EDP Sciences
Free access
Volume 500, Number 2, June III 2009
Page(s) 817 - 826
Section Interstellar and circumstellar matter
DOI http://dx.doi.org/10.1051/0004-6361/200811132
Published online 11 March 2009
A&A 500, 817-826 (2009)
DOI: 10.1051/0004-6361/200811132

The broad H$\alpha$, [O III] line wings in stellar supercluster A of NGC 2363 and the turbulent mixing layer hypothesis

L. Binette1, 2, L. Drissen1, L. Úbeda1, A. C. Raga3, C. Robert1, and Y. Krongold2

1  Département de physique, de génie physique et d'optique & Centre de recherche en astrophysique du Québec, Université Laval, Québec, Qc, G1V 0A6, Canada
2  Instituto de Astronomía, Universidad Nacional Autónoma de México, Ap. 70-264, 04510 México, D.F., México
3  Instituto de Ciencias Nucleares, Universidad Nacional Autónoma de México, Ap. 70-543, 04510 México, D.F., México

Received 10 October 2008 / Accepted 2 March 2009

Context. Supercluster A in the extragalactic H II region NGC 2363 is remarkable for the hypersonic gas seen as faint extended broad emission lines with a full-width zero intensity of 7000  ${\rm km\,s^{-1}}$.
Aims. We explore the possibility that the observed broad profiles are the result of the interaction of a high-velocity cluster wind with dense photoionized clumps.
Methods. The geometry considered is that of near static photoionized condensations at the surface of which turbulent mixing layers arise as a result of the interaction with the hot wind. The approximative treatment of turbulence was carried out using the mixing length approach of Cantó & Raga. The code MAPPINGS Ic was used to derive the mean quantities describing the flow and to compute the line emissivities within the turbulent layers. The velocity projection in three dimensions of the line sources was carried out analytically.
Results. A fast entraining wind of up to $\approx$$
4300\,$ ${\rm km\,s^{-1}}$ appears to be required to reproduce the faint wings of the broad H$\alpha$ and [O III] profiles. A slower wind of 3500  ${\rm km\,s^{-1}}$, however, can still reproduce the bulk of the broad component and does provide a better fit than an ad hoc Gaussian profile.
Conclusions. Radial acceleration in 3D (away from supercluster A) of the emission gas provides a reasonable first-order fit to the broad line component. No broad component is predicted for the [N II] and [S II] lines, as observed. The wind velocity required is uncomfortably high and alternative processes that would provide comparable constant acceleration of the emission gas up to 4000  ${\rm km\,s^{-1}}$ might have to be considered.

Key words: ISM: HII regions -- line: profiles -- turbulence -- stars: winds, outflows -- stars: formation -- galaxies: clusters: general

© ESO 2009