TAC-curate or Accurate? An Investigation into the Effects of Time-sampling and Undetected Companions on the Accuracy of Radial Velocity Mass Measurements

Precise mass measurements are key to interpreting the transmission spectra of planets smaller than Neptune. Over the last five years, Doppler surveys of TESS planet candidates have diligently worked toward providing the community with such measurements. However, as illustrated by cases such as Kepler-10 c, even mass measurements of purportedly high precision may be subject to inaccuracy due to poor phase coverage or model misspecification. Inaccurate mass measurements pose the threat of misleading future observations planning and misrepresenting planet bulk composition. As TESS follow-up continues to populate the sub-Neptune regime of the mass-radius diagram, care should be taken to ensure that these mass measurements are not only precise, but robust to imperfect time-sampling and undetected companions. In this talk, we investigate the effects of time-sampling and additional, undetected companions on the accuracy of radial velocity (RV) mass measurements. First, we present new RV observations for ten systems with previously published planet mass measurements or mass upper limits. In the most extreme case, we find that a 5σ mass measurement from the literature actually overestimates the true planet mass by more than 100%. Next, we apply a data resampling framework to real RV time series and conduct a suite of injection-recovery tests on synthetic data to explore how observing cadence and model misspecification affect recovered planet mass measurements. These tests provide useful guidelines for informing future RV observing proposals and justifying time requests.