betaFeedback

WorksPhysics and Astronomy

SHARP(ening the) view of unresolved globular clusters in ultra diffuse galaxies

Abstract

Ultra-Diffuse Galaxies (UDGs) challenge current galaxy formation models within the ΛCDM framework, particularly regarding their dark matter content and formation histories. Globular Clusters (GCs) serve as excellent tracers for probing the halos of these faint systems; however, current spectroscopic facilities struggle to reach the depth required to fully characterize GC populations beyond the local neighborhood. In this work, we present the scientific potential of NEXUS@SHARP, a near-infrared multi-slit spectrograph proposed on the upcoming Extremely Large Telescope (ELT). We demonstrate that NEXUS is uniquely positioned to overcome current observational limitations by combining high spatial resolution with efficient multiplexing. We show that NEXUS can efficiently help to characterize the GC population in UDGs across diverse environments out to 60 to 70 Mpc - with a few hours of ELT time. These capabilities will allow for precise measurements of radial velocities and stellar populations, enabling robust dynamical mass estimates and distinguishing between competing UDG formation scenario.

Scientific Rationale

The efficiency with which galaxies convert their gas reservoirs into stars depends on their halo mass (Moster et al., [96]; Behroozi et al., [6]). In the absence of astrophysical processes, dark matter (DM) halos of all masses would be expected to contain a baryonic component consistent with the cosmological baryon fraction (Planck Collaboration et al., [101]), Ωb/Ωm∼0.17\Omega_b/\Omega_m \sim0.17, meaning that roughly 17% of their total mass should initially be in the form of gas and, eventually, stars. This value provides a reference against which the efficiency of galaxy formation can be measured through observables such as the stellar-to-halo mass relation (SHMR, Kostov & Bonev, [74]).

In practice, halos retain and convert only a fraction of these baryons into stars due to a variety of internal and external processes that regulate galaxy growth. Gas accretion, radiative cooling, star formation, and feedback from supernovae and active galactic nuclei all modulate how much of the baryonic material can condense into galaxies. Environmental mechanisms such as ram-pressure stripping and tidal interactions further reduce baryon retention, particularly in low-mass systems. These regulatory processes are essential for reconciling theoretical predictions—such as the galaxy stellar mass and luminosity functions in Λ\LambdaCDM—with observations (Somerville & Davé, [118]). Halos with masses comparable to that of the Milky Way are thought to be the most efficient, converting about 20% of their baryons into stars. At higher masses, internal and external processes suppress gas cooling, producing lower star-formation efficiencies in massive groups and clusters. Conversely, dwarf-scale halos retain only about 1–5% of their baryons in stars; this drop is mainly driven by stellar feedback and the halos’ shallow potential wells, which make it easier for gas to escape (Bullock & Boylan-Kolchin, [13]).

Although secondary factors such as galaxy morphology, star formation history, and environment can modulate these trends, their impact is generally modest. Across cosmic time and diverse environments, halo mass remains the dominant parameter governing the overall efficiency of baryon-to-star conversion (Martizzi et al., [89]). This mass-dependent efficiency is a cornerstone of modern galaxy formation models and shapes key scaling relations linking baryonic and DM components. Modern cosmological simulations, including IllustrisTNG (Nelson et al., [99]), EAGLE (Schaye et al., [112]), and FIRE (Hopkins et al., [63]), have provided strong support for this framework, generally reproducing the broad observational trends of the SHMR. However, despite this broad agreement, the models diverge at the low-mass end, where the physical processes governing star formation and feedback are most uncertain. In particular, each simulation relies on different sub-grid prescriptions, often calibrated differently and implemented with varying numerical strategies. As a result, their predictions for the low-mass SHMR, baryon cycling efficiency, and dwarf-galaxy star-formation histories can vary by up to an order of magnitude, highlighting the sensitivity of this regime to unresolved physics and the continuing need for improved constraints from both simulations and observations [see 137 for a review on cosmological models].

The last six decades of research on the massive end of the galaxy mass function have established a bottom-up, hierarchical assembly picture [10]. In this paradigm, small DM halos collapse first and later merge into progressively larger systems, implying that low-mass galaxies are expected to reside in these early, small halos. The advent of wide and deep sky surveys has revolutionized this field by enabling detailed studies of low-mass and low-surface-brightness galaxies across environments. Those observations, together with higher-resolution simulations, have revealed several small-scale challenges to the Λ\LambdaCDM paradigm –including the cusp-core, missing-satellites, and too-big-to-fail problems [e.g. 13].

Extended low surface brightness galaxies have been known since the 1980s [110], but it was only in 2015 that they become popular, when van Dokkum et al. [132] identified dozens of ultra-diffuse galaxies (UDGs, see Figure 1) in the Coma cluster. These were defined as galaxies with central surface brightness μg,0≳24.0\mu_{g,0} \gtrsim24.0 mag/arcsec2^2 and effective radii Re≳1.5R_e \gtrsim1.5 kpc–basically galaxies with the size of the Milky Way but about 100 times less massive. Initially thought to predominantly inhabit dense environments [73, 132, 133, 134, 76, 65, 84], UDGs have since been identified in a wide range of environments, including the field [88, 56, 97, 85, 94].

Color composite image of NGC 1052-DF2

Figure 1. Color composite image obtained with the Hubble Space Telescope of NGC 1052-DF2. Image credit: NASA / ESA / P. van Dokkum, Yale University.

The origin of UDGs remains an open question, with two main formation scenarios. One proposed scenario is that UDGs are "failed" massive galaxies, L∗L^* systems quenched early (z∼2z \sim2), which preserved dwarf-like stellar masses within relatively massive halos. These galaxies would thus appear red, old, metal-poor, DM dominated, and host unusually rich globular cluster (GC) population [132, 127, 68, 123]. Alternatively, UDGs may be "puffed-up" dwarfs, genuinely low-mass galaxies that have expanded due to feedback or environmental processes [3, 33, 23, 7, 142]. Relative to the previous scenario, such objects are expected to be bluer, exhibit extended star formation histories, and have DM content and GC population, resembling those of typical dwarfs.

There are also other ways to form UDGs such as external mechanisms including galaxy interactions, mergers, and environmental stripping (e.g., Ilevleva et al., 2024) Possible origins include collisional debris from galaxy mergers [78], like tidal dwarf galaxies, or gas clumps stripped by ram pressure from infalling systems [140, 102, 71] (Ilevleva et al. 2024). These scenarios yield blue, dusty, star-forming UDGs with low metallicities, little or no DM, and significant UV emission.

While external mechanisms provide a valuable explanation for galaxies with a small amount of DM, they are unable to account for systems such as NGC 1052-DF2 [130] and NGC 1052-DF4 [129]. These two UDGs appear to exhibit a lack of DM in their inner regions, challenging our understanding of galaxy formation within the Λ\LambdaCDM framework [61]. Several explanations have been proposed, with the leading model suggesting that high-velocity interactions separated baryonic and DM while generating the pressure needed to form bright GCs (bullet dwarf scenario, [115, 131]). These galaxies indeed host unusually bright GCs, with top-heavy GC luminosity functions (GCLFs) that differentiate them from normal dwarfs [129] (Shen et al. 2021). Buzzo et al. [19] identified another DM-deficient system, FCC 24, which shares several traits with NGC 1052-DF2 and NGC 1052-DF4, including slow and prolate rotation, quiescence in low-density environments, coeval formation of stars and GCs, flat stellar population gradients, top-heavy GCLFs, and nearly monochromatic GC populations. These shared features suggest the existence and a way to identify a class of DM deficient dwarf galaxies.

In summary, current observations indicate that UDGs likely represent a heterogeneous population encompassing galaxies with distinct intrinsic properties, such as colors, stellar populations, and DM fractions.

Hence, to understand and discriminate among the different formation scenarios, we should collect information on the stellar population (field stars and GCs) and DM content in these galaxies. An example of a coherent analysis of both the stellar populations and dark-matter content of UDGs in the same environment is provided by the LEWIS survey, which studied UDGs in the Hydra I cluster (see e.g., Iodice et al., [65]; Hartke et al., [60]; Buttitta et al., [17]; Mirabile et al., [95]; Doll et al., [34]; Rossi et al., [105])

Here, we focus on the GC population because, among the properties discussed in the next section, GCs act as fossils that trace the history of a galaxy: they serve as tracers of halo mass and of the system’s dynamics, providing a direct diagnostic to distinguish among UDG formation channels (Toloba et al., [122]; Buzzo et al., [18])

Globular cluster properties

GCs are ancient (t≳8t \gtrsim8 Gyr), compact stellar systems with typical half-light radii of re∼2.5r_e \sim2.5 pc, formed during the early phases of galaxy assembly when intense star formation favored the creation of massive clusters (Larsen & Richtler, [77]; Brodie & Strader, [11]; Forbes et al., [43]). As relics of this epoch, they act as fossil records of the host galaxy formation and evolution. The GC systems in external galaxies typically exhibit a consistent luminosity function, which is well approximated by a Gaussian profile with a characteristic peak–known as the turnover magnitude (TOM)–at an absolute magnitude of MVTOM∼−7.5M_V^{\mathrm{TOM}} \sim-7.5 mag (Harris, [57]; Rejkuba, [104]). This near-universality allows the GC luminosity function (GCLF) to serve as a reliable secondary distance indicator (Ferrarese et al., [40]). The width of the GCLF is observed to depend on the host galaxy mass, with more massive galaxies showing broader GCLF (Villegas et al., [136]). The TOM also shows variations across galaxy types. In dwarf galaxies, the TOM is observed to be slightly fainter compared to more massive systems, indicating a dependence on host galaxy mass (Villegas et al., [136]). The situation is more debated for UDGs. Some studies report a TOM consistent with that of typical dwarf galaxies (Saifollahi et al., [109]; Li et al., 2025), while others find significant offsets (van Dokkum et al., [129]; Müller et al., [97]). Thanks to their high luminosity, GCs are detectable at cosmological distances (e.g., Alamo-Martínez et al., [2]).

The number of GCs hosted by galaxies varies depending on the galaxy type, from dwarf galaxies hosting zero to a few dozen, to spiral galaxies with hundreds, and up to elliptical galaxies where this number can reach several thousand. In the case of UDGs, the number of GCs ranges from zero to even ∼100\sim100 (e.g., DF-44, van Dokkum et al., [128]), which is an unexpectedly high number of GCs considering their stellar mass.

Studying the GC population of UDGs, especially estimating the number of GCs in such systems, serves as a valuable diagnostic tool to investigate the formation pathways of the UDGs (see Sect. 1). Deep imaging campaigns targeting UDGs in clusters and isolated environments (Prole et al., [103]; Lim et al., [81]; Jones et al., [70]) have quantified GC populations, revealing that some UDGs host exceptionally rich GC systems for their stellar mass, suggesting the presence of massive DM halos, while others contain GC populations consistent with ordinary dwarf galaxies (Beasley & Trujillo, [5]; Amorisco et al., [4]; Lim et al., [80]; Saifollahi et al., [107]). The total number of GCs in a galaxy correlates with its halo mass (Blakeslee et al., [9]; Harris et al., [58]; Burkert & Forbes, [16]), and is thus providing an indirect tracer of its DM content. This correlation arises because GC formation and survival scale with the depth and assembly history of the host halo, so richer GC systems trace more massive DM halos (Kruijssen, [75]). At the same time, knowing the total number of GCs (NGCN_{\mathrm{GC}}) allows us to estimate the specific frequency, defined as SN=NGC100.4(MV+15)S_N = N_{\mathrm{GC}} 10^{0.4(M_V+15)}, where MVM_V is the galaxy’s total V-band magnitude (Harris, 1991). This quantity measures the galaxy’s efficiency in forming and retaining GCs over cosmic time. A scatter in SNS_N at fixed stellar mass highlights different evolutionary pathways.

Spectroscopic studies of GCs provide valuable constraints on the evolutionary pathways of UDGs by measuring kinematic signatures that can indicate the influence of tidal interactions and environmental effects (Toloba et al., [122]). The velocity dispersion derived from the motions of GCs enables direct estimates of DM content, as higher dispersions generally point to more massive halos (Toloba et al., [123]; Buzzo et al., [19]; Haacke et al., [55]). However, caution is needed in interpreting these measurements: contamination from unbound or intracluster GCs and the use of small samples can lead to uncertainties and potentially overestimate halo masses, especially for UDGs with fewer than 10 GC candidates (Toloba et al., [123]; Doppel et al., [35]). Nevertheless, in systems where stellar kinematic tracers are absent (mainly because of the prohibitive exposure times eventually required to get spectra for the low surface brightness host galaxy), GCs remain an essential tool for probing the DM halos and testing theoretical predictions for their formation and mass assembly.

Open questions and limits of the current observations

Is the scaling relation between the NGCN_{\mathrm{GC}} and halo mass valid for UDGs?

