FEATURE ARTICLE
Spitzer’s Cold Look at Space
To get a clear view of infrared emissions from celestial objects, the Spitzer Space Telescope has been cryogenically cooled—and what sights it has seen
Michael Werner
The Birth of Stars and Planets
Within a few thousand light-years of the Sun, in our corner of the Milky Way galaxy, new stars and planetary systems are forming from the gravitational collapse of dense interstellar clouds. Much of the formation and early evolution of stars occurs at temperatures and densities that are particularly well suited to study in the infrared. Spitzer has proved to be a very powerful instrument both for studying large-scale patterns of star formation and for detailed studies of individual nascent stars.
In addition, Spitzer has reinforced our previous understanding that the early star will be embedded in a circumstellar disk that develops as a result of the conservation of angular momentum in the collapsing cloud. The material in the disk can evolve into a planetary system, and the recent discovery of planets around literally hundreds of nearby, solar-type stars indicates that this occurs very frequently. The dense, planet-forming (or protoplanetary) disk is an ideal target for infrared studies as it is heated by the star and reradiates in the infrared. The protoplanetary disk dissipates as planets form and the star evolves, leaving behind a tenuous, residual debris disk consisting of dust particles that are generated and regenerated by evaporation or by collisions of asteroidal or cometary objects within the planetary system. Spitzer can easily study these debris disks—which are useful indicators of planetary-system evolution—around numerous solar-type stars because their large surface area makes the disks brighter in the infrared than are the stars themselves.
At the other end of the process of planetary production, Spitzer has directly measured the light from fully formed planets circling other stars. Most known exoplanets have been discovered by observing their effects on the star they orbit. Most often, this is done using a technique that exploits the small variations in the star’s radial velocity, as seen from the telescope’s position, in response to the gravitational pull of the orbiting planet. With one or two startling exceptions, however, the light from an exoplanet cannot currently be spatially separated from that of its parent star; the star is too bright and the planet too close.
The exoplanets found by the radial-velocity technique tend to be large planets close to the stars, as these produce the largest signal. Exoplanets of this type have been dubbed “hot Jupiters” because they are of a similar mass to our Jovian neighbor and are greatly heated from being so near their star. In fact, the first exoplanet discovered, 51 Pegasus b, announced in 1995, was a Jupiter-mass planet in an orbit a fraction of the size of Mercury’s; this setup was so different from our solar system and from all expectations that it unleashed a flood of theoretical and modeling papers that has continued to grow as other surprising exoplanet systems are discovered.
Now we know of more than 350 exoplanets orbiting more than 200 stars near the Sun. These objects are in systems containing up to five planets, and the smallest exoplanets that can now be detected are less than five times the Earth’s mass. Even as further discoveries pour in, it is important to start characterizing these exoplanets, to understand both the universal features of planetary systems and any idiosyncrasies that may have been at play in the formation of our own solar system. Although ground-based and Hubble Space Telescope studies have also been important, Spitzer has been the astronomical community’s most powerful tool for characterizing exoplanets. The results continue to show that the formation and evolution of planetary systems is a much richer topic than we had thought.
A large exoplanet close to its star—and therefore heated to, say, 1,000 kelvins or more—can be bright enough in the infrared for Spitzer to detect if it lies within about 200 light-years of the Sun. Spitzer has detected these hot Jupiters in numerous cases by temporally, rather than spatially, separating the light of the star from that of the planet. This technique is applied most simply to exoplanets in an orbit that lies edge-on as seen from Earth. When the planet passes in front of the star (or transits), there is a drop in the infrared signal from the star-planet system due to the physical blockage of the stellar disk, and this phenomenon permits an estimate of the size of the planet.
When the planet passes behind the star (or goes into secondary eclipse), there is again a drop in the infrared signal, but this time it’s because the planet’s contribution is no longer present. The size of this drop relative to the signal from the star measures the amount of infrared radiation from the planet. These observations, which require measurements with a precision of better than 0.1 percent over timescales of hours, are possible because of Spiter’s high sensitivity, along with the high stability and long, continuous viewing periods characteristic of Spitzer’s solar orbit.
The observations are particularly valuable when carried out at multiple wavelengths simultaneously using Spitzer’s spectrograph. Because hot Jupiters are rich in gas, different wavelengths arise from different levels in the atmosphere or different chemical constituents. Spitzer data have allowed the determination of planetary temperatures and of constraints on chemical composition (including the identification of water vapor), atmospheric structure and atmospheric dynamics. Spitzer has already characterized more planets (at least 19 in total) orbiting other stars than exist in our solar system; the planets characterized lie between about 50 and 200 light-years from Earth.
One study compared Spitzer measurements at five wavelengths of the exoplanet HD 189733b with predictions for radiation from a gaseous planet of solar composition at the observed temperature of the planet. The results indicate that HD 189733b does not have a high-altitude temperature inversion as has been inferred for other hot Jupiters, so there must be at least two different classes of atmospheres for this group of exoplanets. The key measurement of temperature inversion will be carried out for numerous other exoplanets during the upcoming Warm Spitzer mission, so we are likely in for more surprises.
Spitzer’s solar orbit permits extended, continuous observations over a complete 40-hour planetary orbit, not just for a few hours around transit or eclipse. Because the exoplanet is tidally locked to its star, the same hemisphere faces the star continually, just as tidal locking keeps one hemisphere of the Moon perpetually facing the Earth. In its edge-on orbit, the illuminated hemisphere comes gradually into view as the planet moves from transit to eclipse.
This led to the first mapping of the day-to-night temperature variation in an extrasolar planet, where days reached around 1,210 kelvins and night dipped to around 970 kelvins. In addition, the fact that the hottest area on the exoplanet is not closest to its encircled star means that there is substantial heat transport in this exoplanet’s atmosphere. It is estimated that winds in excess of 5,000 kilometers per hour are required in the upper atmosphere of the planet to account for the observed redistribution of the stellar heat.
Not all exoplanets are tidally locked in circular orbits, and perhaps the most extraordinary Spitzer result to date in the study of atmospheric structure and dynamics comes from an observation of a “Big Swing.” A planet in a highly elliptical orbit swooped by its star in such an orientation that Spitzer could monitor the temperature response of the planetary atmosphere to the resultant heat impulse. Fortuitously, the geometry was just right for the observations to include a secondary eclipse as well. The temperature of the planet’s atmosphere increased from 800 to 1,500 kelvins in just a few hours, suggesting that the stellar-heating energy is deposited high in the planet’s atmosphere, which gives clues about its structure.
In these early days of exoplanet characterization, one can but marvel at the variety of behaviors exhibited by the planets that have been studied to date. Recall that the results were all obtained without spatially isolating or imaging the exoplanet. Looking ahead, we can anticipate that newly launched and planned instruments will greatly extend these results, leading up to our first direct images of Earth-like exoplanets sometime in the next 20-to-30 years.
» Post Comment