Starspots and Flares are Generally Not Correlated

TL;DR

This study tests whether the solar-like association between sunspots and flares holds for other stars by applying a new flare-detection pipeline (TOFFEE) to a large TESS-based sample and robust spot modeling. By analyzing 62,450 lightcurves from 16,305 stars and 218,386 flares across 14,163 spotted stars, the authors quantify the spot-state at flare times and assess correlations with flare occurrence. After extensive bias control, injection tests, and per-star analyses, the combined result is a flare-positivity of $p = 49.97 \pm 0.21\%$, indicating no strong correlation between flare rate and spot visibility. The work highlights the role of faculae and geometry and shows that solar spot–flare coupling is not a universal feature, with implications for understanding magnetic activity across diverse stars.

Abstract

Sunspots and solar flares are two different manifestations of magnetic activity on the surface of the Sun. On the Sun, flares typically occur close to spots. In this paper we test this the connection between spots and flares on other stars. We detect 218,386 stellar flares on 14,163 spotted stars using a new algorithm called \textsc{toffee}. Inhomogeneous spot distributions mean that as stars rotate they become brighter when less spots are facing the observer, and dimmer when more spots are facing the observer. We determine that flares occur when the star is brighter $49.97\pm 0.21\%$ of the time, i.e. there is an equal preference for the flares to occur when the star is relatively bright or dim. We therefore find no evidence for a correlation between flare rate and spot occurrence, contrary to what is seen on the Sun.

Figures (7)

• Figure 1: Detrending and flare search process of Sector 37 of BY Draconis variable star TIC 378126824 from raw lightcurve (top) to final flattened curve (bottom). Top shows the raw time resolved brightness of the star in black points with the quadratic trend coming from the orbit of the TESS telescope in blue. Middle shows the quadratic subtracted lightcurve overlaid with the trend found by Wotan in red. The bottom curve is the final flattened lightcurve in black and the detected flares. Primary flares are colored in red with the peaks labeled as red stars and the secondary is labeled in blue with a blue star representing its peak.
• Figure 2: Rotation periods of 14,163 stars with a Lomb-Scargle periodogram false alarm rate less than 5%.
• Figure 3: A set of three consecutive flares and their spot fittings. The fitted flare in question is centered and the window displays the segment of data used in its center fit. While fitting, the flare (in grey; see \ref{['sec:fit_criteria']}) is masked and only the surrounding measurements are considered. The modulation at flare time is then interpolated. Fits indicating positive spot modulation at the flare are shaded red; negative spot modulation, blue. Spot amplitude is calculated relative to the window median, which is denoted by a horizontal dashed line. The average normalized spot amplitude of the fits are 0.98 (top), 0.13 (middle), and -0.82 (bottom) respectively.
• Figure 4: Two flares for which two fitted spot modulation curves disagree on whether the modulation at the flare is positive (in red) or negative (in blue), with light shades representing the rightmost fit window and dark shades for the leftmost fit window. This disagreement leads to the exclusion of the flare from our dataset. This vetting procedure is responsible for the V-shaped cutout around a normalized spot amplitude of 0 in the histogram in Fig. \ref{['fig:result']}.
• Figure 5: Cumulative number counts of flares in the 2024Feinstein sample demonstrating the number of flares of amplitude expressed in terms of the spread of the lightcurve, $\sigma$, $A_{\sigma}$, or higher expressed in log-space. Deviations from a straight line in log-log space are emblematic of incompleteness at low amplitude (left of the vertical dashed line) or small number statistics at high amplitude.