MRI turbulence and thermal instability in accretion disks

TL;DR

This paper investigates whether MRI-driven turbulence can delay or prevent thermal instability in accretion disks, finding that turbulence delays but does not prevent the eventual thermal runaway, with implications for observed disk stability.

Contribution

It introduces a statistical model combining MRI turbulence simulations and stochastic equations to analyze thermal instability behavior in accretion disks.

Findings

Turbulent fluctuations delay thermal instability onset.

Disk temperature follows a biased random walk, not exponential growth.

Turbulence weakens but does not fully stabilize the disk.

Abstract

A long-standing puzzle in the study of black-hole accretion concerns the presence or not of thermal instability. Classical theory predicts the encircling accretion disk is unstable, as do some self-consistent MHD simulations of the flow. Yet observations of strongly accreting sources generally fail to exhibit cyclic or unstable dynamics on the expected timescales. This paper checks whether turbulent fluctuations impede thermal instability. It also asks if it makes sense to conduct linear stability analyses on a turbulent background. These issues are explored with a set of MRI simulations in thermally unstable local boxes in combination with stochastic equations that approximate the disk energetics. These models show that the disk's thermal behaviour deviates significantly from laminar theory, though ultimately a thermal runaway does occur. We find that the disk temperature evolves as a biased random walk, rather than increasing exponentially, and thus generates a broad spread of outcomes, with instability often delayed for several thermal times. We construct a statistical theory that describes some of this behaviour, emphasising the importance of the `escape time' and its associated probability distribution. In conclusion, turbulent fluctuations on there own cannot stabilise a disk, but they can weaken and delay thermal instability.