Author Topic: ESA - Gaia updates  (Read 123681 times)

Offline Star One

  • Senior Member
  • *****
  • Posts: 13997
  • UK
  • Liked: 3974
  • Likes Given: 220
Re: ESA - Gaia updates
« Reply #80 on: 09/15/2016 08:46 am »
Are systematic errors those errors introduced into the results by Gaia itself?

No, the parallax error listed in the database doesn't include the systematic error. This is said clearly in, for instance, the main paper presenting the data release (near the beginning of section 3):

Quote
The typical uncertainty for the parallaxes is 0.3 mas, where it should be noted that a systematic component of 0.3 mas should be added (see Sect. 6).

I've seen a number of people online criticise ESA for the failure to get across clearly matters such as this and that their public outreach leaves a lot to be desired.

Offline as58

  • Full Member
  • ****
  • Posts: 835
  • Liked: 300
  • Likes Given: 186
Re: ESA - Gaia updates
« Reply #81 on: 09/15/2016 09:43 am »
I've seen a number of people online criticise ESA for the failure to get across clearly matters such as this and that their public outreach leaves a lot to be desired.

I don't know about public outreach, but I don't know how they could've been more clear about the systematics. The data release page even has this sentence bolded:

Quote
The recommendation is to consider the quoted uncertainties on the parallaxes as ±σϖ (random) ±0.3 mas (systematic). Furthermore, averaging parallaxes over small regions of the sky will not reduce the uncertainty on the mean below the 0.3 mas level.

It seems that a lot of people just rushed into looking at data without bothering to read any of the release notes.

Offline Star One

  • Senior Member
  • *****
  • Posts: 13997
  • UK
  • Liked: 3974
  • Likes Given: 220
Re: ESA - Gaia updates
« Reply #82 on: 09/15/2016 09:56 am »
I've seen a number of people online criticise ESA for the failure to get across clearly matters such as this and that their public outreach leaves a lot to be desired.

I don't know about public outreach, but I don't know how they could've been more clear about the systematics. The data release page even has this sentence bolded:

Quote
The recommendation is to consider the quoted uncertainties on the parallaxes as ±σϖ (random) ±0.3 mas (systematic). Furthermore, averaging parallaxes over small regions of the sky will not reduce the uncertainty on the mean below the 0.3 mas level.

It seems that a lot of people just rushed into looking at data without bothering to read any of the release notes.

I mean getting it across to lay people & amateur astronomers or did they think only professional astronomers were going to look at this. At least NASA tries to cater its information for the general populace, something that ESA seems particular poor at.

Offline Star One

  • Senior Member
  • *****
  • Posts: 13997
  • UK
  • Liked: 3974
  • Likes Given: 220
Re: ESA - Gaia updates
« Reply #83 on: 09/15/2016 05:34 pm »
Further article on Gaia.

Interesting comment underneath that if it keeps going for ten years it could discover 70,000 gas giants alone.

http://www.centauri-dreams.org/?p=36391

Offline as58

  • Full Member
  • ****
  • Posts: 835
  • Liked: 300
  • Likes Given: 186
Re: ESA - Gaia updates
« Reply #84 on: 09/15/2016 06:19 pm »
Further article on Gaia.

Interesting comment underneath that if it keeps going for ten years it could discover 70,000 gas giants alone.

http://www.centauri-dreams.org/?p=36391

This article goes into detail on prospects of detecting exoplanets with Gaia.

 http://arxiv.org/abs/1411.1173

Abstract:
Quote


    We provide a revised assessment of the number of exoplanets that should be discovered by Gaia astrometry, extending previous studies to a broader range of spectral types, distances, and magnitudes. Our assessment is based on a large representative sample of host stars from the TRILEGAL Galaxy population synthesis model, recent estimates of the exoplanet frequency distributions as a function of stellar type, and detailed simulation of the Gaia observations using the updated instrument performance and scanning law. We use two approaches to estimate detectable planetary systems: one based on the S/N of the astrometric signature per field crossing, easily reproducible and allowing comparisons with previous estimates, and a new and more robust metric based on orbit fitting to the simulated satellite data.
    With some plausible assumptions on planet occurrences, we find that some 21,000 (+/-6000) high-mass (1-15M_J) long-period planets should be discovered out to distances of ~500pc for the nominal 5-yr mission (including at least 1000-1500 around M dwarfs out to 100pc), rising to some 70,000 (+/-20,000) for a 10-yr mission. We indicate some of the expected features of this exoplanet population, amongst them ~25-50 intermediate-period (P~2-3yr) transiting systems.

