Download The 2008 RBSE Journal - National Optical Astronomy Observatory

The
2008
RBSE
Journal
The RBSE Journal 2008
The RBSE Journal is an annual on-line publication that presents the research of students
and teachers who have participated in the Research Based Science Education program
RBSE, at the National Optical Astronomy Observatory in Tucson. This program consists
of a distance learning course and a summer workshop for high school teachers interested
in incorporating research within their class and school. RBSE brings the research
experience to the classroom with datasets, materials, support and mentors during the
academic year. The journal publishes papers that make use of data from the RBSE
program, or from its related programs: TOP, New Mexico Skies and the SPITZER
teacher observing program.
These papers represent a select set of those submitted for publication by students and
occasionally RBSE teachers. All papers are reviewed both by the Editor and the
Astronomer responsible for the particular research project. This year we wish to
acknowledge editing help from other astronomers with special expertise in topics that are
outside the RBSE projects. More information about both the RBSE and the TOP
program can be found on our website, www.noao.edu/ education/arbse
I want to thank Dr. Travis Rector, Dr. Greg Rudnick and Dr. Joel Parker for their
generous help in reviewing these papers and working with the young scientists. Special
thanks are due to Kathie Coil for her efficient editing of the final copy.
Dr. Katy Garmany
Editor, RBSE Journal
The RBSE Journal
ii
2008 V2
Table of Contents
AGN
Radio and Starbust Galaxies ................................................................................................................. 4
Alexandra JW Echtenkamp, Andrew S. Bowles, Anthony J. Sinker
Breck School, Minneapolis, MN
Teacher: Chelen H. Johnson, ARBSE 2007
MgII and CIII] in Active Galactic Nuclei...............................................................................................12
Sarah Johnson, Jeffrey Portu, Breanna Heilicher
Breck School, Minneapolis, MN
Teacher: Chelen H. Johnson, ARBSE 2007
Relative Velocities of CIV Winds as an Indicator of Black Hole Mass................................................19
Meghan Dorn
Rush-Henrietta High School, Henrietta, NY
Teacher: Jeffrey Paradis, ARBSE 2007
Population Distribution Analysis of the FIRST Bright Quasar Survey............................................. 24
Tanner Sagouspe and Students of the Astronomy Research Seminar
Central Catholic High School, Modesto, CA
Teacher: Christine Wilde, ARBSE 2007
Finding the CIV Wind in Active Galactic Nuclei................................................................................. 34
Teacher: Chelen H. Johnson, Breck School, Minneapolis, MN–ARBSE 2007
Teacher: Javier Melendez, Brophy College Prep School, Phoenix, AZ–ARBSE 2007
Teacher: Jeffrey Paradis, Rush Henrietta Senior High School, Rochester, NY–ARBSE 2007
Teacher: Thomas F. Sumrall, Forrest County AHS, Brooklyn, MS–ARBSE 2007
Teacher: Christine Wilde, Central Catholic High School, Modesto, CA–ABSE 2007
Teacher: Lynne F. Zielinski, Glenbrook North High School, Northbrook, IL–ARBSE 2007
Clusters
NGC 2367: Its Age and Distance.......................................................................................................... 44
Caitlin S. Colley
Sullivan South High School, Kingsport, TN
Teacher: Thomas Rutherford, TLRBSE 2005
A Search for Exoplanets in NGC 957 ................................................................................................... 52
Bobby Adams, Rebecca Redmon, Veronica Buehrig
Sullivan South High School, Kingsport, TN
Teacher: Thomas Rutherford, TLRBSE 2005
Minor Planets
The Short-term Variation of the Transneptunian Binary Objects (42355) Typhon and (90482)
Orcus ........................................................................................................................................................68
Rebecca Jensen-Clem1 and Jacob Shenker
International Community School, Kirkland, WA
2
Gunn High School, Palo Alto, CA
Teacher: Don McCarthy, RBSE 1996-TLRBSE 2002
The RBSE Journal
iii
2008 V2
Radio and Starburst Galaxies
Alexandra JW Echtenkamp, Andrew S. Bowles, Anthony J. Sinker
Breck School, Minneapolis, MN
Teacher: Chelen H. Johnson, ARBSE 2007
ABSTRACT
Active Galactic Nuclei (AGNs) are compact regions in the center of a galaxy that can
produce more radiation than the rest of the galaxy. However, in the case of radio and
starburst galaxies, AGN do not produce more radiation than the rest of the galaxy. Many
different types of AGN can be formed in many different types of galaxies.(1) The three
strongest sources are radio galaxies, quasars, and blazars. The main focus of research will
be radio galaxies from Right Ascension 07h 18m to 09h 43m as part of the FIRST Bright
Quasar Survey at the Very Large Array with the optical spectra obtained with the Kitt
Peak 2.1-meter telescope. The galaxies were discovered by the Faint Images of the Radio
Sky at Twenty- centimeters using a radio telescope in New Mexico.(2) Research will
consist of looking at all the different types of galaxies within a certain area. From there
the focus will turn to all the identified radio galaxies and starburst galaxies. By looking at
the elements and emission lines in the graphs we were able to determine which were
radio and starburst, and which were not. From there we used the lines to calculate the
redshifts, velocity, and distance. Then we compared and contrasted the characteristics of
the radio galaxies to the characteristics of starburst galaxies.
INTRODUCTION
Radio galaxies are for the most part found in elliptical galaxies. They were discovered in
the 1940s when radio telescopes were used to scan the sky.(3) The radio galaxies are jet
structured meaning they have jets, two lobes, counter jets, streams of electron-filled gas
aimed in different directions from a black hole at the center of a galaxy(4), and hot spots.
Between the lobes is the host galaxy, which is connected by jets. Jets are very important
because they trace the path of material that is ejected from the active galactic nucleus and
into the lobes. One jet is brighter and the lighter jet is the counterjet.(5) This structure
makes the radio galaxies somewhat symmetrical.(6)
The RBSE Journal
4
2008 V2
Figure 1. Parts of a Radio Galaxy.(4)
A Double Radio Source Associated with a Galactic Nucleus (DRAGN) is a radio source
that is produced by jets produced by active galactic nucleus that is not in the Milky Way.
This happens when an accretion disk forms around a black hole and spins, converts
gravitational and rotational energy into excess perpendicular to the disk. Although
DRAGNs are found in starburst galaxies, which produce radio emission lines and are
mainly formed in galaxies that are larger than their host galaxies, such as elliptical
galaxies.(4) They are comprised of lobes, jets, and a core just as other radio galaxies are,
and they also have hot spots in the lobes.
There are two types of radio galaxies, Franaroff-Riley type I (FR I) and type II (FR II).
While the two groups share similar properties, such as their size, they have different UV
properties, infrared properties, kinematics, and host galaxies. FR I are either old or they
don’t have enough material or energy to form stars. While they can no longer form new
material, they are the most evolved of the radio galaxies. The FR II galaxies have higher
redshifts but are less evolved; due to this they are richer groups, meaning that there are
fewer things around the galaxy.(7)
To understand exactly what is researched, the basic knowledge of radio and starburst
galaxies must be understood. A radio galaxy is formed when an AGN produces two
persistent, oppositely directed plasma outflows. The outflows are what will soon to
become the jets of the galaxy. While it is not known exactly what is inside the jets, they
have fast moving electrons and magnetic fields, which make the high radio frequencies.
The emission that occurs moves almost the velocity of sound. The jets are formed
through the winding up of magnetic fields, which create a black hole in the nucleus of the
galaxy. The winding of the black hole converts the energy from the magnetic field into
mass. This initial winding is supersonic, meaning it is faster than the speed of sound.(4)
From the formation stage, the radio galaxy goes through the developmental stage. In this
stage the galaxy grows as the jets stretch from the atmosphere or the AGNs, through the
interstellar medium of host galaxies, lower densities and pressures of the outer halo of the
The RBSE Journal
5
2008 V2
galaxy, inter-galactic medium in surrounding galaxies, and finally to the low-density
intergalactic medium. This growth of the galaxy and stretch of the jets extend outward
and usually end up being bigger than the originating galaxy. The smallest known are only
a few tens of parsecs across, while the largest are known to be up to several megaparsecs.
The average radio galaxy is usually typically hundreds of kiloparsecs across. This is
about twice the size of the Milky Way galaxy. The average life span of a radio galaxy is
20 million years.(4)
Starburst galaxies, the other type of galaxy being studied, are thought to be formed by
close encounters or collisions of other galaxies. These collisions send a shock wave
throughout the galaxy; pushing giant clouds of dust and gas, making them collapse and
form hundreds of massive stars. These massive stars use up their fuel very quickly
causing supernovas, which create more collisions, thus creating more stars. Starbursts are
the most luminous galaxies and are thousands of light years in diameter.(8)
Figure 2. A Starburst Galaxy.(8)
The star formations within the galaxy that ends up creating most of the stars are known as
ultra-luminous clusters. They are about 10-20 light-years across and can have
luminosities up to 100 million times that of the Sun. These clusters are the densest starforming environments known. The thing that sets starburst galaxies apart from the rest is
their high, intense emission lines in the far-infrared. These lines are created by the
ultraviolet that is emitted by the numerous hot stars being formed. These young stars are
absorbed by the dust and remitted with higher wavelengths. These wavelengths rate
second only to AGNs themselves.(9) While we know that starbursts last much less than
the age of the universe, it is very difficult to estimate their age because new clusters are
always being formed. This creates the starbursts extreme luminosity making it hard to see
the older parts of the galaxy.(10)
OBSERVATIONS AND DATA REDUCTION
For our project, we looked at radio and starburst galaxies and then compared and
contrasted them. We looked at the galaxies in the Right Ascension range of 07h 01m to
09h 59m in the FIRST Bright Quasar Survey at the Very Large Array with the optical
spectra obtained with the Kitt Peak 2.1-meter telescope. When we recognized either a
starburst or radio galaxy by its graph, we used the galaxy. Since each galaxy has its own
The RBSE Journal
6
2008 V2
graph, we were able to estimate whether the graph represented either a starburst or radio
galaxy. We disregarded all the galaxies that were clearly not recognized as either radio or
starburst galaxies. After amassing a reasonable amount of galaxies (about 70), we
calculated the ratios between the prominent emission lines.
To find these ratios we divided the higher wavelength by the shorter one. We had to find
the ratios of all the emission lines in order to discover which elements were responsible
for the emission lines. The ratio was then compared to the ones posted on the “AGN
Spectroscopy” packet on page 20. When a ratio in the packet that matched the calculated
ratio was found, we knew that the elements forming the posted ratio were the elements
responsible for the two emission lines that made up our calculated ratio. Based on the
placement of [OIII], Hα, and Hβ lines, we were able to differentiate between starburst
and radio galaxies. Radio galaxies have a small [OIII] line before a large [OII] line.
Starburst galaxies have a large [OII] line preceding a small [OIII] line. These facts helped
us determine which galaxies appeared to be starburst or radio galaxies and then helped us
disregard the other galaxies. After the elements responsible for the emission lines were
found, redshifts were calculated.
To find the redshifts, wavelengths for the elements that were commonly found in AGN
were taken from a list also on page 20 of the “AGN Spectroscopy” packet. Then the
wavelengths were plugged into a formula for redshifts and the redshift was found. The
formula for a redshift is 1+ z =
λobs
where λobs is the emission line that was found, λrest is
λrest
the wavelength associated with the elements responsible for the emission line and z is the
redshift. Since there were more than two emission lines found in most of the AGN, more
than one redshift was calculated for each AGN. In this case, the redshifts were averaged.
These averaged values, or original values if there were only two emission lines found,
were the redshifts used to calculate other characteristics of the galaxies, such as velocity
(1+ z) 2 −1
where v is velocity, c is the
and distance. The formula for velocity is v = c
(1− z) 2 + 1
speed of light (3.0x105 km sec-1), and z is the redshift.
cz(1.+ 0.5z)
. In this equation d is the distance, c is the
H 0 (1+ z)
speed of light, z is the redshift and Ho is Hubble’s constant (75km sec-1 Mpc-1). After all
the calculations were made, radio galaxies and starburst galaxies were compared to one
another.
Distance is calculated by d =
ANALYSIS AND RESULTS
For all the galaxies that we observed, both radio and starburst, the relationship between
velocity and distance was linear, which Hubble’s Law confirms. The vast majority of the
galaxies we observed were of fairly low velocity. The galaxies of further distance/greater
velocity seemed to be outliers in the data set. This is because Hubble’s Law only works
for low-redshift objects. Our comparison of sky location yielded less conclusive results;
The RBSE Journal
7
2008 V2
the data seemed to be scattered somewhat randomly. One observable pattern was that the
starburst galaxies were all concentrated in the right ascension range between 8 and 9, and
the radio galaxies were all located between 7 and 10. Generally speaking, however, there
was a uniform lack of pattern among all observed galaxies. (Figure 4) Nearly every
galaxy we observed had a redshift between 0 and 1; the only exceptions were two radio
galaxies in the 2-3 range and one starburst galaxy between 4 and 5. (Figure 3)
Figure 3: Redshifts of AGN
The RBSE Journal
8
2008 V2
Figure 4. Location of AGN.
Figure 5. Velocities and Redshifts of AGN.
The RBSE Journal
9
2008 V2
Figure 6. Distances and Redshifts of AGN.
DISCUSSION
As demonstrated by our results, radio galaxies and starbursts are similar in nearly all
observable characteristics. While the elements responsible for their respective emission
lines varied, the galaxies we observed had few other distinguishing characteristics.
For future projects, more characteristics for galaxies could be calculated. For example,
luminosity was not found in this project. When we put checked to see if our AGN were
seen with an optical telescope using the SIMBAD website, none of them were found
because nobody has studied the galaxies in this research project. It would be interesting
to see if there is a difference between what we found and the optical telescope. We could
also use the Sloan Digital Sky Survey/ Sky Server to enter in the galaxies that were
studied. From there we could find the recognized galaxies and find further characteristics
as shown by the information on the survey. Another thing we could do is compare and
contrast other types of galaxies besides Radio and Starburst. We could look at Quasars,
BL Lacs, and Elliptical galaxies and get a further understanding of how they function
compared to the already analyzed Starburst and Radio galaxies.
The RBSE Journal
10
2008 V2
REFERENCES
NASA’s HEASARC: Education and Public Information. 6 April 2006. NASA. 2
December 2007. <http://heasarc.gsfc.nasa.gov/docs/objects/agn/agntext.html>
Rector, Travis and Wolpa, Brenda. “AGN Spectroscopy Nature’s Most Powerful
‘Monsters’” Teacher Leaders. The National Optical Astronomy Observatory. 4 April
2004:1-32.
