ARTICLE IN PRESS
Planetary and Space Science 55 (2007) 165–173
www.elsevier.com/locate/pss
The exosphere of Titan and its interaction with the kronian
magnetosphere: MIMI observations and modeling
P. Garniera,, I. Dandourasa, D. Toublanca, P.C. Brandtb, E.C. Roelofb, D.G. Mitchellb,
S.M. Krimigisb, N. Kruppc, D.C. Hamiltond, H. Waitee
a
Centre d’Etude Spatial des Rayonnements, CNRS/Paul Sabatier University, 9 avenue du Colonel Roche, Toulouse 31028, France
b
Applied Physics Laboratory, Johns Hopkins University, Laurel, MD, USA
c
Max-Planck-Institut für Sonnensystemforschung, Lindau, Germany
d
Department of Physics, University of Maryland, College Park, MD, USA
e
Department of Atmospheric, Oceanic, and Space Sciences, University of Michigan, Ann Arbor, MI, USA
Received 31 March 2006; received in revised form 11 July 2006; accepted 17 July 2006
Available online 6 September 2006
Abstract
Titan’s nitrogen-rich atmosphere is directly bombarded by energetic ions, due to its lack of a significant intrinsic magnetic field. Singly
charged energetic ions from Saturn magnetosphere undergo charge exchange collisions with neutral atoms in Titan’s exosphere, being
transformed into energetic neutral atoms (ENAs). The ion and neutral camera (INCA), one of the three sensors that comprise the
magnetosphere imaging instrument on the Cassini/Huygens mission to Saturn and Titan, images the ENA emissions from various ion/
gas interaction regions in the Saturnian magnetosphere. During Cassinis second orbit around Saturn the spacecraft performed the Ta
Titan flyby (26 October 2004), at an altitude of only 1174 km. INCA data acquired during this targeted close flyby not only confirm
model predictions of dominant finite ion gyroradii effects, but also reveal a much more complex interaction. These observations are
analyzed and compared to simulations. A new Titan exosphere model is then proposed.
r 2006 Elsevier Ltd. All rights reserved.
Keywords: Titan; Exosphere; Energetic neutral atoms; Cassini; Magnetosphere
1. Introduction
The Cassini mission, since the Saturn Orbit Insertion in
July 2004, provides a wealth of information about the
environment of Saturn and its satellites, and thus
completes those given by the Voyager missions. Among
the satellites, Titan is the largest one and has a dense
atmosphere, which interacts either with the kronian
corotating plasma, when Titan is inside the magnetosphere,
or with the solar wind, when it is outside the magnetosphere.
The interaction of Titan with its environment is mainly
given by the absence of an intrinsic magnetic field, so that it
can be Mars or Venus-like. In particular, when Titan
interacts with the corotating plasma of Saturn, which is the
Corresponding author. Tel.: +33 5 61 55 6681.
E-mail address: [email protected] (P. Garnier).
0032-0633/$ - see front matter r 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.pss.2006.07.006
case most of the time, a variety of complex phenomena
takes place (Brecht et al., 2000; Kallio et al., 2004). In
particular, energetic kronian ions will undergo charge
exchange collisions with the cold neutral atoms of the
upper atmosphere of Titan, thus producing energetic
neutral atoms (ENAs).
The magnetosphere imaging instrument (MIMI) experiment (Krimigis et al., 2004), onboard Cassini, comprises
three sensors: charge energy mass spectrometer (CHEMS),
low energy magnetospheric measurement system
(LEMMS) and ion and neutral camera (INCA), which
can function as an ENA imager.
