Solar-eclipse science: still going strong.
A total eclipse of the Sun is arguably nature's most spectacular and awe-inspiring phenomenon. Occurring an average of once every 18 months across some narrow strip of the globe and lasting anywhere from a fraction of a second to a maximum of 7 1/2 minutes, a total solar eclipse offers us the best ground-based views of the solar-corona--the Sun's extremely hot outer atmosphere, a ghostly web of magnetically shaped plumes and streamers. Scientists are interested in the corona for four major reasons. First, the corona can tell us a lot about the Sun's workings. All the energy radiated by the Sun passes through the corona. Why its temperature is as high as several million degrees Celsius remains an open question; theories of energy transport via magnetic fields currently dominate.
Second, by studying the solar corona we can learn more about the near-Earth space environment and the disruptions that afflict it. The outer corona continually expands into interplanetary space as a stream of charged particles known as the solar wind, which fills the solar system. Violent eruptions on the Sun, such as solar flares and coronal mass ejections, can cause auroras and navigation and communications blackouts, as well as circuit overloads on Earth-orbiting satellites and power grids on the ground. Changes in solar radiation may also affect Earth's weather and climate patterns.
Third, our Sun is a rather average yellow dwarf star, so by studying it we gain detailed, close-up information that we can apply to stars elsewhere. For example, X-ray observatories, especially Chandra, are detecting coronas around many other Sun-like stars, so by studying the Sun's corona up close we are learning about all these other coronas.
Last, the Sun is a physics laboratory offering conditions far beyond anything we can create on Earth. For example, the corona's density is so low that it would be considered a fantastic vacuum in a laboratory. The corona is a hot plasma (an ionized, electrically conducting gas), which the Sun's magnetic field shapes into the beautiful, long equatorial streamers and delicate polar brushes seen during eclipses. The corona thus reveals complex behaviors of hot, ionized gas locked to magnetic fields--a knotty discipline known as magnetohydrodynamics (MHD). Many processes throughout the universe are governed by MHD, yet they are extremely hard to model on computers, let alone test in the laboratory. Here on Earth, such knowledge is sought by plasma physicists engaged in magnetic-fusion research, which may someday revolutionize power plants.
The solar corona is so important that every opportunity to study it should be fully exploited.
Why are eclipses still important for coronal studies? Modern solar telescopes in space, such as those aboard the Japanese Yohkoh X-ray satellite, the Solar and Heliospheric Observatory (SOHO), and the Transition Region and Coronal Explorer (TRACE), monitor the Sun continuously and in parts of the spectrum that are absorbed by the Earth's atmosphere, such as the extreme ultraviolet and X-rays. None of them need an eclipse to observe the corona; they run all the time.
For example, SOHO's Large Angle and Spectrometric Coronagraph (LASCO) consists of three separate cameras, each with an occulting disk to create artificial eclipses of various coverage in the cameras. C1 used to provide views of the corona from 1.1 to 3 solar radii, C2 gives an overlapping band from about 2.2 to 6 solar radii, and C3 extends from roughly 4.5 out to 30 solar radii. (C1 died when SOHO went haywire for a while in 1998.)
This is impressive coverage, but it still doesn't cover all that eclipse-watchers can observe from Earth. Like any terrestrial coronagraph, C1 used a circular opaque mask at the telescope's focal plane to eclipse (occult) the Sun's brilliant disk, the photosphere. LASCO's two remaining coronagraphs use an external occulting disk in front of the main lens. To limit scattered light within the instrument's field of view, LASCO also has to block out the bright innermost corona--at present, everything below 2.2 solar radii. This is a crucial region from which the solar wind arises. In the SOHO images on the facing page, you'll notice a circle at the center of the occulting disk; the circle represents the size of the Sun's photosphere and shows how much of the corona has been hidden. It is exactly this region of the corona that can be observed from Earth during a total eclipse.
Furthermore, the C1 coronagraph had a rather coarse resolution of 12 arcseconds, but during an eclipse we can resolve detail as fine as about 1 or 2 arcseconds. C2 and C3 have even lower resolutions--25 arcseconds and 2 arcminutes, respectively.
Ground-based coronagraphs on high mountaintops can see very close to the Sun's limb, but still not as close as during an eclipse. Thus, total-eclipse observations still provide essential supplements to both ground- and space-based views.
To provide worthwhile science, modern eclipse expeditions need to exploit this remaining niche. Worthy observations should make use of finer resolution in space or time or wavelength than spacecraft can provide, or they should look in wavelength or spatial regimes that are not covered from space.
One crucial advantage is that while planning a space mission takes many years or even decades, eclipse observations can be planned on short notice, perhaps using equipment not previously available or not "space-hardened."
For example, Spiros Patsourakos (Institut d'Astrophysique Spatiale, France) has used eclipse observations he made in 1998 along with simultaneous observations from SOHO's Solar Ultraviolet Measurements of Emitted Radiation (SUMER) experiment. Together they let him derive, for the first time, the radial velocity of the fast solar wind in a region between plumes at the base of a polar coronal hole. He and his colleagues found a very high velocity, which they interpret as indicating that these interplume regions are the source of the "fast" component of the solar wind. Only with the eclipse data could they reach very close to the limb to capture the data they needed.
One of my own eclipse-expedition team's experiments, supported by the National Science Foundation, images coronal loops through a filter passing light from 13-times-ionized iron gas (Fe XIV) at a rate of 10 times per second. Theoretical calculations show that fast oscillations could be capable of heating the corona to its high temperature. The idea seems worth checking, yet none of the spacecraft studying the corona can observe that rapidly, by a factor of 20 or more. Observations from the Ultraviolet Coronagraph Spectrometer (UVCS) experiment on SOHO are among those indicating that fast waves heat at least coronal holes and contribute to the acceleration of the solar wind.
Solar-B, a Japanese spacecraft being built with major American participation and due to be launched in 2005, will have finer time resolution than is now available from space. But it still will not reach ground-based speeds. And its spatial resolution will be limited to 1 arcsecond, not much better than we now get from the ground and a factor of 2 worse than TRACE's. It will, however, see a hotter component of the corona, 3 to 5 million degrees Celsius, which TRACE cannot observe.
The TRACE observations are fantastically detailed, resolving features as small as 400 kilometers wide on the Sun's surface, corresponding to 0.5 arcsecond. At the August 11, 1999, eclipse in Romania, we used a satellite telephone to coordinate with the TRACE mission-control team so that we could point at the same loops on the solar limb. We have higher time resolution on those loops; TRACE has higher spatial resolution.
Serge Koutchmy (Institut d'Astrophysique de Paris-CNRS) has long specialized in high-resolution images of the corona. Using highly detailed images obtained on Mauna Kea, Hawaii, at the July 11, 1991, eclipse and most recently from Chadagan, Iran, in August 1999, Koutchmy and his colleagues are tracing extremely small coronal elements and following them as they move through the corona. At the 1999 eclipse Koutchmy, together with Frederic Baudin and Karine Bocchialini (Institut d'Astrophysique Spatiale), observed a coronal mass ejection directly above an unusually high loop. They also analyzed the profile of the well-known "green line" emission from 13-times-ionized iron in the coronal spectrum across the entire corona, convincingly showing that the spectral line's width is incompatible with increasing turbulent and/or wave activity above about a half solar radius from the Sun's limb.
Jeffrey Kuhn (now with the University of Hawaii), Robert MacQueen (Rhodes College,Memphis), Alan Ridgeley (Rutherford Appleton Laboratory, England), and their respective colleagues are among those using eclipses to make observations of the solar atmosphere in the infrared. Some spectral lines have been found or suspected at infrared wavelengths of several microns. Further eclipse work is necessary to follow up these observations and discover the conditions in the corona that cause them.
From the 1994 eclipse data, Kuhn, Matthew Penn (U.S. National Solar Observatory), and Ingrid Mann (Max Planck Institute for Aeronomy) reported coronal emission lines at 1.25 and 1.43 microns due to silicon with 8 and 9 of its 14 electrons missing (Si IX and X, respectively). Kuhn and his colleagues also reported their probable detection of eight-times-ionized silicon at 3.93 microns. They observed this emission line during the February 26, 1998, eclipse from an instrument-laden C-130 cargo plane over the Pacific Ocean. If this result is confirmed, the spectral line would be the brightest coronal line yet found in the infrared and could improve the prospects of directly probing and measuring the coronal magnetic field.
