Oslo EAGE 2024

I've started a tradition of reporting on conferences (see previous posts on EAGE23 and Seismic24), taking a bit of a deeper dive than the "Here's us smiling at our booth" or "So good to meet up with old friends" posts. As my own employer it goes without saying that "All views are personal" and I’m happy to discuss further with anyone and correct any mistakes/misrepresentations.

Headlines

• 6200 attendees, real buzz in the conference and exhibition and organisers (staff and volunteers) can give themselves a pat on the back and a holiday. As an EAGE Board member, I solicited instant feedback and comments about what wasn't working so well, and I'll list a few in the next section.
• Opening session debate with global YP participants drove home the message about how important reliable electricity supply is.
• 18 parallel oral paper sessions with up to 16 a day, with posters presented during the extended lunch break. Some of us old-timers remember annual conferences a fraction of the size, and many authors struggled with the new requirements of 15 minutes + 5 minutes Q&A rather than the old 20 + 5. Many delegates looked like rabbits in the headlights as they scanned the multiple pages of talks with FOMO!
• Headphones and open plan presentation rooms separated by just large curtains

1. likened to a silent disco and took some getting used to, but we soon realised the advantages.
2. bored with the talk you're watching? Flick to another channel to see if it's worth jumping over to a different session.
3. worried about getting stuck in a session room? Much easier to drift in and out of a session and watch from the back before you commit.
4. can't find your hall? Tune to the channel and use the signal strength to guide you ... warmer, colder, warmer.

• Most popular talk sessions? FWI, CCS, 4D but note that Machine Learning/AI was everywhere.
• Farewell to the CGG name and branding - now Viridien - which tripped up several employees in conversation. Their stand broadcast the new bright colour scheme - but perhaps too brightly resulting in a Tuesday morning power outage. I just hope they soon drop the green writing on a blue background which is an affront!
• TGS/PGS - is it a merger or a takeover? All will be revealed on July 1st. Ironically the two stands were located across the main thoroughfare at the exhibition's centre of gravity.
• Last outing for Wintershall who went out with a bang (not literally) with their live CO2 injection demonstration.

Feedback

As they say in early schooling, 'two stars and a wish' - there are always areas for improvement. So here's a few suggestions I picked up:

• Queue to get in on the first day at the revolving door: I tried to ask security to open an additional door but they didn't have the authority. Someone more influential than I must have had a word as by Wednesday we weren't having to do the long walk outside and funnel in just one revolving door.
• Queue at toilets: good to see males experiencing what females commonly have to suffer, but rather bizarre for such a large venue. We soon learnt where less busy facilities were situated.
• Cost: I had complaints pre-conference (about Norway being too expensive - yes, but a pretty fantastic country), and comments at the conference to which I can clearly state that the Annual Conference is exceptionally good value compared to other gatherings because of the synergy and discounting between the Exhibition and Conference ... and long may it continue.
• Size: has the Annual become too big? My feeling is no: extended abstracts for any talks you missed can be accessed through EarthDoc (and hopefully authors are considering publishing fuller papers to our journals) and there is a place for both these larger events and smaller focused workshops/conferences.
• Exhibition layout: as always winners (big stands in the middle of Hall D) and losers (those in Hall C and around the edges). Having seen the layout of one big space in Toulouse, this should be less of a problem in 2025.

What I'm not going to mention

In 2023 I had spent time in the Strategic Programme sessions to report back to my employer, so consciously did not attend those in 2024 - refer back to last year's post for key messages. I also tried to stop my natural tendency to drift to 4D talks - the conference is an excellent opportunity to learn about unfamiliar topics after all. I browsed posters but only managed interaction with a couple of authors. I will not discuss the Exhibition stands/presentations, being now in the consultancy world with no software or study $s behind me - each company has effective business development communication channels already. I didn't attend the showcase theatres for International Prospecting, Digital Transformation or Energy Transition. So the following are exclusively main conference presentations and I will only list (first author) to stop repeating et al ad infinitum.

CCS

CCS is where I spent a lot of my time, and it was great to see NSTA's stand poster including some images from a recent TRACS project (we built pressure-sensitivity and CO2 plume models of the Tertiary fans in the CNS/NNS and of the Bunter in SNS - lateral grid cell size 1 x 1 km!).

