Download Measuring distances to the edge of the local group

NOAO Observing Proposal
standard proposal
Date: September 27, 2007
Panel:
For office use.
Category: Resolved Galaxies
Measuring distances to the edge of the local group
PI: Colin Slater
Status: U Affil.: Case Western Reserve University
10900 Euclid Ave., Cleveland, OH 44106
Email:
Phone:
FAX:
Abstract of Scientific Justification (will be made publicly available for accepted proposals):
We propose to measure the distance to the nearby galaxies Sextans A & B and NGC3109 by
searching for resolved Cepheid variables. These distances will be used to determine if the selected
galaxies can reasonably considered part of the local group, or if they should be classified as a
distinct group. We will also use these distances to see if the galaxies are gravitationally bound as
a group, or if they are bound to the local group.
Summary of observing runs requested for this project
Run
Telescope
Instrument
1
2
3
4
5
6
CTIO-4m
Mosaic
No. Nights
Moon
Optimal months
Accept. months
6×.5
dark
Apr - Sep
Apr - Sep
Scheduling constraints and non-usable dates (up to four lines).
NOAO Proposal
Page 2
This box blank.
Scientific Justification Be sure to include overall significance to astronomy. For standard proposals
limit text to one page with figures, captions and references on no more than two additional pages.
We propose to measure the distance to the nearby dwarf irregular galaxies Sextans A, B, and NGC
3109 using Cepheid variable stars. These stars have a well-known period-luminosity relationship
caused by the radial pulsation of the star, which was first characterized by Leavitt in 1912. They
can thus be used as standard candles once the period of a particular star is known. To accomplish
this, we plan to take six observations of the three galaxies, with a spacing of 4-5 nights between
observations. Since the period of Cepheids can range from 1-40 days (Binney and Merrifield 1998),
our timing should be adequate to detect enough Cepheids for a reliable distance determination (We
discuss this in depth in the experimental design.)
We have chosen these galaxies for observations because there is some debate as to whether or not
Sextans A & B and NGC3109 are members of the local group or not (van den Burgh 1994, 1999).
Binney and Merrifield (1998) suggest that there are no certain criteria for local group membership,
and simply acknowledge a consensus list of 35 galaxies, including our three targets. They also
present the idea that all local group galaxies should be dynamically bound to the group itself, but
concede that we do not know which galaxies are bound due to observational constraints.
Van den Burgh, however, proposes that galaxies found to be outside of the zero-velocity surface
should not be considered local group members. It is at this radius that the gravitational effects from
the group and the Hubble flow are balanced. He calculates the radius of the zero-velocity surface to
be 1.18 ± .15 Mpc, which is very close to the distance to our selected galaxies. By measuring the
distances more accurately, we hope to better assess whether or not they should be considered part
of the local group. When combined with measurements of radial velocities (Fouqué et al. 1990,
de Vaucouleurs 1991), we can use these distances to determine whether these galaxies experience
greater effects from the Hubble flow versus the local group potential out near the theoretical zerovelocity surface. It is not possible to determine the space-velocity of the galaxies, since they are
too distant to have any measurable parallax, but the distances and radial velocities will still be
useful in constraining the potential of the local group.
Van den Burgh also suggests that Sextans A and B, Antlia, and NGC 3109 is part of a distinct
group he calls Antlia-Sextans. We hope to use Cepheid distance measurements to determine to
what degree this group is gravitationally bound. Tully et al. (2006) suggest that the AntliaSextans group is not in dynamical equilibrium, but is sufficiently massive (1.9 × 1011 Msun ) to be
stable against tidal disruption. This adds yet more evidence for Antlia-Sextans being a distinct
group.
The dynamical questions about the group all depend on the masses of the target objects. We can
use the absolute magnitude and assumptions of the mass to light ratio to determine the luminous
mass of the galaxies. The obvious constraint on this measurement will be the galaxies’ dark matter
halos, which will need to be inferred by later measurements of the rotation curves of the galaxies.
All of these questions depend on an accurate distance to the members of the Antlia-Sextans group.
We expect that our distance measurements, when combined with radial velocity data, will be able
to constrain the relation of the system to the local group.
References
Alcock, C. et al. 1995 AJ, 109, 1653.
Alibert, T. et al. 1999 A&A, 344, 551.
NOAO Proposal
Page 3
This box blank.
Binney, J., & Merrifield, M. 1998, Galactic Astronomy (Princeton: Princeton Univ. Press).
de Vaucouleurs, G., et al. 1991, Third Reference Catalogue of Bright Galaxies (Springer-Verlag).
Dohm-Palmer RC, et al. 1997. AJ, 114, 2527.
Fouqué, P., Durand, N., Bottinelli, L., Gouguenheim, L., Paturel, G. 1990, A&AS, 86, 473.
Leavitt, H., Pickering, E., 1912 Harvard College Observatory Circular, 173, 1.
Lee, Myung G. 1993 ApJ, 408, 409.
Piotto G, Capaccioli M. 1992, Memorie della Socieá Astronomica Italiana, 63, 465.
Tosi M, Greggio L, Marconi G, Focardi P. 1991. AJ, 102, 951.
Tully, R., et al. 2006 AJ, 132, 729.
van den Bergh 1994 AJ, 107, 1328.
van den Bergh 1999 ApJ, 517, 97.
NOAO Proposal
Page 4
This box blank.
Experimental Design
Describe your overall observational program. How will these observations
contribute toward the accomplishment of the goals outlined in the science justification? If you’ve requested
long-term status, justify why this is necessary for successful completion of the science. (limit text to one
page)
We will need observations of the three target galaxies spaced out by several weeks to properly
measure the period of the stars. Using the largest distance modulus of any of the targets of 25.64 ±
.15 for Sextans B (Piotto 1994), we will be able to detect stars as faint as -4 in absolute magnitude
with the 4 meter telescope. This roughly corrosponds to stars with a period of 8-10 day at the very
minimum, since the stars will vary in magnitude and may drop below the limit of detectability. To
adequatly sample this variation, we would like to make observations approximately once every 4
days to prevent the effects of aliasing (The light curve of Cepheids is non-sinusoidal and hence the
Nyquist theorem does not apply, but it is a rough estimate). With this schedule, we also expect to
avoid Cepheids that are pulsating in harmonic modes. Alcock et al. (2004) estimated that 20% of
the Cepheids with periods less than 2.5 days are pulsating in harmonic modes. Limiting our search
to Cepheids in the fundamental mode should increase the accuracy of our analysis, at the cost of
increasing the exposure time and the timespan of observations required.
We are requesting six nights, since that will set the upper limit on the periods we can detect to
roughly thirty days. This means we will cover most of the middle-range of Cepheid periods, and
more importantly cover the periods that correspond to the luminosities we can detect.
We will be taking data in the B and V bands. This will let us use the period-luminosity-color
relationship, which was shown by Martil et al. (1997) to reduce the scatter in the calculated
magnitudes. In addition, we can use the average metallicities for the galaxies to further constrain
the period-luminosity relationship, as was show by Alibert et al. (1997). Metallicities have been
obtained for Sextans A by Dohm-Palmer et al. (1997), by Tosi et al. (1991) for Sextans B, and by
Lee et al (1993) for NGC 3109.
We should get images of 24th magnitude stars in B and V in 10 minutes, with mutiple exposures
combined to reject cosmic rays. Using only two filters will allow us to observe all three objects
in half of a night. All three objects are within 15 minutes in RA, which will ease scheduling with
other observations on the same nights.
We have decided on the Cerro Tololo 4m primarily for its mirror size and its location in the southern
hemisphere. While Sextans A and B have declinations of -4◦ and +5◦ , NGC 3109 is at -26◦ , and
thus requires a southern-hemisphere observatory. The mosaic on the 4m has sufficiently small pixels
that we should be able to clearly resolve stars in all the galaxies. Although WIYN would be capable
of imaging the Sextans galaxies, we feel that adding NGC 3109 to the target list would sufficiently
improve our results as to warrant observing at CTIO.
NOAO observing proposal LATEX macros v2.14.
NOAO Observing Proposal Standard Proposal
Date: September 19, 2007
Panel:
Category:
Astronomy 306
Galaxy Spectroscopy
Asymptotic rotation velocities of high mass galaxies
PI: Curtis Bunner
Status: S
Affil.: Case Western Reserve University
Department of Astronomy, 10900 Euclid Ave, Cleveland, OH 44106 USA
Email: [email protected]
Phone: ###-###-####
Abstract of Scientific Justification:
I propose a further investigation of the asymptotic rotation velocities of the group of galaxies
first examined by Noordermeer & Verheijen (2007) as a part of a larger project comparing the
three main types of velocity measurement of spiral galaxies: asymptotic (this proposal),
maximum velocity, and the width of the global HI profile. The Tully-Fisher relation provides an
empirical method for measuring cosmic distance scales, which provides an additional tool for
calibrating other standard candles (Cepheids, Type 1A supernovae, etc.). However, because the
Tully-Fisher (TF) relation is empirically derived (not based on theory), any errors or systemic
variations among the data have the potential to drastically affect interpretation of the TF relation.
To address this issue, I propose a more in depth look at a small subset of the data compiled in
Noordermeer & Verheijen (2007); galaxies that had significantly large errors in asymptotic
velocity measurements. The goal of this proposal is to tighten the correlation of mass/luminosity
with asymptotic velocity. The other two parts of this project will be to improve the maximum
galaxy velocity estimates as well as the global HI profile width (data provided by collaborators).
By gathering more data with smaller errors, the goal is to be able to further distinguish among
the three methods of velocity measurement: which provides the most linear relationship, and
what is the significance of the difference between the most linear and the other two?
Summary of observing runs requested for this project
Run
1
2
Telescope Instrument
No. Nights Moon Optimal Months Accept. Months
KP-2.1m Spectrograph 4
New Nov-Dec
October-Jan
KP-2.1m Spectrograph 2
New May-June
April-July
Scientific Justification
The Tully-Fisher relation describes a trend throughout all sizes of spiral galaxies (derived
theoretically for elliptical galaxies). This result was first published by R. Brent Tully and J.
Richard Fisher in 1977 in their seminal paper suggesting the relationship as a means of
calibrating distance measurements. Since then, it has been observed that for medium mass
galaxies (rotational velocities between 90kms-1 and 200 kms-1), the relation was very linear.
When very high and low mass galaxies were added to the relation, however, the relation seemed
to exhibit a “kink” where the slope of the line changed.
The question for scientists was puzzling. The TF relation seemed a very useful tool for
estimating the size of a galaxy based on its rotational velocity, but changed at certain mass limits.
Was this caused by an underlying physical phenomenon that we did not understand the
mechanics of? Or was it potentially a problem with our measuring techniques?
Noordermeer & Verheijen (2007) took an important step by asking: what could cause the
difference? Assuming that the luminosity/mass & rotational velocity are indeed correlated
linearly through any galaxy size, Noordermeer & Verheijen tried to determine if velocity
measurement methodology was potentially responsible. The setup for their comparison was to
measure the maximum rotation velocity of a galaxy (the inner part of the disk of the galaxy), the
asymptotic rotation velocity (the very outer part of the disk), and the global HI profile width
(how the Hydrogen I emission lines are spread from the dual redshift/blueshift of the rotating
galaxy). (See Figure 1)
Noordermeer & Verheijen (2007) set out to find the answer to the question “which of the three
velocity measurements generates the most linear Tully-Fisher relation?” The conclusion of their
investigation, based on empirical findings, was that the asymptotic velocity yielded the most
linear data trend. This does not rule out the possibility that something in the physics has
changed, but rather shows that using certain measuring techniques, a more linear TF relation can
be found for varying galaxy sizes.
