CALIBRATION OF MARINE SEISMIC SOURCES USING A HYDROPHONE OF UNKNOWN SENSITIVITY*

The absolute amplitude of the pressure pulse radiated by a marine seismic source is one of the prime criteria used in evaluating its performance. A technique for this measurement is proposed which is applicable to all sources which radiate a bubble pulse. The technique is described with reference to an air-gun array in which the pulse radiated from a single gun is compared with that radiated by the full array. The advantage of the method is that absolute values of pressure are obtained without any need for a calibrated hydrophone. In theory this may seem a trivial advantage but in practice sensitivity factors for the hydrophone channel cannot always be relied upon. The proposed technique is illustrated by an example.     

THE USE OF AIRGUN PRESSURE PULSES FOR CALIBRATING A LOW-FREQUENCY STANDARD HYDROPHONE*

A new technique is developed for calibrating a low-frequency hydrophone. The technique involves the use of pressure pulses radiated by the Par 0.65 liter airgun when fired at a fixed depth but with various values of initial chamber pressure. The sensitivity of a low-frequency hydrophone, when determined by the proposed technique, is found to be in agreement with that obtained by using the so-called “impulse method”.     

TEST RESULTS OF A NEW TYPE OF EFFICIENT SMALL AIRGUN ARRAY*

Recently the author developed and demonstrated (Safar 1980) an efficient method for operating the airgun. The method involves the generation of a short seismic pulse from the pressure bubble pulses radiated by an airgun when fired several times at the same optimum depth but with different chamber pressures. The purpose of this paper is to present and discuss the test results obtained when implementing the same method using a two-dimensional airgun array. The array consists of seven 0.65 liter airguns fired simultaneously at the same depth but with different chamber pressures. It is shown that the far-field pressure pulse radiated by the seven 0.65 liter airgun array is similar to that radiated by the Flexichoc seismic source. It is concluded that the proposed airgun array can be used as a subarray to form an extremely powerful super-long array suitable for deep seismic exploration. The author would like to thank the Chairman and Board of Directors of the British Petroleum Co. Ltd for permission to publish this paper. Thanks are also due to Mike Symes and Lovell Cox for carrying out the field tests and Seismograph Service (England) Ltd for providing the airguns.     

大容量气枪震源子波时频特性及其影响因素

通过分析福建街面水库气枪实验的近场水听器记录,研究气枪子波时频特性及其受沉放深度和工作压力的影响,并结合气泡模型解释气泡振荡过程。数据分析表明：①气枪子波由主脉冲和气泡脉冲组成。主脉冲振幅大,持时短,频带宽,通常应用于浅部探测;气泡脉冲能量集中在低频段,垂直穿透深,水平传播远,通常应用于深部探测。②随沉放深度的增加,主脉冲振幅变化很小,气泡脉冲振幅增加,初泡比减小,气泡周期减小,低频段主频增加。沉放深度为10m时,主脉冲振幅和初泡比最大,可应用于浅部探测;沉放深度为25m时,气泡脉冲振幅很大,初泡比最小,可应用于深部探测。③工作压力增加时,主脉冲振幅、气泡脉冲振幅、初泡比、气泡周期等随之增大,低频段主频则减小。     

THE RADIATION OF ACOUSTIC WAVES FROM AN AIR-GUN*

A theoretical model is developed for predicting three important parameters of the pressure pulse radiated by an air-gun, namely the rise time, the amplitude of the initial pulse, and the period of the bubble pulse. A knowledge of these three parameters is essential for the efficient design of air-guns arrays. The prediction of the amplitude of the initial pulse is based on the assumption that the initial pulse is radiated by a spherical source with surface area equal to that of the air-gun ports and not by a spherical source with initial volume equal to that of the air-gun chamber, as has been assumed previously. A simple equation is obtained for predicting the period of the bubble pulsation, taking into account the effect of the air-gun body, boundaries such as the sea-surface and seabed and the presence of a number of identical air-guns placed at the same depth and fired simultaneously.     

Marine seismic sources: QC of wavefield computation from near-field pressure measurements

A commercial marine seismic survey has been completed with the wavefield from the n-element (single guns and clusters) airgun array measured for every shot using an array of n + 2 near-field hydrophones, n of which were required to determine the source wavefield, the remaining two providing a check on the computation. The source wavefield is critical to the determination of the seismic wavelet for the extraction of reflection coefficients from seismic reflection data and for tying the data to wells. The wavefield generated by the full array of interacting airguns can be considered to be the superposition of n spherical pressure waves, or notional source signatures, the n hydrophone measurements providing a set of n simultaneous equations for each shot. The solution of the equations for the notional source signatures requires three ingredients: the geometry of the gun ports and near-field hydrophones; the sensitivity of each hydrophone recording channel; and the relative motion between the near-field hydrophones and the bubbles emitted by the guns. The geometry was measured on the back deck using a tape measure. A calibration data set was obtained at the approach to each line, in which each gun was fired on its own and the resulting wavefield was measured with the near-field hydrophones and recorded. The channel sensitivities, or conversion from pressure at the hydrophones to numbers on the tape, were found for each near-field hydrophone channel using the single gun calibration data, the measured geometry, and the peak pressure from each gun, known from the manufacturer’s calibration. The relative motion between the guns and hydrophones was obtained from the same calibration data set by minimizing the energy in the computed notional source signatures at the guns which did not fire. The full array data were then solved for the notional source signatures, and the pressure was computed at the two spare hydrophones and compared with the actual recordings. The rms errors were 5.3% and 2.8% and would have been smaller if the hydrophone channel sensitivities had been properly calibrated beforehand and if the movement of the guns with respect to the hydrophones had been more restricted. This comparison of the predicted and measured signatures at spare hydrophones can, in principle, be done on every shot and we recommend that this be implemented as a standard quality control procedure whenever it is desired to measure the wavefield of a marine seismic source.     

PHYSICAL ASPECTS OF THE “AIRPULSER” AS A SEISMIC ENERGY SOURCE*

During the last few years the airpulser, or air gun, has become very common as an energy source for marine seismic surveys. This paper describes the physical processes which take place during the operation of the pulser and develops theoretical results concerning the energy and frequency of the radiated signal and the amplitude decay of the secondary bubble pulses. The theory takes into account the presence of the airpulser itself which is assumed to be a rigid sphere within the bubble of released air. The theoretical results are combined and compared with measurements made of the pressure within the airpulser, the acceleration of the body of the pulser, and the amplitude and frequency of the signal radiated into the surrounding water. A formula for calculating the bubble frequency is given and a diagram made of the energy partition between mechanical losses, radiated energy, etc. Finally, a comparison is made of the energy release from the airpulser with that from TNT.     

EFFICIENT DESIGN OF AIR-GUN ARRAYS*

A new technique is developed for shaping the pressure bubble pulse radiated by an array of air-guns. It involves the proper spacing of identical units. It is shown that considerable shortening of the pressure bubble pulse can be achieved provided there is sufficient mutual coupling between all the air-guns. The existing air-gun array technique for reducing the bubble pulse involves the redistribution of the energy of the bubble pulses which are produced by an array of variable sized air-guns such that no energy of the bubble pulses is radiated along the perpendicular to the array axis and only the sum of the initial pulses produced by the air-guns forming the array is radiated along that direction. However, the new air-guns array technique involves the damping of the bubble pulses which are produced by an array of identical air-guns by means of mutual interaction and the effective pressure pulse radiated by the array is given by the sum of the damped bubble pulses produced by the mutually coupled identical air-guns. Preliminary field trials gave results consistent with the theoretical predictions. A comparison between the waveforms of the pressure bubble pulses radiated by a single air-gun and by an array of four identical air-guns shows that, due to the presence of mutual coupling between the four air-guns, effective damping of the bubble pulse radiated by the array is about 50% greater than that of the bubble pulse radiated by the single air-gun.     

PERFORMANCE OF 2000 AND 6000 PSI AIR GUNS: THEORY AND EXPERIMENT*

Air guns have been used in various applications for a number of years. They were first used in coal-mining operations and were operated at up to 16000 psi charge pressures. Later, single air guns, operated at 2000 psi, found application as an oceanographic survey tool. Air gun arrays were first used in offshore seismic exploration in the mid-1960's. These early arrays were several hundred cubic inches in total volume and were operated at 2000 psi; they were either tuned arrays or several large guns of the same size with wave-shape kits. Today's arrays have total volumes greater than 5000 cu in. and are typically operated at 2000 psi. Recently, higher-pressure, lower-volume arrays operated at 4000–5000 psi have been introduced; guns used in these arrays are descendants of the coal-mining gun. On first thought one would equate increased gun pressure linearly with the amplitude of the initial pulse. This is approximately true for the signature radiated by a “free-bubble” (no confining vessel) and recorded broadband. The exact relation depends on the depth at which the gun is operated; from solution of the free-bubble oscillation equation, the relation is If P_c,1= 6014.7 psia, P_c,2= 2014.7 psia and P_{O, 1}=P_{O, 2}= 25.8 psia (corresponding to absolute pressure at 25 ft water depth), then Experiments were conducted offshore California in deep water to determine the performance of several models of air guns at pressures ranging from 2000 to 6000 psi and gun volumes ranging from 5 to 300 cu in. At a given gun pressure, the initial acoustic pulse P_a correlated with gun volume V_c according to the classical relation For 1 ms sampled data the ratio varied between 4.5 and 5.5 dB depending on gun model. Pulse width of the 2000 psi signatures indicated they are compatible with 2 ms sample-rate recording while pulse width of the 6000 psi signatures was greater, indicating they are less compatible with 2 ms sample-rate recording. Conclusions reached were that 2000 psi air guns are more efficient than higher pressure guns and are more compatible with 2 ms sample-rate requirements.     

A PULSED PLASMA JET ACOUSTIC SOURCE FOR PROFILING THE OCEAN FLOOR1

A plasma gun source has been successfully used to obtain sub-bottom profiles. The profiles show better penetration than with a 3-5 kHz source and more resolution than with an air gun. The plasma gun source is compact, self-contained, and requires no complex auxiliary equipment. The device was deployed from large research vessels and a small boat. The plasma gun produces sufficient acoustic energy and with its characteristically short pulse length and broad bandwidth it is an attractive sub-bottom profiler, especially in shallow fresh or salt water.     

ON THE DETERMINATION OF THE DOWNGOING P-WAVES RADIATED BY THE VERTICAL SEISMIC VIBRATOR*

It is well recognized that in order to realize the full potential of the Vibroseis technique, one needs to ensure accurate phase locking and a meaningful cross-correlation. To achieve these two important objectives we require an accurate estimate of the compressional stress wave radiated by the vibrator into the ground. In this paper a simple method (subject of a patent application) is developed for predicting the compressional stress waves radiated by a vertical vibrator. The main feature of the proposed method is that it involves the field measurement of the acceleration of the reaction mass and the baseplate, respectively. The method is illustrated by computing the compressional stress waves generated by a typical vertical vibrator radiating into ice, chalk, sand, and mud. It is shown that for a seismic vibrator radiating into hard ground the pressure of the downgoing P-wave is 180° out of phase with the baseplate velocity. It is also shown that when the driving force of the seismic vibrator has a flat amplitude spectrum, the amplitude spectrum of the downgoing P-wave falls off by 6 dB/octave towards low frequencies.     

AN EFFICIENT METHOD OF OPERATING THE AIR-GUN*

A new technique is developed for generating a short seismic pulse from the bubble pulses which are radiated by an air-gun. The new technique, which is useful in well velocity surveys and vertical seismic profiling, can be implemented by firing a single air-gun several times at the same depth but with different chamber pressures. A record obtained by this procedure from a well-geophone clamped at a depth of 2450 m gave a maximum peak-to-peak amplitude within the first 100 ms of the effective seismic pulse at least ten times any later peak-to-peak amplitude.     

Geophone-seabed coupling effect and its correction

By summing geophone and hydrophone data with opposite polarity responses to water layer reverberation, the ocean bottom cable dual-sensor acquisition technique can effectively eliminate reverberation, broaden the frequency bandwidth, and improve both the resolution and fidelity of the seismic data. It is thus widely used in industry. However, it is difficult to ensure good coupling of the geophones with the seabed because of the impact of ocean flow, seafloor topography, and field operations; therefore, geophone data are seriously affected by the transfer function of the geophone-seabed coupling system. As a result, geophone data frequently have low signal-to-noise ratios (S/N), which causes large differences in amplitude, frequency, and phases between geophone and hydrophone data that severely affect dual-sensor summation. In contrast, the hydrophone detects changes in brine pressure and has no coupling issues with the seabed; thus, hydrophone data always have good S/N. First, in this paper, the mathematical expression of the transfer function between geophone and seabed is presented. Second, the transfer function of the geophone-seabed is estimated using hydrophone data as reference traces, and finally, the coupling correction based on the estimated transfer function is implemented. Using this processing, the amplitude and phase differences between geophone and hydrophone data are removed, and the S/N of the geophone data are improved. Synthetic and real data examples then show that our method is feasible and practical.     

LOW-FREQUENCY STANDARD HYDROPHONE * FOR CALIBRATING LINE HYDROPHONE ARRAYS (SEISMIC STREAMERS) AND MARINE SEISMIC SOURCES **

The design of a standard hydrophone with a maximally flat (Butterworth) response in the frequency range 8.0 Hz-1.0 kHz is described. The standard hydrophone has been developed primarily for calibrating line hydrophone arrays (seismic streamers) and marine seismic sources. The standard hydrophone has been used successfully during the past eight years for monitoring the output of a single air gun. It can be used for the calibration of a marine seismic streamer.     

Sea surface shape derivation above the seismic streamer

The rough sea surface causes perturbations in the seismic data that can be significant for time‐lapse studies. The perturbations arise because the reflection response of the non‐flat sea perturbs the seismic wavelet. In order to remove these perturbations from the received seismic data, special deconvolution methods can be used, but these methods require, as input, the time varying wave elevation above each hydrophone in the streamer. In addition, the vertical displacement of the streamer itself must also be known at the position of each hydrophone and at all times. This information is not available in conventional seismic acquisition. However, it can be obtained from the hydrophone measurements provided that the hydrophones are recorded individually (not grouped), that the recording bandwidth is extended down to 0.05 Hz and that data are recorded without gaps between the shot records. The sea surface elevation, and also the wave‐induced vertical displacement of the streamer, can be determined from the time‐varying pressure that the sea waves cause in the hydrophone measurements. When this was done experimentally, using a single sensor seismic streamer without a conventional low cut filter, the wave induced pressure variations were easily detected. The inversion of these experimental data gives results for the sea surface elevation that are consistent with the weather and sea state at the time of acquisition. A high tension approximation allows a simplified solution of the equations that does not demand a knowledge of the streamer tension. However, best results at the tail end of the streamer are obtained using the general equation.     

Moment magnitude (Mw) from hydrophone records of low energy volcanic quakes

Earthquake magnitude calibration using hydrophone records has been carried out at Campi Flegrei caldera, an active area close to the highly populated area of Naples city, partly undersea. Definite integrals of the hydrophone records amplitude spectra, between the limits of 1 and 20 Hz, were calculated on a set of small volcano-tectonic earthquakes with moment magnitudes ranging from 1 to 3.3. The coefficients of a linear relationship between the logarithm of these integrals and the magnitude were obtained by linear optimization, thus defining a useful equation to calculate the moment magnitude from the hydrophone record spectra. This method could be easily exported to other volcanic areas, where submerged volcanoes are monitored by networks of hydrophones and seismic sensors on land. The proposed approach allows indeed magnitude measurements of small magnitude earthquakes occurring at sea, thus adding useful information to the seismicity of these volcanoes.

     

MODELLING OF GI GUN SIGNATURES1

In 1989 a new type of marine seismic source was introduced. This new air-gun, which consists of two air chambers instead of one, is called the GI gun. The main feature of this gun is that the bubble created by the gun is stabilized by an injection of extra air from the second chamber at a later time. This injection mechanism reduces the amplitude of the bubble oscillations, which also means that the acoustic signal from a GI gun shot is characterized by a very clean primary pulse followed by very small bubble oscillations. A method for calculating the acoustic signal generated by a GI gun is presented. Based on the solution of a damped Kirkwood–Bethe equation, the far-field pressure of single GI guns and of arrays of GI guns is calculated. It is shown that the optimal values for injection start time and injection period vary with injector volume and gun depth. It is also shown that the precision in the firing time for the injector should be of the order of 4 ms, while the precision of the injection period should be of the order of 8 ms. Modelled and measured far-field signatures have been compared, and the relative error energy is found to be less than 3.5% for all examples.     

Comparative experimental study on several methods for measuring elastic wave velocities in rocks at high pressure

To measure elastic wave velocities in rocks at high temperature and high pressure is an important way to acquire the mechanics and thermodynamics data of rocks in the earth's interior and also a substantial approach to studying the structure and composition of materials there. In recent years, a rapid progress has been made in methodology pertaining to the measurements of elastic wave velocities in rocks at high temperature and high pressure with solids as the pressure-transfer media. However, no strict comparisons have been made of the elastic wave velocity data of rocks measured at high temperature and high pressure by various laboratories. In order to compare the experimental results from various laboratories, we have conducted a comparative experimental study on three measuring methods and made a strict comparison with the results obtained by using the transmission method with fluid as the pressure-transfer medium. Our experimental results have shown that the measurements obtained by the three methods are comparable in the pressure ranges of their application. The cubic sample pulse transmission method used by Kern is applicable to measuring elastic wave velocities in crustal rocks at lower temperature and lower pressure. The prism sample pulse reflection-transmission method has some advantages in pressure range, heating temperature and measuring precision. Although the measurements obtained under relatively low pressure conditions by the prism sample pulse transmission method are relatively low in precision, the samples are large in length and their assemblage is simple. So this method is suitable to the experiments that require large quantities of samples and higher pressures. Therefore, in practical application the latter two methods are usually recommended because their measurements can be mutually corrected and supplemented.     

考虑多种因素影响的海上地震勘探气枪震源单枪子波数值模拟

在分析Ziolkowski气泡振动模型局限性的基础上,建立了多种实际因素影响下的海上地震勘探单枪子波模型。此模型考虑了气泡壁的热传导作用、枪口节流作用、气泡上浮、液体粘度和枪体本身等对气枪子波的影响。相对于Ziolkowski模型,改进模型所模拟的气枪子波主峰值减小,气泡振动衰减加快,与实测子波吻合性较好。实验分析表明：（1）枪口节流作用控制着气枪脉冲峰值振幅的大小,（2）上浮过程中气泡周围静水压力值减小,气泡振动的周期随之改变,（3）热传导作用和流体粘度是引起气泡振动衰减的主要因素。