T-ReX: a graph-based filament detection method. (arXiv:1912.00732v2 [astro-ph.CO] UPDATED)
<a href="http://arxiv.org/find/astro-ph/1/au:+Bonnaire_T/0/1/0/all/0/1">T. Bonnaire</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Aghanim_N/0/1/0/all/0/1">N. Aghanim</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Decelle_A/0/1/0/all/0/1">A. Decelle</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Douspis_M/0/1/0/all/0/1">M. Douspis</a>
Numerical simulations and observations show that galaxies are not uniformly
distributed in the universe but, rather, they are spread across a filamentary
structure. In this large-scale pattern, highly dense regions are linked
together by bridges and walls, all of them surrounded by vast, nearly-empty
areas. While nodes of the network are widely studied in the literature,
simulations indicate that half of the mass budget comes from a more diffuse
part of the network, which is made up of filaments. In the context of recent
and upcoming large galaxy surveys, it becomes essential that we identify and
classify features of the Cosmic Web in an automatic way in order to study their
physical properties and the impact of the cosmic environment on galaxies and
their evolution.
In this work, we propose a new approach for the automatic retrieval of the
underlying filamentary structure from a 2D or 3D galaxy distribution using
graph theory and the assumption that paths that link galaxies together with the
minimum total length highlight the underlying distribution. To obtain a
smoothed version of this topological prior, we embedded it in a Gaussian
mixtures framework. In addition to a geometrical description of the pattern, a
bootstrap-like estimate of these regularised minimum spanning trees allowed us
to obtain a map characterising the frequency at which an area of the domain is
crossed. Using the distribution of halos derived from numerical simulations, we
show that the proposed method is able to recover the filamentary pattern in a
2D or 3D distribution of points with noise and outliers robustness with a few
comprehensible parameters.
Numerical simulations and observations show that galaxies are not uniformly
distributed in the universe but, rather, they are spread across a filamentary
structure. In this large-scale pattern, highly dense regions are linked
together by bridges and walls, all of them surrounded by vast, nearly-empty
areas. While nodes of the network are widely studied in the literature,
simulations indicate that half of the mass budget comes from a more diffuse
part of the network, which is made up of filaments. In the context of recent
and upcoming large galaxy surveys, it becomes essential that we identify and
classify features of the Cosmic Web in an automatic way in order to study their
physical properties and the impact of the cosmic environment on galaxies and
their evolution.
In this work, we propose a new approach for the automatic retrieval of the
underlying filamentary structure from a 2D or 3D galaxy distribution using
graph theory and the assumption that paths that link galaxies together with the
minimum total length highlight the underlying distribution. To obtain a
smoothed version of this topological prior, we embedded it in a Gaussian
mixtures framework. In addition to a geometrical description of the pattern, a
bootstrap-like estimate of these regularised minimum spanning trees allowed us
to obtain a map characterising the frequency at which an area of the domain is
crossed. Using the distribution of halos derived from numerical simulations, we
show that the proposed method is able to recover the filamentary pattern in a
2D or 3D distribution of points with noise and outliers robustness with a few
comprehensible parameters.
http://arxiv.org/icons/sfx.gif