The central question that initiated this project was whether the TF relation really did become
non-linear with size changes (and linearization found by asymptotic velocity measurements is a
misleading conclusion), or if TF was linear throughout the range and asymptotic velocity
measurements most accurately represented it. The implication of an asymptotic velocity
measurement giving more accurate results is the dark non-baryonic matter component of the
galaxy that contributes the most to a galaxy’s rotational velocity at large radii.
The observations I seek to accumulate through this project have two purposes. The first is to
obtain more accurate asymptotic velocity measurements to improve the currently existing
models. The second is to use the improved model of the TF relation using different velocity
measurement methods to resolve in greater clarity what differences might exist (and how
significant those differences are).
The TF relation provides an exceptionally useful purpose for further exploration. Many of the
current standard candles of astronomy such as Cepheids and RR Lyrae variable stars are only
visible in nearby galaxies. Once we try to use such stars to gauge distances to galaxies too far
away to resolve individual stars in the galaxy, these distance measurements become impossible
to use. Supernovae of Type 1a provide a distance measurement method that can be traced to
much further distances, but are rare occurrences that have the potential to skew the selection of
galaxies measured.
The TF relation does not suffer from these limitations of distance, but instead must confront the
problem of origin: the relationship used to gauge distances to empirically sampled galaxies is
itself derived from empirical observation. The potential for feedback between the TF relation
and the galaxies being observed is worrisome from a scientific standpoint, as there is no truly
objective standard being applied. For example, a group of massive galaxies judged to have
unusually slow asymptotic rotational velocities would drag the TF trend line down, potentially
biasing the distance measurement.
Projects such as that by Noordermeer & Verheijen (2007) perform the important task of trying to
isolate causes of variation, in an attempt to shore up the theoretical predictions for what the TF
relation should be before empirical grounding takes place. It is my goal to add to this larger
project in astronomy to improve a potentially powerful tool for distance measurements.
Figures
From Noordermeer & Verheijen (2007)
References
Noordermeer E., van der Hulst J. M., Sancisi R., Swaters R. A., van Albada T. S., 2007,
MNRAS, 367, 1513 (N07)
Noordermeer E., Verheijen M. A. W., 2007 MNRAS, (accepted for publication)
Spekkens K., Giovanelli R., 2006, AJ, 132, 1426 (S06)
Tully R. B., Fisher J. R., 1977, A&A, 54, 661
Technical Description
Choice of galaxies
I chose galaxies that I would be able to improve upon in terms of the accuracy of the velocity
measurements, which will be instrumental in determining which of the three methods of
measuring galaxy rotation velocities (vmax, vasmp, H1 profile width) is the best choice.
The list of velocity measurements listed in N & J (2007) gives the velocity by each of the three
measurements, as well as the error (±). For this proposal, I will need to be able to gather spectra
with velocity accuracies greater than those already obtained in previous studies to justify
gathering new data. Given my choices of telescopes, the smallest error I will be able to have is
±15 km/s. So, this leaves only 10 galaxies that have asymptotic rotation velocities with errors
greater than this.
The KPNO 4m setup 1 is ruled out from the start because neither of the two signal to noise
options give velocity accuracies better than previous studies. Both the KPNO 2.1m setup and the
KPNO 4m setup 2 have similar velocity accuracies. I would need to be able to achieve a
velocity accuracy of 15 km/s with the detector. Both setups have this option, so in either case I
would have to have a S/N ratio of 20. Having an accuracy of 15 km/s will allow me to improve
upon the current data gathered on the list of galaxies above.
The 10 galaxies I have chosen all fall within the range between 92 and 276 arcsec of angular
size. This means that I will be able to use the KPNO 4m setup 2, as the angular size of the
galaxies will all fit in the telescope's 300" FOV and still have some sky visible on the edges.
Observing comets as chemical tracers of the
early Solar System
Ethan Engle
Department of Astronomy, 10900 Euclid Ave, Cleveland, OH 44106 USA
Email: [email protected]
Phone: 231 590-0192
Abstract of scientific justification
The formation of the solar system from the solar nebula has been notoriously difficult to model.
The solar system is unique compared to other planetary system that we have observed, most of
which have massive Jupiter-sized objects orbiting very closely to the host star. The prominent
formation models tend to predict that volatile chemicals were expelled from the inner solar system
as the inner planets were forming. I intend to observe comets from two families, particularly the
main belt comets and nearly-isotropic long-period comets, to trace the chemical makeup of the
early solar system. Comets from both of these families may have transported volatile chemicals to
the early solar system after the inner planets had formed.
Two main formation models have been proposed. Wetherill (1980) proposed that planetesimals
formed primarily from the agglomeration of dust within the solar nebula. Herndon (2004) showed
that this model did not produce sufficiently massive planetary cores, so he proposed a model in
which the cores of the inner planets condensed within massive gaseous envelopes similar to
present-day gas giants. Solar wind from the Sun’s thermonuclear ignition and temperature
variations dependant on heliocentric distance pushed volatile chemicals to the outer solar system,
leaving iron and magnesium silicates behind in the inner solar system. The solar system may have
formed through an interplay of these two models. Both predict volatile chemical variation with
heliocentric distance.
The two comet families mentioned above occupy different heliocentric distances, and therefore
different temperature and volatile chemical regimes. When passing near earth, these comets rich
in volatile chemicals may have deposited the necessary ingredients for the early oceans and
organic compounds to develop.
Summary of observing runs requested for this project
Run
Telescope
Instrument
No. Nights
Moon
Optimal Months
Accept. Months
1
KPNO 4-m
Setup 2
0.5
half
Dec ‘04 - Jan ‘051
Nov ‘04 - Feb ‘05
2
KPNO 4-m
Setup 1
1
darkest
Dec ‘05
1
Nov ‘05 - Jan ‘06
3
1
We will need to turn back the clock here because this is when the comet made its closest
approach to earth.
Scientific Justification
The conditions within the solar nebula which led to the formation of the solar system are not well
understood, especially considering that our solar system does not resemble many others that have
been have observed (Beer et. al. 2004). In many other observed planetary systems, gas giants
tend to orbit much closer to the central star, preventing small terrestrial planets from forming in
the inner solar system. Two main models have surfaced in the past thirty years which explain how
planetesimals formed within our solar system and why we observe different objects with different
compositions in the inner and outer solar system.
The standard model for planetary formation became popular in the 1960s and was fully developed
by Wetherill (1980). In this model, rocky planetesimals formed as small dust grains agglomerated
within the solar nebula. Within the inner solar system, where the temperatures are cool enough,
iron and magnesium silicates as well as metallic iron could crystallize to form these small grains.
These minerals account for much of the Earth’s composition. The inner solar system was
probably too hot to allow more volatile elements such as hydrogen, carbon, and nitrogen to
condense. As a result, these elements are only observed in solid form in the icy bodies of the
outer solar system (Wetherill 1990).
While the standard model has gained wide acceptance, a few observers, notably Herndon (2004),
have pointed out its flaws. The accretion process in the standard model would not form planetary
cores that are as massive as those that have been observed. These massive cores must contain
enstatite chondrites which can only form at high temperatures and pressures which do not occur
in the standard model. Instead of forming entirely in low pressure environments, the inner planets
must have each formed mostly within thick gaseous envelopes similar in size and composition to
the gas giants. Rocky material would have condensed within this dense gas and rained down to
the center where temperature and pressure were extreme (Herndon 2004). At one point, the Sun
produced an outburst of luminosity and solar wind which blew away the gaseous envelopes that
surrounded the rocky cores of the inner planets, leaving only nonvolatile chemicals (Herndon
2006). This outburst probably resulted from the thermonuclear ignition of the Sun and it would
have separated different volatile chemicals according to heliocentric distance (Figure 1). It is
likely that the solar system formed through an interplay between this model and the standard
model.
I will attempt to use comets to understand the conditions within the young solar nebula as the
planets were forming. Comets are important tools for understanding the composition and
evolution of the early solar system because they preserve the chemical makeup of their
environment at the time and location they were formed (Crovisier 2007, A’Hearn et. al. 2002).
Crovisier (2007) suggests that comets can be divided into two broad categories related to their
orbital properties and area of formation. These two categories are nearly-isotropic and ecliptic
comets. Nearly-isotropic comets are so-called because their orbital inclinations seem to be widely
distributed. These comets have very long orbital periods (> 200 years) and large semi-major axes
ranging from ~100 - 10,000 AU, so they are also commonly called long-period comets. They
originate from the Oort cloud which extends up to 10,000 AU from the Sun (Fernández 1997).
Ecliptic comets have much shorter periods (< 20 years) and they tend to orbit within the plane of
the solar system. Jupiter strongly interacts with these comets so that their semi-major axes are
nearly equal to that of Jupiter. They probably originate from an area called the scattered disc
which is separate from the Kuiper Belt. The scattered disc exists outside Neptune’s orbit and may
extend out to the border of the Oort cloud (Crovisier 2007).
It is not clear exactly how a comet’s place of formation contributes to its chemical makeup.
Other factors such as its orbit may have a stronger influence on its chemical makeup. A’Hearn
(2002) notes chemical differences between short- and long-period comets, particularly C2
abundance, and argues that this discrepancy is due to their different places of formation; he argues
that short-period comets must have formed in the Kuiper Belt and long-period comets formed
within the orbit of Neptune where they were gravitationally ejected by the gas giants to the Oort
cloud. If comets formed at different heliocentric distances, then they must contain different
abundances of volatiles like C, N, and CO. But Crovisier (2007) shows that short-period comets
could not have formed in the Kuiper Belt because the orbits of Kuiper Belt Objects are too stable.
Instead, short-period comets formed together with long-period comets within the orbit of
Neptune before moving to the scattered disc. Any chemical discrepancies between these two
families are probably not due to differences in their place of formation.
According to Crovisier’s argument, the only difference between short- and long-period comets is
the location to which they moved after formation; differences in observed chemical makeup must
result from differences in the evolution of their orbits. The abundance of volatiles in short- and
long-period comets are similar which indicates that both classes of comets formed at similar
heliocentric distances. In addition, multiple close passes to the Sun made by short-period comets
does not seem to alter their chemical makeup. Therefore, comparing the chemical makeup of
short- and long-period comets would do little help us understand the structure and evolution of
the early solar system.
A recently discovered new class of comets will provide a promising sample and solution. Three
comets have been discovered in the main asteroid belt near the temperature region where water
turns to ice (Figure 2). They are very similar to their neighboring asteroids, though their release
of volatiles and dust is characteristic of comets. Computer simulations indicate that these objects
must have formed in the inner solar system instead of among the gas giants like the elliptical and
nearly-isotropic comets (Hsieh & Jewitt 2006). I will compare the chemical composition of a
main belt comet (P/2006 U1 Read) with a typical long-period comet (C/2004 Q2 Machholz) in
order to probe the makeup of the early solar system. I will look for different abundances of
volatile species which is related to the abundance in the solar nebula as the comet formed.
Observations of the chemical makeup of comets that occasionally pass within the inner solar
system are of particular interest because such comets may have contributed to the development of
organic compounds on earth. The young earth, having been stripped of its volatile elements
during the Sun’s thermonuclear ignition, may not have harbored the ingredients or conditions for
organic compounds to develop. Long-period comets rich in icy volatiles may have deposited the
necessary chemicals such at carbon, nitrogen, and water to produce these organic compounds
(Delsemme 1992). Additionally, main belt comets may have contributed enough water for the
early oceans to develop (Hsieh & Jewitt 2006). Observations of long-period and main belt
comets may provide clues about which comet family caused the origin of certain volatile species
on earth.
Figure 1: (Crovisier 2007) The snow lines for different cometary volatiles. The position
of the planets represent the current positions. The plot shows that main belt comets
would show evidence of water and perhaps some CH3OH and HCN. Long-period comets
may be abundant in all the species shown on the plot. These differences would be due to
the different locations of formation of the two comet families.