The ENA imaging through INCA is the only way to get
information about the neutral atmosphere of Titan at
altitudes above 2000 km. This sensor gives the ENA flux,
which is proportional to the line of sight integral of the
energetic ion flux (in the direction towards the imager)
multiplied by the cold neutral density (at the charge
ARTICLE IN PRESS
166
P. Garnier et al. / Planetary and Space Science 55 (2007) 165–173
Fig. 1. H ENA image made by INCA during Ta (closest approach at 15 h 30 min), 26 October 2004, for energies between 20 and 50 keV, at an altitude of
about 8000 km and smoothed, with an 8 min exposure time. The colorscale gives the ENA flux integrated along the lines of sight inside the field of view.
exchange reaction location), and by the charge exchange
cross section. Thus, the ENA imaging gives some
information on both the ion and neutral distributions.
Moreover, the sensitivity of the sensor allows us to image
the extended exosphere of Titan at altitudes above the
sensitivity capabilities of neutral mass spectrometers
(INMS provides data up to about 2000 km altitude).
During the Cassini mission preparation phase, the
interaction of the Titan exosphere with the energetic
kronian ions was analyzed and the expected ENA images
were modeled (Amsif et al., 1997; Dandouras and Amsif,
1999). A first Titan exosphere model was developed for the
ENA production modeling (Amsif et al., 1997). The images
given by INCA for the first flybys (Mitchell et al., 2005; see
Fig. 1 for Ta) confirm the dominant finite ion gyroradii
effects: the shadow induced for the parent ions implies a
left/right asymmetry for the ENA image (see Fig. 2 for
finite ion gyroradii effects and Fig. 3 for the modeled ENA
image). Moreover, the limb brightening effect was also
analyzed, as the maximum integrated optical depth for the
instrument. However, Fig. 1 shows a much more complex
interaction than modeled, with in particular a much higher
altitude for the maximum integrated flux than modeled.
In this study, we present a new ENA calculation model,
and a new Titan exosphere model, deduced from the
MIMI/INCA observations. We will here concentrate on
the conditions of the first flyby of Titan, Ta, on 26 October
2004, where Titan was inside the kronian magnetosphere,
and we will only study the hydrogen ENA flux. The
exosphere modeled is mainly a Chamberlain-type model
for the five main neutrals (H, H2 , N, N2 , CH4 ), while the
boundary conditions at the exobase (temperature and
altitude) come from the latest INMS results. The densities
at the exobase are given, for H, N, N2 and CH4 , from
the latest version of the Titan atmospheric model of
D. Toublanc (Toublanc, private communication; adapted
from Toublanc et al., 1995) which is consistent with the
latest INMS data (Waite et al., 2005) and the Vervack
model (Vervack et al., 2004); the H2 density comes from the
INMS data.
2. ENA production parameters
The ENA calculation model used here is, as a first step,
with only one dimension, since we understand that the left/
right asymmetry in the INCA image (Fig. 1), is due to the
finite ion gyroradii effects (Figs. 2 and 3). These induce
indeed a shadow effect, in certain regions, where the
gyroradii of the ions make them cross well below the
exobase, considered as a lower limit for ENA emission,
before they can become ENA and then be detected by
INCA (see Dandouras and Amsif, 1999). There is also a
small asymmetry between the upper and lower parts of the
INCA image, which will however not be studied with our
1D calculations. Here, we note that this asymmetry might
be related to different north–south sunlighting conditions,
ARTICLE IN PRESS
P. Garnier et al. / Planetary and Space Science 55 (2007) 165–173
167
ϕ1
rcg
C
Exobase
−ϕlim
Titan
+ϕli
lc
ϕ2
ϕ3
Camera
rc1g
rc2g
c2
Fig. 2. Equatorial view of the proton gyration in the Titan vicinity (in the plane perpendicular to the kronian magnetosphere magnetic field in which Titan
is embedded). The dashed circle represents the Titan exobase, and the shadowed area shows where the production of ENAs, having trajectories directed
towards the imager, is not possible. See the text for details or Dandouras and Amsif (1999).
