Podcast: Remaking water infrastructure

[Opening banter]

Jim: Welcome to Blue Sky podcast, season one, episode X, a little production we’re calling Enter the Water Biome. I’m your host Jim Lynch.

Nicole: And I’m Nicole Casal Moore.

Jim: In this episode we’re introducing a project based on the University of Michigan’s Department of Civil and Environmental Engineering: “Remaking Water Infrastructure.” For a layman, lay person like myself, when I hear “water system,” I think of pipes and pumps and treatment plants—a whole network that delivers water to the taps in our homes. But for this project, we’re going to expand our mindset a bit. We’re going to dive into the world of microbial communities known as microbiomes.

Nicole: Ooh, the microbiome! Seriously, microbiomes are amazing. All these little bacteria inside your body, mostly helping you out. Did you know that you actually have more bacteria cells in your body than human cells?

Jim: I did not.

Nicole: We’re chimeras.

Jim: What’s a chimera?

Nicole: An organism made of multiple different organisms.

Jim: Honestly, I’m not sure how to feel about that. That said, it’s definitely fascinating in its own creepy way. When I interviewed Lutgarde Raskin, the professor of civil and environmental engineering, leading this particular project, she was upfront about that aspect—the mixed reactions microbiomes usually get.

Lut Raskin: You know, I think we would like to think that there’s no microbes in the water when it comes out of our tap, right? But there are many microbes in our water and most of them are perfectly fine, but there are some that we don’t want to be there. So when we talk about “microbiome,” we’re talking about the microbes in the water. You know, when you open your faucet what is in there?

Nicole: Is it just the microbes in the water?

Jim: No, in that portion of our interview, she’s walking us through the wide world of microbes, from the ones in water sources like lakes and rivers, to the ones you find in your own gut—those that contribute to the sewer system eventually, I might add—and all the pipes in between.

Nicole: That is a wide world.

Jim: And she’s not just talking about bacteria either team is looking at viruses and protozoa too.

Nicole: Protozoa. Thank you for reminding me about their existence.

Jim: Basically your single celled animals. The bad ones for humans, at least, are the parasites.

Nicole: So is this project primarily about the bugs that make us sick? I asked this cause here in the US we don’t often hear about people getting sick from their water.

Jim: Not often, but it does happen though. And when it does, it can be a pretty big deal. Part of the water crisis in Flint was a Legionella outbreak, and that killed at least 12 people in 2014-’15. But it’s not just about pathogens like that. Professor Raskin’s project is looking at two big things we can do for our water systems by treating microbes as partners, rather than our adversaries.

Nicole: Which makes sense because that’s how it works in our own bodies.

Jim: True, true. The long-term ideal here is to manage microbes, manage them throughout the system to get safe water without having to use chlorine. Now in the near term, Lut thinks it could be possible to—read microbial clues, is one way of putting it, as a way to predict and diagnose problems that you find in your water infrastructure.

Nicole: So microbes could be like sentinels. Is she saying that if we use Flint as an example, we might have been able to predict or diagnose the lead situation, just based on the microbes?

Jim: Yeah, that’s what she’s saying.

Nicole: How exactly does that work?

Jim: So in Flint, the focus wasn’t on microbes, it was corrosion and how corrosion contributed to having lead dissolved into the water. And that was the main focus in Flint, but those metals like lead and copper and iron influence which microbes thrive and develop and so on. And so you have these interactions with metals or with biofilms or with corrosion scales. All of that in my definition of microbiome is included.

Nicole: This word, biofilm-

Jim: Well, it’s actually sort of a strength in numbers approach. You’ve got colonies here, that are sort of grouped together, and then they create their own fortress—a layer that sort of protects them from outside influence and allows them to sort of thrive where they are.

Nicole: So if we’re monitoring the microbes, we could maybe have caught the lead contamination sooner it sounds like?

Jim: Exactly. A surge at a certain microbe population could be like that canary in the coal mine. It’s one of those ideas she’s trying to explore with her Blue Sky project. But they’re going for even more on this project. Lut argues that by monitoring microbes, we could be a lot smarter about fixing our infrastructure when problems crop up.

