Author Topic: Zooite Guide to SDSS Spectra  (Read 39357 times)


  • Hero Member
  • *****
  • Posts: 3860
    • View Profile
Zooite Guide to SDSS Spectra
« on: April 05, 2009, 07:59:56 pm »
A guide to spectral charts from SDSS used by Galaxy Zoo 2:

This guide doesn't work for Hubble Zoo.


Part 1 Introduction - below
Part 2 How SDSS shows  colours
Part 3 The spectra of stars
Part 4 The spectra of galaxies
Part 5 Emission and Absorption Spectra
Part 6 Galaxies with an active nucleus
Part 7 A spectral analysis of 6 round green things 
Part 8 Links to spectra-related threads

Part 1: Introduction

Galaxy Zoo uses the Sloan Digital Sky Survey(SDSS) to present a picture of a galaxy for classification.
[EDIT in 2010:
In 2009, Galaxy Zoo also started using SDSS stripe 82 pictures. They have red, green and blue speckles. They sometimes have spectral charts. Take the picture into SDSS Navigate, and change "stripe 82" to "dr7" in the address bar at the top.
In 2010, Galaxy Zoo also started using Hubble pictures. They have amber and cyan (red and blue) speckles. They sometimes have spectral charts. The spectral charts are more red-shifted, because the objects are further away and sample a smaller range of wavelengths.]

Perhaps after classifying it, you would like to learn more about the galaxy.
So, while you are classifying the shape of the galaxy, you can save it for later study by clicking
"Add to My favourites"
Then you can find it again in
"My galaxies"
Click on the Object Id number (ObjId) learn more about the galaxy from SDSS.

You could post the galaxy in the Forum, and discuss it with the other Galaxy Zooites, so we can all learn about the galaxy.
There is information on how to post a picture, spectrum and ObjId in Threads to help you find your way around Galaxy Zoo forum.
Remember to do a search to see if that object has already been discussed on the forum.

On the SkyServer page, there is often a spectrum of an object.
There is an amazing amount of information in a spectrum of an object. We can deduce:
whether the object is moving towards us and away from us (redshift) and
 internal movement within the object
the atoms and molecules in the object (emission spectra)
the atoms and molecules between us and the object (absorption spectra)

This is a simple guide to colour and the spectra of galaxies.
There are more comprehensive guides on galaxy spectra on the forum:
by NGC3314 on Tutorial bits on galaxy spectra and
by Eigenstate in Introduction to Electronic Spectroscopy and Introduction to Spectral Line Profiles
SDSS has a list of typical spectral charts    
Spectral cross-correlation templates

sample is 3" across
SDSS has tutorials and projects on DR 7 projects

This is the  large telescope in New Mexico, USA used for taking the SDSS pictures. The second picture shows which part of the sky has been observed by SDSS. You can see the Plough, or Big Dipper, in the picture, so most of the galaxies that we see in Galaxy Zoo are from that part of the sky. Image credited to the Sloan Digital Sky Survey

Thanks to Alice for suggestions for making the format  more interesting and legible, to ZookeeperKevin  and Thomas J for their enthusiasm, and to NGC3314 for contributing to the forum (many of his postings have been included in here taken directly from the forum) and for looking over the final guide. - Budgieye

Post suggestions for improvements to this guide on
Galaxy Zoo Forums > The site and the science > Science Questions > Zooite Guide to Spectra by Budgieye

« Last Edit: September 11, 2010, 12:11:53 pm by Budgieye »


  • Hero Member
  • *****
  • Posts: 3860
    • View Profile
Re: Zooite Guide to Spectra Part 2 How SDSS shows star colours
« Reply #1 on: April 05, 2009, 08:00:56 pm »
 Spectral charts, for Newbies Part 2 How SDSS shows colours

Galaxies are made of stars, so let us start simply and look at the colour of stars.

Stars come in different colours:
Blue, white, yellow, orange, red

The colours are similar to the rainbow spectrum.
violet, indigo, blue, green, yellow, orange,  red.

In addition to these rainbow colurs, SDSS also “sees” ultraviolet (UV) and infrared (IR).
ultraviolet (UV), violet, indigo, blue, green, yellow orange, red, far red, near IR
ultraviolet (UV), violet, indigo, blue, green, yellow orange, red, far red, near IR

The colour of a star is its principal colour, but it emits other colours as well.
Blue stars are hot and emit lots of all colours of light, from energetic UV and blue, to low energy zIR.
Red stars are cool, and emit lower energy red light, but cannot emit blue.

A white dwarf star emitsUVbluegreenyellowredfar rednear IR
A blue star emits UVBLUEgreenyellowredfar rednear IR
A  white star emits UVblueGREENyellowredfar rednear IR
A  yellow star  emits   UVbluegreenYELLOWredfar rednear IR
A  red star  emits  greenyellowREDfar rednear IR
A  red dwarf emits       yellowredfar rednear IR
A brown dwarf emits      far rednear infrared

Light travels in waves. Each colour of light has its own wavelength, measured in Angstroms (Å).

