In this episode, Russel Treat talks with Kathleen Smits of Southern Methodist University about how natural gas moves underground after a pipeline leak begins and why that matters for detection, quantification, and repair. The conversation covers how soil moisture, surface conditions, pressure, and gas composition shape what operators measure at the surface. Kathleen also discusses her research on evidence-based guidance for leak management and where that work is headed.

Underground Methane Leak Detection Show Notes, Links, and Insider Terms

• Kathleen Smits Episode guest, Solomon Professor for Global Development at SMU, whose research covers natural gas leak detection, leak quantification, and subsurface methane transport.
• Maguire Energy Institute An energy institute within SMU’s Cox School of Business, where Kathleen is a fellow. newswise
• Subsurface methane transport How leaked gas moves through soil between the pipe and the surface. Kathleen compares the path to spaghetti in a pot rather than a straight line.
• Leak detection Finding that a leak exists, as opposed to sizing it. Methods discussed include walking surveys, driving surveys, and aerial surveys by drone or fixed wing aircraft.
• Leak quantification Estimating how large a leak is, typically from concentration readings taken at the surface or downwind rather than at the pipe.
• Emission rate versus leak rate Leak rate is how much gas leaves the pipe. Emission rate is how much reaches the atmosphere after the soil has stored or degraded some of it. Kathleen’s research shows the two are not interchangeable.
• Advection Pressure-driven bulk flow that dominates gas movement close to the leak, described in the episode as advection driven flow.
• Diffusion Movement of gas from higher to lower concentration, which becomes the dominant transport mechanism farther from the leak.
• Brownian motion The random molecular movement behind diffusion, illustrated in the episode with an odor traveling across a classroom.
• Permeability The ability of gas to move through the soil matrix. Clay has low permeability, which slows gas movement.
• Porosity The volume of gas or liquid a soil can hold. Clay can hold a lot but releases it slowly, like a wet sponge.
• Soil moisture In Kathleen’s study of roughly seventy to seventy-five real pipeline leaks, the dominant influence on methane concentration and migration distance. Wet soil kept gas closer to the leak, while dry soil allowed wider spread at lower concentrations.
• Surface cover conditions Asphalt, wet, or snow-covered surfaces that can cause gas to migrate about three to four times farther than under dry, grass-covered conditions, based on a PHMSA-sponsored study.
• PHMSA The Pipeline and Hazardous Materials Safety Administration, a U.S. Department of Transportation agency. It sponsored Kathleen’s research on how far and how fast gas travels below ground. dot
• Multiphase flow and transport models Models used to confirm the findings from the two hundred fifty to three hundred experiments on gas migration under different surface conditions.
• Soil aeration The process used before or after a repair to remove gas remaining in the soil.
• Bar holes Holes placed in the ground so trapped gas can vent naturally or be pulled out with a vacuum. Kathleen’s group has run nationwide research on the effectiveness of bar hole drilling for identifying below-ground leaks. smu
• Soil-informed inverse models Kathleen’s group’s models that estimate leak rate from surface concentration patterns plus soil type, soil moisture, pipeline depth, and the gas in the pipe. Some operators are testing them in the field.
• Benchmark data sets Shared data sets, under development by a PhD student in Kathleen’s group, intended to let researchers compare quantification methods on the same data.
• Lidar Laser-based remote sensing used in methane leak detection surveys. Russel notes it targets methane rather than the heavier components of natural gas.
• Sniffers and multispectral imagery Other technologies used to pick up indications of a leak.
• Gas composition Natural gas is roughly eighty five percent methane per Kathleen, with ethane, propane, and heavier components varying across the supply chain. Her group’s controlled experiments compared compositions typical of the Permian Basin, the Denver Basin, and distribution-grade gas, and found that a higher density mixture can restrict how detection surveys perform.
• Hydrogen blends Experiments ranging from no added hydrogen to about fifteen percent hydrogen, with analysis of transport and quantification still in progress.
• Computational pipeline monitoring (CPM) Algorithmic tools that help pipeline controllers recognize hydraulic anomalies that could indicate a leak. Russel works with CPMs on liquid pipeline systems using SCADA data, and the linked API Recommended Practice 1130 addresses CPM for liquids. apiapi
• SCADA The supervisory control and data acquisition system that Russel’s leak detection work uses to identify leaks on liquid pipelines.
• Negative pressure wave A leak detection technique Russel mentions among the approaches he has worked with, along with camera and fiber technologies.
• Orifice equation The calculation for gas escaping through a leak when operating pressure and the size and shape of the leak are known. Russel notes this is an oversimplification.
• Cast iron pipe Older pipe material being replaced in many programs, prone to corrosion and small pinhole leaks.
• Leak mitigation Russel’s preferred phrase for the goal of eliminating leaks rather than managing them.
• Prime BEEF Air Force base civil engineer emergency teams that can respond within hours. Kathleen and Russel discuss their shared Air Force civil engineering backgrounds. af
• RED HORSE Air Force heavy repair and construction squadrons that deploy rapidly and operate independently when support beyond normal base civil engineer capabilities is needed. Russel served with a RED HORSE unit in England. af

