Galaxy Color

Galaxy Color2 stars

NGSS Connections

Supports PE HS-ESS1-2 : Earth’s Place in the Universe

Supports DCI ESS1.A: The Universe and Its Stars, PS4.A Wave Properties, PS4.B: Electromagnetic Radiation

Engages in SEP 1:Asking questions, 4:Analyzing and interpreting data, 5:Using mathematics and computational thinking and CCC 1:Patterns, 6:Structure and function: The motion and make up of stars and galaxies provide evidence for the Big Bang theory.

When Edwin Hubble created the tuning fork diagram for organizing galaxy shapes, the photographic technology at the time produced only black and white images.  As soon as the technology was available to produce color images, many new questions in astronomy were opened for exploration.  What can you discover when you apply color to the original Hubble classification system?

Begin by asking a question.  Your question can be general or specific depending upon what you already know about galaxies and color.

Our first step in adding color to the Hubble tuning fork diagram is to gather between 15 and 25 galaxy images from the SDSS database.  It’s a big database so everyone should have a unique starting place.

Find a Starting Place

We will use RA and Dec to define a starting place.  If you completed the activity My Special Place in the Database, you already have a set of coordinates. Use the RA and Dec that you recorded during that activity.

If you do not have a starting place, you can choose one using the Constellations Notebook. Need help? See Constellations Notebook Help.

Search the Database Using SQL

From your starting coordinates, gather images of galaxies that you can place on the Hubble tuning fork diagram.  You can go to the Navigate tool and search the area by hand but a faster way is to send a request to the database using SQL or structured query language.  SQL allows users to see specific images and data associated with objects that meet the request commands.

The first step to conducting a query is to think carefully about your goal.  What do you want your query to do?  Think about the galaxies you have observed previously in SkyServer. Do you want any object that has been identified as a galaxy in the SDSS or are there limits to some characteristic like size and brightness?

A Query to Start With

If you have experience with SQL, you may engineer your query from scratch and compare yours with the sample below.  If you are new to SQL follow the example closely and make modifications in your notebook. Start with a clear understanding of what you want your query to do.

The goal of this query is to return a collection of:

  • 15-25 galaxies
  • galaxy images are large enough to observe their structure
  • all galaxies are from a particular starting place

The first line of the query, called the SELECT line, is the “Go get this” part of the search.  Information in the SDSS database is stored in tables.  This line describes what information you want returned and the table where it is stored.  For this query, we will need information stored in two different tables.  We will designate one “p” and the other “s.”  We will define these letters in the next block of the query, the FROM line.

SELECT p.objid, p.ra, p.dec, p.u, p.g, p.r, p.i, p.z, p.petrorad_r, s.z as redshift

This line says – Return the following information

  • the unique object identifier (objid)
  • the location of the object (RA and dec)
  • the magnitude of the object in all five SDSS filters (u, g, r, i, z)
  • the Petrosian radius as measured in the red filter (petrorad_r), which gives us a measure of how far the light from the object spread out from the center. See Help – Photometric Information for more information.
  • the redshift of the object, which the database knows as “z” but is easier for us to remember as redshift

The second section of an SQL query is the FROM line.  This line defines which table contains the needed information and how it is to be handled.  This is also the line where we will define a starting point in the sky.  Below is an example.

FROM photoObj p join dbo.fgetnearbyobjeq(165,48,120) n on p.objid=n.objid join specObj s on p.objid=s.bestobjid

This line says – The data we want

  • is in the area nearby the coordinates RA 165, Dec 48 in an area that has a radius of 120 arc seconds (dbo.fgetnearbyobjeq(165,48,120) n) or 4 times the diameter of the full moon.
  • is contained in two tables, the photoObj table which we nicknamed “p” and the specObj table, nicknamed “s.”  We want to be sure the data from the two tables match to the same object so we JOIN all the all the requirements ON columns that we know match (=).

The last section of the query is the WHERE block.  Here you add any conditions that constrain the data that is returned.

