Thermosphere - Ionosphere Magnetosphere Coupling
Canada
M. Galand (1), I.C.F. Müller-Wodarg (1), L. Moore (2),
M. Mendillo (2), S. Miller (3) , L.C. Ray (1)
(1) Department of Physics, Imperial College London, London, U.K.
(2) Center for Space Physics, Boston University, Boston, MA, USA
(3) Department of Physics and Astronomy, University College London, U.K.
1.
2.
3.
4.
5.
Energy crisis at giant planets
TIM coupling
Modeling of IT system
Comparison with observations
Outstanding questions
JUPITER
(Gladstone et al.,
2007)
Credit: NASA/JPL/Space
Science Institute
Cassini/ISS (false color)
Cassini/UVIS
(Pryor et al., 2011)
SATURN
Cassini/UVIS
[UVIS team]
Credit: J. Clarke (BU), NASA
Cassini/VIMS (IR)
[VIMS team/JPL, NASA, ESA]
1. SETTING THE SCENE:
THE ENERGY CRISIS AT THE GIANT PLANETS
THERMAL PROFILE
(EARTH)
Exosphere
Key transition region
between the space
environment and the
lower atmosphere
Thermosphere
Texo
500 km
Ionosphere
85 km
Mesosphere
50 km
Stratosphere
~ 15 km
Troposphere
SOLAR ENERGY DEPOSITION IN THE UPPER ATMOSPHERE
Solar photons
ion, e-
Neutral
Thermosphere
Suprathermal
electrons
B
ion, e-
Thermal e-
Ne, Nion
SP, SH
*
+
Exothermic reactions
Ionospheric
e- heating
Te
Airglow
Neutral atmospheric
heating
IS THE SUN THE MAIN ENERGY SOURCE OF PLANERATY THERMOSPHERES?
Exospheric temperature (K)
W
Main energy source:
UV solar radiation
Earth
Main energy source?
Outer planets
CO2 atmospheres
[after Mendillo et al., 2002]
ENERGY CRISIS AT THE GIANT PLANETS
Observed values at low to mid-latitudes
solstice
equinox
Modeled values (Sun only)
[After Yelle and Miller, 2004; Melin et al., 2011 (+poster)]
ENERGY BUDGET OF THE THERMOSPHERE
 HEATING SOURCES
 Solar heating through excitation/dissociation/ionization +
exothermic chemical reactions
 Auroral particle heating via collisions + chemistry
[Grodent et al., 2001]
 “Ionospheric Joule heating” via auroral electrical currents
and ion-drag heating [Vasyliũnas and Song, 2005]
 Dissipation of upward, propagating waves (such as gravity
waves, …)
- Solar EUV/FUV heating*:
0.5 TW (Earth), 0.8 TW (Jupiter), 0.2 TW (Saturn)
- Auroral part./Joule heating*: 0.08 TW (Earth), 100 TW (Jupiter), 5-10 TW (Saturn)
[*: Strobel, 2002]
2. THERMOSPHERE-IONOSPHERE-MAGNETOSPHERE
COUPLING
MAGNETOSPHERE-IONOSPHERE-THERMOSPHERE COUPLING
AURORAL
THERMOSPHERE
IONOSPHERE
Exchange of
particles, momentum & energy
MAGNETOSPHERE
Examples of ITM coupling:
• Angular momentum transfer
• Ion outflow, particle precipitation
MAGNETOSPHERE-IONOSPHERE-THERMOSPHERE COUPLING
• Overall, at high latitudes:
the magnetosphere extracts angular
momentum from the upper atmosphere
through the magnetic field-aligned
currents [e.g., Hill, 1979]
The magnetosphere “swims” on
the ionosphere.
MAGNETOSPHERE
IONOSPHERE
MAGNETOSPHERE-IONOSPHERE-THERMOSPHERE COUPLING
AURORAL
THERMOSPHERE
IONOSPHERE
Exchange of
particles, momentum & energy
MAGNETOSPHERE
Examples of ITM coupling:
• Angular momentum transfer
• Ion outflow, particle precipitation
TIM coupling through current system
TO MAGNETOSPHERE
Field-aligned
current
Field-aligned
current
FROM MAGNETOSPHERE
AURORAL
THERMOSPHERE
S
Ionospheric Joule heating
(resistive + frictional heating)
[Vasyliũnas and Song, 2005]
Q J = J . Ei
MAGNETOSPHERE-IONOSPHERE-THERMOSPHERE COUPLING
ATMOS
MAGNETOSPHERE
Energy redistribution
towards lower
latitudes?
Ion drag fridge mechanism
[Smith et al., 2007; Smith and Aylward, 2009]
EXOSPHERIC
TEMPERATURE
STIM
Stronger
heating
wind
direction
Stronger
cooling
Local time averaged
[Mueller-Wodarg et al., 2011]
Polar sub-corotation due to auroral forcing (westward ion velocities
due to ambient E fields) drives equator-to-pole circulation
Does the ion drag fridge mechanism
rule out auroral energy in solving the
global energy crisis at Giant Planets?
3. MODELING THE THERMOSPHERE/IONOSPHERE SYSTEM
COUPLED FLUID/KINETIC STIM MODEL
Neutral temperatures
* Full ion-neutral dynamical coupling
3D Thermosphere-Ionosphere
Model
Thermospheric densities
(Nn), winds, & temp.
[Müller-Wodarg et al., Icarus,
2006]
*
Ionospheric
densities (Ne, Ni) , drifts, &
temperatures (Te, Ti)
[Moore et al., JGR, 2008]
Electron and ion
densities & temperatures
Electrical conductances
Solar Flux
1D Energy Deposition Model
[Galand et al., JGR, 2009, 2011]
Nn
Beer-Lambert Law
applied to the solar flux
Pe, Pi
Pe, Pi, Qe
Ne, Te
Electric field
Boltzmann Equation
applied to suprathermal
electrons
Incident Auroral
Electron
Distribution
Flexible model which allows us to explore the parameter space and assess the effect of
it on ionospheric/thermospheric quantities, such as Ne, S, Tn.
Pedersen conductance (mho)
STIM RESULT 1: Ionospheric conductances in auroral regions
Q0 = 0.2 mW m-2
12 LT
SOLAR
VALUES
SP 
Q0 (t  t)
with Dt = 25 min


