Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.adass.org/adass/proceedings/adass03/reprints/O1-3.pdf Дата изменения: Sat Aug 28 02:17:59 2004 Дата индексирования: Tue Oct 2 10:09:35 2012 Кодировка: Поисковые слова: total solar eclipse

Astronomical Data Analysis Software and Systems XIII ASP Conference Series, Vol. 314, 2004 F. Ochsenbein, M. Al len, and D. Egret, eds.

Palomar-QUEST: A case study in designing sky surveys in the VO era
Matthew J. Graham, Roy Williams, S. G. Djorgovski, Ashish Mahabal California Institute of Technology, Pasadena, CA 91125, USA Charles Baltay, Dave Rabinowitz, Anne Bauer, Jeff Snyder, Nick Morgan, Peter Andrews Yale University, New Haven, CT 06511, USA Alexander S. Szalay Dept. Physics and Astronomy, The Johns Hopkins University, Baltimore, MD 21218, USA Robert J. Brunner Dept. of Astronomy & NCSA, University of Il linois, Urbana, IL 61801, USA Jim Musser Indiana University, Bloomington, Indiana, USA Abstract. The advent of wide-area multicolour synoptic sky surveys is leading to data sets unprecedented in size, complexity and data throughput. VO technology offers a way to exploit these to the full but requires changes in design philosophy. The Palomar-QUEST survey is a ma jor new survey being undertaken by Caltech, Yale, JPL and Indiana University to repeatedly observe 1 of the sky ( 15000 sq. deg. 3 between -27 27 ) in seven passbands. Utilising the 48-inch Oschin Schmidt Telescope at the Palomar Observatory with the 112-CCD QUEST camera covering the full 4 x 4 field of view, it will generate 1TB of data per month. In this paper, we review the design of QUEST as a VO resource, a federated data set and an exemplar of VO standards.

1.

A new era in astronomy

The new availability of wide-field images from Schmidt telescopes in the 1940's meant that astronomers no longer had to make educated guesses about where to look to find new and interesting phenomena but were now spoilt for choice. The advent of synoptic surveys presents more extreme opportunities; as an illustration, consider the SDSS which over the course of 5 years represents a factor 14 c Copyright 2004 Astronomical Society of the Pacific. All rights reserved.


The Palomar-QUEST survey

15

of a million increase in information over previous surveys; however, the LSST1 (Large Sky Synoptic Telescope, Tyson (2002)) will amass a SDSS every 3 nights. Although overviews of synoptic surveys are riddled with cliches concerning undiscovered countries and uncharted waters, the exploration of the temporal domain results in data sets that are not just more voluminous than before, but far richer and more complex (Paczynski 2001; Djorgovski et al. 2000). This presents challenges to all aspects of astronomy: data gathering, distribution, reduction, analysis, storage, archiving, dissemination and mining. VO technologies are being designed precisely to meet these types of challenges, but to use them requires changes in survey design philosophies. 2. The Palomar-QUEST survey

The Palomar-QUEST survey2 is a ma jor new survey being undertaken by Caltech, Yale, JPL and Indiana University employing the world's largest astronomical camera and the recently refurbished Oschin Schmidt telescope at Palomar to observe a third of the sky ( 15000 sq. deg. between -27 27 ) a minimum of 8 times in 7 passbands to nominally twice the depth of SDSS. The QUEST camera consists of 112 CCDs arranged in four filter strips. Each CCD has 2400 в 600 13µm в 13µm pixels, giving a total of 161 в 106 pixels. At the prime focus of the Oschin Schmidt, QUEST covers a sky area of 4.6 в 3.6 (the effective area is 10 sq. deg) and in a night can survey 500 sq. deg. Two filter sets are used: Johnson UB RI and Gunn riz , with a doubling of Gunn z to afford extra depth. The data rate is 2.45MB/s and with a monthly average of 10 nights' observing, QUEST produces 1TB of data/month. Some of the immediate science goals are searching for high redshift quasars, strong gravitational lensing, supernovae and GRBs, and near-Earth asteroids and trans-Neptunian ob jects. Obviously once there is a sufficient body of repeat observations, searching for new types of variable ob ject and phenomena will play a dominant part; in particular, a rapid response mechanism to transients (see section 4) is planned. 3. QUEST and VO technologies

As this survey is one of the first of the new breed of synoptic surveys, it is being used as a testbed for the VO technologies which will enable astronomers to exploit such surveys to the full. There are currently four areas of attention: 3.1. Data distribution

Different groups want to process the raw data in different ways to optimize the detection of specific types of ob ject. Access requirements to the data are also
1 2

http://www.lssto.org http://hepwww.physics.yale.edu/quest/palomar.html; http://www.astro.caltech.edu/~aam/science/quest


16

Graham et al.

either near real-time or delayed. Data distribution must be secure, fault tolerant (error checking, multiply redundant) and accountable (transaction logging). 3.2. Data processing

The nature of the data is extremely well suited to parallelization, either on a multi-processor machine or in a more general distributed computing environment, e.g. an advanced highly CPU-intensive pipeline would be a suitable Grid-level application. 3.3. Data analysis

The identification of variable ob jects poses many problems: · associating different observations under different conditions (e.g. seeing) with the same identification; · handling ob jects which only appear once (e.g. supernovae) · handling moving ob jects (e.g. asteroids) · optimally characterizing the variability of an ob ject (periodic/aperiodic) · determining the best sampling strategy to maximize the range of temporal baselines covered Other federated data sets will be employed in the data analysis to assist identification, e.g. SDSS, DPOSS, 2MASS. 3.4. Data dissemination ederated data set needs to support both inAccess to data products also needs to be different users: QUEST survey team, collabcommunity.

The deployment of QUEST as a f teractive and batch mode access. transparent to the access rights of orators and the general astronomy 4.

Integration example: 4 minute alert of detection of transient events

To illustrate how QUEST will make use of VO technologies in an integrated fashion, consider one of the pipeline systems under construction (see Fig. 1 for a cartoon depiction): this will produce real time (within four minutes of the data being taken) alerts of transient events (e.g. supernovae). The specific processes which need to mesh are: · Distribution: Every 140s, 112 в 3.1MB raw fits files are produced at Palomar and streamed to Caltech (at 10MB/s) where the CIT Data Broker distributes the data to other sites, the raw image archive and the reduction pipeline · Processing: The CIT Fast Pipeline computes a real-time flat and extracts ob jects - each field produces 10000 ob jects · Analysis: Variable and transient ob jects are detected by comparing the latest observations with the fiducial sky (composed from all QUEST observations and possibly other data sets) in the master archive. They are processed to determine whether they might be asteroids and checked against lists of known variables. Source classification is attempted using other federated data archives.


The Palomar-QUEST survey

17

Figure 1. Cartoon of the real time variable ob ject detection pipeline which will issue an alert within four minutes of detection. · Dissemination: The Alert Decision Engine decides whether an alert should be issued based on decision algorithms and all available data and posts results to the website. 5. Conclusions

Palomar-QUEST is the prototype VO-integrated synoptic sky survey and marks the beginning of an exciting new era in astronomy: the characterization of the variable optical sky. Acknowledgments. MJG, RW, SGD and AAM acknowledge support from the NSF NVO program. SGD, AAM and MJG also acknowledge a partial support from the NASA AISRP program. References Djorgovski, S. G. et al. 2001, in Virtual Observatories of the Future, ed. R. J. Brunner, S. G. Djorgovski & A. S. Szalay (ASP Conf. Ser. 225), 52 Paczynski, B. 2000, PASP, 112, 1281 Tyson, J. A. 2002, SPIE, 4836, 10