Newman, Phil. “Active Galaxies and Quasars.” NASA Goddard Space Flight Center.
NASA. 6 October, 2006. 2 December 2007.
<http://imagine.gsfc.nasa.gov/docs/science/know_l2/active_galaxies.html>
Bridle Alan. “Double Radio Sources Associated with Galactic Nuclei”. NRAO
Charlottesville. 28 June 2006. AstroWeb. 3 December 2007,
<http://www.cv.nrao.edu/~abridle/dragnparts.htm>
Jones, Mark H., Lambourne, Robert J. An Introduction to Galaxies and Cosmology.
Cambridge.
“Radio Galaxies”. 11 February 2007. UT Astrophysics.
<http://csep10.phys.utk.edu/astr162/lect/active/radio.html>
Zirbel Esther. “The Megaparsec Environments of Radio Galaxies.” The Astrophysical
Journal. 20 February 1997: 476:489-509
Starburst Galaxies. 29 August 2006. Chandra X-Ray Observatory. 13 December 2007.
<http://chandra.harvard.edu/xray_sources/starburst.html>
Starburst Galaxy. 2007. The Internet Encyclopedia of Science. 13 December 2007.
<http://www.daviddarling.info/encyclopedia/S/starburst_galaxy.html>
Meurer, Gerhardt R. “Galaxy Evolution.” Extragalactic Astronomy. 2007. Johns Hopkins
University. 13 December 2007.
< http://www.pha.jhu.edu/~meurer/research.html>
Smith Gene. “A Bestiary of Active Galaxies”. 29 September 200. University of
California. 3 December 2007. <http://casswww.ucsd.edu/public/tutorial/AGN.html>
Keel Bill. “Radio Structure in Radio Galaxies”. November 2002. Astr.edu. 3 December
2007. <http://www.astr.ua.edu/keel/agn/vlamaps.html>
Longair M. “Three Radio Galaxies”. NASA. Students for the Exploration and
Development of Space. 3 December 2007. <http://seds.lpl.arizona.edu/hst/3Cgalax.html>
Peratt A. L. 'The Evidence for Electrical Currents in Cosmic Plasma', , IEEE Trans.
Plasma Sci., Vol.18, pp.26-32, 1990.
The RBSE Journal
11
2008 V2
MgII and CIII] in Active Galactic Nuclei
Sarah Johnson, Jeffrey Portu, Breanna Heilicher
Breck School, Minneapolis, MN
Teacher: Chelen H. Johnson, ARBSE 2007
ABSTRACT
This project was aimed at determining the velocity of two component chemicals, MgII
and CIII], in the winds associated with active galactic nuclei (AGNs). We classified
objects based on a thorough examination of spectra of AGNs in the region centered on 07
hours of Right Ascension. Focusing on the MgII and CIII] emission lines in the identified
quasars, we were able to calculate the redshift, relative velocity, and distance of each
element. We are confident that our MgII and CIII] lines are correctly identified: Earth’s
atmosphere cannot produce CIII] or MgII lines, and our MgII lines were consistent
through multiple checks, and were found at predictable wavelengths. Furthermore, since
the lines are redshifted, there is no concern of them being misidentified. MgII has lost an
outer electron shell, while CIII] has lost an inner electron which takes more energy to pull
away. MgII is farther away from the black hole than CIII], thus it is colder and has a
slower velocity. We assumed that the MgII velocity was the quasar’s velocity and the
CIII] velocity was the velocity of the wind. The wind velocity was faster than the actual
velocity of the quasar. Our research showed results in the quasars with MgII and CIII]
lines that were similar throughout all of our data and demonstrated that there is an
indirect relationship between distance from the central black hole and wind velocity.
INTRODUCTION
An Active Galactic Nucleus (AGN) is all of the matter that surrounds a supermassive
black hole. These galactic giants of the Universe are mostly unknown. There are four
different types of AGNs, each with different characteristics. Radio Galaxies are galaxies
that appear as normal galaxies in the optical spectrum, but they emit massive amounts of
radio waves (1). Starburst galaxies are known for forming stars at a very fast rate. Many
times this is caused by a gravitational interaction with another galaxy. BL Lacertae
objects are known for having very weak emission lines, unlike quasars or radio galaxies.
Spiral galaxies are the oldest type of galaxy; they consist of about a hundred billion stars,
which are formed in HII regions and have characteristic arms that spiral out of the center.
For our study, we will disregard the spectra for radio, starburst and BL Lacertae objects
and focus on quasars.
Spectral emission lines can help to differentiate the different types of AGNs.
Spectroscopy allows us to study what types of light we see from an object; it is arguably
the most important aspect of astronomy. Through spectroscopy we can determine the
temperature, velocity, and composition of the object being examined. In terms of
velocity, spectroscopy can be used to determine an object’s velocity toward or away from
us via the Doppler effect. Without spectroscopy, it would be very hard, if not impossible
for astronomers to tell what all of the lights in the sky are.
The RBSE Journal
12
2008 V2
Figure 1. Visible spectrum of Sky with spectroscopy. (1)
To find the shift of an object, the equation λobs = (1+ z) λrest is used, where λobs is the
observed wavelength, and z is the redshift of the object. If z is positive, then there is a
redshift, and if z is negative, a blueshift. A redshift is seen when the object is moving
away from us in the Universe, and a blueshift is seen when something is moving toward
us in the Universe. Galaxies that are very far away move much faster than those that are
close to our galaxy. For nearby objects, Hubble’s Law, v = H 0 d , where v is the object’s
velocity, H0 = 75 km/s/Mpc (the Hubble constant) and d is the distance, is used to
determine the distance of an AGN. The objects in our study are so distant that a more
advanced model, such as the “empty Universe” model needs to be used to calculate
distance which is important because it provides a third dimension to something that we
can only see in two dimensions.
The winds that come off an AGN are crucial in determining the velocity of the AGN, and
can be seen in Figure 2. There is some speculation of a connection between the winds,
especially the CIII] and MgII. To find the velocity of that wind, the equation
(1+ z) 2 −1
is used, where v is the velocity, and z is the redshift.
v =c
(1+ z) 2 + 1
The RBSE Journal
13
2008 V2
Figure 2. Diagram of a typical AGN. (2)
In our project, we examined AGNs specifically near Right Ascension 07 hours. The data
we are using will be from the FIRST Bright Quasar Survey taken at the Very Large Array
with the optical spectra obtained with the Kitt Peak 2.1-meter telescope. The FIRST
Survey is an acronym for “Faint mages of the Radio Sky at Twenty-centimeters”.
Through the analysis of the quasar’s MgII and CIII] emission lines, we determined the
quasar’s velocity, as shown in Figure 2. By analyzing the winds and where they occur,
we were able to see what direction the AGN is moving in and other characteristics of that
region of the sky. The velocity relates to the distance from the continuum source.
The difference in the velocities of the MgII and CIII] lines in the AGN spectrum indicate
the velocity of the wind as the object is moving. MgII and CIII] lines don’t change based
on high or low states of the spectrum, and both are fairly broad lines. In broad-line
profiles, each velocity has a range of ionization, but there is a “preference for higher
ionization at higher velocities” (4). Although MgII and CIII] are low-ionization and lowdensity atoms, CIII] is more highly ionized than MgII. High and low states of AGN cause
ionization radii to expand or contract, and can be caused by changes in the continuum
source radius or velocity. “Peaks in the emission-line profiles can be transient responses
to variation of the continuum source. Averaging over time, there is equal flux from the
blue and red sides of the profile, but the blue side is more likely to have a prominent
narrow peak” (3). The terminal velocity of the wind is determined by the width of the
emission lines and the ionization state.
High ionization lines are produced from an outflowing wind, while low ionization lines
The RBSE Journal
14
2008 V2
tend to come through a thick accretion disk. (3) Ionization increases as atomic radius
increases and the electron shells expand, thus increasing the velocity. Near the source of
the emission, there is strong absorption and minimal velocity; however, for electron
shells with an increasing radius, the width of the emission lines increase and absorption
becomes weaker. “Velocity offset is largest for ions with the highest ionization” (3)
because there is more obscuration, which is directly related to velocity (2, 3).
The quasars in our survey have a spectrum similar to that in Figure 3. (1) However, our
spectra are not as clean as that shown due to the fact that it is a composite of over 100
quasars that have been averaged together. As illustrated, the CIII] and MgII lines stick
out at distinct points and can be easily recognized. By determining the redshift or
blueshift and its relation to the wind velocity of CIII] and MgII winds of some of the
quasars in our region of the sky, we are hoping to be able to give more insight into the
relatively unknown realm of AGNs.
Figure 3. Typical Quasar Spectra. (1)
ANALYSIS AND RESULTS
The purpose of our research was to determine how different elements, specifically MgII
and CIII], and their distance from the black hole affects the velocity of their winds.
Energy from the accretion disks around the black hole in the center of the galaxy heats up
the dust and galaxy, and creating emission lines that, if the galaxy is near enough we can
see. The farther away from the black hole, the more slowly the clouds and dust move, and
the colder they are. The torus circles the accretion disk, and is cold enough to see infrared
radiation, however, it is too thick to see through, and can obscure the view.
The RBSE Journal
15
2008 V2
Table 1. Redshift, Velocity and Distance.
location
bq0713p3656
bq0719p3307
bq0721p2227
bq0722p2856
bq0722p2941
bq0724p2517
bq0726p3019
bq0733p4555.2
bq0758p3222
bq0803p2727
bq0809p4134
bq0810p5025
bq0828p4922
bq0835p4826
bq0836p3455
bq0846p2502
MgII
redshift
1.59
1.37
1.63
1.21
1.2
1.23
1.29
1.18
1.2
1.2
1.22
1.19
1.53
1.37
1.39
1.41
CIII]
redshift
1.67
1.63
1.62
1.51
1.79
1.45
1.05
1.04
1.21
1.22
1.19
1.2
1.49
1.4
1.46
1.44
MgII
velocity
(km/s)
222160
209323
224213
198030
197260
199546
203909
195696
197260
197260
198792
196482
218929
209323
210609
211870
CIII]
Velocity
(km/s)
226189
224213
223707
217809
231695
214316
184671
183757
198030
198792
196482
197260
216668
211243
214913
213714
Relative
Velocity
(km/s)
-4029
-14890
506
-19779
-34434
-14770
19238
11939
-770
-1532
2310
-778
2261
-1920
-4304
-1844
distance
(Mpc)
-8056938
-29777326
1013855
-39556120
-68866985
-29538063
38478918
23879470
-1538019
-3061959
4622273
-1554311
4523909
-3837012
-8605458
-3686149
Relative
Distance
(Mpc)
8056938
29777326
1013855
39556120
68866985
29538063
38478918
23879470
1538019
3061959
4622273
1554311
4523909
3837012
8605458
3686149
In the region of the sky around 07 hours of Right Ascension, we examined 231 AGNs,
24% of which were quasars (see Figure 4. Classification of AGN). However, we
disregarded spectra of quasars that did not meet our requirements of distinct MgII and
CIII] emission lines. We are confident that the CIII] lines are from AGNs because Earth
will never produce CIII] emission lines. Of the 231 spectra, we were able to successfully
analyze 16 (7%) of the spectra (see Figure 5. Classification of AGN with MgII and CIII]
Lines). Of these quasars, seven of the 16 (44%) had relative velocities within -5000 km/s
to -1 km/s (see Figure 6. Histogram of Relative Velocities). This means that the quasar’s
wind is moving toward us at a relatively slow speed.
Figure 4. Classification of Selected AGN.
The RBSE Journal
16
2008 V2
Figure 5. Classification of AGN with Distinct MgII and CIII] Emission Lines.
Figure 6. Histogram of Relative Velocities.
DISCUSSION
The ionization state of each atom affects its velocity and distance from the black hole.
MgII, which is farther away from the black hole than CIII] has lost an outer shell
electron. Because inner electrons take more energy to move, CIII] is a hotter gas, which
The RBSE Journal
17
2008 V2
results in its higher velocity than the MgII (see Table 1. Redshift, Velocity and Distance).
From the positive velocities of the MgII and CIII] lines, we can determine that both the
MgII and CIII] emission lines were redshifted and are in the part of the galaxy that is
moving away from us. MgII is farther away from the black hole than CIII], therefore it
also has a slower velocity. Due to this difference in distance, we assumed that the CIII]
velocity is the velocity of the gas, while the MgII is the quasar velocity because of its
large distance from the black hole.
Further projects could compare the MgII and CIII] lines with CIV emission lines,
examining whether the CIV affects the MgII and CIII]. The difference between
luminosity and distance could also be studied.
REFERENCES
Rector, Travis A. and Brenda A. Wolpa. “AGN Spectroscopy: Nature’s Most Powerful
‘Monsters.’” The National Optical Astronomy Observatory 4 April, 2004.
Johnson, Chelen H. AGN Spectroscopy: Studying the Monsters of Astronomy. 2007.
Hutchings, J.B., G.A. Kriss, R.F. Green, M. Brotherton, M.E. Kaiser, A.P.
Koratkar, and W. Zheng. “Evidence for an Accelerating Wind as the Broad-Line Region
in NGC 3516.” The Astrophysical Journal 559:173-180, 20 September 2001. Breck
Upper School Lib. 3 Dec. 2007
<http://www.journals.uchicago.edu/doi/abs/10.1086/322329>.
Tytler, David and Xiao-Ming Fan. “Systematic QSO Emission-Line Velocity Shifts and
New Unbiased Redshifts.” The Astrophysical Journal 79:1-36, March 1992. Breck Upper
School Lib. 3 Dec. 2007.
The RBSE Journal
18
2008 V2
Relative Velocities of CIV Winds as an Indicator of Black Hole Mass
Meghan Dorn.
Rush-Henrietta High School, Henrietta, NY
Teacher: Jeffrey Paradis, ARBSE 2007
ABSTRACT
Producing almost 1/5 of the energy in the universe, quasars are of great importance to
astronomers (Elvis, 2003). I present spectra of 62 selected quasars and the relative
velocities of particles around them, in or near the torus. Hypothesizing that differences
between relative velocities of elements MgII and CIV are due to the individual size of the
black hole, based on the evidence, I have concluded that this relationship does not exist.
Further study is proposed based on multiple contributing factors within the structure of
the quasar that may have implications on the relative velocities of particles.
INTRODUCTION
Active galaxies are energetic galaxies that emit thousands of times more energy per
second than the Milky Way. Radiation emitted from the black hole in the center is mostly
long-wavelength and non-thermal (Chaisson, 2005). The nucleus of an active galaxy
produces more radiation than the rest of the galaxy and is solely gravitationally driven.
Their observed redshifts indicate extremely large distances from Earth. Discovered in
1963, the currently accepted model is an accretion disk surrounding a super massive
black hole (SMBH). These SMBHs are millions or billions of times more massive than
our Sun. In addition, the luminosity of a quasar is billions of times that of the Sun, often
on the magnitude of 10^38 to 10^42 Watts1. When magnetic field lines, which are
produced by the rotation of the accretion disk, interact with the accretion disk itself, they
release energy that powers the jets that emerge from the center of the quasar. The jets are
perpendicular to the plane of the accretion disk feeding the SMBH and contain gamma
rays and x-rays. The radiation emitted is electromagnetic and mostly in the form of
synchrotron radiation.
ANALYSIS AND RESULTS
My data consists of 62 quasars that are from a database of 1300 galactic spectra taken
from the VLA FIRST Survey (Faint Images of the Radio Sky at Twenty-Centimeters).
This data was obtained from Dr. Travis Rector of the
1
The Sun’s luminosity is just 3.8x10^26W.
University of Alaska. The selection consists of quasars that exhibit carbon IV (CIV),
carbon III (CIII), and magnesium II (MgII) emission lines. The chosen emission lines are
broad and not difficult to pick out of the spectra. By using a small redshift value, these
two elements were easier to pick out in lower redshifted galaxies. I intended to look at
interactions of energy emitted from the quasar jets, with dust in their path. Focusing on
the redshifts of these emission lines, I determined the relative velocity of the CIV line in
relation to the MgII. The significance of the relative velocity is to show the direction of
the movement of the particles being compared. This does not indicate the shift of the
The RBSE Journal
19
2008 V2
entire galaxy, which is no doubt redshifted, but rather the cone of particles interacting
with the jets that are visible. Relative motion of the excited particles could be toward the
observer or away, indicating a blue or red shift. The equation for the redshift of each
emission line is as follows, where z is redshift and lambda means the wavelength. Large
differences in redshifts were ignored due to possible observational error. Values outside
of what is generally expected for these velocities suggest that the measured or calculated
values are inaccurate. A typical quasar spectrum is shown in figure 1.
1+ z =
λobs
λrest
Observed redshift is due to the Doppler Effect which occurs when an object is moving
away from the observer. Using the following equation, the difference in redshift (zMgIIzCIII) of each of the two emission lines calculated the relative velocity of those particles.
This was repeated for each of the quasars, producing a large range of velocities. Z is the
value of the redshifts and c, the speed of light, is a constant (c= 3x105 km s-1).
(1 + z ) 2 − 1
v=c
(1 + z ) 2 + 1
Of all the galactic spectra on file, the 62 quasars selected, all of them were chosen
because they exhibited a similar redshift between the CIII and MgII lines, but the CIV
was slightly askew, resulting in the relative velocity. The quasars needed to have visible
emission lines on the spectra, or there would be nothing to measure. After most spectra
was rejected for this investigation, spectra that had an unusually large difference in
relative velocities were also taken out of the selection. See appendix A for data.
The RBSE Journal
20
2008 V2
Figure 1
Source: Rector, Travis A. and Brenda A. Wolpa. “AGN Spectroscopy: Natures Most
Powerful ‘Monsters’” ARBSE, 1-31
DISCUSSION
To explain this large range, I hypothesized that a quasar having a larger black hole would
have a higher velocity of gas and dust moving away from its jets. From the FIRST
Survey, I retrieved cutouts2 of each quasar that gave me the radio brightness of whichever
particular quasar in mJy/beam.3 That data could then be translated into luminosity using
the inverse square law:
B =
L
4π d 2
The luminosity was then plotted against ( figure 2) the relative velocity to see if there was
any correlation. From Appendix A, the log of the luminosity and relative velocity was
taken in order to scale the data.
The RBSE Journal
21
2008 V2
Figure 2
Luminosity vs Relative Velocity
141
140
139
138
137
136
135
134
133
0
2
4
6
8
10
12
Relative Velocity
Based on the graph no correlation was found whatsoever between the relative velocity
and the black hole, but there is still promising work to be done. The theory of the
Eddington limit proposes that when there is radiation being emitted such as in a jet from a
quasar, there is an outward force of luminosity in addition to the gravity pulling inward
(Heinzeller and Duschl, 2007). Luminosity of quasars would not only be dependent on
2
3
A cutout is a picture and known information about that particular quasar.
These units are millijanskies which are a unit of the flux.
the size of the black hole but also the amount of matter being accreted onto it. My
hypothesis did not take into account how much matter was being accreted onto the black
holes in question, only on the size itself. A smaller black hole could exhibit higher
velocities and greater luminosity if more were being accreted onto it based on the density
of the material near the black hole. It would be sensible that a large black hole would
produce greater velocities, but if the case were such that it had a small amount of matter
being accreted onto that larger black hole, it would not. With two variables, the relation
of the size of the black hole to the relative velocity of the quasar would be very loose if
the second variable was not also taken into account.
A problem that occurred with my data was that some of the relative velocities of the
quasars appeared to be blueshifted, but there is no evidence yet discovered that blueshift
has been observed in a quasar. This is true, due to the fact that the universe is expanding
and all quasars would be moving away from us, but of the 62 selected quasars, those that
show a positive relative velocity indicate a slightly lower redshift in CIV compared to the
CIII and MgII. I propose that we are not viewing a blueshift in the quasar itself, but it is a
possibility that the particle with the lower redshift than the other, producing a small
difference in those redshifts, is moving towards us in the jet that is in our line of sight.
This is indicative that gas from the quasar is moving toward us, rather than the quasar
itself.
The RBSE Journal
22
2008 V2
An explanation for the negative and positive velocities would be the location of the
particles that are interacting with energy from the quasar. If a particle were to be closer to
the jets, it stands to reason that it would be moving faster than a particle farther away
from the jet. Those particles moving faster would appear to have a larger redshift. So,
when the redshifts subtracted and the relative velocity determined to be positive, if the
particles were in another case in opposite locations, the difference would cause the
relative velocity to be negative, but quasars are a point source so it would be near
impossible to see which regions individual particles would be in. A more acceptable
conclusion would be that the viewer was seeing redshifted particles on one side of the
quasar moving away from us, but the particle that appeared less redshifted would be on
the other side, interacting with the jet that is moving towards us, while the quasar was,
overall, moving away.
SUMMARY AND ACKNOWLEDGEMENTS-- My hypothesis that a larger flux
would produce a higher relative velocity was proven to be inaccurate, but there is still
promising work to be done. Further ideas to explore would be to measure the impact of
the shape and angle of the jets, luminosity, and amount of matter accreting onto the black
hole all as variables that determine the velocities of the matter around the jets. Martin
Elvis of the Harvard-Smithsonian Center for Astrophysics has begun to address the
impact of the shape of the cone created by the magnetic field lines interacting with the
energy from the quasar in his article A Structure for Quasars, 2000. He proposes that the
viewing angle dictates the range of velocities that is seen within that particular quasar.
The range of velocities seen within my selection may not be closely related to the size of
the black hole, but may in fact have more to do the structure and geometry of the jets.
This is along the lines of what Elvis is proposing in his article- that the viewing angle of
the quasar will most likely indicate the range of redshifts and therefore velocities seen.
One may go further from the Eddington limit and say that the amount of matter being
accreted onto the black hole in the center of the quasar and the size of that black hole are
only factors that determine the angle of the cones of energy (jets) which would be an
important contributing variable as to the input of what creates the velocities seen.
REFERENCES—
Chaisson, Eric and Steve McMillan. Astronomy Today. New Jersey: Pearson
Education, Inc., 2005
D. Heinzeller and W. J. Duschl. “On the Eddington Limit in Accretion Discs” Monthly
Notices of the Royal Astronomical Society 374 (3), 1146–1154, January 2007
Elvis, Martin. “A Structure for Quasars” The Astrophysical Journal, 545: 63-76, 2000
December 10
The RBSE Journal
23
2008 V2
Elvis, Martin. “Solving Quasars” PowerPoint available online from Fermilab Colloqium
29 October 2003
<http://vmsstreamer1.fnal.gov/VMS_Site_02/Lectures/Colloquium/Elvis/vf001.htm>
Rector, Travis. AGN Spectroscopy: Studying Natures Most Powerful “Monsters.” RBSE
Journal: Tucson, Arizona, 1999
The RBSE Journal
24
2008 V2
Population Distribution Analysis of the FIRST Bright Quasar Survey
Tanner Sagouspe and Students of the Astronomy Research Seminar
Central Catholic High School, Modesto, CA
Teacher: Christine Wilde, ARBSE 2007
ABSTRACT
The students in the Astronomy Research Seminar at Central Catholic High School have
been working to identify the population of objects in the Bright Quasar Survey. We were
given 560 objects to examine; Mrs. Wilde split the class into five main groups that
eventually separated into approximately two groups per primary group. Using Graphical
Analysis on the school computers we worked diligently for three months in opening
assigned columns of Bright Quasar Objects and working together to uncover which
category the object fell under: BL Lac, elliptical galaxy, quasar, radio galaxy, or
starburst. After completing our study of the objects we found that quasars were the most
abundant objects in the night sky with 57%, followed by radio galaxies with 17%,
elliptical galaxies with 13%, starbursts with 8%, and finally with the least, BL Lac
objects with a minuet percentage of 5%. Upon completion of the study we also created a
catalog in which all of the objects studied are listed and placed under their appropriate
category.
INTRODUCTION
Our research question entailed examining the given data and accumulating the number of
various Bright Quasar Objects given in the FIRST Bright Quasar Survey. We hoped to
catalog approximately the first 40 lines of objects, which were 560 objects, and properly
categorize them into the previously given list of possible objects: BL Lac, elliptical
galaxy, quasar, radio galaxy, or starburst. Upon finishing the designated work we
intended to find out what percentage of the night time sky was filled with said amount of
objects. Discussing further into the objects we are able to see the precise differences in
the Bright Quasar Objects. A BL Lac is an active galaxy whose jets are pointed at the
Earth and is in fact one of the two branches of blazar type galaxies. The elliptical
galaxies are one of the three main types of galaxies and are easily characterized by their
elliptical shape. A quasar is an active galaxy in which its jets are not pointed at Earth like
a BL Lac, but in a different direction so they are able to be seen in their entirety.
Starburst galaxies are galaxies in which there are abnormally large numbers of star births
occurring at a given time. They are determined to be starburst galaxies based on the star
formation inside the galaxy in relation to its relative age compared to that of an average
galaxy. Radio galaxies are galaxies that emit large amounts of radio waves from the area
around the super massive black holes at their center.
OBSERVATION AND DATA REDUCTION
The study of the Bright Quasar Objects began once we received the data we were chosen
to examine. Opening the graphs with Graphical Analysis on school computers we began
by splitting into our assigned groups and assigned computers where we would begin the
study. When working in our groups we began to notice trends in comparing as far as
The RBSE Journal
25
2008 V2
numbers in certain objects. Our groups, after running through multiple lines of data,
noticed the common reoccurrence of quasars in the night sky. It was nearly impossible to
go through a set of objects, 15 objects total, without coming across at least three or four
quasars. After all of our calculations were completed we came up with some astonishing
results. Our prior hypothesis was correct in which we presumed that quasars were going
to take up a great portion of the night time sky with an astounding 57%. The rest of the
sky was filled with loosely distributed objects, as shown in Chart 1 in the Analysis and
Results section, with the most being radio galaxies with a 17%, elliptical galaxies 13%,
starbursts 8%, and BL Lacs at 5%.
