Metal-THINGS Survey Overview

• Metal-THINGS is an IFU survey that maps gas-phase metallicity, ionization, and diffuse ionised gas at sub-kiloparsec resolution in nearby galaxies.
• It employs mosaicked integral-field spectroscopy with overlapping pointings and dithers to achieve detailed, consistent measurements across entire galaxy discs.
• The survey extends resolved chemical studies into the low-mass regime, revealing metallicity gradients, stellar feedback effects, and dwarf-galaxy scaling relations.

Metal-THINGS is an integral-field spectroscopy survey of nearby galaxies designed to map gas-phase metallicity, ionization, diffuse ionised gas, and feedback-driven structure at sub-kiloparsec scales. It is presented as a follow-up to the THINGS H I survey, built to exploit the existing multiwavelength context of nearby systems and to extend resolved chemical studies into the low-mass regime, where such measurements have historically been sparse. Published descriptions refer both to an ongoing IFU program targeting 34 nearby THINGS galaxies and to a 25-galaxy analyzed sample observed in 102 pointings, with physical resolutions of about 40–300 pc (Garduño et al., 2023, Valé et al., 11 Jul 2025).

1. Survey architecture

Metal-THINGS uses mosaicked integral-field spectroscopy with the George and Cynthia Mitchell Spectrograph at McDonald Observatory. The survey strategy employs multiple partially overlapping pointings, with dithers to increase areal coverage, so that resolved measurements can be derived across galaxy discs rather than from isolated apertures or global spectra. A central technical characteristic is its physical resolution of about 40–300 pc, which is described as substantially sharper than CALIFA, MaNGA, and SAMI and comparable to the best PHANGS observations (Valé et al., 11 Jul 2025).

Published field studies show the practical observational format of the survey. In NGC 6946, Metal-THINGS assembled 12 pointings into a mosaic with 8856 individual fiber spectra, each fiber subtending 4.2 arcsec, corresponding to about 159 pc at 7.8 Mpc. In that application the spectral coverage was 4400–6800 Å, including H$\beta$, [O III]$\lambda\lambda4959,5007$, [O I]$\lambda6300$, H$\alpha$, [N II]$\lambda6584$, and [S II]$\lambda\lambda6716,6731$, which are the main nebular transitions used throughout the survey’s resolved analyses (Lara-Lopez et al., 2022).

The survey’s scientific scope is broader than radial abundance work alone. Published analyses use the same IFU framework to study gas metallicity gradients, diffuse ionised gas, local star-formation regulation, supernova remnant environments, H I holes, and the nebular environments of ultra-luminous X-ray sources. This breadth is a defining feature of Metal-THINGS: the same data structure supports both galaxy-wide chemical cartography and highly local environmental studies.

2. Spectroscopic workflow and diagnostic framework

Metal-THINGS analyses use a largely consistent reduction and diagnostic pipeline. Stellar continua are fit with STARLIGHT, emission lines are measured after subtraction of the stellar component, and only spectra or fibres with $\mathrm{S/N}>3$ in the relevant lines are retained for quantitative diagnostics. Extinction is corrected from the Balmer decrement under case B recombination. One general survey formulation writes the reddening coefficient as

$$C(\mathrm{H}\beta) = - \frac{1}{f(\lambda)} \log \left( \frac{I(\mathrm{H}\alpha)}{I(\mathrm{H}\beta)} \middle/ \frac{\mathcal{F}(\mathrm{H}\alpha)}{\mathcal{F}(\mathrm{H}\beta)} \right),$$

assuming an intrinsic $\mathrm{H}\alpha/\mathrm{H}\beta = 2.86$ at $T=10^4$ K and $\lambda\lambda4959,5007$0 (Valé et al., 11 Jul 2025).

Diffuse ionised gas is treated explicitly rather than folded into the H II-region population. The survey adopts the Kaplan et al. formalism

$\lambda\lambda4959,5007$1

and uses both a strict DIG threshold, $\lambda\lambda4959,5007$2, and a comparison threshold, $\lambda\lambda4959,5007$3. In the NGC 6946 analysis, the fitted parameters were $\lambda\lambda4959,5007$4 and $\lambda\lambda4959,5007$5, and DIG-dominated fibres under the $\lambda\lambda4959,5007$6 criterion comprised 9.2% of the analyzed sample (Lara-Lopez et al., 2022).

BPT classification is a standard component of the workflow. Metal-THINGS uses the $\lambda\lambda4959,5007$7 versus $\lambda\lambda4959,5007$8 plane, with the Kauffmann et al. line defining star-forming fibres and the Kewley et al. line defining AGN-like excitation; intermediate points are classified as composite. An important interpretive caveat emphasized in the survey is that an “AGN-like” BPT position does not necessarily imply a true AGN, because DIG, shocks, or post-AGB stars can generate similar line ratios (Valé et al., 11 Jul 2025).

Gas metallicities are derived with the Pilyugin & Grebel (2016) strong-line calibration. In the broad survey treatment this is implemented in the PG16 sulphur-based framework, split into upper and lower branches at $\lambda\lambda4959,5007$9, with $\lambda6300$0 as the abundance variable. Survey analyses also derive the ionization parameter from sulphur-line diagnostics, and some targeted studies, such as the ULX work in NGC 925, supplement the strong-line approach with BEAGLE as an independent consistency check.

