Fingerprints of the first black holes? Crosscorrelationg the Near-Infrared and X-ray background in COSMOS
Abstract
Unresolved cosmic backgrounds carry information about the populations of stars and black holes not accessible by any other current observational technique. Studies using both Spitzer and AKARI have revealed large-scale fluctuations in the Cosmic Infrared Background (CIB) after subtracting resolved sources to faint levels. The signal, which is stronger than all foreground contributions, has been attributed to the earliest light in the universe. Other teams, by using Spitzer and CIBER, suggested that this signal could originate from stars tidally stripped from their parent galaxies at low redshift as a result of mergers. Only sensitive multi-wavelength observations can distinguish between these radically different interpretations. Recently our team discovered an intriguing cross-correlation signal between the unresolved CIB and X-ray background (CXB) suggesting significant black hole populations among the CIB sources (C13). The analysis used data from Spitzer and Chandra in an overlapping 8'x45' region of the All-Wavelength Extended Groth strip International Survey (AEGIS), probing the clustering of the underlying sources to angular scales 20', but the experiment is limited by the size and the elongated configuration of the field. Here we propose to use the much more extended multiwavelength observations in the COSMOS field to obtain new constraints on this important debate. Spitzer has recently completed the Spitzer Large Area Survey with Hyper-Suprime-Cam (SPLASH), which observed the full 2 square degree COSMOS field to about half the depth of the AEGIS field, but a 20 times larger sky area. SPLASH, focusing on the COSMOS field with excellent multi-wavelength coverage from space and ground-based observatories, is ideally suited to study the CIB to better precision and at larger scales than available so far. We propose to cross-correlate the Spitzer unresolved CIB to the unresolved CXB observed by Chandra and XMM-Newton. Chandra has covered the full 2 deg2 COSMOS field with a total exposure of 5.6 Ms. XMM-Newton has covered the whole 2 deg2 COSMOS field with an exposure time of about 1.5 Ms . The technique for our cross-correlation analysis has been described in detail by C13. All the known instrumental and non-cosmic backgrounds are removed from the data and all discrete sources are excised using the same mask for both datasets. Then the exposure-corrected images are cross-correlated with each other. Compared to the AEGIS cross-correlation signal we expect a significantly better signal to noise ratio, which together with the larger area, will allow us to reach much larger angular scales, where different models can be constrained. The better photon statistics will also allow to separate the signal into at least four independent X-ray energy bands, thus yielding coarse spectral information allowing to discriminate between different models for the X-ray emitting sources. If the X-ray sources are High-Mass X-ray binaries with stellar-mass black holes descendent from massive Pop III stars at z>7, we would expect the X-ray emission to be only moderately absorbed. If on the other hand, the X-ray emission originates from direct collapse black holes with masses in the range 10^5-6 solar masses, we expect a significant Compton-thick absorption component, similar to the resolved X-ray background produced at lower redshifts. Finally, the new data on the CIB-CXB crosscorrelation will also strongly constrain the intrahalo light scenario proposed as an alternative interpretation. At any rate, these measurements allow diagnostics into the faintest discrete source populations right after they emerged from the dark ages or buried within halos. Until the next generation of telescopes, JWST, WFIRST, and EUCLID, the ground based 30m telescopes and later the ESA X-ray observatory Athena, this is one of the few avenues to constrain the formation and early evolution of stars and black holes in the Universe.
- Publication:
-
NASA ADAP Proposal
- Pub Date:
- 2015
- Bibcode:
- 2015adap.prop...40H