Stellar Flares in GALEX and Kepler

Clara Brasseur

29 July, 2020

EXPANDING THE FRONTIERS OF SPACE ASTRONOMY

​

​

The team

Rachel Osten¹

​

2

Chase Million²

Scott Fleming¹

¹Space Telescope Science Institute

²Million Concepts

Roadmap

1. What are flares and why do we care about them
2. What instruments are we using and what benefit to they give
3. Prior surveys
4. Finding GALEX flares
5. Finding Kepler counterparts
6. Conclusions

​

3

Stellar Flares

• Short-term brightening during magnetic reconnection
• Most dramatic event experienced by cool main sequence stars
• Observed on a range of cool stars across the electromagnetic spectrum
• Probe many layers of the stellar atmosphere
• Impact exoplanet photochemistry

4

NASA/SDO

Stellar Flares: magnetic reconnection

• Change linking so �A->C and D->B
• Reconfigures magnetic field to lower energy state
• Therefore releases energy to environment
• Magnetic energy builds up continuously but is released impulsively

5

Benz and Gudel, 2010, Annu. Rev. Astron. Astrophys. 48:241-87

Stellar Flares: benefit of ultra violet (UV) observations

6

• All stellar atmosphere layers are involved
• UV emission lines probe upper chromosphere and transition region
• Hot blackbody emission peaks in near UV

Stellar Flares: multiwavelength observations

Multiwavelength observations allow energy fractionation exploration.

​

Three ways to get multiwavelength observations:

• Star by star
• Multiplexing in nearby star-forming regions
• Overlapping time domain surveys

​

​

7

(Hawley et al. 2003)

GALEX and Kepler

Galaxy Evolution Explorer

• Orbiting space telescope
• Active between 2003 and 2013
• FUV (~1300-1700Å ) and NUV (~1700-3000Å) detectors
• Took data in ~30 minute “visits”
• FUV camera failed in 2009

8

Kepler

• Orbiting space telescope
• Active between 2009 and 2018
• Optical (430-890 nm) detector
• 30 min and 1 min cadence light curves

GALEX light curves

​

gPhoton

• Suite of Python tools for more flexible access to GALEX data
• GALEX detectors recorded photon events with time resolution of 5 thousandths of a second
• gPhoton allows photon level access and creation of light curves at any cadence

​

For this project we used 10 second binning to produce light curves.

​

​

9

Other flare surveys: Kepler Catalog of Stellar Flares�

• Davenport analyzed every short and long cadence Kepler light curve
• ~2 million flare candidates
• Analysis required stars have
• ≥ 100 flare candidates
• ≥ 10 candidate with �E > 68% completeness threshold
• Result: 4041 flare stars
• Flare duration range�~10 min – 4 hr
• Flare energy range�~10³³-10³⁹ erg

10

GJ 1243

Davenport, 2016 and Davenport et al, 2014

Other flare surveys: finding variable sources with gPhoton

11

• Search for variability in all GALEX NUV data using gPhoton
• 30 second light curve resolution
• Includes all sorts of short duration variability not just flares
• 400 6σ detections

Other flare surveys: finding variable sources with gPhoton

• Binary M dwarf system
• 13 flare detections
• Duration range 5-18 min
• Energy range 10²⁸-10³¹ erg

12

GJ 65

Fleming et al, 2020

Other flare surveys: energy-duration regimes

13

Namekata et al, 2017

Brasseur et al. 2019: Target selection

Initial target requirements:

• Simultaneous(ish) GALEX and Kepler data
• At least 30 min total observation time
• Continuous intervals of at least 5 min in duration
• 10 second binning

​

Result: 34,276 targets

​

​

​

14

Stellar characteristics

• Stars are primarily “sun-like”
• Most R✳︎✳︎✳︎✳︎✳︎_＊ in range ~0.7R_⊙-6R_⊙
• Most T_eff in range ~4,500K-6,500K
• Rotation periods range from ~0.3-66 days*

​

​

​

​

​

​

*only ~14% of stars have defined periods

15

16

GALEX Flare Detection: from ~34,000 targets to ~1,000 flaring stars

GALEX Flare Detection: from ~34,000 targets to ~1,000 flaring stars

17

Flare Detection: from ~34,000 targets to ~1,000 flaring stars

18

Flaring stars: stellar characteristics

• Stars are primarily “sun-like”
• Most R✳︎✳︎✳︎✳︎✳︎_＊ in range ~0.7R_⊙-6R_⊙
• Most T_eff in range ~4,500K-6,500K
• Rotation periods range from ~0.3-66 days*

​

​

​

​

​

​

*only 14% of flaring stars have defined periods

19

Flaring stars: flare characteristics

Short duration�90% under 4 minutes

​

Small�75% are 3.5-10σ above quiescent flux

​

No strong trend towards higher peaks on longer flares

​

​

20

fully observed flares

cut-off flares (duration is minimum)

quiescence near noise floor (peak is minimum)

Flare Energy Distribution

• Flare energy – distance correlation shows sensitivity limitation due to distance
• Lower energy flares have smaller flux enhancement
• Higher energy flares span entire range of flux enhancements

21

Flare Energy Distribution

• Expected evidence of power law
• Index ⍺ in line with literature values
• Similar power law index across distance
• Higher temperature stars show steeper slope

22

Flare population comparison

• Flare population is distinct from other studies
• Evidence of lack of dependence between flare duration and energy
• Active region length scale ~10⁹-10¹⁰ cm
• Magnetic field strength > several hundred Gauss

​

23

Namekata et al. 2017

Summary

• Discovered a previously un-cataloged population of UV flares
• Very short duration, fills in a new energy-duration regime
• Flare rate power law agrees with literature for both stellar and solar flares

24

Looking for Kepler counterparts

• Number of flares with short cadence overlap: 2
• Visibility of those flares in Kepler: None
• Number of flares with long cadence overlap: 1557
• Visibility of those flares in Kepler: No
• Number of Kepler short cadence flares: 1006 (not vetted)
• Number of those flares with overlapping GALEX data: 3
• Number of those flares that are real detections: None

​

25

Using Flare injection

• Inject artificial flares into Kepler light curves
• Model flares with flare template from Davenport 2014
• Inject flares with distribution from GALEX flare population
• Detrend light curve and recover injected flares
• Determine maximum undetectable flare energy
• Place limits on Kepler band energy for GALEX flares

​

26

Injected flare recovery: the flares

27

Injected 853,761 flares

Recovered 5,097 flares

Injected flare recovery: flare energies

Do we recover the correct energies? Mostly.

28

Conclusions

• Previously un-cataloged population of UV flares fills in a new energy-duration regime
• Flare rate power law remains broadly consistent across flare parameters
• Recovery of injected synthetic flares indicate that above a certain threshold flare rate and energy can be recovered quite well.

​

29

Future work

• Use injected flares to place limits on Kepler band energies for GALEX flares
• Explore what this means for flare energy fractionation
• Explore flares on Kepler stars that have coincident GALEX data but are not part of our sample of GALEX flaring stars

​

30

Questions?

References

• Brasseur, C. E., Osten, R. A., & Fleming, S. W. 2019, ApJ, 883, 88
• Benz, A. O., & Güdel, M. 2010, ARA&A, 48, 241
• Davenport, J. R. A. 2016, ApJ, 829, 23
• Davenport, J. R. A., et al. 2014, ApJ, 797, 11
• Fleming, S. W., et al. 2020, in prep.
• Hawley, S. L., Allred, J. C., Johns-Krull, C. M., et al. 2003, ApJ, 597, 535
• Million, C., Fleming, S. W., Shiao, B., et al. 2016, ApJ, 833, 292
• Million, C. 2020, AAS 235
• Namekata, K., Sakaue, T., Watanabe, K., et al. 2017, ApJ, 833, 292

​

​

31

Conclusions

• Previously un-cataloged population of UV flares fills in a new energy-duration regime
• Flare rate power law remains broadly consistent across flare parameters
• Recovery of injected synthetic flares indicate that above a certain threshold flare rate and energy can be recovered quite well.

​

32