A Census of NUV M-Dwarf Flares Using Archival GALEX Data and the gPhoton2 Pipeline

TL;DR

The study leverages archival GALEX data processed with the gPhoton2 pipeline to perform the largest census to date of NUV flares on M-dwarfs, supplemented by XMM-OM data. It derives flare frequency distributions in equivalent duration $\delta$ and absolute energy $E$ across spectral types M0–M6, applying injection-based corrections to account for partial flare detection and biases. The results show a monotonic increase in flare rates with spectral type, with UV flares contributing a non-negligible fraction of the stellar energy budget (up to roughly $\sim33\%$ at M4), and reveal flares with energies up to $\sim10^{34}$ erg. The findings have important implications for exoplanet habitability and abiogenesis, illustrating both potential benefits and drawbacks of UV flares, and they lay out a framework for future wide-field UV surveys with ULTRASAT that could detect millions of M-dwarf flares over short cadences.

Abstract

M-dwarfs are common stellar hosts of habitable-zone exoplanets. NUV radiation can severely impact the atmospheric and surface conditions of such planets, making characterization of NUV flaring activity a key aspect in determining habitability. We use archival data from the GALEX and XMM-Newton telescopes to study the flaring activity of M-dwarfs in the NUV. The GALEX observations form the most extensive dataset of M-dwarfs in the NUV to date, with exploitation of this data possible due to the new gPhoton2 pipeline. We run a dedicated algorithm to detect flares in the pipeline produced lightcurves and find some of the most energetic flares observed to date within the NUV bandpass, with energies of $\sim10^{34}$ ergs. Using GALEX data, we constrain flare frequency distributions for stars from M0 to M6 in the NUV up to $10^5\,$s in equivalent duration and $10^{34}$ ergs in energy, orders of magnitude above any previous study in the UV. We speculate the combined effect of NUV luminosities and flare rates of M4 and later stars could potentially allow abiogenesis on their habitable zone exoplanets. As a counterpoint, we estimate that the high frequencies of energetic UV flares and associated coronal mass ejections would inhibit the formation of an ozone layer, possibly preventing genesis of complex Earth-like lifeforms due to sterilizing levels of surface UV radiation. We also provide a framework for future observations of M-dwarfs with ULTRASAT, a wide FoV NUV telescope to be launched in 2027.

Figures (27)

• Figure 1: Gaia DR3 color magnitude diagram (CMD) plotted for stars from the (a) Gaia DR3 survey and (b) GALEX and XMM-OM cross-matched samples. The red curves demarcate the main sequence bounds $+0.65<M_G<-0.6$ described in section \ref{['sec: Dataset']}.
• Figure 2: Distribution of our M-dwarf sample as a function of spectral type: (a) Exposure durations and (b) Number of GALEX sources; (c) Exposure durations and (d) Number of XMM-OM sources. Black bars show all selected sources on the main sequence as described in section \ref{['sec: Dataset']}. On the top panels, blue bars show the subgroup of GALEX sources detected over at least 3 exposures.
• Figure 3: Distribution of bin sizes in the GALEX sample. The shortest detectable flare duration is equal to the bin size as a flare can comprise of a single lightcurve bin (see section \ref{['sec: flare id']}).
• Figure 4: Histogram of closest match separation of XMM-OM lightcurve positions cross-matched with the TIC sample. The maximum acceptable cross-match is set at 4 arcsec, shown by the dashed black line.
• Figure 5: A flowchart detailing the flare identification algorithm used in this work. See text for details.