CHANG-ES. IV. RADIO CONTINUUM EMISSION OF 35 EDGE-ON GALAXIES OBSERVED WITH THE KARL G. JANSKY VERY LARGE ARRAY IN D CONFIGURATION—DATA RELEASE 1

The theoretical noise level is approximately 6 μJy beam⁻¹ in C-band (with confusion limit and flagging taken into account). Because of a high uncertainty in confusion, the theoretical estimate for L-band of 89 μJy beam⁻¹ is substantially higher than the actual value we achieved, which is around 30–35 μJy beam⁻¹. However, because of variable noise in L-band, we measured the noise in areas far from the source, where it was more uniform (see Section 4.1). Additionally, a few galaxies exhibit higher values for reasons outlined later in this paper.

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 because 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 the NASA/IPAC Extragalactic Database (NED), using the "Hubble Flow corrected for Virgo and the Great Attractor" (HF) distance and a Hubble constant of 73 km s⁻¹ Mpc⁻¹. Because 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 "redshift-independent" distance derivations. Note that we excluded the "Tully 1988" values listed in NED because those are based on the Hubble Flow. 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 of two 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%, and for nine 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.

Eight galaxies in the sample are too large to fit inside the primary beam (PB) of 75 FWHM at C-band. We thus observed these in two pointings, separated by half the diameter of the PB at half maximum, i.e., one-quarter beam from the galaxy center along the disk on either side of the center. 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 galaxies in L-band, where the diameter of the PB is 30'.

Table 2.  Two Pointings of Large Galaxies

Galaxy	R.A. 1	Decl. 1	R.A. 2	Decl. 2
(1)	(2)	(3)	(4)	(5)
N891	02h 22m 3721	42d 22m 412	02h 22m 2961	42d 19m 126
N3628	11h 20m 2450	13d 34m 557	11h 20m 952	13d 35m 501
N4244	12h 17m 3671	37d 49m 409	12h 17m 2261	37d 47m 103
N4565	12h 36m 2658	25d 57m 547	12h 36m 1498	26d 00m 366
N4631	12h 42m 1688	32d 32m 372	12h 41m 5914	32d 32m 216
N4594	12h 40m 0709	−11d 37m 230	12h 39m 5177	−11d 37m 230
N5084	13h 20m 2488	−21d 49m 198	13h 20m 896	−21d 49m 588
N5907	15h 15m 5949	56d 18m 16	15h 15m 48.05	56d 21m 255

Note. Column 1: galaxy name; Column 2: Right Ascension of pointing 1; Column 3: declination of pointing 1; Column 4: Right Ascension of pointing 2; Column 5: declination of pointing 2.

Download table as:  ASCIITypeset image

For the polarization calibrations, the primary calibrator was used as a polarization angle calibrator. One of the NRAO standard zero polarization calibrators was used to calibrate instrumental polarization leakage. We corrected for cross-hand delays, leakage terms, and polarization position angles. Where in some cases the polarization leakage calibrator could not be used (due to loss of scan during observation or being more heavily flagged than the other calibrators), we instead utilized the secondary calibrator where the range in parallactic angle was sufficient, i.e., more than 60°. This was, for example, done for the galaxies in a C-band scheduling block (ID: 4809749), for which the polarization leakage calibrator scan was lost. The Q and U fluxes were calculated in the standard way—see Paper II and references therein. Our tests have shown no difference in the result whether the secondary calibrator or the polarization leakage calibrator was used.

For each galaxy, the Briggs robust weighting (Briggs 1995) was used for both C- and L-band data. Additionally, we imaged a uv-tapered version of this weighting, which we introduced in order to emphasize the broad-scale structures.¹⁵ We achieved an increase in beam size of approximately 80% in C-band and 40% in L-band, by typically applying a 6 kλ Gaussian taper in C-band and a 2.5 kλ Gaussian taper in L-band to the uv distribution. Image products of both these weightings are included in Data Release 1 and are referred to as robust 0 and uv-tapered weightings, respectively.

In the C band, images sufficiently clear of artifacts with uv-tapered weighting could not be achieved for NGC 4438 and NGC 4845, due to particularly strong contaminating sources in the field. In the L-band, the uv-tapered weighting was either challenging to achieve (for the same reason) or unnecessary (the sources were imaged as point sources already at robust 0) for six of our galaxies: NGC 3448, NGC 4192, NGC 4388, NGC 4438, NGC 4845, and UGC 10288, and the attempt was discarded for these. Hereafter any images shown will be of robust 0 weighting, unless otherwise specified.

Each imaging run was cleaned down to a flux density level of 2.5–3σ. With few exceptions, we imaged the entire field (i.e., without specifying regions) because a lower rms could be obtained with this strategy. Additionally, the vast majority of our data do not suffer from significant artifact patterns with spurious sources outside the source region, which could jeopardize this strategy. When needed, one to two self calibrations were performed in order to deal with remaining artifacts.

Our self-calibration strategy was based on iteratively refining the calibration table, by self calibrating at successively deeper thresholds and thus improving the input model, which in turn acts on the original visibilities. At each step, care was taken to (1) check the model to be used for the self calibration that it did not include any artifacts and (2) check that the peak intensity of the source would not decline (such a scenario was aided by opting for a phase-only self calibration at the first iteration or by applying the self calibration at a shallower threshold).

In some cases, our cleaning strategy had to be adapted to the circumstances of the data (the quality of a data set, the strength of emission, contamination in the field, and so on). This particularly applies to data with high dynamic ranges (HDR), from galaxies with strong, often point-like center sources or strong field sources (for example NGC 660, NGC 4594, and Virgo galaxies). Our strategies to deal with these situations include one or more consecutive runs of self calibrations, specifying regions, attempts at peeling for strong field galaxies (see Section 3.3.6), and clean parameter adjustments (including multiscale adjustments). Generally, the best one can achieve in terms of HDR is a signal over noise of the order of 10,000:1 (S. Bhatnagar 2015, private communication). Our dynamic ranges are listed in column 8 of Tables 4 and 5. Additionally, in cases of strong center sources, choosing a smaller cell size will characterize the steep beam better and render a cleaner result (e.g., NGC 4594 and NGC 4845, both in C-band).

