Testing modified Newtonian dynamics in the Milky Way Fabio Iocco,1,2 Miguel Pato,3,4 and Gianfranco Bertone5 1ICTP South American Institute for Fundamental Research, and Instituto de Física Teórica - Universidade Estadual Paulista (UNESP), Rua Dr. Bento Teobaldo Ferraz 271, 01140-070 São Paulo, São Paulo, Brazil 2Instituto de Física Teórica UAM/CSIC, C/ Nicolás Cabrera 13-15, 28049 Cantoblanco, Madrid, Spain 3The Oskar Klein Centre for Cosmoparticle Physics, Department of Physics, Stockholm University, AlbaNova, SE-106 91 Stockholm, Sweden 4Physik-Department T30d, Technische Universität München, James-Franck-Straße, D-85748 Garching, Germany 5GRAPPA Institute, University of Amsterdam, Science Park 904, 1090 GL Amsterdam, Netherlands (Received 28 May 2015; published 21 October 2015) Modified Newtonian dynamics (MOND) is an empirical theory originally proposed to explain the rotation curves of spiral galaxies by modifying the gravitational acceleration, rather than by invoking dark matter. Here, we set constraints on MOND using an up-to-date compilation of kinematic tracers of the Milky Way and a comprehensive collection of morphologies of the baryonic component in the Galaxy. In particular, we find that the so-called “standard” interpolating function cannot explain at the same time the rotation curve of the Milky Way and that of external galaxies for any of the baryonic models studied, while the so-called “simple” interpolating function can for a subset of models. Upcoming astronomical observations will refine our knowledge on the morphology of baryons and will ultimately confirm or rule out the validity of MOND in the Milky Way. We also present constraints on MOND-like theories without making any assumptions on the interpolating function. DOI: 10.1103/PhysRevD.92.084046 PACS numbers: 04.50.Kd, 98.35.-a If Newtonian gravity holds, the rotation curve of the Milky Way cannot be explained by visible (baryonic) matter only, therefore providing evidence for dark matter in the Galaxy, from its outskirts (e.g. [1–3]) down to inside the solar circle [4]. In this paper we derive constraints on modifications of Newtonian gravity based on the same data set as used in Ref. [4]. We focus our attention on modified Newtonian dynamics (MOND), originally proposed by Milgrom [5–7] (see also [8–12]). MOND postulates that below a characteristic value a0 the acceleration a is modified with respect to that predicted by Newton’s law aN by the introduction of an interpolating function μ such that [5,6] μ � a a0 � a ¼ aN; ð1Þ where μðxÞ≃ x for x ≪ 1 and μðxÞ≃ 1 for x ≫ 1 (for modern theories of MOND, see [11]). The theory does not predict the value of a0 nor a specific functional form for μ from first principles, so different proposals have been made in the literature (see e.g. [10,11]). Two widely discussed functional forms for μ are the “standard” interpolating function, μstdðxÞ ¼ xffiffiffiffiffiffiffiffiffiffiffiffiffi 1þ x2 p ; ð2Þ and the “simple” interpolating function, μsimðxÞ ¼ x 1þ x : ð3Þ The limit of large accelerations, i.e. a ≫ a0 or x ≫ 1, is strongly constrained by solar system tests, where MOND is bound to recover Newtonian gravity. The limit of small accelerations, i.e. a ≪ a0 or x ≪ 1, is usually invoked in MOND studies to highlight how the observed flat rotation curves of spiral galaxies can be reproduced in the absence of dark matter. In this paper, we set out to test the most