THE ABUNDANCE PROPERTIES OF NEARBY LATE-TYPE GALAXIES. I. THE DATA

A new method (called the "C method") for oxygen and nitrogen abundance determinations has recently been suggested (Pilyugin et al. 2012). The idea of the C method is very simple. We have compared several combinations of the strong-line intensities in the spectrum of a given H ii region with those in the spectra of a sample of reference H ii regions with known abundances in order to find the counterpart for the H ii region under study. It is assumed that the oxygen and nitrogen abundances in the studied H ii region are the same as in its counterpart. A counterpart can be selected by comparison of four combinations of strong-line intensities: P = R₃/(R₂ + R₃) (excitation parameter), logR₃, log(N₂/R₂), and log(S₂/R₂).

However, there are recent measurements of spectra of many H ii regions where the intensities of [O ii] λ3727+λ3729 or [S ii] λ6717, [S ii] λ6731 lines are not available (e.g., in Sánchez et al. (2012) or in the Sloan Digital Sky Survey (SDSS); see York et al. 2000). It has been argued that the oxygen and nitrogen abundances in H ii regions can be estimated even if the [S ii] λ6717+λ6731 emission line is not measured (Pilyugin et al. 2010) or if the [O ii] λ3727+λ3729 emission line is not available (Pilyugin & Mattsson 2011). The C method can be adapted to such cases (Pilyugin et al. 2013). To find the counterpart for the H ii region under study, one does not need to compare the four combinations of strong-line intensities, but can instead also use only three combinations: (1) logR₃, P, and log(N₂/R₂), or (2) logR₃, logN₂, and log(N₂/S₂). When this first combination of strong-line intensities is used to find the counterpart then the resulting oxygen and nitrogen abundances will be referred to as (O/H)$_{C_{\rm ON }}$ and (N/H)$_{C_{\rm ON}}$. When the second combination of strong-line intensities is used to find the counterpart then the inferred oxygen and nitrogen abundances will be called (O/H)$_{C_{\rm NS}}$ and (N/H)$_{C_{\rm NS}}$.

The data for reference H ii regions with T_e-based abundances are compiled in Pilyugin et al. (2012). The very recent spectroscopic observations of Berg et al. (2012), Zurita & Bresolin (2012), and Skillman et al. (2013) have been added to the compilation. Using these combined data we select a sample of reference H ii regions for which all the absolute differences for oxygen abundances (O/H)$_{C_{\rm ON}}$–(O/H)$_{T_{e}}$ and (O/H)$_{C_{\rm NS}}$–(O/H)$_{T_{e}}$ and for nitrogen abundances (N/H)$_{C_{\rm ON}}$–(N/H)$_{T_{e}}$ and (N/H)$_{C_{\rm NS}}$–(N/H)$_{T_{e}}$ are less than 0.1 dex. This sample of reference H ii regions contains 250 objects and will in the following be used for abundance determinations. This sample will be referred to as the E2013 sample (etalon sample 2013) below.

Furthermore, only blue spectra were observed for H ii regions in a number of galaxies (e.g., Oey & Kennicutt 1993; Zaritsky et al. 1994; Werk et al. 2011), i.e., intensities of [N ii] λ6584 and [S ii] λ6717, [S ii] λ6731 are not available in those cases. The oxygen abundances in those H ii regions can be estimated through the P calibration where only the oxygen [O ii] λ3727+λ3729 and [O iii] λ5007 lines are used (Pilyugin 2000, 2001; Pilyugin & Thuan 2005). We have constructed a new variant of the P calibration. The sample of reference H ii regions, E2013, has been used as calibration data points. To enlarge the number of calibration data points, we have added a number of H ii regions with (O/H)$_{C_{\rm ON}}$ abundances that were chosen in the following way. We determined the (O/H)$_{C_{\rm ON}}$ abundances in H ii regions and obtained radial oxygen abundance gradients across the disks of galaxies (see below). The H ii regions where the deviations of the (O/H)$_{C_{\rm ON}}$ abundances from the general radial abundance trend are less than 0.1 dex were added to the sample of reference H ii regions.

It is well known that the relation between the oxygen abundance and the strong oxygen line intensities is double-valued, with two distinct parts traditionally known as the upper and lower branches of the R₂₃–O/H diagram. We have delimited the upper and lower branches and the transition zone, adopting 12+log(O/H) = 8.3 as the boundary between the upper branch and the transition zone and 12+log(O/H) = 8.0 as the boundary between the transition zone and the lower branch. These boundaries are somewhat arbitrary, but were chosen so as to give the best calibrations with the existing data. Two distinct relations between the oxygen abundance and the strong oxygen line intensities will be established in the following, one for the upper branch (the high-metallicity calibration) and one for the lower branch (the low-metallicity calibration).

The relation between the oxygen abundance Z_P ≡ 12+log (O/H)_P and R₃ and P can be fitted by a polynomial of the form (Pilyugin 2001; Pilyugin & Thuan 2005)

$$\begin{equation} Z_P = k_{0} + k_{1} \log R_3 + k_{2} (\log R_3)^2, \end{equation} \tag{ 1 }$$

where we have used the notation Z ≡ 12+log (O/H) for brevity. To take into account the dependence on the excitation parameter P, the coefficients of Equation (1) are chosen to have the form

$$\begin{equation} k_{j} = a_{j} + b_{j} P. \end{equation} \tag{ 2 }$$

The coefficients a₀, a₁, a₂, b₀, b₁, and b₂ can then be determined by looking for the best fit to our sample of H ii regions. We wish to derive a set of coefficients in Equation (1) which gives the minimum value of 〈Δ(O/H)〉 = $\sqrt{\vphantom{A^A}\smash{\hbox{$(\sum _{j=1}^n (\Delta ({\rm O/H})_j)^2)/n$}}}$. Here Δ(O/H)_j is equal to log(O/H)_{P, j}—log(O/H)$_{T_{e},C_{{\rm ON}},j}$ for each H ii region in our sample. The quantity 〈Δ(O/H)〉 is the average value of the differences between the oxygen abundances determined through the P calibration and the original ones. A few data points with large deviations, in excess of 0.15 dex, are rejected, and are not used in the determination of the final relation.

The obtained upper-branch P calibration (for 12 +log(O/H) ≳ 8.3) is

$$\begin{eqnarray} Z_P & = & 8.334 + 0.533\,P\nonumber \\ & & - (0.338 + 0.415\,P)\,\log R_3 \\ & & - (0.086-0.225\,P)\,(\log R_3)^2.\nonumber \end{eqnarray} \tag{ 3 }$$

The obtained lower-branch P calibration (for 12 +log(O/H) ≲ 8.0) is

$$\begin{eqnarray} Z_P & = & 7.949 - 1.328\,P \nonumber \\ && + (0.926 - 0.440\,P)\,\log R_3 \\ && + (1.220-0.480\,P)\,(\log R_3)^2.\nonumber \end{eqnarray} \tag{ 4 }$$

Hence, here we use the empirical metallicity scale defined by the H ii regions with abundances derived through the direct method (T_e method).

Our final list includes 130 galaxies. Table 1 lists the general characteristics of each galaxy. The first column gives its name. We have used the most widely used name for each galaxy. The galaxies are listed in order of name category, with the following categories in descending order:

NGC—New General Catalogue,

IC—Index Catalogue,

UGC—Uppsala General Catalog of Galaxies,

PGC—Catalogue of Principal Galaxies.

The morphological type of the galaxy and morphological type code T from LEDA are reported in Columns 2 and 3. The right ascension (R.A.) and declination (Decl.) (J2000.0) of each galaxy are given in Columns 4 and 5. The right ascension and declination are taken from the NASA/IPAC Extragalactic Database (NED)⁶. The isophotal radius R₂₅ in arcmin of each galaxy is reported in Column 6. Unless otherwise stated, the R₂₅ values are taken from de Vaucouleurs et al. (1991, thereafter RC3). The position angle and inclination are listed in Columns 7 and 8, and the sources for these values are given in Column 9. The adopted distance and its reference are reported in Columns 10 and 11. The NED distances use flow corrections for Virgo, the Great Attractor, and Shapley Supercluster infall. The isophotal radius in kpc, estimated from the data in Columns 6 and 10, is listed in Column 12.

We have carried out an extensive search of the literature and compiled a sample of measurements of H ii regions in nearby late-type galaxies. We have searched for spectra of H ii regions with the requirement that they include the [O ii] λ3727+λ3729, [O iii] λ5007, [N ii] λ6584, and [S ii] λ6717+λ6731 lines. While we have tried to include as many sources as possible, we do not claim our search to be exhaustive.

