Giant Low-Surface-Brightness Disk Galaxies

• gLSBGs are extreme disk galaxies defined by an HSB inner component and an extended LSB disk, with sizes reaching tens to over a hundred kiloparsecs and masses up to 10^12 M⊙.
• Photometric analyses typically require a two-component model where an inner HSB disk coexists with a diffuse outer LSB disk that dominates beyond 2–3 times the inner scale length.
• Observations and simulations indicate that varied formation channels—including high angular momentum accretion, minor mergers, and tidal interactions—drive their evolution while maintaining low star formation efficiency.

Giant low-surface-brightness disk galaxies (gLSBGs, also GLSB galaxies in part of the literature) are extreme disk systems characterized by exceptionally extended, optically faint outer stellar disks, large H I reservoirs, and high total masses, often of order up to $10^{12}\,M_\odot$. Across the literature, they are commonly described as galaxies with central disk surface brightness fainter than the canonical Freeman value and with disk sizes reaching or exceeding tens to more than one hundred kiloparsecs. A recurrent structural motif is a relatively normal high-surface-brightness (HSB) inner galaxy embedded in a much larger low-surface-brightness (LSB) disk, so that gLSBGs occupy the extreme large-size, low-surface-brightness end of the spiral population rather than forming a wholly disconnected class (Das, 2013, Saburova et al., 2022).

1. Definitions and classification

Operational definitions of gLSBGs are not identical across studies. One review defines a gLSBG by three criteria: central disk surface brightness $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$, exponential scale length $h \gtrsim 10\,\mathrm{kpc}$, and disk stellar mass of order $10^{10}$–$10^{11}\,M_\odot$ (Das, 2013). A GMRT H I study uses $\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$ together with optical sizes comparable to or exceeding those of large HSB spirals, with typical half-light radii $R_e \approx 8$–$23\,\mathrm{kpc}$ and outer isophotal diameters $D_{25} \gtrsim 50$–$100\,\mathrm{kpc}$ (Mishra et al., 2016). In a recent HSC-based census, “giant” is defined by either $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$0 or $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$1, and the LSB subset by deprojected $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$2 (Saburova et al., 2022). In an H I-rich SDSS sample selected through two-component fitting, $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$3 is concluded to be the best criterion to distinguish gLSBGs from normal LSBGs with bulge (Zhang et al., 2023).

These differing thresholds reflect the fact that the class is identified through a combination of diffuseness, physical size, and mass. This suggests that “giant” and “low surface brightness” are not reducible to a single photometric cut. The HSC census explicitly argues that giant disky galaxies represent the large-size end of the volume-density distribution of normal-sized spirals, while the H I-rich SDSS analysis shows that purely photometric diffuseness criteria can include lower-mass systems that are not massive giant disks in a dynamical sense (Saburova et al., 2022, Zhang et al., 2023).

2. Photometric structure and the HSB+LSB double disk

The basic photometric law is the standard exponential disk,

$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$4

or, in intensity form,

$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$5

with $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$6 the scale length (Das, 2013, Saburova et al., 2021). For gLSBGs, however, a single exponential is frequently insufficient. Multiple observational studies and the TNG50 simulation study converge on a two-component description: an inner HSB disk or lenticular-like component plus an outer, more diffuse LSB disk (Sarkar et al., 10 Feb 2026, Hagen et al., 2016, Saburova et al., 2019).

In TNG50, the stellar surface-brightness profile of the seven best-fit double-disk systems is modeled as

$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$7

where the inner and outer exponentials correspond to the HSB and LSB disks, respectively. The reported ranges are $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$8–$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$9, $h \gtrsim 10\,\mathrm{kpc}$0–$h \gtrsim 10\,\mathrm{kpc}$1, $h \gtrsim 10\,\mathrm{kpc}$2–$h \gtrsim 10\,\mathrm{kpc}$3, and $h \gtrsim 10\,\mathrm{kpc}$4–$h \gtrsim 10\,\mathrm{kpc}$5; the outer disk dominates beyond $h \gtrsim 10\,\mathrm{kpc}$6–$h \gtrsim 10\,\mathrm{kpc}$7 (Sarkar et al., 10 Feb 2026). A closely related observational synthesis reports typical two-component parameters $h \gtrsim 10\,\mathrm{kpc}$8–$h \gtrsim 10\,\mathrm{kpc}$9, $10^{10}$0–$10^{10}$1, $10^{10}$2–$10^{10}$3, and $10^{10}$4–$10^{10}$5 (Saburova et al., 2020).

Classical archetypes illustrate the range of the class. Malin 1 has an outer disk with $10^{10}$6 and $10^{10}$7, surrounding an inner SB0/a structure (Lelli et al., 2010). Panchromatic imaging of Malin 1 finds that beyond $10^{10}$8 the disk closely follows an exponential law out to $10^{10}$9, with $10^{11}\,M_\odot$0 and $10^{11}\,M_\odot$1 (Boissier et al., 2016). UGC 1378 is explicitly described as a Milky Way-sized HSB disk embedded in a giant LSB envelope, with $10^{11}\,M_\odot$2, $10^{11}\,M_\odot$3, $10^{11}\,M_\odot$4, and $10^{11}\,M_\odot$5 (Saburova et al., 2019). UGC 1382 is fit by two Sérsic components, with an inner lenticular component of $10^{11}\,M_\odot$6 and an outer LSB disk of $10^{11}\,M_\odot$7, while a pure exponential fit to the outer disk yields $10^{11}\,M_\odot$8 and $10^{11}\,M_\odot$9 (Hagen et al., 2016).

The repeated appearance of an HSB interior plus a giant LSB exterior is central to the modern understanding of gLSBGs. This “double structure” is seen not only photometrically but also dynamically, because the inner and outer regions often differ in stellar density, star-formation efficiency, and dynamical support (Lelli et al., 2010, Sarkar et al., 10 Feb 2026).

3. Gas content, kinematics, and dark-matter halos

gLSBGs are H I-rich systems, but the gas is distributed at low surface density over enormous radii. A review gives $\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$0–$\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$1, $\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$2–$\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$3, and molecular masses only $\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$4–$\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$5, with $\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$6 (Das, 2013). The GMRT H I survey of seven systems finds $\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$7–$\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$8, peak H I column densities of several $\mu_B > 23\,\mathrm{mag\,arcsec^{-2}}$9, and H I disks extending to approximately twice the stellar radius (Mishra et al., 2016). A larger GBT-based H I study of 19 HSC-selected gLSBGs reports that 13/19 have $R_e \approx 8$0, five lie in $R_e \approx 8$1–$R_e \approx 8$2, and one is undetected; it emphasizes that very large optical size, rather than exceptionally high $R_e \approx 8$3, is the defining observational feature (Lah et al., 10 Mar 2026).

The rotation curves are characteristically massive and extended. GMRT data give flat rotation velocities between 225 and $R_e \approx 8$4 (Mishra et al., 2016), while a seven-object spectroscopic study reports $R_e \approx 8$5–$R_e \approx 8$6 and flat curves out to tens of kiloparsecs (Saburova et al., 2021). In the TNG50 analogues, $R_e \approx 8$7–$R_e \approx 8$8, and the circular-speed decomposition is written as

$R_e \approx 8$9

There, baryons carry 60–80% of $23\,\mathrm{kpc}$0 for $23\,\mathrm{kpc}$1, but dark matter carries $23\,\mathrm{kpc}$2 for $23\,\mathrm{kpc}$3 (Sarkar et al., 10 Feb 2026). A reanalysis of Malin 1 and NGC 7589 similarly finds steeply rising inner rotation curves typical of HSB systems, allowing maximum-disk fits with $23\,\mathrm{kpc}$4 for Malin 1 and 2.5 for NGC 7589, so that baryons may dominate the inner dynamics out to $23\,\mathrm{kpc}$5–$23\,\mathrm{kpc}$6 (Lelli et al., 2010).

Mass models uniformly require substantial dark halos. The dynamical estimate

$23\,\mathrm{kpc}$7

yields $23\,\mathrm{kpc}$8–$23\,\mathrm{kpc}$9 in the review literature (Das, 2013). UGC 1378 reaches $D_{25} \gtrsim 50$0 by $D_{25} \gtrsim 50$1, and its rotation-curve decomposition requires a dominant dark halo (Saburova et al., 2019). For the TNG50 double-disk sample, host-halo virial masses are $D_{25} \gtrsim 50$2–$D_{25} \gtrsim 50$3, with stellar-to-dark matter mass ratios $D_{25} \gtrsim 50$4–0.46 and total baryon-to-dark matter ratios $D_{25} \gtrsim 50$5–0.47 (Sarkar et al., 10 Feb 2026). In an observational mass-modeling study using Burkert halos, halo masses inside the LSB radius span $D_{25} \gtrsim 50$6 to $D_{25} \gtrsim 50$7, while six of seven galaxies lie on the high-mass extension of the baryonic Tully–Fisher relation (Saburova et al., 2020).

The halo profile itself remains debated. Pseudo-isothermal, Burkert, NFW, and Einasto forms have all been used in the literature, with some studies favoring cored halos for fitting flat, extended rotation curves or bound satellite orbits around Malin 1, whereas other analyses use NFW successfully for individual systems (Lelli et al., 2010, Saburova et al., 2020, Bustos-Espinoza et al., 16 Feb 2025).

