We have initiated a new survey for local extremely metal-poor galaxies (EMPGs) with Subaru/Hyper Suprime-Cam (HSC) large-area (~500 deg^2) optical images reaching a 5 sigma limit of ~26 magnitude, about 100 times deeper than the Sloan Digital Sky Survey (SDSS). To select Z/Z_sun<0.1 EMPGs from ~40 million sources detected in the Subaru images, we first develop a machine-learning (ML) classifier based on a deep neural network algorithm with a training data set consisting of optical photometry of galaxy, star, and QSO models. We test our ML classifier with SDSS objects having spectroscopic metallicity measurements, and confirm that our ML classifier accomplishes 86%-completeness and 46%-purity EMPG classifications with photometric data. Applying our ML classifier to the photometric data of the Subaru sources as well as faint SDSS objects with no spectroscopic data, we obtain 27 and 86 EMPG candidates from the Subaru and SDSS photometric data, respectively. We conduct optical follow-up spectroscopy for 10 out of our EMPG candidates with Magellan/LDSS-3+MagE, Keck/DEIMOS, and Subaru/FOCAS, and find that the 10 EMPG candidates are star-forming galaxies at z=0.007-0.03 with large H_beta equivalent widths of 104-265 A, stellar masses of log(M*/M_sun)=5.0-7.1, and high specific star-formation rates of ~300 Gyr^{-1}, which are similar to those of early galaxies at z>6 reported recently. We spectroscopically confirm that 3 out of 10 candidates are truly EMPGs with Z/Z_sun<0.1, one of which is HSC J1631+4426, the most metal-poor galaxy with Z/Z_sun=0.016 reported ever.

We present element abundance ratios and ionizing radiation of local young low-mass (~$10^{6}$ M_sun) extremely metal poor galaxies (EMPGs) with a 2% solar oxygen abundance (O/H)_sun and a high specific star-formation rate (sSFR~300 Gyr$^{-1}$), and other (extremely) metal poor galaxies, which are compiled from Extremely Metal-Poor Representatives Explored by the Subaru Survey (EMPRESS) and the literature. Weak emission lines such as [FeIII]4658 and HeII4686 are detected in very deep optical spectra of the EMPGs taken with 8m-class telescopes including Keck and Subaru (Kojima et al. 2019, Izotov et al. 2018), enabling us to derive element abundance ratios with photoionization models. We find that neon- and argon-to-oxygen ratios are comparable to those of known local dwarf galaxies, and that the nitrogen-to-oxygen abundance ratios (N/O) are lower than 20% (N/O)_sun consistent with the low oxygen abundance. However, the iron-to-oxygen abundance ratios (Fe/O) of the EMPGs are generally high; the EMPGs with the 2%-solar oxygen abundance show high Fe/O ratios of ~90-140% (Fe/O)_sun, which are unlikely explained by suggested scenarios of Type Ia supernova iron productions, iron's dust depletion, and metal-poor gas inflow onto previously metal-riched galaxies with solar abundances. Moreover, these EMPGs have very high HeII4686/H$\beta$ ratios of ~1/40, which are not reproduced by existing models of high-mass X-ray binaries whose progenitor stellar masses are less than 120 M_sun. Comparing stellar-nucleosynthesis and photoionization models with a comprehensive sample of EMPGs identified by this and previous EMPG studies, we propose that both the high Fe/O ratios and the high HeII4686/H$\beta$ ratios are explained by the past existence of super massive ($>$300 M_sun) stars, which may evolve into intermediate-mass black holes ($\gtrsim$100 M_sun).