-Pedersen conductivities peak at the homopause  conductances peak near 2.5 keV
- At low energies, conductances are driven by the solar source
[Galand et al., 2011]
Pedersen conductance (mho)
STIM RESULT 1: Ionospheric conductances in auroral regions
Sun only
12 LT
78°S LAT
Pedersen
conductances
mW m-2
Equinox Q0 = 0.2
Summer
Solar max 12 LT Solar min
Equinox
Solar min
0.7 mho
1.6 mho
2.6 mho
SP 
Q0 (t  t)
with Dt = 25 min


-Pedersen conductivities peak at the homopause  conductances peak near 2.5 keV
- At low energies, conductances are driven by the solar source
[Galand et al., 2011]
STIM RESULT 1: Ionospheric Conductances in auroral regions
Auroral electron
mean energy &
energy flux
Earth
Saturn
Jupiter
[Fuller-Rowell and
Evans, 1987]
[Galand et al.,
2011]
[Millward et al.,
2002]
10 keV
1 mW m-2
SP =
4-6 mho
SP =
11-12 mho
SP = 0.10.2 mho
Composition
Altitude range
SP=max(SP)/10 over 70 km
(E) and 500 km (S)
Strength of B field
B field 20 times
stronger at J cp w/ S
Slippage parameter(1) for Jupiter(2) & Saturn(3): k 
 n
 ~ 0.5
 i
(1) Bunce et al. [2003]; (2) Cowley et al. [2004]; (3) Galand et al. [2011]
STIM RESULT 2: sensitivity to vibrationally excited H2 rate
• Charge exchange reaction H+ + H2(v≥4)  H2+ + H
(1)
controls the abundance of H3+ as it is quickly followed by:
H2+ + H2  H3+ + H
• Reaction rate k1* = k1 [H2(v≥4)]/[H2]
– Low k1* means less charge exchange reaction and increase in
ionospheric densities
 k1 = 10-9 cm3 s-1 [Huestis, 2008]
 At low- and mid-latitudes: Moore et al. (2010) found best match
between model and Cassini RSS data for a reduction of
([H2(v≥4)]/[H2]) from Moses and Bass [2000]
 In the auroral regions, expected to be larger: Galand et al. (2011)
assumed 2 x ([H2(v≥4)]/[H2]) from Moses and Bass [2000]
• How does this affect thermospheric circulation?
STIM RESULT 2: sensitivity to vibrationally excited H2 rate
EXOSPHERIC
TEMPERATURE
wind
direction
3 cells
Local time averaged
[Mueller-Wodarg et al., 2011]
4. OBSERVATIONS OF THE THERMOSPHERE/IONOSPHERE SYSTEM
Implication for exospheric temperatures
(1) STIM, 3 cells
(2) STIM, 1 cell
3
(3) Voyager UVS
(Vervack and
Moses, 2011)
1
4
6
+
5
2
(4) Voyager UVS
(Smith et al., 1983)
(5) Cassini UVIS
(Nagy et al., 2009)
(6) UKIRT (Melin et
al., 2007)
[Mueller-Wodarg et al., 2011]
Sample of relevant observations: Earth-based + space missions
Which observations can help constrained the problem?
Type of
observat.
Radio
occultations
H3+ IR
emissions
Physical
quantities
Electron
density
Effective H3+
column
density, temp,
and velocity
vector
UV occultations
UV
emissions
Radio
emissions
Auroral eAuroral eenergy flux
energy flux
& energy
and energy
(generating (within accele
aurora)
region)
In situ measurements
(particle, fields)
Atmosph. densities,
Down/upgoing auroral
Texospheric
particles (if conditions allows),
Magnetic field
strength/direction, Electric
currents
 Combine as many as possible to better constrain the problem
[e.g., Melin,
talk]
JUNO
over the polar regions
Credit: Juno Team
Credit: Juno Team
JUNO observations through the magnetic field lines connected to the
