Substructure from Simulations Can the standard reionization scenario - PowerPoint PPT Presentation

Substructure from Simulations Can the standard reionization scenario explain the current population of satellite galaxies? (How low does galaxy formation go?) James S. Bullock (UC Irvine)

Tyler Kelley ( UC Irvine ) ( UC Irvine —> Texas )

Star-less halos: How low do we go? 8 l . 0 e t a e l i n g S p r Aquarius m dm ~2.e3 M sun m p =2.e4 M sun m p =2.e5 M sun m p =2.e6 M sun

Star-less halos: How low do we go? 8 l . 0 e t a e l i n g S p r H2 cooling limit H cooling limit Aquarius V~16 km/s V~2 km/s m dm ~2.e3 M sun m p =2.e4 M sun m p =2.e5 M sun m p =2.e6 M sun

How Many Do We See? 4 0 1 l + 2 m e K i m o n - r i s G a r H cooling limit V~16 km/s V~30 km/s ELVIS All MW satellites m p =2.e4 M sun Classical MW satellites m p =2.e5 M sun m p =2.e6 M sun

How Many Do We See? 4 0 1 l + 2 m e K i m o n - r i s G a r Dark substructure? V<8 km/s H cooling limit V~15 km/s V~30 km/s ELVIS All MW satellites m p =2.e4 M sun Classical MW satellites m p =2.e5 M sun m p =2.e6 M sun

1995 “…a photo ionizing background suppresses the formation of galaxies with circular velocities v c <30 km/s …” (T vir ~ 30,000 K)

Naively, this is the suppression scale you would “…a photo ionizing expect from from the ionized IGM temperature…. background suppresses the T IGM ~ 10,000 - 30,000 K formation of galaxies with circular velocities v c <30 km/s …” (T vir ~ 30,000 K) Onorbe et al. 2015

Missing Satellites: 1999 Klypin+ 1999 UMaI Sextans Draco Milky Way Sag LMC Carina SMC Sculptor Fornax Bullock/Geha 100,000 light years Moore et al. 1999; Klypin et al.1999

30 km/s Mismatch sets in V~30 km/s Klypin+ 1999

Bullock+2000 Reionization solves the problem. 30 km/s

V max (km/s) Sawala+2014 4 8 17 30 Galaxies get dark at V max ~20-30 km/s because of reionization. sim particle mass: m baryon ~ 10,000 M sun (Every halo is dark below 8 km/s.) Similar results: Gnedin 2000; Hoeft et al. 2006; Okamoto et al. 2008; Ocvirk et al. 2016 [CoDa simulations]

FIRE simulations of Milky Way Hopkins+2018 Garrison-Kimmel+2018 Wetzel+2017 m baryon ~ 5000 M sun ‘Green flash’ is reionization

Dark Matter 600 kpc Garrison-Kimmel et al. 2018

Stars 600 kpc Garrison-Kimmel et al. 2018

Garrison-Kimmel + 2018 10 zoom simulations of Milky Ways ELVIS on FIRE “…do not exhibit the missing satellites problem…” - subhalos get ‘dark’ at V max <20 km/s

Fitts et al. 2017 12 zoom simulations of field dwarfs m baryon = 500 M sun Reionization suppression: V max ~ 20-25 km/s T vir ~ 20,000 K FIRE simulations

If we take this as the V max (km/s) canonical expectation, 4 8 17 30 what should we see when counting very low- mass dwarfs? Similar results: Gnedin 2000; Sawala+2014 Hoeft et al. 2006; Okamoto et al. 2008; Ocvirk et al. 2016 [CoDa simulations]; Fitts et al. 2017; etc.

~40 ultra-faint satellite galaxies discovered since ‘missing satellites’ “Ultra-faint dwarfs” Milky Way Five-fold increase in last M * ~10 3 -10 5 M sun in 14 yrs Roughy half the sky unexplored to SDSS this depth. Certainly not SDSS DES complete beyond 50 kpc. Willman et al. 2005; Zucker et al. 2006; Belokurov et al. 2007; Koposov et al. 2015a; Bechtol et al. 2015; Kim et al. 2015

