HOW STELLAR IRRADIATION AND EROSION CAN SHAPE THE EARLY EVOLUTION OF PLANETARY ATMOSPHERES - @ALINEVIDOTTO 2021 SAGAN SUMMER WORKSHOP - NEXSCI
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
How stellar irradiation and erosion can shape the
early evolution of planetary atmospheres
@AlineVidotto
2021 Sagan Summer Workshop
What is the relevance of a planetary atmosphere?
Do habitable worlds exist? How do exoplanets evolve?
Presence of atmosphere: fundamental for Atmospheric loss sculpts planet and changes
planet to develop life mass-radius distribution
Observations with the Kepler satellite
depletion
Beaugé & Nesvorny (2013)
of small
log(occurrence)
planets
(Porb
Outline
1
How do we know exoplanetas
are evaporating? And why do
they evaporate?
2
The evolution of stellar radiation
& winds (mostly focused on
early ages)
3 Stellar wind interaction with
atmospheres of exoplanets
K010
K025
Indirectly: evaporation through population studies K025
K024
K009
credit: Leo dos Santos
Note.
For small exoplanets with Porb1.7 R and Big
quickly: Rp 
Directly: through transmission spectroscopy
Transmission spectroscopy
During transit, stellar radiation
broadband optical:
is transmitted through the
shallow transit exoplanet’s atmosphere
Lyman-α:
normalised ux
deep transit
1. Take a spectrum of the star at out-of-transit time
2. Take a spectrum of the star during transit
3. Divide the two to nd % of absorption by the
planetary atmosphere time
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 5
fl
fi
ff
Escaping atmosphere of GJ436b (Ly-α)
Transit depth ~ 60% (!)
Lavie et al 2017 Bourrier et al 2016
also: Kulow et al 2014; Ehrenreich et al 2015
Escape rates ≈ 2 x 108 — 109 g/s
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 6
ff
detailed study of the line triplet, including its
ter wavelength; the air wavelength of this line is 10,830 Å). The widths
el3, F. Javier Alonso-Floriano5, Stefan Czesla3, Fei Yan6, Guo Chen1,2,7,
ellen5, Mathias Zechmeister8, Jürgen H. M. M. Schmitt3,
shape, depth, and temporal behavior.
The Saturn-mass exoplanet WASP-69b orbits Infrared: Carmenes, HST
ets Atmosphericof the broadbandevaporation and narrowband in channels
Helium were lines294 Å (12-pixel
Puertas9, Núria Casasayas-Barris1,2, Florian F. Bauer8,9,
an active star with a period of 3.868 days (14).
o9, José A. Caballero4, Stefan Dreizler8, Thomas Henning6, It is a suitable target for atmospheric studies,
to columns) and 98 Å (4-pixel columns), respectively. We fitted both sets
ón9, David Montes10, Karan Molaverdikhani6, Andreas Quirrenbach11, due to its large atmospheric scale height and
s8, Ignasi Ribas12,13, Alejandro Sánchez-López9, high planet-to-star radius ratio, facilitating the
hneider3, María R. Zapatero Osorio14
res of spectroscopic WASP-69b
light curves using the approach described
detection of 5.8 ± 0.3% excess absorption in
the Na D line (15). We used the CARMENES spec- above forWASP-107b
xoplanets can lose part of their atmosphere due to strong stellar irradiation, a
es 1
s .
trograph to observe two transits of WASP-69b
the white-light curve. However, for the planetary transit signals, we only
Downloaded from http://science.sciencemag.org/ on February 11, 2019
can affect their physical and chemical evolution. Studies of atmospheric He 10,833 Å
xoplanets have mostly relied on space-based observations of the hydrogen
on 22 August 2017 and 22 September 2017
(night 1 and night 2, respectively) (see
Spake+2018
2.11
table Non-overlapping bins
I
2 far ultraviolet region, which is strongly affected by interstellar
uling ground-based
n the
. allowed R
high-resolution /R to
p spectroscopy,
s vary as a
we detected excess free parameter, while holding t 0, a/Rs, i, e and
S1 for the observing log). The observationsAllart+2019
spanned approximately 4 hours for each epoch,
All spectral bins
2.10
tP-69b,
a at a Psignal-to-noise
fixed to ratiothe
of 18. Wevalues
measured line reported
blueshifts of in Extended Data Table 1. We fixed the
he helium triplet at 1083 nanometers during the transit of the Saturn-mass which covered the full transit and provided
a before- and after-transit baseline. In total, 66
ar-
atmosphere limb-darkening coefficients in a similar way to the white-light curve
ers per second and posttransit absorption, which we interpret as the escape spectra were recorded, 31 of them out-of-
trailing behind the planet in comet-like form. 2.09
Transit depth (%)
transit spectra.
