Click
Here
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 113, A07217, doi:10.1029/2007JA012899, 2008
for
Full
Article
Interaction evidence between Enceladus’ atmosphere and
Saturn’s magnetosphere
S. Wannawichian,1 J. T. Clarke,1 and D. H. Pontius Jr.2
Received 19 October 2007; revised 13 January 2008; accepted 27 February 2008; published 30 July 2008.
[1] Saturn’s ultraviolet (UV) auroras are highly variable, with much of the activity
believed to be controlled by conditions in the solar wind. Like Jupiter, Saturn is also
expected to have an electrodynamic interaction between the planet’s ionosphere and its
satellites’ atmospheres, a process that is responsible for the moon’s auroral footprint
components at Jupiter. Recent observations from several instruments on the Cassini
spacecraft, e.g., the UltraViolet Imaging Spectrograph (UVIS) and the Cassini Plasma
Spectrometer (CAPS), show strong evidence that Enceladus makes significant plasma
contributions to Saturn’s magnetosphere. Modeling by Pontius and Hill (2006) suggests
that Enceladus’ mass loading region may be comparable in extent to Io’s, whose auroral
footprint emissions are clearly detected. However, the upper limits of Enceladus’ magnetic
footprint emissions are predicted to be only a few tenths of kilo-Rayleighs (kR). We
present results of a search for Enceladus’ auroral footprint using series of UV images of
Saturn’s aurora taken by the Hubble Space Telescope (HST) Space Telescope Imaging
Spectrograph (STIS) in January 2004 and the Advanced Camera for Surveys (ACS)
between February 2005 and January 2007. We detected no auroral evidence of the
electromagnetic connection between Enceladus’ atmosphere and Saturn’s ionosphere, in
agreement with modeling predictions. As our study could detect no signature of currents
linking Saturn’s ionosphere and Enceladus’ interaction region, the implication is that the
field-aligned currents associated with mass loading at Enceladus are too weak to cause
auroral emission.
Citation: Wannawichian, S., J. T. Clarke, and D. H. Pontius Jr. (2008), Interaction evidence between Enceladus’ atmosphere and
Saturn’s magnetosphere, J. Geophys. Res., 113, A07217, doi:10.1029/2007JA012899.
1. Introduction
[2] Interactions between a planet’s magnetosphere and
its satellites have been intensively studied for decades
[Bigg, 1964; Goldreich and Lynden-Bell, 1969], especially
the Galilean satellites of Jupiter [Kivelson et al., 2004;
Clarke et al., 2004]. Io, Ganymede, and Europa are three
large satellites with well established electrodynamic interactions with Jupiter’s magnetosphere. Strong evidence of
these interactions is seen as bright auroral footprints in
Jupiter’s auroral region (about 10– 100 kR, where 1 kR =
109 photons cm2 s1 into 4p steradians), located equatorward of the main auroral emissions at the magnetic
footprint of the field line threading each satellite [Clarke et
al., 2004]. Interactions between satellites and the magnetosphere depend upon collisions between the planet’s corotating plasma and particles in the satellite’s atmosphere or its
surface. Ordinary ion-neutral collisions produce electrical
1
Center for Space Physics, Boston University, Boston, Massachusetts,
USA.
2
Physics Department, Birmingham-Southern College, Birmingham,
Alabama, USA.
Copyright 2008 by the American Geophysical Union.
0148-0227/08/2007JA012899$09.00
Pedersen and Hall currents, while ionization produces a
pickup current as newly charged particles are incorporated
into the plasma flow [Goertz, 1980; Hill et al., 1983]. These
currents are perpendicular to the magnetic field and are
driven by the electric field driving the plasma E B drift.
These perpendicular currents eventually diverge due to
spatial variations, and closure occurs via currents traveling
along magnetic field lines. The local interaction is thereby
coupled to plasma further away along the magnetic field and
eventually to the planet itself. One can also model the
transmission of these currents via a shear Alfvén wave
propagating along magnetic field lines to the ionosphere
[Neubauer, 1980]. The satellite interaction can thus be
studied by looking for auroral emission near the magnetic
footprint of the satellite in the planet’s ionosphere.
