Search of primordial black holes from microlensing observa5ons with - PowerPoint PPT Presentation

Search of primordial black holes from microlensing observa5ons with HSC and OGLE Hiroko Niikura (UTokyo / Kavli IPMU) Collaborators: Masahiro Takada, Naoki Yasuda (Kavli IPMU), Robert Lupton (Princeton), Takahiro Sumi (Osaka), Surhud More (IUCAA), Sunao Sugiyama, Toshiki Kurita (Kavli IPMU), Anupreeta More (IUCAA), Masamune Oguri (UTokyo), Masashi Chiba (Tohoku), Shuichiro Yokoyama (Nagoya), Shogo Masaki (Suzuka) @ Moriond EW, 2019 � 1

� � Target: Primordial black hole as dark matter horizon scale • Searching for dark matter in the local universe � – Massive Compact Object (MACHO), Weakly Interacting Massive Particle (WIMP), Primordial Black Hole (PBH) � primordial � fluctuation • Primordial black hole (PBH) Primordial black hole (PBH) � – Proposed to be generated in the early universe, and can survive as dark matter today if not evaporated (Hawking 1971) � – Previous research still leaves some room for PBH to be a part of dark matter � black hole 2

Constraint on PBH abundance (previous studies) A mass fraction of PBHs to DM � ◉ Open widow on PBH abundance PBHs for LIGO around M PBH =[10 -15 ,10 -9 ] M ◉ � (note: lunar mass ~10 -8 M ◉ ) � ◉ 3

Constraining PBH with microlensing PBH � Subaru HSC M31 PBH microlensing search Use OGLE (Optical Gravitational Lensing Experiment) for PBH search credit: Masahiro Takada 4

arXiv: 1701.02151 --> Nature Astronomy in press (Niikura et al. 2019) Tightest bounds on PBH abundance with HSC observation of M31 Hiroko Niikura, Masahiro Takada, Naoki Yasuda, Sunao Sugiyama, Toshiki Kurita (Kavli IPMU), Robert Lupton (Princeton), Takahiro Sumi (Osaka), Surhud More, Anupreeta More (IUCAA), Masamune Oguri (UTokyo), Masashi Chiba (Tohoku) 5

credit: Masahiro Takada Hyper Suprime-Cam • Maunakea, Hawaii (4200m) � • largest camera � • 3m high � • weigh 3 ton � • 104 CCDs (~0.9B pixels) 6

Andromeda Galaxy (M31) • Large spiral galaxy � • In the northern hemisphere (not accessible from VLT, DES, LSST) � • HSC FoV ~ entire M31 � • ~770kpc ( 𝜈 ~24.4) � • HSC can monitor all stars in the bulge and disk regions of M31 � HSC Image of M31 (HSC FoV=1.8 sq. degrees) 7

PBH microlensing event rate ✓ M PBH ◆ 1 / 2 ✓ ◆ � 1 ◆ ✓ t E ∼ d L θ E d L v PBH ∼ 34 min v PBH 10 � 8 M � 200 km / s 100 kpc 10 � 4 M PBH = 10 � 12 M � Event rate for a single star in M31 10 � 11 M � d 2 N event / d t obs dln t FWHM [events/hrs] 10 � 10 M � 10 � 5 10 � 9 M � 10 � 8 M � 10 � 7 M � Ω PBH 10 � 6 = 1 Ω DM 10 � 7 10 � 8 10 � 2 10 � 1 10 0 10 1 6min 60min t FWHM [hrs] (time scale of ML light curve) Event rate per unit obs. time and per a single star in M31 for a given timescale of light curve � 8

