arXiv subjects

Daniel López-Cano

Publications and source records attributed to Daniel López-Cano.

4 recordsLinked to original sources

arXiv2026

Where the Forest Goes Dark: Halo-centered Characterization of High Column Density Systems with IllustrisTNG

High column density systems of neutral hydrogen (HCDs) --- Lyman limit systems (LLSs), sub-damped and damped Ly$α$ absorbers (DLAs) --- contaminate both the 3D and the 1D statistics of the Ly$α$ forest. Most DLAs are masked out from the analyses, but cleaning algorithms are not perfect, and weaker LLSs are individually undetectable, so the residual contamination is absorbed into nuisance parameters that dilute the cosmological constraining power. Simulations provide the ideal setting to study this effect, allowing absorption features to be directly traced back to the gas producing them. We identify high-column-density structures on the gas cells of the TNG50 hydrodynamical simulation, deblend them in velocity space, and split every sightline into HCD-only and forest-only spectra. Applied to halos at $z \simeq 3$ out to 50 virial radii, it yields each class's covering fraction against impact parameter $b$ and halo mass. The impact parameter defines the type of absorption appearing in the spectrum: DLAs give way to sub-DLAs at $b \approx 0.11\,R_{200c}$, sub-DLAs to LLSs at $0.3$, and LLSs to the Ly$α$-forest at $0.5$, which covers $\simeq 78\%$ of sightlines at $R_{200c}$. In units of $R_{200c}$ the sequence is nearly mass-independent over three decades in mass, so the virial radius sets the scale of the neutral gas distribution. A single Voigt component recovers the column density of essentially every damped system, but not of LLSs, whose absorption features often arise from several separate contributions from gas structures along the sightline. Our detailed characterization of HCDs gives the first step to the construction of advanced techniques to directly forward-model their contribution to Ly$α$ spectra.

astro-ph.CO↗

arXiv2026

J-PAS: Semi-Supervised Sim-to-Obs Transfer for Robust Star--Galaxy--Quasar Classification

Modern studies in astrophysics and cosmology increasingly rely on simulations and cross-survey analyses, yet differences in data generation, instrumentation, calibration, and unmodeled physics introduce distribution mismatches between datasets (``domain shift''). In machine-learning pipelines, this occurs when the joint distribution of inputs and labels differs between the training (source) and application (target) domains, causing source-trained models to underperform on the target. Transfer learning and domain adaptation provide principled ways to mitigate this effect. We study a concrete simulation-to-observation case: semi-supervised domain adaptation (SSDA) to transfer a four-class spectral classifier -- high-redshift quasars, low-redshift quasars, galaxies, and stars -- from J-PAS mock catalogs based on DESI spectra to real J-PAS observations. Our pipeline pretrains on abundant labeled DESI$\rightarrow$J-PAS mocks and adapts to the target domain using a small labeled J-PAS subset. We benchmark SSDA against two baselines: a J-PAS--only supervised model trained with the same target-label budget, and a mocks-only model evaluated on held-out J-PAS data. On this held-out J-PAS data, SSDA achieves a macro-F1 score (balancing precision and recall) of $0.82$ and an overall true positive rate of $0.89$, compared to $0.79/0.85$ for the J-PAS--only baseline and $0.73/0.87$ for the mocks-only model. The gains are driven primarily by improved quasar classification, especially in the high-redshift subclass ($\mathrm{F1}=0.66$ vs.\ $0.55/0.37$), yielding better-calibrated candidate lists for spectroscopic targeting (e.g., WEAVE-QSO) and AGN searches. This study shows how modest target supervision enables robust, data-efficient simulation-to-observation transfer when simulations are plentiful but target labels are scarce.

astro-ph.IM↗

arXiv2024

Characterizing Structure Formation through Instance Segmentation

Dark matter haloes form from small perturbations to the almost homogeneous density field of the early universe. Although it is known how large these initial perturbations must be to form haloes, it is rather poorly understood how to predict which particles will end up belonging to which halo. However, it is this process that determines the Lagrangian shape of protohaloes and is therefore essential to understand their mass, spin and formation history. Here, we present a machine-learning framework to learn how the protohalo regions of different haloes emerge from the initial density field. This involves one neural network to distinguish semantically which particles become part of any halo and a second neural network that groups these particles by halo membership into different instances. This instance segmentation is done through the Weinberger method, in which the network maps particles into a pseudo-space representation where different instances can be distinguished easily through a simple clustering algorithm. Our model reliably predicts the masses and Lagrangian shapes of haloes object-by-object, as well as summary statistics like the halo-mass function. We find that our model extracts information close to optimal by comparing it to the degree of agreement between two N-body simulations with slight differences in their initial conditions. We publish our model open-source and suggest that it can be used to inform analytical methods of structure formation by studying the effect of systematic manipulations of the initial conditions.

astro-ph.CO↗

arXiv2022

The cosmology dependence of the concentration-mass-redshift relation

The concentrations of dark matter haloes provide crucial information about their internal structure and how it depends on mass and redshift -- the so-called concentration-mass-redshift relation, denoted $c(M,z)$. We present here an extensive study of the cosmology-dependence of $c(M,z)$ that is based on a suite of 72 gravity-only, full N-body simulations in which the following cosmological parameters were varied: $σ_{8}$, $Ω_{\mathrm{M}}$, $Ω_{\mathrm{b}}$, $n_{\mathrm{s}}$, $h$, $M_ν$, $w_{0}$ and $w_{\mathrm{a}}$. We characterize the impact of these parameters on concentrations for different halo masses and redshifts. In agreement with previous works, and for all cosmologies studied, we find that there exists a tight correlation between the characteristic densities of dark matter haloes within their scale radii, $r_{-2}$, and the critical density of the Universe at a suitably defined formation time. This finding, when combined with excursion set modelling of halo formation histories, allows us to accurately predict the concentrations of dark matter haloes as a function of mass, redshift, and cosmology. We use our simulations to test the reliability of a number of published models for predicting halo concentration and highlight when they succeed or fail to reproduce the cosmological $c(M,z)$ relation.

astro-ph.CO↗