The NASA Kepler mission, and its successor, TESS, aim to find exoplanets that pass in front of their host stars by detecting the decrease in starlight when a small part of the star is blocked by the exoplanet. In the process, they obtain time series data that can reveal stellar flares - sudden outbursts of energy that occur when the surface magnetic field reconnects.
My collaborators and I look for flares in this time series data, then characterize their energies and overall frequency. We expect the flare frequency to be related to the overall strength of the star’s magnetic field - which itself depends on the star’s rotation, mass, and age. There are three main projects that I’ll introduce here.
flares and age using K2 clusters
As stars age, they typically slow their rotation, which weakens the stellar magnetic field. We expect that the frequency of flares should also decrease with age, and if we can measure the relationship between a star’s flare rate and its age, flares can be a useful age indicator. First, we must examine flares on stars with well-determined ages.
A star cluster is a gravitationally bound group of stars that were typically formed from the same material at the same time. Within most clusters, all the stars have the same age and chemical composition, so we can determine the cluster age from stars that have more obvious age indicators. Ekaterina Ilin is leading a project to examine the flare rate on different types of stars in clusters with K2 data and well-measured ages.
Left panel: The frequency of flares as a function of energy for two star clusters: M44 (age: 125 Myr) and M45 (age: 625 Myr). For both clusters, small flares occur much more frequently than large flares, but the overall flare rate of the younger cluster is higher. (from:
Ilin et al. 2019)
Right panel: A parameterization of overall flare rate as a function of cluster age. The different color points separate stars of different surface temperatures. Overall the data shows a decrease in flare rate as a function of age and an increase for smaller, cooler stars. (from:
Ilin talk at Thinkshop 16)
To reliably identify flares in K2 data, Ekaterina developed the AltaiPony flare finding code, an updated version of Jim Davenport’s Appaloosa code. This fall, she is working to update the first version of the project, which focused on comparing M44 and M45, with four additional clusters and a refined version of the flare injection/recovery testing procedure.
ultracool dwarfs in K2
Despite small masses and cool, mostly neutral atmospheres, many of the coolest stars (and warmest brown dwarfs) still can have dramatic flares. In fact, these stars typically maintain their magnetic field to older ages, so the average ultracool dwarf is typically magnetically active - though often with less overall energy output than more massive stars.
In work led by Rishi Paudel and John Gizis, we looked for flares in all ultracool dwarfs in Kepler’s K2 survey, successfully obtaining short cadence observations for a dozen dwarfs. We found that flares are actually ubiquitous on ultracool dwarfs–the observed dwarfs flared at lower energies than more massive flare stars, but the flares were still common.
TESS ultracool dwarfs
The TESS mission has been underway for over a year, delivering month-long light-curves for the brightest stars in first the southern hemisphere and, as the second year progresses, the northern hemisphere. My collaborators and I have an accepted proposal for TESS to obtain short cadence observations for a carefully selected sample of the brightest M6 and later dwarfs over the entire sky.
Analysis of the data is ongoing, but initial data suggests that the overall flare rate for ultracool dwarfs strongly depends on stellar mass, but may be higher than was found by the initial K2 work.
-
Giant white-light flares on fully convective stars occur at high latitudes
Ekaterina Ilin, Katja Poppenhaeger, Sarah J. Schmidt, Silva P. Järvinen, Elisabeth R. Newton, Julián D. Alvarado-Gómez, J. Sebastian Pineda, James R. A. Davenport, Mahmoudreza Oshagh, and Ilya Ilyin
MNRAS, Oct 2021
White-light flares are magnetically driven localized brightenings on the surfaces of stars. Their temporal, spectral, and statistical properties present a treasury of physical information about stellar magnetic fields. The spatial distributions of magnetic spots and associated flaring regions help constrain dynamo theories. Moreover, flares are thought to crucially affect the habitability of exoplanets that orbit these stars. Measuring the location of flares on stars other than the Sun is challenging due to the lack of spatial resolution. Here we present four fully convective stars observed with the Transiting Exoplanet Survey Satellite that displayed large, long-duration flares in white-light which were modulated in brightness by the stars’ fast rotation. This allowed us to determine the loci of these flares directly from the light curves. All four flares occurred at latitudes between 55\,^∘ and 81\,^∘, far higher than typical solar flare latitudes. Our findings are evidence that strong magnetic fields tend to emerge close to the stellar rotational poles for fully convective stars, and suggest that the impact of flares on the habitability of exoplanets around small stars could be weaker than previously thought.
-
Flares in open clusters with K2. II. Pleiades, Hyades, Praesepe, Ruprecht 147, and M 67
Ekaterina Ilin, Sarah J. Schmidt, Katja Poppenhäger, James R. A. Davenport, Martti H. Kristiansen, and Mark Omohundro
A&A, Jan 2021
Context. Magnetic fields are a key component in the main sequence evolution of low mass stars. Flares, energetic eruptions on the surfaces of stars, are an unmistakable manifestation of magnetically driven emission. The occurrence rates and energy distributions of flares trace stellar characteristics such as mass and age. However, before flares can be used to constrain stellar properties, the flaring-age-mass relation requires proper calibration.
Aims: This work sets out to quantify the flaring activity of independently age-dated main sequence stars for a broad range of spectral types using optical light curves obtained by the Kepler satellite.
Methods: Drawing from the complete K2 archive, we searched 3435 ∼80 day long light curves of 2111 open cluster members for flares using the open-source software packages K2SC to remove instrumental and astrophysical variability from K2 light curves, and AltaiPony to search and characterize the flare candidates.
Results: We confirmed a total of 3844 flares on high probability open cluster members with ages from zero age main sequence (Pleiades) to 3.6 Gyr (M 67). We extended the mass range probed in the first study of this series to span from Sun-like stars to mid-M dwarfs. We added the Hyades (690 Myr) to the sample as a comparison cluster to Praesepe (750 Myr), the 2.6 Gyr old Ruprecht 147, and several hundred light curves from the late K2 Campaigns in the remaining clusters. We found that the flare energy distribution was similar in the entire parameter space, following a power law relation with exponent α ≈ 1.84−2.39.
Conclusions: We confirm that flaring rates decline with age, and decline faster for higher mass stars. Our results are in good agreement with most previous statistical flare studies. We find evidence that a rapid decline in flaring activity occurred in M1─M2 dwarfs around the ages of the Hyades and Praesepe, when these stars spun down to rotation periods of about 10 d, while higher mass stars had already transitioned to lower flaring rates and lower mass stars still resided in the saturated activity regime. We conclude that some discrepancies between our results and flare studies that used rotation periods for their age estimates could be explained by sample selection bias toward more active stars, but others may point to the limitations of using rotation as an age indicator without additional constraints from stellar activity.
-
K2 Ultracool Dwarfs Survey - VI. White light superflares observed on an L5 dwarf and flare rates of L dwarfs
R. R. Paudel, J. E. Gizis, D. J. Mullan, S. J. Schmidt, A. J. Burgasser, and P. K. G. Williams
MNRAS, May 2020
Kepler, K2 long cadence data are used to study white light flares in
a sample of 45 L dwarfs. We identified 11 flares on 9 L dwarfs
with equivalent durations of (1.3-198) h and total
(UV/optical/IR) energies of ≥0.9 x
1032 erg. Two superflares with energies of
>1033 erg were detected on an L5 dwarf (VVV BD001):
this is the coolest object so far on which flares have been
identified. The larger superflare on this L5 dwarf has an energy
of 4.6 x1034 erg and an amplitude of >300
times the photospheric level: so far, this is the largest
amplitude flare detected by the Kepler/K2 mission. The next
coolest star on which we identified a flare was an L2 dwarf:
2MASS J08585891+1804463. Combining the energies of all the
flares which we have identified on 9 L dwarfs with the total
observation time which was dedicated by Kepler to all 45 L
dwarfs, we construct a composite flare frequency distribution
(FFD). The FFD slope is quite shallow (-0.51±0.17), consistent with earlier results reported by Paudel et al.
for one particular L0 dwarf, for which the FFD slope was found
to be -0.34. Using the composite FFD, we predict that, in early-
and mid-L dwarfs, a superflare of energy 1033 erg occurs
every 2.4 yr and a superflare of energy 1034 erg occurs
every 7.9 yr. Analysis of our L dwarf flares suggests that
magnetic fields of ≥0.13-1.3 kG are present on
the stellar surface: such fields could suppress Type II radio
bursts.
-
K2 Ultracool Dwarfs Survey - V. High superflare rates on rapidly rotating late-M dwarfs
R. R. Paudel, J. E. Gizis, D. J. Mullan, S. J. Schmidt, A. J. Burgasser, P. K. G. Williams, A. Youngblood, and K. G. Stassun
MNRAS, Jun 2019
We observed strong superflares (defined as flares with energy in excess of 1033 erg) on three late-M dwarfs: 2MASS J08315742+2042213 (hereafter 2M0831+2042; M7 V), 2MASS J08371832+2050349 (hereafter 2M0837+2050; M8 V), and 2MASS J08312608+2244586 (hereafter 2M0831+2244; M9 V). 2M0831+2042 and 2M0837+2050 are members of the young (700 Myr) open cluster Praesepe. The strong superflare on 2M0831+2042 has an equivalent duration (ED) of 13.7 h and an estimated energy of 1.3 × 1035 erg. We observed five superflares on 2M0837+2050, on which the strongest superflare has an ED of 46.4 h and an estimated energy of 3.5 × 1035 erg. This energy is larger by 2.7 orders of magnitude than the largest flare observed on the older (7.6 Gyr) planet-hosting M8 dwarf TRAPPIST-1. Furthermore, we also observed five superflares on 2M0831+2244 which is probably a field star. The estimated energy of the strongest superflare on 2M0831+2244 is 6.1 × 1034 erg. 2M0831+2042, 2M0837+2050, and 2M0831+2244 have rotation periods of 0.556 ± 0.002, 0.193 ± 0.000, and 0.292 ± 0.001 d, respectively, which we measured by using K2 light curves. We compare the flares of younger targets with those of TRAPPIST-1 and discuss the possible impacts of such flares on planets in the habitable zone of late-M dwarfs.
-
Flares in open clusters with K2 . I. M 45 (Pleiades), M 44 (Praesepe), and M 67
Ekaterina Ilin, Sarah J. Schmidt, James R. A. Davenport, and Klaus G. Strassmeier
A&A, Feb 2019
Context. The presence and strength of a stellar magnetic field and activity is rooted in a star’s fundamental parameters such as mass and age. Can flares serve as an accurate stellar "clock"?
Aims: To explore if we can quantify an activity-age relation in the form of a flaring-age relation, we measured trends in the flaring rates and energies for stars with different masses and ages.
Methods: We investigated the time-domain photometry provided by Kepler’s follow-up mission K2 and searched for flares in three solar metallicity open clusters with well-known ages, M 45 (0.125 Gyr), M 44 (0.63 Gyr), and M 67 (4.3 Gyr). We updated and employed the automated flare finding and analysis pipeline Appaloosa, originally designed for Kepler. We introduced a synthetic flare injection and recovery sub-routine to ascribe detection and energy recovery rates for flares in a broad energy range for each light curve.
Results: We collect a sample of 1761 stars, mostly late-K to mid-M dwarfs and found 751 flare candidates with energies ranging from 4 × 1032 erg to 6 × 1034 erg, of which 596 belong to M 45, 155 to M 44, and none to M 67. We find that flaring activity depends both on Teff, and age. But all flare frequency distributions have similar slopes with α ≈ 2.0-2.4, supporting a universal flare generation process. We discuss implications for the physical conditions under which flares occur, and how the sample’s metallicity and multiplicity affect our results. The detected flare indices, the stellar parameters for M 44 and M 45, and a copy of Table 4 are available at the CDS via anonymous ftp to http://cdsarc.u-strasbg.fr (ftp://130.79.128.5) or via http://cdsarc.u-strasbg.fr/viz-bin/qcat?J/A+A/622/A133. We also published all flares we validated, and stellar parameters used for M 44 and M 45 in the same location.
-
K2 Ultracool Dwarfs Survey. IV. Monster Flares Observed on the Young Brown Dwarf CFHT-BD-Tau 4
Rishi R. Paudel, John E. Gizis, D. J. Mullan, Sarah J. Schmidt, Adam J. Burgasser, Peter K. G. Williams, and Edo Berger
ApJ, Jul 2018
We present photometric measurements of two superflares observed on a very young brown dwarf, CFHT-BD-Tau 4, observed during Campaign 13 of the Kepler K 2 mission. The stronger of the two superflares brightened by a factor of ̃48 relative to the quiescent photospheric level, with an increase in Kepler magnitude ∆ \tildeKp=-4.20. It has an equivalent duration of ̃107 hr, a flare duration of 1.7 days, and an estimated total bolometric (ultraviolet/optical/infrared) energy up to 2.1 × 1038 erg. The weaker of the two superflares is a complex (multipeaked) flare with an estimated total bolometric (UV/optical/IR) energy up to 4.7 × 1036 erg. They are the strongest flares observed on any brown dwarf so far. The flare energies are strongly dependent on the value of the visual extinction parameter A V used for extinction correction. If we apply a solar flare model to interpret the two superflares, we find that the magnetic fields are required to be stronger by as much as an order of magnitude than previous reports of field measurements in CFHT-BD-Tau 4 by Reiners et al. On the other hand, if we interpret our data in terms of accretion, we find that the requisite rate of accretion for the stronger superflare exceeds the rates that have been reported for other young brown dwarfs.
-
K2 Ultracool Dwarfs Survey. III. White Light Flares Are Ubiquitous in M6-L0 Dwarfs
Rishi R. Paudel, John E. Gizis, D. J. Mullan, Sarah J. Schmidt, Adam J. Burgasser, Peter K. G. Williams, and Edo Berger
ApJ, May 2018
We report the white light flare rates for 10 ultracool dwarfs using Kepler K2 short-cadence data. Among our sample stars, two have spectral type M6, three are M7, three are M8, and two are L0. Most of our targets are old low-mass stars. We identify a total of 283 flares in all of the stars in our sample, with Kepler energies in the range log E Kp ˜ (29-33.5) erg. Using the maximum-likelihood method of line fitting, we find that the flare frequency distribution (FFD) for each star in our sample follows a power law with slope -α in the range -(1.3-2.0). We find that cooler objects tend to have shallower slopes. For some of our targets, the FFD follows either a broken power law, or a power law with an exponential cutoff. For the L0 dwarf 2MASS J12321827-0951502, we find a very shallow slope (-α = -1.3) in the Kepler energy range (0.82-130) × 1030 erg: this L0 dwarf has flare rates which are comparable to those of high-energy flares in stars of earlier spectral types. In addition, we report photometry of two superflares: one on the L0 dwarf 2MASS J12321827-0951502 and another on the M7 dwarf 2MASS J08352366+1029318. In the case of 2MASS J12321827-0951502, we report a flare brightening by a factor of ̃144 relative to the quiescent photospheric level. Likewise, for 2MASS J08352366+1029318, we report a flare brightening by a factor of ̃60 relative to the quiescent photospheric level. These two superflares have bolometric (ultraviolet/optical/infrared) energies 3.6 × 1033 erg and 8.9 × 1033 erg respectively, while the full width half maximum timescales are very short,˜2 min. We find that the M8 star TRAPPIST-1 is more active than the M8.5 dwarf 2M03264453+1919309, but less active than another M8 dwarf (2M12215066-0843197).
-
K2 Ultracool Dwarfs Survey. II. The White Light Flare Rate of Young Brown Dwarfs
John E. Gizis, Rishi R. Paudel, Dermott Mullan, Sarah J. Schmidt, Adam J. Burgasser, and Peter K. G. Williams
ApJ, Aug 2017
We use Kepler K2 Campaign 4 short-cadence (one-minute) photometry to measure white light flares in the young, moving group brown dwarfs 2MASS J03350208+2342356 (2M0335+23) and 2MASS J03552337+1133437 (2M0355+11), and report on long-cadence (thirty-minute) photometry of a superflare in the Pleiades M8 brown dwarf CFHT-PL-17. The rotation period (5.24 hr) and projected rotational velocity (45 km s-1) confirm 2M0335+23 is inflated (R≥slant 0.20 R☉ ) as predicted for a 0.06 M☉ , 24 Myr old brown dwarf βPic moving group member. We detect 22 white light flares on 2M0335+23. The flare frequency distribution follows a power-law distribution with slope -α =-1.8± 0.2 over the range 1031 to 1033 erg. This slope is similar to that observed in the Sun and warmer flare stars, and is consistent with lower-energy flares in previous work on M6-M8 very-low-mass stars; taking the two data sets together, the flare frequency distribution for ultracool dwarfs is a power law over 4.3 orders of magnitude. The superflare (2.6× 1034 erg) on CFHT-PL-17 shows higher-energy flares are possible. We detect no flares down to a limit of 2× 1030 erg in the nearby L5γ AB Dor moving group brown dwarf 2M0355+11, consistent with the view that fast magnetic reconnection is suppressed in cool atmospheres. We discuss two multi-peaked flares observed in 2M0335+23, and argue that these complex flares can be understood as sympathetic flares, in which fast-mode magnetohydrodynamic waves similar to extreme-ultraviolet waves in the Sun trigger magnetic reconnection in different active regions.
-
K2 Ultracool Dwarfs Survey. I. Photometry of an L Dwarf Superflare
John E. Gizis, Rishi R. Paudel, Sarah J. Schmidt, Peter K. G. Williams, and Adam J. Burgasser
ApJ, Mar 2017
We report on K2 Campaign 8 measurements of a huge white light flare on the L1 dwarf SDSSp J005406.55-003101.8 (EPIC 220186653). The source is a typical L1 dwarf at a distance of ̃50 pc, probably an old hydrogen-burning star rather than a young brown dwarf. In the long (30-minute) cadence photometry, the flare peak is 21 times the flux of the stellar photosphere in the broad optical Kepler filter, which we estimate corresponds to ∆V ≈ -7.1. The total equivalent duration of the flare is 15.4 hr. We estimate that the total bolometric energy of the flare was 4 × 1033 erg, more powerful than the previously reported Kepler white light flares for the L1 dwarf WISEP J190648.47+401106.8, but weaker than the ∆V = -11 L0 dwarf superflare ASASSN-16ae. The initial (impulsive) cooling phase is too rapid to resolve with our 30-minute cadence data, but after 1 hour the gradual cooling phase has an exponential time constant of 1.8 hr. We use template fitting to estimate that the full time-width-at-half-amplitude of the light curve is <10 minutes and that the true flare maximum reached ̃70 times the stellar photosphere, or ∆V ≈ -8. This flare is comparable to the most powerful Kepler flares observed on the active M4 dwarf GJ 1243.