10.3 Corrections to Non-Photometric Observations

Unfortunately, not all the images have been taken in photometric weather. We need to identify these, and then compute offsets that will bring these images to the same system as the images observed under photometric conditions.

The observations log is shown in Table . It clearly states, that the weather was non-photometric when observing field 210 on night 4, that the weather might not have been photometric when observing fields 427, 215, 28, and 33 just after this, and that there might be problems with the 5 JU fields on night 9.

 

Table: The Notes in the Observations Log about the Weather

Night	Page	Weather	Galaxy fields observed
1	1	Photometric - to be checked	-
1	2	Photometric	-
1	3	Photometric	00, 16, 13, 25, 37
1	4	Photometric	37c, 18, 12, 21, 33, 45
1	5	Photometric	-
1	6	Photometric	-
1	7	Photometric	-
2	8	Photometric	214, 320
2	9	Photometric	428, 632, 29, 419, 519
2	10	Photometric	15
2	11	Photometric	-
2	12	Photometric	-
2	13	Photometric	-
3	14	[nothing written]	-
3	15	Photometric	00, 14, 15, 13
3	16	Photometric	27, 23, 43, 18, 321, 535
3	17	Photometric	535c, 35
3	17	May not be photometric right now, check this	-
3	18	Photometric (?)	-
3	19	Photometric	-
4	20	Photometric	-
4	21	Photometric	46
4	22	Photometric	47, 34, 22, 26, 24
4	22	NON-photometric	210
4	23	Photometric? check this	427, 215, 28, 33
4	24	Photometric	-
4	25	Photometric	-
4	26	Photometric	-
6	25	Photometric	-
6	26	Photometric	213, 39, 36, 44
6	27	Photometric	45, 32, 64, 66, 615, 16
6	28	Photometric	21
6	29	Photometric (?)	-
6	30	Photometric (?)	-
9	42	Photometric - but clouds may come later	-
9	43	Photometric - but clouds on their way!	00, 14, 15, 13, 18
9	44	Non-photometric!	-

 

Notes: The weather was noted on top of each page of the log book, except for page 17 and 22, where it was also noted further down. On night 5, the first page number was set to 21 instead of 27 as it should have been, therefore the strange page numbers for night 6. A ``c'' after the field number means that the observations of the given field were noted over two pages of the log.

One check of the photometric consistency comes from the program galaxies that have been observed more than once in a given filter. All possible pairs of such observations were formed, giving 53 pairs in GR, 42 pairs in JB, and 2 pairs in JU. The order of the two observations within the pairs was chosen to be chronological. About half of the observations are tested in this way see the ``Pair'' column in Table  (p. ). The number of pairs can be calculated from the numbers in Table  (p. ) as $N_{\rm pair} = N_2 + 3N_3 + 6N_4$, where N₂, N₃, and N₄ is the number of galaxies observed twice, three times, and four times, respectively. Note, that the one pair that can be formed from a galaxy observed exactly twice can be uniquely referred to by the name of the galaxy (and the filter).

From the raw magnitudes (cf. Sect. , p. ) instrumental magnitudes were calculated as

$$m_{\rm inst} = m_{\rm raw} -kX + n,$$ (10.3)

where -kX is the extinction correction and +n is the night shift. (This equation is described briefly in Sect. , p. , and in more detail in Ap. , p. .) For each pair of observations, the (instrumental) magnitude difference within a circular aperture, $\Delta m_{\rm circ}(r) \equiv m_{\rm circ,1}(r)-m_{\rm circ,2}(r)$, was calculated at aperture radii 6.26'' and 10.08''.

Magnitude difference at these two apertures for the three filters are plotted in Figure . Most magnitude differences are within $0\hbox{$.\!\!^{\rm m}$ }02$. Only two galaxies, R336 and R337, have large magnitude differences of around $0\hbox{$.\!\!^{\rm m}$ }1$. These two pairs are due to R336 and R337 being located in the overlap region between field 210 and 28. Field 210 is the one for which the observations log clearly states that the weather was non-photometric. Field 28 was observed later on night 4, where the log says that is should be checked whether the weather was photometric.

The only other pairs that have large deviations are R317 and R308. The R317 pair is due to field 33 being observed twice, on night 1 and 4. The R308 pair is due to field 16 being observed twice, on night 1 and 6.

We note that for the 6.26'' aperture there is a small systematic offset of $\sim 0\hbox{$.\!\!^{\rm m}$ }01$. This is most likely a seeing effect, as is shown in Sect.  (p. ). It is not important for the determination of the non-photometric offsets. First of all it is a small effect, and second, the 10.08'' aperture, which was also used, is not significantly affected.

 

From the above we have identified 5 field complexes that need to be checked:

Field complex	Filter	Indication of problems
		Observations log	Galaxy $\Delta(m)$
210/28/16	GR+JB	210: yes; 28: maybe; 16: no	210/28: yes; 16: maybe
427	GR+JB	maybe	n/a
215	GR+JB	maybe	n/a
33	GR+JB	maybe	maybe
00/13/14/15/18	JU	maybe	maybe

For each field complex, aperture photometry using the task phot was performed on all suitable stars in the overlap regions within the complex and with at least one other field that had been observed under photometric conditions according to the observations log. From the raw magnitudes, instrumental magnitudes were calculated in the same way as for the galaxy magnitudes, i.e. using Eq. (). For each overlap region, the magnitude difference for the stars, m₁-m₂, was plotted against m₁, using open boxes as the plotting symbol. These plots are referred to as offset plots. The first offset plots appear in Fig.  (p. ). The title of each of these plots are of the form ``Overlap: image1 (field1) - image2 (field2)'', where m₁ corresponds to image1, and m₂ corresponds to image2. The aperture radius (usually 9 pixels = 4.6'') is noted in the label of the x-axis. If there were any program galaxies in the overlap region, the magnitude differences for these at apertures 6.26'' and 10.08'' (12 and 20 pixels) were also plotted as filled triangles and boxes, respectively. The horizontal dashed line indicates the derived offset.

The offset plots for a given field complex are summarized in what is termed an offset diagram. The first offset diagrams appear in Fig.  (p. ). In these, each observation of a field is shown as a box, and the overlaps are indicated by arrows. Each arrow is labeled by the derived offset. The direction of the arrow indicates how the offset was calculated: if the arrow goes from field 2 to field 1 the offset was calculated as $\langle m_1 - m_2 \rangle$. The idea is, that to go in the direction of the arrow, one has to add the given offset.

How the fields are positioned on the sky relative to each other can be seen from Fig.  (p. ) for GR and JB, and Fig.  (p. ; left panel) for JU.

We now summarize the overlaps studied within the 5 field complexes that we found above needed to be checked. After that, we give the conclusions.