Speaker A [00:00:13]: Doctor Joanna Curran's sediment transport career has unfolded on several scales. Her early work as a grad student and professor involved detailed particle and turbulent scale measurements in experimental flumes, and she's probably best known for this, in particular for her association with the Wilcox and Crow sediment transport function. But since then she's worked as a numerical modeler and river engineer for several consulting firms, and most recently as a field and project scientist for the Seattle district of the Corps of Engineers. And it's this career trajectory that has given her experience in experimental, numerical, and field investigations, which gives her a multiscale perspective on fundamental segment movement questions. And Doctor Curran is also just the kind of scientist who has given a lot of thought to these central questions about what is actually actually driving and limiting sediment transport and morphological change in gravel bed rivers. Our conversation covered log jam porosity hysteresis, hiding the impact of clusters on threshold transport, in addition to Wilcox and Crow and her actual dissertation work on step pool systems. And you'll want to make sure to stick around to the end of the conversation where we talk about the effects of antecedent low flows and interannual flow impacts on sediment transport. If you're not up on that literature, it is potentially paradigm shifting. I'm Stanford Gibson, the sediment transport specialist at HEC. And this week on the RSM River Mechanics podcast, a conversation with Joanna Curran. Joanna Curran, welcome to the podcast. Speaker B [00:01:42]: Thank you. Very happy to be here. Speaker A [00:01:44]: So you're kind of sneaky famous in our community. Not everyone knows that you're the crow and the Wilcox and Crow sediment transport function. So why don't we just start out with how did you get involved with what would become arguably the most famous modern sediment transport function? Speaker B [00:01:59]: I'm very famous in a very small community. It started in grad school. I got involved as a grad student helper in the flume room because I wanted to see how flumes worked. And at that point, there were two of us in Peter's group. Peter Wilcock. Speaker A [00:02:14]: And that's a Johns Hopkins. Speaker B [00:02:15]: Yeah. So at Hopkins it was myself and Steve Kenworthy. Speaker A [00:02:18]: Okay. Speaker B [00:02:19]: And an interesting note on that is, you know, we were the helpers on this project or the grad students being paid off this project, but neither one of us used it for our dissertations. Speaker A [00:02:29]: Yeah, right. Interesting. Speaker B [00:02:31]: We each wanted to do our own thing. And you find that more in geomorph that those of us who go back for PhDs, a lot of times we do it the stubborn way where we're going to find our topic that interests us. And that's what I'm going to work on. And I don't care if it takes a little bit longer to get a PhD, I'm going to do what I want to do. And it serves you well in the long run, but you will spend a little bit longer in grad school. Yeah, so we did those. The work, it was funded by the Department of Justice. So if you're ever reading the papers. Speaker A [00:02:58]: Oh, I did not know that. Okay, we have to talk about that. Why was it funded by the Department of Justice? Speaker B [00:03:03]: So if you ever wonder why they're called the DOJ runs or j seven, j six, J 14. Speaker A [00:03:08]: I did wonder. Speaker B [00:03:09]: It's for justice. Speaker A [00:03:12]: That's part of why we do these podcasts. That's amazing. So what was the motivation? Speaker B [00:03:17]: It was the Snake river adjudication. Speaker A [00:03:18]: Oh, was it really? Speaker B [00:03:19]: And Peter was being paid as an expert witness. And I think through the. I think through the US forest Service, but I'm not exactly positive. They needed to know how much water needed to remain in the river system for the river to maintain a healthy process regime. And then as you do runs, you know, the numbers were whatever the sand content was of the bulk sediment. And if you don't think about it ahead of time, you start nicknaming your flume runs. Speaker A [00:03:45]: Yeah, right. Speaker B [00:03:46]: And then that name sticks because it's the only way you can talk about. Speaker A [00:03:49]: It after a while within your community. And then you go to the paper and it's a little indecipherable. Speaker B [00:03:54]: Right. So they're the DOJ runs. Speaker A [00:03:56]: This is Peter Wilcox, who is supervising this, who is one of the big. Speaker B [00:04:01]: Names in our field, and he's, I think, still at Utah state. Speaker A [00:04:03]: Yeah, that's right. So before we get too far into the details of this transport function, maybe let's just start out with what is a sediment transport function? Speaker B [00:04:12]: A sediment transport function relates the amount of sediment moving through a system to the amount of flow in the system given the stream overall parameters that are put into it. So, for example, some of them use just a D 50 parameter, so they're basing it just on the D 50 and the flow rate. Some use steady, uniform shear stress, so they're basing it on basically slope and depth. But it's a way of trying to get an idea of how much is going to move through a system over a given flow. Speaker A [00:04:46]: So we have parameters of the sediment and parameters of the flow, and we're trying to turn that into this number of how much sediment flux is happening. Speaker B [00:04:57]: It's basically an attempt to parameterize the physical system. In a way that we can then work with it and do something with it. We can't go out and measure all the sediment moving through. It's just not possible. We're trying to get closer. Everybody wants to do direct measurements. It's hard as heck. It's a way of taking a few measurements. And then applying it to the theoretically solid underpinnings from a model. And being able to extrapolate with that. Speaker A [00:05:21]: And then downstream of that in numerical models. That's kind of how we actually compute how sediment moves their system. Speaker B [00:05:28]: Yeah. And it's the only way. If you want to really know what might happen during a flood. You're not going to be out there measuring during a massive flood. Speaker A [00:05:34]: That's right. Speaker B [00:05:35]: It's too dangerous. But that's one reason we build them in the flume. We can simulate the larger events. So that we have an idea of what's going to happen at them. And then build a model then. Yeah. Take it out. Speaker A [00:05:46]: So I'm not sure everyone knows. The painstaking experimental work. That went into this equation. So why don't you describe the bed of many colors. And how you collected the data that made this experiment so special? Speaker B [00:05:58]: Oh, yeah. So bed of many colors actually already existed. Thank goodness it was built by Brian McCardell. And his. The people that were there right before me. I only crossed over with Brian for about a year. He's in Switzerland now. Speaker A [00:06:11]: Okay. Speaker B [00:06:12]: But the bed of many colors was built by. They went out and they got sediment. Some of it from streams in Minneapolis, I believe around Minnesota somewhere. And some of it they bought. Because you can buy aquarium rocks that are pre colored. But you have to sieve everything apart. So all the fractions got sieved apart to make sure they were right. So even if you buy in a certain grain sized distribution. You don't know for sure it's that grain size distribution. So you sieve it all apart. Then you put it in a cement mixer. You throw in acetone and dye. You beat the cement mixer while it's running. Otherwise, all the rocks will stick together. And you endure lots of complaints from the other people in the building. Because we were right below psychology. Who wanted to know what the heck we were doing. They were beating the cement mixer. And then once it's all out, then you have to dry it. And then you have to re sieve it to make sure it's still all the same. And so once all that's done, you've got 16 different colors. So this is, I think they said it took about a year to make it. Speaker A [00:07:06]: Oh, wow. Just to make the. Speaker B [00:07:08]: Just to get the sediment done. Speaker A [00:07:09]: And so each grain class, like a certain band of gravel is one color. Speaker B [00:07:14]: Yes. Speaker A [00:07:14]: And then another band of gravel is another color. Speaker B [00:07:17]: Right. We did 16. There were 16 colors. Speaker A [00:07:19]: Okay. Speaker B [00:07:19]: And I still see them in my head. And some of them, they commented, were not the greatest. Like, if you're ever going to do it, don't use black as a color because it doesn't show up. Well, nice. Magenta and pink should not really be next to each other. Speaker A [00:07:34]: Okay. Speaker B [00:07:34]: They had to reuse some. So we had big green and little green. Speaker A [00:07:37]: Oh, wow. In my dissertation, I did some colored sand analysis, too, and I had some flume runs where I tried to run lots of colors and then I couldn't distinguish them. Like with a camera. Speaker B [00:07:47]: Nope. Speaker A [00:07:48]: And so then I went to just the primary and secondary colors, and orange was still hard. So there is. You have to be a little careful. Speaker B [00:07:55]: Yeah. And we did it all by eye. Speaker A [00:07:57]: Oh, wow. Speaker B [00:07:58]: Because, yeah. No matter how good the computer gets or the camera, you're right, it doesn't see color the same way a human eye does. And it's not going to distinguish. And also, one of the other benefits of using doing it all by eye was you never have to repaint it because you can still see what the size is even when it's only got a fleck of paint on it. Speaker A [00:08:16]: Yeah. Like big green and little green is you can do by eye, but the computer wouldn't be able to. Speaker B [00:08:20]: Right. It wouldn't know which one's which. Speaker A [00:08:22]: Yeah. Speaker B [00:08:22]: So that goes into the one. One of the things that makes it all so special is we could really do a grain size distribution of the surface. Speaker A [00:08:29]: That's amazing. Speaker B [00:08:30]: And you can only do that if you have different colors from a photo point of view. Speaker A [00:08:37]: Yeah. Speaker B [00:08:38]: You either have to pick it up or have colored rocks. Speaker A [00:08:41]: There's just no other way because otherwise, if you just try to use the actual surface image, part of it's going to be buried or it's going to be in a different orientation and you're going to get the diameters wrong. Speaker B [00:08:52]: Yeah. You'll see like a tiny bit of a big red, maybe. Speaker A [00:08:55]: Yeah. Right. Speaker B [00:08:55]: And unless you know exactly what that size is, you don't realize that that's actually exposed on the surface. And so it's kind of like a siren song. Everybody wants to be able to just do a surface count from a picture. It would be great if it worked. Speaker A [00:09:10]: The detail at which you were looking at the, you know, the ratio of the shear stresses is really important. Speaker B [00:09:15]: Yeah. Speaker A [00:09:16]: So you were talking a lot about the surface, and you're being very particular about, like, what's on the surface. And that's because wilcox and crowd is a surface transport function. What is that? And why is that so important? Speaker B [00:09:28]: So it goes back to the theory of where do you think the sediment comes from that's in transport. Is it coming from the bulk bed, or is it coming more from the surface? And bulk bed has been shown to be, when the river reaches full equilibrium transport, everything's transporting to the capacity at which it can be transported. Then you're what you're collecting in your basket. If you're collecting direct samples, what's being in transport should look a lot like the subsurface. It will have gotten to that point where everything's fully mixed. Rivers don't usually get there. They do, but we're probably not there to see it. So it's more likely that the surface is adjusting to moderate the amount of transport and that the transport is limited by what's on the surface. So you've got now an interaction not just between the subsurface bed, but also between the surface and the flow. And so the idea of using a surface based transport equation is to account for the fact that the transport is moderated by that surface. So you want to look at the surface for where it comes from. And it got more popular because it seemed to work better. Speaker A [00:10:28]: So let's take a step backwards because we're talking about, like, the sample, like, the grain classes of the surface and the grain classes of the subsurface. And we're just assuming they're different. But maybe not everyone's familiar with that. So, especially in the gravel bed rivers and the snake that you're dealing with. Why are the differences between the surface and the subsurface? Speaker B [00:10:47]: And mostly because armoring. But I would almost now lump armoring in with everything that's a surface structure. Yeah, because sand bed rivers, it's been recognized for a long time. You get bed forms, they're easier to see. Ripples, dunes, et cetera. That happens in the gravel bed river. We just call it either an armor armor with clusters with transverse ribs. And that surface structure is a result, usually of lower flows or how the flood results, the antecedent history of the flows. What happens is the sand is thought to winnow into the bed surface. So if you kick at a gravel bed surface, you usually find a lot more sand in the subsurface and a lot of times, if you measure transport in gravel bed surface, you'll look down and you'll see mostly gravel on the surface. You'll still catch a lot of sand because it's coming out through little bits. So a lot of what we did in that, determined in a lot of those experiments was that when there's a very low sand content, it's mostly subdivid, you know, trapped under things. It's right behind the big gravels. The sand moves when the gravel moves. And as you start to go from very low sand content, like say five, 6% up towards maybe 1015 percent, maybe even higher, but you start adjusting so that you have more sand on the surface. So it's moving a little bit more independently of those big gravels. First, it only moves with the gravel. And so the sand really has the same incipient shear stress as the gravel. As it gets to be more, it's starting to move at its own shear stress. And then as you get enough sand, you've almost really transformed the surface or the active layer into more like a matrix. And the gravel now is moving at the shear stress of the sand because the sand is able to move and it almost like smooths out the surface like a greased gravel is what we used to call it, because the gravels are now moving with sand. Speaker A [00:12:28]: I've seen that in the math, but I've never heard that described in words. That's really interesting. And so you're looking at the surface because what's on the surface is really what's transporting. Speaker B [00:12:37]: Right. And so as you get more sand on there and you're moving everything now at the reference shear stress of sand, that's the lower one. And that's why the whole bed load starts kind of moving. And if you do that long enough, you know, let's say you dump 100% sand into a gravel bed river, you're going to, for, at least for while that sand is there, temporarily lower the slope of the river because you've now evacuated enough of the bed load. So it was really counterintuitive that you can get higher bed load transport with more sediment going in, but the same flow rate. Speaker A [00:13:07]: Oh, interesting. Speaker B [00:13:07]: So it doesn't track one to one. Speaker A [00:13:10]: Yeah. Speaker B [00:13:10]: And that was one of, you know, it's an old equation now. So that was one of the first findings of like, well, this is why it's not tracking one to one because of the changing shear stresses with the changing grain size distribution of the surface. Speaker A [00:13:21]: So you said it's been a while, but I do think of Wilcox and Crow and kind of the ones that are in that sphere, like Parker. And I think of them as modern transport functions because they intrinsically take into account grain interaction in the, what I'm going to call kind of the boomer equations, the ones that were born a while ago, they were all kind of single grain class equations. And, like, if we use them in a model, we kind of, like, apply them to each grain class individually. But Wilcox and Crow takes these interactions implicitly. And so there's two grain class interactions. The first one I'd love to hear you talk about is you have this fs in the equation where the gravel. The sand impacts the gravel transport. You've touched on that already. But how does the fraction of the sand impact the gravel transport? Speaker B [00:14:13]: Mostly just by changing that overall, we do it through the shear stress. We don't know. Okay, so here's my little rant on shear stress. Speaker A [00:14:19]: Okay. Speaker B [00:14:20]: Grains don't move because of shear stress. And I think that's pretty widely understood that shear stress is not why grains move. If you dig into it, it's really turbulence in the interactions with the flow. Speaker A [00:14:32]: True. Speaker B [00:14:32]: Problem is, we haven't met, and when I say we, I mean we as giant. We haven't managed to parameterize the other, the turbulent forces in such a way that they can go into equations yet. So we're always just using this proxy of shear stress. So how the sand kind of gets in there and why it makes it, our guess is that it makes it smoother, so reduces the friction, and that would go with the shear stress approximation. But it might also mean that it's making it a little bit easier for the sweeps to push it along, or it doesn't have the pockets. Some people have shown that it's not just the bursts that cause the gravel to move, it's also how long they are. So maybe it's not wiggling around and falling back into the safe pocket anymore, because now there's sand there, so it pushes it off, and I'm kind of hypothesizing there. Speaker A [00:15:19]: Yeah. And so the sand, as through these little bit mysterious processes, more sand makes the gravel more transportable. Speaker B [00:15:27]: Yeah. Speaker A [00:15:28]: And that's implicit in your equation. Speaker B [00:15:30]: Yes. Speaker A [00:15:31]: So the premise of this podcast is to capture kind of the wisdom of senior and retirement age scientists. You don't really fit that demographic. We've never talked about our ages, but we're roughly the same mid career demographic, I assume. Speaker B [00:15:49]: Probably we can do that on camera. Speaker A [00:15:52]: But Wilcox and Crow has been around for a couple decades now. If it was a kid, it would be in college. Speaker B [00:15:58]: It's crazy. Speaker A [00:15:59]: So you've had chances to apply it. Other people have applied it. I'm kind of curious about how it's fared in the world. What are some of the best applications of it? Where is it particularly useful, and maybe what are some practical limitations? Speaker B [00:16:10]: And it's amazing. It's paired so well when we published it. I mean, everyone's careful when they publish, and we were very careful to be like, well, this works in this flume situation where we did it, and it seems to work well against these three data sets that have been published that we can apply it to. And that's kind of its little narrow range. And then people took it and they kept testing at different places, and it really did better than the others that were out and quite well. And I thought that was kind of amazing. Yeah, it just kind of showed there are a lot of gravel, sand rivers out there that were in need of stuff because we had, you know, the science had already gotten there better for sand than it did for gravel. I don't know if it's more complicated or just hadn't been studied as much. Speaker A [00:16:46]: Yeah. Speaker B [00:16:46]: So it seems to do very well. When you have a gravel bed river that is low in sand content, usually to the point where it's almost a framework bed. So it's a framework surface. I'm sorry. The bed can be higher. So where you don't have a lot of sand on the surface, I see people. It's not that it's not going to work where there's a lot of sand because it'll just reduce to the one shear stress value. It just seems odd to me to apply it. If your bed is 70% sand. I'm not really sure why you're using a predominantly gravel bed river. It's more of a sand bed river with gravel than it is a gravel bed river with sand at that point. It's not that it won't work, it's just odd. Speaker A [00:17:20]: Yeah. Speaker B [00:17:21]: So the biggest limitation I see is when I see people sort of. Yeah. Apply it outside of its range, and that kind of falls for all transport equations. People still, despite your warnings, my warnings try to apply, you know, they try to shoehorn whatever equation into another situation and then throw up their hands and wonder why it didn't work. Like, well, probably wasn't the best equation there, or you'll see people that try to do that. Well, I've got a whole lot of data sets. I'm just going to apply every equation and see which one's the best and then declare it the best for everything. Well, no, there is no one equation that's going to be the best at every river. You have to understand where your river system is that you're applying it to and then apply the appropriate equation. The other place where sediment transport modeling and the equations may or may not work well, and when you're testing a whole lot of them is if you don't have any data to go into it. So I said that you have to have a good grain size surface distribution. If you're going to use a grain size surface distribution based model, you also have to have something to calibrate. So you can't call it a sediment transport model if you don't have at least one calibration point. There has to be a direct sample, a volume, a change in bed, something that you have done to verify, calibrate that model. And I often have to explain to people, it's kind of like you wouldn't use a hydraulic model without calibrating it to your high water marks. Speaker A [00:18:36]: Well, some people would, but you shouldn't. Speaker B [00:18:39]: You shouldn't. Please do not do that. Speaker A [00:18:41]: Or at least don't have me review the project. Speaker B [00:18:45]: Yeah. And if you can't calibrate it, let's say there's absolutely no data, you've got nothing you can go on because maybe it's a debris flow model or a lahar. That won't be another topic. Then at least don't call it a model. Call it. You're like, better a, better b. It's. You're doing a b testing. Speaker A [00:19:01]: Yeah. Oh, interesting. Speaker B [00:19:02]: And you're determining which one appears to fit better. What would happen if the scenario. You might be looking at the changes. Yeah, like, okay, I'm nothing going to say this is the number that's going to be happening, but I'm going to know that it's going to have a direction up integrate, or direction down a degree. Those are the changes I'm looking at. Speaker A [00:19:19]: There'll be relative differences within, without the management, essentially. So how does hiding work in the transport function? Speaker B [00:19:26]: So, hiding is the exponent in one of the transport equations that goes in there. And I'm trying to remember exactly. It accounts for the fact that some grains are easier to move than others. So the way we put it in was as a ratio of the shear stresses that is then tied to the fraction of sand on the surface. Before that, it had been. I mean, it was there. Parker, I think, was the first to come up with it. I believe. Speaker A [00:19:52]: Yeah, well, there are some older ones. Speaker B [00:19:54]: Okay. Speaker A [00:19:54]: Yeah. But the Parker, I think, was one of the first equations that had it in there. Yeah. Speaker B [00:19:59]: And it was more just a guess. It was like. And then our hiding function will be 1.6. Somewhere between one and two was usually what it was. Speaker A [00:20:08]: I think that profit and Sutherland, they had one for the Ackerson White. But Parker was the first of this kind, I think, that looked explicitly at the ratio of the shear stresses. Speaker B [00:20:17]: Yeah. And this was a really big topic that and like how to measure reference shear stress, things like that back in the. It's really interesting to see that phase of papers that goes through during that time period. Speaker A [00:20:28]: It was a little bit of a golden age. You were writing, Gary Parker was writing a lot of like really classic work at that time. I have binders from that period. Speaker B [00:20:38]: Yeah, there are some. Yeah, some really good ones. But the hiding. So you go into that, ours really tied it to a physical process and that was a big change. So instead of just a number, now it's tied to why that might be happening was, you know, it doesn't capture everything, but it says, okay, if this fraction of sand is on the surface, we know it's gonna affect these shear stresses. That ratio of how the shear stresses change is gonna go into the hiding factor. And that way we didn't have to change a whole lot of the other fundamental equations. Cause you don't really want to change the well known equations, but you can alter the parameter about which you know the least. And by adding some function into that. Speaker A [00:21:13]: Basically like statistically or stochastically, when each grain class starts to move and you're adjusting that based on the prevalence of other particles that are involved in this movement. Speaker B [00:21:24]: Right. Speaker A [00:21:25]: And so you decrease the reference shear stress of the gravel based on how much sand is available. Speaker B [00:21:30]: Right. And the reference shear stress is fitted to the critical shear stress. I think it's 0.08. There's a factor in the equation, account for it, that you'll see and you'll wonder, why is that number there? Or you've never wondered why that number is there? And that's why that number is there. Speaker A [00:21:43]: I have, of course wondered, but also assumed it was fit. Yeah, yeah. Anytime there's a number with like three Sig figs and a transport function, I assume that that was the fit parameter. Speaker B [00:21:53]: Yeah. There's never otherwise three significant figures in a transport number. Speaker A [00:21:56]: That's right. All right, so you already mentioned this, but I was surprised when I went and looked at the literature that this was actually not your dissertation topic. It appears that your main work was on step pool configurations. Speaker B [00:22:12]: Yes. Speaker A [00:22:12]: Could you just start us off by telling us what is a step pool system? Speaker B [00:22:15]: So step pools occur usually in steeper streams, and it's made out of the largest grains in a lot, rather coarse grain size distribution. And if you put it in profile, it would look like a ladder, kind of a series of steps and pools going down. You'll see them. You start to see them all over the place. People were also using them in restoration to try to bridge a discontinuity in the vertical profile. Speaker A [00:22:38]: Okay. Speaker B [00:22:39]: Which I always thought was a really fun phrase, but really means there's too much slope and we got to get across it somehow. Speaker A [00:22:44]: That's right. Fish can only go so steep discontinuity. Speaker B [00:22:48]: In the vertical profile, and I just got interested in them. It's really funny, though, they're right up next to cascades. So if you see a cascade system, and one of the fun part is you can stand at a system because they're usually kind of mixed together. In reality, you can see steps or you can see a cascade, depending on which one you're studying that day. Speaker A [00:23:08]: Can you make the distinction between step pool and cascade? Speaker B [00:23:10]: Cascades are more random. It's not that they're randomly spaced, it's just more random boulders all over the place. You really can tell a step differently. Sometimes you get some random boulders in with it because you will have this sort of defined overflow into a pool that's been scoured just below the step form. And there's usually a hydraulic jump coming out of that. And then it goes for a little while, and then there's another one. Speaker A [00:23:31]: Oh, yes. Speaker B [00:23:31]: So it creates a really nice sequence. Speaker A [00:23:33]: So a step pool sequence definitely has a pattern, whereas cascades, there's just kind of boulders here and there, and the water's moving around it. Okay. But the issue of pattern seems to be at the heart of the work you were doing. Speaker B [00:23:46]: Sort of. I never really thought about it that way. I thought I, when I was a grad student, I thought I was going to find the parameter that causes step poles to form and sets their spacing, and I was going to solve the. Speaker A [00:23:57]: Whole problem, which is always the way dissertation started. Speaker B [00:24:00]: And then probably also world peace. Speaker A [00:24:02]: Yeah. Right. So let me frame this a little bit. For years, one of the cool things in our field is that there is a pretty regular meander spacing for meander rivers. That essentially the spacing of meanders is the wavelength of meanders is like five to seven channel widths. We actually did Calvin Kreach and I just did a paper on this. We did it on the Madeira again, 5.7 or something like that. So the meanders are pretty regular. And it seems like in the literature there was a little bit of a debate about, are the spacings of steps regular or are they maybe more random and stochastic? Speaker B [00:24:37]: Right. And I think what happens, we wanted to find. You want to find a pattern? We love to find patterns. I was so sad that it turned out to not have a pattern, that it turned out to be more stochastic. And I think nowadays people are more okay with stochasticity, but definitely back when. So when the literature I was working on was from mostly eighties or before, because there really wasn't much on step pools. And people were just sort of making hypotheses. They had very few data sets. And they were saying, well, it seems to kind of fit a spacing. That's like what we see other places. And if you wanted to fit that, you can give it enough of an error bar that within, you know, plus or minus. If you say it's four plus or minus one, you've got such a giant air bar, it's gonna fit in there somewhere. Yeah. Speaker A [00:25:21]: Right. Speaker B [00:25:21]: And some people thought they formed out of anti dunes. They never actually saw it. They just assumed it because they were like, well, they seem to be of a higher flow, and if we keep going up in flow rates, it should come after antidunes. Like we. They were again analogizing to sand. Yeah, I was really setting out to look at that antidune thing. So we had two different grain size distributions, and we were going to test basically deterministic testing in the flume. How does this all form? Speaker A [00:25:48]: So you weren't necessarily looking at a statistical process. You're taking an existing hypothesis about how these things formed and trying to reproduce it. Speaker B [00:25:57]: Yeah, I definitely prefer things to be mechanistic. Yes. It has taken me a long time to get used to stochastic. I really prefer to find answers and have mechanisms, but it just doesn't work that way in nature. All the. Well, it doesn't work in our understanding of nature. Speaker A [00:26:10]: Yeah. Speaker B [00:26:10]: Yeah. So it came out to be most predominantly random, is Poisson distribution. So after a short period, so there was distance right downstream of a step that another step won't form. And it's not all that long, but it's probably tied to turbulence. Speaker A [00:26:23]: Right. So there's, I think you called it an exclusion zone. Speaker B [00:26:25]: Yeah. Speaker A [00:26:26]: It's like three to six times the largest particle or something like that. Speaker B [00:26:30]: Right. We tied everything to the largest particle. Speaker A [00:26:32]: Okay. Speaker B [00:26:32]: Like, you don't get a step unless your largest particle is within like two to three or maybe two and a half to three and a half. Takes that many of them to bridge the width of the channel. Speaker A [00:26:42]: Okay, this is mind boggling, but if you go out and look at these systems, the largest particle has to be like a third of the channel width to form on this. Speaker B [00:26:51]: And they're monsters. We didn't really call them. I didn't even call them the D max or D 100. I call them the step forming grain size. Because even if you're at that D 90, you're still within a range. And that might be it. But really it's the biggest ones in that D 90. So they're the step forming ones. And it's all about the channel width to grain size ratio at that point. And it was really well shown in my flume because the flume had smooth sidewalls. And we were really worried when we started out that it was never going to form a step with smooth sidewalls. But that was going to be part of the test. And that is when we formed steps. When we used the grain size distribution that did not include that largest size. Speaker A [00:27:27]: No steps alluded to this. But I did think, reading the paper like scientists like to find order, we like to find equations. And it is interesting that the outcome of your work is that this is a disordered process. Speaker B [00:27:42]: Right. What else is going on that I miss? Is there something there? But I don't know what it would be. Speaker A [00:27:48]: And so what are the necessary components of forming a step? Speaker B [00:27:53]: It's really just that ratio of width to step forming grain size. I mean, you're going to need to have a graded sediment of all the other sizes. So this gets into like how you could then use them, which I've gotten to thinking about more over time, that you can then adjust the width. So let's say you have a river. You want to form steps. You know, you could do the old school method, rip it all up, put in concrete, rebar down to wherever, make sure nothing ever moves. Or you can kind of leave the river alone, but add some giant, you know, take that largest size that's in it. Assuming this is a river that would form stubs, let's hope so. Otherwise, I'm not really sure why you're trying to form steps in it. Right. Speaker A [00:28:32]: This is a classic restoration problem. But let's just assume. Speaker B [00:28:35]: Let's assume it's got the right grain. Speaker A [00:28:36]: That this is supposed to be a step pool system. Speaker B [00:28:37]: Yeah, you're right. Half the time they're not. They put them in. Then all you would need to do is narrow the width in certain locations, and you can do it wherever you want. It doesn't have to form a spacing. It can be where it's most advantageous to have a narrow width, whether it's with, you know, a big rock on the side, root wads to add some habitat, something like that. But that's. Then you're going to force a location where you're most, like, 90% likelihood you're going to get a step there. You might get a few in between, but if you definitely want them there. And what's interesting is Chartrand recently wrote a paper a few years ago that showed you can do the same thing with rifle pool systems, that that's all also controlled by channel width. So instead of just ripping up whole beds of rivers, because we really, I hope everybody realizes that's a very bad idea, because then you compact the riverbed when you drive over it a lot, and then you kill all the hyper react zone and the little bugs that you were really hoping to have there. It's a sad thing. So don't do that, and instead just work on the banks and nudge the river towards the physical processes that set up the bed form system you want. Speaker A [00:29:43]: I don't know which order the podcasts are going to come out, but I'm up here on a little northwest tour, and I also talked to Chris Nygaard, who spent time with Bonneville power looking at hundreds of restoration studies. Speaker B [00:29:54]: Oh, wow. Speaker A [00:29:55]: And I asked him, what did he take away from that? And he said, process over form. You have to emphasize process over form. And that's what I hear you saying as well. Is that okay, the process is, it's the ratio of really big rocks to the width of the channel. So you can add more really big rocks and. Or make some constrictions. And because it's a random process, it's actually more forgiving for restoration. Speaker B [00:30:19]: Right. It'll adjust around it if they're not right there at the first flood, because we've also shown that step pools will kind of adjust. They set during the giant flood, but there will be a few unsteady ones in there that you might see right away afterwards. But they'll work themselves out. They'll roll out, or they'll roll down and join the next one, and you'll get a very much more stable spacing that way. And he's right. And this goes back to what restoration people who are actually educated in it have been saying for a long time, you have to understand the process. The problem is that can take a while. It takes education. It takes worrying about it for a long time. It's not a quick learn by any means. And people want the quick answer. Okay. Speaker A [00:30:58]: So I've walked a few of these step pool systems in some, like, dam removal scenarios in the Napa Valley. And I guess my main question is how permanent are these steps? Once these steps forms, are they there for decades, centuries? Or do they kind of rearrange every ten year event? Speaker B [00:31:18]: These are natural steps. Speaker A [00:31:20]: Yeah, natural steps. Speaker B [00:31:21]: Just making sure they're not rebar. Speaker A [00:31:24]: But yeah. No, like in a natural pool system that maybe we'd be trying to reconnect. Speaker B [00:31:28]: Right. And it would probably be. I think we estimated the 50 year flood and the ones where that's really been tested are Italy and Switzerland because they both have step pool systems that they've been watching long enough and they have seen them reset. Speaker A [00:31:41]: Yeah. Speaker B [00:31:41]: I mean, in the flume, obviously. I made it happen all the time. Speaker A [00:31:45]: Yeah. Right. Speaker B [00:31:45]: And they would kind of set up a stable, even with a constant high flow, set up a stable system, build to a certain point and then all fall apart and then, you know, you're racing to catch them all at the downstream end. Speaker A [00:31:57]: Yeah. Speaker B [00:31:57]: But, yeah, it's usually, it's a pretty, pretty big flood to move because you got to move the boulders. Speaker A [00:32:01]: Yeah. Speaker B [00:32:01]: So some have said hundred year flow. Maybe now it's a 50 year flow. Speaker A [00:32:05]: Okay. So they're. They're persistent but dynamic. Speaker B [00:32:09]: Yeah. Speaker A [00:32:10]: So you're from Tennessee, but we all think of you as a gravel bed person because your dissertation work and then you have ended up in the northwest. So what's unique about gravel bed rivers? What do you love about them? What do you find interesting about them? Speaker B [00:32:25]: Maybe it's because they weren't as well sorted out as sand bed. I mean, sand bed's beautiful. I do love it. And it's a giant systems, and the bed forms are amazing. You can go look at them on, cut them down on beaches and all that gravel's messier in a way. It wasn't an allure to the mountains or anything. No, it's just messier. There's a lot going on. There's a lot of processes we don't understand. I do like mountains and I want to leave everything better off. I hadn't really considered fish much, to be honest, but I just. I like when the big stuff moves. Maybe it's that catastrophe part to it. Maybe that's why I was drawn to step pools, because they moved during these giant floods. Not sure, but it just seems to set up so much and there's so much to learn. Speaker A [00:33:07]: So you've used a term a couple of times which is actually something that both you and I have done some work on, is clusters. What's a gravel cluster? And first, what is a cluster? What do you mean by that? Speaker B [00:33:20]: Yeah. If you start looking at gravel bars a lot and you start looking down at them, you'll see them because they're like everywhere. What happens is it's not a smooth surface of gravel. You'll have mostly a sum of surface of gravel, but you'll have these little, like three to five grains will group together, kind of like they're just holding onto each other or something, and they'll be all over the surface. And that's what we call clusters. Speaker A [00:33:42]: So you're out there and it's all kind of random, except that it's not because there's these clusters of some of the larger particles that it almost seems magnetic. Speaker B [00:33:54]: Yeah, they sort of stick up. Some stick up more than others. People have tried to do morphologies of them. Are they line clusters, pebble clusters? Do they form rings? Because we love to see that pattern. I looked at mostly as clusters. How do they form? Well, maybe not how, but when do they form? Under what flow rates do you get more or fewer? And how were they were spaced? So I looked at whether yet like a single cluster on its own, a double cluster or triple cluster, and I was getting into turbulence a lot. Speaker A [00:34:21]: Yes. Speaker B [00:34:22]: And how that affects the flow, because when we started getting laser systems in flumes or laser systems in general, and then they started moving into the world of research, it was amazing. Everything you could start seeing or start to understand that you never could before, because now you could start to tie the flow to the process a little bit better. Speaker A [00:34:44]: So you did some very detailed turbulence measurements around these clusters? Speaker B [00:34:48]: Yes. I couldn't afford a piv system, but I could get an acoustic doppler velocity. Just the single probe. Speaker A [00:34:56]: Yeah. Speaker B [00:34:57]: And so what we did was we had basically every set of experiments. I do seem to involve multiple sets of sediments with ranges of sand in them because I know how important it is, or I just really like doing it. And on the following limbs, what we did was make the bed, kind of make it flat in the flume, put a flow over it that mobilizes most of the sediment, we call it, because you don't want your bed surface to replicate how you set it up. So you kind of have to do this annoying thing where you run it for a long time to get it to water worked and equilibrium. And then we ran a flood over it, a high flow, and then degradation, no sediment coming in. Lower the flow, just lead it armor and then forms clusters on the armoring