Here are the main wavelengths of light
Gamma rays<          1 Å
X-rays   approx 1 Å
A hydrogen atom              1 Å
UV filter in SDSS       3551 Å
Violet3800-4200 Å
Blue4200-5000 Å
Green5000-5500 Å
Yellow5500-6000 Å
Red6000-7000 Å
IR on SDSS  7000-9200 Å
Microwaves are   300,000 Å
A human hair is 1,000,000 Å wide
Radiowaves are the width of a human hair to kilometers long

SDSS takes pictures in 5 colours  ugriz
which corresponds to                 UVgreenredfar rednear IR
with wavelengths (Å) of             3551Å  4686Å  6165Å  7481Å  8931Å 

(Here is a picture of the CCD camera, showing the red, green and blue filters)

The range of each colour is shown in SDSS system response plot (the filters)

It can only send 3 of these  5 colours to our eyes, because computer screens only receive 3 colours, ie they have red, green and blue pixels.
But SDSS collects its best data in far red (i), red and green.

So, SDSS takes:
far red   (i)  light from the star, and turns it into red
red       (r)  light from the star, and turns it into green
green     (g) light from the star, and turns it into blue

So the stars and galaxies are not their "proper" colours.

What colour will the object be?
You mentally take the area of each part of the spectrum.
The colour that has the biggest area wins.

587734623246614577   red..................near IR
There is more blue, so it is a blue quasar

NGC3314 gives an explanation here
and here is colour information in the asteroids OOTD
Here is an asteroid 588017726549852357 posted by Alice, showing how it moved between exposures with each filter. We are shown only the g, i, and the r  exposures.
Quick astronomy note: all planets and most asteroids orbit from west to east, or right to left in this picture taken from the northern hemisphere.
OOTD Centaur asteroid: furthest from Earth bu waveney
OOTD Five asteroids by geoff
OOTD Asteroid motion in the heavens. by Fermats Brothers

Quick biology lesson on how our eyes see colour:
 if we see blue + green + red light mixed together, we see white
 if we see …… + red light mixed together, we see yellow
 if we see blue light mixed together, we see magenta
 if we see blue + green.........light mixed together, we see cyan

Here are some stars (and Hanny's Voorwerp), arranged from hottest to coolest.
imageclass of starlight from starcolour in SDSSObjId
"White" dwarf stars         
"UV" star  a better name?
emit lots of UV 
some red and IR     
appear blue in SDSS1237667444048658632
Blue starsemit some UV
lots of blue,
some green,
some red and IR
appear blue-white in SDSS1237667323251327051
Hanny's Voorwerpemits green lightbut appears blue in SDSS1237661382772195480
White starsemit all colours of lightappear white in SDSS1237652935641400757
Yellow starsemit green, yellow,
red and IR
appear green in SDSS1237654881262239849
Red carbon starsemit green
red and IR
appear yellow in SDSS1237667254542794964
Red stars emit yellow, red and IRappear orange on SDSS1237667430106661549
Red dwarfsemit red and IRappear red on SDSS 1237660670887526594
Brown dwarfsemit IR appear dim red on SDSS1237678617434915718
Radio sourceemits IR
and radio waves
invisible in SDSS1237651822716716377

There is colour information on the SkyServer Object page.
see Finding information for your target object
see SDSS for Skyserver SR7 Visual Tools

Here is someSkyServer Object information for a red dwarf 1237660670887526594

Filter name           ugriz
magnitude of light      24.11        23.22        21.12        8.78         16.99       
Wavelength          3551468661657481 8931
Colour of light        UVgreenred far rednear IR

Large numbers eg.above 20 mean that there is not much light of that colour in the star.
Little numbers eg. below 15 means that there is a lot of  light of that colour in the star.
(In the same way that a 1st magnitude star is brighter than a 25th magnitude star)
(And a first class train ticket is better than a third class)

The brightness of things in the SDSS has a wide range - something with a magnitude of 15 in a particular filter appears 10,000 times brighter than something with a magnitude of 25 in that same filter. (They don't let bright stars into the spectroscopic observation to keep their stray light from wiping out faint galaxies, for this reason).

Be careful which z you are looking at. There is a z in two places.
Explanation of the photometric z by NGC3314

SkyServer Object information for a hot blue star: the amount of light is compared to a bright blue standard star, with an explanation here

Here is some pictures which were struck by a  cosmic ray.
You can recognize them by their pure blue, green or red colours. They have sharp edges.
These are small, but some are larger, such as Alice's Cosmic Finches
Go into SkyServer and look at the u,g,r,i,z data. There is light in only one column.
The value for g is real            The value for r is real             The value for i is real
1237657775547220153            1237660962941304959           1237661949733766238posted by Tsering           

The path of a photon from a galaxy to SDSS is shown OOTD Friday 9th January, 2009.

Showing colours on a spectral chart

from  The emission spectrum of hot iron

A spectrum is light from an object, spread out into its rainbow colours.
It is difficult to work with a picture, so the light from the wavelengths are drawn out as a spectral chart.