ANALYSIS AND RESULTS
Quasar
Radio
119.0131
509.4139
750.413
751.2919
754.2941
807.3043
809.2912
809.4139
900.3646
910.3759
915.2933
932.284
934.3542
942.401
943.2938
953.3225
953.3917
955.3335
958.3224
1025.4012
1033.3555
1106.3051
46130104
0000m1021
0002m0004.2
0002p0021.2
0004p0000.2
0005m1010
0007p0053
0011p0122.tot
0012m0131
0012m1022
0014m0107
0014m091812_2
0014p0039
729.3046
758.392
824.3342
924.3415
929.3757
930.3439
933.2845
936.3021
943.3614
954.3809
956.362
1002.3453
1005.3414
0012p0125
0058m1114
0059m0215.01
0100m0200
0113p0116
0125m0044
0128m0033
0135m0213.01
0142m0029
0151m0634
0151m0932
0200m085
0203m0242
0218m0033
0225m0743
0245p0108
0246m0642
0247p0023
0249m0828
0256p0039
0317m0054
0319p0005.01
The RBSE Journal
Elliptical
306.0044
748.3709
754.3937
820.3459
821.3107
932.3537
938.3051
951.3614
1001.3052
1038.391
1041.3718
0012p0041
0019m0104
001m0020
001m0041
0035m1019
0041m1108
0043m0026
0105m0033_2
0120m0832
0203m0900
0215m0528
0228p0130
0243p0046
0303p0024
042p2400
0702p5643.2
0703p4436
0703p4443
0704p4439_A
0704p448
0704p5438
0709p4836
0712p5327
0714p3536
26
BL LAC
721.2928
749.3556
807.3043
0050m0929
0051p0126
0127m0151.4
0135m0019
0149p0017
0204m0528
0208m0502
0304m0054.6
0737p2942
0742p2318.4
0746p3926.tot
0803p2437
0820p3640
0824p3916.tot
0854p2223
0854p4408.3
0910p3329
1012p4229
1024p2332
1038p4227
1043p2408
1052p2405
1058p5443
1100p4019
1101p3229
1109p2411
Starburst
926.3453
0004p0051
0013p000.01
00181m1022.2
0028p0055
0031m0136
0038p0010.01
0039m1111
0043m0925
0043m1035
0113p0116
0125m0018
0138m0002.01
0201p0134
0207m0686
0212m0030
0220m0134
0232p0040
0234m0124_A
0247p0023
0249m0828
0711p3015
0712p3627
0714p5343_A
0718p4435
0724p3755
0738p2504
0758p3647
0759.3150.01
0809p3445
0831p3942
0834p1557
0835p2435.01
0845p5519
0849p4028
2008 V2
0015m008_2
0019m0839
0027m0837
0029p0105
002m1039.01
0031m0011
0034m0054_a
0034p0118
0035m1019
0040m146AA
0047m093_2
004m0906
0051p0041_2
0055m1019
0057m0932.01
0059p0006
0100m0055
01013p2212
0102m0853.01
0102m0921
0103m0024
0106m1034.01
0109m0928
0113m0852
0113m1014
0114m0008
0118m0854
0122m0032_A
0122m0935
0125m0005
0125m0018
0128m1032
0129m0054
0129m0829
0129m0914
0130m0135
0130m1019
0130m1046
0131m0841
0132p0026
0134m0133
0135m0213.2
0137m0049
0137m0211
0140m0112
0140m0138
0141m0024
0151m0028
0152m0129
0155p0115
The RBSE Journal
0716p4654
0721p4329.2
0723p3359
0730p2752
0734p4729
0736p3926
0739p5152
0739p5323
0741p2621
0745p3557
0747p4838
0748p334
0749p4152
0755p1447
0801p2608
0802p3940
0804p3833
0804p3853
0805p4714
0805p4810
0817p22423
0818p4635
0823p2448
0823p2852
0825p2340
0829p2225.01
0834p2221
0850p3039.2
0857p3313.2_A
0900p4215.2
0900p4215.3
0902p3957
0903p224.2
0903p2241.2
0903p2447
0913p2447
0914p3059
0932p0055
0933p2222
0933p2456
0934m0151
0937p2314.01
0941p3819
1018p3436
1026p3036
1035p3946
1038p2331
1038p3921
1039p4048m1
1102p2239
0716p4654
0717p5653.2
0723p3731
0724p2830
0726p4706
0726p5726
0727p4816.2
0730p5619
0737p2846
0737p5215
0746p3308
0747p5730
0748p4550
0748p4930.tot
0754p3102
0754p4316_a
0757p306
0800p4639
0801p5045
0805p0627
0822p4711
0832p2233.01
0848p2804.01
0853p4754.3
0854p2811.01
0854p2811.6
0858p1446
08822p4711
0900p4215.2
0900p4215.3
0908p0726
0915p1716
0924p0019
0942p2400
0959p2759
1020m0152
1039p3133
1055p3124
1055p3929
1103p3715
27
0850p2912
0858p1701
0939p2720
1020p3306
1022p2321
1031p2847
1045p0843
1048p2603
1057p4037
2008 V2
0158m0859
02002p0125
0200m0534
0200m085
0200m377 2
0201p0134
0202m0302
0203m0242
0203m0900
0204m0528
0204m0528_a
0204m0721
0210m01015
0210m0517
0212m0100
0216m0444
0218m0605
0219m0737_A
0221p0101
0230m0007
0232m0910_A
0233m0012_A
0236m0121.2
0238m0001
0238m0831
0238p0123_A
0239m0709
0242m0157
0245p0108
0249m0834
0250m0852
0250p0002
0254p4408.3
0256m0119.2
0257p0005
0301p0115.2
0301p0118
0311p0056
0314p0117
0316p0137.2
036m0932
05210p011812
059m0137
0658p5413
0702p4556_A
0705p5448_A
0707p4756
0708p55032
0713p3656
0713p3820
The RBSE Journal
1102p3802
1103p3755
1104p3424
1106p3539
1106p4006
1109p4042
1109p4233
937p2314.01
28
2008 V2
0717p2937
0719p3307
0721p2227
0721p2726
0721p505.2
0721p5241
0722p2856
0722p2941
0722p4558
0723p2859
0724p2517
0724p4159
0725p2429
0725p2819
0726p3019
0726p4010
0726p5037.2
0727p3831
0727p5132
0728p2341.tot
0733p2721
0733p4555.2
0733p5301
0734p2504
0735p2837
0738p2127_A
0739p3043.4
0740p2537
0741p3111
0743p2712
0744.2920.1
0744p2959.4
0744p3208
0744p5149
0745p2614
0745p3142
0745p4734
0748p2200
0748p3006
0749p4510
0752p2017
0758p2624
0758p3222
0800p5010
0803p2727
0804p2722
0805p1529
0806p5041
0807p5117
0809p2753.3
The RBSE Journal
29
2008 V2
0809p3122
0809p4139_2
0809p4723
0810p2321.01
0818p2814
0818p3834
0820p2353
0820p2355
0820p2905
0821p3443_2
0823p2852
0823p4104
0824p4057
0824p5552
0827p3336
0827p5333
0828p4922
082p3134
0830p2708
0830p3213.4
0831p1434
0831p2901
0832p3402
0832p3707
0833p2607
0833p3839
0833p5124
0834p3448
0835p2459
0835p4352
0835p4826
0836p3455
0836p4426
0837p2508
0844p4124
0846p0441
0846p2502
0846p3448
0847p3831
0849.3002.1
0849p2615
0852p4650.3
0857p3313.2_a2
0901p4848
0902p3957
0903p2241.2
0905p2849.2
0907p5515.2
0909p3024
0910p2612
The RBSE Journal
30
2008 V2
0912m0231
0913m0042_A
0917m0000
0918p2325
0919p2914
0922p2236
0924p3547
0929p0041
0934p2902
0934p3153.01
0934p3153.2
0935m0108
0937p3615
0938p2308
0944p2331
0949m0305
0951p2635
0951p2635.01
0952p2240
0952p2352
0952p5048
0956p5152
0957m0253
0957p2356.2
1004p2225
1004p2422
1005p4332
1006p2701
1007m0208
1009p0529
1010.3003.1
1010.4132.01
1012p3309
1013p2449.2
1017p3242
101gm0318
1020p1432
1021p3437
1022p3041.01
1022p3931.01
1025p1551
1030.3102.01
1030p2555
1031p3953.2
1032p3738
1035p3232
1035p3510
1036p3703
1038p3729
1042p3811
The RBSE Journal
31
2008 V2
1044p3656.4
1045p2717
1045p3440
1045p5251
1046p3427
1047p3606
1048p2222
1048p2906
1048p3026
1048p3129
1048p3531
1054p2536.6
1054p2636
1054p2703
1054p3855
1054p4152
1055p2949
1056p3166
1056p3704
1056p5019
1058p3136
1059p4051
1100p2303
1100p2314
1100p2314
1103p3729
1105p3614
1105p5320
1107p3206
1108p2555
1108p3133
1109p2038
904p8233
935p2308
Recorded FIRST Bright Quasar Survey Objects List
BL Lac
29
Elliptical
Galaxy
75
The RBSE Journal
Quasar
Radio Galaxy
Starburst
319
93
44
FIRST Bright Quasar Survey Objects List Total
32
Total
560
2008 V2
350
300
250
200
150
100
50
0
BL Lac
Quasar
Starburst
Chart 1
Once the groups completed their work we compiled our various objects into individual
Excel spreadsheets which were later converged into a single sheet. The end results
showed that an overwhelming amount of quasars dominate the night time sky, tripling the
number of the second closest object, radio galaxies. The BL Lac objects covered the
least amount of the sky studied with only 29 recorded out of the total 560 objects.
DISCUSSION
In our investigation of the night time sky we found that a trend did not exist among the
star placement of the different Bright Quasar Objects. Though quasars were spaced
throughout the entire section of researched data there wasn’t any specific relation
between the assigned columns. Our research found that the numbers of quasars, as well
as other objects, varied from column to column. This was seen after each group
compiled their work for individual sections of objects and saw that some columns
contained as many as eight quasars, the next set of objects carried only five or six
quasars.
SUMMARY AND ACKNOWLEDGEMENTS
We concluded that, though not consistently and equally distributed, quasars take up a
large portion of the night time sky. We feel this is because of the time it takes for light to
travel from the object to Earth has such a delay that we are in fact seeing galaxies from an
earlier time that seem to still be feeding on adjacent stars because they appear to still be
young. There were less BL Lac objects due to the fact that the chance of an AGN to be
pointed directly at Earth so that its jets hit it is slim compared to having the jets facing
anywhere else.
I would like to acknowledge the entire Astronomy Research Seminar class for their
assistance in examining and labeling the objects they were assigned: Roman Acosta,
Gabriel Baduini, Mary Endsley, Zachary Fanelli, Tyler Fountain, Maya Grunder, Patrick
The RBSE Journal
33
2008 V2
Hale, Ryan Hansberry, Nicollette Laroco, Anthony Luis, Ashley Mason, Joseph Mayol,
David Misslbeck, Tyler Padilla, Teddy Pedrozo, Sarah Phillips, Brandon Reno, Caleigh
Smith, Kayla Torres, Jacklyn Trejo, Jacob Wik, and Ethan Wiseman. The entire class
would like to acknowledge Mrs. Christine Wilde for her assistance throughout the
research and her insight that proved helpful when it was required.
The RBSE Journal
34
2008 V2
Finding the CIV Wind in Active Galactic Nuclei
Teacher: Chelen H. Johnson, Breck School, Minneapolis, MN–ARBSE 2007
Teacher: Javier Melendez, Brophy College Prep School, Phoenix, AZ–ARBSE 2007
Teacher: Jeffrey Paradis, Rush Henrietta Senior High School, Rochester, NY–ARBSE 2007
Teacher:Thomas F. Sumrall, Forrest County AHS, Brooklyn, MS–ARBSE 2007
Teacher: Christine Wilde, Central Catholic High School, Modesto, CA–ABSE 2007
Teacher: Lynne F. Zielinski, Glenbrook North High School, Northbrook, IL–ARBSE 2007
ABSTRACT
The focus of this study was to classify quasars that had a blue or redshifted CIV spectral emission
line that was different from usual redshifted spectrum of the quasar. Such a Doppler shifted
emission line in a quasar might be the consequence of a CIV wind emanating from the accretion
disk of active galaxies. The ability to determine the existence of a CIV wind requires a
comparison of the CIV spectral emission line with the CIII] and MgII emission lines. This paper
reveals the fraction of quasars that exhibit this CIV wind and attempts to determine the rate of
motion. Comparative distances and distribution of these quasars were also studied to see if any
other relationships could be found.
INTRODUCTION AND BACKGROUND
Active Galactic Nuclei (AGN) are the source of energy and are considered to be the hot nuclei
within some bright galaxies. Standard features of AGN include a central supermassive black
hole, plasma jets that stream out high energy material, an accretion disk feeding the black hole, an
obscuring torus of thick matter, and broad and narrow line regions considered to be nebular-style
material.
From a close examination of spectra, scientists can determine the characteristics associated with
each AGN type. The term ‘AGN’ encompasses many different types of very bright galaxies.
Typically, AGN are believed to be present in objects that include starburst galaxies, radio
galaxies, quasars, and BL Lacertae (BL Lac) objects. The best way to distinguish between and
identify AGN types is through the spectral comparison of AGN candidate galaxies to normal
galaxies, such as elliptical galaxies. Part of this team’s objective was to do such a comparison. In
addition to identifying galaxy types, the team focused on identifying candidate quasar spectra that
showed a set of uniquely blue or redshifted spectral lines.
In order to make the appropriate comparisons, a set of baseline spectra needed to be chosen.
Normal elliptical galaxy spectral light is believed to come from stars and from nebulae where
stars are being formed within the galaxy.
Comparatively, AGN have a specific set of spectral characteristics that identifies them in a
manner similar to the way human fingerprints exhibit unique patterns. Examples can be found in
the AGN documentation on the RBSE website. These sample spectra were used in this study to
find candidate quasars that might contain the unique spectral lines needed to complete this
investigation.
The RBSE Journal
35
2008 V2
AGN are commonly associated with quasars. Relative to normal galaxies, quasar spectra are
redshifted according to Hubble’s Law due to the cosmological expansion of the Universe. This
team’s first goal was to identify a set of objects whose spectra resembled that of a typical quasar.
In this study, the data were part of the FIRST Bright Quasar Survey, taken at a wavelength of 20
centimeters at the Very Large Array and contains over 1300 spectra. This team created the flow
chart shown in Figure 1 to help characterize and identify each of the FIRST Bright Object Survey
candidate spectra. The flow chart breaks down the specific spectroscopic characteristics that were
needed to find the redshifted quasars needed for this team’s project goal.
Figure 1. Flow Chart for Classification of AGN.
From the redshifted quasars identified, the team looked for specific spectral line identifiers. Of
particular interest were the CIV, CIII], & MgII emission lines, believed to be indicators of hot
gases. All lines can be redshifted in spectra, but when the gas is heated, it may move in a
different direction, causing a blueshift. Some quasars show a CIV emission line that is shifted
differentially relative to the normal redshift of the CIII] and MgII emission lines. These CIV
emission lines are blueshifted.
The RBSE Journal
36
2008 V2
The unusual blueshift in the CIV emission line may be caused by a wind from the AGN’s
accretion disk due to the moving gas inside the quasar. Because the CIV emission line is
blueshifted, the resultant wind producing it is nicknamed the “CIV wind”. Figure 2 shows a
representation of the probably location of the CIV wind relative to the rest of the AGN. It is
purported that high-energy particles, such as ultraviolet and/or x-rays, may be the cause of the
heating in AGN broad line regions, and hence, the cause of the blueshifted CIV wind.
Figure 2, CIV Wind Location on AGN Model. (original image from Max-Planck Institute of
Extraterrestrial Physics.)
CIV winds
In a manner similar to the
direction of the plasma jets
emanating perpendicularly
from the black hole in the
accretion disk, the CIV
wind
also
emanates
outward from the accretion
disk in a perpendicular
direction. However, the
cone of the CIV wind is
believed to be wider than
the plasma jets. This leads
to the research goal for this
project, which was twofold. The first goal was to
determine the fraction of
quasars that exhibit the
CIV blueshift. The second
goal was to determine the
rate at which the CIV wind
is moving.
Observations and Data Reduction
This investigation used the data acquired from the FIRST Bright Quasar Survey, containing
approximately 1300 bright objects. The approach taken to achieve the research goal began by
classifying the AGN in the FIRST Bright Quasar Survey using the flow chart (see Figure 1)
created by the team. Once identified, the quasar candidates were studied to locate the MgII, CIII],
and CIV emission lines. A typical quasar spectrum exhibiting the CIV blueshift is shown in
Figure 3. Note the locations of the MgII, CIII], and CIV emission lines on the figure.
The RBSE Journal
37
2008 V2
CIV
CIII]
MgII
Figure 3. Sample Spectra of Desired Quasar.
Once the candidate quasar spectra were identified, the redshift of the CIV and MgII emission
lines were calculated using the formula:
z +1=
λobs
, where z represents the redshift and λobs represents the observed wavelength, and λrest
λrest
rest wavelength. The motion of the CIV wind was measured and determined using the following
(1+ z) 2 −1
velocity equation: v = c
, where v represents the velocity, z represents the redshift and c
(1+ z) 2 + 1
represents the speed of light
Together, these equations allowed the team to determine the fraction of quasars that exhibit the
CIV blueshift and the rate at which the CIV wind is moving.
ANALYSIS OF RESULTS AND DISCUSSION
Of the 1300 objects available in the FIRST Bright Quasar Survey, the team sampled 519 spectra.