The validity of the NGCN_{\mathrm{GC}}–halo mass relation for UDGs remains a subject of active investigation and debate. Although some studies report that the relation holds reasonably well for UDGs hosting rich GC systems (NGC≥20N_{\mathrm{GC}} \ge20, Forbes & Gannon, [44]), notable outliers appear—especially among systems with few GCs. In this low-NGCN_{\mathrm{GC}} regime, stochastic sampling, and variations in globular-cluster formation efficiency can produce substantial scatter (Georgiev et al., [52]; Harris et al., [59]; Forbes et al., [45]). The diversity of UDG formation channels and environmental histories may further complicate the straightforward application of this scaling relation.

What does the SNS_N of these galaxies tell us?

Figure 2, adopted from Mirabile et al. [95], shows the SNS_N for different dwarf and UDGs in the literature obtained from mainly photometric studies. At fixed luminosity (mass), SNS_N shows a significant scatter that may reflect underlying variations in GC formation efficiency, accretion history, and environmental dependencies. However, at low GC numbers, the GC formation process itself becomes stochastic ([52, 59]). Additionally, the estimate of the total GC population from imaging is affected by several factors. These include the selection criteria adopted for GC candidates, the statistical background subtraction techniques used to account for contamination from Milky Way stars and background galaxies ([20, 29, 21, 62, 94, 82]), as well as observational limitations such as image depth and quality. These effects make direct comparisons across studies more difficult and introduce large uncertainties in the estimated total NGCN_{GC}. Thus, whether the observed scatter at fixed stellar mass is real or the consequence of observational bias remains a matter of debate.

Specific frequency versus magnitude plot comparing Hydra I UDGs with literature values

Figure 2. Specific frequency versus magnitude plot comparing Hydra I UDGs SNS_N (light-blue diamond) with literature values of dwarfs and UDGs from different environments ([93, 53, 59, 81, 84, 67]). All literature measurements are shown as colored dots, except for the Miller & Lotz [93] data, which is displayed as a running mean (solid blue line). The black solid arrow represents the position of UDG 9, a galaxy in the Hydra I cluster for which the population of GCs is statistically consistent with zero. Adapted from: Mirabile et al. [95].

What is the shape of the GCLF in UDGs?

Studies of the GCLF in UDGs reveal intriguing diversities, supporting the existence of multiple formation pathways. Several UDGs exhibit dwarf-like GCLFs, with turnover magnitudes consistent with those of typical dwarf galaxies ([109] (Li et al. 2025)). In contrast, other systems, such as NGC 5846-UDG 1 and VCC 615, show more luminous, massive galaxy-like GCLFs ([97, 31, 92]). Moreover, some dark matter-deficient UDGs, including NGC 1052-DF2, NGC 1052-DF4, and FCC 224, display bimodal GCLFs with an additional bright component associated with UCD2-like objects that are significantly more luminous than canonical GCs ([129, 120]). This diversity suggests that GC-rich UDGs with luminous or top-heavy GCLFs may have formed in high-pressure environments, consistent with the failed galaxy or bullet dwarf scenarios, while systems with dwarf-like GCLFs are more compatible with standard dwarf galaxy evolution.

However, our understanding of the GCLF in UDGs remains incomplete, especially for the small number of large and deep studies, since many UDGs are faint and distant, making accurate measurement of the GCLF turnover is challenging. Photometric uncertainties, contamination, and limited number statistics complicate the determination of the GCLF parameters. It remains unclear whether the observed variations in the GCLF reflect genuine differences in GC formation or are due to environmental effects, selection biases, or stochasticity.

Existing spectroscopic campaigns of GC systems in UDGs. Over the last decade, deep wide-field surveys have uncovered large numbers of UDGs (e.g., [141, 85, 76, 84, 108]), revealing a heterogeneous population with diverse colors, GC richness/populations, stellar populations, and DM fractions. However, as already highlighted, imaging surveys and studies—especially those focused on GCs in such faint systems—ace several challenges. These include dataset incompleteness (e.g., insufficient depth to fully sample the GCLF), selection biases, and poor statistical removal of contaminants to extract reliable measurements. All of these factors can influence the interpretation of results when comparing different studies and may bias our understanding of this intriguing class of galaxies. One way to produce more coherent and homogeneous results is to obtain spectroscopic information for the GC candidates in these galaxies. Unfortunately, spectroscopy results to be extremely telescope-time-consuming over such the faint extra-galactic GCs, resulting in only a limited number of UDGs having been studied spectroscopically ([48], and citation therein), and even fewer with integral-field (IF) spectroscopy (e.g., [98]). For instance, in galaxy clusters, the survey Looking into the faintEst With MUSE (LEWIS, [65, 60, 17, 95]) provided the first homogeneous IF study of 18 UDGs and 6 low-surface brightness (LSB) galaxies in Hydra I galaxy cluster. At the distance of Hydra I (d∼50d \sim50 Mpc), the LEWIS survey managed to get reliable spectroscopy for the brightest GC component in the observed dwarfs. Mirabile et al. [95] obtained the first detailed study of the GCs over four galaxies in the survey—soon to be complemented with the remaining 20 targets—and identified 11 spectroscopic GCs, either host Notes: Col. 1 Galaxy cluster M200M_{200} mass taken respectively from Simionescu et al. [117]; Drinkwater et al. [36]; Tamura et al. [119]; Simionescu et al. [116]; Col. 2 is the number of UDGs (∝M2001.11\propto M_{200}^{1.11} expected in each cluster using Eq. 1 from van der Burg et al. [126]; Col. 3 lists the distance to each cluster: values for Virgo and Fornax are taken from Blakeslee et al. [8], for Hydra from Christlein & Zabludoff [26], and for Perseus from Tully et al. [125]; Col. 4 is the expected GCLF TOM; Col. 5 reports the expected exposure time to reach the TOM with MUSE assuming a G2V Pickles model, with a spatial coadding of 4×44 \times4 pixels, a spectra coadding of 2 pixels, with airmass sec⁡(z)≤1.5\sec(z) \le1.5 and a seeing ≤0.85\le0.85 arcsec; Col.6 reports the expected exposure time to reach the TOM using NEXUS/SHARP@ELT. *Not observable from Cerro Armazones; here reported only for reference.

in the galaxy, or intra-cluster ones. In Virgo (d∼16.5d \sim16.5 Mpc), Toloba et al. [123] studied a sample of 10 UDGs and, using GC velocity dispersions, for the first time, identified UDGs with apparently overmassive dark-matter halos. In Coma (d∼100d \sim100 Mpc), the spectroscopic analysis of 10 UDGs by Gannon et al. [48] shows that six are GC-rich, while the remaining four lack GC information. In the literature, there are other spectroscopic studies targeting UDGs, but only few have acquired spectroscopic information for their GC populations (e.g., [e.g. 42, 37]) Most spectroscopic works to date have focused on the integrated stellar light of UDGs, often limited to small, brighter samples due to the observational challenges posed by their low surface brightness. As a result, detailed dynamical and chemical characterization of GC systems within UDGs remains largely unexplored, restricting our ability to fully understand their DM content and formation histories from these crucial tracers. This situation underscores the importance of new instruments that will enable simultaneous spectroscopy of multiple GCs within UDGs and their field of stars, expanding the kinematic and population studies in these faint systems.

