Next: A Multi-Mission Archive
Previous: Re-Classification of HST Targets
Table of Contents --- Search --- PS reprint
N. A. Oliversen, P. J. Lawton
Hughes STX, Code 631, Goddard Space Flight Center,
Greenbelt, MD 20771
M. E. VanSteenberg
Code 631, Goddard Space Flight Center, Greenbelt, MD 20771
The NDADS near-line archive contains about 1.5 terabytes of space science data from 18 astrophysics and 8 space physics NASA missions. Until recently, in order to request astrophysics data from NDADS, a user would first have to obtain the appropriate catalog (if it existed) from NDADS or the National Space Science Data Center and/or search a database maintained by one of the projects and then make a separate NDADS data request. This multistep process was clearly inefficient and time consuming. We have developed WISARD to improve the accessibility of the archival data, to facilitate multi-wavelength, multi-mission astrophysics research using NDADS and related data archives. WISARD combines the catalog-searching and data-request steps into one activity.
Figure 1: WISARD Processing Flow. All activities in the shaded box retain context from one to the next activity. Thus, WISARD ``remembers'' the results of previous actions within each ``session''. The dotted box and lines represent the next planned functional enhancement to WISARD. These include expanding the Simbad service to treat Simbad as another data archive as well as accessing the bibliographical references in the Astrophysics Data System (ADS) abstract and article service.
Figure 1: GIF 80 Kb
The initial WISARD interface is designed for astrophysics data archives, although the overall architecture could be easily adapted to other disciplines. WISARD currently contains the following databases: Infrared Astronomical Satellite (IRAS) Pointed Observation log; International Ultraviolet Explorer (IUE) merged log of observations (spectral data); IUE Fine Error Sensor (FES) log; Hopkins Ultraviolet Telescope (HUT) and Wisconsin Ultraviolet Photopolarimeter Experiment (WUPPE) logs from Astro-1; public archive logs from the Roentgen Satellite (ROSAT), Advanced Satellite for Astrophysics and Cosmology (ASCA), and the Extreme Ultraviolet Explorer (EUVE) (spectral data). These catalogs represent a subset of the astrophysics data available on NDADS and other archives. Additional catalogs and archive access will be added to WISARD as time permits.
WISARD supports queries on fields which are in-common between the various catalogs (i.e., target coordinates and date of observation); galactic or equatorial coordinate searches; choice of search epoch (1950, 2000); access to the Simbad name-resolver; catalog-specific field searches (e.g., instrument, object class, etc.); search criteria refinements; query unions and intersections on multiple catalogs; default and user-specified display of the query results; and subsequent requests of selected data to the appropriate archives.
The WISARD processing flow is shown in Figure 1. The database(s) to be searched, the type of search (union or intersection) and the type of form (default or custom), are selected from the WISARD main page. A process id or context id, which is tracked through WISARD, is assigned every time a new WISARD session is initiated. The user is then guided through a series of screens based upon his/her selections on the WISARD main page. The database query criteria are entered on the next screen and a link to the Simbad target name resolver is also available. The selected search criteria are next displayed and the database search may be started or the query criteria modified or supplemented. A summary of the query results is next displayed and a default or custom ``print''/display format may be chosen. Large queries are mailed back to the requester. The specific data-sets to be ordered from the archives (NDADS or EUVE) can next be selected. The WWW (NDADS or EUVE) order form is finally displayed (with the specific data-set ordering information pre-set).
WISARD utilizes a master database (table-of-tables) which contains information about each of the databases accessed by WISARD. This master database contains such database-specific information as the project short and full names, data archive ordering information, coordinate accuracy estimates (used for intersection searches), print information, and various documentation URL's pointed to by WISARD. Use of this master database allows new databases/data sources to be added to WISARD relatively easily without large amounts of recoding.
WISARD utilizes the client-server architecture of the World Wide Web (WWW) making use of the standard WWW ``forms'' for passing of information back and forth between the requester and WISARD. Communications with the WWW are accomplished through a combination of Bourne shell, C shell, C, and Interactive Data Language (IDL) programs. The shells are used to extract the WWW environment variables and data request parameters, control some of the WISARD process flow, and start up the IDL sessions.
IDL is used for most of the WISARD processing, including: creation on-the-fly of the various WISARD WWW display pages and forms; searches of the databases; and special processing such as intersection searches and coordinate transformations. The various mission catalogs and the WISARD master database are stored in IDL databases. The IDL database software is very simple and flexible. Individual catalog fields (such as target coordinates and/or dates) can be sorted or indexed to increase the speed of the database searches. Use of IDL databases simplifies the transition between the database searches and the general WISARD IDL processing. IDL was used for the WISARD prototype because the IUE databases were already in IDL and the IUE/WWW merged log SEARCH software had already been developed which could be generalized relatively easily.
WISARD also utilizes the Simbad name resolver software. The Simbad (C client) programs open a remote session on the Simbad/CDS computer to query the Simbad database. Output from the Simbad programs are saved to IDL variables for further processing.
An important part of this effort is the preparation of catalogs of data at the archives. These catalogs come from a variety of sources. The IUE merged observing and FES logs are supplied by the IUE Data Analysis Center (IUEDAC). The ROSAT database was generated by scanning and extracting information (i.e., target coordinates, observation dates, etc.) from the ROSAT public contents files which are delivered to NDADS along with the individual data files. The HUT, WUPPE and IRAS logs were derived from project supplied observation logs. The EUVE and ASCA databases were derived from public archive log information available at the EUVE and ASCA Guest Observer Facilities.
We plan to add a number of enhancements to WISARD as time permits. These include expanding the number of NDADS astrophysics databases; linking to the NED name resolver; expanding the SIMBAD service to return other information on the selected target(s); and adding a link to the ADS abstract service and additional archives. Modifications to WISARD will continue to be made in order to make it easier to maintain the code, documentation and databases.
We are grateful to S. Digel, V. Kargatis, R. Kilgore, J. Lauroesch, J.\ Offenberg, A. Richmond, L. Taylor, and S. Voels for various assistance related to WISARD. The Simbad name resolver software was kindly provided by F. Oschenbein, M. Wenger and D. Egret of the Centre de Donnees Astronomiques de Strasbourg (CDS), France. Some of the underlying IDL software in WISARD is based on software originally developed by D. Lindler for the Goddard High Resolution Spectrograph, the IUE, and the Ultraviolet Imaging Telescope science teams. The IDL database software used by WISARD is available through the IUEDAC homepage.