Click on the H-alpha profile at left to see a movie of the H-alpha line profile variations in the Classical T Tauri star SU Aur. The X-axis of the plot is a velocity scale indicating how fast material is moving in a reference frame centered on SU Aur. The Y-axis is the relative intensity of the line profile. The data used to make this movie was collected at Lick Observatory (pictured above) and at the McMath telescope of the National Solar Observatory (also pictured above). Much of the data shown here is taken on a nightly or more rapid basis, and has been linearly interpolated in intensity from one observation to the next to provide a smooth visual variation. When there are gaps in the data longer than 2 days, a single blank frame has been inserted and the line profile simply appears to jump discontinuously.
The strongest impression one gets when viewing this movie is an extreme variety of line profiles and the sense that most of the variations are chaotic in nature. However, there are some systematic (periodic) variations occuring in the line profile. The longest, continuous set of observations in this data cover 14 nights in February 1988, with a total of 26 spectra taken during this time. These data are shown in the figure below as a 3-dimensional surface plot. In this plot the X-axis is the velocity scale in km/s, the Y-axis is time in days, and the Z-axis is the relative intensity scale. One striking visual impression from this figure is the apparent periodic variations seen in the blue (near) wing of the H-alpha line. Detailed examination of this data does reveal a blue-shifted absorption feature (appearing at about -160 km/s), indicative of the wind flowing away from SU Aur, which varies with a period of about 3 days - the same as the rotation period of the star.
The H-beta line is another highly variable line in SU Aur which probes what is happening in the material closer to the star than is seen in the H-alpha line. The H-beta line profile also shows the same periodic variations in the wind from SU Aur that were seen in H-alpha. However, unlike H-alpha, the H-beta line shows evidence for rotationally modulated accretion onto the star. In addition to blue-shifted absorption features indicative of a wind, both the H-alpha and H-beta line profiles often show red-shifted absorption features indicative of material falling onto the star. These red-shifted absorption features appear more often in the H-beta line profile, and the strength of this absorption feature in the H-beta line varies periodically, again with a period of about 3 days. It turns out that in SU Aur, the strength of the wind and the accretion flows both vary periodically with the rotation of the star, but they vary 180-degrees out of phase from one another.
It turns out that all these facts fit nicely into the paradigm of magnetocentrifugally driven flows developed by Frank Shu and his collaborators ( Shu et al. 1994). The basic picture is shown below. The star is believed to possess a strong, basically dipolar, magnetic field which truncates the surrounding accretion disk. Some of the material in the accretion disk then falls onto the star along this magnetic field. The surface layers of the disk are partially photo-ionized due to its close proximity to the star, and electric currents are set up in the disk which try to exclude the magnetic field from the disk interior. These currents create magnetic field lines which are anchored in the disk and which open up away from both the disk and the star. The surface layers of the disk get captured by these magnetic fields and are flung away from the star in a magnetocentrifugally driven wind. Star formation theory suggests that the plane the accretion disk forms in is always the equatorial plane of the star. When we view the star, we only get to see one half (top or bottom) of the whole scene because the disk blocks are view of the other side. There are no constraints on the orientation of the stellar magnetic field. If the magnetic axis of the star is not aligned with the stellar rotation axis (which is the case for the Earth's magnetic field), then the geometry is such that when conditions are favorable for material to flow onto the star, the wind is inhibited some; and half a rotation period later, the wind can flow freely, but the material falling onto the star is held up a bit. This then leads to the strength of the wind and accretion flows being modulated by the stellar rotation, and also explains why the two vary completely out of phase with one another.
The results of this work have appeared in Giampapa, Basri, Johns, & Imhoff 1993 and in Johns and Basri 1995.