Stanford [00:00:12]:
Well, we've come to the end of our reservoir centric mini season and it seemed fitting to me to wrap up with the final stage of the reservoir sediment life cycle. Dam removable and while it often makes sense to extend the usable life and benefits of reservoirs and to minimize their impact with wise sediment management, a lot of these structures are old enough that they no longer longer meet their intended purpose or the impacts and the liabilities outweigh the benefits, particularly in the face of shifting public values. And when it comes time to decommission a dam, everyone is suddenly interested in sediment transport. This week's guest is particularly well situated.

Stanford [00:00:49]:
To talk about dam removal because she's.

Stanford [00:00:51]:
Not only worked as a technical sediment transport practitioner on the world's largest dam removal, but she's also been instrumental in multiple massive meta analyses that survey the impacts and costs of of dam removals and triage, the analyses that these different classes of projects require. Jennifer boundary works at the Technical Service center of the US Bureau of Reclamation in Denver, Colorado, where she leads the sediment and River Hydraulics branch. In her career with the service center, which is essentially the bureau's in house center of expertise, Jennifer worked on the LwA dam removal and an interagency guidance document that I have bookmarked in my browser because it is my go to recommendation when communities or analysts reach out to me with questions about dam removal analysis. In this interview, we talk about all that, plus the recent interagency work. She's involved in trying to identify and predict the cost drivers of different types of decommissioning projects. I'm Stanford Gibson, the sediment transport specialist at the course Hydrologic Engineering center. And this week on the final episode of our Reservoir Sediment mini season, a conversation about dam removal with Jennifer Boundtree.

Stanford [00:01:56]:
Jennifer Bountry, welcome to the podcast.

Jennifer [00:01:58]:
Thanks. It's great to be here.

Stanford [00:01:59]:
So how did you get involved in river science?

Jennifer [00:02:02]:
Wow, that's a long journey in my story. I grew up in St. Louis and it was on the Mississippi river, so I was used to being on a big river and water skiing and hanging out, you know, as a high schooler in the river.

Stanford [00:02:14]:
You water skied on the Mississippi?

Jennifer [00:02:15]:
Yeah. It's kind of terrifying, actually, when you fall and then you're wondering like what big fish or eel is lurking underneath.

Stanford [00:02:23]:
I didn't know that was a thing.

Jennifer [00:02:24]:
Yeah, yeah. And barges are going by, so it's a whole different skill. Yeah. I went to school for a couple years in Illinois and found myself out in Colorado and graduated with business marketing degree.

Stanford [00:02:36]:
Oh wow. So we're not there yet.

Jennifer [00:02:38]:
So we're not there yet. And, you know, did the whole four year college and then just kind of like the creative side of it, but didn't really love the idea of sales or going to a big city or something like that. So I got super lucky because I started working for this guy, Doctor Maury Albertson, and he kind of crossed paths with engineering and business, which was pretty unusual back then. And he had an office out at this place called the Engineering Research center at Colorado State, where they did all kinds of cool river science.

Stanford [00:03:07]:
Yeah.

Jennifer [00:03:07]:
And I ended up in an office doing sustainable development stuff next to all these people who are doing cool river work. And, I don't know, light bulb went on and totally changed my journey. Went back to school, got my undergrad master's, went through all that process at Colorado State.

Stanford [00:03:22]:
Oh, you started over?

Jennifer [00:03:23]:
Started over. There's very few classes that transfer from a business degree to engineering, but it was worth it, like, to get that fundamental knowledge. And I haven't looked back, and it's just been great. This is my 25 year anniversary, reclamation this month.

Stanford [00:03:38]:
So when was the first time that it occurred to you that dams weren't just something that we might build, but something that we might have to unbuild? Is there like a moment that, that clicked for you?

Jennifer [00:03:49]:
Yeah, pretty early in my career. I'd been here a few years, and my mentor, Washington doctor Tim Randall at the time, and he had been working on this thing called the all while dam removal, and I hadn't heard of it. And he said, well, we don't know when it's going to happen, but we need to start figuring out some of the issues for people who live downstream of the dam. There's some wells, there's some people that live in the floodplain, and we need to start studying this to figure out what we're going to do about them, because the way the river works is going to change. And so that was my real first exposure to, wow, what does it look like when you might pluck out a dam and how might the river change? And it was so fascinating to think about, especially on the elwha, because it was going to completely open up the entire watershed there and change the dynamics.

Stanford [00:04:30]:
So we're going to get to the lwa, but let's start with just how many dams are in the United States.

Jennifer [00:04:36]:
So in the United States, we have something called the national inventory of dams, and there's about 90,000 plus dams in that inventory. And what's interesting is there's millions more small dams that aren't even documented in there.

Stanford [00:04:49]:
What would it take for a dam to get in there? Is there like a criteria?

Jennifer [00:04:52]:
Yeah. For dams to be in the national inventory database, they have to either be 25ft in height and have 15 acre feet of storage, or they have to be 6ft tall and have 50 acre feet or more of storage. So in the real large dams we tend to get real excited about, they're not very common in there, but they do exist.

Stanford [00:05:11]:
But those are pretty sizable structures or impoundments. And so there's many, many small dams that aren't included in that.

Jennifer [00:05:17]:
Yeah, there's dams all over, especially in the east, that just are off the record, off the radar.

Stanford [00:05:22]:
So 90,000 that fit those criteria. And how many dams have we removed in this country?

Jennifer [00:05:29]:
American rivers is a nonprofit, and they have this awesome documentation where they've been trying to keep track of that since 1912, and they've tracked a little over 2000 dams that have been removed.

Stanford [00:05:39]:
What are the most common motivations? Like why do people remove dams?

Jennifer [00:05:43]:
Yeah, it's interesting. We tend to think of dam removal as really exciting and something that is going to help the environment. But one thing to remember is that the dams that have been removed are only about one and a half percent of all the dams we have in the nation. So just want to start by saying that our dams are really vital infrastructure, and they're still super important to the way the US functions and has a healthy lifestyle to deliver water, help protect us from flooding, provide recreation, all those things. But what usually happens is the dams no longer serve their purpose or can be replaced by another facility to meet their needs, or the environmental benefits to remove that dam outweigh the cost to maintain it. Perhaps if there's a dam safety issue or something that would just be really expensive, incorporating fish passage, something like that.

