In the final episode of Beneath the Subsurface Season One, we're focusing on Well Data Products and the full gamut of subsurface intelligence that can be gleaned from leveraging Well Data with Seismic. Caroline Brignac sits down with Jason Kegel, Ted Mirenda and Katie Fearn for a deep dive into the evolution of well data and how it’s used in today’s workflows.
In the final episode of Beneath the Subsurface Season One, we're focusing on Well Data Products and the full gamut of subsurface intelligence that can be gleaned from leveraging Well Data with Seismic. Caroline Brignac sits down with Jason Kegel, Ted Mirenda and Katie Fearn for a deep dive into the evolution of well data and how it’s used in today’s workflows.
EXPLORE MORE FROM THE EPISODE
TABLE OF CONTENTS
00:00 - Intro
01:42 - Evolution of Well Data Products at TGS
03:25 - Production Data & it's Uses
07:38 - Production Data and Thesis Work
09:09 - Longbow: A Well Performance Visualization Tool with Analytics
12:08 - What is Well Performance Data Used For?
15:04 - Validated Well Headers & Interpretation
20:26 - Well Logs and Production Data for Students, Interns & Early Career
22:30 - Historical Production and Well Data
24:43 - The Marriage of Seismic and Well Data: Interpretation
26:48 - Historical Data and Microfiche?!
29:44 - What About Offshore Well Data Products?
34:34 - How Much Gulf Of Mexico Data Does TGS Have?
39:00 - Seismic or Well Data... Why Not Both?
40:20 - Analytics Ready LAS Data (ARLAS)
43:49 - Eye Opening Data for Early Career
48:48 - TGS Projects & Careers
51:37 - Conclusion
EPISODE TRANSCRIPT
Caroline:
00:12
Hello and welcome to Beneath the Subsurface a podcast that explores the intersection of geoscience and technology. This is Caroline Brignac from the well data products group at TGS. In This episode we'll explore our well data products and how they prove to be critical datasets for any exploration and development program. So go ahead and we'll get started with introductions for today's podcast. We've got Jason Kegel with us. Jason why don't you to tell us a little bit about yourself?
Jason:
00:39
Sure. My name's Jason. I work with the geology group here at TGS. I'm a geologist I've been here for six years. I work pretty closely with our well data products and our seismic products.
Caroline:
00:50
Awesome. Thanks Jason. We also have Ted Miranda with us. Ted, why don't you tell us a little bit about yourself?
Ted:
00:55
Sure. Thank you. Caroline. Ted Mirenda. So I'm with TGS. Well, data products group. I've been here for 10 years now. A primary task was to bring production data to TGS and commercialize that product. It's been a lot of work and exciting.
Caroline:
01:12
That's awesome. I'm really excited about having Katie with us. She's a production geologists for a super major. Katie, welcome. Why don't you tell us a little bit about yourself and your experience with TGS.
Katie:
01:21
Thank you, Caroline. I am a recent graduate school graduate and I loved my time at TGS where I got to use Longbow and R360 and then I carried those things that I had learned and into my schoolwork in grad school and it's been awesome.
Caroline:
01:39
So Katie, you started with us as an intern, correct?
Katie:
01:42
Correct.
Caroline:
01:42
That's awesome. Well, we're really excited to have you here and talk a little bit about what your experience with TGS, our products and how you use it in the industry. So one thing that we know a lot about TGS is that it's known as a seismic company. However, TGS offers a wide range of other products such as products in well data. Ted, would you mind telling us a little bit about the well data products division and how it's evolved over time?
Ted:
02:07
Sure. I guess we can step back to 2002 when TGS officially acquired a little company called A2D that gave A2D's the resources to further go out and I believe in acquire Riley's electric log inventory. So that led to the largest commercial well log library. Other resources that TGS provided or enabled was the ability to digitize hard copies and raster logs to LAS. And that library has grown over time where I came into play now 10 years ago after growing the LAS library TGS made the the decision to what's next with well data, well, let's bring in production data. That's when I came into mix. We started building our production data library up. It's been a long challenging project, but it's really paid off. One of the things that critical decision we decided to do was not acquire any production data assets, but build that data from the ground floor up. That meant more work. But in the long run, it's a more valuable product.
Caroline:
03:25
So when you talk about production data, what exactly are you talking about and what does that look like?
Ted:
03:30
Well, we're talking about the full historical production record of every well in the United States. So when you think about different pieces of information that our clients use and need what the well has produced, the reservoir fluids captured from each wellbore is about as important a piece of information as you can have going forward. So we capture that information, really important to tie it to the proper wellbore and a really detailed well header record. There's a lots, a lot of other processes that we do with that as well to then provide the data to our clients.
Caroline:
04:16
So we know that we have, Jason has some experience as well as Katie with this dataset. Would you mind telling us about how you guys use it in your role in the industry?
Jason:
04:25
Sure. So I know at TGS we use the production data quite a bit, looking at our different mapping projects we have. So when we look across the entire, especially United States and look for new areas to shoot on shore seismic, we like to have a really good background information on what companies are actually producing, how much they produced in the past. Can a lot of times tell you where the, where the new plays are and it's always been said that where you found oil before you'll find oil again. And that's been proven over and over again. When we look at the Permian basin, which has been producing since, you know, the 1910, 1920s and today it's one of the biggest basins in the world and we're still finding oil there. So it's nice to really see those historical records of production and where people have gone. On top of that, the Longbow database gives you completion information so you can start really seeing where exactly within the geology has been drilled and how they have done it. So you can get some engineering insight into that as well. Over the years at TGS we've brought all that together to really start looking at new areas where clients want to go and where we can start bringing them seismic.
Caroline:
05:34
So Katie, we knew that you started off as an intern here at TGS a few years ago and we know that you worked with Jason on his team to help sort of guide where we'd go next with our products. What was your experience with the production dataset and Longbow?
Katie:
05:48
So I used the production and information during my project, both at school and at during internship to help me understand the reservoir better so that I could clear up any uncertainties that I was curious about. So for example, I use production data during my time at school to help me understand if there was any reserves left that were not taken out.
Ted:
06:19
Yeah, I know a lot of our clients then use that data to look for bypass opportunities. Another one of the many capabilities of leveraging production data. Jason talked about moving into the completion data side of what we call completion data. Kind of led that evolution. You know, horizontal drilling, unconventional tight reservoirs, fracking, I mean that led to a whole new need for different attributes captured about a well record. So we identify those pretty early on. I had been collecting those and now provide that kind of information to our clients. Not just perf intervals. What is the, what is the producing interval subsurface depth, but the length of the lateral that's being completed and produced correlating production rates, any U R S 2000 foot laterals, another way to really do better well economics and evaluation of assets. So it's, it never ends, you know, the data needs are constantly evolving and changing as industry changes and we follow that path.
Jason:
07:38
So Katie, you said that you use some of our production data with your thesis work, correct. And that was in the, in Louisiana, the Tuscaloosa Marine shale, right?
Katie:
07:47
Yes.
Jason:
07:47
So the Wells and the data that you used there, were they mostly conventional Wells or where we also tried to look at some of the unconventional Wells there too, to define that play that you are looking at.
Katie:
08:01
Right. So I would say the majority, I also focused on the lower Tuscaloosa, which was mostly conventional Wells.
Jason:
08:09
So those Wells, they helped you define that play area and then you had to go deeper and deeper into the log data. Correct. Trying to see exactly what the formation was made up. And you did a sort of a real exploration study of that lower Tuscaloosa Marine shale Longbow helped you kind of understand exactly where the production had become historically and where it might go now and where, where people are drilling currently in the Tuscaloosa Marine shale.
Katie:
08:39
Right. And we also did that with the Austin chalk too. That was another one of our big projects.
Jason:
08:44
Right. And then when you, in the group that was here all from the university of Lafayette worked with us, we also looked up into the Haynesville and looked at some of the smack over units using Longbow quite a bit, looking for trends in conventional plays historically and then seeing where those went unconventionally and if Longbow is the, the main generator of the majority of that data.
Caroline:
09:09
So for those of you listening in who may not be familiar with Longbow is that is our our visualization tool that sits on top of our well performance database. Ted, would you like to add to that?
Ted:
09:19
Yeah, that's right. So you know, production data is a fairly complex data model, right? So you need a tool to search and search your way through that data library, identify Wells that are appropriate to your project assignments so Longbow started out as really as that initial search engine. Hey, you're connecting to almost 5 million Wells, right? In a cloud based database and you're typically going in your assignment, you're going to identify subsets of Wells based on location, geology, formation, operator assets. Hey, examine these assets that are for sale and tell me if it's worth it, right? So Longbow provides that search engine. However, over the, the years and the time, we've incorporated quite a bit of analytics into the search engine. So we're really proud of that. It's if you can think of having a search engine connected to a live database of every well and include analytics, make a bubble in contour map on six month cumulative by zone, you know, all that in one. It saves time. So it's been exciting. We've had great feedback from clients and we are really focused on, Hey, what do our clients want? That's what we put in
Jason:
10:46
When you go. When you talk about analytics Ted, what has been the biggest benefit of forecasting for Longbow?
