Methane Monitors: Overflight observations reveal decline in Permian emissions

View Full Transcript

Hill Vaden:

All right. Welcome back to Energy Sense and S&P Global Commodity Insights Podcast, covering all topics on the intersection of energy and finance. This is your host, Hill Vaden, here with your other host, Sam Humphreys. I almost got your name wrong.

Sam Humphreys:

What I like is we have done over a hundred episodes together. I think that's the first time you've said it wrong. So well done.

Hill Vaden:

Oh, wow. Well, so 1%.

Sam Humphreys:

1%. That's not bad.

Hill Vaden:

That's not terrible.

Sam Humphreys:

I'll take that margin.

Hill Vaden:

So helps this first segue, that if a 1% rate of me getting your name wrong, we are talking to Raoul LeBlanc today about the percentage of declines of methane or methane emissions in the Permian Basin, with some overflight data that he's been working with that describe a pretty compelling story. But can you give people some high points to listen to over the next 30 minutes or so?

Sam Humphreys:

Yeah, absolutely. So this was an excellent explanation of how methane is monitored and how we capture that data, and then what we do with it, and what companies are doing to reduce that number.

So you said there, it was a really significant drop in methane emissions, and I am saying methane or methane, because I'm English. And it is fascinating, because there was this really significant drop, and Raoul goes into an excellent explanation of the various different technologies that have been sort of developed over the last couple of years into how we kept that. So whether it's overflight, which is as simple as something flying over an area to capture data, and cameras and satellites as well, and then the different resolutions and the quality of that data, and also the complexity, then, as to how do we know how much methane is actually being emitted? Because that number is up for debate in certain circles.

So as part of S&P, this report that he's worked on, that we've put out kind of setting a benchmark, so at least we have a number that we can use and track going forwards, which is fascinating, and how AI and technology is sort of playing a role in this space. But what about you? What did you think?

Hill Vaden:

Yeah, I think this is just, I like what you said, but this is just another one of those areas. I mean, one of the things about the energy industry, or in this case, the upstream industry, just throws off so much data. And the more you pay attention to that data, you can innovate around it, and solve problems, and create value in the process. And so this is one of those things that Raoul mentions that 10 years ago nobody was even talking about.

And then people started talking about it and now the data are being acquired and managed, and people are learning things and innovating, and the end result is less emissions, which is good for the atmosphere. So with that, we can hand off to Raoul and we'll let you go.

Sam Humphreys:

Enjoy.

Hill Vaden:

All right. Raoul, thank you for joining us today. The middle of August, and I guess, where are you joining us from today?

Raoul LeBlanc:

I am in the lovely city of Calgary, Alberta.

Hill Vaden:

Okay.

Raoul LeBlanc:

Where it is, I got to say, Hill, I am moving back to Houston in October. I got up this morning, and I went out to walk the dogs, and I had to put on a jacket.

Hill Vaden:

Wow. You might be like the one guy in the world putting on jackets. Sam's been complaining about the heat-

Raoul LeBlanc:

That's right

Hill Vaden:

.. in the UK all week.

Raoul LeBlanc:

Well, that is why I'm not moving back to Houston this week.

Hill Vaden:

Yes. Nobody moves to Houston in August.

Raoul LeBlanc:

I thermometer-

Hill Vaden:

Nobody with a choice.

Sam Humphreys:

Very sensible.

Hill Vaden:

Well, we often talk about the weather, but we're not here to talk about the weather. We're here to talk about methane, or for Sam's purposes, methane emissions in the Permian Basin, and some recent work, Raoul, that you published around methane emissions or methane emissions declining in the Permian Basin, which has been a big focus for some of the Permian operators and the industry as a whole. Can you talk a little bit about what the work you've done is revealing and why it matters?

