Accretion Models

The accretion models considered here are very simple. Matter falls freely and radially in a spherical symmetric geometry. Such a simple accretion scenario is not expected to occur in T Tauri stars where, instead, the accretion is supposed to take place along columns in a magnetospheric accretion picture. The spherically symmetric calculations can nevertheless be instructive.

As expected, spherically symmetric accretion models produce IPC Pa Beta and Br Gamma line profiles. Model line profiles were computed to try to match a sample of observed Pa Beta lines displaying IPC profiles that span different emission strengths. The models were constructed in order to match the emission peak of the lines and they are displayed in Figure 6.6. The results shown are for isothermal accretion models with temperature 7000 K and with constant turbulent velocity equal to 40 km/s. The radius from which infall occurs, which also corresponds to the size of the region modeled, was varied in order to produce line widths (of the emission component) similar to those observed. The adopted values were 10 R_* for YY Ori, 8 R_* for DO Tau, BP Tau and HP Tau, 6 R_* for GI Tau and 3 R_* for GK Tau. The emission strength is matched by varying the mass accretion rate. The mass accretion rates range between to M yr^-1, in good agreement with the results from Gullbring et al. 1998. Figure 6.7 show the comparison between the observed Br Gamma lines for the stars which Pa Beta lines are displayed in Figure 6.6 and the model calculations.

As can be seen from Figure 6.6 although the model profiles are all IPC the redshifted absorption is too deep, and at times too wide, when compared with the observations. The exception is the model profile for GK Tau, in which both the emission and the absorption components are reasonably well matched to the observed profile. It is very difficult with the spherically symmetric models explored here to match simultaneously the strength of both the emission and the absorption components.

The velocities at which the model line peaks occur are usually consistent with those observed and the observed line widths can be reasonably well explained by choosing appropriately the size of the modeled region, i.e. where free fall begins. In these accretion models there is no need to introduce a large turbulent velocity in order to explain the line widths. Such large velocities had to be introduced in order to explain the width of the CW Tau Pa Beta line profile (see the previous section).

The accretion models constructed to match the observed Pa Beta peak intensities and line widths also result in a reasonably good match for these two parameters at Br Gamma (see Figure 6.7). The problem with the strength of the predicted redshifted absorption is even more clear at Br Gamma though, where rather than being shallower than the model profiles, they are in some cases completely absent (eg. DO Tau and BP Tau in Figure 6.7).

It is interesting to note that while the line profile modeling carried out using CLOUDY allows one to match simultaneously the peak intensities of both Pa Beta and Br Gamma, the Br Gamma line profiles resulting from the magnetospheric accretion modeling by Muzerolle, Calvet & Hartmann (1998) produce peak intensities which are significantly higher than what the observations show. Computation of both Pa Beta and Br Gamma line profiles by Muzerolle and co-workers is advisable in order to check the consistency of their accretion model calculations (both Pa Beta and Br Gamma should simultaneously match the observations).

Non-spherically symmetric accretion models imply that the strength, and even the presence, of the redshifted feature depends on the viewing angle of the observations. This might lead to a situation where the emission strength can be matched by tuning the accretion rate and the depth of the absorption feature is controlled by the viewing angle. Also, one should not forget the fact that, in this work, an extra continuum component due to the presence of the accretion shock is not included. According to Hartmann, Hewett & Calvet (1994) the temperature of the accretion shock has a large influence on the presence of the redshifted absorption, as discussed in the previous chapter.

The possibility of producing Type I line profiles with wind models was discussed in the previous section. The conclusion was that the wind models explored here could not produce this type of line profiles. How about infall models?

If one increases the accretion rate, and therefore the density of the flow, the lines get more and more optically thick and finally become roughly symmetric, i.e. Type I profiles. However, the situation is similar to that discussed for wind models, that is, when the lines become generally symmetric the intensity of the line peak is higher than observed by one to two orders of magnitude. A similar situation results from increasing the flow temperature.

Similarly to the wind models, the accretion models explored in this work fail to produce Type I profiles with the observed peak intensities. As for the wind models discussed in the previous section, the spherically symmetric characteristic of the accretion models explored here is the responsible for the far too high peak intensities. The typical mass accretion rates needed to make Type I line profiles (in the context of the accretion models computed here) are one to two orders of magnitude higher than deduced from observations and result in peak intensities for the NIR lines also one to two orders of magnitude higher than observed. A way of decreasing the mass accretion rate is to drop the spherically symmetric characteristic of the models which would be accompanied by a decrease in the model peak intensities. Models with flow geometries other than spherical symmetry have the added complication of being line of sight dependent, as thoroughly discussed in the previous chapter, in the context of the accretion models computed by Hartmann, Calvet, Muzerolle and co-workers.