Hot prominence detected in the core of a coronal mass ejection

Context. We study an erupting prominence embedded in the core of a coronal mass ejection that occurred on August 2, 2000, and focus on deriving the plasma filling factor of the prominence. Aims. We explore two methods for measuring this factor along the line of sight. They are based on a combination of visible-light and ultraviolet spectroscopic observations. Methods. Theoretical relationships for resonant scattering and collisional excitation were used to evaluate the intensity of the neutral hydrogen Lyman-α and Lyman-β lines in two prominence points where simultaneous and cospatial LASCO-C2 and UVCS data were available. Thermodynamic and geometrical parameters assumed for the calculation (i.e., electron column density, kinetic temperature, flow velocity, chromospheric Lyα and Lyβ intensities and profiles, and thickness of the prominence along the line of sight) are provided by both observations and the results of a detailed 1D non-local thermal equilibrium (non-LTE) radiative-transfer model of the prominence, developed in our previous work. The geometrical filling factor was derived from comparing the calculated and measured intensities of the two lines. The results were then checked against the non-LTE model in order to verify the reliability of the methods. Results. The resulting filling factors are consistent with the model in both prominence points when the radiative and collisional components of the total intensity of the hydrogen lines are separated using the Lyα and Lyβ line intensities, which is required to estimate the filling factor. The exploration of the parameter space shows that the results are weakly sensitive to the plasma flow velocity, but depend more strongly on the assumed kinetic temperatures. Conclusions. The combination of visible-light and ultraviolet Lyα and Lyβ data can be used to approximately estimate the line-of-sight geometrical filling factor in erupting prominences, but the proposed technique, which is model dependent, is reliable only for emission that is optically thin in the lines considered, a condition that is not in general representative of prominence plasma.