Explore the vast range of titles available.

To tackle the still unsolved and fundamental problem of the role of Active Galactic Nuclei (AGN) feedback in shaping galaxies, in this work we implement a new physical treatment of AGN-driven winds into our semi-analytic model of galaxy formation. To each galaxy in our model, we associate solutions for the outflow expansion and the mass outflow rates in different directions, depending on the AGN luminosity, on the circular velocity of the host halo, and on gas content of the considered galaxy. To each galaxy we also assign an effective radius derived from energy conservation during merger events, and a stellar velocity dispersion self-consistently computed via Jeans modelling. We derive all the main scaling relations between Black hole (BH) mass and total/bulge stellar mass, velocity dispersion, host halo dark matter mass, and star formation efficiency. We find that our improved AGN feedback mostly controls the dispersion around the relations but plays a subdominant role in shaping slopes and/or normalizations of the scaling relations. Including possible limited-resolution selection biases in the model provides better agreement with the available data. The model does not point to any more fundamental galactic property linked to BH mass, with velocity dispersion playing a similar role with respect to stellar mass, in tension with present data. In line with other independent studies carried out on comprehensive semi-analytic and hydrodynamic galaxy-BH evolution models, our current results signal either an inadequacy of present cosmological models of galaxy formation in fully reproducing the local scaling relations, in terms of both shape and residuals, and/or point to an incompleteness issue affecting the local sample of dynamically-measured BHs.

The fusion research facility ITER is currently being assembled to demonstrate that fusion can be used for industrial energy production, while several other programmes across the world are also moving forward, such as EU-DEMO, CFETR, SPARC and STEP. The high engineering complexity of a tokamak makes it an extremely challenging device to optimise, and test-based optimisation would be too slow and too costly. Instead, digital design and optimisation must be favored, which requires strongly-coupled suites of High-Performance Computing calculations. In this context, having surrogate models to provide quick estimates with uncertainty quantification is essential to explore and optimise new design options. Furthermore, these surrogates can in turn be used to accelerate simulations in the first place. This is the case of Parareal, a time-parallelisation method that can speed-up large HPC simulations, where the coarse-solver can be replaced by a surrogate. A novel framework, Neural-Parareal, is developed to integrate the training of neural operators dynamically as more data becomes available. For a given input-parameter domain, as more simulations are being run with Parareal, the large amount of data generated by the algorithm is used to train new surrogate models to be used as coarse-solvers for future Parareal simulations, leading to progressively more accurate coarse-solvers, and thus higher speed-up. It is found that such neural network surrogates can be much more effective than traditional coarse-solver in providing a speed-up with Parareal. This study is a demonstration of the convergence of HPC and AI which simply has to become common practice in the world of digital engineering design.

The inclusion of high-fidelity simulations of SOL turbulence and transient MHD events such as ELMs in highly iterative applications remains computationally prohibitive, limiting their use in design and control workflows. Understanding these phenomena is vital, as they govern heat flux on plasma-facing components, influencing reactor performance and material lifetime. This study explored FNOs as surrogate models to accelerate plasma simulations from the JOREK MHD and STORM turbulence codes. FNOs were trained on single-step rollouts and evaluated in terms of long-term predictive accuracy in an auto-regressive manner. To mitigate the computational burden of dataset generation, a transfer learning strategy was explored, leveraging low-fidelity simulations to improve performance on high-fidelity datasets. These results showed that FNOs effectively captured initial plasma evolution, including blob movement and density source localization. However, long rollouts accumulated errors and exhibited sensitivity to certain physical phenomena, leading to non-monotonic error spikes. Transfer learning significantly reduced errors for small dataset sizes and short rollouts, achieving an order-of-magnitude reduction when transfering from low- to high-fidelity datasets. However, its effectiveness diminished with longer rollouts and larger dataset sizes, especially when applied to datasets with significantly different dynamics. Attempts to transfer models to previously unseen variables in simulations were unsuccessful, underscoring the limitations in this context. These findings demonstrate the promise of NOs for accelerating fusion-relevant PDE simulations. However, they also highlight key challenges: improving long-term accuracy to mitigate error accumulation, capturing critical physical behaviors, and developing robust surrogates that effectively leverage multi-fidelity, multi-physics datasets.

Simulation-based plasma scenario development, optimization and control are crucial elements towards the successful deployment of next-generation experimental tokamaks and Fusion power plants. Current simulation codes require extremely intensive use of HPC resources that make them unsuitable for iterative or real time applications. Neural network based surrogate models of expensive simulators have been proposed to speed up such costly workflows. Current efforts in this direction in the Fusion community are mostly limited to point estimates of quantities of interest or simple 1D PDE models, with a few notable exceptions. While the AI literature on methods for neural PDE surrogate models is rich, performance benchmarks for Fusion-relevant 2D fields has so far remained flimited. In this work neural PDE surrogates are trained for the JOREK MHD code and the STORM scrape-off layer code using the PDEArena library (https://github.com/microsoft/pdearena). The performance of these surrogate models is investigated as a function of training set size as well as for long-term predictions. The performance of surrogate models that are trained on either one variable or multiple variables at once is also considered. It is found that surrogates that are trained on more data perform best for both long- and short-term predictions. Additionally, surrogate models trained on multiple variables achieve higher accuracy and more stable performance. Downsampling the training set in time may provide stability in the long term at the expense of the short term predictive capability, but visual inspection of the resulting fields suggests that multiple metrics should be used to evaluate performance.

Observational systematics complicate comparisons with theoretical models limiting understanding of galaxy evolution. In particular, different empirical determinations of the stellar mass function imply distinct mappings between the galaxy and halo masses, leading to diverse galaxy evolutionary tracks. Using our state-of-the-art STatistical sEmi-Empirical modeL, STEEL, we show fully self-consistent models capable of generating galaxy growth histories that simultaneously and closely agree with the latest data on satellite richness and star-formation rates at multiple redshifts and environments. Central galaxy histories are generated using the central halo mass tracks from state-of-the-art statistical dark matter accretion histories coupled to abundance matching routines. We show that too flat high-mass slopes in the input stellar-mass-halo-mass relations as predicted by previous works, imply non-physical stellar mass growth histories weaker than those implied by satellite accretion alone. Our best-fit models reproduce the satellite distributions at the largest masses and highest redshifts probed, the latest data on star formation rates and its bi-modality in the local Universe, and the correct fraction of ellipticals. Our results are important to predict robust and self-consistent stellar-mass-halo-mass relations and to generate reliable galaxy mock catalogues for the next generations of extra-galactic surveys such as Euclid and LSST.

We present STEEL a STatistical sEmi-Empirical modeL designed to probe the distribution of satellite galaxies in groups and clusters. Our fast statistical methodology relies on tracing the abundances of central and satellite haloes via their mass functions at all cosmic epochs with virtually no limitation on cosmic volume and mass resolution. From mean halo accretion histories and subhalo mass functions the satellite mass function is progressively built in time via abundance matching techniques constrained by number densities of centrals in the local Universe. By enforcing dynamical merging timescales as predicted by high-resolution N-body simulations, we obtain satellite distributions as a function of stellar mass and halo mass consistent with current data. We show that stellar stripping, star formation, and quenching play all a secondary role in setting the number densities of massive satellites above \\(M_*\\gtrsim 3\\times 10^{10}\\, M_{\\odot}\\). We further show that observed star formation rates used in our empirical model over predict low-mass satellites below \\(M_*\\lesssim 3\\times 10^{10}\\, M_{\\odot}\\), whereas, star formation rates derived from a continuity equation approach yield the correct abundances similar to previous results for centrals.

Model-based plasma scenario development lies at the heart of the design and operation of future fusion powerplants. Including turbulent transport in integrated models is essential for delivering a successful roadmap towards operation of ITER and the design of DEMO-class devices. Given the highly iterative nature of integrated models, fast machine-learning-based surrogates of turbulent transport are fundamental to fulfil the pressing need for faster simulations opening up pulse design, optimization, and flight simulator applications. A significant bottleneck is the generation of suitably large training datasets covering a large volume in parameter space, which can be prohibitively expensive to obtain for higher fidelity codes. In this work, we propose ADEPT (Active Deep Ensembles for Plasma Turbulence), a physics-informed, two-stage Active Learning strategy to ease this challenge. Active Learning queries a given model by means of an acquisition function that identifies regions where additional data would improve the surrogate model. We provide a benchmark study using available data from the literature for the QuaLiKiz quasilinear transport model. We demonstrate quantitatively that the physics-informed nature of the proposed workflow reduces the need to perform simulations in stable regions of the parameter space, resulting in significantly improved data efficiency. We show an up to a factor of 20 reduction in training dataset size needed to achieve the same performance as random sampling. We then validate the surrogates on multichannel integrated modelling of ITG-dominated JET scenarios and demonstrate that they recover the performance of QuaLiKiz to better than 10\\%. This matches the performance obtained in previous work, but with two orders of magnitude fewer training data points.

[ABRIDGED] We aim to provide a holistic view on the typical size and kinematic evolution of massive early-type galaxies (ETGs), that encompasses their high-\\(z\\) star-forming progenitors, their high-\\(z\\) quiescent counterparts, and their configurations in the local Universe. Our investigation covers the main processes playing a relevant role in the cosmic evolution of ETGs. Specifically, their early fast evolution comprises: biased collapse of the low angular momentum gaseous baryons located in the inner regions of the host dark matter halo; cooling, fragmentation, and infall of the gas down to the radius set by the centrifugal barrier; further rapid compaction via clump/gas migration toward the galaxy center, where strong heavily dust-enshrouded star-formation takes place and most of the stellar mass is accumulated; ejection of substantial gas amount from the inner regions by feedback processes, which causes a dramatic puffing up of the stellar component. In the late slow evolution, passive aging of stellar populations and mass additions by dry merger events occur. We describe these processes relying on prescriptions inspired by basic physical arguments and by numerical simulations, to derive new analytical estimates of the relevant sizes, timescales, and kinematic properties for individual galaxies along their evolution. Then we obtain quantitative results as a function of galaxy mass and redshift, and compare them to recent observational constraints on half-light size \\(R_e\\), on the ratio \\(v/\\sigma\\) between rotation velocity and velocity dispersion (for gas and stars) and on the specific angular momentum \\(j_\\star\\) of the stellar component; we find good consistency with the available multi-band data in average values and dispersion, both for local ETGs and for their \\(z\\sim 1-2\\) star-forming and quiescent progenitors.

What determines the nuclear radio emission in local galaxies? We combine optical [O III] line emission, robust black hole (BH) mass estimates, and high-resolution e-MERLIN 1.5-GHz data, from the LeMMINGs survey, of a statistically-complete sample of 280 nearby, optically active (LINER and Seyfert) and inactive HII and Absorption line galaxies [ALG]) galaxies. Using [O III] luminosity (\\(L_{\\rm [O~III]}\\)) as a proxy for the accretion power, local galaxies follow distinct sequences in the optical-radio planes of BH activity, which suggest different origins of the nuclear radio emission for the optical classes. The 1.5-GHz radio luminosity of their parsec-scale cores (\\(L_{\\rm core}\\)) is found to scale with BH mass (\\(M_{\\rm BH}\\)) and [O~III] luminosity. Below \\(M_{\\rm BH} \\sim\\)10\\(^{6.5}\\) M\\(_{\\odot}\\), stellar processes from non-jetted HII galaxies dominate with \\(L_{\\rm core} \\propto M_{\\rm BH}^{0.61\\pm0.33}\\) and \\(L_{\\rm core} \\propto L_{\\rm [O~III]}^{0.79\\pm0.30}\\). Above \\(M_{\\rm BH} \\sim\\)10\\(^{6.5}\\) M\\(_{\\odot}\\), accretion-driven processes dominate with \\(L_{\\rm core} \\propto M_{\\rm BH}^{1.5-1.65}\\) and \\(L_{\\rm core} \\propto L_{\\rm [O~III]}^{0.99-1.31}\\) for active galaxies: radio-quiet/loud LINERs, Seyferts and jetted HII galaxies always display (although low) signatures of radio-emitting BH activity, with \\(L_{\\rm 1.5\\, GHz}\\gtrsim\\)10\\(^{19.8}\\) W Hz\\(^{-1}\\) and \\(M_{\\rm BH}\\gtrsim10^{7}\\) M\\(_{\\odot}\\), on a broad range of Eddington-scaled accretion rates (\\(\\dot{m}\\)). Radio-quiet and radio-loud LINERs are powered by low-\\(\\dot{m}\\) discs launching sub-relativistic and relativistic jets, respectively. Low-power slow jets and disc/corona winds from moderately high to high-\\(\\dot{m}\\) discs account for the compact and edge-brightened jets of Seyferts, respectively. Jetted HII galaxies may host weakly active BHs. Fuel-starved BHs and recurrent activity account for ALG properties. [abridged]

We present an overview of the Middle Ages Galaxy Properties with Integral Field Spectroscopy (MAGPI) survey, a Large Program on ESO/VLT. MAGPI is designed to study the physical drivers of galaxy transformation at a lookback time of 3-4 Gyr, during which the dynamical, morphological, and chemical properties of galaxies are predicted to evolve significantly. The survey uses new medium-deep adaptive optics aided MUSE observations of fields selected from the GAMA survey, providing a wealth of publicly available ancillary multi-wavelength data. With these data, MAGPI will map the kinematic and chemical properties of stars and ionised gas for a sample of 60 massive (> 7 x 10^10 M_Sun) central galaxies at 0.25 < z <0.35 in a representative range of environments (isolated, groups and clusters). The spatial resolution delivered by MUSE with Ground Layer Adaptive Optics (GLAO, 0.6-0.8 arcsec FWHM) will facilitate a direct comparison with Integral Field Spectroscopy surveys of the nearby Universe, such as SAMI and MaNGA, and at higher redshifts using adaptive optics, e.g. SINS. In addition to the primary (central) galaxy sample, MAGPI will deliver resolved and unresolved spectra for as many as 150 satellite galaxies at 0.25 < z <0.35, as well as hundreds of emission-line sources at z < 6. This paper outlines the science goals, survey design, and observing strategy of MAGPI. We also present a first look at the MAGPI data, and the theoretical framework to which MAGPI data will be compared using the current generation of cosmological hydrodynamical simulations including EAGLE, Magneticum, HORIZON-AGN, and Illustris-TNG. Our results show that cosmological hydrodynamical simulations make discrepant predictions in the spatially resolved properties of galaxies at z ~ 0.3. MAGPI observations will place new constraints and allow for tangible improvements in galaxy formation theory.