DE2734091A1 - PROCEDURES FOR SEISMIC INQUIRIES - Google Patents

Description

W.M2926 / 77 20 / hs

Mobil Oil Corporation New York (V. St. * A.)

Seismic investigation method

The invention relates to seismic surveys at sea.

A method for eliminating the influences or the distortion due to multiple reflections or multiple reflections j reverberations) is known (US-PS 3 6Ö9 Ö7O, at which seismic data is used. With the known methods, a work process is necessary by which the reverberation distortion component the data is separated from the component representing the characteristics of subterranean formations. This operation is called dereverberation, and through it a dereverberation operator is obtained which is implemented in the form of an inverse filter can be. With the help of this filter, the seismic

709885/1005

Data processed in such a way that a signal arises which approaches the signal or roughly corresponds to the signal, which would be generated by a desired shot pulse which would interact with the reflectivity of the subterranean formation if there were no reverberation distortion.

Although the dereverberation process has been successfully applied to many marine se ismograms, it is in many areas of application are unsuccessful or cannot be used with a high degree of efficiency.

A method for igniting the sources of a list in a chronological sequence is known (NL-ÖS 76 01 413), by which a resulting acoustic pressure wave is generated, which has a time domain-bound characteristic which is inverse to the distortion due to reverberation. As a result of the generation of seismic pulses, the reflection and the recording of the reflections is linear System wherein the generation of seismic impulses as an inverse distortion operator has the same desired effect how a method generated by inverse filtering. The known method has the additional advantage that the dynamic Area of the recording system that is solely occupied by reverberations is reduced. To a It is to dismiss the listing of sources in a chronological order which form an inverse operator however, the reflectivity of the water bed is necessary

or the bottom of the water and the water depth or the two-energy departure time of the seismic / through this water layer to determine.

In accordance with the invention, an improved method for marine seismic surveys has been developed. Here It is essential that seismic pulses are generated in the process and that the seismograms are recorded correspond to the reflected pulses. The characteristics of the medium through which the seismic impulses passed is then determined from the seismograms

709885 / 100S

• y-

and the determined characteristics are bsw in parameters. Parameters for setting up seismic traces converted. This list is then carried out in accordance with these parameters.

The characteristics that are measured or determined include the reflectivity of the bottom of the water and the depth of the water. Another characteristic which is of interest is the slope of the reflective underground layers or formations. The location of the seismic sources is preferred made in such a way that the energy from the sources is reflected approximately perpendicularly by the inclined formation and so that it intersects the position of the listening devices perpendicularly.

The present invention also makes a Simplified procedure for determining the values of the reflection coefficients and the transit times on board a ship was created.

In one embodiment of the invention, the reflectivity of the water bottom and the two-way or back and forth transit time T _{w is determined} directly from seismograms, which are generated by conventional shooting from sources of a linear list. The measured or determined reflection and the transit time are converted into. a 3-point reverse operator. The list of sources is then restored in a time sequence that this 3-point operator generates.

The sources in the setup can be ignited at the same time to generate conventional seismograms with reverberations or multiple reflections. The reflection and the two-way transit time are determined from these conventional seismograms. The sources will then ignited in groups. The ignition of a second group of sources is delayed from the ignition of the first group by the measured or determined transit time, and the ignition of the third group of sources is about that

709885/1005

Delayed twice the Zeiweg runtime. The number of sources each group is determined by the measured reflection.

The invention is explained below with reference to the drawing, for example.

Fig. 1 serves to explain the occurrence of multiple reflections due to the bottom of the water. Pig. 2 and 3 serve to explain the origin of

Reverberations due to the bottom of the water. Fig. JJ and 5 show installations for seismic investigation at sea.

6 to 9 serve to explain the term two-point and three-point operator (time domain inverse operator).

Fig. Io shows conventionally generated seismograms with assumed values for the yeast fraction R and the transit time T _w>

11 is a flow chart representation and serves to explain the individual stages of the method according to the invention. Fig. 12 shows the correlation function as a function

the time for a selected time window. 13 shows the actual reflection R as a function of the time window length F for various assumed reactions R, where this representation is used to calculate the actual reflection from the assumed reflection. Fig. 1 * 4 shows a system for seismic investigation

for use at sea.

15 shows a schematic seismogram for examining inclined formations.

16 shows a device for seismic investigation.

FIG. 17 shows a modified device for seismic investigation, in which, however, compared to the device in FIG. 16, the erection or arrangement of the sources is inclined.

