154
10 Photometry
and taking an aperture slice from which you can get the FWHM. In MaxImDL,
you need to use the Graph function (found under View. Select Line and draw a
line through a star. Doing so will give you a slice of the PSF, from which you
can obtain the FWHM. In AstroImageJ, this is done by drawing a line through a
star and performing a Profile Plot (found under Analyse). In theory, the FWHM
should be the same for all of your images. If it is significantly different, then do
not use the outlying images.
3. You now need to set the aperture and annulus size. Set this to three times half the
FWHM. In general, it is best to let the software determine the size of the annulus,
but if it does not, five times the aperture radius is normally suitable. If you have
a very dense cluster, you may wish to have a much smaller annulus size in order
to avoid contamination from nearby stars.
4. Using ATP, you will need to set the source algorithm to Model 2 and the Sky
algorithm to Model D.
5. Drop an aperture on each of your reference stars, noting the source count and the
count uncertainty. In APT, this is done by double clicking on the target box. APT
uses a very sophisticated uncertainty model that goes beyond a simple signalto-noise ratio. If you are not using APT to do the photometry, use the SNR to
calculate the uncertainty instead.
6. Using the camera gain, which should be in the FITS header, time normalise the
count, if needed, and apply (10.2). This will give an instrumental magnitude for
each reference star in all fields.
7. Record each instrumental magnitude along with the target star’s position, the filter
in which the image was taken, the air mass, and the magnitude of the standard
star.
10.3.5 Reduction
There now follows a series of steps that will find the various components of (10.4)
for your system. These processes require the plotting of data and the fitting of a line
to that data. In Warner, it is suggested that a spreadsheet be used. You could use
a specifically designed application, such as PhotoRed, or perhaps a mathematical
application such as Matlab to plot the data and fit the line. However, whatever you
use, you should always visualise the data. It is easy for outliers to change the straight
line fitting dramatically. If you see an outlier, try to determine what caused it. It might
be a hot pixel, a cosmic ray, or even a misidentified reference star, and address the
problem if possible. If you cannot correct it, you can remove it from your dataset as
long as you make a note of it in your lab book, with an explanation of why it was
removed.
In this section, I will be assuming that you are using V and R filters for notation,
with V and R being the standard magnitudes of the reference stars. The notations
v and r are used for the instrumental magnitudes, and v and r for the mean
instrumental magnitudes. If you are not using V and R, just substitute.
10 Photometry
and taking an aperture slice from which you can get the FWHM. In MaxImDL,
you need to use the Graph function (found under View. Select Line and draw a
line through a star. Doing so will give you a slice of the PSF, from which you
can obtain the FWHM. In AstroImageJ, this is done by drawing a line through a
star and performing a Profile Plot (found under Analyse). In theory, the FWHM
should be the same for all of your images. If it is significantly different, then do
not use the outlying images.
3. You now need to set the aperture and annulus size. Set this to three times half the
FWHM. In general, it is best to let the software determine the size of the annulus,
but if it does not, five times the aperture radius is normally suitable. If you have
a very dense cluster, you may wish to have a much smaller annulus size in order
to avoid contamination from nearby stars.
4. Using ATP, you will need to set the source algorithm to Model 2 and the Sky
algorithm to Model D.
5. Drop an aperture on each of your reference stars, noting the source count and the
count uncertainty. In APT, this is done by double clicking on the target box. APT
uses a very sophisticated uncertainty model that goes beyond a simple signalto-noise ratio. If you are not using APT to do the photometry, use the SNR to
calculate the uncertainty instead.
6. Using the camera gain, which should be in the FITS header, time normalise the
count, if needed, and apply (10.2). This will give an instrumental magnitude for
each reference star in all fields.
7. Record each instrumental magnitude along with the target star’s position, the filter
in which the image was taken, the air mass, and the magnitude of the standard
star.
10.3.5 Reduction
There now follows a series of steps that will find the various components of (10.4)
for your system. These processes require the plotting of data and the fitting of a line
to that data. In Warner, it is suggested that a spreadsheet be used. You could use
a specifically designed application, such as PhotoRed, or perhaps a mathematical
application such as Matlab to plot the data and fit the line. However, whatever you
use, you should always visualise the data. It is easy for outliers to change the straight
line fitting dramatically. If you see an outlier, try to determine what caused it. It might
be a hot pixel, a cosmic ray, or even a misidentified reference star, and address the
problem if possible. If you cannot correct it, you can remove it from your dataset as
long as you make a note of it in your lab book, with an explanation of why it was
removed.
In this section, I will be assuming that you are using V and R filters for notation,
with V and R being the standard magnitudes of the reference stars. The notations
v and r are used for the instrumental magnitudes, and v and r for the mean
instrumental magnitudes. If you are not using V and R, just substitute.