At the same eclipse, they detected infrared emission from fine dust within 15 solar radii. Sunlight reflected from interplanetary dust farther out provides the faint, hazy zodiacal light seen under dark, rural skies after dusk and before dawn. This dust lies primarily in and near the plane of the solar system. It originates from outgassing comets and colliding asteroids and meteoroids, then gradually spirals in toward the Sun. Kuhn and his colleagues have detected zodiacal dust actually within the corona. Some astronomers had suspected that the inspiraling dust would pile up in rings in the corona, but this was not observed.
Nelson Reginald and Joseph Davila (NASA/Goddard Space Flight Center) and Shadia Habbal (Harvard-Smithsonian Center for Astrophysics) are among those making measurements at eclipses to check their models of the solar wind.
Imaging the Corona
The corona's shape and structure change dramatically from eclipse to eclipse as the Sun goes through its 11-year activity cycle. A major change could be seen even between the eclipse of February 26, 1998, which exhibited an elongated corona typical of solar minimum, and the one 18 months later on August 11, 1999, which had a more-or-less round corona typical of solar maximum.
The corona fades away rapidly from the solar limb (by a factor of about 1,000 in the first solar radius!) so conventional films or even CCD detectors do not cover the whole dynamic range well. For three decades, starting with the work of Gordon A. Newkirk Jr. (High Altitude Observatory) in 1966, photographers have used special radially graded filters to photograph the entire shapes of the magnetically controlled coronal streamers (see the right-hand photograph on page 44). Richard Fisher, Alice Lecinski, and David Elmore are among the HAO scientists who have worked with the equipment.
As late as 1993, when John MacKenty (Space Telescope Science Institute) joined the author in publishing composited unfiltered eclipse photographs, it was unusual to do so. Now computer image processing with Adobe Photoshop has allowed amateurs to assemble compound images that combine very long and short exposures smoothly (S&T: January 1998, page 117). These reveal coronal shapes from the limb all the way to the corona's outermost reaches. Professionals must use other programs, such as IRAF, to combine images in order to maintain the relative intensity values.
The High Altitude Observatory group members, using their radially graded images, have observed motions in thin plumes at the solar limb. Their paper, by Bruce Lites and others, was published in Solar Physics in 1999.
It turns out that only some of the light we see in the corona originates near the Sun itself. Some is added closer to us, by interplanetary dust scattering sunlight forward in our direction. This additional light is relatively unpolarized, while the light reflected to us by electrons in the corona itself is highly polarized (the light-wave vibration is restricted to one plane). Measuring the polarization enables us to separate the "Fraunhofer corona" of dust, which shows the spectrum of the solar photosphere, from the normal electron "K-corona" visible during eclipses. (The K is from the German word for "continuous" since the Doppler shifts of the hot coronal electrons blur out the spectral lines.)
Frederick Clette, Jean-Rene Gabryl, and Pierre Cugnon (Royal Observatory of Belgium) and their colleagues put together a Europe-wide observing campaign, the JOSO-TECONet 99, to measure polarization during the 1999 eclipse. Many amateur astronomers contributed to the effort, taking white-light coronal images through polarizing filters and sending them to Belgium for analysis. Nelson Reginald and Joseph Davila, as well as my own group from Williams College's Hopkins Observatory (especially colleagues Bryce Babcock and Lee Hawkins and student Sara Kate May), also made polarization measurements.
In addition, we are trying to use these observations, along with exposures taken through specially chosen filters, to map the corona's temperature. Putting all these measurements together helps determine the nature of this complex plasma. Our work was supported by the National Geographic Society's Committee for Research and Exploration.
Then there are some observations of whose value I am skeptical. For example, I do not think that the tiny changes in the size of the Sun reported over the last centuries, based on a comparison of the positions of the edges of totality in Edmond Halley's time and at recent eclipses, are valid. Indeed, two papers from a symposium at the International Astronomical Union's General Assembly last August reported that no variations in the solar radius were detectable. One of the papers was written by Jeffrey Kuhn, Marcello Emilio (University of Sao Paulo, Brazil), and Rock Bush and Philip Scherrer (Stanford University); they based their work on observations from the Michelson Doppler Imager (MDI) instrument on SOHO. My own report, published in Solar Physics and written jointly with Brant Nelson (NASA/Jet Propulsion Laboratory), compared observations and predictions at the 1984 eclipse in Papua New Guinea; it showed that the uncertainties are too high for a tiny solar shrinkage to be detected.
Science at a Bargain Price
The advantages of eclipses over space observations include flexibility and cost. New techniques and instruments can be incorporated into an eclipse expedition on short notice. Bulky, state-of-the-art equipment can be transported to remote sites on Earth far more cheaply than to space. Eclipse hardware doesn't have to meet rigorous launch standards for sturdiness. Furthermore, eclipse instruments can be mounted on steady bases (with the Earth as a platform) and can be adjusted at the last minute on the spot.
The SOHO mission, a joint project of NASA and the European Space Agency, costs hundreds of millions of dollars. A well-equipped modern eclipse expedition can be organized for less than a thousandth of that. Even allowing for some eclipses to be clouded out, they are a very cost-efficient way of doing solar research.
So eclipse science is far from over yet! The glories of space observations should not keep us from exploiting these brief opportunities to learn important new things. On June 21st we have such an opportunity in southern Africa (S&T: September 2000, page 32). Then we will have to wait until December 4, 2002, and March 29, 2006, for the next good eclipse prospects. Some of us are sure to be there.
A Pro-Am Solar-Eclipse Conference
Meetings of professional eclipse astronomers, such as those held in Bucharest in 1996 and Paris in 2000, provide forums where researchers can discuss the latest ideas and findings in the field. Too rarely do professional and amateur astronomers get together, but such a gathering was held last October 14-15 in Antwerp, Belgium. Organized by Patrick Poitevin (who runs the Solar Eclipse Mailing List and Solar Eclipse Newsletter on the Internet) and Joanne Edmonds, this first of a new series of professional-amateur solar-eclipse conferences attracted 155 participants from 22 countries.
The event featured exhibits and more than 35 papers. The speakers included Serge Koutchmy (Institut d'Astrophysique de Paris-CNRS), who talked about his team's high-resolution imaging and spectroscopic observations of the 1999 eclipse in Chadagan, Iran; Frederic Clette (Royal Observatory of Belgium), who discussed the results of the professional-amateur measurements of the polarization of the corona during the JOSOTECONet 99 campaign; and Clette's colleague Erwin Verwichte, who described the corona as observed with the SOHO spacecraft's Extreme-ultraviolet Imaging Telescope (EIT).
Eijiro Hiei (Meisei University, Japan), my predecessor as chair of the International Astronomical Union's Working Group on Eclipses, described differences in the coronal structure when the Sun was viewed at the million-degree temperature of the coronal "red line" and the 2-million-degree temperature of the "green line." Hiei also showed a spectacular high-definition TV movie of the 1991 eclipse from Mauna Kea, Hawaii. My own presentation concentrated on the wide variety of scientific advances possible from eclipse studies.
Studies of historical eclipses reveal slight changes in the Earth's rotation over many centuries. John Steele (Durham University, England), gave both his own paper on eclipses in ancient Mesopotamia and that of his absent colleague Francis Richard Stephenson, on eclipses and variations in the Earth's spin rate. Belgian amateur Felix Verbelen described the Maya counting system and his attempts to apply it to the Dresden codex, concluding that, contrary to published reports, total solar eclipses did not coincide with the given Maya dates. E. C. Krupp (Griffith Observatory) spoke of eclipse lore and myths.
Paul D. Maley (International Occultation Timing Association), who organizes expeditions to the edges of the paths of totality, talked about his group's efforts to demonstrate changes in the solar diameter and mentioned the years of unreduced data that IOTA has collected. John Hopper of Pepperell, Massachusetts, spoke about amateur airborne expeditions, while David Berghmans (Royal Observatory of Belgium) spoke about the country's solar- and space-weather activities.
