CERN Accelerating science

Scientific objectives, required observations, and instruments of GREX-PLUS.

Comparisons of survey areas and depths in the wavelength 3--5 $\mu$m bands (left) and 5--8 $\mu$m bands (right). The bluish (orange-ish) colors are $3$ (4) $\mu$m bands in the right panel, while the green-ish (magenta-ish) colors are $\sim5$ (7) $\mu$m bands in the left panel. The five-pointed-stars are the GREX-PLUS survey parameters. The circles, diamonds, pentagons, and squares are Spitzer, AKARI, WISE, and JWST surveys, respectively.

Comparisons of survey areas and depths in the wavelength 3--5 $\mu$m bands (left) and 5--8 $\mu$m bands (right). The bluish (orange-ish) colors are $3$ (4) $\mu$m bands in the right panel, while the green-ish (magenta-ish) colors are $\sim5$ (7) $\mu$m bands in the left panel. The five-pointed-stars are the GREX-PLUS survey parameters. The circles, diamonds, pentagons, and squares are Spitzer, AKARI, WISE, and JWST surveys, respectively.

A comparison of the line sensitivity (1 hr exposure, $5\sigma$) for unresolved narrow line emission between GREX-PLUS high resolution spectrometer (green curve) and a typical one on a ground-based 30 m-class telescope (blue curves). The gap between the wavelengths of 13.5 $\mu$m and 17 $\mu$m of the ground-based case is due to the very low transmission of the Earth's atmosphere.

Structure formation in the $\Lambda$CDM cosmology and the scientific focus of GREX-PLUS.

Examples of rest-frame UV luminosity functions of galaxies at $z=4-13$ from observations \citep[][left]{Harikane22} and cosmological hydrodynamic simulations \citep[][right]{Jaacks12b}.

Examples of rest-frame UV luminosity functions of galaxies at $z=4-13$ from observations \citep[][left]{Harikane22} and cosmological hydrodynamic simulations \citep[][right]{Jaacks12b}.

Volume-averaged neutral fraction as a function of redshift from \citet{Goto21}. See the main text for more details on the lines.

Thomson optical depth as a function of redshift from \citet{Naidu20}.

Zoom-in cosmological hydrodynamic simulation of high-redshift galaxies, resolving individual superbubble with different escape fractions \citep{Ma20}. The left panel shows the large-scale gaseous filament. The middle and right panels show the star particles on top of gas density in region-A of the left panel, color-coded by their stellar ages and $\fesc$, respectively.

THESAN cosmological radiation hydrodynamic simulation. The left box shows the hydrogen neutral fraction, and the right box shows the ionizing radiation field.

Comparison of SHMR from both simulations and observations \citep{Kannan22}. The blue line with a shade is from the abundance matching method \citep{Behroozi19}, and the red data points with error bars are the observational estimates of \citet{Stefanon21}.

Average Ly$\alpha$ transmission as a function of distance from galaxies in the THESAN-1 simulation from \citet{Garaldi22} at $z=5.5$.

Rest-frame UV luminosity functions at $z\sim10$ and $z\sim13$ \citep{Harikane22}. The red circle shows the number density of $z\sim13$ galaxy candidates. The black symbols and the gray shaded region are measurements at $z\sim10$ from the literature (diamond: \citealt{2016MNRAS.459.3812M}, square: \citealt{2018ApJ...855..105O}, pentagon: \citealt{2018ApJ...867..150M}, circle: \citealt{2020MNRAS.493.2059B}, shade: \citealt{2022ApJ...928...52F}). The green star is the number density of GN-z11. Note that the data point of \citet{2020MNRAS.493.2059B} (GN-z11) is horizontally (vertically) offset by $-0.2$ mag (+0.03 dex) for clarity. The gray dashed line is the Schechter function fit \citep{2016ApJ...830...67B}, whereas the gray and red solid lines are the double power-law functions at $z\sim10$ and $13$, respectively, whose parameters are determined by the extrapolation from lower redshifts in \citet{2020MNRAS.493.2059B}.