Figure 2: (Hsieh & Jewitt 2006) The three known
main belt comets as imaged by the University of
Hawaii 2.2-m telescope at Mauna Kea. P/2005 U1
has a mean R-band magnitude of 19.28. Exposure
time was 1.9 hours.
References
A’Hearn, M. F., Millis, R. L., Schleicher, D. G., Osip, D. J. & Birch, P. V. 1995, Icarus,
118,223-270
Beer, M. E., King, A. R., Livio, M. & Pringle, J. E. 2004, Mon. Not. R. Astron. Soc., 354, 763768
Delsemme, A.H., 1992, STI, 21, 279-298
Crovisier, J., 2007, arXiv:astro-ph/0703785v1 30 Mar 2007
Fernández, J.A., 1997, Icarus, 129, 106-119
Herndon, J.M., Solar System Processes Underlying Planetary Formation, Geodynamics, and the
Georeactor, 2006, Neutrino Geophysics
Herndon, J.M, Solar System Formation Deduced from Observations of Matter, 2004,
arXiv:astro-ph/0408151
Hsieh, H.H, & Jewitt D., 2006, Science, 312, 561-563
Wetherill, G.W., 1980, Ann. Rev. Astr. Ap., 18, 77-113
Wetherill, G.W., 1990, Ann. Rev. Earth Planet. Sci., 18, 205-56
Technical Description
Both comets will be observed near their points of closest approach to Earth. At their closest
approaches, C/2004 Q2 will be at ~0.3 AU from earth and P/2006 U1 will be at ~1.5 AU.
C/2004 Q2 will have a bright total magnitude, but its large angular diameter will limit its surface
brightness. Its radius may be as large as 360" and it may reach a visual magnitude of MV = 6.
Therefore, I can expect it to have a surface brightness of ì ~ 20 mag/sq. arcsec. P/2006 U1 will
be very faint with MV = 19, but its radius will be ~3". Its surface brightness will be ì ~ 22.5
mag/sq. arcsec.
Because of its low surface brightness, P/2006 U1 will require one night on the KPNO 4 mtr Setup
1 in order to obtain a spectrum with S/N of ~8.
C/2004 Q2 is bright enough to obtain a high-resolution spectrum with good S/N in one night with
the KPNO 4mtr Setup 2. A spectrum with S/N of ~10 can be obtained in one hour. Therefore, a
half night of observing will be required to gather data for C/2004 Q2. A waning crescent moon is
tolerable because the observations will need to be gathered early in the night before the object
reaches a low altitude.
Brown et. al. (1996) have demonstrated that high resolution spectra can be used to identify
cometary species such as H, O, C2, CN, NH2, C3, H20+, CH, and CH+. In the optical region,
these spectral lines are separated by a ~1-10 angstroms. Therefore, the highest spectral resolution
is desirable. For the KPNO 4mtr Setup1, the spectral resolution is 5D/pixel. This should
sufficient to identify species like H20+ and HCN which I expect to find in the main belt comet
P/2006 U1. The KPNO 4mtr Setup 2 provides a spectral resolution of 1.5D/pixel which will help
to resolve the spectral lines from the large number of volatile species that I expect to find in
C/2004 Q2.
References:
Brown, M.E., Bouchez, A.H., Spinrad, H., Johns-Krull, C.M. 1996, Astron. J., 112, 1197-1202
1
NOAO Observing Proposal
Date: 25 September 2007
Standard Proposal
Panel: For Office use
Category: The Solar System
Observing Kuiper Belt Objects to Understand the Formation
of the Solar System
PI: Kathryn Rennie
Status: UG
Affil.: Case Western Reserve University
Department of Astronomy 10900 Euclid Ave, Cleveland, OH 44106 USA
Email: [email protected]
Phone: 814-440-9484
Abstract of Scientific Justification:
Kuiper Belt objects are believed to have formed in place. In addition, it is also the most
likely place for short-period comets to originate, and thus where protocomets can be
found. Thus, from the size distribution, and density as a function of radius, we can get a
much better picture of what the early solar system was like and how the planets formed.
Also, with the eccentricities and inclinations of the objects, we can better hypothesize
what excited the KBO’s in the early universe. Finally, looking again at the protocomets,
with their orbits, we can also track them to determine the likelihood of any one of them
interacting with earth.
Images will need to be taken of each possible KBO found twice in twenty-four hours to
eliminate the possibility that we are observing asteroids in the main belt. The right
ascensions and declinations will be taken of each object, and then again one year later.
Afterwards, the method of Bernstein and Khushalani (2000) will be used to procure their
semi-major axes, eccentricities, and inclinations. As there are several hypotheses to
explain what happened in the early solar system to create these perturbations from their
original orbits, this data will allow us to further deduce which is correct, and thus what
truly happened in the early solar system.
Summary of Observing Runs Requested for This Project
Run Telescope Instrument No. Nights Moon Optimal Months Accept Months
1 4m-Mosaic
14
All
All
2
SCIENTIFIC JUSTIFICATION:
The Kuiper Belt, which is defined as extending from 30 AU’s to 50 AU’s from the sun,
has been one of the most interesting objects in the solar system to astronomers since
1943. This was the year Edgeworth hypothesized that short-period comets may originate
from this area (Edgworth 1949). Then, in 1951, Gerard Kuiper extended upon this,
hypothesizing that beyond Pluto, we should not only find protocomets, we should find a
large number of them and as well as a large number asteroids in the formation of a ring
(Kuiper 1951). At first glance, a ring of protocomets and asteroids may not seem entirely
interesting, but in reality it is exceedingly exciting. This is because it is most likely a
processed remnant of the protoplanetary disk (Luu 2002).
For more than forty years, the Kuiper Belt remained a theory. In August of 1992,
however, the team of Luu and Jewitt found the first Kuiper Belt Object (KBO), QB1
(Luu 1993). It was found to have a diameter of about 200 km, an albedo of about 0.04
and a nearly circular orbit with a semi-major axis of 43 AU. Looking at these numbers,
we see its albedo is similar to that of comets. However, its diameter is much larger than
cometary nuclei, and thus it cannot be one (Williams 2001). Although QB1 itself is
much too large to be a protocomet, it’s discovery still helped to solidify the Kuiper Belt’s
existence and the theory that short-period comets originate from here and not in the Oort
Cloud like the long-period comets.
Proceeding under the assumption that it is a remnant of the protoplanetary disk, the
Kuiper Belt is believed to have formed in place. If this is true, studying its dynamical
structures can give us an edge in learning about the early conditions in the universe that
brought about the formation of the solar system. In order to do this, however, we first
must find more Kuiper Belt Objects (KBO’s).
To do this, many KBO’s must be imaged, with their right ascension and declination taken
twice in 24 hours make sure they are not asteroids from the main belt, but rather from the
Kuiper Belt (Bernstein 2000). This shows the difference between the two, as main belt
objects move faster due to the fact that they are closer to us. Next, we must come back
one year later and image the same KBO’s again to get their new right ascension and
declination. Once this has been completed, the method of Bernstein and Khushalani
(2000) can be used to find the objects semi-major axes, eccentricities, and inclinations.
This method has been used previously to find orbits of KBO’s, including in 2005 by
Brown, Trujillo, and Rabinowitz (Brown 2005).
As of 2003, 770 trans-Neptunian objects have been found. This may sound like a large
selection, but when compared to the entire area of the Kuiper Belt, this will cover a very
small fraction of it (Morbidelli 2003). Looking at these objects, we can further see that
KBO’s are not distributed evenly throughout. Rather, they seem to be clustered into three
distinct subgroups. These are Classical KBO’s, Resonant KBO’s, and Scattered KBO’s
(Luu 2002).
3
About two-thirds of known KBO’s are Classical. They have semi-major axes between
42<a<48 AU, eccentricities of about 0.1, and perihelia greater than 35 AU (Luu 2002).
Initially, it was suspected that Classical KBO’s were still on their primordial orbits.
However, it has since been discovered that they have excited orbits which can be
concluded from their broad eccentricity and inclination (measures the velocity dispersion)
distributions. The next type, Resonant KBO’s, have orbits that have integer resonances
with Neptune (Luu 2002). Due to these resonances, they have even higher eccentricities
and inclinations than Classical KBO’s. Pluto, for example, has a resonance of 3:2 with
Neptune, as do about 100 other KBO’s found so far (Luu 2002). Resonant KBO’s with
orbits that come near Pluto and Neptune can be thrown off these orbits. These unstable
resonant KBO’s are possibly part of the population for short-period comets. Finally,
scattered KBO’s differentiate themselves from the rest with their highly eccentric and
inclined orbits (Luu 2002). Their origin is unknown, though it is thought they may have
been scattered outwards by Uranus and Neptune (Levinson 1997).
Figure 1, below, shows the differences in eccentricities for Classical and Resonant
KBO’s at different orbital distances from the sun (Luu 2002). Many different hypotheses
have tried to explain what excited these objects, including scattering by Earth-sized
objects in the early Kuiper Belt (Morbidelli 1997), and partial trapping by sweeping 2:1
resonance (Hahn 1999). However, at this point, there is not enough data to determine a
definite source. Adding to this selection of KBO’s, will help to distinguish which
hypothesis may be the correct one, as each gives a different relationship for eccentricity
and inclination distributions (Figure 2). As this will tell us what the early solar system
was like, distinguishing between hypotheses is essential to the study of the Kuiper Belt
and the formation of the planets.
4
Figure 1: Semimajor axis a vs. eccentricity e for Classical and Resonant KBO’s. Filled circles denote
multi-opposition orbits, open circles denote orbits computed from astrometry taken within a single
opposition. The curve marks perihelion q = 30 AU; objects above the line are Neptune-crossers. Vertical
lines mark the locations of mean-motion resonances with Neptune. Orbits with e = 0 are assumed, not
accurately measure. (Figure and description from Luu and Jewitt et al 2002).
Figure 2: predicted orbital distributions of KBO’s: (a) distribution after stellar encounter (before sweeping
resonances), (b) distribution after including sweeping resonances. Figure from Ida et al. (2000b).
Looking at the Kuiper Belt, we can estimate based on the density of objects, its total
mass. This has been found to be about 1/10 the mass of earth (Morbidelli 2004).
However, based on the accretion models, to get the large sizes and size distribution of the
KBO’s seen, the total mass needs to be tens of earth masses (Morbidelli 2004). This is
because, as Stern found in 1995, with the density of the Kuiper Belt now, objects with
diameters of 100km and larger cannot create the number of larger KBO’s seen with
pairwise accretion, with the age of the solar system. In addition, the large eccentricities
and inclinations seen produce large velocities. Thus, they destroy one another instead of
creating larger KBO’s (Morbidelli 2003).
The data collected will be added to the existing data sets, and the true structure of the
Kuiper Belt will be further mapped out. As a side project, with this data, as we will be
tracking the orbits of these objects, we can also determine whether or not there is a
chance of any of them coming into contact with the earth.
REFERENCES:
Bernstein, G. and B. Khushalani, 2000, Astron J, 120, 3323.
Brown M.E., C.A. Trajillo, and D.L. Rabinowitz, 2005, Astrophysical Journal, 635:1.
5
Edgeworth, K. E., 1949, Royal Astronomical Society, 9, 10.
Kuiper, G. P., 1951, Proceedings of the National Academy of Sciences, 13.
Levison, H.F. and M.J. Duncan, 1997, Icarus, 108:18-36.
Luu, J. and D. Jewitt, 1993, IAU Circ, 1.
Luu, J. and D. Jewitt, 2002, Ann. Rev. Astr. Ap., 3-4, 27.
