CHUCK NEWELL: Today, we're going to talk about a key issue, how should you monitor for monitored natural attenuation? How do you get the numbers from the field site to make decisions? This is a rapidly changing area, lots of new thinking about site characterization and long term monitoring. DAVE ADAMSON: And one of those new developments is the concept of high resolution sampling, right? CHUCK NEWELL: That's right, Dave. So this is the idea that a lot of our conventional monitoring techniques were designed for, nice, homogeneous sand, represented by this sand tank experiment here. DAVE ADAMSON: Yeah, but we also have this photo. I think we've shown this before. This is from Fred Payne at ARCADIS, and it's a good representation of the real nature of our subsurface domain. Just look at the geologic media. Pretend those pores are filled with water, and the water has dissolved contaminants that we need to sample to see if the contaminants are degrading, being stored, or are being more dilute. CHUCK NEWELL: That's right. So we also want to know if the plume is shrinking, stable, expanding. So Dave, tell us about our conventional tools, or what you call this first generation sampling? DAVE ADAMSON: Well, this slide shows a summary from a recently completed SERDP project, of ER 1740. It has a chapter on characterization that I helped to write with co-authors Tom Sale and Beth Parker. The original approach, we're basically taking the design of a drinking water supply well and converting it to a monitoring well with a long screen. And early on in the groundwater field, we didn't really snap to this, but it was providing a flow weighted average dominated really by those high permeability zones within that screen interval. CHUCK NEWELL: OK, so that's the ground water. But we're also used to thinking about contaminants coming from landfills, so we were very focused on the idea that this leachate was the most common source of contaminants. So any soil sampling was off in the beta zone of these soils. But tell me more about these flow weighted averages. DAVE ADAMSON: OK, we've got another graphic here. This is an example of a high resolution sampling on the vertical interval from 185 meter elevation to 175 meter elevation, about 30 feet, in this case. So those red dots are discrete groundwater samples showing PCE concentrations in micrograms per meter on the x-axis, and the right hand panel shows the hydraulic conductivity versus elevation. It's highest right there near the top of the screen. CHUCK NEWELL: So most of the wire in the well is coming from the top of that screened interval, where the concentrations are lower than the rest of the screened interval, right? DAVE ADAMSON: Right, right. And so the flow weighted average in this case is around 2000 micrograms per liter. CHUCK NEWELL: I can see some of those red dots. And you notice, that's a log scale x-axis. They've got concentrations up to 40,000 micrograms per liter. DAVE ADAMSON: Yeah, a monitoring well with the typical screen lengths, you miss key features like this. CHUCK NEWELL: OK, well, now let's go to second generation approaches. DAVE ADAMSON: OK. Second generation approaches really have the ability to see what's going on within that 10 foot resolution of that monitoring well. CHUCK NEWELL: Right. So this graphic right here from ITRC, mass flex technology guide speaks to that. Originally built by Fred Payne, it asked the question, what is the minimum size feature that we are able to detect? What the approximate grid spacing that we need? So this gets really to the heart of high resolution sampling. DAVE ADAMSON: Isn't there some phrase that you like that high resolution sampling changed your life? CHUCK NEWELL: Well, it's a bit tongue in cheek, but people talked about this. And it suggests that when you that you can see a completely different picture when your characterization is done at a high resolution, that you can instead of this fuzzy cloud of this sort of general shape of this plume you can see these high flow zones carrying high contaminant loads and then see where these low flow zones are storing contaminants. DAVE ADAMSON: All right, well next, we've got this great image. It's also in that ITRC guidance. It's combining these ideas of high resolution sampling and plume life cycle. In panel A, you have this expanding plume with each red dot being a high flow channel. The longer the channel has been active, the more matrix diffusion you have. So over time, the plume marches forward. On the third graphic from the left on panel A, you see that see the edge of the plume, right? CHUCK NEWELL: OK, and then at some point due to attenuation processes, the plume stabilizes then starts to shrink, right? DAVE ADAMSON: Yeah, and those processes include things like source attenuation, degradation, and diffusion. CHUCK NEWELL: Then as the source is exhausted or remediated, then the plume is sustained by back diffusion, as you see on the bottom panel here. OK, and I think in the next slide we're going to go back to the expanding plume and take a look at that. DAVE ADAMSON: OK, so here you have a mass discharge, mass per time versus distance on the plume axis for three different times. CHUCK NEWELL: OK, so the red box is the plume if it was traveling in a sandbox with no degradation and very little dispersion. The lines, T1, T2, and T3 show the actual plume progression away from the source. And due to degradation and storage in this low permeability zone, you get the resulting plume. Now, this is somewhat of a new concept, but the idea is with high resolution sampling you'll be better able to understand the key attenuation