An inverse-time curve tied to the earlier mean does not fit the 2020–2024 W1 data for WD 0145+234 with the tested visit errors

We tested one inverse-time decay curve for WD 0145+234. We used ten NEOWISE visits from January 2020 through July 2024 and eight reference visits from 2014–2018. We reused a signed SCI data package. The new covariance includes target errors and correlations from a bootstrap of whole comparison sources. The curve limit is tied to the uncertain reference mean. Its pole is bounded by observations used in an earlier heuristic model. The best W1 residual score is Q = 176.81. No score in 10,000 simulations of the fitted model is this large. A separate W2 test gives Q = 77.92. The mismatch remains after each single-visit omission and under alternate calibration and quality choices. Adding independent 2% visit errors gives W1 Q = 73.54. Adding 5% gives Q = 18.64 and weakens the test. A free-limit curve improves W1 Q to 22.34, but it also has excess residuals. Its fitted limit is not evidence for a permanent level. The result applies to the tested monochromatic curve and error model. It does not reject a general collisional cascade or establish a new dust event. The package preserves source data, signed provenance, covariance matrices, fit results, simulation scores, code, figures, and offline reproduction steps.

WD 0145+234 · white dwarf debris · NEOWISE · infrared variability · inverse-time decay · model testing · correlated calibration errors · reproducible astronomy