Astron. Astrophys. 324, 155-160 (1997)

4. Origin of the optical and X-ray variability

Accretion induced activity leads to optical variability while magnetically induced variability leads to correlated optical and X-ray variability such that the ratio between the radiated energies at optical and X-ray wavelengths are of the order unity. Furthermore, magnetically active regions on the stellar surface are expected to be close to cool spots thus corresponding phenomena should be correlated with stellar rotation. Do we find evidence for either behaviour in our data?

It is not possible from our five nights of observations to determine if the nightly X-ray variability is connected to the stellar rotation. It is worth noting that the X-ray light curve is consistent with the previously observed stellar rotational period of BP Tau of 6.1- 8.3 days (Vrba et al. 1986, Simon et al. 1990, Shevchenko et al. 1991, Richter 1992 and Paper I), but it is more difficult to fit the night-to-night variations in the optical data to a period longer than 5.5 days. There is no trend in the data indicating higher X-ray fluxes when the star is in a lower optical state (Fig. 2). If BP Tau contains magnetically active regions on the stellar surface, they are expected to be related to areas close to cool spots on the surface and the star would then show an enhanced X-ray activity when the star is faint. This should also be seen in the optical, especially in the U and B-band. No such anti-correlation between the optical activity and the mean brightness of BP Tau is seen during this or in previous observations of the star (Paper I).

The optical flare observed during the second night had an amplitude of only in the U-band. Since BP Tau has a radius of 2.5 and an effective temperature of 4000 K (Bertout et al. 1988, Gullbring 1994) as compared to 0.3 and 3300 K for UV Ceti type flare stars (e.g. Pallavicini 1990) the same event would have appeared as a very pronounced flare with an amplitude on an M-type flare star. Using our optical BVRI data we determined the temperature of the flare event after subtraction of the spectrum from an underlying K7 star as 6600 K, which is in line with temperatures found previously for events on BP Tauri. This temperature is lower than what is normally observed for flares on dMe stars. We estimated the X-ray luminosity by assuming a temperature and emissivity of the X-ray emitting plasma and by computing the interstellar X-ray absorption from the observed optical extinction. Since the X-ray spectra of TTS are consistent with temperatures of 10⁷ K (Montmerle et al. 1993) we assumed this temperature together with an isothermal hot plasma for the X-ray emission of BP Tau. An adopted optical interstellar extinction towards BP Tau of =0.65 yields a column density of absorbing atoms as cm^-2. Under these assumptions the quiescent X-ray level, during the last part of the first night and the second night, corresponds to a luminosity of about 5 10²⁹ erg/s for a distance of 140 pc to BP Tau (using the energy conversion factor for the PSPC detector as given by Neuhäuser et al. 1995a). These estimates of the X-ray luminosities are considerably lower than the corresponding optical luminosities. For instance, during the optical event of the second night the peak luminosity was 5 10³¹ erg/s. Because no enhancement in the X-ray count rate was observed, a limit on the fraction of flare energy emitted as observable soft X-ray can roughly be estimated as less than a few percent of the optical emission. For flare stars the ratio between energies in the optical and X-ray regions are typically 0.1 to 1 (Byrne 1989). Since the total energy released in the optical flare of BP Tau was a few times 10³⁴ erg one should expect an observable increase in the X-ray count rate if the flare was similar to what is observed for flare stars.

The appearance of the tentative short-term X-ray event observed on night 1 (see Sect. 3.2 and Fig. 4) is reminiscent of the flares observed on dMe flare stars. Since the X-ray data was binned into 400 sec intervals we could not estimate, for instance, the decay time with a high accuracy. The rise and decay time, however, are consistent with values of 5 minutes and 40 minutes (to the approximate -level), respectively. Unfortunately no simultaneous optical data was obtained for the event but the morphological similarity of this X-ray event to the flares observed on flare stars strengthens the idea that at least some of the X-ray emission arises in magnetically active regions.

In Paper I it was proposed that the optical variability, even on short time-scales, are due to the accretion of circumstellar material onto the stellar surface. It is possible to produce X-ray emission from an accretion shock if the inflow of the accreting material is magnetically controlled (Gullbring 1994), but the rather low gravitational potential of T Tauri stars would typically limit the hardness of the arising X-ray emission to below 0.5 keV, and would thus only contribute to the X-ray emission at very low energies.

Optical variability caused by variable accretion can be very energetic. A change in the accretion rate of only a few percent is sufficient to produce variations in luminosities of more than 10³¹ erg/s (see Paper I). Flux variations in the optical caused by the release of magnetic energies (as for solar-type flares) would then be heavily masked by the accretion-induced variations. It is possible that if the optical events as well are produced by the release of magnetic energy, then the corresponding X-ray emission would be very weak and/or absorbed in the energy process. However, the lack of a corresponding increase in the optical activity when the X-ray count rates increases, as observed during the third and especially the fifth night is difficult to envision in this context. One possibility is that the X-ray emission is not produced close to the stellar surface, as for localized stellar flares, but instead arises in magnetic regions further out from the star like in a dipole magnetosphere. Since for CTTS such a magnetosphere would interact with the circumstellar disk, complex magnetic field configurations would be expected and release of magnetic energy would certainly occur (see for instance Aly & Kuijpers 1990, and the discussion by Feigelson et al. 1994). In that case it would be possible to have variable X-ray radiation far from the stellar surface without corresponding optical emission.

© European Southern Observatory (ESO) 1997

Online publication: May 26, 1998

helpdesk.link@springer.de