Stokes Q and U images were produced similarly to I. These maps were used to create polarization intensity and polarization angle maps. The polarization intensity maps were corrected for bias caused by the noise level,¹⁶ and the polarization angle maps are cut off at the 3σ level. We note that if the fraction of polarization intensity over Stokes I intensity is less than 0.5%, the polarization is not believable since the calibration cannot be guaranteed below that level (S. Myers 2015, private communication).

A uniform Faraday screen resulting from the ionosphere is corrected for in the standard polarization data reduction. However, if there is differential Faraday rotation, for example between the location of the primary calibrator and the source, then this is not corrected for.¹⁷ This effect is negligible at C-band. There could, however, be minor effects at L-band, which will be investigated in a subsequent polarization paper.

Our vector maps (panels (d)–(f) of the figures in the Appendix) are also "apparent B vectors," which simply are E vectors rotated by 90°, because internal Faraday rotation in the galaxy has not been corrected for.

Wide-band PB corrections were carried out with the CASA task widebandpbcor after cleaning. This task accesses information about the beam from the visibility data and calculates a known NRAO-supplied model of the PB at each frequency channel in each band. A PB cube is formed, and Taylor terms are found in the same way as for the data set when imaging. Each Taylor coefficient image (two in our case; see Sections 3.3 and 3.4.1) is then corrected.

This task also corrects the spectral index image above a given threshold (we use 5σ), as well as corrects a map of the formal error of the spectral index (see Section 3.4.2). Subsequently, we also applied PB corrections to the polarization intensity maps by dividing the image with the beam created in the widebandpbcor task.

Note that, in this paper, any displayed images are not PB corrected, but any measurements are made from the PB-corrected versions of the maps, unless otherwise indicated. For total intensity images, use of a model for the PB rather than an observation-specific PB introduces negligible error over the scales of our galaxies. However, the effect on the spectral index maps must be investigated in more detail, which we do in Section 3.4.2.

For the eight large galaxies that were observed in two pointings at C-band (see Table 2), each individual pointing was calibrated and imaged separately. Although it would be preferrable to mosaic the pointings onto the same grid while still in the uv plane, this feature was not yet implemented in CASA at the time for the clean settings that we required (i.e., nterms = 2). Instead, we combined the images from the two pointings, forming a weighted (by the PB) average in the overlapping region. The images were combined in the overlapping regions out to 10% of the PB. This rather inclusive choice of cutoff worked well for the D-configuration data, but will likely be adjusted to 50% for the B- and C-configuration data in the future.

This strategy was applied to both the non-PB-corrected images and the two PB pointings. The combined Stokes I image was divided by the combined PB in order to do the PB correction.

The two-pointing spectral index maps, which had already been individually PB corrected, were similarly combined, as well as their accompanying error maps.

We assessed whether the two pointings had differences within the overlap regions, such as in noise levels. In such cases, additional weightings, based on the noise level, would have to be considered. This was however not the case for any of the CHANG-ES galaxies, where separate pointings generally had been observed in turns in the same scheduling blocks or in similar conditions.

For a couple of cases, for example NGC 660 and NGC 4388, bright field sources interfere strongly in the cleaning process, causing a less-than-ideal imaging result fraught with severe artifacts (L-band only). An attempt was made at peeling these interfering sources in those examples, that is, to remove or decrease the intensity of the interfering source so that its impact is lessened in the final deconvolution. The process includes self calibrations with the interfering source centered and subtracting the source from the uv data using the CASA task uvsub in several steps. For more information on peeling, see Pizzo & de Bruyn (2009), as well as the PhD thesis by Adebahr (2013). Table 3 lists the sources we peeled from NGC 660 and NGC 4438.

Table 3.  Peeled Sources in the L-band Fields of Two Galaxies

Galaxy	Source Name	Angular Distance	Intensity Before/After	R.A.	Decl.
		(')	(mJy beam−1)
(1)	(2)	(3)	(4)	(5)	(6)
N660	⋯	15	95/26	01h 42m 188	+13d 27m 486
N4388	M87	75	132/22	12h 30m 48s	+12d 23m 302

Note. Column 1: galaxy name; Column 2: name of interfering source 1; Column 3: distance in arcminutes between galaxy and interfering source; Column 4: intensity of interfering source before/after peeling; Column 5: Right Ascension; Column 6: declination.

Download table as:  ASCIITypeset image

As indicated in Section 3.3, the ms-mfs algorithm allows for spectral fitting across the wide bands that have been used in the CHANG-ES survey. Thus, a single observation over a single wave-band permits the determination of spectral indices, α, as a function of position within the galaxy. As described in Rau & Cornwell (2011) (see also Paper II), a specific intensity, ${I}_{\nu }$, at some frequency, ν, within the band can be fit with a function of the form ${I}_{\nu }\;=\;{I}_{{\nu }_{0}}\;{\nu }^{\alpha \;+\beta \;\mathrm{log}(\nu /{\nu }_{0})}$, where β is a curvature term and ${\nu }_{0}$ is a reference frequency (our central frequency, ${\nu }_{0}$, given in Tables 4 and 5). As implemented, such a fitting function is expanded in a Taylor series about ${\nu }_{0}$, such that the first Taylor term (the map, TT0) corresponds to a map of specific intensities at ${\nu }_{0}$, and the second Taylor term (the map TT1) contains information on α such that $\alpha \;=\;\mathrm{TT}1/\mathrm{TT}0$. The third Taylor term (TT2) allows for the recovery of the spectral curvature, β, where $\beta \;=\;\mathrm{TT}2/\mathrm{TT}0\;-\;\alpha (\alpha -1)/2$.