Counts
per pixel
60
90
80
50
70
40
60
50
30
40
30
20
20
10
10
0
10
20
30
40
50
60
Fig. 3. Simulation of H ENA imaging, from Dandouras and Amsif (1999), adapted by Nathalie Cazajus by introducing the ENA scattering at the carbon
foil. Simulation for an altitude of about 6000 km, energies between 10 and 50 keV, and a 5.75 min exposure time. The colorscale gives the counts per pixel
for the region corresponding to the right part of Fig. 1.
that could result in a non-spherical exosphere, or to a
north–south asymmetry in the magnetic field around Titan,
through the trajectories of the parent ions.
The main objective here is to explain the altitude profile
for the region not concerned by the gyroradii effects, where
the maximum flux altitude is higher than previously
ARTICLE IN PRESS
P. Garnier et al. / Planetary and Space Science 55 (2007) 165–173
168
analyzed. The principle of the model (Fig. 4) is to calculate
the ENA flux integrated along each line of sight cutting the
axis in red, which gives the ENA flux (integrated along the
line of sight) as a function of the line of sight altitude. The
ENA unidirectional flux measured at energy E and
corresponding to the ion species i, is given by the line of
sight integral (Roelof, 1987):
Z 1
X
j ENA ¼
sik ðEÞ
j i ðE; lÞnk ðlÞ dl,
(1)
k
0
where j i ðE; lÞ is the unidirectional magnetospheric ion flux
ðcm s sr keVÞ1 for Hþ at energy E and along the direction
of the line of sight to the ENA camera from point l, nk ðlÞ is
the density of the neutral exospheric gas for species k and
at l, and sik ðEÞ is the energy-dependent charge exchange
cross section between the kronian Hþ and the exospheric
neutral species k (sik ðEÞ values used: Hsieh, private
communication, for N2 , CH4 , and H2 ; McClure, 1966,
for H; Barnett and Reynolds , 1958, for N). The line of
sight integral runs from the ENA camera position to the
infinity, along the look direction opposite to that of the
arriving ENA. Moreover, this formula implies an ‘‘optically thin’’ atmosphere, where any loss of ENAs between
the emission point and the camera position can be
neglected. This point will be more discussed later.
The proton flux is given by the interpolation from the
LEMMS data (Fig. 5), measured during Ta, and which
correspond to the channels for Hþ ions at the energies
studied. The exobase characteristics come from the INMS
data, with for Ta, at an altitude of about 1425 km, and at a
temperature of 148.5 K. The densities at the exobase, which
are inputs for the Chamberlain-type exosphere model
come, as a first step, directly from the atmospheric model
of D. Toublanc (see details in Section 4).
3. The ENA calculation results
line of sight altitude H
B
ENA
H=0
ϕ
H = -2575 km
camera
Titan
exobase
Fig. 4. Geometry of the 1D model for ENA flux profile around Titan. See
the text for details.
This study focuses on the Titan ENA imaging at high
altitude (Mitchell et al., 2005), with the comparison
between Fig. 1 and the model results. The method used
here is an inversion of the ENA image, by adapting the first
step Chamberlain exosphere model, in order to propose an
exosphere model that fits the INCA data. A proper
inversion would be indeed too much complex, and
probably not unique, since several neutral species play an
important role in the production of ENAs.
A simple calculation, with the conditions corresponding
to Fig. 1, leads to an ENA profile with much higher ENA
Fig. 5. LEMMS A0–A3 channels, giving the Hþ flux (cm2 sr1 s1 keV1 ) measured during Ta.
