8 replies to this topic

[han.k]

The Gaia satellite has images about one billion stars in Gaia-G,  red and blue magnitudes. From the latest Gaia DR2 release it is possible to calculate the Johnson-V magnitude and use for photometry.

 

Below an comparison between  the magnitudes in AASVO charts and measured magnitudes of an image taken with a ASI1600 camera  (only an UV-IR block filter).
The magnitude values in the image are measured by ASTAP using the Gaia Johnson-V star database.. The program can do an astrometric (plate) solving and calibrate the measured star flux using a few hundred Gaia stars found in the image. Below two results:

 

Comparison AASVO chart vs a measured image for M5:

 

Comparison AASVO chart vs a measured image for M31:

 

Are the AAVSO reference stars still required after Gaia DR2? I noted that the AAVSO has added the reference "48 Gaia DR2" for some stars,  but are the Gaia satellite measured values not more accurate then the old ground based observations?

 

-- Han

 

[happylimpet]

Presumably the AAVSO comparison stars are 'approved' in that they are known to be not variable - this may not be the case with all Gaia data which is presumably a snapshot or two. Using Gaia data for AAVSO comparison stars should be fine. The Gaia data will have uncertainties if you refer to the actual source data,

 

I doubt Gaia has enough time history to establish constancy - and anyway, all stars vary a bit, so I would stick with the AAVSO stars for now,

[han.k]

It is not a ,,snaphot or two''. If you look in the Gaia documentation they have 51.7  billion records of measured data for 1.69 billion sources.

 

https://www.aanda.or...aa32756-18.html

 

 

An unprecedented accuracy:

https://gea.esac.esa...mentation/GDR2/

 

They have already identified 550737 variable stars.

 

It will only get better with the next and final releases.

[happylimpet]

It is not a ,,snaphot or two''. If you look in the Gaia documentation they have 51.7  billion records of measured data for 1.69 billion sources.

 

gaia records.png

https://www.aanda.or...aa32756-18.html

 

 

An unprecedented accuracy:

gaia accuracy.png

https://gea.esac.esa...mentation/GDR2/

 

They have already identified 550737 variable stars.

 

It will only get better with the next and final releases.

 

 

So about 50 measures for each star. Still arguably not enough to establish constancy. I would only use the AAVSO 'approved' stars, but may be ineteresting/useful to compare their mags with the Gaia ones! 

 

Also, if you're submitting to the AAVSO, bear in mind that other measures they receive will be in relation to their values for the comparison stars, so by using other values, even if actually more accurate, you will introduce a systematic error.

Edited by happylimpet, 31 May 2019 - 08:36 AM.

[Aquarellia]

I fully agree with the fact that the AAVSO magn. references are the only one to be used.

The variation period for a lot of stars are much more long that the Gaïa database history.

Michel

[han.k]

So about 50 measures for each star. Still arguably not enough to establish constancy. I would only use the AAVSO 'approved' stars, but may be ineteresting/useful to compare their mags with the Gaia ones! 

 

Also, if you're submitting to the AAVSO, bear in mind that other measures they receive will be in relation to their values for the comparison stars, so by using other values, even if actually more accurate, you will introduce a systematic error.

That systematic error makes sense. However Gaia keeps on collecting date for at least 5 years. So somewhere in the future it would make sense to use that data.  An other advantage is that it goes much deeper.

 

The formulas for conversion to Johnson-V are in appendix A of this document 

 

I fully agree with the fact that the AAVSO magn. references are the only one to be used.

The variation period for a lot of stars are much more long that the Gaïa database history.

Michel

Gaia DR2 is the result of 22 months observation. With 0.3 mmag accuracy variable stars will be flagged as variable much earlier then one full period. Most likely Gaia will detect variation in standard stars unknown to us.

[Ed Wiley]

The solution is easy, if you do not use AAVSO comp stars simply state which stars you used and any conversions made (e.g., Gaia to Johnson or Cousins) when submitting to the AAVSO database (if that is what you do). Then do a light curve and compare your results with others.No problem if they all line up as expected. I am not sure how tricky transforming might be with data that is converted, you might wish to ask if you transform. 

 

Ed

 

PS, The AAVSO comp team does a great job of vetting their comps. Further, they can get you comps quickly, One time I got a set in less than one hour.

https://www.aavso.or...ble-star-charts

Edited by Ed Wiley, 31 May 2019 - 01:12 PM.

[NJScope]

An important paper to put all this discussion into perspective is "Gaia broad band photometry" (Jordi et al 2010, A&A 523, A48).  Although ccd measurements made through the Gaia filter (G-passband= 400-1000nm) may be very precise, transforms to other photometric systems (Johnson-Cousins, SDSS etc) would produce uncertainty not unlike the conversion of clear filter data to V-filter magnitudes.  First it should be noted that the Gaia passbands (BP, G, and RP) are unique (Jordi et  al. 2010) so that transforms are needed should one desire conversion to other conventional  photometric  systems.  For  example,  when  color  excess (B–V) is known, Gaia G-magnitudes can be transformed to Johnson-Cousins V-mag according to the following expression:
G–V=a+b*(B–V) +c*(B–V)²+d*(B–V)³.  
The  appropriate  Johnson-Cousins  coefficients  (a-d)  can  be  found  in  Table  5.8  of  Documentation  Release  1.1  from  Gaia  Data  Release  2 (https://gea.esac.esa...mentation/GDR2/).  A recent example of this calculation for the HADS variable V460 And was published in JAAVSO (Alton and Stepien, 2019, 47).  So far whenever I (Alton) have performed similar Gaia-G to V-mag calculations on other variable stars there appears to be very good correlation with APASS magnitudes.  However, I would not at this time recommend substituting Gaia-derived magnitude values for other bandpasses that have been already determined with conventional photometric systems using established reference star magnitudes.

[Ed Wiley]

Another approach is to use APASS stars, they are in Johnson B and V and Sloan I. If you want to do photometry in Cousins Ic, then you might ask for comp stars from AAVSO as they convert Sloan I to Cousins Ic.

 

Ed