Download The Monterey ocean bottom broadband station (MOBB)

Instrument Testing and First Results from the MOBB Observatory
Testing
First Results
The Monterey ocean bottom broadband station
(MOBB) was installed on the sea floor in Monterey
Bay, 40km offshore, and at a depth of 1000m from the
sea surface, in April 2002. It is a collaborative effort
between MBARI (Monterey Bay Aquarium Research
Institute) and BSL (Berkeley Seismological
Laboratory).
Tide (counts)
15
MMZ
10
-100
5
0
-120
-5
Tilt (counts)
0
MMN
-5000
-10000
-140
-15000
-160
-20000
-25000
SAO.BK.BHZ
MOBB.BK.BHN
FARB.BK.BHN
JRSC.BK.BHN
SAO.BK.BHN
MOBB.BK.BHE
FARB.BK.BHE
JRSC.BK.BHE
0
Figure 1. The CMG-1T OBS installed in the
titanium pressure vessel. The space between
the inside cylindrical surface of the pressure
vessel and the seismometer is insulated with
multiple layers of Mylar (space blanket)
insulation supported by a cardboard form.
Partly visible on the upper right is the Mylar
covered urethane foam plug which fills the void
between the top of the seismometer and the
bottom of the top end cap of the pressure
vessel (to provide insulation as well as to
reduce the free air volume within the vessel).
Figure 2. Titanium pressure vessel
being purged with dry argon gas.
Visible are the hose used to purge the
air from the pressure vessel and top part
of the Mylar insulation which, along with
a 2 inch thick urethane foam disk, is
used to insulate the top of the
seismometer. By insulating the top and
side of the cylindrical pressure vessel,
but not the bottom, the ~2.2 Watt power
dissipation of the sensor produces a
stable stratification of the argon gas in
the pressure vessel and inhibits the
convective air currents which were
generating a high background noise
PSD on the CMG-1T vertical component
as shown in Figure 3.
MME
-180
-5000
SAO.BK.BHE
-200
-15000
0
0.2
0.5
1
2
5
10
20
50
10
20
100
30
Time (minutes)
Period (sec)
-20000
-25000
100
110
120
130
140
150
160
170
Figure 9. Comparison of 10-100
second band pass filtered absolute
ground velocities recorded at MOBB
and nearby BDSN stations FARB,
JRSC and SAO (see Figure 6) from a
Mw 7.1 teleseism which occured 84.8°
N78°W of MOBB on 26 April 2002 in the
Mariana Islands. The scale for the Ncomponents has been reduced by half.
The comparison shows consistency
between the recording at MOBB and
nearby stations. On the horizontal
components there appears to be some
signal-generated noise following the Swaves and the Love wave, which is
likely associated with ringing in the
shallow mud layers.
180
Day of 2002
Figure 3. Comparison of the background noise PSD
from the CMG-1T OBS, installed in the titanium pressure
vessel, and from the co-sited STS-1 sensors housed in
the BKS Vault. Shown are the PSD’s for the CMG-1T
vertical (large dashed), the CMG-1T horizontals (solid),
and the STS-1’s (small dashed). The colored PSD’s are
for data collected prior (red) and after (blue) installing
the insulation and purging with dry argon gas. The
CMG-1T and STS-1 PSD’s are within ~6 dB of each
other after insulating and purging with argon. That the
CMG-1T and STS-1 PSD track each other over a wide
range of periods is also indicates that the CMG-1T
calibration and transfer functions are correct.
Our second goal is to systematically analyze
body wave and surface wave data for regional
earthquakes to obtain constraints on the
moment tensor solutions and 3D crustal
structure at the western edge of the plate
boundary.
-10000
Figure 7. Guralp CMG-1T OBS mass position data
for the time period 04/11/02-06/27/02. The large
steps are associated with: 1) installation (day 100);
2) recentering (day 112), and a smaller step with a
local Mw 4.95 earthquake on day 134. The tide
signal is clearly visible on the vertical component.
The data indicate that the seismometer package has
been experiencing an exponentially decaying tilt in a
south-southwesterly direction, which is also the
down slope direction (e.g. Figure 5).
Figure 8. Observed background noise PSD at
MOBB. Shown are the Z-component (circles),
N-component (stars) and E-component
(triangles). The USGS high- and low-noise
models (solid lines) are shown for comparison.