ARTICLE IN PRESS
P. Garnier et al. / Planetary and Space Science 55 (2007) 165–173
169
ENA flux profile measured and calculated
5
Flux calculated
Minimum measured flux
Maximum measured flux
Flux calculated
Maximum conditions for ions
Flux calculated
Minimum conditions for ions
4.5
4
ENA flux (cm2.sr.s.keV)1
3.5
3
2.5
2
1.5
1
0.5
0
-2000
0
2000
4000
6000
8000
10000
12000
14000
Altitude (km)
Fig. 6. ENA flux profile measured by INCA during Ta and simulated for various ion conditions. See the text for details.
flux values than measured (factor 7), and moreover a much
lower altitude for the maximum ENA flux than observed
(about 150022000 km for the difference). Actually, beyond
the flux values, this maximum flux altitude seems to be the
most important discrepancy with the INCA data. It is
given by the maximum integrated optical depth, when the
line of sight of the instrument is tangent to the exobase,
which is normally the inferior limit for the ENA emission.
Two ways exist to change this maximum flux altitude, in
order to fit the observations. The first one is by introducing
in the calculations the scattering of the particles when
going through the instrument entry foil. It broadens the
theoretical distribution, and, thus can change the maximum flux altitude by desymmetrizing the calculated flux.
But such a desymmetrization, although necessary to
realistically reproduce the instrument performance, cannot
increase enough this maximum flux altitude. The second
way is to change arbitrarily the minimum altitude for ENA
emission observed by INCA in the conditions of Fig. 1
(spacecraft at an altitude of 8000 km), and put it above the
exobase, so that the real maximum flux altitude can be
reached with an emission cut-off at about 180022000 km
altitude (instead of 1425 km for the exobase). This is
equivalent to the introduction of loss mechanisms below
this cut-off. Moreover, by taking into account the finite ion
gyroradii effects near the nadir (Dandouras and Amsif,
1999), the calculated flux are quite similar to the measured
flux, given the various uncertainties. Fig. 6 shows an
example of ENA flux profile with a cut-off at an altitude of
1900 km, using as entry the exosphere model proposed
(shown in Fig. 7; see Table 1). This result shows, at a very
high altitude, ENA fluxes calculated which are higher than
those observed, but this region corresponds to the edge of
the INCA field of view, where the counts should be taken
with some precaution.
The introduction of such a cut-off, much above the
exobase, thus appears to be necessary. A few mechanisms
can lead to this: the existence of loss mechanisms, either for
the parent ions, or for the ENAs through their absorption,
but also an anisotropy in the ENAs trajectories, which
could thus not be detected by the imager in certain
configurations.
A first loss mechanism can be the finite ion gyroradii
effects, which induce a shadow for the parent ions (see
Fig. 2), and explain the asymmetry observed in the INCA
image. However, the simulations performed by Dandouras
and Amsif (1999; Fig. 3) did not take into account the
plasma corotation (circular trajectories were used for the
ions, instead of cissoidal ones). But, both combined could
increase the loss phenomenon and thus induce an
important cut-off, depending on the geometry. Ion
simulations and analytical calculations about such a role
of the corotation are the subjects of an other paper.
However, it appears that the influence of this mechanism
will highly depend on the geometry of the flyby with, in
some configurations, no induced loss for the ions.
A second loss mechanism for the parent ions can be an
influence of the draped magnetic field, which was studied
ARTICLE IN PRESS
P. Garnier et al. / Planetary and Space Science 55 (2007) 165–173
170
x 104
Titan exosphere model
3
N(4S)
H
H2
N2
CH4
2.8
2.6
2.4
2.2
2
Altitude (km)
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
exobase
0
10-13
10-11
10-9
10-7
10-5
10-3
10-1
101
103
105
107
109
density (cm-3)
Fig. 7. Titan exosphere model for the five main species (H, H2 , Nð4SÞ, CH4 , N2 ) with the exobase parameters shown in Table 1.
Table 1
Titan exobase parameters
Parameter
Value
H c (km)
T c (K)
N c ðNÞ ðcm3 Þ
N c ðHÞ ðcm3 Þ
N c ðH2 Þ ðcm3 Þ
N c ðN2 Þ ðcm3 Þ
N c ðCH4 Þ ðcm3 Þ
1425
148.5
7:1 104
4:6 104
5 105
5:9 107
4:9 106
by Ledvina (2005). The MHD model developed showed
that the increase in the magnetic field value, during the
draping around Titan, reduces the gyroradii of the parent
ions, so as to prevent protons at high energies from
entering into the outer part of the Titan atmosphere.
