Problem Set #2

Problem 1: Mass-Luminosity/Mass-Radius Relations

Data File: mlr.dat (ASCII format with 2-line header)

You have a number of options for doing the least-squares line fits for this problem:

  1. Use lfit, which is available on the Linux workstations in the department. See the lfit help page for details. I know it works.

    To run lfit, type 

       lfit
    If that fails, try the full path: 
       /usr/local/pkg/bin/lfit
    If both fail, come see me.

  2. Use your own programs, but only if already have them (but do NOT write anything special for this class!!!)

  3. Use something else, examples include a line fitter on a Windows PC or Mac (e.g., Excel, SigmaPlot, Kaleidograph, etc), or other programs you know about. In any case, tell me what you used. I do not recommend the line fitting algorithms in sm, they're somewhat buggy.
Please note that for this problem I only expect an unweighted least-squares fit. In reality a more sophisticated fit using the uncertainties on both variables (L and M) would be required, but that would only add unnecessary complication to the problem.

Problem 3: Spectral Classification

The spectra of your 5 stars for Problem 3 are given below in the form of a gzip-compressed tar file. Each of you are assigned 3 stellar spectra that are unique to you, and two "Mystery Spectra" that are to be classified by everyone.

To unpack your tar file type:

 
   tar xvzf xx.tgz
Where "xx" are your initials. Your spectra will have names like 
    xx1.spc  xx2.spc   xx3.spc   mstar1.spc  mstar2.spc
When you write up your homework solutions, please refer to them as "Star #1" through "Star #3", and "Mystery Star #1" and "Mystery Star #2".

Assigned Spectra
Kelly Denney
Subo Dong
Susan Dorsher
Shawn Poindexter
Jose Luis Prieto
The spectra are in 3-column ASCII format as follows: 
   pixel   lambda   relflux
where 
   pixel   = original pixel number (runs 1-2899 or 2898)
   lambda  = wavelength in Angstroms
   relflux = flux in erg/sec/cm^2/Angstrom relative to the flux at 5500A
For this exercise, you want to plot flux vs. lambda, and then use that plot to identify the lines for classification.

Note that these are observed spectra, in these sense that they are given to you as they would appear at the telescope after basic flux calibration, but without any corrections made for atmospheric or interstellar dust extinction along the line of sight. The radial velocities are negligible, so you should expect no Doppler shifts when identifying lines (i.e., don't fool yourself into thinking you can shift the spectra in wavelength to match up lines - these are all "zero velocity" - if what you think is Hydrogen lines don't match the wavelenghts, they aren't Hydrogen, no tricks here).

The MKK Spectral Atlas plates I showed you in class are available in the reading room, and should stay in the reading room so others may use them. You can also look at this website for digitally scanned versions of these plates, but the quality is not always as good as the originals.

Updated: 2003 September 23 [rwp]