After much discussion, the team classified 275 AGN. Also during this time frame, the flow chart
system for classification was developed. As a result, the team would recommend that anyone
attempting to classify similar data should first create a flow chart for AGN classification and then
document the classification of every object studied. Another recommendation is that a very
systematic method of recording data is required for this type of study since AGN are difficult to
classify.
Of the 275 objects classified, 134 quasars were identified; this represents 49% of the objects
classified. In comparison, 15% were classified as BL Lac objects, 14% were classified as
elliptical galaxies, 12% were classified as starburst galaxies, and 10% were classified as radio
galaxies. Figure 4 shows a pie chart depicting the percentage of each AGN object class.
Figure 4. Distribution of AGN Objects.
The RBSE Journal
38
2008 V2
Distribution of
AGN Objects
(N = 275)
Starburst
galaxies
12%
Elliptical galaxies
14%
Radio galaxies
10%
Quasar
49%
BL Lac
15%
Of the 134 quasars classified, 33 fit the team’s profile for quasars containing detectable CIV,
CIII], and MgII emission lines, as well as having the desired blueshift type. This represents
24.6%, nearly one-fourth of the quasars investigated, fit the profile. Another interesting
observation was that approximately half of the quasars exhibiting the CIV wind were redshifted at
48.5%, and half were blueshifted, at 51.5%. In approximately 3% of the total quasars, neither a
redshift nor a blueshift were detected. A possible explanation for this could be the orientation of
the object relative to Earth. In a wide cone, or funnel, the perspective of the observer could dictate
the redshift. All quasars are moving away from us so the blueshift must be accounted for by a
property of the quasar.
After identifying the CIV wind velocity, the relative velocities were calculated. Relative velocity,
in this case, is defined vCIV – vMgII. Figure 5 shows a bar chart graphing the number of quasars
redshifted and blueshifted relative to a range of speeds. The result was that an average relative
velocity for redshifted quasars was –1109 km/s and the average relative velocity for blueshifted
quasars was +581 km/s. The graph clearly illustrated that the majority of the nearby CIV wind
Doppler shifted quasars lay between –500 km/s and +1000 km/s.
Figure 5. Relative Velocity of CIV Wind in Quasars.
The RBSE Journal
39
2008 V2
Relative Velocities of CIV Wind
Redshif
t
Blueshif
t
-5
00
0
-4 to 49
45
9
0
-3 to - 0
99
40
9
00
t
o
-3
49
35
9
0
-2 to - 0
99
30
9
0
-2 to - 0
49
25
9
00
t
o
-1
99 -20
9
0
-1 to - 0
49
15
9
00
to
1
-9
99 00
to 0
-5
-4 00
99
to
0
1
t
50 o 5
0
1
to 0
10
1
00 00
0
t
15 o 1
01 50
0
t
20 o 2
01 00
to 0
25
00
12
10
8
6
4
2
0
relative velocities (km/s)
The results for the relative velocities led the team to question the values. In theory, if the
accretion disk heats the gases creating the CIV wind, then both redshifted and blueshifted CIV
winds should be expected. However, due to the orientation of the quasar relative to Earth, the
redshifted CIV emission lines were expected to be obscured by the torus. The large velocity
range of the CIV cone could indicate that some CIV wind cones are wider or narrower depending
on the speed of the hot gases emanating from the accretion disk.
Considering the presence of both blueshifted and redshifted CIV lines, it seems plausible that the
observed quasars show both the near and far side of the jets emanating from the accretion disk.
The redshifted lines would represent jets on the far side of the quasar, and, similarly, the blue on
the near side. The small blueshift we observed could be accommodated by the true redshift of the
AGN, and the orientation of the jet toward Earth. Generally, the accretion disk and the torus are
considered to be optically thick and would create the tendency to see only the blueshifted
component.
Another possibility to explain the blueshift could be due to a wider angle of the cone and the
position of the observer. Elvis suggests that the cone, or funnel-shaped outflow, is critical to the
observer. The differences between higher and lower angles relative to the accretion disk dictate
the strength of the narrow absorption line (NAL) regions, broad absorption line (BAL) regions
and what he calls, embedded broad emission line regions (BELR). These differences in
perspective may provide the range of velocities and redshift to accommodate our observations.
According to Leighly and Moore, “These profiles are most simply explained if the high-ionization
gas is accelerated toward us in a wind, while the low-ionization gas is emitted on the surface of
the accretion disk or in the low-velocity base of the wind.” This would accommodate large
differences in velocity, and again, depending on perspective, type of shift.
This discussion led to other questions, which should be considered for further study:
The RBSE Journal
40
2008 V2
-
How would position angle and orientation of AGN affect the percentage of CIV wind
detected?
Do other quasar surveys yield similar results?
Using high-energy data (e.g., Chandra) can the intrinsic velocities of the CIV winds
be determined?
In addition, the team asked one more question: Could there be a relationship between these AGN
and their distance and location in the universe? The team took a look at this question, and began
cz (1+ 0.5z)
by using the following equation for calculating the quasar distance: d =
, where c
H 0 (1+ z)
represents the speed of light, z represents redshift, and Ho represents Hubble’s constant at a value
of 75 km/s /Mpc. The calculations revealed that the majority of these quasars were located in a
distance range between 4000 – 5000 Mpc.
Figure 6. Distance to Quasars Containing CIV Wind Doppler Shifts.
Distances to Quasars
14
12
10
8
6
4
2
0
3000-3499
3500-3999
4000-4499
4500-4999
5000-5500
distances in Mpc
In addition, these quasars’ positions in the Universe were plotted using their Right Ascension and
Declination coordinates in two ways. The Figure 7 plot shows them on a straight graph, while the
Figure 8 plot was done on an overlay of a universe map. The blue dots represent the quasars used
in this team’s study. The locations of the quasars are probably a result of the locations of the
original FIRST Bright Quasar Survey objects, and therefore, no real conclusions could be made
relative to location based on this data alone.
The RBSE Journal
41
2008 V2
Figure 7. Quasar Location Plot based on Right Ascension and Declination
0
4
8
12
16
20
24
Figure 8. Quasar Locations Overlaid on a Map of Universe.
(The black dots represent the locations of this quasar set)
The RBSE Journal
42
2008 V2
SUMMARY
Much was learned from this project investigation. Both research goals were accomplished. First,
it was determined that approximately 25% or ¼ of the quasars identified fit the profile for AGNs
containing CIV, CIII], and MgII emission lines, and having the correct blueshift type. The second
goal was to determine the rate at which the CIV wind is moving. From our data, the CIV wind
shifted in a peak region between –500 km/s and +500 km/s. This was unexpected because there
seemed to be a nearly equal distribution of blueshifted and redshifted CIV winds and there also
seemed to be a wide distribution of velocities that may indicate a broad CIV wind cone range. An
unexpected consequence of this investigation was that the blueshifted CIV wind quasars were
present in low-z quasars and that these low-z quasars were commonly found at distances between
3500 and 5000 Mpc. Finally, the team also looked at the distances and locations of the quasars
identified and found that there was no pattern for their distribution.
REFERENCES
Brotherton, et al. “Statistics of QSO Broad Emission-Line Profiles. II. The CIV λ1549, CIII]
λ1909, and MgII λ 2798 Lines.” ApJ 423: 131-142, 01 March 1994.
Elvis, M. “A Structure for Quasars.” ApJ 545: 63-76, 10 December 2003.
Ferland. “The CIII] λ1909 Effective Wavelength-Redshift Relationship in Quasars.”
ApJ 249: 17-22, 01 October 1981.
Gaskell. “A Redshift Difference between High and Low Ionization Emission-Line Regions in
QSOs -- Evidence for Radial Motions.” ApJ 263: 79-86, 01 Dec 1982.
Leighly, Karen & Moore, John. “Hubble Space Telescope. STIS Ultraviolet Spectral Evidence
Of Outflow In Extreme Narrow-Line Seyfert 1 Galaxies. I. Data And Analysis.”
ApJ 611: 107-124, 10 August 2004.
McIntosh, et al. “Redshifted and Blushifted Board Lines in Luminous Quasars.”
ApJ 517: L73-L76, 01 June 1999.
Mueller-Sanchez, Francisco, et al. Nearby AGN Research” Infrared and Submillimeter
Astronomy Group at Max-Planck Institute For Extraterrestrial Physics. 15 May 2007.
<http://www.mpe.mpg.de/ir/Research/AGN/index.php>
Richards, et. al. “Broad Emission-Line Shifts in Quasars: An Orientation Measure for RadioQuiet Quasars?” AJ 124: 1-17, July 2002.
Shuder. “Emission-Line Profiles in Low-Redshift QSOs.” AJ 280: 491-498, 15 May 1984.
Tytler, et al. “Systematic QSO Emission-Line Velocity Shifts and New Unbiased Redshifts.”
ApJS 79: 1-36, March 1992.
The RBSE Journal
43
2008 V2
Wills, et al. “Statistics of QSO Broad Emission-Line Profiles. I. The CIV λ1549 and the λ1400
Feature.” ApJ 415: 563-579, 01 October 1993.
The RBSE Journal
44
2008 V2
NGC 2367: Its Age and Distance
Caitlin S. Colley
Sullivan South High School, Kingsport, TN
Teacher: Thomas Rutherford, TLRBSE 2005
ABSTRACT
The distance to the open cluster NGC 2367 was determined along with its age. The cluster was
examined using data collected with “V” and “B” photometric filters. The images were taken with
a 14-inch telescope located near Cloudcroft, New Mexico. NGC 2367 was found to lie at a
distance of 2570 parsecs (7430 light years) from the earth. The age of the cluster was determined
to be no older than 19.8 million years. Few M-type stars were evident.
INTRODUCTION
An open cluster is a group of stars that are held together by gravity and whose stars lie at
essentially the same distance from the earth. They are all thought to have formed at the same
time, from the same materials, and under identical conditions (Chaisson 2005). However, the
stars in the open cluster do have different masses which make them age at different rates.
The study of open clusters is important because they can provide information on the structure of
the galaxy, provide clues to stellar evolution, and aid in calibrating knowledge of star brightnesses
(Moffat 1972).
The images used in this study were collected in January 2007 using a CCD camera mounted on a
Celestron C14 telescope. The telescope was located at a robotic observatory near Cloudcroft,
New Mexico, called New Mexico Skies. The telescope time was funded by the NOAO/RBSE
Program.
The cluster NGC 2367 was chosen because it had not been researched extensively. The cluster
had been examined in the early 1970’s (Moffat 1972) and more recently in 2005 by McSwain
(2005).
OBSERVATIONS AND DATA REDUCTION
Ten images were taken using each filter (B and V). Each exposure was 60 seconds in length. The
10 images were then stacked using the software MaximDL in order to give an image that was the
equivalent of a single 600-second exposure. This technique improves the signal/noise ratio of the
image; increasing the amount of signal (star images) at a greater rate than it increases the noise
level (Starizona 2008).
The resulting images were processed using the kernal function of MaximDL. This removed
excess hot pixels which were present in the images. The setting for the kernal filter was 10%.
The RBSE Journal
45
2008 V2
Image 1. An image of the open cluster NGC 2367 in B. The brightnesses in the image were
inverted for clarity. The stars of the cluster lie just to the left of center in the “V” like formation.
The brighter stars to the right were thought not to be a part of the cluster based upon their
omission from numerous sources, including the WebDA site.
Originally, it was planned to use the star identification system used by Moffat (1972). However,
Moffat’s image (WebDA) was of such poor quality that it was not possible to determine to which
star a particular reference number belonged. Arbitrary reference numbers were therefore assigned
to the stars as seen in Image 2. The stars labeled 24 and 25 in image 2 were used as the reference
stars.
Image 2. The stars in this study were identified according to the following numbering system.
The MaximDL software compared the brightnesses entered for stars 24 and 25 and then used that
information to calculate to brightnesses for the other stars in the images. It then produced an
The RBSE Journal
46
2008 V2
output file that was compatible with Microsoft Excel, allowing further manipulation of the data to
be performed.
When the brightness of a star in B is subtracted from its brightness in V, the star’s true color
(color index) may be determined according to the following equation:
B - V = Star’s Color Index
Equation 1
However, interstellar dust lies between the Earth and all astronomical objects. Because of this,
light from the cluster was both dimmed (extinction) and reddened as it traveled toward the earth
(Seeds 2005). This means that the stars appear to be both redder and dimmer than they actually
are.
The amount of reddening of the stars in NGC 2367 was 0.331 magnitude (WebDA 2006). This
number was subtracted from each star’s (B-V) value giving the stars’ true color indices. In
addition, extinction was compensated for by multiplying the reddening value (0.331 mag) by 3.2
and subtracting that value from the stars’ V brightnesses (Clemens 2007). These corrected values
may be seen in Table 1.
The RBSE Journal
47
2008 V2
Table 1. The B and V values for the stars in NGC 2367 were corrected for interstellar reddening
and interstellar extinction according to the method mentioned in the text. The apparent duplicate
entries (marked with an asterisk) were due to stars being too close together to measure separately.
Star 3 (marked with two asterisks) was determined to not be a member of the cluster due to its
placement on the ZAMS graph.
Star
NGC 2367-1
NGC 2367-2
NGC 2367-3**
NGC 2367-4
NGC 2367-5
NGC 2367-6
NGC 2367-7
NGC 2367-8
NGC 2367-9
NGC 2367-10
NGC 2367-11
NGC 2367-12
NGC 2367-13*
NGC 2367-14*
NGC 2367-15
NGC 2367-16
NGC 2367-17
NGC 2367-18
NGC 2367-19
NGC 2367-20*
NGC 2367-21*
NGC 2367-22
NGC 2367-23
NGC 2367-24
NGC 2367-25
NGC 2367-26
NGC 2367-27
NGC 2367-28
B
V
B-V
(B-V)RED
(V)RED
11.09
13.39
13.50
10.53
13.54
13.04
13.08
12.93
12.97
10.97
12.73
12.88
10.36
10.36
12.49
12.84
9.82
12.13
10.68
8.78
8.78
13.17
13.46
14.29
12.35
13.11
13.30
11.07
10.61
13.03
11.83
10.30
13.20
12.85
12.69
12.45
12.56
10.83
12.24
12.66
10.29
10.29
12.28
12.24
9.77
11.28
10.48
8.72
8.72
12.86
13.20
13.75
11.77
12.51
12.84
10.61
0.48
0.36
1.67
0.23
0.33
0.19
0.39
0.48
0.41
0.14
0.50
0.22
0.07
0.07
0.21
0.59
0.05
0.84
0.20
0.06
0.06
0.30
0.26
0.53
0.58
0.59
0.46
0.46
0.15
0.03
1.34
-0.10
0.00
-0.14
0.06
0.15
0.08
-0.19
0.17
-0.11
-0.26
-0.26
-0.12
0.26
-0.28
0.51
-0.13
-0.27
-0.27
-0.03
-0.07
0.20
0.25
0.26
0.13
0.13
9.55
11.97
10.77
9.24
12.14
11.79
11.63
11.39
11.50
9.77
11.18
11.60
9.23
9.23
11.22
11.18
8.71
10.22
9.42
7.66
7.66
11.80
12.14
12.69
10.71
11.45
11.78
9.55
Spectral
Type
B
A
M
B
A
B
A
A
A
B
A
B
B
B
B
F
B
F
B
O
O
A
A
A
F
F
B
B
ANALYSIS AND RESULTS
Once the correct (B-V) values were determined for the stars in the cluster, they were classified by
spectral type (O, B, A, F, G, K, and M). The criteria for making this determination may be found
in Table 2.
