Even if massive ($10\,M_\odot \lesssim M \lesssim 10^4 M_\odot$) primordial black holes (PBHs) can only account for a small fraction of the dark matter (DM) in the universe, they may still be responsible for a sizable fraction of the coalescence events measured by LIGO/Virgo, and/or act as progenitors of the supermassive black holes (SMBHs) observed already at high redshift ($z\gtrsim 6$). In presence of a dominant, non-PBH DM component, the bounds set by CMB via an altered ionization history are modified. We revisit the cosmological accretion of a DM halo around PBHs via toy models and dedicated numerical simulations, deriving updated CMB bounds which also take into account the last Planck data release. We prove that these constraints dominate over other constraints available in the literature at masses $M\gtrsim 20-50\,M_\odot$ (depending on uncertainty in accretion physics), reaching the level $f_{\rm PBH}<3\times 10^{-9}$ around $M\sim 10^{4}\,M_\odot$. These tight bounds are nonetheless consistent with the hypothesis of a primordial origin of the SMBH massive seeds.

N-body simulations are essential tools in physical cosmology to understand the large-scale structure (LSS) formation of the Universe. Large-scale simulations with high resolution are important for exploring the substructure of universe and for determining fundamental physical parameters like neutrino mass. However, traditional particle-mesh (PM) based algorithms use considerable amounts of memory, which limits the scalability of simulations. Therefore, we designed a two-level PM algorithm CUBE towards optimal performance in memory consumption reduction. By using the fixed-point compression technique, CUBE reduces the memory consumption per N-body particle toward 6 bytes, an order of magnitude lower than the traditional PM-based algorithms. We scaled CUBE to 512 nodes (20,480 cores) on an Intel Cascade Lake based supercomputer with $\simeq$95\% weak-scaling efficiency. This scaling test was performed in "Cosmo-$\pi$" -- a cosmological LSS simulation using $\simeq$4.4 trillion particles, tracing the evolution of the universe over $\simeq$13.7 billion years. To our best knowledge, Cosmo-$\pi$ is the largest completed cosmological N-body simulation. We believe CUBE has a huge potential to scale on exascale supercomputers for larger simulations.

The cosmic neutrino background is an important component of the Universe that is difficult to include in cosmological simulations due to the extremely large velocity dispersion of neutrino particles. We develop a new approach to simulate cosmic neutrinos that decomposes the Fermi-Dirac phase space into shells of constant speed and then evolves those shells using hydrodynamic equations. These collisionless hydrodynamic equations are chosen to match linear theory, free particle evolution and allow for superposition. We implement this method into the information-optimized cosmological $N$-body code CUBE and demonstrate that neutrino perturbations can be accurately resolved to at least $k\sim1\ h/$Mpc. This technique allows for neutrino memory requirements to be decreased by up to $\sim 10^3$ compared to traditional $N$-body methods.

We analytically model the non-linear effects induced by massive neutrinos on the total matter power spectrum using the halo model reaction framework of Cataneo et al. 2019. In this approach the halo model is used to determine the relative change to the matter power spectrum caused by new physics beyond the concordance cosmology. Using standard fitting functions for the halo abundance and the halo mass-concentration relation, the total matter power spectrum in the presence of massive neutrinos is predicted to percent-level accuracy, out to $k=10 \, h \, {\rm Mpc}^{-1}$. We find that refining the prescriptions for the halo properties using $N$-body simulations improves the recovered accuracy to better than 1%. This paper serves as another demonstration for how the halo model reaction framework, in combination with a single suite of standard $\Lambda$CDM simulations, can recover percent-level accurate predictions for beyond-$\Lambda$CDM matter power spectra, well into the non-linear regime.

Cold dark matter (CDM) could be composed of primordial black holes (PBH) in addition to or instead of more orthodox weakly interacting massive particle dark matter (PDM). We study the formation of the first structures in such $\Lambda$PBH cosmologies using $N$-body simulations evolved from deep in the radiation era to redshift 99. When PBH are only a small component of the CDM, they are clothed by PDM to form isolated halos. On the other hand, when PBH make most of the CDM, halos can also grow via clustering of many PBH. We find that the halo mass function is well modelled via Poisson statistics assuming random initial conditions. We quantify the nonlinear velocities induced by structure formation and find that they are too small to significantly impact CMB constraints. A chief challenge is how best to extrapolate our results to lower redshifts relevant for some observational constraints.