Ted:
10:54
Well, okay, so that is another good point. Production data being the historical production for the wellbore. Again, the reservoir fluid produced once me and my team, I felt we were comfortable and really good at acquiring that data. I always wanted to move into the forecasting realm as well. So we have added to the, to the product feature every single month. Now every, well all active wells get forecasted to their economic limit, giving our clients quick access to EURs. So from that perspective, I can look at historical data for an example like Katie gave about looking for bypass opportunities. Where did prior operators leave hydrocarbon in the ground with forecasting, I can look at, okay, what's the total proposed value of an asset? How much is that asset going to produce? How much remains that's already there in the, in the analytic tool. So, and again, the different analytic tools include besides mapping, probability graphs, scatter plotting and charts. It's the full gamut.
Jason:
12:08
So we have, Katie who has worked with this data as an intern. I work with this data internally with project development and sales. And then I know that I've gone out with you before and we, we sell this data, we try to give our clients opportunities to use this data. Are our clients, strictly exploration type geologists or engineers or do we have other sort of venues where this data's important in the oil and gas industry?
Ted:
12:36
You know, that's a good point. I mean, our clients cover all those gamuts. You know, one thing, again, with production data, it's a valuable piece of information across an integrated oil company. Enterprise exploration, geologists exploration is of course petroleum engineering department, reservoir engineers that have to forecast production. It's really become a big tool also in the A&D world investment banking A&D world at oil companies, business development. And that's what I like about production data. Everybody finds a use and value out of it,
Jason:
13:23
Right? And it seems everybody wants to know how long that well is going to last and where the next well is next to it. It's going to produce as much that really hard to find that information from anything other than production data.
Ted:
13:33
And what's, you know, what's, what's recently happened and I was looking at right, or like writing a paper on this topic. But you know, right now, most of the think tank forecast for supply, they're all like redoing those and lowering them, you know, the Unconventionals. And we, when we started doing our forecast models, we realized that the horizontal Wells had to be looked at differently. And the decline rates on those, those Wells now are, what would I say, exceeding what we thought they would be.
Ted:
14:08
We had this, you know, unconventional production had made perhaps a real the world with the real comfortable setting of endless oil supply and and you see the think tanks now readjusting those forecasts. So our model changed as well. We're looking at studies and how long Unconventionals are really going to produce and readjusting the EURs. And does that also have quite a bit to do with parent child relationships and how they're stacking Wells within the reservoir? It does, and right now that's what everybody's trying to figure out. That is really challenging looking at spacing, refracking spacing, how does another child affect the, the, the parent well and etc. What is the proper spacing that we try and provide the data to our clients to help them do that?
Jason:
15:04
Right. And in some of those cases you said before with our header products that we have, that really has led to Delineating some of the production data with the validated well header. Can you explain a little bit more about how the validated well header helps understand different laterals and how that traces back to production?
Ted:
15:25
Yeah. Yeah. And that's that's another key point, I think what was attractive to building production data here at TGS? You know, you go out and collect production data and for the most part, I mean, when you're getting public production data, the reality is that data is really coming in at a surface level. I mean, what does the state regulator care about? They just want to know how much did operator produce. So your severance, you're paying severance tax below the surface, they're not so much concerned about which zone is that coming from in which borehole? So here at TGS we have, we can leverage our validated well header dataset, which is our proprietary header where we've gone in, looked at the subsurface and identified missing boreholes. So we are in the process of tying our production data now to that validated header. So really moving production data down to the, to the, what we call the 12 digit API level. And that's really making a difference to our clients.
Jason:
16:39
I know it's helped internally where we've gone used the perforation information.
Ted:
16:43
That's right. Yup.
Jason:
16:45
And actually track the perforations. And I'm not sure if you, you might've done this with this, some in your internship, Katie, where we looked at the perforated intervals on the Wells and then when we are doing our cross sections, we would actually see exactly where the perforations were and see where that oil was coming from. And that helps in a lot of situations in basins where you, you don't know a lot about the basin or you're going somewhere new and you're mapping and we'd see, you know, you'd see the Austin chalk and the Buddha and the Eagleford and you try to wonder, well, where exactly in those formations are they getting the oil from? Without those perforations that we'd got from Longbow, we couldn't truly track that back. We've been doing that more and more with the help from interns when you were here a few years ago and also with our newer interns to, to really try to understand that and then provide that on another level through R360 to start understanding where these Wells are actually producing from, which in some states they don't, they don't provide that information.
Ted:
17:42
That's right. And that that really is a really neat project. I know for me and my team at the, and Ted talking about the production data, leveraging Jason and the geoscientists and the interpretation type work you do on your workstations where we can take our production, our perfs, you guys load it in, match it up with the LAS, correlate that production to the actual producing zone. It takes a while to do that, but we're doing that in projects going across different basins and it's really exciting.
Jason:
18:15
No, it's been, it's been very valuable for us that in some of the test information that Longbow has also has in some states like Oklahoma and Texas, let's say, they don't have produced water for a lot of the production. So the only things that you can look back are some of the actual, that the test data that you have where you can find that water. And then a lot of these areas where you're running analytics on some of these Wells to see when they watered out or how much water they have per volume of oil. That's the only place you can get it. And then when you max that match that back to the perforated interval, you can really start understanding some more about those horizons and how much oil or how much oil you have left, but also how much water you're getting out, which is a huge issue right now with a lot of the unconventionals is water not only how much water you're putting in to stimulate if that's what you're doing, but how much formation water you're actually taking out and that could be a, that could be that the factor in having a well that's a good well or not good at all.
Caroline:
19:19
So I know we've touched on production data and the well performance database that TGS offers, but TGS also offers other data like well logs, various types of well logs our validated well header that Jason just mentioned. Katie, I'm curious about your experience as a student getting data from TGS. Can you tell us a little bit of what that was like and how you use other well data with production data to help solve some of the, the issues you guys were running into?
Katie:
19:48
I'm sure. Well, TGS was really helpful because like Jason said, if Jason and Ted said to the state, you don't have to provide good data to the public. So TGS' well logs, their production data was far superior to anything that I saw. So it definitely helped not just at school cause I use this product at UL but I also got to use it in our projects. So it made the uncertainties that were, we were curious about less uncertain. Right. Cause the subsurface is always uncertain.
Caroline:
20:26
So to follow to build on that, Jason, how do you, how do you work more with well logs and production data together, especially when you're working with a group of young interns like Katie and her, her fellow interns
Jason:
20:39
Well one of the things that we do in our group quite a bit is either look for for new areas or sort of redefine basins that have already had had exploration. So the main thing we do when we do that as we get as many well logs as we possibly can. So that's the, the LAS that we have for those areas. Working for TGS is nice because we have access to quite a bit of data. So we pull all those together and we start just doing cross-sections and fence diagrams and make picking our formation tops so that we have a real good general understanding of the basin. As we're doing that, we're also looking at the production data. So each one of those Wells is either a producer or not a producer or maybe it was just a stratigraphic exploration well. But the reason those Wells exist are to make somebody money. So hopefully they're all producers.
Jason:
21:32
So we learned as much from a dry hole as we do from a hole that's not dry. That's where the production data comes in really handy cause we can see exactly how much oil they got out of that well when it was drilled, when it was plugged and abandoned. Some of the issues that might've gone on with it. So we can understand from looking at just some of the well logs themselves than the caliper per se, to see where you had the whole breakup and see where you might've had engineering issues with that well, where they might have crossed faults that might've caused to loss of production in certain areas. And we can tie that back using production to see exactly how these reservoirs work. And we can track that around better to see where explorationists, might need help delineating new fields or new areas. And that's where the seismic comes in with TGS to where we can try to get the seismic out to help limit some of these problems that were we might be seeing in some of the Wells.
Caroline:
22:30
Out of curiosity I know that we offer a long range of historic production data. Recently we just acquired a company called Lasser that goes back far beyond the 1970s. As a geologist, would you say that having a larger dataset going back further in time is more beneficial for you to help solve problems?
Jason:
22:54
Absolutely. So the one thing we've always ran into is not enough data, right? We always want more data. We want to see the complete picture of the entire basin. So having that data that goes farther back in time, that historic production data really helps because we have a lot of those well logs that are sort of historic historics our well logs and our Las don't stop at 1970 or earlier. The production data depending on state isn't necessarily at a strict cutoff of 1970 but that historic data really helps with that production to really start understanding how those wells were drilled. And like I said before exactly what was it producer and what wasn't producer and if it was producing, how long did it produce for? There's been lots of of technology advances that have really increased how much oil you can get out of the ground or gas you can get of the ground.
Jason:
23:45
That's on a purely engineering basis and you can start to see that in the production data, but you can really start seeing that in some of the LAS data when you start looking at the curves and understanding some of the petrophysics behind the Wells. And not only that, you start understanding the basin. So when you look at some of these really old wells, a lot of them are really shallow just to sort of understand that's as far as they could drill to. That's where the technological limit was. But depending on the basin, some, some people in the forties and 50s had drilled all the way to basement. You really want those type of data points when you're understanding the entire basin. The deeper you understand the basin, the more history you can put into it. The more basin modeling you can do. If you can understand the basin from initial infill to present day and the erosion intervals that have been between there. We see that quite a bit in our base in temperature models, which is one of the products that we do that builds off of our LAS data.
Caroline:
24:43
What other tools, interpretation tools do you use internally that TGS helps provide or provides to our clients?
Jason:
24:49
Well firstly I mentioned the basin temperature models. That's one that we, we helped build and we provide to clients and that's a product where we look at the entire basin. We pick the tops in it from 2000 to 3000 Wells from the LAS. And then we do basin temperature modeling on that entire basin with grids and horizons, start understanding the the basin from completed from basement all the way up to the top and understanding the infill. We also provide other products, sort of worldwide called our facies map browser. And this is mainly offshore, but this is looking at sequence stratigraphy within offshore basins.
