A 7 Day Multiwavelength Flare Campaign on AU Mic. IV: Quiescent Gyrosynchrotron and Gyroresonance Radiation from 12 to 25 GHz

TL;DR

AU Mic's quiescent radio emission in the 12–25 GHz range is dissected using five days of simultaneous VLA Ku-band and ATCA K-band observations. The integrated spectra are best described by a two-component model, with a falling gyrosynchrotron component ($α = -0.88 \pm 0.10$) and a flat component ($C = 0.64 \pm 0.14$ mJy); a purely thermal free-free origin is inconsistent with mass-loss constraints, favoring an optically thick gyroresonance origin from multiple regions. Time-resolved analysis shows day-to-day variability in spectral indices and modest circular polarization, implying a distributed magnetically active population rather than a single flare. Electron-kinetic-energy-rate estimates for the gyrosynchrotron component, across plausible magnetic fields ($B \sim 10^2$–$3\times10^3$ G), suggest long-duration energy budgets comparable to flares, consistent with sustained magnetic reconnection and particle trapping. These findings constrain the magnetic topology, particle acceleration, and wind properties of AU Mic, with implications for exoplanet space weather and the broader behavior of active M dwarfs, and they underscore the need for broader-band, simultaneous observations to map high-frequency quiescent emission.

Abstract

We present an analysis of the radio quiescent data from a multiwavelength campaign of the active M-dwarf flare star AU Mic (dM1e) that occurred in October 2018. Using Ku-band data (12 to 18 GHz) from the Very Large Array and K-band data (17 to 25 GHz) from the Australia Telescope Compact Array, we find that the quiescent spectrum can be decomposed into two components: one falling with frequency and one that remains flat. The flat component has a relatively steady flux density of 0.64 $\pm$ 0.14 mJy. The falling component varies in strength, but exhibits a spectral index of $α$ = $-0.88 \pm 0.10$. The falling component is thus consistent with nonthermal, optically thin gyrosynchrotron radiation with a corresponding power-law index similar to flares from AU Mic. While a flat component may arise from thermal, optically thin free-free emission, the observed flux density and inferred mass-loss rate are both too large compared to previous stellar wind and X-ray emission theory and models, necessitating an alternative explanation. This flat component instead matches well with an optically thick gyroresonance component integrated over multiple source regions such that the composite spectra are reasonably flat. The persistence of these components across the rotational period suggests multiple source regions, which may help explain changes in flux density and persistent high-energy electrons.

Figures (4)

• Figure 1: Top: Light curves for the VLA and ATCA are shown, in bins of 1 minute (1 m) and 5 minutes (5 m), respectively. Flares and periods of high noise due to weather effects are grayed out and not used in calculations. The rotational modulation extrapolated from the temporally nearest TESS Sector is shown in yellow Ikuta2023. Bottom: Circular polarization fraction of the VLA data is displayed against the TESS rotational modulation. The gray dashed line marks $\pi_c = 0$.
• Figure 2: Radio spectra integrated over all quiescent times. The VLA Ku-band wideband and the 2 ATCA K-band subbands are split evenly between 4 frequency ranges each.
• Figure 3: Best-fit coronal heating model of thermal free-free emission Cranmer2013 compared to past VLA Cox1985White1994Leto2000 and ALMA MacGregor2013 data. The thin solid line marks the coronal heating model limit where $\dot{M}$ rises above $10^{-8} \, M_{\odot}$ yr$^{-1}$. Spectra from Figure \ref{['fig:FigQuiescentAll']} are plotted in red. Cross markers indicate upper limits.
• Figure 4: Estimated electron kinetic energy rates ($E_\text{KE}(t)$) from the gyrosynchrotron component of the quiescent light curve. Various magnetic field strengths ($B$) are tested within a plausible range for M-dwarf atmospheres.