Jim: Around the world these days, people are trying to tackle infrastructure problems, dealing with the aging pipelines and the aging delivery systems. How is that a solution to this? How do you envision focusing on microbial biomes as a way to deal with the aging infrastructure?

Lut Raskin: Well, oftentimes when you have events, when things go wrong, when mains break or when there’s low pressure events, or when there’s cracks in the system and we lose so much water, our main concern is water quality. And microbial water quality is one of the main concerns. It’s not the only concern but it’s certainly one of the main concerns. That’s why we have that boil water advisory right now, as we speak.

Nicole: Is there a boil water advisory?

Jim: No, not right now. There was in Detroit at the time Lut and I talked.

Nicole: I see.

Lut Raskin: Our idea is that by developing these real-time monitoring tools, if we see something in a particular part of the distribution system that’s unusual, or that gives us a reason for concern, that could be due to an infrastructure problem. And so we would then be really able to pinpoint and solve that infrastructure problem right there, rather than, you know, waiting until the main actually breaks. Usually there is an event before it breaks, right? So you might get some intrusion or you might start picking up those problems.

Nicole: We’ve established that microbes can tell you if there’s a problem developing.

Jim: We have. They’re proposing micro monitoring as a way to help us be proactive with the water systems. Right now, all we really do is react after something bad happens. You get a break, you figure out where it is, and then you replace the whole pipe. Professor Raskin thinks we can do better than that. In a proactive system, you’d start seeing signs of a problem before the break actually occurs. And then when you go to fix it, you might be able to focus just on the damaged section instead of having to replace the whole line.

Nicole: So by knowing what microbes are in the water, you can tell which part of a pipe is compromised?

Jim: Sure. If you have a baseline of what’s supposed to be in the microbiome, and all of a sudden you start seeing a spike of some sort, you know, a certain spike of organisms might tip you off, like, to whether that is the deterioration of a service line or something like that.

Nicole: Okay, I get it. That sounds like a big deal, especially for cash strapped communities that have these old water pipes, but don’t have the money to replace them all. So what would that look like—spotting an infrastructure problem through the microbes?

Jim: I wondered the same thing. Let’s let her explain.

Jim: Let’s, for the sake of argument, if you’re monitoring the biome in say a 60-, 70-year-old pipeline, what would you see pre-break?

Lut Raskin: So there is a baseline microbiome that we know is there and we can characterize, and we can learn what that should be in perfectly fine water quality without any concerns. And when you start seeing changes—one thing, you could see an increase in the total number of microbes that you might detect, or you might start picking up microbes that you definitely don’t want to be there. You might start seeing some of these pathogenic microbes show up in those water samples.

Nicole: If they see more microbes or bad microbes-

Jim: Yeah, they know there was some kind of problem.

Nicole: Removing all judgment about these microbes—how exactly are they going to detect these microbes and count them and identify them?

Jim: Oh, you hit on the biggest catch at this stage: They don’t have a way to constantly keep tabs on microbes right now.

Nicole: I assume that’s something they’re working on as part of this project?

Jim: Absolutely. The big hurdle is how fast can they get analysis results? With current technology, you could get a read on certain pathogens with about an hour of turnaround time. And that’s not exactly what Professor Raskin has as an idea when she talks about real time monitoring.

Nicole: How do they find out what’s in the water?

Jim: Well, what they’re looking at is DNA sequencing. It’s the sort of technology you see on every CSI episode.

Nicole: Really? They’re recognizing microbes by their genomes in an hour? It seems to take a lot longer than that on television.

Jim: They can see what kind of a microbe it is in that amount of time, but they can’t get the particular strain. The DNA sequencing isn’t accurate enough to know whether it’s something that can make people sick.

Nicole: Isn’t knowing it’s there enough to make this work?

Jim: Not quite. E. coli is a good example. Most E. coli is fine. It’s in your colon, helps you digest food. It’s all good. But when you read about an E. coli outbreak in the news, it’s because they’re dealing with one of the bad strains of E. coli called a pathogenic strain.

Nicole: If the one-hour DNA sequencers can’t tell the difference between friendly E. coli and an E. coli that might send you to the hospital, that’s a problem. Are they working on more accurate sequencing?

Jim: Oh, it’s interesting. What they’re working on is a way to give the big data treatment to sequencing results. And it’s way out of my wheelhouse. So I’m gonna go back to the professor on this one.

Lut Raskin: If you sequence a piece of DNA many times, and you use that information with computational algorithms, you could probably get a better estimate of what the real sequence was as opposed to when you sequence it one time, even when the error rate is still relatively high. Those are the strategies that we’re working on to bring those sequencing technologies that can give us data very quickly to that level that we need to identify microbes.

Nicole: They’re going to put DNA sequencing sensors in the water system that can tell you within an hour whether you have a pathogenic bacteria—is that right?

Jim: Yup.

Nicole: And then instead of sequencing a DNA sample once, they’re going to do it multiple times and use algorithms to extrapolate the results?

Jim: Yes.

Nicole: And that’s how they’ll get current sensor technology accurate and fast enough to make this project work.

Jim: That’s one of the long-term goals here—improving the accuracy. They don’t have technology that does exactly what they want it to do as quickly as they wanted to do right here and right now. They’re sort of taking their sampling, their DNA sequencing sensors, and they’re putting it together with algorithms to try and make what we have right now more effective. But even now they can take sort of proxy measurements that give a measure of bacterial activity.

Nicole: And you’re gonna need to explain that. You can’t get away with proxy measurements.

Jim: Let’s start with chlorine. Already it’s something you’ll find in city water from the disinfection process. In the United States we keep it in the water on purpose. So if your chlorine level is lower than expected, that means it’s being used up somewhere in the system to kill bacteria.

Lut Raskin: For example, you could monitor chlorine on a continuous basis all the time. And if you have a dip in your chlorine concentration, perhaps that suggests that maybe there’s a higher concentration of microbes in that system—that the chlorine was being consumed because chlorine was trying to act the way it’s supposed to act.

Nicole: So they can start right now by monitoring the chlorine and then move into identifying bacteria through the DNA later in the project?

Jim: Absolutely. That’s their plan.

Lut Raskin: Assume for a minute now that we have the technology available to real time get all this microbiome data. So one of the challenges we have is we don’t have good risk models. The ideal scenario is that you have no pathogenic microbes in your water. But that’s unrealistic.

Jim: An organization like the EPA has set clean water rules, has set certain levels for certain pathogens. Are you saying we don’t have them for everything we need to have?

Lut Raskin: We actually don’t have that. The EPA tells us to use indicator organisms to monitor water quality because it’s too difficult in many cases to monitor for those real pathogens. And we don’t often have the risk data to tell us what the levels of those pathogens should be.

Nicole: I gotta stop you for a second here. So the environmental protection agency’s clean water laws don’t actually say how much of a pathogen is allowed in the water?

Jim: Yeah. I’m guessing that that’s common knowledge for people in the water treatment business, but it just sounds wrong. Doesn’t it?

Nicole: Yes, it’s frightening.

Jim: According to Lut, they’re not even measuring the pathogens, but instead, water providers look for what are called indicator organisms. They’re like microbes that can be identified quickly without the need for the big time consuming analysis. Things tend to show up when pathogens are around. It’s the way Detroit eventually decided when to call off that boiled water advisory.

Nicole: I don’t know if I like knowing this.

Jim: I’m hearing it for the first time thinking, “Really? We do not have regulated levels of these pathogens?” What they do these days is instead the water providers look for what they call indicator organisms. They’re microbes that they can identify quickly, maybe easier than for them to identify than the pathogens that we’re mostly concerned about. But it comes without the need for the time consuming analysis that those pathogens might have with them.

Nicole: I see, so we’re like looking for their sidekicks.

Lut Raskin: So what Detroit is doing is they are monitoring for indicator organisms right now. And if those indicators are zero, then they assume that the water is safe. We do this all the time.

Jim: Every large system?

Lut Raskin: Every large and small system. So what we would like to do is monitor the real pathogens and then develop risk models to help us decide how low we need to be to have an acceptable risk level.

Jim: You need that as your baseline?

Lut Raskin: We absolutely need that, yes.

Nicole: So figuring out how much of a pathogen you can have in the water without making people sick is a key piece of this project. Are they starting from exactly zero or is there anything they know about maybe the scariest pathogens?

Jim: Well, they’re not starting from zero, but they’re close. Let me have Lut tell you about one of the pathogens that they have some good baseline information on already.

Lut Raskin: So Legionella, because it’s such a common problem, I feel like we are in a pretty good shape. We have a lot of information available and there is this move towards applying that information in our water systems. So we’re in a good position with Legionella or we’re getting there.

Nicole: It sounded like they were saying they only look for these indicator microbes, the sidekicks, but now she’s saying that they have some data for Legionella. So is there like a specific list of pathogens that they do look for directly? Or can you just help me understand what the lay of the land is?

Jim: I think what she’s referring to there is that she’s talking about things like Legionella that historically have provided us a lot of data. There are situations that have cropped up, which have been high profile and sort of forced them, you know, the scientific community, to get a grasp on. There are other ones that have not necessarily impacted larger numbers of people over the years. They can still be harmful, but the research just hasn’t been done on it.

Lut Raskin: There’s a group of organisms that we call nontuberculous mycobacteria, for example.

Nicole: What do we call them? Can you say that slower?

Jim: Nontuberculous mycobacteria, I think that’s right.

Nicole: Most of those syllables sound problematic to me.

Jim: You find them in water and you find them in soil. So for most of us, it’s not problematic. Our immune systems can take care of them. But the concern about these things is on the rise worldwide, especially for older people.

Nicole: So what happens to people who are susceptible to these bacteria?

Jim: Yeah, it’s not good. It’s called nontuberculous, but the length of the treatment is a lot like getting tuberculosis. It’s a year spent on antibiotics.

Nicole: Yikes. So they’re starting with trying to figure out what level of Legionella is okay. And then moving on to these less familiar bacteria, like these nontuberculosis mycobacteria?

Jim: Yeah you nailed it. Seriously first time I ever-

Nicole: I did not nail that.

Jim: Yeah, that’s my impression. Basically they’re going to start with something they know a little bit better, Legionella, and then move on to these other things.

Nicole: They’ve got a really broad scope on this project. It’s impressive. She’s talking about targeting repairs for aging water systems, and then they’re figuring out what level of pathogens make people sick.

Jim: Yep.

Nicole: What about the good stuff? Usually when people talk about biomes, they’re also talking about helpful bacteria, like probiotics. If I trust my sister, I need probiotics to live life to the fullest or something.

Jim: She talked about that too. Not you and your sister specifically, but the things that microbes can do for us.

Lut Raskin: So I’ve alluded to the fact that, you know, we have lots of microbes in our water and most of them are beneficial microbes. So you can start to think about how these beneficial microbes can actually benefit your health. And you can start thinking about engineering your water to contain the right microbes. That again is a bit out there and it’s not going to happen until we have solved our other problems. But if you have a water treatment plant where you have a biofilter, which we do right now, to treat our water and you have microbes helping you with water treatments, those microbes—you can encourage the right microbes to end up in your water.

Jim: Professor Raskin didn’t even really want to talk about turning water into a probiotic. It’s not really about engineering the water system to keep your gut healthy, it’s about engineering a water system that keeps itself healthy.

Nicole: So we don’t have probiotic water? But we have clean water because we support the right bacteria?

Jim: Yeah, if everything came together on this project, the way that they want to do this could allow us to get rid of chlorine city water one day.

Nicole: Do I want that?

