 This morning we're going to hear from three speakers, again focusing on the aspects of pathways of exposure. And for those of you on the board, their biosketches are provided in your briefing materials. Each speaker has about 20 minutes for his presentation. After that, we'll have a few minutes for clarifying questions. We'll open it on the floor only briefly and please direct your questions to the speaker for clarification only. At the end of all three presentations, we do have time for a larger discussion as a panel discussion. The board will have priority in asking questions and having that discussion. But I will reserve a little bit of time at the end to make sure that we do have time for any questions, discussion or input from our visitors or people that are on the phone and also on the live stream. So again, please be aware that this session is being recorded and it is being streamed live over the internet. For those of you who are on the webcast, please note that you will be on mute. And if you do have a question that you would like to rise for discussion, there is a chat function that you can see on your screen on the webcast. Please use that to submit your questions for discussion. So with that, I'll introduce our first speaker. This is Jeff Plumlee. He's with the USGS as the Associate Director for Environmental Health and Acting Associate Director for Energy and Mineral Resources. But he assures us that that has been his position for only the last few years. And prior to that, he's been a very dedicated scientist and has created a lot of information in the area of subsurface contaminants. His presentation is Transdisciplinary Science to Understand Natural Subsurface Contaminants and Their Health Implications. Jeff? Great. Well, thanks very much. Everybody can hear me. I hope. So I thought that I would give a little, seeing the mix of talks that are being presented in these series of seminars, I thought I would take a little bit different tack that actually fits very nicely in with my science career up until a couple of years ago. And that's working across disciplines, transdisciplinary work. And we'll talk about natural contaminants in the subsurface. So natural and contaminant, people argue that might be some sort of an oxymoron. How can it be a contaminant if it's natural? But it's a substance that basically can still have a health effect. And just a quick background for those of you that may not know the USGS. USGS is a natural science agency within Interior Department. We're non-regulatory. We're truly interdisciplinary. We have biologists, geologists, hydrologists, geochemists, microbiologists, remote sensors, folks like that. And it gives us a unique power to look at source, transport, fate, exposure pathways, biological effects like toxicity and other health effects. And then working with public health people and other agencies and other organizations. What are the implications for human health? And these two mission areas that I'm involved with actually environmental health mission area is what I would call the water and biology component. And energy and mineral resources, which is where I spent most of my career looking at mining environmental issues or mining and health issues. That's more of the geology component. And it's more or less the geology component with some water that I'm going to be talking about today. So I'll talk a little bit about exposure pathways for natural contaminants and then give some science examples from folks at the USGS colleagues and also from my career just because that's what I know the best. And then if I have time I'll talk a little bit about a slightly different topic that we find very interesting and compelling at the USGS and looking at pathogens shed from their animal host in the environment as contaminants. So exposure pathways, basically over on the left is a big unmined mineralized area in southwestern Colorado, but you can see the mines that come in and selectively put on underground working. All that material in these mineralized rocks can contain high levels of copper, zinc, which can be toxic to fish, and arsenic and lead and other things. And basically natural weathering can produce degraded environmental conditions, natural acid rock drainage, but then humans when they act on it they actually accelerate the exposures and the weathering processes. So weathering is one key way that contaminants get from the subsurface into the environment. It produces soils which can then be ingested by hand-to-mouth pathways or can become wind-blown dust and inhaled. Erosion, sediment transport, this is a debris flow from the Colorado floods in 2013. That can expose a lot of material that then can become wind-blown. Leaching of contaminants, ground waters reacting with sulfide mineralization can come out, they can then dry up. This is a tailings dam, I don't know if you can see, but it's yellow wind-blown dust that are full of evaporative acid salts that have very high levels of cadmium and zinc. And then human activities can also expose geological materials at the surface. And even the near-surface desert crusts help protect the soil underneath, and when human activities break those apart that can actually lead to dust generation. Things like landslides and then human construction activities as well. So I think everybody knows a lot about arsenic and groundwater and its natural or geogenic source. It first came to light in Bangladesh and there's been a lot of work done on it. Basically attempts to move to a non-polluted groundwater source from contaminated surface waters. They ended up tapping into arsenic rich aquifers and led to arsenic-related health problems for many people. And I think millions are exposed and I haven't seen any updated numbers on how many people actually have health effects. But you can see the EPA drinking water standard, 10 PVB, and essentially everything other than the blue dots is greater than that. And so they can lead to hypercharatoses of skin, skin lesions, cancers, diabetes, and bladder cancer. So this is from my colleague George Bright at the USGS who worked on the problem, one of many. Basically it's an interesting geological problem. Arsenic starts out life in the Himalayas as arsenic and pyrite or silicates. There's some thermal spring waters that gets oxidized as erosion is happening, transforms to arsenic-5, sorb to iron oxide. That gets carried downstream of sediment, but that when they get deposited out of contact with the oxygen, plant debris causes reduction with ferrous iron and arsenic-3 in the water. If you go far enough out, actually the bacterial sulfate reduction in areas where seawater is impinging can actually reform the arsenic-rich pyrite. So it's part of a geological cycle that basically the two wells that were installed tapped into. This is a photo taken by George and you can actually see the reducing conditions in, I'm not sure exactly where this is in Bangladesh, but here in the orange, those are iron oxide-rich sediments with arsenic-sorb that's relatively immobile, but down in the organic-rich sediments with elevated mobile arsenic up to as much as 2,000 ppb. And basically the two wells came down and tapped that. Colleagues in our New Hampshire Water Science Center and in Reston have been looking at geogenic arsenic and groundwater and private well waters in the US. And basically the bedrock, depending on the geology, can be quite enriched in arsenic. And so they developed this geology-based probability map of arsenic in New England bedrock groundwater. Joe Ayotte also has been working with our National Water Quality Database to look at arsenic, to take data on arsenic in well waters and surface waters and basically generate this probability map of arsenic greater than 10 micrograms per liter. Joe and colleagues have been working with the National Cancer Institute in various parts of the US to actually show how the geologic-sourced arsenic going into the water can actually then translate into an increased bladder cancer risk. So this is an example of transdisciplinary science where the earth science, the geologists are working with the hydrologists and the geochemists and the public health people to actually help understand the issue. And now we're also, USGS and others are working on how we can make use of our knowledge of geochemistry and hydrology to help reduce exposures to arsenic-containing water. And there are things that can be done with well placement and things like that. Moving on to a different topic, more in the solids. Everybody knows about asbestos and the occupational exposures commercially and industrially used asbestos. Arianite is a fibrozealite that causes the same sorts of diseases. But in the late 1990s, concerns started increasing about asbestos and arianite that are naturally occurring in the rocks and what happens when those rocks get disturbed and what are the actual exposure and health risks of living in areas or working in areas where, for example, natural asbestos or arianite incur in the near-surface rocks and soils and are disturbed or where excavation may bring them up to the surface. And in the case of North Dakota and Arianite, they're actually using Arianite-containing gravels to gravel the roads in the lock-in formation to where many trucks per day would drive over and generate these dust clouds. So there's a lot of attention being paid to this and there are a lot of headlines, but the question is what are the actual risks? And so asbestos and arianite can occur in specific geologic environments. And my colleague Brad Van Gosen, we had a project back in the early 2000s where we started compiling data. So these are known occurrences of amphibol asbestos, chrysotile asbestos, or arianite. And it only occurs in fairly restricted geologic conditions. So if people are worried about natural occurrences of asbestos, then you don't really, or arianite, you don't really need to worry in some places. And you can see here this, for example, in the western U.S., opiolite belts, same in the eastern U.S. So the geologists can actually start helping understand where the occurrences are, what's the mineralogy. Also more importantly, what are things like the chemistry, the length to width, and other parameters that might influence the toxicity of the fibers that people might be inhaling. And this is actually only one part of, there's a lot more that the earth scientists are not going to be talking about, but there's a lot more that the earth scientists can be doing to actually work with toxicologists to better understand how these geologic materials, how their chemical composition, physical parameters, things like that, and surface chemistry might affect it. So going back, so if we have the geologic occurrences, one of the things we can then do is compare with epidemiological databases. Now one of the things that sticks out right away is that this is malignant mesothelioma deaths from CDC, but you can see very high levels in all the shipyard cities in major urban areas. So it's very clear that occupational exposures produce very high rates. I think that then tends to apply that environmental exposures or occupational exposures to environmental asbestos and arianite might have a lower rate, a substantial lower rate, which actually is a good thing. One of the areas where this has recently been in play is in Clark County, Nevada, where Brenda Buck and colleagues at University of Nevada Las Vegas have found a new occurrence, a new type of asbestos occurrence, where it's filling fractures in the granitic rocks and it's quite abundant. So the question is, if it's there, what are the exposures and how far and what health risks? So we set up some number of years ago this USGS Powell Aeronite Working Group with National Institute of Environmental Health Sciences, Arbery Miller, I think a lot of folks here will recognize Arbery's name, but basically we got folks from NIEHS, USGS, NIOSH, ATSDR, USDA, National NRCS, University of Cincinnati. So this is truly transdisciplinary of geologists working with health scientists, and we are slowly being able to start working on linking our earth science databases with their epidemiology databases. So we're still in progress on that. Shipped over to Coxidioides, the soil fungus that causes valley fever. It's in the hot desert southwest, and it's very expensive. There was a paper from CDC a few years back showing that from 2000 to 2011, the health care costs in California were greater than $2 billion. And this is the endemic range in 2010, but they're now seeing occurrences in southeastern Washington, northern California, and other places as well. So it's interesting is the endemic range changing, and why would that be the case? So this is a naturally occurring soil fungus. It was, I think, first came into prominence in the 94 California Northridge earthquake. USGS seismologist Randy Gibson worked with CDC, and basically these purple spots are landslides in the Santa Susana Mountains. The green are marine sediments with possibly high boron, and these yellow dots are the occurrences of a valley fever outbreak that came within the first couple of weeks after the earthquake. It's basically a fan-shaped distribution of dust coming from these landslides, and there actually was an outbreak in sea otters offshore as well. So based on this, there was a lot of speculation that marine-rich sails and with the chemistry such as boron that might act as microbial sites to enhance the competitive edge for the soil fungus against other pathogens. There was some speculation about that. So there are scientists who are playing a role working with a health scientist to understand what are the soil characteristics that influence the favorability for growth of coxidioides. And colleagues at USGS, Fred Fisher and Mark Boltman have been working on this as well. And it seems like there might be, since we're seeing it now in southeastern Washington, the geological characteristics of the soils are not the same as they were in the Northridge, so it seems like we're moving, it's something a more complex model that we need to be looking at. One of the things is the soil fungus likes hot temperatures. It can't survive the really hot conditions in the very uppermost part, but it can out-compete the others a little bit below the surface. So anything that gets down below that upper layer of soil like human disturbance or erosion during a rainstorm that then takes the spores and transports them into sheet wash and overbank deposits from which dust exposures can occur, that might be something to go on. And we're working, we've got a project now with CDC looking at who, basically there are mechanisms now where they can actually test soil samples and see the presence or absence of the soil spores. That was a new development fairly recently. So it's that kind of work working together. What's the chemical characteristics? Are there certain elements that are enriched? Can we actually see toxins that might be in the soils that might be indicative of the presence and detecting it? Then what can we do to actually help prevent exposures and understand how can we predict soils where it might be present? The last one, and I think I've got a couple more minutes left, is work that we did with CDC and Doctors Without Borders on an outbreak of lead poisoning in northern Nigeria linked to artisanal gold mining. And this is a case where the subsurface contaminants were down below the surface, but they were brought up by these basically people working manually, going down on rope ladders and bringing ores up, distributing them to villages where they would first hand crush the ores. This was all going after gold. They would then grind them with flour mills, the same people that set up a distribution network in late 2009 from the veins to the villages. They would distribute bags of ore and USAID flour, repurposed USAID flour bags, distributed them. People would buy them. They would hand crush them. Then they would grind them. The same people that set up a distribution network actually then purchased flour mills for each of the villages. You can see the dust. They would then fluce and mix with mercury for amalgamation. Usually for artisanal mining, mercury amalgamation is the key thing that's causing the health issues. But in this case, lead poisoning started showing up within six months and kids and over 700 kids died and many thousands and more were probably exhibiting signs of extreme lead poisoning. You can see that little dot down that the person is pointing to in the palm of this child basically is all they get out of a big bag of war, but that's enough to be economically worthwhile for the families to do this. And basically geology is an underlying contributor to the problem. This is the original quartz rich veins that had enough gold. I call them ores with quotes because they weren't really very high grade, but there was enough to process them. Significantly though, these were lead rich, primary lead sulfide. The natural weathering of the gold prior to the mining basically transformed lead sulfide, which is relatively not bio accessible in the stomach acids, into lead carbonate, which is highly bio accessible. So it was the combination of the artisanal mining network and the flower grinding that actually combined with the geology combined to cause the problem. And so basically what we did was characterize and we think our results aided in cleanup. Basically these are scanning electron images on the left. Anything red is a lead rich particle. And the scale bar is 250 micron, so anything less than that is basically ingestible by hand-to-mouth transmission. And actually the flower mills that they were used for grinding, the ores were then when they're not grinding the ores, they were used for grinding their grains and we could actually find lead in the fibers that were being used to grind the grain. So again, this is a case where there's a geologic component and it's the earth scientist working with the health scientist. We did bio accessibility tests, but I think I'm running out of time, so I won't talk about that. We could actually calculate uptake in the different areas per day and basically under very dirty conditions. Children, pregnant women and adults were exceeding the provisional total tolerable lead uptake level. So I think these slides will be available. I'll let folks take a look at them. I have a paper in environmental health perspective to summarize it. Then one last thing, an emerging contaminant possibly is avian influenza virus. What happens when birds, which pick it up from interactions with poultry in China, fly over and come in contact with poultry farms here and they catch it and they have to dispose of large numbers of fowl. What happens? And also the waste ponds, what happens? And so at the request of the Wisconsin State veterinarian, USGS folks started working on detecting avian influenza virus in groundwater. And so I think this is kind of a new spin on contaminants. What's the current and persistence? And if it's there, does it pose a health hazard? Is it transmissible and pathogenic? So with that, basically, without whirlwind, I want to just call attention to our geo health newsletter from the USGS environmental health mission area that talks about a lot of our work, more traditional work, looking at contaminants in the subsurface, effects on biology of contaminants and pathogens, and I'll leave it there. Thanks. Thank you. That was a very wide ranging talk covering a lot of different topics. We do have time for a few questions if anybody needs any further explanation of any of the points that were made. OK. Hearing none. Great. Thank you. Don't run away, Jeff, because we will have a broader discussion of this later. Oh, shucks. I can't. And our next speaker is Prop Kerr Clement. He's a professor in the Department of Civil Construction and Environmental Engineering at the University of Alabama in Tuscaloosa. He's going to give us a presentation titled Challenges in Predicting the Fate and Exposure Pathways of Environmental Contaminants. And this is looking at a lot of modeling type efforts. So I look forward to hearing what you have to say. Thank you. All right. Thank you. And thank you for giving me an opportunity. And you will quickly realize this is kind of an unlikely title for a guy who made money out of building a modeling career. OK. So to kind of set the talk up, I'm going to open up with kind of a philosophical quote. Am I OK with Mike? OK. Thank you. I'm used to yelling at the big classrooms. So I was kind of reading different literature and came across this beautiful quote by Herbert Timers, who won Nobel Prize for this. He said, he introduced this concept of bounded rationality. Basically, he challenges rationality and says, rationality has its own limits. OK. And he says, what we should be looking for is a satisfying solution, not the rational solution. And if you look at the dictionary, satisfied means examining alternatives and finding a best possible solution. So I'm going to leave this, and we'll come back to it in the very last slide. OK. So what I'm going to talk about it is actually a study funded by this board. This was 80 years back. A fairly controversial study. And Susan was there. And it was two years of hard work. I just want to summarize this and kind of leave some impressions of what we can learn from it, actually. Sort of a post-harded, if you will. OK. So before I get into the details, I should first acknowledge people who actually did the job. I was part of this panel. But David Sivitz, fantastic, fantastic. If he led that, it was an honor to work for him. I really enjoyed it. So if you find any brilliant ideas in this presentation, it came from people other than me. OK. But if you find anything wrong, it's probably mine. However, I would never admit mistake, because I'm a professor, right? Because I have a health problem, which is called LBD, which is called listening deficit disorder, which my wife identified 30 years back. I have two girls, and they've been working on it. And actually, it's getting worse, apparently. So I'll leave it there. OK. So let me start with the good news, which is, hey, here's a study you all funded. What did we get out of it? We actually got a bill. So this is beautiful, because I get to see a study going all the way to Senate Bill, which was actually signed by Obama. And they eventually got funded. OK. So now, if you look at the NAS charge based on Lincoln's mandate, it's to provide independent advice to the nation to solve complex problems and inform public policy decisions. And in some way, I kind of feel nice that we were able to contribute that. Now, so that's the good news. That's the beautiful, nice smelling sausage, right? So my purpose of my talk is to kind of open the sausage and show the sausage making process. And what went on? And what can we sort of learn from it? OK. So let me give a quick background on what was the problem. The problem was a classic. So we had a nice talk on natural contaminants. So this is an answer. This is a human-made contaminant base, which is PCE, right? So this was a problem at Camp Lejeune. Camp Lejeune Marine Base is a large base. It's 250 square foot. And there are many, many contaminated sites there. The problem was there was disposal of PC and PC. And basically, that happened here, right at the downstream, there were drinking water wells. OK. Why would they do that? I don't know. That happens all the time. It's like one of those things. So you can see the plume was directly hitting those drinking water wells. So the people are actually exposed. There is no question about it. And there were a lot of health issues, including male breast cancer, which is a very rare event. And overall, the story is very similar to the classic PC stories like Civil Action, which is a very classic Goobman story, right? So that's the problem. So there was a contamination. It happened. When did it happen? It happened way back, right? In the 60s and 50s and 60s. OK. So moving on. This was our charge. Review the scientific evidence on the association between the health effects and contaminated water, which is basically established causation, right? The second one is a lot more interesting. This is the first time we were sitting in a similar room. And this is one star general walks in, OK, with this big entourage. And he looks at this little scientist and actually tells us, hey, tell us what happened. So that's actually very interesting. See how much the public trusts this and predicts the past, what happened in the past, and then tell us how to compensate, which means tell us how to provide, give us a solution. So this was a charge, OK? Now, what they did was this was also the charge given to the team of experts. This was about eight years back. And they said, in order to predict the past, we got to depend on models, right? So I went to the project website, and this is what they said. Water modeling is a scientific method that will help estimate the past that no longer exists today. Because remember it happened in the 50s. And here we are in the 2010s. And water modeling method will help scientists fill the missing data. The models will tell us the past, which means the proposal basically said, here's a crystal ball, and you would go and ask the ball and say, hey, what happened? And it will tell you what happened. This is fantastic use of science, OK? So that was the proposal. So what happened? Well, fast forward 80 years, the project was completed. And one of the project directors actually went to Congress, and he was going to vote and testify in April 2007. He said, effective today, the former Camp Lejeune Marine Center families can find estimated exposure levels of PC, TCE, and degradation products calculated through modeling by visiting the website. So they actually went through a detailed modeling study, found the concentrations, and put that on the web so people can see it. So how was this information received? Extremely well. I worked with a bunch of epidemiologists and toxicologists. By the way, it's the first time I realized epidemiologists and toxicologists do not get along with each other. We had some very, very interesting, to me, an engineer just to listen to that was a fantastic experience. But they both allowed us. They said, the engineers have developed such an excellent model, we can now use the model to predict exposure concentrations and proceed with risk assessment. What about the residents? Oh, we now know that we drank poison water. Fantastic. We got a great review. Now, I thought, okay, let me go to the website and pretend as if I am the Camp Lejeune resident, what would I get out of this? Well, I was born in 63, so I could actually get, if I were in Camp Lejeune, I could say, hey, I was exposed to 58.81 microgram per liter of PC. So accurate, right? And then you can see when I was a teenager, I was lucky, I can even say that because the PC is slightly higher. So we can get some exact numbers. And then they also brought a beautiful summary slide which shows, here's the model prediction. The exposure started in 55 and it went up, went above the MCO, very nice curve, they did Monte Carlo simulations, very good error bounds. Now, when it catches, look where's the data. Nothing much. The model does a beautiful job if there is no data, by the way. When there is data, not really. Primarily because nothing too wrong about modeling because this is 80s, PC was not regulated. There was no technology to measure PC. But overall, it looked very nice. So what they did was, they went ahead and said, here's the proposal to the Navy. We are almost done, but wait a minute, we need a few more health studies to finish a few things and we could not predict some other degradation products. For example, PC degrades to produce vinyl which is a lot more riskier. So we need a little more reactive transport modeling, a little more complex model. If we can do that, we can exactly tell you what happened and the problem is solved. Times can solve the problem. Which is all great news and this is where we came in as a committee to look at what should we do, what should we advise Navy to do. So while we are doing this as a group, one day I was just sitting on my couch and started to have this philosophical question. Are we really that smart to accurately predict the past exposure scenarios using these complex non-water models which includes my own work. I built some of these models myself. So I said, well, should I be really happy or should I be sad? I was kind of confused. Now, at the same time this was 2010, I opened one of our top journals which is called Water Resources Research. That's the top journal. And I found this beautiful paper where they actually modeled a transport in a 40 centimeter column filled up glass beads. And they filled up fine glass bead and coarse glass bead. I actually repeated that since my lab. First glass bead, fine glass bead. Same porosity. And they would send us a pulse of salt from here and measure it. And then they flip it, send it from here and measure it. And then got the breakthrough curve. The basic model for that is we call a convection diffusion equation. It's very simple. Convection is a transport term. Diffusion is a diffusive term. And then doesn't matter how you flip if your porosity is constant, it should give the same result. And it was actually done by Brian Berkovich who is the editor of the journal. So here's a top scientist doing the top work publishing the top journal, right? So here's what they really got. So this is Obama was running for election. So yes, we can. So I was kind of excited at that time. So here you see a beautiful breakthrough curve. White is F2C and black is CPS. Beautiful. So we find out, yes, we can. We agree, you know what to do. Except one of his graduate students said let's repeat the experiment at the different velocities and then everything falls apart. They can't predict actually the model fails. Now think about it. Here's the top scientist trying to model a 40 centimeter column filled with glass beads and he can't predict it. But I go to the National Academy here I have a scientist who is telling me what was the PC concentration 120.57 microgram per liter. Perfect. Is there a problem? So what can we do about it? Well, my conclusion was the basic convection diffusion equation I now call it as a conviction-confusion equation. So the velocity is like a conviction. Somebody will say that's the velocity, you get it. And then D is confusion term. We'll just throw it in and forget about it. Essentially, right? So look at the reality. A PC contamination has got a spill. It dissolves the solution and then the complex geology. We don't even know how much PC is disposed because nobody is going to keep track of how much PC is disposed assuming hey, there will be a modeler who needs this data. So let me record it. No. So we don't know the source, right? So under this condition what the proposal was what should we do? What's the purpose of more health studies and advanced modeling efforts in dissolving a complex policy problem like this? Well, we said, let's go back and look at our charts. What was our charts? Well, review available data and see whether there is causation and what should we what would we do in terms of additional research and how can we compensate the resonance? So our our report which was a 300 page report had an executive summary which said sorry guys, science cannot establish causation in a timely manner. Of course, we defended it. Why? And then he said, stop research. Develop a policy solution based on what you have which we know there were contaminants. You know they were exposed. Come up with a policy solution. Now you can see I remember this doing in Capital June. The residents were totally unhappy. They said, well, these guys were paid by the industry and of course my colleagues who got funding from this project, they were unhappy. Big schools. They lost money, right? Politicians were unhappy. Maybe was not happy. I remember meeting David after this public hearing and he said Prabhakar, you know what? If everybody is unhappy, we probably did something useful. Let's leave it like that. That's all. So now fast forward, several years later there was a policy solution. So there was an act signed by President Obama in August 2012 and then it was actually funded just before President Obama left his office. And the funds basically decided to supplement the eligible veterans who were exposed 30 days cumulative. How did they come up with a 30-day? It's a sausage actually. It's a policy solution. There were the actual modeling results were used and then they eventually come up with some sort of a policy solution and that is the best we can do in a time-constrained manner. So the question is, what can we learn from this experience? Well, I believe rational scientific solution to environmental problem because tons of data, therefore a true scientifically definable solution is literally a mythical solution. It doesn't really exist. So what we need is a bounded rational solution. So what next? Well, let's go back and look at Herbert Simon's quote. When he introduced the concept of bounded rationality, he said, human decision-making process is always limited by available knowledge and our mind's ability to process the information even if it's available. So what we should do is we should always ask the question, how much complexity is needed to derive a satisfied solution or a bounded rational solution rather than a truly rational solution. Now, how do you go about it? Well, here's my first attempt. Now, look at a science or a modeling. So here's a plot of based on my experience the money we invest in modeling whereas the benefit we get. Now, if there is a spill here you came to me, I can tell you at least the spill will go this way or this way. I added in value to you. Then I will charge, put a bell and tell you how fast it goes and say, hey, don't worry about 40 years. I can tell you it'll take about 40 years. You don't have to panic now. I added value. I'm doing some modeling some analysis, but there comes a point you can you start throwing money it platters out and that comes a later point. You throw more money, you confuse the problem. You started losing value. Now, where is this point? I don't know. I really don't know. But at the minimum, we should have in our mind this curve essentially this information curve. And what I also suggest, so with that here are my specific recommendations. I was actually with Charles Papadopoulos company yesterday, gave the version of this presentation and he said a consulting company, he's a national academy member can look at some other old data and perhaps generate these curves, which will be a good information for people to have. So perhaps review past assessment and develop a series of cost-benefit curves that can be used to build a bounded rationality framework. That's one option. And second one is establish an expert panel to review worthiness of large groundwater assessment projects. The beauty of our panel was we reviewed something before it started, so we can actually make some input into it. A completely impartial panel academy can do it, somebody can or AGU national groundwater association can do it. How I don't know. But that's what we stepped. The last but not the least, I'm not sure Susan would like this, it'll be nice to go back to Camp Lejeune and do a post audit. We never do this actually. We do a study, forget about it. We do a study, forget about here's an opportunity for the national academy to go back and say, hey, here's the study we went through it, what can we learn about it? And I believe there is time to learn from that. So with that, I got to talk about football because otherwise they'll fire me. So thank you for your time. Thank you very much for an interesting presentation. Do we have any clarifying questions that we'd like to bring up at this time? So I was challenged a bit to understand the precision that was used in the number that you said about modeling. And as we think about it, you can get as many decimal points as your model can derive. So you have to truncate that at some point. I'm wondering what's their discussion about doing some probabilistic modeling or something like that or distribution of modeling, figuring that out with different parameters. So you do a bit of sensitivity analysis instead of that point estimate of precision. We do that all the time and they actually went back and did it later on, do a sensitivity analysis. But what is still missing is, if I have a ruler to measure this, I would say this is like, I don't know, 8.2 cm, right? I would never say this is 8.2, 4, 5, 6, 7, 8 cm. But in modeling we have different types of models. Analytical model, simple numerical model, complicated models, that's like Bernier-Caliper to a ruler to eyeballing. We have never had a serious discussion on accuracy and precision of models per se. We do still do Monte Carlo some probabilistic analysis but we never really associate a precision to the modeling complexity. That would be another interesting thing to think about. Yeah, and just to follow, then it's also the precision and support of the decision that you need, right? So that's the other way of thinking about. How much precision for the decision? Fantastic point. And how do you do it? I don't know but it's worth thinking about. The settlement was $3 billion? $2 billion, well that's enough digits. How much did all the study cost? All of it? Well, roughly. $100 million? Susan might know. I have no idea. Well, roughly. I would say it's about $100 million. My point is you could spend another $100 million which is what? 5% of $2 billion. Might that not have moved the $2 billion one way or another? Is it the time constraint that you were after or is it the money constraint? Time constraint. Think about it. These people are already sick. Okay. Next question. Could you have implemented a policy decision which looked also at an interim policy solution and then some more studying to get it smarter if you like to bound the rationality further around. You could probably do a phased approach but still I think the thing was when we reviewed this thing the question should be always tied to the policy solution rather than science. At some stages the policy and the science kind of overlaps. When we saw some other proposals we said take it to NSF. It's a great problem actually but it's not a problem to resolve this policy solution. That's a challenge. You've got a great point. We could do a phased approach. Good. We'll have time for further discussion on this. Steve, did you have a clarifying question? Okay. I was struck by your notion that you said that nobody's thinking about precision accuracy and sensitivity. That's just in the world of modeling that I deal with which I'm not a modeler but we've been doing that for decades and it wouldn't a number like that would never it would have gotten trashed early and so I'm just a little struck. I don't know if this is a cross disciplines or... As I said this was the minute they did the first set of modeling they put this website and then they were trashed and they went back and did Monte Carlo analysis. They didn't put error browns. So there are methods to evaluate that there is a systematic way to associate model complexity with precision. There are people working on that in other... Okay. I see some good points for discussion with our panel after our next presentation. Thank you very much Dr. Clement. Appreciate that. So our third and final speaker is going to I think kind of follow very nicely from this presentation. This is Alex Wardle. He's an environmental geologist with the Indian Department of Environmental Quality. Title of this presentation is overcoming barriers to developing risk based cleanup levels. Understanding where the mass of contaminant is and how the mass of contaminant moves. So I think that follows well from what we were just hearing. It says it means the same thing. Or we'll find out from Dr. Wardle. Can everybody hear me okay? Excellent. So I'm a geologist with the Virginia Department of Environmental Quality and I work primarily on petroleum spills. So we've basically been spending the last 30 years trying to get to that founded rational solution that Dr. Clement has been talking about. And what I'll try and do in this talk is to look at the history of our history as an agency in the petroleum program and what we've kind of learnt and what we learnt we need to know better that it will get us to better bounded solutions. Let me see if I can work this. So this is going to be briefly what I'm going to talk about and what can we achieve in our petroleum spills. When we don't achieve them, what are the reasons to characterize a subsurface in order to be able to come up with better solutions? What are the barriers to adopting those new characterization technologies? And I'll end by giving a pitch for the ITRC Advanced Site Characterization Tools team which I'm the joint lead of that just got started in January and we're trying to overcome some of those barriers. So the prologue, this is Bolston in Arlington before the USGS was brought in because three properties had gasoline in their drinking water wells and explosives in their basements and USGS went in, they drilled eight boreholes and recovered thousands of gallons of product from these monitoring wells, recovery wells and their solution at the end was let's get rid of the water wells and seal the basements. And fundamentally, that's kind of what we still do now. So your grandfather would recognize a lot of what we do in petroleum radiation now. So our general objective in Virginia is the key is we want to prevent harm to human health and the environment from petroleum releases. And the key to that is we use a risk-based approach, we look at the actual receptors that we know are affected by renewable likelihood to believe are going to be affected. So we want to know where the source is, we want to know how things could move from the source and how they could get to that receptor and work out whether those are complete or likely to be complete. If they're not, for us it's probably not a problem. And then we're required to remove pre-phase LNAPL as a regulatory standpoint as best as possible. And that's an evolving field for us. So looking at just drinking water wells back in when we started in the mid-80s in our program, we had many drinking water wells reported impacted from gasoline releases. And we have, in Northern Virginia at least, we have had no reports of impacted drinking water wells from gasoline-regulated facility releases for almost 10 years. For residential heating oil tanks, which are often right next to people's drinking water wells, that's a different picture, but we don't regulate those but we do clean them up. So what if when we have a release and we go out and we ask people to do a psych-characterization, are they affected? One of the ways we could look at that and I looked at it was back in the mid-90s we went through a process of closing a lot of cases that had been reported to us because there were a lot of tanks being taken out of service and we had a huge case load and for various policy reasons, a decision was made to try and cut to the quick. So a lot of those investigations were done and some of them maybe weren't as detailed as ideologists would ideally like and some of them were. And then over time people have gone back and they've done property transactions and they've investigated them and we can compare what the conditions were when we closed the site to what they are now. So if you look at the concentration at the time of close and before that ratio if it's high indicates that contamination is increasing, if it's low and below one it's reducing, which is what we would like if we had identified contamination and worked out what whether they were problem sites or not. And what did we find? Summarizing it if no significant psych-characterization had done just a sort of conventional drill-free well state-free soil samples, state-free water samples, we found going back to those sites in the past for depending on the contaminant maybe half of them were worse when we came back to them than they were when we closed them. If we did a little bit more, if characterization was more in depth that ratio went down and if we knew where the source was and the source had been removed and we did a decent characterization and we had pretty good results. When we go to a site and we decide to clean them up what can we actually practically achieve with cleanup? This is based on about 300-odd corrective actions done in the northern region of Virginia over the last 20 years and you can see this this side of this green line is basically it got worse after we did the cleanup. Here is one order of magnitude reduction. We're pretty good at achieving one and two orders of magnitude reduction but any more than that gets kind of tricky for most of our sites and this is the NCL for benzene by the micrograms for liters. So for about 20% of our cases groundwater could achieve the NCL. Now the slide I showed earlier about affected drinking water wells this level of remediation was sufficient to be protective of those drinking water wells and so it is effective it is a rational solution but it's not necessarily achieving what people think we might be needing to achieve. So why is this one of the issues is we historically were focused on just the regulated contaminant the contaminant that has an NCL might be expanded to the priority political list that ETA provides but a lot of times took in our program we're just historically when we first started we were back in the 70s could we seed oil and that was the indicator then we were doing just measures of TPH so 1000 micrograms per liter plus of contaminant not really looking at anything lower than that and then we moved through we looked at the BTEX compounds and now we're starting to look at the additives in the 2000 and now we're starting to look at more breakdown products and historically when you do these investigations you take a couple of samples they're analyzed to a very high level of quality control and precision in the lab but there's only a few and so those results have to be averaged across the site they're extrapolated between different media you look at the soil results see whether you might have an effect on ground water you look at ground water see if you might have an effect on beta concentrations in the soil that might affect people's indoor air quality and you extrapolate use simplistic models to say well if I have it here will it affect this drinking water well this person on a property down there so there's a lot of things a lot of the functions that go into that process and even if we do look at that MCL is that MCL really meaningful in terms of overarching health protection we've our programmatic level that we want to achieve in a drinking water well for benzene is 0.6 the 5M cell and you guys probably know this better than I do is essentially a pragmatic solution that came up for that contaminant in the 1980s if you look at air the parts of the billion are again significantly less than the MCL so the MCL doesn't in and of itself mean it's protected so maybe it's not it shouldn't be used as just the arbitrary threshold that we want to get to clean up we want to use demonstrate clean up so what do we miss typically we can't investigate the source we may not know exactly where the source is if the facility is still operational we can't get underneath the tank if there's been a development over it we can't drill through the building those few samples are not representative we know there's a whole range of contaminants that we're not looking at that are part of the petroleum mix and we can't extrapolate between media and the pathways can't be simply or easily modeled again as Dr. Clement kind of indicated to us so this is an example of you can't use one media to look at another this was a series of about 40 in divestment investigations that came to us and you can see there's really no link between the groundwater concentrations and the soil concentrations and again this is the typical divestment investigation is 304 samples of soil 304 samples of groundwater if historically we've always looked at a 20-30 foot long well in the groundwater maybe 10-15 foot intervals that we're sampling and we've broken that down into individual 