A summary of JWST first year survey results \citep{2023ApJS..265....5H}. The left panel shows the absolute ultraviolet magnitude of galaxies found with HST (gray) and with JWST (red). HD1 and HD2 are galaxy candidates found with ground-based near-infrared images. The black, blue, and green lines are the 5-$\sigma$ limiting magnitudes of the three GREX-PLUS imaging surveys: Deep, Medium, and Wide, respectively. The right two panels show the luminosity function of galaxies at $z\sim12$ and $\sim16$.

The expected numbers of galaxies detected in the GREX-PLUS imaging surveys (left table). In the right four panels, the red solid and dashed lines show the ultraviolet luminosity functions based on the JWST first year results and their extrapolation to $z\sim20$ (H23: \citealt{2023ApJS..265....5H} and D23: \citealt{2023MNRAS.518.6011D}). The vertical and horizontal lines in the panels show the imaging survey sensitivity and area (or survey volume), respectively. The gray lines are on-going JWST surveys, the blue lines are expected Roman surveys, and the green lines are the three GREX-PLUS surveys.

Limiting stellar mass of galaxies as a function of redshift with GREX-PLUS. If we reach 26--27 AB magnitudes at F232 (2.3$\mu$m), F303 (3.0$\mu$m), and F397 (4.0$\mu$m) as specified in the figure, we will be able to trace the stellar mass assembly history back to $z\sim8$ and down to the building blocks of 10$^9$ M$_{\odot}$ or 1/100 of our Milky Way galaxy.

\textit{Left:} Expected peak brightness of pair-instability supernovae (PISN) and superluminous supernovae (SLSN) at high redshifts in the F232 (around $2~\mathrm{\mu m}$) and F397 (around $4~\mathrm{\mu m}$) bands. For pair-instability supernovae, we show two brightest models (R250 and R225). By conducting supernova surveys that reach deeper than 26~AB~mag per epoch, we can discover supernovae that are more distant ($z>7$) than those can be found by Roman. \textit{Right:} Supernova color-magnitude diagram with the F397 and F232 bands. We show pair-instability supernovae at $z>6$, superluminous supernovae at $z>6$, Type~Ia supernovae at $z>1$ \citep{2007ApJ...663.1187H}, and Type~II supernovae at $z>1$ (the Nugent template from https://c3.lbl.gov/nugent/nugent\_templates.html). We can find that the high-redshift supernovae can be separated from low-redshift ones through color.

\textit{Left:} Expected peak brightness of pair-instability supernovae (PISN) and superluminous supernovae (SLSN) at high redshifts in the F232 (around $2~\mathrm{\mu m}$) and F397 (around $4~\mathrm{\mu m}$) bands. For pair-instability supernovae, we show two brightest models (R250 and R225). By conducting supernova surveys that reach deeper than 26~AB~mag per epoch, we can discover supernovae that are more distant ($z>7$) than those can be found by Roman. \textit{Right:} Supernova color-magnitude diagram with the F397 and F232 bands. We show pair-instability supernovae at $z>6$, superluminous supernovae at $z>6$, Type~Ia supernovae at $z>1$ \citep{2007ApJ...663.1187H}, and Type~II supernovae at $z>1$ (the Nugent template from https://c3.lbl.gov/nugent/nugent\_templates.html). We can find that the high-redshift supernovae can be separated from low-redshift ones through color.

\textit{Left:} A summary of background radiation observations. The solid black line and surrounding data show galaxy-integrated light. The gray shaded band is the upper limit to the background radiation from high-energy gamma-ray observations. Data points are the results of direct measurements of the background radiation and are several times higher than the galaxy-integrated light and the upper limit from the gamma-ray observations (adapted from \citealt{2018RvMP...90b5006K}).

A comparison of contributions from direct observations of cosmic infrared background (CIB), zodiacal light, Galactic interstellar dust emission, integrated galaxy light, and integrated radiation from first stars and primordial blackholes. The expected sensitivity of the five bands of the GREX-PLUS imaging surveys is shown by the vertical bands. The lower and upper ends of the bands are sensitivities of the Deep and Wide surveys, respectively (see Table~\ref{tab:GPsurveys}). This is the sensitivity when aiming to detect correlations on the smallest angular scale. For large angle fluctuations and absolute intensity measurements, the sensitivity can be greatly improved by combining many pixels.