We present an IGM HI tomography map in a survey volume of $16 \times 19 \times 131 \ h^{-3} {\rm comoving \ Mpc}^{3}$ (cMpc$^3$) centered at MAMMOTH-1 nebula and three neighbouring quasars at $z=2.3$. MAMMOTH-1 nebula is an enormous Ly$\alpha$ nebula (ELAN), hosted by a type-II quasar dubbed MAMMOTH1-QSO, that extends over $1\ h^{-1}$ cMpc with not fully clear physical origin. Here we investigate the HI-gas distribution around MAMMOTH1-QSO with the ELAN and three neighbouring type-I quasars, making the IGM HI tomography map with a spatial resolution of $2.6\ h^{-1}$ cMpc. Our HI tomography map is reconstructed with HI Ly$\alpha$ forest absorption of bright background objects at $z=2.4-2.9$: one eBOSS quasar and 16 Keck/LRIS galaxy spectra. We estimate the radial profile of HI flux overdensity for MAMMOTH1-QSO, and find that MAMMOTH1-QSO resides in a volume with significantly weak HI absorption. This suggests that MAMMOTH1-QSO has a proximity zone where quasar illuminates and photo-ionizes the surrounding HI gas and suppresses HI absorption, and that the ELAN is probably a photo-ionized cloud embedded in the cosmic web. The HI radial profile of MAMMOTH1-QSO is very similar to those of three neighbouring type-I quasars at $z=2.3$, which is compatible with the AGN unification model. We compare the distributions of the HI absorption and star-forming galaxies in our survey volume, and identify a spatial offset between density peaks of star-forming galaxies and HI gas. This segregation may suggest anisotropic UV background radiation created by star-forming galaxy density fluctuations.

Hyper Suprime-Cam (HSC) is a wide-field imaging camera on the prime focus of the 8.2m Subaru telescope on the summit of Maunakea in Hawaii. A team of scientists from Japan, Taiwan and Princeton University is using HSC to carry out a 300-night multi-band imaging survey of the high-latitude sky. The survey includes three layers: the Wide layer will cover 1400 deg$^2$ in five broad bands ($grizy$), with a $5\,\sigma$ point-source depth of $r \approx 26$. The Deep layer covers a total of 26~deg$^2$ in four fields, going roughly a magnitude fainter, while the UltraDeep layer goes almost a magnitude fainter still in two pointings of HSC (a total of 3.5 deg$^2$). Here we describe the instrument, the science goals of the survey, and the survey strategy and data processing. This paper serves as an introduction to a special issue of the Publications of the Astronomical Society of Japan, which includes a large number of technical and scientific papers describing results from the early phases of this survey.

We present Lya and UV-nebular emission line properties of bright Lya emitters (LAEs) at z=6-7 with a luminosity of log L_Lya/[erg s-1] = 43-44 identified in the 21-deg2 area of the SILVERRUSH early sample developed with the Subaru Hyper Suprime-Cam (HSC) survey data. Our optical spectroscopy newly confirm 21 bright LAEs with clear Lya emission, and contribute to make a spectroscopic sample of 96 LAEs at z=6-7 in SILVERRUSH. From the spectroscopic sample, we select 7 remarkable LAEs as bright as Himiko and CR7 objects, and perform deep Keck/MOSFIRE and Subaru/nuMOIRCS near-infrared spectroscopy reaching the 3sigma-flux limit of ~ 2x10^{-18} erg s-1 for the UV-nebular emission lines of He II1640, C IV1548,1550, and O III]1661,1666. Except for one tentative detection of C IV, we find no strong UV-nebular lines down to the flux limit, placing the upper limits of the rest-frame equivalent widths (EW_0) of ~2-4 A for He II, C IV, and O III] lines. Here we also investigate the VLT/X-SHOOTER spectrum of CR7 whose 6 sigma detection of He II is claimed by Sobral et al. Although two individuals and the ESO-archive service carefully re-analyze the X-SHOOTER data that are used in the study of Sobral et al., no He II signal of CR7 is detected, supportive of weak UV-nebular lines of the bright LAEs even for CR7. Spectral properties of these bright LAEs are thus clearly different from those of faint dropouts at z~7 that have strong UV-nebular lines shown in the various studies. Comparing these bright LAEs and the faint dropouts, we find anti-correlations between the UV-nebular line EW_0 and UV-continuum luminosity, which are similar to those found at z~2-3.

The Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP) is a three-layered imaging survey aimed at addressing some of the most outstanding questions in astronomy today, including the nature of dark matter and dark energy. The survey has been awarded 300 nights of observing time at the Subaru Telescope and it started in March 2014. This paper presents the first public data release of HSC-SSP. This release includes data taken in the first 1.7 years of observations (61.5 nights) and each of the Wide, Deep, and UltraDeep layers covers about 108, 26, and 4 square degrees down to depths of i~26.4, ~26.5, and ~27.0 mag, respectively (5sigma for point sources). All the layers are observed in five broad bands (grizy), and the Deep and UltraDeep layers are observed in narrow bands as well. We achieve an impressive image quality of 0.6 arcsec in the i-band in the Wide layer. We show that we achieve 1-2 per cent PSF photometry (rms) both internally and externally (against Pan-STARRS1), and ~10 mas and 40 mas internal and external astrometric accuracy, respectively. Both the calibrated images and catalogs are made available to the community through dedicated user interfaces and database servers. In addition to the pipeline products, we also provide value-added products such as photometric redshifts and a collection of public spectroscopic redshifts. Detailed descriptions of all the data can be found online. The data release website is https://hsc-release.mtk.nao.ac.jp/.

Using galaxies as background light sources to map intervening Lya absorption is a novel approach to study the interplay among galaxies, the circum-galactic medium (CGM), and the intergalactic medium (IGM). Introducing a new measure of z = 3.1$ HI Lya absorption relative to the cosmic mean, Delta_NB497, estimated from photometric data of star-forming galaxies at 3.3 < z < 3.5, we have made two-dimensional Delta_NB497 maps in the z = 3.1 SSA22 proto-cluster region and two control fields (SXDS and GOODS-N fields) with a spatial resolution of ~ 5 comoving Mpc. The Delta_NB497 measurements in the SSA22 field are systematically larger than those in the control fields, and this HI absorption enhancement extends more than 50 comoving Mpc. The field-averaged (i.e., ~50 comoving Mpc scale) Delta_NB497 and the overdensity of Lya emitters (LAEs) seem to be correlated, while there is no clear dependency of the Delta_NB497 on the local LAE overdensity in a few comoving Mpc scale. These results suggest that diffuse HI gas spreads out in/around the SSA22 proto-cluster. We have also found an enhancement of Delta_NB497 at a projected distance < 100 physical kpc from the nearest z = 3.1 galaxies at least in the SSA22 field, which is probably due to HI gas associated with the CGM of individual galaxies. The HI absorption enhancement in the CGM-scale tends to be weaker around galaxies with stronger Lya emission, which suggests that the Lya escape fraction from galaxies depends on hydrogen neutrality in the CGM.

We present spatial correlations of galaxies and IGM HI in the COSMOS/UltraVISTA 1.62 deg$^2$ field. Our data consist of 13,415 photo-$z$ galaxies at $z\sim2-3$ with $K_s<23.4$ and the Ly$\alpha$ forest absorptions in the background quasar spectra selected from SDSS data with no signature of damped Ly$\alpha$ system contamination. We estimate a galaxy overdensity $\delta_{gal}$ in an impact parameter of 2.5 pMpc, and calculate the Ly$\alpha$ forest fluctuations $\delta_{\langle F\rangle}$ whose negative values correspond to the strong Ly$\alpha$ forest absorptions. We identify weak evidence of an anti-correlation between $\delta_{gal}$ and $\delta_{\langle F\rangle}$ with a Spearman's rank correlation coefficient of $-0.39$ suggesting that the galaxy overdensities and the Ly$\alpha$ forest absorptions positively correlate in space at the $\sim90\%$ confidence level. This positive correlation indicates that high-$z$ galaxies exist around an excess of HI gas in the Ly$\alpha$ forest. We find four cosmic volumes, dubbed $A_{obs}$-$D_{obs}$, that have extremely large (small) values of $\delta_{gal} \simeq0.8$ ($-1$) and $\delta_{\langle F\rangle}$ $\simeq0.1$ ($-0.4$), three out of which, $B_{obs}$-$D_{obs}$, significantly depart from the correlation, and weaken the correlation signal. We perform cosmological hydrodynamical simulations, and compare with our observational results. Our simulations reproduce the correlation, agreeing with the observational results. Moreover, our simulations have model counterparts of $A_{obs}$-$D_{obs}$, and suggest that the observations pinpoint, by chance, a galaxy overdensity like a proto-cluster, gas filaments lying on the sightline, a large void, and orthogonal low-density filaments. Our simulations indicate that the significant departures of $B_{obs}$-$D_{obs}$ are produced by the filamentary large-scale structures and the observation sightline effects.