Objectives

To shed light on the nature of UDGs, deep spectroscopic observations of a large sample of UDGs are needed, as combining the analysis of GC richness with the kinematic and evolutionary properties of field stars (age and [Fe/H]) provides a powerful diagnostic for distinguishing between different formation channels. Such observations also offer the opportunity to measure the DM content of these galaxies independently from their stellar kinematics, GC counts, and GC kinematics (e.g., [e.g. 123, 17, 95]). Here, we focus on the insights that can be gained from studying the GC population.

Perspectives with SHARP

SHARP is a near-infrared (0.95−2.45 μm0.95 - 2.45\ \mu\mathrm{m}) spectrograph ([111, 83]) specifically designed to leverage the capabilities of MCAO-assisted observations at the ESO’s Extremely Large Telescope (ELT), through the MORFEO adaptive optics system ([27]). The instrument comprises two complementary units: NEXUS, a multi-object Spectrograph, and VESPER, a multi-Integral Field Unit system. With a pixel scale of 35 mas (NEXUS) and 31 mas (VESPER), SHARP achieves spatial sampling of ∼150\sim150–250 pc (the scale of giant molecular clouds) across cosmic time, surpassing JWST’s angular resolution by factors of 3-6. The instrument characteristics are described in more detail by Saracco et al. in this volume. In Sect. 2.2 we give the specific SHARP configuration recommended for the science case discussed here.

To address the number of open questions on UDG formation (see Sect. 1.2), particularly through their GC component, deep spectroscopic observations sampling the GCLF at least down to the TOM for a large number of UDGs are necessary. Such observations could provide robust measurements of the total NGCN_{\mathrm{GC}} in each galaxy by confirming cluster membership through radial velocity estimates. In Table 1, we report a list of prototypical galaxy clusters at increasing distance that, if targeted by deep spectroscopic observations with SHARP, could allow to study a large number of UDGs due to their high richness. Nevertheless, isolated UDGs remain of great interest for constraining their possible formation channels.

EnvironmentM200(M⊙)M_{200}(M_\odot)NUDGsN_{\mathrm{UDGs}}DD (Mpc)mVTOMm_V^{\mathrm{TOM}} (mag)MUSE Exp. Time (h)SHARP Exp. Time (h)
(1)(2)(3)(4)(5)(6)(6)
Virgo1.05×10141.05 \times10^{14}∼20\sim20∼16.5\sim16.523.5∼7\sim7∼0.08\sim0.08 (5 min)
Fornax7.0×10137.0 \times10^{13}∼15\sim15∼20\sim2024>100> 100∼0.1\sim0.1 (8 min)
Centaurus/Hydra I2.1×10142.1 \times10^{14}∼45\sim45∼50\sim5026≫500\gg500∼5\sim5
Perseus*6.65×10146.65 \times10^{14}∼150\sim150∼70\sim7026.8≫103\gg10^3∼20\sim20

Table 1. Properties of different environments and expected UDG population

A key motivation for reaching at least the GCLF turnover magnitude is that doing so enables a complete and unbiased census of the GC population in each UDG. Accurate determination of NGCN_{\mathrm{GC}} is essential for testing the empirical scaling relation between GC number and halo mass. Additionally, it allows us to determine whether UDGs conform to the “universal” GCLF shape, or whether they display systematic differences—such as a deficit of low-mass clusters, excess of luminous clusters, or a shift in the turnover luminosity. Moreover, the stellar populations of the GCs can be investigated, providing crucial information on their ages and metallicities. This allows to connect the dynamical and luminosity properties of the GCs with their formation history, helping to disentangle whether observed variations in the GCLF are driven by intrinsic differences in GC formation, subsequent evolution, or environmental effects within the host UDGs. So, the combination of accurate NGCN_{\mathrm{GC}} measurements, dynamical mass constraints, and the GCLF properties provides a powerful multi-dimensional diagnositc of the evolutionary pathways of UDGs through their GCs, helping to determine whether they represent faded dwarfs, failed massive galaxies, environmentally altered systems, or a mixture of these scenarios.

Deep spectroscopic data of GCs in UDGs will not only deepen our understanding of GC systems in currently known UDGs, but will also exploit and complement previous observational efforts. For example, deep imaging surveys (e.g., Ferrarese et al. [39]) with HST, Subaru, and CFHT have provided initial catalogs of GC candidates, but without spectroscopic confirmation, these studies are limited by contamination, distance uncertainties, and incomplete sampling of the GCLF. At the same time, deep spectroscopy will also create strong synergies with ongoing and upcoming facilities and surveys such as Euclid, LSST, and JWST. These missions are dramatically increasing the number of UDGs and other low-surface-brightness systems accessible for study, over wide areas and to unprecedented depths, enabling the construction of catalogs of GC-rich low-surface-brightness galaxies for targeted follow-up with deep spectroscopic observations.

However, obtaining such detailed spectroscopic data for faint systems at these distances presents a significant challenge for current facilities. In the following section, we describe the observational setup and the specific instrument capabilities required to achieve these objectives efficiently, with particular attention to the methods and possibilities allowed by a unique facility like SHARP.

Spectroscopy of GCs in UDGs: Existing observing facilities

With the aim of observing GC populations in UDGs, we present here the case study to target multiple GCs simultaneously to capture GCLF, measure radial velocities, and characterize their stellar populations. Among the best available ground-based instruments for this purpose, we considered MUSE on the ESO VLT, which can target multiple GCs simultaneously and cover the spectral features needed to measure radial velocities and stellar population properties.

The integration times required to reach the depth needed to sample the GCLF down to the TOM at distances typical of nearby clusters are listed in Table 1. Mirabile et al. [95] showed that even ∼6\sim6 hours of exposure with MUSE are insufficient both to reach the GCLF turnover magnitude for UDGs at the distance of the Hydra I cluster (∼50\sim50

Mpc; Christlein & Zabludoff [26] and to obtain spectra with adequate S/N for robust analysis, highlighting the difficulty of building statistically complete GC samples with current-generation facilities. Figure 3 presents an example GC spectrum observed with MUSE@VLT in Hydra I for a source of mH≃22.5m_H \simeq22.5 mag (Mirabile et al. [95]); the red curve shows the pPXF fit (Cappellari [22]). The figure clearly illustrates the low S/N achieved for such sources and underscores the observational difficulty of using advanced IFUs such as MUSE at the VLT, in spectroscopic studies of GC populations at these distances.