1-2 foot sample intervals you can see that this is an average of lots of very different things happening through the groundwater column and this again is for petroleum contaminants which historically have been considered to be stuff that floats near the surface but once it gets into the water column it moves of the water molecule petroleum is a mix say V-tex we didn't look at it before the 1980s that it is we didn't look up until before the 2000 many of the constituents of petroleum and other complex pollutant mixes don't have NCLs many don't necessarily have a health based risk but they smell bad if they're in your water or if they're in air is getting into your basement that's still an issue we can't ignore that as an agency but what do we use as a way to regulate that so it's just a simple example of these two big heating oil USTs been in place since the 1950s discovered there was a spill when they were closed did a conventional ball hole put in a well found that there was product here and some soil contaminant concentrations we wanted to go back and do something a little bit more sophisticated laser induced fluorescent multiple probes and they get a continuous fluorescent image of the product and this is what the plume more likely looked like in the ground you've got perched horizons migrating down through sand gravel channel unit getting to the deeper groundwater next to the river and being able to move and create that floating product so you can see from the competitive investigation if you did a remediation here you might have just treated down here and not known what was going on up here and this eventually would continue to recharge this product down here and you'd be there in perpetuity, never getting below your one or two water the magnitude of cleanup this is an example of complex geological fault this is we know exactly when the spill occurred we know exactly where it is and yet it created some complexity for us the responsible party got to it in a couple of days that's 32,000 tons of soil pretty much looked like it was clean based on the simple model of what's going on we knew there was a well a few hundred feet away based on simple permeability calculations and hydraulic gradients that well wouldn't be affected for 5 to 40 years so pretty simple issue shouldn't be a problem but three months later that freezing water well was contaminated so something more complicated is clearly going on so we looked at the complexity of the geology using surface geophysics to look for preferential pathways in the subsurface downhole geophysics to work out where the fractures are which fractures are providing water which fractures aren't still after 10 years still finding significant recalcitrant contamination went back and did membrane interface probe studies and discovered that even though they'd done remediation of the surface groundwater contamination had migrated at depth and was still there so it was still recharging the groundwater with those contaminants and was not degrading necessarily as quickly as we needed to see to protect those drinking water wells so sort of that simple picture we have a much greater level of complexity and all of these things need to be understood in order to be able to even make a simplistic analysis of what's happening in the subsurface and come to some predictions as to what's going to happen when let alone back in time like Dr. Clement did to find to work out okay how is that going to move, how is that going to disperse how is that going to degrade so looking at this another way we as a program and the most environmental programs have always been very focused on what is the concentration of the contaminant in my particular unit so this is an example of a site these are orders of magnitude but they're pretty close in three different units in the geology when you look at where is the contaminant actually being stored in this weathered unit there's a lot more porosity you can hold a lot more contaminant in the transition zone where it goes from the weathered unit to the bedrock there's pretty good storage there's virtually none in the bedrock which is just the space in the fractures but what's where is it moving very low slow movement of the storage in the satellite really fast storage in the transition zone so if you're going to do an effective remediation you need to do something different here because you want to recover this from here where it's moving so you might be able to rapidly treat this but this is going to be slow and is going to be a different technology so knowing this can we get to that rational objective so this is an example of developing some clean up levels that aren't in the CLs but are hopefully protected so this development occurred back in the 90s early 90s the developer provided two supply wells for the development and as soon as they started developing them to see whether they were suitable they found they had contamination in them and this was in the late 90s 2000s and we had three old gas stations here where we had cases and they were closed because we didn't see any problem though the town is on public water no real issues no real levels of contamination but once they put these wells there could be a problem here they went out, they ended the geophysics to find preferential pathways broadly speaking some fractures oriented in that direction but dipping in that direction and we realised that probably we had a problem about how to deal with it so going in over the years the negotiations happened we wanted to bring this to a close, we had spent about $4 million on the clean up of the various properties the agency was running out of money from our fund so we did a little bit more complicated investigation, did put these multi-level samples in, worked out where things were moving and decided to work for mass flux calculations to work out how what's actually going to happen when we turn these wells on so by understanding what those levels were the relative transmissivity and concentration what is the mass in those units to work out what are the pumping rates of the well if we could get an average of 100 micrograms per litre in the area of the releases once we pump those wells they should get down to an acceptable level and that's pretty much what happened those wells before they were pumping were in the 100 to 200 level we were able, based on those predictions to stop the remediation here and then when the pumping started rapidly fell down until now they're in the 2 to 4 micrograms per litre level which is more than acceptable the town required the developers to put in a treatment system that was capable of achieving this level of cleanup so we are pretty comfortable those wells are protected so in order to make good decisions we need to understand the heterogeneity in the soil and the rock controls what's going on we need to understand that distribution is going to be complex and requires a lot of more detailed but not necessarily precise characterization if we understand those we can understand how the contaminant moves we can understand how it moves in the soil column how it moves in the water it's complicated but it's geologic so we know how that works so these are the sort of new and inverted columns because many of these tools have been around for a long time but with the available computing power we have now and the ability to resolve the information that comes from these tools in a much more precise way we can use these we can use them on site we can get decisions very quickly we can communicate them to effective parties for stakeholders in a very direct way, in a very graphic way that allows people to understand what's going on there are some other newer things and the old industry has been doing that for a long time but as environmental scientists we don't use that and with the drones now I'm quite excited that remote sensing is going to give us stuff that we used to have to have an airplane for we're going to be able to use at our site or large site level on a routine basis so the barriers to use lack of understanding of these tools is one of the main barriers people perceive them as being very expensive they don't know what tool to use and they're not readily available and it's difficult to interpret what's going on on a minor regulatory acceptance as regulators certainly in Virginia we don't have we're not obliged to require any particular tool but when we get to a decision point we're probably always going to come back to eventually we need to have a lab analysis that tells us, gives us a number so in order to overcome these barriers the ITRC has formed this team on an advanced site characterization tool and ITRC is if you're not familiar with it is a group led by the states we bring together federal agencies private companies and state workers to work out what the current environmental science is and how to get that into into our hands as state project managers so it's sort of the work you do we try and use that to come up with pragmatic ways of getting us to be able to apply the great stuff that you guys create for us that's our mission the key thing is when you come into the ITRC it's a whole group of different professionals and different backgrounds but once they're in that group and they are producing that guidance that's what their objective is so the other people working on the EPA is doing a lot of enhanced site characterization training so the STCT has got some great guidance and webinars out ITRC has been working on various aspects of this for a while and you guys here at the National Academy have been doing some very good stuff and I'm just going to leave that for a little synopsis of our team that is working on this for anybody who's interested in participating please join us great thank you very much any clarifying questions from the board? I do have one question you were talking at the beginning and we kind of framed this whole thing as exposure pathways for human health and environmental risks so you did mention in your talk that since I think it was 2004 you've not seen any groundwater intrusion into surface water which is one area to get ecological risk are there any other ecological risks that you would consider from groundwater contamination? from a programmatic view we're typically looking at surface water and we would also look at marshes any sort of part of the water environment is that what you're getting at? yeah precisely because often as eco talks people get pulled into some groundwater discussions and I am frequently scratching my head and saying if you have no hyperic flow into surface water what is the question here so I was wondering if you had had any experience with other types of ecological questions in that regard? typically as a program we don't see it it's not a visible oil then we're not necessarily being made aware of it so again from the MCL perspective there are many contaminants but the MCL maybe one level but the actual for an environmental exposure the level acceptable level for our water programs will be at a well level and we have to bear that in mind when we get to the point of corrective action Tina? yes you mentioned vapor intrusion in your presentation I was wondering if you could talk a little bit more about how you approach that issue because I know that there is quite as I understand it quite a bit of controversy around the use of attenuation factors to predict vapor intrusion versus I think it's the Johnson-Addinger model and different sort of schools of thought about how to look at that issue and focus on when to collect actual data on indoor air yes and that could be a whole talk all of its own as I'm sure you know. The key things from the petroleum perspective the reason why Johnson-Addinger has gone out of favor is because we understand that petroleum degrades once there's oxygen in the soil column petroleum in the vapor form degrades very rapidly and we have the data to support that ITRC did a study a few years ago put all that information together EPA I think is taking it on board with their latest guidance for vapor intrusion and certainly when we've looked at it on a case by case basis we can see that pattern typically if you once you get outside of the mass of contaminant if it's in the soil or the water column and you get you can measure oxygen we can't measure petroleum vapor so that's kind of from an investigation point of view that's kind of being our threshold it's like let's if we think there's a problem let's do subsurface vapor sampling and air sampling oxygen carbon dioxide see what that pattern is that shows it's broken breaking down broken down we know we don't have a problem for the structure if we find it's still there at the level of the structure then we have that question do we go into the structure to take a sample and unfortunately the contaminants we're looking at are ubiquitous within the built environment so once we go inside the house we will find those petroleum contaminants there's no question and so then it's well is this from the subsurface and there's things we can do in terms of looking at unfortunately around here you can look at radon to see well is there migration happening across that basement so you can kind of see well does that pathway exist and you can create that relationship that allows you to make a rational decision as to is this an effect thankfully in our program we're a very protective program so if we think it's an issue we're going to go in and we're going to provide the engineered solution to that property rather than spend a lot of money on characterizing it because the solution is a lot cheaper than the investigation so again talk about that practical, founded solution if I okay let's just do it there's no point in having an argument about the issue sorry it's a little bit more tricky because they don't degrade so does that answer the question so I actually have another question if I may you mentioned preferential pathways and it sounded like you were talking mostly about natural fractures in the rock what about man-made structures, sewer lines utility lines, those kinds of things are those preferential pathways that you look at I know that those have been an issue again I know a little bit more about the vapor intrusion area than I do about the issues you talked about and I was just wondering how it applies yeah I mean that's the way basically the focus is on the geologic pathways because they're more difficult to characterize but one of the critical parts of the characterization effort is where are the utilities in relation to where we know there's contamination because if you have a utility that runs over a source of contamination that's going to be a preferential pathway particularly into a building if you have a water line that goes through the contamination that breaks through the foundation that's clearly going to be an issue so those are the, yes we that's a key part of the process clarifying question one clarifying question maybe it's a broader question as well but you mentioned the fact that kind of the city or one of the municipalities set a cleanup level and then you guys were able to actually with your knowledge of migration through the subsurface in understanding about the hydrology come up with a solution that actually drove that level much lower so is that, do you find that I guess it's a question is that a common activity where you're able to kind of look more where you get a better outcome with more investment of time, effort and resource and understanding the hydrology significantly or do you find that as you kind of indicated earlier that if you know what the problem is and you can kind of figure out the limited set of solutions you run in and you do that, have you looked at that in terms of kind of investment and return on investment at all not systematically but really it's a question of what what are your solutions do you have a simple engineering solution that you can apply so for the town that we were looking out here they said well the simple solution is we make sure that this is the highest number that was ever in those wells so they need to treat it, they need to be able to treat the water that comes out of those wells at that level they had a simple engineering solution so that they were covered from a programmatic point of view that really wasn't ideal for us because maybe that treatment solution would fail, the town would run out of money to maintain it and have longer term protectiveness so we wanted to invest a little bit more time and effort to say okay let's get it to a point where even if that system doesn't exist that well those wells will still be fit for purpose so that was the additional investment which was relatively minor in the grand scheme of this particular process and the importance of the groundwater resource so yeah each one has its balance I guess I was intrigued on that a little more science gives you a much better solution and so trying to kind of conceptualize that in a way and communicate that I think could be very helpful yeah and if you go back to think about the early plots I had of the effectiveness of the site characterization and the effectiveness of the remediation you know whilst they may have been broadly protective of what things are like they weren't really necessarily improving the environment as much as we would like but relatively small investment in time and technology probably would get those numbers to a better place Tina can you hold your question for the broader discussion we'll get to you first so if we could thank our speakers for some very interesting presentations I really appreciate it and ask the three of them to take their spot at the table and we can now open the discussion up to sort of broader thoughts and ideas although if you have specific questions we can address those as well and let's keep in mind that what we're trying to do is address the question what are key challenges and advances for assessing the quality of the subsurface environment and managing risks and from the point of view of this board and the academy who are the players who are the agencies and organizations that may need to have more science to help them understand these questions about subsurface contamination and managing risks so we'll open the discussion up to the board and then in about 20-25 minutes or so we'll see if there are people that are still online that might like to join the discussion as well did you want to get us started? Sure and this is I guess mostly for Dr. Wordle but also for Dr. Clement in both of your presentations we're looking at situations where there was sort of one prime one principle source and in many urban or industrial areas there are multiple sources and you start sort of looking up gradient and finding contamination in that area and you may or may not have given the construct of these programs you may or may not have a responsible party and so things can quickly get even more complicated and I was wondering if you had ideas about how to get at that issue there have been some proposals out there to really look at these issues more on a sort of a groundwater aquifer basis or to really not try to chase down each individual plume that would presumably require some changes in how we sort of think about cleanup programs but I was wondering if you could sort of talk about how that works in your experience so from just from my little universe