Supermassive black holes (SMBHs) self-regulate galaxies, groups, and clusters, yet the pathway that transports gas from halo scales to sub-parsec radii remains debated. In hot, stratified atmospheres, subsonic turbulence can trigger nonlinear thermal instability and a multiphase condensation cascade, producing chaotic time-variable BH `weather'. A key missing link is how the meso-scale (parsecs to kiloparsecs) connects halo rain to the nuclear inflow. We study turbulence-driven condensation and chaotic cold accretion (CCA) in a group-scale halo, quantifying how the stirring level shapes multiphase morphology and thermodynamics, and how this imprints on SMBH feeding down to sub-parsec scales. We ran 3D hydrodynamic hyper-zoom simulations with a GPU-accelerated code, including radiative cooling and driven subsonic turbulence in a hot intragroup halo. Two endpoint runs bracket weak and strong stirring, capturing distinct BH weather states. In both regimes the atmosphere becomes thermally unstable and develops a strongly multiphase medium spanning 8-10 orders of magnitude in temperature and density. Strong stirring delays cold gas accretion and sustains an extended, filament-rich rain pattern to kiloparsec radii (`stormy' CCA), with broader thermodynamic distributions beyond the nucleus. Weak stirring triggers earlier condensation but yields a more compact clumpy rain, with most cold gas confined within 100 pc (`rainy' CCA). At micro-scales the inflow is partly mediated by a clumpy rotating torus. Despite large differences in condensed cold mass, the BH accretion rate is recurrently boosted by up to 100 above the hot-mode Bondi baseline and varies only weakly between the weather regimes, indicating that feeding is regulated primarily by how efficiently multiphase structures couple to the central inflow. Modest changes in turbulence are sufficient to shift the same hot halo between stormy (extended) and rainy (centralized) BH weather, providing a robust quantitative multiscale baseline for interpreting multiphase CCA and its impact on SMBH feeding.
Barbani et al. (Fri,) studied this question.