Raoul LeBlanc:

Sure. So we have been tracking methane emissions. Now, a lot of people have been estimating methane emissions and tracking methane emissions. And one thing I should say, right from the get go, nobody, nobody on this planet, or probably any other planet, knows exactly how much methane has been emitted over any discrete time period for the whole basin. Individuals have maybe some great data for individual areas or individual time periods. Nobody knows.

Okay, so everybody's making an estimate, but there's a huge gradation of those estimates in terms of their quality. So what are we seeing? We think we have the best estimate ever released to the public, and we've been doing it for about three years. We'll talk about how it comes from, but the headline story is that the Permian upstream operators, in terms of their operations, have reduced their methane emissions from 0.93 in 2022 to 0.44 in 2024. So it's cut them in half. And that's a big deal, and it's not terribly surprising when you start to get into it, but it's a good news story, in the sense that people are spending money, they're doing things and it's taking impact.

Sam Humphreys:

So before we get into how these emissions and why [inaudible 00:05:47] matters have been reduced, can you walk us through how we track that? How does someone find out how much methane is in the air, for example?

Raoul LeBlanc:

Yeah, it's a great question, because methane is natural gas, or it's the main component of natural gas. So you can think of it as natural gas. So you're producing oil or you're producing gas, it comes out of the well, and methane gets emitted in two ways. Number one, it can get emitted on purpose, and number two, it can get emitted accidentally.

So remember, this is an invisible, colorless gas. You can't see it, you can't really smell it, and it disperses into the air relatively quickly. So it's hard to know exactly how it is. So the ways you find it are using specialized equipment, and there's a couple of levels. One is you can put a sensor. In reality, because the wind may be blowing away from the sensor, you have to put a bunch of sensors on the site, and kind of hope that there's some wind, because if the methane rises straight up, you won't catch it. So anyway, but there's sensors, and they're on all the time, and that's great, and they can be very sensitive. They don't quantify well.

The next level is something called an OGI camera, which is a specialized camera, and when you look through it, it detects methane, and you can literally see the plume as it leaks from somewhere or releases from somewhere. And so you have that.

Then you have some other technologies that are slightly higher up and try to do a larger area. But then you have the one we'll focus on here today, which is overflights. Okay, so overflights are airplanes, sometimes drones, but that's a different issue. We can talk about mostly airplanes, and they fly over an asset in kind of a lawnmower pattern, and detect these methane emissions. And the way they do it is they look at when sunlight goes and strikes the earth and bounces back, if there's methane in the way, methane absorbs certain wavelengths of light.

So what they do is they look at the reflected sunlight back up in the plane, okay, they're looking at reflected sunlight, seeing where have those wavelengths been removed? There's some methane there. Let's quantify that. It's pretty accurate.

So anyway, to get the bottom line is how do we detect methane? We have several methods. The last one is a satellite that's very high up. It looks over a big area. And so you have this whole panoply of technologies. The problem is none of them is perfect. It's the proverbial, there's no silver bullet problem. Some of them are great at detecting very small. Some of them are good at covering a big area. Some of them, like satellites, cover it maybe every day. Other ones are very infrequent.

So you're not getting the full picture at any time from any technology or any suite of technologies. So there's always an element of estimation, that we have to estimate things. And that's where you get, of course, variability and uncertainty.

Like I said in the beginning, there are two kinds of emissions. There's the ones that happen on purpose, and you say, "What? On purpose? Why would anybody release natural gas on purpose?" Well, a lot of it may have to do with safety. Or there's equipment, there's overpressure, and you need to release that pressure. Or a lot of times it's been because, "Oh, we're performing maintenance on this piece of pipe and we're just going to let this out."

And then, of course, there are the accidents, there are the leaks. Stuff breaks, you don't know it, and it starts to leak methane. And both of those have been in the sights of companies as they have tried to use these technologies to identify where we're having more leaks than we should, and then also, where can we look at those intentional emissions and change out the equipment, or use a better method, or capture it when it gets released and put it back in the pipe? So we've got a bunch of things going on in terms of technologies here in the space.