Thus, for each listed spectrum, we record the measured values of [O ii] λ3727+λ3729, [O iii] λ5007, [N ii] λ6584, [S ii] λ6717, and [S ii] λ6731. The intensities of all lines are normalized to the Hβ line flux. The predicted values of the flux ratio of oxygen [O iii] λ5007/[O iii] λ4959 and nitrogen [N ii] λ6584/[N ii] λ6548 lines are very close to three (Storey & Zeippen 2000). The measurements of the [O iii] λ5007 and λ4959 lines in SDSS spectra confirm this value of the flux ratio (e.g., Kniazev et al. 2004). Therefore, the value of R₃ can be estimated without [O iii] λ4959 line as R₃ = 1.33[O iii] λ5007, and, similarly, the values of N₂ are estimated without the lines [N ii] λ6548 as N₂ = 1.33[N ii] λ6584.

We have taken the de-reddened line intensities as reported by the authors. In some papers only the measured fluxes are reported. In these cases, the measured emission-line fluxes were corrected for interstellar reddening using the theoretical Hα to Hβ ratio (i.e., the standard value of Hα/Hβ = 2.86) and the analytical approximation to the Whitford interstellar reddening law from Izotov et al. (1994).

The spectroscopic data so assembled form the basis of this study. Our total list contains 3904 spectra including 162 spectra of H ii regions beyond the isophotal radius R₂₅. Here we will examine the radial oxygen and nitrogen abundance distributions within the isophotal radius in every galaxy. The radial oxygen and nitrogen abundances beyond the isphotal radius will be discussed elsewhere.

When measurements of the lines [O ii] λ3727+λ3729, [O iii] λ5007, and [N ii] λ6584 were available, the oxygen (O/H)$_{C_{\rm ON}}$ and nitrogen (N/H)$_{C_{\rm ON}}$ abundances in the H ii regions were estimated and used in our study. If the intensity of the line [O ii] λ3727+λ3729 was not measured but the measurements of the lines [O iii] λ5007, [N ii] λ6584, [S ii] λ6717, and [S ii] λ6731 were available, then the oxygen (O/H)$_{C_{\rm NS}}$ and nitrogen (N/H)$_{C_{\rm NS}}$ abundances of the H ii regions were estimated and used. If the measurements of only the oxygen lines [O ii] λ3727+λ3729 and [O iii] λ5007 were available (blue spectrum was observed only) then the oxygen (O/H)_P abundance of the H ii regions was estimated and adopted.

The deprojected radii of the H ii regions were computed using their coordinates (or offsets from the nucleus) as reported in the original papers, as were the position angle and inclination listed in Table 1. In some publications, the positions of the observed H ii regions (or their identifications in catalogs) were not reported, but the deprojected radii were listed instead. In these cases these deprojected radii were used (after correction for the galaxy distance adopted here, if necessary). The fractional radii (normalized to the optical isophotal radius R₂₅) were obtained with isophotal radii from Table 1.

The radial oxygen abundance distribution within the isophotal radius in every galaxy was fitted by the following equation:

$$\begin{equation} 12+\log ({\rm O/H}) = 12+\log ({\rm O/H})_{R_{0}} + C_{{\rm O/H}} \times (R/R_{25}), \end{equation} \tag{ 5 }$$

where 12+log(O/H)$_{R_{0}}$ is the oxygen abundance at R₀ = 0, i.e., the extrapolated central oxygen abundance, C_O/H, is the slope of the oxygen abundance gradient expressed in terms of dex $R_{\rm 25}^{-1}$, and R/R₂₅ is the fractional radius (the galactocentric distance normalized to the disk's isophotal radius R₂₅). We also determined the oxygen abundance gradient expressed in terms of dex kpc⁻¹. If there were data points with large deviations, in excess of 0.2 dex, those points were rejected, and were not used in the determination of the final relation.

The derived parameters of the oxygen abundance distributions are presented in Table 2. The name of the galaxy is listed in Column 1. The extrapolated central 12+log(O/H)$_{R_{0}}$ oxygen abundance and the gradient (the coefficient C_O/H in Equation (5)) expressed in terms of dex $R_{\rm 25}^{-1}$ are listed in Columns 2 and 3. The oxygen abundance gradient expressed in terms of dex kpc⁻¹ is listed in Column 4. The scatter of oxygen abundances around the general radial oxygen abundance trend is reported in Column 5.

Table 2. The Derived Parameters of the Radial Oxygen and Nitrogen Abundance Distributions in Galaxies

