Gaia Sky - Interactive Sky Map based on ESA's Gaia mission

[funk]

There was a post in 2016 about ESA's Gaia mission in this forum. Because it's kinda old and this is more about the results, I started a new one.

For all of you who are interested in astronomy there is a very interesting application called "Gaia Sky" of the Center for Astronomy of University Heidelberg, which charts all the objects and their parameters measured by Gaia.

You can find more informationen and download it on their homepage.

The Gaia mission is still running and the next data release (DR2) with even more accurate data is expected in April 2018.

 

 

Edited March 8, 2018 by funk

[YNM]

Do you happen to work somewhere on it ? I wonder, does it really just sends some data down instead of pictures ?

[Green Baron]

Gaia is an astrometry mission, to measure positions and radial velocities with higher precision than before.

http://sci.esa.int/gaia/

Here's the instruments: http://sci.esa.int/gaia/47354-fact-sheet/

tl, dr: it produces images with 2 identical telescopes and an array of ccds. I am sure the data evaluation is done on earth.

Edited March 8, 2018 by Green Baron

[YNM]

3 hours ago, Green Baron said:

I am sure the data evaluation is done on earth.

Hmm... I remember back then that it's said that the whole thing is done almost "on it's own", ie. it doesn't send back any images down apart from during testing and confirmation period. Which is why I'm asking. Here's an excerpt off an abstract on Gaia's software :

Quote

We present an algorithm configuration for the Gaia on-board autonomous object observation system that makes it possible to observe very bright stars with G = [2.0-6.0). Its performance has been tested during the in-orbit commissioning phase achieving an observation completeness of ~94% at G = 3 – 5.7 and ~75% at G = 2 – 3. Furthermore, two targeted observation techniques for data acquisition of stars brighter than G=2.0 were tested.

And also, if it really just sends down raw images a la Hipparcos, is there any way to access them ? Who knows you could rake in other stuff from it...

Edited March 8, 2018 by YNM

[Starman4308]

Wikipedia suggests it sends back some of the raw data. The total output is on the order of Gbit/sec, whereas its max downlink rate is 5 Mbit/sec. However, since most of its focal plane is just empty space, it apparently just sends a few pixels around each object, presumably plus some information on where that was in the focal plane.

[Green Baron]

3 hours ago, YNM said:

 is there any way to access them ?

Single unprocessed astronomy photos are astoundingly uninteresting :-) Blurry white or light gray spots on a more or less gray background.

esa.int has a a lot of information on Hipparcos.

Information you might already have:

A book on the data evaluation.

The protocol is online http://emits.sso.esa.int/emits-doc/ASTRIUMLIM/GAIA_TTCSCOE/GAIA-ESC-ICD-00515-SGICD.pdf

And the data processing people (Gaia DPAC), but the page runs into an endless loop on my pc here ...

 

That doesn't really answer your question positively :-) ....

[PB666]

15 minutes ago, Green Baron said:

Single unprocessed astronomy photos are astoundingly uninteresting :-) Blurry white or light gray spots on a more or less gray background.

esa.int has a a lot of information on Hipparcos.

Information you might already have:

A book on the data evaluation.

The protocol is online http://emits.sso.esa.int/emits-doc/ASTRIUMLIM/GAIA_TTCSCOE/GAIA-ESC-ICD-00515-SGICD.pdf

And the data processing people (Gaia DPAC), but the page runs into an endless loop on my pc here ...

 

That doesn't really answer your question positively :-) ....

But it could not possibly compile a list of all 13,000 asteroid belt objects if one the one hand its mission was to detect the motion of moving objects but at the same time plot X,Y,Z positions. The data it is sending back are x,P_x and if the program is at all working its got a psuedo ellipse and a time reference to some clock like J2000. To know the relative position of an object based on 'the suns time' when you cannot simultaneously see all those objects means that you have a Newtonian understanding of the motion of those objects with respect to time. (X, Y, Z, T)

 

[Starman4308]

45 minutes ago, PB666 said:

But it could not possibly compile a list of all 13,000 asteroid belt objects if one the one hand its mission was to detect the motion of moving objects but at the same time plot X,Y,Z positions. The data it is sending back are x,P_x and if the program is at all working its got a psuedo ellipse and a time reference to some clock like J2000. To know the relative position of an object based on 'the suns time' when you cannot simultaneously see all those objects means that you have a Newtonian understanding of the motion of those objects with respect to time. (X, Y, Z, T)

 

Presumably each set of observed images is timestamped. One T works for an entire batch of X,Y,I observations (X,Y in the plane, plus an intensity).

[Green Baron]

1 hour ago, PB666 said:

... means that you have a Newtonian understanding of the motion of those objects with respect to time. (X, Y, Z, T)

You have absolutely no idea about my understanding.