Jason:
25:29
This one we also use well data and seismic data where we can and integrate the two of them to, to have a real good understanding and picture of the basin. So the geologists that use this data can jump right in to the basin and have a real good working knowledge of what's going on there. One thing in the industry, I've been in this industry for eight years now and I've seen lots of mergers and you know, lots of layoffs unfortunately with people, but groups shrink and grow all the time. And when they grow, people need to jump into new basins they've never been. So one thing that we provide with some of our well data products like the facies map browser and the basin temperature models easily help people easily get acclimated with basins they may have never worked. It's a, it's a real quick and easy way to understand the stratigraphy and understand some, some components of the basin you might not have thought about before.
Jason:
26:25
Then we've been moving on with the basin temperature model is that the background into TOC models. So actually looking at total organic carbon within the same basin using the background of our basin temperature model and then working with core labs to really understand some of our vitrinite reflectance and core data points. So that's the new thing we're doing particularly in the Permian basin.
Ted:
26:48
And I want to add another point on Lasser that Lasser acquisition, which was a, again, exciting for our team. Jason talked about the need for historical data. Sure. acquiring that data set. Now, the only way you could really replicate that public data is if you went to physically went to the individual railroad commission, district offices and loaded up a bunch of microfiche. So that data's digital. We've got it now. What's really neat is we're running it through our modern QA and QC processes. So adding data production volumes in Texas all the way back to the 30s, and then taking further, taking the lease level production data and allocating it to a well level. Nobody in industry is doing that right now from nobody from a vendor perspective. So that project that's ongoing and will be completed before the end of the year. Having historical production back to the 30s allocated to the well level, excited about that and proud of our team to get that done.
Caroline:
27:55
Not to ask a silly question, but what is microfiche is that what you said?
Ted:
28:01
I said microfiche, yeah.
Jason:
28:01
You don't remember Microfiche? (Laughter)
Caroline:
28:02
You're talking to a millennial.
Jason:
28:04
I feel so old.
Ted:
28:06
The point there is the data is not digital, it's manual, it's on microfilm. Microfiche it's lots and lots of hours of labor to recapture that data in database format. And now that we've got it, it's going to be real exciting.
Jason:
28:27
My experience with microfiche was always in elementary school going to the library. So at the library they always had stacks of microfiche that had historical newspapers from the past and you can still find them and they're really, they're almost like little slides like you remember, do you remember what slides looked like? (Laughter) No, it's done. That's true. It's already 2020. [inaudible] There was a special microfiche reader to see them. And you flip through each one of them. But that's how they always documented historical papers. So we'd go back and have to do research projects and you'd have to go find your little microfiche from the library. And when you looked it up, you would slide through and it was like a little projector screen that read the fiche from like the little, little tiny film and scrolled through the little film. So it is almost like a negative
Ted:
29:17
It's a picture of a document. So I'm not the only millennial in the room. So Katie, I'm gonna make a safe assumption that you did not know what that was?
Katie:
29:23
Nope, no, I had no idea what that was, but I have seen it in movies. So thank you for that visual like connected the two for sure.
Ted:
29:31
That's right. But that, that tells you how you know how-
Caroline:
29:37
How hard to find it, how hard to find that data is.
Ted:
29:39
That's right. There weren't computerized records back then, but we still need the data
Caroline:
29:44
Absolutely.
Katie:
29:44
So you've talked a lot about onshore, so do you offer the same kinds of products offshore as well or what do you, how does it go from onshore to offshore?
Jason:
29:58
That's a good question, actually, because with TGS and with the amount of data that we have onshore as really dense area of log data per se, so we can do areas like the Permian, the Eagleford or the DJ basin and fill them in with 5,000 Wells and pick tops and all 5,000 of those Wells. And they all have temperature points. So we can do our base in temperature models there. Offshore, it gets a little bit more difficult because there are, the data's not so close together and offshore particularly say in the Gulf of Mexico, the geology gets a little more tricky, particularly with basin temperature models because you start dealing with more salt. You start dealing with just having the water to sediment differences that you'll- we understand pretty well, but the more well data you have, the more we can make those interpretive products.
Jason:
30:55
So we have, sort of, different products offshore and like I mentioned before, we have the facies map browser is almost exclusively offshore because we can do that along mainly 2D lines, so long 2D lines that go over large areas and are- usually have a few wells connected to them in exploration areas. So the newest one of those is what we're trying to start now in Mexico and the Mexican side of the Gulf of Mexico where a few years ago we shot a really large 185,000 kilometer 2D survey called Gigante. So we interpreted that whole survey and we shot gravity and magnetics over it. So we actually have a gravity and magnetics model that we've built on that area that helps a lot in exploration, but we've also interpreted all the seismic to pick certain horizons. We would like to go a few steps further and actually understand your stratigraphic facies and your sequence stratigraphy that's in there.
Jason:
31:56
And that's what we're, we're trying to do now with the Mexican side of the Gulf of Mexico. And it's a little bit easier there because there's less wells there and a lot of the operators that are moving in there since they opened up Mexico aren't there. So they don't have as big a knowledge base as they do in the U S Gulf of Mexico. And that big large knowledge base in the U S Gulf of Mexico from the operators that have been there for 40 or 50 years has really limited multi-client type interpretation studies. Because say the Exxons or the Shells or the Chevrons have been in these basins for so long, especially the Gulf of Mexico that they have the working knowledge of those basins and they train their employees on that pretty easily. So they don't necessarily need an outside company like TGS to sort of give them the boost or the the heads up or the, the first step to get into a basin.
Jason:
32:53
Whereas in other basins around the world where we have facies, map browsers, we've had them for a while, we have new companies coming in and going more often. So they sort of like having that extra layer of knowledge that we can offer on shore. In the Gulf of Mexico though we did do a post-well analysis, which is just looking at specific wells and I think we have a little over a hundred now and they're either dry holes or or discoveries and they sort of show the stratigraphy they show why it was a dry hole or why was it a discovery. We match that up with seismic and certain areas so you can see the structures that were being drilled at the time. So we do have that. And then in the Mexico side of Mexico and the Gulf of Mexico, we have production data on both sides now.
Jason:
33:41
So we actually have the contract with the Mexican government to provide not only the seismic but the well log data in Mexico, but also the production data in Mexico. On the U.S. Gulf, we have the contract to deliver log data. So companies that drill in the U.S. Gulf of Mexico, they actually send their log data to TGS. We hold it for the 26 month timeframe. And then we clean that data up. We provide our LAS plus package. We provide that back to the BOEM or BSCE, the government entity that sort of controls the Gulf of Mexico. And then we also provide that to any other company that would like to purchase it. So we're the - TGS is actually, we've had that contract for a little over 10 years now and we've just renewed it this year.
Katie:
34:34
So like how much coverage do you have in the Gulf of Mexico? Data-Wise.
Jason:
34:38
Data-wise? So all of it really. So with the, with the recent acquisition of spectrum, we now have 2D coverage that extends all the way from Florida to the Rio Grande Valley really. So we have 2D coverage that covers, there are, TGS is a seismic company. Our core seismic area has always sort of been 3D seismic anyway, has always sort of been the Mississippi Canyon, DeSoto Canyon, Atwater Valley area. We have lots of 3D seismic. We're currently shooting seismic there. We'll just finished up some new nodal surveys there and doing reprocessing. But we have 2D and 3D coverage across the whole area and well data we have all of it. We have every well that's ever been drilled in the Gulf of Mexico.
Ted:
35:27
On the production data song for Gulf of Mexico. The data's really, really nice from that perspective. I mean every well is reported oil, gas and water, monthly production. Well tests are extensive in the Gulf of Mexico. Perhaps the federal government does a better job of reporting well test data, making sure operators are testing those Wells annually and semiannually and getting that data out to public. So you also get access to certain pressure data in there, you know, flowing tubing pressure, bottomhole pressure, et cetera. So that data sets we like working with that. And now on the Mexico side, you know, we've got full coverage of Mexico petroleum industry. There's about 21,000 Wells with production in Mexico. About 1100 of those are offshore and we have captured and calculated monthly production for all of those Wells. So that was a fun project. Learning to translate certain wellheader attributes from Spanish to English that was fun to do. Converting units of measurement down there from a, you know, average daily rates to total monthly production. Bottom line is that data's now standardized in our library monthly oil and barrels in Mexico, monthly gas and MCF water in barrels. And,looking at the data, there are world-class wells in Mexico, so I think the continued release of data from Mexico. Hopefully we'll stay on track there with the, the government releasing data. Like I said, there's there's been some really gigantic flow rates down there, particularly offshore and no reason to think there's not great opportunity there. Seismic
Katie:
37:36
Where's your seismic that you just shot in Mexico. Where does the location lies?
Jason:
37:40
So the, the 2D seismic that's there, the original Gigante is all offshore and covers the entire Mexican Gulf of Mexico 2D. So it covers everything and it even goes sort of around the horn of the Yucatan near Belize. So it covers everything sort of almost into the Caribbean. We've also been doing looking at reprocessing efforts to extend some of our, to extend the seismic onshore to offshore and the Sureste and Tampico areas. And then we're also looking at 3D programs as well.
Katie:
38:15
Very nice.
Jason:
38:16
So there's quite a bit there. And that's not the only place that we have seismic or well log data. So TGS is actually always, I always try to remind me, we have well log data worldwide. So we have data. Do you know Russia and Africa and Australia and Malaysia all over Europe. And all over South America as well. And seismic too. I sort of focus on Western hemisphere so I know a little bit more about that part, but that's still quite a quite a large area sometime. And we're we're, we're looking at wells and seismic all across, both North and South American.
Ted:
38:53
Don't forget Canada.
Jason:
38:55
And Canada too, we have quite a bit of seismic in Canada as well.
Caroline:
39:00
Nice. So one question I have for the table, we know that as TGS is predominantly a seismic company, but we also offer well data. How does that, how does that really help our clients when we offer two very different and unique datasets together?
Jason:
39:19
I think the biggest part of that is making a complete geologic picture for explorationists. So you need the seismic to really sort of understand areas where we don't have well data and that well data really helps the seismic become better. One of the good examples of that is in some of our reprocessing efforts we're doing offshore, we're incorporating as much well data as we can, particularly Sonic data so that we can really understand the velocity models. And really make sure that we can tie those velocity models when they come out and with our seismic comes out in depth that our wells tie perfectly with them. The more well data we have, the better our seismic is going to be at the end of the day. We've always tied a few Wells that we can here and there, but since TGS has so much well data, it's a real benefit to our clients to be able to use that in the seismic processing and in reprocessing as more wells come out.
Caroline:
40:20
So I'm just curious, you know, we are now offering a new product in the well data group. That's our analytics ready LAS that basically allows us to offer even more data. How do you feel about the machine learning algorithms that we're using in forecasting or with well logs? How do you feel about using that as geologists, Katie and Jason?
Jason:
40:42
So one of the things that we've noticed quite a bit with this is you get a really nice big picture and particularly with analytics ready, we like to call it just ARLAS AR-LAS is that that big picture of that first presentation you can get, particularly when it comes to velocity models in Sonic where you don't have seismic. So one of the great images, and I don't know if I can explain this well through through radio, but one of the great images that you can have is with regular well data you have lots of lots of holes. So we didn't drill every place we could and then every place we drilled through time, we didn't do every log we could do. So a lot of the well logs that we have, particularly on onshore might have one or two curves. They might have a resistivity and a gamma ray or some of the older ones just might have an SP curve.
Jason:
41:32
What can start doing with AR or the analytic ready Las is incorporate sort of Sonics into all of those logs and start understanding where we have those deviations in Sonic across the whole area where it hasn't been drilled. So from a big picture, it really helps you understand how that would tie together where you might want to drill next or what might, what interesting features you wouldn't see where a well isn't drilled without having seismic. And if you have seismic then you can tie them both together as well to kind of have a better understanding of of your depth processing.
Ted:
42:13
And I might add onto that AI question back on the production forecasting a challenge. So we're offering both methodologies now of course we have our, you know, our traditional hyperbolic curve fit type forecasting algorithms that work well and offering the physics based you know, probabilistic spread forecasting new. Your question is how do we think about that? It's like, how does the industry think about that? I know everybody's talking about it. Everyone's trying to figure it out. To me, getting a million forecast in a couple of seconds is impressive. Right? And getting that full spread on each, well a P 10 through a P 99 forecast right at your fingertips. It's powerful stuff.
Caroline:
43:07
Yeah. I'd be really curious to see where machine learning and artificial intelligence takes TGS in the future with other types of derivative products that we end up discovering and producing and really making sure that we're getting these to the industry to reduce cycle time. So I think that's pretty cool.
Jason:
43:22
Yeah, absolutely. Yeah, I think we're, we're already moving in that direction with filling in log curves and in the seismic side trying to understand different seismic bodies. So using machine learning and AI to serve as a tool to understand where salt is in a quicker, more timely fashion or to even start understanding easier ways to define horizons or define some amplitude attributes as well.
Jason:
43:49
[To Katie] So you've seen our data and played with our data and hopefully in the future is you're, you know, experiencing your geology career, you'll get to use it much more.
Ted:
44:01
I think she's just scratched the surface with our data, right. I know all that data.
Jason:
44:06
You had the unique opportunity to use it to come into our -come into the company and see what it was like to have that much data at your fingertips. Can you tell us a little bit about how, what that was like and how, how that's different from then to school to now that you're, you're in the industry.
Katie:
44:24
So I came into TGS knowing nothing, well, not knowing nothing, but you know, minimal. You think you, every time you start somewhere you like think you know something, but you really don't, which I've learned again third time.
Ted:
44:37
Right?
Katie:
44:37
So at TGS, I wouldn't say it was just, I learned how to work with all this data, which was overwhelming at first. It was like I learned how to, I don't know, act, not just like socially in an office, right? But I also learned like what's important, what's not important. It's easy to get bogged down in the details when you go from zero to 100 real quick.
Caroline:
45:03
So you've really had a unique perspective. Especially compared to a lot of us at TGS, you started off in an internship with us getting into the data and learning the data, applying the data. Right. And then I believe maybe you've even used it in your thesis.
Katie:
45:20
Right.
Caroline:
45:20
And now that you're in the industry, what has that looked like for you?
Ted:
45:26
How about, how about how do you access data being an industry now?
Katie:
45:31
When I've looked at data, it tells me, it makes me feel comfortable. It clears up uncertainties.. It's not telling me what's going on, but at least I'd like have more of a general idea. So when I look at these large amounts of data that I get for a project, let's say like I did in grad school, it's okay, I have this data. What does the data tell me? Does it tell me if it's pinching out? Does it tell me if it's, you know, this big chunk or maybe the depositional environment. That's what I looked at a lot in well logs the petrophysics.
Jason:
46:08
No, it's understandable. You get thrown a lot of data in these situations and it's how you put that together, how you can efficiently use it. And that's something that we're always trying to make easier for people. It helps in a lot of situations, particularly in, in super major type of companies or in a lot of different companies, even smaller companies that they have geo techs that efficiently use our data before they give it to you. Right? So a lot of times you never, you'll never get to see the first part of, you know, where did this data come from because it all just ends up on your desktop. Right?
Katie:
46:42
Right. So like I, what I liked about my experience I guess at TGS is I saw the beginnings, right? What a geotech would put it in. So I like got to see that visual fresh or put my own spin on it when we were using Longbow. So making those bubble plots or looking at URs and decline curves. I don't have, I don't, I haven't gotten that experience yet, but I'm a Guppy.
Caroline:
47:10
So it was like you were getting access to data sets such as the, you know, the EURs and the forecasting database that you probably didn't necessarily have access to while you were working on your masters.
Katie:
47:21
Right. And didn't know about until it came to TGS.
Ted:
47:26
And the ability to build that project from scratch. I imagine a lot of times now in industry, you walk in and sit down and there are gigantic projects already existing and workflows established as opposed to like starting at the beginning.
Katie:
47:46
Right. Which is overwhelming. Like I remember Jason was like, Hey, y'all are going to map from Mississippi, Louisiana and Texas. That was very overwhelming. Now I just, you know, you get a project and it, someone's already, most of the time, I don't know picked through it. So you don't, it's not very fresh.
Jason:
48:09
But now you're not afraid of the deep end of the pool.
Katie:
48:10
I don't know about that...
Jason:
48:10
Right. We threw you right in the deep end and I, you can swim. You're ready to go.
Katie:
48:18
Oh no. I'm still learning.
Jason:
48:18
Well that's good. Never wanna stop learning.
Ted:
48:22
We're all still learning.
Katie:
48:22
Right. But I'm really still learning. As a new worker bee.
Jason:
48:30
So Katie, is there anything we haven't seen you in a little while? I know that you're, you're in Louisiana now. Is there anything that you want to ask us that you're interested in from a, from your perspective after you've graduated and are now moving onto bigger and better things that might help you in the future?
Katie:
48:48
Maybe not something that would- maybe wouldn't help me in the future, but also help other people that are looking for jobs. Is, are y'all looking for employment? Like looking to employ anyone or what does that look like? It sounds like you're doing a lot of work. So do you have people to fill these positions or are you, how does that go for y'all? Do you even know?
Jason:
49:10
Well, that's one of those great HR questions where, you know, we're always, we're always just busy enough to need new people. (Laughter)
Caroline:
49:20
And I think with, you know, new departments that were growing especially new datasets like Ted is talking about Mexico and Canada, I feel like it really helps to position us to grow, you know, as a company as a whole. So opportunities are always always coming up. Yeah.
Jason:
49:36
I know particularly with our internship program, we're always looking for, you know, young, exciting new talent that can, that can come in and help us out. But also like you did learn about data from sort of the bottom up and take that knowledge base to other companies. So we don't only like training people to come and stay with us or we're perfectly happy bringing in interns and having them go out in the world and and learn something from us that they can bring somewhere else.
Katie:
50:06
Oh sorry. I would say that that's why I like had not, I think that working at TGS was nice for others to see cause they knew that I had experience I guess with production data, which is a cool talking point I think.
Caroline:
50:22
And just to build off of that, Ted has done a really great job building this new initiative, which is getting our well performance data in the universities to work with people like you, Katie, while you were getting your masters to make sure that we're able to provide data to other other programs and get geologists or young geologists access to data sets that they wouldn't have or wouldn't be familiar with whenever they're entering the workforce.