Offline Bubbinski

Re: ESA - Gaia updates
« Reply #85 on: 09/15/2016 07:18 pm »
Congratulations to the Gaia craft and team!

Wondering if Gaia data can confirm if Proxima b transits its star as seen from our solar system's vantage point.

Also wondering if exoplanets (large gas giants of course) will be found in the Magellanic Cloud dwarf galaxies as a result of this data. (Extragalactic exoplanets!)

Edit: I looked up the distance to the LMC after reviewing the post above mine. LMC is 50 kilo parsecs away and the arXiv article says Gaia can discover planets up to 500 parsecs away (0.5 kilo parsecs). So the answer would appear to be "no" for question 2 (no extragalactic exoplanets from Gaia)

And can this data be used to find ring systems or exomoons?
« Last Edit: 09/15/2016 07:31 pm by Bubbinski »
I'll even excitedly look forward to "flags and footprints" and suborbital missions. Just fly...somewhere.

Offline Star One

  • Senior Member
  • *****
  • Posts: 13997
  • UK
  • Liked: 3974
  • Likes Given: 220
ESA - Gaia updates
« Reply #86 on: 09/15/2016 07:55 pm »
Congratulations to the Gaia craft and team!

Wondering if Gaia data can confirm if Proxima b transits its star as seen from our solar system's vantage point.

Also wondering if exoplanets (large gas giants of course) will be found in the Magellanic Cloud dwarf galaxies as a result of this data. (Extragalactic exoplanets!)

Edit: I looked up the distance to the LMC after reviewing the post above mine. LMC is 50 kilo parsecs away and the arXiv article says Gaia can discover planets up to 500 parsecs away (0.5 kilo parsecs). So the answer would appear to be "no" for question 2 (no extragalactic exoplanets from Gaia)

And can this data be used to find ring systems or exomoons?


Hmmm you're interesting in transits of Proxima b, have a look on the Pale Red Dot Thread.
« Last Edit: 09/15/2016 07:57 pm by Star One »

Offline denis

  • Full Member
  • *
  • Posts: 124
  • EU
  • Liked: 24
  • Likes Given: 9
Re: ESA - Gaia updates
« Reply #87 on: 09/15/2016 08:41 pm »
Congratulations to the Gaia craft and team!
Thanks !  :P

Offline Hungry4info3

  • Full Member
  • ****
  • Posts: 525
  • Liked: 163
  • Likes Given: 81
Re: ESA - Gaia updates
« Reply #88 on: 09/16/2016 12:53 am »
And can this data be used to find ring systems or exomoons?

No. Gaia is using astrometry. It is sensitive to the barycentric motion of the star. It may also be sensitive to transits, but there's a lot of varying estimates on how many we should expect to detect.

Transit searches tend to be rather high-cadence, taking thousands of measurements of a star at a time. Gaia will take 70 measurements per star on average. For a short-period planet, this means three or four measurements may occur during a transit. So we're talking about, statistically, very poor quality detections even of hot Jupiters.

Dzigan and Zucker estimate hundreds to thousands of transiting exoplanets from Gaia photometry.
http://arxiv.org/abs/1205.4725

There is an analogy of this from Hipparcos, where a transit of HD 209458 b was found (with 89 photometric measurements) after the planet's RV-based discovery.
http://adsabs.harvard.edu/abs/2000A&A...355..295R

With the photometric detection of planets restricted to very short-period planets, rings and exomoons are ruled out because they aren't tidally stable (even if, somehow, they were detectable in such sparse data sets).

Offline Star One

  • Senior Member
  • *****
  • Posts: 13997
  • UK
  • Liked: 3974
  • Likes Given: 220
ESA - Gaia updates
« Reply #89 on: 09/20/2016 06:57 am »
Evidence for a systematic offset of −0.25~mas in the Gaia DR1 parallaxes

Quote
We test the parallaxes reported in the Gaia first data release using the sample of eclipsing binaries with accurate, empirical distances from Stassun & Torres (2016). We find a clear average offset of −0.25±0.05 mas in the sense of the Gaia parallaxes being too small (i.e., the distances too long). The documented Gaia systematic uncertainty is 0.3 mas, which the eclipsing binary sample corroborates. The offset does not depend strongly on obvious parameters such as color, brightness, or spatial position. However, with a statistical significance of 99.7%, nearer stars possibly exhibit larger offsets according to Δπ≈−0.16−0.02×π mas.

http://arxiv.org/abs/1609.05390
« Last Edit: 09/20/2016 06:58 am by Star One »

Offline denis

  • Full Member
  • *
  • Posts: 124
  • EU
  • Liked: 24
  • Likes Given: 9
Re: ESA - Gaia updates
« Reply #90 on: 09/26/2016 07:50 pm »
There is a search page, the use of which is presumably self-evident to professional astronomers etc, but which is far from clear to me, and possibly most laymen. There's no how-to-use guide available, for instance. Or at least I couldn't find one! Given BBC news showed some schoolchildren who found a supernova, it's presumably not that difficult to use once you know how. Was there some material given to teachers by ESA, or did this school just happen to have contact with somehow who's familiar with such database search engines? A local university outreach perhaps?

