The MKK system is defined by carefully selected standard stars. All other stars were compared to this select group of bright stars. The MKK book does not give quantitative analysis, instead he describes in words what one should see when looking at a class of star. It is up to the reader to interpret the data and apply it to the star he or she is observing.

A Stellar Spectrum on Photographic Plate, as used by MKK

Note that this is a photographic negative. So, the bright vertical lines indicate where the spectrum is darker due to an absorption line.

MKK looked at many stars, and left behind much data. However the data is in the form of photographic plates, and is impossible to compare with the digital data produced by the Sloan Digital Sky Survey. To complicate matters further, The Sloan was designed to look at dim objects unlike the bright standards MKK used for his system. The Sloan didn't take any images of standard stars. This means that none of MKK's standards were captured by the Sloan, and their spectra must be found elsewhere.

The Dark Sky Observatory is one of the telescopes that has spectra for the standard stars available. A full list of spectra from the Telescope is available here

A Stellar Spectrum from the Dark Sky Observatory, plotted in a spreadsheet program

This spectrum is plotted with the smooth black body curve removed, to make it easier to see absorption lines. On this plot, 1 means the intensity of the star at that wavelength has no absorption, down to 0, where all of the light at that wavelength is absorbed.

The Steward Observatory is one of the telescopes that has spectra for the standard stars available. A full list of spectra from the Telescope is available here

A Stellar Spectrum from the Steward Observatory, plotted in a spreadsheet program

The Sloan Digital Sky Survey is the telescope this project was designed for. The goal is to apply MKK's standards to the mass of publicly accessible data put forth by the Sloan. The Sloan has a different way of gathering data, in fact it has two complementary ways:

• Through direct download
• Through SQL search

Direct Download

Direct download from the database is possible through the use of a Java program written by Chris Stoughton This program is simple to use if you follow these steps:

NOTE: This program requires a Java Runtime Enviroment 1.6 or higher to be installed, install one before you attempt.

In order to determine the version of Java you have, type the following in the command line:

java -version

you should get something like

java version "1.6.0_20"
Java(TM) SE Runtime Environment (build 1.6.0_20-b02)
Java HotSpot(TM) Client VM (build 16.3-b01, mixed mode, sharing)

If you have a Java version less than 1.6, you need to update Java

1. Right click this link and select "Save as"

   NOTE: when downloading the above file, make sure the file you are saving is called "SpSpecToCSV.jar"

2. Save it to a location that is easy to get to because you will access this file when we execute the program through the Command Prompt (Windows) or Terminal (Mac OS X or Linux)
3. • On Windows go to the Start menu and click on "run" then type cmd and press enter
   • On Mac OS X go to Finder > Go > Utilities > Terminal
   • On Linux get to a shell prompt
4. Type in java -jar [exact path to file saved in steps one and two] 51690 306 114 5
   • If you are on Windows, had a username of jbrown and saved the file to 'My documents', you would first type

     cd C:\Documents and Settings\jbrown\My Documents\

     Then you would type

     java -jar "SpSpecToCSV.jar" 51690 306 114 5

     Change everything inside the double quotes (") to the exact path of the downloaded file

     ---

   • If you are on Mac OS X, had a username of jbrown and saved the file to Documents, you would type

     cd /Users/jbrown/Documents

     Then you would type

     java -jar SpSpecToCSV.jar 51690 306 114 5

     ---

   • If you are on Linux, have a username of jbrown and saved the file to Documents, you would first type

     cd /home/jbrown/Documents

     Then you would type

     java -jar SpSpecToCSV.jar 51690 306 114 5

This code will read a data file, and convert it to a format accessible by a spreadsheet program.
The numbers at the end, i.e.

51690 306 114 5

Tell the program which spectrum to retrieve and convert.

51690

Is the 'mjd' or Modified Julian Date, which is a counter from a certain date, it tells you the night the spectrum was taken.

306

Is the 'plate', which corresponds to the actual metal plate with holes drilled in it that was used in the SDSS.

An Aluminum Plate with holes drilled precisely at the location light from distant stars will hit

114

Is the 'fiberid', which corresponds to which hole in the plate the spectrum was taken on.

So, the combination of mjd, plate,and fiberid uniquely identifies one spectrum in the SDSS.

5

Tell the program to average over five values. This helps by smoothing to reduce the effects of noise. Set this too high, however, and the spectrum becomes a line.

SQL Search

SQL (Structured Query Language) is a publicly available database of all the processed data the SDSS has collected over the years. We can find spectra to directly download by writing a SQL Query. For a complete description of how to use SQL to analyze stars, use this document

---

The online tool to find SDSS data is here, with a reference page here

---

This link will allow you to write a query and get information about spectra from the SDSS. A example of a very simple query is this:

1. Click here
2. Click the "clear query" and check that the html button is toggled. The box with white background is the area you will enter your query.
3. Now write the following and then click on the "submit" button:

   Select top 5
   mjd,plate,fiberid
   from sppParams

This will cause another file easily accessed by a spreadsheet program to be created. Opening that file in your spreadsheet program will yield five sets of mjd, plate and fiberid's to use with the direct download.

To add a bit more functionality to the SQL search, you can instruct it to find stars of a certain type e.g.

Select top 5
mjd,plate,fiberid
from sppParams
where sptypea="m0"