hei_stars.tex to appear int the 4th ESO/CTIO Workshop In-Reply-To: <199604281743.NAA20695@sgrastar.astro.umd.edu> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII % % % \documentstyle[11pt,psfig,paspconf]{article} % \begin{document} % My definitions \def\NH{\hbox{N$_{\hbox{H}}$}} \def\NHe{\hbox{N$_{\hbox{He}}$}} \def\Mdot{\hbox{$\dot {\rm M}$}} \def\Rsun{\hbox{R$_\odot$}} \def\Mstar{\hbox{M$_*$}} \def\Rstar{\hbox{R$_*$}} \def\Lsun{\hbox{L$_\odot$}} \def\Lstar{\hbox{L$_*$}} \def\Msun{\hbox{M$_\odot$}} \def\Msunyr{\hbox{M$_\odot\,$yr$^{-1}$}} \def\Teff{\hbox{T$_{\rm eff}$}} \def\Tstar{\hbox{T$_*$}} \def\Vinf{\hbox{V$_\infty$}} \def\kms{\hbox{km$\,$s$^{-1}$}} \def\kpc{\hbox{kpc}} \def\Hz{\hbox{Hz}} \def\mum{\hbox{$\mu$m}} \def\yr{\hbox{yr$^{-1}$}} \def\HeI{He\,{\sc i}} \def\HeII{He\,{\sc ii}} \def\cm{\hbox{\rm cm}} \def\Kel{\hbox{\rm K}} \def\keV{\hbox{\rm keV}} \def\eV{\hbox{\rm eV}} \markboth{Najarro \etal\ }{Quantitative Spectroscopy of the \HeI\ Cluster} \title{Quantitative Spectroscopy of the \HeI\ Cluster} \author{Francisco\,Najarro and Rolf\,P. Kudritzki\altaffilmark{1}} \affil{Universit\"ats-Sternwarte M\"unchen, Scheinerstra{\ss}e 1, D-81679 M\"unchen, Germany} \author{Alfred\,Krabbe, Reinhard\,Genzel and Dieter\,Lutz} \affil{Max-Planck-Institut f\"ur extraterrestrische Physik, 85740 Garching bei M\"unchen, Germany} \author{ D.\,John\,Hillier } \affil{Department of Physics and Astronomy, University of Pittsburgh, 3941 O'Hara Street, Pittsburgh, PA 15260, USA} \altaffiltext{1}{Max-Planck-Institut f\"ur Astrophysik, 85740 Garching bei M\"unchen, Germany} \begin{abstract} We present first results on quantitative infrared spectroscopy of the brightest \HeI\ emission line stars in the Galactic center. The observed \HeI\ and H broad emission lines are caused by extremely strong stellar winds ($\Mdot\sim$ 5 to 80 $\times 10^{-5}\, \Msunyr$) with relatively small outflow velocities (\Vinf $\sim$ 300 to 1000~\kms). The effective temperatures of the objects range from 17,000\,K to 30,000\,K with corresponding stellar luminosities of 1 to $30 \times 10^{5}$\Lsun. Strongly enhanced helium abundances (\NHe/\NH$>.5$) are found. These results indicate that the \HeI\ emission line stars are evolved blue supergiants close to the evolutionary stage of Wolf-Rayet stars. They power the central parsec and belong to a young stellar cluster of massive stars which formed some $5 \times 10^6$ years ago. Statistical evidence for a concentration of dark mass ($\sim$3$\times$10$^6$\Msun) in the Galactic center is derived from the stellar velocity dispersion. \end{abstract} % \keywords{ \HeI\ stars, stellar winds, ionizing flux} \section{ Introduction} \label{sec-intro} The \HeI\ emission line cluster in the Galactic center discovered by Krabbe \etal\ (1991) provides a unique opportunity to test star formation and evolution in connection with the ionization and energetics of the Galactic center. Results of a detailed spectroscopic investigation of the brightest source (AF star) of the cluster in the \hetwot\ line by Najarro \etal\ (1994) revealed that the AF star is a helium-rich blue supergiant/Wolf-Rayet star. It is characterized by a strong stellar wind and constitutes a moderate source of Lyman continuum photons. The latter result clearly suggested that the cluster of \HeI\ stars can make a significant contribution to the total luminosity and the Lyman continuum flux of the inner parsec. Therefore, it was crucial to analyze the spectra of other \HeI\ objects of the cluster and to obtain their stellar parameters to better constrain their role in the energetics of the central parsec. We report here first results of an extensive new study of the Galactic center stellar cluster. Stellar parameters are derived for eight \HeI\ objects. \section{ Summary of observations and first results} \label{sec-obsres} Spectroscopic observations of the \HeI\ cluster have been carried out using CSG4, FAST, SHARP and 3D, and are described in detail in Krabbe \etal\ (1995). \begin{table} \caption[]{ Observed properties for the brighter GC \HeI\ sources. All fluxes are dereddened and have been scaled to a distance of 1\,kpc assuming a distance of 8.5\,kpc to the Galactic Center.} \label{tab-obsothe} \begin{center}\scriptsize \begin{tabular}{|lclllllll|} \hline Star&A(H-K)&F$_{2.2\mu}$&\HeI$_{2.06}$&\bgam & \pap & \HeI$_{2.112}$&\HeII$_{2.19}$&FWHM\\ & & & &\HeI$_{7-4}$&\HeI$_{4-3}$&\CIII+\NIII& & \\ & &(Jy)& \multicolumn{5}{c}{$( \;\;\;10^{-10}\;\; $erg$ \;\; s^{-1} \;\; $cm$^{-2} \;\;\;)$} &\kms\\ \hline \hline 7W &2.7&20.1&5.56&1.10&$<$13 &1.75&$<$.3 &1000\\ 13E1&2.7&139.&11.6&5.66&$<$50 &8.47&$\sim$.9&1000\\ 16NE&2.0&125.&3.02&1.11&$\cdots$&1.21&$<$.3 &600\\ 16NW&2.2&72.4&1.16&.421&$\cdots$&.462&$\cdots$&1000\\ 16C &2.3&115.&2.69&.894&$\cdots$&1.98&$<$.3&600\\ 16SW&2.4&145.&5.04&2.09&$\cdots$&2.02&$<$.5&650\\ 15SW&2.3&66.3&7.53&1.04&$<$7 &.909&$\cdots$&850\\ 15NE&2.3&46.3&6.27&1.30&$<$12 &1.19&$<$.2 &950\\ \hline \end{tabular} \end{center} \end{table} Table~\ref{tab-obsothe} summarizes the observational results obtained for the \HeI\ objects analyzed in this work. Before using the above results as constraints to model the objects some remarks must be made concerning some of the observed values. From Table~\ref{tab-obsothe} we see that the extinction shows high variability on small spatial scales, and therefore individual estimates must be made for each object under consideration. For the wavelength range around the K-band we assume $A(K) = 1.53 ( A(H)-A(K) )$ and $A(\lambda) = A(K) (\lambda/2.2\mu)^{-1.75}$ (Krabbe \etal\ 1995). The measured FWHM of the emission lines were used as an initial guess for \Vinf, the terminal velocity of the stellar wind. The final \Vinf\ value was then obtained by fitting the computed profiles iteratively with the observed ones. Except for IRS\,13E, all \HeII\,2.189\my\ line fluxes are upper limits. All available \pap\ fluxes were used as checks for the consistency of the results since the measurements suffer from a much higher uncertainty. The observed feature at $\lambda$$\approx$2.11\my\ is expected to be a blend of the \hedubt\ and \NIII\ and/or \CIII\ lines (Najarro 1995). Indeed, detailed inspection of the observed spectra indicates that in many of the sources the \hedubt\ we clearly identify a broad blue component centered at $\lambda$$\approx$2.105\my\ as well as two red components at $\lambda$$\approx$2.115\my\ and $\lambda \approx 2.121$\my. Hence, the observed $\lambda$\,2.112\my\ fluxes were considered as upper limits, and, if possible, the true \hedubt\ line flux was estimated by subtracting a $\sim$2\Vinf\ wide region centered at 2.1120\my\ from the