CAMBRIDGE
DEEP
OCEAN
GEOPHONE
JEREMY DUSCHENES, CHRIS POTTS, and MARTYN RAYNER lBullard Laboratories, Department of Earth Sciences, University of Cambridge, Madingley Road, Cambridge CB3 0EZ, U.K.
(Accepted 5 December, 1984)
Abstract. We describe recent mechanical and electronic modifications to the Cambridge Ocean Bottom Hydrophone system, enabling il to record in addition three geophone channels from a deployed, disposable geophone package. Examples of data from seismic refraction experiments show good correspondence between records of ground motion detected by the hydrophone and the vertical geophone. Seismic signals are undistorted by noise from instrument related sources. Clear examples of P to S conversions just below the receiver are observed. Improved recording conditions are achieved by deployingthe geoI:fl~onesin a small pressure vessel as far away as possible from the main instrument package.
1. Introduction The problem of coupling seismometers to the seabed is one which has not always been properly dealt with in attempts to build ocean bottom seismographs (OBSs). As a result, the expected phase correlation between simultaneously recorded hydrophone and vertical geophone records has rarely been seen. Of the two the geophone records have seemed more suspect because true ground motion has often been overwhelmed by cross-coupled instrument motion. Hydrophones are not subject to this problem since they are less sensitive to cross-coupling. They couple well to ground motion via the water and have until recently produced the only records suitable for frequency content and waveform analysis. The Cambridge Ocean Bottom H y d r o p h o n e (COBH), described by Sinha el al. (1981), was designed for short term refraction experiments. It had a single hydrophone as a seismic sensor. Described below are the modifications made to the instrument which resulted in its being able to record three high quality orthogonal g e o p h o n e channels as well as a hydrophone channel. The principal change was the addition of an externally deployed geophone package, similar to that used in the M I T OBS (Duschenes et al., 1981) and elsewhere, Avedik et al. (1978), Byrne et al. (1983). The arguments for using deployed geophones were strengthened by the Lopez Island Experiment, the results of which are collected in a special issue of M a r i n e Geophysical R e s e a r c h e s (Sutton et al., 1981; Lewis and Tuthill, 1981; Trehu and Solomon, 1981; Zelikowitz and Prothero, 1981). The modified instrument is shown in Figure 1. Marine Geophysical Researches 7 (1985) 455-466. @ 1985 by D. Reidel Publishing Company.
456
JEREMY DUSCH1ENESET AL.
ReleaseTransducers%
StrayLine
SyntacticFoam)
& Float
1 I
Hydrophone ~Maln PressureCase
_,,o.o.
Flag
2
Release CableCutter
~
SurfaceFloat
(~AnchorRope
y
Gtophone PressureCase DeploymentArm Anchor
~GeophoneCable .//
E2
Fig. 1. A schematic of the entire instrument. Not to scale.
2. The Instrument The bottom weight underwent the most radical transformation since the geophones had to be in direct contact with the seafloor. The design philosophy was that nothing coming into direct contact with the sea-bed would be retrieved at the end of the experiment. This would improve the recovery rate but made it necessary to keep the bottom assembly as cheap and as simple as possible. T h e C O B H anchor had been a clump of loose chain; the Deep Ocean G e o p h o n e or D O G , as this new instrument came to be called, uses a block of cement, approximately 3 0 c m on a side, containing the chain. It weighs approximately 80 kg in water. The geophone pressure case is 19.5 cm long, as an inner diameter of 18 cm and a wall thickness of 2.5 cm. Made of steel, it has a depth rating of 4000 m with a safety
CAMBRIDGE
DEEP
OCEAN
GEOPHONE
457
Fig. 2. The disposable 3-component geophone gimbal showing the position of the pre-amplifier printed circuit board. The pre-amp is powered by four 1.5 V batteries. The PVC block containing the geophones is free to turn inside the hoop, which itself can rotate freely, thus allowing the geophones to settle in their correct operating positions.
