Chapter 1 – Introduction to Waterflood Design

1.01 Intro to Waterflood Design & Workflows (20 min.) Sample Lesson
1.02 Finding Analogy (16 min.)
1.03 Analogue Pools & Controlling Factors (24 min.) Quiz: 1.03 Analogue Pools & Controlling Factors
1.04 Production Profiles (20 min.) Quiz: 1.04 Production Profiles
1.05 Field Example 2 - East Swan Hills (14 min.)
1.06 Bubble Maps (12 min.) Quiz: 1.06 Bubble Maps
1.07 Gas Collapse and Fill-up (19 min.) Quiz: 1.07 Gas Collapse and Fill-up
1.08 The Effect of Depletion on Volumetric Sweep (11 min.) Quiz: 1.08 The Effect of Depletion on Volumetric Sweep
1.09 Building a Forecast on Excel (10 min.)

Chapter 2 - Volumetric Sweep/Displacement Efficiency

2.01 Volumetric Sweep (20 min.)
2.02 Reservoir Controlling Factors Revisited (13 min.)
2.03 Differences of Water Wet & Oil Wet Systems (17 min.)
2.04 Sweep Efficiency (22 min.)

Chapter 3 - Surveillance Techniques for Waterfloods

3.01 Use of Bubble Maps to Estimate Reservoir Flow Mechanics (12 min.)
3.02 Recovery Factor vs HCPVI Plots (23 min.)
3.03 Measured Pressure Scatter (7 min.)
3.04 Forecasting & Optimization (10 min.) Quiz: 3.04 Forecasting & Optimization

Chapter 4 - Decline Analysis for Waterfloods

4.01 Decline Analysis for Waterfloods (29 min.)
4.02 Selection of Decline Correlation (14 min.) Quiz: 4.02 Selection of Decline Correlation
4.03 Decline Analysis Examples (27 min.)
4.04 Recap Of Lessons So Far (16 min.)

Chapter 5 - Waterflood Case Studies

5.01 Decline Analysis Case Study - Wainwright Field (14 min.)
5.02 Waterflood Response & Water Fencing Case Study - Keybob South Field (16 min.) Quiz: 5.02 Waterflood Response & Water Fencing Case Study - Keybob South Field
5.03 Waterflood Response & Water Fencing Case Study - Keybob South Field Continued (14 min.)
5.04 Waterflood Response & Water Fencing Case Study - Chauvin South Field (12 min.)
5.05 Conformance Plots & Throughput Rate (17 min.) Quiz: 5.05 Conformance Plots & Throughput Rate
5.06 Hall Plots (16 min.) Quiz: 5.06 Hall Plots

Chapter 6 - Waterflood & Injection Induced Fractures

6.01 Injection Induced Fractures (16 min.) Quiz: 6.01 Injection Induced Fractures
6.02 Analysis of Injection Pressures and Rates (23 min.)
6.03 Example of Interference Tests in the Spraberry (11 min.)
6.04 Fracture Length and Complexity (8 min.)
6.05 Course Recap (19 min.) Quiz: 6.05 Course Recap

Waterflood Design, Surveillance and Implementation / Chapter 1 – Introduction to Waterflood Design

Lesson 1.01 Intro to Waterflood Design & Workflows

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. Waterflood Design, Surveillance & Implementation - Lesson 1.01: Intro to Waterflood Design & Workflows 02. What is different about this waterflood course?03. Statement of the Problem...04. Course Outline 05. What is a waterflood?06. Line Drives in Late Stages 07. Vertical Displacement vs. Horizontal Displacement 08. What is typical recovery expected from waterflood?09. Median Values of Recovery Efficiency by Recovery Mechanisms 10. Distribution of Recovery Factor 11. Median Values of Recovery Efficiency by Recovery Mechanisms 12. Distribution of Recovery Factor 13. What is typical recovery expected from waterflood?14. Short Intro to Waterflooding 15. Rule of Thumbs: Alberta Government Reserves 16. Rule of Thumb: Secondary/Primary Recovery Factor (SRF/PRF)17. Recommended Workflow First Pass Estimate

01. Waterflood Design, Surveillance & Implementation - Lesson 1.01: Intro to Waterflood Design & Workflows

OK, so let's just start about what's different about this waterflood course that, let's say, a usual waterflood course that you'd be taking at university or just introductory type of waterflood course.

