Biggest space camera will map Milky Way

Standard Toaster

Biggest space camera will map Milky Way

..And the 1.5-gigapixel camera seems up to the task. Called Astro, its sensitivity dwarfs the Hubble Space Telescope's 16-megapixel main camera and even NASA's planned extrasolar planet finder, Kepler, which will boast an 84-megapixel array.


Clockwork motions
Once launched, the Gaia telescope will rotate in deep space at a point where the gravity of Earth and the Sun balance each other - keeping the spacecraft at a constant distance of about 1.5 million kilometres from Earth. As it spins, the Astro camera will repeatedly acquire images of unprecedented detail and colour so a 3D movie can be made of the positions and motions of stars - and any detected extrasolar planets - in the observed region of the Milky Way....

Would love to see the results of this.

The mission is scheduled to launch in 2011

Flukey

This is an interesting way of looking at the galaxy, posted in After Hours earlier, but of interest to us astronomy buffs.


http://www.micro.magnet.fsu.edu/primer/java/scienceopticsu/powersof10/

ps200306

Well, eleven years after the OP, Gaia released its first data today. ESA announcement press conference here (they screwed up the streaming for the first 15 mins, so it's a bit disjointed):

Today they've release positions and magnitudes for a billion stars, about 1% of the Milky Way. In the future there'll be light curve and space motion data, which today's release has for about 2 million objects. As well as stars there are quasars, asteroids, Kuiper belt objects, and basically everything between magnitudes 11 and 21 picked up by Gaia's 106 CCD gigapixel camera.

All the data is publicly available, although you'll need to download between a quarter and half a terabyte just for the first release, or use the online query tool to get subsets.

Pat Dunne

ps200306 wrote: »

Well, eleven years after the OP, Gaia released its first data today. ESA announcement press conference here (they screwed up the streaming for the first 15 mins, so it's a bit disjointed):

Gaia's 106 CCD gigapixel camera.

Now that is what you call serious pixel envy

Here was me content with 20.2 megapixels on my DSLR

ps200306

I hope as a result of Gaia we'll be hearing a lot more about astrometry in the next few years. Astrometry is the measurement of the positions and motions of stars (whereas photometry is the measurement of their brightness). Maps of the positions of stars come down to us from classical antiquity, but they are probably based on works going back to the early Bronze Age, many thousands of years ago.

Hipparchus of Nicaea is famous as the inventor of stellar magnitudes -- the earliest photometric system in which stars were classified on a decreasing brightness scale of one to six. Around 135 BC, Hipparchus also compiled a star catalogue containing the positions of 850 stars, an incredible achievement. Akin to geniuses like Newton and Galileo, Hipparchus had to construct or invent both physical and mathematical tools to carry out his work.

The original of Hipparchus's catalogue no longer exists, but it was plagiarised and extended by Ptolemy three hundred years later. Ptolemy's work containing 1025 stars is the more famous by virtue of its widespread copying and circulation in the medieval world -- the Arab astronomers called it simply al-magest, "The Greatest" -- but Hipparchus richly deserves the credit for it.

Accurate star measurements have always increased our understanding of how the heavens work. Hipparchus's measurements were accurate enough to clearly show that something had changed since the catalogues of Timocharis of Alexandria and his student Aristyllus, compiled 150 years earlier. The stars had moved! Of course, the stars move across the sky every day, but this was a change with respect to the Sun's regular crossing of the equator at the equinoxes. Hipparchus had discovered the precession of the equinoxes, which we now know to be due to a wobble in the earth's axial tilt. This is also how we know about Ptolemy's plagiarism: he crudely added a fixed -- and wrong -- amount of equinocteal precession to each of Hipparchus's measurements to make his own catalogue.

Hipparchus's work, coming to us via Ptolemy's Almagest, was not superseded until the late 16th century. Medieval Islamic scholars had made some incremental improvements using increasingly accurate mural quadrants -- a kind of wall-mounted protractor. Hipparchus's measurements were accurate to about half a degree of arc. But a whole new level of accuracy was achieved by Tycho Brahe at his Danish castle of Uraniborg. A huge mural quadrant allowed Tycho to make naked eye observations to an accuracy of one arcminute -- a sixtieth of a degree! The resulting 1,005-star catalogue was published in table form in 1598, and in atlas form in Johann Bayer's Uranometria in 1603.

