WO2004063770A1

WO2004063770A1 - Method for quantifying water column timing and amplitude anomalies in a 3 dimensional seismic survey

Info

Publication number: WO2004063770A1
Application number: PCT/US2003/000219
Authority: WO
Inventors: Jack Dewayne Kinkead; Michael Glenn Jones
Original assignee: Pgs Americas, Inc.
Priority date: 2003-01-06
Filing date: 2003-01-06
Publication date: 2004-07-29
Also published as: AU2003207451A1; GB2401182B; GB0419070D0; GB2401182A

Abstract

A method is disclosed for determining timing error between segments of a marine seismic survey (Fig. 2). The method includes picking arrival times of a water bottom reflection event in a plurality of seismic signals acquired in a plurality of seismic survey segments (16). A discontinuity in the picked arrival times is determined between at least two survey segments at locations corresponding to substantially the same water bottom reflection points (42-48). An amount of time adjustment is determined for at least one of the survey segments based on the discontinuity (52).

Description

METHOD FOR QUANTIFYING WATER COLUMN TIMING AND AMPLITUDE ANOMALIES IN A 3 DIMENSIONAL

SEISMIC SURVEY

Background of Invention

Field of the Invention

[0001] The invention relates generally to the field of marine seismic surveying.

More specifically, the invention relates to a method for static correction of marine seismic data which accounts for irregularities in the sea surface, irregularities in the sea floor and the acoustic velocity of water.

Background Art

[0002] Seismic surveying methods and systems known in the art are used for determining structures of and compositions of rock formations below the earth's surface, among other uses. Seismic surveying generally includes deploying an array of seismic sensors at the surface of the earth in a selected pattern, and selectively actuating a seismic energy source positioned near the seismic sensors at a selected location. The energy source may be an explosive, a vibrator, or in the case of seismic surveying performed in the ocean ("marine seismic surveying"), may be one or more air guns or water guns.

[0003] When the seismic energy source is actuated, seismic energy emanates from the source and travels through the earth formations until it reaches an acoustic impedance boundary in the formations. Acoustic impedance boundaries typically occur where the composition and/or mechanical properties of the earth formation change. Such acoustic impedance boundaries are typically referred to as "bed boundaries". At a bed boundary, some of the seismic energy is reflected back toward the earth's surface, where it may be detected by one or more of the seismic sensors deployed on the surface. Seismic signal processing known in the art has as one of a number of objectives the determination of the depths and geographic locations of bed boundaries below the earth's surface. The depth and location of the bed boundaries are inferred from the travel time of the seismic energy to the bed boundaries and back to the sensors at the surface, as well as from the positions of the seismic sensors and the source. Additional information about the physical characteristics of the earth can be obtained from analysis of the amplitudes of the detected energy.

[0004] Marine seismic surveying known in the art includes having a vessel tow one or more seismic energy sources in the water. The same or a different vessel may tow one or more "streamers." Streamers are cables having arrays of seismic sensors. A streamer may extend behind the tow vessel as much as several kilometers in length, and the sensors are arranged along the cable at spaced apart positions. Typically, a seismic vessel will tow a plurality of such streamers arranged to be separated by selected lateral distances from each other. The arrangement of seismic sensors on laterally separated streamers, and the arrangement of seismic sources are selected to enable relatively complete determination of the geologic structures below the sea floor in three dimensions. The sensors used in streamers are typically hydrophones. Hydrophones are a type of sensor which generates an electrical signal or optical signal corresponding to a change in pressure. Hydrophones known in the art include a transducer such as a piezoelectric crystal, that generates an electrical voltage when compressed. Recording equipment located on the seismic vessel is operatively connected to the hydrophones on the streamers. The recording equipment makes a record, indexed with respect to time since actuation of the one or more seismic sources, of the signal generated by each of the hydrophones.

[0005] After recording, the signals from each of the sensors is processed to determine the structures and compositions of the rock formations below the earth's surface. An important aspect of the processing performed on seismic data in order to determine the structures of subsurface earth formations includes "static" correction. Static correction is performed to adjust the timing of the recorded seismic signals for the seismic energy travel time through strata near the earth's surface. The near-surface strata are characterized both by changing acoustic velocity and by variable thickness related to the surface topography of the earth in the location where the seismic data are recorded.

[0006] In marine seismic surveying, static correction is used to correct the seismic data for the effects of the topography of the sea floor, for water depth variations over time, and for variations in the velocity of water both in depth and in time. One example of a static correction technique known in the art for marine seismic surveys is described in U. S. patent no. 4,992,993 issued to Chambers. The method disclosed in the Chambers '993 patent includes recovering acoustic reflections from discrete discontinuities within the water column above water bottom (ocean floor) reflections. A profile of the velocity distribution within the water column is calculated from conventional velocity analysis.