Table 4.  Imaging Results for the C Band

Galaxy	Weighting	Bmaj	Bmin	BPA	${\nu }_{0}$	rms I	Map Peak	DR	rms QU	Pol. Peak
		('')	('')	(°)	(GHz)	(μJy beam−1)	(mJy beam−1)		(μJy beam−1)	(μJy beam−1)
(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
N660	rob 0	9.55	9.14	−18.72	5.99904	8.6	603	70116	6.0	369.1
	uvtap	16.13	16	−58.32		18	614	34111	7.0	554.6
N891	rob 0	9	8.81	−79.46	5.99932	6.5	7.21	1109	6.6	115.0
	uvtap	15.31	14.63	88.05		6.9	6.9	1000	7.2	225.0
N2613	rob 0	15.42	8.3	−11.05	5.99900	10.1	0.643	64	10.4	57.1
	uvtap	19.94	15.82	−11.44		11.8	1.09	92	11.5	63.6
N2683	rob 0	9.36	8.76	11.71	5.99900	9.1	3.27	359	8.8	50.2
	uvtap	15.88	14.81	46.67		9.4	3.31	352	8.9	64.6
N2820	rob 0	9.84	8.91	−3.81	5.99900	5.3	1.699	321	5.3	48.7
	uvtap	16.18	15.36	−3.884		5.2	3.173	610	5.3	97.3
N2992	rob 0	14.33	8.84	−13.3	5.99900	8.4	65.6	7810	8.0	151.7
	uvtap	18.51	16.76	−14.26		9.5	71.7	7547	8.2	202.8
N3003	rob 0	9.37	8.93	−67.02	5.99900	9.4	1.35	144	8.6	41.1
	uvtap	13.78	13.61	16.65		6.0	8.11	1352	9.2	40.0
N3044	rob 0	10.68	9.59	−5.08	5.99900	5.1	5.92	1161	7.0	102.9
	uvtap	13.78	13.61	16.65		6.0	8.11	1352	7.0	138.4
N3079	rob 0	9.32	8.72	−13.89	5.99900	6.5	202.07	31088	5.9	1987.1
	uvtap	16.16	15.44	−9.5		8.0	221.93	27741	5.8	2759.5
N3432	rob 0	11.94	9.97	3.43	5.99900	7.2	2.33	324	7.1	39.2
	uvtap	15	13.14	−179.3		6.8	2.62	385	6.5	49.3
N3448	rob 0	9.1	8.6	3.4	5.99900	5.6	5.36	957	5.4	66.0
	uvtap	16	15.7	−167.6		6.6	8.46	1282	5.9	105.5
N3556	rob 0	9.2	8.7	7.6	5.99900	5.3	3.45	651	5.3	66.8
	uvtap	16	15.	14.1		6.0	6.9	1150	5.7	91.9
N3628	rob 0	9.62	9.47	−31.09	5.99842	9.5	66.96	7048	10.2	174.2
	uvtap	16.83	15.15	75.57		9.5	79.09	8325	9.6	128.4
N3735	rob 0	10.1	8.8	−2.3	5.99900	5.8	2.9	500	5.6	109.5
	uvtap	16.4	15.6	−5		6.2	5.3	855	5.6	220.4
N3877	rob 0	8.85	8.66	39.45	5.99900	6.5	1.43	220	6.4	25.6
	uvtap	16.07	15.27	22.61		6.7	1.94	290	6.6	33.7
N4013	rob 0	9.07	8.88	−31.87	5.99900	6.4	3.88	606	6.4	41.5
	uvtap	16.06	15.25	−7.25		6.6	4.58	694	6.5	62.5
N4096	rob 0	9	8.85	−59.54	5.99900	6.3	0.529	84	6.3	26.9
	uvtap	15.85	15.32	−16.41		6.8	1.1	162	6.6	28.7
N4157	rob 0	9.1	8.74	44.64	5.99900	6.3	1.47	3411	6.0	58.3
	uvtap	15.9	14.99	21.61		6.2	3.6	581	6.1	123.3
N4192	rob 0	9.17	8.99	−11.55	5.99900	6.0	3.807	635	6.0	150.2
	uvtap	15.66	15	83.71		6.0	5.128	855	5.9	222.2
N4217	rob 0	8.91	8.73	29.41	5.99900	6.3	2.63	417	6.3	69.4
	uvtap	16.06	15.23	17.5		6.3	4.52	717	6.2	137.5
N4244	rob 0	9.25	8.91	−4.37	5.99838	5.6	0.693	235	5.9	14.3
	uvtap	15.83	15.07	−1.57		5.8	0.84/1.39	240	5.7	18.6
N4302	rob 0	9.96	8.96	−21.41	5.99900	15.5	1.29	146	15.2	73.1
	uvtap	18.7	18.4	32.98		19.5	1.93	99	17.0	113.3
N4388	rob 0	9.9	8.99	−24.75	5.99900	13.9	18.1	1302	13.5	436.1
	uvtap	16.2	15.97	−23.7		18.0	25.17	1398	16.4	462.7
N4438	rob 0	9.83	9.17	−19.86	5.99900	16.0	22.91	1432	15.0	221.8
	uvtap n/a
N4565	rob 0	9.02	8.82	85.48	5.99834	7.4	2.64	357	7.4	49.7
	uvtap	15.96	14.83	42.18		8.0	3.11	389	7.8	105.0
N4594	rob 0	13.32	8.91	1.1	5.99841	12.9	102	7907	11.4	228.2
	uvtap	18.09	16.07	0.8		14.4	103.3	7159	13.7	245.0
N4631	rob 0	8.88	8.57	−6.88	5.99835	7.7	6.8	883	6.9	122.4
	uvtap	15.71	14.73	86.79		9.2	13.9	1511	9.6	368.6
N4666	rob 0	11.06	9.5	−179.82	5.99900	12.5	7	560	12.0	210.8
	uvtap	14.1	13.54	−176.99		12.5	10.61	849	11.5	332.0
N4845	rob 0	10.98	9.06	−1.4	5.99900	15.0	424.7	28313	14.5	337.3
	uvtap n/a
N5084	rob 0	15.64	8.35	−5.76	5.99841	7.9	29.1	3684	8.0	76.4
	uvtap	20.33	16.23	−11.18		8.6	29.4	3419	9.4	83.3
N5297	rob 0	9.03	8.78	35.28	5.99900	6.3	0.267	42	5.6	24.9
	uvtap	15.66	14.87	21.77		6.1	0.485	80	6.1	29.8
N5775	rob 0	10.08	9.34	−22.79	5.99900	5.0	4.45	890	5.0	104.0
	uvtap	16.1	14.96	75.53		5.8	7.77	1340	5.3	214.2
N5792	rob 0	10.34	9.18	−12.72	5.99900	5.3	6.183	1167	5.3	73.9
	uvtap	16.09	15.54	−87.29		5.4	9.29	1720	5.4	76.9
N5907	rob 0	9.78	8.08	16.32	5.99847	9.3	0.875/5.76	623	9.1	140.7
	uvtap	16.65	14.55	65.68		9.8	1.17/5.22	533	9.6	115.5
U10288a	rob 0	10.96	9.46	−28.47	5.99900	7.0	8.94	1277	5.4	146.6
	uvtap	13.99	13.42	−59.75		8.5	9.39	1105	5.4	169.8