Understanding Underground Methane Leak Detection Full Episode Transcript

Announcer: Welcome to the Pipeliners Podcast, episode 459, sponsored by Flowstate, providers of pipeline leak detection software to remove barriers and empower operators. Flowstate helps teams detect anomalies, reduce false alarms, and improve response with practical tools designed for diverse pipeline operations. Learn more at flowstatesolutions.ai.

The Pipeliners Podcast, where professionals, Bubba geeks, and industry insiders share their knowledge and experience about technology projects and pipeline operations. And now, your host, Russel Treat.

Russel Treat: Thanks for listening to the Pipeliners Podcast. I appreciate you taking the time, and to show the appreciation, we give away a customized Yeti tumbler to one listener every episode. This week, our winner is Luke Haines with Enterprise. Congratulations, Luke. Your Yeti is on its way. To learn how you can win this prize, stick around until the end of the episode.

This week, we speak to Kathleen Smits about understanding underground methane leak detection. Hey, Kathleen, welcome to the Pipeliners Podcast.

Kathleen Smits: Great. Thank you. It’s a pleasure to be here.

Russel Treat: Well, I was very blessed to sit down at a table at the AGA conference here a few months ago, and I asked you what you did. And then that spun into a whole conversation and I was like, oh, I’ve got to get this lady on the podcast. She’s got data that I need to know about. So anyway, with that introduction, would you tell us a little bit about yourself and what you do and kind of how you got into it?

Kathleen Smits: Yeah, sure. Thanks, Russel. I’m the Solomon Professor for Global Development at SMU. I’m also a fellow in the Maguire Energy Institute, also at SMU. And a third part that’s not really related to my pipeline work, but I’m also a lieutenant colonel in the US Air Force Reserves.

My research focuses on natural gas leak detection and quantification, as well as subsurface methane transport. I also do work on environmental remediation and cleanup, basically from my days in the Air Force. I use both controlled laboratory and field experiments to develop the scientific understanding to improve methane detection and emissions estimation. And ultimately, I put that under an umbrella of infrastructure management. I’ve had the opportunity to work with operators, technology developers, and regulators to improve both the safety and the performance of our natural gas infrastructure.

Russel Treat: So what do you do in the Air Force?

Kathleen Smits: I’m a civil engineer in the Air Force.

Russel Treat: Oh my gosh, that’s what I did in the Air Force.

Kathleen Smits: Oh, that’s fantastic. Yeah.

Russel Treat: So are you… I guess you’re in the reserves.

Kathleen Smits: I’m in the Air Force Reserves. I was on active duty for nine years, and then I’ve been in the reserves, I’m aging myself, for the last eighteen years. And crazy enough, the Air Force has finally aligned my professional job with my military job. I’m assigned to the US Air Force Academy for my reserve duty, so I’m the deputy director of Civil and Environmental Engineering. And so I get to come out to the Air Force…

Russel Treat: You’re an academic times two.

Kathleen Smits: I am, yes. I am an academic times two. So I can’t seem to shake it right now.