Concerning finding supernovae, I think you are confusing with the Gaia alert page, which generates "real time" (daily basis?) alerts for objects suddenly brighter than previously measured, such as to trigger further observations and classification from ground. This is not based on the catalogue release but on this page: http://gsaweb.ast.cam.ac.uk/alerts/home


Concerning distances, from reading various tweets it seems computing distance from parallax is a complex problem and simply taking the inverse of the parallax is good only if the parallax error is small (< 1% I saw somewhere). I assume this is why DPAC provides parallax measurements and not directly distance, such that various science groups can estimate distances based on different techniques.
For example, one article concerning converting Gaia's DR1 parallaxes into distances:
https://arxiv.org/pdf/1609.07369v1.pdf

Quote
We infer distances and their asymmetric uncertainties for two million stars using the parallaxes
published in the Gaia DR1 (GDR1) catalogue.
...
except to remind readers that inverting parallaxes to estimate distances is only appropriate in the absence of
noise. As parallax measurements have uncertainties— and for many TGAS stars very large uncertainties—
distance estimation should always be treated as an inference problem.

« Last Edit: 09/26/2016 07:51 pm by denis »

Offline Star One

  • Senior Member
  • *****
  • Posts: 13997
  • UK
  • Liked: 3974
  • Likes Given: 220
Re: ESA - Gaia updates
« Reply #91 on: 09/27/2016 01:05 pm »
TGAS Error Renormalization from the RR Lyrae Period-Luminosity Relation

Quote
The Gaia team has applied a renormalization to their internally-derived parallax errors σint(π)
σtgas(π)=[Aσint(π)]2+σ20−−−−−−−−−−−−−√;    (A,σ0)=(1.4,0.20 mas)
based on comparison to Hipparcos astrometry. We use a completely independent method based on the RR Lyrae K-band period-luminosity relation to derive a substantially different result, with smaller ultimate errors
(A,σ0)=(1.1,0.12 mas)    (this paper).
We argue that our estimate is likely to be more accurate and therefore that the reported TGAS parallax errors should be reduced according to the prescription:
σtrue(π)=(0.79σtgas(π))2−(0.10 mas)2−−−−−−−−−−−−−−−−−−−−−−−√.

http://arxiv.org/abs/1609.06315

Online jgoldader

  • Full Member
  • ****
  • Posts: 760
  • Liked: 322
  • Likes Given: 171
Re: ESA - Gaia updates
« Reply #92 on: 09/28/2016 05:57 pm »
Concerning distances, from reading various tweets it seems computing distance from parallax is a complex problem and simply taking the inverse of the parallax is good only if the parallax error is small (< 1% I saw somewhere). I assume this is why DPAC provides parallax measurements and not directly distance, such that various science groups can estimate distances based on different techniques.
For example, one article concerning converting Gaia's DR1 parallaxes into distances:
https://arxiv.org/pdf/1609.07369v1.pdf

Quote
We infer distances and their asymmetric uncertainties for two million stars using the parallaxes
published in the Gaia DR1 (GDR1) catalogue.
...
except to remind readers that inverting parallaxes to estimate distances is only appropriate in the absence of
noise. As parallax measurements have uncertainties— and for many TGAS stars very large uncertainties—
distance estimation should always be treated as an inference problem.



When Gaia measures the position of a star on the CCD, there's uncertainty in the (x,y) position due to the fitting of the centroid (with some weighting perhaps) of the PSF.  (For the moment, ignore the uncertainties in the conversion of those (x,y) positions to (RA, DEC) positions.) 

If not for proper motion, stars would just move "back and forth" on the sky due to parallax.  But all disk stars are also in motion through space, typically at ~225 km/s or so with random motions thrown in for fun, around the center of the Milky Way.  So if you measure a star from two positions in space, you'll get parallax, yes; but since the star will have moved in 3D space since your first measurement ("proper motion" is that motion projected into 2D onto the celestial sphere) the position differs not only due to parallax, but the star's own proper motion as well.  Proper motion makes the motion of the stars looks like a sort of slanted sinusoidal curve against the sky.  You must have several epochs of observations (at bare minimum several spread over one year, but two years is better) to be able to model the sine curve.  The higher the proper motion perpendicular to the parallax direction, the more "stretched out" the since curve looks (it looks like its wavelength increases); but proper motion near the direction of parallax makes the sine look like a Z that's squashed and stretched sideways.  The higher the parallax (i.e., the closer the star) the greater the amplitude of the sine curve.  (Oh, and some stars are binaries or have planets that make them have extra motions due to their companions, and...)

