SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 332, 25-32 (1998)

Previous Section Next Section Title Page Table of Contents

3. Observations

3.1. Near-infrared photometry

The observations were performed using the ESO/MPI 2.2m telescope at La Silla, Chile. On October 14 and 15, 1995 the 256 [FORMULA] 256 pixel HgCdTe NICMOS-3 array IRAC2 was used to obtain the JHK photometry. A pixel size of [FORMULA] was chosen resulting in a field of view of [FORMULA]. Both nights were photometric.

For each object the field was centered on the IRAS coordinates and the on-line display was used to identify the counterpart using the J and K filter (by default the infrared counterpart to the IRAS source is assumed to be the reddest star in the field). If necessary the telescope was moved slightly to place the star somewhat nearer to the center of the field. Five images were taken in JHK each, one at the center position and four dithered by [FORMULA] in RA and Dec. The five images were median filtered to give the sky emission which was subsequently subtracted from the five object frames. The resulting frames were divided by the flatfield and bad pixels were removed. Flat fields were obtained by subtracting an image of the screen in the dome from one where the screen was illuminated by a lamp.

The assumed counterpart was identified, and four images were shifted by an integer number of pixels to coincide with the counterpart on the fifth image. The five resulting images were then averaged. Software aperture photometry was used to obtain the flux within the aperture.

The typical integration times are 5 [FORMULA] 10 seconds on chip exposure time per image for JHK. For the brighter sources this was adjusted accordingly to avoid saturation. For the weaker sources in J, longer exposure times or a second set of 5 dithered images was taken.

We observed one standard star from the list of standards prepared for the ISO mission, i.e. HD 1274 (listed as having J = 8.807; H = 8.460; K = 8.370). We planned to observe HD 7644 as well but as we found out after the observing run the coordinates listed in the file of standard stars were wrong (listed with a declination of [FORMULA] instead of [FORMULA]). So instead we unknowingly observed a field star. The error was not apparent during the run because the unknown object turns out to have magnitudes similar to HD 7644 and the coordinates listed for HD 7644 were only accurate to [FORMULA]. Both stars were observed at several air masses to obtain the extinction coefficients. Inspection of the count rates of both stars at roughly the same air masses on both nights showed no significant variations. It was decided to combine the standard star observations of both nights to determine the extinction coefficients and zero points. The extinction coefficients determined for both stars were comparable.

They were subsequently averaged and the zero points derived from HD 1274 are used. The unknown object (located approximately at RA = 1h15m36s, [FORMULA], J2000) has magnitudes J = 9.80, H = 9.40, K = 9.27.

The resulting photometry is listed in Table 2. The error quoted is the error due to the uncertainties in the determination of the zero point and extinction coefficient (0.01 in J and 0.02 in H and K) and Poisson noise. In case of upper limits the weakest detected object in the field is taken as a measure of the upper limit. The designation of the sources is ours.


[TABLE]

Table 2. Near-infrared photometry


On October 16 and 17, 1995 L images were obtained for a selected number of sources using the 58 [FORMULA] 62 pixel InSb SRBC IRAC1 array on the 2.2m telescope. A pixel size of [FORMULA] was chosen resulting in a field of view of [FORMULA]. Initially we intended to get M images as well but as we were barely able to detect a M = 3.3 standard star, no further M images were taken.

Standard chopping and nodding techniques were used: [FORMULA] E-W at a rate of 12 Hz. This has the advantage that an object appears twice in the raw data image: once with positive and once with negative signal. These can then be combined to increase the S/N. The standard star HR 77 (L = 2.792) was observed at several air masses during the two nights.

The observations of the targets and the standard star were constructed in such a way that an object would always appear roughly at the same location of the array. So, instead of constructing a formal flatfield based on the pixel-to-pixel variations a flatfield was determined on a `region-to-region' basis. This flat field was determined by comparing the count rates of the standard star in these regions. Measurements made at six positions across the array varied by up to 20%. When applying the region-to-region flatfield this was reduced to about 3.5% (1 [FORMULA]). Because of the relative large scatter in the count rates no useful extinction correction could be derived. The fit was consistent with zero extinction and no extinction correction was applied. The derived L magnitudes are listed in Table 2; some of the error bars are large.

3.2. Optical spectroscopy

On October 16 and 17, 1996, optical spectroscopy was obtained for a sub-sample of sources using the EFOSC1 instrument on the 3.6m telescope. The objects observed are those that had a clear optical counterpart. In addition, two AGB stars in the LMC were observed at the end of the nights when the SMC was at a high airmass. They are TRM 88 (Reid et al. 1990) and LI-LMC 57 (Schwering & Israel 1989b)

Acquisition images in the Gunn z band were obtained and compared to the finding charts. The object was identified and placed in a [FORMULA] slit. Grism `B1000' was used. This results in an useful spectral coverage from 3200 to about 10 000 Å with a dispersion of 25 Å per pixel. Flat fields were obtained by illuminating a screen in the dome. Wavelength calibration was achieved using a He/Ar lamp. As the aim was to obtain spectral types no flux standards were observed. The spectral transmission of the atmosphere plus telescope and detector system was obtained by observing EG 21 and LTT 7379. For comparison the well-known carbon stars Z Psc (spectral type C-N5 C2 4, Barnbaum et al. 1996), TX Psc (C-N5 C2 4+, Barnbaum et al.) and the M3.5 IIIa MK standard HD 1364 (Keenan & McNeil 1989) were observed. Integration times on the targets were between 10 and 45 minutes. The reduction was carried out with MIDAS.

Most of the SMC star and the comparison spectra are shown in Fig. 1. Fig. 2 shows the two LMC stars and the peculiar object S2. S2, S14 and S30 are discussed below. The other cases are straightforward: S9, 10, 18, 20a and 22 are late-type M-giants; S12, 20b and the two LMC AGB stars are carbon stars.


[FIGURE] Fig. 1. The optical spectra of the spectral standards and some SMC objects. To the left the oxygen-rich objects, to the right the carbon-rich objects. Some prominent lines are indicated, including telluric O2 and H2 O.

[FIGURE] Fig. 2. As Fig. 1 for the preculiar object S2, and the LMC stars.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: March 10, 1998
helpdesk.link@springer.de