In the decades to come, a warmer planet could cause sea levels to rise by as much as 59 centimeters (23 inches). That may sound fairly benign. But the organization that devised that estimate — the Intergovernmental Panel on Climate Change — offers an important caveat: If glaciers outside Antarctica or Greenland melt at an accelerated pace, many coastal areas could find themselves swamped.

So that’s why researcher Rob Fatland, a research scientist with Colorado-based Vexcel Corp., a wholly-owned Microsoft subsidiary focusing on geospatial imaging and remote sensing technology is spending parts of three summers tromping around the surface of a vast ice field near the Alaskan city of Juneau. There, he and fellow researchers from the University of Alaska are working on a project colorfully dubbed SEAMONSTER, for Southeast Alaska Monitoring Network for Science, Telecommunications, Education and Research.

Sponsored by a $900,000 grant from the National Aeronautics and Space Administration (NASA), SEAMONSTER is aimed at developing tools for monitoring harsh environments while also furthering understanding of what could be a vital piece of the global-warming puzzle—how the world’s polar environments are responding to a warming planet.

Fatland sees this research as critical to furthering our collective understanding of global climate change, "Our current understanding is that the big ice sheets in Greenland and Antarctica will become increasingly important contributors to sea level rise in coming decades, as a huge amount of cold remote ice is mobilized into the oceans. Our research is aimed at understanding the global effect of the retreat of Alaskan glaciers and whether the great Alaska meltdown might be a precursor of bigger things to come."

The Juneau Icefield is a prime example of a vulnerable ice mass. The fifth-largest icefield in North America, it covers about 3,900 square kilometers (1,506 square miles). A popular tourist destination, the Juneau Icefield includes about 40 major glaciers and 100 minor ones.

Fatland and his co-researchers decided to focus on Lemon Creek, a stream that flows from the Lemon Glacier, one of the Juneau Icefield’s sub-glaciers. The object of the research was two-fold. First, devise sensors that could be left on the glacier or in Lemon Creek itself, measuring temperatures, stream flows and water chemistry. Second, begin to understand the dynamics of the glacier, and gain some understanding into how a sudden melt might effect nearby bodies of saltwater, such as the nearby Prince William Sound.

The impact of such a fast melt could extend beyond raising the sea level. As glaciers move, they grind down the rock beneath them, creating a fine dust called glacial flour, which is entrained in water and carried downstream. Once it reaches seawater, it can become an important source of nutrients for plant life—the crucial bottom of the ocean food chain.

"A big question we are interested in is how glacial melting—driven by climate—affects the water chemistry and the marine life in Prince William Sound, a very productive area for salmon, shellfish and other sea life," Fatland says. "All those things are affected by water from the various nearby watersheds."

Glaciers are tricky things to study, Fatland says. Their remote locations make data recovery difficult. And while views from space, such as synthetic aperture radar, create a "big picture" view of what is happening to the glacier, more information is needed to fully understand glacial dynamics.

According to Fatland, "data from space is incomplete. For example we can use it to see how glaciers flow down valleys but it doesn’t give us the complete picture. We can augment remote sensing information with in situ instruments like global positioning system (GPS) sensors to capture ice motion in fine detail, seismometers to listen to what’s happening in and under the ice, meteorological stations, and more.’

To bridge the data gap, SEAMONSTER relies on a web of sensors and microservers that create a linked network. The system is designed to operate largely autonomously in harsh conditions, relying on solar power and employing a miserly energy-use strategy of operation. The sensors and microservers communicate with one another wirelessly, for example via the 802.11g WiFi protocol.

SEAMONSTER is now in its second year of operation. While it may be years before significant discoveries come from the data collects, it already has yielded some surprises. For instance, a lake that forms and then rapidly drains on the surface of the Lemon Glacier each summer was thought to be the cause of a corresponding flood every summer down Lemon Creek. Instead, researchers learned, the summer flood through Lemon Creek actually comes from water accumulated beneath the glacier. The water discharge apparently relieves back pressure that keeps the lake in place, allowing it to drain several days after the flood maximum, not during as had been assumed.

That might appear to be a small discovery, but it also could be a significant piece of the bigger puzzle of how sub-glacial hydrology functions to govern glacial sliding and hence the transport of ice downhill. And that in turn has considerable relevance to understanding those huge reserves of cold ice in Greenland and Antarctica. Add to this the connection between glacier melt-water and the biogeochemistry of the watershed below and you have a complex and fascinating system about which we are just beginning to understand.

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 could automatically receive data from a remote site as an RSS feed, Fatland says.

Fatland has made use of Microsoft Virtual Earth to help display data from SEAMONSTER in an easy-to-see way. He and his co-researchers also have worked to make the project easily accessible as a teaching tool for students.

In 2008, Rob began work on "Sensors to Science", a Microsoft Research project that will combine SEAMONSTER’s innovative sensor and network development with several related Microsoft environmental science technologies. This innovative research aims to showcase the potential of software to help scientists better understand changes in our global environment with geospatial technologies.