observed 2.11\my\ feature. As pointed by Najarro \etal\ (1994) the well separated \HeI\ and \pap\ components constitute the best indicator in the observed wavelength range to determine whether hydrogen is present or highly depleted in the objects. The {\bf CGS4} spectra of IRS7W, 13E1, 15NE and 15SW do not show the hydrogen \pap\ line after removing the diffuse emission component, which indicates that these objects must be pure He stars (see also Geballe \etal\ 1994). To model the objects we proceeded as in Najarro \etal\ (1994) and used the iterative, non-LTE method presented by Hillier (1987, 1990) which solves the radiative transfer equation in spherical geometry, subject to the constraints of statistical and radiative equilibrium, for the expanding atmospheres of early-type stars. We considered a pure H-He atmosphere with 12 H, 49 \HeI\ (n$\leq$10) and 12 \HeII\ levels. The stellar parameters obtained for the above eight objects are presented in Table~\ref{tab-fitsothe}. Agreement between observed and computed values was always better than 10\%. Below we disscus our results for the analyzed objects. \begin{table}[hbt] \caption[]{Derived stellar parameters for GC \HeI\ sources. \R23~is the photospheric radius at Rosseland optical depth 2/3, \T23~the effective temperature derived for \Lstar\ and \R23\ and $\eta = \Mdot \Vinf / ( \Lstar /c ) $ is the ``performance'' number.} \label{tab-fitsothe} \begin{center}\small \begin{tabular}{|lllllllll|} \hline &\multicolumn{8}{c|}{IRS Object}\\ &\multicolumn{1}{c}{16NE} & \multicolumn{1}{c}{16C} & \multicolumn{1}{c}{16SW} & \multicolumn{1}{c}{16NW} & \multicolumn{1}{c}{7W} & \multicolumn{1}{c}{13E1} & \multicolumn{1}{c}{15NE} & \multicolumn{1}{c|}{15SW}\\ \hline \hline \Rstar~{\footnotesize (\Rsun)} & 85 & 85 & 90 & 70 & 23 & 60 & 45 & 62\\ \Lstar~{\footnotesize ($10^{5}\Lsun$)} & 22.0 & 19.0 & 25.9 & 10.3 & 4.10 & 22.6 & 9.10 & 10.5\\ \Teff~{\footnotesize $(10^{4}$K)} & 2.41 & 2.32 & 2.44 & 2.20 & 3.04 & 2.89 & 2.66 & 2.35\\ \R23~{\footnotesize (\Rstar)} & 1.12 & 1.12 & 1.14 & 1.10& 1.53 & 1.95 & 1.17 & 1.13 \\ \T23~{\footnotesize $(10^{4}$K)} & 2.27 & 2.20 & 2.29 & 1.77 & 2.46 & 2.07 & 2.46 & 2.20 \\ He/H & 1.0 & 3.0 & 1.3 & 1.3 & $>$500 &$>$500&$>$100& $>$100\\ \Mdot~{\footnotesize ($10^{-5}\Msunyr$)} & 9.50 & 10.5 & 15.5 & 5.30 & 20.7 & 79.1 & 18.0 & 16.5 \\ \Vinf~{\footnotesize(\kms)} & 550 & 650 & 650 & 750 & 1000 & 1000 & 750 & 700 \\ $\eta$ & 1.17 & 1.78 & 1.92 & 1.91 & 24.9 & 17.3 & 7.33 & 5.43 \\ V$_{rad}$ (\kms) & -40 & 200 & 300 & -100 & -250 & 0 & -150 & -200 \\ Log Q(H$^{+}$) & 49.6 & 49.5 & 49.7 & 49.0 & 49.2 & 49.9 & 49.4 & 49.3 \\ Log Q(He$^{+}$) &$<45$ &$<45$ & $<45$& $<44$& $<44$ & 45.0 &$<44$ & $<44$ \\ Log Q(He$^{++}$) &$<32$ &$<32$ & $<32$ & $<31$& $<33$ &$<33$ &$<33$ & $<32$ \\ \hline \end{tabular} \end{center} \end{table} \subsubsection{IRS\,7W \& IRS\,13E.