factor of approxmiately 1.35. The end caps are 4.5 cm thick aluminium plates with O-ring face seals, and are held in place by three bolts. A square lug is welded to the side of the pressure case, allowing it to be attached to its deployer. The three geophones are held orthogonally in a self-righting PVC and aluminium gimbal (Figure 2) modified from a design by Ray Davis of the USGS/Woods Hole (pers. comm.). The geophones which are made by Sensor (model S-6, natural frequency 4.5 Hz) are sealed with silicon rubber and immersed in a polybutane oil of high viscosity (approx. 2 • 105 cSt i.e. 0.2 m 2 s I at 5 ~ which couples the gimbal to the pressure case for all frequencies greater than 1/15Hz. A three channel x 25 pre-amplifier with internal batteries is bolted to the interior of one of the endcaps. The 3 signals plus ground (the pressure case is grounded to seawater) are transmitted to the main pressure case through custom made cable which passes through a cable cutter (Figure 3). The geophone pressure case is heavy enough by itself (40 kg in air, 25 kg in water) to act as the instrument's bottom weight, and this is used in the design of the deployment mechanism. The case is suspended from an arm on a 6 foot pole above
458
J E R E M Y D U S C H E N E S ET AL.
Fig. 3. The cable cutter. This unit is mounted in series with the mechanical release and the anchor rope. When the release is activated, the cable cutter drops a few feet until the blade cable, attached to the main pressure case, is drawn taught, and the blade is drawn up through the 4 cables. Only the blade returns to the surface.
the b o t t o m weight (Figure 4). When the bottom weight hits the seafloor, the pressure case continues to drag the rest of the instrument down, but is guided away from the bottom weight by the pole. At a height a b o v e the seafloor of approximately 3 0 c m , the pressure case drops off the arm and the pole and arm retract to the vertical position pulled by the now positive buoyancy of the rest of the instrument (Figure 4). T h e geophone pressure case is left on the seabed with no physical connection to the rest of the instrument except for the thin electrical cables which follow the anchor cable up to the main pressure case. No e~tort is m a d e to measure the orientation of the horizontal geophones. At the termination of the experiment, the instrument is released acoustically and disengages the anchor rope immediately a b o v e the cable cutter. T h e ascending instrument pulls the cutter blade through the cutter frame, severing the four signal cables (Figure 3). T h e frequency response of the geophones is shown in Figure 5, and the recording system is shown schematically in Figure 6. T h e geophone input impedance was 240 KI~. Eight channels for data (three geophones, one of them recorded at two gains for extra dynamic range, the hydrophone recorded at two gains, flutter, and clock) are recorded as frequency modulated signals on 4 tracks
459
CAMBRIDGE DEEP OCEAN GEOPHONE
Fig. 4. A schematic representation of the pole deploying the geophone package on the seabed. The actual release of the pressure case occurs when the distance between the pole and the arm holding the case exceeds the length of wire on the pin holding the case to the arm.
2"0
1-0
SHUNT
u')
DAMPING ~
O,C. IOK
i
58.0 70.5
F-A-~ 0 . 4
~)
0"3
0"2
2
3
4
5
6
10
I
I
20
30
FREQUENCY-
I
I I IIII 50
1 100
Hz
Fig. 5. Output against frequency plot for SM-6 geophone. Output scale is in Volts per inch per second i.e. voltage produced for a given driving velocity (Manufacturers specification).
460
JEREMY
DUSCHENES
ET
AL.
Fig. 6. The recording system shown schematically. Where no geophone input is present, the overall system remains intact, with the 4 channels simply being recorded as null. X, Y, Z = Gcophones, H = Hydrophone, PA = Pre-Amplifier, A = Amplifier, M = Modulator, LPF = Low Pass Filter, HD = Head Driver, BO = Bias Oscillator, T R = T a p e Recorder, C = Clock, PS = Power Supply, L = Program Control Logic.
of a c a s s e t t e t a p e by s u p e r i m p o s i n g 2 c h a n n e l s at d i f f e r e n t ( n o n - i n t e r f e r i n g ) c a r r i e r frequencies on each track. Initially, the 4 d a t a signals (from t h r e e g e o p h o n e s a n d the h y d r o p h o n e ) a r e p a s s e d t h r o u g h p r e - a m p l i f i e r s p o s i t i o n e d as close to the s e n s o r s as possible. T h e signals then t r a v e l to the m a i n e l e c t r o n i c p a c k a g e ( F i g u r e 7) w h e r e t h e y are p a s s e d t h r o u g h an a m p l i f i e r w h o s e g a i n c a n be v a r i e d in 6 d b steps ( 0 - 5 4 d b r a n g e ) . T h e signals f r o m the h y d r o p h o n e a n d o n e h o r i z o n t a l g e o p h o n e are then split, o n e b r a n c h p a s s i n g t h r o u g h a s e c o n d s t a g e a m p l i f i e r w h i c h can also be v a r i e d in 6 d b steps ( 6 - 3 6 d b r a n g e ) . T h e c l o c k runs off a 5 M H z crystal, p r o d u c i n ~ a 100 h o u r b i n a r y c l o c k c o d e .