Stars 600 kpc Garrison-Kimmel et al. 2018

Dark Matter 600 kpc Garrison-Kimmel et al. 2018

Dark Matter 600 kpc Garrison-Kimmel et al. 2018

FIRE Hydrodynamics (dark matter) 100 kpc Garrison-Kimmel+2017

Baryons Matter (A Lot!) FIRE Hydrodynamics Pure N-Body (dark matter) (same halo) 100 kpc 100 kpc Also: Brooks & Zolotov 2014, Zhu + 2016, Garrison-Kimmel+2017

Baryons Matter (A Lot!) FIRE Hydrodynamics Pure N-Body (dark matter) (same halo) NO substructure (V > 5 km/s) within 20 kpc 100 kpc 100 kpc Also: Brooks & Zolotov 2014, Zhu + 2016, Garrison-Kimmel+2017

Most important Factor is Central Galaxy Potential FIRE Hydrodynamics Pure N-body N-body + Gal. Potential 100 kpc Garrison-Kimmel+2017

Substructure within radii of relevance for known GC stream heating is destroyed… Garrison-Kimmel+2017

Phat ELVIS: M v = 0.8-2 x 10 12 M sun 12 high-resolution zoom simulations with Milky Way potentials evolved to z=0 - 12 identical simulations run with Dark Matter only -

Factor of ~10 reduction in substructure within ~25 kpc N(<R) DMO Galaxy R (kpc) Kelley et al. 2018 Also: Garrison-Kimmel et al. 2018

Preferentially destroy halos with percenters <~ 20 kpc DMO N(<R) DMO Galaxy Galaxy R (kpc) Kelley et al. 2018 Also: Garrison-Kimmel et al. 2018

Radial distribution of Satellites V max > 4.5 km/s N(<R) DMO Galaxy Kelley et al. 2018

Radial distribution of Satellites V peak > 10 km/s N(<R) DMO Galaxy Kelley et al. 2018

Radial distribution of Satellites V peak > 10 km/s 14 ultra-faint dwarfs V peak > 10 km/s T vir ~ 3,800 K w/in 50 kpc known Milky Way satellites N(<R) DMO DMO Galaxy Galaxy

Why are there so many ultra-faint dwarf satellites? Median of Galaxy Required to match : Potential Runs V peak = 8 km/s LOW! T vir ~ 2,400 K NOT atomic cooling halos! sky N(<R) correction MW dwarf satellites Nominal re-ionization scale V max = 20 km/s

Radial Profile: full scatter of 12 Galaxy simulations Graus et al. 2018 V max > 4.5 km/s subhalos (M~5.e6M sun ) known MW 100% scatter dwarfs (12 halos) (no sky coverage correction

Apply fiducial “toy” model of how reionization makes halos dark V max > 4.5 km/s Graus et al. 2018 subhalos T vir ~ 20,000 K 100% scatter (12 halos) 20 km/s

“Observed” to account for coverage incompleteness V max > 4.5 km/s Graus et al. 2018 subhalos 100% scatter T vir ~ 20,000 K (12 halos) 20 km/s Average (<1)

“Observed” to account for coverage incompleteness V max > 4.5 km/s subhalos 100% scatter T vir ~ 5,500 K (12 halos) Average 12 km/s Graus et al. 2018

Threshold mass for galaxy formation is apparently very low… V max > 4.5 km/s > In order to explain known MW subhalos galaxies, we must populate majority subhalos with V peak ~12km/s (T vir ~ 100% scatter 5,500 K). (12 halos) > Median halo in the suite needs even Average lower threshold (V ~ 8 km/s) > Do ultra-faint dwarfs really live in such low-mass halos? Graus et al. 2018

CONCLUSIONS 1. Central galaxy potential destroys most DMO galaxy substructure in the inner ~25 kpc of cosmological ‘zoom’ MW halos. - Must be accounted for when making predictions for stream heating & other substructure probes. 2. In order to account for the ~14 Milky Way dwarf galaxies within 50 kpc, we must populate most subhalos with galaxies down to V peak ~10km/s - This is well below the canonical V peak ~20 km/s scale where reionization was thought to start making galaxies go dark! Kelley et al. 2018 Also below the atomic cooling scale (which is V~16 km/s) - Graus et al. 2018

Garrison-Kimmel+2014

Wheeler+15 FIRE-1 Simulations Classical Ancient stars (ultra-faint) 17 km/s 30 km/s

Kelley et al. 2018