Notes. Column 1: galaxy name; Column 2: weighting, where "rob 0" refers to Briggs weighting with robust value set to 0, and "uvtap" refers to a uv-tapered version of the former; Columns 3–5: synthesized beam parameters; Column 6: C-band central frequency in GHz (varying due to differences in flagging); Column 7: Stokes I rms noise; Column 8: peak intensity of the galaxy. When two values are given, the first value is the peak intensity of the galaxy, and the second higher value is the peak of the map when this occurs outside of the galaxy; Column 9: dynamic range in image (map peak intensity over noise); Column 10: Stokes Q and U average rms noise; Column 11: peak intensity of the polarization map (measured from non-PB-corrected maps).

^aMap peak and polarization peak values given for UGC 10288 are of the background source.

Download table as:  ASCIITypeset images: 1 2

Table 5.  Imaging Results for the L Band

Galaxy	Weighting	Bmaj	Bmin	BPA	${\nu }_{0}$	rms I	Map Peak	DR	rms QU	Pol. Peak
		('')	('')	(°)	(GHz)	(μJy beam−1)	(mJy beam−1)		(μJy beam−1)	(μJy beam−1)
(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
N660	rob 0	39.78	36.27	80.22	1.57502	60	422.5	7042	28	191
	uvtap	49.85	47.29	−86.89		75	441.8	5891	28	177
N891	rob 0	36.42	32.48	−74.29	1.57498	60	73.4/78.4	1307	40a	455
	uvtap	44.94	43.11	67.61		95	96.7	1018	44	647
N2613	rob 0	73.43	41.97	−29.89	1.57497	42	11.96/17.2	410	35	176
	uvtap	76.55	51.18	−32.11		45	13.4/16.9	376	40	191
N2683	rob 0	33.9	31.31	−43.8	1.57481	30	8.6	287	27	223
	uvtap	44.45	41.69	−44.23		35	11.7	334	27	218
N2820	rob 0	34.8	30.56	−32.93	1.57479	40	18.54/39.9	998	32	132
	uvtap	43.97	39.88	−36.87		40	23.6	590	32	152
N2992	rob 0	52.08	33.1	−13.55	1.57481	30	193.2	6440	26	139
	uvtap	59.34	45.81	−20.87		32	195.4	6106	27	167
N3003	rob 0	33.86	30.7	−46.95	1.57478	30	8.93	298	27	102
	uvtap	44.64	40.58	−47.84		29	11.5	397	27	92
N3044	rob 0	41.79	32.25	−48.11	1.57479	28	43.96	1570	25	246
	uvtap	50.92	42.39	−61.13		27	20.76	769	27	270
N3079	rob 0	34.88	31.49	−48.61	1.57477	45	427.3	9496	30	488
	uvtap	45.64	41.61	−48.18		45	481.2	10693	30	646
N3432	rob 0	32.62	31.82	−64.03	1.57474	36	12.16/27.83	773	32	181
	uvtap	42.34	41.43	−51.85		40	16.8/27	675	32	195
N3448	rob 0	34.55	31.4	−23.24	1.57475	35	25.46	727	26	114
	uvtap n/a
N3556	rob 0	34.45	31.31	−22.75	1.57475	38	33.87	891	25	372
	uvtap	45.91	41.74	−21.92		42	48.2	1148	26	503
N3628	rob 0	32.27	32.65	−37.4	1.57473	49	249	5082	30	30
	uvtap	47.19	42.47	−41.85		55	267	4855	28	40
N3735	rob 0	36.8	32.1	−15.7	1.57477	32	33.5	1047	25	213
	uvtap	46.6	41.1	−21.7		31	42	1355	25	250
N3877	rob 0	34.12	32.61	−31.6	1.57473	30	7.93	264	27	180
	uvtap	45.41	41.36	−27.75		37	10.2	276	25	188
N4013	rob 0	34.15	31.54	−65.96	1.57472	32	13.8	431	27	103
	uvtap	44.98	41.45	−64.79		38	15.5/24.6	647	26	116
N4096	rob 0	33.85	31.73	−35.65	1.57473	30	7.099	237	27	133
	uvtap	44.75	41.51	−34.28		32	10.33	323	26	180
N4157	rob 0	34.12	31.65	−23.32	1.57488	29	31.42/53.36	1840	28	254
	uvtap	45.19	41.48	−21.07		35	43.8/54.5	1557	27	291
N4192	rob 0	35.52	32.66	−52.17	1.57471	40	17.5	438	28	205
	uvtap n/a
N4217	rob 0	34.01	31.46	−29.04	1.57472	28	24	857	27	264
	uvtap	45.31	41.07	−26.28		30.4	33.1	1089	28	368
N4244	rob 0	33.98	32.46	−47.1	1.57471	30	2.18/9.87	329	27	114
	uvtap	45.43	41.84	−43.9		30	10.1	337	27	93
N4302	rob 0	35.82	33.55	−70.51	1.57470	46	9.73/15.32	333	35	319
	uvtap	47	42.8	−81.78		60	15.11	252	35	409
N4388	rob 0	36.32	33.24	−48.92	1.59888	150	79.92/296.2	1975	65	436
	uvtap n/a
N4438	rob 0	34.7	30.21	−53.59	1.77468	130	79.2	609	50	237
	uvtap n/a
N4565	rob 0	34.5	32.32	−89.06	1.57470	30	9.36/48	1600	27	270
	uvtap	43.27	41.92	−83.71		32	12.58/48.9	1528	24	451
N4594	rob 0	47.89	32.62	−4.6	1.57471	31	81.4	2626	25	209
	uvtap	54.41	44.46	−8.23		32	81.9	2559	25	254
N4631	rob 0	35	32.4	−89.34	1.57488	31	90.1	2906	28	353
	uvtap	32.81	41.72	−82.91		31	123	3955	33	449
N4666	rob 0	37.4	36.08	−30.42	1.57470	23	102.4	4452	25	700
	uvtap	47.95	44.7	−72.66		27	132	4981	25	865
N4845	rob 0	38.58	34.27	−5.22	1.57470	40	224.8	5620	27	556
	uvtap
N5084	rob 0	57.12	31.48	−5.85	1.57472	33	35.6	1079	29	395
	uvtap	65.86	43.94	−8.33		35	36.5	1043	30	426
N5297	rob 0	34.33	32.17	5.23	1.57471	34	4.25	125	27	138
	uvtap	45.16	41.38	−5.49		32	6.11	191	26	143
N5775	rob 0	40.35	35.4	−42.84	1.57468	35	57.5	1643	26	542
	uvtap	51.16	45.04	−63.19		33	75.5	2323	27	645
N5792	rob 0	40.13	35.23	−29.48	1.57469	32	31.92	998	25	1366
	uvtap	49.54	47.71	−39.27		35	34.4	983	25	1461
N5907	rob 0	32.99	30.67	44.87	1.57474	30	9.45/52.79	1759	28	556
	uvtap	42.49	39.93	42.4		30	13.4/31.8	1060	28	603
U10288b	rob 0	40.23	34.34	−31.32	1.57488	39	98.7	2531	35	3027
	uvtap n/a

