Biases in inferring dark matter profiles from dynamical and lensing measurements

The degeneracy between disc and halo contributions in spiral galaxy rotation curves makes it difficult to obtain a full understanding of the distribution of baryons and dark matter in disc galaxies like our own Milky Way. Using mock data, we study how constraints on dark matter profiles obtained from kinematics, strong lensing, or a combination of the two are affected by assumptions about the halo model. We compare four different models: spherical isothermal and Navarro-Frenk-White halos, along with spherical and elliptical Burkert halos. For both kinematics and lensing we find examples where different models fit the data well but give enclosed masses that are inconsistent with the true (i.e., input) values. This is especially notable when the input and fit models differ in having cored or cuspy profiles (such as fitting an NFW model when the underlying dark matter distribution follows a different profile). We find that mass biases are more pronounced with lensing than with kinematics, and using both methods can help reduce the bias and provide stronger constraints on the dark matter distributions.