The RBSE Journal
48
2008 V2
Table 2. The spectral type of a star can be determined by measuring its (B-V) value. A star’s
spectral type is also an indicator of its life expectancy (Washington 2005).
Color Index (B-V)
-0.4
-0.2
0.2
0.5
0.7
1.0
1.6
Spectral Type
O
B
A
F
G
K
M
Life Span (years)
<1 × 106
3 × 107
4 × 108
4 × 109
1 × 1010
6 × 1010
>1 × 1011
When the stars of NGC 957 are plotted on an H-R diagram containing the Schmidt-Kaler ZAMS
(Zero-Age-Main-Sequence) line (1982), an estimate of the age of the cluster can be made. This
involves checking to see where the cluster’s stars are beginning to curve off from the mainsequence (Seeds 2005). NGC 2367 is estimated to be about 19.8 million years old, based on the
lack of a ZAMS turnoff and the presence of an O-type star.
Graph 1. The ZAMS (Zero-Age Main Sequence) line (Schmidt-Kaler 1982) is overlaid with a
graph of NGC 2367’s stars. No turn-off from the ZAMS line is evident. The star to the far right
(Star 3) is not a cluster member.
A star in an open cluster must have its absolute magnitude known in order to have its distance
from earth calculated (Comins 2003). Since this value was unknown, a “stand in” star was
chosen. The star chosen for this was Canopus (alpha Carinae) because its (b-v) value of 0.15 is
the same as the (b-v) value of star 1 (also 0.15).
The RBSE Journal
49
2008 V2
Star 1 has an apparent V magnitude of 9.55. Canopus’ absolute magnitude (V) is -2.5. Stars that
have similar (b-v) values, like star 1 and Canopus, should also have the similar absolute
magnitudes. Using the distance modulus equation (Comins 2003) allows the distance to NGC
2367 to be calculated as shown in equation 2.
d=10(m-M+5)/5
Equation 2
where d is the distance to the star in parsecs, m is the apparent V magnitude, and M is the absolute
V magnitude. The distance to the cluster was calculated to be about 2570 parsecs or 8380 light
years. This is a somewhat greater distance than that calculated by Moffat (1972) who calculated a
distance of 2004 parsecs (6530 light years). Moffat, however, determined this value using a
photoelectric photometer, not a CCD imager—this may explain some of the discrepency.
Graph 2. A frequency graph of the spectral types of stars in NGC2367. Most stars in the cluster
are of the B and A types, although one O type is also present. This study showed only one M
class star, even though this is the most common type of star. The lack of detected M stars may be
due to the presence of a full moon when the images were taken. Furthermore, M-class stars are
faint, making them more difficult to detect than the brighter spectral types.
Spectral Type Distribution of NGC2367
14
12
12
10
Number of Stars
9
8
6
4
4
2
2
1
0
0
G
K
0
O
B
A
F
M
Spectral Types
DISCUSSION
Since there is no apparent ZAMS turn-off, this cluster is relatively young. It is estimated to be
approximately 19.8 million years old after interpolation of the values in Table 1. This is
confirmed by the presence of an O-type star, which provides additional evidence that this is a
young cluster.
The RBSE Journal
50
2008 V2
One star, NGC2367-3 was determined to not be a member of the cluster due to its placement far
from the ZAMS line. Also, the (b-v) value of star NGC 2367-21 was possibly compromised due
to its closeness to the O-type star NGC 21367-20, making it impossible to determine its true
value.
McSwain (2005) determined that the distance to NGC 2367 was 2330 parsecs (7600 light years)
and that it was 5.5 million years old. The age agrees well with the current study’s result of 2570
parsecs. The age of the cluster was estimated at 19.8 million years, which is somewhat older than
McSwain’s value of 5.5 million years. McSwain’s use of a much larger telescope (0.9 meter
versus the 0.36 meter used in this study) would provide a larger sample of cluster members,
which might explain McSwain’s younger age for the cluster.
SUMMARY
Using values measured from New Mexico Skies images, the age and distance of the open cluster
NGC 2367 was estimated to be 19.8 million years and to lie at a distance of 2570 parsecs (8380
light years) from the earth. The cluster’s young age was also evident from the presence of an Otype star. Additionally, there was no turn-off from the ZAMS line.
ACKNOWLEDGEMENTS
Thanks to Mr. Tom Rutherford for his patience, guidance, and the knowledge he instilled in me.
Astronomy was not something I expected to find an interest in, but his infectious spirit towards
this project spread to me and I am glad he opened my eyes to so many things, including how to
find north. I would also like to thank my father for sharing the computer and letting me work in
between his intense solitaire games, and to my mother for lying in our backyard watching the
stars with me.
REFERENCES
Vogt N., Moffat A.F.J. “Southern Open Star Clusters I.” 1972, Astronomy Astrophysics
Supplement Series, Supplement 7, (1972): 133-138. NASA Astrophysics Data System.
Chaisson, Eric and Steve McMillan. Astronomy Today, 5th ed. New Jersey: Prentice Hall, Inc.
2005. 505.
Seeds, Michael. Foundations of Astronomy, 8th ed. Belmont, CA: Brooks/Cole. 2005.
200.
Starizona. “Optimum Exposures”. <http://starizona.com/acb/ccd/advtheoryexp.aspx>. 2008.
Accessed 28 March 2008.
WEBDA. Institute for Astronomy of the University of Vienna. 22 September 2003.
http://www.univie.ac.at/webda/cgi-bin/ocl_page.cgi?cluster=ngc2367.
The RBSE Journal
51
2008 V2
Clemons, Christie and Rachel Reece. “The Age and Distance of the Open Cluster NGC 2345.”
RBSE Journal. 2007. 27 March 2008.
<http://www.noao.edu/education/arbsefolder/files/rbsejournal07.pdf>
University of Washington. “Cluster Color Magnitude Diagrams and the Age of Stars.” 16 May
2005. 30 March 2008. <http://www.washington.edu/labs/claeringhouse/labs/
Clusterhr/cluster.html>
Schmidt-Kaler, Th. 1982, Landolt-Börnstein, Numerical data and Functional
Relationships in Science and Technology, New Series, Group VI, vol. 2(b), ed. K.
Schaifers, & H. H. Voigt (Berlin: Springer Verlag), 14
Comins, Neil F. and William J. Kaufmann III. Discovering the Universe, 6th ed. New York, NY:
W. H. Freeman and Company. 2003. 277, 369.
Mcswain, M. Virginia and Douglas R. Gies. “The Evolutionary Status of Be Stars: Results from
a Photometric Study of Southern Open Clusters.” The Astrophysical Journal Supplement Series,
161:118–146, 2005 November.
The RBSE Journal
52
2008 V2
A Search for Exoplanets in NGC 957
Bobby Adams, Rebecca Redmon, Veronica Buehrig
Sullivan South High School, Kingsport, TN
Teacher: Thomas Rutherford, TLRBSE 2005
ABSTRACT
Approximately 200 stars in the open cluster NGC 957 were examined. The cluster was imaged
using “B”, “V”, and “R” photometric filters attached to a CCD camera. The “R” images were
examined for the shallow, D-shaped changes in brightness that typifies a transiting exoplanet.
The “B” and “V” images were examined in order to determine the stars’ spectral types, allowing
both the age and the distance to the open cluster to be determined. No changes attributable to an
exoplanet transit were detected. NGC 957 was estimated to be less than 10 million years old
based on the lack of a ZAMS turnoff. The open cluster was calculated to lie at a distance of
1380 parsecs (4500 light years) from the earth.
INTRODUCTION
An exoplanet, or extrasolar planet, is a planet that is found outside of the solar system (Comins
2003). As of this writing, 277 exoplanets have been discovered (JPL 2008). The study of such
planets is important for several reasons:
1) Based on the current ideas of how solar systems form, planets should be common
around other stars.
2) The earth’s solar system is made of eight planets and numerous other bodies. The
type of planet depends on its distance from the sun with the inner planets being small
and rocky while the outer ones are large and gaseous. Is this same pattern followed
elsewhere?
3) The solar system contains only certain sizes of planets. Can planets exist outside of
this size range? Can they be larger than Jupiter or smaller than Mercury?
The first detection of a planet outside the earth’s solar system occurred in 1992, with the
discovery of two planets orbiting the pulsar PSR 1257+12 (Bisnovatyi-Kogan, G.S. 1993). The
planets were detected because of the effect that they were having on the timing of the pulsar’s
radio emissions (Wolszczan 1992).
The first detection of a planet orbiting a main-sequence star was in 1995, when 51 Pegasi was
found to have an orbiting planet (51 Pegasi b). This planet orbits its star once every 4.2 days
(Kaler 2005). It was discovered by the effect that it had on the radial velocity of its parent star
(Mayor 1995).
In 1999, the star HD 209458, located in the constellation of Pegasus, was found to have an
orbiting planet. Although this planet was first detected by Doppler shift, it was soon found to
The RBSE Journal
53
2008 V2
actually cross the face of its parent star. This exoplanet transit confirmed that the drop in the
brightness of the star that occurred as the planet crossed its face was detectable from earth
(Malatesta 2004).
Typical exoplanet transits last from about 1-4 hours (Gary 2007), although there can be transits
outside of this range. Exoplanet transits are very shallow, resulting in a brightness drop of only a
few percent or less. They also have periods ranging from 1-21 days (Gary 2007), meaning that
only one transit would be detectable during the observing run. Since at least two transit minima
are needed to determine a period for a transiting planet no such determination can be made from
this data, due to the brevity of the observing run.
However, if the period of an exoplanet transit can be determined from additional studies, then the
planet’s orbital inclination, its mass, its parent star’s mass, its size, its distance from its star, and
its orbital period can be determined. Also, other information, such as the composition of the
planet’s atmosphere, can be measured (Charbonneau 2006).
OBSERVATIONS
On the nights of October 29 and 30, 2007, the 0.9-meter WIYN telescope, located at Kitt Peak,
Arizona, was used to search for transiting exoplanets in the open star cluster NGC 957. This
open cluster is located in the constellation Cassiopeia close to its border with Perseus. Its
astronomical coordinates are RA 02h 33.3m and DEC +57d 34m 7.6s. The Double Cluster
composed of h and χ Persei lies nearby.
Image 1: An image of the open cluster NGC 957 in blue light (B). The vertical spikes, which
appear on the brighter stars, are the result of “blooming” in the CCD camera’s pixels. The true
brightnesses and colors of these stars could not be determined.
The RBSE Journal
54
2008 V2
Attached to the 0.9-meter telescope was the S2KB CCD camera with which all images were
taken. This camera has a pixel size of 21 microns arranged in an array that measures 2048 x
2048 pixels per side. This gives an image scale of 0.6 arc seconds per pixel. The camera is
cooled with liquid nitrogen in order to eliminate the thermal noise produced by each pixel. The
S2KB camera has a column defect that does not affect the results. This defect is the vertical bar
near the mid-line in Image 1.
NGC 957 was chosen as the target cluster because it transited high above the telescope during
the time of the study. This minimized the effects of changes in airmass during the observing run.
Exoplanet searches require very high precision since the changes in brightness during a transit
are generally very small, on the order of only a few percent (Naeye 2004).
Filters −
This project was undertaken using “B”, “V”, and “R” photometric filters. There were several
reasons for these choices:
1) Images taken in “B” and “V” allow the color indices, and therefore the spectral
types, of the stars, to be accurately determined according to the following equation:
Color Index = brightness (B) – brightness (V)
Equation 1
2) Images taken in “R” suffer less from atmospheric scintillation than images taken at
shorter wavelengths, although this is less of an issue with longer exposures than it is
with shorter ones.
3) The longer wavelengths, as seen through an “R” filter, penetrate interstellar dust with
less absorption, resulting in less dimming of the target stars.
4) The CCD camera used was also more sensitive to longer wavelengths than to shorter
ones, allowing fainter stars to be detected.
5) The use of the standard filters chosen allows the results to be compared to the results
of other observers.
Exposure Time −
It was necessary to reach a signal/noise (S/N) ratio of at least 100 due to the fact that the
brightness changes of the parent star will be on the order of 0.01 magnitudes (1%) or less. A
high S/N ratio will permit greater sensitivity to such small changes in brightness.
According
to
the
program
“CCDTIME”
(provided
by
NOAO
at
http://www.noao.edu/gateway/ccdtime/), the previously mentioned setup would reach a
magnitude limit of 18.5 with an S/N ratio of 100 in approximately 171 seconds when using an
“R” filter. In “B”, an exposure time of 225 seconds was required to reach the same limiting
magnitude and S/N ratio. In “V”, the exposure was the same as in “R” (171 seconds) in order to
match the previous values.
The RBSE Journal
55
2008 V2
The faintest members of NGC 957 are of about magnitude 17.5 (WebDA). By having the CCD
exposures go to magnitude 18.5, the faintest known cluster members would be detected and
possible fainter members might also be picked up.
The following procedures were followed each night:
1) At the beginning of the observing run, images in “B” and “V” were collected. This
insured that some usable data was collected even if the observing run was cut short
due to weather or other factors.
2) When the target was near the zenith, “B” and “V” data was again collected.
3) Near the end of the observing run, “B” and “V” data was once again collected.
4) At approximately 5-6 minute intervals, images in “R” were taken. These images
comprise the majority of data collected.
Periodically throughout the observing run, observing requests (service observations) from other
observers had to be fulfilled. Those obligations required moving the telescope to a new target,
changing filters, refocusing, and then acquiring data for those observing projects. This limited
the amount of time that was available for the scheduled observation run.
DATA REDUCTION
Data reduction involved making photometric measurements of the target stars in each image.
The photometric software utilized included Meade AutoStar Suite and MaximDL.
