DESIGNING THE STUDY

The first step in the project was to create a list of candidates for observation under specific parameters. These parameters included star type, magnitude range, stars embedded in nebulae, and to be visible from the Coude Feed Spectrograph at the time of observing.
The star type was under the parameters of T Tauri, the original type of star needed for observing, but after literature searched many sub-classes of T Tauri type stars were found, such as FU Ori, Herbig Ae/Be, and VSOs. The list was constructed accordingly. Also, the stars had to be embedded in some sort of nebula, because that is what many times can either be the cause or is the effect of unnatural behaviors in young stars.
The other major constraints were the apparent magnitude and the limiting RA and Dec. The instruments allowed to us at Kitt Peak only allowed us to see down to near 14th magnitude, and the object must be visible in the Arizona sky for them to be observed.
Once the parameters were defined, we searched through catalogues using VizieR and Simbad Astronomical Database to find stars which were YSOs within the RA and declination range, and with a magnitude of 14 or lower. We also used the Starry Night software program to figure out the RA and Dec. ranges for Kitt Peak on the observing nights. The position of the moon in the sky was not a limiting factor because we would be observing during dark time.

Instrumentation and Setup

NOAO had accepted the proposal from our school and on the nights of November 1st and 2nd, we and our teacher observed at Kitt Peak National Observatory on the Coude Feed Spectrograph. Over two nights we observed for approximately 20 hours.
The Coude Feed Spectroscope at Kitt Peak is one of the greatest in the world, but unfortunately is mounted on one of the most diminutive telescopes on the mountain, one with a 1.6m mirror. It images onto a CCD camera, which takes photon counts of incoming light and then translates them into an image. It uses a CCD chip, which takes the received light from the stars it images and turns it into electrons. Those photons are then counted and then stored into a specific “bucket” or pixel. Since the camera is sensitive to temperature (higher temperatures create higher noise), it must be kept at a very low temperature to limit the noise. To ensure such low temperatures, a dewar cools the CCD to -105° C by means of nitrogen gas. This had to be refilled twice each day.
Due to limitations in the set-up of the spectroscope only a small range could be studied. The range of 3700 – 5100 Å in the blue end and 6200 – 9000 Å in the red end were chosen because the critical lines for determining wind speeds, abundances, and metallicities are located at that area of the spectrum. The resolution limit of the spectroscope was 1 Angstrom and both the blue and red filters were used, blue on the first night, red on the second night.

Calibration techniques

Certain procedures are taken to assess the reliability of the instrumentation and to standardize it. In order for this to occur, a series of calibrations techniques are completed.
The first of three calibrations is called a ‘zero bias’. The CCD camera will produce a spectrum to distinguish real photons of light from bad pixels. Next in the series of calibrations was to create a flat field with a Quartz-Iodide lamp. Each pixel in the CCD chip has a different ability and resolution, and the detector may be more sensitive at one end of the spectrum than at the other. The quartz lamp provides a steady light source that can be used to normalize all pixels to the same sensitivity when the data is reduced. The last calibration, depending on which end of the spectrum was being observed, was with either the Fe-Ar hollow cathode or the Th-Ar hollow cathode. The Fe-Ar and Th-Ar lines are used to calibrate the wavelength scale over the observed range. The Fe-Ar lamp is used in the blue end, whilst the Th-Ar lamp is used in the red end.
The spectra of a few standard stars we used to calibrate the instruments further, because a reliable star of no variability can be used to test the equipment’s capability.

Data Collection

Two separate computers were used to control both the telescope and spectrograph. They were both SUN workstations running UNIX. These machines were powerful enough to handle the massive amounts of information that was being transmitted between each other, the instruments, and the network.
The first computer controlled the telescope. Coordinates for the star to be observed were entered, as well as the name and additional notes (which included the magnitude). The user would then select the star to be observed from the newly made list, and then the telescope would slew to the area of the star. All information was also recorded on an observer’s log sheet.
An auto guider was used to track the star. In order to use the guider effectively, the telescope must be pointed very close to the star. A small video camera displays the telescope’s field of view. Once there, an intensifier was used to amplify the light. This is helpful to identify stars with dim magnitudes that push the limits of the telescope. The guider is then ‘locked’ on the star by using a keypad, which allows one to manually move the telescope at very small intervals with a push of a button. Once in place, the guider will track the star’s path throughout the integration, to prevent any loss of photons.
A second computer operated the spectrograph. On this computer, the user would input commands such as integration time and the nature of the object or the task, and start the exposure. This computer also displayed the spectrum complete with the photon count.
A good signal-to-noise ratio was very important as well to ensure good spectra, so a photon count of around 10,000 was ideal. The observed stars had fairly dim magnitudes, so it was necessary simply to take one long exposure rather than multiple short exposures and stack them. However, not all stars had the same visual magnitude. Thus, the integration times varied from 300-900 seconds and one to two exposures were taken of each star to try to come as close to the photon count as possible.
Back Forward