Heidelberg Extragalactic Astronomy Research Group

Go to the webpage of the SDSS "image list tool". Enter the following coordinates:
183.744129 11.864383
then click "Get Image".

Click on the image, and in the window that opens up, click on "Explore". There, download the FITS image of this field in the g and z band. Unzip it with gunzip filename.

For both images, determine the sky background level with IRAF/imstat, and subtract it with IRAF/imarith.

Open both images with DS9. Look at the star with coordinates 12:14:54.4 +11:51:58.5 in the z-band, and the star with 12:14:50.9 +11:51:48.6 in the g-band. When you adjust the contrast such that the overall noise level appears similar in the g and z images, both stars seem to have about the same brightness. Confirm this impression by measuring their magnitudes with IRAF/imexam. This is an interactive command: hold your mouse on top of the star and press 'a'. The magnitudes should come out fairly similar.

However, when you look at their magnitudes (in the respective band) as given on the SDSS webpage, you will find that the z-band star should be much brighter than the g-band star.

Take a look at the SDSS filter curves. Roughly how much more photons are 'caught' by the g-band as compared to the z-band? Correct the z-band for this lower efficiency by multiplying your z-band image with this factor, using IRAF/imarith.

Due to this scaling, the z-band star now indeed appears much brighter, as long as the contrast is not adjusted to similar noise levels like before — the noise has of course also been multiplied by this factor. It is now reflecting the true situation: the signal-to-noise is much worse in z than in g!