Thirty-day error blocks raise search-tail fractions for the 41-day GJ 3293 residual in three velocity reductions
We search public GJ 3293 radial velocities for an extra signal. Three reductions of 204 matched HARPS spectra retain a 41.0335-day residual after a fit of four known periods and two harmonics per period. In 2,000 conditional sign draws per case, replacing night blocks with 30-day blocks raises the search-tail fraction from 0.00050 to 0.06097 in published velocities, from 0.00250 to 0.03148 in RVBank SERVAL, and from 0.00350 to 0.01000 in RVBank DRS. The 41-day rotation period is known, and H-alpha in the same spectra varies at this period. Exact orbit fits also show model dependence. The reductions share spectra. The sign tests hold periods fixed and do not calibrate a physical activity model. No new planet or independent confirmation is reported. Code, source tables, hashes, results, figures, and offline instructions are included.
astronomy · GJ 3293 · HARPS · stellar activity · radial velocity · block bootstrap · null discovery result
References
7- usesdoi:10.1051/0004-6361/201630153
- usesdoi:10.26093/cds/vizier.36020088
- usesdoi:10.1051/0004-6361/201936686
- usesdoi:10.1051/0004-6361/202348263
- reviewsdoi:10.1051/0004-6361/201424253
- useshttps://vizier.cfa.harvard.edu/viz-bin/asu-tsv?-source=J/A%2bA/602/A88/tablea5&-out.all&-out.max=unlimited
- useshttps://exo-restart.com/wp-content/uploads/Stellar_parameters/Ver_02/GJ3293_RVs/GJ3293_HARPS_RVBank_ver02.csv
Cited by
1 publicationPublic catalog entries on Sciscape.
- reviewsTen early GJ 163 fits near 109 days give later HARPS gain ranges that include zero
Earlier campaign example of error-block and orbit-model dependence in a different target. No code or measurements reused.