Galaxy	12+log(O/H)$_{R_{0}}$	O/H Gradient	O/H Gradient	σ(O/H)	12+log(N/H)$_{R_{0}}$	N/H Gradient	N/H Gradient	σ(N/H)
		dex $R_{25}^{-1}$	dex kpc−1	dex		dex $R_{25}^{-1}$	dex kpc−1	dex
NGC 0012	8.56 ± 0.02	0.015 ± 0.042	0.0011 ± 0.0032	0.033	7.74 ± 0.04	−0.008 ± 0.087	−0.0006 ± 0.0066	0.068
NGC 0055	8.04 ± 0.05	−0.053 ± 0.094	−0.0059 ± 0.0104	0.075	6.50 ± 0.07	−0.037 ± 0.131	−0.0042 ± 0.0145	0.091
NGC 0099	8.55 ± 0.04	−0.319 ± 0.072	−0.0202 ± 0.0045	0.055	7.70 ± 0.07	−0.666 ± 0.137	−0.0422 ± 0.0087	0.105
NGC 0224	8.72 ± 0.02	−0.352 ± 0.039	−0.0172 ± 0.0019	0.077	8.17 ± 0.04	−0.996 ± 0.072	−0.0486 ± 0.0035	0.136
NGC 0234	8.65 ± 0.01	−0.154 ± 0.022	−0.0109 ± 0.0015	0.027	7.91 ± 0.02	−0.329 ± 0.046	−0.0234 ± 0.0033	0.058
NGC 0253	8.55 ± 0.13	−0.125 ± 0.164	−0.0090 ± 0.0119	0.055	7.60 ± 0.35	−0.241 ± 0.441	−0.0174 ± 0.0319	0.149
NGC 0300	8.51 ± 0.02	−0.519 ± 0.040	−0.0842 ± 0.0065	0.069	7.47 ± 0.04	−0.955 ± 0.080	−0.1548 ± 0.0129	0.130
NGC 0428	8.20 ± 0.06	−0.035 ± 0.109	−0.0040 ± 0.0124	0.059	6.95 ± 0.11	−0.118 ± 0.184	−0.0135 ± 0.0210	0.099
NGC 0450	8.35 ± 0.03	−0.285 ± 0.057	−0.0340 ± 0.0067	0.046	7.24 ± 0.06	−0.589 ± 0.112	−0.0702 ± 0.0133	0.092
NGC 0493	8.51 ± 0.04	−0.450 ± 0.070	−0.0393 ± 0.0061	0.053	7.53 ± 0.07	−0.893 ± 0.124	−0.0780 ± 0.0109	0.093
NGC 0575	8.58 ± 0.01	−0.195 ± 0.026	−0.0189 ± 0.0025	0.026	7.82 ± 0.03	−0.530 ± 0.061	−0.0514 ± 0.0059	0.060
NGC 0598	8.48 ± 0.02	−0.359 ± 0.041	−0.0459 ± 0.0053	0.073	7.47 ± 0.04	−0.691 ± 0.084	−0.0883 ± 0.0107	0.145
NGC 0628	8.78 ± 0.01	−0.572 ± 0.027	−0.0379 ± 0.0018	0.060	8.17 ± 0.03	−1.292 ± 0.064	−0.0856 ± 0.0042	0.137
NGC 0753	8.67 ± 0.08	−0.212 ± 0.106	−0.0087 ± 0.0043	0.065	8.01 ± 0.20	−0.399 ± 0.290	−0.0163 ± 0.0118	0.177
NGC 0783	8.68 ± 0.07	−0.189 ± 0.104	−0.0118 ± 0.0056	0.031	7.91 ± 0.30	−0.274 ± 0.434	−0.0171 ± 0.0272	0.130
NGC 0925	8.48 ± 0.02	−0.473 ± 0.037	−0.0334 ± 0.0026	0.067	7.45 ± 0.04	−0.941 ± 0.071	−0.0665 ± 0.0050	0.128
NGC 1055	8.62 ± 0.04	−0.140 ± 0.100	−0.0088 ± 0.0063	0.059	7.89 ± 0.05	−0.414 ± 0.118	−0.0261 ± 0.0074	0.070
NGC 1058	8.62 ± 0.01	−0.352 ± 0.027	−0.0757 ± 0.0058	0.055	7.72 ± 0.03	−0.751 ± 0.069	−0.1613 ± 0.0149	0.137
NGC 1068	8.64 ± 0.02	−0.121 ± 0.051	−0.0116 ± 0.0049	0.070	7.94 ± 0.04	−0.141 ± 0.100	−0.0136 ± 0.0096	0.126
NGC 1090	8.61 ± 0.03	−0.284 ± 0.103	−0.0137 ± 0.0050	0.026	7.89 ± 0.04	−0.775 ± 0.131	−0.0373 ± 0.0063	0.033
NGC 1097	8.74 ± 0.01	−0.357 ± 0.054	−0.0159 ± 0.0024	0.030	8.25 ± 0.05	−0.997 ± 0.209	−0.0445 ± 0.0093	0.114
NGC 1232	8.78 ± 0.04	−0.573 ± 0.068	−0.0253 ± 0.0030	0.040	8.12 ± 0.10	−1.166 ± 0.172	−0.0515 ± 0.0076	0.103
NGC 1313	8.21 ± 0.03	−0.159 ± 0.051	−0.0273 ± 0.0087	0.071	6.84 ± 0.05	−0.269 ± 0.096	−0.0462 ± 0.0165	0.132
NGC 1365	8.64 ± 0.02	−0.253 ± 0.033	−0.0079 ± 0.0010	0.064	7.91 ± 0.04	−0.622 ± 0.080	−0.0193 ± 0.0024	0.137
NGC 1512	8.79 ± 0.03	−0.461 ± 0.048	−0.0374 ± 0.0039	0.066	8.42 ± 0.05	−1.410 ± 0.083	−0.1144 ± 0.0067	0.115
NGC 1598	8.74 ± 0.03	−0.402 ± 0.062	−0.0343 ± 0.0053	0.042	8.20 ± 0.09	−1.014 ± 0.178	−0.0865 ± 0.0152	0.122
NGC 1637	8.64 ± 0.01	−0.115 ± 0.023	−0.0165 ± 0.0034	0.062	7.78 ± 0.03	−0.170 ± 0.067	−0.0236 ± 0.0096	0.160
NGC 1642	8.64 ± 0.01	−0.154 ± 0.024	−0.0094 ± 0.0015	0.030	7.86 ± 0.02	−0.244 ± 0.050	−0.0148 ± 0.0031	0.062
NGC 1672	8.67 ± 0.02	−0.216 ± 0.070	−0.0135 ± 0.0044	0.055	8.00 ± 0.05	−0.464 ± 0.158	−0.0290 ± 0.0098	0.119
NGC 2336	8.80 ± 0.04	−0.390 ± 0.056	−0.0156 ± 0.0022	0.041	8.14 ± 0.08	−0.725 ± 0.121	−0.0290 ± 0.0049	0.088
NGC 2403	8.48 ± 0.02	−0.524 ± 0.043	−0.0538 ± 0.0044	0.079	7.47 ± 0.03	−1.005 ± 0.069	−0.1032 ± 0.0071	0.128
NGC 2441	8.54 ± 0.01	0.017 ± 0.040	0.0011 ± 0.0027	0.036	7.67 ± 0.03	0.100 ± 0.080	0.0068 ± 0.0054	0.072
NGC 2442	8.64 ± 0.05	−0.079 ± 0.072	−0.0058 ± 0.0052	0.030	7.87 ± 0.14	−0.119 ± 0.204	−0.0087 ± 0.0150	0.087
NGC 2541	8.45 ± 0.02	−0.608 ± 0.053	−0.0526 ± 0.0046	0.030	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 2805	8.51 ± 0.03	−0.432 ± 0.045	−0.0168 ± 0.0017	0.040	7.53 ± 0.06	−0.869 ± 0.096	−0.0338 ± 0.0037	0.087
NGC 2835	8.50 ± 0.04	−0.314 ± 0.060	−0.0324 ± 0.0062	0.057	7.51 ± 0.10	−0.633 ± 0.149	−0.0652 ± 0.0154	0.142
NGC 2841	8.99 ± 0.08	−0.607 ± 0.172	−0.0363 ± 0.0103	0.022	8.81 ± 0.20	−1.488 ± 0.450	−0.0890 ± 0.0269	0.052
NGC 2903	8.82 ± 0.03	−0.523 ± 0.072	−0.0321 ± 0.0044	0.061	8.51 ± 0.06	−1.588 ± 0.152	−0.0975 ± 0.0093	0.121
NGC 2997	8.80 ± 0.06	−0.502 ± 0.121	−0.0318 ± 0.0076	0.060	8.44 ± 0.13	−1.503 ± 0.274	−0.0950 ± 0.0173	0.129
NGC 3020	8.39 ± 0.02	−0.558 ± 0.041	−0.0554 ± 0.0041	0.011	7.27 ± 0.07	−1.053 ± 0.167	−0.1045 ± 0.0167	0.045
NGC 3023	8.34 ± 0.08	−0.444 ± 0.201	−0.0361 ± 0.0163	0.054	7.20 ± 0.14	−0.889 ± 0.372	−0.0722 ± 0.0302	0.100
NGC 3031	8.58 ± 0.02	−0.156 ± 0.045	−0.0110 ± 0.0032	0.043	7.84 ± 0.06	−0.464 ± 0.116	−0.0327 ± 0.0081	0.110
NGC 3184	8.66 ± 0.02	−0.176 ± 0.037	−0.0130 ± 0.0027	0.066	7.86 ± 0.04	−0.338 ± 0.086	−0.0249 ± 0.0063	0.137
NGC 3198	8.60 ± 0.04	−0.425 ± 0.076	−0.0237 ± 0.0042	0.063	7.94 ± 0.11	−1.145 ± 0.317	−0.0638 ± 0.0176	0.091
NGC 3227	8.64 ± 0.05	−0.254 ± 0.077	−0.0158 ± 0.0048	0.056	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 3239	8.12 ± 0.01	−0.153 ± 0.012	−0.0260 ± 0.0021	0.005	6.78 ± 0.04	−0.275 ± 0.064	−0.0467 ± 0.0108	0.026
NGC 3310	8.37 ± 0.01	−0.101 ± 0.023	−0.0117 ± 0.0027	0.040	7.27 ± 0.02	−0.254 ± 0.060	−0.0295 ± 0.0070	0.104
NGC 3319	8.50 ± 0.02	−0.516 ± 0.043	−0.0408 ± 0.0034	0.024	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 3344	8.72 ± 0.02	−0.618 ± 0.039	−0.0984 ± 0.0063	0.029	8.08 ± 0.17	−1.413 ± 0.252	−0.2254 ± 0.0407	0.044
NGC 3351	8.82 ± 0.01	−0.210 ± 0.031	−0.0186 ± 0.0027	0.038	8.51 ± 0.03	−0.678 ± 0.094	−0.0599 ± 0.0083	0.116
NGC 3359	8.40 ± 0.06	−0.271 ± 0.108	−0.0179 ± 0.0071	0.099	7.43 ± 0.09	−0.688 ± 0.154	−0.0453 ± 0.0102	0.130
NGC 3486	8.60 ± 0.02	−0.421 ± 0.043	−0.0447 ± 0.0046	0.018	7.80 ± 0.04	−0.928 ± 0.088	−0.0985 ± 0.0094	0.036
NGC 3521	8.83 ± 0.04	−0.666 ± 0.096	−0.0390 ± 0.0056	0.048	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 3621	8.68 ± 0.02	−0.460 ± 0.035	−0.0446 ± 0.0034	0.050	8.05 ± 0.05	−1.298 ± 0.095	−0.1260 ± 0.0092	0.133
NGC 3631	8.71 ± 0.03	−0.414 ± 0.077	−0.0256 ± 0.0048	0.024	7.94 ± 0.05	−0.719 ± 0.145	−0.0445 ± 0.0090	0.046
NGC 3718	8.60 ± 0.19	−0.221 ± 0.253	−0.0119 ± 0.0136	0.066	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 3820	8.64 ± 0.01	−0.102 ± 0.019	−0.0116 ± 0.0022	0.019	7.92 ± 0.05	−0.342 ± 0.093	−0.0389 ± 0.0105	0.092
NGC 3893	8.73 ± 0.16	−0.437 ± 0.254	−0.0371 ± 0.0215	0.028	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 3938	8.79 ± 0.05	−0.577 ± 0.092	−0.0413 ± 0.0066	0.031	8.15 ± 0.07	−1.204 ± 0.141	−0.0861 ± 0.0101	0.048
NGC 4030	8.77 ± 0.02	−0.258 ± 0.036	−0.0156 ± 0.0022	0.010	8.05 ± 0.08	−0.344 ± 0.177	−0.0208 ± 0.0107	0.052
NGC 4088	8.71 ± 0.04	−0.366 ± 0.093	−0.0305 ± 0.0078	0.025	7.97 ± 0.06	−0.698 ± 0.116	−0.0582 ± 0.0097	0.032
NGC 4109	8.63 ± 0.03	−0.402 ± 0.099	−0.0269 ± 0.0066	0.056	7.87 ± 0.05	−1.152 ± 0.154	−0.0770 ± 0.0103	0.081
NGC 4254	8.77 ± 0.02	−0.353 ± 0.034	−0.0316 ± 0.0031	0.036	8.09 ± 0.05	−0.716 ± 0.100	−0.0641 ± 0.0090	0.106
NGC 4258	8.54 ± 0.03	−0.161 ± 0.073	−0.0083 ± 0.0037	0.031	7.59 ± 0.13	−0.232 ± 0.282	−0.0119 ± 0.0144	0.120