However, hybrid codes (Modolo et al., 2005) show a
different evolution for the protons, which can enter very
deeply into the atmosphere, since the increase in the
magnetic field simulated during the draping is not sufficient
to stop them. Furthermore, the observations by the
LEMMS instrument show that such ions are detected well
under the exobase.
Moreover, the Titan flybys show important ENA flux
measured locally much below the exobase (down to
1000 km altitude at least). Thus, ENAs exist at low
altitude, probably after being firstly created in the exosphere, but they are not detected by remote sensing when
the spacecraft is at high altitude. It seems that these ENAs
should either be absorbed along the line of sight, or their
trajectories are such that they cannot reach the detector.
Regarding the creation and the absorption of the ENAs,
an analytical approach was developed by Roelof (2005) for
calculating the thickness of the atmosphere against charge
exchange, or the ratio, for an impact parameter, between
the flux of ENAs produced and the flux of parent ions. We
made some calculations to estimate the thickness of the
atmosphere against charge exchange by collisions with
neutrals at such energies ð20250 keVÞ, using in parallel
analytical and numerical calculations. But both showed
that the absorption of ENAs by charge exchange collisions
with neutrals can only take place at altitudes below
150021600 km, so that a cut-off at 1900 km (like in
Fig. 6) cannot be explained by such reactions. However,
others processes should also be taken into account for the
absorption of ENAs, like the photoionization by solar UV
or the electron impact ionization, and which might induce
an important absorption, and thus explain the cut-off
around 1900 km altitude. A complete study of the ENAs
absorption will be shown in a future paper.
A last mechanism could play an important role in this
cut-off: a change in the ENAs parent ion trajectories due to
the draping. The interaction between Titan and the
magnetosphere induces in particular, as indicated previously, a draping of the magnetic field. As a consequence,
the trajectories of the parent ions can be changed by this
new configuration of the local magnetic field. In this
ARTICLE IN PRESS
P. Garnier et al. / Planetary and Space Science 55 (2007) 165–173
region, the pitch angle distribution of 20250 keV protons
seems to be mainly centered around 90 , thus with an
evolution of the ions in the equatorial plane, where the
spacecraft orbit took place during the first flybys. If for
example the ion trajectories were moved away from this
plane by the draping around Titan, it could prevent the
ENAs produced by charge exchange from being detected
by INCA at high altitude (but they could still be detected
locally). Actually, the LEMMS and CHEMS data indicate
a change in the protons pitch angle distributions near the
closest approach (during T5, 16 April 2005), towards
0 =180 . Hybrid simulations (Modolo et al., 2005) show
also such an evolution in the pitch angle distribution. This
can be explained by the very large gyroradii (more than 1
Titan radius) of these ions: as a consequence, the protons
cannot follow the draping of the magnetic field, whose
draping occurs along a much smaller scale than the
gyroradius.
The 1D ENA flux calculation model, taking into account
the carbon foil scattering, gives results which are quite
consistent with the observations (Fig. 6). The low measured
flux values at high altitudes (Fig. 6) should be moreover
taken with precaution, since we reach the edge of the INCA
field of view. However, these calculations need to introduce
a cut-off for the ENA emission well above the exobase
(at 1900 km altitude in this case). This limit could be
explained by several mechanisms developed in this section.
The exospheric neutral density profiles also influence the
production of ENAs, not only the flux values, but also to a
certain extent the cut-off needed to fit the observations.