stage. And then we had the probe and we set up a grid around the clusters. So depending on the size of the cluster, you had to go out like we went up, I think two on each side, and set a grid, a three dimensional grid, so the probe would move around the cluster and then up through the profile. Oh, wow. So that it could measure, because it measures the three dimensional velocities, but it would only measure it at points. So I had to fill in all those points. I say I, but I had my grad student do this, who then did something I'm still rather amazed by. But he had an incentive to not stand there and do that himself, I guess, that he didn't want to. We had Labview and we had stepper motors. He programmed the whole thing to run on its own. Speaker A [00:36:25]: Oh, wow. So why are clusters important? What do you think? Both kind of morphologically or biologically? Why are they important in gravel reverse? Speaker B [00:36:33]: Oh, for me, I've always thought of morphological mortg, but they do have biological importance. Morphological morphs, they moderate how that bed interacts with the flow and then how everything transports. So I think some of the mess that exists still around sediment transport predictions is due to clusters on the bed and how those clusters are moderating the flows and how that then impacts when it will break up. So we saw that there is a difference with how stable the whole bed is depending on how the clusters are set up and how many double or triple clusters you have will impact when the clusters move and then the overall stability of the bed. So at the end of each, after we did the armoring thing and measured around them for days on end, then we would put the flow rate back up and we would just how long it took for that armor to break. And it broke. And it was the same flow that had to equilibrium. So we weren't changing that. But you could see a difference in how long it took the armor to break depending on how many clusters there were. Speaker A [00:37:28]: Well, that's really cool. So what I'm hearing you say is that when I compute sediment transport flux, I'm really just looking at the individual particles and maybe using Wilcox and Crow looking at some particle interactions, but it's still particle to particle, but there are these kind of larger scale components to the bar morphology, that the initiation of motion is a more macro scale process. Speaker B [00:37:56]: Yeah. And it would be akin to bed forms in the sand, red rivers, except we don't really know what to do with them yet in gravel bed, but we're starting to get closer. Yeah. And then they're important because I think bugs like to live there. Speaker A [00:38:06]: I think that's a great physical scientist answer. I think you're right. Okay, so one of the things I want to do with this podcast is kind of like, pick some discrete insights that kind of made you who you are and that maybe we can transfer to new river scientists. And so is there like one paper insight or a big idea that changed the way you looked at rivers? Speaker B [00:38:33]: Okay, so this is what I've thought about a lot, and my first instinct was not really, but I. But about every few years, some paper I get really enamored of comes out, and sometimes it'll be tied to what I'm working on at that time. Or there'll be a paper that I think really pulls it together well or changes things. And then thinking about it more, I realized it was, for me, really, it was a suite of papers that started showing connections between turbulent flows and aquatic systems. And there's some by Nikora. Was one person doing it. Hardy and Lane are the three main names from that kind of era. And people have continued with it since then, of course. Speaker A [00:39:16]: So I don't completely understand what those words mean in that order. Can you break it down a little bit? Speaker B [00:39:24]: They were some of the first ones to be able to measure around the coherent flow processes that are created downstream of something that's in the flow. Did change the way I was doing things. You're right. I started looking at the turbulence and how that connects back to the grains on the bed surface, and then also to emergent vegetation. How do you get that vortex street setup? What does that mean for the way the sediment deposits on the bed or then erodes from the bed? And the reason the cluster spacing matters is because of the way the turbulence generated around the clusters interacts. If they're close enough, it'll actually dampen everything out, and there'll be more stable on the bed. If the clusters are single, it's standing proud. It's less likely to add to the stability as much as if it's dampening it. Speaker A [00:40:14]: So you have these turbulence. So this sounds like it directly influenced the cluster work, because you have these turbulence feedbacks, but it depends on the number of clusters and the spacing, how much the rest of the bed feels, the turbulence, which, as you suggested, is the real thing that shear stress is kind of a heuristic for. Speaker B [00:40:34]: Yeah. Does it create a skimming flow over it? And basically, do you get the transfer of energy from the near bed into the outer bed also? And I probably could have started thinking about the river or the flow as a big set of energy before that, but it really kind of solidified that because we had the roughness. We know that uses up some energy in the system. But now I really think about it, is what parts are using up? What energy? Speaker A [00:40:54]: Oh, interesting. Speaker B [00:40:55]: And so when you get erosion based on how much excess energy is in the system to erode, as opposed to being taken up by going around root wads or over clusters or, I don't know, through a pool. Speaker A [00:41:08]: Yes. So I have this theory that anyone who's been as prolific as you that there was some report or finding or paper that you put out that you thought was really cool and really important that just never really hit. I mean, not just like Wilcock and Crow hit, but just like it just never got the attention you thought it deserved. And so one of the things I like to do is say, bring our attention to it. Is there something that, is there a finding or a paper that you have out there that you think really should get more attention than it has? Speaker B [00:41:44]: Yeah. And I think it's the emergent vegetation work. Speaker A [00:41:47]: Okay. Speaker B [00:41:47]: And I've looked back at that paper. Speaker A [00:41:49]: That'S with Kevin Waters. Speaker B [00:41:50]: Yeah. And if I ever get off my butt, there'll be more papers. But I haven't really gotten around to getting at the flow ones. But the work remains relevant from what I can, you know, I keep up with literature and it's still very relevant. We looked at how sediment transports through patches of emergent vegetation, and the motivation was looking at rivers. Not so much these gravel bed rivers, but more rivers on this side of the Mississippi divide of the continental divide. And you get patches of vegetation growing in them, whether it's Texas wild rice or some sort of, I'm just going to call them macrophytes. That covers all vegetation, I think. And at what point does that grow so much? The thought was that it was choking off channels in irrigation. I think that was one of the reasons the Rio Grande was drying up. People were worried that the vegetation in the river was taking up too much water. It was causing sediment to deposit. But then at the same time we'd be like, wait, it has a lot of benefits. For an aquatic ecosystem to have macrophytes in it. So we started wondering at what point, and is there, maybe we thought there would be a point at which the patch would cause deposition or not the size of the patch or the density of it, and would there be like a flushing flow? What kind of thing could you do with that? And it was really interesting work. It was back when 3d printers were new, so we 3d printed the vegetation. Speaker A [00:43:07]: Oh, cool. Speaker B [00:43:08]: Yeah. Because we found that it really matters. First we wanted to scale it. We had to give up a little bit of the scaling to get the right processes. And that's one thing I'll see. Especially in urban firms, people try to over dimensionally scale things when what you're really trying to get at is process nothing. Exact dimensional scaling of stuff. That's another rant. Yeah, we printed the vegetation, and if you just use wooden dowels, they kind of stand still. And that's better for simulating, like a pier or pylon. But the emergent vegetation wiggles. The first time we got in the flume and we saw it wiggle, we were like, oh, yeah, it does do that in nature. Yeah, we do see that happen. Okay. That wiggling. It was also fascinating that the density of the vegetation really mattered. I mean, we thought it would, and it did. But when some of the vegetation patches increased erosion and not deposition, and then we realized that all had to do with the way the vortices were forming, whether they were able to go through how the flow was interacting, whether it was going through it, whether it was going around it more. And I still see that as applicable across the board. And I'm seeing it also a lot now that I've moved out here with wood and rivers, it's kind of setting up some similar processes. So if you're looking at an ELJ to put in a river and how porous you want it to be, the issue there is the porosity. So if you want to put logs in the river and have some sort of porosity, because you're trying to achieve a type of still things you have to think about if you're trying to achieve a certain outcome. We don't have a lot of literature yet on wood and porosity. It's growing, but maybe we look over at emergent vegetation literature, of which there is more. Oh, interesting, because putting something in and what density do you want it to achieve? The result you're trying to get? It may not all translate, but it will give some information and help inform the designs. Speaker A [00:44:53]: Yeah, that isn't an application I even thought of when I read that paper is that, you know, when you put in an engineered logjam, you can put it in as a like, complete barrier, or you can put it in with some holes, which some porosity, so that flow can go through it and those will have very different morphological implications. Speaker B [00:45:12]: And I have to give credit here because I didn't think of it either until I got out here. And then I've been doing some work with Xiaofeng Liu and his team out at Penn. He's at Penn State, and they've been looking a lot at porosity. And looking at the results is when we started talking about it, and I was like, this is really similar to emergent vegetation results. And we've got to bring the two together. Speaker A [00:45:31]: So speaking of the work that you've done with Kevin Waters, you passed along a paper to me from down and soar, and it seems really connected to another paper you did with Kevin Waters. The main thing that I'm just fascinated about from this realm of work that you're in is the idea of stress history and the idea of year to year transport stationarity. Can you just tell us a little bit about what stress history is and how it might change our paradigm of sediment transport? Speaker B [00:46:06]: Oh, my gosh. If we ever get a handle on it, it could change a lot, which makes it so cool. I mean, we're due for another upheaval. Speaker A [00:46:15]: Yeah. Speaker B [00:46:16]: And a lot of people looked at it different ways. But as you have these lower flows, the bed rearranges. Some people have shown how it rearranges. Speaker A [00:46:24]: Sorry, you didn't just misspeak. But when you have lower flows, the bed rearranges. That's not the way that I generally think of rivers. Speaker B [00:46:32]: People were taught, especially in their undergrad, the big flood moves the rocks, and it does move the rocks. We're not saying it doesn't, but. And this is like anything you do with sediment transport. You can't get it all in just a few classes, and especially not in an undergrad. Yeah, I don't think I knew what it was in undergrad. But back to stress history of beds, the bed, the little grains, they shift, they rotate, the flow pushes it so it's more streamlined, along with the flow. And some work out of England showed that it happens in the subsurface too, which changes the way the hyperrhic flows will pass through, or can. And they did that using CAT scans. Aside from just that, just with subsurface. But on the surface, it's a little bit easier to understand. And will probably happen more often because you have these flows that do they align the bed. Like I said, the step pools, the steps form in the big flood, but the lower flows afterwards kind of rearrange this to make them more stable. What I liked about the down and soar paper was it tried to tie it through a very detailed set of indirect bed load measurements to what had happened before and whether it had been low flows or high flows. And it kind of ties back to those low flows rearranging the bed. It brings back to the idea of the antecedent rainfall. You have to look for infiltration, whether or not you've had wet soils or dry soils. Speaker A [00:47:49]: Yes. Speaker B [00:47:49]: Yeah. Same idea for flows. Whether or not you've had a whole sequence of high flows. So everything may be loose, might be a word for it, or if you've had a whole lot of these low flows and everything's rearranged and much more stable. And it's just like, no, I'm here and I don't want to move. Speaker A [00:48:02]: And so this was super counterintuitive to me. But the data in your paper and the down and soar paper were really compelling, is that if you have a particular flow year, you can't just like, randomly pick that out of your proverbial jar of marbles. If you have lower flows the previous year, or like maybe near bank fall is what I was seeing, then your bank's going to, like, harden and it'll be. You'll have lower transport than if you actually have a big flow the year before. Speaker B [00:48:34]: Right. And it's. You're right. It's really interesting and it makes it more understandable why it is so hard to predict. Speaker A [00:48:42]: Yes. Speaker B [00:48:43]: It's not that you throw up your hands and say, it can't be done. You say, okay, well, now I really should have been considering all these things as well, and understanding how exits, not just. It's the whole system. Yeah. Speaker A [00:48:54]: So when we turn off the mics, we need to talk about a sediment transport function that looks back in time. Speaker B [00:49:00]: Oh, my gosh. I know. Speaker A [00:49:02]: Yeah. Speaker B [00:49:02]: I mean, I'm dying to add some sort of cluster parameter to sediment transport functions to try to adjust for that. And I still don't know what it would look like, but it feels like it should be adjusted for somehow. Speaker A [00:49:13]: Yeah. So one of the things that I think is really interesting is that we've actually talked a lot about your earlier work. And you started out maybe the first major portion of your career in the laboratory, but now you're. For many years, you kind of been in the private sector. Now you're in the public sector. You're really working in the field. You're doing modeling. You're out there. And so I guess one of the things I'm really interested in is like your career trajectory. And as someone who's more of a like field and model practitioner now, how does your early work in the laboratory inform that? That's kind of a pretty unique perspective. Speaker B [00:49:48]: Yeah. And I'm glad I had the flume work. I kind of miss it sometimes. But you miss whatever you're not doing when you're not doing it. That's right. And you have to have all three. I hear people because people don't understand the flume is so necessary for setting up what could be happening, for getting at those processes because it's too messy to go straight to nature. You can't control enough stuff. And I've spent half this time telling you you have to bring in all these other things. And I'm like, no, you can't control them. But when you want to try to understand how they're interacting and which 1 may be a dominant parameter, you got to do it in a flume. And I'm not as big a fan of stream tables. Stream tables are great if you're doing plan form or vegetation and how that interacts with the plan form, maybe giant slugs of sediment. I really prefer flumes because a lot of what I look at is sediment interaction. And I think I alluded to this before. You can't, you can scale sediment to get it really, really small. So it has a dimensional scaling similar to the size of your width. But you're not going to get the process correct. And it's the process that matters. Not that your sediment was dimensionally scaled. Speaker A [00:50:56]: Right. Speaker B [00:50:57]: And so the flume is where you get all that and it's where you can push a few more of the boundaries. Also, it's good for understanding what might happen if I did x. Like I'm not going to go out and do all my experiments in the White river. That would be fun, but it would take a lot of earth moving and destruction. But I can do it in a flume and get some idea of what might happen. Or let's take step pools is a classic example. I got that all the time. Why didn't I study them in the field? And I think because I'm studying how step pools form and they form during the hundred year flood. And by the way, I'd die if I tried to measure that. Speaker A [00:51:31]: That's right? Speaker B [00:51:31]: Yeah, that's right. So the flume is essential and then you need the modeling to push boundaries even further to say, okay, well, you got all these things in the flume. But now what if we did have that 100 year flood? What would happen? And can I reproduce it well enough that I can then feel really good about what I'm doing in the field? And studying in the field is more fun because you do get all those messes and you do get to say, did I get it close to right? You have to have it for calibration for like, have I got a clue as to how these things form? If they weren't measuring the steps when they reform in the Rio Cardan, we wouldn't have these data sets to say, oh, it is actually random. We would just have flume data saying it's random. Other people saying, well, no, I kind of see a pattern. But no, we're like, okay, yeah, they've been resetting and they're not setting to anything in particular. Speaker A [00:52:14]: It's interesting that you, that you have, you frame it as kind of a three legged stool because that's always the way I've thought about it, too, is you've got like, you got the laboratory in sight, you've got the numerical insight and then you've got the field in sight. And the three of those are in constant conversation. Speaker B [00:52:30]: Right? Yeah. We're not going to go out and change something happening at a fish passage unless we have tested it out first. Speaker A [00:52:38]: Right. Right. I. Speaker B [00:52:39]: We just don't get to experiment on where there's endangered species, nor do we want to. Speaker A [00:52:45]: No endangered species in the flume. So this is a question I'm trying to ask from time to time in these is let's imagine that you had to restart your career in sediment transport and like in some sort of time machine, you could go back and tell that past you. Two things about rivers. What are the two things that are kind of so important that you've learned over time that you would want to give them to yourself as like someone who is restarting in this career? Speaker B [00:53:13]: Yeah. I thought it was going to be what to know about your career. Speaker A [00:53:16]: Oh, that's fine, too. Yeah, I can reframe it. That's all right. Speaker B [00:53:19]: I think rivers would be fun because I think it would be. Well, I wish I'd taken more math and that's so nerdy. Or maybe it's not that I'd taken it, but that I had a knowledge of it. Speaker A [00:53:29]: All right. Speaker B [00:53:29]: And some things have just changed over time. I wish I could do more the robotics type stuff, but I don't really know how to do that. And I want to play with it, but I spend more time doing other things. But yeah, because we can do a lot more automated things. And I'm trying to get into more of that stuff because I think that's really where we're going to learn a lot of stuff. If I had to start it all over again, other than you can always try to understand the processes better. One good thing I did that I would recommend if anybody does when they're getting an education in it is took most of the classes, actually twice. And, you know, the first time really took it for the grade, everything like that. It's kind of unfair if you just sit in the first time and then take it the second time for a grade. But the first time I did it for real. And the second time it was more because I was studying for my orals comprehensive exams. And it was easy to go back and sit in because in this case, Peter was teaching the class again. So I'd sit in on it and. And you hear it again. Maybe I took notes again, probably did, because it keeps me paying attention. And it's such a complex topic. It takes doing it over and over to really start to see how it all connects. Speaker A [00:54:40]: I actually thought I was the only one who did that. I did the same thing. And I just think that once you take the class, you suddenly realize what all the pieces are that need to fit together. And you look back and you're like, oh, that was actually a magical lecture, and I just missed it. And so I would actually go back the next year and sit in on those lectures. I remember one of my professors was like, look, this lecture is so good. Stanford's back for it. Speaker B [00:55:10]: But for career wise, this is what we tell us to. We thought about this a little bit at home with nieces and nephews, going off to college. Okay, you know, I did change my major a lot, but I was always. You always have this fear, like, especially if you're in college or grad school. Everything's. That's the most important decision you're ever going to make. Oh, my God. What. What can possibly. You can never overthink it. But, yeah, you can. Because the big thing I learned is if you don't like what you're doing, you can just quit and get a different job, which I've done. And I don't mean like, if it's too hard, quit. No, you want to be challenged, but you want to be fulfilled. Speaker A [00:55:42]: Right. Speaker B [00:55:43]: And right now, I love working here. But, you know, I had to put. There were all different reasons I liked where I was or where I wasn't at a place. But you don't want to sit at a job. That's not pushing your boundaries. Or not challenging you. Thinking, well, I'll just get hobbies or something. You want your job. You want your career to do that. So just quit and go get. Go do something. There's lots of firms. There's lots of universities. Maybe the corps would hire you. But no, there's lots of agencies, too. We've got great state agencies. That one thing I love about the core. And the state agencies especially, at least out here. Is people are just. We're all trying to do better with everything. And it's fun to be around people who constantly want to be doing better. Speaker A [00:56:25]: That's great. Yeah. Well, we're lucky you landed with us, Jarenka. Thank you for being on the podcast. Speaker B [00:56:33]: Thank you. Speaker A [00:56:35]: So we did talk about how to modify transport equations for antecedent conditions. After that conversation. As well as like a half dozen and other sediment analysis features. We should try to get into HZ rats. So if you have thoughts on that topic in particular, reach out. We may try it sooner than later. Also, while we're not quite halfway through releasing season one. We are almost done recording and producing this season. So it's actually a good time to start thinking about season two. We have some really fun ideas for season two. And anyone who's still listening. And has kind of made their way all the way through this episode. You're gonna love it. But if you have ideas about who else we should talk to in season two. Let us know in the episode. Comments next week we're talking to John Remus. Who has spent his career working on the Missouri river. Including some remarkable prototype experiments on that river. He also is the leader of a team of sediment and river mechanics experts. That I like to call the sediment Avengers. Which I could explain. But it's more fun just to make you wait till the next episode. This podcast was funded by the regional sediment management program of the Corps of Engineers. And is part of their tech transfer initiative portfolio. That we like to call RSMU. Congratulations, by the way, to Dave Perkey. Who was just promoted to the lead of the RSM program. We also received funding from HH and Cset. And the flood and Coastal R and D program. Mike Loretto edited this episode. And wrote music for the season. These are informal conversations. And the views expressed. Do not necessarily reflect the positions of the US Army Corps of Engineers, their partners, or the offices or centers of the guests or host. Thanks to doctor Joanna Curran for talking with us today. And this week we'll dip into the bag of long standing podcast traditions by closing with an outtake. Speaker B [00:58:18]: I had this theory that I should never I couldn't pass my comprehensive exams unless I could derive the Navier Stokes equations. Speaker A [00:58:24]: I had the same theory, and they didn't ask. Speaker B [00:58:25]: They didn't ask me either. But I did. I did it so many times. Speaker A [00:58:29]: Me too. I set coffee shops and just directly Stokes. Oh my God. Speaker B [00:58:34]: I had no idea we were on opposite sides of the country. Just arriving Namius Stokes.