Stanford [00:06:31]:
One of the things that I've heard is that a lot of times, a lot of the attention goes to the environmental benefits, but the thing that really drives the removal is liability, that some of these structures are considered dangerous. And so that's what actually brings the money to the table in the partnership. Is that something you found?

Jennifer [00:06:47]:
Yeah, it can, and it just really varies. Each dam kind of has its own personality. But sometimes the dams, like the El wa dams we were talking about, one of those did have some dam safety issues. They didn't have fish passage. So those things can really add up. And it's really an economic decision, often for those dam owners, of what makes the best future for them. Can they afford to fix the dam and maintain it, or does it just make more sense to decommission that structure?

Stanford [00:07:13]:
And they will, like, leverage that with nonprofits, and the multiple benefits will add up to say, hey, it's economically feasible to take this out.

Jennifer [00:07:20]:
Yeah, it really can. It depends if it's federally owned or private. And the nonprofits, you know, have their own set of criteria and analysis that they use to figure out where they're going to invest.

Stanford [00:07:29]:
Most common obstacles to dam removal kind.

Jennifer [00:07:32]:
Of, as we were talking about, one of the biggest obstacles for dam removal is really that decision whether to remove the dam or not. It's not clear cut in many cases. Sometimes the dam is abandoned, and we just don't have anyone who wants to take ownership of that to even figure out what it would take to decommission other times. It's a really tough decision for people who have operated that dam for a long time, people landowners who are used to living around the facility, using the lakes that exist behind the dam. So it's really hard to figure out when it makes sense to remove the dam. So that's kind of the first obstacle. And then once that decision is made, it becomes sometimes complicated to figure out what needs to be done besides just removing the dam itself. So there can be roads, bridges, people who live along the lake, there can be really sensitive habitat downstream, all kinds of reasons that just kind of have to slow down and think about what would it look like to take out the dam? And is there anything we need to do about that to help make sure it's safe and we don't have any unexpected outcomes?

Stanford [00:08:30]:
So this is the River Mechanics podcast, where we spend a lot of time talking about sediment. I like to think that maybe we're the only podcast that does that, but this is the fourth episode in a mini season on reservoir sedimentation management. And so we've talked a lot about how over the course of time, a lot of sediment can build up behind these dams. And so how often is the potential fate of the sediment behind the dam one of the big obstacles or discussion points for the stakeholders?

Jennifer [00:09:01]:
It's a great question. I think every dam removal that I've been involved in, sediment is one of the first things people ask about. Sometimes people don't even know how much sediment is behind the dam. So that's where we start to see if sediment really is an issue or if it's just perceived to be a problem. And it's really important to kind of slow down when it is present to understand how much is there and what it looks like.

Stanford [00:09:23]:
Okay, so several times a year, I'll get a call or an email from an engineer or planner or someone who finds themselves in a community that's talking about taking out a dam. And one of the first things they want to talk about is the sediment behind the dam. That's why they're calling me. And the first thing I do is I recommend Randall and Bountree 2017, which is this document that you and Tim Randall, who you mentioned, authored this. But you really built this document with a broad interagency team. And I'll link to this on the podcast. But can you just tell us about the dam removal analysis guidelines for sediment, this interagency document, how did that come to be?

Jennifer [00:10:00]:
Yeah, it's such an interesting question. We started hearing that no matter where you were, no matter what size the dam was, no matter how much sediment was there, everyone was having to do a lot of analysis, the same level of modeling, tons of data collection. It was really expensive and it was really time consuming. And we had been involved in some dams that really made sense, but some of the other ones were really small and could barely even find any sediment behind the dam. So we started wondering, was there something we could create that would help educate and share some lessons? We thought if we could get together all the experts in the field who've been doing dam removals all across the US, maybe we could unpack what was happening and where sediment was an issue and where it wasn't. And we could put together a framework that would help have the conversations with regulators who require those sediment analysis to understand when it really was needed or when you might get like a path. We got together people from academia, private consultants, different federal agencies, tribes that have been involved. We got together on both the west coast and the east coast. It became quickly apparent that things were different and we needed to really understand why.

Stanford [00:11:06]:
Okay, so in my experience, you know, if I get in front of a group stakeholders, it's like what you described. The 800 pound gorilla in the room is the sediment behind the dam. And what I find is in almost any room of stakeholders, there's some people who think that the sediment's just not a problem. And then there's another group of people that thinks, you know, this is a PhD project that needs to happen. And we need to do a bunch of numerical and physical modeling, and we need to do a lot of work to evaluate the risks to downstream ecosystems. And I think one of the most important things from those meetings and that document. One of the reasons is the first thing I have people look at is that your finding is that really not all dam removals are created equal. There isn't a one size fits all analysis package. So what are the steps you recommend to determine what a responsible level of sediment analysis is for a dam removal?

Jennifer [00:11:52]:
Yeah, it's a great question. And it was really interesting when the light bulb went in in the workshops, because the first thing you know, of course, was to figure out, do you even have sediment there? That was super important. And we talked a lot about what's the best data collection methods. You know, do you probe, do you go out with big boats and do you do surveys? How do you figure all of that out? And we shared those experiences and then we realized, well, it's not just that you have sediment there, it's about what happens to the fate of that sediment when the dam is removed. And does anyone care? And so we came up with the idea of risk similar to what engineers do in the field of dam safety, and that became a really important way to look at it. Risk is probability times consequence. So the probability was really founded on how much sediment you had there and how does it compare to what the river can handle or an average annual load? And then we take that times consequence of, does anyone care about the impacts if the sediment were released? And that's a conversation with stakeholders. That's a really important part of the process that isn't just running a model, but actually going out and talking to the folks who may be affected by the dam removal and then looking at how the risk may be very small or very high.

Stanford [00:13:03]:
All right, well, let's take those in turn. Probability and consequence come together to form risk. Let's look at consequence. What are some of the consequences that you want to account for or look for downstream of the dam or upstream of the dam?

