next up previous
Next: Discussion Up: No Title Previous: Introduction

   
Observations and Reductions

MR 2251-178 was observed on August 30, 1998 using the TTF at the f/8 Cassegrain focus of the 3.9-meter AAT. The TTF instrument consists of a pair of modified high-finesse ($N
\sim 40$) Queensgate etalons (blue and red) which can be tuned to provide narrowband imaging anywhere within the wavelength range 400 to 960 nm, through an arbitrary bandpass, with resolving powers of 100 to 1000.2 The observations of MR 2251-178 were made with the red side of the TTF, using a mediumband ( $\Delta\lambda = 26.0$ nm) blocking filter centered at 707 nm, tilted by $16\arcdeg$. Two 600-second exposures were obtained at each of two etalon spacings. The exposures were dithered amongst pointings on a $15\arcsec$ grid. The average atmospheric seeing of $1.3\arcsec$ was oversampled by the $0.37\arcsec$ pixels. The night was photometric. MR 2251-178 was also observed on September 3, 1998 in direct imaging mode, using a standard I-band filter.

Fits to a number of emission lines from observations of a calibration lamp (CuAr) were used to determine the relationship between wavelength, spatial position, and etalon gap spacing. The free spectral range of the etalon, i.e., the distance between orders, was found to be 265.8 Å, well-matched to the bandwidth of the blocking filter. The system was used in the 26th order of interference, with a 9 micron etalon gap, at an effective finesse of 39.7. This translates to a spectral resolution of 6.8 Å, or $R \sim 1040$, with an effective bandpass of $\sim12$ Å. The pair of etalon spacings produced imagery with central wavelengths on the optical axis of 6983 Å and 6986 Å, corresponding to redshifted velocities of 60 and 195 km s-1 relative to the quasar. The field is essentially monochromatic: pixels $\sim 1\arcmin$ from the optical axis have a central wavelength only 1.3 Å (60 km s-1) bluer than pixels on-axis. The optical axis is located approximately $35\arcsec$northwest of the quasar.

The data frames were bias-subtracted and flatfielded in the typical manner. An azimuthally-symmetric sky frame was produced from a mean radial sky spectrum for each image and subsequently subtracted. Images at each of the two etalon spacings were then aligned and combined using 9 stellar objects in the field. An image mask was used to simultaneously remove three ghost reflections from each image. The data were flux calibrated using observations of the standard star EG 21 ([Stone & Baldwin 1983]).

The I-band observations were reduced following standard CCD procedures. The summed continuum image was matched to the narrowband imagery using a number of stars and subsequently subtracted. A few remaining stellar residuals and cosmetic defects were identified and repaired by hand.


next up previous
Next: Discussion Up: No Title Previous: Introduction
Patrick Shopbell
1999-08-26