|
The OGLE-2005-BULGE-390 field, showing the microlensing event. Over the course of a few weeks in July, 2005, the indicated star brightened and dimmed due to gravitational lensing by an invisible foreground star. |
(Media resources, including press release, image, videos, and other links available here)
Two astronomers at Perth Observatory have played a crucial part in the discovery of a new planet around a distant star, significantly more Earth-like than any other planet discovered so far. The planet, which is only about 5 times as massive as the Earth, orbits its parent star every 10 years, with an orbital radius of about 3 times the Earth-Sun distance in our solar system. The star it orbits (its 'sun') is much like our Sun, except it is only 1/5th as massive, and much closer to the Galactic center. It was discovered using a technique known as gravitational microlensing, and was reported in Nature as the joint effort of three independent microlensing campaigns: PLANET/RoboNet, OGLE, and MOA, involving a total of 73 collaborators affiliated with 32 institutions in 12 countries (France, United Kingdom, Poland, Denmark, Germany, Austria, Chile, Australia, New Zealand, United States of America, South Africa, Japan). Andrew Williams and Ralph Martin at Perth Observatory were founding members of the PLANET (Probing Lensing Anomalies NETwork) in 1995, and have been involved with the group ever since.
 |
The Perth-Lowell Automated Telescope dome at Perth Observatory. Mounted on a 15m tower, the telescope is situated high enough above the ground to put it into a clear, non-turbulent airstream, reducing the distortion caused the by the atmosphere. The telescope has been running automatically since 1993, and has weather sensors to allow the system to pause and close the roof when cloudy, and restart when the weather clears. |
(The sky above Perth Observatory at 1am on 2005/8/10, at the peak of the planetary anomaly. The main research telescope is in the tower at the far right.)
Using the brightness of the parent star, and the orbital distance, this planet has a calculated surface temperature of only 50°Kelvin or so (220°C below zero). Due to its low mass and low temperature it's probably solid: An icy or rocky planet, similar to (but larger than) Pluto in our solar system, rather than gaseous like Jupiter, Saturn, Uranus and Neptune.
More than 170 planets outside the solar system (known as extra-solar planets or exoplanets) have been discovered to date. Almost all were found using the 'Radial Velocity' technique, which detects planets by measuring the tiny back-and-forth movements of the parent star as the planet orbits. This technique is heavily biased towards finding large planets which are very close to their parent star. If a distant star had a set of planets with masses and orbits identical to the ones in our solar system, they would be almost impossible to detect using the radial velocity technique with current technology.
This new planet (with the unglamorous identifier of "OGLE-2005-BLG-390Lb") is probably the smallest detected so far. Only one other known extra-solar planet comes close: Gliese 876d, at 7.3 ± 1 Earth masses, is within the statistical errors on mass but in a very different orbit - orbiting its sun every 2 days, at 1/50th of the Earth-Sun distance.
Microlensing uses a different strategy, and can find planets more like the ones we are familiar with. In 1915, Einstein predicted that gravity should bend light, and an expedition to Western Australia in 1922 proved it: observations of stars almost directly behind the Sun during a solar eclipse showed exactly the distortion predicted. The gravity of distant stars bends light as well, distorting and magnifying the sky behind them, but this effect is only visible if there is another star almost perfectly in line with the 'lens' star. Finding these near-perfect 'gravitational lenses' has only been possible since the mid-90's. Now, large telescopes run by the Polish-American OGLE (Optical Gravitational Lensing Experiment) team and New Zealand-Japanese MOA (Microlensing Observations in Astrophysics) collaboration, find hundreds of these events every season.
 |
A diagram of the Milky Way, showing a typical microlensing event. The source star is a bright (giant class) star in the bulge of the Milky Way, far enough above or below the disk to be visible past the dark dust clouds in the plane of the main disk of the galaxy. The lens star is a dim (main-sequence) star like our sun, part-way in between us and the source star, inside the disk itself.
(Original image from ESA, modified by Andrew Williams) |
While the gravity of a single star generates a perfect lens effect, two or more masses give a distorted magnification pattern, like projecting light through an oddly-shaped lump of glass. This can range from a large effect (like looking through the base of a wine glass), all the way down to small-scale changes, like a small flaw in an otherwise perfect lens. Although the nearby star (whose gravity is bending the light) is often so dim that it's not visible at all, the exact pattern of 'distortion' seen in the brightness of the background star can be used to calculate the masses and separation of the elements of the gravitational lens.
Dr Andrew Williams, at Perth Observatory, says "We let the gravity of a dim nearby star act as a giant natural telescope for us, magnifying a distant, bright star. A small 'defect' in this gravitational lens revealed the existence of a planet around the nearby star. We don't see the planet, or even the star that it's orbiting, we just see the effect of their gravity"
What an observer on Earth would see (given an imaginary telescope in space, with perfect optics) is illustrated in some computer simulations. They show:
This one shows the whole event - the distortion caused by the planet is the short-lived 'glitch' in the larger yellow image just as the source (red circle) is leaving the Einstein Radius (blue circle). This one zooms in on the planetary anomaly, and slows down so you can see it in more detail. The planet itself is shown as a small blue dot. They are in MPEG-4 format, so you'll need an MPEG4 viewer, eg a recent version of Quicktime.
The existence of a planet around a lens star often causes a distortion that lasts only a few hours. In order to be able to catch and characterize these, nearly-continuous round-the-clock high-precision monitoring is required. This is achieved by the PLANET network of 1m-class telescopes:
Since 2005, PLANET has been collaborating with RoboNet, a UK-operated robotic telescope network consisting of the Liverpool 2.0m (Roque de Los Muchachos, La Palma, Spain) and the Faulkes North 2.0m (Haleakala, Hawaii, USA). This network will be enhanced by the Faulkes South 2.0m (Siding Springs, Australia) from 2006.
Perth Observatory plays an essential role in the PLANET collaboration because of our position on the map. Detecting short-lived fluctuations like the one caused by this planet requires observations every hour or so, 24 hours a day. Observations from Perth Observatory (and Canopus Observatory in Tasmania) fill the large time-zone gap between Chile and South Africa in the Southern Hemisphere. Here is a view of the Earth from above the South Pole, showing the PLANET telescopes and their distribution around the planet to provide continuous 24-hour coverage.
 |
The light curve (a graph of brightness versus time) of the lensing event. Each point represents the brightness in a single image, the color of the point shows which telescope took that image. The regular cycle of colors shows how observing is taken over by the next telescope in turn as the night ends at each site. The inset in the top right is an enlargement of the deviation from a 'perfect' lensing curve caused by the gravity of the planet. The dark blue points indicate images from Perth Observatory. Note that the blue (Perth) points are crucial, ruling out the best non-planet model for the event - the grey dashed curve representing a binary star as a lens, instead of star and planet. |
The OGLE search team discovered the event OGLE-2005-BLG-390 on 11 July 2005, allowing PLANET to start taking data. A light curve consistent with a single lens star peaking at an amplification of about 3 on 31 July 2005 was observed, until PLANET member Dr. Pascal Fouque, observing at the Danish 1.54m at ESO LaSilla, noticed a planetary deviation seen in the data taken on 10 August. An OGLE point from the same night showed the same trend, while Dr. Andrew Williams reported that the last half of the planetary deviation, lasting about a day, had been covered by images from Perth Observatory. The MOA collaboration was later able to identify the source star on its frames and confirmed the deviation with two data points.
The observed light curve - the graph of the brightness of the background star over time - doesn't give the mass of the planet directly. Instead, it gives the ratio between the mass of the planet, and the mass of the star it orbits. Since this lensing star, like the planet, is too faint to resolve, its mass can't be measured directly. Instead, the mass of the lensing star was derived from its location, plus other clues from the light curve, along with a statistical model of the distribution of stars within our galaxy. This statistical model is responsible for most of the uncertainty in the planet properties. Improved technology over the next decade or so should allow the lens star to be resolved directly, and pin down the parameters more precisely.
OGLE-2005-BLG-390Lb is only the third extra-solar planet resulting so far from microlensing searches. This is due to the fact that planets of Jupiter-mass or above, which are much easier to detect, appear to be rare around M-dwarfs, the stellar type most commonly found as the 'lens' stars in microlensing. This rarity has been indicated independently by microlensing and radial-velocity searches. While the other two microlensing planets have masses of a few times that of Jupiter, the discovery of a sub-Neptune mass object so soon is a strong hint that these smaller objects, in contrast, are quite common, but much harder to detect.
Computer simulations of planet formation using core-accretion models predict that giant planets like Jupiter and Saturn begin as lower-mass planetary cores of rock and ice, which then accrete large amounts of Hydrogen and Helium gas from the proto-planetary disk. However, the host star for this new planet is an M-dwarf star, like most stars in our Galaxy. These stars have less than half the mass of the Sun, and the core accretion theory predicts that planets in these systems will usually form too slowly to grow to the mass of Jupiter. For most stars, the simulations predict a large fraction of planets with masses below 10 Earth-asses, at orbits between 0.1 AU and 10 AU. By coincidence, these orbital separations match well the range preferred by microlensing, making it an ideal technique for studying this population down to Earth mass.
The discovery of a sub-Neptune mass planet should encourage the intensification of microlensing planet searches, using current and additional facilities from the ground - or even with a space-based campaign in the near feature - by providing an observational hint that further low-mass planets will be detected. It shows that microlensing is a straight path to the discovery of a twin Earth.