arXiv subjects

S. Charlot

Publications and source records attributed to S. Charlot.

At least 55 records · Page 3Linked to original sources

arXiv2022

The Near-Infrared Spectrograph (NIRSpec) on the James Webb Space Telescope I. Overview of the instrument and its capabilities

We provide an overview of the design and capabilities of the near-infrared spectrograph (NIRSpec) onboard the James Webb Space Telescope. NIRSpec is designed to be capable of carrying out low-resolution ($R\!=30\!-330$) prism spectroscopy over the wavelength range $0.6-5.3\!~μ$m and higher resolution ($R\!=500\!-1340$ or $R\!=1320\!-3600$) grating spectroscopy over $0.7-5.2\!~μ$m, both in single-object mode employing any one of five fixed slits, or a 3.1$\times$3.2 arcsec$^2$ integral field unit, or in multiobject mode employing a novel programmable micro-shutter device covering a 3.6$\times$3.4~arcmin$^2$ field of view. The all-reflective optical chain of NIRSpec and the performance of its different components are described, and some of the trade-offs made in designing the instrument are touched upon. The faint-end spectrophotometric sensitivity expected of NIRSpec, as well as its dependency on the energetic particle environment that its two detector arrays are likely to be subjected to in orbit are also discussed.

astro-ph.IM↗

arXiv2022

The Near-Infrared Spectrograph (NIRSpec) on the James Webb Space Telescope II. Multi-object spectroscopy (MOS)

We provide an overview of the capabilities and performance of the Near-Infrared Spectrograph (NIRSpec) on the James Webb Space Telescope (JWST) when used in its multi-object spectroscopy (MOS) mode employing a novel Micro Shutter Array (MSA) slit device. The MSA consists of four separate 98 arcsec $\times$ 91 arcsec quadrants each containing $365\times171$ individually addressable shutters whose open areas on the sky measure 0.20 arcsec $\times$ 0.46 arcsec on a 0.27 arcsec $\times$ 0.53 arcsec pitch. This is the first time that a configurable multi-object spectrograph has been available on a space mission. The levels of multiplexing achievable with NIRSpec MOS mode are quantified and we show that NIRSpec will be able to observe typically fifty to two hundred objects simultaneously with the pattern of close to a quarter of a million shutters provided by the MSA. This pattern is fixed and regular, and we identify the specific constraints that it yields for NIRSpec observation planning. We also present the data processing and calibration steps planned for the NIRSpec MOS data. The significant variation in size of the mostly diffraction-limited instrument point spread function over the large wavelength range of 0.6-5.3 $μ$m covered by the instrument, combined with the fact that most targets observed with the MSA cannot be expected to be perfectly centred within their respective slits, makes the spectrophotometric and wavelength calibration of the obtained spectra particularly complex. These challenges notwithstanding, the sensitivity and multiplexing capabilities anticipated of NIRSpec in MOS mode are unprecedented, and should enable significant progress to be made in addressing a wide range of outstanding astrophysical problems.

astro-ph.IM↗

arXiv2021

Modelling the M*-SFR relation at high redshift: untangling factors driving biases in the intrinsic scatter measurement

We present a method to self-consistently propagate M$_{*}$ and SFR ($Ψ$) uncertainties onto intercept, slope and intrinsic scatter estimates for a simple model of the main sequence of star forming galaxies where $Ψ= α+ β$M$_{*} + \mathcal{N}(0,σ)$. From simple idealised models set up with broad-band photometry from NIRCam filters at $z\sim5$, we test the method and compare to methods in the literature. Simplifying the $Ψ$ estimate by basing it on dust-corrected MUV can help to reduce the impact of template set degeneracies on slope and intercept estimates, but act to bias the intrinsic scatter estimate. We find that broad-band fluxes alone cannot constrain the contribution from emission lines, implying that strong priors on the emission-line contribution are required if no medium-band constraints are available. Therefore at high redshifts, where emission lines contribute a higher fraction of the broad-band flux, photometric fitting is sensitive to $Ψ$ variations on short ($\sim$ 10 Myr) timescales. Priors on age imposed with a constant (or rising) star formation history (SFH) do not allow one to investigate a possible dependence of $σ$ on M$_{*}$ at high redshifts. Delayed exponential SFHs have less constrained priors, but do not account for $Ψ$ variations on short timescales, a problem if $σ$ increases due to stochasticity of star formation. A simple SFH with current star formation decoupled from the previous history is appropriate. We show that, for simple exposure-time calculations assuming point sources, with low levels of dust, we should be able to obtain unbiased estimates of the main sequence down to log(M/M$_{\odot}$) $\sim$ 8 at $z\sim5$ with the James Webb Space Telescope while allowing for stochasticity of star formation.