The increase in the noise level at periods
longer than ~20 seconds is probably due to the
ocean currents interacting with the highly
compliant seafloor sediments.
Tangential
Radial
MOBB is the first step towards extending the
land-based broadband network in northern
California off-shore to better characterize the
seismicity, tectonics and structure of this
region.
Vertical
Strike=51 ; 142
15.00 sec
MHC Max Amp=8.02e-05 cm
VR=88.2
Deployment
30.00 sec
Mo
=3.86e+21
Mw
=3.7
0˚
0˚
Percent DC=82
BDM Max Amp=3.35e-05 cm
VR=80.8
CVS
Rake =-18 ; -178
Dip =88 ; 72
38˚ 30'
33
Percent CLVD=18
BKS Max Amp=3.71e-05 cm
VR=95.8
24 cm/s
Var. Red=9.10e+01
30.00 sec
P
0˚
CMB Max Amp=2.27e-05 cm
VR=77.0
A)
30
MOBB Max Amp=2.31e-04 cm
VR=92.8
12 cm/s
T
0.4
Strike=44 ; 135
Dip =85 ; 78
30.00 sec
BKS Max Amp=3.71e-05 cm
VR=91.1
B)
36˚ 40'N
122˚ 00'W
121˚ 50'W
km
0
36˚ 00'
-123˚ 30'
50 km
-123˚ 00'
50
-122˚ 30'
2.2
1
T
MOBB Max Amp=2.31e-04 cm
VR=94.6
-122˚ 00'
-121˚ 30'
21
0˚
0˚
15
180˚
36˚ 30'N
121˚ 40'W
Figure 5. Location of MOBB and its
predecessor MOISE in Monterey Bay,
California (Stakes et al., 1998), against
seafloor and land topography. Fault lines
are from the California Geological Survey
database. MOBB is located at 1000 m
below sea-level.
P
SAO
MOBB
36˚ 30'
122˚ 10'W
0˚
30.00 sec
0.4
1
CMB Max Amp=2.27e-05 cm
VR=-3.0
10 km
122˚ 20'W
1.6
Variance=6.44e-11
Var. Red=9.46e+01
15.00 sec
122˚ 30'W
3.4
4
24
BDSN sites
37˚ 00'
MOBB
=3.8
Percent ISO=0
30.00 sec
MOBB location
MOISE
Mw
2.8
2
1.6 2.
Percent CLVD=55
SCCB
36˚ 50'N
=4.96e+21
Percent DC=45
BDM Max Amp=3.35e-05 cm
VR=71.3
MHC
Mo
12
JRSC
Rake =-12 ; -174
1.6
37˚ 30'
15.00 sec
2.2
MHC Max Amp=8.02e-05 cm
VR=40.9
1
90˚
WENL
1.6
37˚ 00'N
2.8 3.4
Vertical
270˚
FARB
Radial
-121˚ 00'
Figure 6. Location of MOBB in relation to nearby
Berkeley Digital Seismic Network (BDSN) stations in
the Central Coast Ranges. Also shown are the
Holocene faults and the background seismicity.
FARB is located on the Farallon Islands. MOBB is
just west of the Holocene trace of the offshore
Palo Colorado-San Gregorio fault.
Figure 10. Deconvolved ground velocity
records at MOBB of the 04/23/02 M 3.63 San
Andreas fault event (53.4 km N68°E of MOBB).
The very large S wave pulse on the horizontal
components as well as the subsequent ringing
are likely due to site response and need to be
investigated further. On the vertical
component, the water reflection of the P wave
is clearly seen 1.5 seconds after the P wave.
In spite of these strong site effects, these data
can be used in moment tensor studies as
illustrated in Figure 11.
Ultimately the experience gained through
MOBB can help design better future near-offshore ocean floor broadband seismic station
deployments to complement the land based
networks in the western US, either
permanently, or in the intermediate time scale
(1-2 years), for example in complement to
such programs as the USArray of Earthscope.
Already, the MOBB instrument preparation
and deployment experience acquired over the
last two years will benefit similar experiments
planned in the context of preparation for the
Keck program.
