CITA Extragalactic Sims: Current Snapshot

CITA Extragalactic Sims: Current Snapshot Low-z Clusters & Groups 320 realizations (M > 2 x 10 13 Msun/h, z < 1.25) High-z Clusters & Groups + Subgrid Biasing Single realization (M > 2 x 10 13 Msun/h, z < 4.5) Alvarez, Bond, Stein Bahmanyar, Battaglia, Pham, van Engelen Halo Catalogs: [x, y, z, vx, vy, vz, Mvir] Maps: [tsz, ksz, κcmb, CIB] questions/requests: malvarez@cita.utoronto.ca Mapmaking pipeline SZ, CIB, lensing Alvarez et al. (2017) Input: profile table, halo catalog, unlensed primary Output: lensed CMB, SZ, CIB Battaglia et al. (2012) + = halo catalog and/or linear density tabulated profile and/or bias parameters multifrequency maps

CITA SZ/CIB Sims Alvarez, Bond, Stein, Battaglia Systematics & foregrounds 1. Baryonic physics effects on halo mass 2. tsz - CIB (important for ksz power) 3. Cosmic variance at low z and high Mhalo 4. External data set calibration 5. Scatter/anisotropy in mass-observable * bullets increase with z at fixed M 6. Neutrino effects on halo mass * beyond linear + LCDM hmf 7. Velocity bias of halos (for ksz cf RSD) Currently being addressed 1. Spherical gas/stars/pressure/dm profiles from a single sim (Battaglia et al.) 2. Planck model for CIB x tsz 3. Large ensembles * ~500 Mhalo > 2e13 and z<1.3 done 4. CMASS HOD mocks Science Requirements 1. Σm ν / σ 8 (z) / w(z) 2. fσ 8 (z) 3. Δz reion / τ es 4. (P, ρ) vs (M halo, z) Going forward (priorities?) 1. More realistic halo model * correlate with substructure and tides 2. Scan astrophysical parameter space * hydro sim calibration (eg CCA galform) * couple to external obs (eg CCAT-p) 3. External structure * the cosmic web * missing baryons 4. Beyond LCDM * neutrinos, modified gravity 5. Correlated external data set mocks * LSST, HIRAX, Euclid, WFIRST, 6. Reionization ksz/tsz 7. Integration into end-to-end data analysis pipeline for validation

CITA Lensing Sims Alvarez, Stein, Bond, van Engelen, Battaglia, Codis, Liu Systematics & foregrounds 1. Baryonic physics effects on mass dist. 2. CIB non-gaussianity & lensing reconst. 3. External data set calibration 4. Scatter/anisotropy in mass-observable * bullets increase with z at fixed M 5. Neutrino effects on halo mass * beyond linear + LCDM hmf 6.??? Currently being addressed 1. Spherical gas/stars/dm profiles from a single sim (Battaglia et al.) 2. Planck model for CIB x tsz 3. Lensing of primary with Borne 4. Comparing LPT for matter outside halos to N-body 5. CMASS HOD mocks Science Requirements 1. r 2. Σm ν / σ 8 (z) / w(z) 3. ρ vs (M halo, z) Going forward (priorities?) 1. More realistic halo model * correlate with substructure and tides 2. Scan astrophysical parameter space * hydro sim calibration (eg CCA galform) * couple to external obs (eg Herschel) 3. External structure * the cosmic web * missing baryons 4. Beyond LCDM * neutrinos, modified gravity 5. Correlated external data set mocks * LSST, HIRAX, Euclid, WFIRST, 6. Integration into end-to-end data analysis pipeline for validation

CITA Lensing Sims Alvarez, Stein, Bond, van Engelen, Battaglia, Codis, Liu

Point source and atmosphere effects on tsz science Kevin Huffenberger Florida State University

Setup Predict ILC-multifrequency reconstruction of Compton-y power spectrum. Fast... can run suites of instrument models. Multi-component sky model with tsz, ksz, CIB sources, Radio sources. Source flux cuts computed self-consistently.

l(l + 1)Cl/2π (µk 2 ) 10 3 10 2 10 1 10 0 10-1 Sky @ 150 GHz cmb cibc cibp radp ksz tsz @ 150 10-2 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 multipole l

