Next Contents Previous

CHAPTER 3. REDUCTION METHODS

The 21 cm line data obtained with aperture+synthesis instruments like the Westerbork Synthesis Radio Telescope (WSRT) are not accessible to interpretation in their original form. Various data handling steps are necessary to produce astronomically intelligible results. In this chapter we will describe briefly the most relevant data handling steps for the WSRT data we have obtained and the problems associated with.them. In section 1 we describe the observing procedure, in section 2 the problems associated with the telescope configuration, in section 3 we discuss methods for the determination of column densities and radial velocities, and in section 4 we discuss the problems associated with the large beam sizes and their effect on the derived astronomical quantities.

1. Observing procedure

Most of the observations described in this thesis were obtained with the WSRT and the 80 channel filter spectrometer. A detailed description of this system is given by Allen et al. (1974). We give here a brief outline of the main principles of operation. During one 12 hour period the object (galaxy) is tracked and measured with 10 interferometers. The interferometer pairs are made up either by telescopes 5 to 9 and A and B or by telescopes 0 to 4 and A and B. In the former case the shorter spacings are measured, in the latter case the longer spacings. If only data at the shorter spacings are used maps can be obtained with an angular resolution of about 1'; for full resolution, i.e. about 0.5', data at longer and shorter spacings have to be combined.

In one 12 hour period 8 frequency channels per interferometer are measured simultaneously. The filter shapes are approximately gaussian, with a full width at half maximum of 27.2 km s-1 in velocity. They are spaced in velocity at intervals of 40 km s-1. To obtain full velocity coverage over a range of 280 km s-1 the filter bank has to be shifted in frequency, by a new setting of a local oscillator, with the equivalent of 20 km s-1. Hence, if the radial velocity range of a galaxy is 500 km s-1 four halfdays of observations are needed to obtain full velocity coverage at the shorter spacings and another four halfdays to obtain full velocity coverage at the longer spacings. Full velocity coverage is essential: only then the zeroth and first moment of the profiles are an adequate measure of the HI column density and the intensity weighted mean velocity (see Allen et al.).

For each of the filter channels the data in the UV-plane are calibrated and Fourier transformed into the plane of the sky using standard procedures (Högbom and Brouw, 1974); the result is a set of channel maps which, in the velocity range of the galaxy, are spaced in velocity at intervals of 20 km s-1. In Table 1 we have collected some observational parameters of the galaxies for which we have performed (partly in collaboration with others) the data handling. In some cases we dealt only with shorter spacings measurements, and in other cases we obtained full resolution data. Note that with the planned new WSRT line receiver full resolution line data of a small object can be obtained in only one halfday.

TABLE 1. Observational parameters of galaxies discussed in this thesis.

Galaxy Holmberg size Spacings Beamwidth Grating ring sizes Number of LO settings Number of halfdays Number of channel maps

NGC 4736 15.0' × 13.3' 36, 108, --, 1404 m 25" × 38" 10' × 15' 4 8 32
NGC 4151+ 9' × 8' 36, 108, --, 684 m 51" × 80" 10' × 16' 3 3 24
NGC 5033 12.3' × 5.8' 36, 108, --, 684 m 51" × 85" 10' × 17' 4 4 32
NGC 3198 11.9' × 4.9' 36, 108, --, 1404 m 25" × 35" 10' × 14' 4 8 32
NGC 5055 16.0' × 10.1' 36, 72, --, 720 m 49" × 73" 20' × 30' 4 8 32
NGC 2841 11.3' × 5.7' 36, 108, --, 684 m 51" × 65" 10' × 13' 5 5 40
NGC 7331 13.5' × 7.0' 36, 108, --, 1404 m 25" × 45" 10' × 18' 4 8 32

see Bosma, Van der Hulst, Sullivan (1977)
+ see Bosma, Ekers, Lequeux (1977)

Next Contents Previous