Time Lapse Seismic Data Correction for Overburden Effects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 4D seismic data processing techniques face challenges in accurately correcting for overburden and recording variations, leading to decreased repeatability and difficulty in observing subtle changes in reservoir fluid movements due to assumptions about constant propagation effects and signal-to-noise ratios, which are often violated by seasonal and daily near-surface changes.
Innovation Solution
A method that normalizes each time lapse seismic survey within itself, using a limited frequency range to derive corrected reservoir amplitudes, avoiding frequencies with significant overburden signal-to-noise ratio variations, and does not require a baseline survey, cross-equalization, or coordinate adjustments, thereby improving data accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional cross-equalization processing is applied to correct overburden effects, then overburden amplitude variations are reduced, but noise is propagated and residual anomalies remain due to frequency bandwidth limitations and signal-to-noise ratio variations
Solution Approach 1:
The patent changes the processing approach from conventional cross-equalization to a new method that calculates correction factors using the ratio of RMS amplitude at the reservoir to RMS amplitude in the overburden, specifically targeting frequency ranges where overburden signal-to-noise ratio is acceptable. This parameter-based correction approach resolves the contradiction by adapting the correction methodology to frequency-dependent noise characteristics.
Solution Approach 2:
The patent creates a corrected copy of the seismic data by applying amplitude corrections derived from overburden-free frequency ranges. Instead of directly modifying the original data, it generates a corrected version that eliminates overburden effects while preserving the original signal characteristics in ranges where noise is minimal.
2Loss of information
If a broad frequency bandwidth is used for correction, then more signal information is preserved, but noise from frequency ranges with poor signal-to-noise ratio is propagated
Solution Approach 1:
The patent segments the frequency spectrum into different ranges based on overburden signal-to-noise ratio characteristics. It identifies and isolates frequency ranges where the overburden does not significantly attenuate or distort the signal, then applies correction only using data from these clean frequency ranges, effectively separating useful signal information from noisy frequency components.
Solution Approach 2:
The patent applies different processing quality standards to different frequency ranges. Instead of uniform processing across all frequencies, it selectively applies correction methodology only to frequency ranges where the overburden signal-to-noise ratio is acceptable, allowing high-quality correction where possible and avoiding noise propagation in problematic frequency ranges.
3Reliability
If permanently cemented sources and detectors are used, then recording repeatability is improved, but overburden changes above and below the installed systems still affect the data
Solution Approach 1:
The patent extracts and removes the overburden effects from the seismic data through mathematical correction. By calculating the ratio of reservoir amplitude to overburden amplitude and applying this as a correction factor, it effectively takes out the overburden propagation effects from the recorded signal, leaving only the reservoir-related information.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Time lapse or 4D seismic data are corrected for geologic overburden and seismic recording system effects. The data from a survey at one time of interest is processed within a selected frequency band and the reservoir level is normalized by the overburden. The results are used to extract reservoir amplitudes from the data of that same survey. Frequencies where overburden signal-to-noise ratios vary dramatically between time lapse surveys may then be avoided in processing of data from the area of interest.