Prompted by the many controversial claims involving the NGC 1052 group, including that it hosts two dark matter-free galaxies with overluminous and monochromatic globular cluster (GC) systems, here we map out the large-scale structure (LSS) of GCs over the entire group.
To recover the LSS, we use archival optical CFHT imaging data. We recover two GC density maps, one based on universal photometric properties of GCs from simple stellar population models, and one based on the properties of spectroscopically confirmed GCs in DF2 and DF4 (the two dwarf galaxies with overluminous GC populations). Both selection methods reveal overdensities around the massive galaxies in the group, as well as around NGC 1052 itself, that are coincident with the positions of previously identified stellar streams and tidal features. No intragroup GCs are found connecting these structures to any of the dwarf galaxies. We find, however, two other dwarfs in the group hosting GC systems. These include RCP32 with 2 GCs with ages equivalent to the GCs around NGC 1052, and DF9 with 3 GCs with ages similar to the GCs around DF2 and DF4.
We conclude that the GC distribution in the group does not strongly support any formation scenario in particular. It favours, nonetheless, scenarios relying on galaxy-galaxy interactions and on the coeval formation of GCs around the DM-free dwarf galaxies. These may include the recently proposed bullet-dwarf formation, as well as high-redshift tidal dwarf galaxy models.
The NGC 1052 group has been the topic of intense debate over several years due to claims that it hosts dark matter (DM)-free low-surface brightness (LSB) dwarf galaxies (van Dokkum et al. 2018, 2019). This came as a surprise since both dwarf (Strigari et al. 2008b,a) and LSB galaxies (de Blok & McGaugh 1997) are believed to be some of the most DM-dominated galaxies in the universe. This is because their gravitational potentials are too weak to counteract the outward pressures of stellar feedback, resulting in decreased star formation efficiencies.Two galaxies claimed to be largely DM-free within their stellar components, NGC 1052-DF2 and NGC 1052-DF4 (DF2 and DF4, hereafter), were also found to have extremely extended sizes (effective radius, π e > 1.5 kpc) for their surface brightnesses, being better classified as ultra-diffuse galaxies (UDGs). Additionally, they were found to host a population of ultra-luminous globular clusters (GCs) (van Dokkum et al. 2018, 2019; Shen et al. 2021). All of these peculiar properties are accompanied by the fact that the galaxies are very close in projection, but far away from each other in line-of-sight tip of the red giant branch (TRGB) distance (2 Mpc, Shen et al. 2021). The puzzle created by these complications has culminated in skepticism about the DM-free nature of these galaxies (Trujillo et al. 2019; Montes et al. 2020) and have also led to the development of simulations and theoretical methods to understand the possible formation pathways of such galaxies. Assuming that it is not a coincidence that the galaxies share such unusual properties while being so close in proximity, a common formation scenario for them must simultaneously explain: 1) their lack of DM, 2) their large sizes and 3) the presence of overluminous GCs.One of the most commonly proposed scenarios involves tidal stripping by a more massive galaxy (Ogiya 2018; MacciΓ² et al. 2021; Jackson et al. 2021; Ogiya et al. 2021; Moreno et al. 2022). This scenario potentially explains the DM depletion (Haslbauer et al. 2019) and extended sizes of the galaxies, and it is expected to leave behind a trail of stripped GCs, but so far it is not capable of explaining the overluminous GC population around DF2 and DF4.
Alternatively, Trujillo-Gomez et al. (2021), attempting to explain both the low DM content and the peculiar GC systems, proposed that the galaxies were formed by a combination of an early, intense burst of star-formation that creates a rich GC system with a top-heavy GC luminosity function (GCLF), and feedback that expands the galaxy. The expansion leads to an LSB galaxy residing in a diffuse DM ‘core’ creating the illusion of a DM-free galaxy. While it is an interesting possibility, this model also implies that all GC-rich LSB galaxies or ultra-diffuse galaxies should have unusual GCLFs and be DM-poor – in contradiction to the observations (e.g., Toloba et al. 2018; Saifollahi et al. 2022; Gannon et al. 2022, 2023).Recently, Silk (2019), Shin et al. (2020) and Lee et al. (2021) have proposed that DM-free galaxies with overluminous GCs could form in a ‘mini bullet cluster’ event (Clowe et al. 2006), where a high-speed collision in the host group would be able to separate baryonic- and dark-matter. These studies were the first ones capable of explaining aspects of the DM depletion and GC population, although not the spatial distribution of gas and GCs. van Dokkum et al. (2022a), building on this scenario, suggested that not only DF2 and DF4 were formed in the aftermath of this high-speed interaction, but also seven to nine other dwarf galaxies in the NGC 1052 group. The collision would have happened ∼8 Gyr ago, separating the baryonic and DM content of the progenitor galaxies. As a result, the gas would have formed a trail of DM-free galaxies (forming a near-linear distribution in projection) along with many massive GCs, while the DM itself would lie at the ends of the trail in two DM-dominated galaxies (see van Dokkum et al. 2022a, figure 1). This scenario can potentially explain all of the unusual properties of these galaxies, but it makes strong predictions. One of the most important predictions is that all of the galaxies in the trail (except the DM-dominated ones), as well as any formed GCs, should have the same ages and metallicities –hence, colours– since they would have been formed by the same process, from the same material and at the same time.