Notes. Column 1: galaxy name; Column 2: weighting, where "rob 0" denotes Briggs weighting with robust value set to 0, and "uvtap" refers to a uv-tapered version of the former; Columns 3–5: synthesized beam parameters; Column 6: L-band central frequency in GHz (varying due to differences in flagging); Column 7: Stokes I rms noise; Column 8: peak intensity of the galaxy. When two values are given, the first value is the peak intensity of the galaxy, and the second higher value is the peak of the map; Column 9: dynamic range in image; Column 10: Stokes Q and U average rms noise; Column 11: peak intensity of the polarization map (measured from non-PB-corrected maps).

^aA previous observation (an extra 10 minutes on source) was included in the L-band Stokes Q and U imaging to increase sensitivity. It was, however, excluded from Stokes I imaging due to artifact contamination. ^bMap peak and polarization peak values given for UGC 10288 are of the background source.

Download table as:  ASCIITypeset images: 1 2

In principle, any number of Taylor terms could enter into such an expansion, with increasing numbers of terms improving the spectral fit (see Rau & Cornwell 2011 for examples). In practice, however, the effective number of terms is limited by the signal-to-noise ratio (S/N) and the calibration precision of the data. For the CHANG-ES galaxies, tests that we have run gave poorer results with three terms than with two; that is, the rms noise in the images is higher for such fits, when cleaning to the same threshold.

The curvature maps also show large and sometimes unrealistic variations, point to point. While globally averaged values of β can still be useful (see Paper II for an example), we have adopted two Taylor terms for the CHANG-ES project, as indicated earlier. A two-term Taylor fit then takes the simple form ${I}_{\nu }\;=\;{I}_{{\nu }_{0}}\;+\;{I}_{{\nu }_{0}}\;\alpha \;((\nu -{\nu }_{0})/{\nu }_{0})$. A map of random errors describing the accuracy of the fit, ${\rm{\Delta }}\;\alpha$, is also formed.¹⁸

The beauty of the spectral fitting is that it is carried out by the flux component, which means that a uniform resolution is achieved over the band, with a restoring beam applied to the TT0 and TT1 maps at the end of the fitting process. Also, the S/N of the entire band applies to each flux component. If one were to form a spectral index in the classical way within a band, one would need to break up the band into its various channels (or groups of channels), smooth each channel to the poorest resolution of the lowest band frequency, and accept the very high noise of these channels as a cutoff when forming α maps.

Clearly, the method, as implemented in CASA, is far superior for any individual band.¹⁹ Nevertheless, there are limitations, as we outline below.

Because the PB varies with frequency, it imposes its own spectral index, ${\alpha }_{\mathrm{PB}}$, onto the α maps. This effect is significant; for example, at the 50% level at L-band, ${\alpha }_{\mathrm{PB}}=-1.6$. We use the CASA task widebandpbcor, described in Section 3.3.4, to correct for ${\alpha }_{\mathrm{PB}}$, and at this point, we cut off the α maps at 5σ, where σ is the rms of the image map (i.e., the TT0 or ${I}_{{\nu }_{0}}$ map).

Once PB-corrected α maps are formed, they can show large variations around their peripheries, and the corresponding error maps, ${\rm{\Delta }}\;\alpha$, show correspondingly large errors. In addition to the 5σ cut originally applied, we also cut off all α maps wherever the corresponding error map exceeds a value of 1.0. The choice of this cutoff is arbitrary; however, various tests showed that more aggressive cuts (e.g., where ${\rm{\Delta }}\;\alpha \;\gt \;0.75$) resulted in undesirable "holes" in some α maps. Note that this step does not change the PB correction of the α maps. It simply discards data points that are known to have large errors.