commonMOND scenarios with the latest data on the rotation curve of the Galaxy. We use a comprehensive compilation of kinematic tracers of the Milky Way and a state-of-the-art modeling of the baryons, both presented in Ref. [4]. Our results and analysis technique are complementary to previous MOND analyses of Milky Way data [13–16] (see also [17–19] for constraints from the vertical force in the discs of galaxies, including our own). The quantitative study of alternative formulations of MOND or other theories of modified gravity is left for future work. There are two observational inputs required to test MOND in our Galaxy: the observed gravitational acceleration a and that predicted by Newtonian gravity aN . These are the two key ingredients of our analysis. The acceleration a is obtained from the rotation curve through a ¼ Rω2 c, where R is the Galactocentric radius and the angular circular velocity ωc is set by the compi- lation of rotation curve measurements of Ref. [4]. This is a PHYSICAL REVIEW D 92, 084046 (2015) 1550-7998=2015=92(8)=084046(6) 084046-1 © 2015 American Physical Society http://dx.doi.org/10.1103/PhysRevD.92.084046 http://dx.doi.org/10.1103/PhysRevD.92.084046 http://dx.doi.org/10.1103/PhysRevD.92.084046 http://dx.doi.org/10.1103/PhysRevD.92.084046 comprehensive compilation of kinematic tracers optimized to R ¼ 3–20 kpc and that includes an unprecedented set of published data on gas kinematics, star kinematics and masers. A detailed description of the data used and their treatment can be found in the supplementary information of Ref. [4] (cf. Table S1 therein). The Newtonian acceleration aN ¼ Rω2 b is set by the three-dimensional density distri- bution of baryons in our Galaxy for which we adopt the survey of observation-based models presented in Ref. [4]. This survey includes seven alternative morphologies for the stellar bulge [20–25], five for the stellar disc [19,26–29] and two for the gas [30,31], while the normalization is set by microlensing observations for the bulge [32,33], the local total stellar surface density for the disc [19] and the CO-to-H2 factor for the gas [34,35]. For further details, please refer to the supplementary information of Ref. [4]. We use the morphologies of bulge, disc and gas in all 70 possible combinations, thus including all configurations of the baryonic component of the Galaxy present in the literature. In this way, we bracket the current uncertainty due to baryonic modeling and our conclusions do not rely on a specific model of the visible Galaxy but are solid against baryonic systematics. We adopt a distance to the Galactic center R0 ¼ 8 kpc, a local circular velocity v0 ¼ 230 km=s and the peculiar solar motion ðU;V;WÞ⊙ ¼ ð11.10; 12.24; 7.25Þ km=s [36]. The impact of varying these Galactic parameters is quantified later on, showing that our conclusions are robust against current uncertain- ties. Only kinematic tracers with Galactocentric radii above Rcut ¼ 2.5 kpc are considered, amounting to N ¼ 2686, 2687, 2715 individual measurements for R0 ¼ 7.98, 8, 8.68 kpc, respectively (cf. below for the adopted uncer- tainty on R0). The effect of pushing the radius cut to Rcut ¼ 4.5 kpc is also quantified in order to avoid any influence of the bar (see e.g. Ref. [37]); in this case, N ¼ 2159, 2162, 2267 for R0 ¼ 7.98, 8, 8.68 kpc, respectively. Figure 1 shows ωc, i.e. the observed rotation curve, and ωb, i.e. the rotation curve expected from baryons under Newtonian gravity. We are now in place to compare these two observation-based quantities and set constraints on MOND by assessing the discrepancy from the Newtonian scenario. The rotation curve predicted by a given MOND theory with fixed μðxÞ and a0 is obtained by solving Eq. (1) for a ¼ Rω2 mond. We then quantify the goodness-of-fit of the obtained ωmondðRÞ curve using a two-dimensional chi- square against the rotation curve data ωcðRÞ. The use of this test statistic is mainly motivated by the sizeable uncertainties on both ωc and R (see Ref. [4] for a technical discussion). The reduced chi-square as a function of the assumed a0 is shown in the upper panels of Fig. 2 for each baryonic model for μ ¼ μstd (left) and μ ¼ μsim (right). The results R [kpc] 5 10 15 20 25 [k m /s /k pc ] ω 1 10 210 = 2 .5 k pc cu t R = 4 .5 k pc cu t R = 8 k pc 0 R rotation curve data baryonic bracketing best fit std μ best fit sim μ FIG. 1 (color online). The rotation curve of our Galaxy. The red data points indicate the angular circular velocity of all tracers in the rotation curve compilation, while the grey band brackets the contribution of all baryonic models under the assumption of Newtonian gravity. For further details, see Ref. [4]. The MOND best-fit rotation curves found later on in our analysis for the fiducial baryonic model I [19,20,30] (cf. Fig. 2) are also shown for the standard (simple) interpolating functions with a0 ¼ 3 × 10−10 ð1.5 × 10−10Þm=s2. We stress that these values of a0 are incompatible (marginally compatible) with the range of values suggested by external spiral galaxies for the standard (simple) interpolating functions (cf. Fig. 2) and we present these best fits here for illustration purposes only. We have adopted here R0 ¼ 8 kpc, v0 ¼ 230 km=s and ðU;V;WÞ⊙ ¼ ð11.10; 12.24; 7.25Þ km=s [36]. FABIO IOCCO, MIGUEL PATO, AND GIANFRANCO BERTONE PHYSICAL REVIEW D 92, 084046 (2015) 084046-2 for what we shall call fiducial baryonic model I [19,20,30] are highlighted in black. Also shown are the values of a0 favored by the observation of rotation curves in external galaxies, namely a0 ¼ ð1.27� 0.30Þ × 10−10 m=s2 for μstd and a0 ¼ ð1.22� 0.33Þ × 10−10 m=s2 for μsim [38], as well as the 5σ exclusion line to guide the eye. In the case of the standard interpolating function, an acceptable best fit is obtained for most baryonic configurations. However, even in those cases, the preferred values of a0 lie in the range ð2.5–5.0Þ × 10−10 m=s2, clearly above the typical values found in external galaxies, which disfavors the standard interpolating function as being able to accommodate the rotation curve of external galaxies and that of the Milky Way at the same time. Instead, in the case of the simple interpolating function, the best-fit values of a0 drop signifi- cantly to ð1.0–3.5Þ × 10−10 m=s2, in line with the range inferred from external galaxies for a subset of baryonic models. Nevertheless, most baryonic models still prefer somewhat large values of a0. These results are roughly in line with previous fits of MOND to Milky Way data [13,14], but here we have expanded upon such analyses by using a comprehensive compilation of the available kinematic data and a wide range of baryonic models. So far we have kept fixed R0 ¼ 8 kpc, v0 ¼ 230 km=s and V⊙ ¼ 12.24 km=s. These values lie well within the ranges encompassing most current measurements, namely R0 ¼ 8.0� 0.5 kpc [39–42], v0¼230�20 km=s [42–47] and V⊙ ¼ 5.25–26 km=s [36,42,47,48]. It is important to understand how the uncertainties on R0, v0 and V⊙ affect the results. For concreteness, let us focus on two specific measurements: (i) R0 ¼ 8.33� 0.35 kpc [39], provided by the monitoring of the orbits of S-stars around the central supermassive black hole, and (ii) Ω⊙ ≡ v0þV⊙ R0 ¼ 30.26� 0.12 km=s=kpc [43], provided by the proper motion of Sgr A�. Combining (i) and (ii) with the allowed values of V⊙ reported above, we have the following 1σ configurations: (a) R0¼7.98 kpc, v0 ¼ 214.52 km=s for V⊙ ¼ 26 km=s; (b) R0¼7.98 kpc, v0¼237.18 km=s for V⊙¼5.25 km=s; (c) R0¼8.68 kpc, v0 ¼ 235.62 km=s for V⊙ ¼ 26 km=s; ]2 [m/s0a 0 0.2 0.4 0.6 0.8 1 -910× /N2 χ -110 1 10 210 σ5 std μ=μ ex te rn al g al ax ie s ]2 [m/s0a 0 0.2 0.4 0.6 0.8 1 -910× /N2 χ -110 1 10 210 σ5 sim μ=μ ex te rn al g al ax ie s ]2 [m/s0a 0 0.2 0.4 0.6 0.8 1 -910× /N2 χ -110 1 10 210 std μ=μ ex te rn al g al ax ie s =12.24 km/ssun=230.00 km/s, V 0 =8.00 kpc, v0R =26.00 km/ssun=214.52 km/s, V 0 =7.98 kpc, v0R =05.25 km/ssun=237.18 km/s, V 0 =7.98 kpc, v0R =26.00 km/ssun=235.62 km/s, V 0 =8.68 kpc, v0R =05.25 km/ssun=258.45 km/s, V 0 =8.68 kpc, v0R =4.5 kpccutR ]2 [m/s0a 0 0.2 0.4 0.6 0.8 1 -910× /N2 χ -110 1 10 210 sim μ=μ ex te rn al g al ax ie s =12.24 km/ssun=230.00 km/s, V 0 =8.00 kpc, v0R =26.00 km/ssun=214.52 km/s, V 0 =7.98 kpc, v0R =05.25 km/ssun=237.18 km/s, V 0 =7.98 kpc, v0R =26.00 km/ssun=235.62 km/s, V 0 =8.68 kpc, v0R =05.25 km/ssun=258.45 km/s, V 0 =8.68 kpc, v0R =4.5 kpccutR FIG. 2 (color online). Fitting MOND to the rotation curve of our Galaxy. The left (right) panels show the reduced chi-square of the MOND scenario with the standard (simple) interpolating function for different values of a0. The upper panels convey the results for all baryonic models adopting R0 ¼ 8 kpc, v0 ¼ 230 km=s and ðU; V;WÞ⊙ ¼ ð11.10; 12.24; 7.25Þ km=s [36]. The black line in each upper panel marks the reduced chi-square obtained for fiducial baryonic model I [19,20,30]. The bottom panels correspond to fiducial baryonic model I [19,20,30] with different combinations of Galactic parameters and radius cut. The range of a0 found for each interpolating function from the rotation curves of external galaxies [38] is encompassed by the vertical dashed lines, while the thick red line indicates the reduced chi-square corresponding to a 5σ exclusion. TESTING MODIFIED NEWTONIAN DYNAMICS IN THE … PHYSICAL REVIEW D 92, 084046 (2015) 084046-3 (d) R0¼8.68 kpc, v0¼258.45 km=s for V⊙¼5.25 km=s. We use configurations (a) through (d) to quantify the impact of current uncertainties on our results. This is shown explicitly in the bottom panels of Fig. 2 for fiducial baryonic model I [19,20,30]. It is possible to conclude that, within the same baryonic model, the best-fit values of a0 are relatively insensitive to the actual local circular velocity v0 provided the Sgr A� constraint is met [cf. (a) and (b)]. Moreover, a value of R0 on the high end of the currently allowed range shifts the favoured a0 to higher acceler- ations, i.e. away from the range inferred from external galaxies [cf. (c) and (d)]. These considerations hold for both μstd and μsim. Therefore, the conclusions drawn in the previous paragraphs do not change qualitatively when varying the Galatic fundamental parameters. The same applies when pushing the radius cut to 4.5 kpc, as also shown in the bottom panels of Fig. 2. In Fig. 3 we perform a data-driven test for MOND-like theories that can be applied directly to rotation curve and photometric data. By mapping the modified acceler- ation with the rotation curve data through a ¼ Rω2 c and the Newtonian acceleration with the baryonic distribution through aN ¼ Rω2 b, we reconstruct μ directly from Eq. (1) with no dependence on a0, namely μða=a0Þ ¼ aN=a. Then, the resulting measurements of μ are binned in linear intervals of size Δa ¼ 10−10 m=s2 requiring at least five rotation curve measurements per bin. This is shown in Fig. 3 by the red bars for fiducial baryonic model I [19,20,30], by the blue bars for an additional baryonic model II [20,26,30] and by the grey band for the bracketing of all baryonic models implemented in our analysis. In the same plot we superimpose μstdða=a0Þ and μsimða=a0Þ for a0 ¼ 10−10 m=s2. As clear from the figure, the standard interpolating function is in tension with the data for both fiducial baryonic models, whereas the simple interpolating function remains viable for fiducial baryonic model I. Note that for a flat rotation curve vc ¼ v0 the acceleration reads a ¼ v20=R, so the kinematic tracers at small R correspond to large a and vice-versa. Therefore, the importance of baryonic modeling grows for large a since that corresponds to the region of the Galaxy where the baryons become more and more important. Although at present uncertainties are still sizeable, Fig. 3 provides a powerful nonparametric test of the MOND paradigm. Our analysis is complementary to that of Ref. [15], which covers a slightly different accel- eration range, and that of Ref. [16], which is more focussed on the outer Galaxy. From Figs. 2 and 3, it appears that the simple interpolat- ing function provides a better fit to Milky Way data, in agreement with earlier studies which adopt a different setup and approach, e.g. [13,14]. Let us recall that this functional form is strongly disfavored by solar system tests and it has to be modified in the large acceleration limit (see e.g. [11,38]). Modifications of the simple interpolating function ]2a [m/s 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 -910× /a N )= a 0 (a /a μ 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 = 2 .5 k pc cu t R = 4 .5 k pc cu t R = 8 k pc 0 R fiducial model I fiducial model II baryonic bracketing )2 m/s-10=10 0 (a std μ )2 m/s-10=10 0 (a sim μ FIG. 3 (color online). The MOND interpolating function as inferred directly from the rotation curve of our Galaxy. The red and blue bars represent the binned 1σ measurement of μða=a0Þ ¼ aN=a for fiducial baryonic models I [19,20,30] and II [20,26,30], respectively, while the grey band encompasses the 1σ measurements of all baryonic models. The slight shift in a of the red and blue bars is for visualization purposes only. Overplotted are the standard and simple interpolating functions for a0 ¼ 10−10 m=s2. For reference, we also indicate the values of a corresponding to R0, Rcut ¼ 2.5 kpc and Rcut ¼ 4.5 kpc in the case of a flat rotation curve vc ¼ v0. We have adopted here R0 ¼ 8 kpc, v0 ¼ 230 km=s and ðU;V;WÞ⊙ ¼ ð11.10; 12.24; 7.25Þ km=s [36]. FABIO IOCCO, MIGUEL PATO, AND GIANFRANCO BERTONE PHYSICAL REVIEW D 92, 084046 (2015) 084046-4 have been proposed in the literature (i.e., the family of ν-functions [11] or the “improved simple” function [38]) by making μ converge to unity for a=a0 ≳ 10. We note that, for the range of accelerations to which we are sensitive (cf. Fig. 3), these modified functions and the simple function are virtually indistinguishable for typical values of a0 and the results presented here for μsim would also apply to those modified functions. As our measurements of the Galactic rotation curve and our understanding of the distribution of baryons improve, we expect these tests to provide more and more stringent constraints on both μ and a0. For the time being, however, we can conservatively state—based on an up-to-date compilation of rotation curve data and a comprehensive collection of data-inferred morphologies of the baryonic component—that MOND variants employing the standard interpolating function do not fit simultaneously the rotation curves of external galaxies and that of the Milky Way. Let us finally point out that we have adopted in our analysis a wide compilation of baryonic models to bracket the current uncertainty on the morphology and composition of our Galaxy. In the coming years, astronomical surveys such as Gaia [49], APOGEE-2 (SDSS-IV) [50], WFIRST [51], WEAVE [52] and 4MOST [53] will eventually help narrow down on the baryonic distribution, thus providing a decisive step toward testing the MOND paradigm in the Milky Way. ACKNOWLEDGMENTS The authors thank Benoît Famaey for useful comments on the manuscript. F. I. acknowledges support from the Simons Foundation and FAPESP process 2014/22985-1, M. P. from Wenner-Gren Stiftelserna in Stockholm, and G. B. from the European Research Council through the ERC Starting Grant WIMPs Kairos. [1] T. Sakamoto, M. Chiba, and T. C. Beers, Astron. Astrophys. 397, 899 (2003). [2] X. X. Xue, H. W. Rix, G. Zhao, P. Re Fiorentin, T. Naab, M. Steinmetz, F. C. van den Bosch, T. C. Beers, Y. S. Lee, E. F. Bell et al., Astrophys. J. 684, 1143 (2008). [3] P. R. Kafle, S. Sharma, G. F. Lewis, and J. Bland-Hawthorn, Astrophys. J. 794, 59 (2014). [4] F. Iocco, M. Pato, and G. Bertone, Nat. Phys. 11, 245 (2015). [5] M. Milgrom, Astrophys. J. 270, 365 (1983). [6] M. Milgrom, Astrophys. J. 270, 371 (1983). [7] M. Milgrom, Astrophys. J. 270, 384 (1983). [8] R. H. Sanders and S. S. McGaugh, Annu. Rev. Astron. Astrophys. 40, 263 (2002). [9] J. D. Bekenstein, Phys. Rev. D 70, 083509 (2004). [10] J. Bekenstein, Contemp. Phys. 47, 387 (2006). [11] B. Famaey and S. S. McGaugh, Living Rev. Relativity 15, 10 (2012). [12] M. Milgrom, Mon. Not. R. Astron. Soc. 437, 2531 (2014). [13] B. Famaey and J. Binney, Mon. Not. R. Astron. Soc. 363, 603 (2005). [14] S. S. McGaugh, Astrophys. J. 683, 137 (2008). [15] S. R. Loebman, Ž. Ivezić, T. R. Quinn, J. Bovy, C. R. Christensen, M. Jurić, R. Roškar, A. M. Brooks, and F. Governato, Astrophys. J. 794, 151 (2014). [16] J. W. Moffat and V. T. Toth, Phys. Rev. D 91, 043004 (2015). [17] C. Nipoti, P. Londrillo, H. Zhao, and L. Ciotti, Mon. Not. R. Astron. Soc. 379, 597 (2007). [18] O. Bienaymé, B. Famaey, X. Wu, H. S. Zhao, and D. Aubert, Astron. Astrophys. 500, 801 (2009). [19] J. Bovy and H.-W. Rix, Astrophys. J. 779, 115 (2013). [20] K. Z. Stanek, A. Udalski, M. Szymanski, J. Kaluzny, M. Kubiak, M. Mateo, and W. Krzeminski, Astrophys. J. 477, 163 (1997). [21] H. Zhao, Mon. Not. R. Astron. Soc. 283, 149 (1996). [22] N. Bissantz and O. Gerhard, Mon. Not. R. Astron. Soc. 330, 591 (2002). [23] E. Vanhollebeke, M. A. T. Groenewegen, and L. Girardi, Astron. Astrophys. 