7b fit. Additional to 1.71 mm. details of the fitting procedure are given in Methods.
The wavelength region surrounding the He I
rs, high-resolution spectroscopy The near-infrared coverage providesfeature is affected by emission and water vapor
2.08
a frequently used tool for inves- access to exoplanet atmospheric features that
2
absorption lines originating from within Earth’s
ble
lanet The
atmospheres inferred
(1–4). Numer-
gh-resolution spectrographs have
cannot values
be observed infor
the the
visible transit
range,
ing the triplet of metastable He lines around
includ- depth, (Rp/R s) , in each
atmosphere (fig. S1). Although these lines are
spectrally separated from the He I triplet, we
wavelength
I
2.07
sit channel
ed on telescopes specifically for
are shown in Fig. 1 and Extended Data Table 2. These results
1083 nm. This feature has been proposed as a corrected for the effect of water absorption using
Downloaded from http://science.sciencemag.org/
ence (5–8). One of these spectro- tracer for atmospheric evaporation (9), a pro- the European Southern Observatory (ESO) tool
ms, constitute
ENES (Calar Alto high-Resolution
arfs with Exoearths with Near- the
and atmospheric
extreme ultraviolet (EUV) (10.0 totransmission
cess whereby intense x-ray (~0.5 to 10.0 nm)
92.0 nm) spectrum.
Molecfit (16) and for the sky emission lines using
2.06
an empirical model derived from the data (17).
ed of the Calar
lescope
Nortmann+2018
The
tical Échelle Spectrographs)
broadband
(8)
Alto Ob- transmission
to expand, resulting in aspectrum is consistent with a previous
irradiation from a host star causes atmospheres
of hot gas exoplanets
After this correction, we performed continuum
normalization and brought the spectra2.05 to the
eonwavelength transmission spectrum for WASP-107b, obtained using the WFC3 G141
spectrograph simultaneously bulk mass flow away from the planet. The con- stellar velocity rest frame. We then computed
range from 0.52 to tinuous mass loss most strongly affects smalla master out-of-transit spectrum (Fout), which
Downloaded from http://science.sciencem
e near-infrared range from 0.96 sub-Neptune–sized planets and may be capable was used to normalize all spectra, following
2.049
t is grism, which covers the 11,000–16,000 Å wavelength range . The latter
10,600 10,700 10,800 10,900 11,000 11,100
per exhibits a muted water at 1083absorption bandof centred at 14,000 Å, with an Wavelength (Å)
Fig. 2. Transmission spectrum between the second and third contacts of WASP-69b,
showing planetary absorption in the He triplet
I nm. (A) The excess absorption
b
ike
helium in the weighted-mean averaged transmission spectrum (black points) from two transit
observations otherwise
of WASP-69b (22 flat August spectrum
2017 and 22 September implying
2017) (see Fig.an opaque cloudFig.
1). The deck. After applying
2 | Narrowband transmission spectrum of WASP-107b, centred
Common feature: relatively active stars:
−1
best-fitting model (red line) shows a net blueshift of −3.58 ± 0.23 km s . The predicted positions
of the heliuma correction
triplet lines (1082.909 for stellar
nm, 1083.025 activity
nm, and 1083.034 nm) variations
are indicated as betweenon the G102
10,833 Å. Eachand G141
spectroscopic channel has been shifted one pixel away
from the next one. Non-overlapping bins are shown by blue symbols. All
needs EUV of to lift thecoincides
atmospheres
vertical dashed blue lines. (B) The residuals of the data after subtraction of the model are
hshowing
a in planetary
shown observation
black, epochs
and the red line indicates (see
the zero level.
absorption in the He triplet at 1083 nm.Methods),
Fig. 2. Transmission spectrum between the second and third contacts of WASP-69b,
the G102
(A) The excess absorption of spectrum
error barsaligns
on of the exoplanet WASP-69b (black) and itsI extended helium atmosphere (gray-blue) at the different contact points. Shown
with
are 1σ. The peakthe the spectrum with the 23S helium
on Februa
werobservations cloud deck
Summer
of WASP-69b
elium, 22 ± 3 min after T4.
