Transcript Exoplanet Working Group - IAG-Usp

Exoplanet Working Groups
CoRoT Brazil Workshop
Natal 2004 : oct 29th – nov 2th
CoRoT-Brazil
Pierre BARGE
• Summary on exoplanet discoveries
• Challenge on the terrestrial planets
and the place of CoRoT
• Present organizing of the Exo WG
CoRoT-Brazil
Exoplanet search main results
(RV from 1995 to 2004)
• 7% of dwarf stars around Sun host giant planets (EGP)
• Stars with planets have spectral types from F to M
• Statistical studies become possible ( 130 EGPs)
• EGP are preferentially found around metal-rich stars
• Orbital periods range from ~1.5 days to some years
• Large eccentricities are common ( 0 < e < 0.927)
• Planet masses range from 14M to 10MJup
• Mass distribution peaks toward small planets
• Density of the planets is determined in some cases
• 10 multi-planets discovered (some commensurabilities)
• Some planets are found in binary star systems
CoRoT-Brazil
Mass/period segregations
Min-mass/period
 R: single dwarf stars
(binaries and evolved * removed)
● heavy planets ( >2 )
○ intermediate (0.75 > ; <2 )
∆ light planets (< 0.75 )
Orbital period distribution
-red: heavy
-grey: light (<0.75)
CoRoT-Brazil
(After Udry et al. 2003)
Parent star metallicity
(After Santos et al. 2003)
 Signature of the core instability scenario ?
 Result of engulfed migrating planets ?
CoRoT-Brazil
Summary on Giant Planets
– Commonly form around stars (single or binaries)
– Have masses in a wide range (0.6 < m < 10 mJup )
– Can be found in the inner part of the system
– Can have orbits with very large eccentricities
 Strong differences with our Solar System
We still do not know how do they form !
• Core instability in a layered nebula
• Gravitational instability in a gas nebula
CoRoT-Brazil
About terrestrial planets
– Well defined problem adressed in terms of
kinetic equations and numerical simulation, both.
– The standard formation scenario is accretion
by planetesimal accumulation (Safronov 1969)
Planetesimals  moon sized bodies (105 yrs)
or bigger planetary embryos
Good agreement and common consensus
Final stage: Embryos  T. Planets
Depends on the presence and location of G. planets !
CoRoT-Brazil
Close-in terrestrials:
a very “hot” question
• Planets with mass similar to that of Uranus were
recently discovered by RV method (14 – 20 m)
• Are they Uranus like (migrated/evaporated) or big
terrestrial ?
 Their density (radius) is required
CoRoT will permit:
•
•
•
•
To answer the above question
To discover other such planets
To test a number of emerging models
To start statistics of terrestrial planets …
CoRoT-Brazil
Exoplanets:
Two Working Groups
E.W.G.
E.C.O.W.G.
(Exoplanet Working Group)
(Exo. Complementary Obs. WG)
Coordination of sci. activities
Coordination of obs. Effort
• Transit detection
• Stellar “noise”
• Works on specific topics
(planetary formation, physics of
Gas Giants, atmospheres and
wind, magnetosphere, tidal
effect, dynamical stability,
planets in binaries, …)
CoRoT-Brazil
•
•
•
•
•
Preparatory observations
Follow-up
Complementary observations
……
Scientific data base (Exodat)
Cf. Magali’s talk
E.W.G.
Objectives and Strategy
• Objectives
– To optimize the impact of CoRoT data on exoplanetary science
– To organize the scientific activity in various working teams
– To make people work together
– To stimulate exchanges between seismo and exo communities
(stellar activity as a noise, metallicities and spectral types, ….)
The difficulty lies in beginning to work with no data …!
• Strategy
Brain storming during Planet workshops
 Specific works decided during Exo sessions at CWs
CoRoT-Brazil
The Planet Workshops
• PW1: “Planetary formation: toward a new scenario”
(june 2-3 2003):
• PW2: “Planetary transit detection: stellar noise and false alarms”
(dec 8-9 2003):
• PW3: “Close-in exoplanets: the star-planet connection”
(may 13-14 2003):
• PW4: “Automatic Spectral Classification for large data sets”
(reported)
• PW5: … to be defined at the next CoRoT Week in Granada
CoRoT-Brazil
Some specific works
• Simulation of the stellar activity
Two different approaches:
– Rotational modulation by dark spots and active
regions calibrated on Virgo-Soho data
– Microvariability deduced from a spectral
analysis of Sun variations
• Simulation of light-curves and transits
• Blind test of the detection algorithms
using simulated light-curves
CoRoT-Brazil
Points raised at PW2
on transit detection
• How to compare and merge the capabilities of
the various methods ?
• How to build up again a detected transit ?
(least square fitting, Bayesian,….)
• Estimate others false alarms possibilities
• How to face stellar noise ?
 Appropriate filtering …..
 Use of colors (CoRoT, Eddington? )
CoRoT-Brazil
Main conclusions of PW2
• Eclipsing binaries
– probably one of the main sources of confusion
– also good targets for planet search !
• Radial velocity follow up
– Not a method to remove false alarms
– Can remove confusing situations
– Adds important information (mass)
• Testing detection algos. would require working
on the same light-curves and blindly
 Proposal: Free exchanges of light-curves
between the various teams
CoRoT-Brazil
Detecting transit blindly
(1st test – CW5)
This test involved various teams in our groups (initiated
during CW5):
• To produce simulated light-curves which account for:
–
–
–
–
Instrumental noises
Noises from the stellar variability
Planetary and stellar signals (possible ambiguities)
A sample of 1000 LCs were produced (secret: 1 person)
• To look for possible transits using different
detection algorithms
– Five different teams were involved (open to all CoIs)
– To work on a common set of LCs for relevant comparisons
CoRoT-Brazil
Conclusions of 1st blind test
• Very different detection methods tested
• False detections seem specific to the algo. used
• Stellar micro-variability is not the main limitations
• The method used to detrend the signal is almost as
important as the detection algorithm itself
• In some cases detrending can produce artefacts
• Background eclipsing binaries are source of confusion
• Characterization of the transits requires other analysis of
the signal
• CoRoT detectivity limitation:
(1.1 R 3days) on M0 dwarf stars
Results are to be published and LCs will be available on request
CoRoT-Brazil
Conclusions of PW3
PW3 was devoted to the Close-in Exosolar planets and the
relations they have with the host star.
• A lot of interesting points were addressed:
–
–
–
–
Existence of extremely hot giant planets (3 confirmed)
Evaporation rate of hot jupiter planets is strong
Origin of the overmetallicity of stars (primordial or not?)
How such planet form ? Migration ?
– Relations with the host star
(tidal effect, radiative and magnetic interactions)
– Possible existence Hot Uranus, big rocky planets (primordial or
evaporated remnants), big liquid planets, ….
 Many questions CoRoT will help to solve soon !
CoRoT-Brazil
Next work within EWG
To detect transit blindly using 3-color lcs
• This test will involve the detection teams of EWG
• LCs will be simulated using the instrument-model
(M.Auvergne) to account for realistic noises
• Transits and ambiguities will be included as in the first
blind test
• This will be a good opportunity to test how color
information can improve CoRoT detection capabilities
CoRoT-Brazil