Sparse hyperbolic inversion of seismic data

Details Sparse hyperbolic inversion of seismic data Sparse hyperbolic inversion of seismic data

1999-06-25

2001-04-17

Patidar, Jay (Department: 2862)

Data processing: measuring, calibrating, or testing

Measurement system in a specific environment

Earth science

C702S014000

Reexamination Certificate

active

06219621

ABSTRACT:

FIELD OF THE INVENTION
This invention relates generally to the field of seismic prospecting and, more particularly, to seismic data processing. Specifically, the invention is an improved method for suppressing multiple reflections, especially water-bottom multiples, in seismic data.
BACKGROUND OF THE INVENTION
In the oil and gas industry, seismic prospecting techniques are commonly used to aid in the search for and evaluation of subterranean hydrocarbon deposits. In seismic prospecting, a seismic source is used to generate a physical impulse (known as a “seismic signal”) that propagates into the earth and is at least partially reflected by subsurface seismic reflectors (i.e., interfaces between underground formations having different acoustic impedances). The reflected signals (known as “reflections”) are detected and recorded by seismic receivers located at or near the surface of the earth, in an overlying body of water, or at known depths in boreholes, and the resulting seismic data may be processed to yield information relating to the subsurface formations.
The seismic energy recorded by each seismic receiver is known as a “seismic data trace.” Seismic data traces typically contain both the desired primary reflections and one or more unwanted noise components that can interfere with the processing and interpretation of the desired the primary reflections. These noises may include random noise, multiple reflections, converted waves, and surface waves. Conventional noise suppression methods typically exploit some difference between the noise to be suppressed and the primary reflections (e.g., different frequency band or different dip). Frequently, however, these differences are not sufficient to allow a clean separation between the desired primary reflections and the targeted noise.
In the case where the unwanted noise is due to multiple reflections (hereinafter referred to as “multiples”), some prior art noise suppression techniques take advantage of differences between the normal moveout (NMO) velocities (hereinafter referred to as “moveout velocities”) of the primary reflections and the multiples in a common-midpoint (CMP) gather of seismic data traces (i.e., a set of seismic data traces having the same midpoint but different source-to-receiver offset distances). Methods such as the Radon transform and velocity-stack inversion can be particularly useful in distinguishing the primary reflections from the multiples. These methods decompose the seismic data traces in the CMP gather into a set of moveout parabolas or hyperbolas in which the multiples can be identified and suppressed on the basis of moveout velocity differences.
Conventional velocity-based multiple suppression techniques can often be used to suppress multiples such as the peg-leg multiple illustrated in FIG.
1
.
FIG. 1
shows a seismic source
10
and a seismic receiver
12
located at or near the surface
14
of a body of water
16
. A subsurface reflector
18
is located a distance below the water bottom
20
.
FIG. 1
shows two raypaths from seismic source
10
to seismic receiver
12
. (For simplicity,
FIG. 1
assumes a constant seismic velocity from source
10
to receiver
12
for both raypaths.) For primary raypath
22
(solid line), the seismic signal propagates downwardly from seismic source
10
to reflection point
24
on subsurface reflector
18
and then upwardly to seismic receiver
10
. In other words, for primary raypath
22
, the seismic signal is reflected only once. However, for peg-leg multiple raypath
26
(dashed line), the seismic signal is reflected three times, once at reflection point
28
on subsurface reflector
18
, once at point
30
on the surface
14
of body of water
16
, and once at point
32
on the water bottom
20
.
As would be well known to persons skilled in the art, multiples resulting from reverberation of the seismic energy within the water layer (hereinafter referred to as “water-bottom multiples”) may involve several reflections in the water layer, and these reflections may occur at either or both ends of the raypath. Water-bottom multiples may have amplitudes comparable to the amplitude of the related primary reflection. Therefore, it is important to remove these multiples from the seismic data. In shallow water, water-bottom multiples can be easily removed by deconvolution techniques. In deep water, conventional velocity-based multiple suppression techniques can satisfactorily eliminate these multiples. However, velocity-based techniques are unable to satisfactorily suppress water-bottom multiples in intermediate water depths—ranging from about 100 meters to about 300 meters—because the difference between the moveout velocity of the primary reflection and the moveout velocity of the multiple is insufficient for satisfactory separation.
Prior art multiple suppression techniques do not make use of all available information. For example, an estimate of the water depth is usually available, either from direct measurements (e.g., by using sonar), or by examination of near-offset traces for multiple reflections. In the marine environment, the shape of the seismic energy generated by the source (i.e., the seismic wavelet) is usually known. Prior art multiple suppression techniques do not make use of this information in the multiple suppression process.
From the foregoing, it can be seen that there is a need for an improved method for suppressing multiple reflections, especially water-bottom multiples occurring in intermediate water depths, during processing of seismic data. Preferably, such a method should make use of any additional information which may be available, such as the water depth and the shape of the seismic wavelet. The present invention satisfies this need.
SUMMARY OF THE INVENTION
The present invention is an improved velocity-based method for suppressing multiple reflections in seismic data. The inventive method includes extra parameters in its model domain for the inversion process. For multiple elimination in intermediate water depths, the two-way water-bottom reflection traveltime and the water-bottom reflection coefficient are used in addition to the conventional subsurface velocity model. The permissible ranges of these two extra parameters may be obtained from the autocorrelation of short offset seismic data and from a priori geological information. Because these water-bottom parameters are explicitly included in the inversion model, reliance on differences in moveout velocity between the primary reflections and the multiples is obviated.
The inventive method inverts the data gather to determine a best-fit sparse representation in the model domain. The optimization is done using a L
p
-norm to be consistent with a sparse model representation. Preferably, a Lanczos subroutine is used in the matrix inversion step of the inversion process. In a preferred embodiment, a limited number of traces are used in the initial steps of the inversion, while in the later steps, a smaller model domain is used. This leads to an increase in the computation speed and, accordingly, a reduction in computing time and cost. After the data gather has been inverted into the model domain, one or more filtering operations are used to suppress the multiples, and the filtered model domain data are then transformed back to the data domain.
The inventive method may also be used to suppress linear noises, such as refraction arrivals, and dispersive noises, such as ground roll on land seismic data, by making appropriate modifications to the model domain.


REFERENCES:
patent: 5132938 (1992-07-01), Walters
patent: 5821753 (1999-02-01), Stanley et al.
patent: 5873050 (1999-02-01), Schneider, Jr. et al.
Hampson, D., 1986, “Inverse velocity stacking for multiple elimination,” Journal of the Canadian Society of Exploration Geophysicists, 22, pp. 44-55.
Levin, F. K., and Shah, P. M., 1977, “Peg-leg multiples and dipping reflectors,” Geophysics, 42, pp. 957-981.
Nichols, D., 1994, “Velocity-stack inversion using Lpnorms,” Stanford Exploration Project Report 82.
Shuey, R. T., 1985, “A sim

Affiliated with

Also associated with

Rating

Sparse hyperbolic inversion of seismic data does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Sparse hyperbolic inversion of seismic data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Sparse hyperbolic inversion of seismic data will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2434926