Stanford [00:00:13]: I honestly can't think of a better way to introduce reservoir sediment management. Than the first two episodes of the season. Conversations with Doctor George Annandale and Doctor Greg Morris. Who are without question two of the top global experts on the topic. But you will notice that most of the examples in those conversations focus on reservoirs in Asia, Europe, Africa, South America, and the Caribbean. There was very little discussion about reservoir sediment management in the continental US. And there are reasons for that. We face some idiosyncratic challenges to these practices here in the US. But we also have a lot of aging reservoirs, most of which are already operating past their design life. The Corps of Engineers alone has over 500 reservoirs in our portfolio. Today's guests have been involved in more reservoir sediment management plans and projects in the US than anyone I know. Doctor John Shelley and Doctor Paul Boyd are the regional technical specialists for sediment transport in the corps of engineers. Kansas City and Omaha districts which together account for the entire Missouri river watershed. And they are national subject matter experts on reservoir sediment management. Paul's leading an initiative to evaluate and rank the sediment impacts across our agency's reservoir portfolio. And he studied flushes and potential sediment management alternatives on and around the Missouri river. And John, who also teaches river engineering at the University of Kansas. Has been involved in one of the most innovative reservoir management plans in the core. John and Paul are also two of my closest collaborators and, frankly, friends in our agency. We've taught sediment and reservoir modeling together in several countries, including occasionally our own. And we just put out an ASCE paper on reservoir release case studies in the US. And we have conversations like this all the time. So when we gathered to prep our latest reservoir sediment modeling class, I thought I'd turn on the mics and capture one of those conversations. About the unique challenges and opportunities around managing sediment in the US. I'm Stanford Gibson, the sediment specialist at the Corus Hydrologic Engineering center. And this week on the RSM River Mechanics podcast, a conversation with doctors John Shelley and Paul Boyden. Paul Boyd and John Shelley, welcome to the podcast. Paul [00:02:16]: Thank you. John [00:02:17]: Thank you. Stanford [00:02:18]: So Jon, you have a classic quote that I replicate all the time about the actual failure mode of dams. Can you tell us a little bit about how you pitch that? John [00:02:29]: Yeah. At the core of engineers, we are very concerned about dam safety. And we have high consequences of a dam failing. But a probability of something ridiculously small. And at the same time, the reservoir is filling up with sediment. So there's a 100% probability of failure for the dam. It's just a matter of when. But it's failure by sedimentation. Stanford [00:02:54]: And so we're very concerned about failure modes of these dams that are approaching the level of a comet could fall from space on the dam. John [00:03:03]: Right? Right. Stanford [00:03:04]: But all of our dams on a long enough time scale have a 100% failure mode that we're not thinking a lot about. John [00:03:10]: That's right. Stanford [00:03:11]: So when I was in grad school in the late nineties, I remember I walked past a flyer on the wall, you know, how, like, in grad school, like other grad schools, would advertise, you know, their program? And there was one graduate school, I don't remember what it was, but they were advertising a PhD program, and the whole thing was about how to remove sediment from reservoirs. And I remember thinking, wow, that's an interesting problem. Seems like it has a very simple solution. You just, like, open the gates and blow out the sediment. Little did I know that that's actually a little bit more complicated. But both of you have reservoirs in your district that kind of open up the gates and blow out sediment, and we call that pressure flushing. Paul, do you want to tell us a little bit about how you use pressure flushing in your district? Paul [00:03:54]: Sure. The corps operates three projects called the Tri Lakes projects in suburban Denver, Colorado. And the one that operates a pressure flush annually is Cherry Creek. And it is on the southeast side of Denver there. And it runs into Cherry Creek, which then goes to the South Platte river and then through Denver and on its way all the way over to the Platte and then the Missouri. The Cherry Creek reservoir catches a fair number of fines that come off that front range watershed right there. And those fines kind of drift, shall we say, and settle on top of five outlet gates, there's five kind of mid level outlet gates, and they're used, obviously, when we need to discharge pools get too high, we need to open them to send water down Cherry Creek. Back in the 1980s, they had, even earlier than that, some operational testing for the reservoir had been in, say, 20 years. And they had a little bit of a problem getting those gates open. Did a little investigation and found that some of those fines, the silts and clays, were kind of packing and restricting the ability of those gates to freely move when they need to. So the corps implemented a pressure flushing activity, pressure implying that we're not draining the reservoir, we're keeping the pool there. We're trying to use as little water as possible to achieve the goal. And the goal is to open these gates, create a little higher velocity, flush out some of these sediments, so that if and when we need to use these gates. They are fully operational and we don't have any restrictions there. So, so the core has operated that on and off, but almost annually for about the last 35, 40 years, in May, got a kind of a well oiled machine of public outreach to let the public know that there will be increased discharges, because at some of those flows, the water does come out of the channel. But it's a. It's a great example of one, a very cost effective method for managing that segment. Otherwise we'd be out there in a boat or some sort of dredge and trying to stir up and remove those materials, and they do it over just half a day. It's well understood by the public. They know about when it's going to happen. Every year there's press releases and outreach, and it's a great success story on using a management activity that uses as little water as possible, is very low cost and is very low risk. Stanford [00:06:14]: So we talked to George Andale and Greg Morris about this, about the distinction between a pressure flush, where you try to keep the reservoir full, you don't try not to use water, and like a drought on flush, where you're actually going to go run a river. One of the things they talked about is the pressure flush doesn't actually move that much sediment. It's a local management option. Paul [00:06:32]: Absolutely. Stanford [00:06:33]: And you guys actually went in to try to measure how much sediment it moved? Paul [00:06:37]: We did. Stanford [00:06:39]: So Cherry Creek's an enormous reservoir with pretty good size, quite a bit of sediment in it. How much sediment does the pressure flush tend to move? Paul [00:06:46]: It depends, as every good engineering answer starts. But in a collaboration we did with reclamation and their office, the technical service center out there, our core RSM program funded a study a couple years ago, 2017 and 2018. I think we were all out there watching this, where we did a high density multi beam survey, developed a terrain, did a flush, came back to the terrain, and let's just say that the impact of that flush on the volume of sediment in the reservoir is within the noise of the data collection. Stanford [00:07:22]: You didn't see it in the repeated data. Paul [00:07:24]: Right? You can't. It's not an action where you are going to count on it to help reduce storage loss. That's all. All that coarser material is coming in at the other end of Cherry Creek, forming a delta there. These are the fines coming down. We made some rough estimations in the end. We're talking a few dozen yards type of thing, but within the noise. So it really is a management activity to reduce management risk, and it does not provide. The cone of influence as measured in that survey, is probably 40, 50ft out from the gates. That's about it. Stanford [00:08:03]: So John, you have several reservoirs that do pressure flushes. Can you tell us how they work in your district? John [00:08:08]: Yeah. So every few years our operations and maintenance staff will go out to a few of our lakes and just put down a measuring tape and see how high the silt has built up next to these gates. These gates, they're not really service gates like Paul was describing. They're actually emergency drawdown gates. Specifically thinking about Blue Springs Lake. There's a sewer line that runs under the lake. These lakes, a couple of them are built in urban areas, so there's already urban infrastructure crossing the area. So if they ever need to service that sewer line, they're going to have to drain the lake in order to do it. And so they have these gates there that they can open and drain the lake all the way down. But the same issue as Paul was talking about, if enough sediment builds up next to the gate, then they can't operate the gate. I think every year they measure it, but every few years there's enough silt built up that then they'll do a pressure flush. So they just open the gates and they just have somebody downstream watching. And when the water starts to run clear, which is about ten minutes, then they say, okay, that's it, close the gates. We videoed this in 2018. Got a very sparsely watched YouTube video. Stanford [00:09:18]: We're gonna put it on the page for this podcast. Paul [00:09:21]: That's good. John [00:09:22]: It was fun to make, but it was really fun to be there. One of the interesting things I caught on video, somebody making this comment that like, wow, it's the smell that I notice, and that's true. There's a real putrid smell, you know, like the smell you get when things are anoxic. And when that release started, you could instantly smell that, you know, anoxic kind of putrid smell. And the fish downstream, the carp start jumping out of the water. So we got that on video, too. Stanford [00:09:50]: Well, this was the most dramatic thing I was at that flush. The most dramatic thing for me wasn't the smell, but it's actually a very small outlet pool. And as soon as it started, I don't know, maybe 30 asian carp started jumping right out of the water. And this is in the video, you should go see it. But John, why are asian carp jumping out of the water? Do they not like the smell? Like, why are the asian carp jumping out of the water? John [00:10:14]: Yeah, you know, I actually, I actually don't know. I know that in the Missouri river, we always talk about it being the vibration. Stanford [00:10:21]: Oh, really? John [00:10:22]: That causes them to jump. The vibration of the boats will cause the carp to jump, you know, literally into your boat or into the side of your boat. So I don't know if it has to do with the vibrations of the increased flow or if it actually has to do with the low dissolved oxygen. Right. Stanford [00:10:35]: But dissolved oxygen is an issue that's often associated with these flushes. What's the dissolved oxygen story? John [00:10:41]: The water at the bottom of the lake tends to have low dissolved oxygen. And then also the sediment right at the bottom of the lake has deposited in these low oxygen environments and used up all the oxygen long before the flush happens. And so I think it's a combination of the sediment and the water already having low do that causes those initial releases to have low dissolved oxygen. Stanford [00:11:04]: So we're going to talk about drawdown flushing in Spencer Dam in a little while. But while we're on the topic of dissolved oxygen, there were some decisions and management options that happened revolving around dissolved oxygen at Spencer. Can you tell us a little bit about that, Paul? Paul [00:11:18]: Yes, absolutely. And we'll get to this a little later. But Spencer Dam had biannual twice a year flush that had been doing for decades, and they had worked with the state and the wildlife management organizations there to develop a plan, really to mitigate. Right. You don't, if you put a big plume of low do water, you have fish that suffocate. Right. And that's what we're trying to avoid. And one of the methods to mitigate that is the old, the solution to pollution is dilution. Right. So in this case, in their management action for flush, they draw down very slowly initially so that we only have a little bit of that low do water heading downstream, mixing with other water that's coming, you know, from surface water sources. And the state of Nebraska who monitored that during the flush, had found that if they do that very slowly, over about 24 hours before they have a very active flush and they get on it. Right. And open everything up, they see that the fish can sense that and they start just kind of heading downstream into areas where the do is sufficient so that by the time really open things up and you have a big plume of low do water, you've got miles of essentially dilution area. So that low do water is mixed with other normal do water and it drastically reduces fish kill issues for a flush. So every situation is different, but that's something that's monitored in a lot of flushing activities. Stanford [00:12:49]: The reservoirs they're flushing in the northwest are paying particularly close attention to this because they're actually flushing those for fish passage. And so they want to make sure that, you know, the medicine isn't worse than the disease. Right. So they're monitoring the do very carefully, and it does drop in the early parts of that flush. But the thing that's interesting about Spencer is all these other ones we're talking about, there's a lot of fines, which I would expect to have high oxygen demand. But at Spencer, there isn't like a deep pool that would have low oxygen and most of the sediment is sand. So where do you think the oxygen demand is coming from in the Spencer flush? Paul [00:13:27]: Right. That's a great point. Traditionally, as John's talking about, if you have this deep, low water and some of our very large reservoirs, if we did some sort of, you know, draw down flush, we'd have a major concern. You've got tens of thousands of acre feet of this cold, low DL water. In the case of Spencer, it's a very shallow reservoir. It essentially filled up a couple of decades after they built it. So it's a very shallow area. However, it's in a area, the sandhills of Nebraska, where we do get some very flashy storm events. And I, as we saw when we did some surveying back there almost a decade ago now, material deposits in that shallow reservoir in lifts, you'll see a sand lens, and then you'll see a lens of a lot of organics, you know, mud and sticks and leaves and all of that. That's coming from some sort of flood event. It looked like tiramisu cake, right? You see that in your head, right? It's a layer cake, like a tiramisu. And you've got sand muck, sand muck, sand muck. Just in the few months it sits there, a year that sits there, some of those organics start to decompose and they have an oxygen demand. And when you flush through them and you re entrain all of those rotting organic materials, you can see a bump in oxygen demand and you spread that out. So there can be multiple sources that are impacting your oxygen demand in the downstream channel. Stanford [00:14:48]: Okay, so you both have reservoirs in your area of responsibility that do these pressure flushes. But while they are management actions, they're more structural management actions. You're trying to keep the gates clear. You're not actually moving a significant amount of sediment. John, I've heard you call these silt flushes. Why are they silt flushes? John [00:15:08]: The coarser sediments, the sands and gravels, tend to fall out of the river right when it hits the lake. So at the upstream end of the lake, they fall out in a delta, and then the finer sediments can make it all the way down to the dam. And so these actions, very often, if you're doing a pressure flush, will just be releasing silts in clays because that's the only sediment that made it all the way to the damn. Stanford [00:15:30]: So you won't actually touch most of the deposited material? John [00:15:34]: That's right. Well, it depends on. Paul [00:15:38]: Not yet. Right. Eventually, it'll be a sand flush, then it'll be a gravel flush, right? John [00:15:44]: That's right. Right. Stanford [00:15:45]: So we've talked a little bit about Spencer. Spencer Dam is on the Niobrara river, which feeds into the Missouri river. Can you just tell us a little bit about the history of Spencer Dam and why it was such great laboratory for so long? Paul [00:15:59]: Spencer Dam was put in 1920, as I recall, by a local power district. It ended up being owned by Nebraska Public Power District through 2019. And so it is a dam put on the Niobrara river. The Niobrara river, in that reach around the Missouri, is the number one contributor of sediment. It's about 60% of the sediment supply to the Missouri. And that reach, that reach is between Gavin's Point Dam and Fort Randall Dam. So the last two, it's part of the Missouri cascade and then on down the Missouri past John's house here in Kansas City, St. Louis, and all the way out. Stanford [00:16:35]: If you go look at some of the classic literature on sand transport, a surprising amount of that was done on the Niobrara river, all the way back to, like Colby. There's some really classical work on the Niobrara because it's such a classic sand bed river. Paul [00:16:49]: It is. It's a classic sand bed river. The sandhills of Nebraska extend roughly 300 miles up from the mouth. The Niobrara is spring fed, so it has a wonderful base flow condition that's at the mouth close to 1000 cfs, and it's a wonderful laboratory. It's like having the controls you had in the sediment flume in situ. Right. Because it's got this base flow, the material is predominantly fine sand, all in that 0.2 to 0.3 millimeter, like the local folks call it, sugar sand. For the listeners, pour a little table sugar out really close. It's almost that color, too. It's a really pale yellow so, Paul. Stanford [00:17:28]: Tell us what happens when you build a dam on a river that drains the sandhills of Nebraska. Paul [00:17:34]: Right. Well, going back to my hydraulics and fluid mechanics class, something about when you slow, the velocity, transport capacity goes down. It filled up very quickly. Long and short story, they put it in in 1920, and by the late 1940s, it was essentially full, and they had to do something about it. It does have a small hydropower facility. That was the whole intent of it. Starting in the early fifties, it just had kind of an ogee crest roller, uncontrolled spillway. They went in, did a major retrofit, and put a couple of tanner gates and then a lower gate in it, which was initially for ice. They'd cut chunks of ice away from the gates and then send it down this low gate, but they did. With that retrofit, they started doing a drawdown flush. It took a couple weeks, twice a year, and the whole goal there was just to recapture enough storage to have a little tiny area in front of the powerhouse so that you had clean water, sediment free water for hydropower. And it would take, with flushing, drawing down, and opening all the gates for about two weeks in the spring, April, 2 weeks, fall, October, you'd get enough to get you through to the next spring, and then that little area would fill up and you flush it again. And then it would fill up and you flush it again. So it's one of the few drawdown flushes probably in the US, with a very long and storied history. Starting in the 1950s and all the way up through 2019, that part of Nebraska experienced what was called a bomb cyclone event in early March of 2019. There was a lot of rain over frozen ground. I forget the exact date. That was the second week of March there. But they had an event so large, with so much ice in it, that it overtopped the earth. An embankment went through, knocked the gates out, and so it was a complete loss. There have been ongoing studies about where all that sediment's going. As you can guess, where it's going. Stanford [00:19:27]: To, say, unintentional dam removal. Paul [00:19:29]: It is, yes, it is an unscheduled dam removal at this point right now. Stanford [00:19:33]: Of a full reservoir? Paul [00:19:34]: Yes, of a full reservoir. Couple tens of thousands of acre feet of sediment. You know, it's tricky to determine how much of that is going to be fully re entrained. A lot of it is now perched and with the channel cutting down. But we're seeing the classic signs of a head cut migrating upstream. It was about a ten meter embankment, 30ft or so. And at the slope of that river, you know, we're starting to see bank caving. And some of that material come as far as about 10 miles upriver as the river reestablishes a new equilibrium grade after the failure. Stanford [00:20:07]: So several years ago, we identified this as essentially a scale laboratory for what flushing could be like in some of the bigger reservoirs. And so again, RSM, the regional sediment management program, who is also funding this podcast, funded a study where we just went and monitored and instrumented, and we worked with the USgs to actually measure several of these flushes. You've been on the ground several times for these flushes. What are some of the things that you've just learned from watching a bunch of these? Paul [00:20:35]: Sure. The number one thing would be, I think, the