Uphole-Calibrated Velocity Modeling for Near-Surface Seismic Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional seismic acquisition methods struggle to accurately map the low velocity weathering layer in the near surface, leading to distorted imaging of deep subsurface structures and increased risk of drilling dry wells due to the sparse and challenging spatial interpolation of uphole data.
Innovation Solution
A machine learning model is used to process uphole seismic survey data, training on travel time and velocity data to generate a calibrated high-resolution velocity model for the weathering section, integrating with seismic first arrival travel times to improve near surface velocity mapping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional seismic acquisition is used to target deep subsurface reservoirs, then deep structure imaging is achieved, but the low velocity weathering layer in the near surface cannot be correctly mapped, resulting in distorted imaging of deep structures
Solution Approach 1:
The velocity model is segmented into two distinct parts: a high-resolution near-surface weathering layer velocity model derived from uphole surveys, and a deep subsurface velocity model from conventional seismic acquisition. The shallow layer is processed separately with specialized techniques to avoid contamination by deep structure information, thereby resolving the contradiction between measuring near-surface velocities accurately and maintaining deep structure imaging reliability.
Solution Approach 2:
Different processing qualities are applied to different depth zones: the near-surface weathering layer (0-200m) receives high-resolution processing using uphole survey data with fine spatial interpolation, while the deep subsurface uses conventional seismic processing. This local quality approach allows optimized velocity modeling for each zone's specific characteristics, improving both near-surface velocity measurement precision and deep structure imaging accuracy.
2Measurement precision
If uphole surveys are performed to map the weathering layer, then near surface velocity calibration is improved, but the sparse spatial distribution makes interpolation difficult and introduces bull eyes artifacts
Solution Approach 1:
The interpolation process uses parameter changes by transforming the sparse uphole survey data into a regular grid through kriging algorithms, adjusting the grid resolution and interpolation parameters to match the seismic survey geometry. This converts the irregular sparse sampling into a regular fine-grid velocity model, resolving the contradiction between maintaining high velocity calibration precision and reducing interpolation complexity.
Solution Approach 2:
A kriging algorithm acts as an intermediary between the sparse uphole survey measurements and the final velocity model. The kriging process mediates the interpolation by using spatial autocorrelation and statistical principles to smoothly distribute the sparse measurements across the entire survey area, eliminating bull eyes artifacts while preserving velocity calibration precision.
3Stability of the object's composition
If large smoothing is applied to the velocity model, then spatial continuity is improved, but the localized high-resolution velocity calibration from uphole data is invalidated
Solution Approach 1:
The velocity model is segmented into shallow (0-200m) and deep zones, with the shallow zone retaining high-resolution localized calibration from uphole data without smoothing, while the deep zone receives appropriate smoothing for continuity. This segmentation allows spatial continuity to be maintained in the deep subsurface without invalidating the high-resolution near-surface velocity calibration.
Solution Approach 2:
Different smoothing strengths are applied locally: strong smoothing is applied to the deep subsurface to ensure spatial continuity, while the shallow weathering layer maintains high-resolution localized calibration. This local quality differentiation resolves the contradiction between achieving spatial continuity and preserving localized velocity calibration precision.
4Adaptability or versatility
If uphole data is integrated with seismic first arrival travel time data, then comprehensive velocity modeling is achieved, but the integration remains difficult due to different data characteristics
Solution Approach 1:
The data integration is segmented by depth: uphole survey data is used to calibrate the shallow velocity model (0-200m), while seismic first arrival travel time data is used for the deep subsurface velocity model. The two datasets are integrated through a unified processing framework that handles each zone's specific data characteristics appropriately, achieving comprehensive velocity modeling while managing processing complexity through segmentation.
Data Source
AI summary
The construction of an uphole-calibrated velocity model from uphole seismic survey data using a machine learning model. Uphole seismic survey data may be processed to obtain seismic travel times sorted in a midpoint-offset domain. The machine learning model may be trained with pairs of training data that include travel time vs offset and uphole time, travel times vs offset and uphole velocity, and travel times vs. offset and seismic velocity determined from an interval velocity interpretation of uphole times. The trained machine learning model may output calibrated pseudo uphole velocities having a vertical resolution comparable to the existing upholes.