02. What is different about this waterflood course?

Generally there you're going to be focusing on various aspects like displacement efficiency, areal sweep efficiency, and then you'll get into sort of case studies. Now, this course is more geared for an intermediate-type engineer. So you probably had some exposure to the introductory petroleum tech courses. We're going to start by looking at typical recovery factors. We're going to start with case studies. All of throughout the course notes are practical examples. There's almost no just purely simulation-type stuff or purely theoretical. I think it's important to understand and be able to diagnose your observations and move those observations to analysis and theories. And then we're going to do a lot of case studies throughout.

03. Statement of the Problem

So before we get dive in it too much, you have to realize that part of any waterflood study, the fundamental cornerstone is good reservoir engineering analysis. And we're trying to answer some questions and for that, we also need a combined team. So we need to understand what's the original oil in place and we need to understand how it's distributed. Unfortunately though, geological studies just by themselves can't quantify reserves and oil rate increases that you can achieve from optimization. So it's the integration of geology petrophysics (geophysics) with reservoir engineering and production engineering. One thing I would just say is, in general, is try to keep the focus of the large-scale trends first and then go to the small scale. One of my criticisms of waterflood courses is generally they tend to focus at the very small scale, the lab focus, and then move upwards but sometimes they miss the more large-scale trends.

04. Statement of the Problem...

So one of the problems that we have in the flooding literature, this is goes with EOR. I would say this is true with thermal as well as waterflood, is that there's a large amount of published simulation and analytical type studies for waterflood. But the literature is largely silent on reservoir surveillance, sort of monitoring, and improving existing type of waterflood. So this course I originally built for Shell and it's addressing that is what do we look for? How do we optimize existing fields? So often the operating engineer is left with a rate and reserve forecast. It often overestimates the initial reservoir performance because it oversimplifies things. And so one of the things that we're really trying to do is diagnose the problem in these fields.

05. Course Outline

So we're going to start off by talking about what kind of waterflood recoveries we would expect. So we'll look at the recovery factors associated. I'm going to cover controlling factors. Now, I'll warn you, there's a lot of factors initially and it'll be kind of overwhelming but I think you'll see how I deconvolve those, the waterflood response into trying to understand the controlling factors. And then we're going to talk about production profiles. Now, understanding production profile to me is the cornerstone of a good waterflood analysis. If you don't understand production profile, it's very hard to optimize something that you don't understand. So we're going to go into a number of field projects. We're going to start off with a relatively small and simple type project, Conrad Sawtooth. And then we're going to move to a second field example, which is a little bit larger, a little bit more heterogeneous type reservoir; it's East Swan Hills waterflood. We're going to then move to talking about, well, how long if you have existing primary recovery, how long that'll take. So gas collapse and fill-up and response. And we'll talk about the depletion effect on volumetric sweep and a little bit into a pilot design. And then we'll go into the third analysis of production signatures.Czyzewski, 1992. The East Swan Hills Unit Waterflood Optimization Study: A Multi-Disciplinary Approach.

06. What is a waterflood?

So what is a waterflood? Well, a waterflood basically is your trying to maintain reservoir pressure in a field or a reservoir. So you have injectors and here I've shown a fairly highly stylized case, so you have an injector here and then you have a producer here and this is obviously idealized 3D case. And so your injector location often depends on what kind of flood pattern you have and we're going to talk about that next. There's also a structural component where you generally want to inject low in waterfloods, and it also depends on which wells water out and when they water out. The big unknown that we have in many waterfloods is heterogeneity and heterogeneity will always give surprises to you. So there isn't a waterflood that I've ever worked on that didn't have its surprises. And like I've mentioned before, I've worked on hundreds of fields worldwide and I can say that there's always surprises. So what happens is you may start off; injector millet locations may start off about very idealized type of performance and then move to more performance driven field development plans. Now, the basic foundation is we're trying to inject water and we're trying to push water towards the producers, maintain pressures and maintain oil saturation at the producers.