Baryonic feedback can significantly modify the spatial distribution of matter on small scales and create a bulk relative velocity between the dominant cold dark matter and the hot gas. We study the consequences of such bulk motions using two high resolution hydrodynamic simulations, one with no feedback and one with very strong feedback. We find that relative velocities of order $100\ \kms$ are produced in the strong feedback simulation whereas it is much smaller when there is no feedback. Such relative motions induce dipole distortions to the gas, which we quantify by computing the dipole correlation function. We find halo coordinates and velocities are systematically changed in the direction of the relative velocity. Finally, we discuss potential to observe the relative velocity via large scale structure, Sunyaev-Zel'dovich and line emission measurements. Given the nonlinear nature of this effect, it should next be studied in simulations with different feedback implementations/strengths to determine the available model space.

Constraining neutrino mass remains an elusive challenge in modern physics. Precision measurements are expected from several upcoming cosmological probes of large-scale structure. Achieving this goal relies on an equal level of precision from theoretical predictions of neutrino clustering. Numerical simulations of the non-linear evolution of cold dark matter and neutrinos play a pivotal role in this process. We incorporate neutrinos into the cosmological N-body code CUBEP3M and discuss the challenges associated with pushing to the extreme scales demanded by the neutrino problem. We highlight code optimizations made to exploit modern high performance computing architectures and present a novel method of data compression that reduces the phase-space particle footprint from 24 bytes in single precision to roughly 9 bytes. We scale the neutrino problem to the Tianhe-2 supercomputer and provide details of our production run, named TianNu, which uses 86% of the machine (13,824 compute nodes). With a total of 2.97 trillion particles, TianNu is currently the world's largest cosmological N-body simulation and improves upon previous neutrino simulations by two orders of magnitude in scale. We finish with a discussion of the unanticipated computational challenges that were encountered during the TianNu runtime.

Reconstruction techniques are commonly used in cosmology to reduce complicated nonlinear behaviours to a more tractable linearized system. We study a new reconstruction technique that uses the Moving-Mesh algorithm to estimate the displacement field from nonlinear matter distribution. We show the performance of this new technique by quantifying its ability to reconstruct linear modes. We study the cumulative Fisher information $I(<k_n)$ about the initial matter power spectrum in the matter power spectra in 130 $N$-body simulations before and after reconstruction, and find that the nonlinear plateau of $I(<k_n)$ is increased by a factor of $\sim 50$ after reconstruction, from $I \simeq 2.5 \times 10^{-5} /({\rm Mpc}/h)^3$ to $I \simeq 1.3 \times 10^{-3}/({\rm Mpc}/h)^3$ at large $k$. This result includes the decorrelation between initial and final fields, which has been neglected in some previous studies. We expect this technique to be beneficial to problems such as baryonic acoustic oscillations, redshift space distortions and cosmic neutrinos that rely on accurately disentangling nonlinear evolution from underlying linear effects.

Measurements of neutrino mass in cosmological observations rely on two point statistics that are hindered by significant degeneracies with the optical depth and galaxy bias. The relative velocity effect between cold dark matter and neutrinos induces a large scale dipole into the matter density field and may be able to provide orthogonal constraints to standard techniques. We numerically investigate this dipole in the TianNu Simulation, which contains cold dark matter and 50 meV neutrinos. We first compute the dipole using a new linear response technique where we treat the displacement caused by the relative velocity as a phase in Fourier space and then integrate the matter power spectrum over redshift. Then, we compute the dipole numerically in real space using the simulation density and velocity fields. We find excellent agreement between the linear response and N-body methods. Utilizing the dipole as an observational tool will require two tracers of the matter distribution that are differently biased with respect to the neutrino density.

