SPHEREx: An All-Sky Spectral Survey

Designed to Explore ▪ The Origin of the Universe ▪ The Origin and History of Galaxies ▪ The Origin of Water in Planetary Systems

The First All-Sky Near-IR Spectral Survey A Rich Legacy Archive for the Astronomy Community with 100s of Millions of Stars and Galaxies

Low-Risk Implementation ▪ Single Observing Mode ▪ No Moving Parts ▪ Large Technical & Scientific Margins SPHEREx Addresses Three Main Astrophysics Questions

Map the large scale structure of galaxies to study the process of Inflation in the early universe, addressing NASA’s objective to Probe the origin and destiny of the Universe.

Time Since Big Bang Measure the total light production from stars and galaxies across cosmic history, addressing NASA’s objective to Explore the origin and evolution of galaxies.

Time Since Big Bang

Determine how interstellar ices bring water and organics into proto-planetary systems, furthering NASA’s objective to Explore whether planets around other stars could harbor life. Stages of Star Formation SPHEREx Team

Science Team Collaborators

Name Institution Role Responsibility Experience Name Institution N/C* Responsibility Jamie Bock Caltech/JPL PI Principal investigator, overall US PI of HReorslacnhdel /dSeP IPRuEt taenr d PlancJkP./HLF I;CIBER LN4 Cosmo parameter fitting management PI; BICEPT2i/mKe Eckif lCeor -PI JPL Synergies with Euclid Matt Ashby CfA Co-I Pipeline development Spitzer/IRNACic otelaasm F lagey IfA LN4 Ice spectral catalog Peter Capak IPAC Co-I Galaxy spectral fitting modules PI of COSYMaOnS G sounrvge y UC Irvine LN4 Galaxy clustering theory Asantha Cooray UC Irvine Co-I Galaxy Formation L4 lead CIBER, SEpiltizsearb, Hetehr sKcrhaeul saena lysis Stanford LN4 Bispectrum simulations Olivier Doré JPL/Caltech PS Project scientist; Planck, WDMaAnPie sl cMieanstistet,r sE uclid ScieCncaelt eTceha m SNp ectral redshift fitting Inflationary Cosmology L4 lead Phil Mauskopf ASU Cluster catalog Chris Hirata OSU Co-I Inflation and Cosmology studies Euclid Science Team N Woong-Seob Jeong KASI Co-I KASI PI NISS PI, MBIeRrItSra nd Menneson JPL IcCe catalog follow-up Phil Korngut Caltech Co-I Deputy instrument scientist CIBER insHtriuemn eNngt udeyveenlo pment JPL InNs trument characterization Dae-Hee Lee KASI Co-I Ground test equipment CIBER, NKISaSr,in M ÖIRbIeSr dge velopment CfA LC4 Ice science interpretation Gary Melnick CfA Co-I Galactic Ice L4 lead SWAS PI Anthony Pullen CMU LN4 Galaxy clustering theory Roger Smith Caltech Co-I Detector array development H2RGs foAr ldviviseers Re aacsctraonoemlli y apps JHU CNo smic magnification sims Yong-Seon Song KASI Co-I Cosmology interpretation PI of KASVI poalkrteicri pTaotilolsn in DESI CfA LC4 Ice pipeline modules Stephen Unwin JPL Co-I Galactic ice science SIM and TSPaFlm-I aDnP SH;a EbxiEb P Dep ProgA rSgcoiennnteis t GNa laxy catalog simulations Michael Werner JPL Co-I Legacy survey science Spitzer PSK atrin Heitmann Argonne GNa laxy catalog simulations Michael Zemcov Caltech Co-I Instrument scientist CIBER anMd aQrucAo DV iinesrtoru ments Stanford InNt ensity mapping sims * Funding Source: N=NASA C=Contributed Competed in Dec 2014 SMEX; Selected by NASA for a Phase-A study in July 2015. Phase A report due in July 2016. Phase A down-selection ~ December 2016 SPHEREx: Simple Instrument, Large Margins Deployed Wide-field thermal telescope shields 20 cm eff. aperture

High-throughput spectrometer Passive uses a linear-variable filter in front cooling of each detector array 185 cm system

BCP 100 spacecraft Deployed Solar panels Parameter Value Telescope Effective Aperture 20 cm Parameter Performance Margin Pixel Size 6.2" x 6.2" Spacecraft Ball BCP 100 N/A Field of View 2 x (3.5° x 7.0°); dichroic Science Data 73 Gb/day 97% Spectrometer Linear-Variable Filters Downlink Resolving Power and R=41.5 λ=0.75 - 4.1 µm Pointing Stability 2.1" (1σ) over 200 s 43% Wavelength Coverage R=150 λ=4.1 - 4.8 µm Pointing 22.7” (1σ) 164% 2 x Hawaii-2RG 2.5 µm Control Arrays 2 x Hawaii-2RG 5.3 µm Pointing Agility 70º in 116 s (large slews) 29% Point Source Sensitivity 18.5 AB mag (5σ) with 8.8’ in 6 s (small steps) 233% Observatory Mass 173.6 kg (MEV) 53% (MEV Performance) 300% margin to req’t Observatory Cooling All-Passive 171.8 W (MEV) 36% 2.5 µm Array and 80 K with 700% margin Power Solar Array Power Optics Temperature (Req’t) on total heat load 234 W N/A 5.3 µm Array Temperature 55 K with 450% margin Output (EOL) (Req’t) on total heat load Payload Mass 68.1 kg (CBE+31% Ctg) Thermal Science Payload Power 27.8 W (CBE+30% Ctg) Margins: High-Throughput LVF Spectrometer

Linear Variable Filter

Focal Plane Assembly

Methane on Pluto 1.0

0.8

0.6

0.4I/F

0.2

0.0 1.6 1.8 2.0 2.2 2.4 Infrared Spectral Image Wavelength (µm) LVFs used on ISOCAM, HST-WFPC2, Spectra obtained by stepping source over the New Horizons LEISA, & OSIRIX-Rex (2016 launch) FOV in multiple images: no moving parts Mapping the Full Sky with SPHEREx All-Sky Coverage

Designed for uniform coverage after 25 months of observations with 4 independent surveys. Deep Surveys at Poles

Except for two 100 sq. degree regions SPHEREx observes the sky simply by pointing the spacecraft At the poles with are ~30x deeper. over multiple orbits to obtain complete spectra. An opportunity for unique science SPHEREx Creates an All-Sky Legacy Archive Object # Sources Legacy Science Reference Detected 1.4 billion Properties of distant and heavily obscured galaxies galaxies Galaxies 120 million Study (H, CO, O, S, H2O) line and PAH emission by galaxy σ(z)/(1+z) type. Explore galaxy and Simulation AGN life cycle based on < 0.03 COSMOS and Pan-STARRS Galaxies σ(z)/ 9.8 million Cross check of Euclid photo-z. Measure dynamics of groups (1+z) and map filaments. < 0.003 QSOs > 1.5 million Understand QSO lifecycle, environment, and taxonomy Ross et al. QSOs 0-300 Determine if early QSOs exist. (2013) plus Follow-up spectro-scopy simulations at z > 7 probes EOR through Lyα forest Clusters with ≥ 25,000 Redshifts for all eRosita Geach et al., clusters. Viral masses and 2011, SDSS 5 members merger dynamics counts Main >100 million Test uniformity of stellar 2MASS catalogs mass function within our sequence stars Galaxy as input to extragalactic studies Mass-losing, Over 10,000 Spectra of M supergiants, Astro-physical OH/IR stars, Carbon stars. Quantities, 4th Notable Features of the SPHEREx All-Sky Survey dust forming of all types Stellar atmospheres, dust edition [ed. stars return rates, and composition A.Cox] p. 527 • High S/N spectrum for every 2MASS source of dust • Solid detection of faintest WISE sources Brown dwarfs >400, incl. Atmospheric structure and dwarfarchives.o composition; search for rg and J.D. • Catalogs ideal for JWST observations >40 of types T hazes. Informs studies of Kirkpatrick, and Y giant exoplanets priv. comm.

Stars with hot >1000 Discover rare dust clouds Kennedy & produced by cataclysmic Wyatt (2013) New ideas recently brought to our attention dust events like the collision which • Redshifts for the all-sky eRosita X-Ray survey produced the Earth’s moon • Photo baselines for wide-field transient survey Diffuse ISM Map of the Study diffuse emission from GLIMPSE survey interstellar clouds and (Churchwell Galaxy • Mapping 3D distribution of Galactic ices nebulae; (H, CO, S, H2O and et al. 2009) PAH emission) SPHEREx Large Volume Galaxy Survey SPHEREx Surveys Maximum Cosmic Volume Catalog Split into Redshift Accuracy Bins

σz/(1+z)

SPHEREx Large-Volume Redshift Catalog • Largest effective volume of any survey, near cosmic limit • Excels at z < 1, complements dark energy missions (Euclid, WFIRST) targeting z ~ 2 • SPHEREx + Euclid measures gravitational lensing and calibrates Euclid photo-zs Survey Designed for Two Tests of Non-Gaussianity • Large scale power from power spectrum: large # of low-accuracy redshifts • Modulation of fine-scale power from bispectrum: fewer high-accuracy redshifts SPHEREx Tests Inflationary Non-Gaussianity

• Non-Gaussianity distinguishes between multi- and single-field models

• Projected SPHEREx sensitivity is δfNL < 1 (2σ)

- Two independent tests via power spectrum and bispectrum • Competitively tests running of the spectral index • SPHEREx low-redshift catalog is complementary for dark energy SPHEREx Measures Cosmic Light Production Two Ways to Measure Cosmic Light Production What Constitutes Cosmic Light Production?

Moseley & Zemcov 13+ billion years of galaxy Diffuse emission between collisions and mergers galaxies from tidally Science 2014 stripped stars

Inflation fraction of a trillionth of a second

Cosmic microwave background ~380,000 years

First stars & galaxies ~400 million years Present universe ~13.8 billion years

1) Photon Production in Galaxies 1) Individual Galaxies & Redshifts Nucleosynthesis & black holes, peaks at z ~ 2 Large telescope for point source sensitivity 2) First Stars and Galaxies 2) Large-Scale Patterns in the Background Epoch of Reionization z > 6 Small telescope with fidelity on degree scales 3) Intra-Halo Light large telescope for point source sensitivity → the amplitude of large-scale (clustering) 4) Surprises? fluctuations proportional to total light production E.g. Light from particle decay SPHEREx Measures Large-Scale Fluctuations Fluctuations in Continuum Bands Fluctuations in Spectral Lines

•Emission lines encode clustering signal •SPHEREx has ideal wavelength coverage at each redshift over cosmic history and high sensitivity •Amplitude gives line light production •Multiple bands enable correlation tests •Multiple lines trace star formation history sensitive to redshift history - High S/N in Hα for z < 5; OIII and Hβ for z < 3 •Method demonstrated on Spitzer & CIBER - Lya probes EoR models for z > 6 - Hα and Lyα crossover region 5 < z < 6 12 SPHEREx Studies Reionization [with 21-cm]

[example plot]

Hα [Errors to be calculated for HERA/SKA-low sensitivity]

Lyα

Fig. 13.— The cross-correlation coefficient of the CII and 21-cm emission for 1m and 10m aperture at z =6,z =7andz =8.Theerror bars of r are also shown (red solid), and the blue dashed ones are the contribution from the 21-cm emission with the LOFAR (left panel), (a) andSPHEREx will map Ly with the SKA (right panel). Weα emission during find the 21-cm noise dominates the errors(a) atSPHEREx Ly z=6 and 7. α intensity maps ideal for 7. OUTLINE OF A CII INTENSITY MAPPING pendentcross-correlation with 21-cm spatial pixels in the spectrometer, and the band- reionization over the whole sky. EXPERIMENT width offluctuations each spectral element. Changing the survey area (b) Apart from z~6, SPHEREx alone will not We now discuss the requirements on an instrument de- or bandwidth will affect the minimum k values probed in signed to measure the CII line intensity variations and the(b) 3-d powerCorrelation coefficient negative at large spectrum as well as the number of modes have sensitivity for a detection of the the power spectrum at z>6. Here we provide a sketch used in the error calculation. To generate concrete exam- of a possible instrument design including calculated noise ples in thisscales to positive at small scales. discussion, we concentrate on calculating the power spectrum [for Hopkins & Beacom requirements based on current technology; a detailed de- CII power spectrum at z = 6, 7 and 8 and assume that sign study will be left to future work as our concept is we will(c) makeDirect measure of bubble sizes! use of 20 GHz effective bandwidths at fre- SFRD].further developed. quencies centered at these redshfits; such effective bands An experiment designed to statistically measure the are readily achieved by summing neighbouring high res- CII transition at high redshift requires a combination of olution spectral bins. At z = 7, for example, a 20 GHz a large survey area and low spectral resolution; because bandwidth corresponds to a ∆z =0.62. Averaging over of this, the use of an interferometer array such as the one a larger spectral window to reduce noise will make the suggested in Gong et al. (2011) for CO seems impractical. cosmological evolution along the observational window Instead we opt for a single aperture scanning experiment non-negligible. employing a diffraction grating and bolometer array in To understand the effect of spatial resolution on this order to provide large throughput. measurement we consider three possible primary aper- For proper sampling of the CII signal on cosmologi- ture sizes of 1, 3 and 10 m; the resulting instrumental cal scales and cross-correlation with 21-cm experiments, parameters are listed in Table 1. These apertures cor- we would require at minimum a survey area of 16 deg2 respond yield beams of 4.4, 1.5, and 0.4 arcmin and co- and a free spectral range (FSR) of 20 GHz. At a red- moving spatial resolutions of 8.3, 2.8, 0.8 Mpc/h at an shift of z = 7, a box 4 degrees on a side and 20 GHz observing frequency of 238 GHz (z = 7), respectively. deep corresponds to a rectangular box with a comoving These apertures probe the linear scales (k<1.0 h/Mpc) angular width of 443 Mpc/h and a depth along the line and are well matched to 21-cm experiments. In this ex- of sight of 175 Mpc/h. However, since larger FSRs are perimental setup light from the aperture is coupled to a easily achieved with diffraction grating architectures and imaging grating dispersive element illuminating a focal would allow for better measurement of the reionization plane of bolometers, similar to the Z-Spec spectrometer signal and separation of its foregrounds, the instrumental but with an imaging axis (Naylor et al. 2003; Bradford concept presented here covers the 220 GHz atmospheric et al. 2009). window with an FSR of 125 GHz. Concretely, covering We assume a fixed spectral resolution of 500 km/s ∼ from 185 to 310 GHz with a spectral resolution of 0.4GHz (= 0.4 GHz as discussed above) in each of the three cases allows measurement of CII in the range 5.1 z 9.3 giving a spatial resolution of 3.5 Mpc/h and a maximum with a smallest redshift interval ∆z of 0.01. ≤ ≤ k mode of k 0.91 h/Mpc. Current technology can al- ≈ The integration time per beam on the sky required to low fabrication of 20, 000 bolometers in a single focal ∼ survey a fixed area depends on a number of parameters plane; for the instrument concept we freeze the number including the size of the aperture, the number of inde- of detectors at this value. The spectrometric sensitivity NASA Small Explorer in Phase-A Study

All-sky near-IR spectral survey

• Unprecedented measurements of the Non- Gaussian signature from Inflation

• Survey of the role of organic ices in molecular clouds and young stellar systems

• Probing galaxy formation through precise measurements of extragalactic background light anisotropy

• Rich spectral catalog for the astronomy community Summary paper: Doré et al. arXiv 1412.4872

What can you do? Perform theory predictions and write papers on your sciences with SPHEREx. Provide input to the SPHEREx Science Team on data requirements and tools for your sciences with SPHEREx data Details soon at http://spherex.caltech.edu