WIN2017; Astroparticle physics and cosmology An Improved Measurement of the Cosmic Microwave Background B-mode Polarization Power Spectrum at Sub-Degree Scales with the POLARBEAR Experiment 23 Jun 2017 from 9:50AM to 10:10AM 1 POLARBEAR is a “Stage 2” CMB experiment CMB-S4 Science Book (arXiv:1610.02743) 2 Chile, Atacama POLARBEAR to Simons Array Atacama Cosmology Telescope CLASS (ACTPol to AdvACT) South Pole BICEP-2/Keck Array/ BICEP-3 South Pole Telescope (SPTpol to SPT3G) 3 102 DASI QUaD CBI QUIET-Q MAXIPOL QUIET-W BOOMERanG BICEP1-3yr 101 CAPMAP BICEP2-3yr ) WMAP-9yr POLARBEAR 2 K µ 100 )( π (2 / 10-1 BB ` C BICEP2 POLARBEAR 10-2 + 1) r=0.20 ⇥ ( ⇥ 10-3 March, 2014 10-4 10 100 1000 Multipole Moment, ell 4 102 DASI QUIET-Q CBI QUIET-W MAXIPOL BICEP1-3yr BOOMERanG ACTPol 1 10 CAPMAP BK14 ACTPol ) WMAP-9yr SPTpol 2 QUaD POLARBEAR K µ 100 )( π (2 / -1 10 SPTpol BB ` C POLARBEAR BK14 10-2 + 1) ⇥ ( ⇥ r=0.07 10-3 2016-2017 10-4 10 100 1000 Multipole Moment, ell 5 102 DASI QUIET-W CBI BICEP1-3yr MAXIPOL ACTPol BOOMERanG BK14 101 CAPMAP SPTpol ) WMAP-9yr POLARBEAR 2 QUaD Simons Array QUIET-Q K µ 100 )( ⇡ (2 / 10-1 BB ` Simons C Array 10-2 + 1) ` ( ` 10-3 ~2020 r=0.01 10-4 10 100 1000 Multipole Moment, ell 6 7 POLARBEAR Collabora�on UC Berkeley UC San Diego KEK McGill University SISSA Shawn Beckman Kam Arnold Yoshiki Akiba Matt Dobbs Carlo Baccigalupi Darcy Barron Kevin Crowley Takaho Hamada Adam Gilbert Nicoletta Yuji Chinone Tucker Elleflot Masaya Hasegawa Josh Montgomery Krachmalnicoff Ari Cukierman George Fuller Masashi Hazumi Davide Poletti Giuseppe Puglisi Tijmen de Haan Logan Howe Haruki Nishino Dalhousie Neil Goeckner-Wald Brian Keating Yuuko Segawa Scott Chapman U Manchester John Groh David Leon Osamu Tajima Colin Ross Gabriele Coppi Charles Hill Lindsay Lowry Satoru Takakura Kaja Rotermund Andrew May William Holzapfel Frederick Matsuda Sayuri Takatori Alexei Tikhomirov Oliver Jeong Martin Navaroli Daiki Tanabe Lucio Piccirillo Adrian Lee Gabriel Rebeiz Takayuki Tomaru Lawrence U of Sussex Dick Plambeck Max Silva-Feaver Berkeley NL Julien Peloton Chris Raum U. Melbourne Praween Siritanasak Julian Borrill Paul Richards Christian Reichardt Grant Teply Reijo Keskitalo UC Irvine Aritoki Suzuki Calvin Tsai Federico Bianchini Theodore Kisner Chang Feng Anh Pham Ben Westbrook Alex Zahn Akito Kusaka Nathan Whitehorn Eric Linder Cardiff University Laboratoire Imperial College Alex Madurowicz Peter Ade Astroparticule & Andrew Jaffe Blake Sherwin CU Boulder Cosmologie Daisy Mak Raymond Tat NASA Goddard Nils Halverson Dominic Beck Nathan Miller Greg Jaehnig Josquin Errard Institute Kavli IPMU Argonne NL Hayley Roberts Maude Le Jeune D’Astrophysique Yuto Minami Radek Stompor Spatiale Amy Bender Nobuhiko Katayama Católica (PUC) Giulio Fabbian David Boettger Rolando Dunner And many more in years past… POLARBEAR Collaboration 9 Observing CMB polarization at the James Ax Observatory in Atacama desert in Northern Chile on Cerro Toco at an altitude of 5,200m since January 2012 Atacama Chile Cerro Toco 5,200m 10 POLARBEAR since 2012 (deployed in 2011) 5,200m @ Atacama Desert, Chile 2yrs for small patches targeting lensing >3yrs for large patch targeting inflation 11 Small patch Large patch for lensing B-mode w/ HWP Time nd rd (hours) 1st season 