Failure Files 2.0: Geohazards

0:00 Rhett: Alright, as we continue our series on the Failure Files, we come to the conclusion and we go back to a familiar failure. So, we've been through Macando, Challenger, Type A, Boeing, and throughout that, we've seen a pattern of kind of missed opportunities and blind spots. I think that as we go back and take a look at this 2023 Geohazards failure, that pattern continues. It's a great opportunity awe return to an emerging threat to stop and talk about the importance of paying attention to signs that can help us see these failures before they occur. Thanks for joining us. 0:57 Rhett: Chris, today on this episode of Pipeline Things as we continue on our Failure Files series, we go back to the threat that's close to me. The one that we've had guests on multiple times. In fact, we actually we covered, 1:12 Christopher: I think I know who this is! 1:17 Rhett: We covered a failure similar to this in the last Failure Files that we did and so we naturally come back to it. Christopher: Are we gonna have a guest? Rhett: No Christopher: Is it who I think it is? Rhett: No, we don't do guests on the guest on the Failure Files. You know that. The Failure Files is just you and I. What's also cool is this is a fairly recent failure in terms of what we cover. 1:39 Rhett: We're going back in time to February 12th, 2023, and for the audience, we're going to be talking about a geohazards failure. Christopher: So not that far back? Rhett: Yeah. So really no, not that far back. I mean, a little over two years ago, right? Recent enough where most of us can generally remember what we were probably doing in our lives at that point in time. Christopher: Yeah. Rhett: So, two years ago, the Gulf South Index 381 pipeline, which is operated by Boardwalk, ruptures and we have a release, but not an ignition of natural gas. Which I think there are a lot of points in this story that are like, I think for the audience, real eye-openers in terms of lessons learned, like in preparing for this, this was probably one of the easiest episodes I've ever had to check off oh man, this is things that operators should be aware about. Check, check, check. And really good lessons learned. I don't mean obvious things like, Christopher: Yeah. Rhett: Oh, the operator did something wrong and they should have known this, but things like, oh, yeah, there's a bunch of operators, Christopher: Yeah. Rhett: That would fit this category. 2:43 Rhett: But point number one, we have a release in a natural gas pipeline, no ignition, Christopher: Yeah. Rhett: And the PHMSA report notes that the cost to repair was $4.8 million dollars. That was the first thing that really surprised me. I was like, we got to $4.8 million in cleanup costs, and we didn't even have a rupture, we didn't have a loss of life, that that's a surprise. And the reason I chose to open with that is to put it in perspective, right? So, I know for a lot of operators when we talk about geohazards, and we talk about it's a new threat. So, there's always I think sticker shock and resistance to the idea of developing programs to address a new threat and spending additional resources. It's worthwhile just to stop before we get into the failure and say, look, $4.8 million is significantly more than I've ever seen any operators spend in their first year standing up a geohazards program. Christopher: You could buy a lot of IMU data with that. 3:37 Rhett: You could buy a lot of IMU data, you could do a lot of bending strain, you could run a lot of LiDAR with $4.8 million. So again, it's worth. So, but let's go back. So, this incident in February of 2023 happens on a pipeline in Jackson, Mississippi, which is second point. We are not in California, we're not on the West Coast, we're not in the Appalachians. We're not anywhere you would typically associate with geohazards; we're in deep South. And what happens here is the Index 381 pipeline is originally an 18-inch pipeline that was built in 1939. 4:09 Rhett: Yeah, right. So, one of our many very young pipelines in the United States. Christopher: That's circa World War II, right? Rhett: Yes, but at this particular location, the pipeline was rerouted in 1952. So, what happened is US Silica is the owner of the facility at the time that interacts with this pipeline, wanted to construct a, like a detention pond in this area and Gulf South needed to route the pipeline around. So, it's real cool. So, if you actually look at the map, the pipeline is straight as an arrow, except for this one place where it goes woooop and routes around. And when it routes, they don't use 18-inch material, they use 20-inch material. So, it's an 18/20 dual diameter pipeline. And only 20-inch section is right here where they routed around this pond. 4:53 Rhett: And you can actually see this. So, if you go, you can pull all this up in Google Earth to the audience. I encourage you if you pull up the PHMSA report, all this data is there, including the images. And everything in this show, I might say, we took all exclusively from the PHMSA report. It's actually a really good report. So, what's interesting about this pond, Chris, is it's not a pond, it's like a hole in the ground. It's a pond that they built up above the natural elevation. So, they built it up with like dams and detention ponds that are above natural ground elevation. So, if you were walking on the pipeline right-of-way, you would have this big earthen dam that's like 20 or 30 feet going up from the right-of-way around. 5:33 Rhett: And what's also, again, another really fun fact that plays into this is if you go back and look at the satellite imagery of this area, we do this as part of our bending strain assessments. We're often using Google Earth satellite historical imagery. They have 26 different time stamps of this site between 1992 and 2022. So, over a 30-year span, there is, I would say that's an above average number of satellite shots of this area. It really gives you a picture in an unusual way that you can kind of get a decent grasp of historically what's happening at the site. Which you'll notice if the audience does that is you can see quite clearly going back to 2003, 6:17 Rhett: That the owner of this pond, US Silica, which is just to the north of the pipeline, was having stability issues with the dam. So, you can actually see in the satellite imagery, and the PHMSA report notes this, between 2007 and 2010, you can see them doing work on the dam between 2017 and 2018, you can see it again. And then in 2019, you can actually see the ground cracks in the satellite imagery adjacent to this dam, right? So, as you can imagine the pipeline is south of the pond these big earthen dams to the north and what's happening is continually this dam is sinking sliding into the pipeline right-of-way they see cracks the dam's not high enough to retain the water behind it so US Silica comes back and is continually repairing this dam over and over again to try and stabilize it. 7:07 Rhett: And in the aftermath, they basically admitted to making at least three attempts to stabilize the retention pond between 2020 and the time of the failure. So again, I just gave you going all the way to 2003, but even between 2020 and 2023 when this failure happened, they had been at that site at least three times trying to stabilize it. Christopher: Why would they make an above grade retention pond? I'm just curious about that, right? 'Cause the water table can't be that high in Mississippi? Rhett: So, I don't know the details about why. Christopher: That's just interesting, right? 'Cause you think about like, why is there an above grade retention pond? Rhett: I mean, it's a chemical, US Silica, so it could be that it's some type of a chemical thing and maybe they have like a liner that they put above ground. Point is, is that for some reason or another, this pipeline is built above or this, this retention pond is built above natural grade. Christopher: And it's leaking. 8:00 Rhett: Well, the, the, the pond's not leaking but the retention pond is falling. Christopher: Yes, it's- Rhett: Right. The retention retaining wall, the structure of the dam is falling. Now what's really interesting about this and this is where the story gets interesting is that Gulf South was aware of the work going on. They had actually identified from aerial patrols. That's one of the few times that we've seen this. The aerial patrols performed in July and September 2022 actually noted earth equipment on the right-of-way. Gulf South does what any prudent operator would do and sent foot patrols out to go inspect the right-of-way. Christopher: Yep. Rhett: I mean, these are all positives for the operator. Christopher: Yep, standard practice. Rhett: But unfortunately, when the foot patrols came back, they said, "Oh, they're not operating directly on our right-of-way. They're working on the dam adjacent to our right-of-way.” 8:52 Rhett: Which I think one of the major questions from this failure that happens is how close does work have to be happening to the right-of-way in order to be a concern to the pipeline, right? Yeah. Christopher: So, but we've kind of talked about this a little bit before in previous episodes, right? It's like when you stand up these programs, like, what is the training effort to have boots on the ground to identify what geohazards would be, right? Rhett: And awareness, it's a challenge, right? I mean, I think, you know, I mean, again, we could, but I think an operator that it- that isn't actively looking for this and sends field