Spectrum of a GC with SNR of 7 Å⁻¹

Figure 3. Spectrum of a GC (black line) with SNR of 7 Å−1^{-1}, along with the pPXF best-fit model (red solid line). Regions with strong sky residual lines are masked from the fit (grey shaded areas). The main absorption features (Hα_{\alpha} and Hβ_{\beta}) are drawn with blue vertical dotted lines. Figure and caption adapted from: Mirabile et al. [95].

On the other hand, future ELT instruments such as MICADO (Davies et al. [32]), HARMONI (Thatte et al. [121]), and ANDES (see https://andes.inaf.it/) will offer powerful spectroscopic capabilities but limited multiplexing. MICADO will offers long-slit coverage, HARMONI is a single-IFU spectrograph, with a FOV that reaches a few square arcseconds across, and ANDES will provide a high-resolution single long slit plus IFU mode. These modes either provide only single-slit coverage or IFU fields of only a few arcsec2^2, so they cannot simultaneously capture multiple GCs across the typical angular extent of a UDG; this limits their efficiency for building statistically complete GC samples.

An exception is MOSAIC@ELT (Pelló et al. [100]), a fiber-fed multi-object spectrograph, designed to obtain spectra for up to 200 targets within a FOV ∼40\sim40 arcmin2^2. However, MOSAIC will operate with a Ground Layer Adaptive Optics system, which corrects only for the ground layer of atmospheric turbulence and therefore delivers a more modest improvement in image quality compared to the full MCAO correction provided by MORFEO for SHARP. Similarly, next-generation telescopes such as the Giant Magellan Telescope (GMT) (Burgett et al. [14]) and the Thirty Meter Telescope (TMT), will host optical/near-infrared multi-object spectrographs with larger fields of view (several arcminutes), which will be suitable for the study of GCs in UDGs. However, because GMT and TMT have smaller primary mirrors than the ELT, reaching the same S/N on the faintest GC targets will require longer exposure times. Within this landscape, the NEXUS@SHARP/ELT concept occupies a valuable niche. Briefly, to highlight the main characteristics of interest for the scientific case proposed here, NEXUS is a near-infrared multi-slit spectrograph with adaptive optics (AO) correction, offering a field of view of approximately 1.2′×1.2′1.2' \times1.2' and a fine pixel scale of ∼0.035\sim0.035 arcsec/pixel. It covers a broad wavelength range from 0.95 to 2.45 μm2.45\ \mu\mathrm{m}, divided into four channels roughly corresponding to the Y, J, H, and K bands. Its configurable slit system can deploy ∼30\sim30 slits (incidentally a good match for the expected GC4 population in GC-rich UDGs) of variable width along two possible axes, with adjustable lengths. NEXUS achieves spectral resolutions of R∼6000R \sim6000, ∼2000\sim2000, or ∼300\sim300, making it flexible for both kinematic and stellar population studies.

NEXUS results to be ideally suited for GC studies in UDGs because it combines high spatial resolution, thanks to AO correction with efficient multiplexing. Its multi-slit capability allows simultaneous observation of multiple GCs within the typically compact fields of UDGs, substantially improving observational efficiency. The AO correction delivers diffraction-limited performance, enabling precise measurements of faint clusters. Together, the capabilities of NEXUS@SHARP and the large collecting area of the ELT enable the study of GC systems in galaxies out to distances of ∼60\sim60 to 70 Mpc within reasonable observing times.

Observational requirements for NEXUS

Using the SHARP Exposure Time Calculator5 (version 0.6), we estimated the exposure times required for spectroscopic observations of GCs in UDGs. These calculations are based on stellar population models from Bruzual & Charlot [12], assuming a metallicity of Z=0.004Z = 0.004, an age of 10 Gyr, and a Chabrier initial mass function (Chabrier, [24]). We assumed a spectral resolution R=2000R = 2000, airmass sec⁡(z)≤1.5\sec(z) \le1.5, and a slit width of 200 mas. The target signal-to-noise ratio, obtained at the reference wavelength of 1.6 μm1.6\ \mu\mathrm{m}, was set to ∼5\sim5 at the TOM of the GCLF, which corresponds to detecting half of the GC population.

Last column in Table 1 summarises the resulting exposure times for NEXUS as a function of distance, providing a practical guide for planning observations that can reliably detect and characterise GC populations in UDGs. These estimates highlight the efficiency of NEXUS/SHARP@ELT to achieve the required depth to reach the TOM in relatively short exposure times compared to current-generation facilities, enabling the sampling of the GCLF up to the TOM, measurement of radial velocities, and characterization of stellar population properties for multiple GCs within a single pointing.

In summary, simulations with the current NEXUS ETC indicate that NEXUS will enable GC studies in UDGs out to distances at least three times larger than those currently accessible—probing a volume of the local Universe roughly ten times bigger. A single NEXUS pointing will provide radial velocities, total GC counts, and stellar-population information for multiple GCs, enabling statistically meaningful samples of UDGs across environments (clusters and the field). Measurements of GC system radial velocities and total numbers will permit independent estimates of galaxy dark-matter content and tests of scaling relations (see Sect. 1.2). Comparing GC ages and metallicities with those of the host galaxy stellar population can help discriminate between different formation channels: in the failed-galaxy scenario GCs and field stars should show similar ages and metallicities, whereas in a puffed-up/diffuse dwarf origin GCs are expected to be systematically more metal-poor than the host stellar population (e.g., Gannon et al., [47]).

Conclusion

As described in the introductory Section, UDGs can be efficiently investigated through the spectroscopic characterization of their GC populations. Present studies, however, are limited to the nearby Universe within ∼20\sim20 Mpc, as larger distances require prohibitive telescope efforts even for a single target. Probing deeper, thus sampling larger volumes of the local Universe and increasing the sample size across a broader range of host environments (clusters, groups, interacting regions, outskirts, field, etc.), is a mandatory requirement for fully constraining galaxy formation models at the faintest surface brightness levels. In this work, we have demonstrated that the proposed NEXUS@SHARP spectrograph at the Extreme Large Telescope represents a transformative step forward for this field. Our proposed scientific case analysis indicates:

  1. Feasibility: NEXUS can reach the GCLF turnover magnitude (MKTOM∼−7.5M_{K}^{TOM} \sim-7.5) with signal-to-noise (∼5\sim5) in remarkably short exposure times (from the ∼5\sim5 minutes at the distance of Virgo, to ∼5\sim5 hours for Centaurus/Hydra UDGs), a result unachievable with current facilities.