0˚
BKS
BRK BRIB
15.00 sec
BDM
0.4
37˚ 10'N
18 cm/s
˚
38˚ 00'
Figure 4. Installation of the seismometer
package inside the PVC caisson. The top of the
package is buried at least 10 cm below the the
seafloor level. Burial of the seismometer
package in the seafloor sediments is crucial to
reducing the tilt noise induced by ocean currents.
The package was later completely covered with
tiny (0.8 mm) glass beads to further isolate it
from the effects of water circulation.
˚
Variance=3.22e-11
POTR
The installation was completed during three dives (911 April, 2002), with the help of the MBARI ROV
(Remote Operated Vehicle) Ventana and ship Point
Lobos. The site was revisited on 22 April, to check
the functioning of the system and 3 Mb of data were
then retrieved. The ship and ROV returned to the site
two months later, on 27 June, and the data recording
and battery modules were replaced, in the first of a
series of such dives planned over the next three years.
30
30 cm/s
Percent ISO=0
30.00 sec
Tangential
Many regional and teleseismic earthquakes have
been well recorded and the mass position signals
indicate that the instruments are progressively settling.
Preliminary analysis of the data retrieved during the
2002 dives are presented.
Our first goal is to access the data quality and
possible improvements, through postprocessing and/or installation adjustments.
We plan to evaluate the long term time
evolution of background noise, as the system
continues to settle and stabilize, and the
shorter term noise fluctuations in relation to
tides and currents as recorded by the currentmeter as well as the Digital Pressure Gauge
(DPG).
60
Because of the extreme sensitivity of the seismometer,
air movement within the pressure vessel must be
minimized. We describe extensive testing and
insulation procedures performed at BSL. Among
others, the top of the pressure vessel was thermally
insulated with two inches of insulating foam and
reflective Mylar. The sides were then insulated with
multiple layers of reflective Mylar space blanket, and
the vessel was filled with argon gas to inhibit
convection.
JRSC.BK.BHZ
Tilt (counts)
The Ocean-bottom MOBB station currently comprises
a three-component seismometer package, a currentmeter, and a recording and battery package. A
differential pressure gauge (DPG) with autonomous
recording (e.g. Cox et al., 1984) was deployed in the
vicinity of the seismometer package during the data
recovery dive in September 2002. The seismic
package contains a low-power (2.2W), threecomponent CMG-1T broadband seismometer system,
built by Guralp Systems, Ltd., with a three-component
24-bit digitizer, a leveling system, and a precision
clock. The seismometer package is mounted in a
cylindrical titanium pressure vessel 54 cm in height
and 41 cm in diameter, custom built by the MBARI
team and outfitted for underwater connection.
FARB.BK.BHZ
-80
AGU 2002 Fall Meeting
Poster S71A-1048
MCC Hall C
Sunday 8 December 0830-1700
Future Plans etc.
MOBB.BK.BHZ
10-100 sec BP Ground Velocity
Introduction
R. Uhrhammer, B. Romanowicz, D. Neuhauser, D. Stakes, P. McGill, T. Ramirez
PSD (dB)
Berkeley Seismological Laboratory
215 McCone Hall #4760
Berkeley, CA 94720-4760
Phone: 1.510.642.8504
Email: [email protected]
Figure 11. Results of moment tensor
inversions for the M 3.63 regional event
shown in Figure 10. Top: inversion using
4 stations of the BDSN and MOBB (BDM,
BKS, CMB are band passed between 0.
02 and 0.05 Hz; MHC and MOBB,
between 0.05 and 0.1 Hz). Bottom:
results of inversion using only MOBB,
showning the good fits of the single
station solution to the other BDSN data.
Figure 12. Distribution of current velocity
data as a function of azimuth. The contour
label units are fractions of the average
density distribution of the current. The two
dominant maxima (centered at 60 and 240
degrees, i.e. orthogonal to the continental
shelf) are associated with the semi-diurnal
tidal currents. The third directional peak is
roughly parallel to the coastline and appears
to be associated with the dominant ocean
circulation.
A significant contribution to this project is
provided by MBARI through the continuing
support of the operation and maintenance of
the MOBB site (at the minimum: exchange of
data loggers, checking and recentering the
seismometers and timing adjustments, every
three months for the next three years), and by
BSL, through the systematic archival of the
data at the NCEDC, at no cost to the project.