Hill Vaden:

All right, and so I'm right that the one that you've been working most closely with is the overflight technology, seems to indicate that emissions are down relative to production, and this is as production grows, so this is on an absolute basis. Am I right, that on an absolute basis we see emissions falling, and on a relative basis we see emissions falling?

Raoul LeBlanc:

Yeah, I mean you're getting a big drop in emissions at the same time. Your drop in emissions is quite large, at over 50%, or about 50%, and your production growth is there, it's up maybe 10%. So it's not enormous production growth. And so obviously if you're reducing your emissions by 50% while your production's growing, both your absolute is coming down, which is great, and obviously the intensity is going down even more, so per barrel produced.

And that's important. People say, "Oh, we have these emissions. Isn't that terrible?" Yes, but it always has to be balanced. I mean, you're getting something for that emission, which is you're producing oil and gas. Now, the best situation is you have no emissions and you produce all of that, right? Because remember, any methane emission could have been put in the pipe, sold, and used on your stove, or to heat your house, or to generate power.

Hill Vaden:

So what's contributing to that decline?

Raoul LeBlanc:

So it's a bunch of things, and this is where it gets kind of interesting, and we can see the data, and the data are telling us there's just fewer leaks. And by the way, we're able to look at the distribution of data. So some leaks are giant, by the way. Some leaks will be, they're measured typically in kilograms per hour.

And so you've got a lot of leaks that are above a hundred, which is big, and some that are above 500 or a thousand. And even if they're not on for very long, because they're so big, they make up a big chunk of the total methane emissions [inaudible 00:11:49]. And then you have leaks all the way down to less than one kilogram per hour. And so you have a big wide variety of leaks.

And I don't know, my go-to analogy is water. Okay? It's like think of water, how much water is leaking. If you have a main that burst, you're going to lose lots and lots of water, but we have tens of thousands of leaks in everybody's house, from toilets or sinks that are slowly leaking.

So the same thing applies to these methane emissions. And so we were able to look at the distributions in 2022, 2023, and '24 and say, "Okay, where's the progress coming from?"

So first thing to note is the progress is coming from across the board. We are eliminating the number of big leaks, number of medium leaks, and the number of small leaks. The number of small leaks is probably not going down quite as much, because the easiest ones to find, and the ones that always pay off to do something about are the bigger ones. And so we're getting stuff there. Now, that's a question of the numbers and the distributions, right? Well, why is this happening?

So the reason it's happening is, remarkably, in some senses, mundane, which is people started paying attention to it. It's almost embarrassing that we're making this enormous progress. Why? Well, particularly the field personnel, it has become just a normal part of their job.

First of all, they've been given tools and technology to do it. So now you have a sensor that goes off, or the airplane goes over and almost immediately they'll send a message to the company and they'll say, "Hey, this compressor at this well pad is leaking." Okay, now we know, we can do something about it. Whereas before, it might've waited around until some annual or quarterly inspection for you to go find it. So one is you're seeing things, and therefore you're fixing them.

Second one is your field personnel. Once you get called out two or three times over the course of the year to the same place that has the same leak, you're like, "You know what? Maybe we ought to fix this." So you're identifying it. And then they've just made it part of their job. It's in their goals now for most of the companies. You have a methane leakage, methane emissions goal. You want to hit that. It's like a safety goal, or a production goal, and incentivizing people, building it into the culture. That is starting to happen. And we've heard that in all our private discussions with people. We were at CERAWeek, and I was asking people actively this question, and this is the kind of things they're telling me. They have all these anecdotes about how it's working.

So normalization as a part of field operations is a big one. And then detection activity, the better the detection is, the more we're going to do something about it and the quicker we're going to do something about it. That's the big thing. There's always going to be a leak, but if you can catch that leak within a few days or a few hours, it's enormously different from leaks that would literally go on for half a year or a year,

But it's cause we didn't know then. Because they had to take some sniffer out there, and it would take five hours to do one well site. And I have hundreds of these things. There was just no time.