[3] In general, auroral emissions in a planet’s ionosphere
provide remotely observable signatures of processes within
its magnetosphere. The emissions take place due to excitation
of the neutral atmosphere caused by collision of high-energy
magnetospheric particles. The Earth’s auroral emissions are
produced mainly by solar wind activity, while Jupiter’s main
auroral oval is driven by the outward transport of Iogenic
plasma in Jupiter’s strong magnetic field [Hill, 2001, 2004;
Cowley and Bunce, 2001; Clarke et al., 2005a], with some
evidence for solar wind influence [Gurnett et al., 2002;
Nichols et al., 2007]. Saturn’s auroral oval is influenced
A07217
1 of 8
A07217
WANNAWICHIAN ET AL.: ENCELADUS-SATURN INTERACTION
more by solar wind pressure than Jupiter’s, as demonstrated
by an offset toward the midnight sector of about 3 –4°,
similar to terrestrial aurora [Clarke et al., 2005a; Milan et al.,
2003; Whalen et al., 1985]. For comparison Jupiter’s main
oval appears to be shifted toward midnight by about 1°
[Grodent et al., 2003]. Saturn’s auroral emissions, like
Jupiter’s, vary more slowly than the Earth’s in terms of local
brightening and latitude and longitude variations. In addition, Saturn’s auroral features move at 20– 70% of Saturn’s
rotation rate and have unique local variations, e.g., during an
auroral storm, the emissions brightened toward the pole and
in the dawn sector [Clarke et al., 2005a].
[4] While the study of Saturn’s aurora progresses via an
increasing number of observations, intriguing evidence of a
significant plasma source in the vicinity of Enceladus and
the E-ring region has been found. Three sets of data from
the Voyager Ultraviolet Spectrometer [Shemansky and Hall,
1992] showed the H Ly a emission brightest within a few
RS from Saturn, implying a strong source of neutral hydrogen near Saturn. The corresponding predicted density of OH
in the plasma sheet near 4.5 RS (between the orbits of
Enceladus and Dione) is about 40 cm3. Later, the Hubble
Space Telescope (HST) Faint Object Spectrograph [Hall et
al., 1996] measured OH fluorescent emissions and found
the average OH number density to be 350 –700 cm3 at
path lengths of 3 – 4 RS, with the assumption that the gas is
confined to equatorial distances between 1.2 and 2.8 RS. In
addition, Jurac and Richardson [2005] modeled the plasma
and neutral environment in Saturn’s magnetosphere using
Voyager plasma and ultraviolet H observations and HST
OH measurements as constraints. They concluded that a
total injection of 1028 water molecules per second is
required near Enceladus’ orbital distance and the E-ring.
[5] More recently, the Cassini spacecraft has returned
evidence for the important role that Enceladus plays in
Saturn’s magnetosphere. Early flybys revealed pronounced
magnetic perturbations that suggested a strong electromagnetic interaction [Dougherty et al., 2006], and these findings
prompted closer approaches on later orbits. On 14 July
2005, the Imaging Science Subsystem (ISS) [Porco et al.,
2006] observed jets of fine icy particles feeding a large
plume ejected from the south polar terrain (55°S). Stellar
occultation observations by the Ultraviolet Imaging Spectrograph (UVIS) confirmed the existence of a regionally
confined water vapor plume and suggested a source of
>150 kg s1 of water molecules [Hansen et al., 2006].
Burger et al. [2007] employed a Monte Carlo neutral model
to model the water gas density and column density measured by Cassini’s Ion and Neutral Mass Spectrometer
(INMS) and UVIS. They estimated a plasma injection rate
of 2 – 3 kg s1 from interactions between magnetospheric
plasma and a plume source of 1028 H2O s1. Contributing
this amount of mass to Saturn’s magnetosphere is more than
sufficient to supply the estimated mass loss rate of the Ering of 1 kg s1 [Hansen et al., 2006].
[6] In contrast, the Cassini Plasma Spectrometer [CAPS]
indicated significant plasma deflections at distances up to
30 Enceladus radii (where 1 REN = 250 km) from the
satellite [Tokar et al., 2006], whose explanation requires a
much larger mass loading rate. As reported in more detail by
Pontius and Hill [2006], the estimated source magnitude is
A07217
proportional to either the Pedersen conductance of Saturn’s
ionosphere or the local Alfvén conductance, depending on
the coupling model employed. The plausible range of those
parameters implies a minimum total mass injection rate of
order 100 kg s1. According to this evidence, Enceladus
could be a very significant plasma source for Saturn’s
magnetosphere. Khurana et al. [2007] modeled the magnetic
field perturbations by assuming a simple representation for
the electric currents. Their results suggest a plasma source
rate <3 kg s1, which agrees more with the neutral plume
studies. However, this estimate depends strongly on their
assumption that the interaction volume has a diameter of
5 REN, which cannot explain the observed flow deflection.
[7] In this work, another piece of the puzzle will be
explored to help us understand the interaction between
Saturn’s magnetosphere and Enceladus’ atmosphere. We
present the results of an intensive search for Enceladus’
auroral footprint from our archive of HST Saturn UV
auroral images. We use the Z3 model for Saturn’s magnetic
field [Connerney et al., 1982, 1983] to locate Enceladus’
footprint positions and predict their locations as a function
of the satellite’s orbital longitude (OLG). We derive an
upper limit to Enceladus’ footprint’s brightness as well as a
rigorous interpretation of the uncertainty in the analysis.
