Cosmicflows-3: Cosmography of the Local Void
R Brent Tully, Daniel Pomarede, Romain Graziani, Helene M Courtois, Yehuda Hoffman, Edward J Shaya
DDraft version May 22, 2019
Preprint typeset using L A TEX style AASTeX6 v. 1.0
COSMICFLOWS-3: COSMOGRAPHY OF THE LOCAL VOID
R. Brent Tully,
Institute for Astronomy, University of Hawaii, 2680 Woodlawn Drive, Honolulu, HI 96822, USA
Daniel Pomar`ede
Institut de Recherche sur les Lois Fondamentales de l’Univers, CEA, Universite’ Paris-Saclay, 91191 Gif-sur-Yvette, France
Romain Graziani
University of Lyon, UCB Lyon 1, CNRS/IN2P3, IPN Lyon, France
H´el`ene M. Courtois
University of Lyon, UCB Lyon 1, CNRS/IN2P3, IPN Lyon, France
Yehuda Hoffman
Racah Institute of Physics, Hebrew University, Jerusalem, 91904 Israel
Edward J. Shaya
University of Maryland, Astronomy Department, College Park, MD 20743, USA
ABSTRACT
Cosmicflows-3 distances and inferred peculiar velocities of galaxies have permitted the reconstruction of the structureof over and under densities within the volume extending to 0 . c . This study focuses on the under dense regions,particularly the Local Void that lies largely in the zone of obscuration and consequently has received limited attention.Major over dense structures that bound the Local Void are the Perseus-Pisces and Norma-Pavo-Indus filaments sepa-rated by 8,500 km s − . The void network of the universe is interconnected and void passages are found from the LocalVoid to the adjacent very large Hercules and Sculptor voids. Minor filaments course through voids. A particularlyinteresting example connects the Virgo and Perseus clusters, with several substantial galaxies found along the chainin the depths of the Local Void. The Local Void has a substantial dynamical effect, causing a deviant motion of theLocal Group of 200 −
250 km s − . The combined perturbations due to repulsion from the Local Void and attractiontoward the Virgo Cluster account for ∼
50% of the motion of the Local Group in the rest frame given by the cosmicmicrowave background.Key words: large scale structure of universe — galaxies: distances and redshifts INTRODUCTIONThe average place in the universe is in a void. TheLocal Void (Tully & Fisher 1987) subtends 40% of thesky and begins 1 Mpc away, at the fringe of the Lo-cal Group. Over the eons, matter evacuates from voidsand builds up in adjacent sheets, filaments, and knots,the components of the cosmic web (Bond et al. 1996).Most of the matter that makes up our galaxy and that of our neighbors must have come out of the Local Void soour relationship to that structure is fundamental to at-tempts to understand details of the local neighborhood(Shaya & Tully 2013; Carlesi et al. 2016).There is increasingly good information about the kine-matics of nearby galaxies from distance measurementsusing the tip of the red giant branch technique thatconclusively demonstrates the motions of galaxies awayfrom the Local Void (Karachentsev et al. 2015; Rizzi a r X i v : . [ a s t r o - ph . C O ] M a y et al. 2017; Shaya et al. 2017; Anand et al. 2018). Stud-ies of the nearby region provide a unique opportunity:only nearby are deviant velocities comparable to cosmicexpansion velocities to the degree that these motions canbe cleanly separated in individual cases. So motions areobserved consistent with expansion of the Local Void.Are these motions of an amplitude that theory wouldanticipate?The Local Void has been difficult to study because itis located behind the center of the Milky Way. It is solarge that it easily protrudes on both sides of the galac-tic plane, but much of it is obscured. This paper givesattention to a way to study the morphology of the LocalVoid that is relatively insensitive to direct observation. Cosmicflows-3 (CF3) is a collection of 18,000 galaxy dis-tances (Tully et al. 2016) that, although deficient in thezone of obscuration, captures the essence of structureall-sky through two alternative analyses. Both analy-ses assumes that structure forms from Gaussian initialfluctuations within a Λ Cold Dark Matter universe withmatter and energy densities characterized by Ω m = 0 . Λ = 0 . V pec , derived from distancemeasurements, d , where to first approximation peculiarvelocities are decoupled from observed velocities, V obs ,as V pec = V obs − H d , with H the value of the Hub-ble Constant consistent with the ensemble of the data.With the current collection of distances the appropriatevalue is H = 75 km s − Mpc − . The direct productsare the three-dimensional velocity field and associateddensity field in the linear regime.The other method, found to be compatible with theWiener Filter procedure and used in the model describedin this paper, follows the work by Lavaux (2016) and isdescribed in detail by Graziani et al. (2019). In simpleterms, peculiar velocities imply a distribution of den-sity perturbations that, in turn, imply a velocity field.A Bayesian procedure is used to estimate the poste-rior probability of a specific velocity field given the lin-ear theory relationship between densities and velocitiesthrough the observed distances with assigned errors.In addition to constraints on the velocity field andcorrelated distances, the model solves for a velocitydispersion parameter, σ NL , that accommodates depar-tures from linear theory, and an effective Hubble Con- stant. The model begins with a fiducial value of H =75 km s − Mpc − (Tully et al. 2016) but searches forthe optimum of a parameter h eff anticipated to benear unity (whence H = 75 h eff ). There are uncer-tainties in both velocities and distances. Those on ve-locities are relatively minor and are approximated by σ cz = 50 km s − . The errors in distances, in the mod-ulus, are much more substantial. In recognition that Cosmicflows-3 is a heterogeneous collection of distances,Graziani et al. (2019) give separation to five sub-sampleswith each one described by a distinct selection function.A model that abides by these constraints is sampled bythe Markov Chain Monte Carlo (MCMC) method of theGibbs sampling algorithm (Lavaux 2016), whereby eachfree parameter is drawn from its conditional probabilitygiven specification of the other parameters. The proce-dure is described in detail by Graziani et al. (2019) butin brief: (a) the parameter h eff is sampled, marginal-ized over the velocity field; (b) the conditional prob-ability of the parameter σ NL is drawn with the otherparameters fixed; (c) a constrained realization of thedensity field is drawn assuming a ΛCDM power spec-trum (Hoffman & Ribak 1991); (d) a new set of dis-tances is established from the sampled constrained re-alization with probabilities set by the current values of h eff , σ NL , and the velocity field, within priors on thedistances. The process is carried through ∼ MCMCsteps until convergence. The procedure has been carriedout on multiple constrained realizations and mock cat-alogs. With the current analysis, Graziani et al. (2019)find h eff = 1 . ± .
01 and σ NL = 280 ±
35 km s − .The resultant model makes predictions for the mor-phology and motions of structure locally within theΛCDM framework and linear perturbations. Ourpresent interest is in voids. It will be asked to whatdegree the overall model is in agreement with the excel-lent knowledge we have of the motions of very nearbygalaxies. MORPHOLOGIES OF NEAREST VOIDSThe Local Void does not have a simple shape. More-over, as the void is followed to shallower levels it mergeswith adjacent voids, as part of a continuous networkthat extends beyond the volume that can currently bemapped. The three-dimensional interplay between com-plex high and low density structures is visually confus-ing. We should not have the ambition to get into greatdetail.As a prelude, previous efforts to identify nearby voidscan be mentioned, derived from regions of emptinessin maps of the distribution of galaxies in redshift sur-veys. Among the earliest were the seminal studies of theBo¨otes Void by Kirshner et al. (1981) and the void infront of the ”Great Wall” Coma and Abell 1367 clusters(Gregory & Thompson 1978). On very large scales thereis the pioneering work by Batuski & Burns (1985) andEinasto et al. (1994) on the concentrations and absencesof rich clusters. More nearby and pertaining to the dis-tribution of individual galaxies, of note is the work ofKauffmann & Fairall (1991) and Fairall (1998) who, inthe latter reference gives a list of 33 void-like regionswithin 8,500 km s − . Elyiv et al. (2013) have produceda more quantitatively rigorous catalog of 89 voids within3,000 km s − ; spherical regions with no known galax-ies brighter than M K = − .
4. Typically these enti-ties are modest in size with radii ∼ and to two interactive models. The com-plex three-dimensional nature of large scale structure ismost easily dissected with the capabilities of zoom andmotion of a movie and interactive models.2.1.
Local Void
Reigning in the focus to nearby, consider the struc-ture represented in Figure 1. Here we see a smootheddescription of over dense regions in our vicinity extend-ing to ∼ ,
000 km s − . The Local Sheet with ourMilky Way at the origin of the plot lies at a density lessthan the lowest grey contour. Major knots are identified:the Virgo Cluster, the Perseus-Pisces filament (Haynes& Giovanelli 1988), the Coma Cluster within the GreatWall climbing to the Hercules complex (de Lapparentet al. 1986) and, nearer home, the Great Attractor re-gion (Dressler et al. 1987) with the Pavo-Indus filamentrising above it connecting to a feature we call the Arch(Pomar`ede et al. 2017) that caps the Local Void andprovides a connection to Perseus-Pisces.The structure shown in Figure 1 is entirely derivedfrom an analysis of departures from cosmic expansionfrom samples of galaxies with measured distances. Thespecific rendition shown in this figure is extracted fromthe quasi-linear construction described by Hoffman et al.(2018). Thanks to the large scale coherence of velocityflows, loss of information in the zone of obscuration hasminimal impact on the derived model and features arerobust within ∼ ,
000 km s − where the density of testparticles with distance measures is high.With the upper left panel of Figure 2, the same ref-erence perspective is preserved but we move in closer.Here and in following figures unless explicitly stated, thelayered surfaces are density iso-contours of the Grazianiet al. (2019) reconstruction derived from Cosmicflows-3 distances. Over or under densities, δ ( r ), follow from thegradient of velocities, v , in linear theory: δ ( r ) = −∇ · v /H f (1)where f is the growth rate of structure assuming stan-dard ΛCDM parameters. The over density surfaces be-gin at δ = 0 .
75 in grey and progress through increasinglystrong shades of red with δ levels 1.00, 1.25, 1.50, 1.75,2.00, 2.25. The under dense levels are − . − . − . https://vimeo.com/326346346/35088b5dd8 https://sketchfab.com/models/f0a44df256aa4faf93391887d66010e2and https://sketchfab.com/models/78885b3d303d4b6e99cfe099b43929fb Figure 1 . Overview of the structure surrounding the Local Void. Isosurfaces of density are inferred from the velocity fieldconstructed from the Wiener Filter treatment of Cosmicflows distances, with the densest peaks in red and less dense filamentsin grey. The Milky Way is at the origin of the colored arrows, 5,000 km s − in length, oriented in the frame of supergalacticcoordinates (red toward +SGX, green toward +SGY, blue toward +SGZ). The Local Void fills the empty region above theMilky Way in this plot. This view inward from a location at positive values of SGX, SGY, and SGZ will be referred to as thereference orientation. specified in the figure caption. The overdense contoursare stripped away in the lower right panel to fully revealthe Local Void.We introduce a naming convention that will be ad-hered to in subsequent figures. The names of familiarstructures are retained. Otherwise, features are givenconstellation names appended with a tag set by theirredshift in units of 10 km s − , with the tags of un-der densities preceded by a minus sign and those ofover densities preceded by a plus sign. Here in the Lo-cal Void, Lacerta − . − .
89 at supergalactic SGX, SGY, SGZ of[+1650, − − ≈ [+22, −
9, +22] Mpc. Andromeda − . − .
53 at[+2100, − − − ≈ [+28, − −
4] Mpc and,in the most familiar part of the Local Void, Aquila − . − .
13 at SGX, SGY, SGZ of[ − − − ≈ [ − −
3, +9] Mpc in ourimmediate vicinity only 10 Mpc away. More removed,UMi − . − .
93 at [+3100, +1700,+1200] km s − ≈ [+41, +23, +16] Mpc. Details regard-ing these minima are accumulated in Table 1.The deepest minima in the Local Void lie at verylow values of SGY; i.e., they lie close to the equato-rial plane of the Milky Way in regions of obscuration.The void manifests a tilt toward positive SGX, toward Figure 2 . The heart of the Local Void. The deepest parts of the void are mapped by surfaces of density − . − . − has removed the Arch to give an unrestricted view of the void (02:32). In the lower right panel, the LocalVoid contours are shown alone, looking in from positive SGY (02:42). the space in front of the Perseus-Pisces filament which isthe well documented domain of a void (Haynes & Gio-vanelli 1986). The CF3 velocity information resolvesambiguity in mapping based on redshift surveys, ag-gravated by galactic obscuration, and clearly identifiesthe Local Void and the void foreground of the Perseus-Pisces complex as parts of the same feature. The ”hy-pervoid” HV1 defined by the union of 56 small sphericalvoids by Elyiv et al. (2013) reasonably approximates ourLocal Void. The rough dimensions of the Local Voidat the isodensity contour − . − = 69,51,60 Mpc, hence a volumeof ∼ × Mpc .A personalized tour of the Local Void stripped of overdense boundaries (Figure 2, lower right panel) can be ex- perienced by accessing the first interactive model. Thesuperimposed orbits were derived from
Cosmicflows-3 distance constraints using numerical action methods(Shaya et al. 2017). The orbits are calculated in co-moving space coordinates following the center of massof the sample. The orbits from z = 4 to today dramat-ically illustrate the evacuation of the Local Void. Seealso the sequence in the video frozen in the frame imageof Figure 3. 2.2. Hercules Void https://sketchfab.com/models/f0a44df256aa4faf93391887d66010e2 Figure 3 . Orbits derived from the numerical action methods of Shaya et al. superimposed on the Local Void iso-densitycontours. Orbits systematically descend out of the void (06.11). In this figure only, the green-blue (SGY-SGZ) coordinatearrows have length 3500 km s − . The Local Void is not isolated from other voids, butbefore investigating its growth at lesser under densi-ties let us become familiar with the other two principaldensity depressions within 8,000 km s − . The one inthe north galactic hemisphere (supergalactic SGY > − . − − ≈ [ −
16, +53,+67] Mpc. A list of secondary minima is given in Ta-ble 1. Those called ”arrowhead” bracket the Arrowhead mini-supercluster (Pomar`ede et al. 2015). The minimaat negative SGZ are parts of what has been called theSouthern Local Void (Einasto et al. 1983). These depres-sions link up as one considers less negative density levels.Increasing the isodensity cut to more positive values, theentire region behind the traditional Local Supercluster(de Vaucouleurs 1953) and in front of the Great Wall (deLapparent et al. 1986) is revealed to be under dense. AsLindner et al. (1995) point out, though, this volume isnot devoid of galaxies. The region is laced with tenu-ous filaments making connections that span the widthof the region. An example of a filament network con-nection between the Virgo and Coma clusters is illus-trated by Tully & Trentham (2008). The full dimensionof what we are calling the Hercules Void at isodensity − . − = 155,104,187 Mpc, a volume of ∼ × Mpc . Atthe isodensity contour of − . Sculptor Void
South of the galactic plane (SGY <
0) the most adja-cent dominant depression has been called the SculptorVoid (Fairall 1998). Its domain is illustrated in Figure 5and the density minima within this extensive void arelisted in Table 1. The nearest basin at density − . − − − − ≈ [ − − −
23] Mpc (Reticulum − . − .Our Sculptor Void is very large, with rough dimen-sions at the isodensity level − . − = 193,87,136 Mpc, enclosing ∼ × Mpc . At isodensity level − . − − . − .
52. Atits near side this void links with the Sculptor Void nearthe Canis Major − . − . − . should be launched in order to immersively ex-perience the panapoly of nearby voids. https://sketchfab.com/models/78885b3d303d4b6e99cfe099b43929fb MORPHOLOGY OF THE WALLSA detailed discussion of the over dense regions willbe left to another day, but we give attention here tothe immediate walls around the Local, Hercules, andSculptor voids.With the Local Void, the dominant bookend bound-ing features are the Perseus-Pisces complex at SGX ∼ +4500 km s − (Haynes & Giovanelli 1988) and at SGX ∼− − the Pavo-Indus arm rising out of the re-gion called the Great Attractor through the Norma Clus-ter (Dressler et al. 1987; Kraan-Korteweg et al. 1996).At right angles, the most prominent features at posi-tive and negative supergalactic latitudes are the Archat SGZ ∼ +4200 km s − and the Centaurus-Puppis-PPfilament at SGZ ∼ − − (Pomar`ede et al. 2017).These ceiling and floor of the Local Void are poorly doc-umented in redshift surveys because of galactic obscu-ration but coherent velocity patterns provide robust re-constructions.The orthogonal directions of ± SGY, by contrast, aretoward the uncontaminated and well observed galac-tic poles. In the galactic north the Local Void isbounded by de Vaucouleurs’ Local Supercluster atSGY ∼ +2500 km s − running through the Virgo andCentaurus clusters (de Vaucouleurs 1956). For all theimportance we have given this structure it is not verysubstantial. The Local Void easily makes connectionsthrough this region with the Hercules Void. The LocalVoid limits are even more porous at negative SGY. DeVaucouleurs’ Southern Supercluster in Fornax and Eri-danus and the strand extensions described by Courtoiset al. (2013) are the weak separators from the Sculp-tor Void at SGY ∼ − − . We will return laterto a discussion of wispy galaxy filaments bounding andpermeating the Local Void.Turning attention to the Hercules Void, dominantstructures on the back side are the Great Wall at SGY ∼ +7000 km s − (de Lapparent et al. 1986) merging intothe Hercules complex at SGZ ∼ +7000 km s − (Bah-call & Soneira 1984) and further merging into the Ophi-uchus (Johnston et al. 1981) and Libra+8 structures atSGX ∼ − − Cancer filament and the Arrow-head mini-supercluster (Pomar`ede et al. 2015). Abovethe supergalactic equator (SGZ >
0) essentially every-where locally densities are below the mean. It is throughthis space that the Hercules and Local voids connect.Shaya et al. (1995) have recorded the kinematic mani-festation of this general under density in the ubiquitousflow toward negative SGZ of nearby galaxies (see Fig-ure 3 and related interactive model and video sequence).
Figure 4 . The Hercules Void. The deepest density minimum are shown with contours of blue at the same density levels as inFig. 2. The locations of local minima are identified by red dots and names. Major bounding overdensities are identified. Thisextended void lies to the foreground of the high density complex of clusters in Hercules. Multiple lesser extrema lie throughoutthe extended void that occupies the space from behind the traditional Local Supercluster and Great Attractor complex to theforeground of the Great Wall. The left panel shows a view from positive SGX and SGZ, slightly rotated from the referenceviewing direction (video frame time 07:15). The view in the right panel is almost along the negative SGY axis, close to theviewing direction in the lower left panel of Fig. 2, but with a foreground clip at SGY=+2200 km s − to afford minimal obstruction(07:25). At present, the full Hercules Void is poorly constrainedon the +SGX side.
Cosmicflows-4 , the next edition ofour catalog of distances will provide more satisfactorycoverage of this part of space.On the other side of the sky, the Sculptor Void is heldin the far side embrace of the Southern Wall (Pellegriniet al. 1990) running at roughly SGY ∼ − − from the Perseus-Pisces region all the way to structureat the celestial South Pole. This latter feature appearsto be very important and we expect to discuss it in de-tail in a future paper. The nearer side of the SculptorVoid is bounded by the minor structures on the negativeSGY side of the Local Void discussed above. There iseasy penetration between the Sculptor and Local voidswhere de Vaucouleurs’ Southern Supercluster peters outbeyond the Fornax-Eridanus complex. In detail, we seethe Southern Wall as bifurcating into what we call theNorth Fork (Pegasus+8.5 filament) and the South Fork(Grus-Pisces Austrinus+10 filament). The North Forkconnects through Capricornus+7 to Ophiuchus forminga roof over the Local Void extending to above the Her-cules Void. At large values of negative SGX and SGYthe Sculptor Void boundaries are at the challenging lim-its of our reconstruction and dissolve in places into whatwe call the Eridanus Void. THE V-WEB REPRESENTATION OF VOIDSThe construction of structure up to this point in thediscussion have been based on a model of the den-sity field derived from the divergence of the three-dimensional velocity field in accordance with linear the-ory. An alternative representation is derived from a cal-culation of the shear of the velocity field at a given lo-cation (Hoffman et al. 2012).Σ αβ = − ( ∂ α v β + ∂ β v α ) / H (2)where partial derivatives of the velocity v are deter-mined along directions α and β of the orthogonal su-pergalactic Cartesian axes, normalized by the averageexpansion given by the Hubble Constant, H . Eigenval-ues indicating collapse have negative values.The eigenvectors of the shear define the principal axesof collapse and expansion. Knots, filaments, sheets, andvoids are associated, respectively, with 3, 2, 1, and 0positive eigenvalues. These four domains can be sepa-rated by surfaces of the eigenvalues. We refer to suchrepresentations as the cosmic velocity (V) web (Hoffmanet al. 2017; Pomar`ede et al. 2017).The current interest is in the voids, locations withexpansion along all three axes. Figure 7 presents an al-ternate to the density isocontour plot of Figure 6; the Figure 5 . The Sculptor Void. Shades of yellow are used at density levels consistent with the previous 2 figures. Once again,local minima are identified as well as prominent features on the bounding walls. The bottom view is from near the northsupergalactic pole, positive SGZ, with a foreground clip at SGZ=+3000 km s − to remove obstructions (08:31). The SouthernWall is a defining boundary at negative SGY. In the top panel, the view is in from near the positive SGY axis. A foregroundclip at SGY=+2000 km s − and an extraction of the immediate area around the Virgo Cluster provides windows onto the void(08:21 and 08:49). V-web representation of voids and sheets. Here, regionswith expansion on three axes (voids) are shown as solidcolors, consistent with the schema in previous figures,while regions with expansion on only two axes (sheets)are shown by transparent surfaces in related colors. Wehave chosen an arbitrary eigenvalue level for the displayof the sheet isosurface that roughly parallels the arbi-trarily chosen density isocontours of previous figures. The alternative V-web and isodensity representationsare similar (of course they are drawn from the samedata and analysis) but there are curious differences. Thedeepest basins have the same locations. However it isinteresting as an example to give attention to the sheet-topology link between the Hercules and Sculptor voidsbypassing the Local Void seen in the top panel of Fig. 7(near the feature named Sagittarius − . Figure 6 . All and only the voids. Surfaces of all voids in the
Cosmicflows-3 model at the density level − .
7. The Local Void iscolored black, the Hercules Void is blue, the Sculptor Void is yellow and all other voids are colored green. The view is from thereference orientation, with the Milky Way at the origin of the red, green, blue arrows (10:06). that minor filaments separate the Local Void from atunnel connecting the Sculptor and Hercules voids. STRUCTURE DEFINED BY A REDSHIFTSURVEYIt is worth briefly to compare structural features de-fined by
Cosmicflows-3 velocities with the redshift spacedistribution of galaxies. The current interest is in un-derdense regions where there are relatively few galaxies.We give attention alternatively to the walls that boundvoids and to the minor strands of galaxies that can per-meate voids and give separation to adjacent minima.Our comparisons are made with the redshift compila-tion V8k that was described by Courtois et al. (2012).The sample consists of 30,124 galaxies within a cube thatextends from the origin ± − on the cardinalaxes in supergalactic coordinates. This sample coversthe entire sky reasonably uniformly except at the Galac-tic plane and provides relatively dense coverage locally where we can most meaningfully make comparisons.A wide angle comparison between the V8k redshiftsample and the iso-density outline of the Local Void canbe seen in the accompanying video in frames following03:28. Almost all the individual galaxies lie outside thecontours of the void although several filamentary fea-tures adhere closely to the void boundaries (emphasizedtransiently in blue in the video). The few exceptionswhere filaments penetrate the void are worth commen-tary.A very sparse filament dramatically spans between theVirgo Cluster and the Perseus Cluster in a very directroute that takes it through a deep minimum in the Lo-cal Void. The commencement is the Local Sheet thatincludes the Milky Way and extends from the proxim-ity of Virgo to the NGC 1023 Group. In the NearbyGalaxies Atlas (Tully & Fisher 1987) the continuationis called the Perseus Cloud, basically receiving a newname just because of the tenuousness and severe obscu-1 Figure 7 . V-web representation of voids. Voids (expansion on 3 cardinal axes) are represented by solid surfaces, with the LocalVoid in black, the Hercules Void in blue, the Sculptor Void in Yellow, and other voids in green. Sheets (expansion on 2 axes,collapse on the third) are represented by transparent surfaces at an arbitrary eigenvalue. Locations and names of deepest densitytroughs are carried over from previous figures. In the top panel, the view is from the reference direction while in the lower panelthe scene has been rotated to a view from negative SGX, SGY, positive SGZ. Table 1 . Locations of density minima within the nearest voidsVoid Density SGX SGY SGZ SGX SGY SGZ Descriptionkm s − km s − km s − Mpc Mpc MpcLocal − .
89 +1650 −
700 +1650 +22 − − . − .
53 +2100 − −
300 +28 − − − . − . − −
200 +700 − − − . − .
93 +3100 +1700 +1200 +41 +23 +16 UMi − . − . − −
16 +53 +67 Hercules − . − . −
200 +6400 +5000 − − − .
55 +2100 +3500 − −
16 UMa − .
3; Lower ArrowheadHercules − . − − −
41 +53 −
72 Sextans − . − . −
200 +5000 − − −
23 Leo − .
2; ComaHercules − . − −
16 +41 +28 Serpens Caput − . − .
12 +1200 +5400 − −
67 Leo − . − .
00 +3500 +4500 +700 +47 +60 +9 UMa − .
8; Far ArrowheadSculptor − . − − − − − −
23 Reticulum − . − . − − − −
28 +35 Capricornus − . − .
53 +1700 − −
35 +3 Pisces − . − .
43 +1700 − −
60 +53 Pegasus − . − . − − − −
53 +40 Telescopium − . − .
32 +2600 − −
28 +4 Pisces − . − . − − − −
72 +35 Pisces Austrinus − . − . − − − −
55 +9 Pavo − . − . − − − −
60 +9 Sculptor − . − . − − − − − −
67 Puppis − . − . − − − − − −
60 Canis Major − . − .
38 +1100 − − − −
29 Cetus − . − . − − − − − −
48 Chamaeleon − . − .
17 +700 − −
116 +123 Aquarius − . − . − − − −
135 +37 Indus − . − . − − − − − −
91 South Pole − − but the V8k sample reveals theextension all the way to the vicinity of the Perseus Clus-ter. This wispy filament is isolated in the two panels ofFigure 8. Remarkably, it passes very close to the deepdensity minimum Andromeda − . L K = 11 .
36 assuming a distance of 32 Mpc. Atleast three more large galaxies with log L K >
11 residein the vicinity: NGC 1169, NGC 1186, and PGC 11586.The second case that draws our attention involves en-tities called the Pegasus Cloud and Pegasus Spur in theNearby Galaxies Atlas. Associated galaxies are illus-trated in Figure 9. The Pegasus Spur lies closely outsidethe mid iso-density contour confining the Local Void.The Pegasus Cloud is particularly interesting becauseit coincides with a higher density ‘tunnel’ between theabysses Lacerta − . − . CONSEQUENCES OF THE LOCAL VOID ONOUR MOTIONIt is to be noted that Lacerta − .
4, the deepest part ofthe Local Void, is aligned with the anti − apex of the cos-mic microwave background dipole. Hoffman et al. (2017)have already brought attention to the likely importanceof a very large underdensity (the Dipole Repeller) ingenerating the 631 km s − motion of the Local Groupwith respect to the microwave background frame. Wesuggest that the Local Void makes a significant additivecontribution.The apex of the Local Group motion in supergalac-tic coordinates is toward SGL = 139,
SGB = −
31. Themajor influences that give rise to this motion can be sep-arated into nearby, intermediate, and far domains. Wewill use two distinct approaches to evaluate the sourcesof our deviant motion.We begin with the first of these. The nearby domainwas studied in detail by Shaya et al. (2017) with nu-merical action orbit reconstructions, illustrated in Fig-ure 3 and following 05:50 in the accompanying video.This domain extends to 38 Mpc ∼ − andincludes the traditional Local Supercluster dominatedby the Virgo Cluster. It excludes the so-called GreatAttractor region, the densest part of our Laniakea Su-percluster. Structure beyond 38 Mpc was representedin the numerical action model by external tidal fields asgiven by a Wiener filter linear theory rendition based on Cosmicflows-2 distances (Tully et al. 2014).Taking the average of the orbits of the Milky Way andM31, we find a Local Group SGX, SGY, SGZ motionwith respect to the center of mass within 38 Mpc to be[ − − − . It can be seen from the posi- tion of the Local Group within the embrace of the LocalVoid (Figures 2 and 3) that the velocities toward nega-tive SGX and SGZ can be attributed in large measureto repulsion from the Local Void. The amplitudes areconsistent with values determined locally (Anand et al.2019) (Anand et al. 2019b in press).The Local Void might contribute to the positive SGYmotion but in this direction the Virgo overdensity mustdominate. In the Shaya et al. model the spherical vol-ume centered on the Virgo Cluster extending to theLocal Group is a factor 1.39 above mean density. Inthe spherical approximation, this overdensity would at-tract us at ∼
300 km s − . Tidal squeezing (from a fila-ment running into Virgo parallel to SGX) and distending(from voids at ± SGZ) are complications to this estimate.In the numerical action model the differential Virgo − Local Group velocity is 200 km s − . In this model, theSGY motion in the near region of ∼ +300 km s − isroughly the sum of a pull of ∼
200 km s − from theVirgo overdensity and a push of ∼
100 km s − from theLocal Void.The intermediate domain, from 38 Mpc to roughly100 Mpc is dominated by the competition between theLaniakea and Perseus − Pisces attractors. The core ofLaniakea (the Great Attractor) contains the clustersCentaurus, Norma, Hydra, Pavo II, A3365, A3537,A3574, and S753. Tugging in the opposite direction arethe chain of clusters including Perseus, Pisces, A262,A347, and NGC 507. Laniakea is clearly winning in in-fluence at our position. To a reasonable approximation,it is the intermediate domain that dominates the tidalforces on the inner 38 Mpc zone as calculated from theWiener filter model based on
Cosmicflows-2 distances.The bulk motion of the inner region due to these mostlyintermediate zone influences has the SGX, SGY, SGZvector components [ − − − (Shayaet al. 2017). The SGX and SGY components reflectthe competition between Laniakea and Perseus − Pisceswhile the SGZ component reflects the great extent ofthe Local Void, reaching well beyond the 38 Mpc limitof the near region.The far domain must account for the remainder. Thesum of the inner and intermediate zones produce LocalMotion of [ − − − , leaving a residualof [ − − −
28] km s − , attributable to a distantpull from the Shapley Concentration and push from theDipole Repeller (Hoffman et al. 2017). The separationbetween the intermediate and far domains is approxi-mate. Part of the attribution to the intermediate do-main may arise from the far domain.Uncertainties in the one-dimensional components ofvelocity deviations in these various ranges are estimatedto be at the level of ±
40 km s − . The estimate is ap-proximate because our break-out of influences is approx-4 Figure 8 . Two rotated views of the Perseus Cloud filament passing from the Virgo Cluster, past the Milky Way, through thedeep Local Void minimum of Andromeda-2.3, to the vicinity of the Perseus Cluster. The image in the inset is of the giantlenticular galaxy NGC 1161, and its spiral companion NGC 1160, deep within the Local Void. (video frames 04:44 to 05:21) Figure 9 . Two views of the filaments Pegasus Cloud (galaxies in blue) and Pegasus Spur (galaxies in magenta) that threadthrough the Local Void. The Pegasus Cloud penetrates the Local Void between the Lacerta − . − . − .
8. (video frames 05:23 to 05:42)
Table 2 . Sources of Local Group MotionNumerical Action AnalysisZone SGX SGY SGZ Sumkm s − km s − km s − km s − Near ( <
38 Mpc) −
122 316 −
190 388Mid (38 −
100 Mpc) −
212 95 −
106 255Far ( >
100 Mpc) − − −
28 100Cumulative −
410 353 −
324 631Wiener Filter AnalysisLocal Void − ±
59 43 ± − ±
36 197 ± ±
50 281 ± − ±
22 282 ± − ±
58 329 ± − ±
37 391 ± − ±
49 355 ± − ±
42 632 ± − − ∼
250 km s − .Alternatively, the impact of respectively the LocalVoid and the greater Virgo Cluster can be evaluatedfrom the Wiener filter model. In the case of the Lo-cal Void, the influence at our position is found summingover the volume defined as the Local Void below the con-tour of δ = 0. In the case of the Virgo Cluster, the sum-mation is over a sphere centered on the cluster extendingin radius to our position. Statistical uncertainties aredetermined by averaging over multiple constrained real-izations (Hoffman & Ribak 1991; Zaroubi et al. 1999).The numerical results are gathered in Table 2 along withthose from the numerical action analysis.The most directly comparable results between the twoanalyses (besides the cumulative values) are the numeri-cal action “near” row and the Wiener filter “LV+Virgo”row. The results are in statistical agreement. The com-parison is approximate since in detail the contributionsare not the same. Give consideration to the distinctSGX, SGY, SGZ components. There is agreement that,in sum, the Virgo overdensity and Local Void are com-bining to give us a deviant motion of about 300 km s − toward positive SGY with Virgo dominant at the levelof 65 − FINAL THOUGHTSThe model of nearby structure in the universe thathas been presented is derived strictly from the radialvelocities of test particles assuming deviations from uni-form expansion arise from Newtonian gravity in a ba-sic ΛCDM cosmology. There is a manifestly reasonableagreement with the alternative of structure inferred fromredshift surveys. Much remains to give the compari-son a quantitative foundation. We have introduced the morphology of structure determined from the 18,000 dis-tances of
Cosmicflows-3 with a study of voids becausesuch regions are simpler than high density regions. Thereconstruction by Graziani et al. (2019) has relativelycoarse resolution of 6 . /h Mpc. Higher resolutioncan be achieved at the expense of computation time.Studies of high density regions will benefit from highspatial resolution where the quasi-linear regime can beprobed with the techniques discussed by Hoffman et al.(2018) and numerical action methods can probe the fullynon-linear regime (Shaya et al. 2017).Even if voids are simpler, with dynamics in the linearregime, their morphologies have been much less well un-derstood than that of high density structures. Yet welive at the edge of a void. Nearby our measurementsof distances of galaxies are plentiful and associated pe-culiar velocities are well determined. This information,processed through the Wiener filter, allows us to definethe properties of the Local Void with considerable preci-sion even where much information is lost due to Galacticobscuration.The Local Void does not have a simple shape. Now,for the first time, we have a map that reveals its com-plexity. It is unambiguously demonstrated that the Lo-cal Void, familiarly prominent at positive SGZ, and thevoid in front of the Perseus-Pisces filament (Haynes &Giovanelli 1986) at positive SGX are parts of the sameextensive under density. This linkage has not been evi-dent because of intervening obscuration.Attention has been drawn to the substantial contri-bution of the Local Void to our motion reflected in thecosmic microwave background dipole anisotropy. TheLocal Group has a deviant motion due to the Local Voidof 200 −
250 km s − with respect to the center of masswithin 38 Mpc (and another 200 −
250 km s − due tothe Virgo overdensity). The full effect of the Local Voidas it extends to the zone beyond the 38 Mpc near re-gion has a repelling influence that explains most of theSGZ component of our motion in the cosmic microwavebackground frame. What is left over after accounting forthese nearby actors is ∼
300 km s − , directed mostly to-ward negative SGX, and attributable to structure in themid and far shells discussed in the previous section.We have revisited the known fact (Lindner et al. 1995)that chains of galaxies can thread through voids. Usu-ally the constituents are small galaxies. The case wepresent of the Perseus Cloud that traverses the LocalVoid from the Virgo Cluster to the Perseus Cluster isparticularly noteworthy, not only because we are partof it. Remarkably, it passes through one of the lowestdensity parts of the Local Void and near that location isa significant gathering of substantial galaxies. Regret-tably, these systems lie at a low Galactic latitude andhave been poorly studied.7Immediately beyond the Local Void lie two much big-ger under dense regions that we call the Hercules Voidand the Sculptor Void. In fact the voids are all inter-connected by necks that are below the mean density.Boundaries can be arbitrary and will undoubtedly besources of contention. We use our discoverer’s preroga-tive and give names to outstanding features.Our cartography reveals hints of the complexity ofoverdenities in the region we are exploring within 0 . c .Even our discussion of voids represents only a first pass.There is much more to be learned as the density of databecomes richer and we gain confidence in reconstructionswith increasing resolution.Support for this program is provided by Grant No.80NSSC18K0424 from the US National Aeronautics andSpace Administration and from multiple awards fromthe Space Telescope Science Institute, most recentlyHST-AR-14319, HST-GO-14636, and HST-GO-15150.HC acknowledges support by the Institut Universitairede France and the CNES.8 REFERENCES.Even our discussion of voids represents only a first pass.There is much more to be learned as the density of databecomes richer and we gain confidence in reconstructionswith increasing resolution.Support for this program is provided by Grant No.80NSSC18K0424 from the US National Aeronautics andSpace Administration and from multiple awards fromthe Space Telescope Science Institute, most recentlyHST-AR-14319, HST-GO-14636, and HST-GO-15150.HC acknowledges support by the Institut Universitairede France and the CNES.8 REFERENCES