Both α and ${\rm{\Delta }}\;\alpha$ maps are computed for each map pixel (cell). However, these pixels are not independent. For example, at L-band, there may typically be 64 cells beam⁻¹, and at C-band, 50 cells beam⁻¹ (both uv tapered). Consequently, the "per pixel" errors are much greater than the "per beam" errors, and the α maps, as produced, show variations pixel to pixel that are larger than the beam-averaged variations would be. In addition, maps of ${\rm{\Delta }}\;\alpha$ show artifacts, i.e., regions of low ${\rm{\Delta }}\;\alpha$ that have a "thread-like" appearance through the map. These artifacts depend on the choice of multiscales that are used during imaging and shift position if the map is remade with different scales. If the adopted scales are changed, these "threads" also shift position.

To ameliorate the above issues, we have "smoothed/averaged" our maps of α and ${\rm{\Delta }}\;\alpha$ over the size of a beam. This is accomplished by artificially inserting into the map header a Gaussian beam size whose area (Gaussian weighted) is equivalent to the pixel area and then convolving the map to the correct clean beam size.²⁰ The result does not change the spatial resolution of the images but recognizes the nonindependence of the pixels and strongly minimizes pixel-to-pixel variations. For example, the rms of an α map could decrease by as much as a factor of two after such a convolution. These are the final maps that are shown in the appendix, panels (h), (i), (k), and (l).

In summary, our final spectral index maps and all calculations performed on them apply to α after

• 1.  
  PB correction,
• 2.  
  a 5σ cutoff,
• 3.  
  a cutoff wherever ${\rm{\Delta }}\;\alpha \;\gt 1.0$, and
• 4.  
  convolving with a Gaussian to smooth over pixel-to-pixel variations within any clean beam.

As indicated above, α is determined via fitting flux components over the band during the imaging process. Fitting errors should, for the most part, be accounted for by the ${\rm{\Delta }}\;\alpha$ maps, but those maps consider random errors only. Therefore, a variety of tests have been carried out on the in-band spectral index maps to help us understand realistic uncertainties, and we outline them here. Largely qualitative descriptions are provided, with a quantitative summary at the end. In the following, we refer to the final α maps, as described above, unless otherwise indicated.

Edge Effects. In spite of the cutoffs and smoothing outlined in the previous section, values in the spectral index maps (as well as their errors) tend to become extreme at the edges (for example, see the L-band panels (i) and (l) of NGC 3735 in the appendix).

PB Correction. The NRAO-supplied PB model cannot take into account variations in the PB that are specific to any given observation. The true PB will change with flexure in the antennas and therefore with position on the sky. It will also rotate on the sky as the source is tracked, and therefore any nonaxisymmetry in the PB will also introduce error.²¹ As a result, PB-correction errors, which increase with distance from the pointing center, must be understood. Bhatnagar et al. (2013) have shown that errors in the α maps are negligible out to the half-power point but increase significantly beyond that distance (see their Figure 4, right). For L-band, our largest α map (NGC 4565) extends to a radius of 6.8 arcmin, which is well within the half-power beam width of the L-band PB (radius of 15 arcmin, or FWHM of 30 arcmin). At C-band, however, the PB radius is only 3.75 arcmin, and consequently α maps with radii larger than this will show such errors in their outer regions.

We have investigated such potential PB errors in two ways.

The first is to examine α maps using cases in which the observations of a given galaxy were carried out in two different observing sessions, such as NGC 660 (see Table 1). As such, it is possible to form two different α maps for the same galaxy and same sky pointing but observed on different days. This has allowed us to compare the resulting α maps and to compare the variations between those days with the random errors as given in the ${\rm{\Delta }}\;\alpha$ maps. We tested both low and HDR cases and compared the two results quantitatively.

The second is to examine our C-band observations of large galaxies where two offset pointings were observed (see Section 2.2.1). A comparison of α made for the individual pointings allowed us to investigate α errors at larger distances from the PB center.

As an example, Figure 1 (panel (a)) shows a map of the absolute pixel-by-pixel difference between the PB-corrected spectral index measurements in the two C-band pointings of NGC 891. An increase of the differences with distance from the map center is clearly apparent. We formed the spectral index error (shown in panel (b)) by computing the PB-weighted deviation from the average spectral index of the two pointings.²²

Zoom In Zoom Out Reset image size

This error map is for the most part comparable to the error calculated by the ms-mfs algorithm (c), as the map of the ratio of these two errors (d) illustrates. A major exception to this is the disk of the galaxy, but only in those parts that are located outside the 70% PB level (i.e., where the PB gain is less than 0.7) of either pointing. The average error ratio in these narrow regions around the midplane is ∼5 (with individual pixel values up to ∼15); that is, here the error originating from the pointing differences is on average ∼400% higher than the error resulting from the ms-mfs spectral fitting (for comparison, the average error ratio of NGC 4565 is 3.3). Such high ratios are primarily a consequence of the small ms-mfs-based errors in regions of high S/N, yet the increase of the pointing differences with distance from the pointing centers is significant. In particular, beyond the half-power point, the midplane errors in panel (b) of Figure 1 increase up to ∼0.5, whereas in panel (c) they remain below 0.1 throughout the disk, as shown by the displayed contours. While the errors in panel (b) are in rough agreement with Bhatnagar et al. (2013), in the sense that they do not exceed 0.1 out to approximately the half-power point (not considering the above-mentioned edge effects), the error ratio map suggests that the simple PB model used by the widebandpbcor task already shows significant inaccuracies at the 70% level.

Resolved-out Structures. Although the method of calculating the spectral index ensures that the resolution is common across the band, if there are structures that are resolved out at one end of the band compared to the other, this is not accounted for in the spectral index maps. For example, there may be structures resolved out at the high-frequency end of the band, but not at the low-frequency end, which would steepen the spectral index.

Comparison with Classic Spectral Index. A comparison was made between the in-band spectral index and a classically formed spectral index from a given band. The classically formed α maps were made from the spectral index maps after the PB correction was applied but not after applying further processing (i.e., PB correction and a 5σ cutoff were applied, but not a cutoff based on the error map, nor smoothing to the size of the beam, as described in Section 3.4.2); they were formed by imaging the lowest end of the band and the highest end of the band, smoothing to a common spatial resolution and then combining in the classic way. The results were in agreement within errors.

Two-pointing PB Cutoffs. For "two-pointing" galaxies (large galaxies at C-band), the maps, as indicated in Section 3.3.5, were combined over regions in which the value of the PB exceeded 0.1 (10% of the peak). When mosaicking is carried out, each point is weighted by the PB such that points that are farther from the beam center are weighted down, and therefore the difference in total intensity maps, whether one cuts off points where the PB is >10% or where the PB is >50%, is entirely negligible. However, since α maps are known to increase quite strongly below the 50% level, we remade our mosaicked α maps with a 50% cutoff to compare it to the 10% cutoff values used in this paper. For these results, we found only minor differences (for example, if the two maps are subtracted, the rms in the regions between the 50% and 10% PB cutoffs is <0.1).

In summary, our final α maps should present realistic results with the following cautions:

• a.  
  Extreme values around the edges are artifacts and should be ignored.
• b.  
  Measurements of α should not be quoted "per pixel" but rather averaged over a beam when carrying out evaluations or comparisons.
• c.  
  The calculated ${\rm{\Delta }}\;\alpha$ maps typically underestimate the true errors by $\approx$20% for galaxies of small angular size.²³
• d.  
  The largest source of uncertainty in ${\rm{\Delta }}\;\alpha$ relates to the PB model correction away from the pointing center, which most strongly affects our largest galaxies in C-band. Spectral index errors increase significantly with distance from the pointing center, such that beyond the 70% PB level of either pointing (i.e., at distances greater than 2'25'' from the respective pointing center at C-band) the ${\rm{\Delta }}\;\alpha$ maps typically underestimate the true errors in the disk by a factor of ∼5 (see Figure 1).

The data of polar ring galaxy NGC 660 have the highest dynamic range in both bands, due to a strong central source. Consequently, the resulting images have higher than expected rms values, particularly in L-band. Nevertheless, the achieved signal to noise of the C-band map with robust weighting reached 70,000, on the higher end of what best can be expected (see Section 3.3.2).

In spite of careful cleaning, including peeling²⁶ of the nearest of the two strongly interfering field sources in L-band (see Section 3.3.6), artifacts are still present. This could potentially be affecting the spectral index results, where regions of flat spectral index are crossing the disk. However, these regions could also be an effect of the possible AGN (see Section 5.2.1).

We note that the polarization intensity is low, with its fraction of polarization over Stokes I intensity just a 10th of the 0.5% that is considered a believable signal (see Section 3.3.3).

Both of these galaxies are well known to have significant halos, which we can also see in our data. Moreover, they both show very flat spectral indices at C-band throughout the disk, where the errors in spectral indices are the lowest. Neither data set showed any particular problems and were single pointings. Bearing in mind the summary regarding spectral index in Section 3.4.3, this suggests that thermal emission may be more strongly dominant for these galaxies in their disks.

For NGC 5775 at L-band, some broad-scale polarization features have been found in the field that are most likely from foreground Galactic emission (this is more obvious when a larger field than shown in the Appendix is displayed). The Galactic coordinates of this source place it almost directly over the Galactic center, although at reasonably high Galactic latitude. Indeed, the Galactic coordinates of NGC 5775 $(l=359\buildrel{\circ}\over{.} 4,b=52\buildrel{\circ}\over{.} 4)$ indicate that we view it through the northernmost tip of the Fermi bubbles that extend 55° above and below the Galactic center (Ackermann et al. 2014). Substantial polarized emission at 2.3 and 23 GHz has been found to coincide with the Fermi bubbles, including ridge-like filamentary structures crossing through this particular location (Carretti et al. 2013). Our polarization images are likely picking up this same extended foreground structure at lower frequency.

These galaxies are highly affected by contamination from strong sources in the field, such as M87, rendering higher than normal rms noise and artifacts hard to clean out. This may have an effect on both spectral index results and polarization.

We note that the C-band spectral index is steep in the inner disk for NGC 4192 (a rather weak source at the low end of our flux density cutoff for the survey) and that the uncertainties are high.

NGC 4388 was particularly affected by M87, and we made an attempt at peeling the source in L-band, as well as self calibrating on M84 (the other disturbing source in the field), which rendered some improvement. The resulting rms at L-band was too high for us to detect significant polarized flux (panel (f)).

Also, NGC 4438 is strongly affected by residual side lobes from M87. Despite attempts of self calibration and peeling, the rms could not be brought down to lower than almost three times the expected value. L-band observations of NGC 4438 differ from the other galaxies, in that only the upper half of the frequency range (spectral windows 16–31) was used for imaging, with a central frequency of 1.77 GHz. Q and U images were made over the same upper half of the frequency range in order to match the total intensity image, as well as the spectral index image.

The strong center of NGC 4594 (the Sombrero galaxy) resulted in residual side lobes that made it difficult to detect the weak east–west disk for the two-pointing data set in C-band. Careful selection of self-calibration inputs eventually revealed the disk, and subsequent merging of the two pointings further helped to cancel out artifacts from the center.

In L-band, a plume is seen to the north of the core and is roughly in the direction of the jet observed at much higher resolution by Hada et al. (2013). An unusual polarization structure and a flat spectral index associated with the core and toward the north are also observed. A more conservative spectral index cutoff could be beneficial for this data set (for example, a 10σ cutoff was adopted for the in-depth study of NGC 4845; see Irwin et al. (2015)).