personnel, which is gonna be a normal typical response, the field personnel are unlikely to say, “Oh, yeah, that, that dam they keep working on that's falling is falling into the right-of-way, and the right-of-way is sliding with the dam near the toe of it, right? That, that's, it's unlikely. I mean, Chris, I don't think if I walked out there and I know a decent amount, I might necessarily draw that conclusion, you know, as an engineer, I think it would definitely take a, somebody with a specialty for that. Christopher: It kind of ties us back to, and this is just an open-ended question, it ties us back to our previous episode where it was on the Macando one, as it's, you know, did they go out to investigate or did they go out to confirm? Rhett: And that's the question I read, right? Like the investigations we're trying to answer is this impacting the pipeline, right? And, and, Christopher: Or are we confirming they're not on our right-of-way? Rhett:Exactly, right. 10:11 Rhett: And, and, and that, again, it's worthwhile. It's an important lesson in this failure that adjacent to the right-of-way is something we really have to be working on. So now that I've set the stage for you, you kind of know what happens, right? So, in 2023, what happens is this landslide that was built or this manmade landslide, this earthen dam has been continually falling and as it's falling, it's pushing into the pipeline. Christopher: It’s moving the earth. Rhett: They keep going back and restabilizing it. Christopher: Keep moving the earth. Rhett: And they keep pushing the pipeline down. Christopher: Keep moving the earth. Rhett: Well, you can only push the pipeline so far out of alignment before it ruptures. And that's what happened in 2023. But there's still more information. So now that we've had the failure, I want to talk about what we gleaned after because there are lessons after the fact as well, right? The actual failure. I think if, if I were to stop and ask the audience right now, I told you that we have an earthen dam that has been sliding into the right-of-way and it's pushing the pipeline out of the right-of-way. Where do you expect the failure to happen? Most people are going to say probably within the portion of the pipeline that was pushed out of a way. Christopher: Ehhhh. Yeah, but if you're in integrity, you can kind of think of it a little bit differently, right? Rhett: Maybe. Christopher: Yeah. Rhett: But I don't even, even I was a bit surprised. Christopher: If there was a girth weld defect there where it was getting pressed, maybe sure, because now you have like a local concentration of stress. But if the girth welds are intact, then where you're pushing may not be where you're getting the most strain, right? 11:30 Rhett: So, in this case, the failure occurs 160-foot west of the actual landslide. Now it's interesting this this isn't altogether different than what we saw when Marathon had its Edwardsville failure. We actually saw the failure happen near the toe, not near the greatest point of movement in the pipeline. But 160 feet away is pretty far away. Um, and what's cool here, I noted is that the, the PHMSA report actually captured the movement of the pipeline. And again, so I work in geohazards, people are always calling me to tell me how much pipelines moved. They're like, we cut the pipeline, and it moved six inches. It moved eight inches it moved 12. Look, Christopher: That's a lot. Rhett: Well, to be honest, but I pretty routinely see vertical or horizontal movement of six, eight or 12 inches. And I tell integrity engineers that like, look, you can't necessarily infer a whole lot because six, eight, or 12 inches might not matter, depending on how big the excavation was. 12:21 Rhett: But what's a surprise here is the axial separation was 14 and a half inches. So, I want to put that in perspective of the audience. The pipe, the two pipe ends when, when they failed, pulled apart 14 and a half inches. I have never seen 14 and a half inches of axial separation. That is the largest that I've ever seen in the axial direction. It speaks to the fact that as a result of this pipeline being continually pushed out of the way, even though it was 160 foot down the right-of-way, it was imparting a significant amount of axial strain at this particular location. 12:56 Rhett: And of course, you know, again, geohazards rarely ever produce a failure absent ofsomething else going on. They're almost always follow the perfect Swiss cheese model of multiple things have to line up. We saw it in Hillsboro; we're going to see it here. In this particular location, they found a 26, it's a 20-inch pipe, they found a 26-inch-long incomplete well penetration. So, think, you know, missing root pass type thing, over 41% of the circumference. That's a pretty egregious girth weld defect by almost all standards. Christopher: Yeah. Rhett: But I also want to say it's typically what we see. Like a lot of times when there's a geohazard failure, it doesn't happen in a perfect weld. It almost always happens with a weld that as something else that looks like this. This is a vintage construction, Monday, Friday, girth weld, the guy— Christopher: Yeah, but it's not just vintage. Yeah, it's not just vintage. I mean, this is 1939. 13:55 Rhett: No, this is the repaired section was built in 1952. Christopher: Oh, that's right. Rhett: Right, so it's still vintage. Christopher: But even still. Rhett: I mean, it's old, but that's a lot of our pipeline infrastructure exists, particularly the gas world exists from that time. Christopher: Yep. Rhett: So, some other fun facts that are worth talking about before we get into the lessons learned. They ran an MFLA tool in 2021. Christopher: So that means they found it, obviously. 14:18 Rhett: You got it wrong, no. There are- Christopher: You’re kidding me Rhett: -no reportable, even after the fact, no reportable girth weld anomalies here. And audience, I'm gonna tell you to hang on 'cause that's gonna be something that Chris and I are gonna discuss, but this is, for me, this was not a surprise. Like the PHMSA report specifically noted that there were no reportable features. But my first response says, yep, that does not shock me from an FLA. Now, maybe till some of our, our listeners out there, maybe you are surprised. Christopher: Yeah, Rhett: But I want to get in. So, I want to go back 26 l- inches long, 26 inches, 40% at the deepest point, but no MFLA reportable feature. Second important fact, and this is probably one of the big ones is the, a follow up bending strain analysis was performed using data that was gathered in ‘17 and ‘21, right? So again, remember going back, I told you, you can see all the way back to 2003, this puppy was moving. US Silica admitted to doing a lot of work between ‘20 and ‘23. We now have a glimpse between ‘17 and ‘21. And the bending strain data identified a clear as day horizontal bending strain feature with a maximum bending strain of .207% AND three feet of movement between the ‘17 and ‘21 data, right? Now, I gotta have to pause so some of you might be thinking hey, Rhett, .2 percent's not really that high. I know a thing or two and three feet's not necessarily freakout stage, 15:45 Rhett: But that was between ‘17 and ‘21. This failure doesn't happen until ‘23, so we know we had additional movement that occurred after ‘21 and up to ‘23, right? And so again, bending strain was performed after the fact, not prior to the failure, but the bending strain analysis would have, would have picked this up with no problem whatsoever. Had it been performed before and definitely put the operator – Christopher: On alert. Rhett: - on alert, right? And so again, so that is the data around the story. When we come back Chris, we're gonna take a break here in a moment. There's some really, I think, very clear blind spots here that a lot of operators might find themselves susceptible to. To our audience, I want to say hang on, because I want to boil those down to really clear takeaways that you can think about in your IMP plan, or what maybe how you're approaching geohazards integrity that we should all learn from. So, hang on, we'll be right back. 16:45 Rhett: All right, welcome back as we get into the lessons learned on our Failure File series dealing with the 2023 Geohazard failure in the Gulf South public line in Jackson, Mississippi. So, and that's where I want to start. Christopher: You know what I really like about the analogy of blind spot that you used? Because whenever you're checking for blind spots, a lot of times you're looking at your review mirrors. Rhett: Okay. Christopher: And we get to look back and learn from these guys. It's a really good analogy. So, what can we learn? Rhett: Maybe, you know, I'm teaching my son how to drive. You need to adjust your own mirrors to make sure that you can see clearly, right? So, a lot of times I get in the car after my wife's driven and the mirrors are all out of place. Christopher: Mirrors are all off. Rhett: All completely off. Thank God you have a little setting now for one and two. Christopher: You know what? This is unprepped. I wonder if we can do this. I wonder if I typically use three mirrors, right? For my rear-view mirrors. I wonder if we can identify three key takeaways out of this- Rhett: Easily. Christopher: - that we can use as mirrors. Christopher: Let's do this for though. Rhett: I have four though. Well, look, Christopher: I guess is that looking forward? I don't know. Rhett: No, but- Christpher: The co-pilot maybe? Maybe the co-pilot has a different advantage. Rhett: Sometimes the rear-view mirror has that Both: Little mirror. Christopher: Yeah. Rhett: In the corner, right? And my rear-view mirror doesn't even have a rear-view mirror. It's a rear-view camera now, which I really prefer. Christopher: You're just showing off, you’re just showing off. Rhett: Flexing a little bit for sure. Christopher: We're going to go with the fourth, which is that little mirror on the driver's side. Rhett: The first one. look, we don't need to belabor it. Christopher: Let's go. Rhett: I said Jackson, Mississippi. To our audience out there, nobody associates Jackson, Mississippi with landslides. Christopher: Yeah, Rhett: Man-made or otherwise. 