A large effort went into trying to get mkapfile to fit the profiles of the defocused stars, and compute the aperture correction. This did not succeed.
Instead, aperture corrections at 12 and 15 pixels were manually read from the growth curves that mkapfile produced. These growth curves, one for each image, were based on the above 4 stars. An example of such a curve is shown in Figure . The resulting aperture corrections are shown in Table , and plotted in Figure , the two left panels. In the top right panel, the difference between the two is shown.
From the plots it can readily be seen, that the aperture correction is not the same for the 4 filters, and that it is not the same from night to night. Further, the difference between the aperture corrections at the two apertures is not the same for the 4 filters and from night to night.
The M67 field of this study had 23 stars in common with the standard stars of Jørgensen (1994). Besides of these standard stars, 36 stars were selected as ``program'' stars on the following three criteria: they should be somewhat bright (we required a peak > 500 ADU in the GR image d1262), they should not be too close to other stars, and they should have a unique name. The reason for including these program stars was: 1) the derived magnitudes for these stars could be useful in another context (they could perhaps serve as tertiary standard stars), and 2) the more stars, the better statistics for the test of the aperture correction, see below.
phot was run on these 59 stars, using two apertures:
12 and 15 pixels. cbox was increased to 20 pixels.
The reason for using two apertures was to test the aperture corrections
at these two apertures.
Since per definition
As can be seen, the form of the ``curves'' is qualitatively the same, but there is a small offset. This offset is probably due to contributions from neighboring stars for the sample of 59 stars. This contribution is much less in the sample of 4 stars, since they were selected from being isolated. This can be checked by inspecting these 4 stars in Figure and . They are almost always in the lower range of the 59 points. The mapping between names and serial ID number for these 4 stars is F81=ID1, F153=ID26, F170=ID28, and I27=ID38. F81 with and ID of 1 is particularly easy to locate. The conclusion is, that the manually determined aperture corrections pass the test of equation ().
The M67 field is quite crowded. For example, F128/ I198 and F129/ I199 are 25 pixels (13'') apart, F129/ I199 and F130 are 23 pixels (12'') apart, and F106/ I11 and its fainter neighbor I10 are only 16 pixels (8'') apart. This also shows up in the magnitude difference plots in Figure and . The mapping between names and serial ID number for these 4 stars is F128/ I198=ID15, F129/ I199=ID16, F130=ID17, and F106/ I11=ID7. Especially F106/ I11 is very contaminated. The above speaks in favor of a small aperture
Also, the images are quite defocused. Inspection of screen dumps of the images used in Jørgensen (1994) shows, that that these images were less defocused than the images of this study, despite that both used a defocusing of +200 encoder steps. This is probably because the filters are now placed in the DFOSC, instead in the instrument adapter as they previously were.
Jørgensen (1994) used two apertures for her defocused images, 7.52'' and 9.40''.
It was decided to use an aperture of 12 pixels (6.09''), and to use the corresponding aperture corrections listed in Table . As previously mentioned, the background region was 42-56 pixels, i.e. 21.31-28.41''.
Finally, we can compare the aperture corrections for the focused and the defocused images. This in done in Figure . The variation with night is quite similar for the focused and defocused images.
Properties of E and S0 Galaxies in the Clusters HydraI and Coma
Master's Thesis, University of Copenhagen, July 1997
Bo Milvang-Jensen (email@example.com)