Seasonal linear activity fits remove 24 to 34 percent of a 58.43-day wave's energy in 496 HD 85512 HARPS nights

We test how linear activity correction affects a pure 58.43-day orbital wave in 496 selected HD 85512 HARPS RVBank version 02 nights. With the measured activity indicators held fixed, exact projection bounds show that seasonal slopes remove 23.53 to 33.87 percent of the wave's weighted energy after a two-span quadratic baseline. Separate slopes in two long spans remove 5.57 to 9.00 percent. These ranges cover every phase. Most controls retain at least 65.2 percent of the energy. A stronger seasonal baseline gives a lower bound of 58.9 percent relative to that baseline alone. A joint activity and orbit fit recovers noise-free injections. Its worst formal coefficient error is 1.230 times the baseline-only error under the stated independent noise model. The bounds measure a correction response, not planet detection or exclusion. Unknown stellar covariance, proxy errors and shared spectra limit physical inference. No new planet is found. Original code, aggregate results, a figure, source hashes and external retrieval steps are included.

astronomy · HD 85512 · HARPS · activity correction · signal retention · linear projection · planet search · null discovery result