This is the spectrum of hot hydrogen taken on Earth.

Here is the spectrum of a galaxy which has hot hydrogen.

« Last Edit: October 22, 2013, 09:20:39 pm by Budgieye »


  • Hero Member
  • *****
  • Posts: 3860
    • View Profile
Re: Zooite Guide to Spectra Part 3 The spectra of stars
« Reply #2 on: April 05, 2009, 08:02:07 pm »
Spectral charts,  for Newbies Part 3 The spectra of stars

[EDIT  dr8 has a wonderful site showing on the spectra of many types of stars ]

The y axis (the axis going up) is the amount of light.
The x-axis (horizontal) is the the wavelength of light (ie the colour).
So the chart is a graph of the amount of light at each wavelength.

We don't have to know the units of the amount of light. Just that 3 is small and 1000 is big.
(If you want to know, it is flux: the amount of energy, falling on a square centimeter, per second, for each wavelength)

4000  Å is the wavelength of blue light which has smaller waves with higher energy.
7000  Å is the wavelength of red light  which has larger waves with less energy.

For example, a spectrum with high numbers at the blue end will have come from a high energy i.e. hot object.

Here are the spectra of some stars, from hottest to coolest.


1237667444048658632   red.............near IR
a "white" dwarf, a very hot star that produces alot of UV light.
The exposed core of a star that has exploded, with a surface temperature of approximately 50,000K
The graph is so smooth that the computer can't decide what it is. See OOTD Wednesday 21st Jan 2009 for information on how things shine.
Posted and much discussed in Mysterious blue blob wanting expert advice by Alice, Thomas J, laihro, and astrobrainic

1237667323251327051   red.............near IR
blue star, it emits UV, blue and green light.

1237652935641400757   red.............near IR
white star emits all colours of light in approximately equal amounts. The blip at 5600Å is an artifact of sky conditions.

1237654881262239849   red.............near IR
yellow star (it looks green) found by starry nite

1237667254542794964   red.............near IR
red star (carbon star),  the "yellow" star in the middle
These cooler stars have a jagged spectrum. This is due to absorption of light by the molecules that can survive in the cooler atmospheres of these stars.
posted by stellar190


1237667430106661549   red.............near IR
red star. The spectrum is looking very jagged.

1237660670887526594   red.............near IR
red dwarf
posted by Pat

587731513688392555   red.............near IR
brown dwarf, found by starry night
It is dim, the spectrum hovers around zero. There is a small amount of IR light.
posted by starry nite


This looks like a mistake, but it is a radio source.
There isn't a spectrum, but if you look at the i and z on the SkyServer page, you can see that it is emitting infrared light.

filter bandugriz
17.00 means the object is bright in the z band (near infrared)            
If you click on ADS in Navigate, you will see a reference to a published paper on radio sources.

1237662262169305112   red.............near IR
blue and red star together, a U-shaped spectra
posted by starry nite
There are more like this in Spectroscopic binaries

This is an object that moved, posted by dthomas02  : explanation here
But we can see what data we get from the background sky, (explanation from NGC3314)
The spectrum stays near the zero line, responding to random bits of light.
See the green line? It tells of the viewing conditions on the night.
The peak at 5577Å is from airglow, a sort of weak permanent aurora (Northern Lights) in which atoms (in this case, oxygen) in the upper atmosphere get jostled and radiate away the energy they have gained.
The wriggly bit at 7000-9000 Å is from water in the atmosphere.
Some spectral charts from other locations may have interfering light at 5890 and 5896 Å from sodium streetlights.  The SDSS site at Apache Point is far from any sizeable city for just this reason. (El Paso is about 200 km to the southwest, with mountain ranges helping to block its most direct light pollution).

Here are some threads about stars
White Dwarf Stars
A cool White Dwarf Star
Cataclysmic Variable Stars (white dwarfs with emission peaks)
Carbon Stars
dMe Stars (red dwarfs with emission peaks)
Brown Dwarf Stars

« Last Edit: October 22, 2013, 10:11:05 pm by Budgieye »


  • Hero Member
  • *****
  • Posts: 3860
    • View Profile
Re: Zooite Guide to Spectra Part 4 The spectra of galaxies
« Reply #3 on: April 05, 2009, 08:03:27 pm »
Spectral charts,  for Newbies Part 4  The spectra of galaxies

The spectra of galaxies are different to spectra of stars.
A galaxy has billions of stars, a black hole and energetic gas contributing to its spectrum.
Also, galaxies are much, much further away, and the wavelengths of light coming from it are shifted towards the red end of the spectrum: this is called the redshift.

The redshift value is the z value on the SkyServer Object Explorer (the red z)
The redshift value is calculated by looking for “the four thousand Angstrom break”  at 4000Å on the spectrum.

The stars in our galaxy have a redshift of zero. They are not moving away from us or towards us with any great speed.

1237652935641400757   red.............near IR
This is a white star.
Look at the 4000Å break in the spectral chart of the star. At 4000Å the average values make a step upwards, in this case from 0.5 to 1.5
The 4000Å spectral break is at 4000Å, so it has a redshift of zero.
An object with a redshift of zero must be in our galaxy, or very close to our galaxy, so is probably a star.