498, 95 (2009). [24] M. López-Corredoira, A. Cabrera-Lavers, T. J. Mahoney, P. L. Hammersley, F. Garzón, and C. González-Fernández, Astron. J. 133, 154 (2007). [25] A. C. Robin, D. J. Marshall, M. Schultheis, and C. Reylé, Astron. Astrophys. 538, A106 (2012). [26] C. Han and A. Gould, Astrophys. J. 592, 172 (2003). [27] S. Calchi Novati and L. Mancini, Mon. Not. R. Astron. Soc. 416, 1292 (2011). [28] J. T. A. de Jong, B. Yanny, H.-W. Rix, A. E. Dolphin, N. F. Martin, and T. C. Beers, Astrophys. J. 714, 663 (2010). [29] M. Jurić, Ž. Ivezić, A. Brooks, R. H. Lupton, D. Schlegel, D. Finkbeiner, N. Padmanabhan, N. Bond, B. Sesar, C. M. Rockosi et al., Astrophys. J. 673, 864 (2008). [30] K. Ferriere, Astrophys. J. 497, 759 (1998). [31] I. V. Moskalenko, A. W. Strong, J. F. Ormes, and M. S. Potgieter, Astrophys. J. 565, 280 (2002). [32] P. Popowski, K. Griest, C. L. Thomas, K. H. Cook, D. P. Bennett, A. C. Becker, D. R. Alves, D. Minniti, A. J. Drake, C. Alcock et al., Astrophys. J. 631, 879 (2005). [33] F. Iocco, M. Pato, G. Bertone, and P. Jetzer, J. Cosmol. Astropart. Phys. 11 (2011) 029. [34] K. Ferrière, W. Gillard, and P. Jean, Astron. Astrophys. 467, 611 (2007). [35] M. Ackermann, M. Ajello, W. B. Atwood, L. Baldini, J. Ballet, G. Barbiellini, D. Bastieri, K. Bechtol, R. Bellazzini, B. Berenji et al., Astrophys. J. 750, 3 (2012). TESTING MODIFIED NEWTONIAN DYNAMICS IN THE … PHYSICAL REVIEW D 92, 084046 (2015) 084046-5 http://dx.doi.org/10.1051/0004-6361:20021499 http://dx.doi.org/10.1051/0004-6361:20021499 http://dx.doi.org/10.1086/589500 http://dx.doi.org/10.1088/0004-637X/794/1/59 http://dx.doi.org/10.1038/nphys3237 http://dx.doi.org/10.1038/nphys3237 http://dx.doi.org/10.1086/161130 http://dx.doi.org/10.1086/161131 http://dx.doi.org/10.1086/161132 http://dx.doi.org/10.1146/annurev.astro.40.060401.093923 http://dx.doi.org/10.1146/annurev.astro.40.060401.093923 http://dx.doi.org/10.1103/PhysRevD.70.083509 http://dx.doi.org/10.1080/00107510701244055 http://dx.doi.org/10.12942/lrr-2012-10 http://dx.doi.org/10.12942/lrr-2012-10 http://dx.doi.org/10.1093/mnras/stt2066 http://dx.doi.org/10.1093/mnras/stt2066 http://dx.doi.org/10.1111/j.1365-2966.2005.09474.x http://dx.doi.org/10.1111/j.1365-2966.2005.09474.x http://dx.doi.org/10.1086/589148 http://dx.doi.org/10.1088/0004-637X/794/2/151 http://dx.doi.org/10.1103/PhysRevD.91.043004 http://dx.doi.org/10.1103/PhysRevD.91.043004 http://dx.doi.org/10.1111/j.1365-2966.2007.11835.x http://dx.doi.org/10.1111/j.1365-2966.2007.11835.x http://dx.doi.org/10.1051/0004-6361/200809978 http://dx.doi.org/10.1088/0004-637X/779/2/115 http://dx.doi.org/10.1088/0004-637X/779/2/115 http://dx.doi.org/10.1086/303702 http://dx.doi.org/10.1086/303702 http://dx.doi.org/10.1093/mnras/283.1.149 http://dx.doi.org/10.1046/j.1365-8711.2002.05116.x http://dx.doi.org/10.1046/j.1365-8711.2002.05116.x http://dx.doi.org/10.1051/0004-6361/20078472 http://dx.doi.org/10.1086/509605 http://dx.doi.org/10.1051/0004-6361/201116512 http://dx.doi.org/10.1086/375706 http://dx.doi.org/10.1111/j.1365-2966.2011.19123.x http://dx.doi.org/10.1111/j.1365-2966.2011.19123.x http://dx.doi.org/10.1088/0004-637X/714/1/663 http://dx.doi.org/10.1086/523619 http://dx.doi.org/10.1086/305469 http://dx.doi.org/10.1086/324402 http://dx.doi.org/10.1086/432246 http://dx.doi.org/10.1088/1475-7516/2011/11/029 http://dx.doi.org/10.1088/1475-7516/2011/11/029 http://dx.doi.org/10.1051/0004-6361:20066992 http://dx.doi.org/10.1051/0004-6361:20066992 http://dx.doi.org/10.1088/0004-637X/750/1/3 [36] R. Schönrich, J. Binney, and W. Dehnen, Mon. Not. R. Astron. Soc. 403, 1829 (2010). [37] L. Chemin, F. Renaud, and C. Soubiran, Astron. Astrophys. 