Chapter 1 - Overview of Quantitative Interpretation

1.01 Introduction (25 min.) Sample Lesson Quiz: 1.01 Introduction
1.02 Rock Physics (40 min.) Quiz: 1.02 Rock Physics
1.03 Well Data (48 min.)
1.04 Case Studies (36 min.)

Chapter 2 - Seismic Data

2.01 Seismic Data (19 min.)
2.02 Seismic Data - Well Ties (18 min.)
2.03 Seismic Data Condition and Conditioning (36 min.)
2.04 Case Study - Can Interpolation Improve QI? (27 min.)
2.05 AVO (25 min.)
2.06 Velocity Modeling (23 min.) Quiz: 2.06 Velocity Modeling
2.07 Sonic Log (37 min.) Quiz: 2.07 Sonic Log
2.07a AVO Tutorial (29 min.)
2.08 Inversion (31 min.) Quiz: 2.08 Inversion
2.09 Pre-stack Inversion (19 min.)
2.10 Case Study (17 min.)
2.11a Stochastic Inversion (9 min.)
2.11b Inversion Tutorial (16 min.)

Chapter 3 - Classification Using QI

3.01 Multi-Attribute Analysis (23 min.)
3.02 Classification Intro (27 min.)
3.02a Classification (32 min.) Quiz: 3.02a Classification
3.02b Classification Example (1:12 hours)

Chapter 4 - Advanced Topics

4.01 4D Reservoir Monitoring (9 min.)
4.02 Multi-Component Data (43 min.)
4.03 QI Summary (21 min.)

Practical Quantitative Interpretation (PQI) / Chapter 1 - Overview of Quantitative Interpretation

Lesson 1.01 Introduction

Access All SAGA Wisdom Training content Subscribe

Already a member? Sign in

Access All SAGA Wisdom Training content Subscribe

Already a member? Sign in

Transcript

01. Lesson 1.01: Introduction 02. Who is Laurie Weston?03. Practical QI Outline 04. Introduction 05. Quantitative Interpretation 06. Geophysics Jargon 07. Big Picture QI 08. Why QI?09. Geology is Complex 10. Size Doesn’t Matter 11. The Effect of Time: Remote Monitoring 12. The QI Transformation 13. Useful Information for Well Planning 14. QI Workflow 15. QI as a Race Car 16. Stay Humble 17. A Note About Software

01. Lesson 1.01: Introduction

All right, well this is the Practical QI course, a comprehensive overview for interpreters. And I'm Laurie Weston.

02. Who is Laurie Weston?

So who am I? Well, I have a degree in Physics and Astronomy. And during that degree I was on the co-op program where I had multiple work terms at a radio observatory because I was interested in astronomy at Dome Petroleum, which doesn't exist anymore, but was a pretty exciting oil company here in Calgary and Chevron and Shell. So at the end of my degree I had the opportunity to choose between a couple of offers and I chose Shell Canada. The main reason was that I was gonna get to go overseas. I wanted adventure, I wanted some interesting things to do. So some of these pictures here you'll see are from my adventures overseas. One of the first things I did when I went to Shell, Canada was go on a training course in Holland for four months and got to experience some of the exciting Shell world. However, I went back to Calgary, started doing processing and was looking for more excitement than what I was getting. So I moved to a British company called LASMO and that's where I traveled and you can see in some of these pictures. So at the top here, this is some ruins in the Syrian desert. These were Roman ruins. We were driving across the desert to go to a seismic shoot out on the Euphrates Groban in between the Euphrates and the Tigris rivers in ancient Mesopotamia. So this was pretty exciting for me. And this was the view from my office in London, England. That was St Paul's Cathedral. This was the view from my office on another day, the I.R.A. bombs went off in London and blew out the glass in several of the windows. So this was exciting in a different way. And then on another trip we went to Morocco and this was in North Africa. In Morocco we were looking for surface oil seeps. So this was looking at just a pipe that the local farmer had stuck in the ground and oil was seeping into it. So this is all very enlightening for me and super exciting. At that seismic shoot, I had an entourage of little kids. All these little Syrian boys following me everywhere. They were thrilled to get their picture taken. And I was told to dress conservatively. You'll see that I'm in a skirt here. I wore the skirt to the well site, which really kind of amused all the drilling hands who were all from Canada, by the way. And I've always thought that I should dress appropriately so that's what this picture represents. And this was when I went up to Fort McMurray, to the oil sands. So in Wascana Energy and Nexen and Oil Sands Imaging, these are all related to the oil sands. And that's where I started to realize that seismic data had a lot more to offer than I was already learning. And so the QI process began. That was back in 2007, I started my company called Oil Sands Imaging. In 2012 we realized that this process was, the QI process I'm going to describe today, was actually relevant to all kinds of plays, not just oil sands. So we changed the name of the company to Sound QI. Within a month we were taken over by Canadian Discovery and that happened in 2012 and I became the director of geophysics at Canadian Discovery. And then a nasty downturn hit and things all changed. By January 2017, our team was disengaged, I guess you could say from Canadian Discovery and Sound QI was formed. Sound QI Solutions or Sound QI 2.0, as I like to call it. So in 2017, right in the middle of a bad downturn, we formed a company and now three years later, we've grown and we're still doing what we love to do. So I'm going to describe in this course some of the things that contribute to the quantitative interpretation process, so that's QI.

