Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.sai.msu.ru/dept/adm/lamzin/aa1179.pdf Äàòà èçìåíåíèÿ: Fri Sep 28 16:47:17 2012 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 23:03:23 2012 Êîäèðîâêà:

A&A 372, 922­924 (2001) DOI: 10.1051/0004-6361:20010564
c ESO 2001

Astronomy & Astrophysics

A possible dependence of DF Tauri's photometric activity on the relative orbital positions of the binary components
S. A. Lamzin1 , S. Yu. Melnikov2 , K. N. Grankin2 , and O.V. Ezhkova2
1 2

Sternberg Astronomical Institute, Moscow V-234, 119899 Russia Ulugh Beg Astronomical Institute and Isaac Newton Institute Uzbekistan Branch, Astronomicheskaya 33, Tashkent 700052, Uzbekistan e-mail: smeln@sai.msu.ru; kn@astrin.uzsci.net; oezh@astrin.uzsci.net

Received 20 February 2001 / Accepted 20 March 2001 Abstract. We constructed a DF Tau historical lightcurve to analyze the character of stellar variability from 1900 to 2000. This time interval exceeds the orbital period P 80 yrs of the double system. At least two sections of enhanced photometric activity are present in the lightcurve, with the time interval between them being close to P/2. They correspond to the sections of the DF Tau companion orbit intermediate between apo- and periastron. A decrease in stellar activity occured near the last periastron epoch. We conclude that long-term variations of DF Tau activity are the result of modulation of primary circumstellar disc accretion rate by orbital motion of the companion. We predict that DF Tau activity will increase significantly in the near future. Key words. stars: individual: DF Tau ­ stars: pre-main sequence ­ stars: variables ­ stars: binary ­ stars: circumstellar matter ­ accretion

1. Intro duction
DF Tau is a clasical T Tauri star (CTTS): the equivalent width of the H line in its spectrum 50 ° (Herbig & Bell A 1988). DF Tau is a binary (Chen et al. 1990), that consists of two stars of M V spectral type (Ghez et al. 1997). Thi´ aut et al. (1995) derived the following set of companeb ion orbital parameters: semi-ma jor axis a = 0.19 ± 0.03 , orbital period P = 82 ± 12 years, excentricity e = 0.8 ± 0.3 and inclination angle i = 63 ± 5. Ghez et al. (1997) found from Hubble Space Telescope observations that the relative contribution of the companion radiation to the observed spectrum strongly decreases at < 6000 ° and is A less than 10% at 3000 ° On the other hand DF Tau's A. optical variability amplitude increases monotonically from R to U band, so one can conclude that the primary is responsible for the variability as well as for the continuum and line emission observed in DF Tau's spectrum (Lamzin et al. 2001). The inclination angle of the system is relatively small, so it seems rather unlikely that variability occurs due to variations of circumstellar extinction. According to the current paradigm, CTTS activity is the result of disc accretion onto magnetized low mass young star ­ see e.g. Herbst et al. (1994), Mathieu et al. (2001) and references
Send offprint requests to : S. Lamzin, e-mail: lamzin@sai.msu.ru

therein, ­ which means that DF Tau's primary has a much more active accretion disc than the companion. It is natural to expect that the primary accretion disc structure will vary as the companion moves along a highly excentric orbit, resulting in a variation of the accretion rate and character of DF Tau's optical variability. The aim of this Letter is to compare DF Tau's optical lightcurve with variations of the distance between the binary system components.

2. Results
DF Tau's lightcurve for 1900­2000 is shown in the upper panel of Fig. 1. Photographic magnitudes were used to construct the lightcurve up to the beginning of the 70th year ­ see Kholopov (1951), Badalyan (1962), Tsesevich & Dragomirezkaya (1967, 1973). Photoelectric B -magnitudes observed from the end of the 50th and up to the beginning of the 80th year were adopted from Mendoza (1968), Za jtseva (1970), Za jtseva & Lyutij (1976a,b), Kuan (1976), Nurmanova (1983), Shaimieva & Shutemova (1985). These photoelectric and photographic data overlapped in part and are in good agreement. B -magnitudes which correspond to the period from the beginning of the 80th year and up to 1997 were adopted from the Herbst database


S. A. Lamzin et al.: The dependence of DF Tau photometrical activity

923

1980 12.5 13 13.5 11.5 12 12.5 45000

1985

1990

1995

50000 JD-2.400.000

Fig. 1. DF Tau's historical lightcurve based on photographic and photoelectric (B band) observations (upper panel) and variation of the distance between components of DF Tau with time normalized to the semi-ma jor axis of the companion orbit (low panel). See text for details.

Fig. 2. DF Tau B (upper panel) and V (low panel) lightcurves averaged over observational seasons. The second order p olynomial fit is shown with a dashed line. See text for details.

http://www.astro.wesleyan.edu/pub/ttauri (Herbst et al. 1994) and subsequent data are unpublished observations of the authors (K.G., O.E. and S.M.) in the framework of the ROTOR programm ­ see Shevchenko (1989). One can see strong light variations within each observational season: flare-like events and quasicyclic variations with a timescale 7 days were observed ­ see Chelli et al. (1999) and references therein. It also follows from Fig. 1 that DF Tau's average brightness and variability amplitude were especially large during the 3­5 year intervals around 1917 and 1963. This feature was discovered by Tsesevich & Dragomirezkaya (1967), who proposed the hypothesis of the periodic character of DF Tau's activity with a timescale of 44 years ­ see also Strelkova (1973). Note that this value is close to half of the orbital period found by Thi´ aut et al. (1995). eb We calculated the variation of the distance between DF Tau's components as a function of time, adopting Thi´ ebaut et al.'s (1995) orbital parameters ­ the curve normalised to the semi-ma jor axis of the secondary orbit a is shown in the bottom panel of Fig. 1. One can see that stellar activity enhancement occured before and after the eb epoch of periastron ­ according to Thi´ aut et al. (1995) the latest epoch was 1983 ± 11 yr. The second nontrivial feature of the lightcurve is the decrease in stellar activity near the periastron epoch. To demonstrate this fact more clearlyweplot in Fig.2the B and V lightcurves of the star averaged over observational seasons. In both cases, points can be well approximated with a second order polynomial fit (dashed lines), the minima of which correspond to the end of July, 1987. It follows from Fig. 1 that decrease in stellar activity during the previous periastron epoch i.e. in the beginning of the 20th century, was less noticeable, but may be the result of low accuracy of the photographic data. Note that

during the subsequent epoch of enchanced activity, stellar brightness reached mpg = 9.2m at JD 2421890.74, i.e. exceeded minimal brightness of this epoch at 4.5m . Tsesevich & Dragomirezkaya (1967) noted that this value is not erroneous. If long term variability of DF Tau indeed has the period equal to the orbital period of the system, one can expect that the new epoch of enchanced activity should occur at the beginning of 21st century. We suggest that a "super flare" (B 6.5m ) occured on January 8 2000 (Li et al. 2001) ­ see solid triangles in the upper panel of Fig. 1 ­ and is the first manifestation of the new epoch of DF Tau's activity. We believe therefore that DF Tau's photometric activity level indeed depends on the distance between the components of the binary system, which means that the accretion rate onto the primary is modulated by the orbital motion of the companion.

3. Interpretation
Theory and observations indicate that two kinds of discs can be present in young binary systems: circumstellar (CS) and circumbinary (CB) ­ see Mathieu et al. (2001) and references therein. Tidal interaction restricts the outer radii Rd of CS discs to Rd R/2 in the case of circular orbit and the CB disc's inner radius to 2R, where R is the distance between binary components (Lubow & Artimowicz 2001). For a minimum distance between DF Tau's components of 0.2a ­ see lower panel of Fig. 1 ­ the primary CS disc radius cannot exceed this value. If the distance to DF Tau 100 pc, then Rd < 5 a.u. It is not known if DF Tau has a CB disc: an upper limit of Md < 10-3 M for the mass of DF Tau cold matter was found by Jensen et al. (1994) from submillimiter observations. If the average accretion rate onto the pri mary is Mac 3 â 10-9 M /yr (Lamzin et al. 2001), then


924

S. A. Lamzin et al.: The dependence of DF Tau photometrical activity the chief of the young star monitoring program ROTOR, on the results of which this paper is significantly based. We also would like to thank V. M. Lipunov and V. G. Surdin for helpful discussion, the referee W.Herbst for helpful comments as well as our colleagues from Variable Stars subdivision of Sternberg Astronomical Institute who helped us to search for archival photometrical data.

this cold matter will be depleted after tac = Md /Mac 3 â 105 years ­ this value is of the same order as DF Tau's age found by Ghez et al. (1997). Let us estimate the lifetime td of the primary CS disc of radius Rd assuming that it is thin (H/R 1) and has an isothermal atmosphere. If Vr and T are the radial velocity and temperature of the gas near the disc edge, respectively, then it can be derived from the equations of disc accretion theory ­ see e.g. Shakura & Sunyaev (1973) and Lipunov (1989) ­ that Rd GM Rd 450, yrs 1/2 1/2 30 td M Rd , Vr T T where M is stellar mass, = 8.31 â 107 erg g-1 K-1 is the universal gas constant, and ­ Shakura-Sunyaev parameter. M , Rd and T are expressed in M , a.u. and K respectively. It is usually assumed that 0.1-0.001 in the case of CTTSs ­ see e.g. Lynden-Bell & Pringle (1974), Lubow & Artimowicz (2001). If so, then the circumstellar disc lifetime td is in agreement with the tac value and it seems resonable to interpret the enchancement of the primary activity as the result of a strong gravitation interaction between the companion and the primary's CS disc during a relatively short period, when the companion passes near the periastron. Apparently, reorganisation of the circumstellar disc structure, which takes some time t, results in the increase of the accretion rate in the innermost disc regions. The t-value should be comparable with the rotation period of the disc external regions (Lipunov 1989) i.e. 20-30 years in our case. One can explain in this way the 1917 and the current peak of DF Tau's activity, but the 1960 activity peak is not so clear in the framework of this scenario.

References
Badalyan, N. S. 1962, Soobscheniya Byurokanskoi Observatorii, 31, 57 Chelli, A., Carrasco, L., Mujica, R., Recillas, E., & Bouvier, J. 1999, A&A, 345, L9 Chen, W. P., Simon, M., Longmore, A. J., Howell, R. R., & Benson, J. A. 1990, ApJ, 357, 224 Ghez, A. M., White, R. J., & Simon, M. 1997, ApJ, 490, 353 Herbig, G. H., & Bell, K. R. 1988, Lick Observatory Bull., 1111 Herbst, W., Herbst, D. K., Grossman, E. J., & Weinstein, D. 1994, AJ, 108, 1906 Jensen, E. L. N., Mathieu, R. D., & Fuller, G. A. 1994, ApJ, 429, L29 Kholopov, P. N. 1951, Peremennie zvezdi, 8, 83 Kuan, P. 1976, ApJ, 210, 129 Lamzin, S. A., Vittone, A. A., & Errico, L. 2001, Astron. Lett., in press Lipunov, V. M. 1989, Astrophysics of Neutron Stars (Moscow, Nauka) Li, J. Z., Ip, W. H., Chen, W. P., Hu, J. Y., & Wei, J. Y. 2001, ApJ, 549, L89 Lubow, S., & Artimowicz, P. 2001, in Protostars and Planets IV, ed. A. Mannings, A. P. Boss, & S. S. Russel (Tuscon: University of Arizona Press), 731 Lynden-Bell, D., & Pringle, J. E. 1974, MNRAS, 168, 603 Mathieu, R. D., Ghez, A. M., Jensen, E. L. N., & Simon, M. 2001, in Protostars and Planets IV, ed. A. Mannings, A. P. Boss, & S. S. Russel (Tuscon: University of Arizona Press), 703 Mendoza, E. E. 1968, ApJ, 151, 977 Nurmanova, U. A. 1983, Peremennie Zvezdi, 21, 777 Shaimieva, A. F., & Shutemova, A. F. 1985, Peremennie Zvezdi, 22, 176 Shakura, N. I., & Sunyaev, R. A. 1973, A&A, 24, 337 Shevchenko, V. S. 1989, Herbig Ae/Be stars (in Russian) Tashkent, FAN Strelkova, A. P. 1973, Astron Tsirc., N. 755 Thi´ aut, E., Balega, Y., Balega, I., et al. 1995, A&A, 304, eb L17 Tsesevich, V. P., & Dragomirezkaya, V. A. 1967, Problemi kosmicheskoi fiziki v.2 (Kiev University, Kiev), 110. Tsesevich, V. P., & Dragomirezkaya, V. A. 1973, RW Aurigae stars (Naukova Dumka, Kiev), in Russian Za jtseva, G. V. 1970, Peremennie Zvezdi, 17, 294 Za jtseva, G. V., & Lyutij, V. M. 1976a, Peremennie Zvezdi, 20, 255 Za jtseva, G. V., & Lyutij, V. M. 1976b, Peremennie Zvezdi, 20, 266

4. Conclusion
We analyzed the historical lightcurve of DF Tau which covers a time interval larger than the orbital period P of this binary system. There are two sections of the lightcurve (near 1917 and 1963) separated by a time interval close to P /2 where photometric activity of the primary was especially strong. Apparently, the new cycle of enchanced activity begins now. These activity cycles correspond to the sections of the companion orbit intermediate between apo- and periastron. At the same time DF Tau's activity was minimal near the periastron epoch in the middle of 1980. We conclude that the CS disc accretion rate onto the primary is modulated in some way by orbital motion of the companion. Further progress will require more accurate information on DF Tau's orbital parameters.
Acknow ledgements. We dedicate this article to the memory of V. S. Shevchenko, who was the initiator, the organizer and