Russel Treat: Well, I’m also a civil engineer. I did four and a half years in the Air Force as a civil engineer, two as a base engineer at Tinker Air Force Base in Oklahoma City, and two as a combat engineer in RED HORSE in Wethersfield, England.

Kathleen Smits: That’s fantastic. I’ve worked with a lot of RED HORSE folks, and big shout out to them. And also your one before that, with the Prime BEEF folks. They’re fantastic, doing all kinds of contingency engineering out there. So all kinds of fun and interesting things, how to get something to work when you don’t have what you need.

Russel Treat: Absolutely. That is military engineering right there.

Kathleen Smits: It is. How do you get stuff done?

Russel Treat: That’s absolutely it. And then how do you break it? We’re very good at that too.

Kathleen Smits: So yeah.

Russel Treat: Yeah. Anyway, sorry. I’m going down a rabbit hole here. I’ll try to rein us back in.

Kathleen Smits: Sure.

Russel Treat: So I want to clarify one thing for the listeners. The nature of the research that you do as an academic is empirical research, meaning kind of base science, right?

Kathleen Smits: Yeah, it really is. What I’m looking at is I’m developing different scientific and then engineering frameworks, and that looks at everything from, and just like you’re saying with empirical research, I look at how does a leak form? After the leak forms, how does it move through the ground? What does the emission look like after it comes out of the ground? How do we then detect the leak? How do we quantify the leak? After we repair the leak, how do we remove any residual gas that may still be hanging out in the subsurface? So yeah, it goes through that.

Russel Treat: And that’s a very kind of specialized type of geotechnical engineering.

Kathleen Smits: It is. That’s one way to look at it because it’s very continuum based. So it’s connecting the subsurface transport all the way to what’s happening on the surface, to the above ground detection and then the quantification.

Russel Treat: Yeah. So the thing that was fascinating to me when we sat next to each other at the AGA conference was the work you’re doing in below ground leak detection, and particularly the work you were doing looking at how natural gas migrates through soil strata. So maybe you could tell us a little bit about some of the research you’re doing there, and maybe some of the key things that you’re learning from that research.

Kathleen Smits: Yeah, sure. So I guess I’ll start with a bit of a misconception, so to speak. When a natural gas leak begins underground, I think what most people think is that the gas travels maybe in a straight line from the pipe to the surface, but what it really looks like is spaghetti moving through the ground. If you just had a big pot of spaghetti, that’s how those gas molecules would move.

And when I started thinking about those things, I linked it to another misperception that we have, which is that natural gas leaks begin underground, but almost every single technology that we use measures gas after it’s reached the atmosphere. So that means we’re trying to understand a below ground process using these different above ground observations. You have a pipe that’s underground. You have a leak that’s underground. There’s a lot of physics, just like you were saying with geotech engineering, that are going on underground, and almost all of our measurements happen after the gas has already left the ground. So when we understand that the gas is not traveling in a straight line, but then also that we’re taking a lot of our measurements above ground, I think there’s a bit of a disconnect there with our understanding.

So first, why does gas travel the way it travels? It doesn’t travel in a straight line in the soil because of the differences in soil type. Soil moisture plays a huge role in it. The soil permeability. Soil permeability is the ability of gas to transport through the soil matrix. Then we have complexities on the surface. So pavement conditions, we have competing utilities under the ground. Maybe there’s fractures. All of that is playing into where the gas is going, and then how it’s presenting itself on the surface. And that of course is impacting our leak detection success.

Russel Treat: Yeah. So one of the things we talked about over lunch, as I got to think about this, is, if you have rocky soil, that’s pretty easy to visualize what’s going to happen. Sandy soil, probably kind of the same. But when you start getting to clay soils and wet soils, and how wet is it at this time of year? When’s the last time it rained? Is the ground fracturing? All of a sudden that starts getting extremely complex, because those soils, I would think, would tend to entrap a certain amount of the gas.

Kathleen Smits: That’s very true. Yeah, because soil moisture fundamentally changes where the gas goes. We did a study with utilities a few years back where we analyzed more than, it was seventy or seventy-five pipeline leaks across the United States. So these were actual leaks. And we found that soil moisture was the dominant influence on both the methane concentration and also the migration distance that the gas was going. Wet soil tended to restrict any of the lateral spreading, and we had, of course, higher concentrations closer to the leak. When we had drier soil, it allowed the gas to spread further with lower concentrations.

If you think about it long enough like I do, it’s pretty intuitive that that would happen. But what we were interested in is why, and then of course what the implication was for both leak detection and leak repair. We had the transport close to the leak primarily driven by pressure, so we called it advection driven flow, while diffusion became a really important factor farther away. And where that happened was really influenced by the soil moisture. So in our wetter soils, we had that methane accumulation close to the leak, and then the dry soil spread it much further.

But what it really showed was that the same leak can look entirely different at the surface depending on the soil moisture conditions. And what we were really interested in, and what I’m constantly trying to get my students to think about, is what does that mean for operators? What does that mean for how they’re interpreting methane concentrations, and how can they potentially include site conditions in that? So that’s one of our interests, accounting for that moisture and the soil property.

Russel Treat: Yeah. If I’m asking you questions that you don’t know about, just tell me.

Kathleen Smits: Sure, Russel.

Russel Treat: I tend to do this free form, my brain just goes where it goes. But one of the things that comes up for me is, I don’t know a lot about the details of how a dig is performed and what work is performed in order to make sure the dig is safe. I certainly understand about trenching and sidewalls and slopes and all that kind of thing, all the things you do to prevent cave-ins.

Kathleen Smits: Yeah.

Russel Treat: But one of the things I’ve never thought about is testing the soil you’re excavating to determine what your explosive risk is. I mean, once I talked to you, I was like, oh, wow, that makes perfect sense. You would need to do that. But I don’t even know if that’s a common practice.

Kathleen Smits: Yeah. I can’t comment on that with utilities or folks, but what I do know they try to do is a process that’s called soil aeration. After a leak is found, and finding the leak is only part of the problem, before they do a repair, and then sometimes after repair, gas often remains in the soil, like you were talking about before with clay soils. Clay soils have a really low permeability. They have a high porosity, so the volume of gas or liquid that can go into clay can be really high, but it doesn’t want to move through the soil.

If you think of a sponge, for example, sponges are kind of a strange analogy, but it acts similarly to clay. You can get a lot of water in a sponge, but then you can just set it on the side of your kitchen sink for long periods of time, and it just stays in there until it evaporates. It doesn’t flow through it. And clay acts a lot like that.

So with soil aeration, what utilities often use to remove the trapped gas is they put in a series of holes in the ground, and they call those holes bar holes. They can either attach some sort of vacuum to those holes or just let it naturally aerate out. What our work is doing is helping to develop evidence-based guidance so that utilities can remove the methane faster and more safely and more efficiently, because when they do see really high concentrations, yeah, it can become a safety risk.

Russel Treat: Yeah. So I’m going to lean way over my skis, okay, because I had a really good geotechnical engineer in college and I haven’t done that kind of work since the Air Force really, so for a very long time. But when I think about this, I think about, if I leave the clay soil in the ground, it tends to retain its water, which means it would retain the gas.

Kathleen Smits: That’s right.

Russel Treat: But if I pull that clay soil out and it’s a sunny day and I’m piling it up and spreading it around, that clay soil is going to dry out and that gas is going to escape.

Kathleen Smits: That’s right.

Russel Treat: And at some level, depending on the nature of the clay and its porosity and the amount of water it can hold and how quickly it would dry out, I could actually dig up dirt and two days later have an explosive environment that I wouldn’t anticipate, potentially.

Kathleen Smits: Yeah, yeah, yeah. That is something that could potentially happen. And it’s something to think about, what are those gas concentrations and how can they be reduced? With our soil aeration work, the questions we’re trying to answer are things like, how many boreholes should we put in? How should we space them? How far apart should they be depending on the soil type, just like you’re saying, with clay versus a sandy clay with its moisture content?

Russel Treat: Yeah, yeah. Right.

Kathleen Smits: And then what kind of vacuum pressure would we need to put on that subsurface to get the gas out as quickly as possible? And how long would we have to do that? Operators, from my understanding, do this and they have a lot of good intuition, but what we’re trying to do is say, how can we have some evidence-based guidance around that?

Russel Treat: Right.

Kathleen Smits: We can replace “oh, it just seems to work” with some science and engineering design behind it.

Russel Treat: So I want to dig in a little bit to something else you talked about, and I’m going to frame it a different way. I do work in leak detection now. The work I do in leak detection is primarily around CPMs on liquid pipeline systems and using the SCADA system to identify leaks. But I’ve also done work in negative pressure wave and various kinds of camera technologies and fiber technologies and so forth, just trying to understand that and understand how you build a comprehensive program.

Kathleen Smits: Yeah.

Russel Treat: One of the things about these kinds of natural gas leaks is that the way that we would typically identify them is actually less reliable than what I would see on the liquid side because of this behavior. So there is a disconnect between what I would pick up as an emission rate and what I would pick up as a leak rate if I were able to directly measure the leak.

Kathleen Smits: Yeah, I think you’re right, because with gases, when we’re talking about leak quantification, okay, when we detect something, detection says, okay, you have a leak. Quantification is then asking, how big is that leak? And I’m not familiar like you are, Russel, with the liquid lines, but with gas lines, I think it’s a hard question because you’re trying to infer an underground leak size based on either surface concentration measurements or above ground measurements, and depending on the leak and how it’s being detected, they use different methods.

So just to back up a bit, when leak detection happens, there are different methods to do that. That could be a walking survey or a driving survey, or perhaps there’s some sort of aerial survey going on, either a drone or a fixed wing aircraft.

Russel Treat: Yeah. And there are different technologies that are being used. There are sniffers, and there’s lidar, and multispectral imagery, and various things, different technologies that will pick up different things that would indicate a leak.

Kathleen Smits: Exactly. And so they’re measuring a concentration. It could either be a concentration on the surface or some sort of downwind concentration measurement. And then they’re utilizing that concentration and a math equation to then be able to say this is how big that leak is. Or perhaps from a regulatory perspective, they’re not interested in quantifying the leak size. Maybe they’re interested in quantifying the emissions.

But one thing I think my research has really shown is that an emission and a leak rate are not the same. They’re not immediately interchangeable. When we look at a leak rate, we’re interested in understanding how much is actually coming out of the pipe. But like we were talking about a few minutes ago, that gas then moves through the soil. The soil can act as a great attenuator of the gas and store it for a period of time, or the gas can actually be degraded due to different microbial action, for example. And then we’re taking constant remote concentration readings and trying to figure out how big that leak is. But a different question is understanding the emissions that are coming out of the ground. That’s not necessarily the same as looking at what the actual leak rate is.

So we’ve done some different studies to demonstrate that, and how many measurements we need, how long we would need to take concentration readings, what other information is needed, whether that be information about the atmospheric conditions or the soil conditions. Do we need some sort of fancy transport model, or can we simplify our model to get a good estimate of either the emissions, the emission rate, or the leak rate? But what we’ve really been trying to do in our communities, that I think is starting to transfer to both operators and regulators, is to show that those are two different things.

One thing we’re working on right now, I have a PhD student trying to create some benchmark data sets. With my work, like you were saying, being a lot of empirical experiments and such in the field, I think one of the biggest challenges in the research community with quantification or emission work is that everyone’s testing their own algorithm or their own math equation on different data. So what we’re trying to do is focus our research on creating data sets that allow researchers to compare methods more objectively. I say more objectively because someone will always have a problem with how we’re collecting the data or how we’re using our physics in order to analyze it. But that’s what we’re trying to achieve with it.

Russel Treat: So what would you say is the most interesting thing that was unanticipated that you’ve discovered when you’ve been doing this research? That’s like the thing that’s kind of not intuitive, that you were like, I didn’t expect that. Anything like that that comes to mind?

Kathleen Smits: So a couple of things. One, a few years back we had a PHMSA-sponsored grant that was looking at how far and how fast gas can transport below ground under different environmental conditions. It was very much focused on first responders, understanding how to interpret these different concentrations of gas in high leak events. And what we were able to show was that when we have different surface conditions, and I’m talking about asphalt or wet conditions or snow-covered conditions or combinations thereof, some sort of restrictive surface condition, that can cause the gas under all kinds of subsurface conditions to migrate about three to four times farther than the equivalent dry, grass-covered conditions.

This is a very generalized finding. But in order to be able to have such a finding, that we have gas migrating a certain amount further than equivalent dry soil conditions, we had to run something like two hundred and fifty. It was between two hundred and fifty and three hundred equivalent experiments, and then prove it over and over again with our multiphase flow and transport models. So just the mere size of it. It’s a very obvious thing that if you put a cap on something, you’re going to make the gas move more in a lateral direction. That’s pretty obvious to anybody who’s taken a basic physics class. But how far it was moving was interesting to us. So that was an interesting finding, I think.

Russel Treat: Yeah. I mean, if you’re explaining this, it makes perfect sense. What’s going on in my head is I’m thinking about this as to what extent operating pressure and size of leak impact this. So I would assume, and you may prove me uneducated in this area, but I would assume that a smaller, lower pressure leak tends to migrate more than a larger, higher pressure leak.

Kathleen Smits: Yeah.

Russel Treat: In general, because a higher pressure leak is going to cut, it’s going to cut an escape path.

Kathleen Smits: That’s exactly right. That’s exactly what happened. Sometimes I say, and I’m speaking from a research perspective, it’s not really very interesting looking at leaks that are super big, because at some point they’re all just going to come out of the ground and they’ll be detected for one reason or the other. There’s almost like a sweet spot with that combination of the leak rate and the soil conditions and the location close to the leak. When the gas is coming out of a pipeline due to some sort of crack or opening in the pipeline, the main driver that’s moving the gas is pressure. Whenever you have a leak, the gas will want to move from a high pressure location to a lower pressure location. That could be up, down, sideways, et cetera, and that’s based on that gas molecule just saying, I want to move to a lower pressure location. That pressure driven flow only influences the gas for about three to six feet, let’s say. And I know that’s a generality, so don’t quote me on that number, but just depending on…

Russel Treat: Sure, I get your point. It’s relatively local to the pipeline itself.

Kathleen Smits: Yes, exactly. It’s really local. But then when you have just more and more gas, it has to go somewhere. So at the boundaries of that pressure driven flow, the gas starts to move due to other mechanisms. I was mentioning earlier, it’s moving due to diffusion. Diffusion is just random molecular movement of the gas. It’s called Brownian motion. I’ve explained it to my students in class that if you were in a classroom and nobody’s moving in the classroom and somebody in the back of the classroom passes wind, I’m eventually going to smell it up in the front because of diffusion. Those gas molecules are going to just go to a lower concentration area.

And that’s why a lot of times you’ll find natural gas, which is predominantly methane, approximately eighty five percent methane. And being methane, it’s lighter than air, so a lot of times operators think, well, it’s just going to go up and out, which in general it does. It wants to utilize its buoyancy to go up into the atmosphere. But because of diffusion, it’s going to go outwards and it’s going to go downwards as well as it spreads around.

Russel Treat: So if I’m working in a gas utility in the northern states where they have a hard freeze, and that hard freeze is in place for months, then when all that ground begins to thaw out, if I’ve had leaks that have been diffusing through the soil, I might find them when all the ice melts.

Kathleen Smits: Yeah, absolutely. And so I think from a public safety connection there, understanding how your surface condition is impacting your leak concentration is really, really important.

Russel Treat: Well, right. And there are all these projects going on to get rid of old cast iron pipe that was used many, many years ago. And that stuff is prone to corrosion and consequently prone to lots of little pinhole leaks, which raises these kinds of risks. It’s so interesting to me, Kathleen, that I’ve done almost five hundred episodes at this point, and I’m still finding whole new areas of science, operations, and risk management that I had no idea were even out there. This being one of them, right? I mean, once you hear about it, you’re like, oh, well, that makes perfect sense. Sure, that’s a risk. Sure, somebody should be looking at that.

Kathleen Smits: Absolutely. And I’ve been very blessed that I have the opportunity to do that, and to partner with operators and other scientists, regulators, solution providers. Ultimately, I think everyone’s aligned on the goal that we want to make our energy infrastructure safer, and we want to make it more efficient, and we want to make it more environmentally responsible. And I think the whole future of, I don’t know if you want to call it pipeline leak management or whatever, is moving more towards how do we take this information and work it into our decision making?

Russel Treat: Yeah. I like using the word leak mitigation.

Kathleen Smits: Yeah, exactly.

Russel Treat: We don’t want to manage leaks. We want to get rid of them.

Kathleen Smits: That’s a good point. Thank you.

Russel Treat: But yeah, so I want to ask another question. If I have a pipeline and I know its operating pressure and it has a leak and I know the size and shape of that leak, it’s a pretty simple matter to calculate and quantify the amount of gas that I’m losing, right? Just because it’s basically an orifice equation.

Kathleen Smits: Yeah, it is, exactly. Basically it’s an orifice equation.

Russel Treat: Yeah. It’s an oversimplification, but that’s basically what you have. But if I want to quantify a leak and I want to quantify it based on some concentration I’m reading at the surface, is that even possible? I mean, I would think what you’d have to do is know, well, what concentration am I reading? What soil do I have? And then what are the other contributing conditions that would cause diffusion, so that I could say, well, I’m probably reading this concentration and it’s probably this much diffused.

Kathleen Smits: Yeah. We’ve come up with, my research group and I, we’ve come up with two different, I call them soil-informed inverse models. And so if…

Russel Treat: That’s great. This sounds like a name that a PhD would come up with. We need a better name.

Kathleen Smits: It’s really descriptive.

Russel Treat: It’s a great name. It just sounds like a PhD came up with that name.

Kathleen Smits: Yeah, exactly. So if we just have, like you’re saying, Russel, if we just have the concentrations on the surface, the question we’re trying to answer is what more information would we need to get a good enough answer for our leak rate. When we work with operators and ask, well, what’s good enough, we always get a lot of “it depends,” based on who they are and whether they’re working in upstream or midstream or downstream. If we talk to a regulator, they have another answer of what good enough would be.

But what we’re trying to understand is, if we have concentration readings, we know where they’re located. I often call it just the concentration contours on the surface. If you can imagine near a leak location, you may have higher concentrations, and then as you move away with distance from that leak location, you get lower concentrations. That’s pretty obvious. But if we have that pattern, and then we also know just a little bit about our soil, the most important parts being the soil type and then the moisture, can we use that information along with some knowledge of perhaps how deep the pipeline is and then what the gas is in the pipeline, because that varies if we’re talking upstream versus downstream, can we use a relatively simple model in order to come up with a good enough estimate of our leak rate?

We’ve put together a couple of models to do that, and they work relatively well for the soil conditions that we’ve tested. And there are some operators that are utilizing those in the field and giving us feedback so that we can continue to make them better.

Russel Treat: Have you looked at hydrogen at all and how it would behave differently?

Kathleen Smits: Yeah, we actually, I think it was about a month ago, the beginning of June, we ran some experiments. We were actually using different hydrogen blends, anything between no added hydrogen up to about fifteen percent hydrogen.

Russel Treat: Okay.

Kathleen Smits: And then how that impacts the gas transport. And then we’re also looking at the quantification with that, but that work is so new we haven’t even had time to analyze all of that.

Russel Treat: Yeah. Well, and I would assume, I think it’s about fifteen hundred miles of pure hydrogen pipeline in the US. It’s around industrial facilities. And I would think that it’s all the same thing, except that it travels further, faster because it’s a smaller molecule.

Kathleen Smits: It’s a very, very small molecule. Yeah, absolutely. And what we’re most interested in, as I was getting at a bit with the composition of natural gas, is that it’s not chemically identical through the entire supply chain. So it contains different amounts of methane, ethane, propane.

Russel Treat: Pipeline quality gas, and I’m doing air quotes, Kathleen can see me because we have our cameras on, but the listeners can’t, is only eighty to ninety percent methane.

Kathleen Smits: Yep.

Russel Treat: It’s got a whole bunch of other heavier stuff in it.

Kathleen Smits: Right. And the composition affects what you’re talking about with the leak quantification, but it also affects the leak detection. We did some controlled experiments, and we published this last year, where we looked at different compositions of gas, compositions that are generally seen, for example, in the Permian Basin versus the Denver Basin versus a utility distribution-grade gas. We were able to show that the gas mixture can impact how far the gas is moving under the ground, which changes the surface configuration of the gas. And when we then link it to leak detection, it impacts the probability of leak detection success. A somewhat practical takeaway of that is that when we have a higher density gas mixture, that can restrict how the leak detection surveys perform.

Russel Treat: Yeah. Because we’re sniffing for methane and we’re not sniffing for those other things, or we’re using lidar for methane. We’re not using lidar for those other things, typically.

Kathleen Smits: That’s true. Yeah. And again, that’s another rabbit hole you can go down, because that technology is fascinating as well, what it can do, and yet what the limits are in terms of a human being actually working with it on a day to day basis.

Russel Treat: Yeah. Well, hey, Kathleen, if anybody is interested and would like to get their hands on some of the research papers, where would they go to find that sort of thing?

Kathleen Smits: Yeah. I don’t know if we can link them here.

Russel Treat: Yeah, we’ll definitely do that. In fact, if you will send me some links, I will add that to the show notes on the Pipeline Podcast Network website and link it into this episode on that page, and that way people can find that and go to those resources if they’re interested.

Kathleen Smits: Yeah. And I publish all of my work open access, so you don’t have to have subscriptions to journals or anything. Anybody can access any of my work.

Russel Treat: Well, I’m very glad that I happened to sit down with you. I’m even more glad that I introduced myself and asked, what do you do? Because that has ended up being a fascinating conversation.

Kathleen Smits: Absolutely. And it’s been a pleasure visiting with you. This is awesome. I really enjoyed it. And I’m glad that you do this podcast, because I didn’t even know it existed before we started talking. I have a commute to work and I’ve been listening to a lot of the content.

Russel Treat: Oh, wow. Awesome.

Kathleen Smits: Yeah. And you have all kinds of folks on your show. So it runs the gamut of all the things pipelining.

Russel Treat: Absolutely.

Kathleen Smits: It makes it great. And I’ve introduced it to my research team, and they’ve been picking up on different stuff because a lot of times they don’t have the opportunity to talk to operators. We do, of course, when we’re doing some fact finding or we’re out in the field partnering with them. But a lot of what we do is controlled experimentation. So that is helpful, just getting them to hear the lingo and understand why they’re doing certain things. So it’s fantastic.

Russel Treat: Yeah. And understanding the distance between the empirical research and the applied technology and practice. There’s a long journey between those two things.

Kathleen Smits: Really long journey. Absolutely.

Russel Treat: Yeah. I’ve done a lot of that. In fact, I got a model from my good friend Chris Williams, and he’s allowed me to kind of adapt it, but it’s a journey, particularly in our world, because you have to create standards and practices so that people can actually use this stuff.

Kathleen Smits: Absolutely. Maintain it.

Russel Treat: Yeah, very much so. Thank you so much for your time. Great to talk to you. And when you do your next kind of interesting research, ping me, I’ll bring you back on and we’ll get even deeper and geekier.

Kathleen Smits: Absolutely. Yeah, we can go on a deep dive into any of these topics. I’m always interested in that. So yeah, that’s great.

Russel Treat: All right. Thank you so much.

Kathleen Smits: Thanks, Russel. Take care.

Russel Treat: I hope you enjoyed this week’s episode of the Pipeliners Podcast and our conversation with Kathleen. Just a reminder, before you go, you should register to win our customized Pipeliners Podcast Yeti tumbler. Simply visit pipelinepodcastnetwork.com and enter yourself in the drawing.

If you have ideas, questions, or topics you’d be interested in, or if you’d like to be a guest, please let me know on the Contact Us page at pipelinepodcastnetwork.com or reach out to me on LinkedIn. Thanks for listening. I’ll talk to you next week.