rdnoise in zerocombine & darkcombine
Guest wrote on Aug 06, 2008
Hi,
I don't understand why for the zerocombine and darkcombine tasks of the noao.imred.ccdred package, the parameter rdnoise is set according to
(rdnoise= o.) ccdclip: CCD readnoise (electrons)
in the examples in 'A User's Guide to CCD Reductions with IRAF' , Philip Massey, 1997, and also by the users in the IRAF.net forum. I thought that the readnoise depends on the model of the CCD camera used.
Can someone enlighten me? Thank you.
Siti :)
I don't understand why for the zerocombine and darkcombine tasks of the noao.imred.ccdred package, the parameter rdnoise is set according to
(rdnoise= o.) ccdclip: CCD readnoise (electrons)
in the examples in 'A User's Guide to CCD Reductions with IRAF' , Philip Massey, 1997, and also by the users in the IRAF.net forum. I thought that the readnoise depends on the model of the CCD camera used.
Can someone enlighten me? Thank you.
Siti :)
Francisco Valdes wrote on Aug 06, 2008
Hello Siti,
A task has to have some default before a user actually uses it. The default in this case is zero. However, it is true that you are supposed to set this to the readout noise for your detector which is usually something provided by the instrument engineer. My other comment is that rdnoise, gain, and snoise are only used for certain of the rejection algorithms, such as ccdclip, during combining. Some that are more empirical, such as reject="avsigclip", don't need these values and, instead, estimate the noise levels from the data itself. So if you don't have the values or they are uncertain you should avsigclip instead of ccdclip for rejection of pixels as statistical outliers.
Yours,
Frank Vlades
A task has to have some default before a user actually uses it. The default in this case is zero. However, it is true that you are supposed to set this to the readout noise for your detector which is usually something provided by the instrument engineer. My other comment is that rdnoise, gain, and snoise are only used for certain of the rejection algorithms, such as ccdclip, during combining. Some that are more empirical, such as reject="avsigclip", don't need these values and, instead, estimate the noise levels from the data itself. So if you don't have the values or they are uncertain you should avsigclip instead of ccdclip for rejection of pixels as statistical outliers.
Yours,
Frank Vlades
Guest wrote on Aug 06, 2008
Dear Frank,
Thank you for your clarification. I have since looked it up and found a detail explanation when I did 'phelp combine'.
Siti :D
Thank you for your clarification. I have since looked it up and found a detail explanation when I did 'phelp combine'.
Siti :D
Last post on Aug 06, 2008