Jennifer [00:13:17]:
Yeah, it actually turned out to be both. So the first area people thought about was the reservoir. You know, was it just going to be a big, stinking mud flat? That was a common perception. And would that look like that for. And it was particularly important to people who were used to growing up around the lake, kayaking, fishing, boating on it, using that area. And then the next part, looking downstream, what might happen? Where was the sediment going to go? If it were sands and gravels, it might drop out on the river, raise it up and cause more flooding. So that was a really important concern. If it got into the water column and affected water quality, it could affect people who are diverting water for cities for irrigation, any of those features. And then lastly, environmental trying to pay careful attention to what species were using the river downstream and how they might be affected by that sediment.

Stanford [00:14:05]:
And so then let's go back to the probability. You said that the probability of an impact is really tied to how much sediment and maybe what kind of sediment is there. You know, one of the things, the episode we did just a couple before this with Greg Morris, he said that the question of how big is a reservoir is a question that doesn't make sense outside of context. You know, the question is how big is the reservoir relative to the watershed, or how big is the reservoir relative to the flow? And that's really the tack that you took because it's a similar question. How much sediment is in the reservoir is a contextual question. I think that's one of the most important things that I want people to get from that document you had. Can you tell us a little bit about how you decided to answer the question, is there a lot of sediment?

Jennifer [00:14:46]:
Yeah, it's so important because if you took the same amount of sediment and put it on a big river like the Mississippi where I grew up, versus a small creek up in the hills of a watershed, that river is going to be able to handle that sediment very differently. So it really became important to look at the relative reservoir sediment volume. And by that we mean taking the volume of sediment that's trapped in the reservoir and comparing it to the average annual load that the river would move. And rivers are different every year. Some years they have floods and some years they're in drought. But if you think about is there one year of sediment and then the reservoir filled up, or is it like decades of sediment, it's going to make a big difference in how well that river can handle it. And if you need to do anything different, when you remove the dam to slow it down.

Stanford [00:15:32]:
So if the dam has been there 50 years, it seems like there would be 50 years of sediment, but not necessarily how would different dams fall into those different categories?

Jennifer [00:15:42]:
Yeah, it's a perfect question. You really have to get a conceptual model. Where is your dam? Are there other dams upstream that might be trapping sediment already that otherwise would come to your reservoir? Or are you in a really productive watershed that has had a wildfire and you might even be in excess of a typical load? So getting out there and measuring the sediment is one of the fundamental first steps you have to do to make any headway on figuring out how you might remove the dam and whether sediment is going to be a big deal or a little deal.

Stanford [00:16:11]:
And then with small dams, small dams might have become run of river many, many years ago.

Jennifer [00:16:16]:
Right. A lot of our smaller dams, they might fill in one storm. It might have some short term impacts. And those can be really important, especially if you have like an important fish spawning area downstream or a water intake immediately downstream. But it's important to look at the context that that might not last a very long time. And maybe there's something we can do with the project to mitigate that and get through that short term impact and then kind of move forward as opposed to something that might have a chronic restoration of sediment load that's going to change the way the river performs, you know, downstream of the dam for many years to come.

Stanford [00:16:49]:
So a giant reservoir with no reservoirs upstream of it and I stationary land use, if it was 50 years old, it would have 50 years of sediment in it. But a small reservoir might have filled river after ten years. And so even though it's been there 50 years, it could have ten years of sediment in it.

Jennifer [00:17:06]:
Exactly. So you can't just assume that the life of the dam is the amount of sediment that you have in the reservoir by going out and doing those early measurements and then comparing that to the average annual load either. If you haven't measured, that's amazing. It doesn't happen in very many places, but there's also some really great ways described in the guideline that we can compute that and then start to get a feel for. How does that compare to the energy or the stream power of the river, what we call it?

Stanford [00:17:32]:
Okay, so we have a number that is how much sediment is in the dam. It's a number that is x times the annual load. So we have ten years of annual load or we have 50 years of annual load or we have 80 years of annual load. What do we do with that number to determine the risk level and the level of analysis that we should do with the damage?

Jennifer [00:17:51]:
Yeah, that's where the really kind of creative, fun part comes. One thing that really surprised me was early on, we were really focused on really big dams, big reservoirs that had decades of sediment because those have a lot of impacts and have to be carefully modeled and understood before anything happens with the dam removal. But what came up a lot, because so many of the dams that have been removed are small, it became really important for the group that working on the guideline to identify what do you do if you don't have a lot of sediment? So they came up with this idea that the probability of sediment impact could actually be negligible. It didn't just start with small, medium, large and this negligible. We talked about it like a 1040 easy form on your taxes. So you kind of just get a pass. Like you don't have a lot going on. You shouldn't have to do a lot of data collection analysis, but you have to have some due diligence to say, yeah, like maybe this, a little bit of sediment is kind of equal to what a gravel bar might be downstream. There's just not a lot to worry about there. You need to focus more on the construction aspects of the dam removal. But then if you aren't negligible, then you have to decide if you're in that small, medium or large based on the years of sediment. And then we actually created a matrix where it becomes qualitative. But you're starting to contrast it with those consequences. And it might matter if you're talking about consequences from fine sediment that affect like water quality or water treatment or if you're talking about sands and gravels, coarse sediment that's going to affect flooding or where the river might migrate to and cause some bank erosion or even where the fate of it's going to be. I mean, how fast it'll get there. So the fine sediment will get there pretty quickly to the next downstream reservoir coast, wherever it might be. Sands and gravels might create a new beach area or could have negative confluences. That's the really interesting part about doing that matrix and cross comparing and then you figure out your risk. So you could have a lot of sediment but not a lot of consequences people are worried about. So you're going to wind up in a lower risk or you could have not very much sediment, but there's some really sensitive areas or very high profile, controversial questions downstream. So you might end up in a higher risk with a lot less sediment.

Stanford [00:19:59]:
I think about maybe the prototypical example of a lot of sediment. Low risk would be maybe Marmot dam, where it was just this giant dam in the middle, the Oregon wilderness. There was a lot of sediment, but there weren't a lot of potential impacts downstream.

Jennifer [00:20:13]:
Yeah, that's right. So it went into a canyon where the sediment was quickly moved by the river and processed. Not a lot of land use or homeowners in that canyon. So that was really nice. They did have some small temporary impacts where it did hit the confluence of the river downstream and it was a fishing area, but that was easily mitigated with some dredging. So that was a great example with, with a lot of sediment, low consequences.

Stanford [00:20:37]:
I think on the other side of that, I've been periodically involved in a dam removal analysis where it's a series of small dams and each one of them probably has less than ten years of sediment, maybe even less than five years of sediment in it, but it goes right through an urban corridor. So multiple dams with what we would call low to moderate probability, but because the consequences are so high, if all of that sediment were to pile up in the same place and it was to induce flooding, the consequences could be high. So it moves it from this low risk process from a physical standpoint to actually a more higher risk category.

Jennifer [00:21:13]:
Yeah, San Clemente dam was like that, and so a fair amount of sediment, but it just wasn't an option to release that sediment downstream, given the communities that were down there. And the impacts would just be too high. So they actually stabilized all of the sediment, even with the dam removal, and rerouted the river to a new path to avoid releasing all that sediment. So urban areas can be really tricky and it is really important to not have any unintended consequences. So that's the whole idea of the risk analysis, is not just assuming based on your sediment, what might happen and what the risks are to really look at who cares and who might be affected. And is there anything we can do about it to help reduce that?

Stanford [00:21:53]:
Well, let's actually ask that question. Is there anything we can do about it? If you're in a situation where you have more sediment than you can release downstream in a pulse, what are the different management alternatives to take out a dam in one of those kind of high risk settings?

Jennifer [00:22:07]:
Yeah, it's a great question. The first thing you can look at is how much of that sediment in the reservoir is actually going to be released.

Stanford [00:22:15]:
That's a great point.

Jennifer [00:22:16]:
So sometimes the reservoirs are much wider than the natural river will be once the dam is gone. And so some of that sediment may be left behind, but it really depends on how erodible that sediment is and how big the flood is that's going to be passing through that reservoir. So that's one of the first things we do is, is 100% of that sediment going to make it and actually cause those impacts we're worried about?

Stanford [00:22:37]:
That's the opposite of what we've been talking about with sluicing and drought on flushing, is that if you do a drought on flush, sometimes you want to mobilize more sediment and you can't because it's limited to the channel. On the flip side of the dam removal, sometimes it's a benefit. It's a benefit. Okay. So the first thing you want to do is determine how much of the sediment actually is going to mobilize.

Jennifer [00:23:01]:
Yeah. And how quickly then, then you can look at, we encourage people to use the guidelines in a way of, let's kind of assume you're going to pull that dam instantaneously. And as much sentiment as you think is going to get out, is going to get out pretty quickly. That's kind of our bookend, you know, what might happen. And if that's not acceptable, then it's time to circle back and look at those strategies that you're talking about. So we can certainly mitigate some of the sediment in the reservoir itself that can be dredged and removed. There can be ways to move the sediment to different floodplains or, you know, restore areas that have been lost. We can also look at a slower dam removal. And that's a really important strategy that had been used on some of the bigger dams where we need to just slow down that pace of how fast the dam is going to drop down and the river is going to activate that sediment and push it downstream. And that can be done, you know, over a period of days, weeks, months or even years like we did on the Ah river.

Stanford [00:23:57]:
So like a series of partial dam removals vertically.

Jennifer [00:24:00]:
Yeah. So you can exactly notch the dam and then pause, give the river some time to process the sediment and then do another. And whatever that rate is, is dependent on each site. Each one kind of has its own optimum. That's where those modeling tasks really come in handy.

Stanford [00:24:16]:
So you said that sometimes when the river erodes, it will leave terraces. What are the considerations for those terraces and vegetation? Are there things that we have to think about?

Jennifer [00:24:26]:
Definitely. We don't typically want a lot of uncontrolled banks left behind. So, for example, in the Eloah river, that delta that was in place was 90ft thick. And if the river had just down, cut through 90ft quickly with an instantaneous dam removal, would have left unstable banks. And that can mean uncontrolled sediment release for many decades to come. And that's really hard to mitigate or control what might happen for downstream impacts. So by understanding in your conceptual model with your vegetation experts how those terraces might be left and then how vegetation might be reestablished on those, can actually come up with a new landscape. It's never going to look exactly like it was predium, but something that meets the goals for that area to control how much sediment would or would not come out the vegetation has been a really huge part of dam removals that I didn't recognize, you know, early on in my career. But what we learned is that you have to out compete the invasives when you create that new, fresh surface. The invasives are really good, even in national parks, at running to that spot.

Stanford [00:25:30]:
They're early succession. Invasives tend to be early succession, yes.

Jennifer [00:25:33]:
And so working with those vegetation experts to get the right cocktail of vegetation out there on the surface and making sure it's successful and out competing those invasives is one of the most important parts.

Stanford [00:25:44]:
And what about mechanical removal? It's something I hear from time to time. It sounds really expensive.

Jennifer [00:25:50]:
It can be or it cannot. It really depends.

Stanford [00:25:52]:
Can you just tell us what it is?

Jennifer [00:25:54]:
Mechanical removal is when you're using construction equipment to actually dredge out sediment from the former reservoir once the water is dropped down. So you're taking the sediment out of the equation. So it's not going to be released. You're usually taking it somewhere off site, or at least in an area where the river is no longer going to access. So putting it in a safe new home so that it doesn't become part of the issue. Sometimes if the area is hard to access access, it can be really expensive and you have to take the sediment a long distance. That can be really expensive. But if it's a small site, it's pretty accessible. Sometimes people taking out the dams have their own equipment, and then it can actually kind of be no big deal to mess around with the sediment and put it in a position. Pilot channels are really common. A lot of times people are concerned where the river might land once the dam is gone and they would like to put it back in the place it was prior to the dam or in some safer location away from landowners. And so that can be a great tool, not dredging all the sediment, but just kind of kickstarting the river in a place where you want it to be.

Stanford [00:26:58]:
One of the most challenging dam removals I've worked on is upstream of an alluvial fan, and it's been there for 100 years. The river channel has incised and a community is built around that river channel. And so even if we went in and removed all the sediment, just taking the dam out and reestablishing sediment continuity would cause a flooding hazard downstream. And so there's a really potential problematic situation. But at the very least, it has to start with mechanical removal.

