Chapter 1 - Well Site Gas For Reservoir Evaluation

01-01 - Introduction to Well Site Data and its Applications (21 min.) Sample Lesson
01-02 - Mercer (1974) Gas Classification (32 min.) Quiz: 01-02 - Mercer (1974) Gas Classification
01-03 - Using (normalized) Well Site Gas Data to Spot Pay (29 min.) Quiz: 01-03 - Using (normalized) Well Site Gas Data to Spot Pay
01-04 - Wellsite Gas Composition Vs Reservoir Compositions (35 min.) Quiz: 01-04 - Wellsite Gas Composition Vs Reservoir Compositions
01-05 - Re-Cap, Some Things To Think About And Some Comments On Set-Up (39 min.) Quiz: 01-05 - Re-Cap, Some Things To Think About And Some Comments On Set-Up
01-06 - Summary (11 min.)
01-07 - Selected Real World Examples - OBM Isabella UK North Sea HPHT (18 min.)
01-08 - Selected Real World Examples - WBM Rosebank UK West Of Shetlands (11 min.)
01-09 - Selected Real World Examples - WBM ROC-1 Bedout Sub-basin, Australia Oil Or Gas? (12 min.)
01-10 - Selected Real World Examples - Montney Play Canada WCSB (16 min.)

Well Site Gas For Reservoir Evaluation / Chapter 1 - Well Site Gas For Reservoir Evaluation

Lesson 01-01 - Introduction to Well Site Data and its Applications

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 to Well Site Data and its Applications 02. Learning Objectives 03. Training Chapter Contents 04. Lesson 1: Introduction to Well Site Gases 05. What are well site gases?06. What are well site gases? (2)07. How the data are normally reported 08. How the data are normally reported (2)09. How the data are gathered 10. How the data are gathered (2)11. Gas measurement 12. Gas measurement (2)13. Gas measurement (3)14. Real World Example 15. What are wellsite gases related to and how do they need correction 16. What are wellsite gases related to and how do they need correction (2)17. Data Input 18. What Causes Mudlogging Mud Gas Response to Vary?

01. Lesson 1.01: Introduction to Well Site Data and its Applications

Hi, my name is Julian Moore. I'm a geoscientist and geochemist with APT. Today I'll be giving a course on wellsite gas for reservoir evaluation. This is a great method that's been around for a long time, and there's been some great papers published on the methodologies you can employ. But it's also perhaps underutilized in the industry as a generalized statement. It may be underutilized because, as a technique, it falls between several different disciplines such as operations geology,,... and so no one domain or discipline owns this method. That said, because of that, it becomes a great method to integrate across different data types. So today I will be giving a course on wellsite gas for reservoir evaluation.

02. Learning Objectives

So the learning objectives in the course are: Mainly to provide an awareness of the data. So how the data is acquired, how it's measured, how it's reported by service companies; We also very briefly describe how the gases originate in the subsurface; We'll introduce the Mercer Gas Classification system, which was published way back in 1974, but it's still a really, really useful way of thinking about the anomalies we see in wellsite gas data; We'll describe non-formation gas events. So these are events that don't relate to formation gases but are important to understand if you're going to use the data for formation evaluation because they forms; We'll describe a methodology of how you can use the gas data to spot pay zones in a well; And then we'll also discuss how the gas compositions vary and how we can relate those gas compositions to PVT; Lastly, we'll run through some scenarios where the data can go wrong. So this will include looking at reservoir vs. non-reservoir zones, looking at the impact of coring, and thinking about what the phenomena can do to well site gases; And then right at the end, we'll run through some real-world case studies to describe some of the methods and techniques that have been described in this course. These examples will include an HPHT well from the UK North Sea, a example again from the west of Shetland in the UK. Then we look at Roc-1 which is a well in Australia, where we look at phase prediction.

03. Training Chapter Contents

So the course is composed of 4 different lessons: The first will be an introductory lesson looking at the acquisition of the data and the measurement of the data and what the data relates to; In the second lesson, as I mentioned, we'll run through the Mercer gas classification system; In Lesson 3, we'll look at how the data isalized and how we can use that to identify pay zones. In the next lesson we'll look at gas compositions and how we can relate the measured gas composition to the compositions in the reservoir and those compositions that we're particularly interested in; In the final lesson we'll do a bit of a recap on all the things that we've run through, and then we'll add in some additional things that you might want to think about if you're working with wellsite gas; And then finally we'll run through the real-world examples, as I've already described.

04. Lesson 1: Introduction to Well Site Gases

So an introduction to wellsite gas.

05. What are well site gases?

So formation evaluation is interested in using gases that are released when we drill a volume of rock. There are other ways to add gas to the system, non-formation gas events, and we will need to understand those. These occur when you're drilling in an underbalanced condition, and as I've already mentioned, these will forms. So we need to understand those and remove those from any evaluation. They can be in their own right very useful as diagnostics for understanding the pressure state and understanding the drilling window with respect togradients. That's not what we're going to talk about in this course today. That's a course in itself. And the other point to make is wellsite gas data is really ubiquitous. So it's collected in virtually every well because it's safety critical. Certainly from the 70s onwards it should be available. In many legacy wells it will probably be available as a paper log, usually in a TIFF or format that may be amenable to digitization, and in the last 20 - 30 years the data itself should be available digitally so we can do alot more with it.

06. What are well site gases? (2)

When we drill a formation, the gas that we measure will occur from the saturation in the formation. But we can also get anomalous gas levels due to a number of other processes. One would be where we have a washout, and that adds additional volume to the wellbore and increase the gas that's being measured. The second event is one that we've already sort of touched on. It's where you have an influx due to the wellbore bottomhole pressure being less than the formation pressure, and we can get some gains (some increases), and that can add gas to the measured gas volume at the. And lastly, where we have a very narrow drilling window, we may actually be fracturing the formation and that will lead to some losses. There can be then some. And these flowback events, which will beally dominated, can add some volume to the gas that's being measured. So you have to be aware of all of these mechanisms when we come to integrate and interpret our gas data to interpret the formations that have been drilled.

07. How the data are normally reported

So how is the gas dataally reported? Well, it usually comes in a fairly basic format. So you probably are aware or familiar with a mud log or a comp log. And there's an example here on the screen which is a very standard mud log. It's a sort of basic service. And we will pull out some basics, so (perhaps a gamma ray), some drilling parameters such as the ROP, the total gas in the compositional data (and we'll describe those in more detail in a few slides), a track and some well site lithology descriptions. So it's very basic and there's virtually no interpretation of what that data means on the mud log itself. There are more advanced services that you can pay for. These are usually more advanced because of either the design of the gas trap and/or the detectors that you're using. And these may be a little bit more advanced in that they have some interpretation, particularly with respect to phase, but they're usually just these sort of blocked flags and there's very little discussion of the data and how you might interpret it and any issues with the data that you may need to be aware of.

08. How the data are normally reported (2)

So currently in industry, the gas data itself is rarely discussed in alot of detail, it's reported in quite a basic and simple way, and the data is not really worked very hard. And many people who have sat in oil companies will have large legacy datasets that you can do useful things with.

09. How the data are gathered

So we're just now going to run through some basics around how the data are gathered. As we drill the well, the mud returns are sent to the. This may be an onshore rig, it may be an offshore rig... the setup will be very similar. It's described here in our cartoon: So on the left-hand side we've got the formation that's being drilled; We have the bit action which we'll describe in more detail later on; The BHA; The annulus; The BOP; The mud will return to the topsides, it will descend down a and it will go into a mud ditch with a gas trap. The gas-air mixture will be collected and taken on a gas line to a series of detectors, and we'll run through these in more detail in the next few slides.
So things that are important to bear in mind: Surface losses... and frankly, the surface losses are unknowable; The flow line length, slope, and position all can impact the gases that we measure; The gas trap type, which we'll describe in more detail shortly; The nature and the length of the mud lines; And the detectors. So they're all things that will materially influence the amount of gas that we measure. In the materials that are included with this course, we've got a Word document, and that just describes the basis on which these different detector types work. So there's a resource there that's been provided.

10. How the data are gathered (2)

So if we look at the gas trap itself in a little more detail, there's a cartoon that describes the generalized setup of these bits of equipment. The gas trap itself can be anything from a floating gas trap in the possum belly, through to more sophisticated constant volume traps which are being used more and more in the industry. The traps themselves may be heated or unheated. They all have a fixed flow rate where we're mixing a fixed volume of air with the mud that's flowing through the trap. There are 2 types of detectors that will be used: There's the total gas analyzer which is run in a continuous mode. This usually is an FID, but sometimes it's a detector (but more rarely). And then there'll be a compositional measurement. Almost always it's a FID these days, and that will be run on a cycle time... perhaps 60s. We may use GC-MS also, but that's usually only available in the more advanced gas systems which we'll have to pay more money for. These are the most common ways we set up these systems, but there are others. There's an infinite ways of doing this gas analysis.

11. Gas measurement

So just to go through the setup in a little more detail. As we already mentioned, there's usually a continuous measurement. This may be referred to as the total gas analyzer or the total hydrocarbon analyzer. And this is really measuring the total amount of combustible gas. It's usually measured in a methane equivalent or a BTU (British Thermal Unit) measurement. And it gives us no compositional data, so it doesn't give us any information about methane or ethane or the actual composition. It's just a number. It may be reported as percent, as ppm, or as units.

12. Gas measurement (2)

After the total gas measurement, there's usually a compositional measurement (as I mentioned). This is nearly always a flame ionization detector. So we'll separate the gas in a column, and it may be separated due to the granular packing within the column or simply through interaction with the column wall itself. And what we're doing is we're separating the hydrocarbon compounds on the basis of their vapor pressure or essentially their boiling point. More volatile compounds will elute first, less volatile compounds will elute at a later time. Methane is the most volatile of the gases that we measure, and that will elute first. Subsequently the gases will be separated on what we call the Kovats index. We'll measure methane first, and then ethane, then propane, then the butanes and then the pentanes.

13. Gas measurement (3)

The flame ionization detector is combusting these hydrocarbon compounds in a flame of hydrogen, and that creates carbon ions. And then very sensitive electrical detectors can measure the current fluctuations. And those current fluctuations are proportional to the number of carbons in the molecule that's been combusted. So we can invert that data from a measurement to a composition.

14. Real World Example

So here's a real-world example. This is a setup on a deep water rig that was drilling a well in deep water in offshore East Africa. So we have the schematic here with the various components of the gas system. In this system we have a degasser to ensure that the mud is clean. So when the mud returns to the annuli to be pumped back down, it's free of any gas that was entrained while you were drilling the previous well. This improves the reliability and the repeatability of the measurements that we're making. The mud heater is a vital bit of equipment often in deep water wells where you have long risers and the water column can be quite cold, so it reduces the temperature of the mud and that restricts or reduces the ability of the mud to release heavier components in the gas trap. So we can apply heating to the mud, and that increases the ability of the mud to release those heavier hydrocarbons so they're available for detection in the gas trap itself. As I described earlier, there will be a gas line that will link the gas trap to the detector equipment. These may or may not be insulated, and they may even be heated depending on the environment you're in. There'll be a gas distribution system, and as I've already described, this will be linked to a continuous measurement. In this case it's an FID. And then a secondary dual FID which is making the compositional measurements.

15. What are wellsite gases related to and how do they need correction

So what are the gases related to that we measure? So when we drill well, the things that influence the amount of gas that we measure are influenced by a number of things, and they both relate to the way the well has been drilled and the geology that we're drilling. On the drilling side, the controls are the hole diameter, the amount of rock that has been cut by unit of time, and the amount of drilling fluid that's flowed back to the. That's all we need to worry about from the drilling side, and when we get into the Mercer classification approach, we'll get into this in more detail. And then from the geological side, the things that we're interested in are the, the hydrocarbon saturation, and the gas-oil ratio of the petroleum in the pore space.

16. What are wellsite gases related to and how do they need correction (2)

If we want to use the wellsite gas for reservoir evaluation, what we want to do is diminish the drilling signature as much as possible and enhance the geological signature as much as possible. So we can do somealization for these drilling parameters to reduce that drilling signal and therefore enhance the geological signal. There's various equations that you can find in publications that provide formulae for doing this. There's a bibliography, again, included in the associated materials in this course, and you'll find in some of the papers that are in that bibliography the functions that you can use to normalize the gas data. On the geological side, there should be a relationship between the amount of gas that we measure and the. The pore volume that's petroleum-bearing is simply a function of (the) × (the hydrocarbon saturation). The formation waters in zones of low-hydrocarbon or zero-hydrocarbon saturation will have some gas. There will be some gas in solution in those waters, but the concentrations will be much, much lower than any zones that have got a high-hydrocarbon saturation which may be pay zones and therefore of interest to geoscientists.

