letters - UCLA IGPP

Vol 453 | 19 June 2008 | doi:10.1038/nature07077

LETTERS Jovian-like aurorae on Saturn Tom Stallard1, Steve Miller2, Henrik Melin3, Makenzie Lystrup2, Stan W. H. Cowley1, Emma J. Bunce1, Nicholas Achilleos2 & Michele Dougherty4

Planetary aurorae are formed by energetic charged particles streaming along the planet’s magnetic field lines into the upper atmosphere from the surrounding space environment. Earth’s main auroral oval is formed through interactions with the solar wind1, whereas that at Jupiter is formed through interactions with plasma from the moon Io inside its magnetic field (although other processes form aurorae at both planets2,3). At Saturn, only the main auroral oval has previously been observed and there remains much debate over its origin. Here we report the discovery of a secondary oval at Saturn that is 25 per cent as bright as the main oval, and we show this to be caused by interaction with the middle magnetosphere around the planet. This is a weak equivalent of Jupiter’s main oval, its relative dimness being due to the lack of as large a source of ions as Jupiter’s volcanic moon Io. This result suggests that differences seen in the auroral emissions from Saturn and Jupiter are due to scaling differences in the conditions at each of these two planets, whereas the underlying formation processes are the same. There are three competing theories describing the process by which Saturn’s main auroral oval is formed: first, that the oval is like that of the Earth, mapping to the boundary between closed field lines and those open to the solar wind, where a shear in rotational flow is expected, requiring a ring of upward-directed current4; second, that the oval is associated with centrifugal instabilities in the outer magnetosphere, where variations in solar wind dynamic pressure lead to varying angular velocities, driving currents into the ionosphere5; and third, that the oval is an analogue of that at Jupiter, formed by internal magnetospheric processes driven by the rapid rotation of the planet2,3. Jupiter’s equatorial plasma sheet initially co-rotates with the planet, but as the plasma diffuses away this co-rotation breaks down, resulting in a strong circuit of electric currents. This forces electrons to precipitate along the magnetic field lines into the atmosphere, forming an auroral oval at the jovian latitude mapping to the co-rotation breakdown. The morphology of Saturn’s aurorae has been examined in detail with the use of Hubble Space Telescope images of the ultraviolet emission from the planet6–8, showing that the aurorae are strongly influenced by changes in the dynamic pressure of the solar wind9. Ground-based infrared spectroscopic studies have used emission from the molecular ion H31 to observe Saturn’s auroral structure10 and to measure the associated temperature11 and ion winds12 within the upper atmosphere. Recent measurements of the ion wind have shown that the main auroral oval is located significantly poleward of the latitude at which ions no longer co-rotate with the planet, which means that, by definition, Saturn’s main auroral oval cannot be Jupiter-like in formation13. However, an excess of emission equatorward of the main auroral oval has been observed in a significant proportion of the infrared observations; this aurora could have an analogous origin to that of the main auroral oval on Jupiter.

To test this, it is necessary to isolate this secondary emission from that of the main auroral oval. Long-slit spectrometer measurements cutting perpendicularly through Saturn’s auroral region result in contemporaneous profiles of the infrared intensity and corresponding line-of-sight ion velocity (Figs 1 and 2)14. Whereas the ultraviolet morphology is dominated by the emission from the main auroral oval, the infrared emission has significant amounts of emission that are not associated with this oval. Assuming that the infrared and ultraviolet main auroral ovals have the same morphological structure13, a model of the main oval can be based on a statistical analysis of the typical ultraviolet oval15. By subtracting this modelled main auroral oval from the infrared intensity profile, the residual emission can be calculated (Fig. 3). This results in three clear regions of secondary auroral emission, one poleward and two equatorward from the location of the main oval. The relative excess of infrared emission in Saturn’s polar cap has previously been noted14. This is likely to be caused by significant emission across the polar region, analogous with that of Jupiter, where emission occurs across the entire polar region in both the infrared and ultraviolet16,17. Corresponding ultraviolet emission across Saturn’s pole, if it exists, is too weak to have been detected. Of the two intensity peaks equatorward of the main auroral oval, the peak on the dawn side has an intensity ,25% of the main auroral oval and the peak on the dusk side has an intensity ,18% of the main auroral oval. Both these intensity peaks have their maximum values positioned at exactly the same location as the breakdown in co-rotation, measured in the corresponding velocity profile, with the dusk emission centred on the breakdown in co-rotation and the dawn emission somewhat extended, decreasing gently towards the pole. The effect of changing the modelled main oval position on the secondary oval brightness is discussed in more detail in the Supplementary Information. This correspondence between the position of the peak secondary emission and the breakdown in co-rotation in the ionosphere is direct evidence that the breakdown in co-rotation within the magnetospheric plasma is driving a current system strong enough to produce an H31 aurora, a weaker variant of the main auroral oval seen on Jupiter. This newly identified aurora forms in two distinct regions within the slit, one on each of the dawn and dusk sides, at slightly lower latitudes than the main auroral oval. Given that this flank emission is seen repeatedly in successive runs13,14 and that there is significant emission equatorward of the main oval at noon10, this constitutes strong evidence of the presence of a second auroral oval. Auroral electron beams have been detected fairly deep inside the magnetosphere, supporting an internally driven auroral component18. These are the first auroral emissions from Saturn that can be directly related to the main auroral oval at Jupiter. So far there has been no published identification of such emission within the ultraviolet data set, the closest comparison being a limb-brightened

1 Department of Physics and Astronomy, University of Leicester, Leicester LE1 7RH, UK. 2Atmospheric Physics Laboratory, Department of Physics and Astronomy, University College London, Gower Street, London WC1E 6BT, UK. 3Space Environment Technologies, Planetary and Space Science Division, 320 N. Halstead Street, Suite 110, Pasadena, California 91107, USA. 4Space and Atmospheric Physics Group, Department of Physics, Imperial College of Science, Technology and Medicine, London SW7 2BW, UK.

1083 ©2008 Macmillan Publishers Limited. All rights reserved

LETTERS

NATURE | Vol 453 | 19 June 2008

‘auroral zone O’ (‘O’ for ‘outer’)19. The potential for such oval emission was previously predicted, for ultraviolet emission, to form a dim narrow oval too weak (,1 kR and ,1u wide) to be detected with current observation techniques20. Saturn’s main auroral oval is typically 20–50 kR, so a secondary oval at 20% brightness is at about the limit of detectability with Hubble. This suggests that the strength of the current formed by breakdown of co-rotation at Saturn is stronger than has been previously modelled and that future observations, either in the infrared or in the ultraviolet, should be better able to determine the exact location of the secondary oval, allowing these models to be improved. With this discovery, our understanding is that Saturn’s aurorae are more akin to those of Jupiter. Observations of Jupiter have previously shown that a significant region of solar wind control exists poleward of the main auroral oval21, a region where there is significant emission in both the infrared and ultraviolet17,22. It therefore seems likely that Jupiter’s polar aurora is at least partly ‘saturnian-like’ in origin; if this is so, a secondary oval will be located within the polar region. Thus, although the general morphology of the aurorae of the planets is significantly different, the processes by which they are formed seem to be similar. Specific morphological differences, such as the

strong dawn brightening seen at Saturn during solar wind compressions8, still distinguish the aurorae from those at Jupiter and have much to tell us about the interaction of the planets with the solar

6

Saturn’s centre

12

8 0 6

–2 4

–4

2

–6

0

Saturn’s centre

10 0.8

Normalized intensity

Velocity (km s–1)

2

1.0

Intensity (×10–17) (W m–2 µm–1)

4

Figure 2 | The location of the spectrometer slit on Saturn’s southern aurora. The slit position (bold) was estimated by comparing the auroral data with a modelled main auroral oval. The longitudinal grid is demarcated in steps of 30u and the latitudinal grid in steps of 15u, with the body of the planet (in grey) shown for clarity. The main auroral oval model was based on an axisymmetric model oval that approximates the observed time-averaged ultraviolet auroral oval15: ,1u wide, located at 75u latitude. This was convolved to appear as though measured from the ground, broadened by an effective seeing of 1.8 arcsec. The intensity structure, consisting solely of the modelled main auroral oval, thus excludes emission from the breakdown region or diffuse aurora in the modelled intensity.

0.6

0.4

0.2 –6

–4

–2

0 2 Offset (arcsec)

4

6

Figure 1 | H31 aurora emission and the associated line-of-sight velocity. The intensity profile (thin) is plotted with the velocity in the inertial frame (bold) and, for reference, the rotation of the planet (dashed). Two vertical dot–dashed lines mark the location where the ionosphere begins to subrotate from the neutral atmosphere. Two vertical three-dot–dashed lines mark the location of the main auroral oval within the slit. The edge of the plot delineates the limb of the planet. This data was taken on 2003 February 6 as a part of an extensive set of observations of the auroral/polar regions of Saturn, with the NASA Infrared Telescope Facility. The long-slit spectrometer CSHELL23 was aligned west–east on the planet, with the field of view of the slit crossing the auroral region. The resultant high-resolution spectrum was centred on emission from the H31 n2 Q(1,02) line at 3.953 mm. This line was fitted with a gaussian at each spatial position, and the gaussian peak brightness and position were used to calculate the intensity and relative velocity of this line by using a method originally applied to Jupiter22, and adapted since to Saturn12,14. The intensity structure shows three peaks, the outermost two marking the point where the slit cuts through the main auroral oval. The extended flanks of the profile also bulge outwards with additional emission. The velocity structure shows a ‘three-tiered’ velocity structure14, with a core region that co-rotates with the planet, flanked by two regions that significantly sub-rotate, which is typical for profiles taken in periods of rarefied solar wind conditions. The errors associated with fitting the data are shown by grey regions surrounding the intensity and velocities, with darker grey where these regions cross. This is the calculated standard deviation of the fitting procedure at each position on the profile.

0

–6

–4

–2

0 2 Offset (arcsec)

4

6

Figure 3 | ‘Jovian-like’ aurorae on Saturn. The residual intensity on the flanks of the observed oval (bold) and in the polar cap (bold and grey), are found by subtracting the line-of-sight corrected intensity structure, shown in Fig. 1 (thin), from the modelled main auroral oval intensity, shown in Fig. 2 (dashed). The dot–dashed and three-dot–dashed vertical lines are the same as those in Fig. 1. The main auroral oval is modelled as an intensity profile, produced by combining the light entering a slit positioned over the seeing-distorted main auroral oval model. Because the H31 aurora is actually formed within an atmospheric ‘shell’, where the observed intensities are integrated along a column of atmosphere whose depth varies with position on the planet, we have applied a line-of-sight correction to the spectral data. The residual intensities are calculated by subtracting the modelled main auroral oval from this corrected intensity profile. The errors in the secondary oval intensity are given as a grey region and are the combined standard deviation in fitting the spectra and in positioning the modelled intensity profile. The ultraviolet auroral intensity is relatively stable across the short integration time over which Hubble images are typically taken, as is the infrared aurora when integrated over a period of several hours; however, these do vary over longer timescales. Changing the location of modelled oval equatorward reduces the peak intensity of the secondary oval but never eliminates the secondary aurora completely, and the peak intensity is always located at the breakdown in co-rotation.


LETTERS

NATURE | Vol 453 | 19 June 2008

wind. However, it is no longer reasonable to consider Saturn’s aurorae a ‘hybrid’ of those of the Earth and Jupiter, but rather that the aurorae of Jupiter and Saturn are variants of the same formation processes. Received 4 December 2007; accepted 29 April 2008. 1. 2. 3. 4.

5.

6.

7. 8. 9. 10. 11.

12. 13. 14. 15.

16.

Dungey, J. W. Interplanetary magnetic field and the auroral zones. Phys. Rev. Lett. 6, 47–48 (1961). Hill, T. W. The Jovian auroral oval. J. Geophys. Res. 106, 8101–8108 (2001). Cowley, S. W. H. & Bunce, E. J. Origin of the main auroral oval in Jupiter’s coupled magnetosphere–ionosphere system. Planet. Space Sci. 49, 1067–1088 (2001). Cowley, S. W. H., Bunce, E. J. & O’Rourke, J. M. A simple quantitative model of plasma flows and currents in Saturn’s polar ionosphere. J. Geophys. Res. 109, A05212 (2004). Sittler, E. C., Blanc, M. F. & Richardson, J. D. Proposed model for Saturn’s auroral response to the solar wind: Centrifugal instability model. J. Geophys. Res. 111, A06208 (2006). Trauger, J. T. et al. Saturn’s hydrogen aurora: Wide field and planetary camera 2 imaging from the Hubble Space Telescope. J. Geophys. Res. 103, 20237–20244 (1998). Ge´rard, J.-C. et al. Characteristics of Saturn’s FUV aurora observed with the Space Telescope Imaging Spectrograph. J. Geophys. Res. 109, A09207 (2004). Clarke, J. T. et al. Morphological differences between Saturn’s ultraviolet aurorae and those of Earth and Jupiter. Nature 433, 717–719 (2005). Crary, F. J. et al. Solar wind dynamic pressure and electric field as the main factors controlling Saturn’s aurorae. Nature 433, 720–722 (2005). Stallard, T. et al. The H31 latitudinal profile of Saturn. Astrophys. J. 521, L149–L152 (1999). Melin, H., Miller, S., Stallard, T., Trafton, L. M. & Geballe, T. R. Variability in the H31 emission of Saturn: Consequences for ionisation rates and temperature. Icarus 186, 234–241 (2007). Stallard, T., Miller, S., Trafton, L. M., Geballe, T. R. & Joseph, R. D. Ion winds in Saturn’s southern auroral/polar region. Icarus 167, 204–211 (2004). Stallard, T. et al. Saturn’s auroral/polar H31 infrared emission II: A comparison with plasma flow models. Icarus 191, 678–690 (2007). Stallard, T. et al. Saturn’s auroral/polar H31 infrared emission I: General morphology and ion velocity structure. Icarus 189, 1–13 (2007). Badman, S. V., Cowley, S. W. H., Ge´rard, J.-C. & Grodent, D. A statistical analysis of the location and width of Saturn’s southern auroras. Ann. Geophys. 24, 3533–3545 (2006). Stallard, T., Miller, S., Millward, G. & Joseph, R. D. On the dynamics of the Jovian ionosphere and thermosphere II: The measurement of H31 vibrational temperature, column density, and total emission. Icarus 156, 498–514 (2002).

17. Grodent, D. et al. Jupiter’s polar auroral emissions. J. Geophys. Res. 108, 1366–1374 (2003). 18. Saur, J. et al. Anti-planetward auroral electron beams at Saturn. Nature 439, 699–702 (2006). 19. Grodent, D., Ge´rard, J.-C., Cowley, S. W. H., Bunce, E. J. & Clarke, J. T. Variable morphology of Saturn’s southern ultraviolet aurora. J. Geophys. Res. 110, A07215 (2005). 20. Cowley, S. W. H. & Bunce, E. J. Corotation-driven magnetosphere–ionosphere coupling currents in Saturn’s magnetosphere and their relation to the auroras. Ann. Geophys. 21, 1691–1707 (2003). 21. Stallard, T., Miller, S., Cowley, S. W. H. & Bunce, E. J. Jupiter’s polar ionospheric flows: Measured intensity and velocity variations poleward of the main auroral oval. Geophys. Res. Lett. 30, 1221–1224 (2003). 22. Stallard, T., Miller, S., Millward, G. & Joseph, R. D. On the dynamics of the Jovian ionosphere and thermosphere I: The measurement of ion winds. Icarus 154, 475–491 (2001). 23. Greene, T. P., Tokunaga, A. T., Toomey, D. W. & Carr, J. S. CSHELL: A high spectral resolution 1–5 micron cryogenic echelle spectrograph for the IRTF. Proc. SPIE 1946, 313–324 (1993).

Supplementary Information is linked to the online version of the paper at www.nature.com/nature. Acknowledgements We thank the NASA Infrared Telescope Facility (IRTF) telescope operators for their continued support and expert advice in making these observations possible. The authors are part of the Europlanet European planetary science network, supported by the European Union’s Framework 6 programme. This work was supported by the UK Science and Technology Facilities Council, with postdoctoral fellowships for T.S., N.A. and E.J.B., and a senior fellowship for M.D. T.S. is now funded by an RCUK Fellowship. H.M. was supported by a postgraduate studentship from the UK Engineering and Physical Sciences Research Council. S.W.H.C. was supported by a Royal Society Leverhulme Trust Senior Research Fellowship. T.S., H.M. and M.L. are visiting astronomers at the IRTF, which is operated by the University of Hawaii under Cooperative Agreement no. NCC 5-538 with the NASA Science Mission Directorate, Planetary Astronomy Program. Author Contributions T.S. designed the study, collected and analysed data and wrote the paper. S.M. collected and aided data analysis. H.M. and M.L. aided data analysis. S.W.H.C., E.J.B, N.A. and M.D. provided the magnetospheric context. All authors discussed the results and commented on the manuscript. Author Information Reprints and permissions information is available at www.nature.com/reprints. Correspondence and requests for materials should be addressed to T.S. ([email protected]).
