𝗦𝗵𝗼𝘂𝗹𝗱 𝗩𝗲𝗿𝘁𝗶𝗰𝗮𝗹 𝗦𝗲𝗶𝘀𝗺𝗶𝗰 𝗣𝗿𝗼𝗳𝗶𝗹𝗲𝘀 (𝗩𝗦𝗣𝘀) 𝗯𝗲 𝘂𝘀𝗲𝗱 𝗶𝗻 𝗰𝗮𝗿𝗯𝗼𝗻 𝘀𝘁𝗼𝗿𝗮𝗴𝗲? The question might seem to be bait, but we ask it sincerely in our latest Canadian Society of Exploration Geophysicists - CSEG RECORDER article. For context, I and my co-authors, Eric Street, Graham Hack, Jason Schweigert, and Matthew Allen suggest in our #Theseus and #24D carbon storage MMV methods that VSPs are not generally needed in onshore commercial carbon storage. Carbon Alpha's Theseus 24D enters the fray of reducing seismic costs in monitoring through changes in imaging areas, the integration of 2D and 3D seismic, and the jettisoning of techniques that are not required through the long life of these projects. The aim is similar to sparse nodal techniques but preserves the continuous imaging of 2D and 3D seismic. And one of the techniques we 'Jettison' is VSP surveying. Our suggestion is likely upsetting on the first read, because VSPs do have great value. They are used heavily in pilots, and have a history of use in Western Canada and the world. At presentations at GeoConvention Partnership and Carbon Capture Canada this year, we know we stirred some controversy by making these suggestions. But science is about challenging ideas and learning, so we wanted to follow up on the controversy and discuss the question of VSPs in commercial scale carbon storage in greater detail. This led to our Recorder article, which can be found here: https://lnkd.in/gMMKERxS Please have a look at the article. Our advice is far from as hard and fast as the few words in this post make it seem.
Lee Hunt
2mo
A commenter has suggested that the argument regarding VSPs cannot stand because of missing subsurface information. I'll address this here, because it is important. Assumptions regarding a lack of information about the geological model, logging, & seal integrity are incorrect in the context of the paper (which contains the argument - not this post). We follow the CSA Z741 standard for carbon storage, which has a series of stages, including Characterization, which must be complete before the monitoring plan for the Develop & Operate phase can be created. In those stages, significant effort is made to understand the subsurface. This includes evaluating the suitability of the storage complex, its injectivity and seal integrity. Typically seismic & logging are incorporated in these early stages to help assure these things. In the paper, we talk about the importance of site specificity and characterizing the site before planning the MMV program and deciding if a VSP is required. In the paper, we are very careful to point out that there may be a specific reason to include a VSP survey. Our 'No VSP' advice is not advice to never use a VSP. In any case, MMV planning cannot be complete without adequate subsurface evaluations.
Don Lawton
2mo
Hi Lee - as you might expect, I disagree strongly with this. Walk-away VSP;s provide really important data early in the lifetime of a CCS project, particularly through high resolution imaging of the storage complex and overburden proximal to the injector well. DAS is ideally suited for this and we have had many discussions with well completions engineers about the VOI that WVSPs provide, and well completions can be designed to minimize any well integrity issues that always are concerned. The other benefit is that the outer corridor stack from VSP data will be multiple-free and can be used as a guide for 2D and 3D surface seismic processing once the plume extends beyond the imaging limit of the WVSP. In the Western Canadian Sedimentary Basin, multiples can really mess with 4D seismic imaging in the Paleozoic section. In our experience, most CCS operating companies within the JIP agree with our position on WVSP data. Don
Jason Eleson
2mo
Interesting thesis. I am not a geophysicist, but might suggest some items that could help buttress your argument: 1) provide a quantifiable measure of typical VSP resolution "near wellbore" vs. resolution from predicted/known distal plume extents from commercial CCS programs 2) quantify imaging loss from VSP in the distal areas where plumes will migrate to vs. what could be achieved from 2D/3D data 3) provide examples where 2D or "24D" seismic would provide or have provided benefits in legacy commercial programs, e.g. the Sleipner plume didn't behave as initial simulations indicated, and instead displayed a rapid migration to the north. The 3D footprint was big enough to track and image this movement, but perhaps VSP could not? Something to keep in mind: there are many programs being developed in the US that lack a baseline 3D program. From research that Rachel Frohman and I have been doing on available, unredacted Class VI regulatory submissions from Class VI operators, only 56% of known CCS programs have a baseline 3D survey, and only 16% of those surveys were acquired specifically for the CCS programs; the rest are legacy seismic acquired for different purposes. What value would VSP have for areas with legacy 3D or no 3D?
