3D dynamical evolution of the interstellar gas in the Gould Belt

The dynamical evolution of the Gould Belt has been modelled in 3D and compared to the spatial and velocity distributions of all HI and H-2 clouds found within a few hundred parsecs from the Sun and to the Hipparcos distances of the nearby OB associations. The model describes the expansion of a shock wave that sweeps momentum from the ambient medium. It includes the effects of the Galactic differential rotation and its gravitational torque, as well as interstellar density gradients within and away from the Galactic plane, possible fragmentation and drag forces in the late stages, and an initial rotation of the system. The evolved Belt geometry and velocity field have been fitted to the directions and velocities of the nearby clouds using a maximum- likelihood test. In order to do so, local clouds have been systematically searched for in the available HI and CO surveys. The likelihood function also included a distance estimate for a subset of well- known clouds. The best fit to the data yields values for the current Belt semi- axes of ( 373 +/- 5) pc and ( 233 +/- 5) pc, and an inclination of 17.2degrees +/- 0.5degrees. These characteristics are consistent with earlier results, but a different Belt orientation has been found because of the presence of new molecular clouds and the revised distance information: the Belt centre currently lies ( 104 +/- 4) pc away from the Sun, towards the Galactic longitude l(centre) = 180.4degrees +/- 2.0degrees, and the ascending node longitude is 1(Ohm) = 296.1degrees +/- 2.0degrees. The Belt characteristics are independent of initial rotation. The present Belt rim is found to coincide with most of the nearby OB associations and H-2 clouds, but the Belt expansion bears little relation to the average association velocities and the younger ones are surprisingly found farther out from the Belt center. An initial kinetic energy of (1.0 +/- 0.1) 10(45) J and an expansion age of ( 26.4 +/- 0.4) Myr are required, in good agreement with earlier 2D estimates. The factor of 2 discrepancy that exists between the dynamical Belt age and that derived from photometric stellar ages could not be solved by adding a vertical dimension in the expansion, nor by adding drag forces and fragmentation, nor by introducing an initial rotation. Allowing the Belt to cross the Galactic disc before reaching its present position would require a longer age of 52 Myr, but the very poor fit to the data does not support this possibility.