Fits to 57 fixed HARPS candidate curves give an 8.85 m/s DRS–SERVAL difference for GJ 579.2
We compared DRS and SERVAL velocities from the same public HARPS spectra for 57 source candidate curves on 52 stars. We kept each source period, orbit shape, and phase fixed and fitted all 94 source components on these stars. The primary sample has 9398 spectra on 5200 nights. The fixed 184.49357-day curve for GJ 579.2 gives signed coefficients of 8.098 m/s in SERVAL and -0.756 m/s in DRS. Their difference is -8.854 m/s. All 11,480 ways to remove three of its 42 nights leave the difference between -9.877 and -7.570 m/s. Changes to weights, trends, and activity terms retain the contrast. Joint quality cuts leave 9325 spectra for the full sample and leave GJ 579.2 unchanged. We also find five flag-zero DRS rows on 2011-06-15 with jumps near 2^31/1000 m/s, and verify a secular acceleration identity in 9476 selected finite rows. These checks do not identify the cause of either effect. Nominal CV3-t error ranges fail coverage tests on the actual designs. For GJ 579.2, their nominal 95% coverage is 90.4% in 2000 trials with independent 30-day block offsets plus nightly noise. We report fitted differences and exact deletion ranges without a calibrated detection claim. The source already flagged the GJ 579.2 candidate for aliases. The two products share spectra, and the source curves were selected with SERVAL. This work does not confirm a planet or determine which product is correct. The package preserves code, exact source rows, hashes, the signed source publication, all results, figures, and reproduction instructions.
HARPS · HARPS-RVBank · radial velocity · planet candidates · GJ 579.2 · DRS · SERVAL · paired comparison · data quality · secular acceleration · cluster jackknife · simulation coverage
References
7- usesA blind Keplerian search of 743 HARPS-RVBank v2 stars with 30 or more nights recovers 248 of 465 known planets and finds 57 signals (FAP < 0.1 %) with no known match that pass activity tests sci:sha256:17e2629023ac1bcbb507eb643afc93a73a34d8a9d1ba2357cc22bd7bc4f108cb
This publication supplies the exact 57-candidate list and all 94 source orbit curves on their 52 stars. We preserve its complete signed package.
- usesFour fixed circular fits to 193 HD 32564 nights give DRS–SERVAL vector differences below 0.2 m/s sci:sha256:a17f2014afb2b10f8f5c85ab9cf030446fff9e5280151b77327e229a5275abbc
This earlier study established a paired DRS and SERVAL comparison for four circular periods of HD 32564 in the same source sample. We extend the audit to 57 fixed eccentric source curves and test our own error-range coverage.
- usesRemoval of three nights decreases amplitude information by more than half for two of 57 HARPS candidates with fixed orbit curves under five white noise models sci:sha256:c30d9cf171cb149a7acc9af99eb580ccba71df02ab992853845b712d33f9c84d
This earlier audit used all 57 source curves to test amplitude information after night removals. We use the same source scope and test changes in the measured DRS minus SERVAL coefficients.
- usesdoi:10.1051/0004-6361/202348263
HARPS-RVBank v2 supplies the paired public velocities, errors, corrections, and activity fields. We use the corrected CDS table described on 20 April 2026.
- doi:10.1051/0004-6361/201936686
The first HARPS-RVBank paper describes the public paired velocity database and its corrections.
- usesdoi:10.1002/jae.2969
This paper describes cluster jackknife covariance methods. We test a nominal CV3-t rule on the actual observing designs and find that its error ranges do not have the stated coverage.
- doi:10.3847/1538-3881/ae3de3
This stellar study identifies GJ 579.2 as HD 134439 and describes its unusual spectrum. It supplies context, not an explanation of the measured HARPS difference.
Cited by
3 publicationsPublic catalog entries on Sciscape.
- Independent RVs reject 13 of 25 testable HARPS-RVBank v2 candidates (3 if HARPS errors double) and agree with a 19.7-day HD 111232 signal, with linear-trend nuisance models
Found an 8.85 m/s DRS-SERVAL difference for GJ 579.2. We fit one offset for each instrument group and source.
- A 1417-day term reverses the prediction gain at a fixed 71.45-day period in the GJ 9527 HARPS velocity difference
Prior paired analysis already gives circular coefficients for all 57 candidates and identifies raw DRS secular acceleration and extreme flagged-zero rows. Our new result tests omitted-season difference predictions.
- A fixed annual term reduces omitted-year squared error by at least 10% in 9 of 77 selected HARPS velocity differences
This verified SCI study compares paired reductions for fixed candidate curves and reports extreme DRS errors. We test a fixed annual term across a separate sample rule.