The C-band spectral index shows an unexpected flattening from the central to the outer disk for this two-pointing data set. This gradual deviation has not been detected in previous observations of this galaxy (Hummel & Dettmar 1990), but it is within the errors pointed out in point (d) of Section 3.4.3.

Despite its small angular size well within the symmetric PB, the C-band spectral index displays an asymmetry between the two halves of the disk, such that it is flatter on the southwest side compared to the northeast. This asymmetry corresponds to an asymmetry in the polarization at the same frequency. Interaction with a nearby companion, NGC 4668 (to the southeast) could be a factor, or motion through an intergalactic medium. No other technical issues (antenna flexure, solar influence) are problematic.

It is worthwhile to note that this galaxy displays an interesting variation in flux density level, and further exploration, being published in CHANG-ES V (Irwin et al. 2015), indicates with little doubt that the source is indeed variable. CHANG-ES V also explores the spectral index and polarization results in more detail.

Due to the low declination of this galaxy, the C-band observations were divided up into three scheduling blocks in order to (1) avoid shadowing and (2) increase uv coverage. The scheduling block in which the first pointing was observed did not have a polarization leakage calibrator scan, and instead the secondary calibrator used for NGC 4192 was used to calibrate polarization leakage (the secondary calibrator of NGC 5084 did not have a sufficiently large parallactic angle span). Part of the second pointing was strongly affected by shadowing, and only 21 antennas were used for these scans.

The C-band images show emission extending east and west of the point-like center, as well as southward from the eastern extension; these are likely real because their intensity is greater than 10σ (and they roughly follow the disk). However, the extensions seen above and below the center may be spurious because they seem to align with weak residual side lobes. In L-band, we find no apparent polarization from the source. The spectral index is flat in L-band and flat to slightly positive in C-band (but the positive trend coincides with higher error values).

UGC 10288 fell within the intensity cutoff of the CHANG-ES sample by piggybacking on the previously unresolved strong background source to the west of the disk. We refer to Paper III for full details on the observations and analysis of this galaxy.

About 10 galaxies have bright nuclei at 22 μm, raising the possibility of significant AGN contamination. To give an idea of the potential contamination, we refer to the study of the AGN contribution to mid-infrared (MIR) continuum emission based on observations of the nuclei of nearby AGNs by Tommasin et al. (2010). They find lower limits of 45%–73% for the AGN contribution to the 19 μm continuum, depending on the type of AGN. For cases where we believe an AGN to be present, we crudely correct our SFRs by assuming that 100% of the nuclear 22 μm emission arises from the AGN. A literature search suggests that all 10 candidates are starburst nuclei except for two, NGC 3735 and NGC 4388, with perhaps the most compelling evidence coming from the IR line ratio study of nuclei by Pereira-Santaella et al. (2010).

We explored this issue further by making a rough determination of the nuclear 3.4–4.6 μm color from the WISE WERGA images, a measure that distinguishes well between AGNs and starbursts (Jarrett et al. 2011, Figure 26, and references therein). We estimated the color of the central source in two different ways, both of which include simplifying approximations. The resolutions are 59 and 65 at 3.4 and 4.6 μm, respectively. First, we fit a two-dimensional (2D) Gaussian with a constant background to the central peak in each band. We tried this with two different box sizes: 9'' × 9'' and 12'' × 12''. We then turned the flux density ratio of the Gaussians into a color. The second method was to Gaussian-convolve the 3.4 μm images to the 4.6 μm resolution, make a color map, and examine the color in a central box of size 6'' × 6''. The reason these methods are approximate is first that the PSFs have significant Airy disks, so some of the flux density is not in the central peak and not described by a Gaussian, and second that a constant background is not necessarily the best assumption in the first method since the galaxy has structure as it cuts through the fitting box. However, the results are reasonably consistent from the two methods.

Although Figure 26 of Jarrett et al. (2011) represents a range of redshifts, our resulting colors clearly fall into either the spiral/starburst or AGN regions, and the classifications are generally consistent with the evidence from the literature, with two exceptions.

The first exception is NGC 3735, which has a starburst-like nuclear color of about 0.0–0.3 mag (the range is from the two different methods and shows the greatest discrepancy for any of the 10 galaxies), whereas MIR line ratios from Pereira-Santaella et al. (2010) clearly indicate an AGN. In this case, we tend to favor the MIR over the near-infrared (NIR) evidence because of the possibility of extinction hiding the NIR emission from the AGN in an edge-on geometry. If it is an AGN, we can only estimate the SFR by measuring the disk flux density outside a radius of 13''. This reduces the SFR to $1.1{M}_{\odot }\;{\mathrm{yr}}^{-1}$, although strictly this is a lower limit.

The second exception is NGC 660, where a color of 1.2–1.3 clearly indicates an AGN. If indeed an AGN, the nuclear flux density can be crudely subtracted via a Gaussian fit to the central peak in the 22 μm image, leaving a lower limit SFR of 0.61–0.81 ${M}_{\odot }\;{\mathrm{yr}}^{-1}$, depending on the fit parameters. Yet the literature (e.g., Bernard-Salas et al. 2009; Pereira-Santaella et al. 2010) indicates a starburst, as well as a LINER galaxy. Thus, we will continue assuming NGC 660 to be a starburst rather than an AGN galaxy, and we retain the originally derived values in Table 6.

The nucleus of NGC 4388 does appear to be AGN-dominated (see also, e.g., Falcke et al. 1998), with a color of 1.1 mag. The 22 μm flux density is dominated by this source, and the faint disk emission is not well resolved from it. Hence, to estimate the SFR, we turn to the Spitzer 24 μm image, where the disk is better resolved. However, in this case the nuclear source so overwhelms the disk that the uncertainty in the flux density of a Gaussian fit precludes an estimate of a disk SFR, as was done for NGC 660. At best, we can crudely estimate a disk flux density outside a radius of 7'' from the nucleus. Hence, the SFR of 0.07 ${M}_{\odot }\;{\mathrm{yr}}^{-1}$ is a lower limit.