And the last thing I would say is better equipment. So we started designing equipment with methane in mind. If you go back to 2010 or 2012 or whatever, and you look at the equipment, methane wasn't on the radar, so it wasn't like a zero bleed device, it was a high bleed device, because that was the way it was. And the last thing I'll say is it's kind of cool, and it sounds kind of AI wonderland, but there's a mode of, "I got to find and fix it." Okay, got to detect it, send somebody out, fix it. Sometimes it's turn the wrench, close the valve. Other times it's replace something fine.

But that's find and fix, and eventually that turns into whack-a-mole. What you would rather do is be a little bit smarter, and do predict and prevent. So if you can use your operational data, remember, oil and gas, digital oil field, that was a thing 20 years ago. It's turning out loads of data all the time. So if you look at the detections and you say, "Hey, you know what? It looks like this particular type of heater treater develops problems after year five." Okay, I can do something about that ahead of time. Or if I see conditions that always precede a big leak, well, I can maybe go out there and try to prevent it. So structural problems, that's kind of the way for the next wave after find and fix is predict and prevent.

Sam Humphreys:

So if we look at it from a company standpoint, you're checking for emissions, particularly methane, and you're reporting on it. What at the moment, there's obviously been lots of changes in the U.S. What is the significance for reporting that at the moment? There's been some changes again, is this tax breaks for reporting it or not reporting it, or is it just a best practice?

Raoul LeBlanc:

Yeah, it's a great question, because this space has become very controversial. And whereas you might make the argument that CO2 is a byproduct of economic activity, and getting things done, and going somewhere, get on a plane, yeah, there's no real reason or no real benefit to methane leakage, right? And so it's always been what I feel is a different paradigm for the companies, which is, it's wasteful. Here's the resource I am actually trying to produce and sell, and have done it.

Now, there's a point at which it becomes too expensive to justify fixing a tiny, tiny leak. You don't want to spend $10,000 to fix a $1,000 problem, but we're far away from that still. So it has been, I would say, controversial, because there's been a lot of discrepancy around what the true leakage rate is. We can talk about that in a minute, but you're asking on the incentive side, on the political side, what's driving the companies?

And the truth is that almost all of the activity has been voluntary efforts by companies to reduce their methane. They made targets several years ago. Now, it gets controversial because the IRA had a methane fee, and by the way, that methane penalty was quite steep, and it was above a certain ... You were sort of allowed a certain threshold, but it was very steep if you had to pay it on a dollar per MCF basis.

The current price in the market is somewhere around $3.00, $3.50, but the penalty was something north of 20 bucks per unit. So it was a big deal. Now, the current administration came in and they have essentially canceled that, so it's not going to happen. But I'll point out that all of our data from '22, '23 and '24 took place during a period when there was no penalty.

So at that point, the only penalty was, well, you lost out on your economic value to sell the gas. Now, you had to report these numbers to the EPA for their greenhouse gas repository. But the methodology that the EPA set up, and I am grateful to the EPA in 2012 for doing something, because it started getting a track record, but the methodology was not good. It relied on spreadsheet calculations. Basically it said, "Hey, how many valves do you have? And we estimate that valve leakage is this. So multiply the number of valves you have times this, and there's your number."

Again, it wasn't a bad thing to do in the beginning, but it didn't ... The big, big development in methane reduction has been this kind of engineering approach versus observed data. And that's the revolution in the last three or four years, we now have loads of satellites, we now have lots of flyovers. We now have lots of sensors that are physically trying to observe it, rather than just trying to estimate whatever that number was going to be. And that penalized, the problem is, A, and we were on public record saying this, basically the EPA methodologies and all the spreadsheet methodologies chronically underestimate the actual leakage.

And so the EPA was actually trying to fix a lot of those methodologies to A, include observation, and B, bump them up. But the point is, so you had under-reporting. And then you've got this slew of people who started observing and they said, "Guys, we have a real problem here. Everybody's been saying it's at x, and the answer is really like 5x."

And it became, that's what has led to a lot of the controversy, and a lot of the sort of rage out there over these numbers. Now, I will say this, it's compounded by, in its very significant way, methodological choices. So Amanda, let's say you and I went out and we agreed on the amount of methane being emitted in a particular area of the Permian Basin, and we are 100%, we agree that this is the number. I got to tell you, I can get an order of magnitude difference in just the ways that we express this and what we include in that number.

It's crazy. It is. So it can go from 1% to 10%, even if we agree on the number. Much less if we don't agree on the number. And that's been a real eye-opener. And it's one reason that the public dialogue has not really gone anywhere on this, because everybody's talking at cross-purposes and comparing apples and oranges.

So we're hoping that this Permian benchmark that we've now been putting out for three years establishes a number that at least is a steady, fixed number, based on observation. And we can talk about why we think it's the best number, but we believe it represents actually a number people can hang their hat on. And then you start looking at performance. Differentiation, and improvement or backsliding. Okay, that's the conversations we ought to be having.

Hill Vaden:

So if we're looking at the Permian Basin over the past several years on this, I guess a couple of related questions on this. Is it the operators bringing innovation into the space? Is it oil field services bringing in innovation? Is it some guy from his garage who's fancied some kit that he's bringing in wholly new? Who gets credit for the improvements? Or is it just people being more responsible with things that they were and continue to do? And then a followup to that question, we're talking about the Permian, are we seeing the same types of trends or can we expect the same types of trends in other basins, whether that be basins in the U.S. or basins outside of the U.S.?

Raoul LeBlanc:

Yeah. Okay, let's take those, those are both good questions. So one I would say is the innovation ecosphere that has led to these new technologies has not been from the traditional oil field services. Now, there's probably a role for them to scale them and own them in the long run, but it's crazy. They're all little startups. Some of whom, by the way, are backed by the oil companies.

But they look to an innovation model which was, "Hey, let's get startups out here, get technologies from other places and adapt them." Look at remote sensing technologies like they have in satellites, and then you build a dedicated satellite for methane. So it's been fun to watch that kind of innovation space and incubation happen with little startups. There are garage startups that are up and at 'em, and there are other ones that got back by OGCI, who is the Oil and Gas Climate Initiative, which is basically the 15 largest oil and gas producers on the planet, and they have their venture capital arm.

So I would say them, but the real deal is you can invent all these technologies, but we went through a period from about 2020 through 2022, I would say, where it was a horse race time. It was beauty contest. Yeah. One company, independent that I talked with, they said, "We are testing six technologies at the same time, and we're going to put this technology A over here on 13 pads, and we're going to put technology B on 12 pads over here, and this, and we're going to compare them and see what we get. And we're going to carefully try to calibrate that through the most detailed measurements we can on a sample." So there was a real sort of winnowing to some way, and what people found out was what you'd expect, "Hey, this one works great in this situation and I'm going to go with this," or, "I need a combination of this. I need flyovers every quarter, and I need OGI cameras every month on these types of sites," or whatever it is. People have come with that mix.

That period of deployment and testing is largely over, and people have picked what they think is their winners. There's technologies that continue to improve, obviously. The algorithms are improving every day. And what's interesting is you're getting more coverage and more deployment. So it's happening now, and I feel like as of '22 onward, we're now in results mode. We're out of testing, we're into mass deployment, and we're getting results, and now we're going that next step, which I mentioned before, which was, "Oh, let's get the data from this and see if we can be smarter about it, and try to go further using machine learning and AI."

Your second question about deployment where. I would say this, first of all, there's a couple of things you should know. Satellites and overflights generally do not work over water, and they also have real problems with northern latitudes, and sometimes with snow cover. So a place like the North Sea can't really utilize a lot of these tools, so the Gulf of Mexico or offshore Brazil, the Golden Triangle, Angola.

So that's a bit of a problem. But in terms of deployment, North America, both Canada and the U.S., first of all, places like the North Sea don't really have a problem. They've been using other kinds of equipment, and it's relatively squeaky clean. We've got some interesting data on that that Chris Kennett runs, and places like Australia are generally good. But I would say this, the rest of the world is far, far behind in terms of deployment of these technologies. And so the satellites are going around and they're seeing a lot of things, and generally finding that, not in every case, but there are some areas of the world that are very problematic.

But the biggest problem is we just don't know. It was interesting. Well, S&P did this great study, it's been made public, about LNG, right? And you can read it. And we looked at LNG gas, the gas that was supplying LNG basins around the world, and we worked a deal with some satellite companies to try to access imagery of those. So these are all the LNG feed basins around the world, and you all know the countries. So we looked at that, and we got as much data as we could get. And I got to tell you, we had an example where you looked across, for example, all of North Africa and the Middle East, and we had something like seven billion pixels. In the Permian Basin alone, we had 330 billion pixels.

Sam Humphreys:

Quite different.

Raoul LeBlanc:

Yeah, the sheer amount of looking has been enormously higher in North America. It's that most heavily sort of observed space. So I think there's a lot of room to make progress in other parts of the world too, because again, once you start looking, you start finding and you start fixing.

It's kind of interesting. It brings up this whole, this is my crazy idea, kind of hear me out. But one of the big problems that we are going to have, and this applies, by the way, to emissions generally, is a misuse of capital. Because let's say, like the North Sea. The North Sea was kind of ahead of people on methane emissions, and they have their emissions down, let's say, I'm going to make up the number, because I don't know it, but a good number is probably something like 0.1%.

And you have, then, the U.S. spends a lot of money, and it gets down to 0.2, something like that. If you have a lot of countries around the world that are at two, three, 4%, it makes very, very little sense for the incremental dollar to go toward trying to get high-hanging fruit in these bases. It's a real problem, because you'll have a lot of capital being expended for very little effort, or very little, sorry, very little result, when really, what you would like in an optimized world is to be able to spend that money on the easiest, low-hanging, no-brainer type of methane reductions. But that's not the way it's likely to work.

Sam Humphreys:

All right. Well, just looking at time, we could talk about this for a lot longer. We always like to sort of finish these conversations with a look forward, sort of maybe in the next six to 12 months. So if you're talking to everyone, which you are, what is a couple of, or maybe one point that you think we should be watching for in this space that could change things, or we should just be looking out for near term?

Raoul LeBlanc:

So we think that companies will hit, all the companies that have announced goals, or 90% of them, on methane, will hit their methane goal. So I think that's good.

The real deal is now, we have to look under the hood. And we have a publication that's coming out, which I'd be glad to ... That's my hook to try to get back on the show. But we have another publication coming out, which is going to start looking under the hood at the distribution of companies. How are individual companies doing, which companies may be doing better? Public, private, conventional, unconventional, what's the source of the problem?

So as you get further and further down, I think the thing to watch will be how do companies differentiate themselves? Because we are seeing a situation in which you have people making fantastic progress, and then you have a cohort of laggards that really don't look good. How does the market, how does the industry, and how do we deal with those and help ... The people who spend the money should get the credit for what they're doing. And if the basin average improves, that's great. But in reality, if you're doing the best, and you're taking it the most seriously, and getting the best results, or have the cleanest operations, you need to find a way to show some return on that investment. It incentivizes people to keep going. So that's something to look for is how does differentiation play out as we move down the road toward meeting these goals and going beyond?

Hill Vaden:

All right, well that's plenty to look forward to, and it's always great having you on this podcast. We'll find another opportunity to do it again as that next level of study comes out.

Raoul LeBlanc:

Thank you.

Hill Vaden:

Well, thanks so much for joining us Raoul.