speed or the pace at which sediments are transported in a drawdown event that's often dictated by the cohesiveness of the materials. It's driven by that. But we do find things don't happen until they happen. And when they happen, they really happen. Right? And so we can stand there, and I know both of you have been on the catwalk over the gates with me there and watched it, and nothing's happening. Nothing's happening. And then you will have some sort of threshold, localized geomorphic threshold, just in one little spot. And the system will just fall apart and banks will begin to cave and a head cut will race up there. And that threshold, based on environmental conditions and what's happened the last six months, it's very hard to predict where that threshold is going to be. And it's very hard to predict how once that threshold exceeded that, the reservoir evolution of the headcut is going to develop. Stanford, you and I were up there for a couple years, and we saw the same approximate channel form in the same location two or three times. And then John and I said, you got to come up and see this thing. It's fantastic sandbox to learn about this. So we stood on the catwalk and I. And all my supreme knowledge, having been up there two or three times, I'm like, and now you're going to see that channel is going to form, like right there, just like it did the last three times we're up there. And about four minutes later, the channel goes that way completely, not the way that I just said it was going to go. And so there was some sort of local condition or some sort of anomaly in the deposition in the last six months that made it go this way instead of that. And that, you know, that makes it extremely hard to predict. However, when we're. We're looking at this as a management action. We're looking for the bulk response, I'm looking for total volume displaced. We get a lot of volume that moves downstream. And another observation is that deposited volume downstream seems to clear out of a system like that really quickly. Like by the time we're doing a flush, six months later, that whole downstream channel is back to the way it was. You'd never believe that after two weeks you've got a 6ft of sand deposited in these huge sandbars covering dozens of acres right below the dam. And you come back four months later, it's all gone. Stanford [00:23:03]: And we've got some great time lapse videos of this that we're gonna run on the podcast website. But I think the thing that I observe from these events is you talked about how there's like a layer cake. The stratigraphy of the reservoir tells the story of the inflows. John, one of the things you said I've been obsessed with my whole career is how sediment gradation changes as a function of flow. And you'd think that, like, larger flows transport larger grain classes, and they do, but they transport finer grain classes more. And so you have these silt and clay lenses and what's generally a very clean sand reservoir, and those things control. And so you'll actually blow out the sand down to this mud flat, and then it'll hang on the mud flat. The rate of scour through the mudflats determine where the channel is going to form. And it's very stochastic. I thought that was really interesting. So when I talked to Greg Morris, I asked him, why are you doing all the work you're doing outside of the US? And he said, because that's where they're doing reservoir management. And I asked him to describe reservoir management in the United States. And he used the word nascent, which I think is an excellent adjective. I'm sitting here with, I think, the two guys who are doing the most reservoir sediment management work in the United States. Why is it that we're not doing as much of this work as we're seeing in certainly Asia and Europe and South America? John [00:24:34]: You know, there are several reasons why. One of the reasons is we built so many dams. So many of them were large dams, and we built them all kind of a compressed couple of decades. And so we've enjoyed the benefit of these dams being large dams filling up with sediment, and it's happening underwater, and nobody really knows, nobody can see it for several decades now. I think part of it is just that now that we're 50 years into it and we're starting to see the impacts, we're starting to notice that the storage volumes have gone down quite a bit. Now we're just starting to care about it more. When it was, you know, a long ways away, the lakes were plenty big. We just didn't care as much. I think also in the intervening years we have the Clean Water Water act, which has a general paradigm with regards to sediment. That sediment is a pollutant. You've got to keep it out of the water, which to me is ridiculous. That means all rivers on earth are made out of pollution. And it doesn't have to be interpreted that way, certainly, but that is the way that a lot of government agencies have thought about sediment. Keep it out of the water. Keep it out of the water. Kind of going hand in hand with farm programs to help keep soil on the land, to keep the land productive has been the flip side of, oh, and we want to keep it out of the water because it's bad. Stanford [00:25:50]: So keeping it out of the water is a surrogate for keeping it on the land. But it also kind of ignores the sediment that's already in the water. John [00:25:57]: Right. The sediment that would naturally be flowing in the water. You know, all the various species of fish, they're used to a certain amount of sediment, and when you build a dam, it cuts that sediment load off. And so downstream from virtually every dam, you have unnaturally clear water. And the species have changed because of that. Many of the natives fish are completely absent now from the rivers downstream. Or at least the numbers are much, much reduced. Stanford [00:26:21]: This is counterintuitive. I think that especially anglers think about a nice, cold, clean fishery downstream of a dam being pristine, exactly what you'd want. How does turbid water advantage native species? John [00:26:36]: If you look at the native species thinking specifically on the Kansas river, they have little tiny eyes, whereas these sort of, the non native species have great big eyes. And if they have clear water, the non native species, they can see better, they can out compete or prey on the native species, whereas the native species, it's more of an olfactory. So more by smell do they navigate and they hide in the turbidity. And so they need that turbidity to survive and to compete. Stanford [00:27:02]: So if the water was historically turbid, then the native species are not going to rely on those senses as much. But then once the turbidity is lost, they lose their competitive advantage. John [00:27:16]: That's right. Sometimes it's also the interesting federal system of federal government owning maybe the flood control benefit. And by owning, that's the mission that the corps cares about, flood control navigation being primary oftentimes. And then it might be the local government who cares more about the water supply and the recreation. And because of that split, who's going to pay for something that's going to cost millions of dollars? Stanford [00:27:48]: So this was something that I basically learned once I started hanging out