Speaker
Description
The infall of the LMC into the Milky Way (MW) has caused significant disequilibrium throughout the MW. In particular, it has moved the MW's centre of mass and deformed its dark matter halo. There are only a handful of tailored MW—LMC simulations that can capture all aspects of this dynamical disequilibrium.
I will show how we can utilise much simpler rigid MW—LMC models to still constrain many properties of interest e.g., the masses of the MW and LMC.
I will use a ‘Simulation-Based Inference’ (SBI) approach to infer model parameters of the MW—LMC system. SBI is a powerful tool as it does not require likelihood functions to be defined. Instead, one can run many forward simulations to mimic observed quantities, and use data from surveys e.g., Gaia, DESI to extract the properties of the MW and LMC which are allowed given these observations.
I will show that using an SBI pipeline trained on a set of millions of rigid MW—LMC simulations is able to: i) accurately reproduce true parameters from MW—LMC that also include deformations despite that not being in our model and ii) place constraints on the MW mass, LMC mass, reflex velocity apex, and much more using stellar halo kinematic data from MW surveys.
This first application of SBI to the MW—LMC system shows a promising avenue to infer properties of both the MW and LMC without the need to run computationally intensive simulations.