In galaxies, the 4000Å break doesn’t occur at 4000Å ! ??? ??
The break is shifted toward the red end of the spectrum.
The z value measures how far towards the red end of the spectrum the break is shifted, and tells us how far away the galaxy is.
Explanation of redshift by NGC3314 in Tutorial bits on galaxy spectra

The redshift is caused by the galaxy moving away from us at high speeds which stretches the wavelengths of light.
We can use the redshift to estimate distance.
Explanation about z here on the z thread   
A redshift to distance calculator
OOTD Friday 20th February, 2009 Systems : are galaxies interacting?


588017730300543029   red.............near IR
z=0.0065 or 89.3 million light years (Mly)
This one is close enough to have a name, UGC 06876  Nearby galaxies are white or blue. The z value is printed on the graph of the spectrum. The spectral break is close to 4000Å
Posted by miago

587733603718987891   red.............near IR
z=0.0771    distance is 1,007 million light years. The spectral break is at 4300Å

587730847428510150   red.............near IR
z=0.132 or distance = 1,600 million light years. Most of the colours have been shifted into yellow. The spectral break is at 4500Å

588017627222573280   red.............near IR
z=0.525 or distance = 5,200 million light years. The colours have been shifted into orange. The spectral break is at about 6000Å
Posted by gymgoki

588017978355745177   red.............near IR
z=0.884,  distance pf 7,200 million light years
The spectral break is at 7500Å This galaxy is the farthest I could find on the thread "Biggest red shift z" for a "normal" galaxy.
Posted by AlexandredOr

SDSS can't detect ordinary galaxies that have a red shift greater than about z=1.2 The light from these galaxies is being shifted off the end of the chart far into the infrared.

Some distant active galaxies, such as quasars, emit lots of X-rays and UV light, and these can be redshifted down to visible light, and therefore we can still see them on SDSS.

Negative Redshift, or Blueshift

A nearby galaxy can have a negative redshift if it is coming towards us. You can see this in Object of the Day . The galaxy M90 has a negative redshift, z=-0.0008. The light we see coming from this has been blueshifted, or moved to the left in the chart. So M90 is coming towards us.

OOTD Friday 27th March 2009: Too close for comfort? lists 18 galaxies that are moving towards the Milky Way.
OOTD Tuesday 13th Jan 2009 They are coming for us
OOTD Wednesday 21st October, 2009 Blueshift irregulars

587738568167653516   red.............near IR
z= -0.002  , possibly 30 million light years away, it is hard to tell with small redshifts
Posted by LankyYankee

Here is another negative redshift, courtesy of LankeeYankee The spectral break is at about 3992 Å
It is part of the nearby Virgo cluster, and is coming towards us.
Notice that the computer has incorrectly decided that the object is a star.

Very small redshifts in spectra from nearby galaxies are written like this
7.45E-3 instead of 0.00745

« Last Edit: October 22, 2013, 10:34:59 pm by Budgieye »


  • Hero Member
  • *****
  • Posts: 3860
    • View Profile
Re: Zooite Guide to Spectra Part 5 Emission and Absorption Spectra
« Reply #4 on: April 05, 2009, 08:04:26 pm »
Spectral charts, for Newbies Part 5 Emission and Absorption Spectra

The orbits of electrons of  atoms and molecules can emit and absorb light
They use the same wavelengths of light (the same colour) whether they are emitting or absorbing.
If you can determine the line or wavelength, you can find what atom or molecule is involved.

Emission lines make peaks, like this "^" on a spectral chart.
Absorption lines make valleys, like this "V" on a spectral chart.


Emitted light makes bright lines in a spectrum, which are drawn as peaks "^" in a spectral chart.

Compare a spectrum of hot hydrogen with a spectral chart of a galaxy with hot hydrogen in it.
See how 4 of the bright lines of the spectrum correspond to some of the peaks on the spectral chart from the galaxy.
We can conclude that there is hot hydrogen in this galaxy.

For the quantum theory of light emission see    
OOTD Saturday 17th October 2009 Hot hydrogen.. a billion light-years away

See this OOTD Thursday  2nd April 2009 for examples of emission spectrum.

Normal stars are not hot enough to have emission lines!!!
Galaxies which consist of only normal stars will not have emission lines either!!!

From NGC3314: "Normal stars have either no emission lines or emission lines so weak that they do not show up in data of this spectral sampling.
Emission lines generally come from excited gas, either:
gas illuminated by hot stars,
gas outflowing from stars,
gas shocked by supernova explosions, or
gas lit up by the hot surroundings of a black hole."
gas heated by merging of galaxies
Picture of galaxy with gas temperature of millions of degrees

The most common gas is  hydrogen gas, so hydrogen will produce the main peaks.
They are the hydrogen alpha (Hα), the hydrogen beta (Hβ) and then hydrogen gamma (Hγ) peaks.