A talk from Shell on injection into and ultra depleted gas reservoir SOCSNL (Dekker) highlighted positives (no loss of gas production, demonstrated tank, high perms) and areas of focus (hydrate formation, possible irreversible porosity reduction, far-field storage capacity pressure constraint).

Another talk discussed how Shell had screened and then matured (play, lead, prospect, licence application) NL saline aquifers (Abbenhuis) and whether the stores were in closed (K Block) or open (P&O Blocks?) pressure systems.

Equinor (Smith) showed simple modelling of different depositional environments for saline aquifers since their uncertainty is inherently greater than depleted fields (less well data and no production history). As you might expect, channels confine and sheets dissipate pressure. When pressure exceeded 400 bar, injection was reduced - why 400 I asked? Ah, that would be telling.

Osdal (Equinor) showed us the 4D over Sn?hvit in the Barents where CO2 is separated from production and reinjected in an adjacent aquifer fault block. Snippets I picked up were that initial injection in the underlying excellent sand Tub?en formation failed because of poor lateral pressure connectivity and so injection has been predominantly in the St? 2 formation which has 500 mD perms. 4D softening is seen for both CO2 plumes (replacing water) but also near the depressurizing field when residual gas comes out of solution. Simple wedge modelling demonstrated that plumes of CO2 less than 15 m thick can exhibit signal interference with the residual gas signal. The 4D also showed gas-water hardening at relay ramps proving sand-sand juxtaposition.

With a focus on keeping monitoring costs low, Park (Stanford) showed that using Deep Learning to learn noise characteristics from multiple baseline surveys produced cleaner plume results than a single pre-injection baseline.

I used the headphone technology to start listening to Butar (SLB) describe the Smeaheia store in the Sognefjord Formation east of Troll, then tune in to Ziolkowski (Edinburgh Uni) next door describing MTEM for monitoring. Several questions on this on spatial resolution (horizontal = half receiver line spacing; vertical = 10 m) but 4D differences can resolve more than 3D and MTEM has different sensitivity to seismic. It has not been tried in practice and would (probably) require permanent installation in challenging brine environment.

Finotti (Sintef) gave an expansive keynote on multiple EU funded studies (LINCCS, ACCSESS) with M€ budgets including low OPEX floating offshore capture units, capture on vessels, leakage monitoring, large scale injection tests, feasibility studies for linking wood pulp, cement and waste to energy factories in Sweden, Poland, Germany, Switzerland to Norwegian stores, building transnational networks in Greece and the Balkans. I was left overwhelmed by the pace of change, and noted the following comments during questions - the margins are lower than oil and gas, so there needs to be a redistribution of risk which means building more trust - I commented trust = contracts, and can we build on current legal frameworks? Yes e.g. LNG contracts. There is a need for stores to have backup arrangements with other stores in case there is a need to go offline - similarly I would presume the curtailment of supply would also require backup to keep stores operational.

Pestman (Rose & Assoc) gave a thorough guide through the Quantitative Risk Assessment of storage projects. This is worth a read (abstract on EarthDoc) for anybody involved in CCS - I learnt about peak risk, happening in the first years of the injection phase when the effective of the store is being tested. The recommendations were to define the list of potential events, evaluate their annual chance of failure and financial impact and aggregate all events on an annual basis.

I learnt from Masoudi's (Uni of Oslo) poster about how supercritical CO2 injection can evaporate formation water and salt crystals can form - this can also lead to more brine being drawn in with more crystallization. They had a simple experimental setup to demonstrate this.

Hydrogen storage

Going a bit off piste I was also interested in hydrogen storage in man-made salt caverns - I never knew that we had one under Teesside from ICI days I presume. Moscardelli (Bureau of Econ Geology) discussed the need to increase the current capacity (3 in Gulf Coast salt diapirs), consider the geomechanical stresses of cyclical injection/production, beware the impact of H2 in presence of sulphur and interaction with pipelines and establish the markets. The subject area requires the Venn diagram of Subsurface/Engineering/Economics. Three areas were considered Permian in W Texas (Salado/Castilo Formations), Gulf Coast Jurassic salt domes and N Sea Zechstein. Offshore storage is not yet considered practicable, with higher stresses and the problem of brine disposal in initial leaching phase. They have an open source screening tool (STARR BEG then select Digital Tools). A rule of thumb - you need 400 ft thick salt with an 800 ft overburden. Leaching requires a lot of water, but comparisons with O&G are favourable. Hydrogen from wind turbines could be stored in caverns, converted to ammonia and/or piped to market. Hydrogen costs were put at $3.57/$0.84 per kg of H2 for green/blue in Texas whereas $5 for green in Europe.

