Research Feature: Brian Wood Explains Upscaling

Posted: April 2, 2009

Photo of Brian Wood.
Brian Wood is an associate professor of environmental engineering.

Microbiologists often gather data at the cellular or subcellular level – scales so small it can be hard to relate the understanding gained to the more complex environments of colonies of cells living in natural and engineered systems.  Environmental engineering professor Brian Wood focuses on this problem – he uses a process called upscaling to synthesize microscale observations and use them to better predict macroscale behaviors. In this Web interview, Brian explains upscaling, describes some of his research and explains how upscaling can make a contribution to studies of the subsurface biosphere.

What is upscaling?

Upscaling is trying to understand how the essential features at the microscale affect macroscale observations.  Here is an example I often use to help people understand.  Say you have one mole of an ideal gas in a box and you know the temperature and your question is “What is the pressure of that gas?”  You could compute an answer in two ways.  The first way is based on information about the individual molecules within the gas.  If you knew the position and momentum of all of the 6.02x1023 molecules, you could determine the pressure by computing the mechanical impact of all of those particles on the sides of the box.  That is really hard.  The second way you could do the calculation is to use the equation PV=nRT.  Which is easier? It turns out that through averaging of all of momentum equations for the individual molecules, you can derive the much simpler equation.  This is the basic principle of upscaling – you are identifying some sort of regularity in the system or “scaling law” that allows you to take many of the details you are observing at the microscale and replace them with a more compact representation. 

How can upscaling make a contribution to research about the subsurface biosphere?

There is a lot of molecular-scale research going on in biology in general and in research about the subsurface.  The problem is that if you learn something about a cell in the lab, you don’t know how that translates to an observation of a real problem in the environment.  We are finding that there are a number of different characteristic time and length scales that define different processes – for example the movement of microbes in porous media.  Our goal is to identify scaling laws that will allow us to incorporate insights gained through microscale work and better predict bulk behavior at the macroscale of the environment or of engineered systems. 

My main message is that it is possible to relate fundamental microscale (cellular, subscellular) processes to how those processes work in more complicated environments.  Biofilms are a good example – say you understand how a cell works, and you then put those cells in a biofilm, and the biofilm in a porous medium – you might say, “I’m no longer sure how the cell works anymore.”  Well that’s not really true, you still understand how the cell works it’s just that you don’t know how the information is filtered by the more complicated system.  Upscaling creates possibilities because it allows you to draw those relationships in a defensible way.

Illustration of upscaling.
This illustration shows the different length scales associated with microbial transport in porous media. Upscaling can be used to relate observations made at the microscale to the macroscale behavior that is observable in the field.

You do both experimental and theoretical work – can you give an example of project where you are combining these approaches?

One project I am working on right now is with Roseanne Ford at the University of Virginia.  We are looking at the impact of bacterial chemotaxis on bacterial transport in the subsurface.  Chemotaxis describes how bacteria respond when they encounter a gradient of chemical attractant—it describes how the bacteria move up a gradient to find a food source. 

Roseanne has been working on experiments in microflow cells to look at how bacteria respond at the sub-pore scale to chemical attractants.  I have been working on the theory of the process.  Together with our post-doctoral researchers, we were able to take a theoretical understanding of how the bacteria respond – something that is very complex – and upscale this microscale phenomena to a macroscopic representation of the chemotaxis process that we are able to apply to engineered systems and to the environment.   

How has your research been influenced by new technologies?

There has been a huge revolution in imaging technologies things like MRI methods, x-ray tomography and synchrotron x-ray tomography have come a long way in recent years.  The resolution you can get with these techniques is amazing – sometimes you can image things down to a few microns.  Before we were able to measure things at a microscale, there really wasn’t a lot of point in asking some of these questions about how the microscale is related to the macroscale.  But now we can ask a question like “How does the microscale structure of biofilms in porous media influence the behavior of a bioreactor?” and we can collect the data to answer it.

The other thing that has changed in terms of technologies is computing.  It has become incredibly fast.  The things we can do on our desktops now are things that used to take a bank of Crays.  That means we not only can we make measurements at the microscale, but we can actually conceive of modeling the direct numerical representation of microscale processes.  For example, we can consider solving the Navier-Stokes equations for a softball size chunk of porous media to numerically test Darcy's law.  Twenty-five years ago people thought that was impossible. 

 

To learn more about Brian's research, contact him or read some of his recent publications:

Chastanet, J., and B. D. Wood. 2008. Mass transfer process in a two-region medium. Water Resources Research 44:W05413.

Ramirez, J. M., E. A. Thomann, E. C. Waymire, J. Chastanet, and B. D. Wood. 2008. A note on the theoretical foundations of particle tracking methods in heterogeneous porous media. Water Resources Research 44:W01501.

Saini, G., and B. D. Wood. 2008. Metabolic uncoupling of Shewanella oneidensis MR-1, under the influence of excess-substrate and 3, 3?, 4?, 5 tetrachlorosalicylanilide (TCS). Biotechnology and Bioengineering 99:1352-1360.

Wood, B. D. 2007. Inertial effects in dispersion in porous media. Water Resources Research 43, W12S16.

Wood, B. D., and R. M. Ford. 2007. Biological processes in porous media: from the pore scale to the field. Advances in Water Resources 30:1387-1391.

Wood, B. D., K. Radakovich, and F. Golfier. 2007. Effective reaction at a fluid-solid interface: Applications to biotransformation in porous media. Advances in Water Resources 30:1630-1647.