Transcript Introduction: exoplanets, atmospheres, project goals

PTYS 195A
Observing Exoplanets:
Recording, reducing, and analyzing ground-based data
Rob Zellem
PhD Candidate
Lunar & Planetary
Laboratory
University of Arizona
About me!
• I am a giant nerd
– I love Star Wars!
• I am a Planetary Astronomer
– I am trying to find alien life
• Grew up in Nashville, TN
– Went to grade school and high school there
• Went to college at Villanova University
– Degree in Astronomy and Astrophysics
• MSc in Space Science at University College London
• PhD candidate at LPL
– Structure and composition of exoplanets
– Graduating at the latest this May
Course Website
• www.lpl.arizona.edu/~rzellem/PTYS_195A.ht
ml
• Will post lectures and assigned reading
• Grades will be on D2L
– Assuming I can figure it out….
Office Hours
• Space Sciences 247
• By appointment (drop-ins okay if not busy)
• [email protected]
Course Objectives
• Exoplanetary research indicates the existence of planets, a majority of
which are completely unlike those in our Solar System. Of the more than
1055 transiting exoplanets discovered to date, over 105 are giant planets,
known as “hot Jupiters”, that have near-Jupiter masses and orbit close to
their host stars. Due to their large planet-to-star contrast and being hosted
by bright stars, these exoplanets are accessible with the University of
Arizona’s 61” Kuiper telescope.
• Here students will record their own data on the 61”. We will learn how to
reduce and analyze data from this platform.
• This class will culminate in a small group presentation on a specific
exoplanet, potentially for inclusion in a published paper and Fall 2016 AAS
poster presentation.
• Students will also develop presentation, critical thinking, and critiquing
skills.
Class Format
• The course will be conducted as a seminar, with a particular focus at each
meeting. Students are expected to have read the assigned material in
advance and be ready to participate in the discussion.
• Two students will be assigned to lead each discussion with a short
presentation on the required reading at the beginning of each class.
Weekly reading will typically consist of about one published, peerreviewed paper.
• Throughout the semester we will reduce a common dataset.
• In the middle of the semester, we break down into smaller groups, which
each group assigned a specific target. This group will reduce this target
and present their results and previously-published data in an 8 minute
presentation at the end of the semester.
Textbook
• NONE! YAY!
• Published papers available when you are on
UofA’s network
Grades
• 10% for being at the telescope for one
observing run
• 10% weekly presentation
• 10% weekly quizzes
• 30% class-wide data reduction project
• 40% final project (data reduction results 50%)
and PowerPoint presentation (50%)
Telescope Signup Sheet
Weekly Presentations
• 2 students will give a short (8 minute) presentation on the weekly
assigned reading and help facilitate a discussion about the methods and
results. Students will be graded with the following rubric with a 0 for not
meeting the requirement and 2 for meeting the requirement:
• Read the paper
• Addressed major paper concepts
• Explained major concepts clearly and concisely
• Facilitated discussion with peers and/or answered questions adequately
• Ask questions in other students’ presentations (when not presenting)
• BONUS (+1): Found a critique about the paper, only available to presenting
students
Weekly Quizzes
• At the beginning of each class, there will be a
short quiz with questions on the previous
class or the reading due for the present class.
The aim for the daily quiz is to reinforce major
class concepts and to provide the instructor
with a proxy attendance grade. The lowest 2
quiz grades will be treated as extra credit.
POP QUIZ TIME!!!
Extra Credit
• The lowest 2 quiz grades will be treated as extra
credit, to be added to the “Quiz” portion of the final
grade calculation. In addition, there will be
opportunities to go to public talks throughout the
semester. Each student will be expected to take
notes on the topic and get the signature of the
instructor or the speaker for credit. Each opportunity
will be treated as 1 additional point applied to their
final numerical grade.
Academic Integrity
• It is strongly recommended that all students read the
University of Arizona’s Code of Academic Integrity. All
students in this course are expected to abide by this code.
• We will operate with the “3 strike method” for academic
integrity issues: 1st offense will result in a 0 on the assignment
in question, 2nd offense will result in a loss in a letter grade for
the class (e.g., a student’s grade will be reduced from an “A”
to a “B”), and 3rd offense will result in a course failure. ALL
offenses will be reported to the University according to their
Academic Integrity policy.
Planetary Transit Technique
Measures dimming of star
light as planet passes in front
of (or behind) the star
Star-light dims less than 1%
Like looking for a firefly next
to a lighthouse
Gives us the size
(radius)
Planetary Transit Technique
Advantages:
a) Relatively cheap
b) Can determine the size
of the planet
Disadvantages:
a) Bias towards large
planets and in short period orbits
b) False detections due to stellar variability
c) Planet’s orbit must be seen edge-on from the observer
point of view (so the planet passes in front of the star)
Kepler Mission
• Launched in 2009
• Mission objective:
to discover Earthlike planets
orbiting other stars
• As of February
2014, 961
confirmed planets
– 2903 unconfirmed
planet candidates
NASA
ight curves were ext ract ed from t he dat a by using
lyt ic equat ions of Mandel & Agol (2002) t o genermodel t ransit (Pearson et al. 2014). A Levenbergardt (LM) non-linear least squares minimizat ion
et al. 1992) provided an init ial local fit of t he& Agol (2002) model light curves t o our dat a.
t he analysis, t he t ime of mid-t ransit (Tc ) and
t o-st ar radius (R p (λ)/ R S ) were left as t he only
ramet ers. T he eccent ricity (e), argument of pen (ω), scaled semi-major axis (a/ R s ), quadrat ic
arkening coefficient s (µ1 and µ2 ), and t he orbit al
(Pb) of t he planet were fixed t o t he values list ed in
2. T he linear (µ1 ) and quadrat ic ( µ2 ) limb darkoefficient s in each filt er were t aken from Claret
men (2011) using t he st ellar paramet ers (Tef f =
, log g= 4.452 (cgs), [F e/ H ]= 0.450 ) from Toral. (2008), which agree wit h t he more recent valTeske et al. (2013a). T hese solut ions were t hen
s seed values for t he EXOFAST (East man et al.
Different ial Evolut ion Markov Chain Mont e Carlo
CMC) (Braak 2006) t o find a global best -fit . We
d t he t ransit wit h t he DE-MCMC using 20 chains
6
links. T he Gelman-Rubin st at ist ic (Gelman &
1992) found t hat all chains converged t o ≤ 1%, as
d in Ford (2006).
uncert aint ies indicat ed by t he DE-MCMC were
d wit h a residual permut at ion, or “ prayer bead” ,
d (Sout hwort h 2008). T he residual permut at ion
s a series of repeat ed st eps: (1) t he best -fit model
act ed from t he dat a; (2) t he residuals are circuhift ed and added t o t he dat a point s; (3) a new
und; and (4) t he residuals are shift ed again, wit h
t t he end wrapped around t o t he st art of t he dat a.
Exoplanet Atmospheres
• Transits allow the study of exoplanet
atmospheres
– Can study how light varies at different
wavelengths – tells us about atmospheric
structure and composition
F i gu r e 1. U band light curves showing t he t ransit of X
Values are normalized t o one, for each dat e of observed t
Rob’s Thesis
• Observe transits of HD 209458b
– One of the two brightest exoplanets
– Hot Jupiter
• M = 0.714 Mjupiter
• R = 1.38 RJupiter
• 0.04747 AU away from its host star
– 25 times closer to its star than the Earth is to the Sun
– 9.5 times closer to its star than Mercury is to the Sun
• 3.52474859 day orbital period
– ~150 light-years away
UofA’s 61” Kuiper Telescope
of mid-t ransit (Tc ) and
S ) were left as t he only
ty (e), argument of peaxis (a/ R s ), quadrat ic
and µ2 ), and t he orbit al
xed t o t he values list ed in
adrat ic ( µ2 ) limb dark• Transits
allow the study
were t aken
from Claret
ellar paramet ers (Tef f =
gas giant atmospheres
e/ H ]= 0.450 ) from Torit h t he more recent valhese solut ions were t hen
OFAST (East man et al.
rkov Chain Mont e Carlo
nd a global best -fit . We
E-MCMC using 20 chains
bin st at ist ic (Gelman &
ns converged t o ≤ 1%, as
Exoplanet
Atmospheres
by t he DE-MCMC were
at ion, or “ prayer bead” ,
he residual permut at ion
ps: (1) t he best -fit model
t he residuals are circu-
of
Griffith et al. 2014
Transit
• What is it measuring?
Transit
• The atmosphere + the planet’s disk
Transit
• The atmosphere + the planet’s optically-thick
disk
Transit
• The atmosphere + the planet’s optically-thick
disk
Transit
• Amount of atmospheric absorption will
change with wavelength
Transit
• Amount of atmospheric absorption will
change with wavelength
Beer’s Law
I = I0e
-t
Transit
• So a planet’s radius will change with
wavelength due to absorption by different
molecules in its atmosphere
So….
• If we measure the transit of an exoplanet at
different wavelengths…
– We can measure how its radius varies with
wavelength
– Indicates its atmospheric structure and content
• Atmospheric structure = how temperature varies with
altitude
• Atmospheric content = what molecules are present
Example!
• Detection of H2 scattering
Zellem et al. (in prep.)
Example!
• Detection of H2 scattering
Another Example!
• Detection of water, methane, and carbon
dioxide in a hot Jupiter’s atmosphere
Swain et al. (2009)
Measuring radii at the 61”
Benneke & Seager (201
• Planet has same signature in the infrared (IR)
despite differing atmospheric contents
• Signal very different in the optical
Why are the IR signatures the same?
• In the IR, a small planet with a thick
atmosphere can block as much light as a large
planet with a small atmosphere
– Hot Jupiter atmospheres are opaque in the IR
Why are the IR signatures the same?
• In the IR, a small planet with a thick
atmosphere can block as much light as a large
planet with a small atmosphere
– Hot Jupiter atmospheres are opaque in the IR
=
However, not the same in the visible
• In the visible, the planet’s atmosphere is now
transparent, so a small planet will look
different than a large one
However, not the same in the visible
• In the visible, the planet’s atmosphere is now
transparent, so a small planet will look
different than a large one
≠
Rob does a spectroscopy trick
• IT’S AN ILLUSION, MICHAEL.
Measuring radii at the 61”
Benneke & Seager (201
• Planet has same signature in the infrared (IR)
despite differing atmospheric contents
• Signal very different in the optical
Looking for New Planets
Looking for New Planets
Looking for New Planets
Looking for New Planets
Looking for New Planets
Looking for New Planets
Looking for New Planets
Looking for New Planets
Looking for New Planets
Looking for New Planets
Transit Timing Variation
(TTVs)
Looking for Exomoons
Looking for Exomoons
Measuring Exoplanetary Magnetic
Fields
Measuring Exoplanetary Magnetic
Fields
In the UV
In the B
Hubble Magnetic Field Detection
Fossati et al. (2010)