Well today, we're going to talk about some relatively new concepts regarding contaminant transport, and how it affects monitored natural attenuation. And that topic is contaminants that are stored in low permeability zones. >> Yeah, this is a little different from the conventional MNA topics where people talk about dispersion and dilution and degradation. So Chuck why are we talking about matrix diffusion? >> Well basically, we got some great insights and some really neat thinking from Dr. John Cherry at the University of Guelph. I sometimes attend their university consortium meetings that are organized by Dr. Cherry, Dr. Beth Parker, Dr. Tom Sayo. And Dr. Cherry was talking in his presentation about this large site North of Los Angeles that basically releases a chlorinated solvent in this large sort of mountain. And the University of Guelph team has done a lot of research out there showing how these chlorinated solvents were released at the top of this mountain from years of rocket testing. But none of these solvents had been observed in the seeps leading the mountain. So the big picture is that this aquifer itself the material in this mountain can itself store these contaminants in low permeability zones such as the clays in consolidated aquifers. Or the matrix of fractured rock sites and if that storage capacity is high enough then a lot of that contaminant mass can be sequestered away really from in these low permeability departments and away from the transmissive zone. >> We're not really talking about permanent storage, but something that can be sequestered over long periods of time. Something like decades or even longer, right? >> That's right, and Dr. Cherry rightfully noted that this itself is an attenuation process. The bad news, it's really hard to remove these contaminants from what we call low k zones, or low permeability zones. But the contaminants are removed from flowing groundwater systems, from drinking water in the aquifer itself for a long time. >> All right, well I think you have a couple of videos to help demonstrate this process. >> Sure, so we're first going to start with advection dispersion. This first one's a really neat movie of a sand tank model developed by Kevin Masarik from the University of Wisconsin, Stevens Point. And it shows a plume migrating from left to right in a porous media. This is speeded up, you can see this green dye in the aquifer, and it is flowing by convection dispersion through the pores in these sand grains. And then you can sort of see it cleans up after a while, they have stopped releasing the dye. But they key point is that if you own a couple pore points you can clean this up pretty quickly. >> Wow, that's pretty neat, but I guess as they say in the infomercials wait, there's more! >> There is more. So if we really think about this, as we try to clean up these offers, do this remediation, we've begun to realize that this matrix diffusion process is a big deal. >> Okay, and you have another video to explain this, right? >> That's right, and so this one's from Colorado State University. We'll sort of go through this, Tom Sale and the grad student, Lee Ann Doner. Tom Sale talks about this one as being sort of the best way to visual matrix diffusion, more important than any paper he's written. What are the four gray zones in there, David? >> Those are the low k zones, right? >> Low permeable, they're bentonite. Then there's green dye that comes across this tank, the tank's about a meter long, for 26 days. So here's 19 days, 20 days, and then they clean up the aquifer. They assume this has been cleaned up by remediation or the source has been removed but wait, the green's still there. Where is it coming from? >> Well it's back diffusing off of those low permeability zones, right? >> Yeah, so it's loaded up for 23 days, here's day 45, 48. Still there, and so it's this back diffusion process that can provide these, what we call long tails, of these contaminants. We've completely cleaned out the sand, but this contamination still is present in those low permeability zones. And that's the challenge we have and there's two twists to it. Number one, it's hard to clean that stuff up but it is an attenuation process. >> And this is different than absorption, right? It isn't necessarily that the contaminants are sticking on to there and be sequestered in there. It's a different process. >> Whole separate process. Absorption is contaminants sticking on to the same grains of say a sand of an aquifer. This is actually diffusion or material that's going into these clays, silts or whatever. So this is a key thing is that we see a lot of aquifer, and so let's go to this example here and this is a really great picture from Fred Payne of ARCADIS. That says this is what we have out there in real life, and you have all these different units when the big guy deposited this stuff. You just see tremendous differences in hydraulic conductivity. Lots of what we call these low k zones in these unconsolidated medias and at fractured rock sites. So at most sites wherever that plume travels part of the mass will diffuse into these low k zones in the beginning and then back out, this back diffusion, once the transmissive zone start to clean up here. >> And this is relatively common thing that we see at sights all over the place, right? >> That's right, and so this is just another way to look at this. You might have this red as the plume that's advancing into these sands. The sand is sort of this tan color. What's the gray there, Dave? >> The gray is the low permeability silts, in this case. >> That's right. So you can see at the top hand the plume is moving, and in the middle panel the plume is there, but you see these red arrows. The stuff is diffusing into these clays, both the top and the bottom. And once you clean out that bottom part, what happens? >> Well once you clean out that bottom part, you've basically removed this source, but the plume is flushing away, and the transmission zone gets cleaned up. But the material basically now is going to diffuse in the opposite direction, it's going to move out of those into the more transmissive areas. And so you have reversal of that diffusion gradient. >> So that's matrix diffusion. And that's what's shown on that bottom panel, and the key point in this week's lecture is that this is a type of attenuation process. It's different than the destruction or degradation of contaminants, but it's like sorption to aqua materials but a lot more powerful. These material are sequestered for a time, and high concentration material is released slowly. So it changes the strength or the dose of a plume slug making it on average, weaker over time, but it lasts a lot longer. I think we can draw that and sort of show what happens. Let's see if I can get this to work. >> All right, let's give it a shot. >> So we'll go over here, and let's just say, that I've got a slug release of a contaminant. So I'm looking plan view right here, and so here's this plume. I'm looking down from the