Tycho's accurate measurement of planetary positions allowed Johannes Kepler to deduce the elliptical orbits of the planets about the Sun, another step up in our knowledge of how the heavens moved. We now knew that the planets had their own independent motions in three-dimensional space. Hipparchus had to invent his own spherical geometry to deal with stellar and planetary positions, but at least he only had to reckon with stars that were glued to the surface of a celestial sphere. All change was angular. There was no concept of distances to stars. (He did compute a distance to the Moon by two different methods, but the Sun was too far distant for his techniques and accuracy).

So how can we know anything about the distances to stars, let alone about their true motions in space? Actually, the second question is easier to deal with than the first. Observe the following:



To us earth-bound humans, and even to our space-based telescopes, the sky looks just like it did to Hipparchus. It appears to be a sphere surrounding us with the stars painted on its surface. If we look and measure carefully, over time we see that some stars change their angular position on the celestial sphere. This movement is called a star's proper motion. Now we all remember our primary school trigonometry, and its grandly named secondary school successor, vector calculus. We know that any given motion vector can be decomposed into two orthogonal components:



What we need to know is the radial velocity of a star along our line of sight, and its transverse velocity perpendicular to our line of sight. From these two we can calculate its true space motion from the simple Pythagorean identity:



Unfortunately we know neither of these directly. All we know is the star's angular proper motion. If we knew the distance to the star we could use trigonometry to convert the proper motion to a transverse velocity. We'll come to that later.

Measurement of the other component, the radial velocity, went from strictly impossible to surprisingly easy over the course of the nineteenth century. In 1842 Christian Doppler proposed his eponymously named effect, that waves emitted from a moving source would appear shifted in wavelength to a stationary observer. A sound source would be altered in pitch, whereas a light source would be altered in colour. Doppler himself tried to explain the colours of binary stars in this way. At around the same time, in the first half of the 19th century Fraunhofer and others were making giant leaps in spectroscopy. Using an advanced version of Newton's prism they split light into constituent colours and observed characteristic dark and bright lines and bands in the spectrum. These seemed to be associated with particular materials in the light source. The explanation for spectral emission and absorption lines had to wait for quantum mechanics in the 20th century, but the upshot was that certain elements produced lines of extremely precise wavelength in their spectra. These lines could be observed in slightly altered positions in the spectra of stars, and a Doppler shift could be used to derive a radial velocity.

Thanks to quantum physics and the transparency of space, Doppler motions for stars can be monumentally accurate. We can measure radial motions down to walking speed! It's so accurate that the limitations on radial motion measurements of stars are down to the random boiling and churning of their hot surfaces toward us. And it is not limited by distance. We can measure the speed of supernova outbursts and whirling black hole accretion discs from half way across the universe, at distances of billion of light years.

If only the same were true of transverse motions! Let's get back to an earlier problem -- calculating the distance to a star using parallax. This is very much like observing the trees on the side of the road change their apparent position against a distant background mountain while you drive. The "driving" in this case is going to be done by the Earth in its annual motion around the Sun. Over the course of six months we will move from one end of a 200 million mile baseline to the other:



We measure the angular movement of the star against background stars which are assumed to be so far away as to appear stationary. Now we have a triangle formed by the Earth's orbital baseline and the angle subtended by the star. Simple trigonometry gives us the distance to the star.

Wait, though! This distance calculation depended on the angular movement of the star. If the star has its own independent space motion, won't there be a separate proper motion that throws things off? Yes and no. In the above picture, the star's direction is perpendicular to our line of sight to the Sun and it will trace out a circular motion against the background stars over the course of a year. More usually, the star is in some random direction and traces out an ellipse. Either way, we know what exact shape to expect and any deviation can be deemed to be a separate proper motion.

So now we've got the parallax movement of the star. By triangulation we can calculate its distance. Separately we can extract out the proper motion of the star. Using the distance we calculated we can turn this into a transverse velocity. Spectrographic measurements give us the radial velocity. We combine the two to give the space motion. And now we know the precise position and velocity of a star within our galaxy. Eureka!

