Kepler Q1 to Q16 do not resolve positive orbit-depth covariance in selected KOI-2700 transits

We test whether local transit contrasts of KOI-2700 retain positive covariance over one to five orbits. Kepler release 25 gives 653 accepted orbits and 287 adjacent pairs in each of two flux products. The aperture lag-one covariance is 1,925 ppm squared, with an approximate 95 percent block interval of -13,269 to 16,249. Corrected flux gives -2,850, with an interval of -23,416 to 18,075. Neither product resolves positive covariance. Wider bands retain 1,217 orbits and give the same null result. Two other periods also give intervals that include zero. Off-transit controls show no clear positive excess. An added first-order autoregressive contrast signal with correlation 0.6 and RMS 200 ppm passes a conditional one-lag threshold in 265 of 300 trials. At 100 ppm, only 5 of 300 trials pass. These are injections into measured contrasts, not raw flux. Noise, selection and quarter fits limit the test. We do not measure grain lifetime or find a new planet. The package gives original code, aggregate evidence, a figure, exact input hashes and a retrieval recipe. Raw data remain outside the package.

astronomy · KOI-2700 · Kepler · dust tail · temporal covariance · null result · injection recovery