Документ взят из кэша поисковой машины. Адрес оригинального документа : http://star.arm.ac.uk/preprints/2014/653.pdf
Дата изменения: Thu May 29 13:07:00 2014
Дата индексирования: Sun Apr 10 02:41:22 2016
Кодировка:
Поисковые слова: cygnus

Mon. Not. R. Astron. Soc. 000, 000000 (0000)

Printed 28 May 2014

A (MN L TEX style file v2.2)

The M4.5V flare star AF Psc as seen in K2 engineering data
Gavin Ramsay1, J. Gerry Doyle1 1
Armagh Observatory, Col lege Hil l, Armagh, BT61 9DG, UK

Accepted 2014 May 28. Received 2014 May 6; in original form 2014 April 9

ABSTRACT

We present the light curve of the little studied flare star AF Psc (M4.5V) obtained using engineering data from the K2 mission. Data were obtained in Long Cadence mode giving an effective exposure of 29 min and nearly 9 d of coverage. A clear modulation on a period of 1.08 d was seen which is the signature of the stellar rotation period. We identify 14 flares in the light curve, with the most luminous flares apparently coming from the same active region. We compare the flare characteristics of AF Psc to two M4V flare stars studied using Kepler data. The K2 mission, if given approval, will present a unique opportunity to study the rotation and flare properties of late type dwarf stars with different ages and mass. Key words: Physical data and processes: magnetic reconnection stars: activity Stars: flares stars: late-type stars: individual: AF Psc, KIC 5474065, KIC 9726699

1

INTRODUCTION

NASA's Kepler mission was launched in March 2009 and sp ent the next 4 years making near continuous flux measurements of over 160,000 stars in an area of sky covering 115 square degrees in the constellations of Cygnus and Lyra (Borucki et al. 2010). Although the prime science driver for the mission was the discovery of Earth sized planets around Solar typ e stars, it provided a wealth of information on objects as diverse as pulsating stars, accreting systems, sup ernovae and flare stars. With the loss of two of the satellites four reaction wheels, the mission has now evolved into the K2 mission (Howell et al. 2014). Funding p ermitting, this will result in a series of p ointings along the ecliptic plane, each lasting 75 days. In the planning stage for the K2 mission, several engineering tests are b eing made. Data from the Feb 2014 tests have recently b een made publically available. The p ointing of the original Kepler was such that the p ointing accuracy was much b etter than the pixel scale on a timescale much shorter than the three month quarters (Koch et al. 2010). In the K2 mission, by p ointing at fields in the ecliptic plane, photon pressure from the Sun acts as the only source of force and the two remaining reaction wheels remove the build up of angular momentum. This causes the stars to shift by measureable amounts on the CCD detectors. However, the Kepler team found that K2 gives photometry which is within a factor of 24 of the original Kepler data (see Howell et al 2014). The almost continuous light curves which Kepler is able to provide makes it ideal for the investigation of many typ es
c 0000 RAS

of phenomena including stellar flares. For instance, Balona (2012) rep orted observations of flares from stars with early A and F sp ectral typ es, while Maehara et al. (2012) and Shibayama et al. (2013) rep ort `sup er-flares' from Solar typ e stars. At lower masses, Walkowicz et al (2011) made a study of flares from cool stars and Gizis et al. (2013) rep orted flares from an L dwarf. In Ramsay et al. (2013), we rep orted observations of two M4V stars which showed intense flare activity. Here we rep ort on observations derived from K2 engineering data on another flare star, AF Psc.

2

AF PSC

AF Psc was discovered as a high amplitude (>6 mag) flare star nearly 40 years ago (Bond 1976). It was included in a large sp ectroscopic sample of nearby M dwarfs and a sp ectral typ e of M4.5V was determined and a distance of 11 p c derived by parallax measurements (Riaz, Gizis & Harvin 2006). AF Psc is around the same brightness (V =14.4) as KIC 9726699 (g =13.9) but much brighter than KIC 5474065 (V =18.1) b oth of which were studied by Ramsay et al (2013). It does not app ear to have b een the sub ject of a previous dedicated optical photometric study, although several flares in the UV were observed on AF Psc using Galex (Welsh et al. 2006). These authors also rep orted the detection of oscillations on a timescale of 30 sec during flares, which they interpreted as b eing due to `slow sausage mode' waves.

2

Figure 1. The light curve of AF Psc made using K2 in engineering data where each point has an effective exposure of 29 min. The data has been phased (=0.0 is defined by minimum flux) on the 1.08 day period which is clearly present in the data. The vertical lines at the top of the panel note the time of flares in the light curve. The three consecutive flares occuring at 0.12 are noted by dashed vertical lines.

3

K2 DATA

The detector on b oard Kepler is a shutterless photometer using 6 sec integrations and a 0.5 sec readout. The observations of AF Psc were made in long cadence, where 270 integrations are summed for an effective 29.4 min exp osure. This contrasts with the observations made of the two M4V targets KIC 5474065 and KIC 9726699 (Ramsay et al 2013) which were made in short cadence where the effective exp osure is 58.8 sec. Observations were carried out in engineering tests from MJD 56692.57 to 56701.50 (2014 Feb 4th to Feb 13). The coverage was therefore 8.9 days in duration. During this time interval there were frequent thrusts of the spacecraft to ensure p ointing accuracy with one large shift occuring on MJD 56694.86 (or 2.3 days into the time series). A 50в50 pixel array is downloaded from the satellite for each target. To extract a light curve of AF Psc we used the PyKe software (Still & Barclay 2012)1 which was develop ed for the Kepler mission by the Guest Observer Office. We exp erimented by extracting data from a series of different combinations of pixels. We found that a mask centered on AF Psc, but including a relatively faint (USNO-B1 gives R 18.9) nearby (20 arcsec) star consisting of 140 pix-

els gave the optimal results. If a smaller numb er of pixels are used we find that there are small discontinuities present in the light curve which is the result of small shifts in the p osition of the stellar profile over the CCD. We also exp erimented with subtracting the background (which increased in a nearly linear fashion over the course of the observations) in different ways. We found that using the median value of each time p oint to represent the background gave the b est results. Finally we removed time p oints which were not flagged `SAP QUALITY==0' (for instance times of enhanced solar activity).

4

RESULTS

1

http://keplergo.arc.nasa.gov/PyKE.shtml

We normalised the extracted light curves by dividing the data by the mean background subtracted flux. The light curve of AF Psc shows a clear modulation on a p eriod of 1.08±0.08 days (Figure 1). Given AF Psc is an M4.5V star, this modulation is caused by the rotation of sp ots or active regions on the photosphere. The first few rotation cycles have smooth profiles, but then b ecome more complex (double horn shap ed at maximum) which indicates that active regions app ear and dissapp ear on relatively short timescales. In Ramsay et al. (2013), we used an automatic routine to identify flares in the Kepler data of two M4V stars. Given the quality of the present light curve is relatively lower (and
c 0000 RAS, MNRAS 000, 000000

AF Psc as seen in K2 data

3

Figure 3. The cumulative energy distribution of flares (in the Kepler band-pass) as seen in AF Psc, KIC 5474065 and KIC 9726699.

Figure 2. Top Panel: The light curve of AF Psc phased and binned on the rotation period of 1.08 days. Lower Panel: The energy of the flares plotted as a function of rotation phase.

the time coverage much shorter) we decided to manually identify flares in the light curve of AF Psc. We searched for events which showed a rapid rise in flux and an exp onential decline which is characteristic of stellar flares. Given each exp osure is 29 min, very short duration flares are likely to either b e missed completely or seen as a flux increase in only one time p oint. We were rather conservative in identifying p oints as flares and did not (for instance) flag the enhanced flux p oint at 2.05 rotation cycles (Figure 1) as this coincided with a significant shift in the stellar profiles. In some cases it was rather sub jective whether an event was a flare or not for instance we did not identify features in rotation cycle 2 (Figure 1) as flares, but rather the general variation seen in the rotation curves of active stars. We identified 14 flares in AF Psc (which are marked in Figure 1) over the whole 9 days of data. To determine the luminosity of the flares we assume that a star with sp ectral typ e M4.5V has an luminosity L = 2.5в 1030 erg s-1 (see Ramsay et al 2013 for details). We then sum up the area in the flare assuming this base luminosity. Given we do not use any model for the flare profile and that the time resolution is rather low, there is some degree of uncertainty in the estimate for each individual flare. However, it does indicate the general characteristics of the flare b ehaviour of this source. We find that the flares seen in AF
c 0000 RAS, MNRAS 000, 000000

Psc have a luminosity in the range 0.1 - 3.0 в 1032 erg in the Kepler band-pass. This compares to L = 1.1 - 7.3 в 1032 ergs for KIC 5474065 and L = 0.01 - 2.2 в 1032 ergs for KIC 9726699. We show the folded and binned light curve of AF Psc in Figure 2 together with the rotation phase and luminosity of each flare. We note that the most luminous flares were seen b etween =0.10.3. Indeed, in Figure 1 it is seen that there are three consecutive rotation cycles where a flare is seen at 0.12. This indicates that there is a region on the star which is active over more than one rotation cycle. We show the cumulative flare-frequency distribution in Figure 3 for AF Psc and KIC 5474065 and KIC 9726699. Interestingly, despite b eing a more rapid rotator than KIC 5474065 (1.08 d compared to 2.47 d), the flare-frequency distribution of these sources are very similar. The distribution of AF Psc goes to lower luminosities since it is a much brighter source (and hence less luminous flares may have b een missed in KIC 5474065). In contrast, KIC 9726699, although having a very similar sp ectral typ e (M4V) is a more rapid rotator (0.59 d). AF Psc does not show the high amplitude flares seen in KIC 5474065 but this mayb e due to the shorter timeline of the data (8.9 d compared to 24.2 d) and the longer exp osure time for each photometric p oint.

5

CONCLUSIONS

If the K2 mission proceeds as hop ed, it will cover a numb er of clusters (Howell et al 2014) which have ages ranging from the very young (Taurus-Auriga Association at 2 Myr), to the not-so-young (the Pleiades at 125 Myr) to the p ositively old (M67 at 3.6 Gyr). The engineering data presented here of AF Psc demonstrate that K2 has the photometric accuracy to identify the rotation p eriod and flare rate of M dwarf stars even in long cadence mode and over a time interval significantly shorter than that planned for K2 in full op eration. K2 will give an unique opp ortunity to determine how the stellar rotation p eriod and flare rate of late typ e dwarfs are effected by age, mass and metallicity. West et al 2008) showed that a marked change in activity levels occurs

4
around the sp ectral typ e M4. K2 will allow the the determination of key observables for dozens of stars with sp ectral typ e in the range M0M8 and hence address the underlying physical reasons for this.

6

ACKNOWLEDGEMENTS

This pap er includes data collected by the Kepler spacecraft using 2-wheel mode. Funding for the Kepler spacecraft is provided by the NASA Science Mission Directorate. Some of the data presented in this pap er were obtained from the Mikulski Archive for Space Telescop es (MAST). STScI is op erated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-26555. Supp ort for MAST for non-HST data is provided by the NASA Office of Space Science via grant NNX09AF08G and by other grants and contracts. This work made use of PyKE, a software package for the reduction and analysis of Kepler data. This op en source software pro ject is develop ed and distributed by the NASA Kepler Guest Observer Office. We thank Tom Barclay, Martin Still and Steve Howell for useful advice. Armagh Observatory is supp orted by the Northern Ireland Government through the Dept Culture, Arts and Leisure.

REFERENCES
Balona, L., 2012, MNRAS, 423, 3420 Bond, H. E., 1976, IBVS, 1160 Borucki, W. J. et al., 2010, Sci, 327, 977 Gizis, J. E., Burgasser, A. J., Berger, E., Williams, P. K. G., Vrba, F. J., Cruz, K. L., Metchev, S., 2013, ApJ, 779, 172 Howell, S. B., et al., 2014, accepted PASP, (arXiv:1402.5163) Koch, D. G., et al., 2010, ApJ, 713, L79 Maehara, H., Shibayama, T., Notsu, S., Notsu, Y., Nagao, T., Kusaba, S., Honda, S., Nogami, D., Shibata, K., 2012, Nature, 485, 478 Ramsay, G., Doyle, J. G., Hakala, P., Garcia-Alvarez, D., Brooks, A., Barclay, T., Still, M., 2013, MNRAS, 434, 2451 Riaz, B, Gizis, J. E., Harvin, J., 2006, AJ, 132, 866 Shibayama, T., Maehara, H., Notsu, S., Notsu, Y., Nagao, T., Honda, S., Ishii, T. T., Nogami, D., Shibata, K., 2013, ApJS, 209, 5 Vidotto, A.A., Jardine, M., Opher, M., Donati, J.F., Gombosi, T.I., 2011, ASPC 448, 1293 Walkowicz, L. M., et al 2011, AJ, 141, 50 Welsh, B.J., et al., 2006, A&A 458, 921 West, A.A., Hawley, S.L., Bochanski, J.J., Covey, K.R., Reid. I.N., Dhital, S., Hilton, E.J., Masuda, M., 2008, AJ 135, 785

c 0000 RAS, MNRAS 000, 000000