In order to reduce the number of stars that had to be examined, those unlikely to have planetary
systems were ignored. The following criteria were used to make this determination:
1) Planetary systems are thought to be unlikely to form around highly energetic stars,
since the high power output of these stars forces potential planetary material, such as
dust grains, outward away from the star (Vu 2006). Therefore, only stars of spectral
types F, G, K, and M were examined for possible planetary transits. The stars’
spectral types were determined using the “B” and “V” data collected during the
observing run according to Equation 1.
2) Although “B” and “V” data was collected three times during the run, only the “B”
and “V” data from the second set of observations was used. This observation
occurred when NGC 957 was at its highest elevation above the telescope and so the
least amount of atmospheric reddening and extinction would have occurred.
3) Some stars were overexposed on the data images. The pixels were saturated and so
the true brightnesses of these stars could not be determined. This also meant that
their true spectral types could not be determined.
The RBSE Journal
56
2008 V2
4) Possible variable stars were checked against the database of the AAVSO and were to
be excluded from the search if they had proven to be known eclipsing binaries. A
check of the AAVSO chart centered on this cluster showed that there are no known
variables present in the cluster.
In order to further differentiate eclipsing binary stars from a true exoplanet transits, the following
presumptions were made:
1) Eclipsing binary stars would have a greater decrease in brightness while they transit
their primary. This would be due to the fact that stars are larger than planets and so
the eclipsing member would block out more of the parent star’s light.
2) The onset and ending of the eclipse will be steeper and shallower with an exoplanet
than with a binary star—once again, this is due to the larger size of the star.
As a preliminary step, the images of known transiting exoplanets were examined. This enabled
the researchers to become familiar with the type of light curve, as seen in Image 2, produced
when a planet crosses the face of its star.
Image 2: A light curve of an actual exoplanet transit of the star X0-1, located in the
constellation Corona Borealis. This is a composite of data taken on two separate occasions with
two different 14-inch telescopes (Gary 2007).
Only the numbered stars shown in Image 3 were used in this study, although some of the other
stars in the image may be members of the cluster. A list of these stars may be found at the end of
the paper in Table 5.
Three reference stars in each image were used for all photometric comparisons. These are the
stars marked 118, 113, and 24 in Image 3. AAVSO charts were examined for the magnitudes of
these reference stars. However, no reference stars were marked for the cluster on the AAVSO
charts so measurements from the WEBDA online database were used.
Image 3: A labeled picture of the open cluster NGC 957 (Gimenez 1980). Stars 118, 113, and
24 were used as reference stars for this project. Only the numbered stars in the image were used
in this study.
The RBSE Journal
57
2008 V2
MaximDL allowed the measurement of stellar brightnesses simultaneously in multiple images.
A set of 5-10 images was loaded into the program, and the previously mentioned reference stars
were selected. The software would then find those stars as well as measure the target stars in
every image. This greatly increased the speed with which the data could be analyzed.
Following the measurement of the stars’ brightness values, those values were entered into an
EXCEL spreadsheet for further analysis. A sample of the information may be found in Table 5.
Due to its large size the complete data table may be found online at
http://www.scde.k12.tn.us/metadot/index.pl?id=9427
The RBSE Journal
58
2008 V2
Table 5. A sample of the data from the observing session is shown. Stars 5-10 are shown along
with the time (JD 2454400 +) of the observations. The total data set was much too large to
include in this table. The stars noted in red are those that were omitted from the study.
NGC r1
5 12.23
0
6 12.49
9
7 11.07
1
8 14.12
7
9 13.58
4
10 13.72
9
r2
r3
12.25
2
12.54
0
11.04
3
14.12
4
13.58
4
13.72
1
12.24
9
12.50
7
11.05
8
14.11
9
13.58
5
13.73
1
r4
Time r1
(JD)
12.27 3.775
9
12.51 3.775
9
11.08 3.775
6
14.12 3.775
4
13.58 3.775
0
13.72 3.775
5
Time r2
(JD)
3.779
Time r3
(JD)
3.783
Time r4
(JD)
3.787
3.779
3.783
3.787
3.779
3.783
3.787
3.779
3.783
3.787
3.779
3.783
3.787
3.779
3.783
3.787
The light from distant objects must travel through interstellar dust on its way to the earth. The
amount of dust varies from region to region in the sky, but it has a similar effect on the starlight
that passes through it.
Shorter wavelengths, such as blue, are absorbed more readily than longer, redder, wavelengths.
This has the effect of making the stars appear redder than they really are. In addition, some of
the starlight is actually blocked by the dust; therefore, the stars also appear dimmer than they
truly are (extinction) (Swinburne 2008).
In the following equation, the object’s true color (E(B-V) ) can be determined by subtracting its
true color index (B-V)0, or reddening value, from its measured color index (B-V).
E(B-V) = (B-V) - (B-V)0
Equation 2
In the case of NGC 957, a reddening value of 0.842 magnitude was taken from the WebDA
website as being the appropriate correction for stars in the cluster. This value was subtracted
from all measured (B-V) values in NGC 957, giving the true color indices of the stars.
Once the reddening factor was known, a correction for extinction could be made. This involved
taking the reddening factor and multiplying it by 3.2 as is shown by the following equation. The
resulting answer is then subtracted from the (V) brightness of the stars in the cluster in order to
determine their true brightnesses (Clemons 2007). Once this information is known, then the
stars’ true spectral types could be determined as seen in Table 1 (Washington 2005).
AV = 3.2 E(B-V)
The RBSE Journal
Equation 3
59
2008 V2
Table 1. The spectral type of a star can be determined by measuring its (B-V) value. A star’s
spectral type is also an indicator of its life expectancy (Washington 2005).
Color Index (BV)
-0.4
-0.2
0.2
0.5
0.7
1.0
1.6
Spectral
Type
O
B
A
F
G
K
M
Life Expectancy
(years)
<1 × 106
3 × 107
4 × 108
4 × 109
1 × 1010
6 × 1010
>1 × 1011
RESULTS
When the data was analyzed and graphed, many of the stars displayed similar changes in
brightness during the observing run, as may be seen in Image 6. This indicated that the changes
are not intrinsic to the stars, but are instead caused by external factors. While the actual shapes
of the variations were different, stars in the same part of the cluster tended to show similar light
curve fluctuations.
Image 6. Many of the stars exhibited similar light curves, as seen below, indicating that the
brightness changes are due to outside factors. These changes varied according to where in the
cluster the star is located with stars near one another exhibiting similar changes.
The RBSE Journal
60
2008 V2
The RBSE Journal
61
2008 V2
Only one star, 176, as seen in Image 7, showed a light curve that was different from those of the
other stars. With only a partial light curve, it is not possible to tell exactly what is responsible
for the change in brightness although variability in the star is the most probable reason.
IMAGE 7. Star 176 was the only star to show a brightness change that was not like at least one
of the other stars, probably due to the star itself being variable.
The brightness drop is unlikely to be due to an exoplanet transit because of the depth of the
magnitude change. The change in brightness is at least 0.23 magnitude (V) in size as can be seen
in the graph. It was not possible to determine the actual brightness change from the data at hand,
since the brightness curve may have been climbing when the first measurement was made.
Since “B” and “V” data were collected, it was also possible to make a determination of the age
of NGC 957 as well as its distance from the earth. This process involves plotting the “V”
magnitudes of the stars in the cluster against their color indices (spectral types). A brightness
correction is then applied in order to overlay the NGC 957 stars over the ZAMS (Zero Age Main
Sequence) line (Schmidt-Kaler 1982). This is seen in Image 5.
IMAGE 5. An H-R diagram showing the stars of NGC 957 compared to the Schmidt-Kaler
ZAMS line. The group of stars to the right is thought not to be cluster members, since they do
not lie near the ZAMS line and are also separated from the other stars by a gap.
The RBSE Journal
62
2008 V2
Since all stars in a cluster lie at the same distance from the earth and are of the same age, the age
of the cluster can be determined by comparing the spectral types (color indices) of its members
to that of the ZAMS line. The point where stars begin to curve off of the main sequence line
toward the upper left gives an indication of the age of the cluster. Young clusters have few stars,
if any, moving off of the main sequence while older clusters have more stars leaving the main
sequence and moving off toward the upper right. Note that NGC 957 appears to have no stars
that have moved off of the main sequence, implying that the cluster is relatively young, perhaps
less than 10 million years.
In addition to its age, the distance to the open cluster can also be determined from the “V” data
using the distance-magnitude equation given by Coumins (2002).
d=10(m-M+5)/5
Equation 4
Where d is the distance to the star in parsecs, m is the apparent magnitude of the star, and M is
the absolute magnitude of the star.
Since the true distance to the cluster is unknown, the absolute magnitudes of its stars are also
unknown. Therefore, a star (whose absolute magnitude is known) of similar spectral type as one
of the cluster members was chosen to represent that cluster member. Using equation 4, this
allows an estimate to be made of NGC 957’s distance from the earth.
The star Vega was chosen as the representative star. Its color index (B-V) is 0. Star NGC 957114 has a color index of 0.001 and so the two stars were deemed to be of the same spectral class.
Inserting the values for Vega (M) and star 114 (m) into equation 4 gives a distance to the cluster
of 1380 parsecs (4500 light years). This compares reasonably well with the distance value of
1850 parsecs determined by Gimenez (1980).
No apparent exoplanet transits were detected during the course of the October 2007 observing
run. All of the stars exhibited brightness changes that were inconsistent with those expected of
an exoplanet transit. Only one star, 176, exhibited a change in brightness not found in the other
stars; this change was of too large an amount to be attributed to a transiting planet.
Many other stars had large variations in their brightnesses during the observing run making
detection of a planetary transit unlikely, if not impossible. Further information may be found
below.
DISCUSSSION
The lack of detected exoplanet transits is attributed to the following reasons:
1) Useable data was only collected for one night, which is a very short amount of time
for this type of project.
2) The 0.9-meter WIYN exhibited many mechanical problems: issues with the CCD
shutter on the camera; the dome shutter failed to track along with the telescope and so
The RBSE Journal
63
2008 V2
had to be moved manually; the focus on the CCD camera was not stable; other
observing requests resulted in time being taken from the exoplanet search.
Of these, the most serious issue was the time spent off-target for other observing
projects—this resulted in large gaps in the data and made it difficult to detect the
short-term changes in brightness that an exoplanet would tend to make.
3) By the second night, clouds were an issue resulting in the data from that night to be
unusable.
SUMMARY
When the star images were examined, it appeared that most of the images showed a change in
brightness at about JD +3.89. A check of the telescope data logs indicated that this occurred just
after the telescope returned to NGC 957 after performing a service observation for another
astronomer. Also, while doing this service observation, shutter problems with the CCD camera
as well as moonlight in the dome were also noted. It is felt that these factors are responsible for
the apparent changes in brightness that occurred to almost every star in the cluster at this
moment.
There was no detected evidence of an exoplanet transit during the course of the observing run.
Many of the stars did exhibit changes in brightness during the run, but none of them showed the
characteristic pattern of a typical exoplanet transit. The brightness changes tended to occur
among groups of stars indicating that some other factor was causing them.
ACKNOWLEDGEMENTS
Thanks to the following groups and individuals who helped make this project possible:
Dr. Ray Bloomer, King College, Bristol Tennessee
Dr. Katy Garmany, NOAO, TOP program, Tucson, Arizona
Dr. Jack Rhoton, East Tennessee State University, Johnson City, Tennessee
Mr. and Mrs. Buehrig; Mr. and Mrs. Redmon.
Sullivan South High School Academic Booster Club.
Joe Smith, Anthony Peters, Wendy Ratliff, Martha Rhoton, and Geri St. Clair, Sullivan South
High School
The RBSE Journal
64
2008 V2
TABLES
Table 2. These stars were found to have saturated the CCD detector and so their true color
indices and brightness changes could not be determined. They were omitted from the study.
1
5
6
7
12
13
19
47
50
56
63
66
70
86
94
98
102
105
116
137
145
151
153
156
167
168
171
188
189
192
194
196
209
Table 3. These stars were found to be of spectral types O, B, or A. Stars of these spectral types
are very energetic and are not thought to have planets. They were omitted from the study.
8
34
62
85
111
153
188
204
9
36
65
87
112
155
189
205
11
37
67
88
113
156
191
206
12
37
76
91
115
167
192
208
18
48
77
92
118
168
193
21
51
79
93
139
169
194
23
53
80
97
140
177
195
27
55
81
101
142
178
196
28
59
82
103
146
179
197
30
60
83
107
151
180
199
32
61
84
109
152
184
200
Table 4. These stars were omitted from the study due to lack of data— telescope pointing errors
resulting in them being out of the field of view of the CCD in over half of the images.
124
131
125
132
126
133
127
134
128
135
129
136
130
141
Table 5. These stars were believed to be outside of the cluster NGC 957.
93
35
73
184
The RBSE Journal
157
150
46
146
142
5
83
166
65
39
164
118
68
2008 V2
REFERENCES
Comins and Kaufman. Discovering the Universe, 6th ed. 2003.
W. H. Freeman and Company, New York.
Kaler, Jim. “The Planet Project,” accessed October 3, 2005.
<http://www.astro.uiuc.edu/~kaler/sow/51peg.html>
Bisnovatyi-Kogan, G. S. “Planetary System around the Pulsar PSR: 1257+12,” Astronomy and
Astrophysics V.275, No. 1/Aug (I), (161, 1993)
<http://articles.adsabs.harvard.edu/full/1993A%26A...275..161B>
Linda Vu. October 3, 2006 NASA Spitzer Space Telescope Newsroom. Retrieved on March 18,
2008. <http://www.spitzer.caltech.edu/Media/happenings/20061003/>
David Charbonneau & Timothy Brown, et. al. “When Extrasolar Planets Transit Their Parent
Stars,” March 14, 2006, accessed March 18, 2008. <http://fr.arxiv.org/PS_cache/astroph/pdf/0603/0603376v1.pdf>
<http://www.nature.com/nature/journal/v378/n6555/abs/378355a0.html>
A. Wolszczan & D.A. Frail. Nature, January 9, 1992 (355, 145 – 147); doi: 10.1038/355145a0
<http://www.nature.com/nature/journal/v355/n6356/abs/355145a0.html;jsessionid=207CF901D
35BB7766FD61C4129DCFFE2>
Kerri Malatesta. Variable Star Of The Season Fall 2004: The Transiting Exoplanets HD 209458
and TrES-1. <http://www.aavso.org/vstar/vsots/fall04.shtml>
Robert Naeye. Amateur Detects Exoplanet Transit, September 3, 2004, Sky & Telescope.
Accessed March 24, 2008.
