The Turndown of the Baryonic Tully-Fisher Relation and Changing Baryon Fraction at Low Galaxy Masses

The ratio of baryonic-to-dark matter in present-day galaxies constrains galaxy formation theories and can be determined empirically via the baryonic Tully-Fisher relation (BTFR), which compares a galaxy's baryonic mass (M (bary)) to its maximum rotation velocity (V (max)). The BTFR is well determined at M (bary) > 10(8) M (circle dot), but poorly constrained at lower masses due to small samples and the challenges of measuring rotation velocities in this regime. For 25 galaxies with high-quality data and M (bary) less than or similar to 10(8) M (circle dot), we estimate M (bary) from infrared and H i observations and V (max) from the H i gas rotation. Many of the V (max) values are lower limits because the velocities are still rising at the edge of the detected H i disks (R (max)); consequently, most of our sample has lower velocities than expected from extrapolations of the BTFR at higher masses. To estimate V (max), we map each galaxy to a dark matter halo assuming density profiles with and without cores. In contrast to noncored profiles, we find the cored profile rotation curves are still rising at R (max) values, similar to the data. When we compare the V (max) values derived from the cored density profiles to our M (bary) measurements, we find a turndown of the BTFR at low masses that is consistent with ? cold dark matter predictions and implies baryon fractions of 1%-10% of the cosmic value. Although we are limited by the sample size and assumptions inherent in mapping measured rotational velocities to theoretical rotation curves, our results suggest that galaxy formation efficiency drops at masses below M (bary) similar to 10(8) M (circle dot), corresponding to M (200) similar to 10(10) M (circle dot).