} \label{sucsuc-irs7w} IRS\,7W is the hottest (\Teff$\approx$30kK) but also the least luminous (not considering the AF star) analysed object of the new sample. The stellar parameters derived are similar to those of WN8 and WN9 stars (Hamann \etal\ 1993, Crowther \etal\ 1995a). Its observed near-IR spectrum clearly indicates a later spectral type than WN8 (Hillier 1985), and probably corresponds to a WN9 or WN10 type. However, unlike WNL stars, H is much too depleted in IRS\,7W, He/H$\geq$500. It is important to \begin{figure}[thb] \hbox{\hspace{3.9cm}{\bf 7W}\hspace{4.5cm}{\bf 13E}} \vspace{-.25cm} \hbox{\hspace{.8cm}\psfig{figure=$HOME/vax/tesi/figures/irs7w_hei205.ps,height=4.2cm,angle=90} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs13_hei205.ps,height=4.2cm,angle=90}} \vspace{-0.35cm} \hbox{\hspace{.8cm}\psfig{figure=$HOME/vax/tesi/figures/irs7w_bga.ps,height=4.2cm,angle=90} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs13_bga_fud.ps,height=4.2cm,angle=90}} \vspace{-0.35cm} \hbox{\hspace{.8cm}\psfig{figure=$HOME/vax/tesi/figures/irs7w_hei212.ps,height=4.2cm,angle=90} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs13_hei212.ps,height=4.2cm,angle=90}} \vspace{-0.2cm} \vspace{-0.15cm} \caption[]{ {\bf left (from top to bottom):} IRS\,7W observed {\bf CGS4} (solid) and computed (dashed) \hetwot, \bgam\ and \hedubt\ profiles. {\bf right (from top to bottom)} IRS\,13E observed {\bf 3D} (solid) and computed (dashed) \hetwot, \bgam\ and \hedubt\ profiles. Diffuse emission has been removed for both objects.} \label{fig-7w13tot} \end{figure} note that the hydrogen abundance is strongly constrained by the absence of H \pap. This, together with the high degree of extension of the atmosphere \R23/\Rstar$>$1.5 and its high ``performance'' number ($\eta$$\approx$25) relate the object to WNE-w stars (Hamann \etal\ 1993), though the effective temperature of IRS\,7W is low compared with other such stars. IRS\,13E is also a rather hot source (\Tstar$\approx$29kK) but because of its enormous degree of extension \R23/\Rstar$\approx$2, V(\R23)$>$\Vinf/3, it presents a much lower \T23\ value. This is essentially caused by the extremely high rate of mass-loss of the object (\Mdot$\approx 8 \times 10^{-4}$\Msunyr). Like IRS\,7W, IRS\,13E shows no stellar hydrogen features in its spectrum which confirms its highly evolved status, and hence it may be classified as a WN9 or WN10 star. Interestingly, its performance number ($\eta \approx$17) is slightly lower than that of IRS\,7W, but still as high as those of WNE-w stars, and very similar to the values obtained for the AF star (Najarro \etal\ 1994). For the \HeII\,2.189\my\ line we obtain a much weaker line flux (factor of 3) than observed. We attribute this result to the neglect of line blanketing (Hillier 1995). Nevertheless, it is important to stress that the presence in emission of both the \hetwot\ and \HeII\,2.189\my\ lines constitutes an important constraint for the effective temperature of the object as this occurs within an small effective temperature range ($\Delta$T$\leq$2000\,K). Figure~\ref{fig-7w13tot} shows the general good agreement of our computed line profiles for IRS\,7W and 13E with the observed \hetwot, \bgam\ and \hedubt\ profiles. Both \hetwot\ lines (including the strong electron scattering wings observed in IRS\,13E) are well reproduced. For IRS\,13E the computed absorption dip is clearly deeper than the observed one. This discrepancy may be due to stellar rotation. The fits to the observed \HeI\,(7$-$4) lines confirm the absence of H in the winds of IRS\,7W and IRS\,13E. Our models reproduce as well the central part ($\pm$\Vinf) of the 2.11\my\ observed feature corresponding to the \hedubt\ line and reveal the important contribution of the \NIII/\CIII\,(8-7) lines (especially in IRS\,13E). \subsubsection{IRS\,16 sources} \label{sucsuc-irs16} The main and perhaps most interesting result from our analysis of the the IRS\,16 sources, \ie\ IRS\,16\,NE, C, SW and NW, is that they show similar He/H ratios (He/H$\sim$1, see Table~\ref{tab-fitsothe}). Since the values obtained for the stellar luminosities and effective temperatures of the objects are also very close (only IRS\,16NW has a slightly lower \Lstar\ and \Teff), we may conclude that the brighter IRS\,16 sources have similar ages. \begin{figure} %\vspace{14.5cm} \vspace{-.55cm} \hbox{\hspace{.4cm}\psfig{figure=$HOME/vax/tesi/figures/irs16ne_hei205.ps,width=4.3cm} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs16ne_bgam.ps,width=4.3cm} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs16ne_hei212.ps,width=4.3cm}} \vspace{-.35cm} \hbox{\hspace{.4cm}\psfig{figure=$HOME/vax/tesi/figures/irs16c_hei205.ps,width=4.3cm} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs16c_bgam.ps,width=4.3cm} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs16c_hei212.ps,width=4.3cm}} \vspace{-.35cm} \hbox{\hspace{.4cm}\psfig{figure=$HOME/vax/tesi/figures/irs16sw_hei205.ps,width=4.3cm} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs16sw_bgam.ps,width=4.3cm} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs16sw_hei212.ps,width=4.3cm}} \vspace{-.35cm} \hbox{\hspace{.4cm}\psfig{figure=$HOME/vax/tesi/figures/irs16nw_hei205.ps,width=4.3cm} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs16nw_bgam.ps,width=4.3cm} \hspace{-.5cm}\psfig{figure=$HOME/vax/tesi/figures/irs16nw_hei212.ps,width=4.3cm}} \vspace{-.35cm} \caption[]{ Observed {\bf 3D} profiles for the IRS\,16 sources compared with the calculations (\hetwot, \bgam\ and \hedubt\ from left to right). Diffuse emission has been removed.} \label{fig-16sourtot} \end{figure} This result is also supported by the almost identical values obtained for the performance numbers (see Table~\ref{tab-fitsothe}), since $\eta$ constitutes a powerful indicator of the evolutionary status of the star. The presence of hydrogen in the observed spectra is consistent with a WNL evolutionary phase, though the effective temperatures obtained for the IRS\,16 objects is rather low when compared to the values derived for other later type (WN9-10) WNL stars (Crowther \etal\ 1995a). On the other hand, except for the enhanced He content, the stellar parameters of the IRS16 sources (\Rstar, \Lstar, \Mdot\ and \Vinf) also resemble those of some known LBVs such as P~Cygni (\eg\ Najarro 1995, Langer \etal\ 1994). Recent work by Langer (private communication) shows that the inclusion of pulsational instabilities in the evolutionary calculations of massive stars leads to a more efficient mixing, and hence we may conclude that the IRS16 objects are just finishing their LBV phase. We retain, nevertheless, the Ofpe/WN9 classification for the IRS16 sources as their He content is close to that of the AF star and they show very similar stellar parameters to other known Ofpe/WN9 objects (Crowther \etal\ 1995a). Figure~\ref{fig-16sourtot} shows the excellent agreement between the observed {\bf 3D} and computed \hetwot, \bgam\ and \hedubt\ profiles for the IRS\,16 sources. The obvious P~Cygni shape of the observed \hetwot\ profiles, confirms the wind nature of the lines. The presence of a blue absorption dip in the observed IRS\,16SW \bgam\ line may be attributed to the problematic subtraction of the diffuse nebular emission. Once more, our results are able to reproduce the \hedubt\ component quite well. Despite the relatively low S/N ratio, Fig.