Jennifer [00:27:25]:
Yeah. And that's a fast track way to help recreate a landscape that doesn't have to deal with how much sediment might be eroded or released or what it might look like. And particularly on some of our smaller dams, it can definitely be incorporated with not a lot of cost. The other thing we can do is consider partial dam removals. Sometimes we can not lower the dam all the way down to the original bed to get some of the benefits of a dam removal, but actually trap some of the sediment and keep it behind. Or we can leave some of the sides of the dam, some of the.

Stanford [00:27:54]:
Lateral pieces you take control of where the channel is going to form and what's going to be left behind.

Jennifer [00:27:59]:
Yeah. So not all dam removals look the same. It doesn't have to always be a complete and full removal. Sometimes that's important, but in other cases we like to leave some. Sometimes their historical structures. Preservation for the local community can be a really interesting part that they're the. They're excited about.

Stanford [00:28:15]:
So I was at the first workshop that you did to develop this document, and I think the thing that I really took away from that is my imagination around dam removal had been formed by these, like, big headliners. What I learned from that meeting is all the east coast people who've been taking out hundreds of dams and they're small, and this whole idea that a dam removal could have a negligible sediment impact, and if so, then we should scale the analysis to the potential risk. That really changed my paradigm, is the reason why I recommend anyone working on dam removal start with this document.

Jennifer [00:28:49]:
That's great. Yeah. It was so eye opening to work with the east coast folks because we had been working on these really exciting big salmon passage dam removals in the west, but the east coast ones, what was so interesting about those is they're removing so many in a watershed together, and that wasn't something we had to deal with as much on the west coast. So they're having to look at the whole watershed picture and removing numerous dams to restore a stretch of stream. And they had a lot of unknown history, too. There might be a lot of buried structures. That dam may have had five versions before the one that we see on top of the river. And it was kind of like an exciting mystery when you start to go in and remove the dam. I remember one I was on, there actually was old plates in China, and there was more of that than rocks and gravels.

Stanford [00:29:38]:
I wonder, what's the critical shear stress of China?

Jennifer [00:29:41]:
I'm such a nerd. I was totally wondering that. I was like, how would I model that? You know, it's kind of angular, but it's very flat.

Stanford [00:29:49]:
It's got good lift.

Jennifer [00:29:51]:
I was actually totally excited about that research to come.

Stanford [00:29:55]:
So like states like Wisconsin or Pennsylvania have taken out a lot of dams and part of what I think is that damn removal builds confidence for damn removal, especially on that low to negligible risk. Because you can develop intuition for comps. If you don't have intuition for comps, every damn removal seems incredibly risky. But once you've done a number of them in similar settings, then folks who have done that have intuition that this is going to be a negligible impact and we can do this at a low cost.

Jennifer [00:30:24]:
Yeah, Pennsylvania had a really neat setup with their fish and game commission and we'd still always recommend that you do your due diligence and measure, go out there and ensure that you don't actually have a lot of sediment. It's negligible. But once you get to that phase, they were able to actually out of the Fish and Game commission, which is often one of regulators that were so worried about with dam removal in terms of managing the conversation about RSM and impacts, ok or not. But since they were actually had their eye on the prize and they were the ones pro on the dam removal, they were able to have those conversations as a community of people working on the dam removal. They all got used to doing it together and they knew what the expectations were, so they were able to get through the process a lot quicker. And I think that shared experience was what made them so successful.

Stanford [00:31:13]:
That has been one of my surprises is that you think about dam removal as something that's often a benefit to ecology and fish and wildlife, but often it's the resource communities that are. They're hesitant and that it seems like the states that are really doing it well. There's been a real collaboration there.

Jennifer [00:31:31]:
It's true, and it's hard because there's a lot of change over in staff. But that's one of the best things you can do if you can start to build those relationships with people you might not normally spend time with as a technical modeler, it really does pay off because they probably are concerned for a reason, for a good reason. Their regulatory processes are in place for a reason. But we also want to be careful not to push the conversation into. Well, why not? There's a lot of stories. Mill Town dam removal is a great example where they had a super fun site, a lot of contaminants, and dam removal was just off the table for years. But eventually they were able to have the conversation well what would it look like? What if we do take out the bad sentiment and then are able to restore the river back to something that will benefit the environment? So they got through that process, but it doesn't happen overnight. And it takes special people to have patience and actually just have an open, objective mind to exchange.

Stanford [00:32:30]:
I've been on a proposal to monitor dam removal, and we put in this proposal every year for five years, expecting that the dam was going to come out this year. And every year we won the proposal. And then by mid year we'd have to go back and say, yeah, this isn't going to happen this year. I think one of the lessons is that it takes patience.

Jennifer [00:32:49]:
Yeah, it really does. And it's such a unique problem. One of the things that we've been learning is like that the understanding of the project to actually figure out what the dam removal will look like is a lot more than traditional projects where you might be able to, you know, repeat, repeat, repeat cookie cutter style. So it just takes a lot more knowledge up front to really have those conversations.

Stanford [00:33:10]:
Is that because, like, you were saying that each of these dams and reservoirs has their own personality?

Jennifer [00:33:15]:
Yeah, absolutely. Like, we, you know, have to really figure out what's going on with that site, with that sediment, with that water, and then who lives around that facility, which is different in every case. And what, what's important to that?

Stanford [00:33:29]:
So kind of back to the big headlining dam removals. I understand that one of the ways you got into dam removals was through your involvement with the largest anthropogenic dam removal in history, the Elwa. Can you just tell us a little bit about that project and, like, what are some of the things that you learned from that?

Jennifer [00:33:47]:
Yeah, the Elwa was absolutely a project of a lifetime, just a pleasure and an opportunity for me to be on the. It was a project with two dams. So the first dam, elbow dam, was 32 meters high. It was built in 1913 by a canadian entrepreneur who wanted to bring power to the local town of Fort Angeles, Washington. And that was really important at the time because power was hard to come by. And they built the dam illegally because all dams are supposed to have fish passage.

Stanford [00:34:15]:
Oh, I didn't know that.

Jennifer [00:34:16]:
They just said, well, we'll just build you a hatchery and all will be well. Of course, you know, that can have a lot of helpful pieces to fisheries, but can also be challenging. So once they had the first dam built, not too much longer, in the 1920s, they said, we might as well build another one for hydropower. And they built Klein's canyon dam further upstream, which was 64 meters high, twice the size. Twice the size and more power. So people were pretty excited about that. But for decades, the lower Elo Colum tribe that lives at the mouth of the Elwha river was super impacted, of course, by the dams and the loss of the salmon fishery and to the coastal ecosystem there. So they fought and finally got a congressional act called the Elwa act to come up with the idea to not only remove the dams, but to fully restore the Elwha river ecosystem, which was a tall order. Nothing had been done with dams of this size. And there was a lot of sediment, 21 million cubic yards of sediment, 21 million to figure out. So it was before my time. But Bill Jackson, Gary Smiley, Tim Randall at reclamation had some meetings of the mines and just decided there was absolutely no way it would be affordable to dredge that sentiment. It was up in Olympic National park, too far to take and just too much of it. We mentioned earlier it was 90ft thick.

Stanford [00:35:35]:
In Klein's Canyon dam, the delta, the sediment deposit is 90ft thick.

Jennifer [00:35:39]:
Yeah. So how to remove that dam, what to do with all that sediment was pretty important. So that project sponsored by the lower Ellcom tribe took decades. But finally, in 2011 to 2014, the dam removals happened.

Stanford [00:35:52]:
What are some of the processes you observed that either you expected and maybe more interestingly, we were surprised by?

Jennifer [00:36:00]:
Yeah, it was a really interesting approach to a project because the team knew there was uncertainty. It was clear that dam removal couldn't happen instantaneously with that tabletop exercise we were talking about earlier. So the team did the best available numerical modeling at the time. They did physical models and played with how fast can we remove the dam or how slow and how much pause do we need after each of those increments to give the river some time to move that sediment around? What would that look like? And then what was really neat is working with USGS. They did a field experiment in the nineties and actually drew the reservoir down 18ft and then watched to see if what they thought would happen and what.

Stanford [00:36:38]:
Oh, wow. So was that drawdown enough to pass sediment downstream or just looking at how it would rearrange in the delta?

Jennifer [00:36:44]:
Yeah, it didn't pass any sediment downstream, just rearranging it in the delta. But because that delta was 90ft thick, that represented a lot of the sediment. A really important step. Because the idea was by using phase dam removal, some of it would be left behind to become vegetated, part of the new natural landscape, and some of it would be moved downstream. So that was really cool.

Stanford [00:37:04]:
You know, we talked about how on the negligible side of things that, you know, sometimes people do too much analysis, I think on these big dam removals, physical modeling, and then these prototype experiments are really essential to kind of understand how this particular dam is going to process a sediment.

Jennifer [00:37:21]:
Yeah. And it gets back to that risk idea and the sediment analysis guidelines. If you're not in those negligible or low risk cases, you not only need to do one type of analysis or model, you need to do several lines of evidence because all of them are going to have their shortcoming. But by having the miracle model, the field experiment, the physical model, we learned something from all of those. And then we went in and said, okay, we're going to come up with 15 foot increments and there's going to be pauses of two weeks. And in some cases we actually had pauses of six to eight weeks because they wanted everything to slow down when fish were moving in and out of the river. So that was called fish windows. And we knew even with all that great science and planning, there was still uncertainty. And because it took so long for the dam removal to actually start, more sediment came in. That's what I like. I think the most challenging thing in dam removal is it's always like three years out, three years out for these big projects, and then things change.

Stanford [00:38:16]:
If it took ten years, that's going to be an additional 1015 percent, huh?

Jennifer [00:38:20]:
Yeah. Especially because we had a really big flood. So it totally brought in a lot more sediment and changed all of the predictions that we had.

Stanford [00:38:29]:
The probability and risk equation.

Jennifer [00:38:30]:
Yeah, we had to update everything because there was mitigation built downstream. So we had to make sure it could still handle the new sediment loads or if we had to completely change. But what we did there was what we call adaptive management. The idea was to do real time monitoring and predictions all throughout the dam removal. And if something got off track from what we thought would happen, we'd have to stop, regroup and change our tactics.

Stanford [00:38:51]:
So you had a learning feedback loop. Yes. Where you did the best predictive work you could do. But these processes are chaotic. They're not fully understanding. Yeah, right, right. It was totally novel. And so you wanted to make sure you were in a learning loop.

Jennifer [00:39:06]:
Yeah, that was incredibly valuable. And Andy Ritchie was our project geomorphologist on the project. He came up with something called plane Cam. So we didn't even have enough budget for aerial photos or lidar back then. And he came up with this way to stick a camera on a local plane. They would fly weekly and he could turn around real time aerial images and topographic surfaces that we were able to use. It was one of the best contributions of the whole project.

Stanford [00:39:30]:
What are some processes that might have happened faster or slower than you had initially expected?

Jennifer [00:39:36]:
Yeah, it was such a whirlwind to think back. And for a time the reservoir was still in place. So we're watching the dams are being lowered, sediment is moving, as we expected, rivers moving it farther downstream with each notch of the dam going down.

Stanford [00:39:50]:
Still in the reservoir, still largely being.

Jennifer [00:39:53]:
Trapped in the reservoir. Elwha dam came out in about eight months. And remember the upper dam, Klein's canyon, took three years. So Elwha had the first point where sediment reached the dam and started being released downstream. And there was definitely an impact, and we could see it filling pools downstream fines on the floodplain. But it wasn't anything that even came close to exceeding levees that have been built, water treatment plants that have been built to handle more sediment. So it was just kind of like this slow pause, getting really, like, fascinated by watching how a piece of wood might change the river path or how a certain, you know, amount of drawdown would affect where the river went. And what was really cool at that time was just the scales of everything was the same. So you could be looking at the big Elwha river or a little tributary coming in, and they were all doing this incision and whitening incision and widening and braiding processes. But then at Klein's Canyon dam, it was really close to the point where we'd almost lost the reservoir. And it was just kind of like your big moment. Right? Like, what do we do? Do we go with the planned amount of dam removal? Do we wait and see what happens? You know, we wanted it to be some kind of pop to get the process started.

Stanford [00:41:02]:
Yeah.