with you guys, is that you see a reservoir and you think, oh, that's a core reservoir, that's a bureau reservoir, that's a state reservoir. A lot of these reservoirs, different stakeholders own different portions of the pool. And depending on if you've got a coarse delta that is taking away the top of the pool, which would be like flood risk management or fines that are depositing in the bottom of the pool, which would be water supply, different stakeholders are impacted by sedimentation on different timescales and will be kind of alternately motivated to do that. But they all have to agree on the management practices. John [00:28:25]: That's, I guess one of the reason why there's problems getting these projects maybe funded is because sediment management might not compete well just for its ecosystem benefit. Even though there is an ecosystem benefit, it's not going to compete as well as actually doing a restoration project, planning a bunch of trees and whatnot. Sure, there's a flood risk management benefit, but it's not going to compete as well as raising a levy or something that's only for flood risk management. Being a multi use type project means unless you can count the benefits from multiple business lines and combine them, it may not actually end up scoring very well or rising to the top of people's concern for any one authorized purpose. Paul [00:29:04]: We've been going through an exercise at Lewis and Clark, which we'll talk about a little bit. On the economic justification of any sediment management action, we have revised wording indicating that the federal agencies need to use lifecycle economics to look at reservoir benefits, costs and future costs. Through this exercise, we've looked at a number of different models of how to calculate what the benefits are from the project. What are the costs that damage costs that have been incurred by the project, and what are the future costs that we're not taking into account in our traditional benefit cost ratio right now, one of the big pushes there is to consider, like what does the end of life scenario for a reservoir look like? Is it a decommissioning and a perpetual dam safety monitoring program for every year? And we just open the gates. Or on the opposite of that, is it a complete removal and some sort of decades long metering of 150 years of sediment that is deposited because you can't all let it go in a year that might be slightly above the clean water standard. A huge federal project 100 years in the future. What does it cost to do that? How do we capture that? Because if we don't manage now, we should be able to say if you don't do this in 100 years. You have a $1 billion dam removal and mitigation action you have to undertake. If you spread out that management and create some sort of as close as you can to equilibrium. And so either extend the life infinitely or at least extend the life significantly. We can mitigate that cost in the future. And the traditional model really doesn't consider that. What we're trying to do is we're not building new benefits. We're protecting existing benefits. Which that's not a new benefit that doesn't really fit into the model construct. And so this idea of life cycle economics is here are our benefits. That these are what they are right now. Stanford [00:31:11]: They're disappearing. Paul [00:31:12]: They are disappearing. And here is how they are disappearing. And eventually the hydropower goes away. And we lose the recreation. And we have to do something about the flood risk benefit and water supply and removing water intakes. So if we add all of those up. And we look honestly at the costs to do those things. So lifecycle kind of says, let's look at all those costs into the future. Let's be honest that there is a huge decommissioning or removal cost some time in the future. And let's bring that back to today and say, what do we need to spend to mitigate and eliminate that future cost? And then when you start talking about that, then you start talking in most large projects, real millions of dollars that could be spent annually to mitigate the billions in the future. Now, of course, our problem is everybody says, well, that billion in the future isn't a real cost right now. Tell me when it is. I'm like, well, it is a real cost because we're heading there. And one of the short sighted actions of our predecessors was not planning for sediment management. Had this been planned for? We're not having this discussion. It's kind of our responsibility to say, well, but we understand the shortfalls in the planning and design process. We have an opportunity to fix them. Because we're in kind of that transition. There was a building period. And now we're in a maintenance and losing loss, and then the next two generations are going to be in that you got to pay for everything with zero benefits. Tim Randall has alluded to this. He's kind of broken this down in a really thoughtful way in kind of like four different phases of reservoir life for these big projects. And we're in the transition and preparing for our kids and our grandkids to say, well, we've got to pay to do something, but there's no benefits left. John [00:32:59]: You also have to pay for replacing the benefits, like in the state of Kansas, where maybe two thirds of the population depends on these federal reservoirs for their water supply. And not only do you have to deal with a reservoir full of sediment, but how do you replace that absolutely critical need to supply, you know, water during, you know, drought periods? Some people say, well, you just, you're going to build a new lake at some point. Most of the good areas for lakes are taken. Certainly they're really, you know, areas for big lakes. So you could build smaller lakes. But if you were to build a new lake, I would hope that we could build in sediment management to the new lakes. And if you're going to do that, why not just manage the sediment in your existing lakes? Stanford [00:33:42]: So you guys have alluded to the fact that our generation of water engineers in the United States were playing a hand someone else dealt us. And it's not usually a great hand because sediment management wasn't something that was really foreseen. But around the world, people are building new dams. And the reason that the three of us are here together in Kansas City right now is that we're doing a reservoir management workshop in Southeast Asia. We've done several of these. And so what is your message to people who are building new dams? What are the lessons learned from kind of what we did as far as if you're going to put in a new dam, and with all the controversy that's surrounding that, what are your recommendations? John [00:34:25]: I guess it's like the Jacob Marley warning, don't do it the way we did it. Stanford [00:34:30]: The ghost of sediment future. John [00:34:32]: That's right. That's right. As you're building the dam, put the infrastructure in so you can manage the sediment. You may not feel like managing the sediment right away. It might be ten or 20 years before you want to start passing sediment. But if you have the gates down at the bottom of the lake, if you've got the conduit built for the hydro suction pipe or whatever you're going to do, if you build it in while you're building the dam. It's much, much less