Periodic Reporting for period 1 - EPOCH-OF-TAURUS (Early phases of planetary birth sites -- environmental context and interstellar inheritance)
Période du rapport: 2022-11-01 au 2025-04-30
The aim of this ambitious research project is to produce the most realistic computer simulations of the assembly of gaseous protoplanetary accretion disks, and to understand which of their traits are inherited from and/or affected by their direct interstellar context. Owing to ground-breaking instruments such as VLT/Sphere or the ALMA telescope array, we now have a first extensive census of disk populations. Moving beyond the core characterisation of relatively isolated disks in the calm Class II stage, the time has come to shift the focus towards the wider context of these systems, that is, the actively star-forming stellar associations, such as the archetypal Taurus, Orion or Lupus regions. Stellar ages of disks with substructure of (likely) planetary origin point to the fact that planet formation is not merely an ubiquitous process, but figuratively speaking happens within the blink of an eye. This mandates to abandon the assumption of the disk as a quiescent entity detached from its surroundings, and instead place it in the context of a collapsing cloud core. Key aspects here are i) the external UV radiation field that can drive powerful photochemical reactions on the surface, ii) perturbations from stellar flybys, iii) gas self-gravity, and iv) magnetic field lines that are self-consistently anchored in the local interstellar medium (ISM); the latter aspect requiring adaptive-mesh technology, provided by the NIRVANA III code, co-developed by the applicant. At the same time, the early appearance of planets poses questions about the solid constituents potentially being inherited from the ISM and “primed” during the protostellar precursor phase. Finally, with the pivotal exchange of angular momentum during the collapse regulated by non-ideal MHD effects, the evolution of microphysical coefficients (i.e. through an ionisation chemistry with recombination on small grains) needs to be followed through the collapse phase.
Because PPDs are initially quite massive in comparison to their host star, their own weight can lead to gravitational instability (GI) and subsequent fragmentation. The interaction of GI with magnetic fields is still poorly understood. To study these topics, we extend the NIRVANA magnetohydrodynamics code to the realm of self-gravity — specifically for the case of spherical-polar meshes. These are mandatory to properly keep track of the disk angular moment content, which in turn critically determines protoplanetary disk evolution. An obstacle is posed by the fact that solving Poisson's equation requires prior knowledge of the gravitational potential on the domain boundary. Standard multipole-expansion is shown to fail for the inner cavity in the polar mesh. As a remedy, and similar to Baron Münchhausen, who pulls himself out of the swamp by his own hair, we use James' trick, employing the already existing multi-grid solver to obtain a surface correction density (i.e. for the preliminary situation where a vanishing gravitational potential is assumed on the boundary/periphery). Translating this "screening charge" into the surface gravitational potential is done via a Green's function convolution method. We describe a novel method to obtain the required discrete kernel function in spherical polar geometry. Care has to be taken to obtain proper convergence with numerical resolution. For this purpose, we devise a new test-problem and provide a semi-analytic solution. Moreover, we demonstrate decent scaling behaviour of the implementation, up to several thousand independent compute cores.
Moreover, PPDs are often found to be extremely thin with respect to their vertical extent. Thus, in a complementary approach, we aim to describe the effects of self gravity (SG) in a vertically-integrated two-dimensional description — enabling highly resolved simulations at moderate computational cost. To this end, our research team pioneers a novel kernel-based framework, that greatly supersedes previous attempts using a simple softening-parameter. We demonstrate that — depending on the ad-hoc choice of that parameter — that class of approximation is bound to either severely under-estimate the effect of SG on certain length scales, or conversely, over-estimates it by equal amounts. We have developed an efficient implementation of the kernel-based two-dimensional approach, and have carefully benchmarked its accuracy against a three-dimensional reference solution. Employing this new framework, we will investigate the clumping of gravitationally unstable gas under various cooling regimes, with the aim to firmly establish the conditions that demarcate the boundary towards fragmentation.
Moreover, PPDs are often found to be extremely thin with respect to their vertical extent. Thus, in a complementary approach, we aim to describe the effects of self gravity (SG) in a vertically-integrated two-dimensional description — enabling highly resolved simulations at moderate computational cost. To this end, our research team pioneers a novel kernel-based framework, that greatly supersedes previous attempts using a simple softening-parameter. We demonstrate that — depending on the ad-hoc choice of that parameter — that class of approximation is bound to either severely under-estimate the effect of SG on certain length scales, or conversely, over-estimates it by equal amounts. We have developed an efficient implementation of the kernel-based two-dimensional approach, and have carefully benchmarked its accuracy against a three-dimensional reference solution. Employing this new framework, we will investigate the clumping of gravitationally unstable gas under various cooling regimes, with the aim to firmly establish the conditions that demarcate the boundary towards fragmentation.
Equipped with the self-gravity framework laid out above, we have performed global simulations of marginally gravitationally unstable PPDs, where stirring/clumping due to self-gravity is dynamically regulated by the complex interplay of i) Keplerian shear, ii) turbulent eddies and iii) the (lacking) efficiency of the radiative cooling. In this regime, we have successfully confirmed the presence of a dynamo, and obtained the cornerstone properties of a mean-field contribution to the dynamo, acting on large-scale magnetic fields. The characterisation of the mean-field dynamo is aided by accompanying (and simplified) local shearing-box simulations, where we employ the powerful test-field method (TFM) framework to directly obtain the turbulent transport coefficients entering the mean induction equation. By these means, we were able to demonstrate that the previously identified magnetic dynamo is of the mean-field type, and that it exists in the strict kinematic limit. This has important implications for the potential of the GI dynamo in bootstrapping the coherent and moderately strong seed-fields required for the magnetorotational instability (MRI) to set in during a later evolutionary stage. This possibility is intriguing, because this would enable magnetic turbulence (and hence substantial angular momentum redistribution) irrespective of the inherited amount of net-vertical magnetic flux during the formation of the PPD, that is, out of its natal collapsing core. If, corroborated by future research, the timescales for this pre-amplification are indeed found to be sufficiently short to overpower the initial level of any pre-existing coherent field, this would lead the way to a profound understanding about which environmental factors can shape young PPDs, and which are rendered less important.