Lut Raskin: The reason we’re adding chlorine to our water is to kill microbes and kill pathogenic microbes. But what if we would treat our water to the extent that we don’t have any pathogenic microbes in the finished water—we have this amazing distribution system that doesn’t allow any pathogenic microbes to enter? Why do we need to add chlorine? If we could run our systems like that, we could save a lot of money, right? We don’t have to add this chemical to our water. We avoid all the unintended consequences with using chlorine, which there are many. For one thing, we produce disinfection byproducts that are carcinogenic, that we don’t want in our water, but we have them at low levels because we need to use chlorine. And chlorine eats away at our infrastructure and causes all kinds of problems. You might think that’s farfetched, but it’s not really. There are countries where they don’t add chlorine or any other residual disinfectant to their water before they distribute it.

Nicole: That’s really interesting because the quantum engineering project is also exploring disinfecting water without chlorine.

Jim: That’s the project with the UV light, right?

Nicole: Yeah, they’re working on an energy efficient UV LED

Jim: They could work together, right? Zetian’s group purifies water with its UV light and then Lut’s team keeps it clean with their water system that contains all the right microbes.

Nicole: If they both succeed, sure. But that’s a pretty big if.

Jim: You gotta think big Nicole.

Nicole: Well, these Blue Sky projects are all high risk, high reward endeavors and the water biome alone is huge in scope. How does her team even get started?

Jim: Well, it just so happens she has access to a local water system.

Lut Raskin: With the technology that’s available we think we can do what I’ve called near real-time monitoring of a limited number of locations in the Ann Arbor drinking water system. You know, there’s no way by the end of year one that we will have a hundred nodes set up. That would be very unrealistic. But we would like to start with a couple of nodes and demonstrate proof of concept and then go from there.

Nicole: Ann Arbor is convenient I’m sure, but why not go somewhere like Flint?

Jim: Cities that have problems may not be as willing to go public with data from these types of sensors. And right now, Flint is still going through line replacement all around the city.

Lut Raskin: You couldn’t do this research anywhere because there’s a lot of places where people are too afraid to collect information that might be damaging to them and they don’t want their general public to find out about these things. And so they rather do nothing than invest in this research. And we are very fortunate in the city of Ann Arbor to have people who are excited about this and, you know, forward thinking enough to help us do this work. I grew up in a family that was very environmentally conscious and, you know, I always talk about my dad being an organic farmer before that word existed. He wasn’t [actually] a farmer, but we had our own garden and he was very much into organic gardening. I had, from home, that mindset. And then, you know, most engineers love science and math. So I went into a field that at that time wasn’t called environmental engineering, but had a lot of elements of it.

Nicole: I can see that. In organic farming you’re using the pests predators instead of using pesticides, chemicals. And instead of just killing the microbes with chlorine, you bring in the ones that will get rid of the pathogens.

Jim: Yeah, absolutely. Sort of like organic farming at the microbial level.

Jim: Was there an event or something that sort of pushed you on this path? I mean, obviously you were here in Ann Arbor when Flint became an issue several years ago, but was there something that drew you to this type of project?

Lut Raskin: Well, I’ve been working on projects like this for my whole career. This project developed partly because we have a lot of work going on in this field already. And we started to kind of dream what would be our ideal scenario out of all this research that we’re doing. And so our main constraint is collecting more data. Collecting real time microbiome data is one of the constraints for making progress.

Jim: Living the dream.

Nicole: In an environmental sense, yes, it is, isn’t it? An army of microbes at work in our water making it safer and also better tasting.

Jim: That would be impressive. After talking with Lut, I’m more than happy to get rid of the chlorine.

Nicole: And then at the infrastructure element, being able to prevent main breaks and do more targeted repairs—all from measuring microbes in the water system. Who knew? So what happens next with this project?

Jim: Well, they’re currently working on software to pull more information out of this DNA analysis that they do. They’re also meeting with the people who run water infrastructure, the providers, the treatment plants, to make sure that the technologies they’re working on solve real problems that matter. We’re going to be dropping in on that meeting somewhere down the line. A little later on, we’re going to catch up with one of Lut’s collaborators on the project. That’s Branko Kerkez. He does the sensor installations. And we hope that you listeners are going to come along for the ride.

Nicole: Thanks for listening to Michigan Engineering’s Blue Sky Podcast with me, Nicole Casal Moore.

Jim: And me, Jim Lynch. We’ll see you next time.