Support for a 32.761-day extraction difference changes with the model in 195 GJ 536 HARPS exposures

We test corrected DRS minus SERVAL velocities at the reported 32.761-day period of GJ 536 c. A fixed model with mask-era offsets, known waves, a trend and Earth-velocity terms gives a 0.450 m/s difference amplitude in 195 HARPS RVBank version 02 exposures. Conditional block sign tests give null tail fractions of 0.092 to 0.112. Adding activity indices or using only the M2 mask changes the support. Across six corrected sample-model settings, amplitudes are 0.142 to 0.536 m/s and tail fractions are 0.03675 to 0.782. Thus the evidence does not establish a coherent extraction effect. Raw-product controls give weaker support. The products share spectra; no independent observation confirms the fitted difference. We do not confirm or reject the known planet or find a new one. Code, aggregate results, source hashes, a figure and retrieval steps are included. External source rows remain outside the baseline bundle.

astronomy · GJ 536 · HARPS · SERVAL · radial velocity · extraction sensitivity · null result