level
Workshop
(22 August 2017inferred
- E ects
terial away from the planet could be the cause
of stellar from
irradiation
and 22 September
and the
helium in the weighted-mean averaged transmission spectrum (black points) from two transit
, second (T2),2021
thirdSagan
(T3), and fourth (T4) contacts of the broadband planet transit
2017) (see
and
erosion G141
1). The spectrum (Fig. 1).
onalso
Fig.
the moment
planetary
absorption line
when the tail has passed the
atmospheres
at the level of 1.04 ± 0.09% (24). A companion
at 10,833 Å. Aline Vidotto 7
best-fitting model (red line) shows a net blueshift of −3.58 ± 0.23 km s−1. The predicted positions
ff
What are the physical causes of evaporation in these close-in planets?
Atmospheric Loss processes
Gronoff et al 2020
• Several loss processes, divided
into “kinetic” and “ uid” types
‣ kinetic: individual particles are
lost
‣ uid (or hydrodynamic): bulk
out ow
• Simple “assessment”
‣ bloated planet (ie, low gravity)
and highly irradiated (ie, close-
in or active star): bulk out ow
‣ bulk out ow is very important
for youngish / close-in
exoplanets!
‣ 2 suggested “avenues” of
evaporation for these planets:
*Plus: stellar wind stripping (due to ram pressure), impacts that can blast
➡ photo-evaporation (external)
atmosphere into space, condensation of atmosphere, stellar tidal forces (Roche
lobe over ow), etc ➡ core heating (internal)
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 9
fl
fl
fl
fl
ff
fl
fl
Hydrodynamic escape via photo-evaporation: how does it happen?
o n
ot
ph eV)
U V .6
X >13
(E
Ionises Hydrogen uses 13.6eV…
o n
o t
ph eV)
U V .6
X >13 … and excess energy
(E
converted into thermal
motions
ic h
- r
e n e
og er Outcome: increase
d r ph
Hy mos atmospheric temperature
at
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 10
planet
ffHydrodynamic escape via core luminosity: how does it happen?
after formation: planet
is contracting… … gravitational energy of
the contracting core is
being released
re
h e
s p
o Outcome: increase
t m
a atmospheric temperature
planet
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 11
ffWhy is the high temperature so important?
High T generates a gradient
of pressure → force that
drives the planetary out ow
ρa = planet gravity + tidal + thermal forces
al
m
e ρGMp 3ρGM⋆r dP
er
or
c du
th
a l f ρu =− + −
tid dr r2 a3 dr
re
he
sp i ty
o av
t m r
a planet g Momentum equation does not care
what is causing the gradient of
*in uid dynamics, mass is replaced by density and forces are given per unit volume.. pressure (core heating, stellar
photoionisation, or something else!)
** eqs assume 1D geometry, spherical symmetry and steady state
This distinction enters in the energy
Interested in the derivation? https://rdcu.be/cjs9q equation.
Check Section 5.2.1 of Vidotto 2021, Living Reviews in Solar Physics
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 12
fl
ff
flEnergy-limit escape: a particular limit of hydrodynamic escape
Eirradiation,input = Ekinetic,output
1
assumption
cross-section:
a fraction ε of the … and is used to πReff2
energy ux is accelerate the
intercepted by the out ow to its
planet cross-section… terminal velocity Reff
· 2
2
Muterm
ϵFEUV(πRe ) = FEUV at
2 distance a
evaporation rate (g/s)
2
FEUV(πRe )
2
·
1/2 ME = ϵ
assumption
2GMp
( Rp )
The terminal velocity is
GMp /Rp
on the order of the vesc =
surface escape velocity
ff
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Allan & Vidotto 2019 Aline Vidotto 13
ff
fl
fl
ffEnergy-limit escape: a particular limit of hydrodynamic escape
2
· FEUV(πRe )
ME = ϵ cross-section:
GMp /Rp πReff2
LEUV
FEUV =
4πa 2 Reff
2
· ϵ LEUV(Re /a)
ME =
4 GMp /Rp
FEUV at
distance a
Mp
ρ̄ = 4
• Evaporation rate is larger for planets:
2
· 3ϵ LEUV /a πRp
3
‣ orbiting at close distances (small a)
ME ≃ 3
16 Gπρ̄ ‣ smaller average densities (e.g., gas giants)
‣ orbiting stars that are active (large high-energy
luminosities): usually young(er) stars
• Important form of escape in hot Jupiters & planets
orbiting young stars
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 14
ff
ff
ffMass loss estimates for exoplanets
© Maeve Upton
433 exoplanets orbiting low-mass stars - Energy limit
15 16
log(potential energy of plane
per unit mass [erg/g])
log(escape rate [g/s])
14
12
13
8
12
4
30 25 20 15
log(stellar high-energy rate
deposited @ planet [erg/s]) after Ehrenreich & Desert 2011
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 15
ff
tEUV Stellar irradiation and stellar wind erosion
+X- shape atmospheric evaporation
irra rays
diat
ion
photo-evaporation
stel
lar w
inds
2
The evolution of stellar radiation
& winds (mostly focused on
early ages)
©Garlick/U.of WarwickThe BIG picture: evolution of winds/activity of cool dwarf stars
activity &
planet detection
1. Wind removes
angular momentum Effects on planets
4. Evolution of 2. Rotational
magnetism/activity evolution of stars
Dynamo 3. Interior properties
Age of stars
Seismology of stars (& planets)
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 17
ffHow do winds, activity & planets evolve?
old stars age young stars
Open questions
•How has the solar wind evolved in
the past 4 billion years?
•What is the implication for young
exoplanetary systems?
•How do stellar winds a
ect
exoplanets (magnetosphere and
atmosphere)?
for solar-like stars only solar-like stars (MS)
M dwarfs/evolved stars
Sun
upper limits
Vidotto 2021, Living Reviews in Solar Physics
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 18
ff
ffPlanetary evaporation in evolutionary timescales: stellar X/EUV ux
Tu et al 2015
Observations of stellar
Stellar rotation ⟷ age ⟷ Stellar activity
rotation at different ages
(a)
(b) (b)
Gallet et al 2015 fast
EUV luminosity (erg/s)
Xray luminosity (erg/s)
fast
rotation rate (x solar)
slow
OD OD
slow
age (Myr)
age (Myr)
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 19
ff
flThe extreme ultraviolet and X-ray Sun in Time: evaporation of
planetary atmosphere Tu et al 2015
Atmospheric evaporation of a 0.5M⊕ planet @ 1au
Matm,initial=0.5% M⊕
slow
H content of the planetary
atmosphere is very di erent if
orbiting:
• slowly rotating star: 45%
retention of initial atmosphere
fast
• rapidly rotating star: entire
atmosphere is lost < 100 Myr
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 20
ff
ffEscape affects the internal structure of the planet
Kubyshkina et al 2020, Kubyshkina & Vidotto 2021
Free tools!
If you don’t have access to a
hydrodynamical model, python interpolator
no e tools from Daria Kubyshkina available at
scap
e
doi.org/10.5281/zenodo.4643823
ene
rgy
limit
Planetary evolution with MESA &
hyd atmospheric escape
rod
yna * Run your own model! inlists publicly
m ics
available
https://doi.org/10.5281/zenodo.4022393
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 21
ffPredicted observational signatures of atmospheres of close-in planets
Evaporation rates higher Ly-α H-α
at young ages
~20% mass
lost through
evolution
planet @
0.045au Symmetric pro les: At younger ages:
lack of stellar wind • broader (& saturated) • Hα transits with depths
interactions ~3 - 4% in excess of
Allan & Vidotto 2019 mid-transit Lyα line at
geometric transit
line-centre
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 22
ff
fiEUV
+X-
irra rays
diat
ion
photo-evaporation
stel
lar w
inds
3 Stellar wind interaction with
atmospheres of exoplanets
©Garlick/U.of WarwickStellar wind
wi n d stripping
la r
e l Carolan, Vidotto et al
s t
in prep.
orbital motion planet’s
shadow
star (nightside)
bow shock
tail shaped by the stellar
wind interaction + orbital
motion
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 24
ffCreate atmospheres
(HD219134, Vidotto+2018)
Erode
atmospheres How can stellar
(young Mars,
Kulikov+2007) winds affect
atmospheres of close-in
exoplanets?
Prevent escape
(Vidotto & Cleary 2020,
next slides)
Do nothing?How stellar out ows in uence planetary mass loss Vidotto & Cleary 2020
Stellar wind shapes planetary out ow, a ecting Stellar wind squashes planetary out ow,
observational signatures, but not escape rates reducing/preventing atmospheric escape
stellar
stellar wind “would-be”
wind
sonic radius
sonic radius
interface interface
between out ows between out ows
Stellar winds can reduce or even suppress mass loss from their exoplanets
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 26
fl
fl
ff
fl
fl
fl
fl
ffStellar winds con ne atmospheric escape of close-in planets
Vidotto & Cleary 2020 Con ned
Models that do not
consider the e ects of
stellar winds can over-
predict atmospheric
escape rates.