Williams, I. and A. Fitzsimmons, 2001, 10-13.
Morbidelli, A, and G.B. Valsecchi, 1997, Icarus, 128:464-68.
Morbidelli, A M.E. Brown, and H. F. Levinson, 2003, Earth, Moon and Planets, 92.
Morbidelli, A and H. F. Levinson, 2004, Nature, 1-3.
Hahn, J.M. and R. Malhotra, 1999, Astron J, 117:3041-53.
TECHNICAL DESCRIPTION OF OBSERVATIONS:
From Luu and Jewitt, September 2002, we see the range of apparent R magnitudes over
which KBO’s can be found is approximately 21-26. Assuming a signal-to-noise ratio of
20, we see the observing times are 30 seconds – 300,000 seconds = 0.5 minutes – 5,000
minutes. Thus we see for bright KBO’s, the exposure time is considerably less than the
exposure time needed for the faintest ones.
Next, the KBO’s most likely follow a differential power law size distribution,
n(r )dr = Γr − q dr , where n(r)dr is the number of objects found with a particular radius and
gamma and q are both constants. From this equation, we can get a rough estimate of the
number of objects at each size estimate in the Kuiper Belt to be approximately 10^10
objects with a radius > 1km, ~ 3*10^4 objects with a radius >100km, and ~10 with a
radius > 1000km (Levison 1994). Thus, as we can see, there are only a few extremely
large KBO’s, which leads to the conclusion that there are many more faint KBO’s than
bright ones.
The 4m-Mosaic telescope, with a field view of 36x36 arcminutes is needed to view the
KBO’s as it has a wide field-of-view as well as a limiting magnitude around 25. This is
still one magnitude brighter than some of the faintest KBO’s, but there should still be
plenty of objects brighter than this. Thus, this telescope will be able to supply the images
needed to obtain the orbits of the KBO’s. To obtain this data, with the exposure times
listed above, two weeks of observing time is needed in order to acquire a reasonable
number of Kuiper Belt objects as most of the ones that will be found will have apparent
magnitudes fainter than 21, and consequently much longer exposure times than thirty
seconds.
Looking at these magnitudes further, we see an exposure of a 25th magnitude object
would take 13.153 hours to complete. Due to the fact that these objects are faint, we
must observe them when it’s dark out. Thus, even though the limiting magnitude of the
telescope is near 25, we cannot observe that faint with a single exposure. Looking
further, we see a 24.5th magnitude takes 5.25 hours to complete one run. This is a much
more reasonable timeframe. Thus, this is approximately the faintest object we will be
able to observe.
6
A smaller telescope is unusable to view KBO’s as they are extremely small, and thus
won’t be visible. Also, as we don’t know ahead of time where the KBO’s are, a smaller
field of view will not help us. Rather, it would only increase the time required to find
each one.
ASTR 306 Observing Proposal
Date: September 26, 2007
PI: Laura Boon
Affil.: Case Western Reserve University
Department of Astronomy, 10900 Euclid Ave, Cleveland, OH 44106 USA
Email: [email protected]
Abstract of Scientific Justification:
I propose to take spectra of elliptical galaxies to measure their rotation
curves. Using an elongated slit spectrometer.
I will measure the velocity dispersion of stars in elliptical galaxies to
determine the amount of dark matter present. Knowing the concentration
of dark matter in elliptical galaxies I will compare it to the concentration of
dark matter in spiral galaxies. The concentrations of dark matter in
elliptical galaxies should support the theory that elliptical galaxies are
formed by the collision of spiral galaxies.
Observing Runs Requested:
Telescope Instrument
No.
Nights
Spectrometer 7 nights
Moon
Optimal Months Accept. Months
AngloAustralian
2.1-m
Darkest
May-August
May-August
Scientific Justification:
The formation of elliptical galaxies is still a highly debated topic. One theory
states that elliptical galaxies are formed through the collisions and mergers
of spiral and other galaxies (Bender). There is strong observational
evidence for this theory.
Easiest to see is the location of elliptical galaxies in clusters.
Approximately 80% of the galaxies in clusters are elliptical galaxies, while
80% of field galaxies are spirals (Kutner) Even out of the 80% of elliptical
galaxies the majority of them are found in the center of the cluster, with the
highest density of galaxies. This close proximity to one another makes it
natural that the galaxies would interact with each other.
More evidence for the formation theory is the presence of ripples in the
galaxies. These ripples were first observed by Malin in 1979(Kennicutt)
Since then ripples have been found in half of all elliptical galaxies
(Schweizer). As seen in Figure 1 ripples are incomplete rings around the
galactic center, created by a higher density of stars. The ripples could
have formed one of two ways; there were many smaller mergers each one
of these smaller mergers created one of the rings that is visible today. Or
the ripples might have been created when two galaxies of similar size
collided to form the elliptical we see today (Schweizer). Although there is
not agreement on how the ripples formed there is no question that they
formed though some galaxy interaction. Finally, there is evidence that the
core and outer stars rotate independently of each other (Franx). Efstathiou
discovered this strange phenomenon when he was looking at the rotation
curve of NGC 5813. He found that the core was rotating with at peak
velocity of 89 km s-1 just outside of the center this velocity drops to 8 km s-1
(Efstathiou). The difference in the velocities of the galactic core and
surrounding region is evidence of a collision that altered the rotational
velocity of the stars in the galaxy. When he looked closer in to this idea,
he found that there are many galaxies that have this property. The
frequency in which this happens shows that it is not just a special case in a
single galaxy.
Looking at the rotation curves of spiral galaxies we find that the rotation
velocity of stars does not decrease with distance as expected. According
to theory the velocity should decrease at a certain radius. This however
did not match the data taken by astronomers. A study done at the
Carnegie Institution found that rotational velocity of the stars remained
constant (Kutner). Using the relation v(r) = GM the concentration of dark
r
matter in spiral galaxies has been measured using the rotation curves of
the stars around the galactic center. These studies found that 90% of
spiral galaxies are dark matter (Chaisson). Since we know the amount of
dark matter in spirals we can use that to compare to the amount of dark
matter in elliptical galaxies to support the claim that elliptical galaxies are
formed though collisions of spirals. The concentration of dark matter in
elliptical galaxies should be comparable to that in spirals.
Once we have the spectrograph of each elliptical galaxy that information is
used to find ! or the velocity dispersion of the stars in the galaxy. Sigma
is used instead of velocity because the stars do not move in a set orbit;
stars as a population have too much random motion. The relation between
5r! 2
(Bender). Where M is the mass of
! and the mass is given as M =
G
the galaxy including the dark matter. The ratio of mass of the galaxy with
dark matter to without dark matter is given by
M wodm
. Giving us the
M wdm
percentage of dark matter in each galaxy. As stated before the average
spiral galaxy is made of 90% dark matter. This would mean that if elliptical
galaxies are made of 87-90% dark matter it supports the theory that they
are formed from spirals. If the percentage is much greater then 90% it
would not support the theory but rather disprove it, showing that there is
more dark matter in the elliptical then there are in spirals.
Figure 1:
NGC 3610 showing the ripples seen in elliptical galaxies involved in a merger or
collision
From Kennicutt
References:
Chaisson, Eric, Steve McMillan. Astronomy Today. United States of
America: Prentice Hall, 2002.
Bender, R. “New Clues to Structure and Formation of Elliptical Galaxies.”
from Proceedings of the International Conference, May 29- June 2
1989.
Efstathiou, G., Lake, G., & Negroponte, J. 1982.
Franx, M. Illingworth, G., & di Zeeuw, T. 1991
Kennicutt, R. C., F. Schweizer, J. E. Barnes. Galaxies: Interactions and
Induced Start Formation. Germany: Springer_Verlag Berlin
Heidelberg, 1998.
Kutner, Marc L. Astronomy A Physical Perspective. United Kingdom:
University Press, Cambridge, 2003.
Parker, Barry. Colliding Galaxies: The Universe in Turmoil. New York:
Plenum Publishing Corporation, 1990.
Schweizer, F., “Colliding and Merging Galaxies” Science(January 1986)
Description of Observations:
There are seven elliptical galaxies that were picked to observe.
Five of them are in the Virgo Cluster
Elliptical
Surface
Brightness
M89
21.44
M49
21.86
M59
21.60
M60
21.75
M87
21.45
The other two other elliptical galaxies
Leo 1 with a surface brightness of 21.19
M105 with a surface brightness of 22.97
Each of these elliptical galaxies will take one night of observing.
To take this data I need to use the 2.1 meter telescope with a long slit
spectrometer. This scope has a wide enough field of view to just fit the
entire major axis of each of these galaxies in its field of view. The elliptical
galaxies that I am measuring range from 300" to 540", this wide field of
view will allow me to take a spectrum of the entire galaxy at one time. This
is more time efficient, both to take the data, and analyze the data to take
one spectrum of each galaxy. The velocity dispersions of elliptical galaxies
range from 150-200 km/sec, allowing me to use the smaller telescope and
still obtain an S/N of about 15, not ideal but that is small enough for the
calculations needed for an accurate calculation of the dark matter present
in the galaxy.
NOAO Observing Proposal
Date: September 20, 2007
Standard Proposal
Category: High
Redshift Galaxies
Observing Spectra and Finding Chemical Abundances of Galaxies at
Redshift z~1.0
PI: Lauren Boucher Status: UG Affil: Case Western Reserve University
Department of Astronomy, 10900 Euclid Ave. Cleveland, OH 44106
Phone: 937.304.6022
Email: [email protected]
Abstract of Scientific Justification:
We propose to carry out single slit spectroscopy on multiple galaxies at redshifts of
approximately z=1. The spectroscopy data will allow us to detect the presence of metals
(Z>8) in distant spiral galaxies and to measure the relative chemical abundances and
relative metallicity of the galaxies. The early universe was dominated by light elements
and the evolution of heavier elements occurred during nucleosynthesis in the cores of the
first cycle of star formation. Stars from this first generation of stellar evolution, like the
older stars present in globular clusters and those found in the halo of galaxies, had lower
metallicities because they formed at a point in time when metal abundances were lower.
Galaxies that are at higher redshifts are being viewed at a younger age than those found
in the local universe. According to theories of stellar evolution and the evolution of
galaxies, galaxies at higher redshifts should have lower metallicities because the galaxies
are younger and have gone through less star formation and production of heavier
elements. The data from the spectra of several redshift z=1 galaxies will be compared to
spectra of local galaxies.
In order to obtain these spectra, we propose to target several spiral galaxies at redshift of
approximately z=1. This sampling of spiral galaxies will allow us to view galaxies that
contain active star formation. We will be observing spiral galaxies of type Sb and Sc,
which have higher star formation than other spirals and ellipticals. We are requesting 10
dark nights on the 4m telescope in order to reach objects of peak surface brightness 22.5
magnitudes.
Run
1
2
3
4
5
Summary of observing runs requested for this project
Telescope Instrument No. Nights Moon
Optimal Months Accept. Months
KPNO-4m
10
Darkest Mar-Jun
Mar-Jun
Scheduling constraints and non-usable dates:
NOAO Proposal
Page 2
Scientific Justification:
High redshift galaxies allow us to learn more about stellar and galaxy evolution because
we are looking into the past and viewing these objects at a younger age. Studying the
chemical evolution of galaxies brings together ideas about stellar evolution and theories
of star formation. In order to study these chemical abundances and metallicities, we need
to examine the spectra of spiral galaxies, home to active star formation, at varying
redshifts. We can observe the change in chemical abundances of galaxies as they evolve
(i.e. at different redshifts) and also learn about the types of stars that they may possess.
