the dynamic Milky Way in the Gaia era

Equilibrium phase-space models have enabled much of what we know about the structure of the Milky Way, including global or bulk properties of Galactic dark matter. Kinematic data from the recent second data release from the Gaia mission has shown that these assumptions of time-independence and equilibrium are invalid given the precision of the Gaia data: time dependent phenomena are important from the solar neighborhood, to the outer Galactic disk, to the orbits of stellar streams in the inner halo. I will discuss recent work that highlights what we can learn when we relax strong assumptions about equilibrium in dynamical inferences about the Milky Way.

Transcript

1. Adrian Price-Whelan Lyman J. Spitzer, Jr. Fellow Princeton University the dynamic Milky Way in the Gaia era
2. why do we study the Milky Way in detail? full position, velocity (at present day) detailed chemical abundances stellar ages {x0 , v0 , [Fe/H], [X/Fe], τ} for ~109 stars! By ~2022, we’ll know
3. star formation the ISM stellar structure exoplanets Milky Way assembly … dark matter why do we study the Milky Way in detail?
4. the Milky Way as a dark matter lab Bullock & Boylan-Kolchin 2017 dwarf galaxies dark substructure? the hope: measure the global and small-scale structure of dark matter in the Milky Way CDM CDM FDM WDM z = 0 power spectrum halo mass function
5. {x0 , v0 , [Fe/H], [X/Fe], τ}N → ρ(x, t) dark matter! the Milky Way as a dark matter lab
6. {x0 , v0 }N → ρ(x, t) “Since the age of the Galaxy is about 10 Gyr, most disk stars have completed over forty revolutions, and it is reasonable to assume that the Galaxy is now in an approximately steady state.” — Binney & Tremaine (2008)
7. “Since the age of the Galaxy is about 10 Gyr, most disk stars have completed over forty revolutions, and it is reasonable to assume that the Galaxy is now in an approximately steady state.” — Binney & Tremaine (2008) {x0 , v0 }N → ρ(x)
8. {x0 , v0 }N → a(x) ∇ ⋅ a = − 4πG ρ
9. how do we measure the acceleration field from a kinematic snapshot? if I gave you the positions and velocities of all Solar System planets right now, could you tell me the force law? only with strong assumptions!
10. Bovy, Murray, Hogg (2010) Solar system is long-lived not observed at a special time non-resonant phase-mixed (angles uniformly distributed) planets are bound spherical symmetry how do we measure the acceleration field from a kinematic snapshot?
11. how do we measure the acceleration field from a kinematic snapshot of the Galaxy? long-lived ? not observed at a special time ? non-resonant ? phase-mixed ? symmetries ?
12. how do we measure the acceleration field from a kinematic snapshot of the Galaxy? Distribution function (DF) moment methods (e.g., Jeans modeling) Assume: equilibrium, time-independent, non-resonant/ chaotic, phase-mixed, geometric symmetries Orbit ensemble methods (e.g., Schwarzschild modeling, forward-modeling the DF) Assume: steady-state, integrable, tracers fair sample see also: all of Binney & Tremaine (2008)
13. but duh: we know that those assumptions are wrong in detail! many early Gaia results have shown that these assumptions are bad precision — bias
14. the Gaia Mission Gaia’s first look at the Milky Way stellar streams and their utility perturbations to the GD-1 stream from dark substructure Overview
15. the Gaia mission * *PS, it’s not an acronym anymore *PPS, I have no affiliation with the Gaia consortium
16. Astrometry — μ, Low-res spectrophotometry — BP, RP 3-band photometry — G, GBP, GRP Radial velocity Gaia
17. Gaia focal plane source detection astrometric field (+G mag.) BP RP Radial velocity spectrometer (RVS) transit direction (10 sec.) cosmos.esa.int/gaia
19. Gaia scan pattern
20. Astrometry — μ, Low-res spectrophotometry — BP, RP 3-band photometry — G, GBP, GRP Radial velocity Gaia DR1 : September 2016 DR2 : April 2018 DR3 : first half 2021 DR4 : end of 2022 Anthony Brown (Gaia): April/May 2019 Spitzer lectures
21. No proprietary period! Astrometry (proper motion + parallax) ~1.3 billion, G < 21 at epoch J2015.5 Low-res spectrophotometry BP, RP 3-band photometry - G, GBP, GRP ~1.4 billion mean flux/mag Radial velocity ~7.2 million, G < 12.5 median RV Gaia DR 2 (April 25, 2018) cosmos.esa.int/gaia