587738568703017016   red.............near IR
Here is an elliptical galaxy with no excited gas. There are no emission peaks.
All colours are present in equal amounts, and the galaxy is white. Posted by Lanky Yankee.

587738195572949357   red.............near IR
Here is a galaxy with some activity. There is an emission peak at 6800Å.
Hydrogen gas is falling into the black hole in the nucleus, getting very hot.
The galaxy is white, but the nucleus is red. The hydrogen alpha emission in the red part of the spectrum gives the nucleus its red colour.
Posted by AlexandredOr

587739158720544838   red.............near IR
This is a quasar, which is a galaxy with a very active nucleus. It has a large hydrogen alpha peak, a hydrogen beta peak, and a peak of excited oxygen.
Large amount of matter (stars?) are falling into the black hole. Some matter is ejected out of the top and bottom of the quasars, forming jets.
Explanation of the width of quasar emission lines by EigenState and NGC3314
Posted by zeta2008

588848899912040466   red.............near IR
A blazar
The ion jet is pointing right towards us
from "Method identifies blazar candidates"
SDSS J115404.55-001009.8
spectrum goes straight across, (with values high above zero) and no apparent emission peaks
This is apparently how to recognize a blazar
[.... ]from here)  [......]The stronger the blazar component is, the less there is to see in an optical spectrum - it was a frustrating struggle for the discoverers to even get their redshifts and find out what kind of objects these are.

587736809916399664   red.............near IR
A Wolf-Rayet galaxy with excited hydrogen and oxygen gas.  The hydrogen alpha emission is represented by red light, and the oxygen emission is represented by blue light, so the colour of the gas is magenta (blue-red).
Matter is being forced away from the nucleus, forming jets.
Look at the magnitude of the units on the y-axis. :o Posted by Titanium Dragon

Exploring that nice magenta colour.
This colour is a blend of red and blue, called magenta.
The blue colour comes from the OIII emission line, a green wavelength, presented to us as blue.
The red colour comes the hydrogen alpha line, a far red emission, presented to us as red. If the hydrogen alpha line is redshifted to 7000Å or greater, it appears as red in the galaxy photo, and the magenta colour appears.

587735346962497616   red.............near IR
The OIII peak makes a blue colour.
The hydrogen peak makes a red colour.
blue + red = magenta (also called purple or even "pink")
posted by sdrew123


587745244701655168   red.............near IR
posted by Mukund Vedapudi
This is a Compact Blue galaxy.
There is much blue light between 4000-5000Å, so there are many young blue stars.
It has emission lines of oxygen and hydrogen.
The oxygen peaks makes blue colour.
The hydrogen peaks makes green colour.

OIII galaxies:Some galaxies with active nuclei have emission peaks coming from hot oxygen gas.
OIII emission peaks are light emitted by oxygen atoms when two electrons are torn away from it.
This is known as doubly ionized oxygen.
The lines are at 5007Å and 4959. This is green light, but SDSS presents green light as blue, so galaxies emitting this hot ionized gas are blue when close by (z>0.1)

88010359621353529   red.............near IR
This is a blue OIII object, a "blue pea". The big emission peak of OIII is near 5000Å.
posted by Galaxy Hunters Inc

587738410863493299   red.............near IR
This is a similar looking chart, but the galaxy is farther away (z=0.14) and the OIII emission line is redshifted past 5500Å into the part of the chart where red light is presented as green, so the object looks green. These objects are nicknamed  "peas", because they look like the green garden vegetable.
posted by Rick Nowell

Research is still on-going in this newly discovered type of galaxy. More info on:
OOTD Friday 21 May 2010 Spectra of Peas! Blue, purple, white, green and red
OOTD Tuesday 14th October, 2008
topic:OIII Objects and
The story of the peas: writing a scientific-paper
Galaxy Zoo Green Peas: Discovery of A Class of Compact
Extremely Star-Forming Galaxies

OOTD Saturday 19th September, 2009
See the thread Give peas a chance

Here is a list of emission lines used by SDSS.


If there is gas or dust is in between the galaxy and us, the gas or dust will absorb some of the light.
Different atoms and molecules absorb different wavelengths of light.
This absorption makes black gaps in a spectrum, which are drawn as valleys "V" in a spectral chart.
Cooler stars and galaxies have many valleys, and from these, you can determine what atoms or molecules are present. 

587734623784075325   red.............near IR
Here is an elliptical galaxy with little excited gas. All colours are present in equal amounts, and the galaxy is white.
There are no emission peaks here. The "peaks" that you see are gaps between large absoption valleys.

The spectrum of our sun has 2 black lines at "D", where sodium (Na+) absorbs light.

The spectral chart of this galaxy shows absorption by Na+

The computer labels the important wavelengths for us.

The green lines mark wavelengths where absorption commonly occurs.
Common atoms are Na+, K+ and Mg++, also N
G in the spectra of stars and galaxies doesn't denote a specific element, but a specific Sun-like spectral feature. In this case it comes from many weaker things blended together - the CH molecule, Fe lines, and more.

The black lines mark weaker or uncommon features. 