  1. Efficiency: The combination of Adaptive Optics and multi-slit capabilities allows for the simultaneous characterization of multiple GCs within a single UDG, drastically reducing the telescope time required for statistical completeness.

  1. Scientific Impact: These observations will provide the necessary kinematic data (e.g. GC system velocity dispersion) and total NGCN_{GC} to derive dynamical masses free from many of the assumptions that affect stellar mass–based estimates. The combination of precise NGCN_{GC} measurements and dynamical mass determinations therefore allows a self-consistent evaluation of the scaling relation within the same dataset. Agreement between the halo masses inferred from GC counts and from dynamical modeling would reinforce the validity of the scaling relation in UDGs. Conversely, any systematic discrepancies would immediately point to a breakdown of this relation, offering new physical insight into GC formation efficiency, halo assembly, and baryon–DM coupling in low-density galaxy environments. Additionally, these

observations will help to understand the nature of the GCLF in these galaxies, and resolve the debate over the UDGs formation scenario.

By enabling the systematic spectroscopic survey of GCs in low-surface-brightness systems, NEXUS@SHARP will act as a definitive tool for constraining the formation channels of the most elusive galaxy populations in the local universe.

Acknowledgments

The SHARP team acknowledges support by Bando Ricerca Fondamentale INAF 2022, Techno-Grant "SHARP" - 1.05.12.02.01 and Bando Ricerca Fondamentale INAF 2024, Large-Grant "SHARP" - 1.05.24.01.01. Part of this work was supported through the INAF “Astrofisica Fondamentale” GO-grant 2024 n. 12 (PI: M. Cantiello). MM acknowledges support form the ESO Studentship Program. The authors thank Gannon et al. [48] for the compilation of their catalogue of UDG spectroscopic properties. The catalogue includes data from: [90]; [132]; [5]; [87]; [139]; [88]; [127, 128]; [72]; [130]; [122]; Gu et al. (2018); [81]; [106]; [1]; [41]; [45]; [86]; [25]; [38]; [30]; [129]; [124]; Iodice et al. (2020); [28]; [98]; [50]; [80]; [97]; Shen et al. (2021); Ji et al. (2021); Huang & Koposov (2021); [46, 51]; [91]; [31]; [135]; [138]; [109]; [68]; [49]; [42]; [123]; [114].