NGC 4303	8.78 ± 0.03	−0.577 ± 0.073	−0.0452 ± 0.0057	0.062	8.24 ± 0.08	−1.370 ± 0.189	−0.1073 ± 0.0148	0.156
NGC 4321	8.74 ± 0.03	−0.219 ± 0.048	−0.0118 ± 0.0026	0.038	8.04 ± 0.06	−0.355 ± 0.104	−0.0192 ± 0.0056	0.083
NGC 4395	8.19 ± 0.05	−0.164 ± 0.082	−0.0200 ± 0.0100	0.079	6.89 ± 0.08	−0.346 ± 0.143	−0.0420 ± 0.0174	0.136
NGC 4490	8.29 ± 0.08	−0.318 ± 0.254	−0.0444 ± 0.0355	0.058	7.09 ± 0.15	−0.560 ± 0.449	−0.0781 ± 0.0629	0.102
NGC 4501	8.92 ± 0.12	−0.424 ± 0.242	−0.0299 ± 0.0171	0.054	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 4535	8.71 ± 0.03	−0.299 ± 0.064	−0.0163 ± 0.0035	0.025	7.97 ± 0.03	−0.594 ± 0.055	−0.0324 ± 0.0030	0.022
NGC 4559	8.53 ± 0.03	−0.573 ± 0.056	−0.0467 ± 0.0046	0.047	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 4625	8.58 ± 0.01	−0.130 ± 0.044	−0.0443 ± 0.0151	0.042	7.84 ± 0.05	−0.680 ± 0.143	−0.2318 ± 0.0488	0.127
NGC 4631	8.39 ± 0.06	−0.333 ± 0.125	−0.0194 ± 0.0073	0.077	7.18 ± 0.09	−0.480 ± 0.169	−0.0280 ± 0.0098	0.093
NGC 4651	8.68 ± 0.04	−0.469 ± 0.096	−0.0343 ± 0.0070	0.049	8.02 ± 0.10	−1.133 ± 0.233	−0.0829 ± 0.0170	0.119
NGC 4654	8.66 ± 0.04	−0.364 ± 0.083	−0.0318 ± 0.0073	0.044	7.90 ± 0.11	−0.797 ± 0.229	−0.0695 ± 0.0200	0.120
NGC 4656	8.06 ± 0.16	−0.551 ± 0.280	−0.0305 ± 0.0155	0.084	6.62 ± 0.24	−0.634 ± 0.409	−0.0351 ± 0.0227	0.123
NGC 4713	8.53 ± 0.09	−0.771 ± 0.250	−0.1325 ± 0.0428	0.095	7.60 ± 0.14	−1.548 ± 0.397	−0.2662 ± 0.0679	0.150
NGC 4725	8.83 ± 0.37	−0.472 ± 0.853	−0.0223 ± 0.0404	0.070	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 4736	8.57 ± 0.03	−0.066 ± 0.170	−0.0086 ± 0.0224	0.065	0.00 ± 0.00	0.000 ± 0.000	0.0000 ± 0.0000	0.000
NGC 4861	8.00 ± 0.06	−0.013 ± 0.092	−0.0030 ± 0.0210	0.090	6.54 ± 0.13	−0.058 ± 0.200	−0.0132 ± 0.0454	0.193
NGC 5033	8.64 ± 0.06	−0.278 ± 0.094	−0.0095 ± 0.0032	0.090	8.00 ± 0.10	−0.599 ± 0.130	−0.0206 ± 0.0045	0.072
NGC 5055	8.87 ± 0.02	−0.490 ± 0.031	−0.0337 ± 0.0021	0.007	8.42 ± 0.10	−1.259 ± 0.186	−0.0867 ± 0.0129	0.044
NGC 5068	8.57 ± 0.04	−0.355 ± 0.098	−0.0626 ± 0.0173	0.050	7.53 ± 0.11	−0.451 ± 0.263	−0.0794 ± 0.0463	0.134
NGC 5194	8.88 ± 0.03	−0.324 ± 0.053	−0.0223 ± 0.0037	0.062	8.52 ± 0.05	−0.704 ± 0.102	−0.0484 ± 0.0070	0.117
NGC 5236	8.78 ± 0.01	−0.221 ± 0.028	−0.0256 ± 0.0032	0.050	8.34 ± 0.03	−0.649 ± 0.058	−0.0752 ± 0.0068	0.104
NGC 5248	8.64 ± 0.03	−0.142 ± 0.076	−0.0069 ± 0.0037	0.072	7.74 ± 0.05	−0.132 ± 0.155	−0.0064 ± 0.0075	0.146
NGC 5457	8.71 ± 0.01	−0.840 ± 0.029	−0.0293 ± 0.0010	0.076	8.01 ± 0.03	−1.760 ± 0.056	−0.0613 ± 0.0020	0.142
NGC 5474	8.19 ± 0.02	−0.008 ± 0.086	−0.0019 ± 0.0182	0.100	6.82 ± 0.03	−0.261 ± 0.144	−0.0553 ± 0.0303	0.156
NGC 5668	8.52 ± 0.03	−0.471 ± 0.061	−0.0364 ± 0.0047	0.075	7.48 ± 0.05	−1.032 ± 0.127	−0.0796 ± 0.0098	0.147
NGC 6384	8.87 ± 0.04	−0.531 ± 0.068	−0.0228 ± 0.0029	0.029	8.41 ± 0.12	−1.172 ± 0.197	−0.0504 ± 0.0084	0.083
NGC 6691	8.66 ± 0.02	−0.152 ± 0.052	−0.0077 ± 0.0026	0.050	7.98 ± 0.06	−0.414 ± 0.157	−0.0211 ± 0.0080	0.142
NGC 6744	8.88 ± 0.03	−0.726 ± 0.060	−0.0266 ± 0.0022	0.034	8.66 ± 0.07	−1.996 ± 0.157	−0.0731 ± 0.0057	0.087
NGC 6946	8.72 ± 0.06	−0.337 ± 0.099	−0.0337 ± 0.0099	0.061	8.05 ± 0.15	−0.751 ± 0.237	−0.0751 ± 0.0237	0.148
NGC 7331	8.67 ± 0.08	−0.222 ± 0.193	−0.0097 ± 0.0084	0.044	7.75 ± 0.09	−0.056 ± 0.213	−0.0024 ± 0.0092	0.048
NGC 7495	8.61 ± 0.01	−0.260 ± 0.027	−0.0140 ± 0.0015	0.028	7.83 ± 0.02	−0.616 ± 0.062	−0.0333 ± 0.0033	0.064
NGC 7518	8.74 ± 0.02	−0.127 ± 0.045	−0.0122 ± 0.0043	0.034	8.30 ± 0.07	−0.514 ± 0.136	−0.0496 ± 0.0131	0.104
NGC 7529	8.64 ± 0.04	−0.232 ± 0.067	−0.0303 ± 0.0087	0.040	7.97 ± 0.13	−0.752 ± 0.223	−0.0979 ± 0.0290	0.133
NGC 7570	8.58 ± 0.03	−0.086 ± 0.045	−0.0217 ± 0.0114	0.029	7.78 ± 0.06	−0.219 ± 0.106	−0.0549 ± 0.0265	0.068
NGC 7591	8.71 ± 0.01	−0.146 ± 0.029	−0.0085 ± 0.0017	0.024	8.13 ± 0.06	−0.400 ± 0.114	−0.0234 ± 0.0066	0.093
NGC 7678	8.55 ± 0.04	−0.079 ± 0.078	−0.0048 ± 0.0047	0.030	7.64 ± 0.14	−0.081 ± 0.248	−0.0049 ± 0.0149	0.094
NGC 7793	8.50 ± 0.02	−0.305 ± 0.048	−0.0662 ± 0.0104	0.071	7.52 ± 0.05	−0.752 ± 0.095	−0.1631 ± 0.0207	0.135
IC 0010	8.14 ± 0.08	0.252 ± 0.185	0.4294 ± 0.3025	0.094	6.90 ± 0.15	0.355 ± 0.338	0.6035 ± 0.5534	0.165
IC 0193	8.59 ± 0.02	−0.139 ± 0.064	−0.0083 ± 0.0039	0.039	7.80 ± 0.05	−0.306 ± 0.151	−0.0183 ± 0.0090	0.092
IC 0208	8.58 ± 0.01	−0.168 ± 0.035	−0.0165 ± 0.0034	0.036	7.78 ± 0.03	−0.422 ± 0.083	−0.0416 ± 0.0082	0.086
IC 0342	8.83 ± 0.04	−0.513 ± 0.087	−0.0500 ± 0.0085	0.045	8.46 ± 0.07	−1.465 ± 0.131	−0.1426 ± 0.0128	0.068
IC 1132	8.65 ± 0.03	−0.390 ± 0.059	−0.0323 ± 0.0049	0.077	7.69 ± 0.06	−0.598 ± 0.106	−0.0495 ± 0.0087	0.128
IC 2204	8.62 ± 0.01	−0.160 ± 0.033	−0.0159 ± 0.0033	0.024	7.89 ± 0.03	−0.387 ± 0.086	−0.0384 ± 0.0086	0.064
IC 5201	8.35 ± 0.07	−0.501 ± 0.147	−0.0316 ± 0.0093	0.074	7.19 ± 0.12	−0.908 ± 0.242	−0.0573 ± 0.0153	0.122
IC 5309	8.70 ± 0.03	−0.058 ± 0.066	−0.0059 ± 0.0067	0.059	8.08 ± 0.09	−0.211 ± 0.206	−0.0215 ± 0.0210	0.170
UGC 00223	8.53 ± 0.02	−0.099 ± 0.053	−0.0105 ± 0.0057	0.029	7.72 ± 0.06	−0.295 ± 0.152	−0.0313 ± 0.0162	0.084
UGC 01087	8.60 ± 0.01	−0.246 ± 0.023	−0.0192 ± 0.0018	0.026	7.83 ± 0.03	−0.609 ± 0.058	−0.0475 ± 0.0045	0.066
UGC 01862	8.43 ± 0.02	−0.094 ± 0.039	−0.0194 ± 0.0081	0.040	7.42 ± 0.04	−0.316 ± 0.090	−0.0651 ± 0.0186	0.092
UGC 02023	8.13 ± 0.19	−0.088 ± 0.280	−0.0358 ± 0.1128	0.075	6.84 ± 0.29	−0.197 ± 0.434	−0.0800 ± 0.1752	0.115
UGC 02216	8.08 ± 0.02	−0.068 ± 0.052	−0.0099 ± 0.0077	0.024	6.72 ± 0.04	−0.124 ± 0.094	−0.0181 ± 0.0137	0.041
UGC 02345	8.04 ± 0.03	−0.893 ± 0.085	−0.0885 ± 0.0084	0.041	6.63 ± 0.11	−0.846 ± 0.255	−0.0838 ± 0.0253	0.131
UGC 03701	8.41 ± 0.01	−0.087 ± 0.046	−0.0076 ± 0.0040	0.042	7.41 ± 0.03	−0.329 ± 0.102	−0.0288 ± 0.0089	0.092
UGC 04107	8.56 ± 0.01	−0.231 ± 0.041	−0.0219 ± 0.0039	0.045	7.72 ± 0.03	−0.498 ± 0.087	−0.0470 ± 0.0083	0.096
UGC 04305	7.89 ± 0.07	−0.225 ± 0.137	−0.0573 ± 0.0352	0.075	6.41 ± 0.13	−0.307 ± 0.246	−0.0784 ± 0.0629	0.139
UGC 05100	8.58 ± 0.05	−0.078 ± 0.075	−0.0056 ± 0.0053	0.047	7.90 ± 0.13	−0.220 ± 0.204	−0.0156 ± 0.0145	0.119
UGC 06410	8.58 ± 0.02	−0.148 ± 0.040	−0.0113 ± 0.0031	0.057	7.78 ± 0.05	−0.454 ± 0.093	−0.0345 ± 0.0071	0.123
UGC 08091	7.58 ± 0.04	0.177 ± 0.067	0.5487 ± 0.2003	0.055	6.03 ± 0.08	0.255 ± 0.129	0.7891 ± 0.3905	0.108
UGC 09562	8.22 ± 0.05	−0.193 ± 0.143	−0.0585 ± 0.0433	0.080	6.80 ± 0.06	−0.206 ± 0.144	−0.0626 ± 0.0437	0.068
UGC 09837	8.45 ± 0.02	−0.412 ± 0.076	−0.0370 ± 0.0068	0.078	7.32 ± 0.04	−0.820 ± 0.140	−0.0738 ± 0.0126	0.141
UGC 10445	8.27 ± 0.04	−0.082 ± 0.077	−0.0103 ± 0.0096	0.050	7.03 ± 0.08	−0.197 ± 0.141	−0.0246 ± 0.0176	0.091
UGC 12709	8.47 ± 0.08	−0.706 ± 0.218	−0.0434 ± 0.0134	0.082	7.30 ± 0.15	−1.089 ± 0.418	−0.0669 ± 0.0257	0.157
PGC 29167	8.34 ± 0.10	0.152 ± 0.169	0.0507 ± 0.0565	0.040	7.25 ± 0.01	0.005 ± 0.020	0.0017 ± 0.0065	0.005
PGC 41318	8.45 ± 0.04	−0.168 ± 0.143	−0.0281 ± 0.0239	0.033	7.28 ± 0.17	−0.563 ± 0.568	−0.0941 ± 0.0945	0.130
PGC 44772	8.59 ± 0.03	−0.268 ± 0.045	−0.0216 ± 0.0036	0.043	7.79 ± 0.07	−0.705 ± 0.115	−0.0569 ± 0.0093	0.109
PGC 45195	8.14 ± 0.04	−0.152 ± 0.059	−0.0122 ± 0.0047	0.035	6.81 ± 0.08	−0.277 ± 0.104	−0.0223 ± 0.0084	0.061
PGC 46182	8.59 ± 0.03	−0.126 ± 0.078	−0.0089 ± 0.0055	0.035	7.92 ± 0.11	−0.462 ± 0.313	−0.0327 ± 0.0221	0.141