The magenta dot-dash lines are the wavelengths of strong airglow emission features from our atmosphere (such as [OI] 5577 Å). These are the most likely wavelengths for upward or downward artifacts in a faint-object spectrum in case of imperfect sky subtraction, so any peaks or valleys at these locations may be artifacts.

The other dashed lines show wavelengths of expected (emission or absorption) spectral features at the estimated redshift of the object.
The green spectrum at the bottom shows the blank-sky spectrum which has been subtracted, so you can see whether that was a large correction or not (and whether there was some problem with the automatic processing).


587735665854841008    red.............near IR
In this spiral galaxy, the dust lane blocks the light from the nucleus.

Explanation from NGC3314 “That's the kind of spectrum you see when a lot of the starlight from a galaxy has been filtered through interstellar dust. Since many of the dust particles are smaller than the wavelength of visible light, it blocks bluer light more effectively than red light (just like the particles in our atmosphere - starlight behind dust is reddened the same way our local star appears reddened at sunrise and sunset). This kind of spectrum - sloping strongly to the red but with the absorption line ratios typical of bluer populations - shows up in a lot of galaxies with strong dust lanes. It also shows up in some red spirals even when we see them face-on, but that's another thread.”

The spectrograph can only target an area 3" across. 
See OOTD Tuesday 3rd Feb 2009 for a picture of the spectrographic plate.
The arms of a spiral galaxy have more young blue stars, so there is more blue in the spectral chart from the arm.

spiral arm 587739407876554999
spiral galaxy nucleus 587739407876620430
posted by jip26

Spectroscopy becomes very complicated
Topic: Zooite Guide to Spectra by Budgieye

For more information:

SDSS has examples of  spectra of 8 diverse objects at

Here are two  OOTDs on Thurday 19th March 2009 (with pictures!) on the invention of the spectroscope and spectra and another OOTD on Thursday 2nd April 2009 absoption spectra

Another OOTD Thursday 8th October 2009 The spectrographic plates at Apache Point

Objects with a high OIII peak OOTD Friday 16th January, 2009 and OOTD Tuesday 14th October, 2008

I Zwicky 18  is studied in OOTD Tuesday 2nd December, 2008. This galaxy shows a sudden increase in star-forming activity, a "starburst" galaxy.
There are massive hydrogen alpha lines within NGC 3690 (ARP 299) OOTD Saturday 2nd Febrary, 2008

The importance of dust in the evolution of galaxies in Wanted! Galaxy pairs which overlap but are not merging

OOTD Wednesday 3rd June 2008 Metals and star populations

Blue ellipticals and  red spirals OOTD Tuesday 9th December, 2009
Blue galaxies on OOTD Wednesday 14 November 2008

SDSS has  information on introductory emission and absorption and advanced emision and absorption

Wikipedia has a good description of the  Emission spectrum .

This is a simple guide to colour and the spectra of galaxies.
There are more comprehensive guides on galaxy spectra on the forum:
by NGC3314 on Tutorial bits on galaxy spectra and
by Eigenstate in Introduction to Electronic Spectroscopy and Introduction to Spectral Line Profiles
SDSS has tutorials and projects on DR 7 projects

« Last Edit: May 27, 2010, 05:03:55 am by Budgieye »


  • Hero Member
  • *****
  • Posts: 3860
    • View Profile
Re: Zooite Guide to Spectra Part 6 Galaxies with an active nucleus (AGN)
« Reply #5 on: April 05, 2009, 08:05:12 pm »
Spectral charts,  for Newbies Part 6                    Galaxies with an active nucleus (AGN)

Galaxies with an active nucleus can be seen from farther away than normal galaxies.  The extreme heat and energy of these objects makes them emit ultraviolet light and X-rays, and this invisible short wavelength light is redshifted into visible light which can be seen by SDSS.

The nuclei of many galaxies have a black hole that is feeding on gas and stars. As objects spin around the hole, they emit energetic radiation eg. X-rays.  Depending on the angle  from which we view the galaxy, they are called quasars, blazars, BL Lacs, Liners and Seyferts. Here is more information in this OOTD Friday 14th November 2008
Here are images of a Seyfert galaxy on OOTD Thursday 15th January, 2009
More information: Quasars. What they are and how to find them
what is a quasar? and Black holes and AGN

Active galaxies have big emission peaks on their spectra.
For nearby galaxies, the hydrogen alpha peak at about 7000Å gives an red colour to the nucleus of the galaxy .
If there are hydrogen beta and gamma peaks at 4000Å and 5500Å they give the galaxy a magenta colour.

