The final 82 HD 48948 measurements provide 62.9–83.4 percent of 38-day amplitude information with white noise and seasonal rotation at 42.6 days

We measured how the final observing season helps separate a 38-day velocity curve from stellar rotation in public HD 48948 data. We used all 189 measurements in the YARARA table from Dalal et al. (2024). The mean model has seasonal constants and global curves at 7.34013 and 151 days. A rotation curve has a separate amplitude and phase per season. Matrix projection gives exact amplitude information bounds over all fixed phases of the 38-day curve. The main case uses rotation at 42.6 days, white noise, and equal variance. The final 82 measurements provide 62.9–83.4 percent of the information after the rotation fit. They are 43.4 percent of the rows. Relative measurement-error weights give bounds of 52.9–80.7 percent for YV2 and 56.7–80.6 percent for YVA. Seven fixed noise shapes and eight mean models are preserved. Conditional velocity tests favor an added 38-day term in the basic model with white noise. The tests weaken with strong noise correlation or rotation changes within a season. This is a retrospective sampling audit. It does not confirm a planet or estimate the true stellar noise. The package contains source tables, provenance, code, full results, figures, and numerical checks.

HD 48948 · radial velocity · stellar rotation · observation times · planet candidates · Fisher information · seasonal sampling · reproducible research