expensive. And you keep your flexibility and your options open. Stanford [00:34:59]: It's like $10 million now, $300 million in 30 years. One of the things that people often ask me when they kind of hear about this, this issue is, well, why don't we just dredge all that sediment? You know, the core is really good at dredging. We spend a third of our budget on dredging. Why don't we just kind of move that coastal infrastructure upland and dredge all the sediment? And actually, one of the things we talked to Greg about is that sometimes that could even be economically viable. John, can you tell us a little bit about how dredging has worked in Kansas? John [00:35:31]: Yeah. Back in 2015, the Kansas water office, so state agency paid for some dredging to be done at John Redmond Lake. They got out two and a half million cubic yards of sediment. But at a cost of about $20 million, it ended up being $6.7 a cubic yard. Everyone was excited because, like, wow, this was. This is actually a pretty good rate for dredging. Compared to other dredging that, you know, that's been done. That's a really good rate. Economy of scale. There's a lot of sediment. They took the sediment, they dredged it, and they put it in some confined disposal facilities. And everyone was happy and like, okay, hey, let's do round two. And they said, no, sorry, that's just too expensive. Even though the per cubic yard rate was. Was pretty good, it was just too expensive. They had to take out a bond for that $20 million and spread that over multiple years. So it wasn't economically sustainable. Stanford [00:36:22]: How much of the annual sediment load was removed? John [00:36:25]: They removed a little less than three years worth of sediment. So before the bond was due to be paid off, they had already refilled all that area that they had gained. And they just said, this is too expensive. So the area they dredged, that's a benefit, I guess, that carries on in the future, because they're always two and a half million tons more volume than they would have had. So it's not like it filled in, and now it was no use in doing it. But when you're trying to manage these very, very large sediment loads, you have to do something less expensive. Just totally break the bank if you tried to do that every single year. Stanford [00:36:57]: And so, Paul, you and I met over Gavin's point. Gavin's point is the smallest of the Missouri dams. But the Missouri dams are on a different scale. Paul [00:37:06]: It's all relative, right? It's the smallest one on the Missouri, which makes it about the 10th largest in the entire area core portfolio. Stanford [00:37:13]: And it's 30% full of sediment. And this is the project we met on. And every once in a while someone will ask you to cost out dredging for Gavin's point. And so you did that. How much would it cost just to hold serve? Just to remove the sediment that's coming in on an annual basis. Paul [00:37:32]: So we've got to determine what serve is. Holding serve in flow just in the lake, ignoring the deposits upstream to the Niobrara, there is just shy of 4 million cubic yards. So it's a little hard to wrap your head around. It's about 2600 acre feet. So if you want to try and put that into a context, football field's about an acre, 2600ft, about half a mile. So take a football field and make it half a mile tall, which is twice as tall as the Empire State Building. Per year. Per year. Every year. Stanford [00:38:08]: And this is just the trib load too. Paul [00:38:11]: This is just. Yes. Stanford [00:38:12]: Because there's a dam upstream on the Missouri. Paul [00:38:14]: Yes. So there are five more large dams on the system. And so Gavin's point, that football field half a mile high, is only two to 3% of the total sediment that's being trapped in the system. All of the rest of it is in huge reservoirs. Now the numbers are a lot larger for what's trapped, but they're exponentially larger reservoirs. So this is more of an acute problem there. And you're right, Stanford. Absolutely. There's one thing the core does well is dredge. Right. We've got huge infrastructure. We've got some of our own. We work with the largest contractors in the world to do this. Yeah. The question was posed if you wanted to hold, serve and say, protect. What we have at Gavin's point right now, which is 70 years of sediment deposited in there, about a third of the lake now filled. You want to keep all the rest of it. You still got 15 miles to the dam. So that gives you a relative distance there moving 4 million yards, 15 miles, 4 million yards. It's just not practical to put an upstream disposal area every year. Right. So upstream disposal has not been a consideration because in a few years you'd cover a couple counties along each side, a couple of feet deep. So downstream reintroduction, not disposal, it gets used kind of as a dirty term there. It's a disposal of sediment, not sediment that would have been there and will eventually. So we partnered with New Orleans District Corps of Engineers, who does very large both river and coastal projects to help us size a dredge project. To do this, they came up with a couple of different options. And the long and the short of it is, even at that huge economy of scale, you're looking at roughly 50 to 100 million in capital expenses just buying and building the system. And then you're looking at, best case scenario, $50 million annually, plus maintenance, plus escalation, plus that. And you're going to need 150 million the first year to get it off the ground. So we cannot find a way to say that there's $50 million of benefits a year to protect. In that case, we had Tim Randall on. We talked about Bureau of Reclamation's prize challenge on trying to find a better mousetrap to move sediments and all of that. So that has been one major area of focus across the federal agencies. Is there a better mousetrap that moves it at $2 a yard? If it does, then we're going to revisit that discussion. Stanford [00:40:50]: So, John, on smaller reservoirs, downstream release does seem to change the value proposition. So we wrote a paper, the three of us, with Roland Hotchkiss, who is one of the leaders in this, you've been associated with him on mill site dam. Can you tell us a little bit about what's going on there to change that value proposition? John [00:41:09]: Yeah. So mill site dam is the small irrigation reservoir, Mill site Lake. And what they're doing that's unique is rather than putting the sediment in a confined disposal facility somewhere on the land, is that they're actually discharging the sediment into the river downstream. They've done enough measurements upstream to know how much sediment is coming in, and so they know how much they can release in order to match the load, essentially removing the footprint of the dam and having sediment continuity from upstream to downstream. Now, it's not 100% that they're getting, but that's, you know, they're kind of limited to that. They are going to stay within the amount that's coming in from upstream. Stanford [00:41:47]: And so that's the construct of their permit is that if the dam wasn't there, this is the amount of sediment that would be in the river. And so it makes sense from every point of view to release the amount of sediment that is at least coming in. John [00:42:00]: That's right, yeah. And just by doing that, doing nothing else, no other changes to the technology at all. If all you do is pass the sediment downstream instead of try to dispose of it somewhere, quote unquote, then you could cut the cost in half. So if it's $7. Maybe it goes down to three and a half dollars. Or maybe it goes down from ten to five or. Or even more than cutting it in half. Some dredgers have told me that even more than half of the total cost is the disposal of the sediment. So if you can reuse the sediment by returning it to the natural system, you can save a lot of money. Stanford [00:42:33]: And so the complicated part of that is not the technology, but the permits. John [00:42:37]: In many cases. In some cases, it's the technology or it's the infrastructure. For instance, at Gavin's, the gates are not down at the bottom of the lake. And so that limits you in what you can do. You couldn't do a drawdown flush, for instance, because the gates are not at the bottom. If you drew all the way down, you would still have a lake. Whereas a lot of lakes in Kansas actually do have the gates down at the bottom. Tuttle Creek Lake, one that is been of special concern in the state of Kansas and the Kansas City district, has gates right at the bottom. And that's where the normal releases happen. So that opens up other opportunities for passing sediment less expensively. Stanford [00:43:14]: Well, let's talk about Tuttle. I think Tuttle's going to be the big legacy of your career. And Tuttle is where you're really taking some, I think, novel looks at how dredging could be improved. Why don't you tell us a little bit about the Tuttle story and then some of the technology that you're going to implement there? Yeah. John [00:43:31]: So Tuttle Creek Lake is filling up 6 million cubic yards per year. Is what's the sediment load filling up that lake. And as Paul said, this is year after year forever. If you were to manage it with traditional dredge and dispose on the land, that's just too expensive. A lot of the cost of dredging comes from the disposal, like I said. And another large cost of the dredging comes from all the energy you have to put in to entraining, sucking up the sediment, and then just transporting it for thousands of feet or multiple miles through a pipeline. What we will be doing in the near future is called water injection dredging. Instead of pumping sediment thousands of feet through a pipeline, we're going to pump clean water tens of feet from the dredge on top of the lake, just down into the reservoir sediment. And it fluidizes. It kind of resuspends a little bit the sediment on the bottom of the lake. But that sediment hugs the bottom of the lake and it flows downslope, just driven by gravity and a difference in density. This has been done in a lot of other situations. It's been done in ports and harbors, navigation channels a little bit in the United States, more extensively in Europe and elsewhere, but it's never been done anywhere in the world in LA. But we've done a lot of testing in Tuttle Creek Lake. And the sediment's good, the bed slope of the reservoir is good, the sediment is clean and so it's okay to put it downstream and the rivers free flow all the way to the ocean. The Tuttle Creek lake discharges to the big blue, which flows into the Kansas, to the Missouri, to the Mississippi, all the way to the ocean. So it's not just going to be moving it from one lake to another. Stanford [00:45:05]: Right. You're not just filling someone else's reservoir. John [00:45:07]: That's right. Stanford [00:45:08]: So we talked to Greg and George quite a bit about turbidity currents. Is this kind of like you're inducing a turbidity current? John [00:45:15]: That's exactly right, yep. Inducing a turbidity current in maybe places where it wouldn't usually have transported far enough to get all the way to the lake. By resuspending that sediment, I'm reforming a turbidity current, then we can get it to pass through the dam. Stanford [00:45:29]: You've done a lot of work on these deposits. They're like largely fine deposits. And you had an interesting insight about the cores and the depth density. Can you tell us a little bit about that and what you think the management implications are of some of the data you've taken from the stand? John [00:45:44]: Yeah, a few years back, it went out with Rob Wells from USDA and we collected ten foot deep cores in Tuttle Creek Lake, cores of sediment. And then we kind of sliced it up like hockey pucks. And we tested the erodibility and the bulk density and the critical shear stress of those different hockey pucks as we went deeper into the core. And the ones on top were very, very erodible. Low critical shear stress. Basically, you could probably blow on it and it would erode away. Right. But as you got a little bit deeper, it got much, much less erodible. It would take more force and then it would erode more slowly. Even once you started applying that shear stress. The implications are the longer you wait, the harder the sediment is to remove because it has time to consolidate and it becomes less erodible. You know, we haven't quantified it economically, but there is likely an economic benefit to managing as you go, rather than waiting for some big event in the future to do a whole bunch of. Stanford [00:46:42]: And so wading doesn't just make it expensive, because you're gonna have to do it all at once. But wading actually makes it more expensive because it makes it more difficult, just harder to erode. John [00:46:52]: And then as the sediment comes in, you can pass it downstream and you can say, yeah, I'm restoring sediment continuity. This kind of a natural condition. If you were gonna take five years worth of sediment and put it downstream, that wouldn't be natural in anybody's estimation. And so just the disposal then would be a very expensive activity. Stanford [00:47:09]: So we've written a paper, Shelley et al. The et al is the three of us, and Roland Hotchkiss. It's a series of case studies of reservoir sediment management in the United States. And this is something, John, that you really believed in. Can you tell us what was the main argument of the paper and the main reason you wanted to publish it? John [00:47:26]: Yeah, this actually started while I was sitting in Roland Hotchkiss's office. The RSM regional sediment management program had funded me to go out there and work with him on guidelines for reservoir sediment management on the environmental advisory board at the time, and was thinking about advice to give the corps, I guess, on how to manage our sediment. And he said, we should write a paper. And I thought, I think that paper's already been written. Other people have written papers with case studies on reservoir sediment management on multiple continents and whatnot, with a list of distinguished people that have visited all those. Stanford [00:48:02]: Lakes, some of those who've been on this podcast. Right. John [00:48:04]: Some of your previous guests here. But what hadn't really been done is focusing just on the United States because of that paradigm, that clean Water act keeps sediment out of the water. It's a pollutant. Many people don't even consider passing sediment to the downstream channel. Stanford [00:48:20]: They just don't think it's possible. John [00:48:22]: They don't even, in their brainstorming, no one even says that, because everybody knows you can't put sediment in water. And so because of that, they're leaving off the table what could be very cost effective, environmentally friendly options. And because the options that are left on the table are so expensive, they end up doing nothing. And so the purpose of the paper was to show that there are examples in the United States, even with the current permitting structures we have in place, where people are able to pass sediment, and they're doing it in ways that are environmentally permitted, environmentally sound and cost effective. And that was really the point, just to give people some good examples. So it could rise to people's levels of thinking and at least be a consideration when they're talking about sediment. Stanford [00:49:08]: The sediment management might be nascent in the United States, but it is nascent. That is something that is done. Permits are issued. People are doing it. John [00:49:16]: That's right. Stanford [00:49:17]: And so we had a series of case studies. We'll link to the paper in the notes. But I wrote about Fall Creek. There's a video that we'll link to in the notes. But let's just kind of wrap up with Paul. One of the projects we haven't talked about, but is one of the big annual drawdowns that happens in the United States is Guernsey. Can you tell us just a little bit about Guernsey? Paul [00:49:34]: Sure. Guernsey Reservoir in very southeastern Wyoming, it's a bureau of reclamation project. It is drawn down to manage sediment. It has advantageous physical properties to it as a fairly narrow, long, winding channel. Stanford [00:49:49]: Why is that advantageous? Paul [00:49:50]: Simply because you're trying to rechannelize a river. For example, Cherry Creek is a big, round circle of a reservoir. If you were to draw it down with those gates, you'd have this channel, but you'd have all these perch sediments that are not available to this cutting river channel. When we have a very narrow reservoir because of the surrounding geology to it, we've got all those sediments concentrated here, and we're cutting right down through them. And so what we would call our drawdown flush efficiency can be very high. The number of tons of sediment moved per unit water can be a lot higher in that kind of condition. And Guernsey is a little bit of an interesting situation. There's multiple benefits to doing the drawdown flush there. One is increasing storage capacity, moving some of that sediment downstream, whatever environmental benefits there are also to that. And right downstream of Guernsey is a very extensive irrigation project with a number of earthen canals, and that sediment is distributed out into those canals. And as those fines settle, there is a discussion and a general understanding by, you know, the irrigators that they feel there's a benefit there to almost kind of sealing their channels with this very fine material. And so when you add all of those benefits together, you have quite the justification to do a drawdown flush there. So they do that in every fall, try to temper, prepare those channels there. And that one's been very successful and been around for decades. Stanford [00:51:25]: And so I'm just reading from the paper here. The cost of moving the sediment is basically the lost hydropower, and it costs out at about 8.3 cents per cubic meter, as opposed to John Redmond, these numbers will be different because they're si $8.7 per cubic meter. So we're talking about two orders of magnitude if you can use the energy of the water. All right, John, any final thoughts on the paper? John [00:51:49]: Just the point of the paper is that people can do this. You don't have to let your reservoirs just fill up with sediment. You can pass sediment downstream. You might have to go to bat or go to war for your permit. Stanford [00:52:01]: Or form good collaborative relationships with the resource agencies, which you've done. John [00:52:05]: That's right. That's right. Yeah. So it could be harder, it could be easier. But you're going to have to jump through the hoops, do the collaboration with the other agencies that are concerned about the ecology downstream. That's. We are right. Absolutely. But you can do it. You can pass sediment downstream and manage your reservoirs long term going into the future. Stanford [00:52:25]: Well, doctor Pellboyd and doctor John Shelley, thanks for being on the podcast. John [00:52:30]: Thanks for having me. Paul [00:52:30]: You're most welcome. Stanford [00:52:34]: That was fun. I want to thank Paul and John for taking the time to share their experiences. We will be having really important interagency conversations about reservoirs and sediment in this country for like the next 30 to 50 years. They won't be easy conversations. They'll require some difficult trade offs. But I really appreciate how John and Paul have started to build the foundation for these conversations. Before I preview our next guest, there is one topic that we haven't covered, but that I've been thinking about. We have talked a little bit this season about how climate change will affect reservoir sediment management. But we haven't talked yet about how reservoir sediment management could impact climate change. While dams and reservoirs have well documented costs and impacts, they also represent an important and frankly underrated resource in our race to decarbonization, particularly in the intermediate timescale of the next 30 to 50 years, while other decarbonization technologies are ramping up to scale. So extending the life and reducing the morphological impact of this infrastructure that is already generating hundreds of terawatts of carbon free electricity per year, not only has the long term supply imperative that we talked to George about, but has a critical intermediate time scale role to play at least as a multi decadal stopgap for decarbonization. But we are at the point in the design, life and life cycles with some dams where the costs, risks and impacts do outweigh the benefits. And so the final chapter of our reservoir sediment mini season surrounds the question, what do you do with all the sediment in a reservoir when a community is done with a dam and wants to decommission it. But they haven't been using these techniques to manage it all along. And so we're left with decades or in many cases over a century of sediment to manage, but now over a relatively short period of time. To tackle that question, we'll talk to Jennifer Boundfree with the US Bureau of Reclamation, who co authored an incredibly helpful interagency report on dam removal triage and was involved in the largest anthropogenic dam removal in history. In the meantime, I'll be posting some of the videos of actual reservoir sediment management actions that we talked about in the episode on the podcast website, which is linked in the episode notes. These are informal conversations and the views expressed do not necessarily reflect the official positions of the US Army Corps of Engineers or the districts or center of the guests or host. This initiative was funded by the regional sediment Management R and D program led by Dave Perkey. Mikeretto edited this season and wrote the music. Thanks for listening.