Expected surface density of quasars (per 1000 deg$^2$) for several cases of lower redshift cut, as a function of limiting magnitude. Four missions are considered here, i.e., Hyper Suprime-Cam (HSC) SSP survey (thin lines), {\it Euclid} (dashed lines; overlapping with other lines at $z > 7 - 10$), {\it Roman Space Telescope} (medium lines), and GREX-PLUS (thick lines). The light-blue shading represents the surface densities providing $>$1 quasars in the whole sky; in other words, no quasar is expected on average in the lower part of this plot.

Star formation histories inferred from optical spectroscopic observations of nearby early-type galaxies \citep{2010MNRAS.404.1775T}. The results suggest that massive galaxies with a dynamical mass of $\log(M_\mathrm{dyn}/{\rm M}_\odot)\sim12$ experienced a short burst of star formation at $z\sim4$. The figure is taken from Figure 9 of \cite{2010MNRAS.404.1775T}; 'Environment and self-regulation in galaxy formation'.

Examples of multi-wavelength images ($g$, $r$, $i$, $z$, $y$, $J$, $H$, ch1, ch2, N2, N3, N4, S7, S9W, S11, L15, L18W, and L24, from top left to bottom right) for optically dark IR galaxies reported in \cite{2020ApJ...899...35T}. R.A. and decl. are relative coordinates with respect to objects in the AKARI NEP-wide catalog \citep{2012A&A...548A..29K}. White circles in the images also correspond to the coordinates of the AKARI NEP-wide catalog.

Typical SEDs of optically dark IR galaxies found in the AKARI NEP \citep{2020ApJ...899...35T} as a function of redshift up to $z = 4$. The detection limits of ongoing and forthcoming missions are over-plotted.

Left: schematic of the observation of an AGN torus region using near-infrared absorption lines. The thermal emission from the dust sublimation layer is the dominant continuum emission source over the galaxy, and by spectroscopy of absorption lines to it, an effectively high spatial resolution can be achieved. Right: example of a CO vibrational rotational absorption spectrum (column density $N(\mathrm{CO})=10^{18.5}\,\mathrm{cm^{-2}}$, temperature 600\,K, velocity width 50\,km\,s$^{-1}$). Lines of different rotational levels appear at intervals of $\sim$0.01\,$\mu$m around the rest wavelength of 4.67\,$\mu$m.

Left: expected spectrum of the outflow described in this text observed from the same line of sight as the propagation direction with S/N=2.7 for the continuum level. Right: velocity profiles for the lines of different rotational levels $J$ averaged for each low and high excitation regime. The binning is applied with a velocity width corresponding to the required resolution (30\,km\,s$^{-1}$), rather than the nominal wavelength resolution of the high-dispersion spectrometer.

Voigt profile fits (blue curves) to the 4.7~$\mu$m CO bands at $z_{\rm abs}$ = 2.4185 toward a quasar SDSS J143912.04+111740.5 (black histograms). The vertical lines with numbers are the locations of different CO transitions from different $J$ levels. The figure is taken from \citet{2008A&A...482L..39S}.

Synthesized spectrum around the 4.7~$\mu$m CO ro-vib bands at $z$ = 2.2. We assume log($N_{\rm CO}$/cm$^{-2}$) = 16, T$_{\rm kin}$ = 100~K, and n(H$_2$) = 18 cm$^{-3}$, following \citet{2008A&A...482L..39S}, \citet{2019MNRAS.490.2668B}, and \citet{2019A&A...625L...9G}. Blue curve denotes the intrinsic spectrum with thermally broadened line width, while green and red curves are those after applying convolution with $R$ = 30,000 and adding noise to produce S/N = 20 pixel$^{-1}$ spectrum. Both R and P branches are clearly detected. We iterate model fits to the synthesized spectrum using four free parameters (T$_{\rm CMBR}$, T$_{\rm kin}$, n(H$_2$), and log$N_{\rm CO}$) after changing a seed for noise. The accuracy of T$_{\rm CMBR}$ measurement ($\delta$T) is very dependent on S/N ratio; $\delta$T $\sim$ 1~K or 0.1~K if S/N = 20 or 1000 pixel$^{-1}$.