π Men c: expected strong
Not con ned atmospheric escape, but
none was detected!
π Men c (Garcia-Munoz et al 19)
Con ned by stellar wind?
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 27
fi
fi
fi
ff
ff
fiStudies of stellar wind con nement in 3D Carolan, Vidotto et al 2020b
• Vidotto & Cleary 2020: 1D radiative
hydrodynamics simulations →
cannot include stellar wind e ects
• 3D hydrodynamic simulations of
typical hot Jupiter & warm Neptune
density
(amu/cm3)
HD209458b
sonic
surface
GJ3470b
stellar wind
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres
being injected in Aline Vidotto 28
the grid
ff
ff
fiLower escape rates after disruption of sonic surface Carolan, Vidotto et al 2020b
sonic
surface Increasing stellar wind mass-loss rate sonic
surface
Closed (undisturbed sonic surface) Open (disrupted sonic surface)
Stronger stellar winds:
• lower volume occupied
by planetary
atmosphere
planet • for open sonic surfaces:
lower planetary escape
rates
2x solar wind 10x solar wind 20x solar wind
How does this a ect
3 4 5 6 7 8 9 observational signatures?
log(density[amu/cm3])
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 29
ff
ffThe effects of stellar winds on Lyα synthetic observations
Increasing stellar wind mass-loss rate
Carolan, Vidotto et al 2020b
Atmospheric escape rate (g/s)
d
lar
in
w
so
no
2x
fac c Hot
Jupiter
sur soni
e
en
op
• Observational signatures strongly
a ected by the presence of
stellar winds even when
lar
planetary escape rates are not!
so
0x
10
Lyα absorption (blue + red wings) [%]
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 30
ff
ffThe dichotomy of AU Mic b Carolan, Vidotto et al 2020a
• AU Mic b: Neptune-size planet orbiting a 22 Myr-old, pre-main sequence M dwarf (Plavchan et al 2020)
Strong wind of AU Mic (10 to 1000x
High EUV ux from the star causes
the solar wind mass-loss rate)
strong evaporation in AU Mic b
prevents/reduces evaporation
Lyα absorption in blue and red wings (%)
stellar wind: 10x solar stellar wind: 1000x solar
Transit signatures
“erased” by
stellar winds →
transits to probe
stellar wind
evaporation rate: reduced by 50% Stellar wind mass-loss rate (x solar)
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 31
fl
ffEffects of stellar activity on planetary escape
Temporal variations in the exosphere of HD189733b
X-ray count rate (1/s)
Flare
Transit
time (hours)
Lecavelier +12,
Bourrier +13
Credit: NASA, ESA,
Possible scenarios:
L. Calçada, SDO
1.Change of stellar wind properties (passing of a
Coronal Mass Ejection)
2.Increase of stellar energy input into the planet’s
upper atmosphere
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 32
ffHazra, Vidotto et al, submitted
Case 1: Quiescent Case 2: Flare
(“normal” Stellar wind (“normal” SW + high Fxuv)
+ “normal” Fxuv)
Hazra, Vidotto et al,
submitted
Case 3: CME Case 4: CME and Flare
(“strong” SW + Fxuv) (“strong” SW + high Fxuv)Hazra, Vidotto et al,
Simulated Ly-alpha transits of HD189733b submitted
data from Bourrier et al 2013
signature within
the full line
signature only
considering the
blue wing
nt
nd
are
are
E
CM
ce
wi
ies
E+
no
ly
qu
on
CM
• Flare alone does not change evaporation signi cantly
• CMEs are more e ective at removing planetary material
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 35
fl
fl
ff
ff
fiWhat’s next? Effects of magnetic elds on evaporation
“Local” escape simulations of
magnetised hot Jupiters (© Carolan)
2021 Sagan Summer Workshop - E ects of stellar irradiation and erosion on planetary atmospheres Aline Vidotto 36
ff
fiConclusions
Atmospheric escape and the evolution
of planets depends on the XUV history
of the host star.
Stellar wind can “erase” Ly-α transit
signatures in young systems
Stellar winds play important role in
atmospheric evaporation: from
retention to stripping of atmospheres
Variation in stellar out ows (quiescent
wind vs CMEs) can affect planetary
evaporation momentarily
@AlineVidotto
flYou can also read