587729386083582038   red.............near IR
Active Galactic Nucleus (AGN), and the spectrum shows a large hydrogen alpha peak. The nuclei usually have an orange or magenta colour at this redshift.
This is "a Seyfert it has large 'relativistic' jets coming out equally from both sides.”   
z=0.124  1,600 million light years away
posted by BL Lacertae and Geoff Roynon

587725552265724093   red.............near IR
This is a nearby quasar. The light emitted from the nucleus is green and IR, which is presented to us by SDSS as blue and red, which makes the colour magenta in our eyes.
z=0.11   1,400 million light years away
posted by elizabeth


587729155208708182   red.............near IR
a white quasar, The hydrogen alpha peak is near the end of the chart. There is an even mix of all visible colours, so it looks white.
z=0.310  3,400 million light years away

OOTD showing double-peaked emitters from two close black holes 587730022261129248 posted by c_cld

587735665310498940   red.............near IR
Shows the CIV peak and the MgII peak
The hydrogen alpha peak has moved past the end of the chart. Most of the colour on the chart is blue, so the quasar looks blue.
z=1.590 9,500 million light years away
posted by Mauromarussi

587727214416037041   red.............near IR
................................................................a Lyman alpha break................................
a red quasar (yes, it looks yellow), we are seeing X-rays from the quasar stretched out to be red light. If you see Lyα (Lyman alpha) on the chart, you will know that you are looking at a far away quasar. The CIV peak is still there, but the MgII peak has redshifted past the end of the graph. Most of the light is red and IR, which SDSS presents as green and red, which looks yellow to our eyes.
z=4.328   12,000 million light years away


587727180613353723   red.............near IR
z=5.414  12,600 million light years away
farthest away quasar on SDSS with a spectrum
The Lyman alpha peak is sliding off the end of the chart.
Explanation of by NGC3314 on Lyman alpha peak
The smaller peaks in the middle are not random flucuations of incoming light, they are the emissions know as the Lyman alpha forest. The light coming towards us was captured and re-emitted by clouds of neutral hydrogen. Since the clouds are closer, the peaks have a lower red-shift. The formation of the early universe can be mapped by using these peaks.
posted by starry nite

"Quasars with redshifts larger than ≈ 5.7 cannot be found by the SDSS spectroscopic survey because at these redshifts the observed wavelength of the Lyman  emission line is redward of the i band; at this point quasars become single-filter (z) detections"  from D. Sneider et al, 2005.

just visible, but no spectrum, redshift value from other sources on NED
posted by Mukund Vedapudi

Possible galaxy at

The most distant quasar visible on SDSS. The light from this quasar has been travelling for 12,800 million years (12.8 billion)
There is perhaps a speck of light visible. Certainly, there is a single-filter (z) detection.

There is a chatty description of these distant objects in High Redshift Galaxy Surveys.

dthomas02 did a study of quasar colours

There is an excellent display of quasar spectra arranged in redshift order, by SDSS at Redshift Gallery : Quasars

Let us look at what other telescopes are finding.

What's here? Not a lot you say?

Well that is why it was chosen for Hubble to look at it for 11 days.
The result is the Hubble Deep Field:

Click for a larger view.

This was the first of three deep views into the "nothing".  This was followed by Hubble Deep Field South and Hubble Ultra Deep Field (both of these are outside the SDSS footprint).

The race to find highest redshifts is on. Here are the problems:
1.The objects are far away, and so they are dim.
2. The first galaxies formed were small. Later, they combined to form bigger galaxies.
3. The light is absorbed by clouds of hydrogen. Between z values of 6-20, (12.7-13.5 billion light years away) space was opaque and and tended to absorb light. This was the "The Dark Ages", just before the Reionization Era
4. The light from these objects is redshifted down into far IR light that cannot pass through the Earth's atmosphere. So space telescopes must be used, such as the Hubble Space Telescope IR camera, the Spitzer Space Telescope and the future James Webb Space Telescope.

Objects farther away than 12,000 million light years away are going to be hard to see. What objects can we see?

z=3.09 Astronomy Picture of the Day July 2, 2009 showed a Lyman-alpha blob, a large cloud of hydrogen gas heated by a black hole, 11.5 billion years away. On SDSS, it is 587731186729550732

z=4.0 to 5.76 Hubble Ultra Deep Field images small merging galaxies.

(between z= 6 to 30, clouds of cool unionized hydrogen are making observations difficult)

z=6.5 Spitzer space telescope detects a galaxy that was large,  mature and fully formed only 850 million years after the Big Bang  HUDF-JD2

z=6.96 The most distant galaxy at 6.96 was confirmed by spectroscopy in 2006 by the ground based Subaru Telescope. This is 12,880 million (12.8 billion) light years away.

z~7.3 Another distant galaxy (as of 2008) seen by Hubble Space Telescope and Spitzer Space Telescope is A1689-zD1, which is a lensed galaxy magnified by the gravitational lens around the galaxy cluster Abell 1689. It was imaged with the Hubble Near Infrared Camera.

z=8.2. In May, 2009, a Gamma Ray Burst (from the collapse of a star to a black hole) was detected by the Swift Space Telescope, and an IR spectrum was taken by ground based telescopes and the redshift was confirmed at z=8.2 OOTD Saturday, 2nd May, 2009. This is ~13,000 million light years away (13 billion)

z~8.5 In 2010 Hubble found 5 small  galaxies previously unseen in the Hubble Ultra-Deep Field photo, , after the new Wide-Field Camera 3 (WFFC3) was installed. These galaxies were formed 600 million years after the Big Bang.

