First Impressions

Andrew Long reports on the first few days of IMAGE 2021 as a virtual attendee.

wave imagewave image

​​​​​​In a return to physical conference events in the US, total registrations numbers at IMAGE21 are reportedly about 5 000 - about half attending in person and the other half paying for a highly flawed virtual experience (more on that later).

As was the case for the traditionally massive OTC conference earlier this year, several large oil companies and service companies either withdrew or significantly scaled down their planned participation in IMAGE21. Notable vendors absent from the physical exhibition hall include the likes of Schlumberger, CGG, DUG, and ION, and some other booths are shared between different companies.

As various LinkedIn posts testify, those physically present in Denver are delighted to engage once again with others, but the same photos also suggest skeleton staff representations.

Small Live Audiences

According to those in Denver, the majority of physical participants seem to be scheduled speakers, and technical sessions correspondingly have small live audiences. The combined SEG-AAPG technical program is nevertheless quite vast and rich in geological case study content plus a variety of ‘new energy’ themes.

For the roughly 2 500 virtual participants, IMAGE21 is off to a rocky start, and the use of virtual conference platforms seems to have gone backward since SEG2020. Once you get past a cute 3D animation of figures walking around in a virtual conference lobby, the functionality falls off rapidly.

Unless you connect in real-time and watch the sessions as they happen, almost no recorded content is accessible after three days of IMAGE21.

This means those in other global time zones who paid to be a virtual participant really have no benefit from their investment—they could simply have paid a fee to view recorded sessions when they eventually appear in the SEG Technical Library (if that is indeed the plan).

Personally, my ambition was to follow each day of presentations on a roughly 12-hour delay; catching up whilst participants in Denver were sleeping. The IMAGE21 organizers tried hard to pull out all stops to maximize live participation, delaying the release of the PDF abstracts in the Technical Library, and only making details of the final technical program available in the final days before the event. All in all, it was a chaotic process to follow, and it is frustrating to now experience the lack of online functionality.

In contrast, the hybrid EAGE event scheduled to occur in late October has encouraged virtual participation by releasing its technical program and completing an Early Bird registration process many weeks ago.

I suggest that, if the SEG and AAPG want to retain their membership, they do more to accommodate their global membership, rather than try to coerce US-based members to register in person and largely ignore everyone else.

Time to Completely Revisit the Role of Large Physical Conferences

Assuming that the world moves into coexistence with the coronavirus and its mutations, hybrid conferences and workshops seem here to stay—just as does flexible working-from-home arrangements.

Even if the unavailability of the recorded sessions is only temporary, one has to ask whether technical geoscientists need large events such as IMAGE21 to access pre-recorded content that can in principle be made available any time after they are recorded?

One clear trend I have observed in the many virtual and hybrid workshops dedicated to specific themes is that pre-recorded presentations enable non-native English speakers to nevertheless record fluent and highly accessible presentations. 

Brief Technical Insights

For those who did travel to Denver, the largest marine seismic content has again focused on full waveform inversion (FWI) benefits; most notably with applications to ocean bottom node (OBN) data—with an (understandably) overwhelming bias towards GOM and Brazil case studies.

FWI has rapidly become integral to most depth imaging workflows, and the ‘twist’ in recent years is the pursuit of so-called ‘FWI Imaging’, wherein a pseudo-reflectivity image is generated in complex regimes where traditional cascaded velocity model building and migration is affected by various artifacts and challenges.

Day 1

The most profound development in the FWI sessions on Day 1 was the ‘Simultaneous Inversion’ presentation by Yang Yang of PGS. FWI and least-squares migration (LSM) are elegantly combined into a process that simultaneously optimizes both the velocity model and the reflectivity image.

Day 2

Several presentations on Day 2 continued the industry growth in multi-source acquisition with various shot blending strategies. Increased efficiency will be a necessary enabler for lower-cost OBN seismic, and multi-source shooting is also upgrading the resolution and quality of towed-streamer seismic.

For spectacular examples, check out the acquisition case study by Martin Widmaier, and the related quantitative interpretation (QI) case study by Cyrille Reiser. And for something completely different, see Stian Hegna’s use of the continuous energy generated by vessels to pursue seismic imaging. Although there are almost no IMAGE21 presentations available to view yet, you can access all the PDF abstracts in the SEG Library.


Little seems to have changed on the Machine Learning and AI front, and the most common explicit applications for these platforms are to augment the interpretation of salt bodies during seismic model building, for automated facies description, and so on.

Given the admirably vast AAPG technical program content, I wish I could correspondingly comment on how other geophysical technology is being used to benefit the interpretation of geoscience data—for both conventional and new energy applications—but sadly none of those presentations are yet available as viewable recordings. Maybe things will have improved by the end of the week.