References

References

  1. [1]Alabi, A., Ferré-Mateu, A., Romanowsky, A. J., et al. 2018, MNRAS, 479, 3308
  2. [2]Alamo-Martínez, K. A., Blakeslee, J. P., Jee, M. J., et al. 2013, ApJ, 775, 20
  3. [3]Amorisco, N. C. & Loeb, A. 2016, MNRAS, 459, L51
  4. [4]Amorisco, N. C., Monachesi, A., Agnello, A., & White, S. D. M. 2018, MNRAS, 475, 4235
  5. [5]Beasley, M. A. & Trujillo, I. 2016, ApJ, 830, 23
  6. [6]Behroozi, P. S., Marchesini, D., Wechsler, R. H., et al. 2013, ApJL, 777, L10
  7. [7]Benavides, J. A., Sales, L. V., Abadi, M. G., et al. 2023, MNRAS, 522, 1033
  8. [8]Blakeslee, J. P., Jordán, A., Mei, S., et al. 2009, ApJ, 694, 556
  9. [9]Blakeslee, J. P., Tonry, J. L., & Metzger, M. R. 1997, AJ, 114, 482
  10. [10]Blumenthal, G. R., Faber, S. M., Primack, J. R., & Rees, M. J. 1984, Nature, 311, 517
  11. [11]Brodie, J. P. & Strader, J. 2006, ARA&A, 44, 193
  12. [12]Bruzual, G. & Charlot, S. 2003, MNRAS, 344, 1000
  13. [13]Bullock, J. S. & Boylan-Kolchin, M. 2017, ARA&A, 55, 343
  14. [14]Burgett, W., Bernstein, R., Ashby, D., et al. 2024, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol.
  15. [16]Burkert, A. & Forbes, D. A. 2020, AJ, 159, 56
  16. [17]Buttitta, C., Iodice, E., Doll, G., et al. 2025, A&A, 694, A276
  17. [18]Buzzo, M. L., Forbes, D. A., Jarrett, T. H., et al. 2025a, MNRAS, 536, 2536
  18. [19]Buzzo, M. L., Forbes, D. A., Romanowsky, A. J., et al. 2025b, A&A, 695, A124
  19. [20]Cantiello, M., Cappacioli, M., Napolitano, N., et al. 2015, A&A, 576, A14
  20. [21]Cantiello, M., Grado, A., Rejkuba, M., et al. 2018, A&A, 611, A21
  21. [22]Cappellari, M. 2017, MNRAS, 466, 798
  22. [23]Carleton, T., Errani, R., Cooper, M., et al. 2019, MNRAS, 485, 382
  23. [24]Chabrier, G. 2003, PASP, 115, 763
  24. [25]Chilingarian, I. V., Afanasiev, A. V., Grishin, K. A., Fabricant, D., & Moran, S. 2019, ApJ, 884, 79
  25. [26]Christlein, D. & Zabludoff, A. I. 2003, ApJ, 591, 764
  26. [27]Ciliegi, P., Agapito, G., Aliverti, M., et al. 2021, The Messenger, 182, 13
  27. [28]Collins, M. L. M., Tollerud, E. J., Rich, R. M., et al. 2020, MNRAS, 491, 3496
  28. [29]D’Abrusco, R., Cantiello, M., Paolillo, M., et al. 2016, ApJL, 819, L31
  29. [30]Danieli, S., van Dokkum, P., Conroy, C., Abraham, R., & Romanowsky, A. J. 2019, ApJL, 874, L12
  30. [31]Danieli, S., van Dokkum, P., Trujillo-Gomez, S., et al. 2022, ApJL, 927, L28
  31. [32]Davies, R., Alves, J., Clénet, Y., et al. 2018, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 10702, Ground-based and Airborne Instrumentation for Astronomy VII, ed. C. J. Evans, L. Simard, & H. Takami, 107021S
  32. [33]Di Cintio, A., Brook, C. B., Dutton, A. A., et al. 2017, MNRAS, 466, L1
  33. [34]Doll, G., Buttitta, C., Iodice, E., et al. 2026, A&A, 707, A88
  34. [35]Doppel, J. E., Sales, L. V., Benavides, J. A., et al. 2024, MNRAS, 529, 1827
  35. [36]Drinkwater, M. J., Gregg, M. D., & Colless, M. 2001, ApJL, 548, L139
  36. [37]Fahrion, K., Beasley, M. A., Gvozdenko, A., et al. 2025, A&A, 697, A145
  37. [38]Fensch, J., van der Burg, R. F. J., Jeřábková, T., et al. 2019, A&A, 625, A77
  38. [39]Ferrarese, L., Côté, P., Cuillandre, J.-C., et al. 2012, ApJS, 200, 4
  39. [40]Ferrarese, L., Côté, P., MacArthur, L. A., et al. 2020, ApJ, 890, 128
  40. [41]Ferré-Mateu, A., Alabi, A., Forbes, D. A., et al. 2018, MNRAS, 479, 4891
  41. [42]Ferré-Mateu, A., Gannon, J. S., Forbes, D. A., et al. 2023, MNRAS, 526, 4735
  42. [43]Forbes, D. A., Bastian, N., Gieles, M., et al. 2018a, Proceedings of the Royal Society of London Series A, 474, 20170616
  43. [44]Forbes, D. A. & Gannon, J. S. 2025, MNRAS, 543, L1
  44. [45]Forbes, D. A., Read, J. I., Gieles, M., & Collins, M. L. M. 2018b, MNRAS, 481, 5592
  45. [46]Gannon, J. S., Dullo, B. T., Forbes, D. A., et al. 2021, MNRAS, 502, 3144
  46. [47]Gannon, J. S., Ferré-Mateu, A., & Forbes, D. A. 2026, PASA, 43, e031
  47. [48]Gannon, J. S., Ferré-Mateu, A., Forbes, D. A., et al. 2024, MNRAS, 531, 1856
  48. [49]Gannon, J. S., Forbes, D. A., Brodie, J. P., et al. 2023, MNRAS, 518, 3653
  49. [50]Gannon, J. S., Forbes, D. A., Romanowsky, A. J., et al. 2020, MNRAS, 495, 2582
  50. [51]Gannon, J. S., Forbes, D. A., Romanowsky, A. J., et al. 2022, MNRAS, 510, 946
  51. [52]Georgiev, I. Y., Puzia, T. H., Goudfrooij, P., & Hilker, M. 2010, MNRAS, 406, 1967
  52. [53]Georgiev, I. Y., Puzia, T. H., Hilker, M., & Goudfrooij, P. 2009, MNRAS, 392, 879
  53. [55]Haacke, L., Forbes, D. A., Gannon, J. S., et al. 2025, MNRAS, 539, 674
  54. [56]Habas, R., Marleau, F. R., Duc, P.-A., et al. 2020, MNRAS, 491, 1901
  55. [57]Harris, W. E. 2001, in Saas-Fee Advanced Course 28: Star Clusters
  56. [58]Harris, W. E., Blakeslee, J. P., & Harris, G. L. H. 2017, ApJS, 836, 67
  57. [59]Harris, W. E., Harris, G. L. H., & Alessi, M. 2013, ApJ, 772, 82
  58. [60]Hartke, J., Iodice, E., Gullieuszik, M., et al. 2025, A&A, 695, A91
  59. [61]Haslbauer, M., Banik, I., Kroupa, P., & Grishunin, K. 2019, MNRAS, 489, 2634
  60. [62]Hazra, N. et al. 2024, in preparation
  61. [63]Hopkins, P. F., Wetzel, A., Kereš, D., et al. 2018, MNRAS, 480, 800 Huang, K.-W. & Koposov, S. E. 2021, MNRAS, 500, 986
  62. [65]Iodice, E., Hilker, M., Doll, G., et al. 2023, A&A, 679, A69
  63. [67]Janssens, S. R., Forbes, D. A., Romanowsky, A. J., et al. 2024, MNRAS, 534, 783
  64. [68]Janssens, S. R., Romanowsky, A. J., Abraham, R., et al. 2022, MNRAS, 517, 858
  65. [70]Jones, M. G., Karanakarana, A., Bennet, P., et al. 2023, ApJL, 942, L5
  66. [71]Junais, Boissier, S., Boselli, A., et al. 2022, A&A, 667, A76
  67. [72]Karachentsev, I. D., Makarova, L. N., Sharina, M. E., & Karachentseva, V. E. 2017, Astrophysical Bulletin, 72, 376
  68. [73]Koda, J., Yagi, M., Yamanoi, H., & Komiyama, Y. 2015, ApJL, 807, L2
  69. [74]Kostov, A. & Bonev, T. 2018, Bulgarian Astronomical Journal, 28, 3
  70. [75]Kruijssen, J. M. 2026, in Encyclopedia of Astrophysics, Volume 4, Vol. 4, 500–534