~\ref{fig-16sourtot} confirms the presence of the \NIII/\CIII\,(8-7) lines in three of the IRS\,16 sources (NE, C and SW), while they seem to be absent in the spectrum of IRS\,16NW. This result is consistent with the lower temperature ($\Delta\Teff \sim$1500\,K) and wind density obtained for the latter, as we expect the strength of the \NIII/\CIII\,(8-7) lines to decrease with decreasing \Tstar\ and wind density. This would also explain the absence of these features in the spectrum of the AF star (lower \Teff) and their relatively large strength in the hotter, denser IRS\,13E and IRS\,7W. \subsubsection{IRS\,15SW \& IRS\,15NE} \label{sucsuc-irs159} IRS15NE and IRS15SW may be considered as ``transition'' objects. These objects have effective temperatures only slightly higher (\Teff $\sim$25000~K) than those obtained for the IRS16 objects suggesting a Ofpe/WN9 classification, the main caveat for a WN9 to WN11 classification being the low effective temperature. \begin{figure} %\vspace{14.5cm} \vspace{-.15cm} \hbox{\hspace{-.2cm}\psfig{figure=$HOME/vax/tesi/figures/cgs4_15sw_206_meet.ps,width=3.8cm} \hspace{-.875cm}\psfig{figure=$HOME/vax/tesi/figures/cgs4_15sw_212_meet.ps,width=3.8cm} \hspace{-.875cm}\psfig{figure=$HOME/vax/tesi/figures/cgs4_15sw_216_meet.ps,width=3.8cm} \hspace{-.875cm}\psfig{figure=$HOME/vax/tesi/figures/cgs4_15sw_palfa_meet.ps,width=3.8cm}} \vspace{-.25cm} \hbox{\hspace{-.2cm}\psfig{figure=$HOME/vax/tesi/figures/cgs4_15ne_206_meet.ps,width=3.8cm} \hspace{-.875cm}\psfig{figure=$HOME/vax/tesi/figures/cgs4_15ne_212_meet.ps,width=3.8cm} \hspace{-.875cm}\psfig{figure=$HOME/vax/tesi/figures/cgs4_15ne_216_meet.ps,width=3.8cm} \hspace{-.875cm}\psfig{figure=$HOME/vax/tesi/figures/cgs4_15ne_palfa_meet.ps,width=3.8cm}} \vspace{-.25cm} \caption[]{ IRS\,15SW and IRS\,15NE observed {\bf CGS4} (solid) and computed (dashed) \hetwot,\hedubt,\bgam\ and \pap\ profiles. Most of the diffuse emission has been removed.} \label{fig-15lines} \end{figure} However, they appear to be pure He sources (as do IRS7W and IRS13E-1) indicating a Wolf-Rayet phase. The latter could be due to the effective mixing through pulsation instabilities cited above. A further indication for this ``transition'' status is given by the performance numbers and terminal velocities shown in Table~\ref{tab-fitsothe}. Figure~\ref{fig-15lines} shows the general good agreement of our model calculations with the observed IRS\,15SW and IRS\,15SW \hetwot, \hedubt, \bgam\ and \pap\ lines. The corresponding large slit width for the {\bf CGS4} pixels (3.1\arcsec) hinders a clean subtraction of the relatively non-uniform background and diffuse emission, and therefore tends to blur the P~Cygni dips of the \HeI\ lines (see Fig.~\ref{fig-15lines}). This effect is even more extreme for the H lines (see the \bgam\ and \pap\ lines in Fig.~\ref{fig-15lines}) where a ``hole'' or a narrow peaked feature at the wavelength of the hydrogen component may result from the presence of a non-uniform background. Despite these problems, our computed profiles are able to reproduce the shapes of the observed profiles satisfactorily (see Fig.~\ref{fig-15lines}) and account for the broader blends due to the \HeI\,(4$-$3) and \HeI\,(7$-$4) lines. We note, once more, the clear presence of the \NIII/\CIII\,2.103\my\ feature bluewards of the \hedubt\ line. \subsubsection{Evolutionary status, ionizing flux and GC mass.} \label{sucsuc-irsflux} Krabbe \etal\ (1995) have analyzed the nuclear star cluster using the star cluster models of Krabbe \etal\ (1994) and conclude that the observations fit a 7$\pm 1 \times 10^6$yrs old decaying burst in which $\sim$30000 stars formed very well. They obtain a model with a bolometric luminosity of L(Bol)$\sim 2.7 \times 10^7$\Lsun\ and a bolometric to Lyman luminosity ratio of ten. Such a model predicts $~15$\,OB stars with L$\geq 3\times 10^5$\Lsun, $~$30 later type WR stars (WNL, WCL and the \HeI\ objects observed) and 2 to 4 KM supergiants with L(K)/L(Lyman)$\sim 3\times 10^{-2}$. Further, the implied supernova rate (one in $4\times 10^4$yrs) agrees with the interpretation of Sgr.~A~East (Genzel \etal\ 1994). Moreover, the presence of a fairly large number ($\sim$10) of moderately luminous late type stars in the central 8\arcsec\ also indicates (see Haller \etal\ 1989) an earlier starburst about $10^8$yrs ago. The above results clearly favor a model in which the central parsec of the Galaxy is powered by a cluster of hot stars, as earlier proposed by Rieke \& Lebofsky (1982) and Allen \& Sanders (1986). The eight objects studied here alone, already supply $\approx 3\times 10^{50}$ Lyman continuum photons per second ($>60$\% of the total Lyman flux as estimated from thermal radio continuum measurements by Genzel \etal\ 1994). Therefore, the hot star cluster can fully account for the bolometric and Lyman ionizing luminosities of the central parsec. None of these objects contribute to the \HeI-continuum. Even IRS13E1 does not provide more than 1\% of the flux required for the He-ionization of the SgrA West HII region, $\log Q($He$^+) \approx 49$ (Krabbe \etal\ 1991). Hence, some hotter O and/or WR stars is required, the presence of which have been recently reported by Krabbe \etal\ (1995). Future, moderate S/N spectra of these objects are crucial to check whether they can or can not account for the \HeI-continuum in the Galactic center. Finally, we may use the derived radial velocities to obtain the velocity dispersion of the objects and therefore, by means of the virial theorem (and/or the Bahcall-Tremaine estimator, Bahcall \etal\ (1981), derive the enclosed mass in the central parsec. Krabbe \etal\ (1995) have obtained velocity measurements from Gaussian fits to the observed \pap, \hetwot, \hedubt\ and \bgam\ lines, and obtained a mass concentration of $\sim$2 to 4$\times$10$^6$\Msun, thus finding a high statistical significance for a dark mass concentration. The latter could be in the form of massive stellar remnants (10\Msun), \ie\ the black hole cluster option suggested by Morris (1993), or more likely as a massive black hole (Genzel \etal\ 1994). \begin{references} %AG Tagung 94 \def\agtag#1{AG Abstract Series, {\bf{10}} ,~#1} %IAU Symposium 163 \def\iaau#1{in: IAU Symp. 163, Wolf-Rayet Stars: Binaries, Colliding Winds, Evolution, eds. K. A. van der Hucht and P. M. Williams, Kluwer, Dordrecht, p.~#1} %IAU Symposium 143 \def\iau#1{in: IAU Symp. 143, Wolf-Rayet Stars and Interrelations with Other Massive Stars in Galaxies, eds. K. A. van der Hucht and B. Hidayat, Reidel, Dordrecht, Holland p.~#1} %Infrarred with Arrays \def\irarr#1{in: Infrared Astronomy with Arrays: The Next Generation, ed. I. McLean, Kluwer, Dordrecht, p.~#1} %IAU Symposium 99 \def\iua#1{in: IAU Symp. 99, Wolf-Rayet Stars: Observations, Physics, Evolution, eds. C.W.H. de Loore and A.J. Willis, Reidel, Dordrecht, Holland p.~#1} %NASA O and WR stars \def\nasow#1{in: O stars and Wolf-Rayet Stars, eds. P.S. Conti, A.B. Underhill, NASA SP-497, p.~#1} %Boulder Munich Workshop \def\bmw#1{in: Properties of Hot Luminous Stars, ed. C. D. Garmany Astron. Soc. Pacific Conf. Series, Vol. 7, p.~#1} %Physics of luminous blue variables. \def\plbv#1{in: Physics of Luminous Blue Variables, eds. K. Davidson, A.F.J. Moffat, H.J.G.L.M.Lamers, Kluwer,Dordrecht, p.~#1} %The Nuclei of Nomal Galaxies: Lessons from the Galactic Center \def\ringb#1{in: The Nuclei of Nomal Galaxies: Lessons from the Galactic Center, eds. R. Genzel, A.I. Harris, Kluwer,Dordrecht, p.~#1} \reference Allen, D.A., Sanders, R.H., 1986, NATURE, 319, 191 % \reference Bahcall, J.N., Tremaine, S., 1981, \apj, 244, 805 % \reference Crowther, P.A., Hillier, D.J., Smith, L.J., 1995, \aap, 293, 172 %Fundamental parameters of Wolf-Rayet stars: I. Ofpe/WN9 stars \reference Geballe, T.R., Genzel, R., Krabbe, A., Krenz, T., Lutz, D., 1994, \irarr{73} % \reference Genzel, R., Hollenbach, D.J., Townes, C.H., 1994, Rep.Progr.Phys., 57, 417 % \reference Haller, J.W., Rieke, M.J., 1989, in The Center of the Galaxy, ed. M.Morris, Kluwer, Dordrecht, 487 % \reference Hamann, W.R., Koesterke, L., Wessolowski, U., 1993, \aap, 274, 397 %Spectral Analyses of the Galactic Wolf.Rayet stars: a comprehensive %study of the WB class \reference Hillier, D.J., 1985, \aj, 90, 1514 %Infrared spectra of WN stars II. WN7 and WN8 stars. \reference Hillier, D.J., 1987a, \apjs, 63, 947 %Modelling the extended atmospheres of WN stars \reference Hillier, D.J., 1990, \aap, 231, 116 % An iterative method for the solution of the statistical and radiative % equilibrium equations in expanding atmospheres. \reference Hillier, D.J., 1995, \iaau{116} % e.s. paper \reference Krabbe, A., Genzel, R., Drapatz, S., Rotaciuc, V., 1991, \apjl, 382, L19 % A cluster of HeI emission line stars in the Galactic Center \reference Krabbe, A., Stenberg, A., Genzel, R., 1994, \apj, 424, 72 % \reference Krabbe, A., Genzel, R., Eckart, A., Najarro, F., Lutz, D., Cameron, M., Kroker, H., Tacconi-Garman, L.E., Thatte, N., Weitzel, L., Drapatz, S., Geballe, T., Sternberg, A., Kudritzki, R. , 1995, \apjl, 447, L95 % \reference Langer, N., Hamann, W.-R., Lennon, M., Najarro, F., Pauldrach, A.W.A., Puls, J., 1994, \aap, 290, 819 %Towards an understanding of very massive stars. A new evolutionary %scenario relating O stars, LBVs and Wolf-Rayet stars \reference Morris, M., 1993, \apj, 408, 496 % \reference Najarro, F., Hillier, D.J., Kudritzki, R.P., Krabbe, A., Genzel, R., Lutz, D., Drapatz, S., Geballe, T.R., 1994a, \aap, 285, 573 % \reference Najarro, F., 1995, PhD Thesis, University of Munich % \reference Rieke, G.H., Lebofsky, M.J., 1982, in AIP Conference Proceedings 83: The Galactic Center, eds. G. Riegler, R. Blandford, New York, 194 % \end{references} \end{document}