Astron. Astrophys. 347, 508-517 (1999)

1. Introduction

Classical T Tauri stars (henceforth called cTTSs) are young, low mass, optically visible pre-main sequence stars with accretion disks. Weak-Line T Tauri stars (henceforth called wTTSs) resemble the classical ones but lack disks, and accretion. It has been a mystery as to how the rotation rates of cTTSs are kept much lower than the breakup velocity, despite the large accretion rates, and despite the fact that cTTSs are still contracting towards the main sequence. Currently, the most favoured model is the magnetic accretion scenario in which a strong magnetic field couples to the star and the disk, so that angular momentum flows outward in the disk, while matter flows inward (Königl 1991; Cameron & Campbell 1993; Cameron et al. 1995; Armitage & Clarke 1996).

Although a few direct measurements of the field strength now exist (Guenther et al. 1999, Johns-Krull et al. 1999), the evidence that most, or all T Tauri stars have strong fields is still rather sparse. As discussed in more detail in Guenther et al. (1999), direct measurements of the field strength are rather difficult, and limited to very few T Tauri stars.

In principle, observations of flares might be used as indirect evidence for the presence of magnetic fields. Since these observations are not limited to certain stars, observations of flares could tell us how common magnetic activity is amongst young stars. Flare-like events have indeed been detected in cTTSs and wTTSs in X-rays. These events are so similar to flares on dMe stars that they are canonically interpreted as enhanced solar-type flares (Feigelson & DeCampi 1981; Walter & Kuhi 1984; Montmerle et al. 1993; Preibisch et al. 1993). While the discovery of X-ray flares on T Tauri stars is not surprising as all late type stars show flares (Schmitt 1994), the enormous energy-output of the events is impressive. The energy released in these flares in the -band are sometimes as large as  erg, compared to  erg for the largest flares of dMe stars (Hawley & Pettersen 1991). X-ray flares on T Tauri stars often last for several hours (Skinner at al. 1997; Preibisch et al. 1993). Using a scaled-up model of a solar flare, and the observed temperatures and decay times, Preibisch et al. (1993) were able to derive a lower limit of the magnetic flux density of 210 G, and an emitting volume of , implying that very strong magnetic fields must indeed be present. Observations of the very rapidly rotating zero-age main-sequence star (henceforth called ZAMSS) AB Dor (age years) implies that the X-ray flares are indeed connected to large star-spots (Vilhu et al. 1993).

Flares not only emit X-rays but all kinds of electromagnetic waves, from -rays down to radio waves (Somov 1992). Recent simultaneous optical and X-ray observations of a solar flare show that the light curves of and soft X-rays (0.25-4.0 keV) are almost identical, implying that the Balmer-lines are probably induced during a flare by the soft X-ray radiation (Johns-Krull et al. 1997a). Although the optical emission is a secondary effect, it is a necessary ingredient of a flare, and if the interpretation of X-ray data is correct, optical flaring must be observable.

After the pioneering work of Haro & Chavaria (1965) it has been well established that young stars have optical flares. Since rapid increases of the brightness in the optical regime and variations of spectral lines on time scales of less than an hour have occasionally been observed in T Tauri stars too (e.g. Bastian & Mundt 1979), the picture seems, at first sight, to be rather consistent.

However, a statistical analysis of broad-band photometry of T Tauri stars by Gahm (1990) indicates that while the colours and the light curves of events on wTTSs were flare-like, the events on cTTSs were not. Gahm correspondingly interprets the events observed on cTTSs as variations of the extinction, or variations of the veiling continuum but not as flares. This conclusion is further supported by a simultaneous spectroscopic and photometric monitoring of two cTTSs and three wTTSs by Gahm et al. (1995). Although three flares on wTTSs were found, no flare was observed in a cTTS. Unfortunately, the only attempt to observe a cTTS (BP Tau) simultaneously in the optical and in X-rays was unsuccessful, as no optical data were taken at the moment when an X-ray flare occurred (Gullbring et al. 1997). A small optical variation had no counterpart in the X-rays. That means that either X-ray flares on cTTSs have no optical counterparts, or have just been missed by chance. Given the low frequency of X-ray flares of about (Gahm 1990) this seems plausible. The question, whether cTTSs have optical flares or not thus remains unsolved.

Flares are not only important as indirect evidence for the presence of magnetic fields in young stars but might be important for ionizing the circumstellar disk. This is because flares lead to an increased X-ray flux, harden the X-ray spectrum, and increase the flux of high energy particles (Glassgold et al. 1997; Tsuboi et al. 1998). Although the flare-activity declines with age, flares could also be important in the later stages of the evolution, because the EUV and X-ray emission of the young Sun might have been important for the formation of a planetary ionosphere (Güdel et al. 1997).

Progress in optical studies of flares from young stars was hitherto severely hindered by the rarity of the events. However, the multi-object spectrographs now allow to study many T Tauri stars simultaneously, thereby increasing the probability for detecting such events enormously. We thus have started a monitoring campaign to search for flares using the UK Schmidt telescope in Siding Springs and the Schmidt telescope in Tautenburg. On the basis of the data taken in our first observing run on the UK Schmidt telescope, we demonstrated that flares can be studied very efficiently using such an instrument, and we have reported on the detection of two flares in wTTSs (Guenther & Emerson 1997, henceforth called paper I). However, up to now, no flares were detected in cTTSs.

The first aim of this paper is to report on a possible detection of a flare in the cTTS FN Tau. In order to discuss whether this event was a flare or not, we compare it with a flare in a ZAMSS, and with three flares observed in wTTSs which were observed with the same equipment. Since wTTSs and ZAMSSs are known to show flare activity, these events serve as typical examples for flares observed spectroscopically on young stars.

The second aim of the paper is to study the frequency of flares in the three groups of stars. For that purpose, we use the properties of these flares to define criteria for the (semi)-automatic detection of flares in our time series of 7674 spectra.

© European Southern Observatory (ESO) 1999

Online publication: June 30, 1999
helpdesk.link@springer.de