. 709885/1005

from

The seismic energy can be / finer in tow List of sources are generated. The resulting acoustic pressure waves are generated by the underground Boundaries of the formations are thrown back and the corresponding reflections are recorded as seismograms. Unfortunately, these seismograms include perturbations that affect the characteristics of the underground Blur formations. These external influences or deformations of the seismograms are often found in investigations at sea strongly pronounced. These seismograms include multiple reflections that can be traced back to the bottom of the water are and, as shown in Fig. 1, arise. Furthermore, these seismograms contain reverberations, their origin is explained with reference to FIGS. 2 and 3.

When the initial shot pulse reaches the bottom of the water, it is partly transmitted or passed on there and partly thrown back at this interface. The transmitted part is reflected by the discontinuities below the water, creating a primary reflection. The reflected part runs back to the surface, where it is in turn reflected to the bottom of the water. Again, part of this pulse is reflected at the interface and another part is left on. This process continues indefinitely. Every time a pulse travels back and forth between the surface and the bottom of the water, a pulse is emitted from the interface at the bottom of the body of water, which follows the initial pulse introduced in the section. These multiple pulses are delayed relative to one another by equal periods of time and their amplitudes are successively reduced by the reflection factor or reflection coefficient at the bottom of the water. The end result of this process is that a seismogram is generated which consists of a single primary reflection, which is a train of equally spaced multiple bsw. Reverberations follow.

In a known dereverberation process (US Pat. No. 3,689

700885 / 100B

a reverse filter is used, which is based on the seismic Data is applied to generate a signal that roughly corresponds to the signal that is generated would be by a desired shot impulse, which with the reflection process of the subterranean formation interacts when the reverberation distortion would not exist.

While the well-known dereverberation process has been used successfully on marine seismic granuns, it is cannot be used in all areas. In particular, when the bottom of the water is hard, the reverberations have such a strong influence that the entire dynamic Area of the recording system is required to record just the keverberations realistically. Accordingly, it is a purpose of the invention to provide a method by which the reverberations be suppressed before recording.

FIGS. ¹ J and 5 show a system for seismic investigation at sea, in which a vehicle 11 traverses a seismic exploration path when investigating an underground formation 12 below the water 13. A linear, horizontal list of seismic sources ΐΊ, 15, l6 and other sources is in the tow of the vehicle 11, a cable 17 being used. Oberf smileenschwinuiier help to maintain the same distance between the sources.

An underwater listening device 18 (hydrophone streamer) is also in tow behind the vehicle A surface support buoy 19 helps to establish the correct spatial Location between the listening device 18 and the sources to be maintained. A seismic trace which occurred during the Seismic survey at sea generated by this facility can be described mathematically as follows:

4 * Rd) * N «)

708885/1005

where S (t) is the seismic trace,

B (t) is the shot pulse, H (t) is the reverberation distortion

or the influence of reverberation,

R (t) is the reflection function and N (t) is the disturbance function (noise).

Equation (1) states that the seismic trace is generated by the shot pulse on which the reverberation influence is superimposed, on which the reflectivity is superimposed, with additional coherent disorders and random noise is added. The process of convolution is indicated by a $ in equation - (1), this operation being carried out by the filtering method. As a result, the seismic trace is the result of the shot pulse which is filtered by the reflectivity of the earth and by the influence of reverberation has been.

Basically, the reverberation influence is a function of both the size and the sign of the reflection coefficients R at the bottom of the water and at the water surface. The reverberations or multiple reflections of the shot pulse are equal relative to one another Time spans are delayed and their amplitudes are continuously reduced by these reflection coefficients. The reflection coefficient at the bottom of the water can either be positive or be negative in its sign, which depends on the nature of the water floor, whereas the reflection coefficient on the water surface is always negative and in its amount is assumed to be the size of the unit (-1). Furthermore, every multiple pulse is im Compared to the previous pulse, reduced by the amount of the water floor reflection R. Accordingly, a dereverberation operator can be described by one / two-point operator:

709885/1005

i-Tw) (2) This operator, which consists of a first pulse of unit size at time t = u and a second pulse of size R at time t = T, can: be represented in two forms. The first form int is shown in Fig. 6 for a: positive water floor reflection coefficient R. The second form is shown in Fig. 7 for a negative water floor coefficient R. Under all circumstances where the sea floor is relatively shallow, those are running below Reverberationseinfluß at the shot point and the up-going Keverberationseinfluß can at the detector approximately equal to each other. thereof As a result, for a flat water bottom, an entire Dereverberaticnseperator I (t) by the Konvulot Lon of Zweipunktoperatorf3 aiii shot point i _c (t) with the two-point operator i. (t) are reproduced on the detector:

Kt) « i _s (t) * i _d (t) (3)

» [5 (t) + R6 (t + T)] * [6 (t) + R6 (t + T)] (4)» & It) + 2R6 (t + T ) + R ² O (t + T ) (5)

In the event that there is a positive seabed reflection R, the dereverberation operator can be represented as in FIG. 8, while ViIv a negative seabed reflection of the dereverberation operator can be represented in FIG. 9. Therefore it can be seen that the entire dereverberation operator I (t) for a shallow water bottom is a three-point operator, which consists of three pulses, the first having the unit size at time T = 0, while the second has a size 2 R at time t = T and the third is the quantity R ² at time t = 2T _w .

By igniting the sources in different groups, whereby different ignition times apply for each group, the resulting acoustic pressure wave has the one shown in Fig. 8

709885/1005

BATH ORIGINAL

JAW? ^ I-IO Ci

and 9 reproduced, time-dependent characteristic. For example, the operator of FIG. 8 is generated by igniting a first group of pulses at time T = 0. The amplitude of the pulse generated by this group is arbitrarily chosen as a unit. A second group of pulses is ignited at a time | r. The amplitude of this pulse, which is generated by the second group, has a magnitude of 2 R compared to the magnitude of the pulse of the first group, where K is the reflection of the bottom of the water. A third group of pulses is ignited at time 2T. The amplitude of the pulse es generated by the third group has a ratio of 1: R ^{c to the} amplitude of the first pulses. It can be shown that the ignition of the sources in such an order as described above produces recordings with suppressed reverberation or suppressed multiple reflections.

In order to ignite the installation of the sources in such an order, it is necessary to determine the reflection coefficient R and determine the two-way transit time T of the acoustic energy through the water layer. According to the invention become R and T from a conventional seismogram obtained, which is generated by simultaneously supplying the sources Ii to 16. The sources are simultaneous fired to determine R and T and on the sources are fired in groups, producing a three-point reverse distortion operator. The seismic Energy from this reverse operator produces seismograms which are essentially free of reverberations.

As an example of the manner in which R and T _w can be determined from a conventional seismogram, consider FIG. 10. In Fig. 10, a conventional seismogram is shown which has been combined, stacked or correlated in accordance with

for different assumed values of R and T. the

BATH ORIGINAL

Make seismic sections under lines 21 and 22 Represents reflections detected by the listening devices 18 as a function of time (ordinate) after the shot have been. The seismic sections under lines 23 to 25 represent the same seismic sections which with an assumed reflection of R = O, 1 have been stacked, using various assumed values of T as a basis. The seismic cut below line 24 is the sum of the conventionally made incision plus that of the conventionally generated cut, which has been multiplied by a factor of 2 (0.1) and delayed by 96 ms, where to The track generated in the conventional way has been added to this section, which is given by the factor (0.1) multiplied and delayed by 192 ms. The cut under the line 25 is the sum of the usual way and Manner generated track, which was multiplied with the same factors, but delayed by 98 ms and 2 χ 98 ms became. The last section in this sentence under line 26 is the sum of the conventional manner generated Track, which was multiplied in the same way, but delayed by 120 ms or 2 χ 120 ms became.

The cuts under lines 27 to 29 are combined, correlated or stacked, the sections having the same time delays, namely 9 ^, 96, 120 me, but with an assumed reflection of R = 0.2. The sections under lines 30-32 are stacked sections with an assumed reflection of R = 0.5 and the same time delays. The values of R and T can be determined from such an analysis by selecting those values of R and T _w which produce the minimum energy in a section. A visual analysis of Figure 10 shows that section 33 has the minimum energy in it. This indicates that the correct value of R is 0.5 and the correct value of the running time T _{w is} 11 * 4 ms. The values R and T _w can in this way and

709885/1005

However, a very large number of computer operations are necessary and it is usually not practical to conduct such analysis on board a ship. However, a simplified method for determining R and T ^ can be carried out on board a ship. Figure 10 shows that the correct value of T can be determined for any assumed value of R. L> he cut 3 ^ namely has the minimum energy in his sentence and indicates a value of T of 114 ms, although the cut 34 was based on an incorrect value for R = 0.2 during the stacking process. Correspondingly, the section 35 has the minimum energy in its set and ciie3 indicates a correct value for T of 114 ms, although again an incorrect value for R = 0.1 was used for stacking. According to the invention, this phenomenon is exploited in that a value is assumed for the reflection and the transit time T is determined. The reflection can then be determined from the known value for T.