2 3 season 10000 season 8000 6000 4000 2000 Note: Included time for calibration 0 2012/07 2013/07 2014/07 2015/06 1st BB result st Previous results just come from 1 season data set nd 60% more data by adding 2 season Observing a larger patch for inflationary science 12 rd 23 Dec, 2013: Detection of lensing by POLARBEAR x CIB rd 23 Dec, 2013: Detection of lensing by POLARBEAR w/ 4pt th 10 Mar, 2014: Measurement of lensing B modes by POLARBEAR w/ 2pt th 9 Sep, 2015: Constraint on Cosmic birefringence & Primordial magnetic field by POLARBEAR 4.0σ 4.2σ 97.2% →4.7σ(comb.) 13 CMB 3.5 m Guard ring Primary mirror CMB “pixel” (637) Receiver 23 uK √s (not seen) 1274 TES (Transition-edge sensor) bolometers Secondary mirror Superconducting detector cooled (not seen) down to 250mK 14 Intensity Dec=90 Total area: ~25 deg2 (FDS Dust Map) RA12 RA=0 RA=-180 RA23 RA4.5 select low dust region 15 Calibration/Data Selection Filtering/Map-making 5000 ) 2 K 4000 µ ) ( 3000 /(2 l 2000 Power Spectrum Estimation l(l+1)C 1000 0 500 1000 1500 2000 2500 Multipole Moment, ell Cosmological parameters 16 Calibration/Data Selection Filtering/Map-making Confirm analysis before unveiling the power spectra Data/Analysis Validation ➡ “Null test” & Systematic Error Estimation Power Spectrum Estimation Cosmological parameters 17 Null test is a powerful tool to detect hidden bias Divide data into 2 subsets, make maps, difference them map0 map1 null map - = (CMB+Noise1) - (CMB+Noise2) = Noise Calculate “null” power spectrum If no systematic, “null” spectrum is really null, but… Calculating for several interesting splits of the data 18 1st_season_vs_2nd_season 1st_half_vs_2nd_half high_gain_ces_vs_low_gain_ces 9 different high_elevation_vs_low_elevation combinations rising_vs_setting of CES’s high_pwv_vs_low_pwv far_from_sun_vs_close_to_sun far_from_moon_vs_close_to_moon sun_above_horizon_vs_sun_below_horizon 2 different left_going_scan_vs_right_going_scan splits of FP q_pixels_vs_u_pixels & L/R scans left_side_pixels_vs_right_side_pixels 19 BB “null” spectrum size of cosmological signal 20 Null test is a powerful tool to detect hidden bias Divide data into 2 subsets, make maps, difference them map0 map1 null map - = (CMB+Noise1) - (CMB+Noise2) = Noise POLARBEAR successfully passed 12 null tests No evidence for systematic contamination & miscalibration in the POLARBEAR data set & analysis 21 Estimate all possible systematic bias by instrument & foreground which could affect B-mode measurement B-mode we want to measure Systematic biases from 9 sources e.g. pointing error, beam error, angle error, analysis technique, and foreground Confirmed all systematics are much smaller than B-mode 22 Dust & synchrotron are estimated by Planck 353 GHz & 30 GHz & WMAP K-band polarization maps extrapolation of angular scales from ell=80 to us extrapolation of frequency to us contamination consistent w/ zero, but dominated by noise in the Planck polarization data at POLARBEAR scales Dusty & radio galaxies Set of simulated galaxies with distribution, intensity and polarization fraction modeled after observation (De Zotti et al, 2005; George et al, 2015; Bonavera et al, 2017) 23 Two independent pipelines performed TOD into maps & power spectrum estimate Pipeline A Pipeline B 24 6000 TT 