03. Practical QI Outline

This is the outline for the Practical QI course. We have multiple chapters, an introduction, rock physics, well data, data, classification, advanced topics and then a summary pulling it all together. So in each of these chapters, there's going to be multiple lessons that dive a little deeper into each of the subjects. However, we don't go that deep. Each of these and you'll see as I start talking about them, each of these is actually a full course in itself. And what you get here is a practical overview of all the elements that go into QI. And as an interpreter, this course is geared for interpreters. As an interpreter, you need to pull it all together. You need to have a little bit of knowledge. You have to know a little bit about the pitfalls in each of these categories so that you can bring it together into an environment where you can make decisions and the team has confidence in those decisions. So that's what this course is for it. This course is for you as an interpreter to help contribute valuable information to the decisions the team needs to make about oil and gas exploration and development.

04. Introduction

Starting with an introduction. So first of all, what is QI? And then why is it useful? Why should we learn about QI and why should we bring it to our teams?

05. Quantitative Interpretation

I like to look at this as a number of ingredients that make us a delicious cake. It's a geology cake, it's delicious. And all these ingredients are available to everybody but the quality of the ingredients is important. The order in which you add them is important. How much you mix them, the kind of tools you use, everything is important. And that's what QI is all about. QI is about taking ingredients that are actually available to everybody. But the way you put them together, your biases, your interpretation, your assessment of the results is what makes a difference. So that's what this course is all about, is how to make a delicious geology cake that everybody will want to sample.

06. Geophysics Jargon

So one of the things that I want to start with right off the bat is the jargon that everybody uses. If you look at any discipline, we've all got our jargon. And unfortunately, it's only those people who are familiar with these things that actually understand them, that can relate to them and if you use these kind of words in mixed company, you're gonna lose people. They're going to stop understanding what you're saying and you're going to lose your audience. So these words are all definitely part of geophysics for sure but we need to try to relay these concepts in more of a plain language type of method. So that's another aspect to the course is learning not only what the words mean, but also how to convey these things intuitively so that people understand what you're talking about. If you use these words, you're gonna confuse people. So let's try to stay away from the jargon, even words like relative and trace and trim and training. All of these things actually have distinct meanings within the geophysical environment and people might project their own understanding of what these things mean. So it's best to avoid them or explain them if you have to use them and you can't get around it.

07. Big Picture QI

So the big picture QI. Let's just look at a giant 10,000 ft overview of what encompasses QI. And so this is really what this big picture is. Here's attributes. So that's one thing we're going to talk about in great detail about seismic attributes.
Well data analysis and rock physics. This is really how we incorporate and integrate well information into our interpretation of seismic data.
And classification. Classification is how we pull it altogether. And this is where we get the rest of the team. This is where we integrate all the disciplines and actually, it's the fun part. And it's where we start to relay that data information into actual concepts for decision making.

08. Why QI?

So why do we do QI in the first place? Well, the bottom line is QI makes money. To drill and complete one horizontal well in the I think it's, I've heard it's around $10 million. So to acquire and process a 3D program, about $150,000 per square mile, depending on where you are, depending how deep your objective is but in the Montney, it's about $150,000 a square mile. And the Montney is a shale gasin Canada. To drill and core one vertical well in the oil sands and again this is in Canada, is around $500,000. To drill 8 * 16 vertical wells per square mile. And these vertical wells are drilled purely for evaluation purposes. They're not drilled for production. So they're just drilled to figure out what kind of reservoir we're dealing with. So that could be $4M - $8M per square mile. And one horizontal well pair for this kind of play, this kind of production is at least $2M depending on the length and whether it's drilled from a pad or not. So to shoot and process a 3D seismic program in this environment is $1M per square mile. So already the cost is minimal compared to the actual development of these kind of plays.
But what you can get from this adds immeasurable value because seismic data is an encrypted photograph of the subsurface.
And imagine what it would be like if a surgeon was about to start an operation without an X-ray, without knowing what they were cutting into. Mistakes in both cases, in the oil industry and in the medical industry are costly.
So QI increases the odds of drilling more good wells, fewer bad wells.
QI can also have an impact on booked. As recently as 2010, this was considered a reliable technology under certain conditions. We have to prove certain things, but it is allowed under FCC regulations as a reliable technology for the booking of reserves, oil and gas reserves.