Ted:
50:48
That's right. So, you know, we're happy to donate donate our products, donate production data and Longbow to the universities. As you know, at ULL they brought it into the geoscience and engineering groups. And now we're sitting on the, what the 20 workstations in the lab and part of the curriculum. So it's exciting at the same time, giving the students access to these data products learning actual, you know, working product tools. When they do get hired and hit the, hit the workforce, they're ahead of the game and ready to go. Now, from my selfish perspective, it helps to get feedback and make the products better. So it's a win win for both.
Caroline:
51:37
Well, thanks everyone for coming out today and having this conversation, you know, hanging out, covering a lot of really awesome topics, kind of, you know, exploring where TGS is headed next, where we've been, where we're going. Katie, you know, especially thanks to you for coming all the way from New Orleans to sit with us and kind of give us your insight and your opinions and let us know how it's, how the journey has been for you. So thanks, Jason. Thanks Ted looking forward to the next, the next episode.
Katie:
52:01
Thank you for having me.
Jason:
52:03
Yeah, thanks Katie, it's been great
Ted:
52:04
Thank you.
Caroline: 00:12 Hello and welcome to Beneath the Subsurface a podcast that explores the intersection of geoscience and technology. This is Caroline Brignac from the well data products group at TGS. In This episode we'll explore our well data products and how they prove to be critical datasets for any exploration and development program. So go ahead and we'll get started with introductions for today's podcast. We've got Jason Kegel with us. Jason why don't you to tell us a little bit about yourself?
Jason: 00:39 Sure. My name's Jason. I work with the geology group here at TGS. I'm a geologist I've been here for six years. I work pretty closely with our well data products and our seismic products.
Caroline: 00:50 Awesome. Thanks Jason. We also have Ted Miranda with us. Ted, why don't you tell us a little bit about yourself?
Ted: 00:55 Sure. Thank you. Caroline. Ted Mirenda. So I'm with TGS. Well, data products group. I've been here for 10 years now. A primary task was to bring production data to TGS and commercialize that product. It's been a lot of work and exciting.
Caroline: 01:12 That's awesome. I'm really excited about having Katie with us. She's a production geologists for a super major. Katie, welcome. Why don't you tell us a little bit about yourself and your experience with TGS.
Katie: 01:21 Thank you, Caroline. I am a recent graduate school graduate and I loved my time at TGS where I got to use Longbow and R360 and then I carried those things that I had learned and into my schoolwork in grad school and it's been awesome.
Caroline: 01:39 So Katie, you started with us as an intern, correct?
Katie: 01:42 Correct.
Caroline: 01:42 That's awesome. Well, we're really excited to have you here and talk a little bit about what your experience with TGS, our products and how you use it in the industry. So one thing that we know a lot about TGS is that it's known as a seismic company. However, TGS offers a wide range of other products such as products in well data. Ted, would you mind telling us a little bit about the well data products division and how it's evolved over time?
Ted: 02:07 Sure. I guess we can step back to 2002 when TGS officially acquired a little company called A2D that gave A2D's the resources to further go out and I believe in acquire Riley's electric log inventory. So that led to the largest commercial well log library. Other resources that TGS provided or enabled was the ability to digitize hard copies and raster logs to LAS. And that library has grown over time where I came into play now 10 years ago after growing the LAS library TGS made the the decision to what's next with well data, well, let's bring in production data. That's when I came into mix. We started building our production data library up. It's been a long challenging project, but it's really paid off. One of the things that critical decision we decided to do was not acquire any production data assets, but build that data from the ground floor up. That meant more work. But in the long run, it's a more valuable product.
Caroline: 03:25 So when you talk about production data, what exactly are you talking about and what does that look like?
Ted: 03:30 Well, we're talking about the full historical production record of every well in the United States. So when you think about different pieces of information that our clients use and need what the well has produced, the reservoir fluids captured from each wellbore is about as important a piece of information as you can have going forward. So we capture that information, really important to tie it to the proper wellbore and a really detailed well header record. There's a lots, a lot of other processes that we do with that as well to then provide the data to our clients.
Caroline: 04:16 So we know that we have, Jason has some experience as well as Katie with this dataset. Would you mind telling us about how you guys use it in your role in the industry?
Jason: 04:25 Sure. So I know at TGS we use the production data quite a bit, looking at our different mapping projects we have. So when we look across the entire, especially United States and look for new areas to shoot on shore seismic, we like to have a really good background information on what companies are actually producing, how much they produced in the past. Can a lot of times tell you where the, where the new plays are and it's always been said that where you found oil before you'll find oil again. And that's been proven over and over again. When we look at the Permian basin, which has been producing since, you know, the 1910, 1920s and today it's one of the biggest basins in the world and we're still finding oil there. So it's nice to really see those historical records of production and where people have gone. On top of that, the Longbow database gives you completion information so you can start really seeing where exactly within the geology has been drilled and how they have done it. So you can get some engineering insight into that as well. Over the years at TGS we've brought all that together to really start looking at new areas where clients want to go and where we can start bringing them seismic.
Caroline: 05:34 So Katie, we knew that you started off as an intern here at TGS a few years ago and we know that you worked with Jason on his team to help sort of guide where we'd go next with our products. What was your experience with the production dataset and Longbow?
Katie: 05:48 So I used the production and information during my project, both at school and at during internship to help me understand the reservoir better so that I could clear up any uncertainties that I was curious about. So for example, I use production data during my time at school to help me understand if there was any reserves left that were not taken out.
Ted: 06:19 Yeah, I know a lot of our clients then use that data to look for bypass opportunities. Another one of the many capabilities of leveraging production data. Jason talked about moving into the completion data side of what we call completion data. Kind of led that evolution. You know, horizontal drilling, unconventional tight reservoirs, fracking, I mean that led to a whole new need for different attributes captured about a well record. So we identify those pretty early on. I had been collecting those and now provide that kind of information to our clients. Not just perf intervals. What is the, what is the producing interval subsurface depth, but the length of the lateral that's being completed and produced correlating production rates, any U R S 2000 foot laterals, another way to really do better well economics and evaluation of assets. So it's, it never ends, you know, the data needs are constantly evolving and changing as industry changes and we follow that path.
Jason: 07:38 So Katie, you said that you use some of our production data with your thesis work, correct. And that was in the, in Louisiana, the Tuscaloosa Marine shale, right?
Katie: 07:47 Yes.
Jason: 07:47 So the Wells and the data that you used there, were they mostly conventional Wells or where we also tried to look at some of the unconventional Wells there too, to define that play that you are looking at.
Katie: 08:01 Right. So I would say the majority, I also focused on the lower Tuscaloosa, which was mostly conventional Wells.
Jason: 08:09 So those Wells, they helped you define that play area and then you had to go deeper and deeper into the log data. Correct. Trying to see exactly what the formation was made up. And you did a sort of a real exploration study of that lower Tuscaloosa Marine shale Longbow helped you kind of understand exactly where the production had become historically and where it might go now and where, where people are drilling currently in the Tuscaloosa Marine shale.
Katie: 08:39 Right. And we also did that with the Austin chalk too. That was another one of our big projects.
Jason: 08:44 Right. And then when you, in the group that was here all from the university of Lafayette worked with us, we also looked up into the Haynesville and looked at some of the smack over units using Longbow quite a bit, looking for trends in conventional plays historically and then seeing where those went unconventionally and if Longbow is the, the main generator of the majority of that data.
Caroline: 09:09 So for those of you listening in who may not be familiar with Longbow is that is our our visualization tool that sits on top of our well performance database. Ted, would you like to add to that?
Ted: 09:19 Yeah, that's right. So you know, production data is a fairly complex data model, right? So you need a tool to search and search your way through that data library, identify Wells that are appropriate to your project assignments so Longbow started out as really as that initial search engine. Hey, you're connecting to almost 5 million Wells, right? In a cloud based database and you're typically going in your assignment, you're going to identify subsets of Wells based on location, geology, formation, operator assets. Hey, examine these assets that are for sale and tell me if it's worth it, right? So Longbow provides that search engine. However, over the, the years and the time, we've incorporated quite a bit of analytics into the search engine. So we're really proud of that. It's if you can think of having a search engine connected to a live database of every well and include analytics, make a bubble in contour map on six month cumulative by zone, you know, all that in one. It saves time. So it's been exciting. We've had great feedback from clients and we are really focused on, Hey, what do our clients want? That's what we put in
Jason: 10:46 When you go. When you talk about analytics Ted, what has been the biggest benefit of forecasting for Longbow?
Ted: 10:54 Well, okay, so that is another good point. Production data being the historical production for the wellbore. Again, the reservoir fluid produced once me and my team, I felt we were comfortable and really good at acquiring that data. I always wanted to move into the forecasting realm as well. So we have added to the, to the product feature every single month. Now every, well all active wells get forecasted to their economic limit, giving our clients quick access to EURs. So from that perspective, I can look at historical data for an example like Katie gave about looking for bypass opportunities. Where did prior operators leave hydrocarbon in the ground with forecasting, I can look at, okay, what's the total proposed value of an asset? How much is that asset going to produce? How much remains that's already there in the, in the analytic tool. So, and again, the different analytic tools include besides mapping, probability graphs, scatter plotting and charts. It's the full gamut.
Jason: 12:08 So we have, Katie who has worked with this data as an intern. I work with this data internally with project development and sales. And then I know that I've gone out with you before and we, we sell this data, we try to give our clients opportunities to use this data. Are our clients, strictly exploration type geologists or engineers or do we have other sort of venues where this data's important in the oil and gas industry?
Ted: 12:36 You know, that's a good point. I mean, our clients cover all those gamuts. You know, one thing, again, with production data, it's a valuable piece of information across an integrated oil company. Enterprise exploration, geologists exploration is of course petroleum engineering department, reservoir engineers that have to forecast production. It's really become a big tool also in the A&D world investment banking A&D world at oil companies, business development. And that's what I like about production data. Everybody finds a use and value out of it,
Jason: 13:23 Right? And it seems everybody wants to know how long that well is going to last and where the next well is next to it. It's going to produce as much that really hard to find that information from anything other than production data.
Ted: 13:33 And what's, you know, what's, what's recently happened and I was looking at right, or like writing a paper on this topic. But you know, right now, most of the think tank forecast for supply, they're all like redoing those and lowering them, you know, the Unconventionals. And we, when we started doing our forecast models, we realized that the horizontal Wells had to be looked at differently. And the decline rates on those, those Wells now are, what would I say, exceeding what we thought they would be.
Ted: 14:08 We had this, you know, unconventional production had made perhaps a real the world with the real comfortable setting of endless oil supply and and you see the think tanks now readjusting those forecasts. So our model changed as well. We're looking at studies and how long Unconventionals are really going to produce and readjusting the EURs. And does that also have quite a bit to do with parent child relationships and how they're stacking Wells within the reservoir? It does, and right now that's what everybody's trying to figure out. That is really challenging looking at spacing, refracking spacing, how does another child affect the, the, the parent well and etc. What is the proper spacing that we try and provide the data to our clients to help them do that?
Jason: 15:04 Right. And in some of those cases you said before with our header products that we have, that really has led to Delineating some of the production data with the validated well header. Can you explain a little bit more about how the validated well header helps understand different laterals and how that traces back to production?
Ted: 15:25 Yeah. Yeah. And that's that's another key point, I think what was attractive to building production data here at TGS? You know, you go out and collect production data and for the most part, I mean, when you're getting public production data, the reality is that data is really coming in at a surface level. I mean, what does the state regulator care about? They just want to know how much did operator produce. So your severance, you're paying severance tax below the surface, they're not so much concerned about which zone is that coming from in which borehole? So here at TGS we have, we can leverage our validated well header dataset, which is our proprietary header where we've gone in, looked at the subsurface and identified missing boreholes. So we are in the process of tying our production data now to that validated header. So really moving production data down to the, to the, what we call the 12 digit API level. And that's really making a difference to our clients.
Jason: 16:39 I know it's helped internally where we've gone used the perforation information.
Ted: 16:43 That's right. Yup.
Jason: 16:45 And actually track the perforations. And I'm not sure if you, you might've done this with this, some in your internship, Katie, where we looked at the perforated intervals on the Wells and then when we are doing our cross sections, we would actually see exactly where the perforations were and see where that oil was coming from. And that helps in a lot of situations in basins where you, you don't know a lot about the basin or you're going somewhere new and you're mapping and we'd see, you know, you'd see the Austin chalk and the Buddha and the Eagleford and you try to wonder, well, where exactly in those formations are they getting the oil from? Without those perforations that we'd got from Longbow, we couldn't truly track that back. We've been doing that more and more with the help from interns when you were here a few years ago and also with our newer interns to, to really try to understand that and then provide that on another level through R360 to start understanding where these Wells are actually producing from, which in some states they don't, they don't provide that information.
Ted: 17:42 That's right. And that that really is a really neat project. I know for me and my team at the, and Ted talking about the production data, leveraging Jason and the geoscientists and the interpretation type work you do on your workstations where we can take our production, our perfs, you guys load it in, match it up with the LAS, correlate that production to the actual producing zone. It takes a while to do that, but we're doing that in projects going across different basins and it's really exciting.
Jason: 18:15 No, it's been, it's been very valuable for us that in some of the test information that Longbow has also has in some states like Oklahoma and Texas, let's say, they don't have produced water for a lot of the production. So the only things that you can look back are some of the actual, that the test data that you have where you can find that water. And then a lot of these areas where you're running analytics on some of these Wells to see when they watered out or how much water they have per volume of oil. That's the only place you can get it. And then when you max that match that back to the perforated interval, you can really start understanding some more about those horizons and how much oil or how much oil you have left, but also how much water you're getting out, which is a huge issue right now with a lot of the unconventionals is water not only how much water you're putting in to stimulate if that's what you're doing, but how much formation water you're actually taking out and that could be a, that could be that the factor in having a well that's a good well or not good at all.
Caroline: 19:19 So I know we've touched on production data and the well performance database that TGS offers, but TGS also offers other data like well logs, various types of well logs our validated well header that Jason just mentioned. Katie, I'm curious about your experience as a student getting data from TGS. Can you tell us a little bit of what that was like and how you use other well data with production data to help solve some of the, the issues you guys were running into?
Katie: 19:48 I'm sure. Well, TGS was really helpful because like Jason said, if Jason and Ted said to the state, you don't have to provide good data to the public. So TGS' well logs, their production data was far superior to anything that I saw. So it definitely helped not just at school cause I use this product at UL but I also got to use it in our projects. So it made the uncertainties that were, we were curious about less uncertain. Right. Cause the subsurface is always uncertain.
Caroline: 20:26 So to follow to build on that, Jason, how do you, how do you work more with well logs and production data together, especially when you're working with a group of young interns like Katie and her, her fellow interns
Jason: 20:39 Well one of the things that we do in our group quite a bit is either look for for new areas or sort of redefine basins that have already had had exploration. So the main thing we do when we do that as we get as many well logs as we possibly can. So that's the, the LAS that we have for those areas. Working for TGS is nice because we have access to quite a bit of data. So we pull all those together and we start just doing cross-sections and fence diagrams and make picking our formation tops so that we have a real good general understanding of the basin. As we're doing that, we're also looking at the production data. So each one of those Wells is either a producer or not a producer or maybe it was just a stratigraphic exploration well. But the reason those Wells exist are to make somebody money. So hopefully they're all producers.
Jason: 21:32 So we learned as much from a dry hole as we do from a hole that's not dry. That's where the production data comes in really handy cause we can see exactly how much oil they got out of that well when it was drilled, when it was plugged and abandoned. Some of the issues that might've gone on with it. So we can understand from looking at just some of the well logs themselves than the caliper per se, to see where you had the whole breakup and see where you might've had engineering issues with that well, where they might have crossed faults that might've caused to loss of production in certain areas. And we can tie that back using production to see exactly how these reservoirs work. And we can track that around better to see where explorationists, might need help delineating new fields or new areas. And that's where the seismic comes in with TGS to where we can try to get the seismic out to help limit some of these problems that were we might be seeing in some of the Wells.
Caroline: 22:30 Out of curiosity I know that we offer a long range of historic production data. Recently we just acquired a company called Lasser that goes back far beyond the 1970s. As a geologist, would you say that having a larger dataset going back further in time is more beneficial for you to help solve problems?
Jason: 22:54 Absolutely. So the one thing we've always ran into is not enough data, right? We always want more data. We want to see the complete picture of the entire basin. So having that data that goes farther back in time, that historic production data really helps because we have a lot of those well logs that are sort of historic historics our well logs and our Las don't stop at 1970 or earlier. The production data depending on state isn't necessarily at a strict cutoff of 1970 but that historic data really helps with that production to really start understanding how those wells were drilled. And like I said before exactly what was it producer and what wasn't producer and if it was producing, how long did it produce for? There's been lots of of technology advances that have really increased how much oil you can get out of the ground or gas you can get of the ground.
Jason: 23:45 That's on a purely engineering basis and you can start to see that in the production data, but you can really start seeing that in some of the LAS data when you start looking at the curves and understanding some of the petrophysics behind the Wells. And not only that, you start understanding the basin. So when you look at some of these really old wells, a lot of them are really shallow just to sort of understand that's as far as they could drill to. That's where the technological limit was. But depending on the basin, some, some people in the forties and 50s had drilled all the way to basement. You really want those type of data points when you're understanding the entire basin. The deeper you understand the basin, the more history you can put into it. The more basin modeling you can do. If you can understand the basin from initial infill to present day and the erosion intervals that have been between there. We see that quite a bit in our base in temperature models, which is one of the products that we do that builds off of our LAS data.
Caroline: 24:43 What other tools, interpretation tools do you use internally that TGS helps provide or provides to our clients?
Jason: 24:49 Well firstly I mentioned the basin temperature models. That's one that we, we helped build and we provide to clients and that's a product where we look at the entire basin. We pick the tops in it from 2000 to 3000 Wells from the LAS. And then we do basin temperature modeling on that entire basin with grids and horizons, start understanding the the basin from completed from basement all the way up to the top and understanding the infill. We also provide other products, sort of worldwide called our facies map browser. And this is mainly offshore, but this is looking at sequence stratigraphy within offshore basins.
Jason: 25:29 This one we also use well data and seismic data where we can and integrate the two of them to, to have a real good understanding and picture of the basin. So the geologists that use this data can jump right in to the basin and have a real good working knowledge of what's going on there. One thing in the industry, I've been in this industry for eight years now and I've seen lots of mergers and you know, lots of layoffs unfortunately with people, but groups shrink and grow all the time. And when they grow, people need to jump into new basins they've never been. So one thing that we provide with some of our well data products like the facies map browser and the basin temperature models easily help people easily get acclimated with basins they may have never worked. It's a, it's a real quick and easy way to understand the stratigraphy and understand some, some components of the basin you might not have thought about before.
