A Workflow to Quantify Velocity Model Uncertainty

Bell, A.; Lorenzo, Rosa Irene Fregel; Martin, T.; Berg, D. van der; Caselitz, B.P.

doi:10.3997/2214-4609.201601098

Cited by 4 publications

(1 citation statement)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For this we use a pseudo-random approach to model uncertainty. The workflow uses repeated and randomized sampling of the model space to determine estimates of the uncertainty in any model (Bell et al, 2016). The arbitrary perturbations to the model are constrained by an automated measure of the wavelength and magnitude modifications recoverable by the data used for the inversion.…”

Section: Methodsmentioning

confidence: 99%

Sample Size Automation in a Pseudo-random Model Uncertainty Workflow

Martin¹

2019

Second EAGE/PESGB Workshop on Velocities

Self Cite

View full text Add to dashboard Cite

Velocity model building (VMB) using tomography produces one credible realization of an earth model, which, in turn, generates one conceivable subsurface image. The inversion, by its nature, is highly non-linear, and can lead to uncertainty with a single model and image approach. Uncertainty can be quantified by using a model population, rather than a single realization. In this scenario, all models must equally explain the data by producing flat gathers from the inversion. Defining what is an appropriate sample size for a nonlinear system using a pseudo-random approach to model uncertainty is critical for cost and turnaround. We automate a real-time constraint on the expanding model population using statistical relevance to the attributes produced through the uncertainty process. Analysis using cumulative distribution functions (CDFs) of the deviation in the model population define an automated threshold. The sample size threshold is met when there is no additional statistical relevance for the output attributes; the process stops and the model uncertainty metrics defining spatial reliability of the data are output. We demonstrate this method on data from the North Sea.

show abstract

Section: Methodsmentioning

confidence: 99%

Sample Size Automation in a Pseudo-random Model Uncertainty Workflow

Martin¹

2019

Second EAGE/PESGB Workshop on Velocities

Self Cite

View full text Add to dashboard Cite

show abstract

Velocity model building in the major basins of the eastern Mediterranean Sea for imaging regional prospectivity

2018

View full text Add to dashboard Cite

Hydrocarbon prospectivity in the eastern Mediterranean is challenged by the presence of the Messinian salt layer, which varies in complexity across the area. The macro salt layer geometry itself varies from the deep abyssal planes in the middle of two major basins (the Herodotus and Levant basins), where it shows simple top and base relief to more complicated geometries near the continental salt-free shelf. Building depth velocity models for imaging the eastern Mediterranean basins requires accurate determination of the complex salt top and base, as well as accurate estimation of their velocities. Since 2006, many 2D and 3D megasurveys using conventional and dual-sensor acquisitions have been carried out across the area. Regional and relatively simple velocity models are used to image the Messinian salt layer extensions from offshore West Egypt to offshore Lebanon, passing by the Cypriot waters and the Eratosthenes Seamount. The data acquired include shallow and deepwater surveys, and technologies such as full-waveform inversion are applied to produce highresolution images for the shallow post-Messinian section. Shallowwater imaging challenges such as acquisition footprint issues and low-angle illumination are solved by imaging with multiples using dual-sensor acquisition data. The Messinian salt layer is modeled with a simple constant regional velocity, while the pre-Messinian section is modeled with a simple combination of velocity gradient models. The regional model-building scheme introduces a good correction for the pre-Messinian structure and provides reliable multiclient data ready for outlining new prospects.

show abstract

Automation of marine seismic data processing

Long¹,

Martin

2020

The Leading Edge

View full text Add to dashboard Cite

Marine seismic data sets contain highly redundant information. Data analytics and machine learning-based solutions should provide opportunities to reduce turnaround and improve confidence levels in output data volumes. A proof-of-concept (POC) thrust regime example from Indonesia illustrates that parameter testing can almost be eliminated if existing project parameter data can be mined from a database. Where quality control (QC) is required for complex challenges such as noise removal, supervised classifiers are a platform that can enable rapid global quantitative decisions based on relevant data attributes, moving behind the subjective art of observational QC. Finally, many early processing steps depend on reasonable knowledge of the velocity model in addition to the explicit dependence of imaging steps. A POC Monte Carlo-based model building exercise in West Africa used an efficient tomographic platform to demonstrate that turnaround can be reduced from 90 days to only a few days, even when the starting model was significantly wrong. These examples illustrate that a lot is already within our reach, and the development of embedded feedback loops will improve the level of automation further, particularly if humans can learn to let the data speak for itself.

show abstract

A Workflow to Quantify Velocity Model Uncertainty

Cited by 4 publications

References 3 publications

Sample Size Automation in a Pseudo-random Model Uncertainty Workflow

Sample Size Automation in a Pseudo-random Model Uncertainty Workflow

Velocity model building in the major basins of the eastern Mediterranean Sea for imaging regional prospectivity

Automation of marine seismic data processing

Contact Info

Product

Resources

About