astro-ph.GA↗

arXiv2021

Constraints on the dust extinction law of the Galaxy with Swift/UVOT, Gaia and 2MASS

We explore variations of the dust extinction law of the Milky Way by selecting stars from the Swift/UVOT Serendipitous Source Catalogue, cross-matched with Gaia DR2 and 2MASS to produce a sample of 10,452 stars out to ~4kpc with photometry covering a wide spectral window. The near ultraviolet passbands optimally encompass the 2175A bump, so that we can simultaneously fit the net extinction, quoted in the V band (A$_V$), the steepness of the wavelength dependence ($δ$) and the bump strength (E$_b$). The methodology compares the observed magnitudes with theoretical stellar atmospheres from the models of Coelho. Significant correlations are found between these parameters, related to variations in dust composition, that are complementary to similar scaling relations found in the more complex dust attenuation law of galaxies - that also depend on the distribution of dust among the stellar populations within the galaxy. We recover the strong anticorrelation between A$_V$ and Galactic latitude, as well as a weaker bump strength at higher extinction. $δ$ is also found to correlate with latitude, with steeper laws towards the Galactic plane. Our results suggest that variations in the attenuation law of galaxies cannot be fully explained by dust geometry.

astro-ph.GA↗

arXiv2021

Ultra-faint [CII] emission in a redshift = 2 gravitationally-lensed metal-poor dwarf galaxy

Extreme emission-line galaxies (EELGs) at redshift z=1-2 provide a unique view of metal-poor, starburst sources that are the likely drivers of the cosmic reionization at z$\geq6$. However, the molecular gas reservoirs of EELGs - the fuel for their intense star-formation - remain beyond the reach of current facilities. We present ALMA [CII] and PdBI CO(2-1) observations of a z=1.8, strongly lensed EELG SL2S 0217, a bright Lyman-$α$ emitter with a metallicity 0.05 $Z_\odot$. We obtain a tentative (3-4$σ$) detection of the [CII] line and set an upper limit on the [CII]/SFR ratio of $\leq1\times10^6$ $L_\odot$/($M_\odot$ yr$^{-1}$), based on the synthesized images and visibility-plane analysis. The CO(2-1) emission is not detected. Photoionization modelling indicates that up to 80% of the [CII] emission originates from neutral or molecular gas, although we can not rule out that the gas is fully ionized. The very faint [CII] emission is in line with both nearby metal-poor dwarfs and high-redshift Lyman-$α$ emitters, and predictions from hydrodynamical simulations. However, the [CII] line is 30$\times$ fainter than predicted by the De Looze et al. [CII]-SFR relation for local dwarfs, illustrating the danger of extrapolating locally-calibrated relations to high-redshift, metal-poor galaxies.

astro-ph.GA↗

arXiv2021

The VANDELS ESO public spectroscopic survey: final Data Release of 2087 spectra and spectroscopic measurements

VANDELS is an ESO Public Spectroscopic Survey designed to build a sample of high signal to noise, medium resolution spectra of galaxies at redshift between 1 and 6.5. Here we present the final Public Data Release of the VANDELS Survey, comprising 2087 redshift measurements. We give a detailed description of sample selection, observations and data reduction procedures. The final catalogue reaches a target selection completeness of 40% at iAB = 25. The high Signal to Noise ratio of the spectra (above 7 in 80% of the spectra) and the dispersion of 2.5Å allowed us to measure redshifts with high precision, the redshift measurement success rate reaching almost 100%. Together with the redshift catalogue and the reduced spectra, we also provide optical mid-IR photometry and physical parameters derived through SED fitting. The observed galaxy sample comprises both passive and star forming galaxies covering a stellar mass range 8.3< Log(M*/Msolar)<11.7. All catalogues and spectra are accessible through the survey database (http://vandels.inaf.it) where all information can be queried interactively, and via the ESO Archive (https://www.eso.org/qi/).

astro-ph.GA↗

arXiv2020

Stars and gas in the most metal-poor galaxies I: COS and MUSE observations of SBS 0335-052E

Among the nearest most metal-poor starburst-dwarf galaxies known, SBS 0335-052E is the most luminous in integrated nebular He II λ4686 emission. This makes it a unique target to test spectral synthesis models and spectral interpretation tools of the kind that will be used to interpret future rest-frame UV observations of primeval galaxies. Previous attempts to reproduce its He II λ4686 emission luminosity found that X-ray sources, shocks, and single Wolf-Rayet stars are not main contributors to the He II-ionizing budget; and that only metal-free single rotating stars or binary stars with a top-heavy IMF and an unphysically-low metallicity can reproduce it. We present new UV (COS) and optical (MUSE) spectra which integrate the light of four super star clusters in SBS 0335-052E. Nebular He II, [C III], C III], C IV, and O III] UV emission lines with equivalent widths between 1.7 and 5 Å, and a C IV λλ1548, 1551 P-Cygni like profile are detected. Recent extremely-metal poor shock + precursor models and binary models fail to reproduce the observed optical emission-line ratios. We use different sets of UV and optical observables to test models of constant star formation with single non-rotating stars which account for very massive stars, as blueshifted O V λ1371 absorption is present. Simultaneously fitting the fluxes of all high-ionization UV lines requires an unphysically-low metallicity. Fitting the P-Cygni like + nebular components of C IV λλ1548, 1551 does not constrain the stellar metallicity and time since the beginning of star formation. We obtain 12+log(O/H)=7.45\pm0.04 and log(C/O)=-0.45(+0.03)(-0.04) for the galaxy. Model-testing would benefit from higher spatial resolution UV and optical spectroscopy of the galaxy.

astro-ph.GA↗

arXiv2020

MEGARA-IFU detection of extended HeII4686 nebular emission in the central region of NGC1569 and its ionization budget

We here report the detection of extended HeII4686 nebular emission in the central region of NGC1569 using the integral field spectrograph MEGARA at the 10.4-m Gran Telescopio Canarias. The observations cover a Field of View (FoV) of 12.5 arcsec x 11.3 arcsec at seeing-limited spatial resolution of ~15 pc and at a spectral resolution of R=6000 in the wavelength range 4330--5200 Angstrom. The emission extends over a semi-circular arc of ~40 pc width and ~150 pc diameter around the super star cluster A (SSC-A). The Av derived using Balmer decrement varies from the Galactic value of 1.6 mag to a maximum of ~4.5 mag, with a mean value of 2.65+/-0.60 mag. We infer 124+/-11 Wolf-Rayet (WR) stars in SSC-A using the HeII4686 broad feature and Av=2.3 mag. The He+ ionizing photon rate from these WR stars is sufficient to explain the luminosity of the HeII4686 nebula. The observationally-determined total He+ and H0 ionizing photon rates, their ratio, and the observed number of WR stars in SSC-A are all consistent with the predictions of simple stellar population models at an age of 4.0+/-0.5 Myr, and mass of (5.5+/-0.5)x10^5 Msun. Our observations reinforce the absence of WR stars in SSC-B, the second most massive cluster in the FoV. None of the other locations in our FoV where HeII4686 emission has been reported from narrow-band imaging observations contain WR stars.

astro-ph.GA↗

arXiv2019

Spatially-resolved analysis of neutral winds, stars and ionized gas kinematics with MEGARA/GTC: new insights on the nearby galaxy UGC 10205

We present a comprehensive analysis of the multi-phase structure of the interstellar medium (ISM) and the stellar kinematics in the edge-on nearby galaxy UGC 10205 using integral field spectroscopy (IFS) data taken with MEGARA at the GTC. We explore both the neutral and the ionized gas phases using the interstellar Na ${\small I}$ D doublet absorption (LR$-$V set-up, R $\sim$ 6000) and the H$α$ emission line (HR$-$R set-up, R $\sim$ 18000), respectively. The high-resolution data show the complexity of the H$α$ emission line profile revealing the detection of up to three kinematically distinct gaseous components. Despite of this fact, a thin disk model is able to reproduce the bulk of the ionized gas motions in the central regions of UGC 10205. The use of asymmetric drift corrections is needed to reconciliate the ionized and the stellar velocity rotation curves. We also report the detection of outflowing neutral gas material blueshifted by $\sim$ 87 km s$^{-1}$. The main physical properties that describe the observed outflow are a total mass M$_{out}$ $=$ (4.55 $\pm$ 0.06) $\times$ 10$^{7}$ M$_{\odot}$ and a cold gas mass outflow rate $\dot{M}$$_{out}$ $=$ 0.78 $\pm$ 0.03 M$_{\odot}$ yr$^{-1}$. This work points out the necessity of exploiting high-resolution IFS data to understand the multi-phase components of the ISM and the multiple kinematical components in the central regions of nearby galaxies.

astro-ph.GA↗

arXiv2019

Chronos: A NIR spectroscopic survey to target the most important phases of galaxy evolution across cosmic time

(Abridged summary) Responding to ESA's Voyage 2050 call to define the long-term plan for the future space missions that will address the astrophysics science questions during the 2035-2050 cycle, we propose a dedicated, ultra-deep spectroscopic survey in the near infrared (NIR), that will target a mass-limited sample of galaxies during two of the most fundamental epochs of cosmic evolution: the formation of the first galaxies (at z>6; cosmic dawn), and at the peak of galaxy formation activity (between redshift z=1 and 3; cosmic noon). By way of NIR observations, it is possible to study the Lyman-alpha region in the former, and the optical rest-frame in the latter, allowing us to extract fundamental observables such as gas and stellar kinematics, chemical abundances, and ages, providing a unique legacy database covering these two crucial stages of cosmic evolution. A dedicated, space-based facility will overcome the challenges faced by ground-based telescopes, no matter how large the aperture, or the reduced field of view and low multiplex factor of the best space-based instrument in the near future, namely NIRSpec at the JWST. Our project (codename Chronos) aims to produce about 1 million high quality spectra, with a high S/N in the continuum, where information about the underlying stellar populations is encoded.

astro-ph.GA↗

arXiv2019

The Search for Living Worlds and the Connection to Our Cosmic Origins

One of the most exciting scientific challenges is to detect Earth-like planets in the habitable zones of other stars in the galaxy and search for evidence of life. The ability to observe and characterise dozens of potentially Earth-like planets now lies within the realm of possibility due to rapid advances in key space and imaging technologies. The associated challenge of directly imaging very faint planets in orbit around nearby very bright stars is now well understood, with the key instrumentation also being perfected and developed. Such advances will allow us to develop large transformative telescopes, covering a broad UV-optical-IR spectral range, which can carry out the detailed research programmes designed to answer the questions we wish to answer: Carry out high contrast imaging surveys of nearby stars to search for planets within their habitable zones. Characterise the planets detected to determine masses and radii from photometric measurements. Through spectroscopic studies of their atmospheres and surfaces, search for habitability indicators and for signs of an environment that has been modified by the presence of life. Active studies of potential missions have been underway for a number of years. The latest of these is the Large UV Optical IR space telescope (LUVOIR), one of four flagship mission studies commissioned by NASA in support of the 2020 US Decadal Survey. LUVOIR, if selected, will be of interest to a wide scientific community and will be the only telescope capable of searching for and characterizing a sufficient number of exoEarths to provide a meaningful answer to the question - Are we alone?. This paper is a submission to the European Space Agency Voyage 2050 call for white papers outlining the case for an ESA contribution to a Large UVOIR telescope.

astro-ph.EP↗

arXiv2019

The VANDELS survey: the role of ISM and galaxy physical properties on the escape of Lyα emission in z $\sim$ 3.5 star-forming galaxies

We investigate the physical properties of a sample of 52 Lyα emitting galaxies in the VANDELS survey, with particular focus on the role of kinematics and neutral hydrogen column density in the escape and spatial distribution of Lyα photons. For these galaxies, we derive different physical properties and characterize the Lyα emission in terms of kinematics, EW, FWHM and spatial extension and then estimate the velocity of the neutral outflowing gas. We reproduce some of the well known correlations between Lyα EW and stellar mass, dust extinction and UV \b{eta} slope, in the sense that the emission line appears brighter in lower mass, less dusty and bluer galaxies. We do not find any correlation with the SED-derived star formation rate, while we find that galaxies with brighter Lyα tend to be more compact both in UV and in Lyα. Our data reveal a new interesting correlation between the Lyα velocity and the offset of the inter-stellar absorption lines with respect to the systemic redshift, in the sense that galaxies with larger inter-stellar medium (ISM) out-flow velocities show smaller Lyα velocity shifts. We interpret this relation in the context of the shell-model scenario, where the velocity of the ISM and the HI column density contribute together in determining the Lyα kinematics. In support to our interpretation, we observe that galaxies with high HI column densities have much more extended Lyα spatial profiles, a sign of increased scattering. However, we do not find any evidence that the HI column density is related to any other physical properties of the galaxies, although this might be due in part to the limited range of parameters that our sample spans.

astro-ph.GA↗

arXiv2019

The role of Active Galactic Nuclei in galaxy evolution: insights from space ultraviolet spectropolarimetry

This Astro2020 white paper summarizes the unknowns of active galactic nuclei (AGN) physics that could be unveiled thanks to a new, space-born, ultraviolet spectropolarimeter. The unique capabilities of high energy polarimetry would help us to understand the precise mechanisms of matter and energy transfer and supermassive black holes growth, together with the impact of AGN feedback on galaxy evolution.

astro-ph.HE↗

arXiv2018

Probing the structure and evolution of active galactic nuclei with the ultraviolet polarimeter POLLUX aboard LUVOIR

The ultraviolet (UV) polarization spectrum of nearby active galactic nuclei (AGN) is poorly known. The Wisconsin Ultraviolet Photo-Polarimeter Experiment and a handful of instruments on board the Hubble Space Telescope were able to probe the near- and mid-UV polarization of nearby AGN, but the far-UV band (from 1200 angs down to the Lyman limit at 912 angs) remains completely uncharted. In addition, the linewidth resolution of previous observations was at best 1.89 angs. Such a resolution is not sufficient to probe in detail quantum mechanical effects, synchrotron and cyclotron processes, scattering by electrons and dust grains, and dichroic extinction by asymmetric dust grains. Exploring those physical processes would require a new, high-resolution, broadband polarimeter with full ultraviolet-band coverage. In this context, we discuss the AGN science case for POLLUX, a high-resolution UV spectropolarimeter, proposed for the 15-meter primary mirror option of LUVOIR (a multi-wavelength space observatory concept being developed by the Goddard Space Flight Center and proposed for the 2020 Decadal Survey Concept Study).

astro-ph.IM↗

arXiv2018

The VANDELS ESO public spectroscopic survey

VANDELS is a uniquely-deep spectroscopic survey of high-redshift galaxies with the VIMOS spectrograph on ESO's Very Large Telescope (VLT). The survey has obtained ultra-deep optical (0.48 < lambda < 1.0 micron) spectroscopy of ~2100 galaxies within the redshift interval 1.0 < z < 7.0, over a total area of ~0.2 sq. degrees centred on the CANDELS UDS and CDFS fields. Based on accurate photometric redshift pre-selection, 85% of the galaxies targeted by VANDELS were selected to be at z>=3. Exploiting the red sensitivity of the refurbished VIMOS spectrograph, the fundamental aim of the survey is to provide the high signal-to-noise ratio spectra necessary to measure key physical properties such as stellar population ages, masses, metallicities and outflow velocities from detailed absorption-line studies. Using integration times calculated to produce an approximately constant signal-to-noise ratio (20 < t_int < 80 hours), the VANDELS survey targeted: a) bright star-forming galaxies at 2.4 < z < 5.5, b) massive quiescent galaxies at 1.0 < z < 2.5, c) fainter star-forming galaxies at 3.0 < z < 7.0 and d) X-ray/Spitzer-selected active galactic nuclei and Herschel-detected galaxies. By targeting two extragalactic survey fields with superb multi-wavelength imaging data, VANDELS will produce a unique legacy data set for exploring the physics underpinning high-redshift galaxy evolution. In this paper we provide an overview of the VANDELS survey designed to support the science exploitation of the first ESO public data release, focusing on the scientific motivation, survey design and target selection.

astro-ph.GA↗

arXiv2018

The VANDELS ESO public spectroscopic survey: observations and first data release

This paper describes the observations and the first data release (DR1) of the ESO public spectroscopic survey "VANDELS, a deep VIMOS survey of the CANDELS CDFS and UDS fields". VANDELS' main targets are star-forming galaxies at 2.4<z<5.5 and massive passive galaxies at 1<z<2.5. By adopting a strategy of ultra-long exposure times, from 20 to 80 hours per source, VANDELS is designed to be the deepest ever spectroscopic survey of the high-redshift Universe. Exploiting the red sensitivity of the VIMOS spectrograph, the survey has obtained ultra-deep spectra covering the wavelength 4800-10000 A with sufficient signal-to-noise to investigate the astrophysics of high-redshift galaxy evolution via detailed absorption line studies. The VANDELS-DR1 is the release of all spectra obtained during the first season of observations and includes data for galaxies for which the total (or half of the total) scheduled integration time was completed. The release contains 879 individual objects with a measured redshift and includes fully wavelength and flux-calibrated 1D spectra, the associated error spectra, sky spectra and wavelength-calibrated 2D spectra. We also provide a catalog with the essential galaxy parameters, including spectroscopic redshifts and redshift quality flags. In this paper we present the survey layout and observations, the data reduction and redshift measurement procedure and the general properties of the VANDELS-DR1 sample. We also discuss the spectroscopic redshift distribution, the accuracy of the photometric redshifts and we provide some examples of data products. All VANDELS-DR1 data are publicly available and can be retrieved from the ESO archive. Two further data releases are foreseen in the next 2 years with a final release scheduled for June 2020 which will include improved re-reduction of the entire spectroscopic data set. (abridged)

astro-ph.GA↗

arXiv2017

The distribution and ages of star clusters in the Small Magellanic Cloud: Constraints on the interaction history of the Magellanic Clouds

We present a new study of the spatial distribution and ages of the star clusters in the Small Magellanic Cloud (SMC). To detect and estimate the ages of the star clusters we rely on the new fully-automated method developed by Bitsakis et al. (2017). Our code detects 1319 star clusters in the central 18 deg$^{2}$ of the SMC we surveyed (1108 of which have never been reported before). The age distribution of those clusters suggests enhanced cluster formation around 240 Myr ago. It also implies significant differences in the cluster distribution of the bar with respect to the rest of the galaxy, with the younger clusters being predominantly located in the bar. Having used the same set-up, and data from the same surveys as for our previous study of the LMC, we are able to robustly compare the cluster properties between the two galaxies. Our results suggest that the bulk of the clusters in both galaxies were formed approximately 300 Myr ago, probably during a direct collision between the two galaxies. On the other hand, the locations of the young ($\le$50 Myr) clusters in both Magellanic Clouds, found where their bars join the HI arms, suggest that cluster formation in those regions is a result of internal dynamical processes. Finally, we discuss the potential causes of the apparent outside-in quenching of cluster formation that we observe in the SMC. Our findings are consistent with an evolutionary scheme where the interactions between the Magellanic Clouds constitute the major mechanism driving their overall evolution.

astro-ph.GA↗

arXiv2017

The MUSE Hubble Ultra Deep Field Survey: III. Testing photometric redshifts to 30th magnitude

We tested the performance of photometric redshifts for galaxies in the Hubble Ultra Deep field down to 30th magnitude. We compared photometric redshift estimates from three spectral fitting codes from the literature (EAZY, BPZ and BEAGLE) to high quality redshifts for 1227 galaxies from the MUSE integral field spectrograph. All these codes can return photometric redshifts with bias |Dzn|=|z-z_phot|/(1+z)<0.05 down to F775W=30 and spectroscopic incompleteness is unlikely to strongly modify this statement. We have, however, identified clear systematic biases in the determination of photometric redshifts: in the 0.4 3 they are systematically biased high by up to Dzn = 0.05, an offset that can in part be explained by adjusting the amount of intergalactic absorption applied. In agreement with previous studies we find little difference in the performance of the different codes, but in contrast to those we find that adding extensive ground-based and IRAC photometry actually can worsen photo-z performance for faint galaxies. We find an outlier fraction, defined through |Dzn|>0.15, of 8% for BPZ and 10% for EAZY and BEAGLE, and show explicitly that this is a strong function of magnitude. While this outlier fraction is high relative to numbers presented in the literature for brighter galaxies, they are very comparable to literature results when the depth of the data is taken into account. Finally, we demonstrate that while a redshift might be of high confidence, the association of a spectrum to the photometric object can be very uncertain and lead to a contamination of a few percent in spectroscopic training samples that do not show up as catastrophic outliers, a problem that must be tackled in order to have sufficiently accurate photometric redshifts for future cosmological surveys.

astro-ph.GA↗