The study of element abundances at high redshifts aims to examine three main issues
including: primordial abundances of lighter elements (i.e. He, D, and Li), the increase of
metal content at different cosmic periods, and the element ratios as functions of
metallicity. These last two items are important in determining how the metal content of
the universe increases with time and to verify whether our current understanding of
galaxy and stellar evolution is correct. In order to learn about this relationship, we must
obtain measurements at high redshifts to be able to view the galaxies at different cosmic
epochs (Pettini 1998).
Through the process of nucleosynthesis in the cores of stars, heavier elements were
formed from hydrogen. This first generation of stars took hydrogen, converted it to
heavier elements, and then at the end of their lifecycles, the stars released these heavier
elements back into the interstellar medium with a supernova explosion. A new
generation of stars formed from this metal rich medium and incorporated these heavier
elements, remnants from the previous generation. By extension of this cycle of star
formation, we expect to see that galaxies at higher redshifts (i.e. younger) have either
stars that have not progressed as far through their life cycles and therefore have created
les heavy metals than those in nearby galaxies or there have been fewer successive
generations of star formation at that point in time. From this reasoning we would believe
that we should obtain spectra indicating lower chemical abundances present in high
redshift spirals and the opposite, higher chemical abundances, in closer galaxies, because
the stars are farther along in their life cycles and therefore have developed more layers of
heavy elements. Galaxies at lower redshift are likely to have had more generations of
stars that have expelled their metals formed in the core back into the surrounding gas
through supernova explosions.
In order to study the evolution of metal concentrations in galaxies and learn more about
galaxy evolution, we propose to study the chemical abundances of high redshift z=1
spiral galaxies in comparison with nearby galaxies (z~0). By knowing chemical contents,
abundances and metallicities at higher redshifts, we can determine the extent of chemical
evolution at different times in the universe and in extension the chemical evolution at
different times in galaxy evolution. The sample of galaxies that will be observed
includes spiral type galaxies with Hubble classifications of Sb and Sc. In the Sb and Sc,
there is a higher proportion of active star formation and young stars than that found in
NOAO Proposal
Page 3
ellipticals and S0 and Sa spirals. In order to study stellar evolution models, our work is
best conducted by observing these galaxies that still have active star formation and
comparing them to spirals nearby, which also have active star formation.
In order to determine the chemical abundances and relative metallicities of the galaxies,
we can obtain spectra from redshifted galaxies, which give us a look back into the past.
We measure the presence of metals by looking for known absorption lines for neutral and
ionized elements. Different metals have different absorption lines and when elements are
ionized, this changes their absorption wavelengths (see Table 1). The metals that we will
be focusing on are iron (FeI), sodium, magnesium, ionized calcium, and carbon (Jaschek
& Jaschek 1995). By using the absorption wavelength tables, we can measure the
presence of a given element by looking for the absorption line on our spectra. Also, the
spectra provide information on the intensity of the absorption lines. The redshifted
galaxy spectra will need to be adjusted to compensate for the shift in wavelength of these
known absorption wavelengths.
The chemical compositions of the galaxies may be best compared in the ratios of [Fe/H]
or even other metals in relation to Fe (i.e. [Na/Fe], [Ca/Fe], [Zn/Fe] (Jaschek & Jaschek
1995). Typically, the metallicity of stars is measured as the proportion of iron to
hydrogen [Fe/H]. This proportion is often expressed in terms of the log of the ratio of
iron to hydrogen in a star to the composition of the sun. We can apply this concept to the
galaxies we are studying. Typically, if this ratio is less than -1, we say that the object is
metal poor and if it is greater than -1, that it is metal rich. We can compare the strength of
the Fe lines (or other metal lines) to the strengths of the H lines to obtain metallicities. In
obtaining these ratios for the nearby and distant galaxies, we can compare the two in
order to look for an inverse relationship between metallicity and redshift.
The metallicities for nearby galaxies can be obtained from available data and spectra.
We can obtain the spectra for these galaxies from the Sloan Digital Sky Survey for
comparison. By determining the metallicity of galaxies at different redshifts, we may be
able to form a relationship between age and metallicity or confirm galaxy and stellar
evolution models.
NOAO Proposal
Page 4
Figure 1: This is a spectrum of galaxy at redshift z=1.0640 obtained from the Sloan
Digital Sky Survey. The galaxy is approximately 30” in diameter. The spectra is marked
with common absorption lines for metals and lighter elements Al, MgII, OII, OIII, HeI,
etc.
Wavelength (Angstroms)
Element
5896, 5890
Sodium
5270
FeI
5183-5168
Mg
3968, 3933
CaII
4267
CII
4674
CIII
Table 1: This table is a listing of the wavelengths to expect for the absorption lines of the
elements that we want to observe in our study of the spectra of the galaxies. (Birney,
D.S.1991) More complete tables of absorption wavelengths can be found in
Observational Astronomy by Birney.
NOAO Proposal
Page 5
Figure 2: This is a spectrum of a galaxy obtained from the Sloan Digital Sky Survey with
a redshift of z=0.078. This spectrum includes identifiers for Ca absorption lines and
compared with Figure 1, we can see the effect of redshifting on the expected absorption
lines of the elements. Using these spectra we can compare the presence of metals in the
two galaxies after making the appropriate redshift adjustments.
References:
Birney, D.S. Observational Astronomy. 1991, 198.
Jaschek, C. & Jaschek M. The behavior of chemical elements in stars. 1995.
Pettini, M. Element abundances at high redshifts. Chemical evolution from zero to high
redshift. 1998, 233.
NOAO Proposal
Page 6
Experimental Design:
The sample of spiral galaxies are at redshift z=1 and have a peak surface brightness of
approximately 22.5mag /sq arcsec. The galaxies have diameters ranging from 10-20
arcseconds. We propose to use the KPNO 4m telescope due to the low surface brightness
of the galaxies. The setup of the telescope will be with a slit width of 1” and slit length
of 300”. In order to reach surface brightness of 22.5mag/sq arcsec with a S/N ~ 10 for
the galaxies, we need total exposure time of one full night per galaxy; therefore, for the
ten spiral galaxies in our sample set, we will need a total of 10 full nights of observing
time.
While the 2.1m telescope can view these faint galaxies, we would be unable to obtain the
required S/N ratio to make any use of the spectra from the smaller telescope and therefore
the 4m was chosen. Setup 1 for the 4m telescope was chosen with a resolution of 5
Angstroms per pixel and an unbroadened spectral line equal to 3 pixels at FWHM in
order to obtain a S/N ~ 8 for the given surface brightness over a full night exposure.
Ideally, we would like to obtain a resolution of 1.5 Angstroms per pixel; however, the
S/N would be too low to obtain reliable spectra. The galaxies were chosen with a
maximum z~1 so that they were just within the lower surface brightness limit of the
telescope setup.
The slit will be lined up over the center of the galaxy in order to obtain the spectrum of
the entire galaxy. The galactic centers of the galaxies will be brighter than the outer
regions and provide a better signal even though the overall surface brightness is faint.
Multiple galaxies are chosen in order to produce a range of values for metallicity so that
an average can be found and to avoid anomalous spectra results. The spiral galaxies of
type Sb and Sc are chosen in particular to study galaxies with active star formation.
Spectra will need to be redshift corrected for the appropriate z value
1 + z = (observed / emitted ) in order to make correct determinations of absorption lines.
CWRU Observing Proposal Date: 26 Sept., 2007 Searching for bright Kuiper Belt Objects PI: Lucille H. Frey Department of Astronomy Case Western Reserve University 10900 Euclid Ave., Cleveland, OH, 44106, USA Email: [email protected] Abstract of Scientific Justification: We propose to conduct a survey of bright Kuiper Belt Objects (KBOs). Due to their distance and relatively small sizes, KBOs are difficult to study in detail. A large‐scale search for bright KBOs will allow us to obtain higher resolution spectra for our objects. Spectral data will be used to determine surface composition which, along with orbital information, can identify collisional families. In addition to studying the current composition and distribution of KBOs, these objects can also be used to constrain models of solar system formation. This observing run will be part of a larger survey covering 360 deg2 which will be followed by spectroscopic observations of new KBOs. This data will then be published and compared with previous observations to identify collisional families and study composition. Fields along the ecliptic will be chosen to supplement previous surveys. We are requesting 6 dark nights on the 0.9m to obtain observations of 60 deg2 to a limiting magnitude of 21.9. Summary of observing runs requested for this project Run Tel. Instrument No. Nights Moon Optimal months Accept. Months 1 0.9m Mosaic f/7.5 6 darkest any
any Scientific Justification The existence of a group of bodies beyond the orbit of Neptune was first theorized by Edgeworth and Kuiper nearly 60 years ago (Edgeworth, 1949 and Kuiper, 1951). Since 1992, nearly 1,100 of these objects, variously known as Trans‐Neptunian Objects (TNOs) or Kuiper Belt Objects (KBOs), have been discovered. 1 Despite this large number, our knowledge of KBOs is still very limited. The distance to these objects and their small size makes any study beyond orbital determinations difficult. Pluto, at 14th magnitude, is the brightest known KBO by several magnitudes. A more numerous sample of large (and therefore bright) KBOs would allow more complete spectroscopic studies. While numerous surveys of KBOs have been carried out, see table in Allen, 2002 for a partial list, our sample is far from complete. Spectra can be obtained for the KBOs brighter than around 21st magnitude in the R band (Barkume et al, 2006), allowing us to determine surface composition, albedo, and size. High quality spectra have been obtained for Pluto’s moon Charon using 8m Gemini telescope on Mauna Kea with adaptive optics, showing evidence of cryovolcanism (Cook et al, 2007). In addition to studying current properties and processes, KBOs can also be used to study the history of the solar system. Surface composition, derived from spectra, have been used with orbital parameters to identify a ‘family’ of KBOs which appear to have originated from a single parent body disrupted in a collision early in the solar systems’ history. (Brown et al, 2007) Studying the Kuiper belt also provides constraints on models of solar system formation, as well as information about the current dynamics of the outer solar system. Accurate models must account for the current KBO population in terms of size, composition and orbital resonances. Current estimates of the mass and orbital distribution of KBOs have been combined with dynamical models to suggest that Neptune went through a short phase with a high orbital eccentricity (Morbidelli et al, in press). These simulations assume that mass function of KBOs is not strongly dependent on orbital radii. Morbidelli notes that their conclusions will change significantly should this assumption be proved false by subsequent observations, such as this survey. Our current project seeks to discover new KBOs, determine their orbital and spectroscopic properties, and combine this data with previous observations to identify collisional families and study composition. Fields along the ecliptic will be chosen to supplement those previously observed to a comparable limiting magnitude (Jewitt et al, 1996, Trujillo et al, 2001, Larsen et al, 2001). We will then obtain spectra for new KBOs and conduct follow‐up observations after 1 and 2 years to accurately determine orbital parameters (Jones, 2006). 1