But in the end, you have a bunch of measurements of position, and from those you have to get the amount and direction of proper motion, plus the amount of parallax.  And the quality of your solutions for each of these depends on the quality of your position measurements.  So the uncertainty propagates through everything you do.  Hence, due to your position uncertainties (which include random---i.e., centroiding---errors, which depend in part on brightness and CCD characteristics, as well as systematics), your parallax and proper motion will also have random + systematic uncertainties. 

One of the many wonderful things about Gaia is that the teams will continue to improve the quality of the catalog as they get better at modeling systematics (e.g., thermal and outgassing effects) and accumulate more years of data.
Recovering astronomer

Offline bolun

  • Senior Member
  • *****
  • Posts: 3557
  • Europe
  • Liked: 970
  • Likes Given: 110
Re: ESA - Gaia updates
« Reply #93 on: 11/01/2016 08:33 pm »
Follow the Gaia 2016 data release #1 Workshop live

28 October 2016

On 2-4 November, the European Space Astronomy Centre near Madrid, Spain, will host the Gaia 2016 Data Release #1 Workshop. Many of the talks will be broadcast live.

Livestream: http://livestream.com/ESA/events/6544080

http://sci.esa.int/gaia/58552-follow-the-gaia-2016-data-release-1-workshop-live/

Offline Star One

  • Senior Member
  • *****
  • Posts: 13997
  • UK
  • Liked: 3974
  • Likes Given: 220
Re: ESA - Gaia updates
« Reply #94 on: 11/22/2016 08:19 pm »

Offline bolun

  • Senior Member
  • *****
  • Posts: 3557
  • Europe
  • Liked: 970
  • Likes Given: 110
Re: ESA - Gaia updates
« Reply #95 on: 01/27/2017 08:45 pm »
Gaia turns its eyes to asteroid hunting

24 January 2017

Whilst best known for its surveys of the stars and mapping the Milky Way in three dimensions, ESA's Gaia has many more strings to its bow. Among them, its contribution to our understanding of the asteroids that litter the Solar System. Now, for the first time, Gaia is not only providing information crucial to understanding known asteroids, it has also started to look for new ones, previously unknown to astronomers.

http://sci.esa.int/gaia/58706-gaia-turns-its-eyes-to-asteroid-hunting/

Offline as58

  • Full Member
  • ****
  • Posts: 835
  • Liked: 300
  • Likes Given: 186
Re: ESA - Gaia updates
« Reply #96 on: 02/08/2017 05:02 pm »
There have been some changes in the data release plan. Now the next release is expected in April 2018 instead of earlier plan of Q4 2017. On the plus side, it appears (if I remember correctly) that the release will contain some data that were previously not planned for DR2.

https://www.cosmos.esa.int/web/gaia/release

Offline denis

  • Full Member
  • *
  • Posts: 124
  • EU
  • Liked: 24
  • Likes Given: 9
Re: ESA - Gaia updates
« Reply #97 on: 02/08/2017 08:24 pm »

I find the following wording a bit strange:
Quote
The DPAC project office and ESA are working on the longer term data release schedule. The release planned at three years after the end of the nominal mission lifetime (called 'final release' below) will be maintained, while the number of releases between Gaia DR2 and the final release remains to be decided​.

Assuming no failure, the mission is likely to be extended until the spacecraft runs out of cold gas.
So does that mean they plan a 'final release' 3 years after end of nominal mission lifetime and what comes next is "extra" that will be part of a further release ?
Or is it just because they cannot formally assume there will be an extension ?

Offline as58

  • Full Member
  • ****
  • Posts: 835
  • Liked: 300
  • Likes Given: 186
Re: ESA - Gaia updates
« Reply #98 on: 02/08/2017 08:46 pm »
I think the 'final' release is some sort of mission success criterion. Though an extension is likely (assuming good spacecraft health etc.), they can't officially count on it until it's been formally approved.

Offline jebbo

  • Full Member
  • ****
  • Posts: 940
  • Cambridge, UK
  • Liked: 608
  • Likes Given: 309
Re: ESA - Gaia updates
« Reply #99 on: 02/09/2017 11:38 am »

Tags: gaia 
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
1