But now we are limited by the pointing accuracy of our electromechanical systems. We've come a long way since Hipparchus, but are still severely limited. The problem is the vast distances to stars. Any triangulation is going to involve very tiny angles. Consider the distance unit of the parsec. The very name is derived from the parallax method. It is the distance that subtends a parallax angle of one arcsecond on the Earth's orbital baseline. One arcsecond is sixty times the measurement accuracy achieved by Tycho. And one parsec is about 3.2 light years -- not even the distance to the nearest star!

So here we get to the insane level of pointing accuracy of modern instruments, especially the dedicated astrometry space missions Hipparcos (guess the name origin!) and Gaia. Gaia orbits the Sun in a Lissajous orbit around the Earth-Sun L2 Lagrange point. It is always pointing out into space away from the Sun, so its solar array serves a double function -- as a generator of 2 kW of power for the mission operations and as a sunshield to keep the spacecraft completely thermally stable. The entire spacecraft frame is made of mechanically and thermally stable silicon carbide. Whereas conventional spacecraft use gyroscopes for attitude control, Gaia uses them only for gross movements and uses a low disturbance cold gas micro-propulsion for fine attitude control.

Gaia has two telescopes pointing almost at right angles to each other (actually 106.5 degrees, continuously calibrated by onboard equipment to within half a millionth of an arcsecond!). Its camera has 106 CCD chips delivering an amazing billion pixel image. Gaia slowly spins at an ultra-precise rate, continuously reading out the scan lines from the CCDs. The time between when a star passes through the field of view of one telescope and then the other is measured by an onboard atomic clock. The time, combined with the spin rate, can be converted into a highly accurate angular position. Separate instruments perform photometry and medium resolution spectrometry for radial velocity measurements. This simple video gives a great idea of how Gaia scans the sky as it orbits the Sun, followed by a more detailed on about the continuous scanning technique:

Parallax measurements from Earth have traditionally been limited to stellar distances of just dozens of light years. By achieving an accuracy of 24 microarcseconds, Gaia calculates parallax distances to 10% accuracy at the distance of the galactic centre, 30 thousand light years away. But a positional map of a billion stars in our galaxy is only the start. With information on true motions of the stars we can see how the stellar populations get mixed over time, and backtrack to calculate where the stars were born. Combined with photometric and spectroscopic information which tells us about the ages and chemical compositions of the stars, we can start to figure out how our galaxy evolved over its ten billion year lifetime.

But Gaia isn't fussy about the objects it captures either. Some will be within our own solar system, so we can expect a huge new database of asteroids and Kuiper belt objects. And some will be beyond the confines of our galaxy, giving us a new quasar database and in particular a set of gravitationally lensed quasars which will tie down the expansion rate and other parameters of the entire cosmos.

Let's finish with the cute Gaia logo on the fairing of the spacecraft when it launched at Christmas 2013;

Maximus Alexander

Excellent post, thank you.

ps200306

I flung together a table of angular resolutions achieved by sky surveys in the last two thousand years -- any mistakes are mine*. It's sobering to look at the rate of improvement. In the seventeen centuries from Hipparchus to Tycho it improved by a factor of thirty. In the following four centuries to Hubble it improved by a factor of six hundred. In the quarter century from Hubble to Gaia, it's gone up by a factor of four thousand!





(*I'm intentionally blurring the distinction between angular resolution and positional accuracy for effect :pac: )

Capt'n Midnight

Tycho was still using the Mark I eyeball.

Gaia's 24 microarcseconds

http://sci.esa.int/gaia/58135-gaia-s-second-anniversary-marked-by-successes-and-challenges/

A microarcsecond, which is one millionth of an arcsecond, is equivalent to the edge of a euro coin on the Moon as seen from Earth

Capt'n Midnight

http://www.esa.int/Our_Activities/Space_Science/Gaia/Gaia_creates_richest_star_map_of_our_Galaxy_and_beyond


Title Gaia’s Hertzsprung-Russell diagram