USGS - science for a changing world

Woods Hole Coastal and Marine Science Center

Skip Navigation
OBS Home What is an OBS? Specifications Cruise Reports Related Websites

Cascadia Region Seismic Experiment

Cruise Report for R/V Sonne cruise SO108

An OBS awaiting Recovery.


The main goal of The R/V Sonne cruise SO108 was to investigate the large-scale structure of the Cascadia subduction zone to better assess the earthquake hazard it presents to the Pacific Northwest. The cruise was carried out in April and May of 1996 cooperatively by GEOMAR, Kiel, Germany and the USGS, Marine Program, Menlo Park and Woods Hole. Our ocean bottom seismometers (OBS) and the German hydrophones (OBH) were deployed at 116 stations and collected wide-angle reflection and refraction seismic data along 12 profiles across and along the continental margin off Washington and Central Oregon and on the adjacent ocean floor of the Juan de Fuca plate. 1532 km of multichannel seismic reflection (MCS) data along 13 lines were collected over the Washington continental margin with two of the lines being coincident with wide-angle seismic profiles. Additionally, the largest earthquake to hit Seattle in the last 15 years was recorded by our OBS and OBH. All the wide-angle seismic data and two of the MCS lines have been processed, plotted, and analyzed on board for quality control and for initial scientific interpretations. The cruise ran smoothly and efficiently with the exception of a major storm and the loss of one OBS in the first week of operations. Data quality is good to excellent.

Both wide-angle reflection and refraction data and the MCS data could detect the subducting oceanic crust under the continental margin to a distance of 20 km from shore. Three major differences between the continental margins of Washington and Oregon were discovered. The Oregon margin is much narrower than the Washington margin but the depth of the bottom of the subducting plate near the coast is similar in both areas (18-20 km), hence the dip of the subducting plate is higher under Oregon than under Washington. The dip of the subducting plate may also decrease from south to north under the Washington margin. The second major difference is the average velocity of the sedimentary and crustal rocks overlying the subducting plate. Offshore Washington, the rock velocity never exceeds 5.3 km/s even at depths as large as 10-12 km. Offshore Oregon, the rock velocity is up to 6.5 km/s and is 5 km/s or larger below 5 km depth. These differences represent differences in crustal composition. In Oregon, the Siletz River basalts are thought to extend offshore. In Washington, crustal rocks are probably broken and fractured Eocene to Middle Miocene melange. The boundary between these crustal domain is probably south of the mouth of the Columbia River where the sediment and upper crustal structure change dramatically. The third difference is in the thickness of sediments underlying the continental shelves. Offshore Oregon sediment thickness does not exceed 3 km, whereas offshore Washington a deep (up to 8 km) partly fault-controlled elongate basin underlies the shelf.

The MCS sections show that thrust faults within the accretionary wedge extend downward to the level of the igneous oceanic crust, so the entire 3-km thickness of sedimentary rocks on the ocean plate is being accreted. The frontal thrust verges landward and locally returns a high amplitude reflection, which probably reveals focused flow of fluids. Major thrust faults in the outer 40 km of this wedge verge landward, and at their landward limit, they stack fault blocks of accreted sediment against what may be a backstop of older, indurated rock.


Nine US Geological Survey Ocean Bottom Seismometers (OBS) were used aboard the RV Sonne cruise SO108 offshore Oregon and Washington to record wide-angle seismic reflection and refraction information. Each OBS is housed in a 24-inch aluminum sphere ( Figure1) and contains data-acquisition system designed to record continuous data from a vertical geophone, 2 horizontal geophones, and a pressure compensated hydrophone. The OBS is designed to free-fall to the ocean bottom with an anchor plate attached and to stay at depths of up to 5000 meters for up to 10 days. An acoustic release is incorporated in the instrument to allow release from its anchor upon command from the surface, permitting the buoyant OBS sphere to ascend. Once on the surface, a light strobe and a radio beacon help locate the instrument.

