4 Age structure of LH 101/104

4.1 Transformation to $M_{\rm bol}$, T$_{\rm eff}$ for known spectral type stars

The stellar energy distribution derived by color calibration is not accurate due to uncertainties of the bolometric magnitude ($M_{\rm bol}$) for hot stars with log$T_{\rm eff}$ > 4.4 (Massey et al. 1995). Therefore we need to place these stars in the H-R diagram with the help of spectral types obtained by spectroscopy.

We deduced values of E(B-V), $M_{\rm bol}$ and log$T_{\rm eff}$ which are listed in Tables 1 and 2 where Cols. 1 and 2 give the stellar identification, and Col. 3 the epoch 2000 coordinates. We adopted a distance modulus m - M = 18.5 (e.g. Panagia et al. 1991) for the LMC, and an ordinary ratio R = 3.1 = A_V/E(B-V).

The (B-V)₀ intrinsic color indices, the bolometric corrections (BC) and $T_{\rm eff}$ for different spectral types were taken from Schmidt-Kaler (1982), except for O stars for which the values from Chlebowski & Garmany (1991) were used. For WR stars the $T_{\rm eff}$ is from Hamann et al. (1991) and Koesterke et al. (1991) and BC from Smith $\&$ Maeder (1989).

4.2 Transformation to $M_{\rm bol}$, $T_{\rm eff}$ for unknown spectral type stars

For stars of unknown spectral type, we first determined the color factor Q = (U-B) - 0.72(B-V), a coarse indicator of spectral type and then derived the stellar parameters following the procedure of Massey et al. (1995).

The differences between the observational and the theoretical values of the Q parameter listed in Col. 7 of Tables 1 and 2, show that in a nebula and bad photometric conditions this parameter is not reliable for hot stars, contrary to the spectroscopy classification (e.g. $Q \sim - 0.80$ for the stars 5-31 and 5-52 instead of $\sim - 0.95$) .

4.3 Hertzprung-Russel diagrams of LH 101 and LH 104

The resulting H-R diagrams log $T_{\rm eff}$ vs. $M_{\rm bol}$ for both OB associations are shown in Figs. 6a and b. We discriminated between stars with spectral classifications and those with photometry only, denoting them with different symbols. We also overlaid a few stellar evolutionary tracks and isochrones calculated by Schaerer et al. (1993) for a metallicity of Z = 0.008.

  

Figure 6: a) and b) are the HR diagrams of the OB associations LH101 and LH104. Filled circles, squares and diamonds show data with spectral classification and they indicate luminosity V, III and I respectively. The crosses show data determined only by photometry. Empty diamonds indicate WC types and empty triangles WN types. For O star belonging to systems the $M_{\rm bol}$ obtained from observed integrated magnitude is represented by filled triangles. The stellar evolutionnary tracks with the corresponding stellar mass identified in $M_{\odot}$ and isochrones in dashed lines are overlaid (Shaerer et al. 1993)

In Tables 1 and 2 the apparent magnitudes and the colors of the three WR + O systems refer to the observed integrated magnitude, as is the case for the spectroscopic binary star 5-65 of type O7If. The composite spectra are shown in Fig. 5a and Fig. 5d. The $M_{\rm bol}$ derived from these integrated magnitudes does not have any clear meaning. However these stars should be taken into account in the HR diagram and the IMF. As a first approximation, since spectra of both components are visible, we have assumed that these systems are composed of two stars of similar magnitudes, and that the O components in the WR systems are of luminosity class V.

In the HR diagrams of Figs. 6a and b, we have plotted the positions of the O+O and WR+O systems as they would appear with the $M_{\rm bol}$ values which result from the apparent integrated magnitude and $T_{\rm eff}$ of the O type component. This illustrates how the stars would appear in the HR diagram if we had no information about their binarity.

Arrows are drawn from these positions in the HR diagram to where the components would be located assuming two equal magnitude stars of the observed spectral types. In LH 104 this procedure locates the WR+O systems in the bin $40-60\,M_{\odot}$.

4.4 Age

LH 101 seems to consist of two subpopulations (Table 4). A comparison with the isochrones shows that the younger subgroup has an age of $\leq$ 2 Myr and the older one of $\sim$ 3-6 Myr. As noted by Testor & Niemela (1996), stars outside the nebulosity seem to be somewhat more evolved than those inside the nebulosity. Thus sequential star formation appears to be present even at scales less than 50 pc.

LH 104 consists mainly of a young population spreading out between 2 and 6 Myr (Table 4). The HR diagrams also show in both associations a group of older and less massive stars.

 

Table 4: Distribution of stars with isochrones

 

Notes to Table 4:

Column 1: Masses
Columns 2, 3, 4, 5, 6: Number of stars per 2 Myr interval in each mass bin.