Tesseral Ai

When we talk about the use of AI and machine learning we are also interested in how things like predictive or prescriptive analytics could lower costs. One of the issues we have with geophysics or seismic is the scarcity of the data. This is where synthetic modelling or forward modelling comes into play. Forward modelling is being used to verify the structural and stratigraphic interpretations. The synthetic seismic sections derived from forward modeling can be compared to stacked sections to verify the original interpretation. We can also use this with predictive or prescriptive analytics. In the Permian we are seeing companies using AI and machine learning which has caused the U.S. to pump out 60% more oil a day with 40% fewer workers. By extracting more oil while reducing capital expenses and manpower, companies are lowering their breakeven prices and it has fallen from $90 in 2012 to $40 in 2024. Many are predicting that AI should take that number even lower, boosting oil company margins and cash flow. Most in the oil and gas industry are conservative and that is generally because of the amount of time it takes to produce a conventional field. With the advent of shale plays we are seeing the majority of the production within the first 3 years which has challenged companies to start to look at new technologies. We at Tesseral Ai realize if we could go beyond a stack section in our analysis we could be looking at reservoir characterization. We will be producing migrated gathers which we can do AVO and inversion work with. This comes from my experience with Vastar Resources, a division of ARCO and the years I have spent working on AVO. The one thing we see coming is change and we need to manage that change. With that we need to hold onto our foundation as we change and step outwards. With Tesseral they have been focused on forward modelling but it has been focused on survey design. We want to pivot to be more involved with the full life cycle of the seismic. Soon we will launch our website and in it we will be showing our progression in moving forward with our quarterly releases. Our commitment is to our clients and being small we can be nimble. We are not a processing company but a technology company that provides software. That is our focus. We want to see other companies embrace what we are doing and some service companies provide it as a service.

The following paper looks at how seismic and Tesseral modeling can be used in ore exploration. It is based upon work done by Tesseral. This is not discussed too much in universities. It also touches on how forward modeling can be used for verifying structural and stratigraphic interpretations in geophysics and geological studies as well as reservoir engineering. We hope you enjoy, and to our American clients Happy Thanksgiving!!!!

I was invited to speak at the GeoConvention, focusing on developments from 16 years ago, as learned at Vastar Resources, a division of ARCO, the current state, and future projections. At Tesseral Ai, we are advancing the future of reservoir characterization through forward modeling to confirm structural and stratigraphic interpretations in geophysics and geological studies. Our software produces synthetic migrated gathers, AVO angle stacks, AVO attributes, etc., using forward modeling. This enables interpreters to corroborate their initial interpretations against actual data. These models are instrumental in supporting machine learning applications like predictive analytics, which are crucial for the economic assessment of the play. We believe this will change how people use our software. We still support survey design but when I took the role of president I said modeling needs to be about the full life cycle of the seismic. Hope you learn something from this old presentation. It reflects my growth in reservoir characterization.

The other day we got a chance to see how Tesseral Ai was being used by our client OGS. One of the studies they showed was about how Tesseral software was used to look at the noise from offshore wind turbines for marine mammals. With our software because we can do both acoustic and elastic it was discovered that the noise is elastic. We are making a commitment to meet more with our clients. We are listening more and with that our software is changing quite a bit. By the end of the weekend I will be releasing a paper on mining and seismic. Part of this has to do with understanding the final interpretation. We are discovering more and more how our software is being used to confirm the interpretation. Listening to our clients has allowed us to learn so much. Come December 1 we will be launching our website and we hope everyone sees through our website the value we put on our clients. Each of our presentations ends with a commitment statement.

This weekend I am finishing up Tesseral Ai website and writing a paper on the use of seismic for mining. We want to show how seismic can be used for mining by showing modeling done on Tesseral software. In the past we focused on potential fields and electromagnetic surveys for mining but lately we have been hearing more about the use of seismic for mining and modeling can support this. HiSeis has been pushing the use of seismic for mining. We are a technology company interested in developing the software to model what the results might be. We want to demonstrate this because there are many companies involved with mining. Later we will show Tesseral Engineering which is being used by a construction company. We are a small company, driving kaizen or continious development and know feedback from people push us further on development. We plan a new release every quarter. We have a commitment to our clients which we have in our presentations.

This is a rough copy of a pamphlet we are working on. It shows the type of work we do in synthetic seismic modeling. We are now moving into reservoir characterization, using migrated gathers from the synthetic model. We plan to be able to do thin bed modeling and we are looking into Frequency Dependent AVO which looks at the detection of seismic dispersion due to fluid saturation. It combines the two-term approximation of Smith and Gidlow (1987) with the spectral decomposition method used to achieve high resolution. Frequency-dependent AVO allows us to analyze AVO in thin beds. As many know I have been working on thin bed AVO for a while with Schiefer Reservoir Consulting. I looked at: 1) Using Depth Migration. 2) Utilizing Swan and higher order moveout. 3) Angle stacks that are converted to angle gathers. 4) Spectral enhancement of both high and low frequencies. As I have said many times, low frequencies reduce the sidelobes where thin beds hide and low-frequency relative inversions approximate the deterministic inversion without a wavelet or low-frequency well model which may leak through. Each step of the process was to push the frequencies a little bit. The gain in frequency and higher resolution is incremental. With synthetic models, we can see differences better and begin to understand the subsurface. We can then see all this can feed into AI or predictive analytics.

The understanding of faults is becoming critical with CCS. Most migration packages only image to 60 degrees. Duplex wave migration is designed to image beyond 60 degrees capturing steeply dipping faults which may have come off of the basement. This is important because many of the CCS targets are deep and could be effected by these faults.

It is great to be able to show to our clients where we are going. We are now working on moving into reservoir characterization with our modeling and we are developing tools for our seismic survey design. Tesseral has always been light on what we have had for seismic survey design but believe with the changes we are making that it will improve both areas. It is an exciting time for us and in a couple days we will launch our website.

Someone asked me for some information on geoscience and reservoir engineering. I decided to take this out of a textbook I have written but have not published. This textbook was written as I would attend meetings and learn about various things. I would go back to my desk and research what I did not know. It was how I learned. My exposure to reservoir engineering was through 4Ds that bp was acquiring in the North Sea to ensure gas was not coming out of solution and forming gas caps. The 4Ds were used to determine if injection wells would be used to increase the pressure-keeping the gas in solution.

When I first started, I was a field geophysicist working in the field to QC the data. We would apply geometry to the data and stack it. I was asked to go to the Netherlands to see how a Prakla crew was doing it in Dukkom, Holland. Prakla had developed a workflow and they were also using something they called Spider which did static binning of the acquisition parameters to monitor the survey acquisition parameters quickly. Goal was to duplicate what they were doing in North America. When I got back into Houston we began to work on this. We got it going and went to the field. Spider would do the fold geometry using what was acquired and create the SPS files. It was the first time we were using SPS files. Fast forward and in Tesseral Ai we are working on a project that will use seismic modeling to QC the seismic acquisition parameter changes. Most seismic acquisition is acquired using cableless telemetry systems where we do not see the data. Since the goal of acquisition is seismic imaging we can use modeling to see how changes in the seismic acquisition parameters will change the imaging. Tesseral Ai also includes some seismic design tools to analyze the data. It is strange to be doing what I was doing when I first started in the industry and working on the commercialization of this idea. The seismic modeling can then be used to help guide the seismic acquisition and interpretation. We start to see how this fits together and as we move towards AI and machine learning modeling will become more important. Seismic modeling is becoming more important for our seismic workflow.