From erik Fri Jan  8 08:08:21 1999
Subject: CUO_52.txt On-board photometry
To: gaia-sag@astro.estec.esa.nl (GAIA SAG), Frederic.Arenou@obspm.fr (Arenou),
	indus (Jens Knude), helge (Helge J. Sorensen), cf (Claus Fabricius),
	makarov (Valeri Makarov), smid (Mattia Vaccari)
Date: Fri, 8 Jan 1999 08:08:21 +0100 (MET)


On-board photometry is not recommended for GAIA
===============================================

E. Hoeg          8 Jan. 1999        SAG_CUO_052 


ABSTRACT:
The idea of on-board photometry for GAIA is discussed. The idea 
is rejected because of several disadvantages. Essential advantages 
have not been presented since the idea was put forward three months ago.



The idea
--------
The idea of on-board photometry for GAIA was proposed by Fabio 
for consideration by SAG at the October 1998 meeting and in a 
report shortly after. The idea was not specified or discussed, 
presumably to leave all possibilities open. Requests for a 
specification have not been answered, no doubt because of more 
urgent business. A discussion was again encouraged by Fabio 
in his report of Dec.2, 1998.

The idea has recently been mentioned in a study of the APM algorithm by
Jordi et al. in UB_IWG_001 of 14 Dec. 1998. I intend to comment on
the report when I receive the information I have asked Xavier to 
give for my better understanding.


Definition of on-board photometry
---------------------------------
For the sake of a clear discussion I define the idea as follows.

Broad-band photometry (BBP) with the APM algorithm onboard the GAIA
satellite has been proposed. The proposal is meant as
an alternative to the transmission of samples of patches as
described in CUO_50.

The idea involves detection and estimation of stellar
magnitude and background in each of the four BBP bands. Only the
resulting position, magnitude and background for each star in each
band are transmitted to ground where the cross identification of
objects detected in the ASM and BBP1-4 is carried out.

This purpose is not explicitely described in the UB report or elsewhere,
but this is my understanding from what I have heard and read.


Discussion
----------
Essential advantages of on-board photometry are not known.

Since we require a final GAIA photometry with the optimal accuracy 
inherent in the CCD samples the following essential disadvantages of
on-board photometry can be listed.

1. Algorithms and iterations:  
Since the original samples are not transmitted to ground it is not
possible to redo the photometry lateron with improved algorithms.

2. Calibration:
An optimal photometric analysis is not possible on board the satellite,
especially during the first part of the mission when a good knowledge
of the PSFs, the CCD calibrations, and satellite attitude is not yet
available.

3. Double stars:
An optimal reduction of double and multiple stars is not possible.

4. Use of astrometric information:
It is not possible to make use of astrometry from other CCDs, which is
required in order to achieve the highest accuracy.

5. Faint stars:
It is not possible to make an optimal analysis of faint stars since a
stacking of observations from many transits is required to approach the
Cramer-Rao limit.

Each of these arguments seems strong enough to reject the idea of
on-board photometric analysis for the final GAIA photometry. 
Finally, these arguments do not require supporting simulations or 
any numbers.