Fred Peterson
2mo
Hi Lee, It is good to stimulate conversation but VSPs provide a proven effective measurement reference that enables improved evaluation/comparison and selection of other methods of plume measurement. With a multifunctional fiber cemented outside casing, in addition to low cost VSPs, one can also measure temperature, acoustic noise and/or strain to monitor well and sequestration integrity which are critically important. When VSP is integrated with an evaluation of other MMV methods it is a relatively inexpensive but very valuable tool. If surface based seismic methods evolve sufficiently for reliable plume mapping or leakage detection to the extent needed perhaps VSPs will no longer be useful but I suspect it will be years down the road if it ever happens.
Nancy House
2mo
A good treatment of the question of effectively using technology for CO2 monitoring. Maybe it's a bit provocative in the title. But I agree with the premise. It's important to understand the objective e of any data gathering when designing a survey. As geophysicists, we sometimes get enamored with the capabilities of our methods and forget to think about how they fit into the economic drivers compared to the technical objectives of the data gathering.
Brian Schulte
2mo
In bp, we wanted to think of the initial seismic as being the baseline and to acquire and process it accordingly. When we processed we wanted to take AVO into account so it was included in the design. When considering 2D, it is actually more powerful than most think because it can have a smaller near offset, longer spread and regular offsets which is ideal for AVO and inversion. We can actual model some of this to understand the results beforehand. With Mound Point in the GoM with the model we realized that 35000 feet of offset was required to image the class 2 AVO present. The model can help us to understand what will we achieve with the seismic and the parameters that we are using. It can also help with the processing. This is why Tesseral Ai sees synthetic modeling being a part of the full life cycle of the seismic and is moving into reservoir characterization with our software. Tesseral Ai is also being used to model the VSP and fibre optics for CCS projects.
Robert Pickering
2mo
I recognize some inconsistencies in the process. Some are discussed below; Firstly. Do you have a valid geological depth model to work with? I suspect (know) you don't because you're working with a seismic time model that works with aspects of depth domain geology that time domain seismic is incapable of resolving. Secondly. Does your depth control involve the inconsistencies that exist from a consistent surface datum to the shallowest sonic well log recording? Again I think not! In short, I conclude that you have little valid irrefutable information about what your VSP is designed to test and where the best location for the VSP might be located. If seal integrity is the expected out come of the test, I believe you should build your case on something a bit more substantial than the digital creekology that can be extracted from seismic alone. Comments anyone? Robert Pickering
Randall Hunt
2mo
Lee, I appreciate the innovative way you approach seismic plume monitoring. I have faced a similar dilemma designing an MRV system for a current project. Let’s face it, CCS produces nothing of value, it is purely a cost. Any project that relies on tax credits or carbon taxes for revenue is likely to always be ROI-challenged, meaning costs must be minimized. A system that gets the job done and nothing more is prudent in those circumstances, though not particularly satisfying for us geophysicists…. DAS VSPs are great technology, but as you highlight, lose most of their value after the plume passes . Repeat 3Ds are way too expensive. I think your approach is a good compromise. You also have a great last name. 😎
Jason Schweigert
2mo
It has been great seeing all the great comments and discussion on this post with a lot of great points being made on both sides. Like all seismic methods, VSP's have certain strengths and weaknesses. There is important information that VSP's can provide that no other method can. But in the end, what really matters are the overall objectives of the project. In the case of implementing an MMV plan we need to ask the question: Is a VSP (or any other seismic method) critical to imaging the plume over the entire MMV life cycle (which includes identifying multiples)? In many cases, the answer is yes, but we should never assume that is always the case. There are projects out there where VSP's may not be critical. Proper modelling and analysis should always be done in advance to ensure we only acquire the data we require to reach the projects objectives.
Principal, Geophysics
2moJason Eleson asked some questions that require the three Theseus 24D journal papers to fully answer. I am adding links to those papers here. https://cseg.ca/wp-content/uploads/LEE-HUNT-PART1-reviewed.pdf https://cseg.ca/wp-content/uploads/LEE-HUNT-PART2-reviewed.pdf https://cseg.ca/wp-content/uploads/LEE-HUNT-PART3-reviewed.pdf