[funk]

Finally I'm able to answer. The forum was bugged out for me. First: I'm not an expert I can just repeat what a guy involved in this mission presented on his slides:

Gaia has a 1GPxl camera and the downstream to 3 ground stations (Spain, Australia, Argentinia) is between 3 to 8 Mbit/s at approx. 6 hours a day. From 2014 until Nov 2017 46.6TB of data has been transferred.

There are several tricks to reduce the amount of data that needs to be transferred. Gaia is rotating so the sky is rotating over the focal plane. The onboard cpu detects the objects of both telescopes first and determines where and when they will be pictured on the focal plane. While most of the sky is black, only a square around each star is cut out and stored/transfered (with timestamp). If there are many stars, e.g. when looking at the starry line of the milkyway, storaging is needed and it will be transfered later.

Beyond that it's very interesting what needs to be considered to finally get the correct results. The most obvious is light bending around the sun, Jupiter or other planets. Or the incoming scattered light.

To reach a significant result for the movement, each star is crossing the focal plane up to several hundred times over the years. Every crossing takes about 4.5s on the focal plane. Because a star seems to have an elliptic movement over time it needs many crossings to calculate a significant mean value. Further from each star meeting certain requirements (brightness) Gaia takes a spectra.

If you understand german I can give you a link to the presentation.

Edited March 8, 2018 by funk

[Streetwind]

Gaia is one of my favorite missions-that-no-one-talks-about. I mean, if the video linked in the OP doesn't blow your mind when you realize that every dot is a real star moving according to real, measured data... then how about this:

The huge amount of data that Gaia downlinks on a daily basis, mentioned in the posts above, requires an enormous processing effort. An entire organization exists simply to handle this task. In total, 450 people are working with a world-class supercomputer to process Gaia's raw output into a usable form - and 180 of them are full-time employees that literally do nothing else all week, every week. It is such a complex, gargantuan effort that ESA had to revise their initial schedule of data releases from once per year, to once per two years.

This is why even after and and a half years, everyone is still working with Data Release 1 right now, which was released in September 2016. Everything that came out of Gaia so far, including this Gaia Sky application, is based on that set, and no updates have been released since then. Even so, the Gaia mission manager at ESA said last month that the scientific community is currently producing “something like one paper every day and a half”.

That's a lot of papers.

And now consider the following: Data Release 1 contains only basic measurements for the vast majority of stars, because most of them simply needed more repeat observations to more precisely determine their characteristics than were available at the time. Among 1.14 billion stars in the first dataset, only two million have what ESA calls "advanced measurements", which includes data on movement and such.

In other words: you can see "only" two million stars in that video (and the Gaia Sky application in general). And the scientific community is churning out twenty papers a month mostly just studying those two million stars.

When Data Release 2 becomes available in April, it is expected to finally have "advanced measurements" for over one billion stars. A roughly five hundred fold increase.

 

I do hope that by now, your mind has been sufficiently blown about this mission.

[funk]

To emphasize @Streetwind's statements:

[Green Baron]

3 hours ago, funk said:

If you understand german I can give you a link to the presentation.

Bring it on pls, we can always translate :-)

[YNM]

7 hours ago, PB666 said:

But it could not possibly compile a list of all 13,000 asteroid belt objects if one the one hand its mission was to detect the motion of moving objects but at the same time plot X,Y,Z positions.

That's another question... But given it's limited to G magnitudes (apparently means "green" !), perhaps there's a caveat that reflected light won't be visible at all ? Also, it wouldn't appear to wobble after half an orbit, they'd just disappear off...

3 hours ago, funk said:

The onboard cpu detects the objects of both telescopes first and determines where and when they will be pictured on the focal plane. While most of the sky is black, only a square around each star is cut out and stored/transfered (with timestamp).

Nah, this is the amazing part...

3 hours ago, funk said:

If you understand german I can give you a link to the presentation.

I hope Google Translate is sufficient...

Great info nonetheless ! I was always fascinated with survey obsrvations when I was "in" on the subject. Our only observatorium so far was originally used to survey southern sky's double stars, from astrometric- to spectrographic-type. Then I remembered the optic-cable equipped spectrograph (I forgot which one, was it Alice Springs ?). Then when news of Gaia comes, I'm intrigued by the automatic detection...

Edited March 8, 2018 by YNM

[funk]

 

The first part is more or less about some basics of the Gaia, instruments, launch, orbit etc. @35min the part about measuring starts and links to the second part, which shows basic illustrations for the calculations, error deviation, some simulations and results and finally questions from the audience.

The german subtitle (auto-generated) is mostly accurate. Don't know if it's possible to extract it somehow. At any point I'm willing to help to translate if there are questions.

 

 

Edited March 9, 2018 by funk