4. Star formation, stellar populations, and chemical structure

The dominant star-formation signature of gLSBGs is inefficiency rather than absence. A review gives global $D_{25} \gtrsim 50$8–$D_{25} \gtrsim 50$9, patchy star formation, and gas fractions often exceeding 0.5 (Das, 2013). In a 27-object sample, 26 bona fide gLSBGs span $100\,\mathrm{kpc}$0 to $100\,\mathrm{kpc}$1, $100\,\mathrm{kpc}$2 to $100\,\mathrm{kpc}$3, and $100\,\mathrm{kpc}$4 to $100\,\mathrm{kpc}$5, systematically below the characteristic local value $100\,\mathrm{kpc}$6 for $100\,\mathrm{kpc}$7 galaxies (Du et al., 2023). In the same sample, the offset from the local star-forming main sequence is $100\,\mathrm{kpc}$8 to $100\,\mathrm{kpc}$9 dex at the high-mass end (Du et al., 2023).

This suppression is linked to the H I-to-H$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$00 conversion bottleneck. The same study finds $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$01–$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$02, $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$03–0.4, but only $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$04–$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$05, $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$06–0.1, and $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$07–$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$08. It relates this to low H I surface densities, with self-shielding requiring

$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$09

whereas interferometric maps of several gLSBGs find peak $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$10–$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$11 (Du et al., 2023). UGC 1378 provides a resolved example: the LSB disk has $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$12, an order of magnitude below the classical Schmidt–Kennicutt law, while the HSB disk and central ring have $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$13 (Saburova et al., 2019). UGC 1382 has a gas-depletion timescale $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$14, implying that at its current efficiency it will remain in a green-valley state for many Hubble times (Hagen et al., 2016).

The spatial distribution of stellar populations is commonly inside-out. In the 27-object sample, the inner bulge-dominated regions have lower SFRs, higher stellar masses, and redder colors, while the outer disk-dominated regions have relatively higher sSFR and can lie closer to the main sequence; in some cases the outer disks tend to follow the main sequence (Du et al., 2023). Malin 1 shows flat outer color profiles beyond $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$15, localized age spreads from $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$16 to $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$17, sub-solar metallicity $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$18 in outer regions, and a model present-day $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$19 (Boissier et al., 2016). Long-slit and MUSE studies of several gLSBGs generally find old, metal-rich bulges and younger or lower-metallicity outer disks, with gas-phase metallicity gradients that are shallow to flat. For UGC 1382 and AGC 192040, MUSE yields $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$20 and $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$21, respectively (Saburova et al., 29 Sep 2025). Edge-on systems show gradients spanning 0 to $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$22, with an average $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$23 (Saburova et al., 2024).

The combination of large H I masses, weak molecular content, low $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$24, and comparatively blue outer disks indicates that gLSBGs are not quiescent in a purely passive sense. Rather, they sustain prolonged low-efficiency star formation over very large radii.

5. Formation channels and evolutionary pathways

No single formation mechanism accounts for all observed gLSBGs. A spectroscopic synthesis of six systems concludes that there is a need for diversity of formation scenarios: some could have formed by in-plane mergers of massive galaxies, while for some others the major-merger scenario is excluded by the data (Saburova et al., 2020). A related seven-object analysis formulates three scenarios: two-stage formation in which an HSB galaxy forms first and later grows an LSB disk by accreting gas from an external supply; formation in an unusually shallow and extended dark-matter halo; and a major merger with fine-tuned orbital parameters and progenitor morphologies (Saburova et al., 2020).

Evidence for a two-stage accretion pathway is strongest in systems with dynamically cold outer disks and little sign of kinematic heating. UGC 1378 is explicitly interpreted as an early epoch of HSB disk assembly followed by prolonged, quiescent accretion of low-metallicity gas that inflated a giant LSB disk; its low stellar velocity dispersion is described as inconsistent with a recent major merger (Saburova et al., 2019). Malin 1, in a panchromatic disk-evolution analysis, is reproduced by a long, quiet star-formation history in a halo with very high angular momentum, with best-fit parameters $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$25 and $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$26, and no significant event such as a collision is necessary (Boissier et al., 2016).

