2012ApJ...750...10S

Measuring the relative mass contributions of luminous and dark matter in spiral galaxies is important for understanding their formation and evolution. The combination of a galaxy rotation curve and strong lensing is a powerful way to break the disk-halo degeneracy that is inherent in each of the methods individually. We present an analysis of the 10 image radio spiral lens B1933+503 at z_l= 0.755, incorporating (1) new global very long baseline interferometry observations, (2) new adaptive-optics-assisted K-band imaging, and (3) new spectroscopic observations for the lens galaxy rotation curve and the source redshift. We construct a three-dimensionally axisymmetric mass distribution with three components: an exponential profile for the disk, a point mass for the bulge, and a Navarro-Frenk-White (NFW) profile for the halo. The mass model is simultaneously fitted to the kinematics and the lensing data. The NFW halo needs to be oblate with a flattening of a/c = 0.33^+0.07_–0.05 to be consistent with the radio data. This suggests that baryons are effective at making the halos oblate near the center. The lensing and kinematics analysis probe the inner ∼10 kpc of the galaxy, and we obtain a lower limit on the halo scale radius of 16 kpc (95% credible intervals). The dark matter mass fraction inside a sphere with a radius of 2.2 disk scale lengths is f_DM, 2.2_= 0.43^+0.10_–0.09. The contribution of the disk to the total circular velocity at 2.2 disk scale lengths is 0.76^+0.05_–0.06, suggesting that the disk is marginally submaximal. The stellar mass of the disk from our modeling is log₁₀(M_*/M_☉) = 11.06^+0.09_–0.11 assuming that the cold gas contributes ∼20% to the total disk mass. In comparison to the stellar masses estimated from stellar population synthesis models, the stellar initial mass function of Chabrier is preferred to that of Salpeter by a probability factor of 7.2.