Download table as:  ASCIITypeset images: 1 2

As in the case of the oxygen abundance, the radial nitrogen abundance distribution in every galaxy was fitted by the following equation:

$$\begin{equation} 12+\log ({\rm N/H}) = 12+\log ({\rm N/H})_{R_{0}} + C_{{\rm N/H}} \times (R/R_{25}). \end{equation} \tag{ 6 }$$

Again, if there were data points with large deviations, in excess of 0.3 dex, those points were rejected, and were not used in the determination of the final relation. These different rejection criteria for oxygen abundances, 0.2 dex, and nitrogen abundances, 0.3 dex, are used for the following reason. There is no one-to-one correspondence between nitrogen and oxygen abundances, but instead there is a scatter in N/H at a given O/H. Therefore one can expect that the natural scatter in N/H at a given galactocentric distance can be larger than that in O/H assuming similar uncertainties in both abundance determinations. The derived parameters of the nitrogen abundance distributions are presented in Table 2. The extrapolated central 12+log(N/H)$_{R_{0}}$ nitrogen abundance and the gradient (the coefficient C_N/H in Equation (6)) expressed in terms of dex $R_{\rm 25}^{-1}$ are listed in Columns 6 and 7. The nitrogen abundance gradient expressed in terms of dex kpc⁻¹ is listed in Column 8. The scatter of nitrogen abundances around the general radial nitrogen abundance trend is reported in Column 9.

A list of references for the emission line flux measurements in the H ii regions is given in Table 3.

