The global characteristics of the interstellar medium (ISM) are very complex. Various phases have been detected. They consist either of neutral gas (cold or warm), or plasma (warm or hot) or a mixture of both. The spatial distributions and filling factors of all the phases are not well known. The morphology of the ISM encompasses filaments, sheets, shells and bubbles of various sizes up to several hundreds of parsecs. Some of the largest objects seem associated to the formation and evolution of OB associations. Cyclic regulating processes, involving galactic fountains, have been proposed for understanding the dynamics and energetics of the ISM (e.g. Ikeuchi 1988, Houck & Bregman 1990).
At low galactic latitudes b, on a scale up to about 1 kpc, it is possible to separate two main local HI-features from the very complex galactic background. One is Lindblad's ring or feature A (Lindblad 1967, Lindblad et al. 1973); the other one corresponds to the nearby parts of the local arm (e.g. Sandqvist & Lindroos 1976). Basically, the separation is performed by considering both the radial velocities V and the velocity dispersions .
From these two local features the most puzzling seems to be Lindblad's ring, which is closely related to the Gould Belt (GB). This is a flat system of young stars tilted at about to the galactic plane. The kinematics of the GB presents evidences of expansion. Its age should be not larger than about 60 Myr, while its size is of the order of 800 pc (Westin 1985, Lindblad et al. 1997). At least three large nearby OB-associations belong to the GB, namely Ori OB1, Per OB2 and the Sco-Cen association (e.g. Blaauw 1991, de Zeeuw et al. 1999). There are also large molecular cloud complexes associated to the GB and Lindblad's ring (e.g. Dame et al. 1987, Pöppel et al. 1994 [from here on Paper I]).
Two different scenarios have been proposed for understanding the origin of the GB system (i.e. the stars plus the associated gas). One is based on the occurrence of an energetic explosive event (cf. Blaauw 1965, Olano 1982, and Paper I). The other scenario considers collisions of high velocity clouds with the galactic disk (e.g. Franco et al. 1988, Comerón & Torra 1992, 1994, Lépine & Duvert 1994). For further details about the GB system we refer to the review by Pöppel (1997; in the following its Sect. 4.7, which is coauthored with Marronetti & Benaglia, will be called Paper II).
At intermediate and high latitudes it is expected that most of the observed HI should be local. In contrast to the case of low , there is no unambiguous kinematical criterion allowing an assignation to any one of both local features.
The scenario of an explosive event provides us an interesting tool for trying a kinematical identification of some of the local gas at intermediate and high . Already Olano suggested that the HI with , observed toward the galactic poles, should be backfalling remnants of the assumed explosive event. Moreover, in Paper I it was suggested that the well-known large hole in the distribution of the HI with low velocities, which is observed in the northern hemisphere (e.g. Wesselius and Fejes 1973, Kuntz & Danly 1996), should be a signature of this event in the warm neutral medium (WNM).
The cold neutral medium (CNM) detected by means of the 21-cm line appears as the most indicated phase to be analyzed, because: i) its smaller extent in z makes it more susceptible to a local energetic explosion event near the plane than the WNM. Actually, Lindblad's expanding ring is characterized by a narrow , and therefore belongs to the CNM; ii) unlike the WNM, the CNM was observed in many directions in absorption (e.g. the optical interstellar lines), as well as in self-absorption. This provides a better chance for deriving distances and making optical identifications; iii) usually, the accuracy of the determination of V is much less sensitive to spurious stray-radiation effects for the narrow features of the CNM than for the broad ones of the WNM (in the case of not too weak components).
In Paper I we made a systematic separation of both the neutral phases in the LISM at using the atlas of 21-cm profiles of Heiles & Habing (1974) and its southern extension by Colomb et al. (1980), complemented with other data. The velocity range considered was -40 to +40 km . In this paper the kinematical characteristics of the CNM derived in Paper I will be used as an independent check of the assumption of a former energetic explosive event in the local interstellar medium (LISM). Briefly, we recall the methods applied in Paper I.
For studying the WNM we sampled the HI-data by means of a mosaic of adjacent cells of (in galactic coordinates ) for and of , otherwise. In each cell the emission of the WNM was fitted by one mean broad Gaussian curve ( 10-14 km ). No additional broad components were required.
For studying the CNM we sampled the data from both HI-atlases. By subtracting the mean broad Gaussian curves fitted to the WNM we obtained residual profiles. Their peaks were assumed to correspond to the CNM. A statistical analysis was applied to them for deriving their radial velocities , equivalent Gaussian velocity dispersions and column densities . was referred to the local standard of rest (LSR). The distribution of peaked at 3 km . For the mean value was km . Clearly, the peaks were narrow , as expected of the emission from cold gas (e.g. Kulkarni & Heiles 1987). The galactic distributions of and were mapped with a mean sampling of about 1 square degree (cf. Paper I, Figs. 7 and 8, and Paper II, Figs. 4.32-4.37). In the following, we asume that the peaks detected in the residual HI-profiles in Paper I are a significant statistical sampling of the CNM at , and call them the observed cold clouds (OCCs).
Our aim is to compare the global kinematics of the OCCs with the computed ballistic positions and radial velocities of massive test particles ejected from an assumed local explosion center. The test particles are moving in the local galactic gravitational field, and we assume that they constitute a sample of the expanding shell produced by the explosive event. In Sect. 2 the results of Paper I for the OCCs are presented on new maps, which are more adequate for our aim. In Sect. 3 we derive ballistic orbits for the test particles. In Sect. 4 we compute the expected positions and velocities of the test particles. We plot them on our new maps of the OCCs for comparison. In Sect. 5 we consider the effects of disturbance centers on the orbits of the test particles. We also consider the large HI-hole, as well as the possibility that some HI-complexes with high and intermediate velocities were originated by the assumed explosive event. Finally, in Sect. 6 we discuss the results and give the conclusions.
© European Southern Observatory (ESO) 2000
Online publication: June 26, 2000