Networking the Hudson

Data from the river will create a model for environmental monitoring.

A research consortium that includes the Beacon Institute, IBM, and Rensselaer Polytechnic Institute plans to distribute hundreds of sensors throughout the Hudson River. By collecting information on everything from salinity and temperature to oxygen levels and the presence of fish schools, the sensors will help create a “virtual river” that can aid scientists monitoring aquatic life and pollution levels.

River watch: In this artist’s rendering, a solar-powered autonomous underwater vehicle (foreground) joins forces with fixed sensors tethered to buoys (background).

Some sensors are likely to be mounted on a novel, solar-powered underwater robot developed by RPI, the Autonomous Undersea Systems Institute in Lee, NH, and Falmouth Scientific in Cataumet, MA. Other sensors will be fixed to buoys and suspended at various depths. In some cases, fiber-optic cables will convey data to the surface, where it will be sent ashore wirelessly. “This project is without a doubt a huge advancement [in sensor networks] and is on a much larger scale than anything that has been done before,” says Sandra Nierzwicki­-Bauer, a freshwater biologist at RPI and a leader of the effort.

Because of its scale, the network will demand a massive new data-­analysis system, which IBM will provide. One goal is rapid response to changing conditions, such as a sewage release or a drop in oxygen that could kill fish. Completing the design will take more than a year, but the first sensors will be placed in the river in early 2008. The full installation is expected to take three years.