Present

In a field as fast moving as extrasolar planets, any attempt to summarize the current status is doomed to appear rather antique in a short period of time. Several major discoveries have been announced in the two months between the May 2005 Symposium at ST Scl and the writing of this summary (early July 2005), with more sure to follow once the Symposium proceedings go to press. With that caveat, the status of the field as of the time of the May 2005 Symposium can be summarized by the plot of discovery space shown in Figure 4. Here, the masses of extrasolar planets are shown as a function of the semimajor axes of their orbits. By May 2005, well over 130 planets had been discovered and submitted for publication in refereed journals. The International Astronomical Union's Working Group on Extrasolar Planets maintains a list of extrasolar planets that meet the Group's requirements for inclusion on their web site at http://www.dtm.ciw.edu/boss/iauindex.html. This web site also addresses the question of defining what is and what is not a "planet."

Figure 4. Extrasolar planet discovery space as of May 2005, showing primarily minimum masses of planets discovered by Doppler spectroscopy as a function of orbital semimajor axis. The oblique dashed line shows how the sensitivity limit of Doppler spectroscopy depends on semimajor axis for accuracies of ~2-3 m s_1 and a signal-to-noise ratio of —short-period, massive planets are the easiest to detect. Brown dwarfs (open symbols near the top of discovery space) are infrequent companions to solar-type stars.

Figure 4. Extrasolar planet discovery space as of May 2005, showing primarily minimum masses of planets discovered by Doppler spectroscopy as a function of orbital semimajor axis. The oblique dashed line shows how the sensitivity limit of Doppler spectroscopy depends on semimajor axis for accuracies of ~2-3 m s_1 and a signal-to-noise ratio of —short-period, massive planets are the easiest to detect. Brown dwarfs (open symbols near the top of discovery space) are infrequent companions to solar-type stars.

3.1. Doppler Spectroscopy Nearly all of the planets shown in Figure 4 were discovered by the Doppler spectroscopy method, which yields only a lower limit on the mass of the planet because of the unknown orientation of the planet's orbit with respect to the line-of-sight to the star. If the planetary orbit is being observed nearly pole-on, then the mass of the companion is larger by a factor of 1/sin i (where i is the inclination of the planet's orbit; i = 0 for pole-on) than the minimum mass plotted in Figure 4. Assuming that planetary orbital planes are randomly distributed, the typical true planetary mass should be a factor of 4/n larger than the minimum mass, i.e., ~1.3 times higher.

Given the drive to find Solar System analogues and their spectroscopic suitability, most of the target stars in the ground-based spectroscopic planet searches have been late F, G, and K dwarfs, though these searches have also been extended to early M dwarfs. It is evident from Figure 4 that solar-type stars tend not to have brown dwarf companions, i.e., companions capable of burning deuterium (requiring a mass greater than ~13 Jupiter masses for solar composition), but not hydrogen (implying a mass less than ^75 Jupiter masses). This is consistent with the lack of binary companions with mass ratios much larger than 10:1 (Duquennoy & Mayor 1991)—solar-mass primaries generally do not have brown dwarf companions.

Figure 4 also makes it clear that such stars do typically have planetary-mass companions, though evidently the process that produces these objects does not always adhere to the IAU definition of a planet as being an object less massive than 13 Jupiter masses. Note that the absence of planets with semimajor axes greater than a few AU does not imply their nonexistence, but rather is a result of the need to follow a target star for an entire planetary orbit before announcing a detection in order to minimize the risk of interpreting noisy data as a detection. The spectroscopic surveys are just now entering the phase of their programs where they have been monitoring stars with sufficient accuracy (a few m s_1) long enough (a decade or so) to begin to detect long-period planets similar to Jupiter.

Figure 4 shows that the range of masses of extrasolar planets is considerably greater than in our Solar System: planets with masses ten times that of Jupiter exist, as well as masses smaller than that of Saturn. Recently, the attainment of Doppler spectroscopy precisions of m s_1 means that the lower mass limit has been extended down to Neptune masses by the discoveries of planets orbiting GJ 436 (Butler et al. 2004), Mu Arae (Santos et al. 2004), and p1 Cancri (McArthur et al. 2004). These three might well represent the first examples of a new class of planets, i.e., they could be ice-giant planets, given the similarity of their masses to those of Uranus and Neptune, or they might be super-Earths, rocky planets with masses well above that of Earth and Venus. The former explanation seems to be inconsistent with the occurrence of several gas-giant planets on longer-period orbits in both the Mu Arae and p1 Cancri systems, implying that the Neptune-mass planets formed inside the orbital radii of their gas giants, and then migrated inward. While this explanation seems most plausible, it will remain for a transit detection of a "hot Neptune" to measure one of these planets' mean densities, and so determine whether it is composed primarily of rock or of rock and ice/water.

The orbital radii evident in Figure 4 cover a wide range—from the semimajor axes of 0.02 AU of the "hot Jupiters" out to 5.2 AU for the "cold Jupiters," with a number of "warm Jupiters" orbiting in between. The close-in orbits of the hot and warm Jupiters imply significant post-formational inward orbital migration, given the difficulties in forming gas giants so close to their stars. Perhaps most surprisingly, many of the orbits are highly eccentric, making the low eccentricities of the major planets in our Solar System seem out of the ordinary, rather than the norm. The origin of these eccentricities has become another major theoretical puzzle—are they a result of the formation process or of the orbital-migration process?

It is notable that there have been no discoveries of extrasolar planets to date with the astrometric technique, though there have been two astrometric measurements of previously-known planets (for GJ 876 and p1 Cancri) using the Fine Guidance Sensors of HST (McArthur et al. 2004).

3.2. Transits

The first planet seen to transit its host star was the hot Jupiter orbiting HD 209458, detected by Doppler spectroscopy (Charbonneau et al. 2000; Henry et al. 2000). Because of the short-period orbits of the hot Jupiters, orbiting at roughly 10 stellar radii, the chances of having the orbit of a hot Jupiter aligned so as to lead to a transit are roughly 10%. We had to wait for the tenth hot Jupiter to be discovered by spectroscopy before one of them was discovered to be a transiting planet—hopefully we will be luckier with the hot Neptunes.

HD 209458's planets provided the first strong evidence that many, if not most, of the objects in Figure 4 are indeed gas-giant planets. A transit fixes the orbital inclination, and thus the mass of the planet, and the depth of the transit allows the radius of the planet to be determined as a fraction of its host star's radius. HD 209458's planet's mass is « 0.7 MJ, and it has a radius and a density roughly equal to that expected for a hot Jupiter. In addition, sodium was detected in its atmosphere (Charbonneau et al. 2002), as predicted for a hot Jupiter (Seager & Sasselov 2000).

Figure 5. Spitzer Space Telescope (SST) 24-micron photometry of the star-planet system HD 209458 during a secondary eclipse of the planet by the star, constituting the first direct detection of light from an extrasolar planet (Deming et al. 2005). A similar detection was accomplished for the TrES-1 system by Charbonneau et al. (2005) using SST.

Figure 5. Spitzer Space Telescope (SST) 24-micron photometry of the star-planet system HD 209458 during a secondary eclipse of the planet by the star, constituting the first direct detection of light from an extrasolar planet (Deming et al. 2005). A similar detection was accomplished for the TrES-1 system by Charbonneau et al. (2005) using SST.

While Doppler spectroscopy has been by far the leader at detecting new planets, the transit detection technique has now accounted for the discovery of six, all confirmed by follow-up Doppler spectroscopy. The first planet detected by transit photometry was a hot Jupiter orbiting a star toward the galactic bulge (Konacki et al. 2003), that had been observed to have photometric variations consistent with a transiting planet (Udalski et al. 2002). This planet is known by the name of the transiting event, OGLE-TR-56. The Optical Gravitational Lensing Experiment (OGLE) project (Udalski et al. 2002) at the Las Campanas Observatory in Chile has discovered a large number of possible planetary transits, and four more so far have turned out to be caused by planets, all as a side benefit of the OGLE project. A sixth transiting planet (TrES-1) has been found by a new ground-based transit search program, the Transatlantic Exoplanet Survey (Alonso et al. 2004). Because transit surveys preferentially find short-period planets, all of the planets found to date by transits are hot Jupiters, though they are mostly smaller and denser than HD 209458's planet (Sozzetti et al. 2004).

3.3. Microlensing

A third technique that has found a planet around a main-sequence star is microlensing, where the photometric variations caused by gravitational bending of background starlight by a foreground star can be enhanced for a period of a few days by a planet orbiting at the Einstein radius. The first microlensing detection was accomplished by Bond et al. (2004) associated with the microlensing event known as OGLE 2003-BLG-235/MOA 2003-BLG-53. [Clearly there is a need for more succinct names for some of these extrasolar planets.] The inferred planet has a mass of ~1.5 Jupiter masses and orbits at ^3 AU from the presumed main sequence host star.

Figure 6. The apparent first spatially-resolved image of an extrasolar planet (b) orbiting ~100 AU from the young star GQ Lup (A). The image was obtained with the adaptive optics instrument NACO on the VLT by Neuhauser et al. (2005).

3.4. Pulsar Timing

Astronomers have continued to search for more planetary-mass companions to pulsars by looking for minute variations in the pulsars' precise pulsation periods, as was used by Wolszczan & Frail (1992) to discover the pulsar PSR 1257+12 planetary system. However, searches of over 100 pulsars to date have yielded very little evidence for more planetary-mass companions, unlike main-sequence stars. The exception is the detection of a gas-giant planet-mass companion to a binary star system containing a white dwarf and the pulsar PSR B 1620-26 in the M4 globular cluster by Sigurdsson et al. (2003). Evidently gas-giant planets can form in regions quite different from the galactic disk, including extremely metal-poor environments such as an ancient globular cluster.

3.5. Direct Detections

The field of extrasolar planets took another enormous leap forward in 2005 with the announcement of several direct detections of extrasolar planets. The previous evidence for sodium in the atmosphere of HD 209458's planet (Charbonneau et al. 2002) was obtained by noting a depletion of the host star's light at the wavelengths of the sodium doublet lines during planetary transits, so formally speaking, this discovery did not detect photons from the planet itself, but rather the absence of stellar photons that had been absorbed in the upper atmosphere of the planet. All that changed in 2005, when the Spitzer Space Telescope (SST) enabled the first direct detection of the light from two transiting planets, HD 209458 (Figure 5; Deming et al. 2005) and TrES-1 (Charbonneau et al. 2005). Because the hot planet is relatively bright at mid-infrared wavelengths, when the planet disappears behind the star (the secondary eclipse) the total amount of mid-infrared light is observed to decrease by a measurable amount. When observed out of eclipse, then, the extra photons must be coming from the planet.

While pathbreaking, the SST direct detections of the HD 209458 and TrES-1 hot Jupiters were not direct detections in the sense of spatially resolving the light from the planet from that of the star. That honor appears to have been reserved for the detection of a multiple-Jupiter-mass planet in orbit around the T Tauri star GQ Lup. Neuhauser et al. (2005) were able to obtain an image of GQ Lup's planet (Figure 6) by using the adaptive optics system on one of the Very Large Telescopes (VLT) in Chile. While the mass of the planet is uncertain, it could exceed the 13-Jupiter-mass upper bound for being a planet. The young age of this system (~1 Myr), when combined with recent models of the evolution of newly formed gas-giant planets, implies that this is indeed the first detection of a spatially-resolved extrasolar planet, though at a puzzlingly large orbital separation of ^100 AU. This discovery has produced yet another challenge for the theorists: forming a gas-giant planet at such a large distance is likely to be difficult, implying that GQ Lup's planet may have been kicked outward to its present locale.

Prior to the discovery of GQ Lup's planetary candidate, another excellent discovery was made by the VLT. Chauvin et al. (2004) imaged a roughly 5-Jupiter-mass companion to the brown dwarf 2M1207. Because 2M1207 has a mass itself of only about 25 Jupiter masses, this system has a mass ratio of ~5:1, typical of binary star systems. The two components of the 2M1207 system thus seem most likely to have formed simultaneously during the collapse and fragmentation process that forms binary and multiple star systems. While the very low mass of the companion seems to place it in the planet category, the fact that it is in orbit around a brown dwarf argues against awarding this discovery the prize of the first spatially resolved, direct detection of a "planet." A number of planetary-mass objects had previously been imaged in regions of recent star formation, with masses as low as that of 2M1207's secondary, and these objects are believed to have been formed directly by the same star formation process that leads to main-sequence stars and brown dwarfs. Accordingly, such objects are perhaps better referred to as "sub-brown dwarfs."

Was this article helpful?

0 0

Post a comment