~ ~ Result : Distribution of transient candidates 2014 � HSC-M31 � • Detection: Image difference � � focal plane � Reference image Target image � More than � 15,000 transient � candidates in � one field-of-view � of HSC. � Difference image � (6 min.-) � transient candidates � fake (incl. RR-Lyrae) Cepheid variable asteroid ~15,000 stellar flare eclipsing binary contact binary 21 21 . 0 10 5mes flare 21 . 5 22 2 hours Magnifica5on ! 22 . 0 23 22 . 5 mag 24 mag 10 min. 23 . 0 23 . 5 25 24 . 0 26 24 . 5 27 25 . 0 WD+BD eclipse 0 5000 10000 15000 20000 25000 0 5000 10000 15000 20000 25000 time [sec] time [sec] 9

Analysis : Selec5on of microlensing candidates ( 6 min. -4 hours ) Follow selec5on method by Griest et al., 2014 (Kepler) Total number of events : 15,571 15000 10000 5000 # of count Noise threshold (S/N > 5 for 3 consecu5ve visits) 0 candidates − 5000 (t max out of obs. period) − 10000 0 5000 10000 15000 20000 25000 Apart from CCD edge time [sec] fake events, binary stars 11,703 6000 4000 Fitting of ML lightcurve model 2000 102 for nochiin count 0 (for lightcurves in difference images) − 2000 bad χ 2 − 4000 dof − 6000 0 5000 10000 15000 20000 25000 339 for long including 15 short 227 time [sec] Symmetric shape of peak around the 2500 2000 1500 peak in the light curve flare stars, fake events 1000 count 500 0 − 500 asymmetric peak 89 ccd edge 146 − 1000 significant peaks − 1500 noisy events − 2000 0 5000 10000 15000 20000 25000 time [sec] 6000 3000 92 bad chi2 66 2000 spikes, asteroid or some defects 4000 Visual inspec5on 1000 1 2000 count 0 count 118 noise − 1000 0 bad fiZng − 2000 − 2000 − 3000 − 4000 − 4000 9 seeing 0 0 5000 5000 10000 10000 15000 15000 20000 20000 25000 25000 time [sec] time [sec] 14 flare 134 spikes 15 bimyo 10

－ 1500 counts 1000 500 0 500 － 1000 － time [sec] 25000 20000 15000 10000 5000 0 One remaining candidate.. Total number of events : 15,571 blue: ± 1 σ noise on difference images 移動天体や CCD の欠陥 134 spikes One remaining candidate 11

PBH constraint with HSC M31 ML Search M PBH [ M � ] Nature Astronomy in press � 10 � 15 10 � 10 10 � 5 10 0 (Niikura et al. 2019) Kepler Kepler Femto EROS/MACHO CMB Femto EROS/MACHO CMB 10 � 1 1 f= Ω PBH / Ω DM A mass fraction of PBHs to DM � 10 � 2 n 2 BH Evaporation o i t a r o p a 10 � 3 v • Wave optics 3 E • Finite size of H B source stars 10 � 4 HSC M31 constraint (95% limit) 4 tightest bound on PBH with M PBH =[10 20 ,10 28 ]g � 1 HSC night ⇔ 2 years Kepler data � HSC 95% CL (2days): +finite source size effect +wave effect 10 � 5 5 10 15 10 20 10 25 10 30 10 35 M PBH [g] 12

arXiv: 1901.07120 --> PRD in press (Niikura et al. 2019) Earth mass black hole? PBH constraint with OGLE Hiroko Niikura, Masahiro Takada (IPMU), Shuichiro Yokoyama (Nagoya), Takahiro Sumi (Osaka), Shogo Masaki (Suzuka) Image credit: AAS 13

OGLE: Optical Gravitational Lensing Experiment Credit: OGLE project 5 × 10 7 stars • 1.3m telescope in Las Campanas, Chile � • Long-term microlensing observation (1996-), with cadence either 20 or 60 min � 14

5-years OGLE data (Mroz et al. 2017) OGLE 5yrs PBHs? (halo) 10 2 Number of events Free-floating Astrophysical 2622 events / 5 years planets? 10 1 origin (disk+bulge) 10 0 10 � 1 10 0 10 1 10 2 10 3 t E [day] • 2622 microlensing events: the ML timescale distribution is provided (now >5000 events) 15

