2. OBSERVATION SETUP

2.1. Survey design

The details of the observations are presented in Table 1. The observations in C-band (central frequency 6.000 GHz) cover a bandwidth of 2 GHz (4.979 – 7.021 GHz), in 16 spectral windows and 1024 spectral channels (64 in each spectral window). The L-band (central frequency 1.575 GHz) observations cover a bandwidth of 512 MHz (1.247 – 1.503 GHz, 1.647 – 1.903 GHz), in 32 spectral windows, and 2048 spectral channels. In L-band, we placed the two base bands, of 16 spectral windows each, 144 MHz apart in order to avoid a region of particularly strong and contaminating radio frequency interference (RFI). Note that due to flagging during the reductions, the final central frequencies shown in Tables 4 and 5 for each galaxy are adjusted from the values given here.

2.1.1. Sensitivity

The theoretical noise level is approximately 6 µJy/beam in C-band (with confusion limit and flagging taken into account). Due to a high uncertainty in confusion, the theoretical estimate for L-band of 89 µJy/beam is substantially higher than the actual value we achieved, which is around 30-35 µJy/beam. However, due to variable noise in the L-band, we measured the noise in areas far from the source where it was more uniform (see Sec. 4.1). Additionally, a few galaxies exhibit higher values for reasons outlined later in this paper.

2.1.2. Distances

We have made a few modifications to the distances used in Table 1 of Paper I. Five galaxies have had their distances modified, whereas the rest remain the same as listed in Paper I.

The five galaxies with modified distances are: NGC 891, NGC 4244, NGC 4565, NGC 4631, and NGC 5907. These were adjusted because they have distances derived from the Tip of the Red Giant Branch (TRGB) (Radburn-Smith et al. 2011). Because of the reliability of the TRGB method, and that it likely provides the best distances for edge-on systems, we have adopted those distances here. The most significant distance change was for NGC 4565, which went from a Hubble Flow distance of 27.1 Mpc to a TRGB distance of 11.9 Mpc. For the other four galaxies the changes were minor (see Table 1).

Of the remaining galaxies, the distances of the five Virgo cluster galaxies (denoted with V in Table 1) were adopted from Solanes et al. (2002).

All other galaxy distances were taken from NED, using the "Hubble flow corrected for Virgo and the Great Attractor" (HF) distance, and a Hubble constant of 73 km/s/Mpc. Since a Hubble flow method may not always be accurate for nearby galaxies at low heliocentric velocities, we reviewed the galaxy distances for this paper. We compared the HF distance with the median of various Tully-Fisher (TF) derived distances. The latter are listed in NED as "red-shift independent" distance derivations. Note that we excluded the "Tully 1988" values listed in NED since those are Hubble Flow based. For the majority of the galaxies, the changes between the two methods were well within the uncertainties associated with either method. In particular, it is not uncommon to find a factor 2 difference in the various TF-method published distances. We therefore retained the distances used in Paper I for these 25 systems. The comparison showed that for 16 galaxies the agreement between the adopted HF distance and the median TF distances was better than 25%, for 9 galaxies the difference was larger than 25%. The galaxies with the largest distance uncertainty (> 35%) include NGC 2683, NGC 3432, NGC 4666, NGC 5775, and NGC 5792. See Table 1 for our current distance list.

2.2. Observations

Of the 405 hours which were awarded for the entire CHANG-ES project, 65 hours were set apart for the D-configuration observations in two frequency bands, L and C. The observations were divided up into 13 scheduling blocks, each of which contained scans of one primary gain and phase calibrator (hereafter referred to as the primary calibrator) and one zero polarization calibrator to calibrate polarization leakage from the instrumentation. Additionally, complex gain calibrations were performed using a source (hereafter the secondary calibrator) less than 10 degrees from the target galaxy. The secondary calibrator was observed before and after the galaxy scans, every 20-25 minutes. The bulk of the data was observed during December 2011. Two scheduling blocks were re-observed in March 2013. See Table 1 for observation specifics. One scheduling block observed in C-band, ID 5062559, suffered increased antenna system temperatures at times coinciding with bad weather (snow). Consequently, the rms noises for galaxies NGC 4302, NGC 4388, NGC 4438, NGC 4594, NGC 4666 and NGC 4845 increased to approximately twice the theoretical expectation at C-band.

2.2.1. Large galaxies

Eight galaxies in the sample are too large to fit inside the primary beam of 7.5' FWHM at C-band. We thus observed these in two pointings, separated by half the diameter of the primary beam at half maximum, i.e. one quarter beam from the galaxy centre along the disk on either side of the centre. The two-pointing galaxies are marked with an asterisk beside their names in Table 1. Table 2 lists the coordinates for the separate pointings. One single pointing was sufficient for all galaixes in L-band, where the diameter of the primary beam is 30'.