11 is used to explain a method for Determination of R and T, which can be used on board a ship. A seismogram from one to conventional. Wise executed scoundrel is recorded on a magnetic tape 40 as usual. A part or a The window of the seismogram is played or selected, as is indicated at 41. This time window will subjected to an autocorrelation process by an auto-correlator.

The auto-correlation function is shown in FIG. As is known, the auto-correlation requires point-wise Multiplying the values by itself in a time window to produce a set of autocorrelation coefficients, which are then summed to produce a point of the autocorrelation function shown in FIG. Then it will Shifted in time, the values are in turn multiplied by themselves and added up to form another point of the autocorrelation function. This

. 70Ö885 / 1005

The process is repeated for different time shifts. As an example, the time shift of the value the values in the window are multiplied by themselves to give the autocorrelation coefficient for the delay O produce. This summation gives rise to the value S of in Mg. 12 curve reproduced. Thereupon the values of the time window shifted by a sample time unit, multiplied by itself, by the coefficient at a To generate delay of a sample unit time. These Values are added up to another point of the To form the correlation function according to FIG. In general, the time window is related to the size of the time window shifted by a number of sample times, which is denoted by η.

The autocorrelation coefficient S for the time difference

lag O is combined with the autocorrelation coefficient for a lag of η and for a lag of 2 η for different values of η. The change in the value of η is performed as indicated at 4j5 in FIG. The combination takes place in the multipliers 44. The 0 delay value 3 is multiplied by 2.0625. This value of the correlation function with a delay of η is multiplied by 2.5 and the value with a delay of 2 η is multiplied by 0.5. These products are added in summer 45. The minimum energy, ie the smallest sum in the combined autocorrelation functions, is selected as indicated at 46. The minimum energy sum indicates the correct value n, which is denoted by M. Since M is given in sample time units, this can be _{converted directly into T w} because the sample time unit is known. The sections according to FIG. 10, for example, have a sample period of 2 ms. Assuming that a minimum energy is determined in the combined autocorrelation coefficient when n = 57, the result of the multiplication (at 47 in FIG. 11) is a running time T of 114 ms,

709885/1005

49 273409!

The described determination of the reflection represents an improvement of the method described in the article by Pflüger (Geophysics, 37, 1972, pp. 738-796). In known methods, the reflection is related to the ratio of the amplitudes of the relative, maximum amplitudes of the autocorrelation function, which are multiples the running time T occur. This process creates a Good estimate of the reflection if the time window used in the autocorrelation is very long. However leads the use of a long time window in the autocorrelation leads to further problems, in particular to daft interference, noise and the like in the calculation process to be introduced. For this reason, the window, which is determined according to Ml of FIG. 11, is selected as a relatively small multiple of the time T.

When determining the reflection according to FIG. 11, the Pflüger method was used to establish an estimated value R. The divider 48 according to FIG. 11 forms the ratio

where S ₁ , - the determined value of the correlation function at T in w

and S _{2M is} the value of the autocorrelation function at 2 T _w .

The autocorrelation function is interpolated in order to achieve a finer or smaller sampling rate or a smaller step. For example, the autocorrelation coefficients can be determined at intervals of Ί ms and it is desirable to convert this period into a 1 msec sample size. To do this, it is assumed that the autocorrelation function is a function represents an x / x function. Using this interpolation method, the amplitude spectrum of the data remains and no new frequencies will be introduced.

It can be shown that the estimated value R for the true reflection R in the manner shown in Fig. 13 in Relationship stands. F is the length of the correlation window

709885/1005

in units or multiples of the running times, which are included in the window. As an example, it can be assumed that the autocorrelation window extends from 3 T up to 7 T and R is determined to be 0.4 by the Pflüger method. For P - 4, according to FIG. 13, a value for R is obtained which is approximately 0.425. in practice no reference is made to a diagram in the manner of FIG. Instead, the coefficients which are used to convert the estimated value ft into the true value R are stored in a memory. The coefficients are A, A. and A ^ and are applied to IT as indicated in Figure 11 at 49 and 50, respectively. The coefficients represent least squares of an area obtained by plotting the true value of R as a function of η and the first estimate of R. The coefficients are determined in particular according to 50 of FIG. 11. The value F according to this method is found by dividing the length of the time window 3 by the transit time T as indicated at 51.

The length of the autocorrelation window, which is determined according to FIG. 41, is a function of the two-way transit time, which can be determined with the aid of a depth meter 52. The autocorrelation delay M is obtained by dividing twice the reading of the depth gauge 52 by the speed of propagation in the water, which is assumed to be 4800 fVs. This ratio is determined at 52 A according to FIG. The length of the autocorrelation window is a multiple of the autocorrelation delay n. In general, a time window extends from 5M to 15M for each track, which "is a suitable value."

A better understanding of these work processes can can be obtained from the description of the underlying theory below.

A seismic trace on which a reverberation operator has acted can be described as follows:

3 fe}

708885/1005

where S. is the i-th sample of the seismogram and S. is a Represents sample with time shift n.

According to the method according to the invention, a reasonable value for R is assumed. For example, R = 0.5 so that equation (6) becomes:

Ä -

Using the least squares method, the correct value of R and η is obtained when the sum of the squares of the foregoing is a minimum. This means that the correct value η occurs when the following quantity assumes a minimum:

i-2n

> SS + 2R Υ SS
t— i · in L ~ t χ x-

2n

+ 4R ~) SS L, in i-

2n (8)

A large number of computer operations are required to carry out the above method. To determine the correct value η in this way, approximately 200,000 operations would be required. Around Carrying out a determination for 10 different values of η would accordingly require approximately 2,000,000 operations. According to a feature of the present In the invention, certain simplifications and certain simplifying assumptions can be made. If the length of the

708885/1006

right to
Time window In the time shift η is long, the measurements obtained for two windows which have been shifted by η with respect to one another are the same. Hence the following can be assumed:

^; i ^s in * L ^S in ^S i-

2n - ^S M

I ^S i ^S i-

2n " ^S 2M

Using these assumptions, equation (7) becomes s ² = (1 + 4R ² + R ⁴ ) S + (4 $ + 4R ³ ) S _M + 2 & ² S, _M (12)

using an assumed value of K = 0.5 value adjustment (12)

2 _β 2.0625 S ₀ + 2.5 S _M + .5 S _2M < ¹³⁾

In this way, simplified multipliers 44 can be used.

In order to carry out the method according to the invention, it is necessary to determine the correlation function only once. As mentioned earlier, this can take approximately 200,000 operations. In order to carry out the 3 for 10 different values for η, it is necessary to multiply by S _M and S_ _M with 2.5 and 0.5 for only nine

708885 / 10.05

perform more times. This results in 18 operations for a total of 200,018 operations. This is approximately a ten-fold reduction in the number of operations that would be required without the simplifying assumptions.

Once the values T R have been determined, the charges dismissed in groups, whereby the desired three-point operator is generated or implemented. The sources are in three Groups ignited, which are designated with a, b and c. Group a is ignited at time T = O. The group b becomes the Time t = T and group c becomes at time T = 2T ignited. The ignition circuits 53 to accomplish this can be of the type shown in U.S. Patent 3,687,218.

The number of sources in each group is determined by the value R. If a, b and c are the number of sources in each group denote, these numbers are related to each other, resulting in the determined reflection by the following Equation is given:

a + b + c = total number of sources

A typical list which has been used in practice has 40 sources, which ^{are denoted by the numbers n} 1 ^M to "40". In this example, four sources are used as backup sources, and the sources are ignited in the following abandoned groups:

708885/1006

R -0.3 (a, b, c) = (20,28,39) a «40,39,37,36,33,32,30,29,27,, 23,22,19,18,13 , 12,10,9,7,4,3,1 b »38,34,31,28,24,21,20,16,14,11,8,5,2 c - 25.17

R - ^ 4 (a, b, c) - (23,26,38) a - 40,38,36,33,31,29,27,23,22,19,18,13,11,9,7 , 4,3,1 b - 39,37,34,32,30,28,24,20,16,14,12,10,8,5,2 c - 25,21,17

R -0.5 (a, b, c) - (25.25.37) a - 40.38,36,33,31,29,25,22,18,14,12,10,8,5, 3.1 b ■ 39,37,34,32,30,27,23,21,19,17,13, li;, 9,7,4,2 C - 28.24.20.16

R = 0.6 (a, b, b) ~ (27,24,36) a - 39,37,34,32,28,24,23,19,12,11,9,7,4,2 b - 40,38,36,33,31,30,27,22,20,18,16,14,10,8,5,3,1 C = 29,25,21,17,13

R = 0.7, (a, b, c) = (29,23,35) a «39,37,34,30,27,22,18,14,11,8,4,2 b» 40,38,36,33,31,29,25,23,21,19,17,13 , 10,9,7,5,3,1 c »32,28,24,20,16,12

R «0.8 (a, b, c) = (30.23.34) a - 39,37,32,31,28,23,19,16,11,7,3, b «40,38,36,34,30,27,24,22,20,18,14,12, 10,8,5,4,2,1 c - 33,29,25,21,17,13.9

709885/1006

The ignition process is carried out as follows: When the vehicle makes a run through the test section, the sources are alternatively ignited in a known manner, with all charges being discharged at the same time, then three groups are ignited to represent the three-point operator. Traces from the last five listening devices in the listening device array produced by a conventional shot are used to determine T. This value T is used to implement the delay time for the three-point operator shot that is subsequently used. For each three-point operator shot, the value T from the immediately preceding conventional shot is used.

The value of the reflection R is also obtained from the tracks of the last five listening devices in the arrangement. because if R changes slowly when the vehicle is moving, R is not recalculated for each shot. Instead of this an optimal value of R, which occurs over the underground length of the cable, is used to calculate the number of sources in each group for the three-point operator nuts.

As mentioned above, the orientation of the installation in a certain direction can also be dependent on the measurement of the inclination, with this method the arrangement of the source is preferably made in such a way that that the energy from the sources of the inclined formation is reflected approximately vertically, and onto them Way the arrangement of the listening devices intersects vertically.

First, a conventional check is made to identify the subterranean formation of interest. The inclination of the formation can be determined from the seismic reflection signals generated during this first check. A survey is then conducted to generate seismograms showing increased reflections from that formation. During this

709885/1005

As part of the investigation, the ignition of each source in the array is delayed so that seismic ones are created Wave is directed. That is, then the seismic wave not vertically towards the inclined formation, but instead tapers at an angle. That angle has such a size that if u have seismic energy touches or impinges on the sloped formation is reflected in a vertical direction. The influence of direction can alternatively be obtained by changing the length of the cables between the surface buoy and the By dragging successive sources with different depths, the Anoi'dnung the sourceffi a seismic wave, which is in propagates at a desired angle.

14 shows a device for carrying out a seismic survey at sea above inclined formations. A vehicle bü passes through or crosses a seismic investigation line when monitoring subterranean formations below a waterfront location. A linear, horizontal setup of the seismic sources S _{1 and} S is carried along behind the vehicle 60. Also behind the vehicle 60 at a greater distance than the sources, a horizontal arrangement of acoustic receivers R ₁ to R is towed. If the vehicle 60 crosses or passes through a desired investigation line, the seismic sources S and S 'are ignited at the same time in order to generate a seismic pressure wave in the water layer or in the water. The acoustic receivers R ₁ through R generate electrical signals responsive to receiving seismic reflections from the subterranean formations caused by the generation of the seismic pressure wave. With the aid of a cable 61, these electrical signals are fed to conventional recording devices on board the vehicle 60.

In a typical seismic survey at sea, the seismic sources S ₁ to S are ignited simultaneously and repeatedly when the vehicle 60 crosses or passes through the investigation line. The seismic reflections

709885/1005 'BAD ORIGINAL

tion signals which are obtained by the recorded Si from the acoustic receivers R ₁ to R for each seismic source are recorded in a known manner as side-by-side traces of a seismic graph (FIG. 15). The arrival times of the reflection signals of the receivers R. to R are shown lying on a curve H. This curve is characterized by the hyperbolic function:

^T x

T is the reflection time of a particular seismic Track,

T is the reflection time of an idealized seiso

mix track with a reflection point immediately below of the shot point.

X is the horizontal distance between the point of fire and the receiver which is the particular seismic trace generated.

V is the acoustic velocity of the layer through which the seismic energy passes.

However, if the subterranean formation is inclined at an angle <κ between the horizontal and the formation itself, the times T of seismic reflections from such are inclined Formation changed by the influence of the angle of inclination oc so that the arrival time curve H is represented by the following equation will:

7 2 X ²

7 2 X 2XT _n S im

709885/1005

CJAH

• to

The positive sign applies to an egg trap arrangement, which is inclined downward from a source anorulation.

In U3-PS 3 696 331 there is an example of a method for determining and storing values underground Slopes below the acoustic velocity given in accordance with the equation (Ip).

According to the method according to the present invention, a second crossing is carried out by the vehicle bO along the investigation line, and the values of the inclination and the acoustic speed determined during the first pass are now used to determine the successive dumping of the loads in the location of the seismic sources so that the resulting seismic wave is directed into the water inclined at a desired angle θ along the line of investigation. This desired angle θ is such that the seismic pressure wave shown in F13. 16, designated 65a, occurs upon the inclined formation 62 so as to be reflected _{vertically towards the line of seismic receivers R 1 to R.} Such alignment can be obtained by retarding the ignition timings of successive sources S ₁ through S, the retardation time being given by the following equation:

T-sf. sm θ

X is the distance between the sources

V _w is the speed of the seismic energy in the water

θ is the direction angle

709885/1005

It is a particular feature of the method according to the invention that the direction angle θ is such that the reflected seismic energy from the inclined formation 62 propagates vertically towards the water surface. In this way, poorly, the reflected seismic pressure wave (o'jb in Fig. 16) will intersect the horizontal array of receivers H. bi3 R perpendicularly, thereby permitting the generation of seismograms which have enhanced reflections from the inclined formation. It is noted that it follows from equation (15) that the move-out of the recorded reflection signals from the position of the receivers R ₁ to R is related to twice the sine of the inclination angle ex. Accordingly, in the method according to the invention, the ignition of successive seismic sources S ₁ to S is delayed in such a way that the direction angle θ for the downwardly traveling seismic pressure wave deviates from the vertical by twice the degree of the inclination angle o (of the fornation 62 Ignitions of the seismic sources in order to generate such a seismic pressure wave can be varied continuously during the second pass along the investigation line in accordance with the determined value of the inclination and the speed from the first pass along the investigation line.

As an alternative, the above-described seismic Investigation procedures for use at sea can be carried out during a single pass along the investigation line by igniting the seismic Sources between a first simultaneous Zoning of all sources and a second successively carried out Ignition of all sources is carried out alternately. The first simultaneous ignition is followed by the value of Inclination and speed, as described above, determined and a second consecutive ignition

• 709885/1006 BAD ORIGINAL

alier sources is then carried out to the resulting seismic pressure wave in t.ineni Wirice] - ü to train, which is double the amount cios Nei ^ un ^ ri-

•, ι λ / · *. do. ^ simultaneous angle OList, which has been determined from the first rounding of all seismic sources. This sequence of a first and a second ignition is carried out repeatedly along the entire examination line. Appropriate ignition circuits are used to ignite the sources in a time sequence with a delay time Ü * between each source. An example of such a switching circuit contains the U3-Po t> ud7 21b.

An alternative possibility for obtaining a directional arrangement of the sources is shown in FIG. The linear arrangement of the source is dragged or carried along by the vehicle at an angle θ to the riorisontal, so that the seismic wave generated by the simultaneous ignition of the source has the direction angle θ in be-2, u £ c. to the vertical Change the length of the cables between the surface bojun and the sources in the order of the sources. D.is means that the cable bü is longer than the cable hf , the cable 67 being longer than the cable bö, so the source ^ ₁ hangs lower than the source 3, and the source S ₂ hangs lower than the source S etc.

Conventional analog or wired digital circuits can be used to carry out a method according to the invention can be used .. However, a small general purpose digital computer will be used preferred, which can be carried on board a ship. An example of such a thing Computer that has been used successfully is the Texas Instruments Computer ydü A.

709885 / 1'OOS BATH ORIGINAL

Le e rs e

ite

■ • '■ "Ο

Claims (1)