The result will be compared to the distribution and magnitude of field-aligned currents and hence of energy deposition in the auroral regions along field lines connected to
Enceladus’ vicinity as predicted by the model of Pontius
and Hill [2006]. Our results represent a significant step
toward understanding the connection between Enceladus’
interaction region and Saturn’s ionosphere. An important
question is whether the electrodynamic interaction of
Enceladus with Saturn’s magnetosphere is effectively
communicated with Saturn’s ionosphere by field-aligned
currents, as is the case for Io, Europa, and Ganymede.
The presence of an auroral footprint would definitively
answer this question.
2. Observations and Data Reduction
[8] A collection of far-ultraviolet (FUV) images of Saturn
auroral emissions were obtained during HST campaigns in
January 2004, February 2005, October 2005, and January
2007. The 25MAMA (Multi Anode Micro channel Array)
detector on the Space Telescope Imaging Spectrograph
(STIS) was used to obtain the images [Clarke et al.,
2004] until the summer of 2004. Later continuous observations were made by the Advanced Camera for Surveys
(ACS) instrument [Tran et al., 2002]. The 25MAMA
detector on STIS and Solar Blind Camera (SBC) on ACS
have a similar bandpass (115– 170 nm) including the H2
Lyman bands, the Werner bands, and the H Ly a line. For
clear images the conversion factors from counts per second
per pixel to kR is 0.0013 for 25MAMA (STIS) and 0.0021 for
F115LP (ACS). Filtered images were taken simultaneously
with the 25MAMA detector combined with F25SRF2 filter
(>130 nm) or the ACS added with F125LP (>125 nm) and
F140LP (>140 nm) filters. Consequently, the flux conversion
factors are 0.00053 cts1 s1 kR1 per pixel for F25SRF2
[Grodent et al., 2005], 0.00028 cts1 s1 kR1 per pixel for
F125LP, and 0.00056 cts1 s1 kR1 per pixel for F140LP
(according to the sensitivity curve for ACS).
2 of 8
A07217
WANNAWICHIAN ET AL.: ENCELADUS-SATURN INTERACTION
A07217
Figure 1. (Left) A raw image of Saturn taken by HST/ACS on 17 February 2005. (Right) The emission
due to reflected sunlight is modeled according to a Generalized Minnaert function [Vincent et al., 2000]
and a best fit of the measured north/south banding pattern.
[9] The reduction process for STIS and ACS was done by
the pipeline developed and tested separately at Boston
University and University of Liège [Clarke et al., 2005a,
2005b]. For both STIS and ACS images, the reduction
procedure initially subtracts a dark image, applies a flatfield correction, and corrects the instrument’s geometric
distortion of the raw images. Later, the images are converted
to kR, rotated to orient the planet’s north pole at the top of
the image, and rescaled to make Saturn appear as it would
from a standard distance of 8.2 AU.
from the planet’s atmosphere. After convolving with the
measured point spread function, the complete model is
shown in Figure 1b. The residual images are then obtained
by subtracting this background from the original reduced
image, as shown in Figure 2. Note that the main auroral oval
appears clearly even in the quiet event on 17 February 2005.
This technique has been used in a previous study for the
case of Jupiter’s aurora [Clarke et al., 1998].
3. Analytical Procedures and Theoretical
Interpretation
3.1. Background Modeling for Reflected Sunlight
[10] Saturn’s auroral emission is typically 1 – 10 kR,
comparable to the background due to reflected sunlight in
ultraviolet wavelengths, as shown in Figure 1a. The background brightness is therefore significant and presents a
major challenge for detecting faint emissions. The largescale surface brightness variations resulting from limb
darkening can be modeled by using a modified Minnaert
formulation with empirical coefficients [Vincent et al.,
2000]:
lnð mI=F Þ ¼ C0 þ C1 x þ C2 x2 þ C3 x3 ;
where x = ln(mm0), m and m0 are the cosines of the
observation and solar zenith angles, respectively, and the
coefficients are derived by a least squares fit to the intensity
at a selected latitude range.
[11] To remove the background emission, we use the
Minnaert function including a fit to the north/south banding
emission from observed images. The scaling factor is
chosen so that the mean background level is no lower than
1 kR in the clear images, referring to the emission of H Ly a
Figure 2. Residual image results from subtracting modeled
emission due to reflected sunlight from the original image
with the background level set to 1 kR, referring to H Ly a
emission from the planetary atmosphere.
