Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 321, 497-512 (1997)

Previous Section Next Section Title Page Table of Contents

2. Observations

Suitable targets for our project are TTS with very rich emission line spectra and large mean veiling values, which are the TTS with large accretion rates. We thus selected DG Tau, DR Tau, and DI Cep (Cohen & Kuhi 1979). Published values of the veiling in DG Tau and DR Tau are in range 0.8-2.4 for DG Tau (Basri & Batalha 1990) and 0.7-21 for DR Tau (Basri & Batalha 1990; Guenther & Hessman 1993). DI Cep was selected as much for its favourable R.A. as for its spectral characteristics: its position far away from Taurus/Auriga made it very easy to get longer series of spectra for several objects during August and September.

2.1. Echellette spectroscopy

In order to study the highest possible time scales of the variations, and in order to cover the whole spectral region from 3200 Å to 1.1 µm with a single spectrum, we obtained 93 spectra of DG Tau, 121 of DR Tau, and 117 of DI Cep in 13 nights between 1991 September 10 and 22 using the echellette spectrograph (Solf & Eislöffel 1990) on the MPIA 2.2m telescope at Calar Alto (Table 1). The spectral resolution [FORMULA] with a slit width of [FORMULA] and the GEC-Chip with 22.5µ pixels is about 3000 - large enough to measure the veiling but not enough to study subtle changes in the line profiles. Exposure times were typically 5-9 minutes. Thorium-Argon frames were taken at least every 20 minutes. The flux calibration star BD+284211 was observed each night and secondary flux standards several times per night. Additionally, we observed 32 dwarf, sub-giant, and giant stars of known spectral type between F8 and M5 to serve as spectroscopic template stars for the veiling measurements. The template stars were selected from the MK catalog and checked using the SIMBAD database.


Table 1. Journal of the observations [FORMULA]

Standard IRAF routines were used to flat-field the frames, to remove the stray light and sky background, and for extracting the spectra. The wavelength calibration was confirmed by dawn spectra. Atmospheric absorption lines were removed by dividing the spectra of the TTS by spectra containing the atmospheric absorption lines that were individually rescaled to the same optical depth as in the TTS. The different parts of the spectra were relative flux calibrated using BD+284211 and its tabulated fluxes. The quality of this process was monitored by comparing the derived values of spectroscopic templates with known fluxes. Although none of the nights were photometric, stable seeing conditions resulted in average errors of the spectral energy distribution on most nights of only 5% in the range of Johnson I, R, and V filters, 10% in Johnson B, and 20% in the range of Johnson U. A fully-reduced but still reddened echellette spectrum of DG Tau is shown in Fig. 1. The strongest photospheric absorption line is Li I [FORMULA] 6707 - barely visible in the logarithmic plot due to the unusual strength of the veiling and emission line spectra.

[FIGURE] Fig. 1. Echellette spectrum of DG Tau showing the rich emission line spectrum and the high degree of veiling.

2.2. Echelle spectroscopy

For detailed line profile studies, we also obtained 10 spectra of DI Cep and 3 spectra of DR Tau between 1991 August 22 and 28 (see Table 1) using the fiber coupled FLASH-spectrograph of the Landessternwarte Heidelberg (Stahl et al. 1994) and the MPIA 2.2m telescope on Calar Alto. The spectra cover the region between 4040 Å and 6800 Å with a resolution of 20 000. [FORMULA] served as a flux standard. A number of spectroscopic template stars selected from the MK catalog were observed as well, ranging in spectral class from A0 to M2. Standard IRAF routines were again used for the data reduction. The accuracy of the wavelength calibration with 229 Thorium Argon lines is better than 0.5 km/s.

2.3. Photometry

Broadband relative-photometry in the Johnson U and I bands was carried out simultaneously with the spectra on some nights (Table 1) using the 1.23m telescope on Calar Alto and the GEC CCD-Camera. Data reduction was performed using the IRAF apphot package. Although none of the observing nights where photometric, a relative precision of the I-band photometry of better than 1% was achieved. Unfortunately, due to problems with finding sufficiently bright comparison stars in U and the reddened program stars, no useful ultraviolet data was obtained.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: June 30, 1998