We present a description of cosmic neutrinos as a dispersive fluid. In this approach, the neutrino phase space is reduced to density and velocity fields alongside a scale-dependent sound speed. This sound speed depends on redshift, the initial neutrino phase space density and the cold dark matter gravitational potential. The latter is a new coupling between neutrinos and large scale structure not described by previous fluid approaches. We compute the sound speed in linear theory and find that it asymptotes to constants at small and large scales regardless of the gravitational potential. By comparing with neutrino N-body simulations, we measure the small scale sound speed and find it to be lower than linear theory predictions. This allows for an explanation of the discrepency between N-body and linear response predictions for the neutrino power spectrum: neutrinos are still driven predominantly by the cold dark matter, but the sound speed on small scales is not stable to perturbations and decreases. Finally, we present a calibrated model for the neutrino power spectrum that requires no additional integrations outside of standard Boltzmann codes.

Astrophysical techniques have pioneered the discovery of neutrino mass properties. Current cosmological observations give an upper bound on neutrino masses by attempting to disentangle the small neutrino contribution from the sum of all matter using precise theoretical models. We discover the differential neutrino condensation effect in our TianNu N-body simulation. Neutrino masses can be inferred using this effect by comparing galaxy properties in regions of the universe with different neutrino relative abundance (i.e. the local neutrino to cold dark matter density ratio). In "neutrino-rich"' regions, more neutrinos can be captured by massive halos compared to "neutrino-poor" regions. This effect differentially skews the halo mass function and opens up the path to independent neutrino mass measurements in current or future galaxy surveys.

The relic neutrinos are expected to acquire a bulk relative velocity with respect to the dark matter at low redshifts, and neutrino wakes are expected to develop downstream of the dark matter halos. We propose a method of measuring the neutrino mass based on this mechanism. This neutrino wake will cause a dipole distortion of the galaxy-galaxy lensing pattern. This effect could be detected by combining upcoming lensing surveys with a low redshift galaxy survey or a 21 cm intensity mapping survey, which can map the neutrino flow field. The data obtained with LSST and Euclid should enable us to make a positive detection if the three neutrino masses are quasidegenerate with each neutrino mass of $\sim$0.1 eV, and a future high precision 21 cm lensing survey would allow the normal hierarchy and inverted hierarchy cases to be distinguished, and even the right-handed Dirac neutrinos may be detectable.

We present a new technique to measure neutrino masses using their flow field relative to dark matter. Present day streaming motions of neutrinos relative to dark matter and baryons are several hundred km/s, comparable with their thermal velocity dispersion. This results in a unique dipole anisotropic distortion of the matter-neutrino cross power spectrum, which is observable through the dipole distortion in the cross correlation of different galaxy populations. Such a dipole vanishes if not for this relative velocity and so it is a clean signature for neutrino mass. We estimate the size of this effect and find that current and future galaxy surveys may be sensitive to these signature distortions.

Discovering the mass of neutrinos is a principle goal in high energy physics and cosmology. In addition to cosmological measurements based on two-point statistics, the neutrino mass can also be estimated by observations of neutrino wakes resulting from the relative motion between dark matter and neutrinos. Such a detection relies on an accurate reconstruction of the dark matter-neutrino relative velocity which is affected by non-linear structure growth and galaxy bias. We investigate our ability to reconstruct this relative velocity using large N-body simulations where we evolve neutrinos as distinct particles alongside the dark matter. We find that the dark matter velocity power spectrum is overpredicted by linear theory whereas the neutrino velocity power spectrum is underpredicted. The magnitude of the relative velocity observed in the simulations is found to be lower than what is predicted in linear theory. Since neither the dark matter nor the neutrino velocity fields are directly observable from galaxy or 21 cm surveys, we test the accuracy of a reconstruction algorithm based on halo density fields and linear theory. Assuming prior knowledge of the halo bias, we find that the reconstructed relative velocities are highly correlated with the simulated ones with correlation coefficients of 0.94, 0.93, 0.91 and 0.88 for neutrinos of mass 0.05, 0.1, 0.2 and 0.4 eV. We confirm that the relative velocity field reconstructed from large scale structure observations such as galaxy or 21 cm surveys can be accurate in direction and, with appropriate scaling, magnitude.