09. Geology is Complex

Now, geology is complex. We all know this. It seems though, that sometimes when we're looking at the surface and we have a geological concept in mind and we're about to drill, we tend to simplify and we all do this. We think OK, we've drilled a couple of wells here we're smart people, we know what's going on, we have all kinds of analogs so we tend to oversimplify until we know exactly what's going on.
Well, this is an outcrop. This is an outcrop in the McMurray. So the McMurray is the oil sands reservoir in northern Canada. And you can see we've got unconformities. We've got down lapping beds. We've got various shale lenses. We've got all kinds of multiple channel cuts on this cross-section. Multiple and unconformities, some at the top, some at the base, some in the middle.
And yet we think that we can drill a couple of wells. And these wells are even very close together. They're 400 m apart, which is pretty close together for evaluation wells. And what we find is that we encounter some geology in this well, we encounter some geology in that well, and we simply extrapolate between wells. But when we look at the outcrop and when we look at the data, we see that we have all of those features in the seismic data and we just don't know exactly what they mean. And if we are drilling this kind of spacing of vertical wells, even at that spacing, it's very difficult to extrapolate the correct geology. And when we're looking at seismic data in this type of format, which is the conventional wiggle trace format, we actually can't really see differences between geological layers. We can't see fluid properties, we can't see fractures necessarily. We can maybe see some large faults, but we can't really see anisotopy or orientation of different arrangements of minerals, channels,, all of that kind of thing. So QI is a way of understanding this kind of geology out of this kind of data.

10. Size Doesn’t Matter

And I have to say, size does not matter. This could be a gigantic faulted mine face or it could be fractures on a very small scale. And so geology is similarly complicated whether we're looking at large scale or small scale. And data is actually a way of trying to relate those scales to each other, relate thoseto the well data and see more information between wells.

11. The Effect of Time: Remote Monitoring

We can also use QI and information for monitoring. So this is an example. This was published in 2010. It's a North Sea field and it's a time lapse survey of I believe this is an injection. So we're looking at fluid effects over a period of time. The first one in 2001 and then subsequent surveys, subsequent seismic surveys at a couple of year intervals. Looks like every two years they did another seismic survey. So then they were able to see the effect of the geological processes or the production and injection processes.

12. The QI Transformation

This is what I call the QI transformation. So we looked at data in relation to an outcrop. And that seismic data showed us a lot of complex features and when we relate it to the outcrop, we have an idea of what those things might be, but we really don't know until we've done a deep dive in the data.
So for example, let's look at something that's close to Fort McMurray in Canada so that would be an oil sandsand say we've drilled some wells here. We've shot some seismic and this is what our seismic data looks like. And at the well location, we've got some well logs. So we can see from this gamma-ray log here that we've got a shale layer. This is a shale layer and it's consistent in each well. We've also got a consistently thick sand layer in each well. You can see that there as well. However, in each of these wells, a water leg was encountered at the base. Only it wasn't a consistently thick water leg. In different wells it had a different thickness. And this can happen in oil sands, oil-water contacts or not flat because the oil is immovable so any structural changes actually change the whole orientation of the fluid contacts. And in this particular play, there is actually water at the top of each of these wells too. There's a transition zone from water to oil. So how are we going to find that out? How are we going to predict those subtle features? Those contacts by looking at this kind of seismic data? I don't know if you're familiar with seismic data, but if you are, even if you're an expert in seismic data, you're gonna have real trouble figuring out how to map those contacts.
Since we don't have any real way of accurately and not only accurately but convincingly telling our team where these water contacts are, what we do is QI. And QI is the process of transforming that seismic data. That's why I call this the transformation. It seems like magic. We've transformed our seismic data into geological representations of shale, oil and water. And we've still got the seismic wiggle traces overlain here.
But let's just take those wiegel traces away and also put the representations from core in at each well and that shows us now a geological cross-section. We've actually created this from the seismic data and we validated it at the well information. So we can look at pretty nice matches of where we are seeing those water transitions at the top and the base of each well.