Current catalog of TNOs available at http://www.cfa.harvard.edu/iau/lists/TNOs.html Along with data from subsequent observing runs, this new data will then be published and combined with previous surveys by other groups to identify collisional families and study KBO compositions. Any survey seeking to observe faint objects must reach a compromise between limiting magnitude and areal coverage. We have decided to search for objects brighter than 22nd magnitude in the R band, allowing us to use a smaller telescope with a larger field of view. Previous surveys (see table in Allen, 2002) with comparable or brighter limits have successfully discovered dozens of KBOs. Our parameters were chosen to give an expected detection of 1 – 2 KBOs during our observing run, given previously observed densities (Larsen et al, 2001). Figure 1: Eccentricity and Inclination of possible collisional family members as a function of semi‐major axis. Black dots indicate water absorption line depths of less than 10%, grey dots indicate less than 30%, and white dots greater than 30%. Orbital parameters alone are not sufficient to distinguish related objects from other KBOs, but combined with spectral properties the KBOS with the strongest water‐ice absorption can been seen to be correlated. These objects are believed to have originated along with KBO 2003 EL61 in the collisional break‐up of a parent body. (Brown, 2007) References: Allen, R.L., 2002, Ph.D. diss., U. of Michigan Barkume, K.M, Brown, M.E., Schaller, E.L., 2006, ApJ, 640:L87‐L89 Brown et al, 2007, Nature, 446:7133, 294 Cook et al, 2007, Apj, 663, 1406 Edgeworth, K. 1949, MNRAS, 109, 600 Jewitt, Luu, and Chen, 1996, ApJ, 112, 1225 Jones et al., 2006, Icarus 185:2, 508‐522 Kuiper, G.P. 1951, Proc. of the Nat. Acad. of Sci., 37: 1‐14 Larsen et al, 2001, ApJ, 121, 562 Morbidelli et al. 2007 in press arXiv:astro‐ph/0703558v1 Trujillo, Jewitt and Luu, 2001, ApJ, 122, 457 Experimental Design Our project aims to find bright KBOs over a large area. Without access to a large telescope dedicated to such a task, the most efficient way to conduct a survey is to use the largest available telescope with an acceptably large field of view. Any chosen ‘ideal’ balance between the two is by definition arbitrary, but our decision to use the 0.9m telescope with a nearly 1 deg2 field of view will allow us to obtain a large quantity of data with a reasonable R band limiting magnitude. KBOs range in magnitude from Pluto at 14th magnitude down to as faint as 24th magnitude. Our cut‐off of 22nd magnitude in the R band will enable us to detect the bright end of this range. As noted above, this statement has been proven by previous surveys. With our sequence of observations, described below, we will observe a total of 60 deg2 during 6 nights of observations. Observations will be conducted in the R band, which is most efficient for the typically reddish KBOs. Using simple trigonometry, the apparent motion of KBOs can be determined to be between 1”‐3”/hr, near opposition. This motion is detectable between observations spaced several hours apart, allowing data to be taken during a single night. This eliminates possible errors due to differing weather conditions or instrumental differences between nights. For each field, three 600s exposures will be taken at one hour intervals. In the R band, this exposure time produces a limiting magnitude of 21.9 for a S/N of 20. These images will be processed during the observing run using the Moving Object detection Pipeline (Petit et al, 2004). This system has been used since 1999, with modifications, to identify candidate KBOs in a triplet of images while eliminating known stars. This program outputs a list of candidates to be manually checked. Its quick runtime of about 45 minutes for each triplet of images will allow us to analyze our data as it is obtained and conduct follow‐up observations of candidates during our observing run. Six nights will allow us to observe 60 deg2 along the ecliptic and conduct further observations to confirm or eliminate candidate KBOS. Reference: Petit et al, 2004, MNRAS 341, 471‐480 Observing Kuiper Belt Dwarf Planets’ Sizes in Search of Triton’s
Origin
PI: Sara Cummins
Status: Student
Affil.: Case Western Reserve University
Department of Astronomy, 10900 Euclid Ave, Cleveland, OH 44106 USA
Email: [email protected]
Phone: (513)850-4488
Abstract of Scientific Justification:
Pluto and Triton most likely originate from the same material as evidenced by their size,
surface temperature, and composition (Beatty et al.1999). I propose to test the
predictions of some of the theories regarding Pluto’s and Triton’s origin by studying
several other objects in the Kuiper Belt. Preliminary observations have shown that Eris,
Sedna, and Quaoar have relative sizes greater than that of most other Kuiper Belt objects
so they are classified as dwarf planets. I have chosen them for comparison with Pluto
because it too is a dwarf planet.
Two of the theories my study will address attempt to explain the origin of Triton and the
origin of Pluto, respectively. The first claims that Triton originated in the Kuiper Belt
and was captured by Neptune causing the retrograde orbit of Triton around Neptune
(Goldreich et al. 1989). The second theorizes that Pluto and the other objects in the
Kuiper Belt were formed by many collisions with each other caused by Neptune’s
gravitational pull (Stern 1998).
I propose to measure the apparent magnitude of Eris, Sedna, and Quaoar as part of a
larger study to determine their compositions. I have already obtained data on infrared
emissions from these three dwarf planets so I need a precise measurement of apparent
magnitude to calculate their albedo and size. Next, I will calculate the mass of each
dwarf planet. This will allow me to calculate the density of each and compare with
Pluto’s density which is known to be 2.03 g/cm³ (Windows 2000).
Summary of observing runs requested for this proposal
Run Telescope
1
KP – 4m
Instrument No. Nights Moon
Mosaic
1
darkest
Optimal Months Accept. Months
July-Oct
July-Oct
Scientific Justification
Triton and Pluto share many characteristics such as size, bulk density, composition, and
temperature as described by Beatty et al. (1999). This has resulted in theories claiming
that they interacted with each other when they were formed (Beatty et al.1999). I
propose to calculate the sizes of Eris, Sedna, and Quaoar as part of a larger study to
determine their similarities with Pluto and with each other. My study will focus on
comparing the densities of these dwarf planets to determine if they have similar
composition. I need to measure their visual brightness with great precision in order to
complete my calculations of their sizes.
Pluto’s composition is not completely understood at present. However, according to
NASA (2007), spectroscopic observations have shown that the surface is covered in
methane ice. In addition, its density is 2.03 g/cm³. This reveals that its interior is largely
made up of water ice (NASA 2007). Ice near the core is in phase II because of the
increased pressure while ice closer to the surface is in phase I (Windows 2000). The core
is made of an iron-nickel alloy and rocky elements.
Similar spectroscopic observations described by NASA (2007) of Triton show that its
surface is covered in nitrogen and methane ices with traces of carbon monoxide and
carbon dioxide ices. Triton has a density of 2.066 g/cm³ which is noticeably similar to
that of Pluto. Its density reveals that its interior is formed of water ice and rocky
elements such as silicate in roughly the same proportions as Pluto (NASA 2007).
Many theories have been proposed to explain how Pluto and Triton interacted in the past.
My study will test the theory behind these by determining the composition of Eris, Sedna,
and Quaoar. For example, calculations by Goldreich et al. (1989) have revealed that it
would be possible for Triton to have formed in the Kuiper Belt and then have been
captured by Neptune. It may have collided with a moon already orbiting Neptune and
destroyed it causing Triton’s retrograde orbit around Neptune (Goldreich et al. 1989).
If I discover that Eris, Sedna, and Quaoar have compositions similar to that of Triton, this
theory would be supported. These three dwarf planets are in a 3:2 stable orbital
resonance with Neptune (Bland et al. 2003). As a result, they were probably members of
the Kuiper Belt when Triton was formed. Thus, Triton would have been formed in the
Kuiper Belt.
Another theory that will be tested by my research tries to explain the creation of Pluto
and other dwarf planets. In this theory proposed by Stern (1998), Pluto originally was
very small and had a near-circular, low-inclination heliocentric orbit. However, as more
bodies accumulated in the Kuiper Belt, the gravity of Neptune caused them to collide
with each other. This led to the increase in Pluto’s size, the formation of Triton, and the
formation of the dwarf planets. Pluto itself may have collided with an object between one
hundred to one thousand kilometers in diameter which created Charon but destroyed the
object (Stern 1998).
Although it is clear that some collisions must have occurred in the Kuiper Belt to cause
the fragmentation that is present, the involvement of Neptune and origin of Pluto are not
understood completely (Bland et al. 2003). This relates to my study because if the dwarf
planets I am observing each have a similar composition to that of Pluto and Triton, they
must have been made from similar materials as a result of collisions. Thus, this theory
will be given further support.
My study began in September 2006 by recording data on the infrared emissions of Eris,
Sedna, and Quaoar. Infrared light at a wavelength of 1.2 millimeters was used for these
observations. Next, some preliminary calculations were conducted from this data for
Eris. I chose to calculate Eris’s surface temperature in order to compare it with Pluto’s. I
assumed that Eris has an albedo of .62 when estimating surface temperature because .62
is close to that of Pluto. Also, the distance from Eris to the Sun is known to be 97 AU
(Discovery 2007). These two pieces of information yield a surface temperature of about
25 K. This indicates that Eris is composed of something different from Pluto because
Pluto’s surface temperature is known to be 44 K. However, Eris may have an albedo that
greatly differs from that of Pluto in which case this estimate is incorrect.
The requested time on the 4m will determine the correctness of the above assumption. A
measurement of apparent magnitude is needed for each dwarf planet. This measurement
will reveal how much light Eris, Sedna, and Quaoar reflect while the infrared emissions
already collected reveal how much light is absorbed. By combining this information, the
albedo can be determined for each dwarf planet. Size can be calculated from this because
the amount of light from the Sun falling on any area in the Solar System is known and
simple to determine. Finding the amount of light reflected, apparent magnitude, and
knowing what percent this is, albedo, of the total light reaching the object allows for the
calculation of size.
Next, I will calculate the mass of Eris, Sedna, and Quaoar and use this and their sizes to
determine their average densities and compare to the density of Pluto. If they too have
average densities of 2.03 g/cm³, then they are probably composed of ice and rock in
proportions close to that of Pluto. A larger density will indicate a greater proportion of
rock while a smaller density will indicate the opposite.
Once I have completed the comparison of the Kuiper Belt objects’ densities, I will be
able to infer information about their origin. Since it is widely accepted that Pluto
interacted with Triton in the past, trans-Neptunian objects that have compositions similar
to that of Pluto probably also interacted with Triton (Beatty et al.1999). However, dwarf
planets with densities significantly different from that of Pluto may have been captured in
the Kuiper Belt.
Figures and References
Figure 1: Outcomes of simulated binary-planet encounters. This figure shows the
simulation results of a Triton binary encountering Neptune at various velocities. Two
possible masses are shown, Triton’s current mass and 1 its current mass. Larger
10
velocities are unlikely because Neptune’s orbital speed is 5.4 km . For velocities
s
≤ .35 km , an object of either mass is captured with about 50% probability. At higher
s
velocities, capture of 1.0 mT is rare while 0.1mT is still captured approximately 50% of
the time because of its greater orbital speed. (Agnor et. al 2006)
Figure 2: Orbits for objects in the solar system. As shown in this figure, the orbit of
Pluto passes very close to that of Neptune. This shows that Neptune could disrupt the
objects near Pluto and cause them to collide. These collisions probably increased the size
of Pluto and created Charon and Triton as in the theory proposed by Stern. (NASA 2007)
References
Agnor, C.B., and Hamilton, D.P.. 2006, Nature, 192
Beatty, J. Kelly, Petersen, Carolyn Collins, and Chaikin, Andrew. 1999, The New Solar
System, 294-6
Bland, Phillip A., McBride, Neil, Moore, Elaine A., Widdowson, Mike, and Wright, Ian.
2003, An Introduction to the Solar System, 262-7
The Discovery of Eris, the Largest known Dwarf Planet. 2007,
http://www.gps.caltech.edu/~mbrown/planetlila/
Goldreich, P., Murray, N., Longaretti, P.Y., and Banfield, D. 1989, Science, 500-4
Minor Planet & Comet Ephemeris Service. 2007, http://www.cfa.harvard.edu/iau/MPEph
/MPEph.html
NASA Solar System Exploration. 2007, http://solarsystem.jpl.nasa.gov/
Object Seasonal Observability. 1993, http://imagiware.com/astro/observability.cgi
Stern, S. A. 1998, Solar System Ices, 1-6
Windows to the Universe. 2000, http://www.windows.ucar.edu/tour/link=/pluto/
pluto_composition.html
Technical Description of Observations
Preliminary observations have revealed that Eris has an apparent magnitude of about 19,
Sedna has an apparent magnitude of about 20, and Quaoar has an apparent magnitude of
about 18. These estimates show that I will be able to see these objects with the 4m
telescope at Kitt Peak. However, I need more precise measurements for my calculation of
each dwarf planet’s density.