Later in the discussion section that article states:
The bullet-dwarf collision scenario makes the prediction that any GCs formed in the interaction should have small variations in colour and all be consistent with an age of ∼8 Gyr. The model suggests that the metallicities of the sources should be similar, but does not predict any specific value as it does for the ages. Previous studies, none the less, have shown that DF2 and DF4 have stellar metallicities of [π/H] ∼ −1.1 dex (Fensch et al. 2019a; Buzzo et al. 2022). If the other galaxies/GCs were formed from the same material, there is an expectation that they would have the similar metallicities. The GC colours and the stellar body of DF2 and DF4 were shown to meet these expectations (Buzzo et al. 2022; van Dokkum et al. 2022b). Apart from DF2 and DF4, only RCP32 and DF9 are found to host GC populations (both with πGC ≤ 3). No intragroup structure was identified along the bullet trail, which naturally raises the question of why the collision would form a trail of galaxies, but only a portion of them would have enough gas to form several and mostly overluminous GCs.
The finding of two GCs around RCP32 (one of the DM-dominated dwarfs) with colours similar to those of the progenitor galaxy (NGC 1052) and three GCs around DF9 with equivalent colours (within the uncertainties) to those of DF2 and DF4 GCs are consistent with the predictions of the scenario. . . .
Thus, some of our findings seem to be broadly consistent with the bullet collision, namely: 1) The GC system around RCP32 has stellar populations similar to those of the progenitor galaxy. 2) The GC system around DF9 has populations similar to those of the GCs around DF2 and DF4. 3) The predicted number of GCs around DF9 (3.6) is consistent with the finding of 3 GCs around it. 4) The lack of GCs around other dwarfs is consistent with the prediction that their GC numbers are smaller than 1. While these aspects may be broadly consistent with the scenario, the lack of constraints on the formation histories of any of the other galaxies in the trail prevents us from conclusively adhering to this scenario. . . .
Since the first claim that the galaxies in the NGC 1052 group could be dark matter-free, the most prominent model to explain their formation relied on tidal stripping by a more massive galaxy (e.g., Ogiya 2018; Ogiya et al. 2021; Moreno et al. 2022). Such models were shown to successfully explain the lack of DM and the presence of GCs in the galaxies, but even using different simulations, they were not able to create GCs as massive as the ones seen in DF2 and DF4.
Tidal stripping models, additionally, are expected to leave a trail of stripped GCs between the stripped dwarf galaxy and the massive host (Lee et al. 2010), but such structures are not seen in any of our LSS maps.
On the other hand, GC streams and other GC LSS structures are observed all around NGC 1052, indicating recent interactions between NGC 1052, NGC 1047 and NGC 1402, as suggested by MΓΌller et al. (2019b). Thus, tidal stripping could explain the connections found between the massive galaxies in the group, but do not seem to have any connection with the dwarfs.
Models that are alternative to the existence of dark matter may also be suitable explanations for the existence of these DM-free galaxies (e.g., MOND; Kroupa et al. 2012). They have been proposed as possible formation pathways for DF2 by Kroupa et al. (2018) and for DF4 by MΓΌller et al. (2019a). These models, however, do not make predictions about the globular cluster population of the galaxies nor about their stellar populations, thus, they are more difficult to test with this study. They remain a possibility to be further developed and tested.