23. Gaia astrometry examples Internal structure & kinematics of nearby objects (e.g., the Pleiades)
24. Gaia astrometry examples Proper motions of distant objects, parallax cleans foreground (e.g., 47 Tuc)
25. a Gaia color-magnitude diagram Babusiaux et al. 2018
26. the Milky Way in the Gaia era
27. solar neighborhood, outer disk, inner halo I is not mixed the [ ]
28. Data from HIPPARCOS Hunt et al. 2018 Dehnen 1998 Simulation: MW disk + Bar the solar neighborhood with HIPPARCOS “Hercules stream”
29. Price-Whelan et al. (in prep.) the solar neighborhood with Gaia
30. see also: Trick et al. 2018 Price-Whelan et al. (in prep.) the solar neighborhood with Gaia
31. Katz et al. 2018 radial disk kinematics with Gaia
32. vertical disk kinematics Bland-Hawthorn et al. 2018 simulated smooth disk Antoja et al. 2018 Gaia data
33. Laporte et al. 2017, 2018 simulated disk + Sagittarius dSph Sagittarius pericenter ~ 8–12 kpc! z-vz spiral results from recent (t = -500 Myr) direct impact between Sag and the disk disk velocity structure driven by dark matter wake
34. the Milky Way stellar halo is dominated by an ancient radial merger of a sizable dwarf galaxy II
35. Belokurov et al. 2018 vr v most metal poor more metal rich nearer disk far from disk Gaia “sausage” …
36. Helmi et al. 2018 Gaia–Enceladus
37. the radially-anisotropic stellar halo Kinematically distinct, but dominates r < 30 kpc halo Associated with shells (“Hercules-Aquila,” “Virgo cloud”), not fully phase-mixed Belokurov et al. 2018, Lancaster et al. 2018, Helmi et al. 2018, Kruijssen et al. 2018, Simion et al. 2018 …
38. the Milky Way is being strongly perturbed by Sagittarius, the LMC, and past mergers broken: equilibrium, phase-mixed, time-independent
39. Solar system is long-lived not observed at a special time non-resonant phase-mixed (angles uniformly distributed) planets are bound spherical symmetry back to the Solar System example, Hooke / Newton didn’t have to assume
40. stellar streams
41. orbital plane rotating frame trailing tail leading tail
42. Grillmair 2016
43. Stream stars nearly delineate the orbit of the progenitor ➔ Measure of local acceleration field In principle, requires fewer assumptions Thin streams are also sensitive to perturbations, e.g., dark matter substructure! Fundamental assumption: streams are not phase-mixed see, e.g., Bonaca & Hogg 2018
44. The longest thin stream (now known to be >100º) Metal-poor, old stellar population (fully-disrupted globular cluster?) Relatively nearby @ 8–10 kpc Prominent over background (denser than Pal 5) Grillmair & Dionatos 2006 the GD-1 stream Hints of density variations
45. Under-densities? “Spur” “Blob” the GD-1 stream in Gaia Price-Whelan & Bonaca 2018 these features are not expected from simple models of stream formation
46. Progenitor fully disrupts >400 Myr ago Price-Whelan & Bonaca 2018 model GD-1 stream
47. x y stream simulated in a live DM halo Yoon et al. 2011
48. x y y z Δv ∝ G Msubhalo b vsubhalo Bonaca, Price-Whelan, Hogg (in prep.) stream–subhalo interactions
49. “Spur” the GD-1 stream in Gaia Bonaca, Price-Whelan, Hogg (in prep.)
50. Bonaca, Price-Whelan, Hogg (in prep.)
51. example model of the GD-1 stream with a dark matter subhalo impact Bonaca, Price-Whelan, Hogg (in prep.)
52. Stellar streams provide a powerful way to constrain the Galactic acceleration field, requires fewer assumptions than DF methods But: streams are very sensitive to time-dependent phenomena; we better model that as well (that’s hard!) The payoff will be large: constraints on small-scale dark matter (see: GD-1, PW+Bonaca work) & assembly history of the Milky Way precision inferences with stellar streams
53. No components of the Milky Way are steady-state, time-independent systems The era of strong equilibrium assumptions is in decline It remains to be seen what we can even learn from collections of un-phase-mixed, time-dependent, chaotic stellar orbits Challenge for the next 5–10 years: We need new tools to perform these inferences! (but don’t recycle your copy of Binney & Tremaine just yet…) final thoughts
54. Extra slides
55. Laporte et al. (2017) Johnston, Price-Whelan et al. (2017) simulated disk + Sagittarius dSph + LMC
57. “Spur” “Blob” Spur Blob Control
58. Density model: Uniform background Gaussian for stream Gaussians on either side of stream to capture features Spur Blob Control