from the Incorporated Research Institutions for Seismology (IRIS) Data Management Center (DMC).
Figure S1 shows example records from a moderately large earthquake, filtered to analyze Rayleigh waves. The waveforms are very coherent across the OBS network, and the record quality is virtually the same as that of PLUME land stations. In fact, waveforms are compatible with those at various permanent observatory instruments of the Global Seismographic Network (GSN), GEOFON, and GEOSCOPE. This data example is fairly typical of the records obtained for PLUME, which has facilitated the assemblage of a database that is comparable in size with that of regional experiments on land.
Fig. S2. Vertical-component spectra of 20-min-long segments before and during the earthquake shown in Figure S1. A large gap between the spectra signifies a high signal-to-noise ratio, hence a “quiet” station. For the purposes of earthquake seismology, this approach is more revealing than assembling the standard ambient noise plots, such as those published in the Federation of Digital Seismograph Networks (FDSN) station book. The water depth at each OBS station is given below its name. Colored bars beneath the middle panel mark the two noise bands, the microseism band above 0.05 Hz as well as the infragravity band below 0.01 Hz. The range in between is known as the noise notch.
Many OBSs have excellent signal-to-noise ratios between the infragravity and the microseism bands, often termed the noise notch (Figure S2). Near the infragravity end, some instruments remarkably surpass the quality of the Geotech/Teledyne KS-54000 borehole seismometer that is installed at GSN station MIDW. In fact, some of the better OBS spectra are compatible with land-observatory records on a Wielandt–Streckeisen STS-2. The quality does not quite match that of the very-broadband Wielandt–Streckeisen STS-1 at station KIP, which is one of the best very-long-period stations worldwide as far as vertical signal-to-noise levels are concerned. Also shown are the three OBSs with the lowest signal-to-noise levels. Though the band of useful signal is limited to frequencies above 0.02 Hz, these records still contain enough signal for useful analyses. The strength of the infragravity noise appears to fluctuate greatly among sites, but a simple dependence on water depth is not immediately apparent.
The Hawaiian PLUME project produced a very rich dataset of unprecedented quality for a seafloor network and provided an opportunity to observe ultra-low-frequency seismic signals. We recorded the 28 March 2005 aftershock (M0=111x1020Nm; MS=8.2) of the great 26 December 2004 Sumatra–Andaman earthquake [Laske et al., 2007]. At the quiet GSN site KIP, normal mode 0S4 (f ≈ 0.647 mHz) stands out clearly above the noise floor on the STS-2 recording. Due to the stronger infragravity noise on the ocean floor, this mode could not be expected to be seen on the OBSs. However, at least five OBSs recorded 0S8 (f ≈ 1.413 mHz) above the noise. PLUME may have been the first seafloor experiment to observe ultra-low-frequency normal modes on unburied seismometers.
PLUME also recorded numerous local earthquakes that lie outside the detection range of the land-based network of the USGS Hawaiian Volcano Observatory (HVO). The off-shore events can be found across the entire OBS network, if the data are high-pass filtered above about 5 Hz [Anchieta et al., 2007]. The seismicity patterns are currently being analyzed.
The initial assessment of the DPG data is also very favorable. In many cases, signal-to-noise levels on the DPGs are compatible with those on the vertical seismometer components. Some two-station surface wave dispersion curves obtained from DPG records are internally more consistent than from the corresponding seismometer records.
References
Anchieta, M. C., C. J. Wolfe, G. Laske, J. A. Collins, S. C. Solomon, R. S. Detrick, J. A. Orcutt, D. A. Bercovici, E. H. Hauri, G. L. Pavlis, J. A. Eakins, and F. L. Vernon, (2007), Characterizing offshore earthquakes at Hawaii recorded by the first PLUME temporary ocean–bottom seismometer network, Eos Trans. AGU, 88 (52), Fall Meeting suppl., abstract V33B–1383.
Dziewonski, A. M., and D. L. Anderson, (1981), Preliminary reference Earth model, Phys. Earth Planet. Inter., 25, 297–356.
Laske, G., J. A. Orcutt, J. A. Collins, R. S. Detrick, C. J. Wolfe, S. C. Solomon, D. A. Bercovici, and E. H. Hauri, (2007), Broadband ocean bottom instruments record Earth’s free oscillations during the Hawaiian PLUME experiment. Eos Trans. AGU, 88 (52), Fall Meeting suppl., abstract S23A–1107
Author Information
Gabi Laske, Scripps Institution of Oceanography (SIO), La Jolla, Calif.; E-mail: glaske@ucsd.edu; John A. Collins, Woods Hole Oceanographic Institution (WHOI), Woods Hole, Mass.; Cecily J. Wolfe, University of Hawaii at Manoa, Honolulu, Hawaii; Sean C. Solomon, Carnegie Institution of Washington (CIW), Washington, D.C.; Robert S. Detrick, WHOI; John A. Orcutt, SIO; David Bercovici, Yale University, New Haven, Conn.; and Erik H. Hauri, CIW