Other systems show much stronger merger signatures. Long-slit spectroscopy revealed kinematically decoupled central components in UGC 1922, UGC 1382, and UGC 6614, including counter-rotation (Saburova et al., 2020). In a seven-galaxy study, compact elliptical satellites were found around Malin 1 and UGC 1382, and the authors argue that this favors a major-merger origin for at least a subset of gLSBGs (Saburova et al., 2021). MUSE observations of UGC 1382 and AGC 192040 infer that both galaxies experienced mergers several Gyrs ago, but with different pathways: for AGC 192040, gas accretion from the filament followed by an intermediate-mass ratio prograde merger; for UGC 1382, multiple gas-rich mergers with companions on retrograde orbits (Saburova et al., 29 Sep 2025). Deep studies of edge-on extended LSB disks likewise favor mergers as an essential stage, based on dynamical overheating, elevated Toomre $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$27, and shallow metallicity gradients (Saburova et al., 2024).

The Malin 1 environment has become a particularly detailed test case. Orbital reconstructions using stellar, gaseous, and dark-matter potentials find that its satellites and two giant stellar streams can be linked to past interactions, with pericenter passages of order 0–4 Gyr ago depending on the scenario, and with cored/isothermal halos more favorable to bound satellite orbits than NFW halos (Bustos-Espinoza et al., 16 Feb 2025, Bustos-Espinoza et al., 12 May 2026). These studies argue that minor mergers and tidal stripping enriched the H I supply, induced star formation in the outer disk, and launched extensive stellar streams.

Cosmological simulations now recover several of these ingredients. In TNG50, the seven double-disk analogues are formed through two phases: a $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$28 regime of rapid central collapse and minor-merger-driven starbursts that build the inner HSB disk, followed by a $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$29 regime of relatively quiescent, high-angular-momentum gas accretion along cosmic filaments that forms the extended LSB disk. High halo spin parameters $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$30–0.07 correlate with the most extreme outer scale lengths (Sarkar et al., 10 Feb 2026). This suggests that late high-$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$31 accretion and merger activity are not mutually exclusive; rather, different gLSBGs may sample different mixtures of these processes.

6. Abundance, environment, nuclear activity, and current synthesis

Although gLSBGs are rare, they are not vanishingly rare. A systematic HSC search over $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$32 at $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$33 found 42 giant disks, of which 37 satisfy the LSB criterion $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$34, giving a volume density $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$35 and implying $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$36 such galaxies over the full sky out to $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$37 (Saburova et al., 2022). In that sample, 30 of the 42 giant disky systems are isolated (Saburova et al., 2022). Earlier reviews likewise emphasized that gLSBGs are overwhelmingly isolated, often found near void edges and in low-density filaments rather than in rich groups or clusters (Das, 2013).

At the same time, isolation is not equivalent to complete dynamical quiescence. Deep imaging of Malin 2 reaches $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$38 and traces the stellar disk to $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$39, revealing asymmetric diffuse structures and a candidate ultra-diffuse dwarf satellite at a projected separation of $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$40 (Junais et al., 11 Aug 2025). The new stellar asymmetries overlap a lopsided H I distribution, and the study argues that tidal processes have likely contributed to the formation of the giant disk (Junais et al., 11 Aug 2025). A larger H I study of 19 HSC-selected gLSBGs finds that 9/18 detections have strong spectral asymmetries with $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$41, much higher than the $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$42 reported for ALFALFA galaxies, and concludes that purely smooth accretion is disfavored for much of the population (Lah et al., 10 Mar 2026).

Nuclear activity is also common, though the reported incidence depends strongly on sample definition. One review gives approximately 10–15 percent with low-luminosity AGN (Das, 2013). In the HSC census, the AGN fraction among newly found gLSBGs is 19%–40% by optical and X-ray selection (Saburova et al., 2022). In a seven-object observational sample, six of seven host AGN, and several black holes lie more than 0.5 dex below the $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$43–$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$44 relation (Saburova et al., 2020). Radio-continuum observations detect compact nuclei in all seven galaxies of an AGN-selected sample and infer total star-formation rates of 0.15–$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$45 (Mishra et al., 2014).

A current synthesis, supported by both simulations and observations, is that gLSBGs are rare but real outcomes of disk-galaxy formation in massive halos. EAGLE reproduces a giant-disk abundance of $\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$46, close to the HSC estimate (Saburova et al., 2022). TNG50 produces seven massive HSB+LSB double disks with halo masses, rotation speeds, and scale lengths comparable to observed systems such as Malin 1, UGC 1382, and UGC 1378 (Sarkar et al., 10 Feb 2026). A plausible implication is that the class is best understood not as a single evolutionary channel but as a structural endpoint reachable through multiple histories: high-spin halos with prolonged high-$\mu_0 \ge 22.0\,\mathrm{mag\,arcsec^{-2}}$47 accretion, gas-rich minor accretion in isolated environments, and merger-driven rebuilding in systems that retain or recover dynamically cold giant disks.