Analyzing Online Data on Variable Stars: An Authentic Research Experience for Teachers

Pebble Richwine
Timothy F. Slater
Conceptual Astronomy and Physics Education Research (CAPER) Team, University of Arizona, Steward Observatory

James Bedient (bedient AT - change AT with @)
American Association of Variable Star Observers
University of Hawaii

Analyzing Online Data on Variable Stars: An Authentic Research Experience for Teachers

Abstract: In recent years amateur astronomers have led the way in pursuing research by data-mining Internet resources. Online datasets, such as Stardial, already contain more data than will be inspected visually by the current cadre of professional astronomers. This provides an important and valuable opportunity for amateur astronomers. Astronomy educators are now adapting the some of techniques amateurs have honed for student use. Both the NRC National Science Education Standards and the AAAS Project 2061 Benchmarks clearly emphasize that students should learn science by DOING science. In response, science education is slowly but steadily being transformed into a process-driven, inquiry-based area of study. Astronomy is a natural choice for this type of effort because nearly all of recent primary data and large amounts of secondary data (catalogues and technical publications) are in online electronic form. This data is often easily accessible in the classroom using typical computer equipment. Research results from on-line data are often as robust as those from "real" telescopes—and moreover are available almost instantaneously to students. This paper describes a project directed at enhancing secondary school teachers astronomy research skills using online resources with the goal of improving science teaching and learning.

The NRC National Science Education Standards and the AAAS Project 2061 Benchmarks clearly emphasize that students should learn science by actually DOING science. In response, science education is slowly but steadily being transformed from a classroom based on memorization of facts and formulae into a process-driven, inquiry-based classroom that mimics scientific inquiry. As one avenue, inquiry-oriented science fair-like research projects provide engaging opportunities for students to become personally and directly involved in scientific investigations. Rather than model volcanoes and dish soap comparisons, reformers in science education are calling for authentic investigations that can actually result in publishable papers. It appears that Internet-based astronomy might be well positioned to fulfill such a need. It is generally accepted that the field of astronomy easily captures many students' attention. Therefore, it is only natural that many students already want to embark on astronomy-related science fair projects. Unfortunately a quick web-search demonstrates only limited guidance exists for students on how to participate meaningfully in the enterprise of astronomy.

In a very short time, the amount of astronomical data freely available over the Internet has become significant. Online datasets, such as Stardial, already contain more data than will be inspected visually by the current cadre of professional astronomers. More and more astronomical catalogs have been placed online and linked, search engines are becoming more and more sophisticated. Research results from on-line data have become just as rich as those from "real" telescopes-and in contrast to applying and waiting for data, internet data is available almost instantaneously.

Much larger than Stardial, comprehensive ventures such as the National Virtual Observatory (NVO) are actively preparing for a data avalanche and the astronomy education community is beginning to gear up for how to appropriately bring the accompanying resources to students. As a first stage we are exploring strategies to develop and evaluate various approaches to support and facilitate authentic research experiences for students and teachers, bringing them in to the astronomical enterprise in the meaningful way afforded by a wealth of Internet-based astronomy data.

Perhaps surprisingly, amateur astronomers have pioneered data-mining in astronomical databases. This may be because amateur astronomers have adapted what is available to them, while professional astronomers and graduate students "hunger" for the latest data, sometimes considered more glamourous. Amateur astronomers have used online primary data to discover new variable stars and evaluate the current behavior of known variable stars (e.g. Wils 2003), to prove the non-existence of suspected variable stars (Bedient 2003) and classify eclipsing binaries (Otero 2003). Furthermore, amateur astronomers have located solar system objects of interest such as Near-Earth Objects (NEOs) and Kuiper Belt Objects (KBOs) in online archival imagery, thus contributing to early refinement of orbits soon after discovery (e.g. Stoss et al. 2004). Change in brightness or position provides an opportunity for careful observation and analysis of data - the essence of scientific inquiry. Our goal is to translate some of the growing knowledge base of amateur astronomers on using data-mining techniques to the modern science classroom.

