Photo Credit: Jeffrey Donenfeld

Scientists work in the computer server room in the IceCube Neutrino Observatory laboratory building. The increasing complexity of polar research demands an equally robust cyberinfrastructure to support it. The National Science Foundation is taking steps toward that goal.

Networking

Workshop brings together polar scientists and cyberinfrastructure experts

Antarctica is not as remote as it once was.

Planes reduce travel time to the southernmost continent to just a few hours. It’s possible to exchange instant messages from McMurdo Station to someone with a mobile phone riding the subway in New York City. Sea-going gliders can transmit oceanographic data to a server at a university. Telescopes at the geographic South Pole stream gigabytes of data across the world every day.

Still, the computer power of researchers working in the polar regions – the ability to generate, transmit, analyze and access data – lags in a number of ways.

“A lot of it comes down to the remoteness of the regions,” noted Jonathan Pundsack , managing director of the Polar Geospatial Center at the University of Minnesota , which hosted a Workshop on Cyberinfrastructure for Polar Sciences earlier this year.

The workshop brought together more than 60 polar researchers and computer scientists to discuss the needs of the former and to learn how the latter can enable a more robust cyberinfrastructure (CI) in the future.

“We’re trying to help science move at a faster pace and make discoveries that wouldn’t be possible without the combination of the two,” said Marco Tedesco , Polar Cyberinfrastructure Program manager at the National Science Foundation’s Division of Polar Programs . “It’s a call for a community that I think does not exist yet.”

Tedesco said the September workshop was the first of what he hopes are a series of meetings between scientists representing various disciplines and computer experts that will result in a long-term collaborative effort to build and sustain the sort of cyberinfrastructure needed in the Antarctic and Arctic.

Discussions ranged from big data – increasingly big, complex and unwieldy datasets – to supercomputers to education and training. A report from the workshop is now available online.

“The workshop was fantastic. There was a lot of interaction among the two groups,” Tedesco said, noting that it was important for computer scientists to understand the needs – and challenges – of researchers in the field.

Photo Credit: Nick Powell/Antarctic Photo Library

The South Pole MARISAT-GOES Terminal/RF Building complex is the primary facility supporting intra- and intercontinental communications.

“Field experience is invaluable in obtaining an understanding of what can and cannot be taken for granted in Antarctica in comparison to the stateside laboratory environment,” said John Helly , laboratory director for the Earth and Environmental Sciences at the San Diego Supercomputer Center , who straddles both worlds. He was one of the workshop organizing committee members.

In his role at the San Diego Supercomputer Center, Helly and his team conduct research in the application of high-performance computing resources to environmental and earth science problems, ranging from data acquisition to modeling and analysis. He has also participated in Antarctic fieldwork aboard research vessels on a project studying the ephemeral marine ecosystems created by melting, drifting icebergs. [See previous article — The hotspot: Icebergs become source of nutrients for biological communities as they melt.]

“The polar CI effort is not different than most of the other workshops that I’ve participated in general terms, although the focus on data-sharing and providing open-access to federally funded scientific data is a big shift in the culture,” Helly said. “It’s one we’ve been ambling toward for a long time, but the climate change debate really crystallized it at a societal level.”

Pundsack said the workshop particularly focused on data – “everything is data, data, data” – especially on its generation, storage, accessibility and searchability. Noted Tedesco: “You can have the best data in the world, but it becomes useless if it is not used – and to be used needs to be discovered.”

Another bottleneck for researchers in Antarctica is bandwidth – the ability to transmit gigabytes, even petabytes, of data from research vessels, remote observatories, and even the South Pole Station, which, due to its high latitude, has limited connectivity to satellite communications.

Four main priorities for the next two years emerged from the workshop, which Pundsack summarized:

• Data as a Service: The goals are to provide on-demand data sharing through discovery, access, transportation, and delivery service to the end user.
• Education and Training: A variety of training forms, ranging from informal workshops to formal education, is essential to maintain a sustainable and cutting-edge polar CI to enable polar sciences.
• Communication and Networking: Networking continues to be a major bottleneck in polar sciences. This includes syncing data with data centers when conducting fieldwork and freely moving data for polar research across data centers.
• Community Building and Community Portals: Polar CI is an emerging community crossing many disciplines, and the community needs proper mechanisms to improve the awareness, advance the building and utilization, and sustain the evolution of Polar CI.

Tedesco said he wants the polar research community to be aware that there is a place within the Division of Polar Programs to support polar cyberinfrastructure. He added that while the report from the first workshop will provide community input to the cyberinfrastructure program in the near-term, other ideas are, obviously, also welcome.

“Community building is very important at this moment,” he said.

For more information, visit the workshop website or contact Jonathan Pundsack (pundsack@umn.edu), managing director of the Polar Geospatial Center.