Periodic Reporting for period 4 - INTERSTELLAR (The Interstellar Medium of High Redshift Galaxies)
Reporting period: 2022-04-01 to 2024-03-31
The main goal of INTERSTELLAR has been a thorough investigation of the formation of the earliest galaxies in the Universe, their assembly and evolution during a crucial epoch, referred to as the Epoch of Reionization encompassing the first billion year of cosmic evolution. The impact of these themes on society is very broad as these topics might quite directly speak to the general public. Questions like what is the origin of the Universe and its its evolution, and the understanding of how it became the amazing entity the we observe today are constantly asked in public talks, and form also the basis of a human cultural background all around the word. They are as general as music can be perceived by humans. Finally, from the technological and investment point of view they are very powerful drivers for breakthrough solutions and challenges.
INTERSTELLAR has approached these questions in an innovative and very successful way. The research has been based on a combination of theory, supercomputing simulations, data from the top experimental facilities, both space-born (HST, JWST) and ground based (ALMA, VLT), and cutting-edge data analysis techniques, often based on Artificial Intelligence methods. With these tools, we have produced significant breakthroughs in the field fulfilling our original promises and expectations. We have made precise predictions and used our results to guide observations, acquire and analyse top-notch data or extract them from archives. The final outcome is a coherent and unprecedented picture of the physics of the first galaxies. To manage this extremely interdisciplinary approach, we have built a very diverse, gender-balanced Team including theorists, numerical experts and experimentalists that have worked in a highly coordinated manner to achieve our objectives.
INTERSTELLAR has approached these questions in an innovative and very successful way. The research has been based on a combination of theory, supercomputing simulations, data from the top experimental facilities, both space-born (HST, JWST) and ground based (ALMA, VLT), and cutting-edge data analysis techniques, often based on Artificial Intelligence methods. With these tools, we have produced significant breakthroughs in the field fulfilling our original promises and expectations. We have made precise predictions and used our results to guide observations, acquire and analyse top-notch data or extract them from archives. The final outcome is a coherent and unprecedented picture of the physics of the first galaxies. To manage this extremely interdisciplinary approach, we have built a very diverse, gender-balanced Team including theorists, numerical experts and experimentalists that have worked in a highly coordinated manner to achieve our objectives.
During its entire duration, the project has produced about 90 papers published on international, refereed journals, and a major review of the field published on Physics Reports. In the first part of the project we have concentrated as planned on the characterization of the far infrared line emission in galaxies in the Epoch of Reionization (EoR) as a tracer of the interstellar medium (ISM) of these systems, and its connection with their star formation history. These studies have involved major advances in the study of the physics of molecular clouds, the feedback of massive stars and the radiation transport in inhomogeneous media. We have also lay down the foundations for the implementation of dust modelling in galaxy formation theory and numerical simulations. These theoretical studies have been augmented by the development of cosmological numerical simulations which are necessary to describe the complex physical network, and the range of physical scales present in the interstellar medium. The effects of magnetic fields have been also quantified. Our cutting-edge supercomputer cosmological simulations based on a novel zoom-in technique, including on-the-fly multi-frequency radiative transfer, have for the first time resolved the internal structure of high-redshift galaxies, thus representing a benchmark for other studies that have followed us during the last 5 years. INTERSTELLAR results have led to important achievements. A partial list includes: (a) a theory for the relation between FIR emission lines and the star formation activity in galaxies; (b) a thorough derivation, based on a novel methodology using the [CII] line of the dust content and effects in early galaxies, with a new determination of the grain temperature evolution with redshift; (c) a novel interpretation of extended carbon halos around EoR galaxies, later also independently confirmed by the ALMA Large Program ALPINE collaboration; (d) a detailed characterization of the dynamics of galactic disks up the highest detectable redshifts; (e) a theory for the newly discovered (by JWST) super-early, luminous galaxies; (f) a simulation-based determination of the star formation rate stochasticity in the early galaxies and its effect for their visibility and on the mass-metallicity relation.
INTERSTELLAR has gone beyond the state-of-the art in many ways. We have developed comprehensive theoretical models for the interstellar medium of galaxies which outshine any previous attempts in terms of physical processes included and detailed/continuous successful comparison with available data. Our numerical simulations are simply unprecedented in the field as they resolve the internal structure of the galaxies and embed all the key processes of the ISM, including non-equilibrium chemical networks for various key species, as for example molecular hydrogen and carbon oxide, that regulate the star formation activity of the galaxy. INTERSTELLAR results have reliably predicted the emission properties and spectra in several observable bands, from the UV to the far-infrared, thanks also to unique features like the multi-frequency, on-the-fly radiation transfer. These simulations have been the basis of our contribution to large collaborations using both ALMA (REBELS – PI: Bouwens; CRISTAL – PI: Herrera-Camus) and JWST (PRIMER – PI: Dunlop) Large Programs and have been fundamental to interpret those data. Towards the end of the project we have pushed our numerical work so to increase the statistical weight of our simulated sample of high redshift galaxies. By performing mass production runs, we have been able to put together (thanks to ERC-funded additional parallel supercomputing nodes) the SERRA simulation suite which now includes more than 300 galaxies and it is still growing. This complete statistical sample has allowed us to be the first to quantify a very important quantity, the stochastic mode of star formation in early galaxies. Such “bursty” star formation mode prevailing in the early universe, has dramatic implications, which are only partly understood at the moment, on the observed properties of these pristine systems. The SERRA simulations, coupled with sophisticated theoretical work, has allowed us to take the world lead on this issue. Last but not least, INTERSTELLAR has trained a number of young researchers, some of which are already enrolled in permanent positions in top institutions in Italy.