The primary Kepler Q1 to Q17 test leaves adjacent-orbit memory in KOI-2700 b uncertain under matched noise

We test whether fitted dust transit strengths of KOI-2700 b have memory between adjacent orbits. We use 17 public Kepler DR25 long cadence light curves. We fit a fixed asymmetric profile in transit and control windows. We then remove a separate linear trend from each quarter. The primary sample has 1,266 paired orbit fits and 1,104 adjacent pairs. After projected formal noise and control covariance are subtracted, the excess adjacent covariance is 10,563 ppm squared. A 10,000-resample, 12-orbit block interval runs from -3,013 to 23,936 ppm squared at the approximate 95% level. The excess residual variance is 49,217 ppm squared, with an interval from 29,022 to 69,732 ppm squared. All five tested block choices leave zero in the primary covariance interval. Five of 12 alternative fit settings give positive intervals that exclude zero. Thus, the primary test is uncertain, and interval status depends on the fit setting. A physical interpretation requires matched residual noise in the transit and control windows. The intervals are conditional on the fitted trends. They do not include model selection. Artificial signals confirm the projected calculation. Trend removal gives even an independent signal a small negative covariance. This result does not prove independent dust release or measure a dust lifetime. The package contains the original data, source hashes, code, artificial-signal checks, results, and figures.

KOI-2700 b · Kepler · dust transit · transit variability · covariance · control windows · reproducible analysis