Separate centers improve four template fits beyond two noise controls in 62 K2-22 Campaign 1 orbits

We test fixed transit profiles with a variable amplitude in 62 selected K2-22 Campaign 1 orbits. Separate event centers improve the fit beyond a fitted common center in K2SFF v1 and EVEREST v2.0. Three Gaussian profiles and a 46-minute box all exceed 2,000 trials under each of two sampled-noise models. A fixed box sampled at the same times also falls short. Quadratic baselines retain the result. The main score is 1.62 times the 99th null percentile for K2SFF and 3.49 times for EVEREST with three-cadence noise blocks. Removal of the five largest event gains lowers the K2SFF score below its full-sample threshold. These are conditional tests of simple templates and noise, not a detection of orbital timing changes. Dust shape changes, an untested constant asymmetric profile, transit noise or reduction errors can cause the gains. Both reductions share photons. Prior ground observations support general shape and timing changes, but do not confirm this statistic. No new planet is found. Code, aggregate results, hashes and external-data retrieval steps are included.

astronomy · K2-22 · transit timing · dust · exposure integration · noise controls