EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 698 (June 2025) A&A, 698 (2025) L8 Full HTML
Open Access
Issue
A&A
Volume 698, June 2025
Article Number L8
Number of page(s) 5
Section Letters to the Editor
DOI https://doi.org/10.1051/0004-6361/202554906
Published online 29 May 2025

© The Authors 2025

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.

1. Introduction

According to modern cosmological concepts, the angular momentum (spin) is a fundamental property of a galaxy and contains information about the formation and evolution of stellar systems. The idea that tidal forces are the origin of the galaxy rotation dates back to Strömberg (1934), Hoyle (1951). The tidal-torque theory developed by Peebles (1969), Doroshkevich (1970), White (1984) describes the acquisition of angular momentum by a proto-galaxy in the early Universe through tidal forces from neighboring proto-galaxies and large-scale structures. The main prediction of the tidal-torque theory is that galaxy spins are nearly perpendicular to the minor axis of the shear tensor associated with the distribution of the surrounding matter. In other words, the spin of a galaxy must lie in the plane of the wall of the large-scale structure to which it belongs (Navarro et al. 2004). This occurs because in the frame of the shear tensor eigenvectors, the three components of angular momentum depend on the difference in the eigenvalues in an analytical way, namely Ji = eijk(λj − λk), where Ji is the angular momentum in the i direction, eijk is the Levi-Cevita symbol, and λi is the eigenvalue associated with this direction. The angular momentum in the J2 direction is therefore maximized (since λ3 − λ1 is by definition greatest).

The theoretical picture was confirmed in several studies (Godłowski 1994; Tempel et al. 2013; Chen et al. 2019; Blue Bird et al. 2020). Moreover, modern mass surveys allow us to detect more subtle effects. Using the Sloan Digital Sky Survey (SDSS), Tempel & Libeskind (2013) found a difference in the orientation of elliptical and spiral galaxies. The minor axis of the ellipticals is predominantly perpendicular to the host filaments, while the spin of spirals is aligned with it. Using kinematics from the Sydney-AAO Multi-object Integral field spectrograph (SAMI) survey of the GAlaxy and Mass Assembly (GAMA) fields, Welker et al. (2020) detected a mass-dependent flip in the spin orientation: The spin of low-mass galaxies tends to be along the filament, while the spin of more massive galaxies is preferentially orthogonal to the filament. Based on 3D spins from the MaNGA survey, Kraljic et al. (2021) found a morphological segregation in galaxy orientation: Late-type predominantly low-mass galaxies are aligned, while S0-galaxies are perpendicular. It was also found that the orientation is related to the stellar mass of the bulge. Galaxies with a low-mass bulge tend to parallelize their spins with the filament, whereas high-mass bulge galaxies are more perpendicularly aligned (Barsanti et al. 2022).

On the other hand, a number of studies did not confirm a spin alignment with respect to the large-scale structure or reported only an extremely weak correlation (see for example Pahwa et al. 2016; Krolewski et al. 2019; Antipova et al. 2021). In particular, Karachentsev & Zozulia (2023) studied the angular momenta of 720 galaxies enclosed in the Local Volume of 12 Mpc around the Milky Way. A sample of 27 elite spirals with angular momenta exceeding 0.15 that of the Milky Way stands out among these 720 galaxies. They contribute more than 90% of the total angular momentum of the Local Volume. Nevertheless, their spin distribution in the sky does not show a preferred alignment with respect to the Local Sheet plane. Therefore, the question of a galaxy spin alignment relative to the large-scale structure requires further study.

The Local Supercluster appears to be an ideal laboratory to test this hypothesis. This nearest supercluster with a diameter of 30−40 Mpc is one of the most frequently studied structures in the Universe. As a result of the location of our Galaxy inside the Local Sheet, the Local Supercluster clearly stands out in the distribution of nearby galaxies in the sky as a narrow belt with a noticeable concentration of galaxies around the center in the Virgo cluster. This fact simplifies the identification of the Local Supercluster plane, which is determined by the great circle of the supergalactic coordinate system. There is a wealth of studies in the literature its structure, composition, and kinematics (see for example Makarov & Karachentsev 2011; Kim et al. 2016; Kashibadze et al. 2018).

Navarro et al. (2004) pointed out that the use of edge-on galaxies greatly simplifies the analysis. In the case of arbitrarily oriented disk galaxies, the positional angle, inclination, and side of the disk that is closer to the observer need to be known. For edge-on galaxies, the position angle (PA) is measured with a high accuracy, the inclination is fixed at 90° with an error smaller than 5°, and the position of the spin in the plane of the sky eliminates the ambiguity with the orientation. Moreover, in the case of the Local Supercluster, this further simplifies the task and reduces it to measuring the PA of the major axis of the edge-on galaxy in the supergalactic coordinates. Using this approach, Navarro et al. (2004) discovered that the nearby edge-on disk galaxies are predominantly oriented perpendicular to the Local Supercluster plane, which is consistent with numerical models of the formation of dark matter halo angular momentum within the large-scale structure. This effect is clearly visible in the histogram of supergalactic PAs for edge-on spiral galaxies with cz < 1200 km s−1 (see left panel of Fig. 2 in the original paper by Navarro et al. 2004). However, this conclusion was based on a sample of only 30 edge-on galaxies from the Principal Galaxy Catalog (PGC, Paturel et al. 1997).

The massive deep photometric and redshift surveys over two decades since the work by Navarro et al. (2004) increased the sample of known Local Supercluster galaxies by at least an order of magnitude. In addition, new catalogs of edge-on galaxies (Bizyaev et al. 2014; Makarov et al. 2022) with well-defined selection criteria have recently been published. All these allow us to use the approach proposed by Navarro et al. (2004) with much better statistics and to test the alignment of galaxy spins with respect to the plane of the Local Supercluster at a new level.

2. Sample

Our work is based on the catalog of edge-on galaxies (EGIPS, Makarov et al. 2022) found in the Pan-STARRS11 survey (Chambers et al. 2016), which is the largest list of edge-on galaxies to date. It contains 16551 objects. The EGIPS catalog covers three quarters of the sky above δ = −30°. As a result of the structure of the public Pan-STARRS1 survey and the neural network used to search for galaxies, the EGIPS catalog may undercount the largest galaxies, as well as galaxies of low surface brightness. In order to populate these gaps, we used the edge-on disk galaxies from the SDSS catalog (EGIS, Bizyaev et al. 2014), which contains 5749 objects and a sample of galaxies from the HyperLeda database (Makarov et al. 2014).

The EGIPS and EGIS catalogs specialized in the study of edge-on galaxies, but galaxies from the HyperLeda required additional preparation efforts. HyperLeda galaxies were selected according to the following parameters: vlg < 3600 and t > −3 and mod0 > 25 and inc > 85, where vlg is the object velocity, czLG, in the reference frame of the Local Group centroid (Karachentsev & Makarov 1996); t is the de Vaucouleur numerical code of the morphological type of the galaxy; mod0 is a distance modulus to exclude Local Group members; inc is the inclination estimated from the observed galaxy axial ratio and a characteristic thickness depending on its morphology. Measurement errors, intrinsic variation in characteristic thickness, and misclassifications in morphology can lead to large errors in the inclination estimates. To avoid this problem, we visually inspected all selected galaxies to rule out incorrect cases. In the final sample, we included only spiral galaxies because the inclination angle of S0 galaxies is difficult to determine.

The statistics on the subsamples of edge-on galaxies inside different redshift ranges with respect to the centroid of the Local Group (Karachentsev & Makarov 1996) are gathered in Table 1. The rows count only unique objects added to the previous catalogs. The last row gives the total number of galaxies in each subsample. The third column, phot, indicates the number of galaxies with good photometry. To analyze the alignment, Navarro et al. (2004) used only 30 galaxies with velocities lower than 1200 km s−1. Table 1 shows that the new data sample has increased by almost eight times.

Table 1.

Sample sizes.

3. Analysis

To analyze the PA distribution of the edge-on galaxies in supergalactic coordinates, we used the standard equations of spherical trigonometry to convert PA from the J2000.0 equatorial system into the supergalactic one. In our analysis, the PA was adjusted to the range [0 ° ,90 ° ) because the orientation at an angle α > 90° relative to the supergalactic plane is equivalent to an angle of (180 ° −α) < 90°.

The all-sky distribution of the edge-on galaxies with the indication of their orientations is plotted in supergalactic coordinates in the top panels of Fig. 1. For clarity, we show two layers in separate panels: nearby galaxies with czLG ≤ 1200 km s−1 (top left panel), where we expect to see the alignment effect most clearly, and more distant galaxies in the velocity range 1200 < czLG ≤ 2400 km s−1 (top right panel). For convenience, the bottom panels of Fig. 1 map the distribution of arbitrarily oriented galaxies in the same velocity range, based on the modern 50 Mpc Galaxy Catalog (Ohlson et al. 2024). The panels show the complex network of the large-scale structure in the nearby Universe at different scales. The centers of the figures correspond to the supergalactic coordinates (lSG, bSG) = (90 ° ,0 ° ), which is quite close to the center of the Local Supercluster in the Virgo cluster ∼(102.9 ° , − 2.3 ° ).

thumbnail Fig. 1.

All-sky maps of the galaxy distribution in supergalactic coordinates (Aitoff projection). Top row: Edge-on galaxies in our sample, split into two redshift intervals: czLG < 1200 km s−1 (left) and 1200 < czLG < 2400 km s−1 (right). The major axes are shown as short segments. Bottom row: Nearby galaxies from the 50 Mpc Galaxy Catalog (Ohlson et al. 2024) in the same redshift intervals as above. The colors indicate the redshifts in the Local Group centroid reference frame.

The nearest redshift region czLG ≤ 1200 km s−1 is dominated by the Local Supercluster, and most galaxies are concentrated in a narrow belt along the Supergalactic Plane. In addition, the Leo Spur, a structure at bSG ≈ −15 parallel to the Local Supercluster Sheet, and the near periphery of the Fornax Supercluster at ∼(262.5 ° , − 42.1 ° ) are clearly visible.

To our surprise, the edge-on galaxies in Fig. 1 do not show the expected prominent perpendicular orientation in the nearest czLG < 1200 km s−1 layer, either with respect to the Local Supercluster or to the Leo Spur plane.

For the quantitative analysis, we considered four subsamples. Because galaxies in the virial zone of a cluster should be randomized in velocities, positions, and orientations, in order to purify the statistics of their influence, we excluded from consideration the Virgo cluster galaxies lying within the 6° radius of M 87 with velocities czLG < 3600 km s−1. We split the remaining galaxies into three layers by redshift: czLG ≤ 1200, 1200 < czLG ≤ 2400, and 2400 < czLG ≤ 3600 km s−1. The PA distribution of these subsamples is shown in Fig. 2. The main impression is that the distribution of the galaxies across PA is fairly uniform and shows no statistically significant dominance in orientations. The galaxy disks within the virial zone appear to tend to lie in the plane of the Local Supercluster, but the Kolmogorov-Smirnov test shows that this alignment is statistically insignificant, with a p value of 0.466.

thumbnail Fig. 2.

Supergalactic PA distribution of edge-on galaxies from our final sample inside and outside the virial zone of the Virgo Cluster.

Samples of galaxies with a simple redshift selection criterion are clearly clogged with neighboring structures. To improve the selection of galaxies belonging to the Local Supercluster, we only considered objects lying within the ±2 and ±5 Mpc layers from the Local Supercluster plane. As before, the virial zone of the Vigro cluster was excluded from our consideration. Using photometry data from the 50 Mpc Galaxy Catalog (Ohlson et al. 2024), we separated the galaxies by morphology and luminosity. The distribution of edge-on galaxies by (g − i)0 color shows a pronounced bimodality, and the color (g − i)0 = 0.86 splits them into two roughly equal subsamples of blue and red galaxies. We also divided the sample by the stellar mass M* = 108.7 M⊙ into two approximately equal parts of bright and faint galaxies.

We performed the one-sample Kolmogorov-Smirnov test under the null hypothesis that the PAs are uniformly distributed. The cumulative distributions of supergalactic PA for the subsamples of different morphologies and luminosities, measured in two regions of different redshift depths, are shown in Fig. 3. The results are summarized in Table 2. The tests showed that the supergalactic PA distributions of edge-on galaxies in most of the subsamples do not differ from random at the α = 0.05 significance level. The only exception was the case for faint objects. Galaxies with low stellar masses, M* < 108.7 M⊙, show a predominantly perpendicular spin orientation relative to the Local Supercluster plane with a difference from the random distribution at a significance level better than α = 0.07 in almost all cases we considered. Despite the statistical significance of the result at approximately the 2-σ level, it might be considered a fairly robust detection, as it is observed in three out of four cases, but there are a few caveats. These samples are not fully independent because the larger regions encompass the smaller ones. The deviation from the null hypothesis for the subsample of nearby faint galaxies in the thick slice, czLG ≤ 1200 km s−1 and ±5 Mpc, is statistically insignificant. Its p value differs sharply from the other three values, although it is comparable to those of other uncorrelated subsamples.

thumbnail Fig. 3.

Cumulative PA distributions of edge-on galaxies with respect to the supergalactic plane. The panels show the distributions within the czLG < 1200 and czLG < 2400 km s−1 ranges for subsamples of the bright and faint galaxies, separated by stellar mass at M* = 108.7 M⊙, as well as for the blue and red galaxies, divided by color at (g − i)0 = 0.86 mag. The top row corresponds to the ±2 Mpc layer relative to the supergalactic plane, and the bottom row shows the distributions within the ±5 Mpc layer. The virial zone of the Virgo cluster was excluded from our consideration. The dashed black line indicates the uniform distribution.

Table 2.

Kolmogorov-Smirnov test for different subsamples of edge-on galaxies.

4. Summary

We analyzed the distribution of the galaxy orientations relative to the Local Supercluster plane following the idea proposed by Navarro et al. (2004) using a sample of 1689 nearby edge-on disk galaxies within czLG < 3600 km s−1 compiled from the EGIPS (Makarov et al. 2022) and EGIS (Bizyaev et al. 2014) catalogs and from the HyperLeda database (Makarov et al. 2014). This is larger by about an order of magnitude than in the original article.

We tested different subsamples selected by redshift, Local Supercluster sheet membership, color, and stellar mass. Unlike Navarro et al. (2004), we found no statistically significant alignment of the spins to the Local Supercluster plane either among the nearest galaxies, czLG < 1200 km s−1, or in any other subsample. The distribution of the supergalactic PAs of edge-on galaxies appears to be random, except for a subsample of faint galaxies. The disks with a low stellar mass, M* < 108.7 M⊙, that belong to the Local Supercluster show a tendency to be aligned in its plane. In other words, their spins are oriented predominantly perpendicular to the plane of the Local Supercluster. This effect is observed within the ±2 and ±5 Mpc layers relative to the supergalactic plane in regions of ≤1200 and ≤2400 km s−1 in redshift space at a significance level of 2.4−6.2%, which corresponds to 2.3−1.9σ. This result contradicts the expectation that the spin of low-mass galaxies is aligned along the filaments, as found by Welker et al. (2020) and Kraljic et al. (2021). It should be noted that this correlation breaks down in the subsample of faint galaxies within ±5 Mpc and czLG < 1200 km s−1. Given the relatively small sample size, it cannot be ruled out that this is a consequence of some statistical fluctuation.

The lack of a spin alignment to the plane of the Local Supercluster predicted by theory may arise because the galaxies are part of different filaments and are subject to complex tidal influences from the surrounding large-scale structure. Thus, although tidal force fields should lead to spin alignment, the intersections of filaments (knots) and the presence of sheets create complex time-varying gravitational fields that can lead to a random orientation of spins. The interactions at these junctions can disrupt any initial alignment.

The possible explanation for the lack of any correlations of spins and the Local Supercluster plane is given by Arakelyan et al. (2024). The authors quantitatively estimated the possible spatial and time evolution of the potential of the Milky Way analogs in the HESTIA cosmological simulations (Libeskind et al. 2020) and the role of satellite galaxies and the Local Group in the formation of the potential of the Galaxy. One of the conclusions is that the rotation axis of the galactic disk can slowly change its orientation by an angle of up to 70°. The authors do not know the reason for this behavior.

1

Panoramic Survey Telescope and Rapid Response System (Pan-STARRS).

Acknowledgments

This work was supported by the Russian Science Foundation grant No 24–12–00277. We acknowledge the usage of the HyperLeda database (http://leda.univ-lyon1.fr) (Makarov et al. 2014).

References

  1. Antipova, A. V., Makarov, D. I., & Bizyaev, D. V. 2021, Astrophys. Bull., 76, 248 [Google Scholar]
  2. Arakelyan, N. R., Pilipenko, S. V., Gottlöber, S., et al. 2024, ArXiv e-prints [arXiv:2412.18880] [Google Scholar]
  3. Barsanti, S., Colless, M., Welker, C., et al. 2022, MNRAS, 516, 3569 [NASA ADS] [CrossRef] [Google Scholar]
  4. Bizyaev, D. V., Kautsch, S. J., Mosenkov, A. V., et al. 2014, ApJ, 787, 24 [Google Scholar]
  5. Blue Bird, J., Davis, J., Luber, N., et al. 2020, MNRAS, 492, 153 [NASA ADS] [CrossRef] [Google Scholar]
  6. Chambers, K. C., Magnier, E. A., Metcalfe, N., et al. 2016, ArXiv e-prints [arXiv:1612.05560] [Google Scholar]
  7. Chen, Y.-C., Ho, S., Blazek, J., et al. 2019, MNRAS, 485, 2492 [NASA ADS] [CrossRef] [Google Scholar]
  8. Doroshkevich, A. G. 1970, Astrofizika, 6, 581 [Google Scholar]
  9. GodÅ‚owski, W. 1994, MNRAS, 271, 19 [Google Scholar]
  10. Hoyle, F. 1951, Problems of Cosmical Aerodynamics, 195 [Google Scholar]
  11. Karachentsev, I. D., & Makarov, D. A. 1996, AJ, 111, 794 [NASA ADS] [CrossRef] [Google Scholar]
  12. Karachentsev, I. D., & Zozulia, V. D. 2023, MNRAS, 522, 4740 [Google Scholar]
  13. Kashibadze, O. G., Karachentsev, I. D., & Karachentseva, V. E. 2018, Astrophys. Bull., 73, 124 [Google Scholar]
  14. Kim, S., Rey, S.-C., Bureau, M., et al. 2016, ApJ, 833, 207 [NASA ADS] [CrossRef] [Google Scholar]
  15. Kraljic, K., Duckworth, C., Tojeiro, R., et al. 2021, MNRAS, 504, 4626 [NASA ADS] [CrossRef] [Google Scholar]
  16. Krolewski, A., Ho, S., Chen, Y.-C., et al. 2019, ApJ, 876, 52 [NASA ADS] [CrossRef] [Google Scholar]
  17. Libeskind, N. I., Carlesi, E., Grand, R. J. J., et al. 2020, MNRAS, 498, 2968 [NASA ADS] [CrossRef] [Google Scholar]
  18. Makarov, D., & Karachentsev, I. 2011, MNRAS, 412, 2498 [NASA ADS] [CrossRef] [Google Scholar]
  19. Makarov, D., Prugniel, P., Terekhova, N., Courtois, H., & Vauglin, I. 2014, A&A, 570, A13 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Makarov, D., Savchenko, S., Mosenkov, A., et al. 2022, MNRAS, 511, 3063 [Google Scholar]
  21. Navarro, J. F., Abadi, M. G., & Steinmetz, M. 2004, ApJ, 613, L41 [NASA ADS] [CrossRef] [Google Scholar]
  22. Ohlson, D., Seth, A. C., Gallo, E., Baldassare, V. F., & Greene, J. E. 2024, AJ, 167, 31 [Google Scholar]
  23. Pahwa, I., Libeskind, N. I., Tempel, E., et al. 2016, MNRAS, 457, 695 [NASA ADS] [CrossRef] [Google Scholar]
  24. Paturel, G., Andernach, H., Bottinelli, L., et al. 1997, A&AS, 124, 109 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  25. Peebles, P. J. E. 1969, ApJ, 155, 393 [Google Scholar]
  26. Strömberg, G. 1934, ApJ, 79, 460 [Google Scholar]
  27. Tempel, E., & Libeskind, N. I. 2013, ApJ, 775, L42 [NASA ADS] [CrossRef] [Google Scholar]
  28. Tempel, E., Stoica, R. S., & Saar, E. 2013, MNRAS, 428, 1827 [NASA ADS] [CrossRef] [Google Scholar]
  29. Welker, C., Bland-Hawthorn, J., van de Sande, J., et al. 2020, MNRAS, 491, 2864 [NASA ADS] [CrossRef] [Google Scholar]
  30. White, S. D. M. 1984, ApJ, 286, 38 [NASA ADS] [CrossRef] [Google Scholar]

All Tables

Table 1.

Sample sizes.

In the text
Table 2.

Kolmogorov-Smirnov test for different subsamples of edge-on galaxies.

In the text

All Figures

thumbnail Fig. 1.

All-sky maps of the galaxy distribution in supergalactic coordinates (Aitoff projection). Top row: Edge-on galaxies in our sample, split into two redshift intervals: czLG < 1200 km s−1 (left) and 1200 < czLG < 2400 km s−1 (right). The major axes are shown as short segments. Bottom row: Nearby galaxies from the 50 Mpc Galaxy Catalog (Ohlson et al. 2024) in the same redshift intervals as above. The colors indicate the redshifts in the Local Group centroid reference frame.
In the text
thumbnail Fig. 2.

Supergalactic PA distribution of edge-on galaxies from our final sample inside and outside the virial zone of the Virgo Cluster.
In the text
thumbnail Fig. 3.

Cumulative PA distributions of edge-on galaxies with respect to the supergalactic plane. The panels show the distributions within the czLG < 1200 and czLG < 2400 km s−1 ranges for subsamples of the bright and faint galaxies, separated by stellar mass at M* = 108.7 M⊙, as well as for the blue and red galaxies, divided by color at (g − i)0 = 0.86 mag. The top row corresponds to the ±2 Mpc layer relative to the supergalactic plane, and the bottom row shows the distributions within the ±5 Mpc layer. The virial zone of the Virgo cluster was excluded from our consideration. The dashed black line indicates the uniform distribution.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.