Of great interest to many were the presentations by Peter Kalebwe (University of Zambia, Lusaka), Francis Podmore (University of Zimbabwe, Harare), and South African amateur Peter Tiede on the preparations under way at various sites for the 2001 and 2002 total eclipses in Africa.
Conference organizers Poitevin and Edmonds, who billed the meeting as "a crossroad on solar physics and eclipses of the Sun," plan to hold a similar conference in 2004, when no central eclipse occurs. For more information contact Poitevin, 7A The Drift, Rowlands Castle, Havant, Hampshire PO9 6DG, England; +44-7901-514-097; email@example.com or firstname.lastname@example.org.
Early Eclipse Science
Total eclipses of the Sun have a long and honorable tradition of providing important scientific discoveries. A famous example was the August 18, 1868, eclipse observed in Guntur, India, by the French scientist Pierre Jules Cesar Janssen with the newly invented spectroscope. Just before and just after totality Janssen saw a yellow spectral line emitted by solar prominences that was so bright that he deduced he could observe it even after the eclipse, which he did the following morning. At first he thought it was the familiar close pair of D lines from sodium. But the new line, D3, wasn't at quite the right wavelength. A hitherto unknown chemical element must be producing this line. The English astronomer J. Norman Lockyer independently arrived at the same conclusion. Lockyer named the new element helium, because it had been found only in the Sun--helios in Greek. Helium was not isolated and identified in a laboratory until 1895.
In 1869 the American astronomers Charles A. Young and William Harkness independently detected a faint green emission in the spectrum of the corona during the eclipse on August 7th of that year. It's still known as the coronal "green line" and was said at first to have come from the element coronium. Only when the periodic table of elements was nearly complete did it became obvious that there was no room left in the table for coronium. The problem of identifying it persisted for 70 years. Finally it dawned on scientists that coronium might be a known element in such extreme conditions that it produces a spectrum never seen on Earth. In 1939 Walter Grotrian of Germany pointed the way, and in 1941 Bengt Edlen of Sweden identified coronium as iron with 13 of its electrons missing, an ion known as Fe XIV. To ionize iron so greatly the corona had to be extremely hot, and for the line to appear at all, the density of the corona had to be nearly that of a vacuum.
Many people remember the deflection of starlight by the Sun's gravity, predicted by Albert Einstein's general theory of relativity, as an epochal eclipse experiment. Measurements reported by groups headed by the English astronomer Arthur S. Eddington of the eclipse on May 29, 1919, when the Sun was in front of the Hyades cluster, provided evidence for the weird and counterintuitive theory and made Einstein's name world famous. The experiment was repeated during the eclipse of September 21, 1922, and a half dozen times later.
The most recent observations were conducted in Mauritania on June 30, 1973--but by then Einstein was already on rock-solid ground. In 1974 and 1975 radio astronomers Edward B. Fomalont and Richard A. Sramek achieved greater accuracy when they used widely separated radio telescopes to measure the positions of quasars as the Sun passed in front of them. Such observations can now be carried out day or night without waiting for an eclipse. And, in fact, the Hipparcos astrometry satellite measured the solar gravitational deflection in every star's position across half the celestial sphere! (Nowadays gravitational lensing is routinely observed in many other areas of astronomy as well.) Today's professional eclipse expeditions look for new types of observations, not the same old ones of decades ago.
A veteran of 30 solar eclipses, JAY M. PASACHOFF is Field Memorial Professor of Astronomy at Williams College in Williamstown, Massachusetts, and chair of the International Astronomical Union's Working Group on Eclipses. He is coauthor with Leon Golub of the book Nearest Star: The Exciting Science of the Sun, to be published this spring by Harvard University Press.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||includes related article|
|Author:||Pasachoff, Jay M.|
|Publication:||Sky & Telescope|
|Date:||Feb 1, 2001|
|Previous Article:||Today's science of the sun. Part 1.|
|Next Article:||A year of discovery: astronomy highlights of 2000.|