System and method for three-dimensional iterative filtering of scattered waves in cross spread seismic systems

Patent metadata

Field	Value
Publication number	US-9541660-B2
Application number	US-201314040868-A
Country	US
Kind code	B2
Filing date	Sep 30, 2013
Priority date	Oct 1, 2012
Publication date	Jan 10, 2017
Grant date	Jan 10, 2017

Abstract

A system and method are provided for filtering scatterer noise energy from land seismic waves using three dimensional (3D) iterative filtering in a cross-spread source-receiver pattern. The system and method isolate scatterer noise energy associated with a scatterer based on a filtering process performed using a scatterer referential time delay. Then, the isolated scatterer noise energy can be subtracted from the surface wave data. This process can be repeated for each scatterer in the covered area to remove each of their contributions, and can be performed, for example, after a preliminary filtering of the surface wave data using 3D fk filtering or the like.

First claim

We claim:

1. A method for filtering scatterer noise energy from land seismic waves in a cross-spread source-receiver pattern, the method comprising:
recording waves received at receivers disposed relative to a shot pattern in a cross-spread source receiver pattern;
filtering surface waves from the recorded waves to produce surface wave filtered data;
determining a plurality of scatterers associated with the recorded waves;
selecting a scatterer from the plurality of scatterers;
isolating scatterer noise energy generated by the scatterer by filtering the surface wave filtered data using a scatterer referential time delay associated with the scatterer, wherein the scatterer referential time delay is calculated as
T RD =( d dr,2B +d sd,2B −d dr,2A −d sd,2A )/ v SM , wherein

d dr,B is a first distance from said selected scatterer to a first selected receiver,
d sd,B is a second distance from said selected scatterer to a first selected source,
d dr,A is a third distance from said selected scatter to a second selected receiver, and
d sd,A is a fourth distance from said selected scatterer to a second selected source;
subtracting said scatterer noise energy from said surface wave filtered data; and
repeating each of the steps of selecting, isolating and subtracting for each of the plurality of scatterers.

2. The method according to claim 1 , further comprising:
transmitting seismic signals from one or more sources using a cross spread source-receiver design; and
receiving said transmitted signals at one or more receivers.

3. A method for filtering scatterer noise energy from land seismic waves, the method comprising:
filtering surface waves from acquired land seismic data to produce surface wave filtered data;
selecting a scatterer from a plurality of scatterers;
isolating scatterer noise energy from the scatterer based on a filtering in a scatterer referential time delay;
subtracting said scatterer noise energy from said surface wave filtered; and
repeating each of the steps of selecting, isolating and subtracting for each of the plurality of scatterers.

4. The method according to claim 3 , wherein said step of filtering comprises:
filtering said surface waves using one of a 3D fk filtering method and a singular value decomposition filtering method.

5. The method according to claim 3 , wherein said step of selecting further comprises:
muting said surface waves not scattered by said selected scatterer.

6. The method according to claim 3 , wherein said step of isolating scatterer noise energy from the first scatterer comprises:
determining a position of said selected scatterer;
determining a surface mean velocity applicable to said selected scatterer;
determining a scattered waves arrival-time for said selected scatterer;
windowing scattered waves for said selected scatterer;
determining, for each trace, a scatterer referential delay for said selected scatterer;
adding said scatterer referential delay to each of said traces associated with said selected scatterer to obtain applied scatterer referential data;
performing filtering on said applied scatterer referential data; and
subtracting said scatter scatterer referential delay from said filtered applied scatterer referential data to isolate said scatterer noise energy from the first scatterer.

7. The method according to claim 6 , wherein said step of determining said position of said selected scatterer comprises:
using at least one semblance algorithm to determine said position of said selected scatterer.

8. The method according to claim 6 , wherein the step of determining said scatterer referential delay comprises evaluating the following expression—
Scatterer Referential Delay, τ RD =( d dr,B +d sd,B −d dr,A −d sd,A )/ v SM , wherein

d dr,B is a first distance from said selected scatterer to a first selected receiver,
d sd,B is a second distance from said selected scatterer to a first selected source,
d dr,A is a third distance from said selected scatter to a second selected receiver, and
d sd,A is a fourth distance from said selected scatterer to a second selected source.

9. The method according to claim 6 , wherein the step of adding said scatterer referential delay creates substantially circular shaped wave data.

10. The method according to claim 6 , wherein the step of performing filtering on said applied scatterer referential data comprises:
using one of a Full 3D fk filtering method and a singular value decomposition filtering method on said applied scatterer referential data.

11. The method according to claim 3 , further comprising:
determining a plurality of different surface velocities for a first selected scatterer, and iteratively determining said scatterer noise energy based on each of said plurality of different surface velocities.

12. The method according to claim 11 , wherein the different surface velocities comprise a Rayleigh mode distribution of said surface velocities.

13. The method according to claim 3 , further comprising:
determining whether said scatterer noise energy exceeds a first noise threshold, and performing said subtraction only if said scatterer noise energy exceeds the first noise threshold.

14. The method according to claim 3 , further comprising:
transmitting seismic signals from one or more sources using a cross spread source-receiver design; and
receiving said transmitted signals at one or more receivers.

15. A method for filtering scatterer noise energy from seismic data, the method comprising:
selecting a scatterer from a plurality of scatterers;
determining scatterer noise energy generated by the scatterer in seismic data;
subtracting said scatterer noise energy from the seismic data;
filtering the seismic data prior to selecting, determining and subtracting;
determining a scatterer delay associated with the selected scatterer;
adding the scatterer delay to the seismic data; and
re-filtering the seismic data to subtract the scatterer noise energy from the seismic data.

16. The method of claim 15 , wherein the step of subtracting further comprises:
subtracting the scatterer noise energy for a given trace from the seismic data if the scatterer noise energy for that trace exceeds a threshold.

17. The method of claim 15 , wherein the seismic data is generated using a seismic acquisition system having a cross spread source-receiver pattern.

18. The method according to claim 15 , further comprising:
transmitting seismic signals from one or more sources using a cross spread source-receiver design; and
receiving said transmitted signals at one or more receivers.

Assignees

• Cgg Services Sa

Inventors

• De Cacqueray Benoit
• Roux Philippe

Classifications

• G01V2210/1425

  Land surface · CPC title

• G01V2210/242

  F-k filtering, e.g. ground roll · CPC title

• G01V1/364 Primary

  Seismic filtering (G01V1/37 takes precedence) · CPC title