EIGER IV. The Cool 104 K Circumgalactic Environment of High-redshift Galaxies Reveals Remarkably Efficient Intergalactic Medium Enrichment

We report new observations of the cool diffuse gas around 29, 2.3 < z < 6.3 galaxies using deep JWST/NIRCam slitless grism spectroscopy around the sight line to the quasar J0100+2802. The galaxies span a stellar mass range of 7.1≤logM∗/M⊙≤10.7 , and star formation rates (SFRs) of −0.1<log SFR/M ⊙ yr−1 < 2.3. We find galaxies for seven Mg ii absorption systems within 300 kpc of the quasar sight line. The Mg ii radial absorption profile falls off sharply with radius, with most of the absorption extending out to 2–3 R 200 of the host galaxies. Six out of seven Mg ii absorption systems are detected around galaxies with logM∗/M⊙> 9. The Mg ii absorption kinematics are shifted from the systemic redshifts of host galaxies with a median absolute velocity ≈ 135 km s−1 and standard deviation ≈ 85 km s−1. The high kinematic offset and large radial separation (R > 1.3 R 200), suggest that five out of the seven Mg ii absorption systems are not gravitationally bound to their host galaxy. In contrast, most of the cool circumgalactic medium at z < 1 is gravitationally bound. The high incidence of unbound Mg ii gas in this work suggests that toward the end of reionization, galaxy halos are in a state of remarkable disequilibrium, and are highly efficient in enriching the intergalactic medium. The two strongest Mg ii absorption systems are detected at z ∼ 4.22 and 4.5, the former associated with a merging galaxy system and the latter associated with three kinematically close galaxies. Both of these galaxies reside in local galaxy overdensities, indicating the presence of cool Mg ii absorption in two “protogroups” at z > 4.


Introduction
The commissioning of the JWST has ushered in a new era for spectroscopy of galaxies and intergalactic matter near the Epoch of Reionization (EoR; Rigby et al. 2023).Strong restframe optical lines (e.g., Hα and [O III]) are finally observable at z  3.5; which combined with JWST's efficient spectroscopic modes have enabled large-scale spectroscopic surveys of galaxies at the EoR (Kashino et al. 2023a;Matthee et al. 2023b;Oesch et al. 2023;Wang et al. 2023).By performing carefully constructed spectroscopic experiments, where the galaxy fields also have bright high-z quasars, one can extend the study of galaxies to characterize their gaseous halos (Kashino et al. 2023a).Deep ground-based near-infrared (NIR) spectroscopy of these quasars often reveal intervening metal absorption line systems associated with galaxies along the line of sight: a signature of a diffuse baryonic reservoir of gas around galaxies (e.g., Cooper et al. 2019).These cosmic ecosystems fuel the growth of stellar mass in galaxies and serve as reservoirs of gas recycling.At z < 2, such cosmic ecosystems have been successfully characterized as the circumgalactic medium (CGM) around galaxies (Tumlinson et al. 2017).Over the last two decades, large galaxy and quasar surveys have enabled detailed characterization of the CGM, establishing it as an ubiquitous reservoir of diffuse gas around galaxies (Chen et al. 2001(Chen et al. , 2010;;Bordoloi et al. 2011;Nielsen et al. 2013;Tumlinson et al. 2013;Zhu et al. 2014;Burchett et al. 2016;Huang et al. 2016;Johnson et al. 2017).
For data spanning the last 7 Gyr of the history of the Universe (z < 1), comparison of absorption line systems observed in background spectra of bright background quasars or galaxies with their host galaxy populations have revolutionized our understanding of the late-time CGM and its role in galaxy formation (see Tumlinson et al. 2017 for a detailed review).These studies have revealed that both highly ionized metals (traced by O VI and C IV) and low ionized metals (traced by Mg II) show strong trends with increasing galaxy star formation rates (SFRs) and stellar masses (Chen et al. 2001(Chen et al. , 2010;;Bordoloi et al. 2011;Prochaska et al. 2011;Tumlinson et al. 2011;Nielsen et al. 2013;Liang & Chen 2014;Lan & Mo 2018).Cool circumgalactic gas traced by Mg II shows strong dependence on morphology and the orientation of outflows versus inflows (Bouché et al. 2012;Kacprzak et al. 2012;Bordoloi et al. 2011Bordoloi et al. , 2014aBordoloi et al. , 2014b;;Rubin et al. 2014;Martin et al. 2019;Lundgren et al. 2021).A diffuse warm/hot intragroup medium is also detected around groups of galaxies (Bordoloi et al. 2011;Johnson et al. 2015;Chen et al. 2020;Dutta et al. 2021;McCabe et al. 2021).CGM gas appears Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
bimodal in metallicity, with a portion tracing enriched outflows (∼20%-100% solar) and a distinct metal-poor component (∼5% solar) that may resemble the long sought "cold accretion" entering galaxies from the IGM (Lehner et al. 2013;Wotta et al. 2019).Most crucially, a mass and metal census of the CGM gas at z ∼ 0.2 suggests that the CGM may host a large share of galactic baryons, with the CGM mass content outweighing the total stellar mass content of galaxies (Stocke et al. 2013;Werk et al. 2014;Prochaska et al. 2017), and hosts a massive reservoir of galactic metals, with galaxies having ejected at least as much metal mass as they have retained (Tumlinson et al. 2011;Bordoloi et al. 2014a;Peeples et al. 2014).At these redshifts, most of the CGM gas is bound to the dark matter halo of its host galaxy, suggesting that most of the gas will be recycled back to the interstellar medium (Tumlinson et al. 2013;Bordoloi et al. 2014aBordoloi et al. , 2018;;Ford et al. 2014;Huang et al. 2016).
At z ∼ 2, the CGM contains a large metal reservoir, but more of this gas is kinematically consistent with not being bound to the host galaxy's dark matter halo than its z < 1 counterparts (Rudie et al. 2019).Observations of lensed QSOs and spatially extended lensed arcs reveal that individual CGM gas clouds are small and show large variations even within a single halo (Rauch et al. 1999;Lopez et al. 2018;Rubin et al. 2018;Bordoloi et al. 2022b).A consensus has emerged that the CGM is a ubiquitous feature of galaxies from z ∼ 3 to z ∼ 0 (Rudie et al. 2012;Tumlinson et al. 2017;Péroux & Howk 2020;Lehner et al. 2022).
Within the first gigayear of cosmic history, the rest-frame UV transitions that are instrumental in characterizing the low-z CGM are redshifted into the NIR.Moreover, any transition blueward of Lyα will not be observable owing to the intervening neutral IGM.Therefore, heavy element absorption systems in high-z quasar spectra provide our only access to the chemical enrichment and ionization taking place in this environment.NIR spectroscopy of QSOs at z > 6 (e.g., Becker et al. 2001;Simcoe et al. 2012;Bañados et al. 2018;Cooper et al. 2019;Yang et al. 2020;Davies et al. 2023) has pushed these investigations within the first gigayear of the Big Bang.These studies find that the number density evolution of strong Mg II absorption systems traces the cosmic star formation history of the Universe whereas weaker systems show no evolution out to z ∼ 6 (Matejek & Simcoe 2012).The incidence of detection of low ionization species (e.g., C II and O I) remain high even at high z, whereas the incidence of detection of high ionized species (e.g., C IV and Si IV) drop off sharply beyond z ∼ 5.7 (Becker et al. 2019;Cooper et al. 2019).This might be owing to changes in the UV background in the early Universe or reflect some fundamental change in galaxy properties.Therefore, identifying the galaxies associated with these absorbers at high z might give crucial insight into the latter stages of the EoR.
Prior to JWST, it has been prohibitive to conduct detailed galaxy surveys to identify the host galaxies of these absorbers.But initial studies of the detection of host galaxies suggest a strong correlation between the absorption lines detected at high z with host galaxies.
Recently two works reported the presence of galaxy overdensities near a strong C IV absorption system at z ∼ 5.72, four Lyα emitting (LAE) galaxies (Díaz et al. 2021) and two [C II]158 μm emitting galaxies (Kashino et al. 2023b).Additionally, another host galaxy associated with a z = 5.9 O I absorption system using the [C II]158 μm line was reported (Wu et al. 2021).Cross-correlation of LAE galaxies with deep MUSE observations also suggests a link between strong C IV absorbers and bright LAE galaxies out to z ∼ 4 (Galbiati et al. 2023).
These promising early results suggest that a systematic multiwavelength galaxy survey is warranted to study the CGM of host galaxies at z > 4, where detailed galaxy properties can be studied.
In this work we present the first CGM measurements traced by Mg II absorption around 29 (2 < z < 6) galaxies in the Emission-line Galaxies and Intergalactic Gas in the Epoch of Reioniation (EIGER) survey (Kashino et al. 2023a).We focus on the first observations around the hyperluminous z = 6.33 quasar J0100+2802.
This paper is organized as follows.In Section 2 we describe the observations and summarize the survey strategy.In Section 3 we describe the measurements of galaxy properties and the CGM absorber properties.In Section 4 we describe the results.In Section 5 we present the summary and discussion of the results.Throughout this paper we follow a flat Lambda cold dark matter (ΛCDM) cosmology with H 0 = 67.7 km s −1 Mpc −1 , Ω M = 0.31, and Ω Λ = 0.69 (Planck Collaboration et al. 2020).All magnitudes are listed in the AB system.Unless stated otherwise, all distances are quoted in units of physical kiloparsecs.

