Observations and data analysis

We present results from TAURUS-2 Fabry-Perot imaging spectrograph at the William Herschel Telescope (4.2 m) at the ORM on La Palma (Canary Islands). The output from this set-up, after phase correction and calibration (Lewis & Unger 1991), is a 3-D data cube with x and y being the spatial directions and z the wavelength (etalon step) sampling. Data-cube dimensions were (245, 217, 100) and the chosen etalon-FSR (free spectral range) was 13.4 Å or 610 km s in H ( Å). The spatial scale was 0.26 arcsec pixel, or 6.25 pc at the distance on NGC 4449. The exposure time for each data-cube was one hour. Spectral scanning was Å equivalent to 9 km s in H.

An image photon counting system (IPCS) was employed. With the IPCS the 100 planes are scanned continously with short exposures to avoid the effects of seeing variations during the total integration time. In this way, changes in seeing occurring during the 3600-s exposure time, that would modify the emission-line profile, are minimized. During the observations seeing was about 1.0''. A complete description of the observations and data reduction is given in Fuentes-Masip (1997). Each TAURUS data-cube contains a total of 53165 spectra from an area of 1 arcmin, hence the analysis procedure is an enormous task. Specific software, MATADOR, has been used for data analysis (see details in Gavryusev & Muñoz-Tuñón 1996; and Muñoz-Tuñón et al 1995).

Two overlapping fields in H were taken to cover an effective area of . Interferential filter widths with FWHM = 15Å were used to limit the wavelenght range, thus avoiding the overlapping of different interference orders.

The underlying continuum is fitted and subtracted from the calibrated data-cubes. The collapsed H map that results from adding all z planes, free from continuum, has been used to unveil the ionized structure of the galaxy. Several independent methods were then used to isolate and catalogue individual H II regions, as well as to obtain their size and associated luminosity. The regions were identified in several ways: visually, detecting relative maxima of emission, using the flux gradient map and using the line-width and velocity gradient maps. In all cases the distribution function of the findings was computed to determine the possible errors and/or bias of each technique (see Fuentes-Masip 1997 for details). Flux calibration was carried out using an O III CCD image taken at the 1-m Jacobus Kapteyn Telescope, also at the ORM. The complete catalogue will be published elsewhere (Fuentes-Masip et al. 1997).

© Copyright Astronomical Society of Australia 1997