3 of 8
A07217
WANNAWICHIAN ET AL.: ENCELADUS-SATURN INTERACTION
A07217
Figure 3. (upper) The upper limit of the emission from Enceladus’ magnetic footprint is defined by the
average value of the brightest point sources (approximately the size of PSF) near the predicted location
from the Z3 magnetic field model [Connerney et al., 1982, 1983]. Three-dimensional images are
projected onto a two-dimensional map of planetary latitude and longitude. The mapped location of the
footprint in the original image (white square) is zoomed in (yellow rectangle). (lower) One example from
observation in February 2005 shows the background brightness distribution as well as the distribution of
the bright spot with average over the cell size of PSF. The mean value of the background is 1.05 kR,
while the upper limit of brightness is 3.10 kR.
3.2. Magnetic Mapping and Footprint Prediction by
Magnetic Field Model
[12] We use the Z3 magnetic field model [Connerney et
al., 1982, 1983] to map the magnetic latitude and longitude
of Enceladus’ magnetic footprint in Saturn’s ionosphere,
which should be close to the ionospheric terminus of the
coupling system. The Saturn auroral UV images taken by
HST are then projected onto a planetocentric latitudelongitude grid (Figure 3). We improve the signal-to-noise
ratio by summing all the images taken within the same HST
orbit. Since Enceladus’ orbital velocity is slower than
Saturn’s rotation, its magnetic footprint shifts eastward with
later observing times. To compensate, each image is shifted
accordingly so that the predicted magnetic footprints coincide before the images are added.
[13] We note that the ionospheric signature may not
appear exactly at Enceladus’ magnetic footprint as identified by the Z3 model, or any global field model, because it
does not include magnetic field perturbations induced by the
coupling currents themselves. Instead, the footprint should
lag in longitude by an amount that depends upon the
strength and the nature of the coupling. This is observed
at Jupiter, where Io’s footprint is observed to lag sometimes
by as much as several degrees, although the viewing
geometry and magnetic field model uncertainties make a
precise assessment of longitude difficult [Gérard et al.,
2006a; Clarke et al., 1998].
[14] Theoretically, the magnitude of any such lag is
expected to depend upon Io’s magnetic latitude and hence
its position within the plasma torus, which varies substantially due to the tilt of Jupiter’s magnetic field. Given the
close alignment of Saturn’s dipole and rotation axes, any lag
in Enceladus’ footprint would probably be less variable.
Hence by aligning the mapped footprint, we expect that any
signatures would lag by a consistent amount, so that
combining multiple images would still accumulate any
positive signature in a common location. The lower source
rate suggests that any lag would be smaller as well, so that
any auroral emissions would not be far from the mapped
footprint and should still fall within the regions we analyze
4 of 8
WANNAWICHIAN ET AL.: ENCELADUS-SATURN INTERACTION
A07217
A07217
Table 1. Upper Limit to Emission Near Enceladus’ Magnetic Footprint Compared to Those of Io, Ganymede, and Europa [Clarke et al.,
2002]
Brightness Near Enceladus’ Magnetic Footprint
Io
Ganymede
Europa
8 – 30 Jan 04
17 Feb 05
26 Oct – 2 Nov 2005
13 – 26 Jan 07
Several 100 kR*
Few 10 kR*
Few 10 kR*
1.81 – 2.99 ± 1.08 kR
2.83 – 3.39 ± 1.04 kR
2.49 – 4.13 ± 1.04 kR
2.24 – 3.83 ± 1.05 kR
3.3. Coupling Currents
[15] We can calculate the expected brightness produced at
Saturn’s ionosphere using the model of Pontius and Hill
[2006] (hereafter referred to as PH06) for the Enceladus
interaction. In earlier models for Io such as those by Goertz
[1980] and Hill and Pontius [1998], field-aligned currents
were concentrated into infinitely thin sheets because the
source region was assumed to terminate abruptly at its outer
edge. The same is true for the model of Khurana et al.
[2007] for Enceladus. Such an infinite current density is
obviously unphysical and gives no insight about the actual
current density into the ionosphere or the energy deposited
there from which auroral brightness could be determined.
[16] The spatially variable source in PH06’s model rectifies this shortcoming. Following Goertz [1980], they
treated mass loading per unit magnetic flux as an effective
pickup conductance, and any change in that conductance
produces a divergence in the pickup current. They approximated a spatially variable conductance using a large sum of
nested constant values separated by infinitesimal changes,
and the electric fields produced by all those infinitesimal
changes are accumulated to obtain the total field. Generalizing their discrete equations gives a smoothly varying
differential equation for the ionosphere electric field
throughout the region where the source varies. Combining
that with the ionospheric Pedersen conductance SP and
taking its divergence gives the field-aligned current density
at the ionosphere.
[17] Upon assuming a Pedersen conductance of 0.1 S,
their nominal solutions (cf. Pontius et al. [2006] based on
Case B in PH06) gives a maximum field-aligned current at
the ionosphere of 0.06 mA m2. Using Knight’s [1973]
model, this value is approximately three times the current
that can be carried in the absence of a magnetic fieldaligned potential drop, based on a hot electron population of
temperature 12 eV and density 0.2 cm3 [Tokar et al., 2006].
To obtain the needed current density requires a potential drop
of only several tens of volts, and the resulting energy
deposition in the ionosphere is 5 – 10 mW m2. This suggests
a maximum UV auroral emission of brightness 0.1 kR.
[18] Several features of this analysis must be noted. First,
Pontius and Hill’s model determines the electric field from
the measured flow deflection, and the result depends on the
ratio of the total source rate to the ionospheric Pedersen
conductance. In contrast, the magnitudes of the currents
themselves depend on the absolute values of those quantities. Hence the absence of an identifiable signature can be
used to place a strict upper limit on SP. Second, their model
allows for coupling to Saturn’s ionosphere or to an intermediate Alfvén wing, and in the latter case the Alfvén
conductance SA takes the place of SP. However, present
estimates imply that SA exceeds SP by an order of magnitude. This would imply a concomitantly larger source rate,
although the source rate of 100 kg/s derived from SP = 0.1
S is already at the upper end of the range of estimations. We
refer the reader to Pontius and Hill [2006] for a complete
explanation.
4. Results: Footprint Brightness and
Uncertainty Analysis
[19] The numerical results of our investigations are
shown in Table 1. According to previous observations, the
interaction between Jupiter and three of its satellites, i.e., Io,
Ganymede, and Europa, is significant enough that we can
clearly see and obtain the brightness of their auroral footprints in UV wavelengths [Clarke et al., 2002]. However, as
shown by the example in Figure 3, there is no prominent
emission in the region of Enceladus’ magnetic footprint.
The mass-loading region, assumed to be a circular area of
30 REN in the equatorial plane maps along Z3 magnetic
field line into a region approximately 0.097 arcsec across
in the ionosphere. This is approximately the size of the
STIS and ACS point spread functions (PSF). The upper
limit of the emission is then calculated by averaging the
intensity of brightest sources near the expected location
with the size of PSF. According to Table 1, the emission
near Enceladus’ magnetic footprint is not more than a
few kR, which is 1– 2 orders of magnitude lower than those
of Io, Ganymede, and Europa [Clarke et al., 2002].
[20] The uncertainties are based on an analysis employing
two independent methods. First, Poisson statistics considers
the uncertainty according
pffiffiffiffito the average count values or the
random error given by N , where N is number of counts. The
second method, total uncertainty, refers to the uncertainty
calculation based on the distribution of the brightness of point
source regions close to the predicted location. This will reveal
any systematic error in the data. After the background has
been subtracted, some low-order variation generally remains
due to the reflected sunlight. This is corrected by linear
interpolation in latitude and longitude over the region surrounding the predicted location of magnetic footprint, such
that the resulting variation is overall uniform at 1 kR. In
addition, the nearly Gaussian distribution of values of point
source regions suggests a random distribution, from which
we can determine the standard deviation. The comparison
from both methods shows that the systematic error (total
measured uncertainty) is more than the random (Poisson)
uncertainty. Therefore our calculations indicated that systematic errors dominate. The main source of systematic error may
be the background modeling, which mostly matches with the
original image, but there may also be other instrumental
effects.
[21] Because the location of Enceladus’ magnetic footprint is expected to change due to the difference between
Enceladus’ orbital velocity and Saturn’s spin, the vicinity of
5 of 8
A07217
WANNAWICHIAN ET AL.: ENCELADUS-SATURN INTERACTION
Figure 4. Upper limit to Enceladus’ auroral magnetic
footprint brightness observed by STIS (January 2004) and
ACS (February 2005 – January 2007) on HST. The error
bars represent the total 1s uncertainty measured from the
brightness distribution of the PSF cells. The background
values correspond to H Ly a from reflected sunlight,
constrained to 1 kR.
the predicted location of footprint emission is tested for that
variation. The criterion to test whether emission is related to
Enceladus is that it remains within the predicted footprint
location in multiple images. However, our data provide no
evidence for any such relation. Furthermore, taking the
brightest feature observed within 10° of the predicted
magnetic footprint as an upper limit on the emission
brightness, those upper limits range over a few standard
deviations.
[22] Comparing observations from January 2004 to
January 2007 reveals no significant change in the upper
limit of Enceladus’ magnetic footprint brightness (Figure 4).
Since varying solar wind conditions are correlated with the
total auroral brightness [Clarke et al., 2004; Gérard et al.,
2006b], we compared the aurora input power observed
during January 2004 to the upper limit of the footprint
brightness. The brightness variation is too small to exceed
the uncertainty of the background (1 kR) and does not
correspond to the different total auroral input powers (see
Figures 4 and 5).