Fig. 7. The main electronics frame showing dual power supply, 4 cassette tape recorder and electronics cards.
CAMBRIDGE
DEEP
OCEAN
GEOPHONE
461
With a flutter correction signal there are then a total of 8 channels. Each of these is frequency modulated - clock, flutter, vertical geophone, and the high gain horizontal geophone channels at a center frequency of 675 Hz with a symmetrical excursion of • and the hydrophone and low gain horizontal geophone channels at 5400 Hz again with an excursion of • 4(1%. The four low frequency channels are low pass filtered at 95(1 Hz, and tile 8 channels are then paired, one high and one low, before being mixed with a 70 kHz bias signal and passed through a head driving circuit to a 4 track tape head. T o save power, the bias oscillator and head drive circuits are only run when a recording is actually being made. Tile recording is done on four tape recorders (Lenco model Mini T B - 5 1 8 ) with special record-only heads installed. With C- 12(1 cassettes this allows four hours of recording. T h e recording window controller, which uses the clock signal as input, can be p r o g r a m m e d to rtm the tape recorders in any order for any combination of 24 ten-minute windows. T h e r e are 6 ten-minute windows per cassette. T h e playback system includes a four track head replay cassette deck, four pre-amplifiers and a playback unit. This splits each signal by high pass filtering it at 325(I Hz and low pass filtering it at 950 Hz, then demodulates each one and low pass filters at 13(1Hz to r e m o v e any modulation harmonics. The resulting eight analogue signals can be displayed on an oscilloscope, or on any eight channel pen recorder, or be sent to a digitizer for further processing. Most other aspects of the instrument remain unchanged from Sinha et al. ( 1981 ). T h e complete technical details of the instrument have been compiled by Potts (1983).
3. Instrument Testing Construction of this instrument was begun in O c t o b e r 1979. High pressure tank tests of the new pressure cases to the equivalent of 4 km depth were conducted at the G P O facility in Ipswich in the spring of 1980. A low pressure tank test of the deployment mechanism was carried out at HMS Vernon in the s u m m e r of 1980. Sea trials were carried out in the Bay of Biscay in N o v e m b e r 1980 and in the North Sea in July 1981 aboard RRS John Murray. T h e D O G system has been used so far on six refraction experiments, including two in rugged ridge-transform environments with an instrument recovery rate of 93%, and a data retrieval rate of 92'/0 from recovered instruments.
4. Seismic Response Characteristics T h e quality of the equipment and the success of an experiment depend on the ability of the instrument to record undistorted ground motion, without which the signals cannot be used for processing more sophisticated than simple first arriwfl modelling. This is dependent on the sensors being isolated from any electrical or mechanical noise generated by the whole system, and on the system itself not being
462
JEREMY
DUSCHENES
ET
AL.
susceptible to improper cross-coupling of ground motion to instrument motion. The final test of this equipment came during its use in N o v e m b e r / D e c e m b e r 1981 when three major refraction lines were shot in the western Mediterranean. Described below are some of the features which would not necessarily be obvious on a record section, but which indicate that the instruments provide high fidelity recordings of ground motion. 4.1. SIMILARITIES BETWEEN H Y D R O P H O N E AND G E O P H O N E SIGNALS H y d r o p h o n e s a n d vertical g e o p h o n e s o p e r a t i n g side by side o n a n O B S should g e n e r a t e similar r e c o r d s of P - w a v e s a r r i v i n g with a n e a r - v e r i t i c a l a n g l e of i n c i d e n c e at the seafloor. A n e x a m p l e of such an explosive s o u r c e g r o u n d w a v e arrival as r e c o r d e d by the h y d r o p h o n e a n d d e p l o y e d vertical g e o p h o n e of the C a m b r i d g e D O G is s h o w n in F i g u r e 8. T h e p h a s e - f o r - p h a s e c o r r e l a t i o n is striking. G r o u n d m o t i o n is b e i n g m e a s u r e d s i m u l t a n e o u s l y in two different places several
Z o 0 u) uJ
(n
H 1 SEC
<
0.5
IO .i II.
