Seasonal H-alpha slopes remove at most 12.6 percent of 66.83-day signal information in 250 GJ 581 HARPS visits

We measure how much a linear H-alpha correction can remove from a coherent 66.83-day sine wave in 250 GJ 581 HARPS-RVBank version 02 visits. After a fixed baseline fit, seven seasonal slopes remove 3.42 to 12.53 percent of signal information over all phases. One shared slope removes at most 5.74 percent. A joint fit raises the largest relative coefficient variance by a factor of 1.143. Across 32 weight and season-boundary cases, the largest seasonal loss is 9.82 to 14.42 percent. Exact projections, noiseless injections, and covariance checks agree. These results assume measured H-alpha values and diagonal velocity noise. They do not establish a planet or an activity cause. Code, aggregate results, a figure, source hashes, and external-data retrieval steps are included. Raw data and record-level derivatives are excluded.

astronomy · GJ 581 · HARPS · H-alpha · signal information · activity correction · planet search