\textbf{(a:)} A modeled spectrum of low-mass, metal-poor galaxy at $z=5$ (gray) along with the available filters of Roman Space Telescope (up to $2\,\mu$m) and GREX-PLUS (from $2$ to $8\,\mu$m based on the current plan). The infrared bands allow us to perform systematic searches of low-mass galaxies over $z=2-6$ by capturing the characteristic intense emission lines such as H$\alpha$ imprinted on the broadband photometric colors as shown in Panel (b). Combining with Lyman-break / Ly$\alpha$ provided by Roman, such a low-mass galaxy search can be extended to $z=6-8$. \textbf{(b:)} Simulated distributions of low-mass, metal-poor galaxies at $z=4.3-5.6$ (green) on the color-color diagrams using the GREX-PLUS bands, ensuring their efficient selection and discrimination from the other populations of evolved / different redshift galaxies (blue), QSOs (orange), and galactic stars (yellow) solely using the photometric data.

Theoretically-predicted infrared fluxes and colors of YSOs at the distance of the LMC (50 kpc), based on \citet{Rob07}. Plot symbols are color-coded depending on YSO's mass; high-mass ($>$10 M$_{\odot}$, blue), intermediate-mass (2 -- 10 M$_{\odot}$, green), and low-mass ($<$2 M$_{\odot}$, red). The solid orange lines indicate the expected spectroscopic (upper) and imaging (lower) sensitivities of GREX-PLUS. The spectroscopic sensitivities of the AKARI LSLMC survey \citep{ST13} and JWST/NIRSpec/IFU (10 minutes on source) are also shown.

Spectral energy distributions of various objects related to star- and planet-formation scaled at the distance of the LMC; (blue) embedded high-mass LMC YSO (ST16), (yellow) embedded low-mass Galactic YSO (Elias 29), (green) Herbig Ae/Be star (AB Aur) as an intermediate-mass protoplanetary disk source, (brown) T tauri star (AA Tau) as a low-mass protoplanetary disk source. The wavelength coverages and sensitivities of selected infrared telescopes are shown by the horizontal lines (solid: imaging, dotted: spectroscopy), where those of GREX-PLUS are shown in red.

Schematic illustration of two different modes of volatile delivery to terrestrial planets. (a) Volatile delivery by planetesimals before and/or after the dissipation of protoplanetary disk gas. (b) Volatile delivery by pebbles following the migration of snow lines before the dissipation of protoplanetary disk gas.

({\it Left}) Model calculations of the distribution of gas-phase water in a protoplanetary disk and their emission line spectra \citep{Notsu+2017}. The 63.37 $\mu$m emission lines detected by Herschel mainly trace the hot surface layer of the outer disk, whereas the 17.75 $\mu$m emission lines, observable by GREX-PLUS, trace the position of the water snowline. ({\it Right}) By analysing emission profiles that trace the Doppler shift due to the Keplerian rotation, the emission region, i.e. the water snowline, can be located using high spectral resolution of R=29,000 even if it cannot be spatially resolved \citep{Kamp+2021}.

({\it Left}) Model calculations of the distribution of gas-phase water in a protoplanetary disk and their emission line spectra \citep{Notsu+2017}. The 63.37 $\mu$m emission lines detected by Herschel mainly trace the hot surface layer of the outer disk, whereas the 17.75 $\mu$m emission lines, observable by GREX-PLUS, trace the position of the water snowline. ({\it Right}) By analysing emission profiles that trace the Doppler shift due to the Keplerian rotation, the emission region, i.e. the water snowline, can be located using high spectral resolution of R=29,000 even if it cannot be spatially resolved \citep{Kamp+2021}.

The infrared absorption spectrum of benzene (C$_6$H$_6$) and polyynes (C$_4$H$_2$, C$_6$H$_2$) detected for the first time, by the SWS (Short Wavelength Spectrometer) of ISO (Infrared Space Observatory) with $R\sim2,000$ toward the proto-planetary nebula CRL618 \citep{2001ApJ...546L.123C}.

