De-ghosting Variable Depth Seismic Data Using Mirror Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for de-ghosting seismic data collected with streamers having variable depths, such as curved profiles, are inadequate as they cannot be generalized to three dimensions, leading to inaccuracies in imaging underwater geological structures.
Innovation Solution
A method involving the generation of migration and mirror migration data, derivation of ghost-free models using linear operators, and adaptive subtraction techniques to produce a de-ghosted dataset, which improves image resolution by accounting for ghost lag models in both the migration and mirror migration data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional de-ghosting methods are used on streamer data with variable depths, then processing can be performed, but the imaging accuracy and resolution deteriorate due to inability to generalize to three dimensions
Solution Approach 1:
The patent transitions from conventional 2D de-ghosting methods to a 3D de-ghosting approach by incorporating depth variable streamer configurations. The method processes seismic data in three dimensions, accounting for streamers with curved profiles and varying depths, thereby improving imaging accuracy while maintaining method generalizability to complex offshore geometries
Solution Approach 2:
The patent changes the parameters of the de-ghosting method to accommodate variable depth streamers. By modifying the processing algorithm to handle depth variations as a variable parameter rather than a fixed condition, the method achieves both improved imaging accuracy for curved streamer profiles and maintained adaptability to different acquisition geometries
2Ease of operation
If streamers are positioned at constant depth horizontally, then acquisition is simplified, but ghost interference increases reducing frequency content quality
Solution Approach 1:
The patent converts the harmful ghost interference into a beneficial processing target. By developing de-ghosting methods that specifically address ghost reflections from horizontal streamers, the technique transforms the problematic ghost energy into removable artifacts, thereby improving frequency content quality while maintaining the operational simplicity of horizontal streamer deployment
Solution Approach 2:
The patent applies parameter changes in the processing domain to counteract the adverse effects of horizontal streamer positioning. By adjusting processing parameters such as ghost lag models and adaptive subtraction thresholds, the method eliminates ghost interference while preserving the acquisition simplicity of constant depth deployment
3Measurement precision
If adaptive subtraction is applied to remove ghosts, then frequency content quality improves, but processing complexity increases
Solution Approach 1:
The patent applies preliminary action by performing adaptive subtraction as an early processing step before migration. By removing ghost interference before subsequent processing operations, the method improves frequency content quality and prevents ghost-related artifacts from propagating through the workflow, thereby reducing overall processing complexity despite the sophisticated nature of adaptive subtraction itself
Data Source
AI summary
Computing device, computer instructions and method for de-ghosting seismic data related to a subsurface. The method may include receiving input seismic data recorded by seismic receivers that located at different depths (zr), generating migration data (du) and mirror migration data (dd) from the input seismic data, deriving a ghost free model (m) based on simultaneously using the migration data (du) and mirror migration data (dd), generating primary (p) and ghost (g) datasets based on the ghost free model (m), simultaneously adaptively subtracting the primary (p) and ghost (g) datasets from the migration data (du) to provide adapted primary (p′1 and p′2) and adapted residual (r′1 and r′2) datasets, and generating a final image (f) of the subsurface based on the adapted primary (p′1 and p′2) and the adapted residual (r′1 and r′2) datasets. In certain embodiments, the input seismic data d includes both hydrophone data and particle motion data.