In the case of (O/H)_P abundances, there is the following problem. The relationship between oxygen abundance and strong oxygen line intensities, the P calibration, is double-valued with two distinct parts usually known as the "lower" and "upper" branches of the R₂₃–O/H relationship. The expression for the oxygen abundance determination in high-metallicity H ii regions, Equation (3), is valid only for H ii regions that belong to the upper branch, with 12+log(O/H) ≳ 8.3. Thus, one has to know a priori on which of the two branches the H ii region lies. We can overcome this problem in the way suggested in Pilyugin et al. (2004). It has been known for a long time (Searle 1971; Smith 1975) that disks of spiral galaxies show radial oxygen abundance gradients, in the sense that the oxygen abundance is higher in the central part of the disk and decreases with galactocentric distance. We thus start from the H ii regions in the central part of disks and move outward until the radius R* where the oxygen abundance decreases to 12+log(O/H) ∼ 8.3. It should be noted that it is difficult to establish the exact value of R* due to the scatter in oxygen abundance values at any fixed radius. An unjustified use of Equation (3) in the determination of the oxygen abundance in low-metallicity H ii regions beyond R* would result in overestimated oxygen abundances, and would produce a false turnover in the slope of the abundance gradients (Pilyugin 2003). Therefore, H ii regions with galactocentric distances larger than R*, i.e., those with 12+log(O/H) ≲ 8.3 were rejected.

The derived radial distributions of the oxygen and nitrogen abundances in 130 galaxies are presented in Figure 1. The oxygen abundances for individual H ii regions are indicated by points. All the data points (including points with large deviations, which are rejected and are not used in determination of the final relation) are shown. The linear best fits (derived via the least squares method) to the points are represented by solid lines. The galactocentric distances are normalized to the isophotal radius. The nitrogen abundances for individual H ii regions are shown by the plus signs. The linear best fits to the points are indicated by dashed lines.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The values of the gradients in a number of galaxies (e.g., in NGC 12, the first galaxy in our list) are rather small and are comparable to (or even lower than) the uncertainties of gradients. It can be assumed that there is no abundance gradient in such a galaxy, and its abundance can be specified by the mean of the central abundance and the abundance at the isophotal R₂₅ radius of the galaxy.

Thus, the radial oxygen and nitrogen abundance distributions across the optical disk in every galaxy are individually fitted by a single relation. It looks like a rather good approximation for the majority of galaxies. However, a small change in slope in the abundance distribution cannot be excluded in the disks of several galaxies (e.g., NGC 925, NGC 3184, NGC 5457). It is interesting to note the following. Pohlen & Trujillo (2006) studied the surface brightness profiles of a sample of late-type (Sb to Sdm) spiral galaxies using imaging data from the SDSS survey. They found that the surface brightness profiles can be divided into three classes. A small fraction of galaxies (around 10%) belong to type I, which comprises those galaxies that have a normal (standard) exponential disk down to the noise limit. The surface brightness distribution of the rest of the galaxies is better described as a broken exponential. About 60% of the galaxies belong to type II, which means that they show a down-bending profile with steeper outer part. About 30% of the galaxies belong to type III, implying that they show an up-bending profile with shallower outer part. Thus, the abundance distribution profile does not seem to follow strictly the slope of the surface brightness profile. It should be noted, however, that the number of measured H ii regions in many galaxies is too small to allow one to detect a small bend in the radial abundance distribution.

NGC 1512. The five H ii regions that are located in the "bridge" between the interacting galaxies NGC 1512 and NGC 1510 (Bresolin et al. 2012) were excluded from our analysis. Dicaire et al. (2008) found that there is a significant discrepancy between the kinematically derived inclination of NGC 1512 and its photometric value (i_phot = 65° and i_kin = 35°). They noted that the explanation is quite clear: the photometric parameter is mainly representative of the bar which contributes a large part of the light resulting in more edge-on values. However, the outer isophotes are much more face-on. Therefore, the kinematic value of the inclination is adopted here.

NGC 2442. The peculiar spiral galaxy NGC 2442 is tidally distorted (Pancoast et al. 2010, and references therein).

NGC 3227. NGC 3227 is a nearby Seyfert galaxy that is interacting with its gas-poor dwarf elliptical companion NGC 3226. Mundell et al. (1995) discovered a cloud of H i close to, but physically and kinematically distinct from, the galactic disk of NGC 3227. They suggested that this cloud (J1023+1952) might be a dwarf galaxy that is either preexisting and being accreted by NGC 3227, or a newly created tidal dwarf galaxy (Mundell et al. 1995, 2004). However, the oxygen abundance of the cloud (H ii regions 24 and 25 from Werk et al. 2011 and of the H ii region measured by Lisenfeld et al. 2008) follows the general trend of the radial abundance distribution in the disk of NGC 3227. The oxygen abundances of the H ii region in the cloud estimated in different ways using the emission line measurements by Lisenfeld et al. (2008) are in satisfactory agreement to each other: 12+log(O/H)$_{C_{{\rm ON}}}$ = 8.40, 12+log(O/H)$_{C_{{\rm NS}}}$ = 8.43 and 12+log(O/H)_P = 8.34. H ii regions 10 and 12 from Werk et al. (2011) follow also the general trend of the radial abundance distribution in the disk of NGC 3227 although those H ii regions may be associated with NGC 3226.

NGC 3239. The irregular galaxy NGC 3239 is a candidate for a merging system (Krienke & Hodge 1990). The oxygen abundances determined in four H ii regions from SDSS spectra as well as the global abundance determined using the integrated spectra from Moustakas & Kennicutt (2006a) are 12+log(O/H) ∼ 8.0 ÷ 8.1, i.e., those abundances correspond to the transition zone between the upper and lower branches of the O/H–R₂₃ diagram. Therefore, the (O/H)_P abundances we derived from measurements by Werk et al. (2011) are not reliable and were not used.

NGC 3310. NGC 3310 is believed to be an advanced merger and its unusual outer structure is probably the result of a recent merger with a smaller galaxy (Conselice et al. 2000; Knapen & James 2009). Spectroscopic observations of six H ii regions in NGC 3310 were presented by Pastoriza et al. (1993). The auroral line [O iii] λ4363 is detected in four H ii regions. The T_e-based abundances in those H ii regions are in the range from 12+log(O/H) = 8.13–8.25. However, the coordinates of the H ii regions are not published, which prevents us from using those H ii regions in radial gradient determinations.

NGC 3359. The NGC 3359 is a giant, very strongly barred spiral galaxy. Martin & Roy (1995) estimated the radial gradient using the [O iii]/Hβ and [N ii]/[O iii] indicators (one-dimensional calibrations). They found the radial abundance gradient break near the corotation radius. Zahid & Bresolin (2011) carried out new measurements and determined abundances through the N2 and O3N2 calibrations following Pettini & Pagel (2004). They concluded that, with a high degree of confidence, a model with a break fits the data significantly better than one without a break. The C-based abundances show no gradient across the entire galaxy. The T_e-based abundances in three H ii regions in NGC 3359 are in agreement with no gradient.

NGC 3718. This galaxy is so peculiar that it is difficult to categorize it morphologically (Tully et al. 1996).

NGC 4559. The H ii regions in NGC 4559 with galactocentric radii larger than 0.4R₂₅ measured by Zaritsky et al. (1994) were excluded since they are in the transition zone.

NGC 4625. This is a Magellanic spiral with R₂₅ ∼ 3 kpc and an extended faint disk reaching four times the optical R₂₅ radius of the galaxy in the ultraviolet.

NGC 5668. Marino et al. (2012) have found that within ∼36'' of the nucleus the oxygen abundance O/H follows an exponential profile while the outer abundance trend flattens out to an approximately constant value and could even reverse.

NGC 7518. This is a hydrogen-deficient galaxy in the Pegasus cluster. Following Robertson et al. (2012), oxygen lines from "blue" spectra (360 nm–560 nm) have been used in the abundance determinations.

NGC 7529. This galaxy of the Pegasus cluster has a normal H i content. There is no determination of the position angle, but the inclination is small. Again following Robertson et al. (2012), oxygen lines from "blue" spectra have been used in the abundance determinations.

NGC 7591. This galaxy of the Pegasus cluster has a normal H i content. As before, following Robertson et al. (2012), oxygen lines from "blue" spectra have been used in the abundance determinations.

IC 10. Due to uncertainties in the position angle and in the inclination of IC 10, the galactocentric distances of H ii regions have been computed without any de-projection (following Magrini & Gonçalves 2009).

IC 5309. A hydrogen-deficient galaxy of the Pegasus cluster. Following Robertson et al. (2012), oxygen lines from "blue" spectra have been used in the abundance determinations.