Simulated IR high-resolution spectrum of benzene (C$_6$H$_6$) $\nu$$_4$ vibration band in the case of the column density $\it{N}$$_{col}$=5$\times$10$^{15}$cm$^{-2}$ and $\it{T}$$_{ex}$=200K, with the wavelength resolution $\it{R}$=2,000 ($\it{left}$) and 28,000 ($\it{right}$) \citep{2017GradThesis.Sci.Nagoya-U.}.

Molecular structures of aromatic hydrocarbons recently detected in the cold dark cloud TMC-1 (\citealp{2021A&A...649L..15C}, \citealp{2021A&A...652L...9C}).

Infrared emission spectra of C$_{60}$ and C$_{70}$ toward the young planetary nebula Tc-1, continuum subtracted spectrum between 5 and 23 ${\mu}$m, by the Spitzer IRS \citep{2010Sci...329.1180C}. The red and blue curves below the data are thermal emission models for all infrared active bands of C$_{60}$ and C$_{70}$ at temperatures of 330 K and 180 K, respectively.

A chemical reaction network for the formation of organic molecules from simple molecules in an interstellar molecular cloud at low temperatures ($T\sim10K$) (\citealp{1992ChRv..92.1473},\citealp{2013ChRv..113.8710A}).

The results of the cross-correlation between the mock and template spectra for the case of 20 transit observations of a mini-Neptune planet GJ 1214b by GREX-PLUS.

Top panel: model thermal emission spectrum of a Jupiter-like planet with equilibrium temperature of 500~K. Middle panel: pressure levels at $\tau =1$. Bottom panel: the result of cross-correlation analysis between the mock spectra and the model spectra, where mock data assumes a $\sim $500~K Jupiter-size planet around a Solar-type star at 20~parsec and that the 1~day observation is performed both at two quadrature phases. Two vertical dashed lines represents the radial velocities at two quadrature phases.

Probability distribution of planetary radial velocity at two quadrature phases based on a mock observation of a $\sim $500~K Jupiter-sized planet around a solar-type star at 20~pc, where each quadrature phase is observed for 1~day.

Top panel: modeled temperature profile of GJ1214b with and without haze layer based on Ohno \& Fortney in prep; the presence of haze layer of GJ1214b is implied from observed near-infrared transmission spectrum \citep{Kreidberg+14}. The haze layer would cause thermal inversion in the upper atmosphere. Bottom panel: cross-correlation coefficient between the mock data of GJ1214b and two modeled spectra with and without haze layer (black and red lines in the top panel). The mock data assumes the atmospheric profile with haze (black line in the top panel) and observations for 2.5 days at each quadrature phase. Two vertical dashed lines represents the radial velocities at two quadrature phases. While the model with haze layer has signals at the correct radial velocities with SNR $>$ 5, the model without haze layer has less significant signals.

An example of infrared spectra of Titan atmosphere observed by Cassini/CIRS. Spectral resolution of CIRS is $\sim$0.5~cm$^{-1}$.

Left: Simulated infrared spectrum of Uranus's atmosphere. A close-up to H$_2$ S(2) is shown in the upper right panel. A spectral resolving power of $\sim$30,000 is assumed. Right: Temperature weighting functions per some specific wavelengths.

: The plot of spectral fraction of H$_2$O ice \citep{2012AJ....143..146B} vs. body size for TNOs (including Charon; red circles), Centaurs (blue squares), and Haumea family members (green triangles).

: Near-infrared reflective spectrum of crystalline H$_2$O ice at 60~K \citep{2009ApJ...701.1347M}. The shaded areas represent wavelength ranges of the planned filters for GREX-PLUS.

Evolution of protoplanetary disks---fraction of stars with NIR disk excess as a function of age. Figure~5 (left panel) of \citet{Yasui2014}, ``Rapid evolution of the innermost dust disc of protoplanetary discs surrounding intermediate-mass stars''.

A variety of molecular emission lines toward a low-mass young stars observed using Spitzer Space Telescope (Figure courtesy: NASA/JPL/J. Carr). These are considered to be associated with an unresolved circumstellar disk at a few au scales \citep[e.g.,][]{Pontoppidan10}.

Imaginary picture of sub-structures of the Galaxy