I was commissioned to build this model in support of a presentation about geotechnical engineering. The goal is to illustrate the flow paths that groundwater takes under an obstruction (e.g. a sheet pile or cutoff wall). So much of engineering is just theoretical work, so it was really cool to see such an elegant example of a geotechnical engineering concept borne out in real dirt and water.

There’s an old joke about the different kinds of engineers that says mechanical engineers design the weapons and civil engineers design the targets. Well it’s even worse than that for geotechnical engineers, who really just care about what’s underneath the targets. Yet, despite dirt’s synonymy with grit, grime, and gossip, its importance in civil engineering can’t be overstated. There’s hardly a single structure out there that doesn’t sit on the ground or at least sit on something that sits on the ground, and there’s really more to earth than first meets the eye. For the most part, geotechnical engineers are content to perform their analyses quietly knowing full well that the general public does not share their devotion to dirt and reverence for rock. But occasionally they find themselves with the desire to educate and inform, in which case, models often speak louder than words.

In the field of civil engineering, it’s often important to be able to characterize the flow of groundwater. Water in the subsurface can have a major impact on civil structures by causing uplift pressure, seepage, and changing the strength characteristics of the soil, among other things. We have fancy computer models that do a good job simulating groundwater flow now, but even today one of the most important tools used by geotechnical engineers is the flow net. Without getting into the nitty gritty details, a flow net consists of two sets of perpendicular lines which create a curvilinear grid: One set are equipotentials or lines which connect points with the same pressure. Once the equipotential lines are drawn, the flow lines are just drawn perpendicular to them forming squares. These kinds of simplified drawings make pretty pictures, but do they really reflect how groundwater flows in real life? It’s always important to ground-truth your calculations, sometimes even with real ground.

This model is designed to do just that. It simulates the flow of groundwater around an obstruction to illustrate the morphology and velocity of the flow. It’s made of quarter-inch acrylic sheets cut to size on the table saw. Solvent welding is used to connect the acrylic sheets into a narrow box. All of the plumbing is composed of aquarium bulkhead fittings and clear nylon tubing. Everything was leak tested before the sand was added.

Here’s how it works: Potassium permanganate is added to three spots in the top of the sand. A pump in a bucket below keeps a constant head pressure on one side of the model. The differential between the sides is what drives the water to flow from one side to the other. Groundwater flowing through the sand creates traces of the flow lines over the course of several hours, illustrating exactly what we estimated earlier with the flow net.

The flow of groundwater hasn’t always been so well-understood. In fact, for some states, regulation of the pumping of groundwater is established on explicit ignorance of its behavior. In a landmark case which cemented the “Rule of Capture” into Texas Water Law in 1904, the courts said that, [groundwater] movements are so secret, occult, and concealed that regulating the use of groundwater would be practically impossible. Fortunately, geotechnical engineers since then have developed a foundation of knowledge around the flow of water in the subsurface.

It can be dirty job, but a large part of geotechnical engineering is relegated to abstract calculations and computer models. I was glad to get a chance to take a concept away from the desk and illustrate it with some real dirt and water. Thanks for watching, and let me know what you think.