meridian flip process

[bryan martin]

would appreciate critique of a process I am using with MF
1. calibrate all images
2. split into before and after flip files
3. reduce data before flip, then load and reduce after flip files. keep AIJ open as it seems to add the data to the data table
4. enter appropriate info for the transit into the plot tables and fine tune the detrends
5. I invert x and y after the flip to create the same orientation and use the same comp star pattern

Am I introducing errors using this approach?

bryan

[karenacollins]

Hi Bryan,

I responded to this by email, but I see that the response did not log here. The contents are posted below.


Yes, this method should work, and it's great that you noticed AIJ appends to the measurements table.  

If your images are plate-solved (i.e. have a WCS solution in the fits header), you can run the full image set at once, if you enable place apertures by RA/Dec in multi-aperture setup. If your images are not already plate-solved, you can plate solve the first image in the sequence and the first image after the flip. Then use RA/Dec placement, but disable "Halt processing on WCS error." Then RA and Dec will be used to place apertures in images that are plate-solved, but processing will continue for images without WCS by retaining the x, y aperture positions from the previous image, and then centroiding (if enabled). This should result in running through the entire image set without needing separate runs or rotating images. But your current method should give identical results, if you prefer to avoid plate-solving.

Karen