Our new elastic least-squares imaging technology solves for reflectivity, Vp, Vs, P-impedance, S-impedance and density. It delivers unsurpassed imaging along with elastic rock properties for quantitative interpretation — directly from field-data input! Watch the video here.
Comparison of the reflectivity derived with DUG Elastic MP-FWI Imaging using raw field data, versus a conventional LS-RTM workflow using extensively pre-processed input. This legacy dataset was acquired in 2006 on the Australian North West Shelf. This region contains rapidly changing shallow velocity variations due to localised channel features and carbonates. (Data courtesy of DUG Multi-Client)
The updated rock property models show a significant increase in resolution. We can readily identify the reservoir location — a beautiful example of a flat spot! (Data courtesy of DUG Multi-Client)
Land seismic data, such as this example, can often exhibit strong elastic effects that violate the assumptions of acoustic imaging. This desert setting has geological complexity both in the near-surface and at the target depth, including high-impedance carbonates. Conventional pre-processing workflows are non-trivial, especially on land data, where stages such as demultiple often require complex modelling and adaptive subtraction workflows. Significantly better illumination and event continuity is obvious when comparing the sections.
Here we consider an OBN dataset from the Gulf of America where water depths are approximately 2 km. Above we see a comparison of the downgoing mirror RTM using pre-processed data and DUG Elastic MP-FWI Imaging reflectivity. We can immediately see the improvements that the MP-FWI imaging brings over the conventional RTM with respect to resolution and illumination. The elastic MP-FWI imaging result has obviously produced the superior image around the high-impedance contrast caused by the salt, and the structures beneath it. (Data courtesy of Shell)
Here, velocity model updates generated using a conventional workflow and DUG Elastic MP-FWI Imaging were evaluated using a Kirchhoff pre-stack depth migration (KpreSDM), using the fully pre-processed data as input. The figure above compares common-offset-vector KpreSDM snail gathers using the initial velocity model and the DUG Elastic MP-FWI Imaging derived velocity model The MP-FWI imaging velocity model demonstrates a clear and significant kinematic improvement (gather flatness) and a reduction of jitter due to the azimuthal variations, particularly around the high-impedance contrasts given by the carbonates.
DUG Elastic MP-FWI Imaging is not only a complete replacement for the traditional processing and imaging workflows, it also replaces the subsequent inversion workflow for elastic rock properties. Elastic MP-FWI Imaging accounts for both compressional and shear waves, handling variations in seismic wave dynamics as a function of incidence angle, including in the presence of high impedance contrasts and onshore near-surface geological complexity. Multiples and converted waves are now treated as valuable additional signal, increasing sampling, resolution and constraining the inverted parameters. When using the full wavefield for simultaneous velocity model building, rock property inversion and true-amplitude imaging, a multi-parameter solution is a necessity.
DUG Elastic MP-FWI Imaging delivers high-resolution elastic rock property outputs — directly from field data. By eliminating the need for conventional pre-processing and imaging workflows, and a post imaging AVA inversion workflow, (along with their assumptions/workarounds that ultimately affect the AVA compliance of the data) more certain predictions can be realised.
Here, two equivalent crossplots of elastic rock properties are compared. The ellipses indicate the expected distribution of the labelled lithology and fluid combinations from statistical rock physics analyses of well log data. Quantitative interpretation requires true amplitudes and high signal-to-noise ratio. Less scatter translates to less uncertainty. (Data courtesy of DUG Multi-Client)
Here we compare the elastic rock property well ties from DUG Elastic MP-FWI Imaging to a conventional processing, imaging and AVA inversion workflow. It is obvious that the Elastic MP-FWI Imaging results are a better match to the well. (Data courtesy of DUG Multi-Client)
Resolving illumination issues caused by complex salt bodies has been a persistent geophysical challenge. DUG MP-FWI Imaging uses all parts of the scattered wavefield to reconstruct the reflectivity and kinematic Earth parameters. Naturally treating both up- and down-going wavefields, along with prismatic waves, multiples and ghosts as valuable information, guarantees significantly better subsurface sampling than primary-only imaging. This in turn delivers much-improved illumination, resolution and ultimately imaging of subsurface geology.
In these land dataset examples quite spectacular uplift was observed. DUG MP-FWI Imaging resolved subtle structural and stratigraphic features that were simply not imaged with a conventional processing and imaging workflow. Even with the additional complexity that comes with onshore data acquisition (such as topography and near-surface effects), DUG MP-FWI Imaging produced far superior results. Significantly better illumination and event continuity is obvious when comparing the sections below, thanks to least-squares imaging of the full wavefield—that includes interbed multiples. Incredible near-surface detail has been resolved, as can be seen on the shallow depth-slices to the right. This example highlights why DUG MP-FWI Imaging is now the technology of choice, and a must-have for any seismic project, either land or marine.
A comparison between imaging results in a complex salt environment, offshore Gulf of Mexico. Conventional reverse time migration (RTM), after pre-processing & regularisation versus DUG MP-FWI Imaging result using field-data input. Superior resolution. Superior imaging. (Data courtesy of Shell)
Shallow-water environments are notoriously challenging for conventional processing workflows due to the presence of strong, short-period multiples that obscure primary signal. In this shallow-water dataset from the Barents Sea, DUG MP-FWI Imaging generated a high-resolution image up to 110 Hz. (Data courtesy of Aker BP)
A comparison of results from a hybrid (towed-streamer and OBN) dataset from the Norwegian North Sea. (Data courtesy of Seismic Partner and Geoex MCG)
Obtain unrivalled results faster with DUG’s Multi-parameter FWI (MP-FWI) Imaging technology. Simultaneous model building and high-frequency, least-squares imaging deliver accurate, high-resolution Earth models using field-data input—without the many time-consuming, subjective, serial steps of a conventional processing and imaging workflow. At high frequency, this revolutionary approach provides reflectivity images for both structural and quantitative interpretation, including angle stacks for AVA analysis. Our new elastic imaging technology solves for three-component reflectivity, Vp, Vs, P-impedance, S-impedance and density.
FWI inverts for high-resolution earth models using the entire seismic wavefield. It is an integral part of our depth model-building strategies for conventional imaging workflows.
A JV project between TGS and DUG, the “Peninsular Malaysia Mega Merge 3D” involved 39 different surveys (including 3 OBC surveys). With data vintages from 1989 to 2014 the combined area was over 17,000 square kilometres. The high-resolution velocity model was built up to a maximum frequency of 35 Hz.
Starting velocity model and final model after FWI. Depth slice and inline with migrated stack overlaid with the respective velocity model. In this OBN example the shallow channels are well resolved after FWI, correcting the imaging distortions at depth. (Data courtesy of Carbon Transition and TGS)
Before and after FWI. Smooth starting velocity model prior to FWI (left) and after FWI, co-rendered with the seismic data (right). (Data courtesy of Shell NZ).
Shallow depth slice of the velocity model prior to FWI versus shallow depth slice of the velocity model after FWI highlighting significant short wavelength detail.
FWI using both high-resolution streamer data and sparse OBN data. Velocity updates beyond 5 km depth are achieved thanks to the long offset diving wave penetration of the OBN data. (LumiSeisTM data courtesy of MCG)
Designed for geoscience, not computer science. Backed by some of the greenest HPC on the planet.
Diving wave penetration QC. The white “bananas” demonstrate the maximum depth of update expected from diving wave FWI.
