Notes. (1) Identifiers; (2) mean LIS line velocities with their 1σ error, adopted from Henry et al. (2015), and measured here for GP 0816 and GP 1458; (3) FWHM of narrow Hβ component, determined by double-Gaussian fits to SDSS line profiles, and corrected for instrumental dispersion. We estimate a fitting error of 50 km s⁻¹; (4) FUV optical depth of dust near the wavelength of Lyα, derived from Hβ flux, FUV continuum flux, and ISM extinction; (5) intrinsic Lyα equivalent width derived from Hβ flux, FUV continuum flux, and ISM extinction; (6)–(12) best-fitting model parameters, obtained from unconstrained Lyα fitting. Median, 10th percentile and 90th percentile values of 20 models with the lowest χ² for each target; (6) Δ z = (z_fitLyα − z_SDSS)c, where z_{fitLy α} is the redshift of the best unconstrained Lyα fit, z_SDSS is the redshift derived from SDSS emission line fits (Table 1), and c is the speed of light; (7) shell expansion speed; (8) Doppler parameter; (9) neutral hydrogen column density; (10) FUV optical depth of dust in the vicinity of Lyα; (11) FWHM of intrinsic Lyα line; (12) equivalent width of intrinsic Lyα line.