07. Pattern Flood

This is just some of the types of pattern floods you have. And again, I really want to stress that when you really look at late-stage mature type fields, that you don't necessarily have these perfect type of patterns. One of the first type that's almost never used is a 4-spot pattern. And you can see there you have 3 injectors which are shown in the triangle and a central well is the producer. And then one of the most common is the 5-spot patterns. And then it goes 7-spots here, 9-spots. And so you move from a 1:1 ratio in a 5-spot pattern to a 2:1 ratio in a 7-spot pattern. And in a 9-spot pattern, you deal with 3:1. And then you can have various types of line drives. So you can see here this is what they call a direct-line drive. And so the injectors are in those locations and they drive producer in a line-drive fashion. There's also a staggered-line drive, which is the wells are not exactly in the same sink, there's an offset to them. And again, you don't necessarily follow all these patterns throughout the life. You tend to take advantage of, once you understand heterogeneity, once you understand the breakthrough, then you get a better, more realistic patterns that you're going to try to use in late stage.

07. Pattern Flood

It's very common in line drives and in late stage to see either fracture communication or communication through high permeability heterogeneity. So probably overwhelming reason why most waterfloods fail is because you have early breakthrough of fluids. And what I'm teaching you here is how to optimize particular fields and account for those surprises that you will undoubtedly have.

09. Vertical Displacement vs. Horizontal Displacement

OK, and then just another definition here is to understand the difference between a horizontal flood, OK, and a vertical flood. Now, we're not going to talk about vertical floods until a bit later, but horizontal floods we tend to use in thin formations or it's usually the default. And so there you're trying to aerially push the water towards the producers. It's influenced by producer-injector displacement efficiency and it's also strongly affected by heterogeneity. And recovery factors for this type of horizontal displacement; our areal sweeps vary from about 15 - 70%. And I would say on average, worldwide is around probably 30 - 40%.
Now, when we deal with vertical floods, it's a little bit different. There what we're trying to do is take advantage of gravity. So this is a cross-sectional view of an injector and a producer. So we tend to inject low in structure and then we're trying to push water up towards the producer. And what we use is gravity as a stabilizing influence. And generally, you'll see very high recovery factors in most gravity stable systems, especially if you're at relatively low rates. And there's lots of fields worldwide that have 50 - 80% recovery factors in them. So, know the difference between types of patterns; 5-spots, 9-spots, 7-spots. And also know the difference between a vertical flood and a horizontal flood.

10. What is typical recovery expected from waterflood?

So what kind of typical recovery factors do we get in conventional reservoirs?

11. Median Values of Recovery Efficiency by Recovery Mechanisms

OK, here is some of the data from probably the most successful old North American type fields. And what we can see is that in general, waterfloods have a kind of a medium recovery factor of about here, about 43 to about 51% in both sandstones and limestones, OK. Where if you compare that to solution gas drive, so just primary depletion sort of think of it like fizzy pop type stuff is you have a recovery factor in sandstones of about 28% and in limestones of about 22% efficiency. So there's a substantial increase in recovery factor fromsolution gas drive to waterflooding. And you can see here we're going to use a rule of thumb that your secondary to primary recovery factor is about 1:1, and that usually means you're doubling recovery factors. Wherever you produce on primary, usually it's about double what you produce with waterflood. Or sorry I should say that the other way: primary production doubled is equal to your waterflood recovery.

12. Distribution of Recovery Factor

And here it just shows you the same thing. It also mentions where I've got the information; it's an old 1984 study. So again, these are fairly old fields and they tend to be larger, bigger, whatever. But you can see here we have on this graph, you can see here we plotted from 1% - 100% recovery factor here on a semi-log scale and then this is just probability scale on the bottom. And you can see here solution gas drives average about 20% recovery factor associated with them. And then if you add some sort of injection schemes or usually gas caps or whatever, then it may go up, pushing up to 30% at the 50% level. But you can see here big changes when you move to the waterflood scale. So at the 50% case level, you're basically producing about 30 - 40% of your recovery factor associated with it.Statistical Analysis of Crude Oil Recovery and Recovery Efficiency.

13. What is typical recovery expected from waterflood?

Now, I apologize for these, but I thought I'd include them for a completeness sake. Again, some of these correlations are quite old, but it gives you something to sort of get your mind wrapped around. And so in this correlation, it's obviously a sort of expanded polynomial type thing. It's just telling you the recovery factor is a function of porosity, water saturation, formation volume factor and permeability as well as water saturation and depletion. So you can use this correlation. Again, it's going to tend to give you high numbers in terms of recovery factor. And then there's a secondary correlation, Guthrie & Greenberger, that's based on 73 fields. These, I should stress are based on the biggest, most permeable fields. They're the ones that were developed first. And so they'll give you a number but I would tend to discount that number somewhat.Arps, 1968. Reasons for Differences in Recovery Efficiency.

14. Short Intro to Waterflooding

So if we want to do, sort of try to understand how much recovery factor that we can improve by using waterflood, one of the things we're going to use is what they call an S/P ratio. And what we mean PRF here is just primary recovery factor; SRF is secondary recovery. And so we typically, if you don't know anything about your waterflood or your primary production, a good S/P ratio is 1:1, that's kind of a standard type of default. When you get into medium-grade crude oils, let's say 15 - 40 cp type range in terms of oil viscosity, then you start getting into S/P ratios of approximately 2. And the reason that the S/P ratio increases in those type of reservoirs is that you have lower drive energy in those type of reservoirs and so what happens is adding a little bit of water drive substantially increases your recovery. In heavier crudes, and I would say over, let's say, 400 cp, your increases recovery factor, S/P ratio goes to about 1. And now what starts to happen is you tend to have viscous fingering from injectors to producers and you tend to have limited coverage in poor volumetric sweep. When you get into tighter crudes, so let's say reservoirs that are less than 10 md or 1 md, then your S/P ratio drops fairly dramatically. And so there you'll be dealing with less than an S/P ratio of 1, usually around 0.7 is a good estimate for anything, let's say above a 0.5 md type reservoir.

15. Rule of Thumbs: Alberta Government Reserves

This just shows you an old; this is in the ERCB. It's from Alberta government, a Canadian source of information about reserves. And I just wanted to kind of show you this table because it reflects the sort of government type standard numbers. And you can see here that in waterfloods here, this second row item, they have about a 15% recovery factor on primary and about 14% onwaterflood or gas flood type things. And that's, again, that sort of confirms your S/P ratio. When you get into waterfloods in heavier oil here, you notice how now you have about a 12% recovery factor here and only about 19% recovery factor. And that's because your boost here with a waterflood in that heavier oil or a moderate crude, you're increasing the recovery factor greater than you normally would because there's usually a little less solution gas drive in those type of reservoirs.

16. Rule of Thumb: Secondary/Primary Recovery Factor (SRF/PRF)

So some rules of thumb, again, just to review. Your S/P ratio is about 1:1 in a conventionallight oil type reservoir. When you move to a medium crude type reservoir, it's 15 - 400 cp in that range, it moves to 2. And then if you go further in terms of oil viscosity, you go back to about a 1:1 ratio in terms of S/P ratio. And then in lower perm rocks, you tend to get an S/P ratio that is lower than 1, and that's generally because your injectivity limited, productivity limited. And what happens in the lower perm reservoirs is you start to get short circuiting through fracture systems and therefore it usually hurts your waterflood performance. And so your S/P ranges from about 0.4 - 0.7.Alberta’s Reserves 2003 and Supply/Demand Outlook 2004-2013

17. Recommended Workflow First Pass Estimate

Now, the recommended workflow that I would use it generally is first of all, you're trying to get an idea of what kind of recovery factor we have. So we've talked a little bit about just getting a first pass estimate of recovery factor.Waterflood is strongly associated with analogy. So one of the things we're going to do is we know we don't totally understandheterogeneity at a first-pass basis. So what we're going to try to do is find analogous fields and use them as a bit of a guide to understand our production profile. So one of the things we're going to talk about a little later is finding an analogy. And what we're trying to do is understand what we're going to have as a production profile and how we're going to react. So later on, we're going to talk about, well once we get seed type of performance, how we're going to react to that. So we can use those API correlations and, like I said, they tend to overestimate recovery factor. We're going to look at our fields and try to understand that what the analogy has in terms of recovery factor. And then we're all trying to generate a production profile because that fits in with our economic analysis.
So this pretty well sums up the overall workflow that we're going to use, at least initially. And the next time we're going to talk about production profiles and trying to understand this a little bit better.

20. Selected References

Kikani, J., 2013. Reservoir Surveillance. Rose, 1989. The Design Engineering Aspects of Waterflooding, Vol. 11. Richardson, Texas: Monograph Series, SPE.Warner, 2015. The Reservoir Engineering Aspects of Waterflooding, Vol. 3. Richardson, Texas: Monograph Series, SPE.Willhite, 1986. Waterflooding, Vol. 3. Richardson, Texas: Textbook Series, SPE.