17. Data Input

So in terms of doing an evaluation using wellsite gas data, the sorts of data that you will need is:
You will obviously need the data in depth. Often if you're working with legacy data, the gas data will already be described in depth. You have to make the assumption that it's been lagged correctly, and there's a variety of ways that you can do this. You can use things like carbide tests where the operations geologists will depth-reference the data (which is collected in time) to a depth in the well. And it's obviously really important that that's done if we're going to use the data for interpretation. The simplest and easiest way of making sure that's been done correctly is simply lining up the gas data with the available petrophysical data that you have. One note of caution is lagging can be much more complicated in horizontal wells where well cleaning is much more difficult. So I would say if you have a horizontal well that you're interested in, paying attention to the lagging is much more important and engaging with your operations geologists and the drilling engineers to understand how the way the data has been lagged is a really valuable thing to do.
So as I already mentioned, to do thealization we will require from the drilling the rate of penetration, the flow rate, and the hole size. And then we'll also want to take the measurements that have been made. And then again, as I've already mentioned, this will be in the form of a continuous measurement, which will be the total gas, and then the chromatogram data from the FID which will be your compositional data. Now, we'll describe this in subsequent slides in more detail, but I'll just introduce this here. The gas that you measure can't really be used in a fully quantitative way. There are methods in the literature that advocate using a number of corrections to try and make the data quantitative, and that's fine if you have all the parameters. But often in legacy scenarios you won't have all the parameterization you need to make the data truly quantitative. So we find it's much, much better to use the data in a qualitative way. The reason why it's so difficult to use the data truly quantitatively is there's a vast number of controls to influence how much gas is actually detected. We have already introduced some of these, so the length of the, the nature of the gas trap, the set up of the gas trap, the detector equipment... all of these things will influence the absolute amount of gas that you measure. So we have to try and avoid being too quantitative with the data. But we can make the data very useful if we use it in a relative way.

18. What Causes Mudlogging Mud Gas Response to Vary?

So what causes the gas that we measure to vary? And there's a very broad range of influences on the quantity of amount of gas that we measure. Some of these are due to the way the well is being drilled. Some are to do with the equipment on the. And some things are due to the geology. If we think about the influences on the topside, a number of workers have published reviews of these controls. And on the figure on the top right-hand side, we see some workers who have published studies of how different commercially available gas traps influence the amount of gas that is being measured. And as you can see there, from Trap A to Trap I, there's a very large variation in the amount of gas that's being measured, and this is purely as a result of the design of the gas trap. The older wells where gas traps with an older design where the volume is not kept constant, we see a very strong control on the amount of gas that's being measured and the volume of mud within the trap. So on the plots on the lower right-hand side we see 2 of the traps that are on Figure 9: Trap A and Trap D (I think it is). If we take Trap A, this shows a 1% methane content in the drilling fluid. Depending on the volume in the gas trap, this can actually result in a measurement of 0.3 to even 2%. So a huge range in gas measurement purely on the basis of the volume of mud within the trap. So this has nothing to do with geology (which we are interested in), it's purely due to variation in the equipment on the topsides, or the influence of variation of mud levels on the topside equipment. People have also looked at the mud type, the mud weight, the PV and the yield point (YP) to see if this influences the amount of gas that's being measured. And workers (particularly Elder) have concluded that there's no influence on these mud parameters on the gas content.
What we'll now do is we'll introduce the Mercer gas classification system. This was published all the way back in 1974. It's a very useful model that allows us to think about what the gas that we measure on the topsides actually means from a subsurface perspective.
Donovan, W. S. "What Causes Mudlogging Mud Gas Response to Vary?." de American Association of Drilling Engineers, Denver, Colorado (2019).WIlliams, Richard D., and Scott P. Ewing. "Improved methods for sampling gas and drill cuttings." SPE formation evaluation 4, no. 02 (1989): 167-172.