Very little is currently known about Eris, Sedna, and Quaoar since their sizes and masses
has not yet been calculated. Although all of these dwarf planets reside in the Kuiper Belt,
they are currently at different distances from the sun. Eris is presently at 97 AU while
Sedna is at 90 AU and Quaoar is at 43 AU (Discovery 2007). I chose these dwarf planets
to compare with Pluto because their relative sizes show that they are larger than most
other objects found in the Kuiper Belt.
The equation to be used for calculating size is the Stefan-Boltzman equation. This states
that L = 4π r 2σ T 4 . This equation will allow for the determination of size because r is
the radius of the object. Luminosity, L, will be obtained from the measured apparent
magnitudes and known distances to each object. T is surface temperature which will be
calculated using the equation for the equilibrium temperature of an object,
Rsun
1/ 4
Tp = Tsun
(1 − a ) . Here, Tp is the temperature of the dwarf planet.
2D
This telescope is ideal for my observations for a variety of reasons. Since I am observing
each of Eris, Sedna, and Quaoar separately, I do not need a very large field of view. For
example, Eris’s relative size is only .04 arcseconds. The 36 x 36 arcminute field on this
telescope will definitely be large enough for my observations. Also, seeing of 1.1 is good
enough for my study. I do not need a more powerful telescope since I am only taking
simple observations of apparent magnitude.
I only need one night for this part of my study because each of my three observations will
take ten minutes. I plan to observe each dwarf planet for this length of time with a signal
to noise ratio of 100.0 to obtain minimal error. It is important to be precise because the
apparent magnitudes I record will be used in many further calculations. With these
parameters, the 4m is able to observe objects that have an apparent magnitude of up to
22.0. This is greater than the rough estimates for all three dwarf planets’ apparent
magnitudes so I should easily be able to study them.
Despite the fact that my observations will take a total of thirty minutes, I need an entire
night because each object is above the horizon at a different time of night in the months
of July through October. This is because the objects have Equatorial Coordinates quite
different from each other (Minor 2007). The differences made finding a time of year
when they will all be visible somewhat difficult which is why only four months out of the
year are acceptable for my observations.
Observing Proposal
Date: September 27, 2007
Distance to M31 through the Cepheid PLC Relationship
Stephen Riley
Email: [email protected]
Affil: Case Western Reserve University
Abstract:
I propose to search the galaxy M31 for type I Cepheid variable stars in order to study their
period-luminosity-color (PLC) relationship and to determine the distance to M31. Although the
PLC has been studied in the past, those studies have used small sample sizes, so this proposal
seeks to study M31 using the Mosaic CCD on the 0.9m. With its 59’ by 59’ field of view, a
much larger sample size than has been used in the past will be obtained. Accurately describing
Cepheid’s properties allows them to be used as standard candles to measure the distance to
nearby extragalactic objects. This is important for two reasons. First, it is important to establish
intermediate distance scales to nearby galaxies (farther than parallax can measure) to establish a
framework from which other distance measuring objects can be calibrated. Ultimately, being
able to measure the distances to many galaxies will allow for a more refined value for Hubble’s
constant, H0. Secondly, the study will lead to a better value for the distance to M31, which is still
not accurately known. This is important because it is the closest spiral galaxy to the Milky Way
and to know the distance to it will allow the absolute magnitudes of many of the interesting
objects in it to be determined.
Run
Telescope
1
2
3
KP-0.9m
KP-0.9m
KP-0.9m
Summary of Observing Time Requested
Instrument
No.
Moon
Optimal
Nights
Months
Mosaic
7
Less 50%
Sept/Oct
Mosaic
7
Less 50%
Sept/Oct
Mosaic
7
Less 50%
Sept/Oct
Acceptable
Months
Aug-Nov
Aug-Nov
Aug-Nov
2
Scientific Justification:
In 1912, Henrietta Leavitt discovered that Cepheid variable stars exhibit a linear relationship
between their magnitudes and the logarithm of their period (“PL” relationship) (Sterken and
Jaschek 1996). This relationship for the V band is expressed in the general form
M 〈V 〉 = δ log10 Pd + ρ
(1.1)
Where M 〈V 〉 is the average absolute V band magnitude, Pd is the pulsation period in days, δ is
the correlation (slope) of the PL relationship and ρ is a zero-point correction term (Feast and
Catchpole 1997). Subsequently, it has been found that the scatter on the PL relationship is
intrinsically due to differences in the surface temperatures of the Cepheids. This scattering about
the mean can be reduced by adding a color correction term such as (〈 B〉 − 〈V 〉 ) (Sandage and
Tammann 1968). The absolute magnitude can be expressed in terms of the PLC relationship as
M 〈V 〉 = α log Pd + β (〈 B〉 − 〈V 〉 ) + γ
(1.2)
(Feast and Catchpole 1997).
Obtaining the absolute magnitude of Cepheids is valuable because it means that the distance can
be found if the apparent magnitude m, of the Cepheid is known using,
d = 10( m − M + 5) / 5 pc.
(1.3)
This is useful because aside from parallax, no way is better understood for finding distances.
Because Cepheids are supergaints they are very bright (M ~ -2 to -6). This makes Cepheids
visible over extragalactic distance scales where they can serve as the basis for measuring the
distances to all nearby galaxies (Sterken and Jaschek 1996). Just as parallax provides the
calibration for Cepheids by independently giving the distance to nearby galactic Cepheids, so can
Cepheids provide a basis for calibrating other standard candles such as type Ia supernovae.
Of great importance to cosmologists and many scientists alike is finding an accurate value for the
Hubble constant H0. Hubble’s constant can be found simply by H 0 = v / d where v is the
recessional velocity of the galaxy. The recessional velocities are rather easily found from
spectroscopic data; it is the distance factor that is often uncertain. Cepheids can accurately
provide the distance to nearby galaxies, and calibrate type Ia supernova to find the distance to
even farther galaxies, leading to a better value for H0. The need to understand the relationship of
Cepheids’ properties is even more important since the start of the HST Key Project, which aims
to measure the Hubble constant to an accuracy of 10%, by observing Cepheids in nearby galaxies
(Freedman et al. 1994).
This proposal seeks to establish a distance to the nearby Andromeda Galaxy (M31), and to
determine the constants for the PLC relationship as given in (1.2). Previous studies conducted by
Feast and Catchpole (1997) attempted to answer this question for the LMC and M31 but only
used a small sample of stars (~29) that had been observed by Hipparcos.
3
Because of M31’s proximity to the Milky Way, many Cepheids can be observed in it. Stanek et
al. (1999) found 35 Cepheids in an 11’ x 11’ field in M31 in 1999 using the McGraw-Hill 1.3m
at MDM Observatory. Considering that the Mosaic camera on KPNO 0.9m has a 59’x59’ field
and M31 is 190’x60’ (NED) than it is reasonable to assume that approximately one third of M31
can be imaged and this could easily provide a sample of over 200 short period Cepheids. This
project will consist of imaging in the V, B and I bands; V and I because that is primarily what
HST images in (Tanvir 1999) and the B so that the reddening extinction can be measured. While
the amplitude of the magnitude change over a period is less in the infrared, the measurements in
infrared are affected considerably less by dust than optical bands (Figure 1) (Tanvir 1999).
It is important to note the distinction between classical Cepheids of the form δ Cephei, which
have the well known relationships described previously, and type II Cepheids of the form W
Virginis for which the relationship is not precisely known. Classical Cepheids are population I
stars while type II Cepheids are population II stars (Petit 1982). Therefore, after a star is
identified as a Cepheid, it will be important to consider other data that is readily available for
many of the variable stars in M31 such as their position, radial velocity, luminosity, and
chemical composition. This will help identify which stars are population II stars that are more
likely the W Virginis type of Cepheids so that they can be excluded from this study. It is often
hard to tell which class a Cepheid is just from its light curve alone (Sterken and Jaschek 1996).
To study the PLC relationship differential photometry will be used to get the magnitudes as a
function of time, the light curves. Standard star fields will be needed however, because absolute
photometry must also be performed to find the true values of the mean and maximum light. In
addition, since the observations will be spread over the course of possibly a few weeks, standards
will allow the light curves from different nights to be seamlessly matched up. The Cepheids
period will be determined from the light curves. This is an important measurement and great care
will have to be taken to assure that the best fits possible are obtained so that the periodluminosity relationship can be accurately described. Finding the mean and maximum magnitudes
in all of the bands will allow for determinations of the color of the stars which can then be used
to find the important β term which will reduce the intrinsic scattering of the PL relationship.
Corrections due to galactic reddening can also be made with the β term as laid out in Sandage
and Tammann (1968). The rest is a matter of plotting data in the absolute magnitude versus log
of the period relationship and finding the parameters of that fit, the slope, δ and α, and the zeropoints ρ and γ to be determined. The final step is to take what is found to be the absolute
magnitude, M<V>, and to match it to the standardized observed apparent magnitude and using
(1.3) this will yield a distance to that Cepheid in M31.
4
Figure 1: Light Curves for a 6.1 day Cepheid, in (top to bottom) K, J, I, R, V and B bands. From
Tanvir (1999)
Figure 2: PL plot of Galactic Cepheids with data from Gieren et al. 1997. The dashed line is the
fit derived from the LMC Cepheids. From Tanvir (1999).
5
References:
Feast, M. W. & Catchpole, R. M. 1997, Royal Astronomical Society Monthly Notices, 286, L1
Freedman, W.L. et al., 1994, Ap. J. 427, 628
Petit, M., 1982, “Variable Stars”, John Wiley & Sons Ltd., 1987.
Sandage, Allan, & Tammann, G. A., 1968, Ap. J., 151, 531
Stanek, et al., 1999, Ap. J, 117, 2810
Sterken, C. & Jaschek, C. 1996 “Light Curves of Variable Stars”, Cambridge University Press,
1996
Tanvir, Nial R. 1999, “Distance Determination with Cepheid Variables” in Post-Hipparcos
Cosmic Candles, Kulwer Academic Publishers, 1999.
6
Technical Description of Observations:
The galaxy to be imaged will be M31, the Andromeda Galaxy. M31 is located at 00 42 44.31
+41 16 09.4 J2000 (SIMBAD). This makes it an object for observing in September or October.
As stated previously, Cepheids have absolute magnitudes of approximately -2 to -6 and if it is
assumed that the distance modulus to M31 is 24.44 (Ribas, Ignasi et al. 2005) than the expected
magnitudes of the Cepheids in M31 (with shorter periods) would be from magnitude 20.44 to
22.44 (V band). With any variation due to amplitude, typically less than half a magnitude, the
faintest magnitude of the search will be about magnitude 23.
The objective will be to obtain five to eight images a night in V band of 900 seconds, giving a
S/N of about 10 for any magnitude 23 objects. In addition, five 1200 second images will be taken
in B and five in IC, which for a 23rd magnitude object will yield a S/N of about 10 and
approximately 5 for B and IC respectfully (calculated with KPNO’s CCDTIME). This totals to 5
hours of light frames a night, and then calibration frames would have to be taken, as well as a
few standard stars.
The 0.9m telescope was selected for this project because it has a wide field which will allow the
imaging of many Cepheids simultaneously. In the case of the Cepheid light curves, it would be
better to get more periods worth of measurements using longer exposures for less frequent data
points, than to get many measurements in a night with a larger telescope but only be able to get
one period worth of data due to less time. It is important to measure over more than one period
for variable stars to get a higher accuracy on the maximums magnitudes. Therefore, 21 days was
chosen as it will get at least two periods on any Cepheid with a period less than 10 days, which
should be about half of the sample found. Since light curves are constantly repeating, it is not
necessary to image for 21 straight days. In reality then it will be most useful to plot the light
curves as a function of phase instead of time. Because many of these stars come close to the limit
of detection and this project involves precise photometry the less moon light the better.