Jason: 26:25 Then we've been moving on with the basin temperature model is that the background into TOC models. So actually looking at total organic carbon within the same basin using the background of our basin temperature model and then working with core labs to really understand some of our vitrinite reflectance and core data points. So that's the new thing we're doing particularly in the Permian basin.
Ted: 26:48 And I want to add another point on Lasser that Lasser acquisition, which was a, again, exciting for our team. Jason talked about the need for historical data. Sure. acquiring that data set. Now, the only way you could really replicate that public data is if you went to physically went to the individual railroad commission, district offices and loaded up a bunch of microfiche. So that data's digital. We've got it now. What's really neat is we're running it through our modern QA and QC processes. So adding data production volumes in Texas all the way back to the 30s, and then taking further, taking the lease level production data and allocating it to a well level. Nobody in industry is doing that right now from nobody from a vendor perspective. So that project that's ongoing and will be completed before the end of the year. Having historical production back to the 30s allocated to the well level, excited about that and proud of our team to get that done.
Caroline: 27:55 Not to ask a silly question, but what is microfiche is that what you said?
Ted: 28:01 I said microfiche, yeah.
Jason: 28:01 You don't remember Microfiche? (Laughter)
Caroline: 28:02 You're talking to a millennial.
Jason: 28:04 I feel so old.
Ted: 28:06 The point there is the data is not digital, it's manual, it's on microfilm. Microfiche it's lots and lots of hours of labor to recapture that data in database format. And now that we've got it, it's going to be real exciting.
Jason: 28:27 My experience with microfiche was always in elementary school going to the library. So at the library they always had stacks of microfiche that had historical newspapers from the past and you can still find them and they're really, they're almost like little slides like you remember, do you remember what slides looked like? (Laughter) No, it's done. That's true. It's already 2020. [inaudible] There was a special microfiche reader to see them. And you flip through each one of them. But that's how they always documented historical papers. So we'd go back and have to do research projects and you'd have to go find your little microfiche from the library. And when you looked it up, you would slide through and it was like a little projector screen that read the fiche from like the little, little tiny film and scrolled through the little film. So it is almost like a negative
Ted: 29:17 It's a picture of a document. So I'm not the only millennial in the room. So Katie, I'm gonna make a safe assumption that you did not know what that was?
Katie: 29:23 Nope, no, I had no idea what that was, but I have seen it in movies. So thank you for that visual like connected the two for sure.
Ted: 29:31 That's right. But that, that tells you how you know how-
Caroline: 29:37 How hard to find it, how hard to find that data is.
Ted: 29:39 That's right. There weren't computerized records back then, but we still need the data
Caroline: 29:44 Absolutely.
Katie: 29:44 So you've talked a lot about onshore, so do you offer the same kinds of products offshore as well or what do you, how does it go from onshore to offshore?
Jason: 29:58 That's a good question, actually, because with TGS and with the amount of data that we have onshore as really dense area of log data per se, so we can do areas like the Permian, the Eagleford or the DJ basin and fill them in with 5,000 Wells and pick tops and all 5,000 of those Wells. And they all have temperature points. So we can do our base in temperature models there. Offshore, it gets a little bit more difficult because there are, the data's not so close together and offshore particularly say in the Gulf of Mexico, the geology gets a little more tricky, particularly with basin temperature models because you start dealing with more salt. You start dealing with just having the water to sediment differences that you'll- we understand pretty well, but the more well data you have, the more we can make those interpretive products.
Jason: 30:55 So we have, sort of, different products offshore and like I mentioned before, we have the facies map browser is almost exclusively offshore because we can do that along mainly 2D lines, so long 2D lines that go over large areas and are- usually have a few wells connected to them in exploration areas. So the newest one of those is what we're trying to start now in Mexico and the Mexican side of the Gulf of Mexico where a few years ago we shot a really large 185,000 kilometer 2D survey called Gigante. So we interpreted that whole survey and we shot gravity and magnetics over it. So we actually have a gravity and magnetics model that we've built on that area that helps a lot in exploration, but we've also interpreted all the seismic to pick certain horizons. We would like to go a few steps further and actually understand your stratigraphic facies and your sequence stratigraphy that's in there.
Jason: 31:56 And that's what we're, we're trying to do now with the Mexican side of the Gulf of Mexico. And it's a little bit easier there because there's less wells there and a lot of the operators that are moving in there since they opened up Mexico aren't there. So they don't have as big a knowledge base as they do in the U S Gulf of Mexico. And that big large knowledge base in the U S Gulf of Mexico from the operators that have been there for 40 or 50 years has really limited multi-client type interpretation studies. Because say the Exxons or the Shells or the Chevrons have been in these basins for so long, especially the Gulf of Mexico that they have the working knowledge of those basins and they train their employees on that pretty easily. So they don't necessarily need an outside company like TGS to sort of give them the boost or the the heads up or the, the first step to get into a basin.
Jason: 32:53 Whereas in other basins around the world where we have facies, map browsers, we've had them for a while, we have new companies coming in and going more often. So they sort of like having that extra layer of knowledge that we can offer on shore. In the Gulf of Mexico though we did do a post-well analysis, which is just looking at specific wells and I think we have a little over a hundred now and they're either dry holes or or discoveries and they sort of show the stratigraphy they show why it was a dry hole or why was it a discovery. We match that up with seismic and certain areas so you can see the structures that were being drilled at the time. So we do have that. And then in the Mexico side of Mexico and the Gulf of Mexico, we have production data on both sides now.
Jason: 33:41 So we actually have the contract with the Mexican government to provide not only the seismic but the well log data in Mexico, but also the production data in Mexico. On the U.S. Gulf, we have the contract to deliver log data. So companies that drill in the U.S. Gulf of Mexico, they actually send their log data to TGS. We hold it for the 26 month timeframe. And then we clean that data up. We provide our LAS plus package. We provide that back to the BOEM or BSCE, the government entity that sort of controls the Gulf of Mexico. And then we also provide that to any other company that would like to purchase it. So we're the - TGS is actually, we've had that contract for a little over 10 years now and we've just renewed it this year.
Katie: 34:34 So like how much coverage do you have in the Gulf of Mexico? Data-Wise.
Jason: 34:38 Data-wise? So all of it really. So with the, with the recent acquisition of spectrum, we now have 2D coverage that extends all the way from Florida to the Rio Grande Valley really. So we have 2D coverage that covers, there are, TGS is a seismic company. Our core seismic area has always sort of been 3D seismic anyway, has always sort of been the Mississippi Canyon, DeSoto Canyon, Atwater Valley area. We have lots of 3D seismic. We're currently shooting seismic there. We'll just finished up some new nodal surveys there and doing reprocessing. But we have 2D and 3D coverage across the whole area and well data we have all of it. We have every well that's ever been drilled in the Gulf of Mexico.
Ted: 35:27 On the production data song for Gulf of Mexico. The data's really, really nice from that perspective. I mean every well is reported oil, gas and water, monthly production. Well tests are extensive in the Gulf of Mexico. Perhaps the federal government does a better job of reporting well test data, making sure operators are testing those Wells annually and semiannually and getting that data out to public. So you also get access to certain pressure data in there, you know, flowing tubing pressure, bottomhole pressure, et cetera. So that data sets we like working with that. And now on the Mexico side, you know, we've got full coverage of Mexico petroleum industry. There's about 21,000 Wells with production in Mexico. About 1100 of those are offshore and we have captured and calculated monthly production for all of those Wells. So that was a fun project. Learning to translate certain wellheader attributes from Spanish to English that was fun to do. Converting units of measurement down there from a, you know, average daily rates to total monthly production. Bottom line is that data's now standardized in our library monthly oil and barrels in Mexico, monthly gas and MCF water in barrels. And,looking at the data, there are world-class wells in Mexico, so I think the continued release of data from Mexico. Hopefully we'll stay on track there with the, the government releasing data. Like I said, there's there's been some really gigantic flow rates down there, particularly offshore and no reason to think there's not great opportunity there. Seismic
Katie: 37:36 Where's your seismic that you just shot in Mexico. Where does the location lies?
Jason: 37:40 So the, the 2D seismic that's there, the original Gigante is all offshore and covers the entire Mexican Gulf of Mexico 2D. So it covers everything and it even goes sort of around the horn of the Yucatan near Belize. So it covers everything sort of almost into the Caribbean. We've also been doing looking at reprocessing efforts to extend some of our, to extend the seismic onshore to offshore and the Sureste and Tampico areas. And then we're also looking at 3D programs as well.
Katie: 38:15 Very nice.
Jason: 38:16 So there's quite a bit there. And that's not the only place that we have seismic or well log data. So TGS is actually always, I always try to remind me, we have well log data worldwide. So we have data. Do you know Russia and Africa and Australia and Malaysia all over Europe. And all over South America as well. And seismic too. I sort of focus on Western hemisphere so I know a little bit more about that part, but that's still quite a quite a large area sometime. And we're we're, we're looking at wells and seismic all across, both North and South American.
Ted: 38:53 Don't forget Canada.
Jason: 38:55 And Canada too, we have quite a bit of seismic in Canada as well.
Caroline: 39:00 Nice. So one question I have for the table, we know that as TGS is predominantly a seismic company, but we also offer well data. How does that, how does that really help our clients when we offer two very different and unique datasets together?