Jennifer [00:41:02]:
And that's when things got really exciting really fast, because you can't stop it at that point. Once you've lost that control of the reservoir, all you can do is stop for the removal of the dam. But the river kept going. So that was one of the biggest surprises, is we did get a lot of sediment downstream. There was no floods that happened during the dam removal period, and the river quickly aggraded, and it just was amazing. It had so much energy to process all that sediment, but it did take time.

Stanford [00:41:32]:
Yeah.

Jennifer [00:41:32]:
And one of the most surprising things, there was so much wood in organics that it actually clogged up the water treatment, diversion and infrastructure downstream. So that had been designed for fine sediment loads, silts and clays. But actually the river built up enough that wood, sticks, leaves, gravel, sands all went into the infrastructure and buried it. And at that point we had to stop dam removal. We didn't know for how long it ended up being a year.

Stanford [00:42:01]:
And in a year it eroded back down.

Jennifer [00:42:03]:
It eroded back down, not to the original, when we hadn't released any sediment, but they also built new pumps that they could supply the water while the river was still dealing with all this sediment, wood and debris. It was tough, especially with all of the crane sitting there. It's a pretty tricky spot to be, to be sure that you're ready to start up again. And what was really interesting was because the river in Lake Mills, which was above glines canyon dam, hadn't reached the valley bottom. We think of rivers to have like an equilibrium slope based on the resistance of what they're running over, if it's bedrock or cobbles and what size sediment is in that river. And it just flattened and flattened and flattened. So even though we had stopped the removal of the dam, the river kept flattening. So it kept eroding and processing and contributing more sediment. So it was good we had the year, but that's kind of why we needed so much time. It just didn't stop altogether.

Stanford [00:42:55]:
And so it's been a little while now. What's the kind of medium to long term result?

Jennifer [00:43:01]:
It was really interesting question because we were asked for legal reasons for mitigation, just in terms of when the project of the actual river process side of the house was done, when are the effects of dam removal over? Every week you could go up and see another grain of sand, see the river move around. And so we started thinking about, well, how do you reach a new normal? When are we just at. It's a river from a rainforest, lots of sediment production, it gets lots of floods. When are we back to what we just might see in a normal El wall river? So we set about trying to figure that out and it ended up being, from the original predictions that we knew after dam removal stopped, that we needed a few floods, particularly because we didn't have any floods during dam removal. So the river cuts down and it widens to build this new floodplain. But you need those floods to really get that floodplain development process, even if the incision has completed. So those floods are really helpful. They took a few years and that's when we were able to say, yeah, it's going to continue to get another grain of sediment to move around in the former landscape. But we think we're kind of back to a new normal. And what's really still happening is the fishery side of the story is still being developed because fish, you know, have such a longer span of recovery. They're still out there learning and monitoring.

Stanford [00:44:16]:
And telling that story over the last 20 years. The thing that surprised me about dam removals is kind of how quickly they get to a new normal. I think that 20 years ago, when we were kind of first imagining these, I thought that it was going to be a massive long term impact. But most of these rivers do metabolize the sediment and get to a new normal relatively quickly.

Jennifer [00:44:37]:
Yeah, it was impressing, especially on a river that has a lot of flow. And I think that made a difference. So we were able to take advantage of that. And that was always the idea is let the river do the work.

Stanford [00:44:51]:
So let's talk a little bit about some of the new work you're doing with some other investigators. One of the questions that people are going to have is just how much does a dam removal cost? So let's say that I'm a stakeholder and I have a dam with liability issues and ecological impacts, and I want to know how much it's going to cost to remove it. Without giving you any other information, what's the range of, like, what's the, what's the range I could be looking at?

Jennifer [00:45:15]:
I'd say tell me more because it could be anywhere from a few thousand dollars to hundreds of millions. So it really does matter how big that dam is, where it's at, and what kind of characteristics are going on in that setting. And each dam removal, as we've been talking about, has its own personality. So you really have to have a longer conversation.

Stanford [00:45:36]:
Let's stratify this a little bit. I got these numbers from one of your papers, but let's break it into three categories. What percent of dam removals are, say, less than half a million and then half a million to 5 million and then greater than 5 million. What percent fits in those buckets?

Jennifer [00:45:52]:
So just as a reminder, we had like 2000 dams that have been removed according to american rivers. And so we work with USGS, Oregon State and the core and we worked to create a database and we had about a third of those dam removals where we could actually find out what the total cost was. So of those numbers from that sample, yeah, 600 plus sites. The way those broke out is about 68% were less than half a million, 26% were between half a million and 5,000,006% were in that elite category of over 5 million.

Stanford [00:46:26]:
So we're talking about 6% in that elite category. But do you have a sense of what percentage of expenditure is in that category?

Jennifer [00:46:35]:
Yeah. So for the database we have with those 600 plus sites, it's worth about one and a half billion dollars of dam removal. And those over 5 million represent about 80% of that total expenditure. So big dams cost a lot.

Stanford [00:46:48]:
This is one of the things I really liked about that paper, is that it showed me just how much my imagination is skewed here. So you've got 6% of dam removals accounting for 80% of the cost.

Jennifer [00:46:59]:
Yeah. It's really was eye opening that so many of the dams that have been removed are not big dams, high dollar. And so we tend to get our eyes focused on those because they have a lot of studies around them and that's where we're looking. But, yeah, a lot of dams are taken out much cheaper.

Stanford [00:47:15]:
It's an availability bias, essentially.

Jennifer [00:47:17]:
Yeah.

Stanford [00:47:18]:
Okay, so you and this interagency team did this work to look at what are some of the cost drivers of the removals? What are some of the big costs that someone like me, who's, you know, worked on some dam removals but hasn't internalized this big picture? What are some of the costs that I might not predict?