auroral ionosphere, close/within the acceleration region (expected to be
2-3 RJ from center [Ray et al., 2009]):
- Electric currents along magnetic field lines
- Plasma/radio waves revealing processing responsible for particle
acceleration [see Hess, tutorial]
- Energetic particles precipitating into atmosphere creating aurora
- Ultraviolet/IR auroral emissions regarding the morphology of the aurora
Outstanding questions
• Can the energy crisis be solved via auroral forcing alone as proposed
here?
– Is the mechanism proposed efficient at Jupiter, Uranus (seasonal
asymmetry [Melin et al., 2011 + poster]), and Neptune?
– At Saturn, beside the solar contribution which is dominant [Moore et al.,
2010], are they additional energy sources at low- and mid-latitudes?
[e.g., break-down in co-rotation of the ions in the ionosphere [Stallard et al.,
2010; Tao, poster; Ray, talk], molecular neutral torus of Saturn through chargeexchange (ENA) [e.g., Jurak & Johnson, 2001], wave heating (super-rotat≠IR)]
 Further constraints on ionospheric densities at different LT [dawn/dusk
RSS, Max Ne SEDs, ground-based IR in H3+ (noon!)]
• What drive the hemispheric differences observed at Saturn in the
magnetosphere and auroral, ionospheric regions?
– Asymmetry in B field? Hemispheric (seasonal) differences in the
atmosphere? If the latter, should reverse now as going out of equinox?
• Is the variable rotation rate observed in the magnetosphere linked
to atmospheric dynamics? [e.g., Jia/Kivelson talk] (two-way MI coupling)
EXTRA SLIDES
PLANETARY IONOSPHERES in the SOLAR SYSTEM
Ionospheres in the solar system
Mendillo et al. (2002)
ORIGIN of the soft X-RAY and UV SOLAR SPECTRUM
[LASP, TIMED/SEE]
Ionization
Heating of thermosphere

n i e   en  e
 in  i 
 P z   
 2
 2
2
2 
B  en   e  in   i 
n
i
2
2


n
e


e
i
eB  z    i 

H
2
2
2
2 
with  
B  en   e  in   i 
m
n
i

Pedersen (sP)
& Hall (sH)
conductivities:

Pedersen (SP) & Hall (SH)
conductances:
SP 
SH 
  zdz
P
ionosphere
  zdz
H
ionosphere
IONOSPHERIC CONDUCTIVITIES IN THE AURORAL REGION
12 LT
Sun only (0.1 mW m-2 including
4x10-3 mW m-2 for photoe-)
Sun + Soft e- (500 eV, 0.2 mW m-2)
Sun + Hard e- (10 keV, 0.2 mW m-2)
IONOSPHERIC CONDUCTANCES IN THE AURORAL REGION
Em = 10 keV
<Q0> = 0.2 mW m-2
0800 UT
 LT-dependence based on
HST/UV brightness analysis
[Lamy et al., 2009]
0400 – 1400 UT
Pedersen
 P Q0 (t  tP )
tP = 23 min

Electrical conductances proportional to Q0 but shifted in time
[Galand et al., 2011]
ENERGY REDISTRIBUTION FROM HIGH TO LOWER LATITUDES
The “Ion Drag Fridge”
Subcorotating Region
Smith et al., Nature (2007)
Smith and Aylward (2009)
Pole-to-equator flow
Equator-to-pole flow
• Polar sub-corotation due to auroral forcing (westward ion velocities due to
ambient E fields) drives equator-to-pole circulation
• NOTE: this is not simply the return-flow of thermally driven pole-toequator winds higher up!
• Therefore, the poleward flow cools the equatorial regions
• Does this rule out magnetospheric energy in solving the Gas Giant energy
crisis?