4. The exosphere of Titan
The exosphere model used initially for the ENA flux
calculation model is a Chamberlain-type model. The
Chamberlain formalism for planetary atmospheres
(Chamberlain, 1963; Chamberlain and Hunten, 1987)
allows us to calculate the densities of exospheric neutrals
at any altitude with only the exobase characteristics, by
using this formula for every species:
NðrÞ ¼ N c expððlc lÞÞ zðlÞ,
(2)
zðlÞ ¼ zesc ðlÞ þ zsat ðlÞ þ zbal ðlÞ,
(3)
where lc ¼ GMm=kT c rc (rc ¼ hc þ RTitan ), M is the mass
of Titan, m the corresponding molecular mass, G the
gravitational constant, T c and hc the exobase temperature
and altitude, and RTitan the Titan radius (2575 km).
N c expððlc lÞÞ (l ¼ GMm=kT c r) represents the hydrostatic equation and zðlÞ is a partition function, including
three different partition functions for each particle
population (escaping, satellite and ballistic orbits).
The parameters T c and hc come here from the INMS
results for TA (about 148.5 K and 1425 km), and the
exobase densities come, as a first step, directly from the
latest version of the D. Toublanc photochemical model
(adapted from Toublanc et al., 1995). This new version of
171
the D. Toublanc model is consistent with the INMS results
for the first flyby, with a good agreement for the
homopause altitude and the density profiles (except for
H2 ), and also with the Vervack model (Vervack et al.,
2004).
A few steps were used to develop this exosphere model.
First, by using only the exobase densities of the D.
Toublanc model, the ENA flux profiles obtained were too
low, compared with the INCA image. The maximum flux
value calculated was about half the measured ENA value.
The study of the INMS results for the H2 density profile
during Ta (most important ENA source above 2000 km
altitude, N2 being the most important up to 2000 km
altitude) lead to multiplication by two its density at the
exobase (we take N c ðH2 Þ ¼ 5:105 cm3 ). The ENA flux
profile thus obtained was higher, with a maximum flux
value which was consistent with the measured value.
Finally, we added the satellite population in the
Chamberlain formalism. This population was previously
not taken into account, since the formalism does not allow
us to consider this population properly (it is not linked to
the exobase, which is needed in the formalism). However,
the satellite population is the main contribution of the
exosphere at high altitude (see Amsif et al., 1997), so that it
plays an important role in the ENA imaging of the
extended exosphere (see Brandt et al., 2005). This population follows a 1=r2 law at high altitude, which is very
consistent with the profiles obtained through the Chamberlain formalism. The Chamberlain partition function for
satellite populations was thus used in the exosphere model,
which lead to the final exosphere model proposed (Fig. 7;
Table 1), used in the previous section to calculate the ENA
flux profile.
Thus, the ENA flux profile is very sensitive to the
exospheric densities (mainly of H2 , N2 , and in a lower
extent, CH4 ). As a consequence, we were interested in
estimating the influence of the introduction of non-thermal
profiles for these species (in particular, considering the high
value of the cut-off) by using the non-thermal escape values
either obtained in the bibliography or by INMS during the
first Titan flybys. Actually, the INMS observations (De La
Haye, 2005) indicate the existence of much more nonthermal escape than expected by previous models, and with
a great variability between the flybys (and also between
inbound and outbound). The introduction of the escape
values calculated before Cassini in our exosphere model
has no consequence on the ENA flux, which is not the case
with some of the INMS results. However, even if the ENA
flux profile can change with the introduction of nonthermal escape, with in particular an increase in the
altitude for the maximum ENA flux, inducing a smaller
cut-off altitude, it is always necessary to introduce a cut-off
well above the exobase. Moreover, reciprocally, the
uncertainties for the ENA flux calculation model (uncertainties on the LEMMS/INCA data and the cross
sections), as well as the relatively weak influence of nonthermal escape on the ENA profiles, imply that the
ARTICLE IN PRESS
172
P. Garnier et al. / Planetary and Space Science 55 (2007) 165–173
inversion of the INCA images could probably not
constrain sufficiently the information on the exospheric
profiles to quantify any non-thermal escape.