UGC 9562 = II Zw 71. The blue compact dwarf galaxy UGC 9562 is a probable polar-ring galaxy, and there are a several luminous H ii regions along its major axis. The oxygen abundances in four H ii regions are around 12+log(O/H) ∼ 8.2, i.e., those abundances correspond to the transition zone between the upper and lower branches of the O/H–R₂₃ diagram. Therefore, the (O/H)_P abundances obtained with measurements from Werk et al. (2011) were not used. The global oxygen abundances determined from the integrated spectra from Kong et al. (2002) are 12+log(O/H) ∼ 8.3.

PGC 029167. The dwarf galaxy "Garland" or PGC 029167 (a tidal dwarf candidate; Makarova et al. 2002) lies within the tidal bridges of neutral hydrogen connecting M 81, M 82, and NGC 3077. This alignment has led to the suggestion that this dwarf formed recently, as a result of tidal interactions within the group. H ii regions in and near Garland exhibit enhanced metallicities compared to other galaxies at similar luminosities. Notably, the oxygen abundances are similar to abundances measured in M 81 and NGC 3077 (Croxall et al. 2009, and references therein).

4.3.1. N/H versus O/H

Figure 2 shows the nitrogen abundance as a function of oxygen abundance in our sample of galaxies. The plus signs are the nitrogen (N/H)$_{R_{0}}$ and oxygen abundances (O/H)$_{R_{0}}$ at R₀ = 0, i.e., the central nitrogen and oxygen abundances. The points show the nitrogen (N/H)$_{R_{25}}$ and oxygen (O/H)$_{R_{25}}$ abundances at the R₂₅ radius of the galaxies. Our data show a well-defined sequence in the O/H–N/H plot. This sequence exhibits the well-known turnover in the sense that the slope at low metallicity is shallower than that at high metallicity. The commonly accepted explanation of this change of slope is that nitrogen can be interpreted as having both primary and secondary components. Nitrogen production is primary at low metallicity, but for 12 +log(O/H) ≳ 8.3, secondary nitrogen becomes prominent, and nitrogen increases at a faster rate than oxygen (Henry et al. 2000).

Zoom In Zoom Out Reset image size

Figure 2 shows that at high metallicity, 12 +log(O/H) ≳ 8.2, the relation between the logarithms of nitrogen and oxygen abundances can be approximated by a linear expression. For (N/H)$_{R_{0}}$ and (O/H)$_{R_{0}}$ abundances, we found the following relation through the least squares method:

$$\begin{eqnarray} 12+\log ({\rm N/H})_{R_{0}} & = & 2.47\,(\pm 0.06)\,\times (12+\log ({\rm O/H})_{R_{0}}) \nonumber\\ && - 13.43\,(\pm 0.50). \end{eqnarray} \tag{ 7 }$$

A similar relation

$$\begin{eqnarray} 12+\log ({\rm N/H})_{R_{25}} & = & 2.50\,(\pm 0.10)\,\times (12+\log ({\rm O/H})_{R_{25}}) \nonumber\\ && - 13.66\,(\pm 0.86) \end{eqnarray} \tag{ 8 }$$

was obtained for the (N/H)$_{R_{25}}$ and (O/H)$_{R_{25}}$ abundances. A comparison between Equations (7) and (8) shows that the relations N/H = f (O/H) for the (N/H)$_{R_{0}}$ and (O/H)$_{R_{0}}$ abundances and for the (N/H)$_{R_{25}}$ and (O/H)$_{R_{25}}$ abundances agree with each other within the errors. The N/H = f (O/H) relation given by Equation (7) is presented in Figure 2 by the solid line. The dashed lines indicate shifts along the Y-axis by ± 0.15 dex. At low metallicity, 12 +log(O/H) ≲ 8.0, the single relation

$$\begin{eqnarray} 12+\log ({\rm N/H}) &=& 0.96\,(\pm 0.08)\,\times (12+\log ({\rm O/H}))\nonumber\\ && - 1.20\,(\pm 0.65) \end{eqnarray} \tag{ 9 }$$

was found using both central abundances and abundances at the R₂₅ radii of the galaxies. This relation is presented in Figure 2 by the dotted line.

Figure 2 shows that there is an appreciable spread in N/H at a given O/H. The variation in N/H is around a factor of two for a given O/H. The scatter can be partially attributed to the errors in the abundance determinations but part of it seems to be true abundance scatter. Two major mechanisms have been proposed for generating a spread. One mechanism invokes a time delay between ejections of the freshly manufactured oxygen and nitrogen into the interstellar medium by a given stellar generation. The N/O ratios may be an indicator of the age of a galactic system, indicating the time that has passed since the bulk of star formation activity (Edmunds & Pagel 1978). Thus, the N/O ratio in a galaxy would then depend on its star formation history. The second mechanism for causing N/H variations at a given O/H is a variation in the efficiency of enriched galactic winds (Pilyugin 1993) or/and in the inflow of gas into the galaxy (Henry et al. 2000). It is believed that galactic winds do not play a significant role in the chemical evolution of large spiral galaxies (Tremonti et al. 2004; Dalcanton 2007). The enhanced N/O ratio in individual H ii regions can be caused by the local pollution in nitrogen by Wolf–Rayet stars (López-Sánchez & Esteban 2010, and references therein). Since we consider "average" N and O abundances based on the abundances of different H ii regions this origin of the scatter in our diagrams seems to be unlikely.

It has been known for a long time that galaxies of different morphological types have different star formation histories, i.e., spiral galaxies with early morphological types have a larger fraction of old stars (Sandage 1986). One can then expect that N/H at a given O/H may depend on the morphological type of galaxy expressed in terms of T type (Pilyugin et al. 2003). On the other hand, the star formation history of a galaxy also strongly depends on its mass (or luminosity) as is epitomized in the galaxy downsizing effect, where the star formation activity shifts from high-mass galaxies at early cosmic times to lower-mass galaxies at later epochs (Cowie et al. 1996). Again, one can expect then that the N/H ratio at a given O/H will depend on the galaxy mass (Pilyugin & Thuan 2011).

There is a relatively tight linear correlation between the absolute magnitudes and the logarithms of the linear diameters of nearby galaxies (van den Bergh 2008). Therefore, the linear diameter of a galaxy can be used as an indicator of its luminosity (and mass). Thus, if the spread in N/H at a given O/H is caused by the time delay between nitrogen and oxygen enrichment and the different star formation histories in different galaxies then the N/H at a given O/H should correlate with morphological T type or/and with linear radius of the galaxy.

Figure 3 shows the residual of Equations (7) and (8) as a function of the morphological T type and linear radius of the galaxy R₂₅. Figure 3 indicates that there may exist some correlation between the N/H at a given O/H and morphological type or/and linear radius of the galaxy. To verify this, we found a relation between N/H and O/H for spiral galaxies (with T < 7.5) where T and logR₂₅ are "secondary parameters," N/H = f (O/H, T, logR₂₅). It should be noted that T and logR₂₅ are not perfectly independent parameters since there is some correlation between them. The relation obtained for central abundances is

$$\begin{eqnarray} 12+\log ({\rm N/H})_{R_{0}} & = & {-}12.67 + 2.39\,(12+\log ({\rm O/H})_{R_{0}}) \nonumber\\ & & -0.022\,T + 0.023\,\log (R_{25}). \end{eqnarray} \tag{ 10 }$$

The values of the coefficients in the terms containing T and logR₂₅ are similar. However, the T value is a more important second parameter than the logR₂₅ value since the variation in T values (from 1 to 7.5) is much larger than the variation of logR₂₅ values (from ∼0.6 to ∼1.5). Therefore, the variation in N/H due to variation in the morphological type is larger than that due to variation in the linear radii of galaxies. For example, the Sab galaxies (T = 2) have nitrogen abundances larger than on average 0.1 dex (i.e., by around 30%) than Sd (T = 7) galaxies with the same oxygen abundances. The relation N/H = f (O/H, T, logR₂₅) obtained for abundances at the optical R₂₅ radii of the galaxies is

$$\begin{eqnarray} 12+\log ({\rm N/H})_{R_{25}} & = & {-}14.67 + 2.59\,(12+\log ({\rm O/H})_{R_{25}}) \nonumber\\ & & + 0.013\,T + 0.178\,\log (R_{25}). \end{eqnarray} \tag{ 11 }$$

In this case the logR₂₅ value is the more important second parameter than the T value. Although the morphological type is the more important second parameter in the relation for central abundances while logR₂₅ is the more important second parameter in the relation for abundances at the optical R₂₅ radii of the galaxies, it is difficult to draw a solid conclusion whether this is physically meaningful since there is a correlation between T and logR₂₅. That the relation between N/H and O/H depends on the additional parameter(s) T and/or logR₂₅ suggests that the scatter in N/H at a given O/H can be caused, at least partly, by the time delay between nitrogen and oxygen enrichment and the different star formation histories in different galaxies.

Zoom In Zoom Out Reset image size

4.3.2. Abundances and Gradients as a Function of Morphological Type and Galaxy Radius

The upper left panel of Figure 4 shows the central oxygen (O/H)$_{R_{0}}$ abundance in a galaxy as a function of its morphological T type (data from Tables 1 and 2). The (O/H)$_{R_{0}}$–T diagram shows that there is a trend in central oxygen abundance with morphological type for galaxies later than Sc (T ≳ 5) such that the central oxygen abundances are lower in galaxies of later types. This trend disappears for early-type spiral galaxies (T ≲ 5).

Zoom In Zoom Out Reset image size

The upper right panel of Figure 4 shows the central oxygen abundance in a galaxy as a function of its isophotal radius R₂₅. Since there is a relatively tight linear correlation between the absolute magnitudes and the logarithms of the linear diameters of nearby galaxies (van den Bergh 2008), this diagram can be considered as some kind of analog of the standard "luminosity–metallicity" diagram. The (O/H)$_{R_{0}}$–R₂₅ diagram shows that there is a weak correlation between the central oxygen abundance and optical radius for small galaxies (R₂₅ ≲ 10 kpc) in the sense that the smaller galaxies have on average lower oxygen abundances. This correlation disappears for large galaxies (R₂₅ ≳ 10 kpc), i.e., the most oxygen-rich galaxies of different radii have similar central oxygen abundances.

Different versions of the luminosity–metallicity diagram have been constructed in earlier studies. B-band luminosity–characteristic oxygen abundance diagrams were considered in Zaritsky et al. (1994), Pilyugin et al. (2004), and Moustakas et al. (2010), where the characteristic oxygen abundance is defined as the abundance at R = 0.4R₂₅. Tremonti et al. (2004) have used the global abundances (in the sense that their abundances do not correspond to the abundances at a fixed galactocentric distance, but instead are some kind of mean abundance for a fraction of a galaxy within the fiber aperture) estimated from SDSS spectra in constructing the luminosity–metallicity diagram. The luminosity–central metallicity diagram was examined in Pilyugin et al. (2007). The flattening of the luminosity–metallicity relation at high luminosities (essentially a plateau) can be seen in all versions of the diagram. Thus the plateau in our (O/H)$_{R_{0}}$–R₂₅ diagram at large radii is consistent with previous results.

It has been advocated that the constant maximum value of the observed central oxygen abundance in the most oxygen-rich galaxies suggests that the observed oxygen abundance in the centers of those galaxies represents the maximum attainable value of the gas-phase oxygen abundance (Pilyugin et al. 2007). The upper-row panels of Figure 4 show that the observed central oxygen abundance in the most oxygen-rich galaxies in our sample is 12+log(O/H)$_{R_{0}}$ ∼ 8.85. The dashed lines in the upper row of panels of Figure 4 show this value of oxygen abundance, which seems to correspond to the maximum attainable value of the gas-phase oxygen abundance in galaxies. The observed central oxygen abundance in the most oxygen-rich galaxies from our sample is a factor of ∼2 higher than the gas-phase oxygen abundance in the solar neighborhood, 12+log(O/H) ∼ 8.5 (e.g., Pilyugin et al. 2006). The maximum attainable value of the oxygen abundance in galaxies obtained here is in agreement with the value from Pilyugin et al. (2007). Because some fraction of the oxygen (about 0.1 dex) is expected to be locked in dust grains (e.g., Esteban et al. 1998), the maximum value of the true (gas + dust) oxygen abundances in H ii regions of spiral galaxies is 12+log(O/H)$_{R_{0}}$ ∼ 8.95.

The lower left panel of Figure 4 shows the central nitrogen (N/H)$_{R_{0}}$ abundance in a galaxy disk as a function of its morphological T type. A comparison of the panels in the left column of Figure 4 shows that the changes in the central oxygen and nitrogen abundances with morphological type of a galaxy are rather similar. The lower right panel of Figure 4 shows the central nitrogen abundance in a galaxy disk as a function of its optical isophotal radius R₂₅. Again, a comparison between the right column panels of Figure 4 indicates that the general behavior of the central nitrogen abundances as a function of optical radius is similar to that for oxygen.

According to the relation between nitrogen and oxygen abundances given in Equation (7), the value of 12 + log(N/H)$_{R_{0}}$∼ 8.42 corresponds to the maximum attainable value of the gas-phase oxygen abundance in galaxies with 12+log(O/H)$_{R_{0}}$∼ 8.85. This value of nitrogen abundance is indicated in the lower row of panels of Figure 4 by the dashed lines. One can see that the value of 12+log(N/H)$_{R_{0}}$ ∼ 8.42 can be adopted as the maximum value of the observed central nitrogen abundance in the most nitrogen-rich galaxies at the present-day epoch. However, in contrast to the case of oxygen, it is not necessary that this value corresponds to the maximum attainable value of the gas-phase oxygen abundance in galaxies because of the time-delay between nitrogen and oxygen enrichment of the interstellar medium.

The panels in the left column of Figure 5 show the oxygen (O/H)$_{R_{25}}$ (upper panel) and nitrogen (N/H)$_{R_{25}}$ (lower panel) abundances measured at the isophotal radius, R₂₅, as a function of morphological T type. A comparison between the panels in the left column of Figure 4 and Figure 5 shows that the changes in the central oxygen (nitrogen) abundances and in the abundances at the optical edge of the disk along the Hubble sequence (or morphological type) are more or less similar (at least qualitatively). However, the oxygen and nitrogen abundances at the R₂₅ radius in four of the late-type galaxies (NGC 4625, IC 10, UGC 223, and PGC 29167) are high and those galaxies show a large deviation from the general trends. NGC 4625 is a Magellanic spiral with R₂₅ ∼ 3 kpc and a very extended faint disk. The dwarf irregular galaxy IC 10 has a very small optical radius, R₂₅ = 0.6 kpc, and a positive radial abundance gradient. One may suggest though that any radial gradient in such small galaxies is hard to define based on H ii regions. It remains a mystery why the oxygen abundance changes by a factor of about two on the scale of 0.6 kpc in this galaxy. IC 10 is one of several dwarf galaxies in which evidence for localized, inhomogeneous enrichment has been found (Kniazev et al. 2005; Koch et al. 2008a, 2008b; Magrini & Gonçalves 2009; López-Sánchez et al. 2011)—perhaps this is a common mode of enrichment in these small objects. In contrast, the irregular galaxy UGC 223 is rather large with R₂₅ = 9.3 kpc. The peculiar dwarf galaxy PGC 29167 (or "Garland") is commonly considered a tidal dwarf galaxy (see comment on this galaxy in Section 4.2).

Zoom In Zoom Out Reset image size

The panels in the right column of Figure 5 show the oxygen (O/H)$_{R_{25}}$ (upper panel) and nitrogen (N/H)$_{R_{25}}$ (lower panel) abundances at the optical edge of the disk as defined by the R₂₅ radius as a function of isophotal radius R₂₅. The positions of the galaxies in the (O/H)$_{R_{25}}$–R₂₅ and (N/H)$_{R_{25}}$–R₂₅ diagrams do not show any obvious trends. In particular, the (O/H)$_{R_{25}}$ and (N/H)$_{R_{25}}$ abundances do not show any appreciable correlation with isophotal radius R₂₅. However, one feature in these diagrams should be noted: the (O/H)$_{R_{25}}$ and (N/H)$_{R_{25}}$ abundances have a maximum value in galaxies with an isophotal radius R₂₅ ∼ 10 kpc and decrease when moving from this value both toward smaller or larger optical radii.

Figure 6 shows the radial oxygen and nitrogen abundance gradients in units of dex kpc⁻¹ as a function of morphological T type (left panels) and of isophotal radius R₂₅ of a galaxy (right panels). Inspection of the left column panels of Figure 5 shows that the values of the abundance gradients in units of dex kpc⁻¹ do not correlate with the morphological type of a galaxy. According to Zaritsky et al. (1994), the lack of a correlation between gradients in units of dex kpc⁻¹ and the macroscopic properties of late-type galaxies may suggest that the relationship between these parameters is more complex than a simple correlation. Indeed, Vila-Costas & Edmunds (1992) have concluded that a correlation for non-barred galaxies is seen. The panels in the right column of Figure 5 show that shallow gradients can be found both in small and large galaxies while steep gradients are seen only in small galaxies in the sense that the smaller a galaxy the steeper its gradient.

Zoom In Zoom Out Reset image size