0.6 Pipeline A Pipeline A 4000 Pipeline B Pipeline B 2000 ) 0.4 0 2 40 EE K µ 30 )( 20 ⇡ 0.2 10 (2 / ) 2 K 0 BB µ 100 TE ` C )( 50 0 ⇡ (2 0 / ` + 1) C -50 ` -100 ( ` + 1) -0.2 ` ( 4 TB ` 2 0 -2 -0.4 -4 0 500 1000 1500 2000 2500 0.6 EB Multipole Moment, 0.4 ` 0.2 We reject the null hypothesis ofno B-mode polarization 0 at a confidence of 3.1σ -0.2 -0.4 including both statistical and systematic uncertainties. -0.6 +0.26 +0.00 0 500 1000 1500 2000 2500 AL =0.60 0.24(stat) 0.04(inst) 0.14(fg) 0.04(multi) Multipole Moment, ` − − ± ± 25 102 DASI QUIET-W CBI BICEP1-3yr MAXIPOL ACTPol BOOMERanG BK14 101 CAPMAP SPTpol ) WMAP-9yr POLARBEAR 2 QUaD Simons Array QUIET-Q K µ 100 )( ⇡ (2 / 10-1 BB ` C 10-2 + 1) ` ( ` 10-3 r=0.01 10-4 10 100 1000 Multipole Moment, ell POLARBEAR successfully measured B modes @ high ell (small angular scales) How about low ell (large angular scales) for inflationary B modes? Actually satellite or only small telescope have measured large angular scales Must demonstrate a large telescope can also measure large angular scales ⇨ demonstration of “1/f noise” mitigation by POLARBEAR! 26 improvement from a few 100 mHz to ~32 mHz With Continuously rotation half-wave plate (HWP), -6 10 successfully removed atmospheric correlated signal No Correlation w/ atmosphere Validated modulation by HWP would be able to mitigate 1/f noise & the residual would be level of r<0.01 knee=32 mHz (l~39) arXiv:1702.07111 S. Takakura et. al. (PB Collab.) 27 What’s Next? Simons Array! A “Stage 3” CMB experiment upgraded from POLARBEAR Simons Array POLARBEAR-1 or… x10 sensitivity CMB-S4 Science Book (arXiv:1610.02743) 28 190mm POLARBEAR Simons Array x6 more detectors (3 telescopes) x (6 x more detectors) 365mm =18 x sensitive than now dual freq. 7,588 bolometers per receiver, factor of 6 increase from current receiver Three telescopes (two new telescopes + one current one) → (3 telescopes) x (6 x more detectors) = 18 x sensitive than now 29 POLARBEAR Simons Array @KEK POLARBEAR-2A 95/150 GHz (3 telescopes) x (6 x more detectors) =18 x sensitive than now POLARBEAR-2B 95/150 GHz @UCB & UCSD POLARBEAR-2C 220/270 GHz POLARBEAR-2 Receiver Frequency Plan 7,588 bolometers per receiver, factor of 6 increase from current receiver Three telescopes (two new telescopes + one current one) → (3 telescopes) x (6 x more detectors) = 18 x sensitive than now Expand frequency coverage for foreground removal (95/150/220/270 GHz) Deploy first receiver (PB2A) in 2017 (this summer!!!) Deploy two more receivers (PB2B, PB2C) in 2018 30 POLARBEAR 102 DASI QUIET-W CBI BICEP1-3yr Simons Array MAXIPOL ACTPol BOOMERanG BK14 101 CAPMAP SPTpol ) WMAP-9yr POLARBEAR 2 QUaD Simons Array QUIET-Q K (3 telescopes) µ 100 x (6 x more detectors) )( ⇡ =18 x sensitive than now (2 / 10-1 BB ` C 10-2 + 1) ` ~2020 ( ` 10-3 r=0.01 10-4 10 100 1000 Multipole Moment, ell -3 -3 Inflation: σ(r=0.1) = 6×10 (4×10 stat.) Neutrino mass: σ(Σmν) = 40 meV (19 meV stat.) w/ BAO from DESI Light relic: σ(Neff) = 0.04 31 POLARBEAR is a “stage 2” CMB experiment, which successfully measured B-mode lensing; Updated the result w/ 2-years data validated 1/f noise mitigation for inflationary B-mode measurement Simons Array is a “stage 3” CMB experiment upgraded from POLARBEAR experiment; Deploy in 2017 & 2018.