Built
A probe, not an opinion
An ESP32 rig that reads the parameters we care about and condenses them into a two-sentence report on the water it just sampled. Bench-tested on tap water first, so we know what a clean baseline looks like.
Loudoun County, Virginia · 39.05°N 77.49°W · Step Up Loudoun Youth
We are six high school students. We built a probe that reads the water, and we are taking it to the ponds and creeks that never make it into anyone's dataset.
Why we started
Solving this will take longer than we have. We think it still starts the same way: pick a body of water, measure it honestly, write down what you found, and do it again next month.
Last year a 64-year-old Virginia man was wading in shallow brackish water near Virginia Beach with a small cut on his leg. The wound was exposed to Vibrio vulnificus, a bacterium that occurs naturally in warm, salty and brackish water, including near the Chesapeake Bay and the Potomac. He died. Most people have never heard of Vibrio, and cases like this are rare, but they are the sharp end of something ordinary. Urban runoff kills aquatic life, pushes fish past the point where they are safe to eat, and knocks the balance of a waterway out of order. It happens because dumping runoff is cheaper than treating it. Stretches of the Potomac are unsafe to swim in today, between the algal blooms, the runoff, and the bacteria that come with them.
Our own start was smaller. One of us, Sudhit, went fishing on rivers near the Potomac and came home with nothing but algae, plastic bags and bottles. He complained to a friend who had been to a different spot and seen exactly the same thing. So we gathered a few more people and started the Aqua Track Project. The more we read about the water in Loudoun County, the clearer it got: stable freshwater ecosystems here are the exception, not the rule. That became the whole focus.
Built
An ESP32 rig that reads the parameters we care about and condenses them into a two-sentence report on the water it just sampled. Bench-tested on tap water first, so we know what a clean baseline looks like.
In progress
Carry the probe to access points along the Potomac and its Loudoun tributaries, plus the neighborhood ponds nobody samples, and log every reading with its location.
Next
Put every reading on a public map under one color-coded index, so a neighbor can see how the water near them is doing without reading a technical report.
Long run
A county with more stable freshwater than it has now, and a record consistent enough that whoever comes after us can pick it up and keep going.
Interactive · demonstration model
Move the sliders to the conditions you want to test. The column responds the way the real thing would, and the panel writes the same kind of two-sentence report our probe produces in the field.
Load a sample
Water quality index
86/100
Good
Probe report
Water quality reads Good at 86 out of 100. Every parameter sits inside the healthy range for a freshwater pond.
This panel runs a simplified index built for the web. The v1 probe reads the physical parameters, pH and oxygen; the nutrient, chloride and E. coli sliders show how test-kit readings would weigh in. One failing parameter caps the score. It is a demonstration of the scoring, not a live feed from a sensor.
Survey chart · sampling not yet begun
Nine reaches across the Potomac basin, drawn from our own route planning. Select one to see what we intend to sample there. Field work is paused for winter. When it restarts, each reach fills in with its measured grade.
Index the map will use
Project log
Updated as things move. Anything filed under Next or Later is planned, not finished. We would rather say so than pad the list.
Now
The ESP32 rig is assembled and bench-tested on tap water. It takes its parameter readings and writes a two-sentence summary of the sample.
The competition is our starting line, not the finish. It gets the project in front of the county so the map has an audience when it goes up.
Cold weather is keeping us off the water, which is why the map is still empty. It fills in as soon as we are back out.
Next
Goose Creek, Broad Run and Difficult Run, plus the neighborhood ponds around Ashburn and Leesburg: the small water that never gets a data point.
Each sampled reach gets its grade and the date it was taken, under the same color-coded index shown above.
Short posts from each sampling trip, so following along does not require checking the site.
Later
Enough repeat readings across Loudoun that the map shows change over time, not just one afternoon's snapshot.
A record and a method consistent enough that the next group of students can pick it up without starting over.
Six students · Loudoun County
We are high school students with backgrounds in leadership and engineering, and a shared irritation about what our local water looks like up close. Nobody is paying us and nobody assigned this.
Take action
That is not modesty, it is arithmetic. What actually moves the number is what happens on the ground upstream of every reach on that chart: driveways, lawns, storm drains. Most of it is small and most of it is free.
A storm drain is not a sink. Whatever goes down it (paint, oil, soapy car-wash water, grass clippings) arrives in a creek untreated, usually within the hour.
Fertilizer and grass clippings on a driveway wash straight into the nearest waterway and feed the algal blooms that pull the oxygen out. Sweep them back onto the grass instead.
Road and driveway salt spikes conductivity in winter and does not break down. A thin, deliberate spread clears ice about as well as a thick one.
Pet waste left near a pond is a direct bacterial load on the water. It is the single easiest input on this list to remove.
The other thing that helps is attention. Follow the sampling, check the map when it fills in, and tell someone upstream of you about it.