Wavelet Estimation Using Multiple Reflections in Seismic Inversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurate estimation of the source wavelet is crucial for full wavefield inversion (FWI) in geophysical prospecting, but existing methods face challenges due to non-uniqueness in wavelet estimation, especially in the absence of well data, and traditional methods rely on direct arrivals which are influenced by surface effects, making it difficult to estimate wavelets for vertically-propagated energy.
Innovation Solution
A computer-implemented seismic processing method that generates and optimizes both the source wavelet and subsurface model in the depth domain, simultaneously simulating and comparing primary and multiple reflections to match recorded waveforms, allowing for wavelet estimation without relying on direct arrivals or dense cross-line sampling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If traditional methods use direct arrivals for wavelet estimation, then the process is simpler, but the accuracy deteriorates due to surface effects and non-uniqueness
Solution Approach 1:
The patent converts the harmful effect of multiples (traditionally considered noise) into a beneficial tool for wavelet estimation. By using multiple reflections instead of direct arrivals, the method eliminates surface effect contamination while providing unique constraints on wavelet parameters through the predictable kinematic relationships of multiples.
Solution Approach 2:
The patent inverts the traditional approach by using multiple reflections (normally eliminated) instead of direct arrivals (normally used) for wavelet estimation. This inversion allows estimation of vertically-propagated energy wavelets without surface effect contamination, directly resolving the accuracy problem.
2Measurement precision
If well data are used to constrain wavelet, then wavelet accuracy improves, but the method becomes less applicable in exploration settings
Solution Approach 1:
The patent makes the seismic data self-sufficient for wavelet estimation by using multiple reflections inherent in the seismic record itself. This eliminates the need for external well data constraints, allowing the method to be applied independently in exploration settings where well logs are unavailable.
Solution Approach 2:
The patent creates a universal wavelet estimation method that works in both exploration settings (without well data) and appraisal settings (with well data). The multiple-based approach provides a common foundation that can be applied universally across different acquisition and processing scenarios.
3Quantity of substance
If sparse sampling is used, then acquisition cost decreases, but wavelet estimation reliability deteriorates
Solution Approach 1:
The patent converts the limitation of sparse sampling into an advantage by using multiple reflections, which are naturally present even in sparsely sampled data. The kinematic relationships of multiples provide strong constraints that maintain wavelet estimation reliability without requiring dense cross-line sampling.
4Object-affected harmful factors
If multiple reflections are eliminated, then primary reflection quality improves, but wavelet estimation opportunity is lost
Solution Approach 1:
The patent converts multiples from harmful contaminants into beneficial estimation tools. Instead of eliminating multiples before wavelet estimation, the method uses them as the primary source of information, thereby eliminating the trade-off between multiple removal and wavelet estimation.
Solution Approach 2:
The patent inverts the traditional processing sequence by using multiples for wavelet estimation before any multiple elimination. This inversion allows the method to benefit from both the estimation power of multiples and the quality improvement from subsequent multiple removal.
Data Source
AI summary
Wavelet estimation method, particularly advantageous for full wavefield inversion (“FWI”) of seismic data, that makes use of both the primary and multiple reflections in the data. The inventive method uses an FWI algorithm to generate a subsurface model from primary reflections (101) in a shallow layer before first arrival of multiple reflections (101). The model is then used to simulate multiples (102). The wavelet is subsequently modified (104) such that the simulated multiples closely match the true recorded multiples (103). The simulated multiples may then be subtracted from the measured data (105) thereby creating a deeper top layer of data substantially free of multiples, and the method may then be repeated to extend the subsurface model to a greater depth (106).