Null test: PBH upper bounds by OGLE M PBH [ M � ] 10 � 10 10 � 6 10 � 2 10 2 10 0 Kepler Caustic CMB EROS/MACHO f PBH = Ω PBH / Ω DM 10 � 1 HSC 10 � 2 OGLE excl. region (95% CL) 10 � 3 Fully take into account the likelihood (Poisson uncertainty) 10 � 4 10 20 10 25 10 30 10 35 M PBH [g]

Detection of Earth-mass PBHs? OGLE data Galactic bulge/disk models (2622 events) Number of events per bin PBH MS M PBH = 9 . 5 × 10 � 6 M � 10 2 f PBH = 0 . 026 BD Free-floating � 10 1 (wide-orbit) planets? WD 10 0 NS 10 � 1 10 0 10 1 10 2 ML LC timescale: t E [days] Earth mass (~10 -6 Msun) PBHs? � 17

Results : Allowed parameter space indicated by detection of Earth-mass PBHs M PBH [ M � ] 10 � 8 10 � 7 10 � 6 10 � 5 10 � 4 10 0 OGLE alone f PBH = Ω PBH / Ω DM 10 � 1 OGLE+HSC HSC excl. region 10 � 2 allowed region (95% CL) 10 � 3 10 26 10 27 10 28 10 29 M PBH [g] 18

Summary & Future prospect (Bai & Orlofsky 2018 [1812.01427]) M PBH [ M � ] 10 � 15 10 � 10 10 � 5 10 0 Kepler Femto EROS/MACHO CMB 10 � 1 Athena 100d HSC OGLE f= Ω PBH / Ω DM 10 � 2 n LOFT 300d? o (Sasaki et al. 2016) i t a (X-ray pulsar r o p a @ SMC) v E H 10 � 3 PBHs for LIGO B 10 � 4 HSC 95% CL (2days) OGLE 95% CL (5years) 10 � 5 10 15 10 20 10 25 10 30 10 35 M PBH [g] Thank you very much for listening! 19

� � backup Observation: monitor M31 with HSC • The wide and deep imaging ~2 min with Hyper Suprime-Cam (HSC); � « Can cover the entire disk M 31 of Hyper Suprime Cam (r-band) and bulge regions of M31 with its one pointing � « 90sec exposure can reach to ~26mag depth for a star � • Observation for 7-hours, t taking images every 2 ~10 11 stars minutes at M31-disk region (r- 1.5 deg ~25mag band), performed in ~7 hours 2014/2017 � ~200 frames 20

backup Detection of transients: difference imaging • Pixel lensing regime: multiple stars in each CCD pixel Reference image (seeing ~ 0.5” ) Target image (seeing ~ 0.7” ) Difference image (seeing ~ 0.7” ) ref. Target diff. diff.-PSF diff. (bad) • tiny objects (< pixel size) � • distorted object � • high residual after PSF subtraction 21

25000 3000 15000 20000 0 time [sec] － 4000 － － 5000 2000 － 1000 0 1000 2000 3000 4000 10000 0 counts 4000 5000 10000 15000 20000 25000 time [sec] － － counts 3000 － 2000 － 1000 0 1000 2000 5000 backup Visual inspection … Visual inspection of 66 candidates to identify junks … asteroid � spike around a bright star � 66 66 ⇒ 65 junks 65 junks � 22

backup Wave effect + finite source size effect • Wave effect: wavelength vs. Schwarzschild radius of PBH Matsunaga & Yamamoto +06 • Magnification magnifica5on suppressed due to 2 y � 0.5 � geo � interference and � � diffraction 1 y � 0.5 � wave � • PBHs with 0.5 M PBH < 2.6 × 10 -12 M sun cannot be detected (y=0.5: source position fixed) 0.001 0.01 0.1 1 10 w • Finite source size effect: R E,PBH < R source (point source limit not valid)