3. Radial chemical structure across the sample

A major published synthesis from Metal-THINGS concerns gas metallicity gradients in nearby galaxies. For the 25-galaxy sample, gradients are modeled in normalized radius as

$\lambda6300$1

with Bayesian linear models implemented in PyMC and Bambi, using 4 MCMC chains of 3000 iterations each. The dominant empirical result is that metallicity typically decreases with galactic radius, consistent with inside-out galaxy growth (Valé et al., 11 Jul 2025).

The survey reports two characteristic breaks in the scaling of gradient with global galaxy properties. For stellar mass, the break occurs at

$\lambda6300$2

with the fitted relation

$\lambda6300$3

For atomic gas fraction,

$\lambda6300$4

another break appears around $\lambda6300$5, with fitted values $\lambda6300$6 for $\lambda6300$7 and $\lambda6300$8 for $\lambda6300$9 (Valé et al., 11 Jul 2025).

Normalization choice is itself part of the survey’s result set. NUV-band effective radii are reported as preferable for galaxies with higher atomic gas content and lower stellar masses, whereas $\alpha$0-band radii are better suited to systems with lower atomic gas fractions and higher stellar masses. $\alpha$1-band radii behave similarly to $\alpha$2-band where available, but the near-IR sample is too small for strong conclusions, and $\alpha$3 generally does not outperform effective-radius normalizations. This suggests that the survey treats radial scaling not as a purely geometric convention but as a physically informative choice linked to gas content and stellar structure.

The role of DIG in these gradients is judged to be modest in most cases. Excluding DIG fibres usually makes gradients slightly steeper because DIG fibres tend to be more metal-poor, but the median change is small; Holmberg I and NGC 4736 are cited as exceptions. A plausible implication is that the survey’s main mass–gas–gradient trends are not artifacts of unresolved DIG contamination, even though DIG remains important for line-ratio interpretation.

4. Stellar feedback, H I holes, and remnant populations

Metal-THINGS has also been used to study the coupling between optical feedback tracers and neutral-gas structure. In NGC 6946, the survey analysis combines 121 H I holes from Boomsma et al. (2008), 225 supernova remnant candidates from Long et al. (2020), and the Metal-THINGS IFU mosaic. Of the SNR candidates, 147 are optically identified nebulae with elevated $\alpha$4 ratios (Lara-Lopez et al., 2022).

The principal spatial result is that the SNRs are concentrated at the rims of the H I holes rather than uniformly distributed through the interiors. The relevant diagnostic compares, for each SNR, the distance to the nearest hole centre, $\alpha$5, with the radius to the rim along the same direction, $\alpha$6. The distribution of $\alpha$7 peaks near zero and is skewed toward positive values, indicating a strong association with shell boundaries rather than cavity centres (Lara-Lopez et al., 2022).

Resolved optical diagnostics reinforce this geometric association. For the star-formation-rate surface density, the inside-versus-rim comparison yields a positive-skewed distribution in

$\alpha$8

with about 67.7% of holes showing enhanced SFR at the rims; a paired $\alpha$9-test gives $\lambda6584$0 and $\lambda6584$1. Extinction behaves similarly: about 61% of holes show higher $\lambda6584$2 at the rims, with $\lambda6584$3 and $\lambda6584$4. Oxygen abundance and ionization parameter show weaker tendencies—58.6% and about 66% positive rim-minus-hole differences, respectively—but the associated $\lambda6584$5-tests do not reject the null hypothesis ($\lambda6584$6, $\lambda6584$7 for abundance; $\lambda6584$8, $\lambda6584$9 for ionization) (Lara-Lopez et al., 2022).

The interpretation advanced in this work is a feedback loop driven by superbubble expansion. Multiple supernovae in massive stellar clusters dozens of Myr ago are proposed to have inflated the H I holes; the swept-up shells then compressed gas at the cavity boundaries, inducing new star formation at the rims. The present-day SNR population would therefore be expected to accumulate preferentially around those boundaries. The analysis also notes that most fibres around SNRs remain SF-like on the BPT diagram and do not show systematic metallicity offsets, indicating that at Metal-THINGS resolution the integrated line emission of a fibre is often not dominated by SNR shocks alone. This directly addresses a common misconception that the presence of an SNR necessarily controls the fibre-integrated nebular diagnostics.

5. Compact accretors and local nebular environments

A distinct application of Metal-THINGS is the resolved study of ultra-luminous X-ray source environments. In NGC 925, the survey provides 13 George Mitchell Spectrograph pointings and 9594 individual spectra over 4400–6800 Å, with 4.2 arcsec spatial sampling, corresponding to about 184 pc per fibre at 9.2 Mpc. These data are combined with higher-resolution PUMA Fabry–Perot H$\lambda\lambda6716,6731$0 observations to analyze three ULX regions (Lara-López et al., 2020).