processes, which include degradation, absorption, and then storage in the case of many of these chlorinated solvent sites. Now, let's list some of the key themes that people have talked about when they talk about this high resolution sampling. What's the first one here? DAVE ADAMSON: Sure. So basically, we need to remember that all compartments and all processes should be considered during site characterization. Don't ignore anything just because you don't think it's relevant. CHUCK NEWELL: Got it. OK. Number two is that this heterogeneity is an important factor in governing contaminant fate and transport. DAVE ADAMSON: Next, high resolution characterization is key to understanding heterogeneity. The scale must be appropriate for the site conditions. CHUCK NEWELL: Sounds good. And there are some new rapid data acquisition tools and methods that can provide valuable data, for when you're trying to get this integrated site approach. DAVE ADAMSON: And then site characterization must be dynamic and adaptive. CHUCK NEWELL: OK, so that sounds pretty good. So those are some of the themes that we're talking about. Want to go through some of the tools then? DAVE ADAMSON: Yeah. So in this case, there's things like membrane interface probe that has constituent specific detection capabilities, so a real powerful real time data acquisition tool, as well as optical streaming tools. CHUCK NEWELL: Like those Ross things that we use at LNAPL sites? DAVE ADAMSON: Yeah. CHUCK NEWELL: There's also a lot more emphasis on direct push technologies for these unconsolidated aquifers, and then sort of a rethinking, maybe we still can get information out of geophysical techniques. DAVE ADAMSON: Rapid field extraction analysis, so mobile labs are becoming very popular in terms of doing this high resolution sampling, as well as core collection techniques, things like cryogenic coring that allows you to capture more of that contaminant mass. CHUCK NEWELL: Very cool stuff. DAVE ADAMSON: Yeah. CHUCK NEWELL: OK, then better tools to measure and understand attenuation rates, and tools for understanding the accessibility of the contaminants and the amendments. DAVE ADAMSON: And then we've got mass flux techniques, so tools that are designed to give you mass discharge numbers, as well, things like the passive flux meter, right? CHUCK NEWELL: Right. And then finally, simple modeling tools to help you think about data interpretation and the visualization tools. So there's some of the key themes. But let's move on to two special topics, OK? First, in some ways we've been saying be careful about monitoring wells. Maybe they aren't the right tool that we've been looking-- we were trashing monitoring wells, in some ways. But were there any scenarios where monitoring wells are really useful? DAVE ADAMSON: Well, we've been talking about using high resolution sampling for site characterization, where you go out and try to understand a site and build your conceptual model. CHUCK NEWELL: But some groundwater experts, such as Brian Looney of the Savannah River National Lab, suggest that for long term monitoring, monitoring wells may be very useful. They can sample this vertical interval and even though the numbers are flow weighted, it can be a more efficient way to observe maybe more of the plume over years and years. So that's this idea that it's OK to do this, in terms of the change looking at changes over time. OK, let's go to the last special topic, using high resolution sampling to get mass discharge and mass flux and how do you calculate rates from the mass discharge data. So here's this US EPA issue paper about calculating monitored natural attenuation rates. It's something I coauthored with some EPA co-authors and other GSI colleagues. But in here, we discussed doing this rate calculation with both concentration and mass discharge data. The key point is that there are two main kind of rates, and they're apples and oranges, and you don't want to get them confused. DAVE ADAMSON: I like apples, so I'm going to start here on the left. This is a concentration versus time rate, natural log of concentration versus time. What does this rate tell you? CHUCK NEWELL: OK, so if the well was close to the source, it really reflects how quickly the source itself is being attenuated, not the plume. So as this mass flux or mass discharge keeps going down. We call this a K point, or a first order rate at a particular point in the plume. And on the right, it's the orange. OK, it's a different type of rate that Dave talked about. It's the natural log concentration versus distance. It measures the degradation of contaminants in that packet of water as it moves down gradient. It represents a degradation rate of dissolved contaminants. And sort of as they leave the plume and sort of head downstream. So the key thing is you don't want to get these rates mixed up, because they tell you two different things. DAVE ADAMSON: All right, well, all this talk about apples and oranges, I'm a little hungry. So let's wrap up and get ready for lunch. CHUCK NEWELL: That sounds great. OK, so we're going to wrap up some of our key points. Second generation are these high resolution tools. It's a collection of key themes and tools that you can use. DAVE ADAMSON: And they allow you to characterize the scale, where you can see high flow zones versus storage zones. CHUCK NEWELL: OK, mass flux, mass discharge, transects is one example of this type of high resolution characterization. DAVE ADAMSON: And you should use high resolution techniques. They're good for characterization. But you can still use modern wells efficiently for long term monitoring.