Jennifer [00:47:33]:
We had to, like, really scratch our heads to figure out how are we going to answer that question team at all worked on dam removals, so we had some ideas of what might be big cost drivers, but we wanted to really see what the data showed us to see if we were being biased as towards some of the bigger, more complex projects or what really did drive up those costs. So we tackled it from three angles. We had this big database with like 666 cases. And we knew the total cost, but we didn't really know what made up that total cost because that information wasn't available. But we did identify which of those studies had certain drivers. And so we came up with ideas like sediment would probably be a big cost driver. You had to like, use cofferdams to manage the care of the stream. Can't work in the wet. Things like, are there contaminants there in habitat, you know, where there were threatened, endangered species? So we knew if they were kind of presence absence. And then we said, well, what can we do to take a deeper dive so we can really kind of test some of those ideas? So Doctor Desiree Tollis at Oregon State helped us do interviews with students, people who had been the leads on dam removals, to take that total cost number and break it down. And one of the things that I hadn't thought about was we tend to really focus on the construction just to get the dam out or what might be happening. But also, you know, what kind of cost is it for all that analysis and all that planning that happens before you get to the point where you're taking the dam out? We knew that litigation could actually be a cost driver and stakeholder attention studies. The more people that are worried about consequences, the more you might have to do to test in addition to the construction. And then the third angle was looking at just the construction bids. So we were able for a few studies to actually get the construction bids and say, what did they actually do? Some of the things that came up there were river restoration features, managing the sediment, vegetation, how hard it was to access, or, you know, that engineering means and methods. So that's kind of how we tackled it. And one of the biggest surprises was vegetation.

Stanford [00:49:32]:
Oh, really?

Jennifer [00:49:32]:
So we talked about that earlier that, you know, trying to out compete those invasives. But when you have a pretty big open landscape, that's pretty big manual effort and sometimes it requires helicopters, sometimes they're hard to get to. You have to grow your own seed. So it's just the amount of resources needed to make sure that that's successful was one of the bigger costs. That was really a surprise to me.

Stanford [00:49:53]:
So in the paper I read you and the team, desiree and yeah, Jeff.

Jennifer [00:49:56]:
Duda from USGS and Kyle McKay from the corps of Engineers and Tim Randall.

Stanford [00:50:01]:
And Al Jansen, your team put together kind of a flowchart with some cost inflection points. I'll say, what are some of those inflection points that actually drove a significant difference in cost in these projects?

Jennifer [00:50:13]:
Yeah, well, this story started with getting a postdoc doctor Suman Jumani used machine learning and she said, you know, it's great that you all have these hypotheses, you know, in sediment, vegetation or cover dams, but, you know, I really want to see what the data shows us, you know, in this machine learning environment. And that's where these flowcharts came out. So the word created these ways to quickly go through a scoping level cost if. Okay, you're that person you asked me about in the beginning, and you maybe know a little bit more about your site, your typical flow, you know, how high your dam is, and maybe you can at least have a tabletop exercise of. Do you think you have a lot of sediment. Do you think it's hard to access? Do you think people care if the dam's gone or not? Kind of walk through some of those and those would be the cost drivers. And where is your dam? Is it in the west, south, east? And does that matter? So this flowchart takes you through those steps. And what was really cool is it helps you bend order magnitude so you can kind of quickly find out. The most expensive dams are the really tall ones with really big flows, and the cheapest are the really small ones with no cost drivers. And then there's a lot in the middle that it makes a difference if you have contaminants, if you are in the west versus the east and those where those all start to play out. So that was really exciting. And she made this thing called a shiny app, too. You can actually go on the web and plug in your data and it can pop out some of the results.

Stanford [00:51:35]:
And it will slot you in like you're in the hundreds of thousands and the single digit millions, the tens of millions, the hundreds of millions put you in that continuum kind of will help.

Jennifer [00:51:44]:
The team or the resource managers know, do we want to keep talking about this? That's right.

Stanford [00:51:47]:
That's right. That's a screening level tool. So that sounds like a great tool. If you were to start with this app or this algorithm and decide, hey, we're still in the ballgame, our community is still in the conversation here, where would you go next to actually determine some sort of cost estimate? That is fantasy.

Jennifer [00:52:04]:
The next important step in the dam removal cost estimating. I learned so much from Al Jansen. Our cost estimating group here is to really do the engineering means and methods. So you have to start thinking about, are you going to take the dam out instantaneously? Is it hard to access? Are you going to need to use a coffer dam? What are the components? Do you need to replace that dam with infrastructure to replace the function of the dam, like a stream side pumping or something like that? And are you going to need river restoration features to help mitigate the impacts of the dam removal? And we built a spreadsheet tool Tim Randall and Al Jansen helped with and myself. And based on those construction bid deep dives, and you can go in there and start to think about unit costs. It matters, interestingly, what state you're in that can affect your unit costs. And that's how you get to reduce the uncertainty and kind of move forward along with your cost estimate.

Stanford [00:52:56]:
All right, well, we'll make sure to link to all those tools in the podcast website.

Jennifer [00:53:00]:
Oh, that'd be great.

Stanford [00:53:01]:
Jennifer Bontree, thank you so much for being on the podcast.

Jennifer [00:53:04]:
Thank you so much. It's been quite a pleasure.

Stanford [00:53:09]:
And that'll wrap up our mini second season on reservoir sediment management. We'll be back in a couple months with a full third season. The third season is mostly recorded. We're just in production. And frankly, I'm really excited about it. In the third season, I kind of branched out from federal practitioners. And worked in more conversations with professors and academics. I kind of transitioned from mainly inviting people I know onto the podcast.

Stanford [00:53:30]:
To reaching out to people I've always wanted to talk to.

Stanford [00:53:33]:
Whether we've met or not. And it's been a lot of fun. We'll be kicking off season three. With conversations with David Montgomery from the University of Washington. And Marcel Garcia from the University of Illinois. And we'll have a full eight to ten episode run. Starting kind of late fall, early winter. Until then, if you have guest or topic recommendations. I have a Google forum on the podcast website. Which is linked in the episode notes. Or feel free to reach out to me on LinkedIn. And speaking of the podcast website. We're posting some remarkable videos of reservoir management activities up there. Including some excerpts from these interviews. And links to the documents and tools we talked about this season. It's worth checking out. These are informal conversations. And the views expressed. Do not necessarily reflect the effect position. Of the US Army Corps of Engineers, the Bureau of Reclamation. Or the offices or centers of the guests or hosts. This initiative is funded by the Regional Sediment Management program. As part of their RSM U tech transfer initiative. Mike Loretto edited this season and wrote the theme music. I'm Stanford Gibson. Thanks for listening.