Salam?o Martins (AtkinsRéalis) showed an early stage screening tool for NE England (Teesside to Lincolnshire) incorporating the sparse wells and the Onshore Geophysical Library data. HydrogenUK's vision has a 12-33 TWh share by 2030, which requires a rapid infrastructure build.

Multiples and diffraction imaging

A bit of lateral learning for me, as I was chairing a session of four papers. I heard two presenters say 'These are just preliminary results' so expect more rounded presentations/publications in the future. Safari (Delft Uni) showed Full Wavefield Migration with incorporation of Q attenuation to improve the result. The most interesting part for me was the question session, with someone asking what about the stability/evanescence in your equation then me asking why have you got that smudgy bit on your plot, which could have been a stupid question but luckily it was due to the cross-talk between reflectivity and attenuation. Alfaraj (Saudi Aramco) demonstrated using diffractions to tune velocity models - it will be great to see future application on real data. Zhang (Delft Uni now Fugro) showed diffraction imaging applied to GPR data over Antarctic ice which nicely identified boulder fields. An interesting study would be time-lapse, but challenges with following the same locations with moving ice and human factors. Li (China University of Petroleum) showed joint RTM of primaries and internal multiples using an up-up, down-down separation filtering that I am still trying to get my head round.

4D

Yes I know, I couldn't help myself. Hawley (Equinor) showed Snorre 4D and how the 4D revealed compartmentalisation and connectivity in a low NTG channelized environment that could not be determined with any other data. Value examples 1M$ to avoid running a PLT, 30-40M$ for a deep injector sidetrack.

Tura (CSM) showed PP and PS 4D from Edvard Greig and how the combined datasets allowed improved separation of pressure and saturation signals.

Bertrand (V?r) showed the challenges of calibrating 4D in steep injectites where water replaces oil and vice versa and in two different formations, requiring individual calibration of each 4D signal. 1D convolution should be replaced by point spread functions for better characterization of these complicated geometries.

Shallow

Aschjem (AkerBP) showed drilling of the Overly Pleistocene prospect by 2/8-19 above the M?lde salt dome. The seismic had clear double flat spots in the Tanum?sen (proven gas) and Ringi?sen (oil) but the discovery was deemed non-commercial due to sand and stratigraphy not working.

Bellwald (NGI) gave a demonstration of the resolution requirements and key technologies for the shallow glacial subsurface.

Inversion

Dingwall (Neptune) presented the results of detailed reprocessing of 2012 CGG (or is it Viridien?) seismic and Qeye facies inversion over the Cygnus Field in UK SNS. Where previously the gathers could be described as a bit mushy and requiring stacking, multiple-free/reduced reflections could now be seen on the input gathers to inversion. The facies classes demonstrated improved characterization and understanding of reservoir performance e.g. the lack of water ingress to producers in the lower Ketch reservoir which previously had been modelled with layered direct connection into the aquifer but now could be interpreted as differentially stacked dipping channels.

Gunning (CSIRO) used a joint inversion of DAS and pressure tomography wells in the Otway CO2 injection test site. Pressure tomography involved injecting water in one well for 6 hours, relax for 18 hours. There was one CO2 injection well and 5 pressure tomography injection wells. This pressure pulse is rapid relative to the CO2 gas injection phase and so the inversion can be treated as effective media. The pressure data needed to have the removal of earth tides and other transients. The Bayesian inversion demonstrated that pressure only inversion did not have resolution to map the plume - resolution set by the well spacing. Seismic DAS inversion only did not detect the thin gas migration at the top of the sands. Only both together could map the plume by bringing together multi-physics. The seismic is biassing the pressure inversion priors.

Other

L?soth (Equinor) showed injectite styles in three formations (Grane/Balder, Utsira and Naust) and how they are triggered by massive buildout of shelf deposits from the west for the first two and east for Naust. This leads to compaction and overpressure.

Preiss (Royal Holloway) has quantitatively mapped out which faults are coupled across the SNS salt with implications for fluid flow and geomechanics.

Butcher (Bristol) put up some crosshole seismic and wavefield separation from the Glasgow Observatory - I will be supplying suggestions for future projects based on my experience in the mid 90s!

And if you have made it this far, congratulations and look forward to seeing you in Toulouse next year for EAGE2025.