S 2.5 dn ds (sr 1 Jy 1 5 ) D3000 (μk 2 ) 10 3 10 2 10 1 10 0 10-1 10-2 10-3 10-4 220 GHz 10-5 10-2 10-1 10 0 10 1 10 2 10 3 10 4 70 60 50 40 30 20 10 Bethermin 2012, modified so total power is not too high CIB counts CIB power Conf. limits with "10m," "7m," "5m" 0 10-2 10-1 10 0 10 1 10 2 10 3 10 4 Flux density S (mjy)

"7m" configuration channels (GHz) 21.0 29.0 40.0 95.0 150.0 220.0 270.0 beam fwhm (arcmin) 9.0 6.5 4.7 2.0 1.3 0.9 0.7 channel map noise (muk-arcmin) 17.4 12.3 12.0 3.3 3.5 11.5 19.7 effective map noise = 2.25 muk-arcmin fsky = 0.50 optional atmosphere: cf Hasselfield

7.0 "7m" dish, flux cut vs. noise 6.5 6.0 220 GHz flux cut (mjy) 5.5 5.0 4.5 4.0 3.5 3.0 2.5 10-2 10-1 10 0 10 1 10 2 270 GHz noise (µk-arcmin)

l(l + 1)ClT 2 cmb /2π (µk2 ) 20 15 10 5 strawso_2016octcaseb_5.0m_fsky50_5yr_drop270 err per multipole noise power tsz compton y 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 multipole l

l(l + 1)ClT 2 cmb /2π (µk2 ) 20 15 10 5 Compton y noise strawso_2016octcaseb_5.0m_fsky50_5yr err per multipole noise power tsz compton y 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 multipole l

l(l + 1)ClT 2 cmb /2π (µk2 ) 20 15 10 5 Compton y noise strawso_2016octcasec_7.0m_fsky50_5yr err per multipole noise power tsz compton y 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 multipole l

l(l + 1)ClT 2 cmb /2π (µk2 ) 20 15 10 5 strawso_2016octcasec_7.0m_fsky50_5yr_atmsimple err per multipole noise power tsz compton y 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 multipole l

1800 1600 1400 1200 2.3 µk, 5.0 m, fsky = 50 w/ l knee = 3400 2.3 µk, 7.0 m, fsky = 50 w/ l knee = 3400 S/N( > lmin) 1000 800 600 400 200 0 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 minimum multipole l min

Are foregrounds important for ksz? 1 When overlapping with spectroscopy, the standard estimators are differential à most components that are uncorrelated with velocities average out as 1/ p N objects Turn bias into noise à may not need as good cleaning as for other science? But how big is the extra noise? And Can this fail in the high S/N regime? Yes, in a few ways! Cancellation can be imperfect for extreme objects (e.g. tsz) Extragalactic can also impose a similar noise floor à need to estimate! Foregrounds can be correlated with velocity Projected fields, power spectrum, 4pt function: foregrounds Simone Ferraro (Berkeley)

Atmospheric noise in T 2 Louis ++ 2016 Simone Ferraro (Berkeley)

Effect of atmospheric noise 3 Optimistic scenario PRELIMINARY Simone Ferraro (Berkeley)

Adding foregrounds + atmosphere 4 ILC on foregrounds only with 5 frequency channels PRELIMINARY! ILC mitigation NO foregrounds/atm RAW foregrounds + atm Simone Ferraro (Berkeley)

Small Angular Scales Discussion Session: Clusters Lindsey Bleem CMB-S4 Workshop 2/28/17

Science Reach/Systematics/ Dependencies What can CMB clusters + LSST clusters do, how much better than S3+LSST? Do we need S4 to reach interesting depth or would more time with existing large aperture facilities (SPT, ACT) be enough? Catalog-based modeling underway; more sophisticated modeling longer term effort! Modeling selection (what is going on in the highest-z clusters? [not foregrounds per se]); false associations; incomplete follow-up (can we get all the telescope time we need to follow up high-z clusters?) Learning a lot from S1-S3 experiments; Telescope access open question! MASS CALIBRATION Optical Weak lensing systematics: shear bias, photo-zs, cluster member contamination, miscentering, profile assumptions Share problems with LSST cluster samples, simulation + data driven efforts underway to tackle these issues! CMB weak lensing biases: tsz, ksz, CIB+Synchrotron from cluster members Correlated Observables - Simulations, targeted followup observations Theoretical Mass Function -> Observed cluster relation (differences in halo catalogs, observer definition of clusters, etc)