The conclusion of the article states:
In this work, we investigated the large scale structure of GCs in the NGC 1052 group in order to test the level of agreement of different formation scenarios proposed for the dwarf galaxies in the group with the recovered GC LSS. These include the ‘bullet-dwarf’ scenario, high-redshift tidal dwarf galaxies, tidal stripping and modified gravity models. Using CFHT data in the π’, π and π bands, we select two sets of GC candidates: one based on SSP models consistent with GCs in general, with a typical range in age of 8−14 Gyr and metallicity of −2.2 < [π/H] < −0.0 dex; and a second narrower one based on the properties of spectroscopically confirmed GCs around DF2 and DF4. We subtracted the GC system of NGC 1052 to be able to see smaller overdensities around all LSB galaxies and any intragroup GC structures. The distribution of GCs around NGC 1052 was modelled to be larger than normal for an early-type galaxy of its mass, reaching a GC number radius of 10.8 galaxy effective radii. If the GC system of NGC 1052 is separated into red and blue GCs, these extend out to 8.8 and 13.2 π e, respectively. The GCs around NGC 1052 have an average colour that correspond to an age of 11.2 ± 1.6 Gyr and [π/H]= −0.8 ± 0.3 dex.
The GC map based on SSP colours revealed overdensities in the centre of the group, around NGC 1052 itself, and GC systems around the giant galaxies in the group. The overdensities around NGC 1052 coincide with the positions of pre-identified stellar streams and LSB features in the group, indicating past tidal interactions. The GC map based on the properties of spectroscopically confirmed GCs around DF2 and DF4 also revealed overdensities in the centre of the group, in the same positions as the previously identified stellar streams. Reassuringly, the GC systems of DF2 and DF4 themselves were also found. They have an average age of 9.1 ± 0.8 Gyr and [π/H]= −1.2 ± 0.2 dex, being thus younger and less metal-enriched than NGC 1052. No prominent intragroup GC structures were found in any map. On the other hand, GC systems were identified around four dwarf galaxies in the group: DF2 and DF4, as already known, but also RCP32 with 2 GCs and DF9 with 3 GCs. The average age of the GCs around RCP32 is consistent with the age of the GCs around NGC 1052, while the GCs around DF9 are consistent, within the uncertainties, with the colours of DF2/DF4 GCs.
If we assume a bullet-like formation and that all of the galaxies in the supposed trail share the averaged specific frequency of DF2 and DF4 –the trail galaxies with most GCs per unit luminosity (ππ =14.6)– we predict a total number of GCs throughout the trail of πGC = 42.3 ± 4.7. Our predictions indicate that the number of GCs around DF9 is πGC = 3.6 ± 1.2, which is consistent with our finding of 3 GCs. All of the other galaxies in the trail were found to have a combined predicted number of GCs smaller than 2. These predictions are consistent with the lack of GCs found in any of the other trail galaxies or in the intragroup medium. In addition, the colours (thus, stellar populations) of the GCs found around DF9 are similar to the ones around DF2 and DF4 and, thus, meet the stellar population expectations of the scenario. On the other hand, the GCs around RCP32 have equivalent ages to the GCs around NGC 1052 itself, which is consistent with the expectation that this DM-dominated dwarf would host the primordial GCs coming from the progenitor galaxy. While this scenario has many aspects of agreement with the recovered GC LSS, the lack of conclusive evidence on the properties of any of the other trail galaxies makes it hard to reach any definite conclusions on their formation scenario.
We discuss how TDGs formed at high-redshifts (0.5 < π§ < 2), when merger and star formation rates were higher, may be able to explain the overluminous population of GCs found in the galaxies. As they would have been formed at high-π§, the metallicities and ages of the GCs may be consistent with the ones observed in this study. TDG models, however, at low or high-redshift, do not seem to be able to explain the 2 Mpc distance between the two DMfree dwarf galaxies in the group, as this mode of formation predicts that the galaxies will live within about one or two hundred kpc from their progenitors. Thus, although promising, models of highredshift TDGs need further exploration in order to correctly explain the physical separation of the galaxies.
We discuss how tidal stripping models are unlikely to explain the dwarf galaxies in the group, as they cannot explain the overluminous GCs in the galaxies. We suggest, however, that tidal stripping could be taking place around the massive galaxies in the group, such as NGC 1052, NGC 1047 and NGC 1042, as stellar streams and tidal features are observed connecting these galaxies and are key imprints of this type of interaction. Additionally, we argue that modified gravity models remain as a possibility to explain these DM-free galaxies, but as they do not make any predictions on the stellar populations or GC systems of the galaxies, they are more difficult to test with our study.
We conclude that the GC distribution in the group does not conclusively point to any formation scenario in particular, but it favours models relying on galaxy-galaxy interactions and on the coeval formation of dwarfs and their GC systems. These include the recently proposed bullet-dwarf formation, as well as tidal dwarf galaxy models. Both scenarios need further exploration and adjustments to be fully compatible with the GC LSS observed in the group.
No comments:
Post a Comment