The galaxy-wide IFU analysis finds a negative metallicity gradient and a radially increasing ionization parameter, fit as

$\lambda\lambda6716,6731$1

and

$\lambda\lambda6716,6731$2

with $\lambda\lambda6716,6731$3. BPT classification shows that most spaxels in NGC 925 are star-forming, including those associated with ULX-1 and ULX-3 (Lara-López et al., 2020).

ULX-1 is the most chemically distinctive case. Two fibres nearest the source have unusually low metallicity for their galactocentric distance by two independent methods. From the S-calibration, the two fibres give $\lambda\lambda6716,6731$4 and $\lambda\lambda6716,6731$5; from BEAGLE, they give $\lambda\lambda6716,6731$6 and $\lambda\lambda6716,6731$7. In the radial interval $\lambda\lambda6716,6731$8, the median metallicity is 8.26 in the S-calibration and 8.42 in BEAGLE, so ULX-1 lies below the local median in both frameworks. Its ionization parameter is described as typical rather than extreme. The paper argues that this localized low-metallicity environment is more consistent with an HMXB population favoring active Roche-lobe overflow than with accretion of a low-mass satellite galaxy, because the low-metallicity gas is not spatially extended (Lara-López et al., 2020).

ULX-3 presents a contrasting environment. It lies in the star-forming region of the BPT diagrams but has higher metallicity and higher ionization parameter than ULX-1. The S-calibration gives $\lambda\lambda6716,6731$9 and $\mathrm{S/N}>3$0; BEAGLE gives $\mathrm{S/N}>3$1, $\mathrm{S/N}>3$2, and $\mathrm{S/N}>3$3. The favored interpretation is that ULX-3 is consistent with a highly accreting neutron star in an evolved stellar-population region (Lara-López et al., 2020).

ULX-2 illustrates an observational limit rather than an astrophysical one: its relevant spaxel had too low $\mathrm{S/N}>3$4 in [N II] $\mathrm{S/N}>3$5 and H$\mathrm{S/N}>3$6 for reliable BPT placement or chemical diagnostics. More generally, the NGC 925 analysis emphasizes that classical BPT diagrams are not foolproof in compact-source environments, because low-metallicity AGN-like regions can masquerade as star-forming. This makes Metal-THINGS valuable precisely because it embeds compact sources within the full galactic metallicity and ionization context instead of treating them as isolated apertures.

6. Dwarf-galaxy scaling relations and broader significance

Metal-THINGS has particular importance in the low-mass, low-metallicity regime. The panchromatic analysis of the dwarf irregular galaxy NGC 1569 combines Metal-THINGS IFU spectroscopy with THINGS H I data, CARMA CO data, and archival DustPedia imaging to derive resolved maps of stellar mass, SFR, gas metallicity, dust mass, H I, molecular gas, total gas, baryonic mass, gas fraction, effective oxygen yield, star-formation efficiency, and depletion time at about 180 pc resolution (Garduño et al., 2023).

The NGC 1569 study recovers several classical resolved scaling relations but with slopes that differ from previous work. The star-formation main sequence has slope 1.21, the total-gas Kennicutt–Schmidt relation has slope 0.96, and the molecular Kennicutt–Schmidt relation has slope 0.58. The relations including metallicity, stellar mass, and gas fraction are flat, with an average $\mathrm{S/N}>3$7 dex. The baryonic-mass versus effective-oxygen-yield relation and the stellar-, gas-, and baryonic-mass versus SFE relations show higher dispersions and lower correlations (Garduño et al., 2023).

The same analysis derives $\mathrm{S/N}>3$8 and $\mathrm{S/N}>3$9, leading to differences of 0.16 dex and 0.95 dex in total and molecular gas surface density, respectively, relative to previously reported values. Using the self-regulated feedback model

$$C(\mathrm{H}\beta) = - \frac{1}{f(\lambda)} \log \left( \frac{I(\mathrm{H}\alpha)}{I(\mathrm{H}\beta)} \middle/ \frac{\mathcal{F}(\mathrm{H}\alpha)}{\mathcal{F}(\mathrm{H}\beta)} \right),$$0

together with

$$C(\mathrm{H}\beta) = - \frac{1}{f(\lambda)} \log \left( \frac{I(\mathrm{H}\alpha)}{I(\mathrm{H}\beta)} \middle/ \frac{\mathcal{F}(\mathrm{H}\alpha)}{\mathcal{F}(\mathrm{H}\beta)} \right),$$1

the paper concludes that stellar feedback plays an important role in generating outflows in NGC 1569 (Garduño et al., 2023).

These dwarf-galaxy results clarify the survey’s broader significance. Metal-THINGS is repeatedly positioned as filling a gap left by large IFU surveys biased toward intermediate- and high-mass systems. Its high spatial resolution and multiwavelength anchoring allow low-mass galaxies to be studied with the same diagnostic vocabulary used for massive spirals, while also exposing regimes in which standard resolved relations change slope, flatten, or acquire larger scatter. This suggests that the survey’s strongest contribution lies not only in measuring metallicities, but in tying together chemical evolution, gas content, diffuse ionised gas, local feedback, and compact-source environments within a unified nearby-galaxy framework.