[Reprinted with permission from the July 10, 2006 edition of Space News]

Sensor Web To Link Scientists to Remote Alaskan Sites
Missy Frederick, Space News Staff Writer
July 10, 2006

Alaska’s unique and harsh terrain provides a wealth of interesting phenomena for scientists of all disciplines to study, but keeping a constant flow of data from remote areas is not easy. A partnership between Microsoft’s Virtual Earth unit and the University of Alaska Southeast in Juneau aims to solve that problem by creating a network that will link a variety of scientific experiments.

The university received an $881,755 grant from NASA to create a wireless data network that will retrieve data from remote areas of the region, according to Matt Heavner, an assistant professor of physics there. The three-year grant will be used to create the Southeast Alaska Monitoring Network for Science, Telecommunications and Educational Research, or Seamonster, a wireless network that will be designed to reduce the cost of data transmissions while providing a permanent network for linking experiments.

Tiny sensors will be placed in multiple locations, according to Rob Fatland, a program manager for Microsoft and formerly a research scientist with Vexcel Corp., which Microsoft purchased in May. The sensors will acquire different types of electronic data and be able to communicate with each other, as well as with a main computer that processes the data.

“It’s like taking a computer and putting it out into the field,” Fatland said. “Some people ask: ‘What’s the point of having a computer sitting out in the middle of a glacier or tree?’ But there’s an explosion going on in this area, and we’re trying to do our little piece of it.”

The project was born out of a similar endeavor that Fatland worked on during graduate school at Penn State University of University Park, Pa. “It seemed like a natural thing to extend the idea of a network of sensors to, in and around Juneau, Alaska,” Fatland said. “It’s a fascinating environment.”

The network originally will focus on projects in the Juneau Icefield and look at expanding in its second and third year into Berners Bay, and potentially farther north into Glacier Bay. Projects in the area are studying trends such as the melting rate of glaciers or the human impact on the remote Alaskan landscape.

One experiment in the first year of the program will be a study of runoff from Lemon Glacier, which is located approximately 24 kilometers from the university campus, Heavner said. Runoff from the glacier goes through both industrial and residential areas of Juneau, and sensors could be placed at various spots along its route to examine the water’s properties and the potential damage it can cause.

At the Mendenhall Glacier, which is visible from the campus, the sensors will be used to determine how much accumulation the glacier gets in the winter, and the flow rate of melting during the summer, Heavner said.

If the project expands farther, northern locations can be monitored for more biological data, such as whale activity and noise issues, Heavner said.

Glaciers in particular can be a tricky thing to study, Fatland said. There can be problems recovering data from remote locations. And while synthetic aperture radar from space creates a “big picture” view of what is happening to the glacier, more information is generally needed to fully understand its dynamics.

“Data from space is incomplete,” Fatland said. “It doesn’t tell you about three-dimensional motion.” Instead, global positioning system (GPS) instruments, as well as “geophones” which listen for echoes, are also needed.

Satellites and the network are essentially complimentary systems for research, with each adding an essential component for understanding a particular environment, Fatland said.

“You can coordinate the ground sensor web with what you see from the satellite,” Fatland said. “I see the satellite as a tent covering huge areas, with the sensor web as a pole that holds the tent up.”

Part of Seamonster’s appeal is that it can help decrease power consumption, Heavner said. The smart sensors save power because they are programmed to shut on and off so they are only being used during prime data collection times.

The vision for Seamonster is that it eventually will become a wireless backbone that any investigator or Earth science student can plug into for their research. Scientists will automatically be able to receive their data each morning as an e-mail attachment, Fatland said.

Fatland originally began working on the project when still employed by Vexcel. Now that Microsoft is on board, the company hopes to incorporate the data obtained by the network into its Virtual Earth software system, which relies on satellite imagery to create a picture of the entire planet.

“If you were to imagine a student or member of the public logging on to Virtual Earth, and ‘flying’ onto the terrain, and then clicking on a blinking item to retrieve their data, that’s what we’re envisioning,” Fatland said.

Fatland hopes collaboration with Microsoft will lead to the development of more sophisticated sensors and communications methods as well.

Seamonster’s initial three-year term will focus on a geographic area of approximately 400 square kilometers, but Fatland said the eventual goal is to expand the network to cover an area of some 700,000 square kilometers. As the network grows, the scientists will look at using long-range radio modems or satellite links to bring the data back to the home base.

The program also serves as an outreach program to the university’s undergraduate students; $14,000 of the grant is devoted toward research opportunities for them in conjunction with the project.

“There’s a great opportunity just to get the students up there at the glaciers, and excited about science,” Heavner said.


[Reprinted with permission from the July 10, 2006 edition of Space News]