Periodic Reporting for period 4 - LEDA (The challenging quest for low-mass dark structures)
Reporting period: 2022-08-01 to 2023-10-31
Dark matter is believed to make up to 85 per cent of the total mass of the Universe; however, most of its properties have yet to be constrained. At the present time, the most tested and the most studied cosmological model for the formation of structures in the Universe is the Cold Dark Matter (CDM) paradigm, where the dark matter is assumed to be cold and collision-less. This model has been very successful in describing the Universe on large scales and in reproducing numerous observational results. Due to the challenging nature of the observations in this regime, however, the CDM paradigm is still largely untested at the smallest galactic and sub-galactic scales, where definite theoretical predictions from different dark matter models mostly differ. All dark matter models predict that structures form via gravitational instability from primordial density fluctuations in the early Universe and grow hierarchically in time through the merging and accretion of smaller objects. As a result, it is expected that the dark matter distribution should be clumpy and that there should be many low-mass dark structures scattered around other more massive galaxies. The specific scale at which structure is expected to be clumpy strongly depends on the assumed physics of the dark matter particles. For example, the CDM model predicts that the Universe should be clumpy all the way down to the smallest sub-solar scales and that a rich population of low-mass structures should exist.
On the other hand, models that assume a warm dark matter (WDM) particle predict the existence of much fewer low-mass structures and the presence of a cut-off mass below which the dark matter distribution should be smooth. Measuring the abundance of small-mass dark structures is, therefore, a key constraint to the nature of dark matter and structure formation physics. Unfortunately, most of these small-mass structures are expected to be completely dark or extremely faint and not directly observable. These small objects can, therefore, only be observed and quantified by using strong gravitational lensing.
The primary aim of this ERC project is to use strong gravitational lensing to constrain the nature of dark matter observationally. To this end, we developed an advanced lens modelling code that can efficiently and robustly analyse large interferometric data directly in visibility space. This novel technique can be used to probe the dark matter distribution on sub-galactic scales within the lens galaxies and along their line of sight. By combing our lens modelling code with high-resolution cm-observations taken with the Global VLBI array we derived one of the tightest constraints on the particle mass of Fuzzy dark matter to date, and probed, for the first time, the subhalo and halo mass functions below 10^7 Msun. From the analysis of sub-mm observations taken with ALMA, we demonstrated that lens galaxies are not simple elliptical and smooth objects as typically assumed. Instead, they are characterised by complex mass density distributions with radial and angular structures. These features need to be taken into account for a robust inference on the nature of dark matter.
On the other hand, models that assume a warm dark matter (WDM) particle predict the existence of much fewer low-mass structures and the presence of a cut-off mass below which the dark matter distribution should be smooth. Measuring the abundance of small-mass dark structures is, therefore, a key constraint to the nature of dark matter and structure formation physics. Unfortunately, most of these small-mass structures are expected to be completely dark or extremely faint and not directly observable. These small objects can, therefore, only be observed and quantified by using strong gravitational lensing.
The primary aim of this ERC project is to use strong gravitational lensing to constrain the nature of dark matter observationally. To this end, we developed an advanced lens modelling code that can efficiently and robustly analyse large interferometric data directly in visibility space. This novel technique can be used to probe the dark matter distribution on sub-galactic scales within the lens galaxies and along their line of sight. By combing our lens modelling code with high-resolution cm-observations taken with the Global VLBI array we derived one of the tightest constraints on the particle mass of Fuzzy dark matter to date, and probed, for the first time, the subhalo and halo mass functions below 10^7 Msun. From the analysis of sub-mm observations taken with ALMA, we demonstrated that lens galaxies are not simple elliptical and smooth objects as typically assumed. Instead, they are characterised by complex mass density distributions with radial and angular structures. These features need to be taken into account for a robust inference on the nature of dark matter.
To constrain the nature of dark matter, it is vital to push the detection threshold of dark haloes below 10^7 Msun, where predictions from different dark matter models differ by more than two orders of magnitude. The mass regime that can be probed with strong gravitational lensing mainly depends on the angular resolution of the data. For example, a substructure with a mass of 10^6 Msun will produce an Einstein ring with a radius of just three milli-arcseconds, or at least distortions on this angular scale. At present, such high angular resolutions can only be achieved using observations of extended gravitational arcs with Very Long Baseline Interferometry (VLBI) observations at cm- and mm-wavelengths. However, analysing these observations can be very challenging. Unlike optical observations in which the sky is directly imaged, interferometers at cm and mm-wavelength observe the sky's Fourier component in an incomplete way. Moreover, high-resolution VLBI observations are characterised by a very large number of data points, making their analysis challenging in terms of computing memory and speed.
As part of this ERC project, we have developed a new gravitational lens modelling code that fits the data in the Fourier space without the need to average or otherwise reduce the data size beforehand. We have thoroughly tested this new technique using simulated observations and demonstrated that it is possible to directly and efficiently model VLBI data, which can contain large numbers of data points. The method is fast, thanks to the use of GPU acceleration, and accurate, with uncertainties on the relevant quantities of just a few per cent. The method can be applied to 3-dimensional data (one frequency and two spatial dimensions) to study the kinematic and physical properties of high-redshift lensed galaxies on sub-kpc scale. It can model both total intensity and polarised data. In the latter case, the effect of an external Faraday rotating screen can be taken into account allowing us to set constraints on the magnetic field of the lens galaxy.
By applying our technique to milli-arcsecond resolution data from the Global VLBI array, we have derived from a single gravitational lens system one of the tightest constraints on Fuzzy Dark matter and probed for the first time the subhalo and halo mass function down to a mass limit of 10^6 Msun. By analysing observations of strong gravitational lens systems with ALMA, we have shown lens galaxies to be complex objects characterised by angular and radial structures beyond a single power-law model. Such structures have to be taken into account to avoid false detection of dark matter haloes within the lenses and along their line of sight and deliver robust constraints on the nature of dark matter.
As part of this ERC project, we have developed a new gravitational lens modelling code that fits the data in the Fourier space without the need to average or otherwise reduce the data size beforehand. We have thoroughly tested this new technique using simulated observations and demonstrated that it is possible to directly and efficiently model VLBI data, which can contain large numbers of data points. The method is fast, thanks to the use of GPU acceleration, and accurate, with uncertainties on the relevant quantities of just a few per cent. The method can be applied to 3-dimensional data (one frequency and two spatial dimensions) to study the kinematic and physical properties of high-redshift lensed galaxies on sub-kpc scale. It can model both total intensity and polarised data. In the latter case, the effect of an external Faraday rotating screen can be taken into account allowing us to set constraints on the magnetic field of the lens galaxy.
By applying our technique to milli-arcsecond resolution data from the Global VLBI array, we have derived from a single gravitational lens system one of the tightest constraints on Fuzzy Dark matter and probed for the first time the subhalo and halo mass function down to a mass limit of 10^6 Msun. By analysing observations of strong gravitational lens systems with ALMA, we have shown lens galaxies to be complex objects characterised by angular and radial structures beyond a single power-law model. Such structures have to be taken into account to avoid false detection of dark matter haloes within the lenses and along their line of sight and deliver robust constraints on the nature of dark matter.
As detecting low-mass structures is observationally challenging, most studies have been limited to the satellites of the Milky Way and the Andromeda galaxies, where kinematic measurements are possible, but which may not necessarily be a fair representation of the Universe. Strong gravitational lensing observations allow us to detect low mass haloes in distant galaxies and at any redshift along their line of sight. So far, due to the limited resolution of available observations, one could only detect dark matter haloes with masses above ~10^9 Msun. This project aims at detecting dark matter haloes with masses as low as ~10^6 Msun by using high-resolution observations of strong gravitational lens systems. This will be the first time that the number of such low-mass objects will be quantified at such cosmological distances. This ERC project will challenge our standard model for small-scale structure formation and have significant implications for the fields of cosmology and galaxy formation.