of the petroleum program in Virginia we are constrained by the laws and regulations to just do it site by site and that's frustrating so that's just the petroleum and then when you add as you say urban environments it's not just petroleum there's the dry cleaner there's the old whatever the decreases that the car repair place used or the industrial activity and some of these may have a program that looks at them and some of them don't and in Virginia we have the volunteer remediation program but again looks at each of those in isolation but we don't and I don't think there are many states that do have a sort of overarching ground water protection purview for me again this is purely a personal on a personal basis that seems a lack it's with the Safe Drinking Water Act ground water sources have certain source protection parameters that communities and localities follow to try and keep those protected but if there's an issue and it falls to the water and water entities to deal with them because the source may not be identified so they may be able to come to us as a regulator and say can you find this and we may be able to say yes or we may say no not really we don't have a program that has money to go and look at that so that from a policy legislation perspective having some way of doing that would be to be the next step environmental protection that we need to go particularly if you look at it from the broader aquifer perspective we don't need to clean up every molecule of water in this aquifer we just need to know that we're protecting where we're using the water in that aquifer whether it's going to a drinking water well it's going to a marsh ecological or a human health perspective in my part of the country that's the highest risk of subsurface contaminant and I thought of this with a vapor intrusion story is radon and we worked hard to try to get a bill through the legislature to measure radon in schools and it failed because it was opposed by the school superintendents because there's no money allocated to solve the problems they didn't want to measure it and know what the problem is and I wonder if there's a lesson there in other areas and or if even radon is a subsurface contaminant as we're thinking about it here well certainly since I talked about geologically sourced arsenic geologically sourced radon is very clearly one that's been upon people's radar screens for a number of years so it is yet another one and the interesting because of the characteristics it can have multiple exposure pathways as well it can become volatilized and go into basements out in Colorado where I spent most of my career on the pike's peak bath lists and when my colleagues looked there and you could actually the biggest exposure pathway was when people would get the water from the wells and they would heat it up for their morning shower and then it would volatilize into the shower and that was a very big exposure pathway and so that's I'm not I don't think it's appropriate for me to weigh in from policy perspectives but certainly from a communication standpoint how can we effectively communicate, how can we understand, at least provide people the tools by which they can understand this and how can we actually get the data that will allow them to understand is there an issue or is there not so that's not really a good answer to your question but I certainly would lump rate on in with the with the geologically sourced naturally naturally occurring but in some cases human enhanced exposures. Yeah, questions Your comments on the Camp Lejeune study actually a couple of interrelated questions it it it seems to me that a key piece of the equation in a situation like Camp Lejeune is the degree of confidence that you have in the causal link between cancers in this case and exposure to the contaminants and if the evidence is suggesting that there really is a causal link it may be that on the exposure characterization side you can tolerate a higher degree of imprecision than in a case where the causal connection really is open to dispute and so my question is is what degree of confidence was there about the causal link between contamination with PCE and TCE and the adverse health effects that the community was reporting So I got to qualify my answer because first of all I'm not a epidemiologist so but having sat on different meetings with the epidemiologist the real bottleneck was tracking people remember in Camp Lejeune these people come and serve for 30 days to 60 days and then they leave and then in some cases the babies were born there they were little children basically they were born there so there was no record to track them to find exactly when they were there and then so there were two pieces right the groundwater pieces different and that's what I looked at it very carefully and there was a piece of tracking people essentially different registry status information they try to do that and then linking exposure to that how many days were there and that was complicated number second was the lifestyle so people leave and they were young 21 to 25 years old and now they are probably 50, 60 years old and they have certain disease in between we don't know what lifestyle they had they could have smoked they could have drank and all kinds of this thing and then Agent Orange for example they could have been exposed to in fact David was involved in that panel too so I remember they were talking about in the battlefield what kind of exposure they had how do you differentiate all this so this whole idea of a linking exposure to specific population was a huge challenge so in fact one of the things they suggested was do a follow up study to go back to some registries to reconstruct some of these things but the epidemiologist in the group looked at it and said we could probably do that but we are now getting into an academic exercise probably we are not going to solve this problem in a timely manner that's a very important point because these guys need solution now so we have this thing it's a great academic exercise but should we do that so that was the kind of a bottleneck to exposure and I remember talking to them at that time you took what 30-35 years to prove show that smoking causes cancer we know the exposure we know the causation I mean still the tobacco industry thought so these are complicated problems obviously and so when time is an issue how do you deal with that that's the challenge Jeff did you want to follow up yes I want to reiterate that our colleagues in the medical community have repeatedly said to me when I work with them correlation does not mean causality and so that's actually a key place where the earth scientists the geochemists and folks can actually play a role in helping cement those links so and using the lead poisoning in Nigeria as one example we were actually able to show what the neurology was just because something's high lead doesn't mean it will actually be really highly bio accessible and the exposure pathway could vary as well so we a lot we provide the tools that can help them understand how people are taking it up is it actually if they're taking it up does that mean it's actually getting absorbed and if it's getting absorbed then we actually have tools that we can use to help understand its distribution in the body so it's again a key part of the disciplinary aspect of the whole process that bringing all the different disciplines tool kits to bear to help understand is there actually causality from the correlation and providing that refined information and particularly in cases of complex exposures as well one of the things we found in Nigeria was that the people are not so they're also getting exposed to the mercury that was the lead that was manifested first but then they're also getting exposed to high arsenic high antimony and high manganese and high crystalline silica so there's all these complex things that they were getting exposed to and what are the combined effects what are the sequential effects what will happen first and all that so there's a lot of complexity that we really need to be working together to help understand thank you we're going to Dom and then Bill and then George and then left I was struck by the ground-water modeling presentation which basically suggested that if you could figure out where the water was going you could figure out where the contaminant was going but of course you've got reactive transport to worry about and in particular you've got sorption to the solid phase you make a one order of magnitude mistake in the partition coefficient you make a one order of magnitude mistake in the time that it takes to go from here to there there are not a whole lot of models that can calculate sorption coefficients to one order of magnitude precision so how worried are you about that a great question so so here's how I look at it so I'm a model developer so some of my code called RT3 which is a reactive time made my life doing reactive transport the problem is I can put all these nice prime I can use different isotherms I can do rate sorption all these things right and then we can do parameters uncertainty to create bounds and all that that's model uncertainty but that is an epistemic uncertainty when I built the model I knew I put in something which I'm not 100% confident about I mean like for example I would go back and even challenge basic Darcy's law I mean Darcy did a little column experiment and defined hydraulic conductivity it's a macroscopic description of our head loss essentially right so there are epistemic uncertainty how do we include that that's the challenge that's where I think an expert final would help somebody would would sit together and say hey