Research

Built against a real river, in the season it is hardest to watch.

Thalweg was not developed against a benchmark dataset. It was developed against the lower Grand River in spring — cold, fast, sediment-loaded and entirely uncooperative — during the lake sturgeon spawning migration.

The study system

Lake sturgeon are a long-lived, late-maturing species from a lineage that has been swimming since the age of the dinosaurs. They are listed as threatened in the State of Michigan, and every spring adults move out of Lake Michigan and up the Grand River toward spawning reaches near downtown Grand Rapids.

The migration begins when the river warms to roughly seven degrees Celsius, and it happens largely at night in water carrying enough sediment that visual methods are not an option. A fixed acoustic station on the migration route can watch the whole of it. That is the deployment Thalweg was built on.

One crossing, in full

Fig. 2 A single target, followed across 143 consecutive pings. Distance from the instrument against time. It did not sweep through the beam the way drifting debris does — it stayed for nearly sixteen seconds, easing barely two metres closer the whole time. Pale marks are individual per-ping echoes; the line is the reconstructed path.

What the instrument measures

A fixed split-beam array is mounted on the bank and aimed across the channel. Because the beam is split, every echo carries direction as well as distance, so a target that crosses can be reconstructed as a path in three dimensions rather than a mark on a chart.

Thalweg turns those recordings into tracks, separates the fish from the very much larger population of things that are not fish, and ranks what is worth a biologist's attention. What it reports is beam crossings — the unit fisheries acoustics actually measures — with a distance, a direction and a duration attached to every one.

Field record

2025 season: Over 600 hours of sonar recorded at a fixed station on the lower Grand River, spanning the spring run from first movement to the end of the migration.
2026 season: A second spawning season at the same site, on rebuilt field hardware — remote access, automatic restart and off-site replication, so the station is managed from anywhere. Recorded, archived, and in analysis.
Publication: The method and the 2025 season are documented in a Grand Valley State University M.S. thesis (Biology, 2026).
Intellectual property: Patent pending — U.S. provisional application filed May 2026. All intellectual property is owned solely by its inventor.

A person in chest waders standing in the Grand River beside a deployed sonar mount, bare trees and grey water in early spring. — Fig. 3 **Deployment, lower Grand River.** The array goes in at low water in late February and begins recording as the river reaches the temperature that starts the run.

What comes next

Two seasons on the Grand River established the instrument. The next step is reach: a river Thalweg has not seen before, run alongside a fisheries or environmental monitoring programme that already knows what its water contains.

That is the collaboration we are actively looking for, and it is the fastest way to put continuous acoustic monitoring in front of the people who need it.

Discuss a deployment