New visible polarization data combined with existing IR and FIR polarization data are used to study how the magnetic field threading the filamentary molecular cloud GF 9 connects to larger structures in its general environment. We find that when both visible and NIR polarization data are plotted as a function of extinction, there is no evidence for a plateau or a saturation effect in the polarization at Av ~ 1.3 as seen in dark clouds in Taurus. This lack of saturation effect suggests that even in the denser parts of GF 9 we are still probing the magnetic field. The visible polarization is smooth and has a well-defined orientation. The IR data are also well defined but with a different direction, and the FIR data in the core region are well defined and with yet another direction, but are randomly distributed in the filament region. On the scale of a few times the mean radial dimension of the molecular cloud, it is as if the magnetic field were `blind to the spatial distribution of the filaments while on smaller scales within the cloud, in the core region near the IRAS point source PSC 20503+6006, polarimetry shows a rotation of the magnetic field lines in these denser phases. Hence, in spite of the fact that the spatial resolution is not the same in the visible/NIR and in the FIR data, all the data put together indicate that the field direction changes with the spatial scale. Finally, the Chandrasekhar and Fermi method is used to evaluate the magnetic field strength, indicating that the core region is approximately magnetically critical. A global interpretation of the results is that in the core region an original poloidal field could have been twisted by a rotating elongated (core+envelope) structure. There is no evidence for turbulence and ambipolar diffusion does not seem to be effective at the present time.
تحميل البحث