An early width fit reduces DM-difference error by 39 percent for 51 later CHIME burst pairs of FRB 20220912A

We test same-day burst pairs of FRB20220912A in CHIME/FRB Catalog 2 release 20251231. A width coefficient fitted to 70 early pairs reduces mean squared DM-difference error from 8.022 to 4.919 in 51 later pairs. The reduction is 38.7 percent, but a 95 percent interval from resampling both fitting and test days is -28.7 to 66.3 percent. No one-day deletion removes the measured gain. Extra property and subset tests show limited precision. DM and width share a burst fit, so fit covariance, burst structure, selection and instrument effects can explain the relation. A slow published linear DM trend cannot alone explain the same-day differences. This is a catalog prediction check, not a new plasma measurement, source or planet. Independent telescope confirmation of the coefficient is absent. Code, retained data, hashes, full results, figures and offline instructions are included.

astronomy · CHIME · FRB20220912A · dispersion measure · burst morphology · time validation