The flux correction accounts for the dips at the catalog times of TOI-4319.01 and TOI-4509.01 in TESS Sector 34

We tested two public planet-candidate events with catalog times 8.92 minutes apart and stars 8.75 degrees apart. Both use TESS Sector 34 Camera 3 CCD 3. Their corrected core depths are 7664 ± 559 and 4005 ± 389 parts per million (ppm). The background-subtracted aperture flux has no comparable decrease. On a common target flux scale, the aperture contrasts are -547 ± 591 and -658 ± 404 ppm. The delivered correction adds 8211 and 4663 ppm. A constant and eight published trend vectors predict this correction with coefficients fitted outside a 16.8-hour event window. Within the central 12 hours, the prediction errors are 32.9 and 12.9 ppm. Baseline changes and calibrated pixel sums retain the main difference. All 24 controls from a fixed metadata selection are preserved. Most have no deep dip; one has an opposite-sign correction feature. The evidence supports a processing origin for the two tested dips. It does not confirm or exclude planets, test other catalog events, or identify a unique upstream cause. The exclusion applies to our coefficient fit, not the original basis construction. The package preserves exact input arrays, source hashes, and native target light curves and basis data. It also includes code, figures, results, and reproduction instructions.

TESS · TOI-4319.01 · TOI-4509.01 · TIC 157115010 · TIC 319610598 · single transit · flux correction · cotrending basis vectors · planet candidate vetting · reproducible data audit