z~10 Hubble telescope with new Wide Field Camera  (WFC)  sees distant galaxies
 see OOTD by Rick  The Farthest Away Yet Found

We will have to wait until  the larger James Webb Space Telescope is launched into orbit in 2014. This telescope "sees" only in the infrared, and will be able to see highly redshifted galaxies of z=15-30.. The mirror will be bigger, and will be able to resolve more detail about the formation and merger of the first galaxies and stars.

z=16 The first galaxies are formed, when the universe was 250 million years old. These galaxies are small, not much bigger than globular clusters, but shining very brightly with new stars.

z=30  First stars formed, when the universe was 100 million years old. These stars were large and hot Population III stars, which only burn for several million years, then explode as hypernovae. These explosions initiate the formation of galaxies. The hydrogen between the stars becomes ionized and absorbs less light, space is easier to see through, so this is known as "The Reionization Era"

The hydrogen clouds absorb so much light, that between z=30 and 1011 is "The Dark Ages"
The z value goes up exponentially, and the next value is:

z=1100  The Cosmic Microwave Background (CMB): the formation of hydrogen
This is nearly at the edge of our space and time.
The  UV light, which was produced about 300,000 years after the Big Bang by the formation of hydrogen atoms from protons and electrons, is now stretched out into 5cm radio waves.
The CMB comes from more than 13,700 million light years away.
see Wikipedia redshift

z=∞  "Big Bang"
Current calculated age of the universe is 13,730 million years old (13.73 billion)
Therefore, the limit for Earth's Universe that we can "see" would be 13,730 million light years away.

If you wish to explore further than that (and get a headache), see wikipedia for Multiverse.

« Last Edit: December 27, 2011, 07:25:50 am by Budgieye »


  • Hero Member
  • *****
  • Posts: 3860
    • View Profile
Spectral charts, for Newbies  Part 7  A spectral analysis of  6 round green things
These 6 objects look similar in the pictures, but have amazingly different spectra.
The objects cannot be identified without a spectrum.

Reminder - they all look "green", but SDSS presents red objects as green.
The light we see from these objects is in the red part of the spectrum at 6000-7000 Å.


587729160052736213   red.............near IR
the redshift is z=0, so it is a star in our galaxy.
The spectrum shows a gradual rise to green, yellow and red, so it must be a yellow star.
posted by starry nite on the green star thread

587725490523340876   red.............near IR
the redshift is z=0 so it is a star in our galaxy
It is a white dwarf, but it is a rare exception since it has cooled down since it was was created.
Most of its light in not in the UV, as is normal for a white dwarf, but in green
posted by mitch and seconded by Tsering

587724197202428120   red.............near IR
This has a redshift of z=0.2953, so it is a distant galaxy 3,300 million light years away (3 billion).
There is a large peak at 6500 Å, in the red part of the spectrum (presented as green by SDSS), made by excited oxygen OIII.
It has an active nucleus producing a hydrogen alpha peak, so it is an active galaxy.
The rest of the galaxy contributes little to the spectrum, the other values are near zero.
This is an OIII galaxy, a type of active galaxy, nicknamed a "pea"

"Green" star vs. OIII pea size and color comparison without a spectrum

587732701783589053   red.............near IR
The redshift is 0.476.  It is a distant and ageing starburst galaxy
The blue starburst emissions are redshifted into red light, and absorption lines show cooling has started.
posted by Hrundi
more information from NGC3314 here

588297863115440297   red.............near IR
The redshift is 1.99
It is probably a type of quasar containing iron, a FeloBAL
posted by mitch

587734621640261792   red.............near IR
The redshift is z=3.9355, so it shines to us from across the universe, 12,000 million (12 billion) light years away
We cannot see visible light from normal galaxies from so far away, so this must be a quasar.
The SDSS computer has printed QSO is printed at the bottom. This means "quasi-stellar object"
UV light from the quasar has been redshifted into red light, making the Lyman alpha peak at 6100 Å. The red light is presented to us as green by SDSS.
posted by StephenT

« Last Edit: August 17, 2010, 08:04:01 am by Budgieye »


  • Global Moderator
  • Hero Member
  • **
  • Posts: 31782
    • View Profile
Re: Zooite Guide to Spectra
« Reply #7 on: April 06, 2009, 11:57:44 am »
Links to spectra-related threads:

Tutorial bits on galaxy spectra: By NGC3314, one of our astronomers, and a great discussion place.

How absorption spectra were discovered: a historical Object of the Day.

How emission spectra were discovered: a historical Object of the Day.

Newbies, post your spectrum questions here: Come and ask in this thread!
« Last Edit: September 24, 2009, 11:56:05 am by Alice »


  • Galaxy Zoo Keeper
  • Admin
  • Hero Member
  • *****
  • Posts: 2020
    • View Profile
Re: Zooite Guide to Spectra
« Reply #8 on: June 30, 2009, 03:32:00 pm »
Stickied. Simply amazing!