578, A14 (2015). [38] G. Gentile, B. Famaey, and W. J. G. de Blok, Astron. Astrophys. 527, A76 (2011). [39] S.Gillessen, F. Eisenhauer, S.Trippe,T.Alexander, R.Genzel, F. Martins, and T. Ott, Astrophys. J. 692, 1075 (2009). [40] K. Ando, T. Nagayama, T. Omodaka, T. Handa, H. Imai, A. Nakagawa, H. Nakanishi, M. Honma, H. Kobayashi, and T. Miyaji, Publ. Astron. Soc. Jpn. 63, 45 (2011). [41] Z. Malkin, arXiv:1202.6128. [42] M. J. Reid, K. M. Menten, A. Brunthaler, X. W. Zheng, T. M. Dame, Y. Xu, Y. Wu, B. Zhang, A. Sanna, M. Sato et al., Astrophys. J. 783, 130 (2014). [43] M. J. Reid and A. Brunthaler, Astrophys. J. 616, 872 (2004). [44] M. J. Reid, K. M. Menten, X. W. Zheng, A. Brunthaler, L. Moscadelli, Y. Xu, B. Zhang, M. Sato, M. Honma, T. Hirota et al., Astrophys. J. 700, 137 (2009). [45] J. Bovy, D. W. Hogg, and H.-W. Rix, Astrophys. J. 704, 1704 (2009). [46] P. J. McMillan and J. J. Binney, Mon. Not. R. Astron. Soc. 402, 934 (2010). [47] J. Bovy, C. Allende Prieto, T. C. Beers, D. Bizyaev, L. N. da Costa, K. Cunha, G. L. Ebelke, D. J. Eisenstein, P. M. Frinchaboy, A. E. García Pérez et al., Astrophys. J. 759, 131 (2012). [48] W. Dehnen and J. J. Binney, Mon. Not. R. Astron. Soc. 298, 387 (1998). [49] J. H. J. de Bruijne, Astrophys. Space Sci. 341, 31 (2012). [50] http://www.sdss.org/surveys/apogee‑2/. [51] D. Spergel, N. Gehrels, C. Baltay, D. Bennett, J. Breckinridge, M. Donahue, A. Dressler, B. S. Gaudi, T. Greene, O. Guyon et al., arXiv:1503.03757. [52] http://www.ing.iac.es/weave/index.html. [53] R. S. de Jong, O. Bellido-Tirado, C. Chiappini, É. Depagne, R. Haynes, D. Johl, O. Schnurr, A. Schwope, J. Walcher, F. Dionies et al., Proc. SPIE Int. Soc. Opt. Eng. 8446, 84460T (2012). FABIO IOCCO, MIGUEL PATO, AND GIANFRANCO BERTONE PHYSICAL REVIEW D 92, 084046 (2015) 084046-6 http://dx.doi.org/10.1111/j.1365-2966.2010.16253.x http://dx.doi.org/10.1111/j.1365-2966.2010.16253.x http://dx.doi.org/10.1051/0004-6361/201526040 http://dx.doi.org/10.1051/0004-6361/201526040 http://dx.doi.org/10.1051/0004-6361/201015283 http://dx.doi.org/10.1051/0004-6361/201015283 http://dx.doi.org/10.1088/0004-637X/692/2/1075 http://dx.doi.org/10.1093/pasj/63.1.45 http://arXiv.org/abs/1202.6128 http://dx.doi.org/10.1088/0004-637X/783/2/130 http://dx.doi.org/10.1086/424960 http://dx.doi.org/10.1086/424960 http://dx.doi.org/10.1088/0004-637X/700/1/137 http://dx.doi.org/10.1088/0004-637X/704/2/1704 http://dx.doi.org/10.1088/0004-637X/704/2/1704 http://dx.doi.org/10.1111/j.1365-2966.2009.15932.x http://dx.doi.org/10.1111/j.1365-2966.2009.15932.x http://dx.doi.org/10.1088/0004-637X/759/2/131 http://dx.doi.org/10.1088/0004-637X/759/2/131 http://dx.doi.org/10.1046/j.1365-8711.1998.01600.x http://dx.doi.org/10.1046/j.1365-8711.1998.01600.x http://dx.doi.org/10.1007/s10509-012-1019-4 http://www.sdss.org/surveys/apogee-2/ http://www.sdss.org/surveys/apogee-2/ http://www.sdss.org/surveys/apogee-2/ http://arXiv.org/abs/1503.03757 http://www.ing.iac.es/weave/index.html http://www.ing.iac.es/weave/index.html http://www.ing.iac.es/weave/index.html http://www.ing.iac.es/weave/index.html http://www.ing.iac.es/weave/index.html http://dx.doi.org/10.1117/12.926239 http://dx.doi.org/10.1117/12.926239