[0007] As previously explained, three dimensional (3D) marine seismic surveying known in the art includes towing a plurality of laterally spaced apart streamers behind a vessel. The plurality of streamers enables acquiring data representing a large area of subsurface coverage with each pass of the seismic vessel along a "sail line." Static correction of 3D seismic surveys may include error in the static correction due to small errors in timing, changes in water velocity, changes in water depth and other error sources between successive acquisition sequences in adjacent sail lines, or in seismic surveys made over the same area at different times.

[0008] One such method for correcting seismic surveys for timing error and other sources of error in overlapping survey segments and/or in surveys made at different times is disclosed in U. S. patent no. 6,438,069 Bl issued to Ross et al. The method disclosed in the '069 patent includes making a seismic survey over a geographic area at at least two different times. A corresponding reflective event proximate to a producing reservoir is identified in the first and second seismic surveys. The first reflective event is described by a first set of event parameters, and the reflective event in the second survey is identified by a second set of event parameters. A crossequalization function is determined such that when applied to the second set of event parameters, a crossequalized set of event parameters is generated. The crossequalization function is determined when the crossequalized set of event parameters falls below a selected threshold.

[0009] Generally speaking, the method described in the '069 patent seeks to identify corresponding reflective events in time-separated seismic surveys made over a same geographic area which do not change over time, and adjusts one or more of the time-separated surveys such that the time-invariant reflective events are substantially the same in each of the time-separated surveys. Subsurface structures which undergo some seismically identifiable change over time, such as petroleum producing reservoirs which undergo pore-fluid displacement or movement, will be more accurately analyzed. While the method disclosed in the '069 patent may be used to determine water column timing differences between surveys, as suggested therein, the method is not ideally suited to determine water column timing variations to a high degree of precision for the purpose of accurately identifying subsurface structures, nor is the method in the '069 patent primarily intended for determining water column amplitude variations. Accordingly, it is desirable to have a method for calculating water static correction that enables accurate calculation of the water column travel time and amplitude variation at each subsurface location in a survey area.

Summary of Invention

[0010] One aspect of the invention is a method for determining water column timing error between segments of a marine seismic survey. The method according to this aspect includes picking arrival times of a water bottom reflection event in a plurality of seismic signals acquired in a plurality of seismic survey segments. A discontinuity in the picked arrival times is determined between at least two of the plurality of survey segments at locations corresponding to substantially the same water bottom reflection points. An amount of time adjustment is determined for at least one of the survey segments based on the discontinuity.

[0011] Another aspect of the invention is a method for determining amplitude error between segments of a marine seismic survey. The method according to this aspect includes picking amplitudes of a water bottom reflection event in a plurality of seismic signals acquired in a plurality of seismic survey segments. A discontinuity in the picked amplitudes is determined between at least two of the plurality of survey segments at locations corresponding to substantially the same water bottom reflection points. An amount of amplitude adjustment is determined for at least one of the survey segments based on the discontinuity.

[0012] Another aspect of the invention is a method is for determining water column timing error between segments of a marine seismic survey. The method according to this aspect includes picking arrival times of a water bottom reflection event in a plurality of seismic signals. The seismic signals are acquired in a plurality of seismic survey segments. The picked arrival times are normal moveout corrected. A truncated data set is made from adjacent survey segments by eliminating multiple seismic signals corresponding to substantially the same water bottom reflection points. The truncated data set thus includes at most one seismic signal corresponding to each one of a plurality of water bottom reflection points. A first gradient is calculated between the moveout corrected arrival times corresponding to adjacent water bottom reflection points from the truncated data set. In one embodiment, the adjacent water bottom reflection points are along a subsurface grid axis orthogonal to a direction of acquisition (movement of a seismic vessel). A second gradient is then calculated between seismic signals corresponding to adjacent water bottom reflection points from the moveout corrected picked arrival times within each individual survey segment. A second data set is made from these individual segment gradients, which also includes at most one gradient value corresponding to each one of a plurality of water bottom reflection points. A gradient difference between the first and second gradients is calculated. The gradient difference is integrated. The integrated gradient difference is added to the arrival times in the truncated data set to generate a pilot surface for static correction.

[0013] Another aspect of the invention is a method for determining water column amplitude anomalies between segments of a marine seismic survey. The method according to this aspect includes picking amplitudes of a water bottom reflection event in a plurality of seismic signals. The signals are acquired in a plurality of seismic survey segments. The picked amplitudes are corrected, in some embodiments, to account for the expected decay in energy through the water column as well as any directivity corrections for the source and receiver arrays. A truncated data set is made from adjacent survey segments. The truncated data set includes at most one amplitude corresponding to each one of a plurality of water bottom reflection points. A first gradient is calculated between amplitudes corresponding to adjacent water bottom reflection points from the truncated data set. In one embodiment, the adjacent water bottom reflection points are along a subsurface grid axis orthogonal to a direction of acquisition (movement of a seismic vessel). A second gradient is then calculated between seismic signals corresponding to adjacent water bottom reflection points from the amplitude corrected picked amplitudes within each individual survey segment. A second data set is made from these individual segment gradients, which also includes at most one gradient corresponding to each one of a plurality of water bottom reflection points. A gradient difference between the first and second gradients is calculated. The gradient difference is integrated, and the integrated gradient difference is added to the amplitudes in the truncated data set to generate a pilot amplitude surface.

[0014] Other aspects and advantages of the invention will be apparent from the following description and the appended claims. Brief Description of Drawings

[0015] Figure 1 shows a top view of a typical marine seismic acquisition system.

[0016] Figure 2 shows a cross-sectional view through the system of Figure 1, particularly showing one of the streamers and seismic energy travel paths.

[0017] Figure 3 is a graph of normal moveout corrected arrival time of a water bottom reflection event at a plurality of water bottom locations along a selected crossline, for a plurality of different survey segments.

[0018] Figure 4 shows water bottom reflection event arrival times as in Figure 3 wherein multiple arrival times for the same water bottom location are removed.

[0019] Figure 5 shows difference between water bottom reflection event arrival times in adjacent water bottom positions when calculated from the arrival times shown in Figure 4.

[0020] Figure 6 shows a difference between water bottom reflection event arrival times in adjacent water bottom positions for the arrival time data shown in Figure 3 with multiple arrival times for the same water bottom location removed.

[0021] Figure 7 shows differences between the quantities shown in Figure 5 and in

Figure 6.

[0022] Figure 8 shows values of an integration of the quantities shown in Figure 7, the integrated differences representing a cumulative timing correction.

[0023] Figure 9 shows the cumulative timing correction of Figure 8 applied to the arrival time data shown in Figure 4, to develop a pilot surface for static correction.

[0024] Figure 10 is a flow chart of an embodiment of a method for static correcting seismic signal amplitudes.

Detailed Description [0025] Figure 1 shows a top view of a typical marine seismic data acquisition system in use. The system in Figure 1 may be used to acquire seismic data that may be processed by a method according to the invention. The system includes a seismic vessel 22, shown moving in a selected direction indicated by reference designator D. The vessel 22 is shown towing a plurality of streamers 28. Each streamer 28 includes a plurality of seismic sensors, usually hydrophones, at spaced apart positions along each streamer 28. The seismic sensors are each shown generally by reference numeral 29. The vessel, 22 is also shown towing a seismic energy source 36, which may be an air gun array of any type well known in the art. While in the present embodiment the seismic source 36 is an air gun array, any other type of seismic energy source may be used with the invention, and therefore the type of seismic energy source is not intended to limit the scope of the invention. The system embodiment of Figure 1 shows the vessel 22 pulling five streamers 28, however, marine seismic acquisition systems known in the art use different numbers of streamers and seismic sources. Therefore, the number of streamers shown in Figure 1, and the lateral spacing between adjacent streamers is not intended to limit the scope of the invention.

[0026] The vessel 22 also typically includes a seismic recording system 24 which may include signal recording devices (not shown separately) for recording signals generated by the sensors 29, a seismic source controller (not shown separately) for controlling actuation of the source 36, and navigational equipment (not shown separately) used to determine the positions of the vessel 22 and the positions of each sensor 29 and the source 36 during the recording of seismic data.

[0027] In acquiring seismic signals, the navigation equipment (not shown separately) in the recording system 24 makes a determination of the position of the source 36 and each one of the sensors 29. The source and sensor positions are recorded, the source 36 is periodically actuated, and a record of the signals generated by each of the sensors 29 is made in the recording system 24. The signal recording is typically indexed with respect to time of each actuation of the source 36. The source 36 is actuated a plurality of times, and signal recordings are made, as the vessel 22 travels along the selected direction D.

[0028] Movement of the vessel 22 along a single path in the selected direction D is referred to as a "sail line." Positions of the sensors 29 are referenced with respect to the sail line as being a particular distance along the subline direction S and the crossline direction X. A reference or index for the particular distance may be the position of the source 36 at the time of each actuation thereof. Alternatively the positions of both the source 36 and each of the sensors 29 with respect to time may be referenced to a known geographic position.

[0029] In a typical marine seismic survey operation, the vessel 22 may move several kilometers to several tens of kilometers along a single sail line, during which vessel movement seismic signals corresponding to a plurality of source actuations are recorded. The signals generated and recorded along a sail line thus correspond to the midpoints of a subsurface area roughly outlined by the arrangement of the source 36 and sensors 29 shown in Figure 1, extended along the length of the sail line. At the end of a sail line, the vessel 22 may reverse direction and travel a new sail line, typically in the same direction, D, or in the opposite direction (parallel to D). The new sail line is generally laterally offset from the previous sail line such that there is only slight, or no, lateral overlap of the subsurface area covered with respect to the mid-point positions of the sensors 29 and source 36 between successive (adjacent) sail lines. The process of surveying along such sail lines is repeated until a selected subsurface area is surveyed.

[0030] Each time the source 36 is actuated, seismic energy travels laterally through the water 14 from the source 36 to the sensors 29 where it may be detected. Methods for determining components in the detected seismic signals resulting from these so-called "direct arrivals" are known in the art. The seismic energy also travels downwardly through the water as will be explained below with reference to Figure 2. [0031] Figure 2 is a cross section of the marine seismic acquisition system in a vertical plane including one of the streamers 28. The seismic energy is shown traveling along ray paths 12, downwardly through the water 14 where it is reflected by the water bottom 10 (sea floor). The reflected energy travels upwardly where it is detected by the sensors 29. It should be noted that the term "sea floor" is generally used to refer to the bottom of an ocean body of water. As used in this description of the invention, however, the term "sea floor" is intended to include within its scope the bottom of any body of water in which a marine seismic survey is performed, including, for example, lakes and very large rivers. Accordingly, any reference to the term "sea floor" is not intended to limit the invention to use with seismic data acquired in ocean water.

[0032] As is known in the art, the seismic energy also travels downwardly through the earth formations 16 below the sea floor 10 until it reaches acoustic impedance boundaries (not shown) in the formations 16. Some seismic energy is reflected from the boundaries and travels upwardly until it is detected by the sensors. The seismic energy that is reflected from the formation boundaries is used to determined structure and composition of the formations 16, among other uses. The seismic energy that travels downwardly through the formations 16 and reflected from boundaries (not shown) is not shown in Figure 2 in order to simplify explanation of the invention. However, the seismic energy must travel both downwardly and upwardly through the water 14, therefore the travel time of the seismic energy through the water 14 at any location within the seismic survey area must be accurately determined in order to accurately determine the travel time of the seismic energy through the earth formations 16 to the boundaries (not shown). Accurate water travel time determination includes accurate determination of both the water velocity and the height (depth) of the water column at each survey position within the survey area. Methods according to the invention are used to determine accurately the water column travel time at each survey position. [0033] In Figure 2, some of the sensors 29 are shown as having spacing from the source 36, along one of the streamers 28, indicated individually by reference numerals FI, F2, F3, F4. The seismic energy is shown traveling from the source 36 downwardly through the water 14 where it is reflected, at points Ml, M2, M3, M4 along the sea floor 10. The points Ml, M2, M3, M4 are located approximately half the distance between the source 36 and the corresponding one of the receivers 29 spaced at FI, F2, F3, F4. The source to sensor distance is known in the art as "offset." The seismic energy travels upwardly, where it is ultimately detected by the corresponding ones of the sensors 29. Seismic energy travels along similar paths between the source 36 and the sensors on the other streamers (not shown in Figure 2).

[0034] Seismic energy reflecting from the sea floor 10 is shown being detected only by ones of the sensors 29 that are relatively close to the source 36. While seismic energy is reflected from the sea floor 10 so that it may be detected by substantially all of the sensors 29, typically the signals from those sensors 29 disposed relatively close to the source 36 are used in the invention.

[0035] As may be inferred by the seismic energy travel paths shown in Figure 2 and the description above, the seismic energy which is reflected from the sea floor 10 corresponds to reflection from known positions along the sea floor 10. The known positions correspond, as previously explained, to locations half way between the positions of the source 36 and the receivers 29 at the time of actuation of the source 36. For the one of the streamers 28 shown in Figure 2, these positions are indicated at Ml, M2, M3, M4. Similar sea floor reflection points exist for signals detected by the sensors 29 on the other ones of the streamers (such as shown at 28 in Figure 1).

[0036] In some embodiments of a method according to the invention, seismic signals are selected, from a subset of all the sensor signals, for processing. Preferably, the selected signals correspond to sea floor reflection points spaced apart by the amount of movement of the vessel 22 in the time interval between successive actuations of the source 36. In this way, substantially the entire sea floor 10 along each sail line has at least one sensor signal, and more preferably not substantially more than one such sensor signal, corresponding to any individual sea floor reflection point for which data are processed according to the invention. The geographic spacing between such sea floor reflection points will be related to, among other things, the speed of motion of the vessel 22, the time interval between successive actuations of the source 36, and the spacing between the source 36 and each of the sensors 29.

[0037] In one embodiment of the invention, a Cartesian coordinate grid comprising a plurality of individual grid segments is generated for a survey area in which static correction is to be applied to seismic signals. The Cartesian coordinate grid preferably has a spacing about equal to the expected distance between the sea floor reflection points in the seismic survey. In one embodiment of the invention, the geographic position of each of the sea floor reflection points (e.g. Ml, M2, M3, M4) is determined from the geographic position of the source 36 and each sensor 29 for each source actuation. For each grid segment, the recorded sensor signal corresponding to a sea floor reflection point closest to a center of each grid element is selected and assigned to a calculation "bin" corresponding to that grid segment.

[0038] After a sensor signal is assigned to each grid "bin", an arrival time of the seismic energy corresponding to reflection from the sea floor 10 is determined for each seismic signal. The arrival time may be determined or "picked" based on an amplitude characteristic of the seismic signal, such as by "zero crossing", amplitude peak or amplitude trough. Irrespective of the amplitude characteristic used to select the arrival time of the sea floor reflection in each seismic signal, it is important that the selected amplitude characteristic be the same for each signal for which the arrival time is picked.

[0039] For each seismic signal, the value of offset (shown in Figure 2 as FI, F2,

F3, F4) is also noted. The value of offset for each signal is then used to perform a normal moveout correction on each signal. Normal moveout correction, as known in the art, adjusts the arrival time of a selected event in a seismic signal to the arrival time which would obtain if the source 36 and the receiver for which the signal is being processed were collocated directly above the sea floor reflection point corresponding to that signal. In the time domain, normal moveout correction is based on offset between the source and the sensor, and the seismic velocity of the medium through which the seismic energy travels. In the invention, the medium is substantially all water, and the velocity thereof may be readily measured or estimated. Methods for performing normal moveout correction are well known in the art. Moveout corrected arrival times for the sea floor reflection event can then be written to or included in each corresponding bin in the grid.

[0040] The process described above is repeated for each sail line, each survey segment or each survey made over the grid area. Typically, there will be at least some geographic (subsurface coverage) overlap between sail lines or surveys made in a particular geographic area. Referring to Figure 3, sea floor reflection event times (time on the ordinate axis) normalized to a minimum time value are shown for four different survey segments (partial coverage of a selected survey area) made in a selected geographic area. Each survey segment is made at the same crossline position (X in Figure 1). The coordinate axis in the graph of Figure 3 indicates position along the crossline direction (X in Figure 1) for each survey segment, SI, S2, S3, S4. Each survey segment SI, S2, S3, S4 may correspond to a single actuation of the source (36 in Figure 2), and the crossline position of each point in each segment SI, S2, S3, S4 corresponds to the sea floor reflection point for each sensor signal.

[0041] As can be observed in Figure 3, there are apparent time discontinuities between adjacent survey segments. The adjacent survey segments also include some overlapping sea floor reflection points. The time discontinuities represent timing error associated with static correction for the water column. In one embodiment of the invention, sea floor reflection event arrival times in multiple survey segments that correspond to substantially the same sea floor reflection points are eliminated for this part of the process. Referring to Figure 4, the reflection times previously shown in Figure 3 are now shown such that multiple reflection event arrival times corresponding to the same position in the crossline direction are eliminated. Elimination of overlapping reflection point arrival times may be referred to as "truncating" the data of one or more of the survey segments. The particular survey segment from which to eliminate overlapping arrival times is not important. While the data shown in Figures 3 include multiple arrival times for some sea floor reflection points resulting from overlap of the coverage of the acquisition system, multiple arrival times for the same sea floor reflection points may also arise from, for example, multiple actuations of the source (36 in Figure 2) along a single sail line. The principles of the invention are equally applicable to multiple surveys made over overlapping geographic areas at different times, or to portions of multiple sail lines which overlap within a single survey operation. After truncating the overlapping data points, the survey segments can be merged into a single "truncated" data set. While the example of Figure 3 shows four such survey segments truncated to form the truncated data set, the method of the invention only requires at least two such survey segments. Accordingly, the total number of survey segments is not intended to limit the scope of the invention. In the present embodiment, a difference between sea floor reflection event arrival times, after normal moveout correction ("corrected arrival times"), at adjacent sea floor reflection points is calculated for the truncated data set. Referring to Figure 5, the corrected arrival time differences between adjacent sea floor reflection points are shown as the points forming curve Gl. The corrected arrival time differences between adjacent sea floor reflection points in the truncated data set are referred to collectively herein as a first gradient. Where there are time discontinuities between the survey segments, as can be observed between adjacent survey segments shown in Figure 4, in Figure 5 the first gradient curve Gl shows distinct peak values, such as at Dl, D2 and D3. [0043] In the present embodiment, corrected arrival time differences between adjacent sea floor reflection points are then calculated from the original corrected arrival time data, as in the previous step, but without eliminating overlapping sea floor points (without truncating) in adjacent survey segments. The differences in corrected arrival times in the untruncated data are referred to as a second gradient. Multiple second gradient values for any sea floor reflection points may be eliminated by truncating the second gradient data sets corresponding to each survey segment. Referring to Figure 6, a second gradient curve G2 is formed from points representing corrected arrival time differences between adjacent sea floor reflection points within each survey segment. The corrected arrival time difference values used to generate the curve G2 in Figure 6 were calculated from the sea floor reflection event corrected arrival times shown in Figure 3, without eliminating the overlapping data, and without calculating arrival time differences between reflection points disposed in different survey segments. Overlapping time difference points between survey segments may be truncated or eliminated in some embodiments, as previously explained.

[0044] Next, differences between the gradients shown in Figure 5 (first gradient), and the time differences shown in Figure 6 (second gradient) are calculated for each sea floor reflection point, without regard to whether there are overlapping data from multiple survey segments. Difference between gradients (gradient difference) is shown at curve G3 in Figure 7. As can be observed in Figure 7, the difference between gradients includes peaks D4, D5, D6 at substantially the same sea floor reflection points as where the time discontinuities are located, the discontinuities being as shown in Figure 3.

[0045] Next, the differences of gradients as shown in Figure 7 are integrated. An example of integration is shown in Figure 8. The integration values shown at curve I in Figure 8 represent a cumulative time adjustment to be applied to each corrected arrival time corresponding to each sea floor reflection point in the truncated data set of Figure 4. Adding the cumulative time adjustment to the corrected arrival time for each sea floor reflection point of the truncated data set provides a corrected sea floor reflection time for each sea floor reflection point as shown at curve T in Figure 9. The corrected sea floor reflection event arrival times for each sea floor reflection point, as shown in curve T in Figure 9 may then be used as a pilot surface to static correct seismic travel times corresponding to each sea floor reflection point. Methods for statically correcting marine seismic data using sea floor reflection event arrival times are well known in the art.

[0046] The example embodiments described above with respect to Figures 3 through 9 include calculating arrival time gradients and gradient differences along a direction orthogonal to a direction along which the seismic vessel (22 in Figure 1) travels during acquisition of seismic data. It should be clearly understood that for purposes of defining the scope of the invention that the direction of acquisition and determining gradients is not important. Time differences may be determined between adjacent sea floor reflection points along any selected direction. Gradients can also be calculated with respect to the true mid-point position of each trace, or with respect to the bin center, which has been previously described.

[0047] Another aspect of the invention is a method for correcting seismic signal amplitudes for error between segments of a seismic survey, or between different seismic surveys made at different times. Embodiments of a method according to this aspect of the invention are similar to the method of the previous aspect of the invention related to methods for determining timing error between seismic surveys or seismic survey segments. The seismic signals used in embodiments of a method according to the present aspect of the invention may be the same seismic signals, or different seismic signals as used in the previous aspect of the invention. In the present aspect of the invention, seismic signals are acquired as explained above with reference to Figures 1 and 2. Selected ones of the recorded seismic signals which correspond generally to at most one sea floor reflection point in a preselected survey grid are used in the present aspect, just as in the previous aspect of the invention. [0048] Referring to Figure 10, an amplitude of each sea floor reflection event in each selected seismic signal is measured or determined at 38. The amplitude of the sea floor reflection event in selected seismic signals are then corrected for travel path length (amplitude decay corrected), at 40, to provide an amplitude which would obtain if the sensor (29 in Figure 1) and source (36 in Figure 1) were collocated directly above the sea floor reflection point corresponding to the particular seismic signal. Methods for correcting signal amplitude for differences in travel path length and orientation to the source and receiver arrays are well known in the art.

[0049] At 42, the amplitude corrected signals from a plurality of survey segments

(or a plurality of at least partially overlapping surveys) are truncated such that within a preselected sea floor survey grid (similar to the grid described fpr methods according to the previous aspect), at most one amplitude value is used. At 44, a difference in corrected amplitude between adjacent sea floor reflection points is determined. The differences in corrected amplitudes is referred to as the first gradient.

[0050] At 46, a difference in corrected amplitude between adjacent sea floor reflection points within each survey segment is determined. The amplitude differences are merged into a single data set in which multiple values corresponding to any one sea floor reflection point are eliminated. The amplitude differences thus determined are referred to as a second gradient.

[0051] At 48, a difference between the first gradient and the second gradient is determined for each sea floor reflection point. The gradient difference is integrated, at 50 to determine a cumulative amplitude correction to add to the amplitudes of the corrected amplitudes determined at 40. Finally, at 52, the integrated gradient difference is added to the corrected amplitudes determined at 40 to obtain a pilot amplitude set. [0052] Methods according to the various aspects of the invention can provide seismic data which are corrected for errors in timing and amplitude between adjacent segments of a single survey or between surveys made at different times.

[0053] While the invention has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of this disclosure, will appreciate that other embodiments can be devised which do not depart from the scope of the invention as disclosed herein. Accordingly, the scope of the invention should be limited only by the attached claims.

Claims

What is claimed is:

[cl] A method for determining timing error between segments of a marine seismic survey, comprising: picking arrival times of a water bottom reflection event in a plurality of seismic signals acquired in a plurality of seismic survey segments; determining a discontinuity in the picked arrival times between at least two of the plurality of survey segments at locations corresponding to substantially the same water bottom reflection points; and determining an amount of time adjustment for seismic signals in at least one of the survey segments based on the discontinuity.

[c2] The method as defined in claim 1 wherein the determining the amount of time adjustment comprises: moveout correcting the picked arrival times; generating a truncated data set from adjacent portions of the at least two survey segments, the truncated data set comprising at most one seismic signal corresponding to each one of a plurality of water bottom reflection points in a survey area; calculating a first gradient between the moveout corrected arrival times corresponding to adjacent water bottom reflection points from the truncated data set; calculating a second gradient between the moveout corrected arrival times corresponding to adjacent water bottom reflection points from the seismic signals within individual survey segments; calculating a gradient difference between the first and second gradients; and integrating the gradient difference.

[c3] The method of claim 1 wherein the survey segments comprise signals recorded in each of a plurality of actuations of a seismic energy source along a single sail line of a seismic vessel. [c4] The method of claim 1 wherein the survey segments comprise signals recorded in each of a plurality of sail lines of a seismic vessel.

[c5] The method of claim 1 wherein the arrival times are picked from a subset of all seismic signals acquired in each of the survey segments, the subset comprising signals from seismic sensors spaced from a seismic source such that movement of a seismic vessel and timing between successive actuations of the seismic source result in water bottom reflection points having substantially only one seismic signal corresponding to each position in a preselected water bottom survey grid.

[c6] A method for determining amplitude anomalies between segments of a marine seismic survey, comprising: picking amplitudes of a water bottom reflection event in a plurality of seismic signals acquired in a plurality of seismic survey segments; determining a discontinuity in the picked amplitudes between at least two of the plurality of survey segments at locations corresponding to substantially the same water bottom reflection points; and determining an amount of amplitude adjustment for at least one of the survey segments based on the discontinuity.

[c7] The method as defined in claim 6 wherein the determining the discontinuity comprises: correcting the picked amplitudes; generating a truncated data set from adjacent survey segments, the truncated data set comprising at most one seismic signal corresponding to each one of a plurality of water bottom reflection points in a survey area; calculating a first gradient between corrected amplitudes corresponding to adjacent water bottom reflection points from the truncated data set; calculating a second gradient between corrected amplitudes corresponding to adjacent water bottom reflection points from the seismic signals within individual survey segments; calculating a gradient difference between the first and second gradients; and integrating the gradient difference.

[c8] The method of claim 6 wherein the survey segments comprise signals recorded in each of a plurality of actuations of a seismic energy source along a single sail line of a seismic vessel.

[c9] The method of claim 6 wherein the survey segments comprise signals recorded in each of a plurality of sail lines of a seismic vessel.

[clO] The method of claim 6 wherein the arrival times are picked from a subset of all seismic signals acquired in each of the survey segments, the subset comprising signals from seismic sensors spaced from a seismic source such that movement of a seismic vessel and timing between successive actuations of the seismic source result in water bottom reflection points having substantially only one seismic signal corresponding to each position in a preselected water bottom survey grid.

[ell] A method for determining timing error between segments of a marine seismic survey, comprising: picking arrival times of a water bottom reflection event in a plurality of seismic signals acquired in a plurality of seismic survey segments; moveout correcting the picked arrival times; generating a truncated data set from adjacent survey segments, the truncated data set comprising at most one seismic signal corresponding to each one of a plurality of water bottom reflection points in a survey area; calculating a first gradient between the moveout corrected arrival times corresponding to adjacent water bottom reflection points from the truncated data set; calculating a second gradient between moveout corrected arrival times corresponding to adjacent water bottom reflection points from the moveout corrected picked arrival times within individual survey segments; calculating a gradient difference between the first and second gradients; integrating the gradient difference; and adding the integrated gradient difference to the arrival times in the truncated data set to generate a pilot surface for static correction.

[cl2] The method of claim 11 wherein the survey segments comprise signals recorded in each of a plurality of actuations of a seismic energy source along a single sail line of a seismic vessel.

[cl3] The method of claim 11 wherein the survey segments comprise signals recorded in each of a plurality of sail lines of a seismic vessel.

[cl4] The method of claim 11 wherein the arrival times are picked from a subset of all seismic signals acquired in each of the survey segments, the subset comprising signals from seismic sensors spaced from a seismic source such that movement of a seismic vessel and timing between successive actuations of the seismic source result in water bottom reflection points having substantially only one seismic signal corresponding to each position in a preselected water bottom survey grid.

[cl5] A method for determining amplitude anomalies between segments of a marine seismic survey, comprising: picking amplitudes of a water bottom reflection event in a plurality of seismic signals acquired in a plurality of seismic survey segments; amplitude correcting the picked amplitudes ; generating a truncated data set from adjacent survey segments, the truncated data set comprising at most one amplitude corresponding to each one of a plurality of water bottom reflection points in a survey area; calculating a first gradient between corrected amplitudes corresponding to adjacent water bottom reflection points from the truncated data set; calculating a second gradient between corrected amplitudes corresponding to adjacent water bottom reflection points from the amplitude corrected picked amplitudes within individual survey segments; calculating a gradient difference between the first and second gradients; integrating the gradient difference; and adding the integrated gradient difference to the amplitudes in the truncated data set to generate a pilot amplitude surface.

[cl6] The method of claim 15 wherein the survey segments comprise signals recorded in each of a plurality of actuations of a seismic energy source along a single sail line of a seismic vessel. [cl7] The method of claim 15 wherein the survey segments comprise signals recorded in each of a plurality of sail lines of a seismic vessel.

[cl8] The method of claim 15 wherein the picked are picked from a subset of all seismic signals acquired in each of the survey segments, the subset comprising signals from seismic sensors spaced from a seismic source such that movement of a seismic vessel and timing between successive actuations of the seismic source result in water bottom reflection points having substantially only one seismic signal corresponding to each position in a preselected water bottom survey grid.

[cl9] A method for marine seismic surveying, comprising: moving a seismic vessel towing at least one sensor streamer and at least one source in a selected pattern through a body of water; periodically actuating the at least one source; recording seismic signals detected by sensors on the at least one streamer; picking arrival times of a water bottom reflection event in a plurality of seismic signals acquired in a plurality of seismic survey segments; moveout correcting the picked arrival times; generating a truncated data set from adjacent survey segments, the truncated data set comprising at most one seismic signal corresponding to each one of a plurality of water bottom reflection points in a survey area; calculating a first gradient between moveout corrected arrival times corresponding to adjacent water bottom reflection points from the truncated data set; calculating a second gradient between moveout corrected arrival times corresponding to adjacent water bottom reflection points from the seismic signals within individual survey segments; calculating a gradient difference between the first and second gradients; integrating the gradient difference; and adding the integrated gradient difference to the arrival times in the truncated data set to generate a pilot surface for static correction.

[c20] The method of claim 19 wherein the survey segments comprise signals recorded in each of a plurality of actuations of a seismic energy source along a single sail line of the seismic vessel.

[c21] The metliod of claim 19 wherein the survey segments comprise signals recorded in each of a plurality of sail lines of the seismic vessel. [c22] The method of claim 19 wherein the arrival times are picked from a subset of all seismic signals acquired in each of the survey segments, the subset comprising signals from seismic sensors spaced from the at least one seismic source such that movement of the seismic vessel and timing between successive actuations of the seismic source result in water bottom reflection points having substantially only one seismic signal corresponding to each position in a preselected water bottom survey grid.

[c23] A method for marine seismic surveying, comprising: moving a seismic vessel towing at least one sensor streamer and at least one source in a selected pattern through a body of water; periodically actuating the at least one source; recording seismic signals detected by sensors on the at least one streamer; picking amplitudes of a water bottom reflection event in a plurality of seismic signals acquired in a plurality of seismic survey segments; amplitude correcting the picked amplitudes; generating a truncated data set from adjacent survey segments, the truncated data set comprising at most one amplitude corresponding to each one of a plurality of water bottom reflection points in a survey area; calculating a first gradient between corrected amplitudes corresponding to adjacent water bottom reflection points from the truncated data set; calculating a second gradient between corrected amplitudes corresponding to adjacent water bottom reflection points from picked amplitudes within individual survey segments; calculating a gradient difference between the first and second gradients; integrating the gradient difference; and adding the integrated gradient difference to the amplitudes in the truncated data set to generate a pilot amplitude surface. [c24] The method of claim 23 wherein the survey segments comprise signals recorded in each of a plurality of actuations of the seismic energy source along a single sail line of the seismic vessel.

[c25] The method of claim 23 wherein the survey segments comprise signals recorded in each of a plurality of sail lines of the seismic vessel.

[c26] The method of claim 23 wherein the picked are picked from a subset of all seismic signals acquired in each of the survey segments, the subset comprising signals from seismic sensors spaced from the seismic source such that movement of the seismic vessel and timing between successive actuations of the seismic source result in water bottom reflection points having substantially only one seismic signal corresponding to each position in a preselected water bottom survey grid.