0.25
H
0.0 10
20
30
40
50
60
~g
ILl
Hz
O n.
0.5
t. ffl m
0.25
Z
o.o
_1
<
n. Hz
o z
Fig. 8. Seismogram and normalised Fourier amplitude spectrum of a ground wave arrival recorded in the Tyrrbenian Sea by both a hydrophone (H) and a vertical geophone (Z). The area under the power spectrum curve between two frequencies represents the proportion of energy in that range. Note the excellent phase correlation between the two. The seismograms have been Low-pass filtered at 30 Hz before plotting.
CAMBRIDGE
463
DEEP OCEAN GEOPHONE
meters apart by two sensors operating on entirely different principles, and the correspondence between the two records leads one to conclude that both are recording ground motion accurately. T h e frequency spectra of these arrivals are also shown in Figure 8. Both show peaks at about 8 Hz. T h e hydrophone records show m o r e energy below 6 or 7 Hz than the geophone, but this is not surprising as the output of the g e o p h o n e (which has a natural frequency of 4.5 Hz) starts to fall off at approximately that frequency (see Figure 5). This is an argument for either using expensive high quality low frequency (less than 2 Hz) vertical geophones or not using vertical geophones at all. T h e spike at 50 Hz is caused by mains pickup during the playback procedure, and the 30 Hz spike represents instruments generated mechanical noise recorded by the hydrophone. Neither is a serious problem as all the seismic energy is clearly below 30 Hz and most of it is below 10 Hz. (The energy associated with these two peaks on the hydrophone fourier spectrum in Figure 8 represents less than 5% of the whole). T h e r e is no c o m p a r a b l e test of horizontal geophone recording. 4.2.
SIGNAL
TO
NOISE
RATIO
T h e signal-to-noise ratio on OBS records is a function of several factors: the ambient noise level, the shot size and range, transmission characteristics along the
IdtLat~d. kU,~I,AL*~,L,--IhA.ALA.~,,t~.~L,~...I i~,,t,.uil, Aldl,iJ ..... r'"-~m w V'.,-.,. Iv~.v.r, vvq,,,~ ~l--rv "v~".v~ yv.r, r--l-..rr- ~1 0.5 0.25 0.0 10
20
30
40
50
60
Hz
1 Sec .....
A,I.~J~.A~
.,,
, ~
.ILA,,I~,A~.
30
40
i.
,.,JJA
.ill
0.5
Z
0.25 0.0
a
10
20
50
60
Fig. 9. Seismogram and normalised Fourier amplitude spectrum of background noise recorded before the shot arrival shown in Figure 8. Scales in Figure 8. The heights of the peaks at 30 and 50 Hz are equal to those in Figure 8, allowing one to make direct comparison of signal to noise in the 0-20 Hz range.
464
JEREMY DUSCHENES ET AL.
ray path, and the level of internal noise recorded. The internal noise can be picked up in one or more of three places in the system. It can be mechanical noise picked up by the geophones and indistinguishable from seismic energy (except potentially for its frequency); it can be electronic noise picked up between the geophones and the tape recorders; it can be electro/mechanical noise picked up from the tape recorder itself as the recording is made. A way of quantifying the S/N ratio is to compare the shape of the frequency spectrum of the instrument during a period when no seismic energy is arriving to its shape when a ground wave arrives. In comparing the frequency spectrum of noise (Figure 9) to the spectrum of the shot immediately after (Figure 8), and using the peaks at 30 and 50 Hz to normalize the comparison, it is clear that the background noise is not interfering at all with the signal level, nor are there any dominant frequencies to distort the signal's frequency content.
4.3. P
TO S CONVERSION
One feature common to many record sections is the arrival of a strong phase on the horizontal components about one second after the first arrival on the hydrophone. T h e offset is not a function of range, nor is it constant from receiver to receiver. It appears that the horizontal phase is a P to S conversion, generated just below the receiver. In some cases a wiggle to wiggle correlation between the source wave P (on H or Z) and its conversion S (on Y or X) is observed (Figure 10), P to S conversion is a well known phenomenon (e.g. Lewis and McClain, 1977; and White and Stephen, 1980) which occurs most efficiently at major velocity discontinuities, in this case either a sediment/evaporite, or a sediment/basalt interface. On some receivers the ratio of converted to unconverted energy is very substantial (severalfold) but this is entirely dependent on local geology, i.e. the presence of a suitable interface for the conversion to occur.