18:05 Rhett: And this really speaks to what the Dirt Merchant and I have talked about a lot, which is geohazards don't occur in places where we often expect them. Right. So, if you're like, oh, deep Gulf South, don't need to worry about geohazards. You're going to miss this type of thing. Christopher: So that's probably because you're not looking for it, right? Again, you're confirming, not investigating. When you send your ILI out or your due assessment, it's, you're confirming, I don't have a problem, not, hmm, Rhett: Yeah. Christopher: Am I sure I don't have a problem? Rhett: Second one for me is again, uh, I'm going to, I'm going to quote the Dirt Merchant is in hindsight, he says, we often have the information we need to identify geohazard failures before they happen. And I actually think that's a great example. I think in this case, Gulf South had all the information, but probably not the knowledge and experience to assemble that information and recognize the warning signs. Christopher: Hold on. Let's be careful here. They had all the signs to what? 18:59 Rhett: But where they could have taken action. They could have identified. Christopher: Give me an example of an action, though. Rhett: Um, so Christopher: To hire an expert to come and do an- Rhett: Well yeah. Christopher: - investigation. Right. So, I mean, because if you think about it. Again, they recognize there was something going on. They sent boots out on the ground. Rhett: Got it. And that's where it stops. Christopher: Okay. Rhett: Right. And that's a challenge. And I'll also say, you know, again, I think a lot of operators that would have done the same thing, which is work near the right-of-way isn’t important, but when you talk about geohazards, near the right-of-way can very much impact on the right way. So, I'll give you a great example of this, Chris. 19:38 Rhett: We were doing a review for a bending strain feature on an operator a couple of weeks ago. Very clear horizontal pattern. He said, "Well, let me go bring up the LiDAR data." And I was like, "Let's do it." And he brings up the LiDAR data and he's like, "There's nothing here." And I said, "Stop, stop, stop." And he had a huge map LiDAR feature that was to the left of the right-of-way and to the right of the right-of-way, and then the right-of-way was perfectly clear. And I was like, “Was this right-of-way regraded?” And he's like, “Yeah, they did some work back when-“ I said, “The regrading work made the LiDAR totally blind.” And you could see like a path. It cut right through the middle of what was a very clear feature that would intersect and went through the right-of-way, but wasn't quote unquote on the right-of-way. 20:19 Christopher: Yeah, I think, I think maybe what, and again, not saying that this is what happened, this is not a fact, but again, it just goes back to, again, separate from this issue, you know, typically what are boots on the ground going to look for, right? It was a third-party damage. If it's third-party damage, then I'm, I'm, I'm hyper tuned to you're on my right-of-way, Rhett: Right. Christopher: Whereas if I'm more tuned to, I'm here for any threat, right? I'm coming to investigate what is happening on my pipeline, which often is more of that integrity approach. It's more like, “Ooh, this is near enough. Do I have enough information to discount that a earth movement threat is happening.” Rhett: You know, that's I mean, Chris, that's great, because what we don't know here in the PHMSA report doesn't allude to is what, what were the people that went to go investigate the right-of-way charged with, right? But I imagine that whatever task they were given, probably framed up what they were going out to look for, right? And so, um, I that's, that's, that's just a pause point for operators is that- Christopher: Yep. It’s just something to think about. Rhett: Yeah Christopher: Right. When you send boots out on the ground for the threat, are you being maybe too hyper focused or they go again and back to Macando, are they to confirm something related to the right-of-way or are they there to investigate the right-of-way? Rhett: Yeah and really think about the fact that geohazards can be adjacent. There's a second important characteristic here, which we talk about. And I think operators are at various aspects, which is both bending strain and LiDAR are different. So, like in this case, and I already kind of alluded to it, 21:43 Rhett: LiDAR might not be effective in this case, Christopher; Sure. Rhett: Because US Silica had been continually repairing the right-of-way. So, every time LiDAR sees it, it's going to see the repairs likely not catch the incident whenever it was in a moved state, which is going to make it very difficult for LiDAR, right? So, I want to speak to a lot of the operators that are using LIDAR out there. You need to be aware that um, let's say remediation efforts can be masking the influences of geohazards on or near your right-of-way if somebody's repairing them. Similarly, I'm going to say bending strain can have its own blind spots when it talks about axially oriented landslides, which is not the case here. So, it's a good point from a lessons learned standpoint to stop and think about, “Hey, a comprehensive geohazard program will often merge both of those attributes together, your surface -based assessments with— Christopher: But there can always be blind spots. Rhett: Your bending strain assessments. Christoher: Right, 'cause if the LiDAR doesn't capture the current geotechnical environment at its worst, then now you're blind to what actually happened to the pipeline, or could have happened to the pipeline. And if you're doing bending strain, maybe you're a little bit blind to— Rhett: Axial oriented, Christopher: Yeah challenges, Rhett: But you but those will pick picked up very well by LiDAR unless it's, you know, again, be remediated and we've seen that. So, Chris, I gave you my my two my two lessons. Christopher: Sure. Rhett: Well, do you want to give the third one? Christopher: Um, Rhett: Are you just going to make it up as we go? Christopher: No, I think one of the topics that we wanted to talk about before we, we, we got to come as some of your blind spots was it's you said there was an MFL inspection on this this line. Rhett: Yes. Christopher: And if you heard, I was like, obviously it founded, right? But that's obviously, you know, sometimes being a little bit facetious there and knowing that we always need to ask more questions. Rhett: Yeah. Christopher: And so, some people would say, well, it's, it's pretty easy to rationalize why MFL should have seen this. And it's a pretty gross feature, right? It's- Rhett: For me, it was the opposite. It was pretty easy to rationalize why MFL DIDN’T see it. Christopher: Which just depends, right, which we're getting to, right? We talk about sales jargon of ultra-high resolution, right? And this is a 26-inch-long feature. That's 48% deep. I mean, this is a gross feature. Why would from a from an orientation perspective, right? Have an axial MFL with a circumferentially oriented flaw. Rhett: It's perfect. 24:04 Christopher: It should have seen it, right? Rhett: And but I think that again, you have to appreciate um, and, and I think there's, there's two things that contribute to this, Chris. Let's talk about both. Cause we were there. I had seen MFLA data be used to quantify or risk rank girth welds for operators where they knew they had poor issues. I had seen girth welds, almost every operatives running on MFL has probably had girth weld related features called out. I've seen it either a corrosion, uh, crossing a girth weld or even other types of features. I've seen construction features called out. I think the challenge there is a, most people might not realize it's often done on a best effort basis, right? Like- Christopher: Yes. Rhett: If the tool passes over the girth weld the right way. And if the analyst is looking for it, they can probably, and I'm using all sorts of qualitative words here, Christopher: Yes. Rhett: They can probably identify it for you on a best effort basis. And a lot of times the vendors are helpful at doing that. Christopher: Yeah, and so all of those things are very anecdotal. So as an operator, let's say you found yourself in this situation where let's go back in time, right? And say, hey, we found that there is something going on on our right-of-way. We send a boots on the ground, boots are gonna confirm, yes, there's active changes there. And the integrity team gets flagged and said, "Hey, do we have any data that we can go look at?" And let's assume, yes, there is some recent feature data that we can go and look at, and you pull up the ILI data and you're like, "Should I expect to see this?" 25:36 Christopher: What I would say is don't just look at the data. I think it's important to call your ILI service provider where you're looking at the dataset and ask them if their ILI specification is qualified for finding girth weld type flaws. Let's assume you were gonna look at girth welds, right? You were gonna assume that, hey, assume I didn't buckle or anything like that, that I'm looking at girth welds for failure. Ask them straight up. This is what you say is this. Is your ILI specification qualified for finding girth weld related flaws? 26:08 Rhett: But let's talk about why that's, and I want the audience to appreciate it, no matter what answer you get, I want them to appreciate why it's such a challenge, right? Christopher: Yeah, we'll get there. I think what I to get to first was it's, it's because what happens is because you said first best effort basis. Rhett: Yes. Christopher: And so, what you'll do is you go to the performance spec and then there'll be POD, POI and sizing, right? Rhett: Yep. Christopher: And they'll say, hey, if the feature is at this long, this wide, this deep, we should see it. But then it depends what type of feature. So, I might get a squiggly line, I might detect it, but I won't be able to identify it, right? So, I won't be able to identify a girth weld type flaw that you're interested in. Rhett: But the problem is that the girth weld, you're always going to get the squiggly lines, right? That's one of the most challenging things, right? So, I want to speak to the audience. You already have a, I'm going to say a magnetic disruption at the girth weld because you usually have extra material deposited in the form of a weld cap, right? So, you already have a signal that's going to be, I'm going to say disruptive from comparative to a base pipe signal adjacent to it. Christopher: Yeah. And so, let's frame that, right? So, we're used to people saying, it depends and guys qualifiers, lots of qualifiers here, it depends on the ILI system that you are looking at and considering, right? Rhett: Yep. Christopher: Because not every ILI system is qualified for the same type of flaws or defects. Rhett: Yes. Christopher: Not every ILI system has the same performance specification or technological basis or measurement principle. So, all those qualifiers aside. You know, it's not uncommon for people to say, Hey, usually we're “blind”. Actually, we can even say that two inches on either side of a girth weld. Rhett: Right. And again, that's because of disruptions to the sensor carriers. It goes over the girth weld because if you have a girth weld, Christopher: Think of it as a speed bump, right? Sensor head hits it. Now you have movement. There's interference. How well can you classify that? Rhett: Absolutely. Christopher: Well, what if there's lack of penetration? Maybe there is no weld cap. 27:54 Christopher: Maybe, maybe the ride is smooth. Rhett: Which usually on the inside surface, but yes, I mean, Christopher: Which is what I'm getting to, Rhett: Oh, and so now the sensor head is right in on the ID. There's no weld cap for the speed bump effect. Rhett: So, what you're talking about now is comparative If we were to look at all the girth welds in the line, we're looking for outliers. We're looking for signal outliers Got it. Christopher: Right. And so, what I'm getting to guys is it's The whole so what of that was to say start with understanding how the ILI system was qualified and, “Is there a performance spec tied to it?” Because if you do that, then it sets an expectation for what you may or may not be able to get out of the data. So that then when you look at the data, you always need to try to find something to compare to and preferably do it on a macro scale, right? Don't do it on a micro scale to like you said earlier, hey, where would you have expected the damage to have happened where the bending was happening or away from it? 28:42 Christopher: Right? So, you want to you want to try to find a like or similar type comparison before you feel like you can draw any conclusions. Rhett: I like the “like or similar” and that's one area again where I would encourage. You said it right. I don't want people to hear you can't or shouldn't use MFLA. Christopher: Yes. Rhett: It's when you want to use it this way, you A, shouldn't just go in with the expectation that it can do it. But B, if you do talk to your vendor, they may be able to do some analysis that could be very helpful in potentially identifying these types of features, but it's just not your standard. Hey, let's look for metal loss. Christopher: It's the idea of what we said earlier. Don't look at the ILI data to confirm one of your personal assumptions. Right. You're going to look at the MFL data and be and have an investigation. “Why do I think this could work? If it did work, what would it look like? What would I do with it if I did identify something?” Rhett: So last, Christopher: So that was point three. Rhett: It was great. But I got to throw the fourth bonus point because I remembered, I wanted to go back to the vending strain, right? Because I wanted to talk about why it was significant. The girth weld that failed, if you look at the plots that are in the PHMSA report, it actually had no bending strain. When I say no, I mean like zip, right? And again, that was in 2021, it's possible that it had bending strain imparted by 2023. But given the location 160 feet away, I'm going to say that I doubt it did either. What that speaks to though is that, so somebody might challenge me and say, “Well, Rhett, how would you have used bending strain?” And I’m going to reiterate. When we went back and did Hillsborough, let's talk about a few things. 30:03 Rhett: Number one, you have a horizontal feature. That's red flag number one. And number two, you have change, right? So those are red flags, again, that I always speak to operators that speak about orientation, orientation, orientation, and stability. When you have a feature that is acting oblique or perpendicular to the pipeline, and you know it's not stable, then your flags go through the roof, regardless of the magnitude of the strain, right? So. Had I seen a feature like this, this would have been what we typically call a priority three to an operator. I would have put them on it and been like, you need to do something immediately. Had we had the opportunity— Christopher: You need to call an SME to go out there. Rhett: Exactly. Christopher: And maybe you'd answered it, but I want to answer the question anyways. I want to ask the question to you anyways. So, you said there was a .207 total strain--horizontal strain feature on this, and we kind of alluded to this earlier. Did it fail where traditionally do you think people would have expected it failed? I mean we kind of started talking about that a little bit it failed four joints away roughly. Rhett: Yeah, Christopher: Right and so when you get it, walk us through that process of saying… walk us through the process of, “Hey, we went we went to you guys or whoever you used. You called the bending strain feature. It's horizontal, roughly .2%.” Rhett: Right. Christopher: You're saying it would be a point… priority three because of all the features that are happening. Would you have sent them to go look at the area four joints away or 160 feet away? Rhett: Man, Chris, that's a really, and that's a fair question. I don't know. So, the extent of the bending strain feature actually did not extend to cover that girth weld. So, I would say that there's a decent chance that the- that any by operator may have gone out there. Only, only looked at the portions of the pipeline that were perhaps displaced. And if they excavated may have only even reinforced those girth welds. That's definitely a high chance. I think the key takeaway there is recognizing that the influence of geohazards on axial strain, particularly when it's oblique like this, can be large and extend definitely, obviously, 100 feet or more into the soil beyond what we might call the “edges” or the “flanks” Christopher: of the landslide. And again, under normal circumstances, if that's a healthy girth weld, there's no issues. There really are not. That would not have been the place where it would’ve failed. But it's also being aware of when you don't have healthy girth welds, right? So that becomes another threat that operators have to be aware of, which is why, you know, one question I have to ask operators is, do you have a history of failures in girth welds in this line? So, yeah. Christopher: So, to catch it, it almost sounds like you would have had to have investigated if there was a weak girth weld in this vicinity. Rhett: Man, it's a challenge. Christopher: How we do that is different. Rhett: I am not gonna say that in hindsight, I would have said, you need to go look at this girth weld, you know, three joints away. I don't I don't think that's fair. I don't think that's fair to go south. But it is it is something to be aware of when we look at them. Rhett: So alright, hey, to our audience, you know, I love geohazards. I think this is again, I would really encourage you that the PHMSA report is short and to the point. It's actually a really good report with lots of lessons learned and images are really good. So hopefully you enjoyed this episode of the Failure Files. Again, so relevant for where we are as an industry with regards to the threat of geohazards. And thanks for joining us. And we'll see you again shortly. 33:43 Rhett: This episode executively produced by Sarah Etier and written by myself, Rhett Dodson. Source material can be found in the form of the PHMSA Failure Reports. And that's it.