Finally, we would like to remind a limitation of the
Chamberlain formalism near the exobase, which can have
some consequence on the lightest neutral species.
The principle of the Chamberlain formalism does not
take into account the particles coming into the atmosphere
from the infinity. These constitute the fourth possible type
of population, which must be as important as the escaping
population to maintain a hydrostatic equilibrium. Removing this fourth type of orbit, by using only the first three
ones like in Eq. (3), allows us to study the atmosphere
above the exobase, which is determined by a nonhydrostatic equilibrium, contrary to the region below.
But the use of such a formalism introduces a discontinuity
in the densities at the exobase, insofar as we come brutally
from a hydrostatic equilibrium, with Maxwellian distributions, to an exosphere without the population of entering
particles from the infinity. The discontinuity is most
important for light species, like H or H2 . For example,
with the Titan exosphere model presented in this study, the
discontinuities can reach 20% for the H density at the
exobase, and then easily propagate at higher altitudes.
As a consequence, the Chamberlain formalism should be
used with precaution around the exobase especially for
light species. No formalism exists that allows us to
calculate properly the densities near the exobase. The
arbitrary introduction of entering particles, with a linearly
decreasing partition function for these particles, can help to
ameliorate the transition at the exobase. As far as the
assumption that no particle enters the atmosphere from the
infinity is correct, this problem affects only the lowest part
of the exosphere and almost only H or H2 . However, the
calculation of any integrated densities, like with the ENA
imaging of planetary atmospheres dominated by those
species, introduces a lack of particles precisely because of
the missing of some neutrals near the exobase. But, for the
ENA imaging of the extended Titan exosphere, the
uncertainties on the instrumental measurements used in
the calculations largely exceed the effect induced by this
lack of particles.
5. Discussion and summary
Following the first Titan flybys by Cassini (mainly Ta),
and the acquisition of the first ENA images of the Titan
exosphere interaction with the kronian magnetosphere, an
analysis of the ENA flux profile is presented, compared
with previous models.
A new 1D ENA flux calculation model is developed here,
in order to understand this H ENA flux profile. We use, as
input for Eq. (1), the LEMMS data for the protons and a
new exosphere model for the main neutrals. The results
show the necessity for a lower limit for remotely sensed
ENA emission well above the exobase (at about
180022000 km altitude).
A few mechanisms may play an important role in the
existence of this cut-off, like the absorption of ENAs
through charge exchange (with adding electron impact
ionization and photoionization), the finite parent ion
gyroradii effects combined with corotation, and also
eventually a change in the ENA parent ion trajectories
due to the draping of the magnetic field around Titan.
The motion of the Cassini orbiter during the acquisition
of the INCA images may also transform to a certain extent
the real ENA fluxes. However, the displacement of the
ENA peak around Titan due to this effect has been
calculated and it is less important than the spatial
resolution of the INCA images.
After a few steps, the exosphere model, based on the
D. Toublanc photochemical model (new version), was
adapted to fit the INMS data up to 2000 km altitude (for
N2 , CH4 and H2 ). The ENA flux profile obtained is very
consistent with the INCA data. A new Titan exosphere
model is thus proposed in this study, corresponding to the
Ta encounter, for the five main species: N2 , CH4 , H2 , H
and Nð4SÞ.
A forthcoming paper will complete this work on the
ENA imaging around Titan. First, in order to explain the
high altitude for the maximum ENA flux, we need to
analyze the absorption of ENAs by both electron impact
ionization and photoionization, quantify precisely the finite
ion gyroradii effects by taking into account the corotation,
and also look at the eventual change in the ENA/ions
trajectories by the draping, thanks to ions simulations and
data analysis. Moreover, the INMS results, with in
particular the observation of non-thermal profiles, will be
studied to refine our exosphere model. Then, a statistical
analysis of all the flybys will follow, since the INCA data
show a great variability at each encounter. Finally, a last
step will be the development of a comprehensive 3D model,
which will allow a comprehension of the ENA imaging
around Titan in various configurations.
Acknowledgments
We are grateful to Prof. Ke Chiang Hsieh for his great
work on the determination of cross sections for various
reactions, which was very useful in this study. We would
also like to thank Ronan Modolo and Gerard Chanteur for
their help on various issues of this study, based on their
hybrid code on the Titan interaction with its environment.
We also wish to thank R. Johnson for helpful discussions
and comments.
References
Amsif, A., Dandouras, J., Roelof, E.C., 1997. Modeling the production
and the imaging of energetic neutral atoms from Titan’s exosphere.
J. Geophys. Res. 102, 22169–22181.
Barnett, C.F., Reynolds, H.K., 1958. Charge exchange cross sections of
hydrogen particles in gases at high energies. Phys. Rev. 109 (2),
355–359.
ARTICLE IN PRESS
P. Garnier et al. / Planetary and Space Science 55 (2007) 165–173
Brandt, P.C., et al., 2005. The neutral H2 cloud around titan obtained
from Cassini/INCA observations. AGU, New Orleans.
Brecht, S.H., Luhmann, J.G., Larson, D.J., 2000. Simulation of the
saturnian magnetospheric interaction with Titan. J. Geophys. Res.
105, 13119–13130.
Chamberlain, J.W., 1963. Planetary corona and atmospheric evaporation.
Planet. Space Sci. 11, 901.
Chamberlain, J.W., Hunten, D.M., 1987. In: Donn, W.L. (Ed.), Theory of
Planetary Atmospheres: An Introduction to their Physics and
Chemistry. Academic, San Diego, CA, pp. 330–415.
Dandouras, J., Amsif, A., 1999. Production and imaging of energetic
neutral atoms from Titan’s exosphere: a 3-D model. Planet. Space Sci.
47, 1355.
De La Haye, V., 2005. Coronal formation and heating efficiencies in
Titan’s upper atmosphere: construction of a coupled ion neutral and
thermal structure model to interpret the first INMS Cassini data.
Ph.D. Thesis, University of Michigan.
Kallio, E., Sillanpaä, I., Janhunen, P., 2004. Titan in subsonic and
supersonic flow. Geophys. Res. Lett. 31, L15703.
Krimigis, S.M., et al., 2004. Magnetosphere imaging instrument (MIMI)
on the cassini mission to Saturn/Titan. Space Sci. Rev. 114, 233–329.
173
Ledvina, S.A., 2005. Ion distributions in Saturn’s magnetosphere near
Titan. J. Geophys. Res. 110.
McClure, G.W., 1966. Electron transfer in proton-hydrogen-atom
collisions: 2–117 keV. Phys. Rev. 148, 47–54.
Mitchell, D.G., et al., 2005. Energetic neutral atom emissions from Titan
interaction with Saturn’s magnetosphere. Science 308, 989–992.
Modolo, R., Wahlund, J., Chanteur, G.M., 2005. Global hybrid
simulations of the corotating plasma interaction with the neutral
environment of Titan (configuration of the TA flyby of Titan by
Cassini). AGU, San Francisco.
Roelof, E.C., 1987. Energetic neutral atom image of a storm-time ring
current. Geophys. Res. Lett. 14, 652–655.
Roelof, E.C., 2005. Theory of ‘‘Optically-Thick’’ ENA emission from
Titan’s exosphere. AGU, New Orleans.
Toublanc, D., et al., 1995. Photochemical modeling of Titan’s atmosphere.
Icarus 113, 2–26.
Vervack, R.J., et al., 2004. New perspectives on Titan’s upper atmosphere
from a reanalysis of the Voyager 1 UVS solar occultations. Icarus 170,
91–112.
Waite, J.H., et al., 2005. Ion neutral mass spectrometer results from the
first flyby of Titan. Science 308, 982–986.