The EIGER Survey
The EIGER survey is a 126.5 hr JWST Guaranteed Time Observations (GTO) program (PID: 1243, PI: S. J. Lilly) that performs NIRCam wide field slitless spectroscopy (WFSS) around six extragalactic fields, each centered on a hyperluminous (6  z  7) quasar.We refer the reader to Kashino et al. (2023a) for a detailed description of the survey design rationale and data reduction methods.Below we briefly summarize different aspects of the observations.

NIRCam Observations of J0100+2802
In this work, we focus on metal absorption in the vicinity of galaxies detected in the z = 6.33 quasar field of J0100+2802.JWST/NIRCam WFSS spectroscopy is performed with the F356W filter using the reverse grisms, which both disperse the spectra horizontally on the NIRCam sensors but with opposite parity.The spectral resolution of the observations is R ∼ 1500.Simultaneously with spectroscopy, F115W and F200W imaging of the field is performed.Direct and out-of-field imaging in the F356W filter is performed to cover the full spectroscopic field of view (Kashino et al. 2023a).
A four pointing mosaic strategy is employed, which ensures that the central 40″ × 40″ has a maximum depth of ∼35 ks.The total spectroscopic field of view around J0100+2802 is ∼25.9 arcmin 2 , and the total exposure time ranges from 8 to 35 ks.In this work we focus only on the central ≈4.6 arcmin 2 of the field, which is covered by NIRCam modules A and B (with reversed dispersion directions).This enables us to identify single emission line objects accurately and measure their redshifts (Matthee et al. 2023a).
The NIRCam imaging data are reduced as described in Kashino et al. (2023a) using the jwst pipeline (v1.8.2; Bushouse et al. 2022a).Additional postprocessing steps are performed to mask strong cosmic-ray hits following Merlin et al. (2022).Astrometry is calibrated by aligning known stars from the Gaia Data Release 2 catalog (Gaia Collaboration et al. 2018).Several known artifacts (e.g., stray-light features, 1/f noise, and residual sky background) are subtracted (Kashino et al. 2023a) before obtaining a final coadded image with a pixel size of 0 03 pixel −1 .
These final coadded images are used to perform aperturematched photometry with SExtractor in dual mode.The F356W image is used as the detection image.All images are convolved to match the point-spread function of the F356W image.Kron aperture magnitudes are measured and photometric uncertainties are estimated by measuring random blank sky variations for apertures of different sizes, scaled to the local variance propagated by the pipeline (see Kashino et al. 2023a).
NIRCam WFSS data reduction is performed using a combination of the jwst pipeline (v1.7.0;Bushouse et al. 2022b) and custom in-house tools as described in detail in Kashino et al. 2023a andMatthee et al. (2023a).To summarize, each individual exposure is processed with the Detector1 step in the jwst pipeline and assigned a world coordinate system value with the Spec2 step.The frames are flat fielded and additional 1/f noise and sky background variation are removed by subtracting the median flux of each column to create the science frames.A continuum map is created by using a running median filter along the dispersion direction.The median filter kernel size is adaptive and has a hole in the center to ensure that it does not oversubtract the emission lines.This continuum map is subtracted from each science frame to create an emission line map for each exposure.We stress that the continuum-subtraction process does not rely on the source position or any trace model.We refer the reader to Kashino et al. (2023a) for a detailed description of this process.
For each object detected in F356W imaging, a 2D spectrum is extracted based on grismconf7 with the V4 trace models. 8We perform additional pixel-level correction to the trace models to optimize the extraction based on our own empirical calibration using the spectra of bright stars.Individual exposures are divided by the relevant sensitivity curve, rectified for small curvature, and resampled onto a common observed wavelength grid (3 μm λ 4 μm, with a pixel size of 9.75 Å).For each individual module, these exposures are coadded with sigma clipping to produce the final 2D spectrum of a galaxy.For a given object position, we extract one independent spectra from each module.This results in two independent spectra obtained from the two NIRCam modules for each source.Since these two spectra have reverse dispersion directions, only emission lines that are truly coming from the object of interest will appear at the same observed wavelength in both the spectra.All other lines shift in wavelength and/or disappear completely.This is the primary diagnostic to remove unrelated emission lines that are coming from other objects.Figure 1, top panels, shows the continuum-subtracted 2D emission line spectra of three galaxies at z ∼ 2.3, 4.2, and 6.3.Common emission lines are independently detected on both modules, further verifying their robustness.

HST Observations of J0100+2802
We obtain Hubble Space Telescope (HST)/Advanced Camera for Surveys imaging of the J0100+2802 field in the F850LP, F775W, and F606W filters (HST PID: 15085 and 13605).The total exposure time for these observations is 24,450 s.We query MAST for the individual flc.fits exposures, which are corrected for charge transfer inefficiency but have not been resampled.We align the individual exposures to the NIRCam F356W mosaic, and drizzle the images to the common pixel grid of the NIRCam mosaics using DrizzlePac (Mack et al. 2022).The routine masks cosmic rays, performs median blotting, and matches the sky of each exposure before calculating the median coadded image.To create PSF-matched images for precise multiband photometry we calculate matching kernels using PSFs derived from TinyTim (Krist et al. 2011) and resample to the 0 03 pixel scale.

