Director’s Thoughts on Astro 2010

In August, the Astronomy and Astrophysics Decadal Survey “New Worlds and New Horizons in Astronomy and Astrophysics,” convened by the National Research Council for the National Academy of Sciences, ranked the LSST as its top priority for the next large ground-based astronomical facility. This so-called “Astro2010” report is important input to the coordinated planning process at the national funding agencies. We are working with the agencies to move LSST into construction, navigating parallel coordinated tracks at DOE and NSF.

From the beginning of this project we envisioned LSST as a community endeavor, with the data and the software open. We had open focused workshops, posters at meetings, and formed community-based LSST science collaborations. These science collaborations, together with project personnel, produced the LSST Science Book. Attendance at our “All-Hands” meetings has increased every year, and it would be an understatement to say that our most recent All-Hands meeting was successful. Held just north of Tucson August 9-13, it overlapped with the 2010 NRC decadal survey Astro2010 public release.

For the 218 attending the All-Hands meeting — many of them students — Friday August 13 at 9am was a thrilling moment. Astro2010 ranked LSST as its highest priority for new ground-based facilities. In their words: “The top rank accorded to LSST is a result of (1) its compelling science case and capacity to address so many of the science goals of this survey and (2) its readiness for submission to the MREFC process as informed by its technical maturity, the survey’s assessment of risk, and appraised construction and operations costs. Having made considerable progress in terms of its readiness since the 2001 survey, the committee judged that LSST was the most ‘ready-to-go.’”

I believe that this ranking reflects the community’s perception that LSST will open new windows on the universe, providing an exciting and wide-spread capability for data mining the sky for new science. That community extends beyond professional scientists to the general public and their curiosity about the sky. For scientists it is LSST’s “Big Picture” and the huge statistical samples covering 20 billion objects that present unique opportunities, many of which are recounted in the Science Book. These opportunities — and more — depend not only on the quantity but the quality of the data. This is where LSST will have to deliver. Like previous surveys, much of the science enabled by our data products will be done at the faint end, where most of the survey volume resides. It was apparent at this recent All-Hands meeting that the science collaborations are enthusiastically engaging with the simulations in an effort to understand in some detail the sensitivity of their science to the parameters of the survey, including photometric and astrometric precision. Not surprisingly, in many cases the science is enabled by uncommon control of data quality and uniformity. This was difficult enough in the SDSS and other recent surveys. Considerable effort was spent on detecting and understanding systematics. LSST will produce a SDSS volume of data nightly!

Many regard LSST as the lighthouse project for looming Big Data challenges that are faced by most areas of science and engineering. Gone are the days when a lone postdoc reviewing raw images can grasp the full landscape of the data completeness and accuracy, or discover that pesky sample bias or systematic error. Novel tools for data exploration and visualization must be utilized. Efficient human-assisted tools for automated data quality assessment must be developed. Indeed the very process of discovery in some cases will be automated. Discovering the unexpected in petabytes of data is an exciting challenge. For us, controlling systematics and understanding selection effects is an LSST system-wide issue, from hardware to survey cadence to software algorithms extending beyond pipelines to data analysis.

This data-to-knowledge challenge at the petascale is one area where our funding partners see potential spin-off. Together with the breakthrough astrophysics, this will be an important theme as we advance through the approval process at NSF and DOE. The recent Astro2010 report is important input to the coordinated planning process at the agencies. We are working with the agencies to make that happen soon, moving along parallel coordinated tracks at DOE and NSF. Three years ago the NSF LSST CoDR review panel report said that we were ready for Preliminary Design Review. With the decadal survey looming, it was a naturally conservative decision for the agencies to wait for the Astro2010 report. For their part (the LSST camera), DOE is beginning to advance us through their Critical Design phases.

At NSF our next review is PDR (stay tuned!), and then Final Design Review. The key element in the NSF approval process is the National Science Board. NSB must approve all major NSF facility construction, the so-called “MREFC” approval, and this has several phases. NSF’s Division of Astronomical Sciences plans to brief the NSB on the Astro2010 report soon. While Astro2010 has now confirmed LSST’s high degree of readiness, this is a necessary but not sufficient condition on our way towards MREFC approval. As illustrated in the accompanying colorful graphic, funding major facilities is a non-linear process involving review committees, the agencies, Office of Management and Budget, Office of Science and Technology policy, and Congress. We will continue to engage the community and work with our funding agencies and private supporters to open new windows on the universe with LSST. We are on our way!

This article written by Tony Tyson.