WHERE p.petrorad_r > 15 and p.r < 19 and s.Class='GALAXY' and s.z >= 0.01

Our query was constructed to meet our goal by

  • choosing objects where a particular amount of the total light is spread out more than 15 arcseconds from the center point as measured with the red filter (p.petrorad_r).  This increases the chances of getting large galaxies.
  • AND have magnitudes in the red filter brighter than magnitude 19 (p.r < 19) We want bright objects so we can observe details.
  • AND have spectra that have been labeled galaxy (s.Class=’GALAXY’).  The SDSS  identifies galaxies on the basis of their spectra.
  • AND have redshifts greater than or equal to 0.01  (s.z >= 0.01) This one is a little added insurance against the query returning stars that are so bright they spill over from their point sources and can be mistaken by the SDSS for galaxies.  Galaxies, being outside our Milky Way, almost always have at least a small redshift.  Stars within our galaxy do not.  If we add a command stating that the returned objects must have at least some redshift, we eliminate bright stars.  Try running this query without this command and see for yourself.

All together our sample query reads:

 select p.objid, p.ra, p.dec, p.u, p.g, p.r, p.i, p.z, p.petrorad_r, s.z
               from photoObj p
               join dbo.fgetnearbyobjeq(165,48,120) n on p.objid=n.objid
               join specObj s on p.objid=s.bestobjid
               where p.petrorad_r > 15 and p.r < 19
               and s.Class= 'GALAXY'
               and s.z >= 0.01

Run Your Query in SkyServer

There are a number of places in SkyServer from which you can run a query. We have placed one of those tools below so that you may see a chart of the returned information and can be sure it runs properly.  Give it a try.

SkyServer SQL Search

If your query does not return between 15 and 25 galaxies, modify your query and try again. Record any changes you made.

Run Your Query in Image List

For this activity, we need color images of galaxies that we place onto the Hubble tuning fork diagram.  Seeing images right away speeds up this process. You need to use a tool called Image List.  Image List requires us to modify the SELECT line to include only name, RA and Dec.

If you are familiar with the SDSS database, you know that there is no “name” column in any table in the database.  We have to tell the Image List what column we are using to identify our objects as well as eliminate requests that are not allowed.  Here is how that is done:

We start with our original SELECT line and get rid of requests that are not allowed.

SELECT p.objid, p.ra, p.dec, p.u, p.g, p.r, p.i, p.z, p.petrorad_r, s.z as redshift

Keep the p.objid as the unique identifier but redefine it “as name.”

SELECT p.objid as name, p.ra, p.dec

The remainder of the query stays the same because it describes which objects we want to see and that has not changed.  Our completed query for Image List is now:

select p.objid as name, p.ra, p.dec
               from photoObj p
               join dbo.fgetnearbyobjeq(165,48,120) n on p.objid=n.objid
               join specObj s on p.objid=s.bestobjid
               where p.petrorad_r > 15 and p.r < 19
               and s.Class='GALAXY'
               and s.z >= 0.01

To submit your query to Image List:

  • Open Image List a new tab
  • Find and select the Use SQL Query to fill coordinates list link, on the right side below the cut and paste box.
  • Replace the query that appears in the box and click submit.  A table of results will be returned.
  • Above the returned information is a Send to List button.  When you click this button the Image List tool populates with thumbnail images of your search results with links to other tools in the SDSS.

Image List Links to Other Tools

Although you can create a table of data using your original query, you can access more information through Explore and Navigate in SkyServer.  The header for each image in Image List links to the Explore tool and the thumbnail image links to Navigate.  The Navigate tool allows you to zoom in and out on your target for closer inspection. Explore contains additional data and calculations.

Colorize the Tuning Fork

From your Image List search results,  place each galaxy on the Hubble classification diagram based upon its shape.  Do you observe any patterns related to color when you do this? If you are working in a group, share and compare your results with others.

Once you have made some observations, can you offer any additions or changes to the tuning fork diagram?

One Good Observation Leads to the Next Experiment

The next part of this expedition requires you to design an experiment to verify one of the patterns you observed while colorizing the Hubble Tuning Fork Diagram or to explore a new question that resulted from the activity.  Formulate one or more questions related to what you observed, or concluded, that could be used as a starting point.

When you are ready to go further try your own Investigations of Galaxy Color.