LSST Observatory - Meetings - Science with LSST and Other Large Surveys

Science with LSST and Other Large Surveys:
Community Access and Utilization of Future Archives
What will LSST do for You?

University of Washington, Seattle
September 20-22, 2004

Poster Summary (9:30 - 10:00 a.m.)
Suzanne Hawley, University of Washington

Name: David R. Alves
Institution: National Research Council, Goddard Space Flight Center
Title: Image Subtraction and Hough-Transform Backmasking for Transient Variability Searches
Abstract:: Massive datasets from time-domain surveys afford the discovery of transient variability like microlensing, stellar explosions, and the unknown. However, probing to ever fainter magnitudes will challenge our ability to separate transient variability from moving objects like asteroids, comets, and nearby stars. Meanwhile, the grid scenario is seeking to harness the more rapid growth of network speed over computing power and storage capacity for scientific application, but only specific types of problems will be amenable to grid-based solutions because of the storage limits of distributed computing resources. The problem of filtering moving objects from time-domain survey data in order to find transient variability may be well suited for the grid as follows. For a time-series of N spatially registered images, N-1 subtractions are summed in quadrature and converted into a binary mask of significantly variable pixels. This is passed through a Hough transform yielding an image in which the number of lines crossing any one pixel is related to the probability of the null hypothesis that the variable pixels are uniformly and randomly scattered. The reverse transform, or backmask, is a statistically rigorous definition of all cospatial pixels that are not traversed by moving objects. I'll show an example of this procedure using images from the Subaru telescope.
Name: Byun, Yong-Ik
Institution: Yonsei University
Title: YSTAR-KAO Detection of Optical Transients
Abstract:: A large number of faint optical transient (OT) have been detected from the database of time series observations made with the 0.5-m YSTAR wide-field telescope stationed in South Africa. Some of these flashes have obvious causes such as satellites and image defects etc., but there exist OTs that are not easily explained. They have stellar look, fainter than 13 R-mag, appearing only once in successive visits of the same night, but not cosmic ray incidents nor asteroids. There is a high chance that these OTs are astronomical events of previously unknown nature, because (1) they have similar look of stars and the event rate is much higher than image defects caused by cosmic rays, i.e. short streaks and single-pixel events, (2) there are occasions of recurrence, and finally (3) some of them (~20 %) do have faint counterparts in DSS plates. LSST should expect to find an enormous number of similar faint OT events. YSTAR-KAO is a joint project of Yonsei University Observatory and Korea Astronomy Observatory, and is partially supported by Korea Research Foundation through the grant KRF-2002-070-C00045.
Name: Niv Drory
Institution: Astronomy Department, University of Texas at Austin
Title: VIRUS: A massively-replicated IFU spectrograph for the HET
Abstract:: The VIsible Replicable Ultracheap Spectrograph (VIRUS) is a massively replicated integral field spectrograph with a wide field of view. VIRUS consists of 132 simple spectrographs fed by 32,000 fiber IFU elements, capable of working from 340 - 850 nm. The huge replication of a simple modular design represents a radical departure from the usual astronomical instrument, which is essentially a prototype where the cost is dominated by engineering. Development work on the prototype VIRUS, which will be mounted on the Hobby-Eberly Telescope (HET), will be described, along with science applications of the instrument. The primary science driver will be to measure the dark energy equation of state via a huge galaxy redshift survey at redshift 1.8 < z < 3.8.
Name: Mark Huber
Institution: LLNL
Title: The Faint Sky Variability Survey -- get your copy today
Abstract:: The Faint Sky Variability Survey (FSVS) has cataloged 23 square degrees in BVI filters from 16--25 mag to investigate variability in faint sources at moderate to high Galactic latitudes. Multi-epoch (10--30) observations were made in V spanning minutes to years to sample a variety of dynamical phenomena. An analysis of the color and variability characteristics for point sources from the FSVS will be presented. Overall, 5--8% of the point sources indicate some form of variability over the range of V = 17.5--22.0 mag. For V brighter than 19 mag, the variable population is dominated by low amplitude (< 0.05 mag) and blue (B-V < 0.35) sources, possibly representing the population of gamma-Doradus stars. The dominant population of variable sources are bluer than B-V of 0.65 and have Main Sequence colors, possibly reflecting larger populations of RR Lyr, SX Phe, gamma-Doradus, and W UMa variables than expected. The processed images and ASCII photometry files will soon be online at the NOAO/DPP site: http://www.noao.edu/dpp/
Name: J. Allyn Smith
Institution: LANL
Title: SDSS Data Quality Assessment & Implications for LSST
Abstract:: We summarize the automated routines used to perform SDSS data quality assessment. Building upon this, we compare the SDSS to the LSST data acquisition and discuss areas of compatibility and concern for LSST data.
Name: Johan Masreliez
Institution: EST Foundation
Title: The Scale Expanding Cosmos Theory
Abstract:: This talk presents a possible alternative to the Standard Cosmological Model (SCM) based on the big bang. I will show that the new model agrees better observations and resolves several fundamental problems with the SCM. On the conceptual plane it provides an explanation to the progression of time, which also give the missing link between General Relativity and Quantum Mechanics. Regarding observational discrepancies it provides explanations in several areas, for example:
The origin of Dark Energy
Large scale structure
The Pioneer anomaly
Number count discrepancies
Angular size discrepancies
The surface brightness test
The supernovae Ia observations
Irregularities in the planetary and lunar ephemerides
Galaxy formation
Name: Kirk Borne
Institution: George Mason University
Title: Data Mining in Distributed Databases for Interacting Galaxies
Abstract:: We will present results from an exploratory data mining project to identify classification features of special classes of interacting galaxies (for example, infrared-luminous galaxies) within distributed astronomical databases. Using a variety of data mining techniques, interaction-specific features are learned, to distinguish this class of galaxies from a control sample of normal galaxies. Subsequently, the corresponding rule-based feature model of that class of galaxies is then applied to the large multi-wavelength astronomical databases that are now becoming available. This distributed data mining activity is a prototype science use case for the Virtual Observatory. We specifically apply multi-archive multi-wavelength data to the problem. We will present examples of both successful and unsuccessful data mining attempts.
Name: Lutz Haberzettl
Institution: Astronomical Institute Ruhr-University Bochum
Title: Reduction pipeline for wide field imaging data
Abstract:: The advent of multichip CCD cameras such as WFI@MPG/ESO 2.2 m telescope, OmegaCAM@VST, MOSAICII@CTIO Blanco 4 m telescope or SUPRIME@SUBARU 8 m telescope imposes new, high demands on data reduction. Preprocessing steps such as debiasing or flatfielding etc. can be done independently on a chip-by-chip basis, allowing for efficient parallel processing on multiprocessor computer clusters. These steps can be tackled using the same well-known algorithms as for single chip cameras. However, accurate astrometric and photometric calibration of multichip CCD data requires techniques going far beyond those for stacking observations of single chip cameras. Furthermore, different sensitivities of the CCDs and gaps between them lead to a very inhomogenous noise respectively exposures time in the coaddition. In order to deal with those problems in the coadded data a weighting scheme is essential.
In our poster we present an automized reduction pipeline for the reduction of multichip CCD data. Our approach for the astrometric and photometric calibration is outlined in more detail. We present a statistically optimized weighting strategy helping to conserve the maximum amount of information during the image coaddition process. Wherever possible our pipeline is based on existing software modules, such as EIS drizzle, the Terapix software suit, FLIPS, Eclipse, Imcat and LDAC which are wrapped together with a number of bash shell scripts. The essentail pipeline modules are all written in ANSI-C and hence it is easily portable to new architectures. It has already successfully been used under SUN Solaris, IBM AIX, Linux (Linux/Alpha, Linux/x86, Linux/AMD64), DEC Alpha, HP-UX and IRIX64.
Name: Mario Juric
Institution: Princeton University
Title: Mapping the structure of the Galaxy using SDSS stellar sample
Abstract:: We will discuss the spatial distribution of ~180 million stars detected by the Sloan Digital Sky Survey. This sample provides a detailed view of the Galaxy, including halo substructure, and allows accurate determinations of thin and thick disk parameters. We will also show how the upcoming surveys, such as LSST, will further improve such mapping of the Milky Way.
Name: Carl Pennypacker
Institution: UC Berkeley
Title: Reaching Many Schools and Science Centers with LSST Data
Abstract:: The LSST will be a tremendous resource for students in classrooms and after-school programs, and also visitors to science centers and other venues. We have found that non-professional learners are eager to undertake aspects of modern astronomy -- and LSST (and the NVO) will provide a rich pipeline of on-going discovery for such explorers. Over the past decade, Hands-On Universe has developed much of the system necessary to distribute and meaningfully engage a large community in LSST and NVO data. In our pilot stages, have trained over 700 teachers in the United States, and a comparable number in other nations in how to use image processing in astronomy. A recent development is the development of a "Collaboratory system," in partnership with Northwestern University, for a shared data/project based learning system. We intend to start more serious expansion now, as our curriculum and software will soon be published and made into a more widely-available product, and can begin to provide good distribution of data to the larger community.
Name: Peregrine M. McGehee
Institution: LANL / NMSU
Title: Tracking stellar magnetic processes in the time domain
Abstract:: The multi-band variability seen in low mass stars and by pre main sequence objects in the T Tauri phase is a manifestation of stellar magnetic activity. We use multi-epoch Sloan Digital Sky Survey imaging acquired in the Orion equatorial region and of the dispersed T Tauri population south of Taurus-Auriga to demonstrate the detection and characterization of magnetically active stars. The optical variability in these objects is a result of both periodic (such as rotational modulation) and aperiodic (flaring and magnetospheric accretion) physical processes. Science goals include characterization of activity levels in M dwarfs and brown dwarfs, study of accretion behavior across the substellar boundary in star formation regions, and detection of high galactic latitude T Tauris. The cadences and sampling required for these studies are discussed
Name: Sean Raymond
Institution: University of Washington
Title: A strategy for finding Near Earth Objects with SDSS
Abstract:: We present a detailed observational strategy for finding Near Earth Objects (NEOs) with the Sloan Digital Sky Survey (SDSS) telescope. We investigate strategies in normal, unbinned mode as well as binning the CCDs 2x2 or 3x3, which affects the sky coverage rate and the limiting apparent magnitude. We present results from 1 month, 3 year and 10 year simulations of such surveys. For each cadence and binning mode, we evaluate the possibility of achieving the Spaceguard goal of detecting 90% of 1 km NEOs (absolute magnitude H <= 18 for an albedo of 0.1). We find that an unbinned survey is most effective at detecting H <= 20 NEOs in our sample. However, a 3x3 binned survey reaches the Spaceguard Goal after only seven years of operation. An SDSS NEO survey could make a significant contribution to the detection and photometric characterization of the NEO population.
Name: Dominik Rosenbaum
Institution: Astronomical Institute of the Ruhr-University Bochum/ Germany
Title: The environments of LSB Galaxies from SDSS
Abstract:: The formation and evolution scenarios that led to the existence of Low Surface Brightness (LSB) galaxies are not well understood yet. Although LSBs have HI components with low surface densities (systematically below the Kennicutt criterion) they can be regarded as gas-rich in general. Hence, the key in understanding LSBs lies in the answer to the question what prevented them from sufficient star formation. Wide-field surveys containing spectroscopic redshifts are well suited for studies on the environments of galaxies. Using the public data releases of the Sloan Digital Sky Survey (SDSS), we investigated the galaxy density around LSBs in comparison to High Surface Brightness (HSB) galaxies. We performed a number counting analysis in three dimensions within spheres of several radii on a sample of HSBs and LSBs with well measured SDSS redshifts. On scales between 2 and 5 Mpc our results show significantly lower galaxy densities in the vicinity of LSBs compared to HSBs. At larger scales than 5 Mpc, LSBs and HSBs share the same clustering properties but on scales below 2 Mpc the galaxy densities in the neighborhood of LSBs lie systematically (but only with a slight significance) below that of HSBs. In the pie slice diagrams LSBs favor the outer rims of the filaments of the Large Scale Structure and some LSBs are even found in voids. These results give strong evidence to a formation and evolution scenario where LSBs were formed in low density environments. Hence, the lack of tidal interactions with companions must have caused the absence of an effective trigger for a sufficient star formation.
Name: Thomas Quinn
Institution: University of Washington
Title: Parallel Visualization/Analysis of Large Data Sets
Abstract:: Analyzing and visualizing data from large surveys has traditionally been done with serial programs, even as we recognize that the data is acquired and archived in a parallel fashion. Even reduced catalogs are now sufficiently large that they cannot be loaded into the memory of a single machine. In addition, to perform a new analysis function, the researcher must add code to the program, recompile, and reload the data. An analysis tool with an embedded programming language could dramatically increase the rate of producing scientific results. Similar challenges arise in the realm of astrophysical simulations, and we have developed Salsa, a parallel, interactive analysis tool for point-based data such as N-body simulations or object catalogs. The user runs a graphical client application on their desktop, which communicates with the server running on a parallel machine. The client controls the display of the data set, which is rendered by the server. The client can enter high level code, like Python or C++, that is executed on the server, implementing new functionality and providing programmatic control.
Name: Luc Simard
Institution: Herzberg Institute of Astrophysics, National Research Council of Canada
Title: A LSST precursor experiment: Data Management for the CFHT Legacy Survey
Abstract:: The CFHT Legacy Survey is in many ways a precursor experiment for the LSST. The data are acquired on behalf of a diverse user community in pursuit of a wide range of science goals such as high-z supernovae, galactic structure, moving objects, weak-lensing, and galaxy evolution. The sheer volume of data produced by MegaPrime/CFHTLS (close to 20 TB/yr) currently represents interesting challenges in acquisition, processing, transfer and archiving as these operations must all be performed as quickly as possible in order to enable synoptic surveys that are coupled to follow-up studies on other facilities. These challenges have led to the design and construction of a new data storage and distribution system at the Canadian Astronomy Data Centre that allows users to query and retrieve images and catalogs via web forms and data proxies. This poster is a description of this system with emphasis on LSST-relevant aspects.
Name: John Bochanski & Mark Claire
Institution: University of Washington
Title: Observing Extrasolar Planetary Transits with LSST
Abstract:: We present simulations and initial "back-of-the-envelope" calculations on the feasibility of observing extra-solar planetary transits with LSST. We characterize the ability of LSST to significantly detect a planetary transit of a given period and duration. The multi-day cadence of the survey allows LSST to be most sensitive to large planets with short orbital periods. The statistics of such planetary systems will be well investigated by the time LSST is operational, but it has a distinct advantage due to its brightness and distance limits. This will enable LSST to extend the extrasolar planet census to larger distances within the Galaxy.
Name: Vera Ellinger Margoniner
Institution: University of California
Title: Deep Lens Survey
Abstract:: The Deep Lens Survey (DLS) is a deep BVRz' imaging survey of seven 2X2 degree fields, with all data to be made public. The primary scientific driver is weak gravitational lensing, but the survey is also designed to enable a wide array of other astrophysical investigations. A unique feature of this survey is the search for transient phenomena. We subtract multiple exposures of a field, detect differences, classify, and release transients on the Web within about an hour of observation. Here we summarize the scientific goals of the DLS, field and filter selection, observing techniques and current status, data reduction, data products and release, and transient detections. Finally, we discuss some lessons which might apply to future large surveys such as LSST.

Title: Shear-selected clusters from the Deep Lens Survey
Abstract:: We present preliminary results from the Deep Lens Survey (DLS) shear-selected cluster search. The DLS is a deep BVRz' survey of 20 square degrees, which will produce a sample of ~ 100 shear-selected clusters. Our follow-up focuses on the X-ray and optical properties of these clusters, which may reveal biases in existing X-ray and optically selected samples. Chandra and/or XMM X-ray observations are soon to be complete for the first fifteen clusters from the sample, and we present mass and X-ray maps of some individual clusters.