1. The APEX Sunyaev Zel'dovich Experiment Observations Matt Dobbs McGill University On behalf of the APEX-SZ collaboration

2. U.C. Berkeley / LBNL Brad Benson Hsiao-Mei Cho John Clarke Daniel Ferrusca Bill Holzapfel Brad Johnson Zigmund Kermish Adrian Lee Martin Lueker Jared Mehl Tom Plagge Christian Reichardt Paul Richards Dan Schwan Helmuth Spieler Ben Westbrook Martin White Oliver Zahn Max Planck IfR Gayong Chon Rolf Guesten Ruediger Kneissl Ernst Kreysa Karl Menten Dirk Muders Martin Nord Peter Schilke Bonn University Kaustuv Basu Frank Bertoldi Florian Pacaud Reinhold Schaaf Cardiff Peter Ade Carole Tucker The APEX-SZ Collaboration C.U. Boulder Amy Bender Nils Halverson McGill University Matt Dobbs James Kennedy Trevor Lanting Onsala Space Observatory Cathy Horellou Daniel Johansson

3. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 3 The Sunyaev-Zel’dovich Effect (slide adapted from NASA publicity figure) CMB photons are used to backlight structure in the universe.

4. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 4 1-2% of CMB photons traversing galaxy clusters are inverse Compton scattered to higher energy. 1-2% of CMB photons traversing galaxy clusters are inverse Compton scattered to higher energy. The Sunyaev-Zel’dovich Effect BIMA Diabolo SuZIE FREQUENCY

5. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 5 Galaxy cluster searches SZ observations do not fade away over large distances. SZ observations do not fade away over large distances. Carlstrom et al., (BIMA)

6. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 6 Sunyaev-Zel’dovich Effect Single Clusters Measure of integrated pressure (total thermal energy) Measure of integrated pressure (total thermal energy) Peculiar velocities at high z Peculiar velocities at high z Distances, H o, H(z) Distances, H o, H(z) Cluster gas mass fractions, structure, etc. Cluster gas mass fractions, structure, etc. Cluster Surveys Exploit SZ redshift independence Exploit SZ redshift independence Measure growth of structure Measure growth of structure Cluster abundance and evolution are critically dependent on cosmology Cluster abundance and evolution are critically dependent on cosmology  Growth based dark energy test (complement to distance based SN test) (compare to x-ray surface brightness)

7. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 7 APEX Telescope 12 m on-axis ALMA prototype built by Vertex RSI 12 m on-axis ALMA prototype built by Vertex RSI Sited at the Atacama plateau, Chile, elevation 16,500 ft Sited at the Atacama plateau, Chile, elevation 16,500 ft Submillimeter observatory Submillimeter observatory 18  m surface accuracy goal 18  m surface accuracy goal 1’ resolution @ 150 GHz 1’ resolution @ 150 GHz 0.4  field of view 0.4  field of view Funded by:

8. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 8

9. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 9 APEX-SZ Experiment Overview PI instrument on APEX PI instrument on APEX First light: December 2005 First light: December 2005 55 detectors, 1 week obs 55 detectors, 1 week obs 280 TES Bolometer channels @ 150 GHz 280 TES Bolometer channels @ 150 GHz 2007, 2008 – roughly 1 month/year of which we typically get 2 weeks good observing 2007, 2008 – roughly 1 month/year of which we typically get 2 weeks good observing Demonstrates new technologies that are scalable to other experiments, i.e., SPT Demonstrates new technologies that are scalable to other experiments, i.e., SPT TES bolometers TES bolometers Frequency domain multiplexed readout Frequency domain multiplexed readout Pulse-tube cooler to eliminate liquid cryogens Pulse-tube cooler to eliminate liquid cryogens Powerful camera for targeted cluster observations Powerful camera for targeted cluster observations Overlaps with northern-hemisphere multi-wavelength observations Overlaps with northern-hemisphere multi-wavelength observations D. Schwan et al., in prep 2009 ~1m

10. Enabling Technology for SZ Surveys “big” bolometer arrays “big” bolometer arrays readout multiplexing readout multiplexing cooling without expendable cryogens cooling without expendable cryogens

11. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 11 TES Detectors Fabbed at UC Berkeley by Jared Mehl Fabbed at UC Berkeley by Jared Mehl 330 element array – 280 wired 330 element array – 280 wired 6 wedges of 55 detectors 6 wedges of 55 detectors Science results typically use ~150 bolometers. Science results typically use ~150 bolometers. March 2009: 2 wedges replaced, optimizing optical coupling, thermal conductivity, bandwidth. March 2009: 2 wedges replaced, optimizing optical coupling, thermal conductivity, bandwidth. NET OLD = 870 μK CMB √s NET OLD = 870 μK CMB √s NET NEW ≈ 500 μK CMB √s NET NEW ≈ 500 μK CMB √s “Spiderweb” Bolometers 0.4 deg FOV

12. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 12 Multiplexed Readout Analog frequency domain multiplexed readout Analog frequency domain multiplexed readout SQUID amplifiers SQUID amplifiers First field implementation First field implementation Developed at LBNL/UC Berkeley/McGill Developed at LBNL/UC Berkeley/McGill Used on APEX-SZ, SPT Used on APEX-SZ, SPT New digital system developed for EBEX, Polarbear, SPTpol New digital system developed for EBEX, Polarbear, SPTpol

13. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 13Cryogenics Mechanical Pulse Tube Cooler (3-4K) Mechanical Pulse Tube Cooler (3-4K) SQUIDs live here. SQUIDs live here.  No expendable cryogens  No nasty fills or LHe delivery issues.  Essential for remote locations 3 stage He 4 He 3 He 3 Absorption fridge (260 mK) 3 stage He 4 He 3 He 3 Absorption fridge (260 mK) Detectors live here. Detectors live here. APEX-SZ Camera shown mounted in cabin with pulse tube lines & ballasts visible.

14. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 14 APEX-SZ Beams & Calibration Daily mappings of Mars Daily mappings of Mars Calibrated against Rudy Model, updated with 1% WMAP data Calibrated against Rudy Model, updated with 1% WMAP data Calibration uncertainty 5.5%: Calibration uncertainty 5.5%: 4% from beam area, 4% from beam area, 3% Mars temporal fluctuations, 3% Mars temporal fluctuations, 1.7% Mars temperature,.. 1.7% Mars temperature,.. Beams: Beams: Gaussian main lobe Gaussian main lobe Near sidelobes increase real beam solid angle Near sidelobes increase real beam solid angle 0.4 deg FOV

15. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 15 Scan Strategy APEX targeted cluster observations are very different from SPT (big sky area) APEX targeted cluster observations are very different from SPT (big sky area) Circular drift scan pattern Circular drift scan pattern 20 circles w/ 5 second period 20 circles w/ 5 second period No turn-arounds No turn-arounds Atmosphere ≡ cosecant baseline Atmosphere ≡ cosecant baseline Cluster signal modulated to frequencies higher than detector 1/f and atmospheric noise Cluster signal modulated to frequencies higher than detector 1/f and atmospheric noise Ground contamination and atmospheric opacity signals can be separated from sky signal Ground contamination and atmospheric opacity signals can be separated from sky signal

16. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 16 Timestream Processing & Map-making Pathological data flagged and removed. Pathological data flagged and removed. 0.2 Hz high pass filter (0.3 Hz for power spectrum) 0.2 Hz high pass filter (0.3 Hz for power spectrum)  removes low frequency noise. Simultaneous fit to cosecant(elevation) function and 20 th order polynomial is subtracted Simultaneous fit to cosecant(elevation) function and 20 th order polynomial is subtracted  removes air mass signal that is modulated by circular scan strategy. Remove 2 nd order spatial polynomial across the array for at each instant in time Remove 2 nd order spatial polynomial across the array for at each instant in time  removed correlated atmosphere Timestreams are RMS weighted and binned into 20” maps Timestreams are RMS weighted and binned into 20” maps Fit to cluster profiles with Markov Chain MC Fit to cluster profiles with Markov Chain MC β model provides adequate parameterization for data β model provides adequate parameterization for data

17. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 17 The Bullet Cluster

18. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 18 The Bullet Cluster Bullet Cluster (1E 0657-56) Jackknife Noise Map Halverson, Lanting et al., arXiv:0807.4208v1 Pathfinder role: First science result released from a large array of multiplexed TES bolometer array. 23 sigma detection no evidence for significant 150 GHz emission from 13.5 mJy @ 270 GHz point source reported by Aztec.

19. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 19 Bullet Cluster Mass weighted electron temperature, using: Mass weighted electron temperature, using: Isothermal Isothermal Clowe et al., 2006, Chandra Clowe et al., 2006, Chandra T E = 10.7± 0.8 KeV T E = 10.7± 0.8 KeV (χ 2 =1.003) (χ 2 =1.003) Xray temperatures range from 10.6 keV with XMM (Zhang et al. 2006) to 13.9 keV with Chandra (Govini et al. 2004) Viklinin 2006, measures ratio of 1.11±0.06 for relaxed clusters White contours: APEX SZE measurement, 100 μK contours Green contours: weak lensing surface mass reconstruction (Clowe et al 2006) Colormap: XMM data, 0.5-2 KeV band

20. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 20 Bullet Cluster Cluster gas profile well-fit by a isothermal elliptical beta model Cluster gas profile well-fit by a isothermal elliptical beta model (adequate for the sensitivity and resolution of these maps – despite the complex nature of this system) Beta model parameters and gas-mass fraction are consistent with those derived from X-ray data Beta model parameters and gas-mass fraction are consistent with those derived from X-ray data SZE DataBest-fit Beta ModelResidual Map Zhang et al (2006) f gas = 0.161± 0.018 (1.42 Mpc) Bradac et al (2006) f gas = 0.14±0.03 (4.9’x3.2’)

21. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 21 A2163 with APEX-SZ & LABOCA Nord, Basu, Pacaud, et al., arXiv:0902.2131 APEX-SZ 150 GHz XMM Newton LABOCA 345 GHz Smoothed APEX-SZ XMM Newton (27±9” offset)

22. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 22 A2163 with APEX-SZ & LABOCA Nord, Basu, Pacaud, et al., arXiv:0902.2131 with new data T=10.5 KeV

23. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 23 A2163 with APEX-SZ & LABOCA Nord, Basu, Pacaud, et al., arXiv:0902.2131 Consistent with XMM (Markevitch, Vikhlinin 2001) r 500 Solid line: APEX + xray Assumes isothermal Shaded region: xray (Squires et al., 1997) WMAP5 Isothermal De-projected profile

24. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 24 APEX 150 GHz Power Spectrum 0.8 sq degrees at 150 GHz with 1’ resolution 0.8 sq degrees at 150 GHz with 1’ resolution 10 nights in Aug/Sept 2007, 2.9 k- bolo-hrs 10 nights in Aug/Sept 2007, 2.9 k- bolo-hrs 12 μK RMS per 1’ pixel 12 μK RMS per 1’ pixel XMM LSS field, centered on XLSSU J022145.2-034614 (5 KeV x-ray cluster) XMM LSS field, centered on XLSSU J022145.2-034614 (5 KeV x-ray cluster) Total Anisotropy < 105 μK 2 at 95%. Total Anisotropy < 105 μK 2 at 95%. σ 8 < 1.18 at 95% σ 8 < 1.18 at 95% Fitting for SZE & Poisson bright point source population Fitting for SZE & Poisson bright point source population Properly accounting for non-Gaussian statistics (limit would be σ 8 < 0.94 assuming Gaussian noise only) Properly accounting for non-Gaussian statistics (limit would be σ 8 < 0.94 assuming Gaussian noise only) Reichardt, Zahn et al., arXiv:0904.3939

25. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 25 APEX 150 GHz Power Spectrum Point Source Power At 150GHz, expect significant power from distant dusty galaxies At 150GHz, expect significant power from distant dusty galaxies Expect 20x less power from radio sources Expect 20x less power from radio sources Negrello et al. (2007) model predicts 1.1 x 10 -5 μ K 2 in the absence of clustering. Negrello et al. (2007) model predicts 1.1 x 10 -5 μ K 2 in the absence of clustering. C l PS ≈ 1 x 10 -5 μK 2 C l PS ≈ 1 x 10 -5 μK 2 Nearly independent of flux cut for masking point sources Nearly independent of flux cut for masking point sources With BLAST 600 GHz data  spectral index α=2.64 +0.4 -0.2. With BLAST 600 GHz data  spectral index α=2.64 +0.4 -0.2. Agrees with MAMBO/SCUBA index, 2.65 Greve et al. (2004) Agrees with MAMBO/SCUBA index, 2.65 Greve et al. (2004) Knox et al., 2004. Knox et al., 2004. Dusty galaxies account for most power in APEX-SZ maps. Dusty galaxies account for most power in APEX-SZ maps.

26. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 26 The Future APEX-SZ on its way home for upgrade APEX-SZ on its way home for upgrade Half detectors at 90 GHz, half at 150 GHz Half detectors at 90 GHz, half at 150 GHz Factor 4 improvement in mapping speed Factor 4 improvement in mapping speed

27. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 27 What’s Next? Couple dozen clusters in the can Couple dozen clusters in the can Gas mass fraction vs. radius Gas mass fraction vs. radius  Constrain H 0, w Scaling relations: (Amy Bender) Scaling relations: (Amy Bender) Redshift evolution – good overlap with Redshift evolution – good overlap with multi-wavelength data  Including high-Z clusters Relaxed vs. complex systems Relaxed vs. complex systems Constraints for cluster simulations Constraints for cluster simulations Isothermal vs. universal temperature profiles Isothermal vs. universal temperature profiles Radial profiles Radial profiles Extended cluster emission Extended cluster emission  widen scan strategy to get beyond r virial

28. Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 Matt.Dobbs@McGill.ca Perimeter SZ Workshop 2009 28Summary APEX-SZ is a 280 element 150 GHz bolometer array operating on the 10m APEX telescope APEX-SZ is a 280 element 150 GHz bolometer array operating on the 10m APEX telescope Technology pathfinder role: TES bolometer array, multiplexed readout, no expendable cryogens. Technology pathfinder role: TES bolometer array, multiplexed readout, no expendable cryogens. New SZ observations of the Bullet and A2163 New SZ observations of the Bullet and A2163 General consistency with expectations from x-ray. General consistency with expectations from x-ray. Power spectrum is consistent with WMAP5, with majority of power from dusty galaxies. Power spectrum is consistent with WMAP5, with majority of power from dusty galaxies. σ 8 <1.18 at 95% σ 8 <1.18 at 95%