Документ взят из кэша поисковой машины. Адрес оригинального документа : http://hea-www.harvard.edu/tanagra/etc/jpb_HEAD2011_1005.pdf Дата изменения: Tue Sep 6 21:27:58 2011 Дата индексирования: Fri Feb 28 06:53:32 2014 Кодировка: Поисковые слова: п п п п п п п п п п п п р п р п р п р п р п р п р п р п р п р п р п р п р п р п

HEAD 2011: Poster 10.05

Project Tanagra: Stellar Flares in Chandra High-Resolution Spectra
Jennifer Posson-Brown, Vinay Kashyap, Steven Saar & Jeremy Drake (Harvard-Smithsonian Center for Astrophysics)

Abstract
We introduce Project Tanagra: Timing Analysis of Gratings Data, a uniform study of archival Chandra gratings observations of active l ow-mass coronal stars. We include ACIS-S/HETG, ACIS-S/LETG, and HRC-S/LETG observations. Gratings data are optimal for timing analysis since they are free from pile-up and allow for joint spectro-temporal analysis. We discuss technique s for timing analysis of gratings data and explore the distribution of stellar flare e nergies and the time variability of individual lines fluxes. Here we present preliminary results from four targets: AU Mic, AD Leo, Procyon and sigma Gem. The project website is: http://hea-www.harvard.edu/tanagra./ This work is supported by CXC NASA contract NAS8-39073 and Ch andra grant AR0-11001X.

Goals
1. Characterize flare distribution · Processes that generate solar & stellar flares appear to be scale-free: distribution of flare energies is power-law. · Power-law distribution has been verified for the Sun over many orders of magnitude of flare energies and range of timescales (Aschwanden et al. 2000, ApJ, 535, 1047). · Power law index 1.8 for Sun but generally > 2 for other active stars. Beyond =2 it becomes possible to attribute all coronal luminosity to increasingly weaker, but more numerous, flares. 2. Spectro-temporal analysis · Examine changes in spectral lines to characterize variabil ity at different temperatures in stellar coronae.

Processing
1. Download data products from Chandra archive, reprocess w ith CIAO 4.3, and extract source and background events from dispersed spe ctrum, 0th order, and transfer streak. 2. Make counts and flux lightcurves for broad band and strong s pectral lines; compare line flux changes to overall luminosity variations. 3. Fit stochastic flare model to photon arrival time data usin g an MCMC-based method. This allows us to find the most likely value of without direct flare detection. (See Kashyap et al. 2002, 2011 and Saar et al. 2011.)

Initial Results
· Four sources analyzed so far, see Table 1 for stellar and obse rvation parameters, and best-fit values. · Example plots shown for AU Mic ObsID 17 (ACIS-S/HETG): 1. High resolution spectra from MEG and HEG first order, and ze roth order. 2. Counts and flux lightcurves for dispersed spectrum and zer oth order. 3. Flux lightcurves for Fe XVII, O VIII, and Ne X lines. 4. Fractional flare count rate versus from MCMC iterations. · An example of line flux variations with temperature, from pre vious work with Capella, is shown in Figure 5.

Figure 3 Background-subtracted running flux lightcurves for specifi c lines in the AU Mic dispersed spectrum (ACIS-S/HETG ObsID 17). Top: Ne X at 12.134 е (left), Fe XVII at 15.013 е (right). Bottom: Fe XVII at 16.913 е (left), O VIII at 18.969 е (right).

Figure 5 An example line flux variability at different temperatures, from previous work with Chandra gratings observations of Capella (Posson-Brown et al. 2011). Plotted is the log of the ratio of fractional variability ( µ ) in a given line, calculated with the secular (broad band) correction, to fractional var iability calculated from the uncorrected line fluxes, as a function of temperature. Th e lines are colorcoded according to grating arm and order: light blue is HRC-S /LEG -1, dark blue is HRC-S/LEG +1, light green is ACIS-S/MEG -1, dark green is ACIS-S/MEG +1, light red is ACIS-S/HEG -1, and dark red is ACIS-S/HEG +1. Ratio values less than one indicate that the fractional variability is greater without the secular correction, implying that the variability is due to the secu lar trend. The log ratio is anti-correlated with temperature (r=-0.52,p=1.03e-7), i ndicating that the secular variability is mainly due to hot lines, while cool lines exhibit significant nonsecular variability.

Source Procyon

Sp ectral Typ e F5 IV-V

Instrument HRC-S/LETG

Figure 1 Background-subtracted spectra for AU Mic, ACIS-S/HETG ObsID 17, from MEG (top), HEG (middle) and zeroth order (bottom).
sigma Gem AU Mic AD Leo K1 I I I M1 Ve M3.5 V ACIS-S/HETG ACIS-S/HETG HRC-S/LETG ACIS-S/HETG HRC-S/LETG

ObsID 63 1224 1461 10994 12042 5422 6282 17 8894 2570 24 975

Observation Date 1999-11-06 1999-11-08 1999-11-07 2009-12-15 2009-12-26 2005-05-16 2005-05-17 2000-11-12 2008-06-26 2002-06-01 2000-01-22 2000-10-24

Exp osure (ks) 70.15 20.93 70.25 72.13 65.84 63.86 58.88 59.56 50.18 45.90 10.12 48.41

MEG: MEG: MEG: MEG:

Best-fit 1.86±0.29 1.84±0.63 2.26±0.50 1.96±0.30 1.87±0.54 2.67±0.22, HEG: 2.67±0.22, HEG: 2.31±0.16, HEG: 2.08±0.54 2.55±0.30, HEG: 2.01±0.64 2.11±0.59

2.78±0.23 2.64±0.54 2.44±0.30 2.60±0.31

Table 1 Source name, spectral type, Chandra instrument, ObsID, obs ervation date, exposure time, and best-fit and standard deviation for stars analyzed thus far in Project Tanagra. (For a full list of Tanagra project sources, see http://hea-www.harvard.edu/tanagra/.) We note that the best-fit values for AD Leo are consistent with previous measurements based on EUVE data (Kashyap et al. 2002, GЭdel 2004).

References
· Aschwanden, M.J., Tarbell, T. Nightingale, R., et al. 2000, ApJ, 535, 1047 · GЭdel, M., 2004, A&ARv, 12, 71G · Kashyap, V.L., Drake, J.J., GЭdel, M., & Audard, M., 2002, ApJ, 580, 1118 · Kashyap, V.L., Sarr, S., Drake, J.J., Reeves, K., Posson-Brown, J. & Connors, A., 2011, SCMA V,
http://astrostatistics.psu.edu/su11scma5/lectures/kashyap_scmav_poster.pdf

Figure 2 Background-subtracted running lightcurves for AU Mic, ACI S-S/HETG ObsID 17. Left: Zeroth order in counts (top) and flux (bottom). Right: Dispersed events, separated by grating arm and order, in counts (top) a nd flux (bottom).

· Posson-Brown, J. & Kashyap, V.L., 2011, AAS, 21822801P
Figure 4 Scatterplot of the MCMC iterations for the flare index and the fraction of the count rate attributable to the flare component for AU Mi c ObsID 17 (ACISS/HETG), MEG (top) and HEG (bottom).

· Saar, S.H., Kashyap, V.L., Drake, J.J., Reeves, K., & Connors, A., 2011, AAS, 21832202S

contact: jpossonbrown@cfa.harvard.edu