LSST Key Player in Sea Change of Data Availability

LSST continues to be recognized as a leader in the new astronomical research paradigm of Data Intensive Astronomy by pushing the envelope on all aspects: data mining, data sharing, and cyberinfrastructure. LSST’s contributions to the advancement of computational systems, the fostering of the next generation of cross-disciplinary scientists, and investments in the developing world’s cyber-infrastructure contribute to narrowing the gap between awareness of increasingly massive data collections and understanding of the knowledge within them.

Data Mining

Discovery of the unusual or unexpected in scientific data often guides the most fundamental breakthroughs in physical sciences. And while the expansion of data and computational resources have enabled new modes of discovery, as scientific data grows in size, our ability to understand it and to find the unexpected rapidly diminishes with current models of analysis.

Jeff Kantor, Project Manager for LSST Data Management, estimates that by the end of its survey LSST will have contributed an additional 100,000-plus terabytes of data to the trove of publically accessible astronomical data. Traditional methods of frame-by-frame searching to look for discrepancies within the data will be overwhelmed.

Data intensive astronomy utilizes statistical data analyses to enable rapid information extraction, knowledge discovery, and scientific decision support. This new paradigm came about to address the disconnect between the potential for exciting scientific breakthroughs to be found in massive datasets and the limitations of traditional analysis.

As an example of LSST’s contributions to the advances in data science necessary for data intensive astronomy, Kirk D. Borne, chair of the LSST Informatics and Statistics science collaboration and an associate professor of astrophysics and computational science at George Mason University, cites an international research effort to create an open-source database technology called SciDB. Designed to address extreme-scale science and inspired primarily by the needs of the LSST, the resulting architecture will enable analytics and knowledge generation on a scale that is practically unattainable with existing systems. Community support will ensure that SciDB and similar technologies will be available to meet the needs of data intensive science.

As much as astronomy needs computational advances, a new breed of scientist also must be fostered. New modes of interdisciplinary collaboration will require training and skills at the interface of astronomy, physics, computer science, statistics, and information science: astroinformatics. Borne recommends the development of astroinformatics as a formal sub-discipline of astronomy education and research. It provides a natural context for the integration of research and education – the excitement and experience of research and discovery are enabled and infused within the classroom through a portable informatics paradigm.

Data Sharing

The image archive produced by the LSST survey and the associated object catalogs that are generated from that data will be made available to the U.S. and Chilean scientific communities with no proprietary period. The LSST Education and Science Collaboration teams have generated ideas for Citizen Science research projects that engage the public in monitoring, classifying, and annotating data for the advancement of astronomical research. By encouraging the involvement of educators, amateur astronomers, and citizen scientists as well as the traditional research community, LSST extends its scientific potential and increases the diversity of those participating in the exploration.

Željko Ivezić, professor of astronomy at University of Washington and LSST Project Scientist, seconds the value of data-sharing. In a recent presentation at The Case for International Sharing of Scientific Data: A Focus on Developing Countries, an international symposium in Washington, DC, he said projects like the Sloan Digital Sky Survey have discovered that data-sharing produces a number of benefits, including the extraction of more science from the dataset and the democratization of expensive and limited astronomy resources throughout a broad community. “More users yield more science,” he observed.

Cyberinfrastructure

If data-sharing enables world-wide, cross-discipline coeval science and empowers small teams in developing countries to “do ‘big’ science,” as Ivezić puts it, then LSST’s collaborations, positioning of its full dataset in Chile and the US, a user-friendly data management system, and investments in Chile’s network and computing infrastructure position the project at the forefront of data intensive astronomy.

At meetings for the May symposium in Santiago, Chile, Towards a Digital Society through Advanced Connectivity Infrastructure, Kantor highlighted collaborations with REUNA and AmLight that will provide La Serena, Chile to Santiago networks and Santiago to Urbana-Champaign, Illinois networks. He also identified the LSST Chilean DAC and the new Chilean National Laboratory for High Performance Computing, networked with most major research universities in Chile by REUNA (with LSST investment), as the basis of a grid-based virtual laboratory for data intensive science, not just for astronomy.

At the same meetings Ed Seidel, Assistant Director for the Mathematical and Physical Sciences Directorate of the National Science Foundation, discussed the new NSF Director’s emphasis on economic development enabled by science, technology, and international collaboration. He cited LSST as a “model” for such projects. By making its data public and contributing to the development of Chile’s cyberinfrastructure, LSST satisfies two goals. LSST enables the production of more and better quality science, and it practices good citizenship by empowering the developing world.

Article written by Robert McKercher and Suzanne Jacoby