  71. [76]La Marca, A., Iodice, E., Cantiello, M., et al. 2022, A&A, 665, A105
  72. [77]Larsen, S. S. & Richtler, T. 2000, A&A, 354, 836
  73. [78]Lelli, F., Duc, P.-A., Brinks, E., et al. 2015, A&A, 584, A113
  74. [80]Lim, S., Côté, P., Peng, E. W., et al. 2020, ApJ, 899, 69
  75. [81]Lim, S., Peng, E. W., Côté, P., et al. 2018, ApJ, 862, 82
  76. [82]Lonare, P., Cantiello, M., Mirabile, M., et al. 2024, arXiv e-prints, arXiv:2412.18015DOI
  77. [83]Mahmoodzadeh, H., Saracco, P., Conconi, P., et al. 2025, Journal of Astronomical Telescopes, Instruments, and Systems, 11, 035002
  78. [84]Marleau, F. R., Cuillandre, J. C., Cantiello, M., et al. 2024, arXiv e-prints, arXiv:2405.13502DOI
  79. [85]Marleau, F. R., Habas, R., Poulain, M., et al. 2021, A&A, 654, A105
  80. [86]Martin, G., Kaviraj, S., Laigle, C., et al. 2019, MNRAS, 485, 796
  81. [87]Martin, N. F., Ibata, R. A., Lewis, G. F., et al. 2016, ApJ, 833, 167
  82. [88]Martínez-Delgado, D., Lásker, R., Sharina, M., et al. 2016, AJ, 151, 96
  83. [89]Martizzi, D., Vogelsberger, M., Torrey, P., et al. 2020, MNRAS, 491, 5747
  84. [90]McConnachie, A. W. 2012, AJ, 144, 4
  85. [91]Mihos, J. C., Durrell, P. R., Toloba, E., et al. 2022, ApJ, 924, 87
  86. [92]Mihos, J. C., Durrell, P. R., Toloba, E., et al. 2025, ApJ, 978, 93
  87. [93]Miller, B. W. & Lotz, J. M. 2007, ApJ, 670, 1074
  88. [94]Mirabile, M., Cantiello, M., Lonare, P., et al. 2024, A&A, 691, A104
  89. [95]Mirabile, M., Cantiello, M., Rejkuba, M., et al. 2026, A&A, 705, A117
  90. [96]Moster, B. P., Naab, T., & White, S. D. M. 2013, MNRAS, 428, 3121
  91. [97]Müller, O., Durrell, P. R., Marleau, F. R., et al. 2021, ApJ, 923, 9
  92. [98]Müller, O., Marleau, F. R., Duc, P.-A., et al. 2020, A&A, 640, A106
  93. [99]Nelson, D., Springel, V., Pillepich, A., et al. 2019, Computational Astrophysics and Cosmology, 6, 2
  94. [100]Pelló, R., Puech, M., Prieto, E., et al. 2024, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 13096, Ground-based and Airborne Instrumentation for Astronomy X, ed. J. J. Bryant, K. Motohara, & J. R. D. Vernet, 1309615
  95. [101]Planck Collaboration, Aghanim, N., Akrami, Y., et al. 2020, A&A, 641, A6
  96. [102]Poggianti, B. M., Ignesti, A., Gitti, M., et al. 2019, ApJ, 887, 155
  97. [103]Prole, D. J., Hilker, M., van der Burg, R. F. J., et al. 2019, MNRAS, 484, 4865
  98. [104]Rejkuba, M. 2012, Ap&SS, 341, 195
  99. [105]Rossi, L., Buttitta, C., Doll, G., et al. 2026, arXiv e-prints, arXiv:2605.2709
  100. [106]Ruiz-Lara, T., Beasley, M. A., Falcón-Barroso, J., et al. 2018, MNRAS, 478, 2034
  101. [107]Saifollahi, T., Trujillo, I., Beasley, M. A., Peletier, R. F., & Knapen, J. H. 2021, MNRAS, 502, 5921
  102. [108]Saifollahi, T., Vogel, K., Lancon, A., et al. 2024, arXiv e-prints, arXiv:2405.13500DOI
  103. [109]Saifollahi, T., Zaritsky, D., Trujillo, I., et al. 2022, MNRAS, 511, 4633
  104. [110]Sandage, A. & Binggeli, B. 1984, AJ, 89, 919
  105. [111]Saracco, P., Conconi, P., Arcidiacono, C., et al. 2025, arXiv e-prints, arXiv:2512.17042arxiv.org/abs/2512.17042
  106. [112]Schaye, J., Crain, R. A., Bower, R. G., et al. 2015, MNRAS, 446, 521
  107. [114]Shen, Z., van Dokkum, P., & Danieli, S. 2023, ApJ, 957, 6
  108. [115]Silk, J. 2019, MNRAS, 488, L24
  109. [116]Simionescu, A., Allen, S. W., Mantz, A., et al. 2011, Science, 331, 1576
  110. [117]Simionescu, A., Werner, N., Mantz, A., Allen, S. W., & Urban, O. 2017, MNRAS, 469, 1476
  111. [118]Somerville, R. S. & Davé, R. 2015, ARA&A, 53, 51
  112. [119]Tamura, T., Makishima, K., Fukazawa, Y., Ikebe, Y., & Xu, H. 2000, ApJ, 535, 602
  113. [120]Tang, Y., Romanowsky, A. J., Gannon, J. S., et al. 2025, ApJ, 982, 1
  114. [121]Thatte, N. A., Melotte, D., Neichel, B., et al. 2024, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 13096, Ground-based and Airborne Instrumentation for Astronomy X, ed. J. J. Bryant, K. Motohara, & J. R. D. Vernet, 1309614
  115. [122]Toloba, E., Lim, S., Peng, E., et al. 2018, ApJL, 856, L31
  116. [123]Toloba, E., Sales, L. V., Lim, S., et al. 2023, ApJ, 951, 77
  117. [124]Torrealba, G., Belokurov, V., Koposov, S. E., et al. 2019, MNRAS, 488, 2743
  118. [125]Tully, R. B., Rizzi, L., Shaya, E. J., et al. 2009, AJ, 138, 323
  119. [126]van der Burg, R. F. J., Hoekstra, H., Muzzin, A., et al. 2017, A&A, 607, A79
  120. [127]van Dokkum, P., Abraham, R., Brodie, J., et al. 2016, ApJL, 828, L6
  121. [128]van Dokkum, P., Abraham, R., Romanowsky, A. J., et al. 2017, ApJL, 844, L11
  122. [129]van Dokkum, P., Danieli, S., Abraham, R., Conroy, C., & Romanowsky, A. J. 2019, ApJL, 874, L5
  123. [130]van Dokkum, P., Danieli, S., Cohen, Y., et al. 2018, Nature, 555, 629
  124. [131]van Dokkum, P., Shen, Z., Keim, M. A., et al. 2022, Nature, 605, 435
  125. [132]van Dokkum, P. G., Abraham, R., Merritt, A., et al. 2015, ApJL, 798, L45
  126. [133]Venhola, A., Peletier, R., Laurikainen, E., et al. 2017, A&A, 608, A142
  127. [134]Venhola, A., Peletier, R. F., Salo, H., et al. 2022, A&A, 662, A43
  128. [135]Villaume, A., Romanowsky, A. J., Brodie, J., et al. 2022, ApJ, 924, 32
  129. [136]Villegas, D., Jordán, A., Peng, E. W., et al. 2010, ApJ, 717, 603
  130. [137]Vogelsberger, M., Marinacci, F., Torrey, P., & Puchwein, E. 2020, Nature Reviews Physics, 2, 42
  131. [138]Webb, K. A., Villaume, A., Laine, S., et al. 2022, MNRAS, 516, 3318
  132. [139]Yagi, M., Koda, J., Komiyama, Y., & Yamanoi, H. 2016, ApJS, 225, 11
  133. [140]Yozin, C. & Bekki, K. 2015, MNRAS, 452, 937
  134. [141]Zaritsky, D., Donnerstein, R., Dey, A., et al. 2019, ApJS, 240, 1
  135. [142]Zheng, H., Liao, S., Gao, L., & Jiang, F. 2025, arXiv e-prints, arXiv:2504.14973arxiv.org/abs/2504.14973

Paper details

vlnc:W_jurgff5g
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 11

downloads · 1

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents