-
Velocity recostruction with graph neural networks
Authors:
Hideki Tanimura,
Albert Bonnefous,
Jia Liu,
Sanmay Ganguly
Abstract:
In this work, we seek to improve the velocity reconstruction of clusters by using Graph Neural Networks -- a type of deep neural network designed to analyze sparse, unstructured data. In comparison to the Convolutional Neural Network (CNN) which is built for structured data such as regular grids, GNN is particularly suitable for analyzing galaxy catalogs. In our GNNs, galaxies as represented as no…
▽ More
In this work, we seek to improve the velocity reconstruction of clusters by using Graph Neural Networks -- a type of deep neural network designed to analyze sparse, unstructured data. In comparison to the Convolutional Neural Network (CNN) which is built for structured data such as regular grids, GNN is particularly suitable for analyzing galaxy catalogs. In our GNNs, galaxies as represented as nodes that are connected with edges. The galaxy positions and properties -- stellar mass, star formation rate, and total number of galaxies within 100~\mpc -- are combined to predict the line-of-sight velocity of the clusters. To train our networks, we use mock SDSS galaxies and clusters constructed from the Magneticum hydrodynamic simulations. Our GNNs reach a precision in reconstructed line-of-sight velocity of $Δv$=163 km/s, outperforming by $\approx$10\% the perturbation theory~($Δv$=181 km/s) or the CNN~($Δv$=179 km/s). The stellar mass provides additional information, improving the precision by $\approx$6\% beyond the position-only GNN, while other properties add little information. Our GNNs remain capable of reconstructing the velocity field when redshift-space distortion is included, with $Δv$=210 km/s which is again 10\% better than CNN with RSD. Finally, we find that even with an impressive, nearly 70\% increase in galaxy number density from SDSS to DESI, our GNNs only show an underwhelming 2\% improvement, in line with previous works using other methods. Our work demonstrates that, while the efficiency in velocity reconstruction may have plateaued already at SDSS number density, further improvements are still hopeful with new reconstruction models such as the GNNs studied here.
△ Less
Submitted 21 February, 2024;
originally announced February 2024.
-
Unravelling the structure of magnetised molecular clouds with SILCC-Zoom: sheets, filaments and fragmentation
Authors:
S. Ganguly,
S. Walch,
D. Seifried,
S. D. Clarke,
M. Weis
Abstract:
To what extent magnetic fields affect how molecular clouds (MCs) fragment and create dense structures is an open question. We present a numerical study of cloud fragmentation using the SILCC-Zoom simulations. These simulations follow the self-consistent formation of MCs in a few hundred parsec sized region of a stratified galactic disc; and include magnetic fields, self-gravity, supernova-driven t…
▽ More
To what extent magnetic fields affect how molecular clouds (MCs) fragment and create dense structures is an open question. We present a numerical study of cloud fragmentation using the SILCC-Zoom simulations. These simulations follow the self-consistent formation of MCs in a few hundred parsec sized region of a stratified galactic disc; and include magnetic fields, self-gravity, supernova-driven turbulence, as well as a non-equilibrium chemical network. To discern the role of magnetic fields in the evolution of MCs, we study seven simulated clouds, five with magnetic fields, and two without, with a maximum resolution of 0.1 parsec. Using a dendrogram we identify hierarchical structures which form within the clouds. Overall, the magnetised clouds have more mass in a diffuse envelope with a number density between 1-100 cm$^{-3}$. We find that six out of seven clouds are sheet-like on the largest scales, as also found in recent observations, and with filamentary structures embedded within, consistent with the bubble-driven MC formation mechanism. Hydrodynamic simulations tend to produce more sheet-like structures also on smaller scales, while the presence of magnetic fields promotes filament formation. Analysing cloud energetics, we find that magnetic fields are dynamically important for less dense, mostly but not exclusively atomic structures (typically up to $\sim 100 - 1000$~cm$^{-3}$), while the denser, potentially star-forming structures are energetically dominated by self-gravity and turbulence. In addition, we compute the magnetic surface term and demonstrate that it is generally confining, and some atomic structures are even magnetically held together. In general, magnetic fields delay the cloud evolution and fragmentation by $\sim$ 1 Myr.
△ Less
Submitted 17 July, 2023;
originally announced July 2023.
-
The Nature of the Motions of Multiphase Filaments in the Centers of Galaxy Clusters
Authors:
Shalini Ganguly,
Yuan Li,
Valeria Olivares,
Yuanyuan Su,
Francoise Combes,
Sampadaa Prakash,
Stephen Hamer,
Pierre Guillard,
Trung Ha
Abstract:
The intracluster medium (ICM) in the centers of galaxy clusters is heavily influenced by the ``feedback'' from supermassive black holes (SMBHs). Feedback can drive turbulence in the ICM and turbulent dissipation can potentially be an important source of heating. Due to the limited spatial and spectral resolutions of X-ray telescopes, direct observations of turbulence in the hot ICM have been chall…
▽ More
The intracluster medium (ICM) in the centers of galaxy clusters is heavily influenced by the ``feedback'' from supermassive black holes (SMBHs). Feedback can drive turbulence in the ICM and turbulent dissipation can potentially be an important source of heating. Due to the limited spatial and spectral resolutions of X-ray telescopes, direct observations of turbulence in the hot ICM have been challenging. Recently, we developed a new method to measure turbulence in the ICM using multiphase filaments as tracers. These filaments are ubiquitous in cluster centers and can be observed at very high resolution using optical and radio telescopes. We study the kinematics of the filaments by measuring their velocity structure functions (VSFs) over a wide range of scales in the centers of $\sim 10$ galaxy clusters. We find features of the VSFs that correlate with the SMBHs activities, suggesting that SMBHs are the main driver of gas motions in the centers of galaxy clusters. In all systems, the VSF is steeper than the classical Kolmogorov expectation and the slopes vary from system to system. One theoretical explanation is that the VSFs we have measured so far mostly reflect the motion of the driver (jets and bubbles) rather than the cascade of turbulence. We show that in Abell 1795, the VSF of the outer filaments far from the SMBH flattens on small scales to a Kolmogorov slope, suggesting that the cascade is only detectable farther out with the current telescope resolution. The level of turbulent heating computed at small scales is typically an order of magnitude lower than that estimated at the driving scale. Even though SMBH feedback heavily influences the kinematics of the ICM in cluster centers, the level of turbulence it drives is rather low, and turbulent heating can only offset $\lesssim10\%$ of the cooling loss, consistent with the findings of numerical simulations.
△ Less
Submitted 19 April, 2023;
originally announced April 2023.
-
The Virial Balance of Molecular Clumps and Cores in Colliding Magnetized Flows
Authors:
Michael Weis,
Stefanie Walch,
Daniel Seifried,
Shashwata Ganguly
Abstract:
We simulate the formation of molecular clouds in colliding flows of warm neutral medium with the adaptive mesh refinement code {\sc Flash}. We include a chemical network to treat heating and cooling and to follow the formation of molecular gas. For the forming molecular clumps and cores in four different simulations with varying initial magnetic field strength between 0.01 - 5$\,μ$G, we carry out…
▽ More
We simulate the formation of molecular clouds in colliding flows of warm neutral medium with the adaptive mesh refinement code {\sc Flash}. We include a chemical network to treat heating and cooling and to follow the formation of molecular gas. For the forming molecular clumps and cores in four different simulations with varying initial magnetic field strength between 0.01 - 5$\,μ$G, we carry out a full virial analysis including all time-independent surface and volume terms as well as the time-dependent term. The initial magnetic field strength influences the fragmentation properties of the forming cloud because it prohibits motions perpendicular to the field direction and hence alters, or even suppresses, the formation of filamentary substructures. Molecular clump and core formation occurs anyhow. As a result, with increasing field strength, we find more fragments with a smaller average mass; yet the initial field strength is dynamically not relevant for the fragments which constitute our molecular clumps and cores. %yet the magnetic field overall is dynamically negligible for the fragments which constitute our molecular clumps and cores. The molecular clumps are mostly unbound, probably transient objects, which seem to be weakly confined by ram pressure or thermal pressure, indicating that they are swept up by the turbulent flow. They experience significant fluctuations in the mass flux through their surface, indicating that the Eulerian reference frame gives rise to a dominant time-dependent term due to their ill-defined nature. We define the cores to encompass molecular gas, which is additionally highly shielded. Most cores are in gravitational-kinetic equipartition and are already well described by the common virial parameter $α_\mathrm{vir}$ (as can be seen from the Heyer relation), while some undergo minor dispersion by kinetic surface effects.
△ Less
Submitted 24 August, 2022;
originally announced August 2022.
-
SILCC-Zoom: the dynamic balance in molecular cloud substructures
Authors:
S. Ganguly,
S. Walch,
S. D. Clarke,
D. Seifried
Abstract:
How molecular clouds fragment and create the dense structures which go on to form stars is an open question. We investigate the relative importance of different energy terms (kinetic, thermal, magnetic, and gravity - both self-gravity and tidal forces) for the formation and evolution of molecular clouds and their sub-structures based on the SILCC-Zoom simulations. These simulations follow the self…
▽ More
How molecular clouds fragment and create the dense structures which go on to form stars is an open question. We investigate the relative importance of different energy terms (kinetic, thermal, magnetic, and gravity - both self-gravity and tidal forces) for the formation and evolution of molecular clouds and their sub-structures based on the SILCC-Zoom simulations. These simulations follow the self-consistent formation of cold molecular clouds down to scales of 0.1 pc from the diffuse supernova-driven interstellar medium in a stratified galactic disc. We study the time evolution of seven molecular clouds (five with magnetic fields and two without) for 1.5-2 Myr. Using a dendrogram, we identify hierarchical 3D sub-structures inside the clouds with the aim to understand their dynamics and distinguish between the theories of gravo-turbulent fragmentation and global hierarchical collapse. The virial analysis shows that the dense gas is indeed dominated by the interplay of gravity and turbulence, while magnetic fields and thermal pressure are only important for fluffy, atomic structures. Over time, gravitationally bound sub-structures emerge from a marginally bound medium (viral ratio $1 \leq α_{\rm vir}^{\rm vol} <2$) as a result of large-scale supernova-driven inflows rather than global collapse. A detailed tidal analysis shows that the tidal tensor is highly anisotropic. Yet the tidal forces are generally not strong enough to disrupt either large-scale or dense sub-structures but cause their deformation. By comparing tidal and crossing time scales, we find that tidal forces do not seem to be the main driver of turbulence within the molecular clouds.
△ Less
Submitted 5 April, 2022;
originally announced April 2022.
-
Symmetry Group Equivariant Architectures for Physics
Authors:
Alexander Bogatskiy,
Sanmay Ganguly,
Thomas Kipf,
Risi Kondor,
David W. Miller,
Daniel Murnane,
Jan T. Offermann,
Mariel Pettee,
Phiala Shanahan,
Chase Shimmin,
Savannah Thais
Abstract:
Physical theories grounded in mathematical symmetries are an essential component of our understanding of a wide range of properties of the universe. Similarly, in the domain of machine learning, an awareness of symmetries such as rotation or permutation invariance has driven impressive performance breakthroughs in computer vision, natural language processing, and other important applications. In t…
▽ More
Physical theories grounded in mathematical symmetries are an essential component of our understanding of a wide range of properties of the universe. Similarly, in the domain of machine learning, an awareness of symmetries such as rotation or permutation invariance has driven impressive performance breakthroughs in computer vision, natural language processing, and other important applications. In this report, we argue that both the physics community and the broader machine learning community have much to understand and potentially to gain from a deeper investment in research concerning symmetry group equivariant machine learning architectures. For some applications, the introduction of symmetries into the fundamental structural design can yield models that are more economical (i.e. contain fewer, but more expressive, learned parameters), interpretable (i.e. more explainable or directly mappable to physical quantities), and/or trainable (i.e. more efficient in both data and computational requirements). We discuss various figures of merit for evaluating these models as well as some potential benefits and limitations of these methods for a variety of physics applications. Research and investment into these approaches will lay the foundation for future architectures that are potentially more robust under new computational paradigms and will provide a richer description of the physical systems to which they are applied.
△ Less
Submitted 11 March, 2022;
originally announced March 2022.
-
On Synthetic Absorption Line Profiles of Thermally Driven Winds from Active Galactic Nuclei
Authors:
Shalini Ganguly,
Daniel Proga,
Tim Waters,
Randall C. Dannen,
Sergei Dyda,
Margherita Giustini,
Timothy Kallman,
John Raymond,
Jon Miller,
Paola Rodriguez Hidalgo
Abstract:
The warm absorbers observed in more than half of all nearby active galactic nuclei (AGN) are tracers of ionized outflows located at parsec scale distances from the central engine. If the smallest inferred ionization parameters correspond to plasma at a few $10^4$~K, then the gas undergoes a transition from being bound to unbound provided it is further heated to $\sim 10^6$~K at larger radii. Danne…
▽ More
The warm absorbers observed in more than half of all nearby active galactic nuclei (AGN) are tracers of ionized outflows located at parsec scale distances from the central engine. If the smallest inferred ionization parameters correspond to plasma at a few $10^4$~K, then the gas undergoes a transition from being bound to unbound provided it is further heated to $\sim 10^6$~K at larger radii. Dannen et al. recently discovered that under these circumstances, thermally driven wind solutions are unsteady and even show very dense clumps due to thermal instability. To explore the observational consequences of these new wind solutions, we compute line profiles based on the one-dimensional simulations of Dannen et al. We show how the line profiles from even a simple steady state wind solution depend on the ionization energy (IE) of absorbing ions, which is a reflection of the wind ionization stratification. To organize the diversity of the line shapes, we group them into four categories: weak Gaussians, saturated boxy profiles with and without an extended blue wing, and broad weak profiles. The lines with profiles in the last two categories are produced by ions with the highest IE that probe the fastest regions. Their maximum blueshifts agree with the highest flow velocities in thermally unstable models, both steady state and clumpy versions. In contrast, the maximum blueshifts of the most high IE lines in thermally stable models can be less than half of the actual solution velocities. Clumpy solutions can additionally imprint distinguishable absorption troughs at widely separated velocities.
△ Less
Submitted 15 May, 2021; v1 submitted 11 March, 2021;
originally announced March 2021.
-
When Freeze-out occurs due to a non-Boltzmann suppression: A study of degenerate dark sector
Authors:
Anirban Biswas,
Sougata Ganguly,
Sourov Roy
Abstract:
Exponential suppression or commonly known as the Boltzmann suppression in the number density of dark matter is the key ingredient for creating chemical imbalance prior to the usual thermal freeze-out. A degenerate/quasi-degenerate dark sector can experience a different exponential suppression in the number density analogous to the radioactive decay law leading to a delayed freeze-out mechanism of…
▽ More
Exponential suppression or commonly known as the Boltzmann suppression in the number density of dark matter is the key ingredient for creating chemical imbalance prior to the usual thermal freeze-out. A degenerate/quasi-degenerate dark sector can experience a different exponential suppression in the number density analogous to the radioactive decay law leading to a delayed freeze-out mechanism of dark matter known as the co-decaying dark matter. In this work, we study the dynamics of a multicomponent dark matter from thermally decoupled degenerate dark sector in a hidden U$(1)_{X}$ extension of the Standard Model. We compute the relic density of dark matter frozen-out through the co-decaying mechanism by solving four coupled Boltzmann equations. We demonstrate how temperature $T^\prime $ of the dark sector changes due to all types of $3\rightarrow 2$ and $2\rightarrow 2$ interactions along with the eternal expansion of the Universe. We find that $3\rightarrow 2$ interactions enhance $T^\prime$ by producing energetic particles in the dark sector while the excess heat is transferred by $2\rightarrow 2$ interactions to the entire dark sector. As the direct detection is possible only through the feeble portal couplings, we investigate the neutrino and $γ$-ray signals from dark matter annihilation via one step cascade processes and compare our results with the measured fluxes of atmospheric neutrinos by Super-Kamiokande and diffuse $γ$-rays by Fermi-LAT, EGRET, INTEGRAL collaborations. We find that the present scenario easily evades all the existing bounds from atmospheric neutrino and diffuse $γ$-ray observations for degenerate dark sector. However, the constraints are significant for quasi degenerate scenario.
△ Less
Submitted 18 June, 2021; v1 submitted 4 November, 2020;
originally announced November 2020.
-
On the wind-driven relaxation cycle in accretion disks
Authors:
Shalini Ganguly,
Daniel Proga
Abstract:
A disk wind can cause perturbations that propagate throughout the disk via diffusive processes. On reaching the inner disk, these perturbations can change the disk luminosity, which in turn, can change the wind mass loss rate, $\dot{M}_w$. It has been argued that this so-called "wind driven relaxation cycle" might explain the observed variability in some disk accreting objects. Here, we study the…
▽ More
A disk wind can cause perturbations that propagate throughout the disk via diffusive processes. On reaching the inner disk, these perturbations can change the disk luminosity, which in turn, can change the wind mass loss rate, $\dot{M}_w$. It has been argued that this so-called "wind driven relaxation cycle" might explain the observed variability in some disk accreting objects. Here, we study the response of the innermost mass accretion rate $\dot{M}_a$ to the loss of matter at different rates and radii. We allow the wind launching radius, $R_L$, to scale with $\dot{M}_a$. We computed a grid of time-dependent models for various $\dot{M}_w$-$\dot{M}_a$ and $R_{L}$-$\dot{M}_a$ dependencies. We find that the disk behaviour significantly differs for the 'variable $R_L$' case compared to the 'fixed $R_L$' case. In particular, much stronger winds are required to destabilize the disk in the former than the latter case. However, the $\dot{M}_a$ amplitude does not grow significantly even for unstable cases because the oscillations saturate at a low level either due to disk depletion or due to the wind being launched at very small radii, or both. This result implies that disk winds are unlikely to be responsible for state transitions as those require large changes in the inner disk. Despite modest changes at the inner disk regions, the disk surface density at large radii can vary with a large amplitude, i.e., from 0 to a few factors of the steady state value. This dramatic variation of the outer disk could have observable consequences.
△ Less
Submitted 11 January, 2020;
originally announced January 2020.
-
Gravity-Induced Geometric Phases and Entanglement in Spinors and Neutrinos: Gravitational Zeeman Effect
Authors:
Banibrata Mukhopadhyay,
Soumya Kanti Ganguly
Abstract:
We show Zeeman-like splitting in the energy of spinors propagating in a background gravitational field, analogous to the spinors in an electromagnetic field, otherwise termed the Gravitational Zeeman Effect. These spinors are also found to acquire a geometric phase, in a similar way as they do in the presence of magnetic fields. However, in a gravitational background, the Aharonov-Bohm type effect…
▽ More
We show Zeeman-like splitting in the energy of spinors propagating in a background gravitational field, analogous to the spinors in an electromagnetic field, otherwise termed the Gravitational Zeeman Effect. These spinors are also found to acquire a geometric phase, in a similar way as they do in the presence of magnetic fields. However, in a gravitational background, the Aharonov-Bohm type effect, in addition to Berry-like phase, arises. Based on this result, we investigate geometric phases acquired by neutrinos propagating in a strong gravitational field. We also explore entanglement of neutrino states due to gravity, which could induce neutrino-antineutrino oscillation in the first place. We show that entangled states also acquire geometric phases which are determined by the relative strength between gravitational field and neutrino masses.
△ Less
Submitted 1 October, 2020; v1 submitted 28 February, 2018;
originally announced February 2018.
-
Statistical Significance of spectral lag transition in GRB 160625B
Authors:
Shalini Ganguly,
Shantanu Desai
Abstract:
Recently Wei et al (arXiv:1612.09425) have found evidence for a transition from positive time lags to negative time lags in the spectral lag data of GRB 160625B. They have fit these observed lags to a sum of two components: an assumed functional form for intrinsic time lag due to astrophysical mechanisms and an energy-dependent speed of light due to quadratic and linear Loren tz invariance violati…
▽ More
Recently Wei et al (arXiv:1612.09425) have found evidence for a transition from positive time lags to negative time lags in the spectral lag data of GRB 160625B. They have fit these observed lags to a sum of two components: an assumed functional form for intrinsic time lag due to astrophysical mechanisms and an energy-dependent speed of light due to quadratic and linear Loren tz invariance violation (LIV) models. Here, we examine the statistical significance of the evidence for a transition to nega tive time lags. Such a transition, even if present in GRB 160625B, cannot be due to an energy dependent speed of light as th is would contradict previous limits by some 3-4 orders of magnitude, and must therefore be of intrinsic astrophysical origin . We use three different model comparison techniques: a frequentist test and two information based criteria (AIC and BIC). From the frequentist model comparison test, we find that the evidence for transition in the spectral lag data is favored at $3.05σ$ and $3.74σ$ for the linear and quadratic models respectively. We find that $Δ$AIC and $Δ$BIC have values $\gtrsim$ 10 for the spectral lag transition that was motivated as being due to quadratic Lorentz invariance vio lating model pointing to "decisive evidence". We note however that none of the three models (including the model of intr insic astrophysical emission) provide a good fit to the data.
△ Less
Submitted 30 July, 2017; v1 submitted 5 June, 2017;
originally announced June 2017.