Jason: 39:19 I think the biggest part of that is making a complete geologic picture for explorationists. So you need the seismic to really sort of understand areas where we don't have well data and that well data really helps the seismic become better. One of the good examples of that is in some of our reprocessing efforts we're doing offshore, we're incorporating as much well data as we can, particularly Sonic data so that we can really understand the velocity models. And really make sure that we can tie those velocity models when they come out and with our seismic comes out in depth that our wells tie perfectly with them. The more well data we have, the better our seismic is going to be at the end of the day. We've always tied a few Wells that we can here and there, but since TGS has so much well data, it's a real benefit to our clients to be able to use that in the seismic processing and in reprocessing as more wells come out.
Caroline: 40:20 So I'm just curious, you know, we are now offering a new product in the well data group. That's our analytics ready LAS that basically allows us to offer even more data. How do you feel about the machine learning algorithms that we're using in forecasting or with well logs? How do you feel about using that as geologists, Katie and Jason?
Jason: 40:42 So one of the things that we've noticed quite a bit with this is you get a really nice big picture and particularly with analytics ready, we like to call it just ARLAS AR-LAS is that that big picture of that first presentation you can get, particularly when it comes to velocity models in Sonic where you don't have seismic. So one of the great images, and I don't know if I can explain this well through through radio, but one of the great images that you can have is with regular well data you have lots of lots of holes. So we didn't drill every place we could and then every place we drilled through time, we didn't do every log we could do. So a lot of the well logs that we have, particularly on onshore might have one or two curves. They might have a resistivity and a gamma ray or some of the older ones just might have an SP curve.
Jason: 41:32 What can start doing with AR or the analytic ready Las is incorporate sort of Sonics into all of those logs and start understanding where we have those deviations in Sonic across the whole area where it hasn't been drilled. So from a big picture, it really helps you understand how that would tie together where you might want to drill next or what might, what interesting features you wouldn't see where a well isn't drilled without having seismic. And if you have seismic then you can tie them both together as well to kind of have a better understanding of of your depth processing.
Ted: 42:13 And I might add onto that AI question back on the production forecasting a challenge. So we're offering both methodologies now of course we have our, you know, our traditional hyperbolic curve fit type forecasting algorithms that work well and offering the physics based you know, probabilistic spread forecasting new. Your question is how do we think about that? It's like, how does the industry think about that? I know everybody's talking about it. Everyone's trying to figure it out. To me, getting a million forecast in a couple of seconds is impressive. Right? And getting that full spread on each, well a P 10 through a P 99 forecast right at your fingertips. It's powerful stuff.
Speaker 2: 43:07 Yeah. I'd be really curious to see where machine learning and artificial intelligence takes TGS in the future with other types of derivative products that we end up discovering and producing and really making sure that we're getting these to the industry to reduce cycle time. So I think that's pretty cool.
Jason: 43:22 Yeah, absolutely. Yeah, I think we're, we're already moving in that direction with filling in log curves and in the seismic side trying to understand different seismic bodies. So using machine learning and AI to serve as a tool to understand where salt is in a quicker, more timely fashion or to even start understanding easier ways to define horizons or define some amplitude attributes as well.
Jason: 43:49 [To Katie] So you've seen our data and played with our data and hopefully in the future is you're, you know, experiencing your geology career, you'll get to use it much more.
Ted: 44:01 I think she's just scratched the surface with our data, right. I know all that data.
Jason: 44:06 You had the unique opportunity to use it to come into our -come into the company and see what it was like to have that much data at your fingertips. Can you tell us a little bit about how, what that was like and how, how that's different from then to school to now that you're, you're in the industry.
Katie: 44:24 So I came into TGS knowing nothing, well, not knowing nothing, but you know, minimal. You think you, every time you start somewhere you like think you know something, but you really don't, which I've learned again third time.
Ted: 44:37 Right?
Katie: 44:37 So at TGS, I wouldn't say it was just, I learned how to work with all this data, which was overwhelming at first. It was like I learned how to, I don't know, act, not just like socially in an office, right? But I also learned like what's important, what's not important. It's easy to get bogged down in the details when you go from zero to 100 real quick.
Caroline: 45:03 So you've really had a unique perspective. Especially compared to a lot of us at TGS, you started off in an internship with us getting into the data and learning the data, applying the data. Right. And then I believe maybe you've even used it in your thesis.
Katie: 45:20 Right.
Caroline: 45:20 And now that you're in the industry, what has that looked like for you?
Ted: 45:26 How about, how about how do you access data being an industry now?
Katie: 45:31 When I've looked at data, it tells me, it makes me feel comfortable. It clears up uncertainties.. It's not telling me what's going on, but at least I'd like have more of a general idea. So when I look at these large amounts of data that I get for a project, let's say like I did in grad school, it's okay, I have this data. What does the data tell me? Does it tell me if it's pinching out? Does it tell me if it's, you know, this big chunk or maybe the depositional environment. That's what I looked at a lot in well logs the petrophysics.
Jason: 46:08 No, it's understandable. You get thrown a lot of data in these situations and it's how you put that together, how you can efficiently use it. And that's something that we're always trying to make easier for people. It helps in a lot of situations, particularly in, in super major type of companies or in a lot of different companies, even smaller companies that they have geo techs that efficiently use our data before they give it to you. Right? So a lot of times you never, you'll never get to see the first part of, you know, where did this data come from because it all just ends up on your desktop. Right?
Katie: 46:42 Right. So like I, what I liked about my experience I guess at TGS is I saw the beginnings, right? What a geotech would put it in. So I like got to see that visual fresh or put my own spin on it when we were using Longbow. So making those bubble plots or looking at URs and decline curves. I don't have, I don't, I haven't gotten that experience yet, but I'm a Guppy.
Caroline: 47:10 So it was like you were getting access to data sets such as the, you know, the EURs and the forecasting database that you probably didn't necessarily have access to while you were working on your masters.
Katie: 47:21 Right. And didn't know about until it came to TGS.
Ted: 47:26 And the ability to build that project from scratch. I imagine a lot of times now in industry, you walk in and sit down and there are gigantic projects already existing and workflows established as opposed to like starting at the beginning.
Katie: 47:46 Right. Which is overwhelming. Like I remember Jason was like, Hey, y'all are going to map from Mississippi, Louisiana and Texas. That was very overwhelming. Now I just, you know, you get a project and it, someone's already, most of the time, I don't know picked through it. So you don't, it's not very fresh.
Jason: 48:09 But now you're not afraid of the deep end of the pool.
Katie: 48:10 I don't know about that...
Jason: 48:10 Right. We threw you right in the deep end and I, you can swim. You're ready to go.
Katie: 48:18 Oh no. I'm still learning.
Jason: 48:18 Well that's good. Never wanna stop learning.
Ted: 48:22 We're all still learning.
Katie: 48:22 Right. But I'm really still learning. As a new worker bee.
Jason: 48:30 So Katie, is there anything we haven't seen you in a little while? I know that you're, you're in Louisiana now. Is there anything that you want to ask us that you're interested in from a, from your perspective after you've graduated and are now moving onto bigger and better things that might help you in the future?
Katie: 48:48 Maybe not something that would- maybe wouldn't help me in the future, but also help other people that are looking for jobs. Is, are y'all looking for employment? Like looking to employ anyone or what does that look like? It sounds like you're doing a lot of work. So do you have people to fill these positions or are you, how does that go for y'all? Do you even know?
Jason: 49:10 Well, that's one of those great HR questions where, you know, we're always, we're always just busy enough to need new people. (Laughter)
Caroline: 49:20 And I think with, you know, new departments that were growing especially new datasets like Ted is talking about Mexico and Canada, I feel like it really helps to position us to grow, you know, as a company as a whole. So opportunities are always always coming up. Yeah.
Jason: 49:36 I know particularly with our internship program, we're always looking for, you know, young, exciting new talent that can, that can come in and help us out. But also like you did learn about data from sort of the bottom up and take that knowledge base to other companies. So we don't only like training people to come and stay with us or we're perfectly happy bringing in interns and having them go out in the world and and learn something from us that they can bring somewhere else.
Katie: 50:06 Oh sorry. I would say that that's why I like had not, I think that working at TGS was nice for others to see cause they knew that I had experience I guess with production data, which is a cool talking point I think.
Caroline: 50:22 And just to build off of that, Ted has done a really great job building this new initiative, which is getting our well performance data in the universities to work with people like you, Katie, while you were getting your masters to make sure that we're able to provide data to other other programs and get geologists or young geologists access to data sets that they wouldn't have or wouldn't be familiar with whenever they're entering the workforce.
Ted: 50:48 That's right. So, you know, we're happy to donate donate our products, donate production data and Longbow to the universities. As you know, at ULL they brought it into the geoscience and engineering groups. And now we're sitting on the, what the 20 workstations in the lab and part of the curriculum. So it's exciting at the same time, giving the students access to these data products learning actual, you know, working product tools. When they do get hired and hit the, hit the workforce, they're ahead of the game and ready to go. Now, from my selfish perspective, it helps to get feedback and make the products better. So it's a win win for both.
Caroline: 51:37 Well, thanks everyone for coming out today and having this conversation, you know, hanging out, covering a lot of really awesome topics, kind of, you know, exploring where TGS is headed next, where we've been, where we're going. Katie, you know, especially thanks to you for coming all the way from New Orleans to sit with us and kind of give us your insight and your opinions and let us know how it's, how the journey has been for you. So thanks, Jason. Thanks Ted looking forward to the next, the next episode.
Katie: 52:01 Thank you for having me.
Jason: 52:03 Yeah, thanks Katie, it's been great
Ted: 52:04 Thank you.