All Figures

	Figure 1: Arae (unresolved star + uniform disk) model upper limit diameters (in mas), as a function of wavelength, from the 2005 MIDI data described in Paper I for the June 16 (solid line) and June 17 nights (dotted line). Note that the large values obtained for larger wavelengths are essentially due to poor calibration and large error bars rather than a true physical effect.
Open with DEXTER


In the text

Figure 2: Arae (unresolved star + uniform disK) model diameters's (in mas) dependence on the baseline position angle. Solid line: AMBER measurements in the continuum at 2.1 m. Dashed lines: MIDI measurements. The length of the plot corresponds to the values of the error bars (note the very large error bars for the MIDI data). The dotted line is the direction of the majoraxis of the envelope estimated from polarization measurements (PA = 172 ) obtained by McLean & Clarke (1979) and Yudin et al. (1998).

Open with DEXTER


In the text

Figure 3: "Toy'' models used to fit the AMBER and MIDI measurements. The 4 upper pictures are a cut of the circumstellar disk in a plane defined by the observer line of sight and the stellar rotational axis (the observer is on the right for each picture); the corresponding projections into the skyplane with the interferometric data points from MIDI and AMBER over-plotted are the central pictures, whereas a "3D artist view'' is plotted into the lower row for each model.

Open with DEXTER


In the text

	Figure 4: Flatness ratio of the projected envelope versus opening angle calculated with our simple disk model for different inclination angles. Horizontal lines show flatness from models 1 and 2.
Open with DEXTER


In the text

	Figure 5: Arae's Spectral Energy Distribution (SED) from various sources in the literature (see text). Dashed line: emission from the central star only assuming a blackbody with  = 4.8 ,  = 18 000 K, and d=105 pc. Fit of the SED with SIMECA, taking into account the envelope free-free and free-bound contribution (solid line) between 0.3 and 20 m.
Open with DEXTER


In the text

	Figure 6: Example of the SIMECA density distribution as a function of the stellar latitude from the pole ( ) to the equator ( ) at 10 R*.
Open with DEXTER


In the text

	Figure 7: Left: rotational velocities as a function of the stellar latitudes from the pole ( ) to the equator ( ) at various stellar radii (over-plotted on the graph). Right: terminal velocity as a function of the stellar latitudes.
Open with DEXTER


In the text

Figure 8: Intensity map in the continuum at 2.15 m obtained with SIMECA for our best model parameters. The inclination angle is 55 , and the central bright region is the flux contribution from the thin equatorial disk, whereas the smoother regions originate in the stellar wind. The brightness contrast between the disk and the wind is globally 30, but can reach 100, if you compare the inner region of the disk with the outer parts of the wind.

Open with DEXTER


In the text

Figure 9: Line profiles modeled with the SIMECA code using a thin disk + polar wind scenario. Modeled profiles are dotted lines, whereas observed ones are in solid lines. For Br, the modeled profile (dotted line) obtain with SIMECA is convoluted with a 15  Gaussian, corresponding to the AMBER spectral resolution of 1500. The convoluted Br profile we obtain is the dashed line superimposed on the observed one (solid line).

Open with DEXTER


In the text

	Figure 10: Differential visibility across the Br line profile for the B0, B1, B2, and B3 baselines. The first picture from the top is the Br line profile. The solid lines are the fits we obtain with SIMECA from our best models whereas the AMBER/VLTI data are the points with error bars.
Open with DEXTER


In the text

	Figure 11: Phase of the visibility across the Br line profile for the B0, B1, B2, and B3 baselines. The solid lines are the fits we obtain with SIMECA from our best model, whereas the AMBER/VLTI data are the points with error bars.
Open with DEXTER


In the text

	Figure 12: Intensity maps across the Br line profiles within spectral channels of 15 , from which the differential visibility modulus and phase are estimated. Note that to increase the image contrast, the central star and the continuum emission have been subtracted.
Open with DEXTER


In the text

Figure 13: parameter as a function of the equatorial velocity, taking into account the stellar photosphere deformation due to its fast rotation. The solid line corresponds to physically possible cases, whereas the dashed line corresponds to rotational values larger than the "breakup'' velocity. The dash-dotted line also corresponds to unrealistic values, but in this case we have not taken into account the stellar photosphere distortion larger than 1.5  at the equator.  = 351 km s-1 corresponds to the minimum vsini found in the literature for this star, i.e., 288 km s-1, by Uesugi & Fukuda (1970) with .

Open with DEXTER


In the text

	Figure 14: from our models and the AMBER data plotted as a function of the  parameter.  is the minimum value obtained for a model with  = 0.48, which is very close to the Keplerian rotation ( = 0.5).
Open with DEXTER


In the text

	Figure 15: Arae differential visibilities ( left) and phases ( right) across the Br line profiles for the B0, B1, B2, and B3 baselines. The solid lines are the fits we obtain with SIMECA from a rotating and expanding scenario described in Sect. 6.6, whereas the AMBER/VLTI data are the points with error bars.
Open with DEXTER


In the text

	Figure 16: Arae differential visibilities ( left) and phases ( right) across the Br line profiles for the B0, B1, B2, and B3 baselines. The solid lines are the fits we obtain with SIMECA from the model 3 described in Sects. 4.3.2 and 6.6, whereas the AMBER/VLTI data are the points with error bars.
Open with DEXTER


In the text