A07217
emissions. Applying this logic to other models is somewhat
tempered by the need to specify the spatial distribution of
current density, which no other model provides at this time.
For example, the model of Khurana et al. [2007] treats
field-aligned currents as infinitely thin sheets.
[24] The flow disturbance itself corresponds to the perturbation electric field that drives electric currents, and
PH06’s model is specifically designed to model that flow
pattern. Taking that electric field as given, the field-aligned
current density would be larger if the ionospheric conductance were higher. The source rate would then have to be
increased proportionally, and the value of 100 kg s1 is
already higher than other model predictions. Hence the
model we use represents an upper limit on current density,
and using an estimated conductance of 0.1 S, the emission
due to field-aligned currents at Enceladus’ magnetic footprint cannot be more than 0.1 kR. Our observational study
places an upper limit on emission near Enceladus’ magnetic
footprint of a few kR, consistent with the assumed values.
The auroral emissions in the Knight model scale as the
square of the field-aligned current density, and therefore
provide a limit on the combined parameters that determine
the current density. For example, a higher assumed ionospheric conductivity would lead to stronger currents and
brighter auroral emissions.
[25] Even though some auroral emissions were detected,
the statistical significance is not strong enough to support
the identification of Enceladus’ auroral magnetic footprint
emission. There are occasional features of statistical significance that could correspond to Enceladus’ magnetic footprint, at least in principle. To definitively associate an
emission feature with Enceladus’ magnetic footprint, the
feature should persist over multiple observations and should
move with Enceladus’ position, not with the planet’s rotation. However, we find no such feature. In our search, the
emissions appear randomly at different places on different
days.
[26] After subtracting the background due to reflected
solar continuum from the original images, the remaining
reflected solar Lya is 1 kR. The upper limit of the
5. Discussion
[23] The existence of significant electrodynamic interactions between certain satellites and their host planet has long
been recognized. For the case of Io and Jupiter, a large body
of evidence implying a tight interaction has been obtained
including bright prominent auroral magnetic footprints
[Clarke et al., 2004]. For the case of Enceladus and Saturn,
a number of observations suggest Enceladus as a significant
contributor to Saturn’s magnetospheric plasma. Therefore a
qualitatively similar interaction is expected. From theoretical calculations based on plasma deflection measured by
CAPS, the interaction region at Enceladus is comparable in
extent to Io’s highly concentrated atmosphere, but with a
smaller total source [Pontius and Hill, 2006]. The implied
source in that model is at the upper end of available
estimates, and any smaller source would produce weaker
Figure 5. Comparison between aurora brightness and
Enceladus’ magnetic footprint’s brightness observed by
STIS instrument on HST during the campaign on January
2004.
6 of 8
A07217
WANNAWICHIAN ET AL.: ENCELADUS-SATURN INTERACTION
brightness varies from 2 – 4 kR, statistically a few times
higher than one standard deviation of the background level
(1 kR) (Figure 3, lower panel). That implies an intriguing
signal-to-noise ratio for the emission’s brightness. However,
the brightness variation is not responsive to the changing
auroral input power (Figure 5), as happens for the main
auroral oval [Clarke et al., 2005a, 2005b]. This result
suggests that our observed upper limit of footprint brightness has no connection with the main auroral oval.
[27] Our study reveals no observed signature of a major
connection between Enceladus’ atmosphere and Saturn’s
ionosphere. It is possible that there are other emissions
related to Enceladus’ magnetic footprint farther from the
predicted location due to the uncertainty of the magnetic
field model and/or lags in longitude due to perturbations in
the magnetic field. Consequently, precisely predicting the
location of the magnetic footprint emission may be difficult.
[28] According to the evidence, the observed large mass
pick-up rate from the vicinity of Enceladus shows a strong
interaction between Saturn’s magnetospheric plasmas and
Enceladus’ atmosphere. However, our search for an
Enceladus auroral footprint suggests that the communication
from Enceladus’ interaction region to Saturn’s ionosphere
cannot be directly observed with current instrumentation.
We cannot identify any persistent emissions at or near the
calculated magnetic footprint, so we can only determine an
upper limit on the brightness of any such feature.
6. Conclusion
[29] The calculations and observation of the plasma and
neutral environment near Enceladus and the E ring [Jurac
and Richardson, 2005; Porco et al., 2006; Hansen et al.,
2006; Tokar et al., 2006] suggest that Enceladus is a
significant plasma source in Saturn’s magnetosphere. However, the theoretical calculation of electrodynamical interactions at Enceladus [Pontius et al., 2006] predicts that
auroral emissions due to mass loading from Enceladus
would not exceed the upper limit we find of a few kRs.
Our results did not detect any emission at or near the
predicted magnetic footprint of Enceladus that could be
identified as the signature of a strong connection between
Enceladus’ mass loading region and Saturn’s ionosphere.
[30] Acknowledgments. We acknowledge helpful discussions with
Thomas Hill, and we acknowledge programming assistance from Jonathan
Nichols. This work is based on observations with the NASA/ESA Hubble
Space Telescope, obtained at the Space Telescope Science Institute, which
is operated by the AURA Inc. for NASA. This research was supported by
NASA grant HST-GO-10506.01-A and HST-GO-10862.01-A from the
Space Telescope Science Institute to Boston University.
[31] Wolfgang Baumjohann thanks Donald Gurnett and another reviewer for their assistance in evaluating this paper.
References
Bigg, E. L. (1964), Influence of the satellite Io on Jupiter’s decametric
radiation, Nature, 203, 1008 – 1009.
Burger, M. H., E. C. Sittler Jr., R. E. Johnson, H. T. Smith, O. J. Tucker, and
V. I. Shematovich (2007), Understanding the escape of water from Enceladus, J. Geophys. Res., 112(A6), A06219, doi:10.1029/2006JA012086.
Clarke, J. T., L. Ben Jaffel, and J.-C. Gérard (1998), Hubble Space Telescope imaging of Jupiter’s UV aurora during the Galileo orbiter mission,
J. Geophys. Res., 103, 20,217 – 20,236.
Clarke, J. T., et al. (2002), Ultraviolet emissions from the magnetic footprints of Io, Ganymede and Europa on Jupiter, Nature, 415, 997 – 1000.
A07217
Clarke, J. T., D. Grodent, S. W. H. Cowley, E. J. Bunce, P. Zarka, J. E. P.
Connerney, and T. Satoh (2004), Jupiter’s aurora, in Jupiter. The Planet,
Satellites and Magnetosphere, edited by F. Bagenal, T. E. Dowling, and
W. B. Mckinnon, pp. 639 – 670, Cambridge Univ. Press, New York.
Clarke, J. T., et al. (2005a), Morphological differences between Saturn’s
ultraviolet aurorae and those of Earth and Jupiter, Nature, 433, 717 – 719.
Clarke, J. T., et al. (2005b), HST UV Imaging of Saturn’s Southern Aurora
during Simultaneous Cassini Imaging of the Northern Aurora, Eos Trans.
AGU, Fall Meet. Suppl., Abstract SM12A-06.
Connerney, J. E. P., N. F. Ness, and M. H. Acuna (1982), Zonal harmonic
model of Saturn’s magnetic field from Voyager 1 and 2 observations,
Nature, 298, 44 – 46.
Connerney, J. E. P., M. H. Acuna, and N. F. Ness (1983), Currents in
Saturn’s magnetosphere, J. Geophys. Res., 88, 8779 – 8789.
Cowley, S. W. H., and E. J. Bunce (2001), Origin of the main auroral oval
in Jupiter’s coupled magnetosphere – ionosphere system, Planet. Space
Sci., 49, 1067 – 1088.
Dougherty, M. K., et al. (2006), Identification of a dynamic atmosphere at
Enceladus with the Cassini magnetometer, Science, 311, 1406 – 1409.
Gérard, J. C., A. Saglam, D. Grodent, and J. T. Clarke (2006a), Morphology
of the ultraviolet Io footprint emission and its control by Io’s location,
J. Geophys. Res., 111(A4), A04202, doi:10.1029/2005JA011327.
Gérard, J. C., et al. (2006b), Saturn’s auroral morphology and activity
during quiet magnetospheric conditions, J. Geophys. Res., 111(A12),
A12210, doi:10.1029/2006JA011965.
Goertz, C. K. (1980), Io’s interaction with the plasma torus, J. Geophys.
Res., 85, 2949 – 2956.
Goldreich, P., and D. Lynden-Bell (1969), Io, a Jovian unipolar inductor,
Astrophys. J., 156, 59 – 78.
Grodent, D., J. T. Clarke, J. H. Waite, S. W. H. Cowley, J.-C. Gérard, and
J. Kim (2003), Jupiter’s polar auroral emissions, J. Geophys. Res.,
108(A10), 1366, doi:10.1029/2003JA010017.
Grodent, D., J.-C. Gérard, S. W. H. Cowley, E. J. Bunce, and J. T. Clarke
(2005), Variable morphology of Saturn’s southern ultraviolet aurora,
J. Geophys. Res., 110(A7), A07215, doi:10.1029/2004JA010983.
Gurnett, D. A., et al. (2002), Control of Jupiter’s radio emission and aurorae
by the solar wind, Nature, 415, 985 – 987.
Hall, D. T., P. D. Feldman, J. B. Holberg, and M. A. McGrath (1996),
Fluorescent hydroxyl emissions from Saturn’s ring atmosphere, Science,
272, 516 – 518.
Hansen, C. J., L. Esposito, A. I. F. Stewart, J. Colwell, A. Hendrix, W. Pryor,
D. Shemansky, and R. West (2006), Enceladus’ water vapor plume,
Science, 311, 1422 – 1425.
Hill, T. W. (2001), The Jovian auroral oval, J. Geophys. Res., 106, 8101 –
8107.
Hill, T. W. (2004), Auroral structures at Jupiter and Earth, Adv. Space Res.,
33, 2021 – 2031.
Hill, T. W., and D. H. Pontius (1998), Plasma injection near Io, J. Geophys.
Res., 103, 19,879 – 19,886.
Hill, T. W., A. J. Dessler, and C. K. Goertz (1983), Magnetospheric models,
in Physics of the Jovian Magnetosphere, edited by A. J. Dessler, pp. 353 –
394, Cambridge Univ. Press, New York.
Jurac, S., and J. D. Richardson (2005), A self-consistent model of plasma
and neutrals at Saturn: Neutral cloud morphology, J. Geophys. Res.,
110(A9), A09220, doi:10.1029/2004JA010635.
Khurana, K. K., et al. (2007), Mass loading of Saturn’s magnetosphere near
Enceladus, J. Geophys. Res., 112(A8), A08203, doi:10.1029/
2006JA012110.
Kivelson, M. G., F. Bagenal, W. S. Kurth, F. M. Neubauer, C. Paranicas,
and J. Saur (2004), Magnetospheric interactions with satellites, in Jupiter.
The Planet, Satellites and Magnetosphere, edited by F. Bagenal, T. E.
Dowling, and W. B. Mckinnon, pp. 513 – 536, Cambridge Univ. Press,
New York.
Knight, S. (1973), Parallel electric fields, Planet. Space Sci., 21, 741 – 750.
Milan, S. E., M. Lester, S. W. H. Cowley, K. Oksavik, M. Brittnacher, R. A.
Greenwald, G. Sofko, and J.-P. Villain (2003), Variations in the polar cap
area during two substorm cycles, Ann. Geophys., 21, 1121 – 1140.
Neubauer, F. M. (1980), Nonlinear standing Alfvén wave current system at
Io: Theory, J. Geophys. Res., 85, 1171 – 1178.
Nichols, J. D., E. J. Bunce, J. T. Clarke, S. W. H. Cowley, J.-C. Gérard,
D. Grodent, and W. R. Pryor (2007), Response of Jupiter’s UV auroras
to interplanetary conditions as observed by the Hubble Space Telescope
during the Cassini flyby campaign, J. Geophys. Res., 112(A2), A02203,
doi:10.1029/2006JA012005.
Pontius, D. H., and T. W. Hill (2006), Enceladus: A significant plasma
source for Saturn’s magnetosphere, J. Geophys. Res., 111(A9), A09214,
doi:10.1029/2006JA011674.
Pontius, D. H., G. Richards, and C. Dooley (2006), Modeling magnetic
field perturbations near Enceladus, Eos Trans. AGU, Fall Meet. Suppl.,
Abstract SM31C-01.
7 of 8
A07217
WANNAWICHIAN ET AL.: ENCELADUS-SATURN INTERACTION
Porco, C. C., et al. (2006), Cassini observes the active south pole of
Enceladus, Science, 311, 1393 – 1401.
Shemansky, D. E., and D. T. Hall (1992), The distribution of atomic hydrogen in the magnetosphere of Saturn, J. Geophys. Res., 97, 4143 – 4161.
Tokar, R. L., et al. (2006), The interaction of the atmosphere of Enceladus
with Saturn’s plasma, Science, 311, 1409 – 1412.
Tran, H. D., et al. (2002), On-orbit performance of the ACS Solar Blind
Channel, Proc. SPIE Int Soc. Opt. Eng., 4854, 686 – 693.
Vincent, M. B., et al. (2000), Jupiter’s polar regions in the ultraviolet as
imaged by HST/WFPC2: Auroral-aligned features and zonal motions,
Icarus, 143, 205 – 222.
A07217
Whalen, B. A., et al. (1985), Waterhole auroral arc modification experiments Electrodynamics response, J. Geophys. Res., 90, 8377 – 8386.
J. T. Clarke and S. Wannawichian, Center for Space Physics, Boston
University, 725 Commonwealth Avenue, Boston, MA 02115, USA.
([email protected])
D. H. Pontius Jr., Physics Department, Birmingham-Southern College,
900 Arkadelphia Road, Birmingham, AL, USA.
8 of 8