sky and let's say about this dissolved plume is bent release and I have a well that's right here. And if you look at this plume, the ground water flow is from left to right then, if I'm looking at the concentration at this well x right here and I'm looking concentration versus time. According to the vection dispersion, I would see a perfect bell curve just goes up and then it goes down. But then if you have a lot of silts and clays in that unit then you're going to get a different response right, Dave? You want to draw that one? >> All right. >> What would if look like? >> Let me get the pen here. I think if I draw my c versus t here, little neatly, you'll see a little bit different curve, start off maybe not quite as high of a peak. And then you see as diffusion becomes most dominant process, we're tailing off that concentration as you move longer and longer curves in time. >> The long tail. >> The long tail. >> So that's the key thing we're trying to look at. >> Okay. >> Now another way to visualize this is courtesy of researchers again from the University of Guelph, Dr. Beth Parker and Steve Chapman. They've done sort of a lot of work in this area, a lot of work at fractured rock sites. And they have observed and monitored these solvent plumes and these fractured rock systems where there's a lot of interface area between the transmissive zone and the matrix itself. For example, a particular model that we're looking at here, and this is one that they use to simulate plumes that are moving through fractured rock systems. So if we go to this particular graph right here we've got the z axis. This is like a vertical slice in the subsurface. So that's depth on the z axis, then on the x we have these meters. And each one of this lines in here is actually a fracture. >> So the fracture's the high permeability area. >> That's right. And they're all different sizes, the little graph here talks about the distribution, some big fractures, some small fractures. But then they run these models of this source on the top left, you see the red. That's this steam [INAUDIBLE] source that's producing these contaminants in here. Let's look and see what the plume does according to this model once you account for matrix diffusion. So in here, we've got a plume. This is the no biodegradation, this is their FRACTRAN model, this is after 20 years, that plume has migrated. Let's go after 50 years, it looks like something like this. It's gone almost 500 meters. Quite a bit of distribution here, but you start to see this stuff is soaking into some of those fractures and things like that. So they can do a lot of things with these models. And one is they can say that, say for example after 100 years it looks something like this. What would it look like with biodegradation in that matrixes? Is that important? >> Well yeah, I think it is. I would expect these greeners to become less green, more blue, and the plume to sort of shrink back a little bit. >> And that's what we have here. With a 10 year half life and the fractures in the sandstone matrix. So one thing about matrix diffusion piece is that, this half life is really important. If there is any degradation in that matrix that can be a big deal for us. Okay, let's do another example. Let's go to HydroGeoSphere, another model that this team ran. It's an unconsolidated site that we're looking at here. They have a series of four clay layers and three sand layers. The red is the stean apple and you can sort of look at the same pattern here, these are called type sites. And to model this though, you really need some fine grids. So down here Dave, what's the bottom panel? Is that all the layers they have? >> Yeah, that's the z direction, depth direction, and you're looking at about 2 to 5 meters right there. >> That's right and the model can do the math almost centimeters, too. >> Yeah. So yeah, you see the much finer lines showing the resolution there on that model. >> Right. So here's this, I'll just show you some other simulations that are in the SERDP report that they developed. But you have different concentrations, what do they look like? >> Yeah, at the bottom you're near zero and these are milligrams per liter actually converted to an equivalent pour water concentration up to about one milligram per liter, in this case. >> So in the model they've got this dean apple source now and they're going to let the model run. So this is after one year is the panel you're seeing here. This is after two years and the plume is getting larger in the sand zones, right? Remember the sand zones are in the middle, there are these three clay layers. The red is that the high concentration? >> That's the high concentration in this model. >> Okay, and then loading this thing up. Here's at five years it's the plume's already gone off the edge of the model, right? >> Yeah. >> But you can start seeing the clay layers that stuff in the middle starting to soak up some of that material, right? And that's this matrix diffusion, and after ten years they say that's this loading. And the whole system, that whole slice has got some contamination in it. Ten years, is that a common time for a loading period at an actual site or is that a little low? >> I'd think it would be maybe a little bit low, but I think for the purpose of this it worked. >> A lot of the sites we deal with are 40, 50 years old. So in some ways this might be an optimistic case. >> Yeah, yeah, so that's with the source on. We then have data for when you're turning that source off. >> They assume that the denapple source has been cleaned up in ten years. Now let's see what happens. So here's one year after the source is removed. There's actually this semi-detached plume that those red bars that are moving on. Let's look at 2 years they are sort of moving out there. Here's 5years, here's 10 years and with 10 years you can see there's still a lot of mass throughout that whole zone. Those orange lines in there aren’t in the sands, there in the clays. Let’s keep going, here’s 20 years, 30 years, 50 years, and then finally, we have 100 years. And even after 100 years of this flushing, and so this is 10 years of loading and 100 years of this sort of clean up period, you still have contaminants that are moving contaminants into those sands and those aquifers. So now let's wrap up and go through some of the key points. Number one is matrix diffusion is an important attenuation process because of these storage effects. >> Yeah, and then in some ways you can think of this process as acting like a capacitor in a circuit. It soaks up and stores the contaminants then slowly releases them over time. >> Okay, you're saying that because you took circuits in your education in college. >> That's what my transcript says, yeah. >> Okay, great. But the other point is, we sort of touched this a little bit, and we'll have a whole lecture about this in this week, the last lecture. That even a small amount of biodegradation in that matrix is important for the long term management of these plumes.
