Mapping the Galaxy
The field of astronomy that deals with measuring the positions of celestial bodies in the sky is known as astrometry. Over the course of many centuries astronomers have relied on astrometry to compile ever more detailed maps of the heavens. By monitoring how the positions of stars and other astronomical objects vary over time, it is possible to infer their distances from us and their motions through space – both of these are essential for investigating the physical nature of these distant bodies.
Tiny changes in the position of stars contain information about their distance – this is encoded in the parallax, an apparent annual shift of stars on the sky caused by Earth's motion around the Sun. Over time, stellar positions also slowly change due to the stars' real movement through the Galaxy. The motion perceived across the plane of the sky is termed the proper motion. This provides two of the three components of a star's velocity through the Galaxy; the third component – the radial velocity – can be inferred from the red- or blue-shift of the light in its spectrum.
Gaia will perform its unique census of stars in the Galaxy using the billion-pixel camera at its heart, which collects the light focused by the satellite’s dual telescope system. As the satellite spins, the two telescopes scan great circles on the sky. They feed three instruments: one for astrometry (to determine the positions and motions), one for photometry (to measure the colours of the stars) and one for spectroscopy (to measure their radial velocity and see what the stars are made of).
The astrometric instrument is the core element in the focal plane, providing input to determine the position of stars and other astronomical sources to unprecedented precision; in turn, the positions are used to derive stellar parallaxes and proper motions. This instrument also measures the brightness of all stars in the Gaia G-band, using a bandpass that covers the visible portion of the spectrum, between 330 and 1050 nm. This band was chosen to optimise the collection of starlight, which in turn maximises the precision of the measurements. It also matches the overall sensitivity of the Gaia astrometric instrument thus providing a measure of the brightness as seen by that instrument.
The photometric instrument separates the light from the stars into their constituent colours and does this over a shorter (from 330 nm to 680 nm) and longer (from 640 to 1050 nm) wavelength range. These measurements will in turn be used to determine the colour of each star – the portion of the spectrum where it emits most of its light – which is key to determine its temperature, age, and other properties.
Finally, the radial velocity spectrometer provides spectra for a subset containing the brightest stars observed by Gaia; these spectra are used to estimate the radial velocity of stars – their velocity along the line of sight with respect to Gaia.
For each celestial object, all the information recorded by the instruments is compressed into data packets and stored on board. The data packets are transmitted to ESA's ground stations in Spain, Australia, and Argentina, and are then processed by the Data Processing and Analysis Consortium (DPAC), who are responsible for turning them into scientifically useful data.
At the core of this process is a mathematical procedure called AGIS – the Astrometric Global Iterative Solution. Put simply, this is the way to collate the billions of pieces of information that come from the satellite and convert these into the Gaia map of the Galaxy.
It can be thought of like a giant jigsaw puzzle with hundreds of billions of pieces that have to be combined, very accurately, before the complete picture emerges. For Gaia, the assembly of this gigantic picture is a complex process that needs to be performed over a number of years in order to reach the expected measurement goal of 20 microarcsecond accuracy for a magnitude 15 star. (A magnitude 15 star is about 4 million times fainter than Sirius, the brightest star in the sky.)
In practice, Gaia's focal plane tracks the position where light from each star falls as it passes across the CCDs and records the time of each transit. To transform this information into astronomically useful quantities, namely measurements of stellar positions on the sky, scientists need to know where Gaia was pointing at each time, as well as its relative position with respect to the Sun and other planets in the Solar System. In addition, they also need to take into account the light bending caused by the Sun and the major planets as well as the exact position of each element in Gaia's optical system and focal plane.
This calibration process is carried out in an iterative fashion. In the data processing, the images of stars recorded on the CCDs are compared to models of how such an image should look. As more data are collected and fed to the algorithms, the models are constantly improved and, in turn, this improves the estimates of the stellar positions on the sky. Other information can also be added in later steps of this process, such as the colour of stars measured by the photometric instrument, since stars of different colours will give rise to slightly different images on the CCDs.
For future data releases, the process will be repeated for more and more data until the estimated position, brightness, colour and other parameters of each source are locked into a consistent and coherent framework. However, in the first data release, based on less than a quarter of the total amount of data that will be collected by Gaia over its entire mission, this iterative loop was performed using only a small subset of the stars – those in common with the earlier Hipparcos and Tycho-2 catalogues for which the position was already known to good accuracy. In the following releases, the subset of stars used for the calibration will be increasingly larger, allowing scientists to estimate more and more parameters to even greater accuracy for all surveyed sources.