Claims I 1J method for seismic investigation, marked by generating seismic pulses from a source towed by a marine vehicle, by recording seismograms representing the reflected seismic pulses by determining the characteristics of the medium through which the seismic impulses pass from the seismograms by converting the measured values of the characteristic into control parameters for controlling or defining an arrangement seismic sources and by defining an arrangement of seismic sources which are provided by the marine vehicle be carried in accordance with the control parameters. 2. The method according to claim 1, characterized in that when measuring the water floor reflection and the water depth from the recorded seismograms is determined. 3. The method according to claim 2, characterized in that that the measurements of the reflection and the water depth are converted into a time-related operator, which for the reversal of the reverberation influences-in is characteristic of the water layer between the water bottom and the water surface. * J. Method according to claim 3, characterized in that that the arrangement of the sources in groups is ignited in such a way that the acoustic impulses from the sources are combined to generate a resulting acoustic pressure wave, which is a characteristic of a time-dependent Operator in its expansion to and from an underground formation. 5. The method according to claim 4, characterized in that that the sources are ignited simultaneously to seismic . 709885/1008 ORIGINAL INSPECTED To generate impulses from which the seismograms are recorded in order to determine the characteristics to be measured to enable. 6. The method according to claim 5, characterized in that the sources alternatively simultaneously and then in groups are ignited, the group ignition generating an acoustic pressure wave, which a time-dependent operator is equivalent to. 7. The method according to any one of claims 4-6, characterized in that a first, second and third group is ignited in such a way that the front flanks of the pressure waves from these reinforce each other, to the first, second and generate third acoustic impulses in the water. Ö. Method according to claim 7> characterized in that that the energy produced by the second and third groups of sources is so determined in relation to the energy which is generated by the first group that there is a multiplication factor between them, which in relation to the measured reflection of the water bottom. 9. The method according to claim b, characterized in that the energy generated by the second group to the The energy produced by the first group is related to a factor of two between the two measured water floor reflection is present and that the energy generated by the third group to that by the first group generated energy squared of the measured Reflection is dependent. 10. The method according to any one of claims 7 ~ 9> characterized by retarding the second group of liners
pulse after the first group of pulses by a time T, which is related to the determined water depth, and by delaying the ignition of the third group of the source from the ignition of the first group of the source by an amount of 2T w. 709885 / 1.005 11. The method according to any one of claims M-IO, characterized in that the number of ignited sources of each group is determined in accordance with the measured reflection. 12. The method according to claim 4, characterized in that that the resulting acoustic pressure wave has a time-dependent characteristic, which for the reversal the reverberation influence in the water layer between the water bottom and the water surface is characteristic. 13. Method according to claim 12, characterized in that the water bottom reflection and the water depth are determined by generating aato correlation coefficients with a delay of 0 from a time window of the seismogram, since the autocorrelation coefficients of the window are generated with a delay of n Autocorrelation coefficients of the window are determined with a delay of 2 _n test units, that the above is repeated for different values of η, that the autocorrelation coefficients generated for each value of η are combined and that that value for η is selected which has the minimum in the combined autocorrelation functions, whereby this value is indicative of the water depth. 14. The method according to claim XJ, characterized by forming the values S _o , / & _{2 of} the autocorrelation function of the window at the test times 0, η or 2n. 15. The method according to any one of claims 13 or 14, characterized in that the autocorrelation functions can be combined by summing the values of the autocorrelation functions according to the following equation: «1,06.25 ά + 45" S ₁ + 709885/1006 16. The method according to any one of claims 13-1!?, Characterized by converting the number of test units η into the running time T, which is the two-way running time the acoustic energy through the water. 17. The method according to any one of claims 13-lb, characterized characterized in that the water floor reflection coefficients are generated by making an estimate the reflection is performed from the following equation: 2S2 by the length of the window in units of the number η of the two-way transit time T in the window by the following equation Time window length in ms T _w in ms is determined by extracting the value of the coefficients A, A., A _D from the following "" expressions: ² A ₀ -0.0048 (J -0.0174 (I) ² A ₁ - -4OOO6 + 0, 0786 (n) -0, 0022 (n) ² A -0.0023 + 17.055 (J) - 25.691 (J) ² , and wherein the reflection R is determined from the following equation becomes: ^ Lö.Verfahren according to any one of claims 1-17, characterized characterized in that the water depth is determined with the aid of a depth gauge and that the time window of the seismogram is selected with a length which is twice the reading of the depth gauge divided by the speed the seismic energy in the water. 19. The method according to claim 1, characterized in that the inclination of the underground formation of interest is determined from the seismograms and the arrangement of the 709885/1005 seismic sourcesJLn depending on the measured Inclination is set so that the seismic pulses from the sources are oriented at such an angle that the resulting seismic wave, which is reflected from the formation of interest, runs vertically in the direction of the water surface. 20. The method according to claim 19, characterized by a linear arrangement of the seismic sources to directional generate seismic impulses. 21. The method according to claim 20, characterized in that that the seismic pulses are generated by delay egg generated seismic pulses which by each source in the lineup can be obtained by selecting a time delay between the ignition of the sourcei to direct the seismic energy at your desired angle. 22. The method according to claim 21, characterized in that the delay is selected by: where T "is the time delay, ^ X the distance between the sources, V is the velocity of the seismic impulses in the water and θ is the angle of direction. 23. The method according to claim 22, characterized in that the direction angle θ is twice the angle of inclination oc of the formation of interest is chosen. 2M. Method according to claim 20, characterized in that that the seismic impulses are aligned by setting up the source linearly at an angle is towed, which deviates from the horizontal, with the sources of the installation ignited at the same time will. 25 · The method according to claim 20, characterized in that that the seismic pulses are oriented at an angle from the normal which is times the Corresponds to the measured slope of the subterranean formation. 709885/1006 26. The method according to any one of claims 2o-25 _f, characterized in that the seismic pulses are generated at spaced points along an examination line which is inclined downward from the arrangement of the listening devices or hydrophones. 709885 / 10.0 5