News & Views | Published:

Solar physics

The Sun under a microscope

Fine details of the filamentary structure of sunspots are revealed in new observations. These high-resolution measurements herald the quality of data to be expected from a new generation of solar telescopes.

Many scientific questions about the Sun can only be answered by observations made with high spatial resolution1,2. At the solar surface, turbulent convection and magnetic fields interact in a complicated manner, generating fine-scale structures. This is especially evident in a sunspot, with its dark central core, or umbra, surrounded by a region of radial filaments, known as the penumbra. High-resolution observations have revealed much of the intricate thermal and magnetic structure of the penumbra, but have also hinted at structure on scales below the resolution limit of existing telescopes3.

On page 151 of this issue, Scharmer et al.4 report new observations of the penumbral filaments with unprecedented spatial resolution (Fig. 1). They have detected dark central cores, less than 90 km wide, within many of the filaments (which are typically 150–180 km wide). This discovery is the first scientific result from the new Swedish 1-m Solar Telescope, which became operational in May 2002.

Figure 1: Darkness in the light.

Around the dark core of a sunspot, a filamentary structure develops, driven by intense magnetic fields and gas flow. This image from Scharmer et al.4 shows in unprecedented detail the structure of the filaments themselves, in particular their dark central cores, for which there is as yet no explanation.

It is appropriate that this first result is for sunspots, which have long been a test bed for studying magnetohydrodynamics — how conducting fluids, such as plasmas or liquid metals, behave in magnetic fields — under astrophysical conditions. The filamentary penumbra has proved especially difficult to understand, with its complicated magnetic-field geometry and associated gas flows and oscillations5,6. The filament cores discovered by Scharmer et al.4 are likely to be an important key to understanding the penumbra. But their results are based on narrow-band imaging, which reveals only the intensity (thermal) pattern of the dark cores. As the authors point out, a full understanding of these features will require measurements of the magnetic field and flow velocity in the cores using the more difficult techniques of spectroscopy and polarimetry.

An essential feature in these new observations is the use of an adaptive optics (AO) system, operating in real time to correct for the distorting effects of the Earth's atmosphere on sunlight. In an AO system, a reference image is used to detect the distortions, and then the sections of a multi-segment mirror are moved independently to correct for them. The system used by Scharmer et al. at the Swedish 1-m Solar Telescope has 19 mirror elements. In contrast to night-time observations, where a steady, point-like source (a single star) is available as a reference, solar AO systems have solved the more difficult problem of how to use an extended, low-contrast, time-varying image such as a solar granulation pattern as a reference. An AO system is essential for any large ground-based solar telescope to achieve consistent high resolution near the limit of its power (its diffraction limit).

Adaptive optics systems have breathed new life into some older solar telescopes. A 24-element AO system has been in use for three years at the Dunn Solar Telescope at the US National Solar Observatory (NSO) in Sunspot, New Mexico. Other AO systems are also in use at the German Vacuum Tower Telescope on Tenerife in the Canary Islands and at the largest existing solar telescope, the NSO's McMath–Pierce Telescope on Kitt Peak in Arizona, where sensitive measurements of solar magnetic fields are being made at near-infrared wavelengths. An 80-element AO system is under development for use both at the Dunn telescope and at the Big Bear Solar Observatory in California.

High-resolution imaging in solar astronomy relies not only on angular resolution but also on spectral (wavelength) and temporal resolution. Good angular resolution typically reveals only the horizontal surface structure of the Sun. To sample different heights in the solar atmosphere, measurements at different wavelength positions within a spectral absorption line are needed. In general, the smaller the solar feature, the more rapidly it changes with time. So, as angular resolution is increased, temporal resolution must also be increased to follow these changes. To achieve high angular, temporal and spectral resolution simultaneously requires a high throughput of light, and this may well dictate a larger telescope aperture than would angular resolution alone.

The need for large aperture is a driving factor in the design of new solar telescopes. A German consortium is reconfiguring a telescope on Tenerife into the GREGOR telescope with a 1.5-m aperture, with completion expected in 2005. The most ambitious project is the Advanced Technology Solar Telescope (ATST) at the NSO. The ATST will have a 4-m aperture, which will provide high angular resolution (less than 0.1 arcsecond) at wavelengths where very sensitive magnetic measurements can be made. The ATST will also observe out to relatively long wavelengths in the thermal infrared region, where magnetic fields in the solar corona can be measured. Design studies and site testing for the ATST are now being carried out, and first light is scheduled for 2010. Of course, the ideal high-resolution instrument would be a large solar telescope in space — but the largest project at present is a telescope with a relatively small aperture of 50 cm, scheduled for launch on the Japan–US–UK Solar-B satellite in 2005.

Is there a limit to how small a structure we might expect to observe on the Sun? Theoretical arguments suggest that there are magnetic structures on scales as small as a kilometre, or less, in the Sun's photosphere (its visible surface). Computer simulations of photospheric magnetoconvection show very small structures, but the simulations have not yet achieved sufficient resolution to determine the limiting size. The horizontal mean free path — in other words, the average distance travelled without interacting — of a photon in the solar photosphere is about 50 km, and so this might be expected to be the smallest observable length scale, because of the smoothing effect of radiative energy transfer. But sophisticated radiative-transfer calculations show that fine structures as small as a few kilometres should in principle be directly observable7. It seems, then, that the quest for higher resolution in solar astronomy — and with it, the increase in understanding of our own star — is likely to continue well beyond the telescope projects now under way.


  1. 1

    Rimmele, T. R., Balasubramaniam, K. S. & Radick, R. R. (eds) ASP Conf. Series 183, High Resolution Solar Physics: Theory, Observations, and Techniques (Astron. Soc. Pacific, San Francisco, 1999).

  2. 2

    Thomas, J. H. Nature 396, 114–115 (1998).

  3. 3

    Sánchez Almeida, J. & Lites, B. W. Astrophys. J. 532, 1215–1229 (2000).

  4. 4

    Scharmer, G. B., Gudiksen, B. V., Kiselman, D., Löfdahl, M. G. & Rouppe van der Voort, L. H. M. Nature 420, 151–153 (2002).

  5. 5

    Thomas, J. H. & Weiss, N.O. (eds) Sunspots: Theory and Observations (Kluwer, Dordrecht, 1992).

  6. 6

    Schmieder, B., del Toro Iniesta, J. C. & Vázquez, M. (eds) ASP Conf. Series 118, Advances in the Physics of Sunspots (Astron. Soc. Pacific, San Francisco, 1997).

  7. 7

    Bruls, J. H. M. J. & von der Lühe, O. Astron. Astrophys. 366, 281–293 (2001).

Download references

Author information

Correspondence to John H. Thomas.

Rights and permissions

Reprints and Permissions

About this article

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.