No Arabic abstract
The epoch of reionization (6 < z < 10) marks the period in our universe when the first large galaxies grew to fruition, and began to affect the universe around them. Massive stars, and potentially accreting supermassive black holes, filled the universe with ionizing radiation, burning off the haze of neutral gas that had filled the intergalactic medium (IGM) since recombination (z~1000). The evolution of this process constrains key properties of these earliest luminous sources, thus observationally constraining reionization is a key science goal for the next decade. The measurement of Lyman-alpha emission from photometrically-identified galaxies is a highly constraining probe of reionization, as a neutral IGM will resonantly scatter these photons, reducing detectability. While significant work has been done with 8-10m telescopes, these observations require extremely large telescopes (ELTs); the flux limits available from todays 10m class telescopes are sufficient for only the brightest known galaxies (m < 26). Ultra-deep surveys with the Giant Magellan Telescope (GMT) and Thirty Meter Telescope (TMT) will be capable of detecting Lyman-alpha emission from galaxies 2-3 magnitudes fainter than todays deepest surveys. Wide-field fiber spectroscopy on the GMT combined with narrow-field AO-assisted slit spectroscopy on the TMT will be able to probe the expected size of ionized bubbles throughout the epoch of reionization, following up degree scale deep imaging surveys with the Wide Field Infrared Space Telescope. These data will provide the first resolved Lyman-alpha-based maps of the ionized intergalactic medium throughout the epoch of reionization, constraining models of both the temporal and spatial evolution of this phase change.
We calculate Lyman Alpha Emitter (LAE) angular correlation functions (ACFs) at $z simeq 6.6$ and the fraction of lifetime (for the 100 Myrs preceding $zsimeq6.6$) galaxies spend as Lyman Break Galaxies (LBGs) or as LBGs with Lyman Alpha (Ly$alpha$) emission using a model that combines SPH cosmological simulations (GADGET-2), dust attenuation and a radiative transfer code (pCRASH). The ACFs are a powerful tool that significantly narrows the 3D parameter space allowed by LAE Ly$alpha$ and UV luminosity functions (LFs) alone. With this work, we simultaneously constrain the escape fraction of ionizing photons $f_{esc}=0.05-0.5$, the mean fraction of neutral hydrogen in the intergalactic medium (IGM) $langle chi_{HI} rangle leq 0.01$ and the dust-dependent ratio of the escape fractions of Ly$alpha$ and UV continuum photons $f_{alpha}/f_c=0.6-1.2$. Our results show that reionization has the largest impact on the amplitude of the ACFs, and its imprints are clearly distinguishable from those of $f_{esc}$ and $f_alpha/f_c$. We also show that galaxies with a critical stellar mass of $M_* = 10^{8.5} (10^{9.5})M_{odot}$ produce enough luminosity to stay visible as LBGs (LAEs). Finally, the fraction of time during the past 100 Myrs prior to z=6.6 a galaxy spends as a LBG or as a LBG with Ly$alpha$ emission increases with the UV magnitude (and the stellar mass $M_*$): considering observed (dust and IGM attenuated) luminosities, the fraction of time a galaxy spends as a LBG (LAE) increases from 65% to 100% (0-100%) as $M_{UV}$ decreases from $M_{UV} = -18.0$ to $-23.5$ ($M_*$ increases from $10^8-10^{10.5} M_{odot}$). Thus in our model the brightest (most massive) LBGs most often show Ly$alpha$ emission.
The spatial fluctuations of the extragalactic background light trace the total emission from all stars and galaxies in the Universe. A multi-wavelength study can be used to measure the integrated emission from first galaxies during reionization when the Universe was about 500 million years old. Here we report arcminute-scale spatial fluctuations in one of the deepest sky surveys with the Hubble Space Telescope in five wavebands between 0.6 and 1.6 $mu$m. We model-fit the angular power spectra of intensity fluctuation measurements to find the ultraviolet luminosity density of galaxies at $z$ > 8 to be $log rho_{rm UV} = 27.4^{+0.2}_{-1.2}$ erg s$^{-1}$ Hz$^{-1}$ Mpc$^{-3}$ $(1sigma)$. This level of integrated light emission allows for a significant surface density of fainter primeval galaxies that are below the point source detection level in current surveys.
Detection of the redshifted 21cm-line signal from neutral hydrogen in the intergalactic medium (IGM) during the Epoch of Reionization (EoR) is complicated by intense foregrounds such as galactic synchrotron and extragalactic radio galaxies. The 21cm-Lyman-$alpha$ emitter(LAE) cross-correlation is one of the tools available to reduce the foreground effects because the foreground emission from such radio sources is statistically independent of LAE distribution. LAE surveys during the EoR at redshifts $z=6.6$ and $7.3$ are ongoing by the Subaru Hyper Suprime-Cam (HSC). Additionally, Prime Focus Spectrograph (PFS) will provide precise redshift information of the LAEs discovered by the HSC survey. In this paper, we investigate the detectability of the 21cm signal with the 21cm-LAE cross-correlation by using our improved reionization simulations. We also focus on the error budget and evaluate it quantitatively in order to consider a strategy to improve the signal-to-noise ratio. In addition, we explore an expansion of the LAE survey to suggest optimal survey parameters and show a potential to measure a characteristic size of ionized bubbles via the turnover scale of the cross-power spectrum. As a result, we find that the Murchison Widefield Array (MWA) has ability to detect the cross-power spectrum signal on large scales by combining LAE Deep field survey of HSC. We also show that the sensitivity is improved dramatically at small scales by adding redshift information from the PFS measurements. The Square Kilometre Array (SKA) has a potential to measure the turnover scale with an accuracy of $6times10^{-3}~{rm Mpc^{-1}}$.
Heating of neutral gas by energetic sources is crucial for the prediction of the 21 cm signal during the epoch of reionization (EoR). To investigate differences induced on statistics of the 21 cm signal by various source types, we use five radiative transfer simulations which have the same stellar UV emission model and varying combinations of more energetic sources, such as X-ray binaries (XRBs), accreting nuclear black holes (BHs) and hot interstellar medium emission (ISM). We find that the efficient heating from the ISM increases the average global 21~cm signal, while reducing its fluctuations and thus power spectrum. A clear impact is also observed in the bispectrum in terms of scale and timing of the transition between a positive and a negative value. The impact of XRBs is similar to that of the ISM, although it is delayed in time and reduced in intensity because of the less efficient heating. Due to the paucity of nuclear BHs, the behaviour of the 21~cm statistics in their presence is very similar to that of a case when only stars are considered, with the exception of the latest stages of reionization, when the effect of BHs is clearly visible. We find that differences between the source scenarios investigated here are larger than the instrumental noise of SKA1-low at $z gtrsim 7-8$, suggesting that in the future it might be possible to constrain the spectral energy distribution of the sources contributing to the reionization process.
Detecting $rm H_I$ 21cm line in the intergalactic medium (IGM) during the Epoch of Reionization (EoR) suffers from foreground contamination such as Galactic synchrotron and extragalactic radio sources. Cross-correlation between the 21cm line and Lyman-$alpha$ emitter (LAE) galaxies is a powerful tool to identify the 21cm signal since the 21cm line emission has correlation with LAEs while the LAEs are statistically independent of the foregrounds. So far, the detectability of 21cm-LAE cross-power spectrum has been investigated with simple LAE models where the observed Ly$alpha$ luminosity is proportional to the dark matter halo mass. However, the previous models were inconsistent with the latest observational data of LAEs obtained with Subaru/Hyper Suprime-Cam (HSC). Here, we revisit the detectability of 21cm-LAE cross-power spectrum adopting a state-of-the-art LAE model consistent with all Subaru/HSC observations such as the Ly$alpha$ luminosity function, LAE angular auto-correlation, and the LAE fractions in the continuum selected galaxies. We find that resultant cross-power spectrum with the updated LAE model is reduced at small scales ($ksim 1 rm Mpc^{-1}$) compared to the simple models, while the amplitudes at large scales ($k lesssim 0.2 rm Mpc^{-1}$) are not affected so much. We conclude that the large-scale signal would be detectable with Square Kilometre Array (SKA) and HSC LAE cross-correlation but detecting the small scale signal would require an extended HSC LAE survey with an area of $sim 75 rm deg^2$ or 3000 hrs observation time of 21cm line with SKA.