Ground-based Spectroscopy of Quasar J0100+2802
Deep ground-based optical and NIR spectroscopic observations are performed on the z = 6.33 quasar J0100+2802 using both the Magellan/FIRE and the VLT/X-shooter instruments.The target has been observed for a total of 16.8 hr with 5.8 hr of Magellan/ FIRE (PI: R. Simcoe) and 11 hr of VLT/X-shooter (program ID: 096.A-0095; PI: M. Pettini) observations, respectively.Additionally, high-resolution (R ≈ 50,000), Keck/HIRES observations are performed on J0100+2802, to cover the optical (0.86 μm λ 9.9 μm) part of the spectrum (Cooper et al. 2019).The total integration times for the HIRES observations are 3.8 and 3 hr, respectively, in two different grating setups.
These observations are self consistently reduced with the PypeIt data reduction pipeline (Prochaska et al. 2020), and a final coadded, flux-calibrated spectrum is produced for both the instruments.We refer the reader to Eilers et al. (2023) for a detailed description of the data reduction method.The final FIRE/X-shooter spectra have a median signal to noise ratio (S/N) ∼ 138/126 per resolution element across the full wavelength range.The 16th and 84th percentile S/N per resolution element for FIRE/X-shooter spectra is 75/32 and 204/214, respectively.The variation of S/N per resolution element as a function of observed frame wavelength is presented in Table 1.

Methods
Here we describe how emission line galaxy properties are measured and how absorption lines are identified and analyzed.

Galaxy Redshifts
We create a candidate emission line galaxy list using the following two criteria.We search for emission line objects (with S/N per module > 7) in the central ≈4.6 arcmin 2 of the EIGER footprint, which is covered by both the NIRCam grism modules (A and B).This choice ensures that the total S/N of each emission line object is >10.We further restrict our search to objects whose spectrum contains a verified emission line that would plausibly be at a redshift within Δz ≈ 0.02 of the identified intervening metal absorption lines.For each identified object, we extract 2D galaxy spectra as described in Section 2.2.Additionally, we look for higher-redshift galaxies near the quasar identified with [O III] and Hβ emission lines (Matthee et al. 2023a).This yields a sample of 127 objects within ≈108″ of the J0100+2802 quasar.Monte Carlo simulations of injecting and recovering synthetic emission lines in the central ≈4.6 arcmin 2 of the J0100+28 field yield a spectroscopic completeness of 50% for an emission line flux of 1.6 × 10 −18 erg s −1 cm −2 , and 90% for an emission line flux of 2.8 × 10 −18 erg s −1 cm −2 (R. Mackenzie et al. 2024, in preparation)).We quantify the mass completeness of the spectroscopic sample presented in this work by using mock observations from the JWST Extragalactic Mock Catalog (JAGUAR; Williams et al. 2018).We compute the spectroscopic mass completeness as a function of stellar mass bins in different redshift slices.In each redshift slice, the mass completeness of an individual stellar mass bin is defined as the ratio between the number of galaxies above the F356W flux detection threshold to the total number of galaxies in that bin.At 〈z〉 ∼ 6 we find that the spectroscopic sample presented in this work is 50% mass complete at * M M log 8.5  , and at 〈z〉 ∼ 4.2 the galaxy sample is 50% mass complete at * M M log 8.13  .Each spectrum is individually inspected using a custom Python application programming interface (API) called zgui (Bordoloi et al. 2022a), which is used to extract an 1D spectrum for each module independently.We identify individual emission lines and fit a Gaussian profile to measure the redshift of the galaxy.The photo-z posterior distribution function for each object is also inspected to identify any contaminating foreground object.For the same object, it is crucial to inspect the spectra from both modules to identify and mask out any contaminating feature from other galaxies.Since the two spectra are extracted from the two NIRCam modules with reversed dispersion directions, only "real" lines associated with the extracted object will appear consistently at the same wavelength in the two modules.Each object is visually inspected by at least two individuals, and only objects with secure redshifts are considered for analysis.This creates a total sample of 87 galaxies with 0.4 < z < 6.8. Figure 2, left panel, shows the spatial distribution of these galaxies within 300 physical kiloparsecs of the J0100+2802 quasar sight line.Each individual galaxy is color coded to reflect its spectroscopic redshift.Independent redshift measurements from both modules yield a typical redshift accuracy of ≈140 km s −1 for each galaxy.Since we focus only on the CGM host of the Mg II absorbers along the J0100+2802 sight line, we further select only galaxies within 300 kpc from the J0100+2802 quasar sight line and at a redshift lower than the quasar (2.3 < z < 6.33).The lower redshift limit is chosen to match the lowest-redshift Mg II absorber detected along this line of sight.This yields a final sample of 29 galaxies.Figure 2, right panel shows the redshift and impact parameter distributions of this final sample of galaxies, which has a mean redshift of 〈z〉 = 4.5478 ± 0.201.Throughout the rest of the paper, we will only focus on this sample of galaxies.
The gray shaded regions (Figure 2, right panel) mark the redshift ranges where no strong galaxy rest-frame optical emission lines shift into the observed wavelength range of the EIGER NIRCam/grism spectroscopy.Our survey is most sensitive to the three redshift windows 2.3 < z < 2.7 (using He I, [S III], and Paγ lines), 4 < z < 5.1 (using Hα and [S II] lines), and 5.3 < z < 7 (using [O III] and Hβ lines).Future, ground-based spectroscopic follow ups or additional JWST spectroscopy with different gratings will enable us to cover these missing redshift ranges.
We note that since our galaxy search is explicitly within Δz ≈ 0.02 (Δv ≈ ±6000 km s −1 ) of the identified absorption lines, the galaxy sample is not selected blindly without any knowledge of the absorption systems.This galactocentric (e.g., Bordoloi et al. 2011;Tumlinson et al. 2013) approach is essential to characterize the covering fraction of the CGM gas or to measure the total metal mass budget of the CGM (Tumlinson et al. 2011;Bordoloi et al. 2014c).Owing to challenges of identifying single emission line galaxies, and in mitigating contamination from other sources, we restrict this work to focus only on galaxies within Δz ≈ 0.02 of the identified absorption line systems.This search window is large enough that we can still detect galaxies not associated with these absorption systems.However, this work does not search  for all galaxies outside the selection window and does not attempt to quantify the Mg II absorption covering fraction and metal mass budget around the EIGER galaxies.A complete galactocentric analysis of the CGM of the full EIGER survey will be presented in a future work.

Emission Line Measurement and SED fitting
We follow the procedure introduced in Matthee et al. (2023a) to measure the emission line fluxes of the strongest lines in a galaxy spectrum (e.g., Hα, He I 10830 Å, and [O III]) from the grism emission line observations, which is summarized as follows.We start with the 2D emission line spectra (top panels in Figure 1), and select a spectral region within ±50 Å of the emission line of interest in the rest frame.We collapse this emission line in the spectral direction and fit the spatial profile with single or multiple Gaussian profiles.This spatial profile is used to extract an 1D continuum-filtered spectrum for the galaxy optimally.We follow Matthee et al. (2023a) and rescale the noise level of the 2D emission line spectrum, by evaluating the standard deviation of empty sky pixels and setting it equal to the mean noise level of our 1D spectrum.This procedure is performed independently on each module.Figure 1, bottom panels, show three representative optimally extracted 1D spectra of galaxies at z ≈ 2.3, 4.2, and 6.3.The strong emission line features in each spectrum are marked.
We use these optimally extracted 1D spectra to fit the emission lines of interest.We fit the emission lines with Gaussian profiles (between one and three, depending on complexity) and measure their total line flux.
These line fluxes are used to perform spectral energy distribution (SED) fits, along with the photometric data at the spectroscopic redshift of the galaxy.We use broadband photometry from three HST (F6060W, F776W, and F850LP) and three JWST (F115W, F200W, and F356W) images (see Sections 2.2 and 2.3).We use the SED fitting code prospector (Johnson et al. 2021) to perform fits to these six photometric measurements along with the F356W grism line fluxes.Following Matthee et al. (2023a), we assume a 5% error on the spectrophotometric calibration of the observations.Prospector fits models the total stellar mass, gas-phase and stellar metallicities, the star formation history of the galaxy, dust attenuation, and the ionization parameter.We assume a Chabrier (2003) initial mass function and use MIST isochrone models (Choi et al. 2016;Dotter 2016).We use a delayed-τ star formation history model, and apply a dust attenuation correction following Calzetti et al. (2000).
Figure 3 shows the stellar mass and SFR estimates for these galaxies from the SED fits.Each circle represents a galaxy and is color coded as a function of its spectroscopic redshift.The estimated stellar masses span four decades in range (  ).We estimate the halo mass of the galaxies using the abundance matching relation from Behroozi et al. (2019).In this work, we quantify the virial radius of a galaxy as R 200 , the radius at which the halo mass density is 200 times the critical density of the Universe.We write it as where M halo is the halo mass, G is the universal constant of gravitation, and H(z) is the Hubble parameter at the redshift of interest.The uncertainties of the stellar masses and abundance matching relations are propagated through to the halo mass and virial radii estimates.The R 200 measurements are within 5% of the R vir estimates derived from Bryan & Norman (1998), well within the uncertainties of the R 200 estimates.We create a falsecolor JWST/NIRCam (F115W, F200W, and F356W) RGB image of each galaxy presented in this work.Each image is a 5″ × 5″ stamp and is presented in Figure 4.

Absorption Line Measurements
We visually inspect the spectra of the quasar J0100 obtained with the FIRE, X-shooter, and HIRES instruments and search for intervening absorption line systems.We use a Python-based Figure 2. Redshift and impact parameter distributions of the galaxies within 300 kpc of the quasar J0100+2802.Left panel: the distribution of all spectroscopically confirmed galaxies within 300 kpc from the quasar sight line.Each square represents a galaxy color coded as a function of its redshift.The quasar J0100+2802 is in the center of the figure (black star).Right panel: redshift and impact parameter distributions of the 29 galaxies analyzed in this work.Prominent emission lines used for the redshift measurements are shown.These galaxies are chosen to be at z < 6.33 and at impact parameters < 300 kpc.The gray boxes denote redshift ranges where no strong galaxy emission lines shift into the observed frame of the NIRCam 3.5 μm WFSS spectra.
API from the rbcodes package (Bordoloi et al. 2022a) to identify and tabulate these systems.Along the J0100+2802 quasar line of sight, 22 unique intervening absorption line systems are identified within 2.3 < z < 6.33.These lines include strong absorption systems traced by Al II, Al III, C II, C IV, Fe II, Mg II, Mg I, O I, Si II, and Si IV transitions.Among these, there are 16 unique Mg II absorption systems between 2.3 < z < 6.14.We compute the rest-frame Mg II equivalent width detection limit (3σ detection threshold) per two resolution elements for the X-shooter and FIRE spectra of J0100+2802.The mean 3σ detection threshold for the X-shooter and FIRE spectra are 11 and 18 mÅ, respectively.The 50% equivalent width completeness limits, computed as the median 3σ detection threshold per two resolution elements within different redshift bins, are presented in Table 2.
In this paper we focus primarily on the hosts of Mg II absorption line systems within the redshift windows of 2.3 < z < 2.7, 4 < z < 5.1, and 5.3 < z < 6.3 (see Figure 2).These redshift ranges correspond to the observer-frame wavelength ranges containing optical galaxy emission lines covered by the EIGER NIRCam/grism spectroscopy.The galaxy properties are presented in Table 3.We use a semiautomated framework to measure the absorption line strengths and kinematics associated with each identified foreground galaxy as follows: we first shift the reduced final quasar spectrum to the rest frame of the foreground galaxy, using the spectroscopic redshift of the galaxy as described in the previous section.We focus on common atomic absorption lines in predictable observed frame wavelength ranges.We quantify a detected absorption system to be associated with a host galaxy if it is within 300 physical kiloparsecs of J0100 +2802 sight line and within ±400 km s −1 of the systemic redshift of the galaxy.Our emission line search criterion (galaxy emission line S/N > 7 per module) can detect galaxies at * » M M log 7.1  (Figure 3).However, it is possible that some faint emission line galaxies are missed in this search.We adopt a conservative approach and only focus on reliably detected emission line galaxies in this work.The search for fainter (lower S/N) emission line galaxies will be carried out in a future work incorporating all the six quasar fields of the EIGER survey (R. Bordoloi et al. 2024, in preparation).
We extract short slices of quasar spectra around ±600 km s −1 of the systemic redshift of the galaxy, for each line of interest.These lines include Mg II, Fe II, C IV, Si IV, etc.We continuum normalize each slice using a multiordered Legendre polynomial and measure the rest-frame equivalent width and apparent optical depth (AOD) column density of each transition.We visually inspect each transition to confirm its presence, and set the velocity range for the AOD column density integration.We use the identified absorption line list to minimize contamination from other intervening absorption line systems, and when setting the velocity integration range.We attempt to identify every detection feature within ±600 km s −1 of the lines of interest.Most such features are not associated with the Mg II host galaxy and are positively identified to be associated with other intervening absorption line systems.We pay particular attention to the identified Mg II absorption doublet and ensure that the AOD ratios between the doublet range from 2:1 to 1:1 and require that the Mg II absorption profiles are aligned within 200 km s −1 from other detected metal absorption lines in that system.
Additionally, we fit Voigt profiles to the Mg II absorption doublet to quantify its kinematic component structure and to estimate column densities for severely blended lines.This is a crucial step, as the Voigt profile fitting improves on the AOD column density measurement using information about the line shape and location to constrain the fit.Further, for several saturated Mg II absorption features line saturation is taken into account as the line spread function is accounted for in these fits.
We use the Python-based Bayesian Markov Chain Monte Carlo (MCMC) Voigt profile fitting toolbox rbvfit (Bordoloi et al. 2023) to perform simultaneous fits to the Mg II absorption doublet.This approach fits the column density (N), Doppler b parameter, and velocity offset v for each component simultaneously for each Mg II doublet.We assume flat priors on each of these parameters with reasonable physical bounds.The number of components and the initial guess of the velocity offsets are obtained via visual inspection of the data.The fitting procedure generates posterior distributions for the model parameters.We chose the median of each distribution as the best-fit model parameter and the 16th and 84th percentiles as the corresponding upper and lower bounds on the best-fit parameters.One advantage of using a Bayesian MCMC approach over a frequentist χ 2 minimization approach is that our approach yields marginalized posterior distributions of the fitted parameters.This results in accurate column density estimates, even if simultaneously the Doppler b parameters are not well constrained in the moderate-resolution spectra used in this work.The best-fit Voigt profile parameters of the Mg II absorption systems presented in this work are reported in Table 4 and Figure 7.

Results
In this section we present the variation of Mg II absorption strength with galaxy properties, their radial profiles, and the absorber kinematics.

Distribution of Mg II Absorption Around Galaxies
We first characterize the spatial extent of Mg II absorption around the EIGER galaxies.We examine how the Mg II restframe equivalent width (W MgII2796 ) varies as a function of impact parameter (R) and the normalized virial radius (R/R 200 ) around the EIGER galaxies.Figure 5 shows the 1D Mg II radial absorption profiles around the 29 galaxies as function of R (left panel) and R/R 200 (right panel), respectively.Seven unique Mg II absorption system associated with host galaxies are detected, and the galaxy with the closest impact parameter to the absorber is assigned as the host galaxy (gray circles).No absorption is detected around 12 galaxies, which are marked with downward arrows; these measurements show the 2σ limit on the nondetection.Seven of the galaxies associated with a nondetection are at R < 200 kpc, suggesting that Mg II absorption is patchy at z > 4. All galaxies are color coded as a function of their redshift.In particular, for Mg II absorption systems at z ∼ 4, multiple galaxies are detected within ±400 km s −1 of the absorber redshift.All these associated galaxies are also presented in Figure 5.We will discuss the environment of the Mg II absorbers in the next section.The error bars on the x-axis (right panel) denote the uncertainties on the R 200 estimates.
Focusing on the closest galaxy (gray circles), two immediate observations stand out in Figure 5, left panel.The strongest absorbers are detected at close impact parameters (R < 100 kpc) of their host galaxies.Further, there are several absorbers detected at high impact parameters (R > 150 kpc).These absorbers are at the outer edge of the radial profiles of the Mg II absorption systems observed at z ≈ 2 (gray shaded region; Dutta et al. 2021).However, this trend is not taking into consideration that these galaxies all have different masses and virial radii, or the possibility that even fainter galaxies below our detection limit exist at closer distances.
False-color JWST/NIRCam F115W/F200W/F356W broadband images of each galaxy are presented in Figure 4.The galaxies exhibit diverse morphologies.Galaxies at z < 3 typically show well-formed disks and prominent inner bulges.Most of the z > 3 galaxies exhibit complex morphologies, with several of them exhibiting tidally disturbed features and several individual knots, indicating either merger or discrete star formation events along the galaxy disk.In particular, the stamp of galaxy EIGER-01-10308 stands out (Figure 4) as a merging galaxy with tidal streams clearly visible within 2″ of the galaxy.This galaxy is associated with the  strongest Mg II absorption system reported this work (Appendix A1), and we discuss this system in detail in the next section.
As a fraction of their inferred virial radii (Figure 5, right panel), most Mg II absorption extends out to ∼2-3 R/R 200 of the host galaxies.This spatial extent is much larger than typically observed around galaxies at z < 1 (Chen et al. 2010;Bordoloi et al. 2011;Churchill et al. 2013;Nielsen et al. 2013;Werk et al. 2013).The Mg II absorption strengths also show a steep decline with R/R 200 .We quantify this radial falloff by considering only the closest galaxies with detected Mg II absorption, and fit a power law, finding

( ) (
) .This best-fit power law with 68% confidence interval is presented as the dashed line with gray shaded region in Figure 5, right panel.When nondetections are included in the fit, the radial profile becomes slightly steeper but not too different than the previous fit ).We note that -+ 71 19 13 % (5/7) of the Mg II absorbers are detected outside the inferred virial radii of the host galaxies.
Figure 6 presents the variation of Mg II absorption as a function of stellar mass (top left panel), SFR (top right panel), and specific star formation rate (sSFR; bottom panel).The symbols are color coded to be consistent with Figure 5.We note that -+ 86 18 9 % (6/7) of the detected Mg II absorption systems are associated with higher stellar mass galaxies ), with the strongest absorption systems associated with the highest stellar mass systems.The two highest-redshift Mg II absorption systems are associated with lower-mass / * M M log  ≈ 7.1 (z ∼ 5.33) and / * M M log  ≈ 9.1 (z ∼ 6.01) galaxies.Further, the strongest Mg II absorption systems are associated with galaxies with the highest SFRs.These results tentatively suggest a correlation between the absorption strength of these strong high-z Mg II absorption systems and the star formation activity in their host galaxies, similar to what is observed for strong Mg II absorption at z ∼ 1 galaxies (Bordoloi et al. 2011).However, the variation of Mg II absorption with sSFR of the host galaxies (Figure 6; bottom panel) does not show a strong trend with increasing sSFR.Although the strongest Mg II absorption is detected around the most massive and vigorously star-forming galaxies, these are not the galaxies with the highest sSFR in this sample.This suggests that within this highly star-forming galaxy sample, with higher stellar masses host the strongest Mg II absorption.We add the caveat that the sample presented in this work exclusively consist of high-star-forming galaxies ( >log sSFR 9).Therefore, detailed comparison of galaxy sSFR and CGM absorption would require a sample of galaxies with <log sSFR 9.This will be explored in a future work once the full EIGER survey is complete (R. Bordoloi et al. 2024, in preparation).A unique facet of these high-z Mg II absorption systems is that the majority of them are beyond the inferred virial radii of their host galaxies.At these distances the Mg II absorbing gas may not be bound to the gravitational potential of their host galaxies.We explore this in the next section.

CGM Kinematics and Environments at High Redshift
In this section we focus on the kinematics and environments of the seven observed Mg II absorption profiles and discuss whether the observed absorption profiles are consistent with being bound to the dark matter halos of their host galaxies.We obtain the Mg II column densities and the best-fit Voigt profiles as described in Section 3.3.Figure 7 shows the best-fit absorption profiles for both Mg II transitions.The different panels show the Mg II absorption associated with each system.The vertical red ticks mark the positions of individual Voigt profile components.We use these components to quantify the velocity distribution of the Mg II absorption systems.
Figure 8, left panel, shows the velocity centroids of individual Voigt profiles fitted to the Mg II absorption components as a function of R/R 200 from their respective host galaxies.The symbols are color coded to show the Mg II column density of each component.The vertical range bars show the velocity range over which the equivalent width of the system is calculated.They are effectively the full width at zero optical depth of each Mg II absorption system.Both the thermal and bulk motion associated with the Mg II absorption systems are incorporated within the full velocity widths and therefore represent the maximum projected velocities of the absorption systems.The horizontal range bars represent the uncertainties in the estimated R/R 200 values.Figure 8, right panel, shows the distribution of Mg II absorption components relative to the host galaxy redshifts.The distribution of absorption components show a large velocity spread from −400 to 300 km s −1 .The absorption component velocities are offset from the systemic velocities with a median absolute velocity of 135 km s −1 and a standard deviation of 85 km s −1 .This is different than what is observed in the CGM of low-z galaxies, where most of the CGM absorption systems cluster around the systemic velocities of their host galaxies and their velocities are almost always consistent with being less than the associated virial velocities (Tumlinson et al. 2013;Bordoloi et al. 2014c;Huang et al. 2016).
We further investigate if the Mg II absorption detected around the EIGER galaxies is consistent with being bound to the dark matter halos of the host galaxies, as shown in Figure 9.We present the Mg II absorption component velocities normalized to the escape velocity associated with the host galaxy at that impact parameter as a function of R/R 200 .The horizontal error bars represent the uncertainties associated with the inferred R 200 values.The vertical range bars denote the velocity range over which the equivalent width of the system is calculated (normalized to the escape velocity of the system).It is clearly seen that five out of seven Mg II absorption systems are detected at R > 1.3 R 200 .Two absorption components have velocities higher than the projected escape velocities of these systems.This suggests that these absorption systems are not consistent with being bound to the dark matter halos of the host galaxies.Only two absorption systems (associated with EIGER-01-10308 and EIGER-01-09950) are at R < R 200 , and their component velocities are less than the escape velocities associated at these impact parameters.Only these two z ∼ 4 systems are consistent with being bound to the dark matter halo of their host galaxy.These two systems are also associated with higher galaxy overdensities around them (Figure 10).Looking at both the absorption component kinematics and the R/R 200 distributions of these absorption systems, we conclude that the CGM kinematics at high z is significantly different than what is observed at z < 1.At low z the bulk of the CGM absorption systems are consistent with being bound to the dark matter halos of their host galaxies, unlike the CGM of the EIGER galaxies.This suggests an evolution in CGM gas kinematics as a galaxy evolves from the early Universe to today.In the early Universe, the CGM gas could easily escape from the individual galaxy halos and chemically enrich the IGM around the galaxies.But as the galaxies became larger at low z, the CGM becomes increasingly bound to the host galaxies.
We finally explore the environments around the seven EIGER CGM host galaxies and quantify if these are isolated host galaxies or if they have companion galaxies.Figure 10 show the impact parameter to each galaxy at the systemic redshift of the galaxy noted in each panel.We plot all galaxies within 300 kpc from the J0100+2802 sight line, and within ±600 km s −1 of the host galaxies.Each galaxy is color coded as a function of its stellar mass.The vertical range bars show the associated R 200 of each galaxy.We describe the environment of each system below.EIGER-01-10308: this galaxy, at z ∼ is a tidally disturbed merging system.The tidal streams and different components of the merger can be clearly seen in highresolution JWST imaging (Appendix A1, panel (a)).The merging system has an integrated stellar mass of / * M M log  = 10.17, and an SFR ∼ 70 M e yr −1 .We extract individual NIRCam spectra of the two smaller merging components (Appendix A1, panel (b)) and compute their individual redshifts.These two components are at ∼−23 and −80 km s −1 from the main galaxy, respectively.These individual components are marked as gold stars in Figure 10, top left panel.The galaxy EIGER-01-10308 resides in a local overdensity and there are seven additional galaxies within 200 kpc from the quasar sight line.The galaxies are within 220 kpc of each other and have a velocity dispersion of 228 km s −1 .This large overdensity of galaxies at close physical and kinematic separations may be part of a protogroup at 〈z〉 ≈ 4.2234.The additional seven galaxies are at higher impact parameters than EIGER-01-10308 and at R > R 200 .The Mg II absorption associated with EIGER-01-10308 show complex kinematics with six distinct individual absorption components identified with a velocity spread of ≈500 km s −1 (Figure 7).Absorption is also detected in the Fe II, Mg I, Al III, and Si II transitions (Appendix A1, panel (c)).The galaxy is at low impact parameter with R/R 200 < 1 and the velocity of the absorption components are less than the escape velocity associated with this impact parameter.This system is one of the two absorption systems in this work that has Mg II absorption kinematics consistent with being bound to the dark matter halo of the host galaxy.
EIGER-01-09950: at z ∼ 4.5, there are three additional galaxies within 300 kpc of this galaxy (Appendix A2).The galaxy EIGER-01-09950 has a stellar mass of / * M M log  = 9.88, and an SFR ∼ 29 M e yr −1 , respectively.There are two other galaxies at very close kinematic separations: EIGER-01-09078 at a separation of 16 kpc from EIGER-01-09950 and a velocity offset of 368 km s −1 and EIGER-01-08811 at a separation of 22 kpc from EIGER-01-09950 and a velocity offset of 122 km s −1 .EIGER-01-09078 and EIGER-01-08811 are 40 and 50 kpc away from the sight line of quasarJ0100+2802, respectively.A third, more massive galaxy (EIGER-01-06193) is detected 235 kpc away from EIGER-01-09950 and at a velocity separation of 232 km s −1 .Galaxy EIGER-01-06193 is at an impact parameter of 260 kpc from the sight line of quasar J0100 +2802.These galaxies all lie close to each other (within 232 kpc and 166 km s −1 ) and could form a protogroup at z ∼ 4.5192.The Mg II absorption profile is again complex for this system (Figure 7), with five identified distinct absorption components.The absorption spans a velocity range of ≈350 km s −1 .We also detect absorption in the Fe II, Mg I, Al III, and Si II transitions in this system (Appendix A2).The strongest Mg II absorption component is offset from the systemic redshift of galaxy EIGER-01-09950 by ∼95 km s −1 , and kinematically lines up with the nearby galaxy EIGER-01-09078.But EIGER-01-09078 has a much lower stellar mass (see Table 3), and the impact parameter to it is higher than the inferred virial radius associated with it (Figure 10).Since EIGER-01-09950 is the closest galaxy to the line of sight, and kinematically the projected Mg II velocity is lower than the escape velocity of the host galaxy, we conclude that the Mg II absorption is consistent with being bound to the dark matter halo of the host galaxy.
EIGER-01-10424: at z ∼ 4.64, a faint galaxy is detected at an impact parameter of 72 kpc from the J0100+2802 quasar line of sight.This galaxy has a stellar mass of / * M M log  = 9.18, and an SFR ∼ 13 M e yr −1 .The galaxy is next to a bright z ∼ 1 foreground galaxy detected in Paα emission (see Figure 4).A ground-based MUSE spectrum of the bright foreground galaxy shows the [O II] emission doublet, confirming it as a low-z galaxy.There are two faint emission components detected for the target galaxy, suggesting that it is a merging system at z ∼ 4.64, however, owing to the position angle of the NIRCam grism spectra, the two components cannot be spatially resolved.We note that since the emission line redshift for this system is estimated from a single Hα emission line, it is possible that this emission is associated with the foreground bright galaxy at z ∼ 1.However, the ground-based seeing does not allow us to check for [O II] emission associated with the fainter components in the MUSE data cube.The presence of metal absorption  (Dutta et al. 2021).Galaxies associated with Mg II absorbers are seen out to 300 kpc but most of the associated galaxies are well beyond the inferred virial radii of the galaxies.The vertical dashed line (right panel) marks R/R 200 = 1.The dashed orange line (right panel) with shaded region shows the best-fit power law to the EIGER Mg II radial profile normalized to the inferred virial radii of the galaxies.and the emission lines in the NIRCam grism spectra has led us to conclude that EIGER-01-10424 is the host galaxy of the Mg II absorption detected at z ∼ 4.6.This system only shows a weak Mg II absorption doublet, kinematically offset from the host galaxy's systemic redshift by ∼−350 km s −1 (Figure 7), and beyond the inferred virial radius of the host galaxy (Figure 10).We therefore conclude that the absorption is not consistent with being bound to the host galaxy.
EIGER-01-06898 and EIGER-01-20300: these galaxies are associated with Mg II absorption at z ∼ 6.015 and 5.33, respectively.EIGER-01-06898 is at an impact parameter of 201 kpc and has a stellar mass of / * M M log  = 9.1 and an SFR ∼ 8.4 M e yr −1 .EIGER-01-20300 is at an impact parameter of 19 kpc and has a stellar mass of / * M M log  = 7.1 and an SFR ∼ 1.1 M e yr −1 .Both these galaxies are "isolated," in the sense that no companion galaxy within ±600 km s −1 and 300 kpc of them is detected.Both galaxies are at an impact parameter beyond the R 200 radius of their host galaxy (Figure 10).For EIGER-01-06898, we detect Mg II and C IV absorption offset from the systemic redshift of the galaxy by ≈−136 km s −1 .Around EIGER-01-20300, we detect Mg II, C IV, Si IV, and Si II absorption, kinematically offset from the systemic redshift of the host galaxy by ≈120 km s −1 .We will report and quantify the high-ionization C IV absorption profiles in an upcoming publication (R. A. Simcoe et al. 2024, in preparation).For both these systems, the absorption is kinematically offset from the systemic galaxy redshift and is detected beyond the inferred virial radius of their corresponding host galaxy.We therefore conclude that the Mg II absorption detected around these galaxies is not bound to the host galaxy's dark matter halo.
EIGER-01-06569 and EIGER-01-09351: these two galaxies are associated with Mg II absorption at z ∼ 2.67 and 2.31, respectively.The galaxy EIGER-01-6569 is at an impact parameter of 270 kpc and has a stellar mass of / * M M log  = 10.17 and an SFR ∼ 75 M e yr −1 .EIGER-01-09351 is at an impact parameter of 172 kpc and has a stellar mass of / * M M log  = 10.64 and an SFR ∼ 75 M e yr −1 .In both of these systems, any lines blueshifted more than the Mg II absorption doublet will be in the Lyα forest and not observable along this high-z quasar sight line.Both of these Mg II systems are weak absorption systems (Figure 7) and are beyond the inferred virial radius of their host galaxy.In both cases, the Mg II absorption is consistent with not being bound to the dark matter halo of its host galaxy.
In all the detected systems, only the galaxies EIGER-01-10308 and EIGER-01-09950 have associated Mg II absorption consistent with being bound to the dark matter halo of the host galaxy.In both cases, there is a galaxy overdensity, suggesting that these galaxies reside in two galaxy protogroups.In all other cases, where no galaxy overdensity is seen, most of the Mg II absorption is consistent with not being bound to the dark matter halo of their host galaxy.This is significantly different than the cool CGM detected at z < 1, where most of the CGM gas is consistent with being bound to its host galaxy.

Discussion and Conclusions
The commissioning of JWST has opened a new discovery space to study the CGM of high-z galaxies.In this work, we present deep NIR (3.5 μm) JWST/NIRCam WFSS spectroscopic observations of the field of the z ∼ 6.33 quasar J0100 +2802 from the EIGER survey to characterize the cool CGM (traced by Mg II absorption) around 29, 2.3 < z < 6.3 galaxies.The JWST WFSS spectroscopy is accompanied by deep JWST NIR and HST optical broadband imaging and deep groundbased high-resolution spectroscopy of the quasar.This work builds on the initial EIGER survey papers that characterized the properties of a large sample of [O III]-emitting galaxies at z = 5.33-6.93(Kashino et al. 2023a;Matthee et al. 2023a).Our main conclusions are summarized as follows: 1. Using JWST/NIRCam 3.5 μm grism spectroscopy, we discover 29 galaxies within 300 kpc of the sight line of the quasar J0100+2802 in three redshift windows, 2.3 < z < 2.7, 4 < z < 5. , and exhibit a strong correlation between their SFRs and stellar masses.All the galaxies presented in this work are star forming.3. The identified galaxies show diverse morphologies, from tidally disturbed mergers to well-formed disks.Most of the z > 3 galaxies show complex morphologies of either several clumps or tidally disturbed features.4. We identify the CGM of the host galaxies of seven Mg II absorption systems within an impact parameter of 300 kpc.Identifying the closest galaxy to the quasar line of sight as the host, we find that strongest Mg II absorption is detected at close impact parameters (R < 100 kpc).The Mg II absorption strength drops off as a function of galactocentric radius from the host galaxies, characterized by a power-law falloff.This radial falloff is slightly shallower than the Mg II radial absorption profile observed for z < 1 galaxies.5.There are 12 galaxies within 300 kpc for which no Mg II  absorption is detected at a mean detection threshold of 10-20 mÅ.This shows that at high z, the cool CGM traced by Mg II absorption is patchy. galaxies, respectively.Similarly, the strongest Mg II absorption systems are also associated with the most star-forming galaxies.9.The Mg II absorption kinematics are not symmetrically clustered around the systemic velocities of their host galaxies.The absorption components velocities have a large velocity spread (from −400 to 300 km s −1 ) around the systemic redshifts of the host galaxies.The absorption components show a median absolute velocity of 135 km s −1 and a standard deviation of 85 km s −1 .10. Five out of the seven Mg II absorption systems are associated with host galaxies at R > 1.3 R 200 .Moreover, two absorption components show projected velocities higher than the escape velocities of the host galaxies.We conclude that five out of seven absorption systems have cool CGM gas, consistent with being unbound to their host dark matter halos.11.We highlight the CGM around two particular Mg II absorption systems (z ∼ 4.2 and 4.5) because they are associated with host galaxies at R < R 200 , and with absorption gas kinematics consistent with being bound to the dark matter halos of the host galaxies.These z ∼ 4.22 and z ∼ 4.5 Mg II absorption systems exhibit complex kinematics spanning ≈500 and 350 km s −1 , respectively.12.The Mg II absorption system at z ∼ 4.22 is associated with a morphologically disturbed merging galaxy with three distinct merging components within 80 km s −1 of each other.This galaxy is within a local galaxy overdensity where seven additional galaxies are observed within 200 kpc of the quasar sight line and within ±500 km s −1 of the host galaxy.These galaxies might be part of a galaxy protogroup at z ∼ 4.22.13.The Mg II absorption system at z ∼ 4.5 is associated with a galaxy with two close kinematic companions within 16-22 kpc of the host galaxy.Both of these companion galaxies are within a velocity separation of <370 km s −1 from the host galaxy.A third massive companion galaxy is detected 235 kpc from the CGM of the host galaxy.These galaxies might be part of a galaxy protogroup at z ∼ 4.5.14.The two strongest and kinematically most complex Mg II  absorption systems (at z ∼ 4.22 and z ∼ 4.5) are both part  The components that are at R/R 200 < 1 and with velocity/v escape < 1 are from two individual galaxies at very close impact parameters.For the other five galaxies, the Mg II absorption is consistent with not being bound to the host dark matter halos.Mg II absorption associated with EIGER-01-09950 is offset by 0.1 R/R 200 along the x-axis for visual clarity.
Figure 10.Environments around the galaxies for the Mg II absorption system with the closest impact parameter as a function of impact parameter.The vertical ranges mark the inferred virial radius of the individual galaxies.Only two systems are within the virial radius of the associated galaxies.The galaxies are color coded as a function of their stellar mass.The gold stars in the top left panel show the velocity components of three distinct components that are merging to form the full system.For the galaxy EIGER-01-10424, although marked as a single system, the imaging reveals two components that are merging to form a single system.
of two local galaxy overdensities.The Mg II absorption detected in these systems may be part of the intragroup gas associated with these two protogroup galaxies at high z.
In summary, we present the first results characterizing the cool CGM around 2.3 < z < 6.3 galaxies in the first field of the EIGER survey.We examine CGM hosts of seven Mg II absorption systems and find that most of the high-z Mg II absorption is not consistent with being bound to the dark matter halos of the host galaxies.This is in contrast to what is seen for the CGM of z < 1 galaxies (Tumlinson et al. 2017).In particular, extensive HST/Cosmic Origins Spectrograph CGM surveys of z < 0.2, L * and sub-L * galaxies show that at low z most of the CGM is kinematically consistent with being bound to their host galaxies.These differences arise owing to a combination of lower gravitational potentials of high-z galaxies and a much higher Hubble parameter in the earlier Universe.These findings indicate that the galaxies in the early Universe were much more efficient in distributing metals produced in stars out of galaxies and chemically enriching their IGM.Such chemically enriched gas could be deposited in nearby galaxies at a later time, providing fuel for the next generation of stars within them.
These observations will enable direct comparisons of CGMgalaxy correlations in next-generation simulations.Several simulations have looked into the statistical properties of metal absorption line systems at high z.These works reproduce the absorber statistics of strong absorption systems but significantly overproduce weak metal absorption systems (Rahmati et al. 2016;Finlator et al. 2020;Hasan et al. 2020), perhaps suggesting that the feedback prescription used in these simulations were too strong at high z.There are also tensions between the observed metal absorption line statistics of cool Mg II gas and those produced in simulations at high z (Keating et al. 2016).
In the TECHNICOLOR DAWN simulation, a correlation between high-z absorbers and galaxies within a few hundred physical kiloparsecs was found (Finlator et al. 2020;Doughty & Finlator 2023).A two-point correlation function analysis between the metal absorption systems and galaxies show a strong correlation at small impact parameters (R < 100 pkpc) that qualitatively trace the Mg II absorption strength-R anticorrelation presented in this work.Interestingly, Doughty & Finlator (2023) find that there is a strong correlation between metal absorber type/strength with the local galaxy overdensity than the stellar mass of the host galaxies.Several strong systems in this work are also observed in local overdensities and better statistics are needed to explore these correlations.With new JWST observations providing direct observational constraints on CGM-galaxy correlations, quantitative comparison in the future would enable direct constraints on the feedback prescriptions being implemented in these simulations.
These results reinforce the power of JWST/NIRCam grism observations to conduct high-z galaxy spectroscopy campaigns efficiently.By combining a high-fidelity JWST spectroscopic campaign with deep group-based NIR spectroscopy of z > 6 quasars, we demonstrate an efficient program design to census the cool CGM around high-z galaxies in the EIGER survey.In an upcoming paper, we will focus on detailed properties of the CGM of high-z O I and C IV absorption systems (R. A. Simcoe et al. 2024, in preparation.).We will further extend this work to the full six quasar fields of the EIGER survey (R. Bordoloi et al. 2024, in preparation) to quantify the CGM-galaxy correlations in the EIGER survey better.

Figure 1 .
Figure1.JWST NIRCam/F356W grism spectra of three galaxies at z ∼ 2.3 (top left), z ∼ 4.2 (top right), and z ∼ 6.3 (bottom).The 2D emission line spectrum from each module is presented for each galaxy.The 1D optimally extracted spectrum is shown in the main panel for each galaxy with prominent emission lines marked out.The red shaded regions show the 1σ uncertainties for the 1D spectra.

)
and almost two dex in SFR.Most of the high-z (z > 5) galaxies have low stellar mass (

Figure 3 .
Figure 3. Stellar mass and SFR estimates of the galaxies used in this work.Each circle is color coded to reflect the spectroscopic redshift of the galaxy.The error bars represent the 16th and 84th percentile uncertainties in the stellar masses and SFRs obtained from the SED fits.

Figure 4 .
Figure 4. False-color JWST/NIRCam F115W/F200W/F356W stamps of the galaxies within 300 kpc of the background J0100+2802 quasar.Each galaxy has a confirmed spectroscopic redshift.The yellow arrow is a position vector directed toward the quasar line of sight from the host galaxy.The location of the position vector is arbitrary in each stamp.

Figure 5 .
Figure5.Mg II radial absorption profile as a function of impact parameter (left panel), and normalized virial radius (right panel).In both panels, the squares represent detections and the squares with downward arrows represent the 2σ upper limits of the nondetections.Each square is color coded as a function of the redshift of the galaxy.The gray circles show the closest galaxy associated with a Mg II absorber at a distinct redshift.The horizontal error bars show the uncertainty on the normalized virial radii.Gray shaded regions (left panel) indicate the best-fit Mg II absorption radial profile for z ≈ 2 galaxies(Dutta et al. 2021).Galaxies associated with Mg II absorbers are seen out to 300 kpc but most of the associated galaxies are well beyond the inferred virial radii of the galaxies.The vertical dashed line (right panel) marks R/R 200 = 1.The dashed orange line (right panel) with shaded region shows the best-fit power law to the EIGER Mg II radial profile normalized to the inferred virial radii of the galaxies.

Figure 6 .
Figure 6.Variation of Mg II absorption strength with galaxy stellar mass (top left panel), SFR (top right panel), and sSFR (bottom panel).At each discrete absorber redshift, the closest galaxy is marked with the gray circle.Each galaxy is color coded to show its redshift.Horizontal error bars show the uncertainties on the stellar masses, SFRs, and sSFRs.The strongest Mg II absorption systems are detected near the most massive and most vigorously star-forming galaxies.

Figure 7 .
Figure 7. Normalized quasar spectra of J0100+28002 showing the detected Mg II 2796 (left panel) and 2803 (right panel) absorption profiles along with their corresponding Voigt profile fits (blue lines).The vertical red ticks mark the individual Voigt profile components.For each system, the corresponding galaxy ID, redshift, and the integrated total column density are shown.The plus signs bracket the velocity range over which the equivalent width is computed.

Figure 8 .
Figure 8. Left panel: the Mg II absorption component velocity centroids with respect to the systemic redshift as a function of impact parameter normalized to the virial radius of the galaxy.The vertical range bars indicate the maximum projected kinematic extent of Mg II absorption for each system.The horizontal bars show the uncertainties in the R/R 200 estimates.Each component is color coded to show the Voigt profile fitted column density.Mg II absorption associated with EIGER-01-09950 is offset by 0.1 R/R 200 along the x-axis for visual clarity.Right panel: the distribution of individual Mg II absorption components in each Voigt fit.

Figure 9 .
Figure9.Individual Mg II absorption component velocity normalized by the escape velocity of their associated galaxy at that R, as a function of the R/R 200 of that galaxy.The components are color coded to show their column densities.The components that are at R/R 200 < 1 and with velocity/v escape < 1 are from two individual galaxies at very close impact parameters.For the other five galaxies, the Mg II absorption is consistent with not being bound to the host dark matter halos.Mg II absorption associated with EIGER-01-09950 is offset by 0.1 R/R 200 along the x-axis for visual clarity.

Figure A1 .
Figure A1.Compilation of CGM absorption around the merging galaxy system EIGER-01-10308 at z ∼ 4.22.(a) A false-color JWST broadband image of the system is presented.Merging components are marked with a number.A tidal tail between components 1 and 3 can be clearly seen.(b) Extracted NIRCam WFSS 1D spectrum associated with each marked merging component is presented in the individual subplots.Hα emission components are fitted with Gaussian profiles to measure the redshifts of the merging components.(c) Mg II, Fe II, Al III, and Mg I absorption associated with this system is presented with their corresponding Voigt profile fits.

Figure A2 .
Figure A2.Compilation of CGM absorption around three kinematically close galaxies at z ∼ 4.5.(a) A false-color JWST broadband image of three galaxies.All galaxies are within 50 kpc of the quasar sight line and <370 km s −1 of each other.(b) 1D NIRCam WFSS spectra associated with each marked galaxy.The Hα emission components are fitted with Gaussian profiles to measure the redshifts of the merging components.(c) Mg II, Fe II, Al III, and Mg I absorption associated with this system is presented with their corresponding Voigt profile fits.

Table 1
Signal to Noise Ratio Per Resolution Element a The 16th, 50th, and 84th percentile S/N per resolution element is reported.

Table 2
Mg II 50% Equivalent Width Completeness Limit [Units of mÅ]

Table 4
Voigt Profile Fit Parameters of the Mg II Absorption