Cruise Report for R/V Sonne cruise SO108 Deployment Information

The experiment included 5 deployments of 7-8 instruments each :

Deployment 1 - Lines 1, 2, 4, 5, 6, 102, 506 - Rosette

#OBS Latitude Longitude Depth BTime BShift ETime EShift Drift TT#
A2 44 30.036 126 28.655 2898 96:04:22:01:27 4.5 - - - -
A4 - - - - - - - - -
A1 44 30.004 126 21.582 2889 96:04:22:01:56 6.0 96:04:26:06:48 0.9 -5.1 A1
C3 44 27.513 126 22.012 2901 96:04:22:01:47 2.2 96:04:26:07:55 -3.9 -6.1 C3
A3 44 28.187 126 14.666 2898 96:04:22:02:00 -0.3 96:04:26:09:10 -0.2 0.1 A3
C9 44 34.100 126 14.337 2871 96:04:22:01:32 -1.5 96:04:26:10:18 -5.8 -4.3 C9
C1 44 30.867 126 09.043 2895 96:04:22:01:54 3.1 96:04:26:12:23 1.5 -1.6 C1
C4 44 35.915 126 06.358 2892 96:04:22:01:59 3.9 96:04:26:13:42 2.3 -1.6 C4
A8 44 41.123 126 06.673 2871 96:04:22:01:41 8.7 96:04:26:14:28 20.8 12.1 A8

Deployment 2 - Lines 7, 8, and 9 - Oregon E-W, N-S coastal, N-S seaward

#OBS Latitude Longitude Depth BTime BShift ETime EShift Drift TT#
A4 44 38.276 124 41.798 234 96:04:26:22:46 2.1 96:05:01:12:30 -2.1 -4.2 C4
C1 44 38.343 124 35.990 193 96:04:26:22:50 1.4 96:05:01:12:46 0.8 -0.6 C1
A3 44 38.353 124 30.195 127 96:04:26:23:02 -0.7 96:04:29:07:07 -1.1 -0.4 A3
A8 44 38.439 124 23.957 75 96:04:26:22:39 2.3 96:04:29:08:09 4.8 2.5 A8
A1 44 38.343 124 17.911 79 96:04:26:23:12 0.3 96:04:29:09:15 -2.4 -2.7 A1
C9 44 38.279 124 11.407 59 96:04:26:22:56 3.8 96:04:29:10:34 1.2 -2.6 C9
C4 44 38.302 124 27.129 100 96:04:29:00:21 2.7 96:05:01:19:59 -1.9 -4.7 C3

Deployment 3 - Line 10 - SW Washington

#OBS Latitude Longitude Depth BTime BShift ETime EShift Drift TT#
A4 46 38.987 125 38.931 1989 96:05:02:02:28 5.4 96:05:04:11:18 5.8 0.4 C3
C1 46 38.987 125 31.588 2129 96:05:02:02:31 0.6 96:05:04:12:54 0.8 0.2 C1
C4 46 38.992 125 24.936 1913 96:05:02:02:34 -2.2 96:05:04:13:54 -1.3 0.9 C4
C9 46 38.981 124 29.004 108 96:05:02:02:40 1.3 96:05:05:08:28 0.9 -0.4 C9
A1 46 38.992 124 24.003 84 96:05:02:15:25 8.4 96:05:05:06:46 8.1 -0.3 A1
A8 46 38.984 124 18.995 68 96:05:02:15:30 8.1 96:05:05:07:24 14.1 6.0 A8
A3 46 38.996 124 14.028 48 96:05:02:15:36 -2.1 96:05:05:05:17 -2.7 -0.6 A3

Deployment 4 - Line 11 - Washington N-S

#OBS Latitude Longitude Depth BTime BShift ETime EShift Drift TT#
A4 45 50.969 124 10.285 99 96:05:04:21:04 3.7 96:05:07:19:19 1.1 -2.6 C3
C1 45 56.586 124 13.036 105 96:05:04:21:08 0.8 96:05:07:20:08 1.1 0.3 C1
C4 46 03.282 124 16.698 105 96:05:04:21:14 -1.4 96:05:07:21:33 -0.9 -0.5 C4
C9 47 42.495 125 05.971 376 96:05:05:08:45 0.9 96:05:08:10:50 0.6 -0.3 C9
A8 47 49.939 125 09.787 353 96:05:05:07:52 0.5 96:05:08:12:03 14.4 13.9 A8
A1 47 56.335 125 13.045 176 96:05:05:07:01 8.2 96:05:08:13:08 7.6 -0.6 A1
A3 48 03.297 125 16.574 150 96:05:05:06:21 -2.7 96:05:08:14:06 -3.3 -0.6 A3

Deployment 5 - Line 12 - Olympic

#OBS Latitude Longitude Depth BTime BShift ETime EShift Drift TT#
A1 47 34.343 124 39.949 67 96:05:08:13:25 7.6 96:05:10:16:07 7.1 -0.5 A1
A3 47 33.336 124 45.195 93 96:05:08:15:23 -3.3 96:05:10:16:15 (-3.9) -0.6 A3
A8 47 32.463 124 49.746 121 96:05:08:12:33 4.5 96:05:10:16:58 1.7 -2.8 A8
C1 47 31.581 124 54.815 171 96:05:08:01:24 0.6 NO DATA - - A2
C4 47 26.262 125 22.896 1492 96:05:07:21:52 -0.9 96:05:10:21:45 -0.5 0.4 C4
C9 47 25.271 125 28.287 1484 96:05:08:11:07 0.6 96:05:10:23:01 0.1 -0.5 C9
A4 47 24.247 125 33.942 1478 96:05:07:19:39 1.1 96:05:10:23:49 -1.8 -2.9 C3

GMT = Data logger_time + Shift ; A negative Shift means GPS clock pulse behind (is later than) Data logger Time in UTC (7 hours ahead of local time).
Drift = (EShift - BShift) in milliseconds.
Correction_Shift = BShift + [Drift * (Orig_Starttime - Btime)] in milliseconds.
Corrected_Trace_Starttime = Orig_Starttime + Correction_shift.
The value of Correction_shift for each SEGY trace is (added to the value) stored in the trace header under Total_Static

During the Rosette and the Oregon deployments, hence a total of 75 records were generated. All instruments recorded 4 channels on all the lines except for 6 instruments on the Oregon deployment, which were programmed to record only channels 1 and 4. Sadly, the first OBS deployed in this experiment has not been recovered, and this represents a real loss to the USGS OBS program. Another instrument did not record data on one deployment.

Data quality were good on at least one channel in each instrument in each deployment . Several sedimentary and crustal arrivals can be identified on all but 4 OBS records from shallow waters.


The instruments were brought into the seismic laboratory approximately 4 hours prior to deployment where the battery power was connected, all components of the system were tested, the data logger were programmed, and the spheres were sealed and vacuumed. Deployment was done through the starboard-side crane at mid-ship. A small stand was set on deck and the anchor was placed on top of it. The OBS was placed on top of the anchor and the two were bolted together. During later deployments, a floating ("swimming") line with an end-float was wrapped tightly around the base of the sphere, and the end of the floating line was tucked between the base of the sphere and the anchor. Once on location, the OBS and its anchor were lifted up by the crane and into the water where it was released using a slip knot and a pin.

Recovery was done using the same starboard-side crane which lifted the OBS on board. A tag line was attached to the OBS with a Happy Hooker or a lifting hook. In later deployments, the floating line was caught by throwing a rope with a hook, and dragging the floating line with the OBS closer to the side of the ship. The use of the floating line shortened considerably the time needed for the 3000-ton ship to maneuver alongside the OBS. The OBS was lifted to deck, was washed in fresh water, and carried back to the lab, where the post-deployment checks were done. The data logger was brought to the lower-deck lab, checked for time drift against the GPS clock, and attached to the computer. While the data, recorded on the data logger, were being downloaded, new batteries were installed into the OBS in anticipation of redeployment at a later time. The sphere was then closed and tied down to the main deck. Once the data was recovered, the data logger was placed on a charging station to recharge the gelcell battery and to keep the oven-controlled oscillator at operating temperature. Some data loggers were programmed and immediately placed back into their sphere, because of the short time between recovery and the next deployment.


Step 1 - Data were down loaded from the data logger to a PC computer, one file per data logger. The raw data were written on one 8 mm Exabyte tape for each deployment, and a duplicate tape of the raw data was made.
Step 2 - Shot instance and navigation data, supplied by GEOMAR, were converted from UKO format to our format.
Step 3 - OBS location and depth from our log were entered to NAVD file, and *.src files with trace header information were created for each OBS for each seismic line.
Step 4 - *.src files were used in program OBSTOOLS to create SEGY files from the raw data. Specifically, the header file was edited to give the length of each trace, the project number and location, OBS name etc. The program uses shot instant from the *.src files to determine the start of each trace, and the difference between shot location and the OBS location to determine shot-receiver offset. SEGY files for each deployment were written as a SEGY binary tar tape.
Step 5 - The SEGY tapes were read into Promax processing software and processed. The processing sequence included:

1. tar tape into UNIX workstation hard disk.
2. Read SEGY disk input file..
3. Spiking deconvolution with an operator length of 80 ms..
4. Bandpass filter (Butterworth) - 4-7-18-32 Hz..
5. 2-D spatial filtering (not used in this cruise because of poor navigation)..
6. Trace equalization (not used in this cruise)..
7. Linear moveout correction (6 km/sec)..
8. Automatic gain control with a window of 2000 msec..
9. Trace header math with the equation.
cdp = offset in meters + no. in meters; where no. > total length of line..
Trace header math with Line no. and OBS no..
10. Plot data by cdp (effectively by offset)..

Because a plotter was not connected to the SUN workstation with the Promax system during cruise SO108, data were written to a DAT tape in SEGY format, read by a different SUN workstation into GEOMAR`s GEOSYS software and plotted by D. Klaeschen using this software.

Steps 1-4 were performed on a 386-PC. Step 5 was performed on a SUN Sparc-10 workstation on board.


Predeployment - In initial testing, an oscillator on one data logger failed, prohibiting one OBS from being deployed during each of the subsequent deployments. That OBS was used for spare parts during the experiment.

Deployment 1 - 8 OBS were deployed along 6 lines in a rosette shape. Only 5 lines were shot due to severe weather. The gun arrays were battered and we were forced to stop shooting twice. Guns operated at partial capacity for periods of time. One OBS (A2) could not be ranged and was never recovered and is either at the bottom or floating. 3 hours at night and 4 hours in daylight were spent looking for it. >40 release commands were sent from 4 azimuths and the ship searched in a ~4 x 4 n.m. area around the deployment location. It was the first OBS deployed in the experiment. Weather during shooting was stormy with winds of up to 50 knots and 8-12 ft. waves. Windy but calm seas during deployment and calm weather during recovery.

Deployment 2 - Oregon lines - 6 OBS were deployed at intervals of 4 n.m. along the landward end of the E-W line (Line 7). Calm seas and light wind during deployment, shooting, and recovery. Guns fired at almost full capacity. Ship navigation froze during deployment as we approached shore and at the 3-mile limit we were 2.5 miles off, hence, we did not deploy the planned seventh instrument. Exact positions of instruments were found during recovery. 4 instruments were recovered, 2 instruments near the junction with Line 9 were left at the bottom after locating and ranging on them. The instrument, not deployed on the E-W line, was deployed at the crossing with Line 8. The remaining 3 OBS were recovered without incident after the shootings of Lines 8 and 9.

Deployment 3 - SW Washington line - 3 OBS were deployed in the lower continental slope and 4 instruments were deployed along the landward end of the line. The 3 instruments in the lower continental slope were recovered before we left the line to meet a supply boat. The remaining 4 OBS were recovered when the ship reached the area again after deploying some OBS and OBH along the N-S Washington line. All guns fired. Calm seas and light wind during deployment, shooting, and recovery.

Deployment 4 - From the meeting point with the supply boat off Astoria, we continued south to the southern end of the N-S line and deployed our 3 OBS already on board. The remaining OBS, picked from the SW Washington line during the deployment, were deployed at the northern end of the line. Deployment, shooting, and recovery was carried out amidst fishing boats and crab buoys. Two of the guns (a 580 and a 500 ceased shortly after beginning of shooting so we had only 5200 of air for shooting. The northern end of the line was relocated slightly farther offshore at the shelf edge to avoid deploying in waters that are too shallow.

Deployment 5 - Olympic Line- Upon the recovery of the last OBS from Deployment 4, the ship sailed to the eastern end of the Olympic Line, and within 3 hours, the first OBS was deployed. 3 more OBS were deployed in shallow waters (67-171 m), and 3 OBS were deployed in a flat section of the continental slope. The line was shot at full gun capacity. Recovery of the OBS started from the eastern end of the line. The sea was very calm during deployment, shooting, and recovery.


Cruise Participants

Photograph of the research vessel RV Sonne.
RV Sonne


Cruise Data

Deployment Map

Record Sections for Line1

Record Sections for Line2

Record Sections for Line4

Record Sections for Line5

Record Sections for Line7

Record Sections for Line8

Record Sections for Line 9

Record Sections for Line10

Record Sections for Line11

Record Sections for Line 12b


Deployment 1

1. A2 not recovered. 3 hours at night and 4 hours in daylight spent looking for it. >40 release commands sent from 4 azimuths. No clear response from ranges or releases. It was the first OBS deployed in the experiment.
2. C4 strobe leaked, strobe damaged and replaced with fresh unit. This delayed sighting upon recovery.

Deployment 2

1. On recovery C4 geophone pack found to be knocked off gimbals, reassembled.
2. C9 vertical geophone wire broken in sensor pack, found and fixed after recovery. Channel known to be bad on deployment but no time to fix it. Data QC does not indicate a problem.
3. After recovery, A1 hydrophone found to be wired to RS-232 in upper sphere. Fixed.
4. A4 no signal from hydrophone at preamp board. Fixed, FM.
5. C3 could not be deployed during deployment 2-5 because of a bad oscillator.

Deployment 3

1. A8 vertical examined with scope for any noise problem. No problem found.
2. A1 low signal at data logger input for horizontal 1. No repair attempted.

Deployment 4

1. A8 vertical still bad. Swapping sensor pack with C3 during deployment 5 remedied the problem. Problem appears to be with geophone pack. Probably the vertical geophone has a damaged suspension and must be replaced.
2. A1 has weak hydro and low horz1.

Deployment 5

1. C1 with datalogger from A2 did not record anything. It was in acquisition mode but the time was way off and no tracks were recorded. Probably the clock was glitched installing datalogger into sphere.
2. A3 lost time reference at recovery or more likely during a frequency check. Estimate drift at +0.3 msecs on the basis of previous deployments.


Trace header format on SEGY files of all OBS files are as follows:

Byte position 1 = trace # ( a number from 1 to the # traces in the file).
Byte position 5 = trace # as above.
Byte position 9 = channel # (1 = vertical geophone, 2 and 3 = horizontal geophones, and 4 = hydrophone).
Byte position 13 = shot # (from navigation data).
Byte position 29 = trace id code ( 1 for seismic data)
Byte position 31 = # of vertically stacked traces ( 1 for 1 trace)
Byte position 33 = # of horizontally stacked traces ( 1 for 1 trace)
Byte position 35 = data use ( 1 for production)
Byte position 37 = Shot-receiver offset (Offsets are negative for all shots occurring south/west of each OBS location and positive for all shots occurring north/east of each OBS location)
Byte position 41 = receiver elevation (negative value of the OBS depth)
Byte position 65 = water depth of receiver (OBS depth)
Byte position 73 = source longitude (in seconds)
Byte position 77 = source latitude (in seconds)
Byte position 81 = OBS longitude (in seconds)
Byte position 85 = OBS latitude (in seconds)
Byte position 89 = coordinate units ( 2 for seconds of arc)
Byte position 109 = delay recording time ( 0 in this case)
Byte position 115 = # samples in this trace
Byte position 117 = sample interval (in microsec, 10,000)
Byte position 119 = gain type ( 1 for fixed)
Byte position 125 = correlated ( 1 for not correlated)
Byte position 141 = alias filter frequency (in hertz)
Byte position 143 = alias filter slope (dB/octave)
Byte position 157 = year of shot
Byte position 159 = day of year (Julian)
Byte position 161 = hour of day
Byte position 163 = minute
Byte position 165 = second
Byte position 167 = time basis code ( 2 for GMT - GMT used for all time references)
Byte position 169 = trace weighting factor (data values are recorded with the least significant bit corresponding to 10`s of microvolts for a weight factor of 14)
Byte position 181 = milliseconds of the shot time

User provided:

Byte position 221 = Line no. (4 Byte integer)
Byte position 225 = OBS no. (4 Byte integer), Examples: A8 = 18; C4 = 34.


Skip Navigation

Accessibility FOIA Privacy Policies and Notices

USAGov logo U.S. Department of the Interior | U.S. Geological Survey

Page Contact Information: WHSC Webmaster
This page last modified on