NOAO Observing Proposal
Panel:
For office use.
Category: Galactic - Other
Date: September 26, 2007
Observing high redshift galaxies for Tully Fisher relationship
PI: Steven Janowiecki
Status: U Affil.: CWRU
11311 Euclid Ave., Cleveland, OH 44106 USA
Email: [email protected]
Phone:
FAX:
Abstract of Scientific Justification (will be made publicly available for accepted proposals):
We propose to study the Tully-Fisher (TF) relationship over a range of redshifts to better calibrate
this correlation for future use and to study the evolution of galaxies as a function of lookback
time (as determined by redshift). This relation in spiral galaxies between rotational velocity and
absolute luminosity has been well studied and is typically employed to find distances to galaxies
with measured spectra. In studying the TF relation up to a redshift of 1 we can better calibrate it
and make it useful for a wider range of observations. Higher redshifts not only correspond to greater
distances, but also earlier times. Thus we will also be examining the effects of evolution on spiral
galaxies. Through the TF relationship we get an independent distance estimate and a look into
kinematic evolution of these galaxies. Our sample of 10 galaxies has already been photometrically
observed and fully reduced. However, since the TF relationship requires kinematic information, we
are requesting 4 nights on the 4m to do spectroscopy on each galaxy. With this spectroscopic data,
we will complete our study and determine the effects of redshift on the TF relationship.
Summary of observing runs requested for this project
Run
1
2
3
4
5
6
Telescope
KP-4m
Instrument
No. Nights
Spectrograph 1
4
Moon
Optimal months
Accept. months
Sep - Nov
Sep - Nov
Scheduling constraints and non-usable dates (up to four lines).
NOAO Proposal
Page 2
This box blank.
Scientific Justification Be sure to include overall significance to astronomy. For standard proposals
limit text to one page with figures, captions and references on no more than two additional pages.
The Tully-Fisher (TF) relationship (Tully & Fisher 1977) is a widely used and powerful tool that
relates the intrinsic luminosities and rotational velocities of spiral galaxies. These parameters
correlate tightly enough that, with a zero point and slope for the relationship, measurements of
rotational velocities can yield estimates for intrinsic luminosities. A distance to the spiral galaxy
can be obtained by comparing this intrinsic luminosity with an observed apparent magnitude.
Since independent estimates of distance are few, this relationship provides an important means for
studying galaxies. It is not simply a rule of thumb, though, as it has a clear physical basis. In spiral
galaxies the stars are orbiting the center and maintain these orbits in patterns such as spiral arms.
This clear and organized rotation dynamically supports the galaxy and allows the assumption that
rotational velocity is coupled with dynamical mass (the mass of the stars that are orbiting in the
galaxy). These stars also have certain mass to light ratios, which correlate the estimate of mass
with an estimate of light emitted from that mass. The TF relationship is the combination of these
connections: a measure of rotational velocity yields an estimate of luminosity.
As mentioned, this relationship has a zero point and slope which need to be calibrated. These
calibrations are most useful when they are performed on a particular sub-sample of spiral galaxies,
such as distant galaxies in clusters (Bamford et al 2005), distant galaxies in the field (Bamford
et al 2006), as well as nearby galaxies. (In a large sample, there are so many sources of intrinsic
scatter that the relationship is very hard to convincingly fit.) For the local galaxy sample, their
distances are known independently, i.e. from Cepheids or Supernovae, etc. It is from these local
well-measured galaxies that the zero point and slope of the TF relationship are established, and
then are applied to more distant galaxies to derive their distances (Pierce & Tully 1992). For the
galaxies in clusters, the dynamic history of each individual galaxy plays a significant role in its
current observed TF relationship. If a galaxy is perturbed or has recently interacted, it will not
be entirely supported by rotation, and the connection between rotational velocity and mass will be
weakened. The scatter in the correlation will be much greater for cluster galaxies.
Examining galaxies at high redshifts provides not only a very distant sample but also a less-evolved
sample, since objects at higher redshift are generally younger than objects at a redshift z = 0. Thus
in our study of high redshift galaxies we will be separating effects caused by the distance and effects
resulting from the different states of evolution. Both of these phenomena will change the slope and
zero point, but in predictable ways. Bohm et al. (2004) observes that the TF slope gets shallower
at z ∼ 0.5, and accounts for this with a different mass to light ratio, which results from the galaxy
being less evolved. As the spiral galaxy evolves, it increases its metallicity, and its mass to light
ratio increases. This effect is particularly strong for high-mass spirals, which are the type we are
more likely to observe at high redshift. Similarly, galaxies observed at great distances will begin
to be noticeably altered by cosmological effects. These effects may necessitate a K correction to
make sure the measured flux is correct, as well as the evolution correction mentioned above, since
galaxies will also get redder with age. While these overall observational effects will be dominated
by the evolutionary changes, there may still be some extra reddening or other distortions in the
observations that are redshift dependent. This will become obvious in our reduction and analysis.
In order to better study the effects of redshift on the TF relation, we have selected 10 galaxies to
observe, with redshifts between 0.77 and 1.1. There is a chance that the galaxies selected will be
biased at higher redshifts, due to the fact that galaxies appear dimmer at greater distances, and
so it is easier to see the brighter galaxies. Thus, if we observe a trend as a function of redshift,
it may only be a trend as a function of luminosity, and not actually redshift. Our chosen sample
of galaxies was selected to avoid this bias by picking a sample with much greater variation in
NOAO Proposal
Page 3
This box blank.
brightnesses than in redshifts. Similarly, we may encounter a bias toward star-forming galaxies, as
those are also brighter and more likely to be observed. Still, we will be able place an upper limit
on the TF relation for this sample of galaxies, even if they are only representative of the bright end
and not a uniform sample.
For our selection of galaxies, we have already obtained high quality optical imaging, and have
derived the necessary inclination angles (to correct for projected rotation velocity, discussed below)
and apparent magnitudes (Janowiecki 2007). All that remains is to obtain a spectrum for each
galaxy in order to compute its rotation velocity. While this computation is a well-established
procedure for nearby galaxies, it becomes a more complicated problem at high redshifts where the
galaxies are small on the sky. Chiu et al (2007) observed similar high redshift galaxies and Figure
1 shows some of their imaging and spectral data for reference. Their spectra are for a particular
emission line along a slit through the galaxy. It is clear from the spectra that one side of the galaxy
is rotating toward us (blue-shifted, relative to the galaxy) and the other side is rotating away from
us (red-shifted, galaxy relative). However, there is not enough resolution to simply fit the rotation
velocity as a function of radius and obtain a rotation curve, so we instead rely on an alternative
method of analysis. The spectra are processed with a routine developed by Simard & Pritchet
(1998) called ELFIT2D. It creates synthetic emission line models and tries rotation curves until the
observed spectra are reproduced. This technique has been successfully applied in similar data sets
(Chiu et al 2007). These generated rotation curves will asymptote (or reach their maximum at)
the value we use for rotation velocity. Once this rotation velocity has been computed, we still have
to correct it to account for the inclination of the galaxy relative to us. The spectra only measure
the component of velocity that is radial to us, and not the true circular velocity. As a galaxy is
tilted further away from us, the projected velocity scales as the sine of the inclination angle until
it is face on, and has no projected rotation velocity. In some of the galaxies that are nearly edge
on, estimates will also be made for the absorption from dust within the galaxy, which would dim
the appearance of the galaxy and skew the TF relationship.
The spectra derived from ELFIT2d will also permit us to re-compute redshifts for the sample,
providing another distance measure and allowing us to look for redshift trends in the TF relation.
Bamford et al. (2006) has undertaken a similar study of TF slope as a function of redshift, and their
results are unable to make a claim about any redshift dependence (Figure 2). Other groups have
claimed to find significant effects that depend on the evolution (as determined through redshift),
but there are many publications on both sides of the issue (Vogt et al 1997 or Simard & Pritchet
1998). These evolution effects will be especially pronounced in our high redshift galaxies, since
they will be substantially less evolved (z ∼ 0.5 corresponds to a lookback time of almost 8 Gyrs,
assuming standard CDM cosmology). With this very early look into these galaxies, the high redshift
TF relation should reflect the changing properties of the galaxies, especially the different mass to
light ratio at younger ages. The bright young stars in the earlier galaxies will emit more light per
mass than the more evolved galaxies will. Our TF fit will show how these galaxies are evolving, in
terms of their mass to light ratio and their kinematic structure - the key parts of the TF relation
We expect our results to have smaller uncertainties than previous work and to more tightly constrain
the TF relationship at high redshifts, in order to make a stronger statement about the TF slope’s
dependence of redshift. Still, much of the scatter in this relation is not a result of instrumental
or measurement uncertainties, but rather is from physical processes in the galaxies, such as the
evolution, perturbation, lack of total rotation support, or geometric alignment factors. Even so,
our calibration of the TF slope and zero point will be useful in future high redshift observations as
a distance estimate, and will also help reveal more of the kinematics of spiral galaxies at various
states of evolution in a variety of environments.
NOAO Proposal
Page 4
This box blank.
References
Bamford, S. P., Milvang-Jensen, B., Aragon-Salamanca, A., Simard, L., 2005, MNRAS, 361, 109
Bamford, S. P., Aragon-Salamanca, A.,Milvang-Jensen, B., 2006, MNRAS, 366, 308
Bohm, A., et al. 2004, A&A, 420, 97
Chiu, K., Bamford, S. P., Bunker, A., 2007, MNRAS, 377, 806
Janowiecki, S. P., 2007, private communication
Pierce, M. J., Tully, R. B., 1992, ApJ, 387, 47
Simard L., Pritchet C.J., 1998, ApJ, 505, 96
Tully R. B., Fisher J.R., 1977, AAP, 54, 661
Vogt, N. P. et al, 1997, ApJ, 479, L121
NOAO Proposal
Page 5
This box blank.
Observing Run Details for Run 1:
Technical Description Describe the observations to be made during this observing run. Justify the
specific telescope, the number of nights, the instrument, and the lunar phase. List objects, coordinates, and
magnitudes (or surface brightness, if appropriate) in the Target Tables section below (required for queue
and Gemini runs).
The targets for this observation are 10 galaxies from the DEEP2 spectroscopic survey (Cuillandre
et al 2001), following the selection criteria of Chiu et al (2007). They are inclined enough to prevent
most of the interference from dust, but not inclined so much as to alter the inclination correction
(from projection) to the rotation velocity measurement. These galaxies are at redshifts between
0.77 and 1.1, with apparent I-band magnitudes from 21 to 21.6, and angular sizes between 2 and
4 arcseconds. In order to obtain a S/N of ∼ 10, we will expose for 3 hours on each galaxy, using
Setup 2 on the spectrograph. The galaxies range in rotational velocities up to about 200 km/s. In
order to obtain velocity uncertainties around 20 km/s (comparable with Chiu et al (2007)), this
S/N will be acceptable, as it yields a velocity accuracy of about 20 or 25 km/s.
Cullandre, J.-C. et al., 2001, in Clowes R., Adamson A., Bromage G., eds, ASP Conf. Ser. Vol.
232, The New Era of Wide Field Astronomy. Astron. Soc. Poc. San Francisco, p. 398
Instrument Configuration
Filters:
Grating/grism:
Order:
Cross disperser:
Slit: 1”
Multislit:
λstart :
λend :
NOAO observing proposal LATEX macros v2.14.
Fiber cable:
Corrector:
Collimator:
Atmos. disp. corr.: