News & Views | Published:

Planetary Science

A new spin on Saturn

Naturevolume 441pages3435 (2006) | Download Citation


Measuring the rotation of a gaseous planet is no easy task. For Saturn, do observations of its magnetic field — which indicate that it is spinning more slowly than thought — mark a revolution in our understanding?

According to the International Astronomical Union (IAU), Saturn rotates once every 10 hours, 39 minutes and 22.4 seconds. Astronomical observations as far back as William Herschel's in the late eighteenth century1 had suggested values for Saturn's rotational period of ten or so hours, but the IAU's seemingly precise value was defined by a periodicity in the kilometre-wavelength radio signal sent out by Saturn and detected by NASA's Voyager spacecraft in 1980. Surprising though it may seem, these radio emissions are still not well understood and — unlike radio emissions from pulsars or from Jupiter — turn out to be imprecisely periodic.

In this issue, Giampieri and colleagues (page 62)2 report the observation, in data from the Cassini mission currently investigating the Saturn system, of a signal in Saturn's magnetic field of period about 10 hours 47 minutes. The authors suggest that this might be the planet's true rotation period.

So why should we care? One obvious reason is that rotation period is a fundamental property of a planet and tells us something about the conditions — such as the total angular momentum — when it formed. But that does not require a highly precise value. A less evident reason is that, with an equatorial radius some 10% greater than its polar radius, Saturn is the most distorted planet in the Solar System. Both the rotation and internal structure of a celestial object contribute to such an ‘equatorial bulge’, and the discrepancy of about 8 minutes (more than 1%) between the accepted and the new value for Saturn's rotational period will thus also affect our estimates of the size of its likely inner core of rock and ice. Although the ‘gas giant’ Saturn is made mostly of hydrogen and helium, understanding the origin and evolution of such planets depends crucially on the nature and internal distribution of their minor constituents3. Saturn's rotation rate also provides the essential reference frame within which to talk about the dynamics of its atmosphere, and in particular the velocity of its very strong east–west winds. And this last point highlights a central issue: what does the rotation rate of a fluid planet actually mean?

The rotation rate of a planet such as Earth is conventionally defined to be that of its solid mantle, and can be measured to exquisite precision using geodetic methods such as the Global Positioning System. Tectonic movements of Earth's solid parts are, on average, slower than its rotation speed by 12 orders of magnitude, but are readily measurable. Larger effects arise from the interplay of tides and Earth's rotation, which causes an inexorable extraction of spin angular momentum, and the Moon to recede from Earth. Even larger, but fluctuating, effects arise from the exchange of angular momentum between Earth's solid regions and its fluid parts — the atmosphere, oceans and core. But the largest of these is still only a millionth of a hypothetical 1% saturnian change over the past 20 years.

Such a large variation in Saturn's rotation might be explained by its atmospheric features, in particular its strong easterly winds of up to 400 m s−1 relative to the IAU period. But a 1% difference in rotation rate of an atmospheric feature corresponds to a ‘wind’ of around 100 m s−1, and, although Saturn's winds may have changed, it is unlikely to have been by that much.

A more reliable measure of a fluid planet's spin than cloud patterns on its surface comes from its magnetic field. If this field is large and emanates principally from a dipole whose axis is tilted with respect to the axis of rotation, there will be a distinctive, readily detectable periodicity in the field and any resulting radio emissions. Because this magnetic field is generated deep down, its lines are embedded deeply in a large fraction of the planetary mass where large variable rotational motions should not occur. An east–west motion of just 1 m s−1 would, for instance, generate a toroidal magnetic field orders of magnitude larger than the observed dipole field, causing a Lorentz force to act and enforce uniform rotation again.

This argument agrees with our understanding of Earth. In Earth's liquid core, movements at depth are undetectable, but the movements at the top, in part revealed by magnetic-field changes over decades, indicate only tiny differential motions. Even the seismologically detected superrotation of Earth's inner core4 corresponds to only about one part in 1010 of Earth's total rotation.

Giampieri and colleagues2 have found a stable periodicity in Saturn's magnetic field. Their observations are, however, incompatible with the expected inverse cubic dependence of field strength on distance from the planet that is expected from a tilted dipole. Saturn's magnetic field (Fig. 1) certainly lacks the substantial dipole tilt that we see so strikingly in Earth, Jupiter and, in a more complicated way, in Uranus and Neptune; but it would be astonishing if Saturn's field was exactly symmetric about the rotation axis. There is no reasonable model for field structure that can explain the observed behaviour solely through the internal field.

Figure 1: Straight-up rotator.
Figure 1

Saturn's magnetic field is almost exactly aligned with its rotation axis, and the dipole field lines emanate from a massive electrically conducting metallic hydrogen core (both axes are tilted by about 27° with respect to the plane of its orbit). Giampieri and colleagues' observation2 of a periodic field variation, although not fully understood, could indicate that Saturn is rotating more slowly than had been assumed.

The observed periodicity might instead result from something that is a proxy for the internal rotation, for example a disturbance generated in Saturn's magnetosphere by a field anomaly (not necessarily dipolar) that is rotating with the core of the planet. The stability of the period would certainly be easiest to explain with the rotation of such a massive object. But although Saturn's deep interior would undoubtedly be a good clock, some region of smaller inertia would not necessarily be a bad one: the observed periodicity might be an indication of some wave-like or advected feature of the planet's ionosphere, and so be quite possibly unrelated to a rotation of the whole planet. In this case, it is curious that all the observed atmospheric motions rotate faster than the proposed rotation.

But the central puzzle would seem to be why Saturn behaves differently from Jupiter. Jupiter has a tilted dipole, a magnetic field that is much larger than Saturn's (even after taking account of the different-sized metallic region at its core), and winds that move both with and against the sense of the planet's rotation. Owing to the much greater tilt of its axis of rotation (27° against 3°), Saturn has appreciable seasons, whereas Jupiter does not; perhaps we need a longer time base to appreciate fully the differences arising from this. (The orbital periods of the two planets are 29.5 years and 12 years, respectively.) However, this can only explain superficial differences. Maybe the Jupiter–Saturn differences are just another example of planetary diversity: the remarkable richness of planets in the Solar System alone has arguably been the most striking outcome of the planetary exploration programme.


  1. 1

    Herschel, W. Phil. Trans. R. Soc. Lond. 84, 48–66 (1794).

  2. 2

    Giampieri, G., Dougherty, M. K., Smith, E. J. & Russell, C. T. Nature 441, 62–64 (2006).

  3. 3

    Guillot, T. Science 286, 72–77 (1999).

  4. 4

    Zhang, J. et al. Science 309, 1357–1360 (2005).

Download references

Author information


  1. the Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, 91125, California, USA

    • David J. Stevenson


  1. Search for David J. Stevenson in:

About this article

Publication history


Issue Date


Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing