Spectral Classification Activity

Often times, astronomers are confronted with information that is not consistent with prior assumptions thus requiring reevaluations of the various systems that have been put in place. As we learn more and more about the Universe, much of our understanding has been codified and categorized, but there is no reason why current schemes must be the only ones. For this activity, you will work with actual astronomical data in an attempt to reconsider one of the many schemes currently used in modern astronomy. Specifically, you will be examining the actual spectra of various stars and using your observations to develop a new stellar classification scheme.

Data

The spectra you will analyze were taken as part of the Sloan Digital Sky Survey; a large multi-wavelength and spectroscopic survey of the sky. To learn more about the survey, visit their website. Every object that has been examined has a website where its data can be viewed. Listed below are the pages for several stars. On each page, you will notice a lot of information, a picture, and a graph. Click on the graph to view a larger version of it. This is the star's spectrum. Astronomers like to have their data in a mathematical form so numbers can be easily compared. This is why technical spectra are viewed as plots rather than the actual "rainbow" one might expect. Please examine each of the spectra below and consider their different properties.

  • Star A
  • Star B
  • Star C
  • Star D
  • Star E
  • Star F
  • Star G
  • Star H
  • Star I
  • Star J
  • Star K
  • Star L
  • Star M
  • Star N

    Things to notice

  • Spectra have two primary types of features: absorption lines and emission lines. These are visible in the data as the various dips and spikes respectively that deviate from the general curve.
  • The line that defines the overall shape of the spectrum is known as the continuum. The shape of the continuum varies from star to star as a result of blackbody radiation, among other things. Perhaps consider looking for any trends that exist between the nature of the continuum and prominence of spectral lines.
  • Not all spectral lines are equal! Some are deep, some are broad, some emission lines are taller than others, not all stars have emission lines, etc. Consider the specific natures of the various lines in these spectra.
  • Not all spectra have the same number of lines. This is significant!
  • There is a lot of information on the plots. The y-axis provides a measure of intensity. Higher values indicate more light. The x-axis is in units of wavelength as measured in angstroms. Indicated on the plot are the wavelengths of various elements. These are illustrated with dashed vertical lines in various colors at the top of which is a the name of the element. Element names with roman numerals refer to ionized versions of a particular element. For example, features marked as CaII are spectral features that come from calcium atoms that have lost one electron. Hydrogen lines are marked with Greek letter subscripts. These are just there as a shorthand to indicate which specific electron transition is responsible for the line.

    What to do

    After having considered the above points and conducted a thorough evaluation of the data, construct a classification scheme. Do this by placing individual stars in groups based on criteria you feel are important. These can be based on anything you observe in the spectra: a certain line strength, number of lines, slope of the continuum, etc. Be creative, but also be consistent! There may be as many or few groups as you want, but keep in mind that your scheme should be usable by others. It is generally a good idea (though not necessary) to have at least two stars per group so that the traits that define the group are easier to identify. Thus, you'll want to avoid schemes that put too many stars in one group or put each star in its own group.

    After you have developed your scheme, please type a one page paper describing the criteria you chose, why you chose the features you did, and how you went about placing individual stars in their respective categories. Please be sure to provide some sort of explanation of the science behind your choices and why it supports your scheme as one that is suitable for stars. On an additional page, please provide a reference sheet for other astronomers so they know how to use your scheme. This sheet should have information such as the name of your classification, the names of groups, and the basic characteristics that define each group. On this same sheet, please list all the stars from your data set that fall under each category.

    Grading

    This activity will be graded on a scale of 1-7. Your scheme will be assessed based on how self-consistent, logical, and practical it is. You will not be judged solely on the scientific merit of your scheme, since this is more an exercise in scientific thinking than knowledge. However, you should still be conscious of the backing theory that you are employing.

    The author greatfully acknowledges an exercise developed by the people at SDSS as his starting point for this activity. The original may be viewed here.

    Home