This effort had its roots in the Towards Other Planetary Systems (TOPS) teacher enhancement workshops (Meech 2000) which ran from 1993 through 2003. Initially funded by the NASA IDEAS program and later by a NSF Teacher Enhancement Program, TOPS provided training in basic astronomy to secondary school teachers and provided the opportunity to conduct research activities using small telescopes (Kadooka et al. 2003). The program was designed collaboratively by K. Meech, an astronomer at the University of Hawaii, T. Slater, an astronomy educator at the University of Arizona and J. Mattei, Executive Director of the American Association of Variable Star Observers. Author PR, a participant at TOPS 2002 and 2003, saw an opportunity to develop a similar teacher enhancement program that was more conducive to helping teachers work with students without access to telescopes, using a "virtual observatory" in the form of online archived images and data to investigate variable stars. Teachers can use the techniques learned in their own classrooms, and encourage and support science-fair-type projects in astronomy.

Variable stars have been chosen because they are stars that change brightness. As such, they provide a unique opportunity to engage students in systematically studying the time-varying behavior of astronomical objects. Research on variable stars is important because it provides information about stellar properties, such as mass, radius, luminosity, temperature, internal and external structure, composition, and evolution. This information can then be used to understand other stars and the structure of the galaxy. Professional astronomers have neither the available time nor the unlimited telescope access needed to gather data on the brightness changes of thousands of variable stars. Thus it is observers utilizing visual, photographic, photoelectric, and now CCD techniques, who are making a real and highly useful contribution to science by observing variable stars and submitting their observations to the American Association of Variable Star Observers (AAVSO) online International Database. Supplementing real-time observations of variable stars with data from image archives is a time-honored tradition in variable star astronomy, with photographic plate collections like those at Harvard College Observatory and Maria Mitchell Observatory still providing fertile ground for researchers. This scenario is even more attractive given the rapidly growing resources and datasets available over the Internet.

The brightness changes of variable stars can range widely from few millimagnitudes to as much as twenty magnitudes over intervals of a few seconds to years, depending on the type of variable star. There are a number of reasons why variable stars change their brightness. Pulsating variables, for example, swell and shrink due to internal forces, while an eclipsing binary will dips in brightness when one component is eclipsed by the other; and then brightens again when the occulting star moves out of the way. The different causes for light variation in variable stars provides the impetus for classifying the stars into different categories. Classification leads to developing explanations for the stars' behavior, and the interrelationship of pressure, temperature and luminosity can be explored. Over 30,000 variable stars are known and catalogued, and many thousands more are suspected to be variable. The large number of uncatalogued and unclassified variable stars presents an exciting opportunity for students to be involved meaningfully in the enterprise of scientific discovery.

Variable stars need to be systematically observed over long periods of time in order to determine the long-term behavior of a star, thus providing professional astronomers with data needed to analyze variable star behavior, and in scheduling observations of certain stars, correlating data from satellite and ground-based observations and to making computerized theoretical models of variable stars. In short, observation of variable stars provides an exciting opportunity to involve students authentically in the enterprise of astronomy as well as directly support students in developing meaningful understanding of the National Science Education Standards - Patterns of Change unifying concept in science.

In this project, we have developed a unique summer workshop approach for participants focused on conducting research on variable stars using existing Internet datasets. The objectives for participants will be to (a) learn to access Internet-databases; (b) identify variable stars and determine their light curves and (c) develop strategies, written procedural guidelines, and resources needed to mentor students in variable star-based science fair projects. Techniques used by amateurs in the resources cited above are used extensively.

In particular, participants will make heavy use of three primary resources: Stardial, an autonomous drift-scan camera on the web (McCullough and Thakkar 1997); the All-Sky Automated Survey, ASAS-3 (Pojmanski 2002); and the AAVSO International Database (Mattei 2003). The next section describes these resources in more detail.

Stardial provides nightly (weather permitting) images of a band of sky between declination 0 and -7, sliced into 15 minute of right ascension segments. The red-sensitive KAF-400 CCD in the camera coupled with the RG-1 filter makes Stardial an ideal detector of red long-period variables. Moreover, the one-time-per-night obsevation tempo makes it ideal for variable star observation, though it has also been used to detect novae, eclipsing binaries, bright asteroids, comets and geosynchronous earth satellites. Stardial provides primary data, that is, images that have not been processed or analyzed. Users need to learn basic photometry techniques in order to extract observations of individual objects. Students using Stardial must also apply knowledge of the celestial sphere to determine when an object might be visible in Stardial images.

The ASAS-3 system is an automated system of four CCD cameras located at Las Campanas Observatory, Chile. Two wide-field cameras, each equipped with a 200mm f/2.8 Minolta telephoto lens and 2Kx2K AP-10 CCD camera, cover 8.8 x 8.8 degrees of the sky in Vand I. A third camera is fitted with a 250mm, f/3.3 Cassegrain lens with a 3-element corrector. It has the same 2Kx2K AP-10 CCD camera and an I filter, with a useable field about 2 degrees in diameter. The fourth camera is an experimental very-wide-field camera with a 50mm lens. The data from these instruments is reduced in near-real time, and V- and I-band photometry data is available almost immediately on the ASAS website. The website features a search engine that allows a user to retrieve photometric data from the ASAS-3 database by specifying either Right Ascension and Declination or a variable star name.

The AAVSO International Database contains ~10.5 million variable star brightness estimates going back over ninety years. It is the largest and most comprehensive digital variable star database in the world. 700 amateur observers from all over the world contribute over 400,000 new variable star brightness measurements to the database every year. The AAVSO is engaged in a NASA-funded effort to validate and place on-line the entire database. This project is 74% complete at this writing. Once complete, the entire database will be available to anyone for instant retrieval online.

Extracting the data from these three databases and combining them can result in a picture showing the behavior of a star over many years. The light curve of GX Aquilae on this page is an example. Participants learn astronomical data reduction techniques in extracting data, and statistical techniques in analysis and display of data, for example, the phase diagram of Stardial data on V539 Aquilae on this page. They will be able to explore the behavior of known stars, and describe and classify uncatalogued variable stars, conducting original science research activity.

Software for data extraction and analysis is available on the web, a circumstance which has favored amateur astronomers and is beneficial to teachers and students. Software for viewing and analyzing FITS-format images is available (FITSView, DS9 and IRIS) as well as software for advanced statistical analysis of light curves (AVE, Zap, TS and WWZ). The availability of the data and the tools to work with it make virtual astronomy a viable classroom project.

Workshop participants make extensive use on online databases and image servers to identify and validate their results. Chief among them are products of the Centre de Données astronomiques de Strasbourg (CDS), the Set of Identifications, Measurements and Bibliography for Astronomical Data (SIMBAD), the VizieR catalogue access tool, and the Aladin image and data viewer. SIMBAD is a database of ~8.5 x 106 cross-referenced identifiers of astronomical objects, providing accurate positions of known objects. VizieR provides access to nearly 4000 catalogues of astronomical objects, searchable in many different ways. The most impotant to variable star hunters is searching by position; VizieR can query all its catalogues and provide information on objects located around a given RA and declination. Important catalogues within VizieR used by researchers of long period variables are the Infrared Astronomical Satellite catalogues and the 2 Micron All-Sky Survey (2MASS). Aladin is an interactive online sky atlas allowing the user to visualize digitized images of any part of the sky and superimpose entries from astronomical catalogs on the field. Aladin can access related data from SIMBAD, VizieR and other archives for all known objects in the field. Aladin is particularly useful for multi-spectral cross-identifications of astronomical sources and checking new data sets by comparison with standard catalogues covering the same region of sky.

The combination of all these tools and techniques should result in an authentic research experience for participants and their students. They must obtain their data, search for objects of interest in it, and then analyze the data to classify the star. They must correlate their data with other information to determine some of the physical parameters of the star. In its final form, our project is helping participants produce as complete a picture of a specific star as can be determined, written in suitable presentation style with the ultimate goal of being suitable for publication.

The goal is to provude an authentic experience of doing real science, not just repeating another lab demonstration by following step-by-step directions. We hope that we can provide inspiration and resources to present these techniques to classroom students as potential science fair projects. We hope amateurs involved in this type of work will be open to assisting teachers attempting virtual astronomy projects in the classroom.


We would like to thank Dr. Karen Meech and the Towards Other Planetary Systems Leadership Teacher Enhancement Program for the use of images from TOPS 2002. TOPS material is based upon work supported by the National Science Foundation under Grant No. ESI-9731083, PIs Meech and Slater. JB would also like to thank Mary Kadooka for her editorial comments.


Bedient, J. 2003, IBVS No. 5478
Kadooka, M., Meech, K.J., Bedient, J. 2003, JAAVSO 31, 39
Mattei, J.A., 2003. Status of the AAVSO International Database. Private Communication
Meech, K.J., 2000. 2000bioa.conf..679M
McCullough, P., Thakkar, U. 1997, PASP 109, 1264
Otero, S. 2003, IBVS No. 5480
Pojmanski, G. 2002, Acta Astronomica, 52, 397
Stoss, R. 2004, MPEC 2004-D15
Wils, P. 2003, IBVS No. 5401

12-Mar-2004 10:47 PM


Enter the forum and ask your questions to the author!

Back to the Meeting index