13. Useful Information for Well Planning

Once we have something like this, imagine how useful it is. So this is taking a small portion of one of those volumes that's been transformed into the, water and gas in an oil sands play. And we've put that small portion, that small subcube of seismic data over one of the planned wells. So this well plan here is 6 well pairs that are being drilled from a central pad. And this being a operation, the bottom well is the oil producer and the top well is the steam injector. And originally, without using seismic these wells are planned based on the results of vertical wells. So we've got a well at the heel here and we've got a well at the toe. And the geology that was encountered in those wells would have been interpolated between the wells. And then that's how the horizontal wells would have been situated is based on that interpolation between the geology at those wells. Now look at the detailed information we have about shale beds. Whether or not they're continuous. If we should avoid the thicker ones. We've got a gas zone at the top here. We've got some shale and actually some water at the bottom and we see that it's a variable contact there. So that's all going to inform better well planning, better horizontal well planning for more optimum production.

14. QI Workflow

This is another big picture of a QI workflow with a little more detail. So I'm easing you into this gently. So we started out with our big picture, the three main components, the well data, the rock physics, the attributes and the classification.
Well obviously, in each of those major components, there's all kinds of different processes and assumptions and computations that need to go on. So in the seismic attributes, we start with our seismic volume. And this is going to be a pre-stack seismic volume. We're going to look at post-stack as well, I'll describe those terms in a minute. I don't want to use jargon, even this early on to this audience. But we're going to talk about what all these terms mean.
And here's some geophysical computations we need to do on that seismic information. And then we create these various attributes. And these attributes represent geological properties that we may not have discovered yet, but we will. We'll figure out what they all mean. And the way we figure out what they mean is by looking at log data, looking at core data and any other information that we can bring into this process that's going to give us some benchmarks and some clues as to what our seismic attributes mean. We look at that in crossplot and we look at relationships between different properties. We call these our attributes, seismic attributes.
We look at the relationships between those attributes and all of this ground truthing information we can bring in. Now, I use the word truth with a grain of salt because none of these things, with the exception of maybe core, but none of these things are really error free. They all have uncertainty. They all have biases. And that's another thing we want to look in a lot of detail at, is how we judge the integrity of all the data information we're bringing in.
And finally, when we get all this all this background information and we know what we're looking at, then we pull it all together in our classification process and we make our geology layer cake.

15. QI as a Race Car

Another analogy here, not the cake analogy, this is a race car analogy. So race cars have a lot of high tech. They've got all of these technological things, the suspension, the brakes, the fuel, the engine, the tires, the body, the transmission. All of that is under the hood.
Even deeper than that is all the theory that goes into creating the technology. So we need all these first principles. And this is all the research. This is the tools that are being used from the research to power that car. But as a driver of the car, we don't necessarily need to be experts in all of these things. But what we need are quick reactions. We need spatial awareness. We need problem solving, multi-tasking, visualization. All of those things are what the driver needs. And that's what we are going to talk about mainly from the point of view of the driver of a QI operation. So you've got in QI, we've got all kinds of sophisticated technology and research backing it up. But as a driver, we need to know how to use it all and how to bring it together into an effective race. That's us.
OK, if this happens and it sometimes does, whoseis it? 9/10 it's the driver's fault. So the driver has a key role here. You can have all the best technology, all the best theory. But if you don't know how to drive the car, you're gonna get messed up results. That's the risk. And that's the risk we want to reduce.
Also, we need to make money. We need a profitable business model, so we need to get around that track fast. We need to be the fastest to get around the track. And we use all this technology effectively.

16. Stay Humble

At the same time, we need to stay humble. We're trying to predict things that we can't see for the most part. So we often have sparse and infrequent observations. We have errors in everything. We have incorrect interpretations of observations. We have theoretical misunderstanding and so on and so on and so on. Until ultimately we have poor results and then we have to cover them up. My personal favorite is that management directives out here by itself, which we mostly ignore. Anyway, this I saw in a technical publication 30 years ago and it's been on my wall ever since as a reminder that what we're dealing with is something that we can't ever totally understand.

17. A Note About Software

So in this course, I just want to talk about the software that I've used. All the well data crossplots that I'm going to show in this course were created using Paradigm's Geolog software, unless I've indicated otherwise. Any synthetic, all the seismic attributes shown were created using Hampson Russell CGG. And the rock physics plots that I'm going to show were created using Matlab and custom code. So that's Sound QI code, unless I've referenced them to a publication of some kind. And the seismic data across plotting that I'm showing in this course was all done in our Sound QI pro software.