<http://www.skyandtelescope.com/news/3309506.html?page=1&c=y>
JPL/NASA Planetquest. Accessed March 26, 2008. <http://planetquest.jpl.nasa.gov/>
Gary, Bruce. “Exoplanet Observing for Amateurs.” Mira Digital Publishing, St. Louis, MO.
2007. <http://brucegary.net/book_EOA/EOA.pdf>
Swinburn University. “Interstellar Reddening.”. Accessed March 25, 2008.
<http://astronomy.swin.edu.au/cms/astro/cosmos/I/Interstellar+Reddening>
University of Washington. “Cluster Color Magnitude Diagrams and the Age of Stars.” May 16,
2005. Accessed March 31, 2008. <http://www.washington.edu/labs/claeringhouse/labs/
Clusterhr/cluster.html>.
Clemons, Christie and Rachel Reece. “The Age and Distance of the Open Cluster NGC 2345.”
RBSE Journal. 2007. 27 March 2008.
<http://www.noao.edu/education/arbsefolder/files/rbsejournal07.pdf>
The RBSE Journal
66
2008 V2
Gimenez, A. & Garcia-Pelayo, J., “RGU photometry of the young open cluster NGC 957.”
Astronomy and Astrophysics Supplement Series, vol. 41, July 1980, p. 9-12.
<http://articles.adsabs.harvard.edu/full/1980A%26AS...41....9G>
The RBSE Journal
67
2008 V2
The Short-term Variation of the Transneptunian Binary Objects (42355)
Typhon and (90482) Orcus
Rebecca Jensen-Clem1 and Jacob Shenker2
International Community School, Kirkland, WA
2
Gunn High School, Palo Alto, CA
Teacher: Donald W. McCarthy, RBSE 1996-TLRBSE 2002
1
ABSTRACT
On the nights of 2007 January 28 and 29 we observed the transneptunian binary (TNB) objects
(42355) Typhon and (90482) Orcus on the WIYN 0.9-m telescope. We observed a slight
dimming in (42355) Typhon, which could be caused by a number of different factors such as
shape, the presence of an eclipsing companion, or brightness variations on the surface of one of
the bodies in the system. The magnitude of (90482) Orcus, however, remained constant,
suggesting that its primary body is more spherical than that of (42355) Typhon, or that none of
the effects listed above are observable over the course of two nights.
INTRODUCTION
The Kuiper Belt is a debris disk of material extending from the orbit of Neptune to about 50 AU,
with some objects in highly elliptical orbits extending to hundreds of AU in the Scattered Disk
component. The Kuiper Belt is believed to be the source of short period comets and Centaurs.
Because they are so distant from the Sun, Kuiper Belt Objects, or Transneptunain Objects
(TNOs), are in “cold storage,” and their compositions provide a record of the early solar system.
The high incidence of binary objects in the Kuiper Belt provides observers with an additional
window of opportunity (Stansberry et al, 2006); binary objects allow us to measure the mass of
the binary system, whereas to obtain the same measurements for assumed single TNOs we would
need to visit them with a spacecraft. Calculations of Transneptunian Binary Object (TNB)
masses/densities along with spectrophotometric data allow observers to discover the main
constituents, ices, of these bodies. If they are dense, we assume them to be primarily rocky,
probably formed in the inner solar system and flung into the outer solar system by the gas giants,
but if they are primarily icy then they probably formed in their current region. The latter scenario
implies that the compositions of TNOs have remained unchanged since the beginning of the solar
system, while the former suggests a greater amount of processing in the inner solar system.
The formation of TNBs is likely different from other binary objects in the solar system because
when compared, TNBs have very similar mass ratios and wide separations, while main belt and
near earth binaries tend to be closely spaced with companions that are only a few tens of percents
of the primary in diameter. Since the discovery of the first TNB in 2001, several different
hypotheses have attempted to explain their formation. Because physical collisions in the Kuiper
Belt are too rare today to account for their observed frequency, TNBs were most likely born
during the formation of the solar system (Nazzario et al, 2005). Two main hypotheses have been
proposed, one which relies on physical collisions and one which suggests only gravitational
perturbations. Goldreich et al (2002) suggest two ways in which three or more gravitationally
interacting bodies could create the stable binary systems observed today: after an unstable binary
is formed by two objects entering one another’s Hill sphere, either dynamical friction from
The RBSE Journal
68
2008 V2
nearby smaller bodies or the close approach of a third, larger body could cause the system to lose
energy and thus become a stable binary system. S. J. Weidenschilling (2002), however, argues
that in the very early solar system physical collisions between planetesimals combined with the
gravitational effects of a third body could produce the binaries we observe today. Since the
publication of those two hypotheses in 2002, observations of TNBs have mostly supported the
conclusions of Goldreich et al, but more observations are needed to determine the exact
mechanism by which gravitational interactions could produce binary systems such as (42355)
Typhon and (90482) Orcus. It is also possible that observational bias has lead us to the
conclusion that most TNBs are similar in mass and widely spaced because such objects are much
easier to detect than smaller satellites orbiting very close to their large companion. The detection
of fainter and smaller satellites may require higher resolution data, but the detection of many
such average binary objects could indicate that Kuiper Belt binary formation occurred in much
the same way as that of the main belt and other systems.
A lightcurve is a graph of object brightness versus time, and is a common technique used by
astronomers to study TNOs. Very occasionally, the two objects in a binary system will eclipse
one another, and this behavior is evident in the lightcurve as we see larger and smaller dips that
represent the eclipses of the larger and smaller objects in a system. Because of the wide
separation between the bodies of a TNB, however, they rarely appear to eclipse each other from
the perspective of the Earth, and nonlinearalities in the lightcurve indicate brightness variations
on the surface of the objects or suggest that they are not perfectly spherical in shape. Once an
object’s exact shape has been determined, we can make assumptions that allow us to infer its
density if we don’t already have mass information.
In this paper, we present photometric data on the TNBs (42355) Typhon and (90482) Orcus that
can be used in future TNB studies to better understand their internal structures and in time to
help place constraints on Kuiper Belt formation theories.
OBSERVATIONS AND DATA REDUCTION
On the nights of 2007 January 28 and 29, we observed (42355) Typhon and (90482) Orcus on
the WIYN 0.9-m, using a standard R filter throughout the observing run. We began our
observation at 0800 UT both nights, and continued until 1300 UT, taking exposures of 10 to 15
minutes, and switching between the two objects several times throughout the night.
We selected six comparison stars for each data set, all of which were visible in each image and
had magnitudes similar to those of the target. We used MaxIm DL’s photometry tool to define
annuluses around the target and comparison stars in each image, identifying the target object in
each image individually. MaxIm DL’s photometry tool uses the following general equation to
obtain the target’s instrumental magnitude:
[Caperture − ηaperture ] [Cannulus /ηannulus ] + Z
m = −2.5log
t
(Eq. 1)
where Caperture is the total number of counts in ADU in the aperture 4 is the total number of
counts in ADU and η refers to the number of pixels in the annulus. The exposure time in seconds
(
The RBSE Journal
)
69
2008 V2
is t, and Z is the magnitude zero point, which MaxIm DL calculates based on the intensity of the
reference star in a given image and the exposure time.
After obtaining these preliminary instrumental magnitudes for each object and reference star, we
needed to further ensure that the targets’ variations were not due to the variability of the
reference stars
ΔReference Star = Reference Star 1 − Reference Star 2 themselves. To
accomplish this, we
subtracted the instrumental magnitudes of two of the reference stars:
ΔTarget = Reference Star 1− Target
(Eq. 2)
Next, we subtracted the target’s instrumental magnitude from that of the first reference star:
Table 1
Average ΔRef (in
Typhon frames)
0.43 +/- .01
Average ΔOrcus
Night1
Average
ΔTyphon
1.50 +/- 0.01
-1.20 +/- 0.01
Average ΔRef (In
Orcus frame)
-0.78 +/- 0.004
Night2
1.40 +/- 0.02
0.46 +/- .01
-1.20 +/- 0.02
-0.78 +/- 0.01
Night1-Night 2
(Absolute Value)
0.11 +/- 0.02
0.03 +/- 0.01
0.01 +/- 0.02
0.01 +/- 0.01
(Eq. 3)
Thus while ΔReference Star remains constant, any change in ΔTarget is probably not due to the
variation of the reference stars.
In order to determine if the brightness variations we observed were statistically significant
between the two nights of our observations, we averaged the values of ∆Target and then ∆Ref for
each night (see table below). Our data was then reduced to one data point per object per night.
We determined that the difference between the averages of ΔTyphon is about five standard
deviations, and therefore we can conclude with over 99.99% confidence that ΔTyphon’s
brightness variation is probably due to changes in brightness of the object itself, and not due to
external sources of error. The opposite can be concluded with regards to ΔOrcus, because we
calculated only .3 deviations between the two nights, and therefore we did not observe any
statistically significant variations in (90482) Orcus’ brightness.
The RBSE Journal
70
2008 V2
Figure 1: DeltaTyphon
-1.2
2E+06 2E+06 2E+06 2E+06 2E+06 2E+06 2E+06 2E+06 2E+06
-1.25
2454129.979
2454128.964
-1.3
DeltaTyphon
-1.35
DeltaRef in Typhon
frames
-1.4
2454129.979
-1.45
-1.5
2454128.964
-1.55
Average Julian Date
Figure 1. – The instrumental magnitudes of ΔTyphon are plotted with the difference in
instrumental magnitudes between two reference stars, showing that ΔTyphon’s magnitude
changes more than that of the comparison stars over the course of two nights. The ΔRef values
have been offset by -1.7 magnitudes in order to compare them more easily to ΔTyphon.
Figure 2: DeltaOrcus
-0.9
2E+06 2E+06 2E+06 2E+06 2E+06 2E+06 2E+06 2E+06 2E+06
-0.95
2454128.945
2454129.916
-1
-1.05
DeltaOrcus
-1.1
DeltaRef in Orcus
frames
-1.15
2454129.916
-1.2
2454128.945
-1.25
-1.3
Average Julian Date
Figure 2. – The instrumental magnitudes of ΔOrcus versus Julian Date are plotted with the
difference in instrumental magnitudes between two reference stars, showing that its lightcurve is
flat over the course of two nights. The ΔRef values have been offset by -0.2 magnitudes in order
to compare them more easily to ΔOrcus.
DISCUSSION
The RBSE Journal
71
2008 V2
Because we did not observe any statistically significant variability in (90482) Orcus, its
lightcurve appears to be flat over the course of two nights. A flat lightcurve could mean that the
primary body of (90482) Orcus is spherical in shape, or that it rotates very slowly and thus
brightness variations due to any irregularities in shape cannot be seen in only two nights. If the
period, however, is close to twenty-four hours, we may be observing the same point in the
lightcurve on two different nights. If that is the case, (90482) Orcus may still have features of
shape and surface brightness that can be detected with more frequent observations.
(42355) Typhon’s lightcurve is not flat because we observed a statistically significant change in
brightness between the two nights of our observations. A change in brightness over such a short
period of time indicates that the primary body of (42355) Typhon could be more potato-shaped
than spherical, appearing brighter when its longer side is visible from Earth and dimmer when it
rotates to show its smaller end. It is also possible that brightness variations on (42355) Typhon’s
surface could have caused such a dimming, or that a third object in the system eclipses one of the
known members of the binary pair. More observations are needed to determine the exact period
of the object and the specific nature of its shape and surface features.
SUMMARY AND ACKNOWLEDGEMENTS
In summary, the flat nature of (90482) Orcus’ lightcurve over the course of two nights suggests
that its primary body is fairly spherical, or that it is not oblong or irregularly colored enough to
yield an observable lightcurve . The slight change in brightness of (42355) Typhon indicates
irregularities in shape or surface brightness, or the presence of the third object in the system.
We would like to thank Dr. Don McCarthy and Dr. Susan Benecchi for their dedication and time
throughout this project. We would also like to thank Dr. Katy Garmany for making our visit to
Kitt Peak completely unforgettable and Dr. David Trowbridge for his kind support and help with
software and data reduction.
Appendix Table 1: (42355) Typhon Individual Observations
Instrumental magnitude
Observation Number
Julian Date1
1
2454128.86000
-1.50 +/- .03
2
2454128.87000
-1.50 +/- .03
3
2454128.96600
-1.50 +/- .03
4
2454128.97900
-1.60 +/- .03
5
2454129.01200
-1.50 +/- .03
6
2454129.02600
-1.50 +/- .03
7
2454129.03900
-1.50 +/- .03
1.50 +/- 0.01
Average of Night 1
2454128.96400
8
2454129.93300
-1.40 +/- .04
9
2454129.94900
-1.40 +/- .04
10
2454129.96200
-1.40 +/- .04
11
2454130.01200
-1.40 +/- .04
1
Not light time corrected
The RBSE Journal
72
2008 V2
12
Average of Night 2
2454130.03700
2454129.97900
-1.50 +/- .04
1.40 +/- 0.02
Appendix Table 2: (90482) Orcus Individual Observations
Observation Number
Julian Date2
Instrumental Magnitude
1
2454128.91100
-1.20 +/- .02
2
2454128.92400
-1.20 +/- .02
3
2454128.93700
-1.20 +/- .02
4
2454128.95200
-1.20 +/- .02
5
2454128.99900
-1.20 +/- .02
Average of Night 1
-1.20 +/- 0.01
2454128.94500
6
2454129.84700
-1.30 +/- .05
7
2454129.86000
-1.20 +/- .05
8
2454129.87400
-1.20 +/- .05
9
2454129.88600
-1.20 +/- .05
10
2454129.89900
-1.20 +/- .05
11
2454129.97400
-1.20 +/- .05
12
2454129.98700
-1.10 +/- .05
13
2454130.00000
-1.20 +/- .05
Average of Night 2
-1.20 +/- 0.02
2454129.91600
REFERENCES
Goldreich, P.; Lithwick Y.; Sari R. 2002. “Formation of Kuiper-Belt Binaries by Dynamical
Friction and Three-body Encounters.” Nature (London), 420, 643-646.
Nazzario, R. C.; Orr, K.; Covington, C.; Kagan, D.; Hyde, T. W. 2007. “Kuiper Binary
Formation.” Advances in Space Research, 40, 280-283.
Stansberry, J. A.,1 Grundy, W. M., Margot, J. L., Cruikshank, D. P., Emery, J. P., Rieke, G. H.,
and Trilling, D. E. 2006. “The Albedo, Size, and Density of Binary Kuiper Belt Object (47171)
1999 TC36” The Astrophysical Journal, 643:556 – 566.
Weidenschilling, S. J. 2002. “On the Origin of Binary Transneptunian Objects.” Icarus, 160, 212.
2
Not light time corrected
The RBSE Journal
73
2008 V2