Method for Positioning Seismic Sources to Flatten CDP Fold in VSP Surveys

1. Field of the Invention

This invention relates to a method for positioning seismic sources and receivers in vertical seismic profile surveys (VSF's) to obtain a uniform distribution of common depth point data.

2. Description of the Related Art

A vertical seismic profile (VSF) is a seismic data gathering activity in which either the seismic receivers or the seismic sources are positioned at various depths in a drilled well. VSP surveys are closely related to horizontal seismic surveys. Both involve the propagation of seismic energy, from seismic sources, through the earth, to seismic receivers. The arrival times and character of the seismic waves are recorded. This data is typically further processed to eliminate or minimize spurious signals, and the resulting record provides significant information about the section of the earth through which the seismic energy passed.

The geometry of the VSP presents particular problems, however, which are not encountered in horizontal seismic surveys. In a horizontal survey, the positions available for placement of seismic sources and receivers are virtually unlimited. In a VSP, one leg of the survey is fixed, and is limited to the location of the wellbore. Depth positions are limited to depths approximately between the surface intercet of the wellbore and the well's total depth. This limitation presents a problem of diminishing fold across the reflectors surveyed.

In both horizontal and VSP surveys, the record of signals received by a receiver or single group of receivers is called a trace. During the processing stage, traces are stacked to eliminate spurious signals and the stacked traces represent a reflection from a common depth point (CDP) or point of reflection on the reflector surveyed. The number of traces stacked, which depends upon the number of source-receiver pairs which have received data reflected from a common depth point, is the fold.

Because of the limited geometry of the VSP, not all common depth points on a reflector will have the same fold, or number of traces stacked to represent them. Conventional practice positions sources and receivers at equally spaced locations along the surface and in the wellbore. Typical spacing between sources or receivers in the borehole is 10 to 25 meters, and typical spacing along the surface of the earth is hundreds of meters. With this equally spaced positioning, CDP fold for a reflector surveyed falls rapidly as the CDP point moves radially away from the borehole. This results in VSP data which has poorer resolution and a higher noise to signal ratio away from the wellbore. VSP data with fold fall-off is also poorly comparable with horizontal sections with which one may wish to "tie" the VSP survey.

Efforts to eliminate the reflection signal's amplitude degradation with increasing distance from the borehole due to CDP fold fall off problem in VSP's have centered on the processing phase of seismic exploration. Data is collected in a traditional manner, and then processed to compensate, somewhat, for a lack of fold. Most commonly, a technique known as amplitude gaining is employed to enhance data received from areas of the reflectors with fewer fold. This technique is not totally satisfactory because it destroys the true amplitude of the traces. True amplitude is useful both in making an accurate tie with horizontal surveys and in newer exploration techniques, such as direct hydrocarbon imaging (DHI). DHI requires retention of true amplitudes in a final record, as these amplitudes may be indicative of the fluid type (i.e. hydrocarbon) in the rock.

The invention provides a method and system for creating a vertical seismic profile (VSP), whereby the spacing interval between units, either seismic sources or receivers, in a borehole increases exponentially as their distance from the surface increases, and the spacing interval between units, either seismic sources or receivers, along the surface of the earth decreases exponentially as their distance from the surface intercept of the wellbore increases.

In a preferred embodiment of this invention, the CDP fold of the data gathered is substantially flattened across the reflectors surveyed. The initial parameters are selected: the depth of the well in which the VSP will be run is known, as is an approximate depth to a reflector of interest. Initial surface and depth spacing intervals are chosen based on aliasing conditions known in the art. One embodiment of the invention then provides that surface locations of either seismic receivers or sources are positioned exponentially closer together as their distance from the wellbore increases. Another embodiment of the invention provides that the distance between depth locations of either seismic sources or receivers in the wellbore increases exponentially with increasing depth in the wellbore. Data received from an apparatus with sources and receivers so positioned exhibits substantially flattened CDP fold prior to processing. Processing techniques which destroy or mask true amplitude are therefore unnecessary to generate data with improved signal to noise ratio away from the wellbore.

FIG. 1 illustrates pictorially the basic components of a vertical seismic profile, with sources and receivers positioned as taught by this invention.

FIG. 2 illustrates pictorially the basic components of a reverse vertical seismic profile, with sources and receivers positioned as taught by this invention.

FIG. 3 illustrates graphically a division of the available points in the surface offset (r) and depth (z) plane into CDP bins. The borders of this figure represent constraints imposed by reflection imaging.

FIG. 4 (Prior Art) This figure graphically illustrates the distribution of CDP stacking fold for a range of CDP offset radii, r.sub.CDP, for an example in which the target reflector is located at a depth of 10,000 ft. (3,050 meters). Line 401 illustrates the distribution of fold for conventional VSP shooting, with equally-spaced source and receiver positions.

FIG. 5 graphically illustrates the distribution of CDP stacking fold for a range of CDP offset radii, r.sub.CDP, for an example in which the target reflector is located at the same depth as in FIG. 4. Line 501 illustrates the improvement achieved by this invention, wherein the source and receiver were positioned according to the most preferred embodiment of this invention, herein described.

These figures are not intended to define the present invention, but are provided solely for the purpose of illustrating certain preferred embodiments and applications of the present invention.

In accordance with one embodiment of the present invention, there is provided a system for sending and receiving seismic signals. A plurality of seismic sources is positioned at or near the surface of the earth, and a plurality of seismic receivers are positioned, as indicated at 2 in FIG. 1, in a wellbore, as indicated at 4. The distance between a pair of adjacent seismic receivers, .DELTA.z, as indicated at 6 in FIG. 1, increases exponentially as a second distance, z, as indicated at 8 in FIG. 1, measured from the closer of the pair of adjacent seismic receivers, indicated at 10 in FIG. 1, to the surface intercept of the wellbore, indicated at 12 in FIG. 1, increases. Thus, in contrast to conventional VSP shooting, where seismic receiver positions are located at equally-spaced intervals, the seismic receivers in this system are positioned increasingly further apart as one goes further from the surface intercept of the wellbore. The sources in this system may be positioned at any locations at or near the surface of the earth.

In a preferred embodiment of this invention, a distribution of CDP fold recorded by this system is substantially equalized, or flattened, across a reflector surveyed, indicated at 14 in FIG. 1.

In a more preferred embodiment of this invention, a plurality of seismic receivers, indicated at 2 in FIG. 1, are positioned such that Equation I is satisfied:

A first distance, .DELTA.z, indicated at 6 on FIG. 1, is measured between a pair of adjacent seismic receivers, indicated by 10 and 15 in FIG. 1. A second distance, z, as indicated at 8 in FIG. 1, is less than or equal to the total depth of the wellbore and is measured from the closer of the pair of adjacent seismic receivers, indicated at 10 in FIG. 1, to the surface intercept of the wellbore, indicated at 12 in FIG. 1. .beta. is a constant, and e is a constant approximately equal to 2.718. The variable is an initial depth spacing interval based on aliasing considerations. To prevent data from having uncertain phase associated with sampling less densely than the Fourier criterion of at least 2 sample locations per wavelength, the standard procedure is to: (1) Determine the minimum speed of sound among the geologic layers for P-waves, S-waves, and Stoneley or "tube" waves in a borehole (or Rayleigh waves, also known as "ground roll," on the surface); (2) Determine the maximum frequency of seismic signals useful for reflection imaging, apply analog filters during field operations to remove the higher frequencies, and then (3) Set the spatial sampling density, in this case z', equal to the minimum velocity divided by twice the maximum frequency.

In a more preferred embodiment of this invention, the constant .beta. in Equation I is calculated by using linear regression to determine the slope of a plot of z versus log.sub.e A(z), where A(z) is determined by solving Equation II: ##EQU1## for a series of values of z=(')(n). The variable Z' is the initial depth interval used in Equation 1 and n is an integer. Of necessity. z is less than or equal to the depth of the wellhore. The variable r is set equal to r', where r is an initial surface offset interval also based on aliasin8 considerations. D, indicated at 18 FlGURE 1, is the depth to a reflector being imaged, and .DELTA.c is a CDP increment, which may be chosen to be r' or r'/2.

In a most preferred embodiment of this invention, the seismic sources are positioned at a plurality of distances, r, indicated by 20 in FIG. 1, and the seismic receivers are positioned at a plurality of depths, z, indicated at 8 in FIG. 1, such that the plurality of distances, r, and the plurality of depths, z, which are available for mapping a reflector at depth, D, are grouped into CDP radial offset bins. CDP radial offset bins are sets of data that represent a common depth point on a reflector being mapped. In the most preferred emobodiment of this invention, the plurality of distances, r, and the plurality of depths, z, are grouped into CDP radial offset bins according to the Equation III: ##EQU2## In this equation, r.sub.CDP, indicated at 22 in FIG. 1, is a distance measured radially from an axis running vertically through the wellbore, indicated by 24 in FIG. 1, to a common depth point, indicated at 26 in FIG. 1, on the reflector being mapped, indicated at 14 in FIG. 1. The variable r is measured from the surface intercept of the wellbore, indicated at 12, to the position of a seismic source at or near the surface of the earth. The depth z is measured from the surface intercept of the wellbore to the position in the wellbore of a seismic receiver. Of necessity, in shooting reflection seismic, the variety of depths, z, and the variety of surface offsets, r, which may be employed are limited by the constraint that an angle .theta., indicated at 28 on FIG. 1, must be .ltoreq.45.degree.. The depth, z, must be less than or equal to the reflector depth D, and the surface offset, r, will be less than or equal to twice the reflector depth, D. D is measured from the surface of the earth, along a perpendicular to said surface, to the reflector, indicated at 14 in FIG. 1.

FIG. 3 illustrates the division of available surface offset and depth positions into CDP bins. The example in FIG. 3 illustrates the situation where a reflector being surveyed lies at a depth of 10,000 feet in a constant velocity material, with a velocity of 8,000 feet per second. The borders of the diagram represent the constraints just discussed. Embodiments noted as most preferred provided the best flattening of CDP fold in experimental tests.

Another embodiment of the invention provides a system for sending and receiving seismic signals comprising a plurality of seismic receivers positioned in a wellbore, and a plurality of seismic sources, indicated at 30 in FIG. 1, placed along or near the surface of the earth, indicated at 32 in FIG. 1. In this system a first distance, .DELTA.r, indicated by 34 in FIG. 1, between a pair of adjacent seismic source positions, indicated at 36 and 38 in FIG. 1, decreases exponentially as a second distance, r, indicated at 20 in FIG. 1, measured from the closer of the pair of adjacent seismic sources, indicated at 36 in FIG. 1, to the surface intercept of the wellbore increases. Thus, in contrast to conventional VSP shooting, where source positions are located at equally-spaced intervals, the sources in this system are positioned increasingly closer together as one goes further from the surface intercept of the wellbore. The receivers in this system may be positioned at any locations in the wellbore.

In a preferred embodiment of this invention, seismic sources in this system are positioned such that a distribution of CDP fold recorded by the system is substantially equalized across a reflector surveyed.

A more preferred embodiment of the invention provides that the plurality of seismic sources, indicated at 30 in FIG. 1, are positioned such that Equation IV is satisfied:

In this equation, .DELTA.r is a first distance, indicated at 34 in FIG. 1, measured between a pair of adjacent seismic source positions, r is a second distance, indicated 20 in FIG. 1, measured from the closer of the pair of adjacent seismic source positions to the surface intercept of wellbore, indicated at 12 in FIG. 1, and r' is an initial surface spacing interval based on aliasing considerations known in the art. The variable r' must be less than or equal to 2 D, .alpha. is a constant, and e is a constant approximately equal to 2.718.

In a more preferred embodiment of the invention, .alpha. is calculated by using linear regression to determine the slope of a plot of r versus log.sub.e A(r), where A(r) is determined by solving Equation V for values of r: ##EQU3## The depth of the seismic receivers, z, as indicated at 8 in FIG. 1, is held equal to zero, and the equation is solved for a series of values for r equal to n(r'), where n is an integer, and r' is the initial surface spacing interval used in Equation IV. D is a depth to a reflector, indicated by 18 in FIG. 1, and .DELTA.c is a CDP increment, chosen to be either r' or r' /2.

A most preferred embodiment of this invention provides that the seismic receivers are positioned at a plurality of depths, z, indicated at 8 in FIG. 1, and the seismic sources, are positioned at a plurality of distances, r, indicated at 20 in FIG. 1 such that the plurality of depths, z, and the plurality of distances, r, which are available for mapping a reflector at depth, D, indicated at 18 in FIG. 1, are grouped into CDP radial offset bins according to Equation III. CDP radial offset bins are sets of data that are treated as representing a common depth point on the reflector being mapped. In this instance, r.sub.CDP, indicated at 22 in FIG. 1, is a distance measured radially from an axis running vertically through the source or receiver of interest in the wellbore, indicated at 24 in FIG. 1, to a common depth point, indicated at 26 in FIG. 1, on the reflector being mapped. The quantity r is measured from the surface intercept of the axis, usually corresponding to the surface intercept of the wellbore, indicated at 12, to the position of a seismic source at or near the surface of the earth. The depth z is measured from the surface intercept of the wellbore to the position in the wellbore of a seismic receiver. Of necessity, in shooting reflection seismic, the variety of depths, z, and the variety of surface offsets, r, which may be employed are limited by the constraint that an angle .theta., indicated at 28 on FIG. 1, must be .ltoreq.45.degree.. The depth, z, must be less than or equal to the reflector depth D, and surface offset, r, will be less than or equal to twice the reflector depth, D. D is measured from this surface of the earth, along a perpendicular to said surface, to the reflector, indicated at 14 in FIG. 1.

FIG. 3 illustrates the division of available surface offsets, r, and depths, z, into CDP bins for an example in which the reflector of interest lies at a depth of 10,000 feet, and in which the material through which the seismic waves pass has a constant velocity of 8,000 feet per second. The borders of this diagram represent the constraints just discussed.

Another embodiment of the invention provides a system for sending and receiving seismic signals comprising seismic sources positioned in a wellbore and seismic receivers, indicated at 102 in FIG. 2, positioned along or near the surface of the earth, indicated at 132 in FIG. 2. A first distance, .DELTA.r, indicated at 134 in FlGURE 2. measured between a pair of adjace*t seismic receivers, indicated at 136 and 138 in FIG. 2, decreases exponentially as a second distance, r, indicated at 120 in FIG. 2, from the closer of the pair of adjacent seismic receivers, indicated at 136 in FIG. 2, to the surface intercept of the wellbore, indicated at 112 in FIG. 2, increases.

In a more preferred embodiment of the invention, the seismic receivers are positioned such that a distribution of CDP fold recorded by the system is substantially equalized across a reflector surveyed. An example of a reflector surveyed is indicated by 114 in FIG. 2.

In a more preferred embodiment of the invention, the seismic receivers are positioned such that Equation IV is satisfied. .DELTA.r is the first distance, indicated at 134 in FIG. 2, measured between a pair of adjacent seismic receivers, indicated at 136 and 138 in FIG. 2. The variable r is the second distance, indicated at 120 in FIG. 2, measured from the closer of the pair of adjacent seismic receivers, indicated at 136 in FIG. 2, to the surface intercept of the wellbore, indicated at 112 in FIG. 2. The variable r' is an initial surface spacing interval based on aliasing considerations known in the art, .alpha. is a constant, and e is a constant approximately equal to 2.718.

In a more preferred embodiment of the invention, .alpha. is calculated by using linear regression to determine the slope of a plot of r versus log.sub.e A(r), where A(r) is determined by solving Equation V for values of r equal to n(r'). The variable n is an integer, and r' is the initial surface spacing interval used in the above equation. In solving this equation, the depth of the seismic receivers, z, as indicated at 8 in FIG. 1, is held equal to zero. D is a depth to a reflector, indicated by 18 in FIG. 1, and Ac is a CDP increment, chosen to be either r' or r' /2.

A most preferred embodiment of the invention provides that the seismic sources are positioned at a plurality of depths, z, indicated 108 in FIG. 2, and the seismic receivers are positioned at a plurality of distances, r, indicated at 120 in FIG. 2, such that the plurality of depths, z, and the plurality of distances, r, which are available for mapping a reflector at a depth, D, are grouped into CDP radial offset bins according to Equation III. CDP radial offset bins are sets of data that are treated as representing a common depth point on the reflector being mapped. In this instance r.sub.CDP is a distance, indicated at 122 in FIG. 2, measured radially from an axis running vertically through the wellbore, indicated at 124 in FIG. 2, to a common depth point, indicated at 126 in FIG. 2, on a reflector being mapped, indicated at 114 on FIG. 2. The distance r is measured from the surface intercept of the wellbore, indicated at 112, to the position of a seismic receiver at or near the surface of the earth. The depth z is measured from the surface intercept of the wellbore to the position in the wellbore of a seismic source.

FIG. 3 illustrates the division of available surface offsets, r, and depths, Z, into CDP bins for an example in which the reflector of interest lies at a depth of 10,000 feet, and in which the material through which the seismic waves pass has a constant velocity of 8,000 feet per second. Again, in reflection shooting, the available r and z positions are limited by the constraint that an angle .theta., indicated at 140 on FIG. 2, must be less than or equal to 45.degree.. The depths, z, must be less than or equal to the reflector depth D, indicated at 118 on FIG. 2, and the surface offsets, r, indicated at 120 on FIG. 2, must be less than or equal to twice the reflector depth D. The borders of FIG. 3 represent these constraints.

Another preferred embodiment of the invention provides a system for sending and receiving seismic signals comprising a plurality of seismic receivers positioned at or near the surface of the earth and a plurality of seismic sources, indicated at 130 in FIG. 2, are positioned in a wellbore, indicated at 104 in FIG. 2. A first distance, .DELTA.z, indicated at 106 in FIG. 2, between a pair of adjacent seismic source positions increases exponentially as a second distance, z, indicated at 108 in FIG. 2, measured from the closer of the pair of adjacent seismic source positions, indicated at 110 in FIG. 2, to the surface intercept of the wellbore, indicated at 112 in FIG. 2, increases.

In a preferred embodiment of the invention, sources are positioned such that a distribution of CDP fold recorded by the system is substantially equalized across a reflector surveyed.

A more preferred embodiment of the invention provides that the seismic sources indicated at 130 in FIG. 2, are positioned such that Equation I is satisfied. In this instance, .DELTA.z, indicated at 106 in FIG. 2, is a first distance, measured between a pair of adjacent seismic sources, indicated at 110 and 116 in FIG. 2. The depth z, indicated at 108 in FIG. 2, is a second distance, less than or equal to the total depth of the wellbore and measured from the closer of the pair of adjacent seismic receivers, indicated at 110 in FIG. 2, to the surface intercept of the wellbore, .beta. is a constant, z' is an initial depth data spacing interval based on aliasing considerations known in the art, and e is a constant approximately equal to 2.718.

In a more preferred embodiment of this invention, the constant .beta. in the above equation is calculated by using linear regression to determine the slope of a plot of z versus log.sub.e A(z), where A(z) is determined by solving Equation II for a series of values of z less than or equal to the depth of the wellbore, where z=(')(n). The variable 'is the initial depth interval used in Equation I, and n is an integer. In solving this equation, r is held equal to r', an initial surface offset interval based on aliasing consideration, D, indicated at 118 FIG. 2, is a depth to a reflector, and .DELTA.c is a CDP increment, which may be chosen to be r' or r' /2.

A most preferred embodiment of the invention provides that the receivers and sources are positioned such that the depths, z, and distances, r, which are available for mapping a reflector at depth, D, are grouped into CDP radial offset bins according to Equation III. In this instance, r.sub.CDP, indicated at 122 in FIG. 2, is a distance measured radially from an axis running vertically through the wellbore, indicated at 124 in FIG. 2, to a common depth point on a reflector being mapped, indicated at 126 in FIG. 2. The distance r is measured from the surface intercept of the wellbore, indicated at 112 to the position of a seismic receiver at or near the surface of the earth. The variable z is measured from the surface intercept of the wellbore to the position in the wellbore of a seismic source.

FIGS. 1 and 2 illustrate positions of both sources and receivers that are positioned according to this invention. Positioning of sources and positioning of receivers may be practiced independently, however, and should result in data of more even fold than conventional spacing. FIG. 5 illustrates results for a 10,000 foot reflector, imaged through rock of a constant velocity, positioning both seismic sources and receivers as taught by the invention. FIG. 4 illustrates results obtained for the same conditions using conventional positioning. Curve 401 illustrates the extreme drop off of CDP stacking fold under conventional VSP spacing, and curve 501 in FIG. 5 illustrates the results achieved through use of this system.

The computer program contained in Table I, written in FORTRAN, illustrates one preferred embodiment of the invention.