Temporal Provenance in Entropy Measurement: A Five-Study Audit of Laboratory-Time Reproducibility

Entropy is inferred through temporally ordered observations, including sampled trajectories, relaxation intervals, fitted delays, finite observation windows, temporal coarse-graining, transition intervals, and short-time approximations. These timing constructions are often reported as experimental or numerical settings, even when they materially determine the resulting entropy estimate. This paper applies a standardized four-gate Temporal Provenance Audit to five open-access studies involving: A driven colloidal particle Plastically deformed amorphous solids Local entropy production in active matter Time-dependent entropy inference from nonstationary stochastic trajectories Anharmonic Brownian gyrators For each study, the audit identifies the operative clock, extracts the materially consequential timing selectors, classifies their provenance, evaluates their effect on the entropy conclusion, and determines whether the complete timing architecture was propagated through to the final physical result. The audited cases include rapid forcing represented through slower acquisition, a multi-stage window-lag-delay pipeline, entropy inference from finite camera trajectories, interval sweeps tested against exact models, and a finite short-time estimator with demonstrated residual bias. All five studies contain attributable timing constructions that materially enter the entropy inference. However, they apply different standards of physical derivation, calibration, sensitivity testing, convergence analysis, and architecture closure. One study demonstrates stability across a tested interval domain. The others retain qualifications involving cross-resolution equivalence, same-data optimization, finite experimental resolution, or finite-time displacement. These findings do not establish that the original observations or calculations are invalid. They demonstrate that procedural reproducibility and timing-independent physical reproducibility are not identical. A calibrated laboratory clock provides an operational coordinate, but it does not by itself establish why a particular interval, lag, window, sampling rate, or limiting approximation adequately represents the underlying entropy process. The paper proposes a Temporal Provenance and Architecture Statement for future publications. This statement would disclose the origin, role, physical scale, uncertainty, sensitivity, operation order, convergence evidence, and validity domain of all timing quantities that materially affect the reported physical conclusion.