I did a practice run using the calibrated WASP12b data series, everything seems to work OK, as described in the user guide, however one thing I presently do not understand :
How AIJ keeps track of the target/comparison stars for all the images ? through the pixel coordinates ? through the WCS coordinates if plate-solved images are used? does it recalculates the centroid of the stars for each new image before performing photometry ? does it allow a small translation and/or field rotation between two successive images ? if yes, of which amount ?
Thank you for your attention
That is all under user control on the Multi-Aperture setup panel. If you want to use WCS coordinates to place apertures, select the option "Use RA/Dec to locate aperture positions". If that option is disabled, pixel coordinates are used to place the apertures
from one image to the next. If the option "Reposition aperture to object centroid" is enabled, then after the initial aperture placement by RA/Dec or pixels coordinates (depending on the Use RA/Dec setting), a centroid operation will be performed to tweak
the placement of the aperture to the centroid of the flux within the aperture radius (if the flux is very low, centroid may fail). Centroid will track target and comparison stars (translation or rotation) as long as they are within one aperture radius of the
initially placed position. In pixel placement mode, the pixel coordinates update from image to image, so as long as the image movement from one image to the next is less than the aperture radius, the stars will be tracked. If the images are plate solved and
the "Use RA/Dec" option is enabled, the stars will be tracked regardless of the the amount of field shift or rotation from image to image, as long as the plate solution is correct within one aperture radius.
Thanks a lot for your answer ! If I understand well, when the option "Reposition aperture to object centroid" is enabled, the tolerance for field shift or rotation from one image to the next is exactly equal to one aperture radius : so if the user is concerned with image movements higher than that, he may choose to slightly increase the "Radius of object aperture" parameter, and/or to defocus if the object is bright enough.
Yes, increasing the radius would allow for more movement, or you could use the alignment feature to align the images using larger apertures, and then run photometry using the size aperture you need to optimize photometric precision.
If all the images are plate solved, use the RA/Dec aperture placement option and you will not have to worry about how far the images move.
A third option is to use the variable size aperture. With that method, you can specify a larger max aperture size near the top (at the normal aperture radius setting control). Then set the variable aperture scaling factor such that you get the size of aperture
that will optimize your photometry. In that mode, the aperture size used for photometry is the multiplication factor size that you set times the average FWHM of the stars in the image. Usually a value between 1.0 and 2.0 gives good results. I'd start around
1.5 and go from there to see what gives you the cleanest light curve.
Defocusing in general will improve photometric accuracy (to a certain extend), unless there are other stars that start blending into the aperture of your target star or one or more of the comp star apertures.