ETOOLS: Tools for Photon Event Data

M. Abbott, T. Kilsdonk, E. Olson

Center for EUV Astrophysics, University of California, 2150 Kittredge Street, Berkeley, CA 94720

C. Christian

Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 20771

M. Conroy, R. Brissenden, D. Van Stone, J. Herrero

Smithsonian Astrophysical Observatory, Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138

Abstract:

1. What is ETOOLS?

ETOOLS is a NASA ADP funded software development project working to produce a package of general purpose tools for reducing event data. The project is a collaboration between the Center for EUV Astrophysics and the Smithsonian Astrophysical Observatory.

In the past, individual NASA high-energy astrophysics missions have developed their own analysis packages for their observers, such as the PROS package for X-ray data and the EUV package for analysis of EUVE data. But such packages often have overlapping functionality or contain tools that are generally useful but not compatible with each mission's data format.

The ETOOLS project represents an attempt to create a single package of tools for the analysis of event data that will be useful with many missions. Our approach to this problem is to:

• Develop a data model identifying the common properties of event data sets,

• Identify the operations needed to manipulate such data,

• Construct a kernel-based library providing such operations,

• Build applications that perform the basic operations and simple analysis tasks,

• Let each project write its own mission-specific applications using the ETOOLS libraries.

2. Event Data

Event data refers to the very rich data sets produced by photon counting detectors such as those commonly used in high-energy telescopes. Some recent examples of missions that have produced or will produce event data are Einstein, ROSAT, EUVE, ASCA, XTE, and AXAF.

In event data, each event (photon) on the detector is tagged with its arrival time, detector position, and other information. These tags (called attributes) are retained through the later processing of the events so that a scientist analyzing the data can use filtering techniques to examine subsections of the complete data set or form secondary data products, such as images or light curves, by binning the data in various manners.

Because it is easily filtered, an event data set lends itself to a very detailed and flexible analysis in which examining small numbers of (or even individual) events can be useful. The types of software needed to carry out such analyses can be quite different than those used with optical image data.

3. What Will ETOOLS Produce?

The ETOOLS project will produce two major software components: a set of applications for performing general purpose manipulations of event data and libraries providing a consistent interface for the applications.

3.1. ETOOLS Applications

The project will develop those applications that are useful with any event data set. We will leave the development of mission-specific software (calibration tasks, specialized data reduction tools, etc.) to individual missions. Categories of applications we will produce include:

Object management

--- We will support storage of multiple data objects making up an event data set in a single file or in multiple files. In either case, the user will interact with a single directory object. Applications will allow the user to create, rename, and delete objects as well as list those present in an event data file.

Table tools

--- Applications will allow users to copy, combine, list, sort, filter, bin, and perform calculations on tabular data. An interactive table editing tool will also be provided. These applications significantly overlap the functionality currently provided by the TABLES package from STScI. We are exploring ways to reduce this overlap

Filter tools

--- The creation and manipulation of filters will be supported by applications that can generate 1-d filters from tabular data and 2-d filters from images. An interactive filter editing tool will also be provided.

Image Display

--- A standard IRAF image display interface will be used for ETOOLS image applications.

Headers

--- A full set of applications will exist for creating, deleting, and editing event-data header objects.

3.2. ETOOLS Libraries

The ETOOLS libraries are as important as the applications. The libraries can be used to extend the functionality of ETOOLS by adding additional applications. By using the libraries, application developers get:

File Formats

--- A key feature of the ETOOLS libraries makes use of library kernels to support any number of physical file formats within the ETOOLS event-data model. Thus, a new format can be used with existing applications by merely adding another kernel to the libraries and rebuilding the applications. One or more kernels will be developed as part of the ETOOLS project.

Compatibility

--- Application will work with existing ETOOLS applications.

4. The Event-Data Model

The ETOOLS libraries, and thus the applications layered on them, are designed to support a particular model of event data. We identify four primary objects that users manipulate while working with event data:

Tables

--- Tables are the fundamental data objects for event data. The list of events is represented as a table, as are the auxiliary (housekeeping) data that accompany the events and describe the state of the instrument as a function of time. Some reduced products, such as spectra and light curves, are also tabular.

Headers

--- Headers store global information about a data set and are similar to FITS keywords.

Filters

--- Filtering the event list is the most common operation performed during analysis of event data. The filters are generally one- or two-dimensional. Users construct filters by direct specification or from auxiliary data or reduced-data products.

Images

--- The most common reduced-data product from event data is an image formed by binning the event list according to two or more of its attributes.

Since event data sets usually consist of many objects of various types, we made an object management facility an integral part of the data model. We adopted a directory-like facility in which the user manipulates the objects by name.

5. Software Environment Support

Just as the ETOOLS applications and libraries are intended to be generic and extendable, it is important that the environment in which they run be as flexible as possible.

IRAF package

--- Applications will be released as IRAF tasks in an IRAF layered package. Graphical interfaces will use the IRAF Widgetserver (Tody 1995).

Host programs

--- Applications will also be executable directly from the OS command shell. We currently plan to support UNIX hosts; other operating systems are under consideration.

Application modules

--- Applications will be developed modularly so their functionality is available through callable subroutines. This structure permits them to be executed from other programs (for example, from within a Tcl interpreter or IDL procedure¹).

6. Further Information

Completion of the ETOOLS project is scheduled for late 1996. We may put out beta versions of the software prior to the final release. All ETOOLS products will be available from ftp://ftp.cea.berkeley.edu and ftp://sao-ftp.harvard.edu and perhaps other sites.

During the project we will occasionally post information on our WWW site at http://www.cea.berkeley.edu/isd/etools describing the ETOOLS software and libraries during their design and development.

Look for us at the 1996 ADASS meeting; we hope to have operational demos of some of the ETOOLS applications. We encourage comments, criticism, and suggestions by all interested parties.

Acknowledgments:

This work is supported by NASA grant NAS5-32698. The Center for EUV Astrophysics is a division of UC Berkeley's Space Science Laboratory.

References:

¹IDL is a trademark of Research Systems Inc., Boulder, CO.