“The different baryonic Tully-Fisher relations at low masses” by C. B. Brook, I. Santos-Santos, and G. Stinson
There have been a number of papers lately focusing on the low-mass end of the BTFR, biases in mass and velocity measurements, and their effect on the overall relation. This paper focuses on the effects of different methods of velocity measurement, using simulated galaxies, for the low-mass end of the BTFR. This paper uses 16 of the MaGICC simulated galaxies to generate fake observations. The simulation is based on ΛCDM, and uses parameters obtained by WMAP3.
See Figure 2 on page 4 for some simulated rotation curves. See Figure 3 on page 5 for some simulated HI line profiles.
See Figure 5 on page 6, and Table 2 on page 7 for corresponding pfit coefficients. Each type of velocity measurement gives a different BTFR. Consistent with many previous papers, the biggest discrepancies occur at the low velocity/mass end of the relation. Observed relations are plotted, using several different methodologies. Vflat shows the velocity measured at the flat part of the rotation curve, and is fairly consistent with the Lelli et. al. relation, though it has a large amount of intrinsic scatter. Vlast shows the circular velocity at the last point of the HI disc, using two different density criteria. The higher density criteria (green) does not have any strong outliers, compared to the lower density (purple), but it also doesn’t appear to have the correct slope. The W₂₀ method gives a completely wrong slope, and the W₅₀ method follows the observed relation but with large scatter. One thing to take away from the two line-width methods is that, while W₂₀ is expected to be a power law, it follows that W₅₀ should not be, based on Figure 4.