Slide from Steve Allen s Talk The discovery space for CMB-S4 Primary discovery space

Science Reach/Systematics/ Dependencies What can CMB clusters + LSST clusters do, how much better than S3+LSST? Do we need S4 to reach interesting depth or would more time with existing large aperture facilities (SPT, ACT) be enough? Catalog-based modeling underway! Modeling selection (what is going on in the highest-z clusters? [not foregrounds per se]); false associations; incomplete follow-up (can we get all the telescope time we need to follow up high-z clusters?) Learning a lot from S1-S3 experiments; Telescope access open question! MASS CALIBRATION Optical Weak lensing systematics: shear bias, photo-zs, cluster member contamination, miscentering, profile assumptions Share problems with LSST cluster samples, simulation + data driven efforts underway to tackle these issues! CMB weak lensing biases: tsz, ksz, CIB+Synchrotron from cluster members Correlated Observables - Simulations, targeted followup observations Theoretical Mass Function -> Observed cluster relation (differences in halo catalogs, observer definition of clusters, etc)

Connecting Simulations to Observations: tsz + ksz: Baryons added in post processing. (Flender+16; ApJ 823, 98F) Forecasts pairwise ksz signal from DES+SPT; DESI+ACTpol Lensing Pipelines Weak Lensing (OpVcal + CMB) PICS: RealisVc Strong Lensing Pipeline (Li+2016; ApJ 828,1, 54) Galaxies in Simulated Clusters! A mul&-wavelength view of clusters in next genera&on surveys (SZ, Op&cal, X-ray) CMB-S4 Workshop, SLAC, 2/27/17 Simulated cluster tsz, weak- and strong-lensing signals 5

Connecting Simulations to Observations: tsz + ksz: Baryons added in post processing. (Flender+16; ApJ 823, 98F) Forecasts pairwise ksz signal from DES+SPT; DESI+ACTpol Key interprevng DES+SPT pairwise ksz signal (Soergel +2016;MNRAS.461.3172S) Lensing Pipelines Weak Lensing (OpVcal + CMB) PICS: RealisVc Strong Lensing Pipeline (Li+2016; ApJ 828,1, 54) Galaxies in Simulated Clusters Galaxy Placement in halos [D. Korytov+ in prep] CMB-S4 Workshop, SLAC, 2/27/17 6

ExaSky: Conservative Conservatice Reproducing Reproducing Kernel Kernel SPH (Particle SPH (Particle Hydro) Hydro) Triple-point shock test with CRKSPH, Frontiere et al 2016 Frontiere, Raskin, Owen, J. Comp. Phys. (2017); arxiv:1605.00725 [physics.comp-ph] Blob test, Frontiere et al 2016

ExaSky: Conservative Conservatice Reproducing Reproducing Kernel Kernel SPH (Particle SPH (Particle Hydro) Hydro) Triple-point shock test with CRKSPH, Frontiere et al 2016 Upcoming Frontiere, Raskin, Owen, J. Comp. Phys. (2017); arxiv:1605.00725 [physics.comp-ph] Blob test, Frontiere et al 2016

Science Reach/Systematics/ Dependencies What can CMB clusters + LSST clusters do, how much better than S3+LSST? Do we need S4 to reach interesting depth or would more time with existing large aperture facilities (SPT, ACT) be enough? Catalog-based modeling underway! Modeling selection (what is going on in the highest-z clusters? [not foregrounds per se]); false associations; incomplete follow-up (can we get all the telescope time we need to follow up high-z clusters?) Learning a lot from S1-S3 experiments; Telescope access open question! MASS CALIBRATION Optical Weak lensing systematics: shear bias, photo-zs, cluster member contamination, miscentering, profile assumptions Share problems with LSST cluster samples, simulation + data driven efforts underway to tackle these issues! CMB weak lensing biases: tsz, ksz, CIB+Synchrotron from cluster members Correlated Observables - Simulations, targeted followup observations Theoretical Mass Function -> Observed cluster relation (differences in halo catalogs, observer definition of clusters, etc)