Seismic Source Signature Estimation via Sparse Inverse Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing seismic data processing methods struggle to accurately generate far-field signatures for seismic source arrays, especially when near-field data is unavailable or contaminated, as they fail to reliably reproduce low-frequency behavior and account for source array interactions.
Innovation Solution
The method employs far-field seismic data from streamer, ocean bottom node, or ocean bottom cable data to extract the source wavelet, using a constrained operator to derive a time-variant signal representing the seismic source, which includes a model of the seismic source array with notional sources scaled by the cube root of gun volume, to separate and correct direct arrival energy and source array effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical modelling packages (Nucleus and Gundalf) are used to generate far-field signatures, then source array geometry and gun characteristics can be utilized, but the low frequency behaviour of the source arrays cannot be reproduced reliably
Solution Approach 1:
The patent introduces an inverse filter as an intermediary component that processes the far-field signature to correct low-frequency inaccuracies. The inverse filter is designed based on the desired low-frequency response and is applied to the far-field signature obtained from physical modeling, thereby mediating between the imperfect physical model and the desired accurate output.
Solution Approach 2:
The patent modifies the far-field signature by applying frequency-dependent corrections through the inverse filter. This changes the spectral parameters of the signature, specifically enhancing the low-frequency content while maintaining the overall signature characteristics, thus improving low-frequency behavior reproduction.
2Measurement precision
If near-field hydrophone (NFH) data is used to obtain source signature characteristics, then source array interactions can be accounted for, but the data may be unavailable or contaminated in legacy and current surveys
Solution Approach 1:
Instead of using near-field data to derive far-field signatures (the conventional approach), the patent inverts the process by using available far-field seismic data to extract and estimate the source wavelet and far-field signature. This inversion approach allows obtaining source characteristics without requiring near-field hydrophone data.
Solution Approach 2:
The method enables the far-field signature estimation to be self-sufficient by using only far-field seismic data that is already available in most surveys. The process does not depend on external near-field measurements, making the system self-serviceable with commonly available data types.
3Measurement precision
If direct recording of far-field from a sensor mounted on a rope and towed below the boat is used, then far-field data can be obtained, but this is not always practical
Solution Approach 1:
The patent makes the far-field signature estimation method universally applicable to standard seismic survey configurations. By using far-field data that is already recorded as part of routine seismic surveys (with sensors on streamers or ocean bottom nodes), the method eliminates the need for specialized far-field recording equipment and procedures, thus improving ease of operation while maintaining data quality.
4Measurement precision
If modelling packages are used to generate far-field signatures, then source array geometry can be incorporated, but source array interactions and contamination cannot be effectively removed
Solution Approach 1:
The patent extracts the far-field signature directly from recorded seismic data by isolating the direct arrival portion. This extraction process separates the source signature from other seismic events and contamination, obtaining a clean representation of the source characteristics without the need for complex modeling of source array interactions.
Solution Approach 2:
The patent converts the presence of direct arrivals (which contain source array effects) into a benefit by using them as the primary data source for signature extraction. Instead of treating source array effects as harmful contamination to be eliminated through modeling, the method uses the direct arrivals as they are, applying minimal processing to extract the signature, thereby turning a potential problem into the solution foundation.
Data Source
AI summary
A method for estimating a time variant signal representing a seismic source obtains seismic data recorded by at least one receiver and generated by the seismic source, the recorded seismic data comprising direct arrivals and derives the time variant signal using an operator that relates the time variant signal to the acquired seismic data, the operator constrained such that the time variant signal is sparse in time.