5. Summary T h e D O G is able to record the ground motion detected by 3 orthogonal geophones, and a hydrophone. A number of features of the seismograms produced indicate that the ground motion detected by sea-bed geophones approximates actual ground motion substantially more closely than had been possible previously. The seismograms include no obvious evidence of vertical to horizontal crosscoupling, but good hydrophone to vertical geophone phase correlation, and a high signal-to-noise ratio. T h e good recording conditions are attained by moving the geophones as far away as possible from any sources of non-seismic noise such as water-current generated whole-instrument motion and tape recorder vibration, and by housing the geophones in the smallest pressure case possible.
CAMBRIDGE
DEEP OCEAN
465
GEOPHONE
Time j
I
~
L J----
:lutter
~~/~,
H Y
~!
~ V ~
~ ........
H
Y
J
X
Fig. 10. An analogue playback of a shot showing the phase correlation between Z, H, and the converted P to S on Y and X.
Acknowledgements
We acknowledge the assistance of Keith Louden, Drum Matthews, Martin Sinha, Tim Owen, Meivyn Mason, Mike McCormack, Roger Theobald, and John Leonard (all at Bullard Laboratories) in completing this project. We also thank the Captains, Officers and Crews of the RRS John Murray and RRS Shackleton. The geophone gimbal is based on a design by Ray Davies of the USGS (Woods Hole). The Commander of HMS Vernon, and Mr. Munro of British Telecom, generously gave us access to their test facilities. This project was funded by the Natural Environment Research Council, under grant GR/4060, and one of us (JD) was supported by Fonds FCAC (Quebec). Earth Science contribution ES535.
466
J E R E M Y D U S C H E N E S ET AL.
References
Avedik, F., Renard, V., Buisine, D., and Cornic, J-Y.: 1978, 'Ocean Bottom Refraction Seismograph (OBRS)', Mar. Geophys. Res. 3, 357-79. Byrne, D. A., Sutton, G. H., Blackington, J. G., and Duennebier, F. K.: 1983, 'Isolated Sensor Ocean Bottom Seismometer', Mar. Geophys. Res. 5, 437-449. Duschenes, J. D., Barash, T. W., Mattaboni, P. J., and Solomon, S. C.: 1981, 'On the Use of an Externally Deployed Geophone Package on an Ocean Bottom Seismometer', Mar. Geophys. Res. 4, 437-450. Lewis, B. T. R. and McClain, J.: 1977, 'Converted Shear Waves as Seen by Ocean Bottom Seismometers and Surface Buoys', Bull. Seismol. Soc. Am. 67, 1291-1302. Lewis, B. T. R. and Tuthill, J. D.: 1981, 'Instrumental Waveform Distortion on Ocean Bottom Seismometers', Mar. Geophys. Res. 5, 79-86. Potts, C. G.: 1983, 'Cambridge Ocean Bottom Seismometer Users Manual', Departmen of Earth Sciences, Bu|lard Laboratories internal report. Sinha, M. C., Owen, T. R. E., and Mason, M.: 1981, 'An Ocean-Bottom Hydrophone Recorder for Seismic Refraction Experiments', Mar. Geophys. Res. 5, 173-187. Sutton, G. H., Lewis, B. T. R., Ewing, J., Duennebier, F. K., Iwatake, B., and Tuthill, J. D.: 1981, 'An Overview and General Results of the Lopez Island OBS Experiment', Mar. Geophys. Res. 5, 3-34. Sutton, G. H., Duennebier, F. K., and Iwatake, B.: 1981, 'Coupling of Ocean Bottom Seismometers to Soft Bottom', Mar. Geophys. Res. 5, 35-51. Tfehu, A. M. and Solomon, S. C.: 1981, 'Coupling Parameters of the M.I.T. OBS at Two Nearshore Sites', Mar. Geophys. Res. 5, 69-78. White, R. S. and Stephen, R. A..: 1980, 'Compression to Shear Wave Conversion in Oceanic Crust', Geophys. J.R. Astr. Soc. 63, 547-565. Zelikowitz, S. J. and Prothero, W. A.: 1981, 'The Vertical Responses of an Ocean Bottom Seismometer: Analyses of the Lopez Island Vertical Transient Tests', Mar. Geophys. Res. 5, 53-57.